View
215
Download
0
Category
Preview:
Citation preview
New Design Strategies for Phosphine Organocatalysts:
Enantioselective Processes involving
Chiral Rhenium Fragments
and
Recycling involving Perfluorinated Pony Tails
Den Naturwissenschaftlichen Fakultäten
der Friedrich-Alexander-Universität Erlangen-Nürnberg
zur
Erlangung des Doktorgrades
vorgelegt von
Florian O. Seidel
aus Nürnberg
ii
Als Dissertation genehmigt von den Naturwissenschaftlichen
Fakultäten der Universität Erlangen-Nürnberg.
Tag der mündlichen Prüfung: 21.01.2009
Vorsitzender der Promotionskommission: Prof. Dr. E. Bänsch
Erstberichterstatter: Prof. Dr. J. A. Gladysz
Zweitberichterstatter: Prof. Dr. H. Gröger
iii
meiner Familie
iv
Die vorliegende Arbeit wurde am Institut für Organische Chemie der Friedrich-Alexander-
Universität Erlangen-Nürnberg in der Zeit von Januar 2005 bis Dezember 2008 unter Anleitung von
Prof. Dr. John A. Gladysz angefertigt.
v
Danksagung
Bei Herrn Prof. Dr. John A. Gladysz bedanke ich mich für den wissenschaftlichen Anstoß
und die Betreuung dieser Arbeit, wobei maximaler Freiraum für die eigene Kreativität und den
Forscherdrang blieb. Weiterhin bedanke ich mich für die wohlwollende Förderung.
Dem gesamten Arbeitskreis Gladysz danke ich für das angenehme Arbeitsklima, sowie für
viele fachliche und außerfachliche Anregungen und Aktivitäten.
Insbesondere danke ich hierbei Sabine Seidel für die Gemeinschaft und fachliche
Diskussion über den Laboralltag hinaus. Dank auch an Inka Wolf durch deren Nachbarschaft am
Schreibtisch der Erforschung der Entropie keine Grenzen gesetzt waren. Auch Dirk Skaper hat den
Büro- und Laboralltag durch seine Anwesenheit stets aufgelockert. Dank schulde ich auch Dr.
Frank Hampel, der Unmögliches möglich machte.
Ganz besonderen Dank schulde ich meinen Eltern, Biene und meinen Freunden für die
moralische Unterstützung und den Beistand während dieser Doktorarbeit.
vi
vii
Zusammenfassung
In dieser Dissertation wird die Entwicklung neuer rheniumhaltiger Phosphine beschrieben,
und deren Wirksamkeit als "Organokatalysatoren" bezüglich enantioselektiver Morita Baylis
Hillman und Rauhut Currier Reaktionen untersucht. Außerdem wird gezeigt, dass achirale fluorige
Phosphine sowohl wirksame als auch rückgewinnbare Katalysatoren für oben genannte Reaktionen
sind.
Kapitel 1 beschreibt kurz die historische Entwicklung der Morita Baylis Hillman und der
Rauhut Currier Reaktionen.
Kapitel 2 zeigt zuerst, dass bereits bekannte chirale rheniumhaltige Phosphine wirksame
Katalysatoren für die Morita Baylis Hillman und Rauhut Currier Reaktionen sind. Anschließend
werden neue rheniumhaltige Phosphine entwickelt und getestet, wobei die Reaktionen bezüglich
Geschwindigkeit und Enantioselektivität optimiert werden.
In C6H6 und PhCl katalysiert das racemische rheniumhaltige Phosphin (η5-
C5H5)Re(NO)(PPh3)(CH2PPh2) (1a; 10 mol%) eine intramolekulare Morita Baylis Hillman
Reaktion ausgehend von (E)-Ph(CO)CH=CHCH2CH2CHO (2a) zu Ph(CO)C=CHCH2CH2CHOH,
wobei Ausbeuten von 90% bis 95% nach 0.8-2.0 h bei Raumtemperatur erhalten werden.
Die ähnlichen Verbindungen (E)-R(CO)CH=CHCH2(CH2)nCHO (2; n/R = b, 1/EtO; c,
1/CH3O; d, 1/p-Tol; e, 1/CH3; f, 1/i-PrS; g, 2/p-Tol; h, 2/CH3; i, 2/EtO; j, 2/CH3O; k, 2/Ph) und
die Bis(enone) (E,E)-R(CO)CH=CHCH2(CH2)nCH=CH(CO)R (3; n/R = a, 1/Ph; b, 1/CH3; c, 1/i-
PrS; d, 2/CH3; e, 2/p-Tol) werden auf Zyklisierungen zu den Morita Baylis Hillman Produkten
R(CO)C=CHCH2(CH2)nCHOH und Rauhut Currier Produkten R(CO)C=CHCH2-
(CH2)nCHCH2(CO)R untersucht. Während einige Substrate (2a,d-h, 3a-c) mit 1a (10 mol%) gute
Reaktivitäten und Produktausbeuten (50-99%) aufweisen, werden andere Substrate (2b,c,i-k, 3d,e)
schlecht oder gar nicht umgesetzt.
Mehrere andere rheniumhaltige Phosphine mit der Formel (η5-C5H4R)Re(NO)(PPh3)-
(CR'HPR''R''') (4; R/R'/R''/R''' = a, H/CH3/Ph/Ph; b, H/Ph/Ph/Ph; c, PPh2/Ph/Ph/Ph; d,
viii
PPh2/H/Ph/Ph; e, H/H/Cy/Ph; f, H/H/Cy/t-Bu) werden synthetisiert, wobei 4a eine neue
Verbindung ist. Diese Phosphine werden als Katalysatoren für die Zyklisierungen von 2a,b,h,i,k
unter den selben Reaktionsbedingungen wie vorher getestet. Die Katalysatoren 4a-f sind weit
weniger reaktiv als 1a, trotzdem werden NMR-Ausbeuten von bis zu 90% erzielt.
Reaktionen von enantiomerenreinen (1a, 4d), diastereomerenreinen (4a) oder
diastereomerenangereicherten (4e) Katalysatoren werden mit verschiedenen Substraten
durchgeführt. Dabei werden mittels chiraler HPLC ee Werte von 3% bis 74% bestimmt. Die besten
Ergebnisse werden mit 1a erzielt (38-74% ee).
Die neuen rheniumhaltigen Phosphine (η5-C5H5)Re(NO)(PPh3)(CH2PRR') (1; R/R' = b, p-
Tol/p-Tol; c, p-C6H4OCH3/p-C6H4OCH3; d, p-C6H4N(CH3)2/p-C6H4N(CH3)2; e, 2-biphen/2-
biphen; f, α-naph/α-naph; g, Ph/α-naph; h, Ph/β-naph) werden auch enantiomerenrein (1b-d) und
diastereomerenangereichert (1g,h) synthetisiert (siehe unten). Bei Testreaktionen dieser als
Katalysatoren (10 mol%) mit verschiedenen Substraten werden im besten Fall quantitative
Produktausbeuten erzielt. Die detektierten Enantioselektivitäten variieren von –11% bis 88%, wobei
1g der selektivste Katalysator von allen ist (53-88% ee).
Kapitel 3 beschreibt die Synthese der neuen rheniumhaltigen Phosphine, welche in Kapitel 2
als Katalysatoren eingesetzt wurden. Diese werden ausgehend von dem bereits bekannten
Methylkomplex (η5-C5H5)Re(NO)(PPh3)(CH3) (5) synthetisiert. Für die Darstellung racemischer
Komplexe wird 5 mit Ph3C+ PF6– umgesetzt, wobei der Methylidenkomplex [(η5-
C5H5)Re(NO)(PPh3)(=CH2)]+ PF6– gebildet wird. Die darauffolgende Addition der sekundären
Phosphine PRR'H (6; R/R' = b, p-Tol/p-Tol; c, p-C6H4OCH3/p-C6H4OCH3; d, p-C6H4N(CH3)2/p-
C6H4N(CH3)2; e, 2-biphen/2-biphen; f, α-naph/α-naph) führt zu den Phosphoniumsalzen [(η5-
C5H5)Re(NO)(PPh3)(CH2PRR'H)]+ PF6– ([1(b-f)-H]+ PF6
–, 87-94%). Nach Deprotonierung mit t-
BuOK erhält man die reinen Phosphine 1b-f in guten Ausbeuten (64-91%). Analoge Umsetzungen,
welche allerdings mit (S)-5 durchgeführt werden, ergeben die enantiomerenreinen Phosphine (S)-
1b-d in vergleichbaren Ausbeuten. In einer ähnlichen Reaktionsabfolge werden die beiden
racemischen sekundären Phosphine PHRR' (6; R/R' = g, Ph/α-naph; h, Ph/β-naph) eingesetzt.
Ersteres führt mit 78% Ausbeute zu dem Phosphoniumsalz [1g-H]+ PF6–, welches aus einem
ix
Gemisch von SReRP/SReSP Diastereomeren besteht. Mittels mehrmaligen Ausfällens aus n-
Pentan/CH2Cl2 werden die SReRP und SReSP Diastereomere getrennt, wobei maximal 94% de
erhalten wird. Beide Diastereomere werden unter Retention der Phosphorkonfiguration deprotoniert
und der oben eingesetzte Katalysator 1g wird erhalten. Mit 6h wird auf ähnlichem Weg Komplex
(SReRP)/(SReSP)-1h (50:50) hergestellt. Ausfällen aus n-Pentan/C6H6 liefert die SReRP and SReSP
Diastereomere mit maximal 92% de.
Der C-stereogene, protonierte Katalysator (SReSC)-[4a-H]+ PF6– wird synthetisiert indem
der bekannte Ethylkomplex (S)-(η5-C5H5)Re(NO)(PPh3)(CH2CH3) mit Ph3C+ PF6– bei –78 °C
umgesetzt wird. Dabei wird der Ethylidenkomplex (sc)-[(η5-C5H5)Re(NO)(PPh3)(=CH2CH3)]+
PF6– gebildet, welcher anschließend mit PPh2H zum Produkt reagiert. Dieses wird mit 55%
Ausbeute erhalten und direkt vor der Katalysereaktion deprotoniert, ohne das freie Phosphin direkt
zu charakterisieren.
Die Synthesen der sekundären Phosphine 6c-h werden beschrieben. Von diesen
Verbindungen sind drei neu (6e,g,h), welche mittels Reduktion der entsprechenden
Diarylchlorphosphine mit LiAlH4 in 32% bis 70% Ausbeute synthetisiert werden.
Kapitel 4 zeigt den erstmaligen Einsatz von fluorigen Phosphinen als Katalysatoren für
Morita Baylis Hillman und Rauhut Currier Reaktionen. Zwei bereits bekannte fluorige Phosphine
P((CH2)3(CF2)n-1CF3)3 (7; a, n = 6; b, n = 8) werden dafür untersucht. Eine mögliche
Rückgewinnung mittels Ausfällen wird getestet. Die Löslichkeit jener Phosphine ist in CH3CN bei
Raumtemperatur eher gering, steigt bei höheren Temperaturen (ca. 50-60 °C) aber deutlich an.
Demzufolge wird 2a in CH3CN gelöst und 7a,b (10 mol%) zugegeben. Die bei 60-64 °C
stattfindende Reaktion wird im Fall von 7b mittels HPLC verfolgt. Nachdem die Reaktionen
beendet sind und die Lösungen abgekühlt werden, fallen 7a,b wieder aus und werden in je einem
neuen, identischen Zyklus wiederverwendet. Das Reaktionsprodukt wird aus der überstehenden
Lösung isoliert. Da die Rückgewinnung von 7a nur moderat funktioniert, wird im folgenden dem
weniger löslichen 7b mehr Bedeutung beigemessen. Der Reaktionsverlauf von 2a mit 7b (10 mol%)
wird über fünf Zyklen hinweg verfolgt. Produktausbeuten von 78% bis 85% werden erzielt, wobei
die Aktivität pro Zyklus kaum abnimmt.
x
Wenn gelöste fluorige Verbindungen abgekühlt werden, adsorbieren diese sehr leicht auf
fluorigen Feststoffen. In anderen Experimenten werden daher den Reaktionsmischungen fluorige
Polymere beigegeben (Teflon® Band und Gore-Rastex® Faser) um den Wiedergewinnungsprozess
effizienter zu gestalten. Je fünf Zyklen werden durchgeführt, wobei Ausbeuten von 66% bis 82%
(Teflon® Band) und 74% bis 82% (Gore-Rastex® Faser) erhalten werden. Allerdings wird ein
Aktivitätsverlust während des vierten und fünften Zyklus festgestellt (Teflon® Band > Gore-
Rastex® Faser). Der Einsatzbereich von 7b wird mit drei Substraten (2f,g, 3c) erweitert. Je drei
Zyklen werden mittels HPLC oder GC verfolgt. Alle Reaktionen werden mit 10 mol% 7b in
CH3CN bei 60-72 °C durchgeführt. Ausbeuten von 71% bis 96% werden erzielt, wobei für jeden
folgenden Zyklus nur geringe Aktivitätsverluste beobachtet werden.
Kapitel 5 beschreibt eine einfache und extrem flexible Synthese für die Substrate 2 und 3.
Ausgehend von leicht erhältlichen Molekülbausteinen wie α-Bromacetylbromid oder
Methylketonen werden die α-Bromcarbonyle R(CO)CH2Br (R = EtO, CH3O, i-PrS, p-Tol, Ph, CH3;
36-90%) synthetisiert. Diese reagieren mit PPh3 zu den Phosphoniumsalzen [R(CO)CH2PPh3]+ Br–
in 60% bis 92% Ausbeute. Anschließende Deprotonierungen mit wässriger NaOH führen zu den
stabilen Yliden R(CO)CHPPh3 in 52% bis 85% Ausbeute. Die Ylide reagieren nach Wittig mit den
Dialdehyden OHCCH2(CH2)nCHO (n = 1, 2) je nach Stöchiometrie entweder zu den Morita Baylis
Hillman Substraten (2; 34-75%) oder zu den Rauhut Currier Substraten (3; 40-82%).
xi
Abstract
In this thesis, new chiral rhenium-containing phosphines are developed, and their
effectiveness as "organocatalysts" for enantioselective Morita Baylis Hillman and Rauhut Currier
reactions are assayed. Achiral fluorous phosphines are also shown to be effective as well as
recyclable catalysts for these reactions.
Chapter 1 briefly describes the historical development of the Morita Baylis Hillman and
Rauhut Currier reactions.
Chapter 2 first establishes that previously reported chiral rhenium-containing phosphines are
effective catalysts for the Morita Baylis Hillman and Rauhut Currier reactions. New rhenium-
containing phosphines are then developed and screened, optimizing rates and product
enantioselectivities.
In C6H6 and PhCl, the racemic rhenium-containing phosphine (η5-C5H5)Re(NO)(PPh3)-
(CH2PPh2) (1a; 10 mol%) effects an intramolecular Morita Baylis Hillman cyclization of (E)-
Ph(CO)CH=CHCH2CH2CHO (2a) to Ph(CO)C=CHCH2CH2CHOH in 90% to 95% yields after
0.8-2.0 h at room temperature.
The related compounds (E)-R(CO)CH=CHCH2(CH2)nCHO (2; n/R = b, 1/EtO; c, 1/CH3O;
d, 1/p-Tol; e, 1/CH3; f, 1/i-PrS; g, 2/p-Tol; h, 2/CH3; i, 2/EtO; j, 2/CH3O; k, 2/Ph) and the
bis(enones) (E,E)-R(CO)CH=CHCH2(CH2)nCH=CH(CO)R (3; n/R = a, 1/Ph; b, 1/CH3; c, 1/i-PrS;
d, 2/CH3; e, 2/p-Tol), are screened with respect to the Morita Baylis Hillman products
R(CO)C=CHCH2(CH2)nCHOH and Rauhut Currier products R(CO)C=CHCH2(CH2)nCH-
CH2(CO)R. Several substrates (2a,d-h, 3a-c) exhibit good reactivity towards 1a (10 mol%) with
good product yields (50-99%), while others show little or no reactivity (2b,c,i-k, 3d,e).
Several other rhenium-containing phosphines of the formulae (η5-C5H4R)Re(NO)-
(PPh3)(CR'HPR''R''') (4; R/R'/R''/R''' = a, H/CH3/Ph/Ph; b, H/Ph/Ph/Ph; c, PPh2/Ph/Ph/Ph; d,
PPh2/H/Ph/Ph; e, H/H/Cy/Ph; f, H/H/Cy/t-Bu) are prepared, the first one (4a) of which is a new
xii
compound. These are screened as catalysts for the cyclization of 2a,b,h,i,k under analogous
conditions. Catalysts 4a-f are much less active than 1a, but with 4a,d,e NMR yields up to 90% are
obtained.
Reactions of enantiopure (1a, 4d) and diastereopure (4a) or diastereoenriched (4e) catalysts
with various substrates are conducted under analogous conditions. Chiral HPLC establishes ee
values ranging from 3% to 74%. The best results are obtained with 1a (38-74% ee).
The new rhenium-containing phosphines (η5-C5H5)Re(NO)(PPh3)(CH2PRR') (1; R/R' = b,
p-Tol/p-Tol; c, p-C6H4OCH3/p-C6H4OCH3; d, p-C6H4N(CH3)2/p-C6H4N(CH3)2; e, 2-biphen/2-
biphen; f, α-naph/α-naph; g, Ph/α-naph; h, Ph/β-naph) are prepared (below) in enantiopure (1b-d)
and diastereoenriched (1g,h) form. These are screened (10 mol%) with various substrates, and give
up to quantitative product yields. Enantioselectivities range from –11% to 88%, with 1g being most
effective (53-88% ee).
Chapter 3 describes the preparation of the new rhenium-containing phosphines used as
catalysts in Chapter 2. These are synthesized from the previously reported methyl complex (η5-
C5H5)Re(NO)(PPh3)(CH3) (5). For racemic complexes, 5 is treated with Ph3C+ PF6– to generate
the methylidene complex [(η5-C5H5)Re(NO)(PPh3)(=CH2)]+ PF6–. The subsequent addition of the
secondary phosphines PRR'H (6; R/R' = b, p-Tol/p-Tol; c, p-C6H4OCH3/p-C6H4OCH3; d, p-
C6H4N(CH3)2/p-C6H4N(CH3)2; e, 2-biphen/2-biphen; f, α-naph/α-naph) give the phosphonium
salts [(η5-C5H5)Re(NO)(PPh3)(CH2PRR'H)]+ PF6– ([1(b-f)-H]+ PF6
–, 87-94%). Deprotonations
with t-BuOK give the pure phosphines 1b-f in good yields (64-91%). Analogous procedures
starting with (S)-5 give the enantiopure phosphines (S)-1b-d in comparable overall yields. Identical
sequences are conducted using the racemic secondary phosphines PHRR' (6; R/R' = g, Ph/α-naph; h,
Ph/β-naph). The first gives the phosphonium salt [1g-H]+ PF6– in 78% total yield as a mixture of
SReRP/SReSP diastereomers. Precipitations from n-pentane/CH2Cl2 allow the SReRP and SReSP
diastereomers to be obtained in up to 94% de. Both undergo deprotonation with retention of
configuration at phosphorus to give the 1g employed for catalysis above. With 6h, (SReRP)/(SReSP)-
1h (50:50) is similarly obtained. Precipitations from n-pentane/C6H6 give the SReRP and SReSP
diastereomers in up to 92% de.
xiii
The C-stereogenic protonated catalyst (SReSC)-[4a-H]+ PF6– is obtained by the reaction of
the previously reported ethyl complex (S)-(η5-C5H5)Re(NO)(PPh3)(CH2CH3) with Ph3C+ PF6– at
–78 °C to generate the ethylidene complex (sc)-[(η5-C5H5)Re(NO)(PPh3)(=CH2CH3)]+ PF6–,
followed by the addition of PPh2H. This material is obtained in 55% yield and directly deprotonated
before catalysis without characterization of the free phosphine.
The syntheses of the secondary phosphines 6c-h are described. Three of them are new
(6e,g,h), and are obtained by the reduction of the corresponding diarylchlorophosphines with
LiAlH4 in 32% to 70% yields.
Chapter 4 describes the first use of fluorous phosphines as catalysts for Morita Baylis
Hillman and Rauhut Currier reactions. Two previously reported fluorous phosphines P((CH2)3-
(CF2)n-1CF3)3 (7; a, n = 6; b, n = 8) are tested for their suitability for recovery by precipitation.
They are only slightly soluble in CH3CN at room temperature, but become much more soluble at
elevated temperatures (ca. 50-60 °C). Therefore, CH3CN solutions of 2a are treated with 7a,b (10
mol%) at 60-64 °C and the reaction with 7b is monitored by HPLC. Upon completion and cooling,
7a,b precipitate and are reused in an identical cycle. The product is isolated from the supernatant.
The recovery with 7a is low, and therefore attention is focused on the less soluble 7b. The rate of
the reaction of 2a with 7b (10 mol%) is monitored for five cycles. Product yields range from 78%
to 85%, with only a slight loss of activity.
In parallel experiments, fluoropolymers (Teflon® tape and Gore-Rastex® fiber) are added to
mechanically facilitate the recycling process, as fluorous compounds easily adsorb onto such
polymers upon cooling. Five reaction cycles are accomplished using each recycling strategy, with
product yields ranging from 66% to 82% (Teflon® tape) and 74% to 82% (Gore-Rastex® fiber).
However, some loss of activity is seen in the fourth and fifth cycle (Teflon® tape > Gore-Rastex®
fiber). The scope of these protocols is broadened with three further substrates (2f,g, 3c). The rates of
three cycles, each with 10 mol% 7b in CH3CN at 60-72 °C, are monitored by HPLC or GC. The
yields range from 71% to 96%, with only a slight loss of activity.
Chapter 5 describes a simple and extremely versatile synthetic route to the substrates 2 and 3
xiv
employed in Chapters 2 and 4. Starting with readily accessible building blocks such as α-
bromoacetyl bromide or methyl ketones, α-bromocarbonyl derivatives R(CO)CH2Br (R = EtO,
CH3O, i-PrS, p-Tol, Ph, CH3; 36-90%) are prepared. These react with PPh3 to give the
phosphonium salts [R(CO)CH2PPh3]+ Br– in 60% to 92% yields. Subsequent deprotonations with
aqueous NaOH give the stable ylides R(CO)CHPPh3 in 52% to 85% yields. The ylides undergo
Wittig reactions with the dialdehydes OHCCH2(CH2)nCHO (n = 1, 2) to give either Morita Baylis
Hillman substrates (2; 34-75%) or Rauhut Currier substrates (3; 40-82%), depending on the
stoichiometry.
xv
Table of Contents
Zusammenfassung..............................................................................................................................vii
Abstract ...............................................................................................................................................xi
List of Schemes.................................................................................................................................xix
List of Figures ...................................................................................................................................xxi
List of Tables .................................................................................................................................xxvii
1 General Introduction ...................................................................................................................1
1.1 Organocatalysis – a New Chapter within Catalytic Chemistry............................................1
1.2 The Historical Rauhut Currier and Morita Baylis Hillman Reaction ..................................3
1.3 The Intramolecular Phosphine Catalyzed Rauhut Currier Reaction....................................5
1.4 The Intramolecular Phosphine Catalyzed Morita Baylis Hillman Reaction........................8
1.5 Future Directions................................................................................................................11
2 Enantioselective Organocatalysis with Rhenium-Containing Phosphines ...........................13
2.1 Introduction........................................................................................................................13
2.2 The Intramolecular Morita Baylis Hillman and Rauhut Currier Reaction:
First Experiments ...............................................................................................................19
2.2.1 Results ........................................................................................................................19
2.2.1.1 A First Reaction Series ......................................................................................... 19
2.2.1.2 Solvent Dependency ............................................................................................. 20
2.2.1.3 Temperature Dependency ..................................................................................... 21
2.2.1.4 Substrate Dependency........................................................................................... 23
2.2.1.5 Reactions with Other Rhenium-Containing Phosphines....................................... 26
2.2.1.6 Enantioselective Catalyses.................................................................................... 29
2.2.2 Discussion ..................................................................................................................31
2.2.2.1 Solvent Dependency ............................................................................................. 31
2.2.2.2 Temperature Dependency ..................................................................................... 33
2.2.2.3 Substrate Dependency........................................................................................... 33
xvi
2.2.2.4 Formation of a Side Product ................................................................................. 34
2.2.2.5 Oxidation of the Catalyst ...................................................................................... 35
2.2.2.6 Enantioselective Catalyses.................................................................................... 38
2.2.2.7 Activity of Different Phosphorus Centers............................................................. 39
2.2.2.8 Conclusions ..........................................................................................................41
2.3 First Series of Improved Catalysts .....................................................................................42
2.3.1 Results ........................................................................................................................42
2.3.1.1 Increased Nucleophilicity by Electron Rich Aryl Substituents ............................ 42
2.3.1.2 Enantioselective Catalyses.................................................................................... 46
2.3.1.3 Preparative Experiments ....................................................................................... 47
2.3.2 Discussion ..................................................................................................................49
2.4 Second Series of Improved Catalysts.................................................................................51
2.4.1 Results ........................................................................................................................51
2.4.1.1 Increased Steric Demand at the Catalytic Center.................................................. 51
2.4.1.2 Introduction of Chirality at the Catalytic Center .................................................. 53
2.4.1.3 Enantioselective Catalyses.................................................................................... 55
2.4.2 Discussion ..................................................................................................................57
2.4.2.1 Discussion of the Second Series Improved Catalysts ........................................... 57
2.4.2.2 Short Summary and Outlook................................................................................. 58
2.5 Experimental ......................................................................................................................60
2.5.1 General Data...............................................................................................................60
2.5.2 General Procedures.....................................................................................................61
2.5.3 Analytic Experiments Listed by Figures ....................................................................63
2.5.4 Preparative Reactions .................................................................................................74
2.5.5 Isolation of Side Product C=CHCH2CH2CHCH2C(CH3)(OH)CH2(CO) (19) .........76
3 Rhenium-Containing Phosphines .............................................................................................77
3.1 Introduction to Rhenium Complexes .................................................................................77
3.2 Results................................................................................................................................81
xvii
3.2.1 Preparation of Rhenium Complexes...........................................................................81
3.2.1.1 Preparation of Racemic Rhenium-Containing Phosphines................................... 81
3.2.1.2 Preparation of Enantiopure Complexes ................................................................ 85
3.2.1.3 Preparation of Diastereomeric Mixtures of P Chiral Complexes ......................... 86
3.2.1.4 Preparation of (SReSC)-[(η5-C5H5)Re(NO)(PPh3)(CHCH3PPh2H)]+ PF6–
((SReSC)-[15b-H]+ PF6–) ....................................................................................... 91
3.2.2 Preparation of Secondary Phosphines ........................................................................92
3.2.2.1 Preparation via Reduction of Secondary Phosphine Oxides................................. 93
3.2.2.2 Preparation via Reduction of Diarylchlorophosphines ......................................... 93
3.3 Discussion ..........................................................................................................................95
3.4 Experimental ......................................................................................................................97
3.4.1 General Data...............................................................................................................97
3.4.2 Preparation of Rhenium-Containing Phosphines .......................................................98
3.4.3 Preparation of Secondary Phosphines ......................................................................113
4 Recycling of Fluorous Phosphines from Organocatalytic Reactions ..................................117
4.1 Introduction......................................................................................................................117
4.1.1 Introduction of Fluorous Concepts and Recycling...................................................117
4.1.2 Introduction of Fluorous Phosphines .......................................................................123
4.2 Results..............................................................................................................................126
4.2.1 Catalytic Reactions with P((CH2)3Rf6)3 (30a) ........................................................126
4.2.2 Catalytic Reactions with P((CH2)3Rf8)3 (30b) ........................................................129
4.2.2.1 Reactions with Substrate C1Ph........................................................................... 129
4.2.2.2 Reactions with Substrates C1S(i-Pr) and C1S(i-Pr)2......................................... 131
4.2.2.3 Reactions with Substrate C2(p-Tol) ................................................................... 133
4.2.2.4 Preparative Experiments ..................................................................................... 134
4.2.3 Recycling of 30b by Precipitation............................................................................135
4.2.3.1 Recycling from Reactions with C1Ph................................................................. 135
4.2.3.2 Recycling from Reactions with C1S(i-Pr), C1S(i-Pr)2, and C2(p-Tol)............. 137
xviii
4.2.4 Recycling of 30b with Fluoropolymer Supports......................................................140
4.2.4.1 Recycling with the Support of Teflon® Tape..................................................... 141
4.2.4.2 Recycling with the Support of Gore-Rastex® Fiber........................................... 143
4.2.5 Thermomorphic Behavior of 30b in CH3CN...........................................................145
4.3 Discussion ........................................................................................................................147
4.4 Experimental ....................................................................................................................149
4.4.1 General Data.............................................................................................................149
4.4.2 Analytic Experiments Listed by Figures or Tables ..................................................149
4.4.2.1 Experiments with Catalyst 30a ........................................................................... 149
4.4.2.2 Experiments with Catalyst 30b ........................................................................... 150
4.4.3 Preparative Reactions ...............................................................................................155
4.4.4 Thermomorphic Behavior of 30b.............................................................................156
5 Preparation of Substrates........................................................................................................157
5.1 Introduction......................................................................................................................157
5.2 Results..............................................................................................................................160
5.2.1 Preparation of the Stable Ylides...............................................................................160
5.2.2 Wittig Reactions of the Stable Ylides with Dialdehydes .........................................162
5.3 Discussion ........................................................................................................................164
5.4 Experimental ....................................................................................................................166
5.4.1 General Data.............................................................................................................166
5.4.2 Ylide Preparation......................................................................................................166
5.4.3 Substrate Preparation................................................................................................171
5.4.3.1 Morita Baylis Hillman Substrates ....................................................................... 171
5.4.3.2 Rauhut Currier Substrates ................................................................................... 173
6 References and Notes ...............................................................................................................177
xix
List of Schemes
Scheme 1.1. Dimerization of an alkyl acrylate (I), mediated by tertiary phosphines; the
"historical" Rauhut Currier reaction.
Scheme 1.2. Addition of PPh3 to acrylonitrile gives a zwitterionic species III which yields to the
ylide IV.
Scheme 1.3. Wittig reaction of ylide IV with aromatic aldehydes.
Scheme 1.4. Reaction sequence published by Morita. The double bond is activated by an electron
withdrawing group (EWG). The moiety R can be aliphatic as well as aromatic.
Scheme 1.5. Reaction cycle for the phosphine catalyzed intramolecular Rauhut Currier reaction.
Scheme 1.6. Intramolecular Rauhut Currier reaction published by Krische.
Scheme 1.7. The Rauhut Currier reaction as a key step for the total synthesis of (–)-spinosyn A
reported by Roush.
Scheme 1.8. Reaction cycle for the phosphine catalyzed intramolecular Morita Baylis Hillman
reaction.
Scheme 1.9. First intramolecular Morita Baylis Hillman reaction carried out by Fráter.
Scheme 1.10. Representative reaction reported by Murphy.
Scheme 2.1. Enhanced basicity of the lone pairs of (η5-C5H5)Re(NO)(PPh3)(CH2SCH3) (1).
Scheme 2.2. First intramolecular Morita Baylis Hillman reactions with C1Ph and 10 mol% of 14a,
carried out in different solvents.
Scheme 2.3. Formation of cyclic side products 19 and 20 from the reaction of C1Me2 with 10
mol% 14b in PhCl.
Scheme 2.4. Possible pathway for the formation of catalyst-derived phosphorus oxide.
Scheme 2.5. Possible future directions for catalysts based upon rhenium-containing phosphines.
Scheme 3.1. Syntheses of rhenium-containing phosphines with chiral centers. Enantioselective
(and diastereoselective) syntheses are possible in each case. a) Ph3C+ PF6–, CH2Cl2.
b) PPh2H. c) t-BuOK, THF. d) n-BuLi, THF. e) PPh2Cl. f) PhLi. g) t-BuLi. h)
HBF4.OEt2, PhCl.
xx
Scheme 3.2. Preparation of the protonated racemic rhenium-containing phosphines [14-H]+ PF6–.
a) Ph3C+ PF6– (1.1 equiv.), CH2Cl2. b) PR2H.
Scheme 3.3. Deprotonation of the racemic phosphonium salts [14-H]+ PF6–. a) t-BuOK (1.5
equiv.), C6H6.
Scheme 3.4. Preparation of the protonated diastereomeric rhenium-containing phosphines
(SReRP)/(SReSP)-[17-H]+ PF6–. a) Ph3C+ PF6
–, CH2Cl2. b) PRR'H.
Scheme 3.5. Preparation of (SReSC)-[15b-H]+ PF6–. a) Ph3C+ PF6
–, CH2Cl2, –78 °C. b) PPh2H.
Scheme 3.6. Syntheses of the secondary phosphines 28c,d (yields overall). a) Mg, THF. b)
OP(OEt)2H. c) DIBAL-H. d) NaOH/H2O.
Scheme 3.7. Synthesis of the secondary phosphines 28e,f. a) Mg, THF. b) PCl3. c) LiAlH4. d)
NaOH/H2O.
Scheme 3.8. Synthesis of the racemic secondary phosphines 29c,d. a) Mg, THF. b) PPhCl2. c)
LiAlH4. d) NaOH/H2O.
Scheme 4.1. General procedure for the syntheses of tertiary fluorous phosphines 30a,b.
Scheme 4.2. Intramolecular Morita Baylis Hillman reaction of C1Ph catalyzed by 10 mol% of
fluorous phosphine 30a.
Scheme 4.3. Cyclizations of substrates C1S(i-Pr) and C1S(i-Pr)2 catalyzed by 10 mol% of 30b.
Scheme 4.4. Cyclization of substrate C2(p-Tol) catalyzed by 10 mol% of 30b.
Scheme 4.5. Reaction sequence with recyclable catalyst system, reported by Yi.
Scheme 5.1. General route to substrates C1R, published by Koo.
Scheme 5.2. General route to substrates C1R, published by Denmark.
Scheme 5.3. General route to substrates C1R, published by Murphy.
Scheme 5.4. Preparation of the stable ylides 34.
Scheme 5.5. Preparation of the substrates CnR and CnR2, by reaction of ylides and dialdehydes.
xxi
List of Figures
Figure 1.1. Count of publications by year containing the topic "organocatalysis". Result from
SciFinder® search (07.2008).
Figure 1.2. Key features of the highly functionalized intramolecular Morita Baylis Hillman
product XVII.
Figure 2.1. General idealized structures of the rhenium fragment [(η5-C5H5)Re(NO)(PPh3)]+ (A)
and (η5-C5H5)Re(NO)(PPh3)(X) (B), upon which many rhenium complexes are
based.
Figure 2.2. Key orbital interactions in coordinatively saturated octahedral LnMPR2 (C) and
LnMCH2PR2 (D, E) complexes.
Figure 2.3. Rhenium-containing monophosphines that have been employed successfully in
catalysis.
Figure 2.4. Adducts of rhenium-containing diphosphines that have been employed successfully
in catalysis.
Figure 2.5. Reaction profiles for the formation of C1Phprod in Scheme 2.2 in different solvents.
The purple graph (♦) shows for comparison the reaction of C1Ph with 10 mol% of
PPh3 in CH2ClCH2Cl.
Figure 2.6. Reaction profiles for Scheme 2.2 at room temperature and 0 °C in CH3CN. The
former experiment is also in Figure 2.5.
Figure 2.7. Reaction profiles for Scheme 2.2 at room temperature and 40 °C in CH2ClCH2Cl.
The former experiment is also in Figure 2.5.
Figure 2.8. Substrates employed herein for catalytic reactions, and product structures. Substrates
with n = 1 (C1R, C1R2) give products with five-membered rings (C1Rprod,
C1R2prod) and with n = 2 (C2R, C2R2) give products with six-membered rings
(C2Rprod, C2R2prod). Many have already been reported in the literature: C1Ph,
25,28,42,43 C2Me,44 C1Ph2,19,20,26,45 C1OEt,25 C1OMe,46-48 C2OEt,25,46,49
C2OMe,47,48 C2Ph, 25 C1Me,28 C1(p-Tol),42,43 C2Me2,26 and C1Me250).
Figure 2.9. Reaction profiles for the cyclizations of various substrates under the conditions of
xxii
Scheme 2.2 in PhCl; the data for C1Ph are repeated from Figure 2.5.
Figure 2.10. Reaction profiles for the cyclizations of various substrates under the conditions of
Scheme 2.2 in PhCl and C6H6; the data for C1Ph involving PhCl are repeated from
Figure 2.5.
Figure 2.11. Rhenium-containing phosphines that were screened as catalysts.
Figure 2.12. Reaction profile for the cyclization of C1Ph with 10 mol% of (SReSC)-15b in PhCl.
Figure 2.13. Reaction profile for the cyclization of C1Ph with 10 mol% of (S)-16b in PhCl.
Figure 2.14. Reaction profiles for the cyclizations of C1Ph (■) and C2Me (∆) with 10 mol% of
(SReSP)/(SReRP)-17a in PhCl.
Figure 2.15. Top rated solvents for the intramolecular Morita Baylis Hillman reactions described
herein and their dielectric constants.
Figure 2.16. Possible substrate-catalyst interaction that may be a factor in solvent rate trends.
Figure 2.17. 1H NMR spectrum of side products obtained from the reaction of C1Ph with 10
mol% 14a at –25 °C in PhCl.
Figure 2.18. Newman projections down the Ph2P-CHR[Re] bonds of catalysts 14a and 15.
Figure 2.19. Qualitative comparison of different types of phosphorus donor groups with respect to
reactivity towards C1Ph.
Figure 2.20. Newman like projection down the η5-C5H4PPh2-Re axis of (S)-16b.
Figure 2.21. New rhenium-containing phosphines 14b-d.
Figure 2.22. Reaction profiles for the cyclizations of various substrates with 10 mol% of 14a-c in
C6H6; the data for C1S(i-Pr) and C2(p-Tol) involving 14a are repeated from Figure
2.10.
Figure 2.23. Reaction profiles for the cyclization of C2(p-Tol) with 10 mol% 14a-c in C6H6.
Conversion refers to the consumption of substrate. The product yield data are
repeated from Figure 2.22.
Figure 2.24. New rhenium-containing phosphines 14e,f.
Figure 2.25. Reaction profiles for the cyclizations of C1Ph (■) and C1S(i-Pr) (▲) with 10 mol%
14f in C6H6. The conversion of 14f to other species in the former reaction is given by
the orange trace (*).
xxiii
Figure 2.26. New rhenium-containing P-stereogenic phosphines (SReSP)/(SReRP)-17c,d.
Figure 2.27. Reaction profiles for the cyclizations of C1Ph (■) and C1S(i-Pr) (▲) with 10 mol%
of (SReSP)/(SReRP)-17c in C6H6.
Figure 2.28. Reaction profiles for the cyclizations of C1Ph (■) and C1S(i-Pr) (▲) with 10 mol%
of (SReSP)/(SReRP)-17d in C6H6.
Figure 3.1. Enantiopure complexes (S)-14b-d, obtained from (S)-(η5-C5H5)Re(NO)(PPh3)(CH3)
((S)-26) by procedures analogous to the racemates (Schemes 3.2 and 3.3).
Figure 3.2. Partial 1H NMR spectrum (CD2Cl2) of an aliquot of the reaction mixture that yields
(SReSP)/(SReRP)-[17c-H]+ PF6– (δ in ppm).
Figure 3.3. Partial 1H NMR spectra (CD3CN) of diastereomeric mixtures of (SReSP)/(SReRP)-
[17c-H]+ PF6– (δ in ppm), de = 92% and –92%.
Figure 4.1. Common orthogonal liquid phases.
Figure 4.2. Concept A: Recycling of a fluorous catalyst from biphasic liquid/liquid systems. I:
lower temperature, biphasic system, catalyst dissolved in phase B, substrate
dissolved in phase A. II: higher temperature, monophasic system, catalyst and
substrate dissolved in the mixed phase AB. III: lower temperature, biphasic system,
catalyst dissolved in phase B, product dissolved in phase A.
Figure 4.3. Concept B: Recycling of a fluorous catalyst from biphasic solid/liquid systems. I:
lower temperature, biphasic system, undissolved solid catalyst, substrate dissolved in
phase A. II: higher temperature, monophasic system, catalyst and substrate dissolved
in phase A. III: lower temperature, biphasic system, solid undissolved catalyst,
product dissolved in phase A.
Figure 4.4. Concept C: Recycling of a fluorous catalyst from biphasic solid/liquid systems with
fluoropolymer support. I: lower temperature, biphasic system, undissolved solid
catalyst which is coated on a fluoropolymer, substrate dissolved in phase A. II:
higher temperature, monophasic system, catalyst and substrate dissolved in phase A,
uncoated solid fluoropolymer. III: lower temperature, biphasic system, undissolved
solid catalyst which is again coated on the fluoropolymer, product dissolved in phase
A.
xxiv
Figure 4.5. General structure for the phosphines P((CH2)3Rfn)3 (30, a: n = 6, b: n = 8) that were
employed for the intramolecular Morita Baylis Hillman and related reactions.
Segment A: catalytically active center. Segment B: methylene spacers for modulation
of electronic and solubility properties. Segment C: fluorous tag, also for modulation
of solubility properties.
Figure 4.6. Reactions of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots of one
were assayed by 1H NMR (■), and the other (smaller scale) by HPLC (▲).
Figure 4.7. Reactions of C1S(i-Pr) with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots of one
were assayed by 1H NMR (■), and the other (smaller scale) by GC (▲).
Figure 4.8. Reactions of C1S(i-Pr)2 with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots of
one were assayed by 1H NMR (■), and the other (smaller scale) by GC (▲).
Figure 4.9. Reaction of C2(p-Tol) with 10 mol% of 30b in CH3CN at 70-72 °C. Aliquots were
assayed by HPLC.
Figure 4.10. A: Reaction mixture prior to heating. B: Reaction mixture after cooling. C: Catalyst
after separation from the reaction mixture. D: Recovered catalyst.
Figure 4.11. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots were
assayed by HPLC.
Figure 4.12. Reaction of C1S(i-Pr) with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots were
assayed by GC.
Figure 4.13. Reaction of C1S(i-Pr)2 with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots were
assayed by GC.
Figure 4.14. Reaction of C2(p-Tol) with 10 mol% of 30b in CH3CN at 70-72 °C. Aliquots were
assayed by HPLC.
Figure 4.15. Photographs from cycle 1 of Figure 4.16. A: Reaction mixture prior to heating. B:
Reaction mixture after cooling to –30 °C. C: Washed and dried Teflon® tape, coated
with catalyst.
Figure 4.16. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C in the presence of
Teflon® tape. Aliquots were assayed by HPLC.
Figure 4.17. Photographs from cycle 1 of Figure 4.18. A: Reaction mixture prior to heating. B:
xxv
Reaction mixture after cooling. C: Washed and dried Gore-Rastex® fiber, coated
with catalyst.
Figure 4.18. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C in the presence of
Gore-Rastex® fiber. Aliquots were assayed by HPLC.
Figure 4.19. Solubility of 30b in CH3CN as a function of temperature. The hysteresis curve was
determined from two independent 19F NMR experiments. Red (■): heating curve.
Blue (▲): cooling curve.
xxvi
xxvii
List of Tables
Table 2.1. Product ee values for reactions of C1Ph with 10 mol% of enantiopure catalysts in
PhCl.
Table 2.2. Product ee values for reactions of various substrates with 10 mol% of (S)-14a in
C6H6 and PhCl.
Table 2.3. Product yields for reactions of various substrates with 10 mol% 14a-d in PhCl.
Table 2.4. Product ee values for reactions of various substrates with 10 mol% (S)-14a-c in C6H6.
The values for 14a are repeated from Table 2.2.
Table 2.5. Preparative and 1H NMR product yields for reactions of various substrates with 10
mol% of 14a in C6H6.
Table 2.6. Preparative and 1H NMR product yields for reactions of various substrates with 10
mol% of 14c in PhCl.
Table 2.7. Nucleophilicity factors (N) depending on para substituents (R) of aryl phosphines
reported by Mayr.
Table 2.8. Product ee values for reactions of C1Ph and C1S(i-Pr) with 10 mol% of catalysts
(SReSP)/(SReRP)-17c and (SReSP)/(SReRP)-17d in C6H6.
Table 3.1. 31P{1H} NMR shifts and coupling constants for the rhenium-containing phosphines
14a-f.
Table 3.2. Oxidation of 14b with time, assayed as described in the text.a
Table 3.3. Comparison of selected physical properties of diastereomers 1 and 2 of
(SReSP)/(SReRP)-[17c-H]+ PF6–.
Table 4.1. Qualitative solubility of 0.050 g 30a in several solvents at different temperatures.
Table 4.2. Reaction of C1Ph with 10 mol% of 30a in CH3CN at 60-64 °C for 2.5 h. Product
yield as a function of cycle (Concept B).
Table 4.3. Reaction of C1Ph with 10 mol% of 30a in CH3CN at 60-64 °C for 6 h. Product yield
as a function of cycle. Catalyst was recovered by precipitation (Concept B) or
adsorption onto Teflon® shavings (Concept C).
xxviii
Table 4.4. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C for ca. 6 h. Yield as a
function of cycle (Concept B).
Table 4.5. Phosphine oxide O=30b formed during the reactions in Figure 4.12-4.14, as assayed
by 1H NMR and 31P{1H} NMR spectra. The result after the first cycle is from an
independent reaction.
Table 4.6. Phosphine oxide O=30b formed during the reactions in Figures 4.11, 4.16, and 4.18
as assayed by 1H NMR and 31P{1H} NMR spectra.
Table 5.1. Yields of substrates obtained from the reactions of succinaldehyde with ylides 34
(Scheme 5.5).
Table 5.2. Yields of substrates obtained from the reactions of glutaraldehyde with ylides 34
(Scheme 5.5).
1
General Introduction
1.1 Organocatalysis – a New Chapter within Catalytic Chemistry
Starting from the beginning of the organic chemistry, it has always been a challenge to
synthesize new compounds. Essential for this is the formation and breaking of atom-atom bonds.
For such transformations catalysts are often involved, achieving in many cases mild and economic
conditions. And very often products are accessible more easily, or even exclusively by catalytic
controlled reactions. A good case in point would be the formation of scalemic or even enantiopure
products. In the decades before the year 2000 mostly two categories of enantioselective catalysts
were recognized in reviews and course curricula. One was the enzyme catalyzed category and the
other the transition metal mediated one. Without doubt both are very important up to now and many
research groups concentrate on these areas.
However, the targets of interest naturally spread and the means for accomplishing catalysis
broadens. Therefore, in the last years the field of organocatalysis has become more and more
important. According to a recent review, organocatalysis is described as the acceleration of
chemical reactions through the addition of a substoichiometric quantity of an organic compound -
the organocatalyst.1 A simple SciFinder® search for the term "organocatalysis" shows the growing
interest for this field (Figure 1.1).
As depicted, starting with the year 2000 and four publications it can be estimated that during
the calendar year 2008 over 900 publications will be released. This does not necessarily show that
this field was unknown before; of course it was not, only the term "organocatalysis" was coined by
MacMillan in 2000.2 One of the earliest reports wherein a reaction was enantioselectively catalyzed
by organic molecules, in this special case by cinchona alkaloids, was that from Pracejus in 1960.3,4
2 1 General Introduction
Figure 1.1. Count of publications by year containing the topic "organocatalysis". Result from
SciFinder® search (07.2008).
However, the increasing number of publications in the last decade reflects the great interest
within the chemistry community, as there is clearly a high potential for future applications. As the
field has spread widely and there is a plethora of reactions involving organocatalysis, a full
overview is not attempted within this introduction. There are several excellent reviews available,
which illustrate the incredible breadth of this subject.1,5-10 For this reason only reactions relevant to
this thesis are highlighted within this introduction.
For many reactions, organocatalysts offer several notable advantages: usually they are small
organic molecules that are cheap, inert towards oxygen and moisture, and often available on large
scales from the "chiral pool". However, the enantiopure catalysts that are obtained from the "chiral
pool" are commonly only available as one stereoisomer. The other stereoisomer is often not
available or only accessible via an extended synthetic pathway. Furthermore, there are some
reactions that can, in comparison to other familiar transformations, only be quite slowly catalyzed.
Examples extensively studied within this thesis include the Morita Baylis Hillman reaction and the
related Rauhut Currier reaction. The latter is often referred to the vinylogous Morita Baylis Hillman
reaction. In this thesis it was of primary interest to design organocatalysts for the Baylis Hillman
and the Rauhut Currier reaction offering both high reactivities and enantioselectivities.
Independently from this the recycling capabilities of selected catalysts were investigated.
2000 2001 2002 2003 2004 2005 2006 2007 2008 0
100 200 300 400 500 600 700
Num
ber o
f pub
licat
ions
Year
1 General Introduction 3
1.2 The Historical Rauhut Currier and Morita Baylis Hillman Reaction
The Rauhut Currier reaction dates back to 1963, when Rauhut and Currier filed a US patent
on the reaction of alkyl acrylates (I) to yield the needed esters of 2-methylene glutaric acid (II),
mediated by tertiary phosphines (Scheme 1.1).11 This work was carried out at the American
Cyanamid Corporation, and was not further described in refereed journals.
Scheme 1.1. Dimerization of an alkyl acrylate (I), mediated by tertiary phosphines; the "historical"
Rauhut Currier reaction.
Rauhut and Currier described this reaction with several acrylates I, having moieties R with
one to twelve carbon atoms. Furthermore, they used several catalysts PR'3 having moieties R'
ranging from alkyl and alkenyl to alkoxy and others. They claimed that catalyst loads of 0.008-20
mol% are suitable for such reactions, depending on the educt. Independently from this, McClure12
and Baizer13 described similar phosphine mediated reactions with acrylonitriles.
In the year 1962, Price reported the initial reaction of PPh3 with acrylonitrile in the course
of investigating polymerization reactions. He postulated a kind of tautomerism leading from the
Michael adduct III to the ylide IV (Scheme 1.2).14
Scheme 1.2. Addition of PPh3 to acrylonitrile gives a zwitterionic species III which yields to the
ylide IV.
OR
O
PR'3 RO
O
OR
O
I II
2
CNPPh3 CNPh3P
CNPh3P
III IV
4 1 General Introduction
In earlier studies, he had employed the zwitterion III to initiate polymerization of the
acrylonitrile. However, he found that in the presence of protolytic solvents like EtOH, a proton of
zwitterion III migrated to give ylide IV. The latter was noted to be thermodynamically more stable.
Two years later, in 1964, Tanimoto thought about the utilization of ylides like IV to effect
Wittig reactions.15 He was able to successfully implement this idea, leading to the olefination
products V. He demonstrated these reactions for the case of PPh3 with several acrylonitrile
derivatives and aldehydes. An example involving ylide IV is depicted in Scheme 1.3.
Scheme 1.3. Wittig reaction of ylide IV with aromatic aldehydes.
Of course, for this kind of reaction a stoichiometric amount of PPh3 is needed and after the
reaction is finished the whole amount of phosphine is oxidized. Therefore, this reaction is not
catalytic. Also only moderate yields were realized (< 56%).
Morita, from the Toyo Rayon Company, knew about the isomerization from zwitterionic
structure III to ylide IV and did similar experiments but with acrylonitriles and acrylates VI. Also,
he added an aldehyde to the mixture to get coupling products. He could successfully intercept III
with the aldehyde, using catalytic amounts of the aliphatic phosphine PCy3 (Scheme 1.4).16
Scheme 1.4. Reaction sequence published by Morita. The double bond is activated by an electron
withdrawing group (EWG). The moiety R can be aliphatic as well as aromatic.
He never observed products that were derived from isomer IV, and he attributed this to the
OPPh3CNPh3P
IV
(Aryl)CHO CN
VAryl
RCHO H2C C
EWG
CH(OH)RVI VII
PCy3H2C C
EWG
H
0.6 mol%
1 General Introduction 5
use of trialkylphosphine PCy3 catalyst. In contrast Price (Scheme 1.2), as well as Tanimoto
(Scheme 1.3), employed the aromatic tertiary phosphine PPh3. However, the selectivities and
conversions they published for a special sample were very low (both ca. 20%).
Several years later, in 1972, Baylis and Hillman from the Celanese Corporation announced a
patent dealing with the same reaction that Morita had found, but with a distinctly improved
procedure.17 Therein they complained about Morita's poor yields, the deactivation of the phosphine
catalyst, and the high amount of side products. To avoid such problems, they employed tertiary
amines that had at least one bridgehead nitrogen atom, for example DABCO. Such strongly basic
catalysts could act as nucleophiles and were inert towards oxidation. In contrast to the preceding
reactions, Baylis and Hillman's normally proceeded at room temperature and only small amounts of
side products were obtained.
However, after the patent of Baylis and Hillman was accredited, for many years no
publications about this kind of reaction were released. Perhaps the patent discouraged many
research groups from investigating this reaction in more detail. Around the year 1990 several
publications were released as a simple SciFinder® search showed. Around 2000 more and more
groups focused onto this organocatalytic reaction and in 2007 more and more publications (231)
appeared. Also, many variants of this reaction have now been established. The next sections will
focus on phosphine catalyzed, intramolecular reactions.
1.3 The Intramolecular Phosphine Catalyzed Rauhut Currier Reaction
Little attention was paid to the Rauhut Currier reaction for many years. Conceivably there
were problems concerning the reactivity and particularly the chemoselectivities for the
intermolecular variant.18 However, an intramolecular variant locates the subsequently reacting
electrophile within the same molecule. The reaction cycle for the intramolecular variant is shown in
Scheme 1.5.
The reaction starts with substrate VIII. This contains two potentially equivalent double
bonds, each activated by an electron withdrawing group (EWG). One of these Michael systems then
undergoes nucleophilic attacked by a tertiary phosphine resulting in the zwitterionic intermediate
6 1 General Introduction
IX, wherein the carbanion is stabilized by the EWG'. This step is mostly reported to be an
equilibrium. Subsequently, the carbanion attacks the second intramolecular Michael system giving
the cyclic, zwitterionic intermediate X, which is also stabilized, but this time by the second EWG''.
Proton transfer gives intermediate XI, which subsequently undergoes an E1-cb type elimination.
This provides product XII while releasing the phosphine catalyst for the next reaction cycle. As
denoted within the cycle several stereocenters (*) are formed. However, two of them are removed
en route from X to XII due to the phosphine elimination. Nonetheless, the product still contains one
chiral center.
Scheme 1.5. Reaction cycle for the phosphine catalyzed intramolecular Rauhut Currier reaction.
In 2001 Krische published a communication wherein he reported a phosphine catalyzed
intramolecular variant of the Rauhut Currier reaction.19 He investigated the chemoselectivities with
educts that featured two independent Michael systems. These systems were connected by a two or
three carbon metyhlene spacer (VIII, n = 1, 2). An example with a two carbon atom spacer is
*EWG'
EWG''
PR3
EWG''
EWG'
n
*
EWG''
EWG'
nR3P
*
*
*EWG'
EWG''
R3Pn
n
VIII
IXX
XII
*
*
*EWG'
EWG''
R3Pn
XI
1 General Introduction 7
illustrated in Scheme 1.6.
Scheme 1.6. Intramolecular Rauhut Currier reaction published by Krische.
As depicted above two different products XII can be obtained, as the catalyst is presented
with independent Michael systems where it can attack. It was shown that very high selectivities
could be realized (> 95:5), depending on the Michael systems' substituents and the spacer chain
length. Also the reported product yields were very good (87%). Employing the same catalyst,
Murphy reported similar results in 2002.20
Roush reported in 2004 the total synthesis of (–)-spinosyn A, wherein one of the key steps
was the formation of a five-membered cyclic alkene (Scheme 1.7).21 This was achieved by an
intramolecular Rauhut Currier reaction that was promoted by eight equiv. of the tertiary phosphine
P(CH3)3. Although this is not a catalytic quantity, in this special case it was more important to
maximize the yields. And in fact, quantitative conversion to the desired product was reported.
Although it does not qualify as a phosphine catalyzed reaction, some attention should be
paid to a cysteine catalyzed variant published by Miller last year (2007).22 Therein he reports
intramolecular Rauhut Currier reactions mediated by thiolates with moderate to good yields. More
important are the excellent enantiomeric excess values (ee; up to 95%) he obtained under optimized
conditions. However, one equiv. of cysteine was required for optimum yields and
enantioselectivities.
O
OEt
O
O O
OEt
OEt
O
O
PBu3
VIII, n = 1
EWG' = CH3(CO)
EWG'' = (CO)OEt
> 95:587%
XII, n = 1
EWG' = CH3(CO)
EWG'' = (CO)OEt
XII, n = 1
EWG' = (CO)OEt
EWG'' = CH3(CO)
10 mol%
8 1 General Introduction
Scheme 1.7. The Rauhut Currier reaction as a key step for the total synthesis of (–)-spinosyn A
reported by Roush.
1.4 The Intramolecular Phosphine Catalyzed Morita Baylis Hillman Reaction
The Morita Baylis Hillman reaction is closely related to the Rauhut Currier reaction. The
reaction cycle for the intramolecular phosphine catalyzed variant is shown in Scheme 1.8.
The cycle depicted below is very similar to that in Scheme 1.5. The difference is that the
carbanion of the zwitterionic intermediate XIV attacks the aldehyde (or ketone) group, which leads
to the zwitterionic alcoholate XV. A subsequent proton shift to give zwitterion XVI, followed by
elimination of the catalyst leads to the highly functionalized product XVII. Again several
stereocenters (*) are formed during the reaction cycle, and one remains in the chiral alcohol product
XVII.
Several phosphine catalyzed intramolecular Morita Baylis Hillman reactions have been
H H
O
OOCH3
OCH3OCH3
O
O
O
O
ON(CH3)2
R'''
R''H
O
OR'
R
OR'''
R''H
O
OR'
R
OP(CH3)3
(−)-spinosyn A
key step
8 equiv.
1 General Introduction 9
reported to date. The first was published in 1992 by Fráter who needed the particular product XIX
shown in Scheme 1.9.23 For his ring closing process the depicted substrate XVIII was found to be
most reactive with catalyst PBu3. However, the selectivity was only 75% and the reaction took 24 h
to finish. Amine catalysts did not afford any product XIX.
Scheme 1.8. Reaction cycle for the phosphine catalyzed intramolecular Morita Baylis Hillman
reaction.
Scheme 1.9. First intramolecular Morita Baylis Hillman reaction carried out by Fráter.
CHO
∗EWG
OHPR3
EWG
n
CHO
∗
EWG
nR3P
∗
∗
∗EWG
OH
R3Pn
nXIII
XIVXVI
XVII
∗
∗
∗EWG
O
R3Pn
XV
O HO25 mol%
75%, 24 h
XIX
EtO
O
EtO
O
XVIII
PBu3
10 1 General Introduction
During the period from 1999 to 2006, Murphy published four reports detailing systematic
investigations of ring closing Morita Βaylis Hillman reactions.20,24-26 He found that phosphines,
usually 20-30 mol% of PBu3, were generally good choices for the intramolecular reaction. He also
employed amines or thiolates as catalysts but the product yields were lower or the reactions did not
occur at all. A typical cyclization is depicted in Scheme 1.10.
Scheme 1.10. Representative reaction reported by Murphy.
Keck investigated the same reaction in 2002 with a substrate similar to XIII (Scheme 1.10),
but with R equal to SEt.27 Employing 10 mol% of P(CH3)3, he obtained 82% yield of a thioester
alcohol after 15 h.
Also Koo published a communication in 2004 reporting PPh3 promoted reactions similar to
those in Scheme 1.10.28 However, he employed one equiv. of the phosphine. Under optimum
conditions and using a reactive substrate that gave a five-membered ring system, it took 12 h to
obtain the product in 98% yield. Many substrates needed longer reaction times and gave lower
yields.
Noteworthy here is a further publication from Miller, this time in the year 2005.29 He
reported an amine catalyzed intramolecular Morita Baylis Hillman reaction with a complex co-
catalyst system. He employed a catalyst mixture of 20 mol% of pipecolinic acid and 10 mol% of a
particular scalemic peptide. The best yield he obtained was 82%. Even more important was the ee
of 80% he obtained. With a subsequent kinetic resolution of the scalemic product mixture, he was
able to isolate a product in 50% yield with > 98% ee.
CHO
HO
20-30 mol%
PBu3
58-80%, 5-16 h
XVIIEWG = R(CO)
R
O
R
On = 1,2
n = 1,2
XIIIEWG = R(CO)
1 General Introduction 11
1.5 Future Directions
The Morita Baylis Hillman reaction, particularly the intramolecular variant, offers a
tremendous breadth with regard to substrate diversity and product scope. In Figure 1.2 the highly
functionalized product molecule XVII is depicted.
Figure 1.2. Key features of the highly functionalized intramolecular Morita Baylis Hillman product
XVII.
Importantly, the product molecule XVII features three different groups. Many further
reactions at these groups are conceivable. The alcohol could be alkylated. If the EWG features a
carbonyl (R(CO)) group, 1,2-addition, 1,4-addition, or reduction to a second allylic alcohol center
could be effected. The double bond could be stereoselectively hydrogenated leading to a second
stereocenter. Or the double bond could be oxidized selectively to a glycol again leading to a triol
with three adjacent stereocenters.
This stereocenter is embedded in a rigid skeleton formed by a five- or six-membered ring
system. Hence, after modification of the functional groups, such molecules could be useful for
docking to receptors, leading to pharmaceutical applications. Therefore, it is critical to develop
reliable synthetic pathways leading to such molecules, ideally with high stereoselectivity.
It was and it will be a challenge to synthesize molecules of the type XVII. In the past the
Morita Baylis Hillman reaction often had slow reaction rates associated with low yields. High
substrate as well as solvent dependencies were found. Furthermore, at the outset of this work in
∗EWG
OH
nXVII
Three functional groups
Rigid skeleton
Stereocenter
12 1 General Introduction
2004, there were no satisfying chiral catalysts for the enantioselective formation of these products.
Therefore, we asked whether it would be possible to develop highly reactive catalysts
leading to satisfying product yields. Additionally, the catalyst should provide high
enantioselectivities.
A second target was to find a way to recycle the catalyst. Until now there is only one
publication dealing with the recycling of a catalyst for the Morita Baylis Hillman reaction.30 It was
released while this work was going on. However, it involves a totally different catalytic system than
that employed within this work, as detailed in section 4.3.
2
Enantioselective Organocatalysis with Rhenium-Containing Phosphines
2.1 Introduction
In the last few decades there have been many publications involving rhenium complexes.
Rhenium-containing phosphines are distinguished by one or more non-coordinating PR2 moieties,
and were first reported by Gladysz and Buhro.31 The general structure of the rhenium center
featured in this early work is depicted in Figure 2.1. Such complexes are often termed "chiral at
rhenium". However, detailed information regarding the enantioselective synthesis of the rhenium-
containing phosphines employed in this chapter are given in Chapter 3. In the following text some
structural and electronic attributes are highlighted, followed by representative rhenium complexes
that have been applied in catalytic reactions other than those treated in Chapter 1.
Figure 2.1. General idealized structures of the rhenium fragment [(η5-C5H5)Re(NO)(PPh3)]+ (A)
and (η5-C5H5)Re(NO)(PPh3)(X) (B), upon which many rhenium complexes are based.
Structure A in Figure 2.1 shows the d-orbital HOMO of the fragment [(η5-
C5H5)Re(NO)(PPh3)]+. This is a sixteen valence electron species and therefore this fragment is a
strong Lewis acid with attractive interactions.32 A variety of adducts with Lewis bases X– have
been reported (B, Figure 2.1). On the other hand, the rhenium fragment is a π-base as well, as the d-
ON PPh3
Re
PPh3
ON
X
125°
125°90°
90°
A B
14 2 Catalysis with Rhenium-Containing Phosphines
orbital HOMO can donate into the π-acceptor system of subsequently bound ligands. Of course, this
donating effect is also present if groups X without low-lying π-acceptor orbitals are bound. This has
particular consequences when such ligands also feature a lone pair, as sketched in the generalized
structure C in Figure 2.2. Repulsive interactions should result between the occupied orbitals, i.e. the
ligand's lone pair and the d-orbital HOMO.31,33 This explains the experimentally observed
increased basicity of the ligand's lone pairs.
Figure 2.2. Key orbital interactions in coordinatively saturated octahedral LnMPR2 (C) and
LnMCH2PR2 (D, E) complexes.
Increased basicity is also observed when a methylene spacer is present between the metal
center and the lone pair bearing atom. The two relevant interactions are illustrated in structures D
and E. Structure D depicts the repulsive interaction between the filled metal-carbon σ-orbital and
the phosphorus lone pair. This results in an enhanced energy, and therefore enhanced basicity and
nucleophilicity, of the phosphorus lone pair. This effect is exactly analogous to the increase of
carbon basicity and nucleophilicity in allyl silanes.34 The other orbital interaction shown in
structure E involves the metal d-orbital HOMO and the carbon phosphorus σ*-orbital. This filled-
empty (attractive) interaction serves to decrease the carbon phosphorus bond order, putting more
electron density on the phosphorus atom and raising the lone pair orbital energy.
These effects were evidenced by the simple experiment depicted in Scheme 2.1.35
L
L
PL
L
L
RR
C
repulsive:metal d-orbital HOMO / phosphorus lone pair
D
repulsive:σ-orbital /
phosphorus lone pair
E
attractive:metal d-orbital HOMO /
σ*-orbital
L
L
L
L
L
P
L
L
L
L
L
CC
P
2 Catalysis with Rhenium-Containing Phosphines 15
Scheme 2.1. Enhanced basicity of the lone pairs of (η5-C5H5)Re(NO)(PPh3)(CH2SCH3) (1).
When the thioether 1 was mixed with the cationic sulfonium salt 2 an equilibrium developed
and the reaction shifted to the side of the cationic sulfonium salt 3 and the free thioether dimethyl
sulfide. This showed that 1 is a stronger base towards [(η5-C5H5)Re(NO)(PPh3)(=CH2)]+ than
dimethyl sulfide, proving the enhanced basicity of the sulfur atom lone pairs in 1.
Going back to Figure 2.1, someone could guess naively that the fragment B has a tetrahedral
structure (four ligands). In fact it is formally an octahedral structure wherein the cyclopentadienyl
ligand occupies three coordination sites of the rhenium center. Therefore, the angles between the
other three ligands are all formally 90°. However, these angles differ in reality by up to several
degrees, which could effect product stereoselectivities.
Summarizing these characteristics, there is a spectator metal center that enhances electron
density at the catalytically active Lewis base center, and hence greater reactivity is anticipated.
Second, the chiral rhenium center with its three different sized ligands (η5-C5H5, PPh3, and NO)
provides the possibility of enantioselective catalysis. In view of the octahedral geometry, there are
significantly different bond angles as compared to analogs with carbon stereocenters.
There has been a long tradition in employing this fragment for catalysis within the Gladysz
group. Some catalytically active rhenium-containing monophosphines are summarized in Figure 2.3.
ON PPh3Re
H2C
S
CH3
ON PPh3Re
H2C
S
CH3
CH3
ON PPh3Re
H2C
S
ON PPh3Re
CH2
CH3
S
CH3
CH3
1 2 3
16 2 Catalysis with Rhenium-Containing Phosphines
Figure 2.3. Rhenium-containing monophosphines that have been employed successfully in
catalysis.
Compounds 4-6 were successfully employed for the Suzuki coupling reactions of PhBr with
phenyl boronic acid.36 For this 4-6 were generated in situ from the corresponding conjugate acid,
the phosphonium salt, using t-BuOK as a base. Then Pd(OAc)2 was added and the resulting
palladium-phosphine adducts were utilized for catalysis without isolation. Biphenyl was formed in
57% to 99% yields. Typically 4 mol% of ligand 4-6 and 1 mol% of Pd(OAc)2 were utilized. These
catalytic systems tolerated many differently substituted bromophenyl substrates.
Catalyst 7 was employed for Suzuki coupling reactions with the above mentioned
substrates.37 It is a bidentate species that is easily synthesized via cyclometalation and subsequently
isolated. Catalyst 7 gave product yields of 86% to 99%. It is a very reactive species and turnover
numbers of nearly 100000 have been reached.
Rhenium-containing diphosphines that are functionalized at the cyclopentadienyl moiety
have been synthesized. These have been used as chelate ligands for rhodium and palladium
catalysts. Selected catalyst precursors are depicted in Figure 2.4.
Complexes 8-12 in all have a diphenylphosphinocyclopentadienyl ligand, η5-C5H4PPh2. An
H2C
Ph3P Re ∗
ONPd
Ph2P
Br
2
ON PPh3Re∗
PR2
ON PPh3Re∗
CH3
R2P
OC CORe∗
CO
R2P
4 5
6 7
2 Catalysis with Rhenium-Containing Phosphines 17
additional PPh2 group was coordinated to the rhenium directly (9, 12) or linked through a single
carbon spacer (8, 10, 11, and 13).
Figure 2.4. Adducts of rhenium-containing diphosphines that have been employed successfully in
catalysis.
Hydrosilylations of aromatic ketones were successfully carried out with catalysts 8, 9, and
PPh2
NORe∗
PPh3
Ph2P
RhCH2
NORe∗
PPh3
Ph2P
Rh
Ph2P
PPh2
NORe∗
PPh3
Ph2P
Pd
CH2
NORe∗
PPh3
Ph2P
Pd
Ph2P
Cl
Cl
Cl
Cl
CH∗
NORe∗
PPh3
Ph2P
RhPh2P
Ph
ON PPh3Re∗
H2C
P
NOPh3PRe∗
H2C
NOPh3PRe∗
CH2
P
ON PPh3Re∗
CH2
Rh
8 9
10 11
12
13
0,1
18 2 Catalysis with Rhenium-Containing Phosphines
10. After workup the corresponding scalemic alcohols were obtained with ee values ranging from
11% to 92% and good yields (50-83%).38,39 Hydrogenations of protected dehydro amino acids
were also performed with catalysts 8, 9, and 10 resulting in good yields (70-98%) of products and
with good to excellent enantioselectivities (40-98% ee).39 Arylations of dihydrofuran were achieved
with palladium based catalysts 11 and 12. However, the best yield was 84% and the highest ee
12%.40
Catalyst 13 contains four rhenium fragments, each of which is singly bonded to a
phosphorus center through a methylene spacer. As the phosphorus centers are bridged by (CH2)n
spacers, a bidentate ligand for the active rhodium center is obtained. This molecule was screened
with a large variety of reactions. These included hydrogenations of protected dehydro amino acids
and hydrosilylations of aromatic ketones. The conversions ranged from 32% to 99%; unfortunately
products with low enantioselectivities were always obtained (0-33% ee).41
2 Catalysis with Rhenium-Containing Phosphines 19
2.2 The Intramolecular Morita Baylis Hillman and Rauhut Currier Reaction: First
Experiments
2.2.1 Results
2.2.1.1 A First Reaction Series
The starting point of this work was the well known racemic complex (η5-
C5H5)Re(NO)(PPh3)(CH2PPh2) (14a)38 and the substrate C1Ph for the intramolecular Morita
Baylis Hillman reaction (Scheme 2.2). Murphy had established that C1Ph reacts smoothly with the
catalyst PBu3 (Scheme 1.10).25
Scheme 2.2. First intramolecular Morita Baylis Hillman reactions with C1Ph and 10 mol% of 14a,
carried out in different solvents.
For all test reactions in this chapter the following sequence and conditions were employed,
unless noted. First, the reaction mixtures were always 0.050 M in substrate and 0.0050 M in catalyst.
Higher concentrations resulted in more side products while lower concentrations decreased the
reaction rate and rendered NMR monitoring more difficult. The reactions were carried out at room
temperature. In some cases, product formation was monitored. One method was by 1H NMR, using
CH2ClCH2Cl (δ(ppm) = 3.73) as internal standard. For such experiments, reactions were conducted
in a NMR tube under N2. The tube was transferred regularly to the NMR spectrometer. In other
CHOPh
OO
Ph OH
10 mol%
ON PPh3Re
CH2PPh2
C1Ph C1Phprod
14a
20 2 Catalysis with Rhenium-Containing Phosphines
cases, product formation was monitored by GC. For such experiments, 0.050 mL aliquots of the
reaction solutions were removed and added to the ethyl acetate solutions of decane, which served as
an internal standard. More information about the monitorings are detailed in section 2.5.2. The
curves shown in all rate profiles were calculated by standard fitting methods.
2.2.1.2 Solvent Dependency
The reaction in Scheme 2.2 was carried out in several solvents. Only the reactions in
CH3CN, CH2ClCH2Cl, PhCl, or C6H6 were successful. Data obtained by GC or NMR monitoring
as described above are given in Figure 2.5.
Figure 2.5. Reaction profiles for the formation of C1Phprod in Scheme 2.2 in different solvents.
The purple graph (♦) shows for comparison the reaction of C1Ph with 10 mol% of PPh3 in
CH2ClCH2Cl.
In general, nonpolar solvents gave faster reactions. The optimum solvent rate-wise was
0 40 80 120 1600
20
40
60
80
Prod
uct Y
ield
(%)
Time (h)
CH3CN
CH2ClCH2Cl / t-BuOH
C6H6
PhCl CH2ClCH2Cl
CH2ClCH2Cl / PPh3
2 Catalysis with Rhenium-Containing Phosphines 21
C6H6 (●); the substrate was fully consumed after ca. 1 h. In PhCl ( ) the reaction was comparable
but a little slower (2 h). It was much slower in CH2ClCH2Cl (■; 120 h) but the reaction went to
completion anyway. A mixture of CH2ClCH2Cl and t-BuOH (▼) rapidly initiated catalysis but the
yield of product reached a maximum of 30% after 20 h. Much substrate remained unconsumed. In
general, in the presence of an alcohol additive, the reaction was much slower or the selectivity
decreased dramatically. Often many unassigned side products were found in addition to unreacted
substrate. The reaction in CH3CN (▲) is depicted above but it did not go to completion also.
Furthermore, many side products were detected by GC but were not further investigated.
To compare the effect of the rhenium fragment upon activity, a comparison experiment with
PPh3 was conducted in CH2ClCH2Cl (♦). As shown in Figure 2.5, reaction was much slower. The
experiment was terminated after one week of reaction time with ca. 20% yield and much
unconverted substrate.
2.2.1.3 Temperature Dependency
Since the reaction in CH3CN gave many side products, a lower temperature was
investigated in hopes of increasing selectivity. Data are shown in Figure 2.6.
Surprisingly, product formation appeared faster initially. However, the starting material was
consumed after 25 h, at which time the product yield was only 25%. Many more side products than
at room temperature were noted. This showed that for optimization purposes CH3CN seemed to be
the wrong solvent.
As the reaction in CH2ClCH2Cl gave few side products, the temperature was raised and the
effect upon rate assayed. Data are given in Figure 2.7. This shows that raising the temperature to 40
°C in CH2ClCH2Cl decreases the rate of product formation. The catalyst was inactive after several
hours and gave only 20% product as analyzed by GC. These measurements further showed that
there was much substrate left beside some unknown side products. To exclude a temperature
dependent equilibrium, a CH2ClCH2Cl solution of C1Phprod and 14a (10 mol%) was kept at 40 °C
for 24 h. No reaction occurred, as assayed by 1H NMR and 31P{1H} NMR spectroscopy.
22 2 Catalysis with Rhenium-Containing Phosphines
Figure 2.6. Reaction profiles for Scheme 2.2 at room temperature (▲) and 0 °C (■) in CH3CN. The
former experiment is repeated from Figure 2.5.
Figure 2.7. Reaction profiles for Scheme 2.2 at room temperature (▲) and 40 °C (■) in
CH2ClCH2Cl. The former experiment is repeated from Figure 2.5.
0 50 100 1500
10
20
30
40
50P
rodu
ct Y
ield
(%)
Time (h)
rt
0 °C
0 20 40 60 80 100 120 1400
20
40
60
80
Prod
uct Y
ield
(%)
Time (h)
40 °C
rt
2 Catalysis with Rhenium-Containing Phosphines 23
The catalyst 14a, as well as C1Phprod remained. An independent NMR experiment, wherein a
CH2ClCH2Cl solution of 14a was kept at 40 °C overnight, showed the ruggedness of the catalyst
under such conditions.
Similar experiments with C1Ph and 14a in PhCl but with a temperature gradient (–20 °C to
room temperature, 12 h) showed only the formation of side products. As a result of these
experiments, an ambient reaction temperature was judged to be optimal. For practical reasons this
meant a temperature range of 20-25 °C for all experiments that were carried out at room
temperature. If the ambient temperature exceeded 25 °C, a cryostat was employed to cool the
mixtures to 20 °C.
2.2.1.4 Substrate Dependency
Given the superiority of C6H6 and PhCl as solvents established in the previous section,
several substrates were tested. These substrates and their abbreviations are summarized in Figure
2.8. References are given for the substrates that are known from the literature; all others are new
and their syntheses are presented in Chapter 5. Their Morita Baylis Hillman products are also
depicted, but not all of them could be obtained with the employed catalysts in this chapter.
Several of those substrates were tested in PhCl with the conditions of Scheme 2.2 (10 mol%
14a). The reaction profiles are given in Figure 2.9. In the course of this work a set of substrates was
found that showed excellent reactivity. These were C1Ph, C1(p-Tol), C1Me, C1S(i-Pr), C2(p-Tol),
C2Me, C1Ph2, and C1S(i-Pr)2. The others (C1OMe, C1OEt, C2OMe, C2OEt, C2Ph, C2(p-Tol)2,
and C2Me2) showed only low or even no reactivity.
It was found that the Morita Baylis Hillman substrates C1(p-Tol) ( ) and C1Ph (■) were
very quickly transformed into their products (3 h and 2 h respectively). The yield in the latter
reaction was slightly lower (95% vs. 90%). The Rauhut Currier substrate C1Ph2 (▼) also showed a
fast reaction, and product formation was complete within 3 h. Both C1Me ( ) and C2(p-Tol) (□)
reacted more slowly, with maximum product yields achieved after 22 h and 48 h, respectively. The
reaction with C1S(i-Pr) (▲) was less selective, as after 29 h the substrate had been consumed, but
24 2 Catalysis with Rhenium-Containing Phosphines
the product yield was only 50%.
The reaction rates increased when the solvent PhCl was replaced by C6H6. The data are
shown in Figure 2.10, with all other conditions the same as before. The data in PhCl are repeated
from Figure 2.9.
Figure 2.8. Substrates employed herein for catalytic reactions, and product structures. Substrates
with n = 1 (C1R, C1R2) give products with five-membered rings (C1Rprod, C1R2prod) and with n
= 2 (C2R, C2R2) give products with six-membered rings (C2Rprod, C2R2prod). Many have already
been reported in the literature: C1Ph,25,28,42,43 C2Me,44 C1Ph2,19,20,26,45 C1OEt,25 C1OMe,46-48
C2OEt,25,46,49 C2OMe,47,48 C2Ph,25 C1Me,28 C1(p-Tol),42,43 C2Me2,26 and C1Me250).
In the case of substrate C2(p-Tol) the reaction rate was only slightly increased by the usage
of C6H6 (red traces, □). For C1Ph this effect was stronger (black traces, ■) and for the substrate
CHOR
On = 1,2
Morita Baylis Hillman substrates
R
O
n = 1,2
R
O
Rauhut Currier substrates
C1Ph, C1S(i-Pr), C1OMe, C1OEt, C1Me, C1(p-Tol)C2(p-Tol), C2Me, C2OMe, C2OEt, C2Ph
C1Ph2, C1S(i-Pr)2, C1Me2
C2(p-Tol)2, C2Me2
CnR CnR2
R
OOH
R
O
R
O
CnRprod CnR2prod
Morita Baylis Hillman products Rauhut Currier products
n = 1,2
n = 1,2
2 Catalysis with Rhenium-Containing Phosphines 25
Figure 2.9. Reaction profiles for the cyclizations of various substrates under the conditions of
Scheme 2.2 in PhCl; the data for C1Ph are repeated from Figure 2.5.
Figure 2.10. Reaction profiles for the cyclizations of various substrates under the conditions of
Scheme 2.2 in PhCl and C6H6; the data for C1Ph involving PhCl are repeated from Figure 2.5.
0 20 400
20
40
60
80P
rodu
ct Y
ield
(%)
Time (h)
C1Ph
C1(p-Tol)
C1Ph2
C1Me
C2(p-Tol)
C1S(i-Pr)
0 5 10 150
20
40
60
80
( ) C1Ph
( ) C1S(i-Pr)
( ) C2(p-Tol)
Pro
duct
Yie
ld (%
)
Time (h)
C6H6
PhCl C6H6
C6H6
PhCl
PhCl
26 2 Catalysis with Rhenium-Containing Phosphines
C1S(i-Pr) it was significantly higher (green traces, ▲). This meant for the latter substrate a reaction
time of ca. 6 h (C6H6) vs. 30 h (PhCl) and furthermore nearly a doubling of the yield (90% vs.
50%).
Two more substrates were tested in C6H6 (C1Ph2, C1S(i-Pr)2). Also these gave very good
yields (92%, 98%). These data are given in the experimental to keep Figure 2.10 more tractable.
Next it was sought to ascertain how the catalyst could be optimized. For this, previously
synthesized rhenium-containing phosphines were screened.
2.2.1.5 Reactions with Other Rhenium-Containing Phosphines
The next series of experiments utilized different rhenium-containing phosphines, as
summarized in Figure 2.11. All complexes except one had been previously characterized either in
the literature ((SReSC)-15a,51 (SReSC)-16a,51 and (S)-16b38) or a doctoral dissertation ((SReSp)/
(SReRP)-17a,b52). The synthesis of (SReSC)-15b is described in Chapter 3.
Figure 2.11. Rhenium-containing phosphines that were screened as catalysts.
In comparison to catalyst 14, 15 and 16 have a second chiral center at the carbon spacer.
Furthermore, the complexes 16 are distinguished by a second PPh2 group. The complexes 17 also
feature a phosphorus stereocenter.
All following reactions were conducted in PhCl and monitored by 1H NMR. The reaction of
the catalysts 15 with the substrates C2Me, C2Ph, C2OEt, and C1OEt showed no formation of the
ON PPh3Re∗
∗
(SReSC)-15a: R = Ph
(SReSC)-15b: R = CH3
RPh2P
H
ON PPh3Re∗
∗RPh2P
H
(SReSC)-16a: R = Ph
(S)-16b: R = H
PPh2
ON PPh3Re∗
CH2PRR'∗
(SReSP)/(SReRP)-17a: R/R' = Cy/Ph
(SReSP)/(SReRP)-17b: R/R' = Cy/t-Bu
2 Catalysis with Rhenium-Containing Phosphines 27
target products. These substrates were recovered after 168 h, together with a smaller amount of side
products. Only the substrate C1Ph was converted. However, the reactivity of the phenyl-substituted
catalyst (SReSC)-15a was so low that only ca. 10% of the C1Ph was transformed into the cyclic
product within 120 h. Much substrate was unconverted (ca. 80%) and some side products were
found. The methyl-substituted catalyst (SReSC)-15b was somewhat more reactive towards C1Ph as
shown in Figure 2.12. However, in comparison to catalyst 14a very long reaction times were
required (ca. 144 h vs. 2 h).
Figure 2.12. Reaction profile for the cyclization of C1Ph with 10 mol% of (SReSC)-15b in PhCl.
The diphosphine catalysts 16a,b were similarly tested. As shown in Figure 2.13, (S)-16b
exhibited some reactivity, although the amount of product obtained was still lower than with
(SReSC)-15b. After one week, the product was present in 40% yield. Some starting material as well
as side products were detected.
The other substrates (C2Me, C2Ph, C2OEt, and C1OEt) were recovered after 168 h,
together with lower amounts of side products. The complex (SReSC)-16a was totally inactive.
Next, complexes 17a,b were screened as catalysts. For these compounds, the corresponding
0 50 100 150 2000
20
40
60
80
Pro
duct
Yie
ld (%
)
Time (h)
28 2 Catalysis with Rhenium-Containing Phosphines
phosphonium PF6– salts52 were first treated with t-BuOK. After filtration through Celite® the crude
red compounds 17a,b were precipitated with n-pentane and used without further characterization.
The diastereomeric excesses (de) of (SReSP)/(SReRP)-17a and (SReSP)/(SReRP)-17b were 91% and
80%, respectively (predominantly SReSP, each).
Figure 2.13. Reaction profile for the cyclization of C1Ph with 10 mol% of (S)-16b in PhCl.
Complex (SReSP)/(SReRP)-17b showed only low reactivity. In the best case (C1Ph), only
traces of product were detected after 168 h. The 1H NMR analyses showed much unconverted
substrate. The substrates C2Me, C2Ph, C2OEt, and C1OEt gave even poorer results. In contrast,
(SReSP)/(SReRP)-17a, in which the t-butyl group of (SReSP)/(SReRP)-17b has been replaced by a
phenyl, showed appreciable reactivity towards C1Ph and C2Me. The reaction data are depicted in
Figure 2.14. In the case of C1Ph, the first NMR assay was carried out after 24 h; the reaction may
have been complete earlier. But anyway the data for C2Me show a much longer reaction time as
compared to 14a (168 h vs. 2 h).
Finally, the rhenium-containing phosphine (η5-C5H5)Re(NO)(PPh3)(PPh2) (18)38 without a
spacer group was tested. The active phosphorus center should be more electron rich as analyzed in
0 50 100 150 2000
20
40
60
80
Pro
duct
Yie
ld (%
)
Time (h)
2 Catalysis with Rhenium-Containing Phosphines 29
section 2.1, and therefore higher reactivity was expected. However, the catalyst was consumed
rapidly and no product was detected. Thus, it seemed that this phosphine was too reactive.
Figure 2.14. Reaction profiles for the cyclizations of C1Ph (■) and C2Me (∆) with 10 mol% of
(SReSP)/(SReRP)-17a in PhCl.
2.2.1.6 Enantioselective Catalyses
As all of the rhenium-containing catalysts that are mentioned above have at least a rhenium
stereocenter, and can be synthesized with a single rhenium configuration, there is the possibility to
perform enantioselective catalysis. The substrate C1Ph was screened with 10 mol% of several
catalysts in PhCl. The ee values of the product C1Phprod are summarized in Table 2.1. The absolute
configuration of the dominant enantiomer was not determined.
This table shows that the best catalyst is (S)-(η5-C5H5)Re(NO)(PPh3)(CH2PPh2) ((S)-14a).
Two other catalysts ((SReSC)-15b, (SReSP)/(SReRP)-17a) gave somewhat lower ee values and one
gave a product that was nearly racemic ((SReSC)-16b). Furthermore, for the catalysts that gave
0 50 100 150 2000
20
40
60
80
Prod
uct Y
ield
(%)
Time (h)
C1Ph
C2Me
30 2 Catalysis with Rhenium-Containing Phosphines
lower ee values, the reaction rates were also slow (Figures 2.12, 2.13, and 2.14).
To investigate the scope of the best catalyst from Table 2.1, analogous reactions were
conducted with 10 mol% of (S)-14a and a variety of substrates in two solvents, C6H6 and PhCl. The
results are shown in Table 2.2.
Table 2.1. Product ee values for reactions of C1Ph with 10 mol% of enantiopure catalysts in PhCl.
Catalyst
(S)-14a (SReSC)-15b (SReSP)/(SReRP)-17aa
(SReSC)-16b
ee(C1Phprod) 45% 28% 35% 3%
a 91% de.
Table 2.2. Product ee values for reactions of various substrates with 10 mol% of (S)-14a in C6H6
and PhCl.
Solvent
C6H6 PhCl
ee(C1Phprod) 62% 45%
ee(C1S(i-Pr)prod) 74% 62%
ee(C2(p-Tol)prod) 38% 38%
ee(C2Meprod) 70% 47%
ee(C1Ph2prod) 42% 56%
ee(C1S(i-Pr)2prod) 52% ----a
a This value was not determined.
With a single exception (C1Ph2prod), the best results were obtained in C6H6. In every case,
reaction was faster in C6H6. Finally, analogous reactions of C1Ph and C1S(i-Pr) with racemic 14a
proceeded at similar rates.
2 Catalysis with Rhenium-Containing Phosphines 31
2.2.2 Discussion
2.2.2.1 Solvent Dependency
The mechanism of the intramolecular Morita Baylis Hillman reaction was shown in Scheme
1.8. An activated carbon-carbon double bond undergoes nucleophilic attack by the catalyst. The
resulting enolate reacts with an aldehyde group. In former times it was supposed that this step
should be rate determining,53 while newer publications discuss the proton shift, occurring prior to
the elimination step, as rate determining.54,55 This remains to be clarified, and in any event may not
apply to intramolecular reactions where the aldehyde is already connected to the enolate.
The Morita Baylis Hillman reaction is known to be very solvent dependent. For example,
Koo conducted the reaction with substrate C1Ph and a stoichiometric amount of PPh3 in t-BuOH
and stirred this mixture at room temperature for 6 h.28 Their reaction in CH3CN on the other hand
needed 22 h. Up to this point, protic and/or very polar solvents were thought to be optimal for
Morita Baylis Hillman reactions, as the very polar zwitterionic intermediates have to be stabilized
during the reaction cycle (Scheme 1.8).
At the outset of this work, the Koo conditions were applied but with 10 mol% (η5-
C5H5)Re(NO)(PPh3)(CH2PPh2) (14a) in place of one equiv. of PPh3. Unfortunately, 14a is only
very slightly soluble in t-BuOH. To improve the solubility, CH2ClCH2Cl was added. A 1:1 v/v
mixture afforded adequate solubility and the reaction of C1Ph gave nearly a 30% yield of C1Phprod.
Some substrate was unconverted and many (unassigned) side products were detected by GC. The
protic solvent and/or hydrogen bonds derived therefrom were viewed as a possible origin of these
side reactions. In contrast, the literature experiments with PPh3 as "catalyst" in t-BuOH gave hardly
any side products.28
Catalyst 14a also gave much slower or no reactions in the presence of other alcohol
additives such as CF3CH2OH and BINOL. Thus, a protic solvent or cosolvent seems to be a bad
choice for catalyst 14a. Other polar solvents were sought that could dissolve the catalyst. CH3CN
was selected. However, many side reactions were detected.
32 2 Catalysis with Rhenium-Containing Phosphines
Given the data in Figure 2.5, solvents with lower polarities were screened. As shown, good
yields could be obtained with CH2ClCH2Cl, which was used as a cosolvent before. However, the
reaction time was quite long. The experiments in PhCl gave better results concerning reaction time
and yield, and the best results were obtained by employing C6H6 as solvent (Figure 2.10). A
"ranking" of the solvents is shown in Figure 2.15.
Figure 2.15. Top rated solvents for the intramolecular Morita Baylis Hillman reactions described
herein and their dielectric constants.
This stands in sharp contrast to the usual solvents reported in most literature sources.10 It
seems that the rhenium-containing phosphine 14a can better promote the above mentioned reaction
in less polar solvents. The dielectric constants of the employed solvents are also summarized in
Figure 2.15 as one quantitative measure of their polarities.56
Possibly it is less necessary to stabilize polar intermediates in reactions catalyzed by 14b.
For example, the zwitterionic intermediates might have intramolecular charge-charge interactions
that somehow lower the energy, as sketched in Figure 2.16.
Figure 2.16. Possible substrate-catalyst interaction that may be a factor in solvent rate trends.
(worst) CH3CN < CH2ClCH2Cl << PhCl < C6H6 (best)
(more polar) (37.5) (10.4) (5.5) (2.3) (less polar)
rhenium-
containing
catalyst
O
R
covalent
bond
R'
bipolar
interaction
substrate
2 Catalysis with Rhenium-Containing Phosphines 33
2.2.2.2 Temperature Dependency
As noted above, many side products accompany the intramolecular Morita Baylis Hillman
reaction of C1Ph in CH3CN. Since 14a is an effective catalyst for the nitroaldol reaction,57 in
which it is believed to act as a Brønsted base, aldol and related condensation products might form.
Since selectivities often increase at lower temperatures, the reaction with 14a was carried out at 0
°C. However, no significant improvement was realized, though the reaction seemed to run faster at
the beginning. Previous studies have shown that the Morita Baylis Hillman reaction rate can be
increased in special cases by lower temperatures.58
In this context, the first enolate generated, a zwitterionic structure like XIV (Scheme 1.8)
could be syn or anti configured (XIVa, XIVb). The syn isomer XIVa may be more stable, as the
charges are closer together, while the anti one XIVb is less stable. The less stable isomer might
react more rapidly with aldehydes to give side products. At lower temperature, the enolate
selectivity might be increased, and/or reactions of XIVb might be suppressed.
2.2.2.3 Substrate Dependency
To delineate the scope of the catalyst 14a, the different substrates in Figure 2.8 were tested.
A distinct substrate dependency was found.
First, the five-membered ring products always formed much faster than the six-membered
ones. This is nicely illustrated for the substrates C1(p-Tol) and C2(p-Tol) in Figure 2.9. This may
reflect a more favorable exo-trig transition state for the cyclization step XIV→XV in Scheme 1.8.59
Other results were less clear cut. For example, C1Ph was transformed quickly to C1Phprod
while C2Ph gave no C2Phprod at all. In the latter case, 31P{1H} NMR monitoring showed
complete consumption of catalyst 14a (CH2PPH2, 8.1 ppm). Instead, several signals at ca. 50 ppm
were detected, suggesting a positively charged phosphorus atom.
Not only the chain length of the substrates plays an important role but also the nature of the
electron withdrawing group (EWG) in Schemes 1.5 and 1.8. If the group is polarizable (Ph(CO), p-
34 2 Catalysis with Rhenium-Containing Phosphines
Tol(CO)) with a distinct ability to delocalize electron density, good reactivity was obtained. Also
the methyl-substituted substrates showed good reactivity.
Other substrates, for example all alcohol derived esters, were very unreactive and full
conversion to the product was never obtained. In contrast, if the alkoxy group was substituted by a
thioalkoxy group (C1S(i-Pr)), the substrate reacted smoothly. Sulfur is of course less
electronegative and more polarizable than oxygen. However, the reason for the much improved
yields remains unclear. The subtle electronic differences between ketone, ester, and thioester
enolates may play a role.
2.2.2.4 Formation of a Side Product
Not all substrates in Figure 2.8 were investigated in detail. There were several reasons. For
example, C1Me was very unstable as a pure compound and even decomposed upon storage at low
temperatures (–70 °C). After four weeks, a very viscous oil was obtained. It is supposed that
addition and condensation products and/or polymers were obtained. On the other hand the bis
(methyl ketone) C1Me2 (Scheme 2.3) showed acceptable stability, but the yields of C1Me2prod
were quite low. A chromatographic workup of the reaction of C1Me2 with 10 mol% 14b (see
section 2.3.1.1) in PhCl at room temperature showed that C1Me2prod subsequently reacted to give
the aldol side products 19 and 20.
The product 20 had been previously reported by Roush.60 Similar to this work, the synthesis
involved an aldol reaction of C1Me2prod that had been prepared by a Rauhut Currier reaction. The
identity of 20 was established by 1H NMR and 13C{1H} NMR spectroscopy. The aldol intermediate
19 has not been previously reported and the structural assignment rests solely on 1H NMR and
13C{1H} NMR. When consumption of C1Me2 was complete, the crude mixture was analyzed by
1H NMR, which indicated the product ratio in Scheme 2.3. The desired product C1Me2prod
dominated.
2 Catalysis with Rhenium-Containing Phosphines 35
Scheme 2.3. Formation of cyclic side products 19 and 20 from the reaction of C1Me2 with 10
mol% 14b in PhCl.
Note that in principle C1Me2prod could undergo a second type of intramolecular aldol
reaction. Murphy showed this to be the major isomer when C1Me2prod was treated with base as
opposed to a nucleophilic catalyst.20 Interestingly, such products were not detected.
2.2.2.5 Oxidation of the Catalyst
In the course of many catalytic experiments, the phosphine oxides of 14a61 and other
catalysts were often detected by 31P{1H} NMR spectroscopy when the reaction was complete.
There are several possible explanations for this.
All solvents were carefully freeze pump thaw degassed. However, some oxygen could be
introduced with the substrate charge. All Morita Baylis Hillman substrates were stored at –70 °C
under air. Most of the substrates are oils or liquids and therefore they could contain dissolved
oxygen. Of course, the samples were evacuated and backfilled with N2 before solvent was added,
O
O
O
O
OO
OH
Aldol Addition Condensation
14b10 mol%
10 : 6 : 1
NMR ratio, crude mixture
C1Me2
C1Me2prod 19 20
36 2 Catalysis with Rhenium-Containing Phosphines
but there might be oxygen residues.
Another possibility is that traces of oxygen diffused through the caps or septa as all these
experiments were conducted in Schlenk flasks, NMR tubes, or GC vials. But this is unlikely as
several times parallel to the catalysis experiment a catalyst solution was stored under the same
conditions as the catalysis mixture. In such catalyst solutions phosphine oxide was never found by
31P{1H} NMR. This raises the possibility that oxidation could be effected by the substrates
themselves. A possible sequence is illustrated in Scheme 2.4.
Consider the zwitterionic intermediate 21 (similar to XV, Scheme 1.8). This intermediate
could either eliminate the tertiary phosphine (Morita Baylis Hillman route) to give C1Phprod or
follow a "reductive elimination" route. For this an enolization to 22 must first take place. At the
same time or subsequently, a cyclic intermediate 23 is formed. The driving force would be the
formation of the very stable phosphorus-oxygen bond. Roush also speculated about phosphorus-
oxygen interactions in the intramolecular Rauhut Currier and following aldol reactions, and
suggested an important role in determining the regiochemistry of certain products.60
The organic residue would be released as a cyclopropene or 1,3-dipole. Although these are
high energy species, the formation of the strong phosphorus-oxygen linkage would to some extent
compensate. The amount of catalyst derived organic product must by necessity be small. However,
as shown in Figure 2.17, two alcohols, 24 and 25, which would be logical products from 23, could
be provisionally identified from the reaction of C1Ph and 10 mol% 14a in PhCl at –25 °C.
Curiously, the analogous reaction at room temperature gave much less phosphine oxide and
significantly higher yields of C1Phprod. Both 24 and 25 have been reported in the literature and the
1H NMR data agree well with that in Figure 2.17.62,63
2 Catalysis with Rhenium-Containing Phosphines 37
Scheme 2.4. Possible pathway for the formation of catalyst-derived phosphorus oxide.
Figure 2.17. 1H NMR spectrum of side products obtained from the reaction of C1Ph with 10 mol%
14a at –25 °C in PhCl.
O
O
Ph
PR3
H
OH
O
Ph
PR3
OH
O
Ph
PR3
MoritaBaylis Hillman
route
OH
O
Ph
PR3
reductiveelimination
route
O=PR3 + organic products
21 C1Phprod
22 23
ppm 2.03 . 0 4.05.06 . 0 7 . 0
24
25 Ph
OH
H H
H
H
Ph
OH
H
H
38 2 Catalysis with Rhenium-Containing Phosphines
2.2.2.6 Enantioselective Catalyses
The data in Table 2.2 show that appreciable enantiomeric excesses can be obtained with
catalyst (S)-14a. Modification of the aryl substituents of this catalyst that give still higher ee values
are described in section 2.3. However, the data in table 2.1, which involve catalysts with additional
stereocenters, merit a preliminary analysis at this stage. The catalyst (SReSC)-[(η5-
C5H5)Re(NO)(PPh3)(CHCH3PPh2) ((SReSC)-15b) contains a carbon stereocenter that is closer to
the reactive phosphorus center than rhenium. Naively, enhanced enantioselectivities might be
expected. However, the methyl substituent influences a number of other factors, such as the
conformational equilibria sketched in Figure 2.18. Whereas 14a should have two PPh2-CH2Re
conformations of approximately equal energy (14a-I, 14a-II), (SReSC)-15b should have one much
more stable conformer 15-II. Otherwise, a substituent must be placed in the interstice between the
phenyl substituents on phosphorus.
In any event, (SReSC)-15b gives C1Phprod with a lower ee value then (S)-14a. The phenyl-
substituted analog, (SReSC)-15a, gives hardly any product at all. It would have been of interest to
probe the opposite rhenium/carbon diastereomers, both of which are known, but this was not done
due to improved results obtained with simpler systems below (section 2.3).
One might naively assume that a phosphorus stereocenter would result in improved
enantiomeric excess, and there has been much recent interest in P-stereogenic phosphorus.64,65
However, the two complexes initially assayed featured cyclohexyl/phenyl and cyclohexyl/t-butyl
substituents at phosphorus. Lower yields and enantioselectivities were obtained. Perhaps aliphatic
substituents are deleterious, a hypothesis reinforced by improved results with aryl/aryl analogs
below (section 2.4).
2 Catalysis with Rhenium-Containing Phosphines 39
Figure 2.18. Newman projections down the Ph2P-CHR[Re] bonds of catalysts 14a and 15.
2.2.2.7 Activity of Different Phosphorus Centers
Some of the catalysts assayed feature diphenylphosphinocyclopentadienyl ligands,
providing a second type of phosphorus center. Figure 2.19 shows a qualitative comparison of the
different types of phosphorus donors screened in reactions of C1Ph. These in turn allow some
tentative qualitative conclusions.
The phosphorus center of (S)-14a is the most active one. Introduction of the phenyl group at
the spacer ((SReSC)-15a) inactivates center 1' as sketched above. If catalyst (SReSC)-16a is also
inactive, someone could guess that also center 2 is inactive. In contrast, (SReSC)-16b showed some
activity which could then be attributed to center 1. Since there is lower reactivity compared to (S)-
14a, perhaps steric demand of the diphenylphosphinocyclopentadienyl ligand hinders the approach
of substrate to the active center 1. A possible model is illustrated in Figure 2.20. The
diphenylphosphinocyclopentadienyl moiety would be expected to prefer the interstice between the
NO and the methylene group.
PhPh
H[Re]
H
PhP
Ph
[Re]H
H
P
PhP
Ph
[Re]H
R
PhPh
R[Re]
H
P
14a-I 14a-II
PhP
Ph
HH
[Re]
14a-III
PhP
Ph
HR
[Re]
15-I 15-II 15-III
[Re] = (η5-C5H5)Re(NO)(PPh3)
(comparablestabilities)
(most stable)
40 2 Catalysis with Rhenium-Containing Phosphines
Figure 2.19. Qualitative comparison of different types of phosphorus donor groups with respect to
reactivity towards C1Ph.
Figure 2.20. Newman like projection down the η5-C5H4PPh2-Re axis of (S)-16b.
The relative basicities of centers 1/1' and 2 are also relevant to this analysis. The reaction
mixture derived from C1Ph and (S)-16b (10 mol%) in PhCl at room temperature was investigated
by 1H NMR and 31P{1H} NMR spectroscopy after 240 h (Figure 2.13). About 40 % of C1Phprod
was detected besides unconverted substrate. The CH2PPh2 phosphorus signal of the catalyst (7.1
ppm) had been replaced by several new signals (45-50 ppm). Possibly, catalyst-substrate adducts
that block the active center formed. However, the phosphorus signal at –17.7 ppm (η5-C5H4PPh2)
was unchanged. This suggests that the latter phosphorus atom might not be involved in the reaction.
To gain additional insight, (S)-16b and HBF4.OEt2 (1 equiv.) were combined in PhCl at room
temperature. The 31P{1H} NMR spectrum showed a dramatic shift for the CH2PPh2 center (7.1
ON
Ph3P
CH2PPh2
PPh2
stericinteraction
(S)-16b
ON PPh3Re
CPh2P H
H
ON PPh3Re
CPh2P Ph
H
ON PPh3Re
CPh2P Ph
H
PPh2
ON PPh3Re
CPh2P H
H
PPh2
center 1 center 1' center 1' center 1
center 2 center 2
observation: activity inactivity inactivity low activity
conclusion: center 1 active center 1' inactive center 2 also inactive center 1 active (but hindered by center 2)
(SReSC)-16a (S)-16b(S)-14a (SReSC)-15a
2 Catalysis with Rhenium-Containing Phosphines 41
ppm to 30.3 ppm), consistent with the formation of a phosphonium salt. However, the
diphenylphosphinocyclopentadienyl phosphorus signal remained unchanged (–17.7 ppm). This
suggests that the CH2PPh2 phosphorus atom is the more basic in (S)-16b and should therefore be
more reactive.
2.2.2.8 Conclusions
Conclusions from the experiments in section 2.2.1 can be summarized as follows. The
catalyst (η5-C5H5)Re(NO)(PPh3)(CH2PPh2) (14a) was the most reactive and also gave the highest
enantioselectivities. Substitution of the phenyl groups of the diphenylphosphino center by aliphatic
moieties (17a,b) resulted in lower rates and enantiomeric excesses. Modification of the methylene
spacer (15a,b) showed neither enhanced rates nor enantiomeric excesses. The introduction of a
second phosphorus donor group at the cyclopentadienyl ligand (16a,b) decreased the reaction rates
and the enantiomeric excesses. Therefore, catalysts for future experiments should provide a basic
structure similar to 14a, in which the phenyl moieties of the CH2PPh2 group are replaced by aryl
groups. The successful implementation of this strategy is described in sections 2.3 and 2.4.
42 2 Catalysis with Rhenium-Containing Phosphines
2.3 First Series of Improved Catalysts
2.3.1 Results
2.3.1.1 Increased Nucleophilicity by Electron Rich Aryl Substituents
The catalyst 14a effected the cyclization of C1Ph to C1Phprod with short reaction times,
moderate catalyst loadings (10 mol%), and acceptable enantiomeric excesses (up to 62%).
Nonetheless, it was sought to improve on this benchmark. As was concluded in section 2.2.3, the
best way to improve the catalyst might be the modification of the aryl groups at the active
phosphorus center. The new rhenium-containing phosphines assayed are depicted in Figure 2.21.
Figure 2.21. New rhenium-containing phosphines 14b-d.
The phenyl substituents at the active phosphorus atom were substituted in the para positions
by electron donating groups (R = p-Tol (14b), p-C6H4OCH3 (14c), p-C6H4N(CH3)2 (14d)). The
syntheses and characterizations are detailed in Chapter 3.
The reaction profiles for selected Morita Baylis Hillman substrates (C1S(i-Pr), C2(p-Tol),
and C2Me) with the new catalysts (14b,c) are depicted in Figure 2.22. For comparison, the data for
14a in C6H6 (blue traces, ) are repeated for substrates C1S(i-Pr) and C2(p-Tol) from Figure 2.10;
the data for C2Me are new. As supported by the reaction profiles, the new catalysts provide distinct
rate enhancements. The data for C2Me are representative and may be analyzed as follows.
ON PPh3Re
CH2PR2
R = b, p-Tol
c, p-C6H4OCH3
d, p-C6H4N(CH3)2
14
2 Catalysis with Rhenium-Containing Phosphines 43
Figure 2.22. Reaction profiles for the cyclizations of various substrates with 10 mol% of 14a-c in
C6H6; the data for C1S(i-Pr) and C2(p-Tol) involving 14a are repeated from Figure 2.10.
Consider the results for the first five hours. The reaction with catalyst 14a is clearly the
slowest (blue trace, ▲). Much faster are catalysts 14c (red trace, ●) and 14b (green trace, ■), which
are quite comparable and likely within experimental error. After ca. 6 h this picture changes. The
reaction with 14a remains slower than that with 14b. Nonetheless, product yields are high in each
case (90% and 85%, respectively). On the other hand, the reaction with 14c slows down and reaches
a maximum product yield of 70%. The other substrates in Figure 2.22 (C1S(i-Pr) and C2(p-Tol))
show similar behavior, as do some uncharted ones (C1Ph, C1Ph2, and C1S(i-Pr)2), for which data
are given in table form in section 2.5.3. However, the reactions with 14b reaches completion earlier
than with 14a.
Added insight into the behavior of 14c is provided by Figure 2.23. Here the consumption of
C2(p-Tol) with catalysts 14a-c is charted, and the product yield data in Figure 2.22 are reproduced.
0 5 10 15 200
20
40
60
80
catalyst:3 ( ) 14a3 ( ) 14b56
( ) 14c
C1S(i-Pr)
C1S(i-Pr)C1S(i-Pr)
C2Me
C2Me
C2Me
C2(p-Tol)
C2(p-Tol)
C2(p-Tol)
Pro
duct
Yie
ld (%
)
Time (h)
44 2 Catalysis with Rhenium-Containing Phosphines
Figure 2.23. Reaction profiles for the cyclization of C2(p-Tol) with 10 mol% 14a-c in C6H6.
Conversion refers to the consumption of substrate. The product yield data are repeated from Figure
2.22.
This graphic shows nicely that the more reactive catalyst leads to more side reactions. For
catalyst 14a (blue traces, ▲) the curve for the product yield (rising trace) was nearly symmetric
compared to the substrate conversion (falling trace). Both graphs cross each other at ca. 50%
showing that after consumption of 50% of C2(p-Tol), 50% of C2(p-Tol)prod was obtained.
Additionally, after 72 h (here not explicitly charted) the conversion to product was essentially
complete (99%). This indicated a low rate of side product formation. For catalyst 14b (green traces,
●) the crossing point of both curves is somewhat lower than 50% and the product yield after
complete substrate conversion was only 95%. Obviously, more side reactions were promoted by
14b. The situation with the most reactive catalyst 14c (red traces, ▼) was even more pronounced.
Substrate consumption was fastest and the crossing point is found at ca. 40%. After 24 h, substrate
conversion was complete but the product yield was only 75%. This indicates a much higher
proportion of side products.
Catalyst 14d gave even greater proportion of side products, and for this reason is not charted.
0 10 20 30 400
20
40
60
80
Sub
stra
te C
onve
rsio
n (%
)
Pro
duct
Yie
ld (%
)
Time (h)
catalyst:5 ( ) 14a5 ( ) 14b5 ( ) 14c5
C2(p-Tol)prod
100
80
60
40
20
C2(p-Tol)
2 Catalysis with Rhenium-Containing Phosphines 45
Substrate conversions were incomplete, and 31P{1H} NMR indicated deactivation of the catalyst.
Nonetheless, initial rates of product formation were faster than with 14a-c. Table 2.3 summarizes
the product yields obtained with 14d and the other catalysts.
Table 2.3. Product yields for reactions of various substrates with 10 mol% 14a-d in PhCl.
Product yields (1H NMR, %)
Substrate Catalyst
C1Ph C1Me C2OEt C2(p-Tol) C2Me C1OEt C1S(i-Pr)
14a 90c 90c ----e 90 ----e ----d 50b 14b 95 60 ----d 80 65a 30b 70
14c 65a 55 15b 80 60c 45b 70c 14d 20a 20 25a 50 55 20a 35b
a Incomplete conversion, 31P{1H} NMR shows a little remaining catalyst. b Incomplete conversion,
31P{1H} NMR shows some remaining catalyst. c Nearly complete conversion, 31P{1H} NMR
shows a little remaining catalyst. d No reaction. e No data available.
Only in one case was catalyst 14d superior. This was for substrate C2OEt, which always
reacted sluggishly and presumably benefits from the enhanced phosphorus nucleophilicity of 14d.
However, the yield remained low (25%).
Figure 2.10 established that with catalyst 14a, C6H6 was a superior solvent to PhCl for
several substrates. The same effect was also found for catalysts 14b,d. Additionally, product
purities were enhanced in C6H6. Additional data are summarized in table form in section 2.5.3.
The Rauhut Currier substrates C2Me2 and C2(p-Tol)2 gave no reactions with catalysts 14a-
c. However, 14d gave some products, as assayed by 1H NMR spectroscopy of the crude mixtures
(C2Me2prod: 40%, C2(p-Tol)2
prod: 20%).20,60,66 However, due to low yields these products were
never isolated and further characterized although the latter is a new compound. In both cases some
inactivated 14d was detected by 31P{1H} NMR directly after combination of the catalyst and the
substrate solution.
46 2 Catalysis with Rhenium-Containing Phosphines
2.3.1.2 Enantioselective Catalyses
As 14a-c contain a chiral rhenium center, many reactions were also performed with
enantiopure catalysts. The reaction products were assayed for enantiomeric excess by chiral HPLC.
The ee values obtained in C6H6 are summarized in Table 2.4.
Table 2.4. Product ee values for reactions of various substrates with 10 mol% (S)-14a-c in C6H6.
The values for 14a are repeated from Table 2.2.
Catalyst
(S)-14a (S)-14b (S)-14c
ee(C1Phprod) 62% 74% 0%
ee(C1S(i-Pr)prod) 74% 78% 41%
ee(C2(p-Tol)prod) 38% 31% 26%
ee(C2Meprod) 70% 59% –11%
ee(C1Ph2prod) 42% 51% ----a
ee(C1S(i-Pr)2prod) 52% 49% ----a
average 56% 57% ----a
a Not determined.
Compared to (S)-14a, the catalyst with the tolyl phosphorus substituents ((S)-14b) gave
higher ee values for the five-membered ring Morita Baylis Hillman products C1Phprod and C1S(i-
Pr)prod (74% and 78% vs. 62% and 74%), paralleling the rate trends. However, the values for the
six-membered ring systems C2Meprod and C2(p-Tol)prod were lower (31% and 59% vs. 38% and
70%). The Rauhut Currier products did not exhibit a well defined trend. One ee value was higher
(C1Ph2prod, 42% vs. 51%) and one was comparable (C1S(i-Pr)2
prod, 52% vs. 49%). The lowest ee
values were obtained for the catalyst with the p-C6H4OCH3 phosphorus substituents ((S)-14c).
2 Catalysis with Rhenium-Containing Phosphines 47
Although this catalyst gave the fastest reactions (but not the highest yields), the ee values were very
random for the five-membered rings as well as for the six-membered systems; curiously, with
C2Me the opposite enantiomer dominated. The enantiopure catalyst (S)-14d was not employed
because of the low product yields and selectivities.
2.3.1.3 Preparative Experiments
All data presented above were obtained from reactions with very small amounts of
substrates (ca. 0.0050-0.010 g) and catalysts as well. Many reactions were carried out in NMR tubes;
the yields were directly calculated from 1H NMR spectra. Such measurements give an idea about
the reaction rates but different yields are often obtained for the purified products. To show that for
the reactions given above pure products can be obtained in significant yields, some reactions were
carried out in larger scales and worked up to give pure products. The NMR yields and the
preparative experiments are compared in Tables 2.5 and 2.6.
The preparative yields were sometimes significantly lower than those obtained from 1H
NMR experiments. Additionally, it was shown that these products are not complete stable on silica
gel, as TLC spots smear from the baseline. In any case, good to excellent yields were obtained with
C6H6 as solvent. This was in large part due to the avoidance of column chromatography. Instead, a
type of silica gel filtration was employed, as detailed in section 2.5.4.
Table 2.5. Preparative and 1H NMR product yields for reactions of various substrates with 10
mol% of 14a in C6H6.
Substrate
C1Ph C1S(i-Pr) C2(p-Tol) C2Me C1Ph2 C1S(i-Pr)2
Yield (1H NMR) 95% 91%b (56%a) 99% 90% 92% (87% a) 98%
Yield (prep.) 91% 99% (53%a) 91% 86% 87% (70% a) 82%
a Obtained in PhCl. b This value may be overstated due to the poor signal to noise ratio and
resolution.
48 2 Catalysis with Rhenium-Containing Phosphines
Table 2.6. Preparative and 1H NMR product yields for reactions of various substrates with 10
mol% of 14c in PhCl.
Substrate
C1Ph C1S(i-Pr) C2(p-Tol) C2Me C1Ph2
Yield (1H NMR) 61% 70% 51% 84% 71%
Yield (prep.) 55% 58% 40% 54% 50%
2 Catalysis with Rhenium-Containing Phosphines 49
2.3.2 Discussion
Because of the good results that were obtained with catalyst 14a, further optimization of this
lead structure was sought. Therefore, the nucleophilicity of the active phosphorus center was
enhanced. Mayr has shown that the nucleophilicities of tertiary aryl phosphines strongly depend on
the substituents in the para position.67 In Table 2.7 the nucleophilicity factors (N) of some tertiary
aryl phosphines are summarized. Higher values correspond to greater nucleophilicities.
Table 2.7. Nucleophilicity factors (N) depending on para substituents (R) of aryl phosphines
reported by Mayr.
As a chlorine substituent withdraws electrons from the ring system (–i effect), the
nucleophilicity is decreased. On the other hand, a strong +i effect effects higher electron density,
and higher nucleophilicity is obtained (CH3). This is superceded upon the introduction of OCH3
and N(CH3)2 substituents, which donate even more electron density to the aromatic systems by
their +m effects. The nucleophilicity of the phosphorus center is raised (OCH3 < N(CH3)2). All of
these trends should be paralleled in the nucleophilicities of the rhenium-containing phosphines.
The chemical shifts of 31P{1H} NMR signals do not necessarily correlate with the
nucleophilicity. But if very similar substituted phosphines are compared they can give an idea about
the electron density at the phosphorus atom. To a first approximation, the lower field the chemical
R N
Cl 12.58 H 14.33
CH3 15.44 OCH3 16.17
N(CH3)2 18.39
P
R
R
R
50 2 Catalysis with Rhenium-Containing Phosphines
shift, the higher the electron density (higher shielding), leading possibly to increased nucleophilicity.
The rhenium-containing phosphines show the following 31P{1H} NMR shifts for the PR2 moiety:
8.1 (14a, Ph),38 5.8 ppm (14b, p-Tol), 5.1 ppm (14c, p-C6H4OCH3), 2.8 ppm (14d, p-
C6H4N(CH3)2) (see also Table 3.1, Chapter 3). This fits to the catalytic experiments that showed in
principle that 14d is the most reactive complex, followed by 14c, and 14b.
With regard to enantioselectivities, the more nucleophilic catalyst (S)-14c gives distinctly
lower ee values than (S)-14a,b. Although (S)-14b usually gives higher ee values than less
nucleophilic (S)-14a, there are exceptions. In the author's opinion there are no obvious single
parameter rationales for these trends, which must be regarded as empirical. In general, the Morita
Baylis Hillman and Rauhut Currier reactions are very sensitive to reaction conditions. Accordingly,
the nature or even the identity of the enantiomer-determining step may vary, rendering
systematization more challenging.
Subsequent efforts directed at catalyst optimization involved aryl groups that did not contain
heteroatoms, as described in the next section.
2 Catalysis with Rhenium-Containing Phosphines 51
2.4 Second Series of Improved Catalysts
2.4.1 Results
2.4.1.1 Increased Steric Demand at the Catalytic Center
As the first series of improved catalysts showed good reactivity, and furthermore in the case
of 14b increased ee values for the five-membered ring systems, the possibility of increased steric
demand and electron delocalization at the catalytic center was considered. The enhanced electron
delocalization should not be provided by heteroatoms but by extension of the phenyl π-systems. It
was decided to synthesize two new rhenium-containing phosphines, one with two α-naphthyl and
one with two 2-biphenyl moieties (Figure 2.24).
Figure 2.24. New rhenium-containing phosphines 14e,f.
The syntheses were analogous to those of the other rhenium-containing phosphines and are
detailed in Chapter 3. When several representative substrates were treated with 14e in C6H6, no
reactions were observed. Even after several days no conversion was apparent. On the other hand the
catalyst with the naphthyl moieties (14f) showed very slow reactions with C1Ph and C1S(i-Pr),
nearly no reactivity towards C1Ph2, and no reactivity towards all other tested substrates (C1Me2,
C2(p-Tol)2, C1S(i-Pr)2, C2(p-Tol), and C2Me). The reaction profiles for C1Ph (■) and C1S(i-Pr)
(▲) with 14f (10 mol%) in C6H6 are depicted in Figure 2.25.
ON PPh3Re
CH2PR2
R = e, 2-biphen
f, α-naph
14
52 2 Catalysis with Rhenium-Containing Phosphines
Figure 2.25. Reaction profiles for the cyclizations of C1Ph (■) and C1S(i-Pr) (▲) with 10 mol%
14f in C6H6. The conversion of 14f to other species in the former reaction is given by the orange
trace (*).
The reactivity of catalyst 14f compared to 14a-d was very low. The complete consumption
of C1Ph took ca. 168 h, but at least conversion to product was high (90%, 1H NMR).
Approximately one third of the catalyst was consumed during this time, as shown by the orange
trace (*) in Figure 2.25. To calculate these values, the 1H NMR signals of the cyclopentadienyl
ligands of the active (s, 5.07 ppm) and the consumed catalyst (s, 5.85 ppm) were integrated (other
cyclopentadienyl signals were not detected). A catalyst solution without any substrate showed no
conversion after 24 h. Also other solutions with unconverted substrates showed only slight catalyst
loss (for example, the C1Me2 mixture showed ca. 8% consumed catalyst after 168 h). The
converted catalyst is assumed to be phosphine oxide, as a sample of 14f was kept under oxygen and
showed the same NMR signals as the consumed catalyst. Additionally, these chemical shifts trend
parallel those reported for the 14a and oxidized 14a (O=14a) analogs, which have been completely
characterized.61
For substrate C1S(i-Pr) catalyst 14f showed low reactivity. However, a NMR yield of 40%
0 20 40 60 80 100 120 140 160 1800
20
40
60
80
0
20
40
60
80
Con
verte
d C
atal
yst (
%)
Pro
duct
Yie
ld (%
)
Time (h)
C1Ph
C1S(i-Pr)
2 Catalysis with Rhenium-Containing Phosphines 53
was reached after 166 h. The reaction was then terminated. Phosphine oxide was again apparent but
is not charted.
Because of the low reactivity of 14f, no enantioselective catalysis was attempted. Other
substituents at the nucleophilic phosphorus atom were sought that could confer higher reactivity and
hopefully increase the enantiomeric excess as well.
2.4.1.2 Introduction of Chirality at the Catalytic Center
A new stereocenter at the nucleophilic phosphorus atom was introduced, as shown for
(SReSP)/(SReRP)-17c,d in Figure 2.26. These are comparable to (SReSP)/(SReRP)-17a,b, but feature
two different aryl substituents, one being a phenyl group and the other a naphthyl residue.
Figure 2.26. New rhenium-containing P-stereogenic phosphines (SReSP)/(SReRP)-17c,d.
The synthesis is again detailed in Chapter 3. Catalysts (SReSP)/(SReRP)-17c and
(SReSP)/(SReRP)-17d were always employed as diastereomeric mixtures. Before the reactions were
started, a small amount of catalyst was assayed by 1H NMR spectroscopy and the diastereomeric
excess (de) was calculated from the integrals of the cyclopentadienyl 1H NMR signals, which gave
different singlets for each diastereomer (sections 3.2.1.3 and 3.4.2). The absolute configurations of
the phosphorus centers were not determined, so in each case one diastereomer is termed
"diastereomer 1" and the other "diastereomer 2". The two diastereomeric mixtures of
(SReSP)/(SReRP)-17c (10 mol%) were tested with substrates C1Ph (■) and C1S(i-Pr) (▲) and the
reactions were again followed by 1H NMR. The results are depicted in Figure 2.27.
ON PPh3Re
CH2PRR'∗
R/R' = c, Ph/α-naph
d, Ph/β-naph
(SReSP)/(SReRP)-17
54 2 Catalysis with Rhenium-Containing Phosphines
Figure 2.27. Reaction profiles for the cyclizations of C1Ph (■) and C1S(i-Pr) (▲) with 10 mol% of
(SReSP)/(SReRP)-17c in C6H6.
All reactions were carried out in C6H6 and in the best case the reaction was finished after 24
h (C1Ph, diastereomer 2). The diastereomers of (SReSP)/(SReRP)-17c (diastereomer 1, 94% de vs.
diastereomer 2, –92% de) exhibited significantly different reaction rates. With substrate C1Ph,
diastereomer 1 gave a slower reaction than diastereomer 2 (yields: 28% vs. 52%, 3.5 h; 78% vs.
86%, 24 h). A similar effect was noticed with C1S(i-Pr), where diastereomer 2 was also the more
reactive (yields: 12% vs. 28%, 6.5 h; 76% vs. 83%, 48 h).
A β-naphthyl group is commonly considered less bulky than a α-naphthyl group.68 Thus,
analogous experiments were conducted with the slightly less hindered nucleophile (SReSP)/(SReRP)-
17d. The reaction profiles are depicted in Figure 2.28. The rates were much faster than with
(SReSP)/(SReRP)-17c. The rate differences between the diastereomers (diastereomer 1, 48% de vs.
diastereomer 2, –92% de) were not as large as with (SReSP)/(SReRP)-17c. For C1S(i-Pr) the rates
were nearly identical (yields: 49% vs. 48%, 2.0 h; 75%, 3.5 h, both). However, C1Ph showed some
differences (yields: 57% vs. 33%, 0.5 h; 91% vs. 71%, 2.0 h).
0 10 20 30 40 50 60 700
20
40
60
80P
rodu
ct Y
ield
(%)
Time (h)
C1Ph Dia_1C1Ph Dia_2
C1S(i-Pr) Dia_1
C1S(i-Pr) Dia_2
(SReSP)/(SReRP)-17c:
de(diastereomer 1) = 94%de(diastereomer 2) = −92%
2 Catalysis with Rhenium-Containing Phosphines 55
Figure 2.28. Reaction profiles for the cyclizations of C1Ph (■) and C1S(i-Pr) (▲) with 10 mol% of
(SReSP)/(SReRP)-17d in C6H6.
2.4.1.3 Enantioselective Catalyses
The ee values of the products from Figures 2.27 and 2.28 were determined by HPLC. The
results are detailed in Table 2.8. The ee values with catalyst (SReSP)/(SReRP)-17c were quite
impressive. Diastereomer 2 (–92% de) gave the products C1Phprod and C1S(i-Pr)prod with ee
values of 85% and 88%. The ee value for diastereomer 1 was also good, with 82% for C1S(i-Pr).
Only the value for C1Phprod (53%) was moderate, the lowest value within this set of experiments.
In the case of the sterically less demanding β-naphthyl moiety, ee values ranging from 61% to 71%
were obtained. However, these were comparable to the results with 14a (Table 2.2).
0 5 10 15 200
20
40
60
80P
rodu
ct Y
ield
(%)
Time (h)
C1Ph Dia_1C1Ph Dia_2
C1S(i-Pr) Dia_1
C1S(i-Pr) Dia_2
(SReSP)/(SReRP)-17d:
de(diastereomer 1) = 48%de(diastereomer 2) = −92%
56 2 Catalysis with Rhenium-Containing Phosphines
Table 2.8. Product ee values for reactions of C1Ph and C1S(i-Pr) with 10 mol% of catalysts
(SReSP)/(SReRP)-17c and (SReSP)/(SReRP)-17d in C6H6.
Catalyst
(SReSP)/(SReRP)-17c (SReSP)/(SReRP)-17d
diastereomer 1a diastereomer 2b diastereomer 1c diastereomer 2d
ee(C1Phprod) 53% 85% 64% 61%
ee(C1S(i-Pr)prod) 82% 88% 72% 71%
a 94% de. b –92% de. c 48% de. d –92% de.
2 Catalysis with Rhenium-Containing Phosphines 57
2.4.2 Discussion
2.4.2.1 Discussion of the Second Series Improved Catalysts
The experiments in this section show that the aryl substituents on the "lead structure" 14a
can be made so bulky that catalytic activity ceases. This extreme is represented by the ortho
substituted phenyl groups (2-biphen) in 14e. When the 2-biphenyl groups are replaced by α-
naphthyl groups (14f), activity is restored. Nonetheless, the rates are sluggish compared to 14a-c.
Given these results, it seemed sensible to "back off" on the steric demand a little bit by
substituting one α-naphthyl moiety by a phenyl ring (17c).68 Indeed, both diastereomers of
(SReSP)/(SReRP)-17c proved to be effective catalysts. Diastereomer 2 exhibited enhanced reactivity
compared to the opposite diastereomer 1. However, as detailed in section 3.2.3.1, it has not yet
proved possible to crystallographically characterize one of these diastereomers or a derivative.
Hence, the phosphorus configurations remain unassigned.
More interesting is the fact that the faster reactions with diastereomer 2 give also higher
enantioselectivities (Figure 2.27 and Table 2.8). This is a pleasant circumstance, as this implies that
the diastereomers don't necessarily have to be separated before use as catalysts. Although this
experiment was not carried out, the enantioselectivities derived from a 50:50 diastereomer mixture
would be expected to be comparable to those of the pure diastereomer 2. In any case, the obtained
ee values are to the author's knowledge the highest ever attained for Morita Baylis Hillman
reactions with a monofunctional phosphine catalysts and without co-catalytic additives.
The β-naphthyl-substituted catalyst (SReSP)/(SReRP)-17d did not give any significant
increase of ee values compared to 14a. Moreover, even the reaction rates were lower. Hence, this
catalyst system is not cost- or labor-effective, given the complexity of the synthesis and
diastereomer separation (section 3.2.1.3). But in any case both (SReSP)/(SReRP)-17d and
(SReSP)/(SReRP)-17c are "working systems" that likely can be further optimized with additional
substituents or by employing other types of aryl groups. One interesting direction would involve
anthracenyl homologs, such as depicted in Scheme 2.5 (top).
58 2 Catalysis with Rhenium-Containing Phosphines
Scheme 2.5. Possible future directions for catalysts based upon rhenium-containing phosphines.
2.4.2.2 Short Summary and Outlook
Intramolecular versions of the Morita Baylis Hillman and Rauhut Currier reactions have
been investigated. For this, numerous substrates were synthesized. Some of them were new and
some were already reported in the literature. All are capable of giving five- or six-membered ring
systems that are highly functionalized.
Catalytic amounts of rhenium-containing phosphines were employed to promote the
cyclizations. Most of these phosphines were new, although some had already been reported in the
Gladysz group. Reactions were generally screened by NMR or GC. However, it was shown that
significant yields could be obtained in preparative experiments. The solvent, temperature, and
catalysts have been systematically optimized. Reaction rates were found to be a strong function of
the electron densities at the nucleophilic phosphorus centers. The enantioselectivities did not
P
OH HO
SPINOL
ON PPh3Re
P
Ph
ON PPh3Re
2 Catalysis with Rhenium-Containing Phosphines 59
correlate with nucleophilicities or steric bulk, but could fine tuned by varying the sizes of the
phosphorus substituents.
Given this successful demonstration that this family of rhenium-containing phosphines can
be very good catalysts for the above mentioned reactions, there remains much room for future
developments. These might include new catalysts derived from novel cyclic diaryl phosphines, such
as the spiro system in Scheme 2.5 (bottom). This sequence would make use of the diol SPINOL,69
and would embed the nucleophilic phosphorus in a fixed conformational environment, possibly
leading to higher reactivities and selectivities.
60 2 Catalysis with Rhenium-Containing Phosphines
2.5 Experimental
2.5.1 General Data
All reactions were conducted under N2. NMR spectra were recorded at ambient probe
temperature on Bruker 300 and 400 MHz FT spectrometers and referenced to the residual solvent
signal (1H: CHCl3, 7.26 ppm; 13C{1H}: CDCl3, 77.0 ppm). IR spectra were recorded on an ASI
React IR®-1000 spectrometer. High performance liquid chromatography (HPLC) analyses were
conducted with a ThermoQuest instrument package (pump/autosampler/detector
P4000/AS3000/UV6000LP; Columns: Chiralcel OD, Chiralpak AD-H, Chiralpak AS-H, 250 × 4.6
each, Daicel). Gas chromatography (GC) was conducted with a ThermoQuest Trace GC 2000
instrument (FID; OPTIMA-5 0.25 μm capillary column, 25 m × 0.32 mm). Elemental analyses were
determined with a Carlo Erba EA1110 CHN instrument.
Solvents were treated as follows: n-pentane (Grüssing), distilled from sodium
/benzophenone; C6H6 (Grüssing), distilled from CaH2; CH3CN (Grüssing), CH2ClCH2Cl and PhCl
(2 × Fluka), distilled from P2O5 and stored over molecular sieves (A4); t-BuOH (Grüssing),
distilled from Mg; n-pentane (Grüssing), ethyl acetate, and hexanes (2 × Staub) for column
chromatography, simple distilled; isopropanol (Roth) and hexanes (Fischer) for HPLC, used
without purification; CDCl3 (99.8%, Deutero GmbH), stored over molecular sieves (A4).
Chemicals were treated as follows: Decane (99+%), PPh3 (99%, 2 × Acros), and t-BuOK
(98+%, Acros), used as received.
TLC was carried out with ALUGRAM® SIL G/UV254 plates (Macherey-Nagel). Column
chromatography was conducted using silica gel 60M (Macherey-Nagel). Celite® 535 (Fluka), used
as received.
2 Catalysis with Rhenium-Containing Phosphines 61
2.5.2 General Procedures
Procedure A: Reactions that were monitored by GC.
A GC autosampler vial (2 mL) was charged with the substrate and flushed with N2. A
solution of the catalyst was added to the substrate. The resulting solution was stirred and aliquots
(0.050 mL) were regularly assayed by GC. For this, the solvent was removed from the aliquot and
an ethyl acetate solution of decane (standard, 0.000500 M) was added.
Procedure B: Reactions that were monitored by NMR.
A Schlenk flask was charged with the substrate and flushed with N2. Then a 0.0125 M
solution of CH2ClCH2Cl (reference for 1H NMR integration) in the reaction solvent was added to
give a 0.100 M substrate solution. An equal volume of a solution that was 0.0100 M in catalyst and
0.0125 M in CH2ClCH2Cl, corresponding to 10 mol% loading, was added dropwise to the stirred
substrate solution. The mixture or an aliquot thereof was then transferred to a NMR tube, and 1H
NMR spectra were periodically recorded. The product yields and substrate conversions were
calculated from integrations of the CH2ClCH2Cl standard and characteristic substrate/product peaks
(product yield: C=CH and/or CHOH; substrate conversion: CHO).
Procedure C: Reactions to determine enantioselectivity by HPLC.
A reaction mixture was prepared in the same way as procedure B but with
enantiopure/diastereoenriched catalyst. After completion of the reaction, five volumes of hexanes
were added with stirring. The mixture was filtered through a plug of silica gel, which was rinsed
with ethyl acetate/hexanes (1:4 v/v). The solvent was removed from the filtrate and the product
purity was assayed by 1H NMR. The ee was determined by chiral HPLC (isopropanol/hexanes).
Procedure D: Reactions conducted with [catalyst-H]+ PF6– salts.
A Schlenk flask was charged with [catalyst-H]+ PF6– and C6H6. Then t-BuOK (2 equiv.)
was added with stirring. After 30 min, the mixture was filtered through a plug of Celite®, which
62 2 Catalysis with Rhenium-Containing Phosphines
was rinsed with C6H6. The filtrate was concentrated and dry n-pentane was added with stirring. A
red solid precipitated, which was isolated by filtration and dried by oil pump vacuum.
A 0.0125 M solution of CH2ClCH2Cl (reference for 1H NMR integration) in the reaction
solvent was added to the red solid to give a 0.0100 M catalyst solution. This solution was added
dropwise with stirring to an equal volume of substrate solution in the same solvent that was also
0.0125 M in CH2ClCH2Cl. The mixture was then transferred to a NMR tube, and 1H NMR spectra
were periodically recorded. The formation of product was determined as indicated in procedure B.
Procedure E: Preparative Experiments
A Schlenk flask was charged with the substrate (typically 0.100-0.150 g) and flushed with
N2. Then a 0.0125 M solution of CH2ClCH2Cl (reference for 1H NMR integration) in the reaction
solvent was added to give a 0.100 M substrate solution. This was equilibrated to 20 °C using a
cryostat. An equal volume of a solution that was 0.0100 M in catalyst and 0.0125 M in
CH2ClCH2Cl, corresponding to 10 mol% of catalyst, was cooled to 0 °C and added dropwise over a
period of 5 min. An aliquot (0.6 mL) was transferred to a NMR tube, and 1H NMR spectra were
periodically recorded. When the reaction was complete (or no subsequent reaction took place), five
volumes of hexanes were added with stirring. The mixture was filtered through a plug of silica gel,
and the plug was rinsed (1:4 v/v ethyl acetate/hexanes). The solvent was removed from the filtrate
by rotary evaporation. Reactions conducted in C6H6 gave satisfactory product purity. Reactions
conducted in PhCl were purified by column chromatography (SiO2; solvents are indicated below).
The product C1Ph2prod, which was obtained from reaction in PhCl, was not purified by column
chromatography, but contained small quantities of PhCl.
Reference NMR spectra for products CnRprod and CnR2prod. Procedures B, D, and E
involved monitoring by 1H NMR. These used peaks from products previously reported in the
literature (C1Phprod,25 C1(p-Tol)prod,43 C1Meprod,28 C1OEtprod,25 C1Ph2prod,19,20,26
C1Me2prod,26,60 C2Meprod,70 and C2OEtprod 25) or isolated in section 2.5.4 (C1S(i-Pr)prod,
C1S(i-Pr)2prod, and C2(p-Tol)prod).
2 Catalysis with Rhenium-Containing Phosphines 63
2.5.3 Analytic Experiments Listed by Figures
Reaction of C1Ph with 10 mol% of 14a38 in different solvents (Figure 2.5). Substrate C1Ph
(0.0048 g, 0.025 mmol), catalyst 14a (0.0019 g, 0.0025 mmol) and solvent (0.50 mL) were
combined as in procedure A.
The GC monitoring for solvent CH3CN gave the following data:
Time (h) Yield (%)
0 0 24 28 72 42 144 46 192 51
The GC monitoring for solvent CH2ClCH2Cl gave the following data:
Time (h) Yield (%)
0 0 3.0 59 19 75 49 80 50 82 119 95 144 95
The GC monitoring for solvent mixture CH2ClCH2Cl/t-BuOH (1:1 v/v) gave the following data:
Time (h) Yield (%)
0 0 2.0 18 22 27 96 29
64 2 Catalysis with Rhenium-Containing Phosphines
Reaction of C1Ph with 10 mol% of PPh3 in CH2ClCH2Cl (Figure 2.5). Substrate C1Ph (0.0048
g, 0.025 mmol), catalyst PPh3 (0.00070 g, 0.0025 mmol) and CH2ClCH2Cl (0.50 mL) were
combined as in procedure A. The GC monitoring gave the following data:
Time (h) Yield (%)
0 0 24 5 72 13 96 16 120 17 144 21 168 24 192 24
Reaction of C1Ph with 10 mol% of 14a in different solvents (Figure 2.5). Substrate C1Ph
(0.0048 g, 0.025 mmol), catalyst 14a (0.0019 g, 0.0025 mmol) and solvent (0.50 mL) were
combined as in procedure B.
The GC monitoring for solvent PhCl gave the following data:
Time (h) Yield (%)
0 0 0.4 60 2.0 90
The GC monitoring for solvent C6H6 gave the following data:
Time (h) Yield (%)
0 0 0.25 50 0.75 95 1.3 95
2 Catalysis with Rhenium-Containing Phosphines 65
Reaction of C1Ph with 10 mol% of 14a in CH3CN at 0 °C (Figure 2.6). Substrate C1Ph (0.0048
g, 0.025 mmol), catalyst 14a (0.0019 g, 0.0025 mmol) and CH3CN (0.50 mL) were combined as in
procedure A, but the catalyst and substrate solutions were precooled to 0 °C. The reaction mixture
was kept at 0 °C for the entire experiment. The GC monitoring gave the following data:
Time (h) Yield (%)
0 0 1.2 13 24.7 25 30.7 25 40.8 25 54.7 25
Reaction of C1Ph with 10 mol% of 14a in CH2ClCH2Cl at 40 °C (Figure 2.7). Substrate C1Ph
(0.0048 g, 0.025 mmol), catalyst 14a (0.0019 g, 0.0025 mmol) and CH2ClCH2Cl (0.50 mL) were
combined as in procedure A, but the catalyst and substrate solutions were prewarmed to 40 °C. The
reaction mixture was kept at 40 °C for the entire experiment. The GC monitoring gave the
following data:
Time (h) Yield (%)
0 0 24 18 48 19 72 18 96 20 120 22 143 23
Reaction of C1Ph with 10 mol% of 14a in PhCl with a temperature gradient from –20 °C to
room temperature (text section 2.2.1.3). Substrate C1Ph (0.014 g, 0.074 mmol), catalyst 14a
(0.0055 g, 0.0074 mmol), and PhCl (1.5 mL) were combined as in procedure B, but the catalyst and
substrate solutions were precooled to –20 °C (ice/NaCl). The reaction solution was allowed to
slowly warm to room temperature within 12 h. A 1H NMR spectrum of an aliquot showed that the
66 2 Catalysis with Rhenium-Containing Phosphines
substrate was consumed, but only side products were detected.
Reaction of various substrates with 10 mol% of 14a in PhCl (Figure 2.9). Substrate C1(p-Tol),
C2(p-Tol), C1Me, C1S(i-Pr), C1Ph2, C2Me, or C1S(i-Pr)2 (typically 0.0080-0.010 g), catalyst 14a,
and PhCl (typically 0.70-1.5 mL) were combined as in procedure B. The NMR monitoring gave the
following data:
Yield (%)
Time (h) C1(p-Tol)prod C2(p-Tol)prod C1Meprod 0 0 0 0 1 73 5 10 3 95 11 27 18 62 83 22 68 86 25 72 85 44 88 86 48 88 88 114 93
Yield (%) Yield (%)
Time (h) C1Ph2prod Time (h) C1S(i-Pr)prod
0 0 0 0 0.33 43 0.25 0 1.0 63 0.5 7 1.8 87 3.0 14 3.0 87 6.0 33
10 40 29 50
Reaction of various substrates with 10 mol% of 14a in C6H6 (Figures 2.10 and 2.23). Substrate
C2(p-Tol), C1S(i-Pr), C1Ph2, C2Me, or C1S(i-Pr)2 (typically 0.0080-0.010 g), catalyst 14a, and
C6H6 (typically 0.70-1.5 mL) were combined as in procedure B. The NMR monitoring gave the
following data:
2 Catalysis with Rhenium-Containing Phosphines 67
Yield (%) Yield (%)
Time (h) C1S(i-Pr)prod Time (h) C2Meprod 0 0 0 0
1.5 49 1.5 8 3.0 72 2.75 10 4.5 84 4.5 18 5.5 91 6.0 20
24 79 48 90
Yield (%) Yield (%)
Time (h) C1Ph2prod Time (h) C1S(i-Pr)2
prod 0 0 0 0
1.25 69 21 50 1.75 75 46 76 2.75 82 69 86 4.3 89 96 92 5.75 92 120 98
Yield (%) Conversion (%)
Time (h) C2(p-Tol)prod C2(p-Tol) 0 0 0
1.33 10 11 3.0 20 23 4.5 29 30 6.0 33 35 24 74 79 48 89 94 72 99 100
Reaction of various substrates with 10 mol% of (η5-C5H5)Re(NO)(PPh3)(CHRP(Ph)2)
((SReSC)-15a: R = Ph,51 (SReSC)-15b: R = CH3) (Figure 2.12). Substrate C1Ph, C2Me, C2Ph,
C2OEt, or C1OEt (typically 0.0080-0.010 g), catalyst 15a or 15b, and PhCl (typically 0.70-1.5 mL)
were combined as in procedure B. The NMR monitoring gave the following data:
68 2 Catalysis with Rhenium-Containing Phosphines
Complex (SReSC)-15a gave no reaction; data for the reaction with (SReSC)-15b (no reaction
occurred with C2Me, C2Ph, C2OEt, or C1OEt):
Yield (%)
Time (h) C1Phprod 0 0 72 65 120 84 168 86 240 88
Reaction of various substrates with 10 mol% of (η5-C5H4PPh2)Re(NO)(PPh3)(CHRP(Ph)2)
((SReSC)-16a: R = Ph,51 (S)-16b: R = H38) (Figure 2.13). Substrate C1Ph, C2Me, C2Ph, C2OEt,
or C1OEt (typically 0.0080-0.010 g), catalyst 16a or 16b, and PhCl (typically 0.70-1.5 mL) were
combined as in procedure B. The NMR monitoring gave the following data:
Complex (SReSC)-16a gave no reaction; data for the reaction with (S)-16b (no reaction occurred
with C2Me, C2Ph, C2OEt, or C1OEt):
Yield (%)
Time (h) C1Phprod 0 0 72 27 120 38 168 41 240 41
Reaction of various substrates with 10 mol% of (η5-C5H5)Re(NO)(PPh3)(CH2PCyR)
((SReSP)/(SReRP)-17a: R = Ph (91% de), (SReSP)/(SReRP)-17b: R = t-Bu (80% de))52 in PhCl
(Figure 2.14). These sequences were begun with protonated catalysts [catalyst-H]+ PF6– (typically
0.040-0.060 g) following procedure D. Substrate C1Ph, C2Me, C2Ph, C2OEt, or C1OEt (typically
0.040-0.060 g), catalyst (SReSP)/(SReRP)-17a or (SReSP)/(SReRP)-17b, and PhCl (typically 1.2-1.9
mL) were then combined as indicated. The NMR monitoring gave the following data:
2 Catalysis with Rhenium-Containing Phosphines 69
Complex (SReSP)/(SReRP)-17b gave no reaction; data for the reaction with diastereomeric mixture
(SReSP)/(SReRP)-17a (no reaction occurred with C2Ph, C2OEt, or C1OEt):
Yield (%)
Time (h) C1Phprod C2Meprod 0 0 0 24 90 23 72 50 96 63 168 76 216 78
Reaction of C1Ph with 10 mol% of 14a in PhCl at –25 °C (Figure 2.17). Substrate C1Ph (0.014
g, 0.074 mmol), catalyst 14a (0.0055 g, 0.0074 mmol), and PhCl (1.5 mL) were combined as in
procedure B, but the catalyst and substrate solutions were precooled to –25 °C. The reaction
solution was kept at –25 °C for 12 h. A 1H NMR spectrum of an aliquot showed that the substrate
was consumed. The mixture was chromatographed (SiO2, ethyl acetate/hexanes, 3:7 v/v). Results:
see text.
Reaction of various substrates with 10 mol% of (η5-C5H5)Re(NO)(PPh3)CH2PR2 (R = p-Tol
(14b), p-C6H4OCH3 (14c), p-C6H4N(CH3)2 (14d), α-naph (14f)) in PhCl or C6H6 (Figures 2.22,
2.23, and 2.25, Table 2.3). Substrate C1Ph, C2(p-Tol), C1OEt, C2OEt, C1Me, C2Me, C1Ph2,
C1S(i-Pr)2, C2(p-Tol)2, C1Me2, or C1S(i-Pr) (typically 0.0080-0.010 g), catalyst 14b, 14c, 14d, or
14f, and C6H6 or PhCl (typically 0.70-1.5 mL) were combined as in procedure B. The NMR
monitoring gave the following data:
Data for the reaction with 14b in PhCl:
Yield (%)
Time (h) C1Phprod C2(p-Tol)prod C1OEtprod 0 0 0 0
70 2 Catalysis with Rhenium-Containing Phosphines
0.25 24 9 -- 0.75 38 -- -- 22 69 75 10 45 87 80 16 96 26 192 30
Yield (%)
Time (h) C1Meprod C2Meprod C1S(i-Pr)prod 0 0 0 0
0.5 9 0 29 22 58 48 72 48 63 144 63
Data for the reaction with 14c in PhCl:
Yield (%) Yield (%)
Time (h) C1S(i-Pr)prod Time (h) C1Phprod 0 0 0 0
1.5 60 0.17 54 2.75 70 1.5 61 18 70
Yield (%) Yield (%)
Time (h) C2(p-Tol)prod Time (h) C2Meprod 0 0 0 0
0.25 3 0.25 2 1.5 19 1.5 30 3.0 32 3.0 49 6.0 43 4.5 63 10 51 5.75 73 23 50 7.3 83 29 51 26 87
Yield (%) Yield (%)
Time (h) C1OEtprod Time (h) C1Ph2prod
0 0 0 0
2 Catalysis with Rhenium-Containing Phosphines 71
22 18 0.25 55 45 28 1.0 64 95 39 1.75 66 192 43 2.9 71
Yield (%)
Time (h) C1Meprod 0 0
0.5 20 22 55
Data for the reaction with 14d in PhCl (monitoring was not feasible due to the complex 1H NMR
spectra; the yields are the highest reached after the given time and could only be estimated):
Product C1Phprod C1Meprod C2OEtprod C2(p-Tol)prod Time (h) 0.5 3.0 48 1.0 Yield (%) 20 20 25 50
Product C2Meprod C1OEtprod C1S(i-Pr)prod Time (h) 48 24 0.5 Yield (%) 55 20 35
Data for the reaction with 14b in C6H6 (no reaction occurred with C2(p-Tol)2):
Yield (%)
Time (h) C1Phprod C1S(i-Pr)prod C2Meprod 0 0 0 0
0.5 97 70 10 2.0 100 90 20 3.5 100 30 6.5 80 24 85
Yield (%) Conversion (%)
Time (h) C1Ph2prod C1S(i-Pr)2
prod C2(p-Tol)prod C2(p-Tol) 0 0 0 0 0
72 2 Catalysis with Rhenium-Containing Phosphines
0.5 70 10 15 18 2.0 70 25 35 48 3.5 75 45 65 70 6.5 75 65 80 91 24 95 85 100
Data for the reaction with 14c in C6H6 (no reaction occurred with C2(p-Tol)2):
Yield (%)
Time (h) C1Phprod C1S(i-Pr)prod C2Meprod 0 0 0 0
0.5 54 64 8.0 2.0 54 90 25 3.5 90 35 6.5 90 48 24 70
Yield (%) Conversion (%)
Time (h) C1S(i-Pr)2prod C1Ph2
prod C2(p-Tol)prod C2(p-Tol) 0 0 0 0 0
0.5 30 20 17 20 2.0 65 20 48 69 3.5 73 60 84 6.5 73 69 92 24 75 100
Data for the reaction with 14f in C6H6 (no reaction occurred with C2(p-Tol), C2Me, C1Ph2, C1S(i-
Pr)2, C2(p-Tol)2, or C1Me2):
Yield (%) Consumed catalyst (%)a
Time (h) C1Phprod C1S(i-Pr)prod 14f 0 0 0 0
1.5 -- -- 21 7.5 6 -- -- 24 20 5 23 72 50 14 27
2 Catalysis with Rhenium-Containing Phosphines 73
168 90 40 35 192 -- -- 35
a calculated from the cyclopentadienyl 1H NMR integrals (section 2.4.1.1).
Reaction of C1Ph and C1S(i-Pr) with 10 mol% of (η5-C5H5)Re(NO)(PPh3)(CH2PPhR)
((SReSP)/(SReRP)-17c: R = α-naph, (SReSP)/(SReRP)-17d: R = β-naph) in C6H6 (Figures 2.27
and 2.28). These sequences were begun with protonated catalysts [catalyst-H]+ PF6– (typically
0.040-0.060 g, de: see tables below) following procedure D. Substrate C1Ph or C1S(i-Pr) (typically
0.0080-0.010 g), the diastereomeric mixture of catalyst (SReSP)/(SReRP)-17c or (SReSP)/(SReRP)-
17d, and C6H6 (typically 0.70-1.5 mL) were then combined as indicated. The NMR monitoring
gave the following data:
Data for the reaction with catalyst (SReSP)/(SReRP)-17c:
Yield (%)
diastereomer 1 de = 94%
diastereomer 2 de = –92%
Time (h) C1Phprod C1S(i-Pr)prod C1Phprod C1S(i-Pr)prod 0 0 0 0 0
0.5 -- -- 11 -- 2.0 16 -- 36 7 3.5 28 5 52 14 6.5 39 12 72 28 24 78 52 86 68 48 86 76 83 72 91
Data for the reaction with catalyst (SReSP)/(SReRP)-17d:
Yield (%)
diastereomer 1 de = 48%
diastereomer 2 de = –92%
Time (h) C1Phprod C1S(i-Pr)prod C1Phprod C1S(i-Pr)prod 0 0 0 0 0
0.5 57 23 33 16
74 2 Catalysis with Rhenium-Containing Phosphines
2.0 91 49 71 48 3.5 75 86 75 6.5 87 83 24 88 88
Reactions with enantiopure/diastereomeric catalysts (Tables 2.1, 2.2, 2.4, and 2.8). Procedure C
was applied for several substrates. If reactions using racemic catalysts 14a-c were already
monitored by procedure A or B, the same reaction times were used with enantiopure catalysts ((S)-
14a-c). If the reaction was only monitored with diastereopure/diastereoenriched catalysts ((SReSC)-
15b; (SReSC)-16b; (SReSP)/(SReRP)-17a,c,d), the resulting mixtures were subsequently treated as in
procedure C.
2.5.4 Preparative Reactions
Procedure E was applied. Products were usually characterized by 1H NMR, 13C{1H} NMR,
and IR spectroscopy. If the compound was new, elemental analysis was also determined. The
products were isolated from reactions conducted in PhCl by column chromatography (SiO2;
solvents are indicated below), except the one involving C1Ph2. Yields: see text (Tables 2.5 and 2.6).
Ph(CO)C=CHCH2CH2CHOH (C1Phprod)25 was obtained (7:3 v/v ethyl acetate/n-pentane) as a
slightly yellow oil. Spectroscopic data agreed with that in the literature.
i-PrS(CO)C=CHCH2CH2CHOH (C1S(i-Pr)prod) was obtained (3:2 v/v ethyl acetate/n-pentane)
as a colorless oil. Elemental analysis calcd (%) for C9H14O2S (186.1): C 58.03, H 7.58, S 17.21;
found: C 57.63, H 7.71, S 16.78.
1H NMR (400 MHz, CDCl3, δ in ppm): 6.89 (dd, 3J(H,H) = 4.8 Hz and 2.4 Hz, C=CHCHH',
1H), 5.16-5.13 (m, CHOH, 1H), 3.74 (sep, 3J(H,H) = 7.2 Hz, (CH3)2CH, 1H), 2.78 (br s, CHOH,
1H), 2.73-2.62 (m, CHH'CHOH, 1H), 2.46-2.27 (m, C=CHCHH', CHH'CHOH, 2H), 1.91-1.82 (m,
C=CHCHH', 1H), 1.39 (d, (CH3)2CH, 3J(H,H) = 7.2 Hz, 6H); 13C{1H} NMR (101 MHz, CDCl3, δ
in ppm): 190.1 (s, (CO)C=CH), 145.2 (s, (CO)C=CH), 144.4 (s, (CO)C=CH), 75.7 (s, CHOH), 34.3
2 Catalysis with Rhenium-Containing Phosphines 75
(s, (CH3)2CH), 31.8 (s, CH2CHOH), 30.9 (s, C=CHCH2), 23.0 (s, (CH3)2CH).
IR (thin film, cm–1):71 3450 (br, νOH), 1648 (s, νCO), 1613 (s, νC=C).
p-Tol(CO)C=CHCH2CH2CH2CHOH (C2(p-Tol)prod) was obtained (4:1 v/v ethyl acetate/n-
pentane) as a slightly yellow oil. Elemental analysis calcd (%) for C14H16O2 (216.1): C 77.75, H
7.46; found: C 77.18, H 7.53.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.58 (d, 3J(H,H) = 8.2 Hz, C6H4, o to CO, 2H),
7.24 (d, 3J(H,H) = 8.2 Hz, C6H4, m to CO, 2H), 6.70 (t, 3J(H,H) = 3.9 Hz, C=CH, 1H), 4.69-4.72
(m, CHOH, 1H), 3.55 (br s, CHOH, 1H), 2.41 (s, CH3, 3H), 2.42-2.18, 1.97-1.82, 1.69-1.60 (3 m, 3
CHH', 6H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm): 199.1 (s, (CO)C=CH), 146.2 (s, C6H4, p
to CO), 142.7 (s, (CO)C=CH), 139.9 (s, (CO)C=CH), 135.0 (s, C6H4, i to CO), 129.6, 128.9 (2 s,
C6H4, o, m to CO), 64.2 (s, CHOH), 29.7, 26.3, 17.5 (3 s, 3 CH2), 21.6 (s, CH3).
IR (thin film, cm–1):71 3460 (br, νOH), 1633 (s, νCO), 1606 (s, νC=C).
CH3(CO)C=CHCH2CH2CH2CHOH (C2Meprod)70 was obtained (7:3 v/v ethyl acetate/n-pentane)
as a colorless oil. Spectroscopic data agreed with that in the literature.
Ph(CO)C=CHCH2CH2CHCH2(CO)Ph (C1Ph2prod)19,20,26 was obtained as a brown oil; the 1H
NMR spectrum of the product from PhCl, which was not chromatographed showed a small amount
of residual solvent. Spectroscopic data agreed with that in the literature.
i-PrS(CO)C=CHCH2CH2CHCH2(CO)S(i-Pr) (C1S(i-Pr)2prod) was obtained (1:9 v/v ethyl
acetate/n-pentane) as a tan liquid. Elemental analysis calcd (%) for C14H22O2S2 (286.1): C 58.70,
H 7.74, S 22.39; found: C 58.52, H 7.62, S 22.41.
1H NMR (400 MHz, CDCl3, δ in ppm): 6.81-6.78 (m, (CO)C=CH, 1H), 3.72 (sep, 3J(H,H)
= 7.0 Hz, (CH3)2CH, 1H), 3.67 (sep, 3J(H,H) = 7.0 Hz, CH(CH3)2', 1H), 3.53-3.44 (m,
CHH'CHH'CH, 1H), 3.08 (dd, 2J(H,H) = 15.2 Hz, 3J(H,H) = 3.3 Hz, CHH'(CO), 1H), 2.48 (dd,
2J(H,H) = 15.2 Hz, 3J(H,H) = 10.2 Hz, CHH'(CO), 1H), 2.60-2.41 (m, CHH'CHH'CH, 2H), 2.25-
2.15, 1.84-1.76 (2 m, CHH'CHH'CH, 2H), 1.32 (d, 3J(H,H) = 7.0 Hz, (CH3)2CH, 6H), 1.29 (d,
76 2 Catalysis with Rhenium-Containing Phosphines 3J(H,H) = 7.0 Hz, CH(CH3)2', 6H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm): 198.3 (s,
CH2(CO)), 188.8 (s, (CO)C=CH), 146.7 (s, (CO)C=CH), 143.1 (s, (CO)C=CH), 47.8 (s, CH2(CO)),
42.0 (s, CH2CH2CH), 34.5, 34.3 (2 s, (CH3)2CH and CH(CH3)2'), 31.6 (s, CH2CH2CH), 29.1 (s,
CH2CH2CH), 23.15, 22.97, 22.94, 22.91 (4 s, (CH3)(CH3)'CH and (CH(CH3)(CH3)')').
IR (thin film, cm–1):71 1683 (s, νCO), 1652 (s, νCO), 1613 (s, νC=C).
2.5.5 Isolation of Side Product C=CHCH2CH2CHCH2C(CH3)(OH)CH2(CO) (19)
Procedure E was applied to substrate C1Me2 (0.027 g, 0.162 mmol) and catalyst 14b
(0.0125 g, 0.0162 mmol) in PhCl (3.2 mL). However, 1H NMR spectra were not periodically
recorded. After 72 h, an equivalent volume of hexanes was added with stirring. The mixture was
filtered through a plug of silica gel, and the plug was rinsed (1:1 v/v ethyl acetate/hexanes). The
solvent was removed from the filtrate by rotary evaporation. The crude product was purified by
column chromatography (4:1 v/v ethyl acetate/n-pentane). First, a mixture of C1Me2prod and 20
was obtained. The spectroscopic data for each agreed with that in the literature.26,60 Second, a
subsequent fraction gave pure 19, the structure of which was established by 1H NMR and 13C{1H}
NMR.
Data for 19: 1H NMR (400 MHz, CDCl3, δ in ppm): 6.64-6.59 (m, C=CHCH2, 1H), 3.36-
3.23 (m, CH2CH, 1H), 2.52 (d, 2J(H,H) = 17.2 Hz, CHH'CO, 1H), 2.48-2.38 (m, C=CHCH2, 2H),
2.34 (d, 2J(H,H) = 17.2 Hz, CHH'CO, 1H), 2.34-2.25 (m, CHH'CH, 1H), 2.17-2.08 (m, CHH'CH,
1H), 1.92 (br s, OH, 1H), 1.61-1.49 (m, CHCHH'COH, 1H), 1.42 (apparent t, 2J(H,H) = 3J(H,H) =
12.8 Hz, CHCHH'COH, 1H), 1.35 (s, CH3, 3H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm):
198.2 (s, CH2(CO)), 143.8 (s, C=CH), 138.4 (s, C=CH), 73.0 (s, CH2COH), 53.6 (s, CH2(CO)),
44.0 (s, CH2COH), 40.8 (s, CHCH2CH2), 33.0, 32.0 (2 s, CHCH2CH2), 31.0 (s, CH3).
3
Rhenium-Containing Phosphines
3.1 Introduction to Rhenium Complexes
There are ca. 18000 publications dealing with manganese complexes. On the other hand, for
the homologous element rhenium there are currently ca. 2000 publications on the subject of
rhenium complexes, as a simple SciFinder® search shows. This is in one sense astonishing, because
from the author's experience, rhenium forms stable organometallic compounds quite easily.
However, rhenium is a rare metal (although it is no rare earth metal), and as compared to others is
quite expensive. Complexes are often crystalline, which facilitates purification and handling.72,73
In the Gladysz group many rhenium complexes have been developed within the last 30
years. Most isolable species are inert against oxygen and moisture. The typical core fragment from
which most of these complexes are derived was shown in Figure 2.1 (Structure B).
In principle the cyclopentadienyl ligand can be modified by lithiation, the PPh3 ligand can
be substituted by phosphine derivatives, and the moiety X can be an alkyl group, a phosphine, an
amine, or something similar.
As reviewed in Chapter 1, phosphines and amines are very useful compounds in
organocatalytic reactions.1,5-10 During recent decades numerous rhenium-containing
phosphines31,33,36-38,41,51,74 and amines75-79 have been developed in the Gladysz group, many of
which were employed as ligands in transition metal catalysis.36-39,41,74 Some of these phosphines
were shown in free form (4-6), or already coordinated to transition metals (7-13) in Figures 2.3 and
2.4.
Several analogous phosphine oxides were also prepared.31,52,61,80,81 These were evaluated
as catalysts for allylation reactions. Unfortunately, the yields as well as the enantiomeric excesses
were very low.61,81 Hence, only the preparation of the complexes has been published.
However, some of the rhenium-containing phosphines were successful as catalysts in
78 3 Preparation of Rhenium-Containing Phosphines
intramolecular Morita Baylis Hillman and Rauhut Currier reactions. Accordingly, improved
second-generation catalysts were sought, and were developed as described in this chapter.
All of these were prepared by established procedures. These are very fundamental for this
project and are therefore sketched in the following introduction.
The synthesis of the basic building block starts with commercially available Re2(CO)10,
which is converted in three steps to the racemic carbonyl complex [(η5-C5H5)Re(NO)(PPh3)(CO)]+
BF4–. This is stable for an indefinite time period under ambient laboratory conditions.82 Indeed,
most complexes that are mentioned in this chapter are eighteen valence electron species that exhibit
excellent chemical and configurational stability.
The racemic salt [(η5-C5H5)Re(NO)(PPh3)(CO)]+ BF4– plays a central role. The rhenium
atom is a stereocenter and the racemate can be resolved into enantiomers via a four step procedure.
The (S) or (R) stereoisomer is subsequently treated in the same way as the racemic complex.
Reaction with NaBH4 leads to a reduction of the carbonyl group to give the neutral methyl complex
(η5-C5H5)Re(NO)(PPh3)(CH3) (26) in either enantiopure or racemic form, depending upon the
educt.82 Starting from this basic building block, numerous conversions are possible, and relevant
literature reactions are summarized in Scheme 3.1.
The abstraction of a hydride by Ph3C+ PF6– to give the methylidene species [(η5-
C5H5)Re(NO)(PPh3)(=CH2)]+ PF6– is an important process.83 Usually this cation is formed in situ,
but it can also be isolated.83 This species easily reacts with a variety of nucleophiles, including
PPh2H. Subsequent deprotonation of the resulting phosphonium salt leads to the free phosphine
(η5-C5H5)Re(NO)(PPh3)(CH2PPh2) (14a). Finally, lithiation using n-BuLi followed by a
nucleophilic substitution on PPh2Cl gives the diphosphine (η5-C5H4PPh2)Re(NO)(PPh3)(CH2PPh2)
(16b).38
Another conversion can be carried out involving functionalization of the methylene spacer
to provide one more stereocenter at the carbon atom (15a, 16a). First, the methylidene species is
formed by the addition of Ph3C+ PF6–. Subsequent reaction with PhLi affords the benzyl complex
(η5-C5H4PPh2)Re(NO)(PPh3)(CH2Ph).
3 Preparation of Rhenium-Containing Phosphines 79
Scheme 3.1. Syntheses of rhenium-containing phosphines with chiral centers. Enantioselective (and
diastereoselective) syntheses are possible in each case. a) Ph3C+ PF6–, CH2Cl2. b) PPh2H. c) t-
BuOK, THF. d) n-BuLi, THF. e) PPh2Cl. f) PhLi. g) t-BuLi. h) HBF4.OEt2, PhCl.
ON PPh3Re
CH2
ON PPh3Re
PPh2
ON PPh3Re
CH2PPh2
ON PPh3Re
CHPhPPh2
ON PPh3Re
CH2PPh2
ON PPh3Re
PPh2
Li Li
ON PPh3Re
CHPhPPh2
PPh2
b) c)
d)
f)a)b)c)
g)
b)
c)
d)
1814a
15a
16a
ON PPh3Re
CH3
(S)-26 or
racemate
h)
PF6
BF4
ON PPh3Re
ON PPh3Re
CH2PPh2
PPh2
16b
e) e)
ON PPh3Re
PPh2
PPh2
27
(PhCl)
a)
ON PPh3Re
CHPhPPh2
Li
e)
80 3 Preparation of Rhenium-Containing Phosphines
A hydride is similarly removed from the benzyl complex, leading to the benzylidene
complex [(η5-C5H4PPh2)Re(NO)(PPh3)(=CHPh)]+ PF6–.84 As detailed in earlier papers, the
benzylidene complex can exist as two geometric isomers, and either can be obtained depending on
the temperature at which the reaction is carried out. Subsequently, the benzylidene complex is
stereospecifically attacked by the nucleophile PPh2H and after deprotonation only one diastereomer
of the phenyl-substituted phosphine species (η5-C5H5)Re(NO)(PPh3)(CHPhPPh2) (15a) is obtained.
15a can be subsequently converted to the diphosphine (η5-C5H4PPh2)Re(NO)(PPh3)(CHPhPPh2)
(16a) by lithiation and reaction with PPh2Cl.51
A third important sequence is initiated by treating 26 with a strong Brønsted acid such as
HBF4.OEt2 in PhCl. After the evolution of CH4, the cationic rhenium species [(η5-
C5H5)Re(NO)(PPh3)(PhCl)]+ BF4– is formed in situ, in which a very labile solvent molecule (PhCl)
is σ/π coordinated. This is treated with the nucleophile PPh2H and then base to give the free
phosphine (η5-C5H5)Re(NO)(PPh3)(PPh2) (18). This is converted, similarly to 14a and 15a, to the
diphosphine (η5-C5H4PPh2)Re(NO)(PPh3)(PPh2) (27).38
Using reactions of the type in Scheme 3.1, many rhenium-containing phosphines have been
developed and some of them were already mentioned in Chapter 2. All of the above steps can be
carried out on multigram scales with very good yields. Additionally, there is full retention of
configuration at the rhenium center.85,86 Hence, any useful enantioselective catalysts can likely be
synthesized on very large scales.
3 Preparation of Rhenium-Containing Phosphines 81
3.2 Results
3.2.1 Preparation of Rhenium Complexes
3.2.1.1 Preparation of Racemic Rhenium-Containing Phosphines
It was sought to extend the previously reported syntheses of the rhenium-containing
phosphines 14a, 15a, and 16a,b to related complexes, and evaluate their catalytic activity. The
preparation of protonated analogs of 14a is depicted in Scheme 3.2.
Scheme 3.2. Preparation of the protonated racemic rhenium-containing phosphines [14-H]+ PF6–. a)
Ph3C+ PF6– (1.1 equiv.), CH2Cl2. b) PR2H.
Starting with the well known racemic methyl complex 26,82 a slight excess of the strong
hydride abstracting reagent Ph3C+ PF6– was added at –78 °C. The methylidene complex [(η5-
C5H5)Re(NO)(PPh3)(=CH2)]+ PF6– formed in situ and was subsequently treated with a
symmetrically substituted secondary phosphine PR2H (R = p-Tol (28b), p-C6H4OCH3 (28c),87 p-
C6H4N(CH3)2 (28d),87 2-biphen (28e), α-naph (28f);88 for the preparation of the secondary
phosphines see section 3.2.2). After the mixture was allowed to warm to room temperature, the
ON PPh3Re
CH3
ON PPh3Re
CH2PR2H
PF6
[14b-H]+ PF6−: R = p-Tol, 94%
[14c-H]+ PF6−: R = p-C6H4OCH3, 88%
[14d-H]+ PF6−: R = p-C6H4N(CH3)2, 93%
[14e-H]+ PF6−: R = 2-biphen, 95%
[14f-H]+ PF6−: R = α-naph, 87%
a)
b)
[14-H]+ PF6−26
82 3 Preparation of Rhenium-Containing Phosphines
protonated rhenium-containing phosphines [(η5-C5H5)Re(NO)(PPh3)(CH2PR2H)]+ PF6– ([14-H]+
PF6–) precipitated with the addition of n-pentane. The products were obtained in very good yields
as orange to red powders (R = p-Tol ([14b-H]+ PF6–), 94%; p-C6H4OCH3 ([14c-H]+ PF6
–), 88%;
p-C6H4N(CH3)2 ([14d-H]+ PF6–), 93%; 2-biphen ([14e-H]+ PF6
–), 95%; α-naph ([14f-H]+ PF6–),
87%). If impurities were found, recrystallization from CH2Cl2/n-pentane was possible at any time.
All complexes were characterized by elemental analysis, NMR (1H, 13C{1H}, 31P{1H}) and IR
spectroscopy, and mass spectrometry. Melting points were also recorded, but most were in fact
decomposition points, as indicated by a darkening of color before or while melting.
The two phosphorus substituents R in [14-H]+ PF6– are diastereotopic, and therefore two
sets of NMR signals were obtained. The substituents R rotate around the phosphorus-carbon bonds
quite fast on the NMR time scale at room temperature. Therefore, for [14(b-d)-H]+ PF6– a set of
four signals for each phenyl ring was observed in the 13C{1H} NMR spectra. Most of them were
31P coupled doublets.
For complexes [14e,f-H]+ PF6–, in which the R groups are less symmetric, the 13C{1H}
NMR spectra were considerably more complicated. It was not possible to reliably differentiate
between singlets and 31P coupled doublets. However, there is no evidence from the pattern of the
signals of hindered rotation or other dynamic behavior.
Next, [14-H]+ PF6– were deprotonated as depicted in Scheme 3.3. This transformation is a
simple acid-base reaction that was carried out in the nonpolar solvent C6H6. As [14-H]+ PF6– are
highly polar salts, they are almost insoluble in C6H6. The t-BuOK can be supposed to be insoluble
also. Thus, after the addition of the reactants a heterogeneous mixture was obtained and with
vigorous stirring the deprotonation took place. Most of the rhenium-containing free phosphines are
highly soluble in C6H6 and therefore red suspensions containing the white salt K+ PF6– were
obtained. This salt was removed by filtration through a small plug of Celite®. As the filtrate
contained some t-BuOH, which could inhibit catalysis as mentioned in section 2.2.2.1, n-pentane
was added to precipitate the products. The deprotonated phosphines (η5-
C5H5)Re(NO)(PPh3)(CH2PR2) (14) were obtained in good yields as orange to red solids (14b, 80%;
14c, 77%; 14d, 85%; 14e, 91%; 14f, 64%).
3 Preparation of Rhenium-Containing Phosphines 83
Scheme 3.3. Deprotonation of the racemic phosphonium salts [14-H]+ PF6–. a) t-BuOK (1.5 equiv.),
C6H6.
These phosphines were all characterized by elemental analysis, NMR (1H, 13C{1H},
31P{1H}) and IR spectroscopy, mass spectrometry, and melting points. Melting again occurred with
decomposition, as the trivalent phosphorus center is sensitive to oxygen and the analysis is
conducted under air.
Again, the 13C{1H} NMR spectra showed a signal pattern analogous to that of the
phosphonium salts [14-H]+ PF6–. More interesting were the 31P{1H} NMR spectra, which were
very simple as each compound showed only two doublets; one for the RePPh3 and one for the
CH2PR2 group. The chemical shifts are summarized in Table 3.1. The shifts for the PPh3 ligands
are more or less similar. No trend is obvious. The same situation is found for the phosphorus-
phosphorus coupling constants of the complexes 14a-f.
Analyzing the chemical shifts of the CH2PR2 groups, there is a downfield trend upon going
from 14a (8.1 ppm)38 to 14d (2.8 ppm). In this series the para substituents become progressively
more electron releasing and the resonances shift progressively downfield. This behavior agrees with
the observation of Mayr that phenyl-substituents can modulate the electron density at the
phosphorus center and hence, the nucleophilicity of triarylphosphines (see also section 2.3.2).67
However, such conclusions cannot be reliably drawn concerning complexes 14e,f, as the extended π
system could have a strong effect on the shielding of the phosphorus atoms.
ON PPh3Re
CH2PR2H
PF6
14b: R = p-Tol, 80%
14c: R = p-C6H4OCH3, 77%
14d: R = p-C6H4N(CH3)2, 85%
14e: R = 2-biphen, 91%
14f: R = α-naph, 64%
a)
ON PPh3Re
CH2PR2
[14-H]+ PF6− 14
84 3 Preparation of Rhenium-Containing Phosphines
Table 3.1. 31P{1H} NMR shifts and coupling constants for the rhenium-containing phosphines 14a-
f.
(η5-C5H5)Re(NO)(PPh3)(CH2PR2), R =
Ph
(14a)a p-Tol (14b)b
p-C6H4OCH3
(14c)b
p-C6H4N(CH3)2
(14d)b
2-biphen2
(14e)c
α-naph2
(14f)c
PPh3 25.8 27.2 27.6 27.2 23.9 25.7 δ (ppm) PR2 8.1 5.8 5.1 2.8 –5.7 –21.2
3J (Hz) (P,P) 8.0 6.7 6.9 6.7 6.9 6.9
a In CDCl3 (data from reference 38). b In C6D6. c In CD2Cl2.
The phosphines 14 were assumed to be sensitive to oxygen. Therefore, a sample of 14b was
kept under air in non-degassed C6D6. From time to time 1H and 31P{1H} NMR spectra were
recorded. These showed new signals, which were presumed to most likely be from oxidation
products, such as the phosphine oxide. In degassed solvents, these signals did not appear. The 1H
NMR spectra showed a new signal at 5.08 ppm (s), which was most likely from the
cyclopentadienyl ligand of the oxidized compound. The ratio of the cyclopentadienyl signal of the
oxidized phosphine compared to that of the unoxidized phosphine was utilized to assay the
operational availability of the catalyst during catalytic reactions. Other groups gave different 1H
NMR signals compared to the unoxidized phosphine, but these were not that diagnostic.
The 31P{1H} NMR spectra of the 14b exposed to oxygen showed two more doublets
appearing at 44.4 ppm (CH2(O=)P(p-Tol)) and 25.1 ppm (RePPh3). These chemical shift trends
parallel those reported for the 14a and O=14a analogs, which have been completely
characterized.61 Table 3.2 shows the ratio of 14b to the oxidation product as a function of time. It
takes 18 h to oxidize 16% of the phosphine. At the beginning, the oxidation is quite rapid, but then
slows down with time. Similar behavior was noted with the other rhenium-containing phosphines.
3 Preparation of Rhenium-Containing Phosphines 85
Table 3.2. Oxidation of 14b with time, assayed as described in the text.a
Time (h)
0.5 6 18
O=14b / 14b 4 / 96 7 / 93 16 / 84
a The amount of phosphine oxide was calculated by integration of the cyclopentadienyl
1H NMR signals.
3.2.1.2 Preparation of Enantiopure Complexes
As some of the complexes 14 showed good to excellent catalytic reactivity (Chapter 2), they
were synthesized in enantiopure form for enantioselective catalysis. These complexes are depicted
in Figure 3.1.
Figure 3.1. Enantiopure complexes (S)-14b-d, obtained from (S)-(η5-C5H5)Re(NO)(PPh3)(CH3)
((S)-26) by procedures analogous to the racemates (Schemes 3.2 and 3.3).
The procedures were carried out almost analogous to the racemic complexes. In general the
mechanism of product formation is the same in each case; however, solubilities can differ. Hence,
ON PPh3Re
CH2PR2H
PF6
(S)-[14b-H]+ PF6−: R = p-Tol, 86%
(S)-[14c-H]+ PF6−: R = p-C6H4OCH3, 95%
(S)-[14d-H]+ PF6−: R = p-C6H4N(CH3)2, 90%
ON PPh3Re
CH2PR2
(S)-14b: R = p-Tol, 88%
(S)-14c: R = p-C6H4OCH3, 87%
(S)-14d: R = p-C6H4N(CH3)2, 90%
(S)-[14-H]+ PF6− (S)-14
86 3 Preparation of Rhenium-Containing Phosphines
crystallization or precipitation from the product mixture was optimized by employing different
solvent mixtures, resulting in slightly changed workup procedures for the enantiopure complexes.
The complexes (S)-14b-d were characterized similarly to the racemic compounds. As expected, the
spectroscopic data were virtually identical. Since the salts (S)-[14(b-d)-H]+ PF6– were very air
stable they were also characterized by optical rotations. These showed that the complexes were at
least scalemic. In any case, the additions of carbon, nitrogen, phosphorus and sulfur nucleophiles to
the methylidene complex [(η5-C5H5)Re(NO)(PPh3)(=CH2)]+ PF6– have been shown to proceed
with essentially complete retention of configuration at the chiral center.75,86
3.2.1.3 Preparation of Diastereomeric Mixtures of P Chiral Complexes
As already mentioned in the Chapter 2, a new stereocenter was introduced at the phosphorus
atom of the rhenium-containing phosphines. This was done by applying the same procedures as
above, but with the racemic secondary phosphines PPh(α-naph)H (29c) and PPh(β-naph)H (29d)
(Scheme 3.4).
Scheme 3.4. Preparation of the protonated diastereomeric rhenium-containing phosphines
(SReRP)/(SReSP)-[17-H]+ PF6–. a) Ph3C+ PF6
–, CH2Cl2. b) PRR'H.
Again, the sequence started with enantiopure (S)-2682 reacting with Ph3C+ PF6–, followed
by the addition of 29c,d. Only two of four possible stereoisomers were expected to be formed, as
the rhenium configuration should be retained. Indeed, mixtures of (SReSP)-17c and (SReRP)-17c or
ON PPh3Re
CH3
ON PPh3Re
CH2PRR'H∗
PF6
(SReRP)/(SReSP)-[17c-H]+ PF6−: R/R' = Ph/α-naph, 78%
(SReRP)/(SReSP)-[17d-H]+ PF6−: R/R' = Ph/β-naph, 83%
a)
b)
(SReRP)/(SReSP)-[17-H]+ PF6−(S)-26
3 Preparation of Rhenium-Containing Phosphines 87
(SReSP)-17d and (SReRP)-17d were detected. Importantly, some reactions were carried out with a
large excess of the racemic phosphine 29c, in hopes that a matched diastereomer might
preferentially form. In these cases, the reaction solution was very slowly warmed from –78 °C to
room temperature overnight. However, 1H NMR spectra of an aliquot showed a diastereomeric
mixture (Figure 3.2). Since the sample did not completely dissolve in CD2Cl2, no integration values
are given.
Figure 3.2. Partial 1H NMR spectrum (CD2Cl2) of an aliquot of the reaction mixture that yields
(SReSP)/(SReRP)-[17c-H]+ PF6– (δ in ppm).
One of the cyclopentadienyl singlets in Figure 3.2 is arbitrarily assigned as diastereomer 1
and the other as diastereomer 2. The shift difference between the diastereomers is 0.34 ppm. This is
quite large and suggests significantly different steric and electronic environments.
Also the solubility and stability of the diastereomers differ. To precipitate a mixture
enriched in diastereomer 1, the volume of the reaction mixture was reduced to one third and n-
pentane was added slowly with vigorous stirring until a precipitate first appeared. Then the mixture
was cooled to –24 °C. After 24 h, the precipitate was isolated by filtration to give diastereomer 1 in
72% de, as assayed by the ratio of the integrals of the cyclopentadienyl 1H NMR signals (CD3CN,
homogenous conditions). After a second identical precipitation, diastereomer 1 was obtained in
92% de. This reaction was repeated several times and in some cases a third precipitation cycle was
necessary to obtain good de values.
5.5 5.06.0 4.5 4.0
CDHCl2
C5H5(Dia_1)
C5H5(Dia_2)
ppm
88 3 Preparation of Rhenium-Containing Phosphines
Noteworthy is the fact that diastereomer 1 never precipitated from the reaction mixture with
the solvent CH2Cl2 itself. Addition of n-pentane was always obligatory. But when highly
diastereoenriched powders of diastereomer 1 were treated with CH2Cl2 or CHCl3 they showed poor
solubilities. These mixtures were soluble in acetone but after several hours the solutions darkened,
indicative of decomposition. Good solubility was also found in CH3CN, so the NMR spectra were
recorded in this solvent. But also here decomposition was detected by 1H NMR as time went by.
Diastereomer 2 enriched mixtures showed a totally different behavior. These mixtures were
obtained from the supernatants from diastereomer 1 precipitations. To precipitate diastereomer 2, a
large excess of n-pentane was necessary. The best obtained de value for this mixture was –92%. In
Figure 3.3 the cyclopentadienyl 1H NMR signals of diastereoenriched mixtures are shown.
Figure 3.3. Partial 1H NMR spectra (CD3CN) of diastereomeric mixtures of (SReSP)/(SReRP)-[17c-
H]+ PF6– (δ in ppm), de = 92% and –92%.
Diastereomer 2 enriched mixtures showed very good solubilities in CH2Cl2 as well as in
CHCl3. Also in CH3CN and acetone very good solubilities were noticed, but decomposition
occurred within hours. The color darkens within tens of minutes. Because of this, the 13C{1H} and
31P{1H} NMR spectra of these mixtures were recorded in CD2Cl2. Hence, the chemical shifts are
not fully comparable with those for diastereomer 1 in the experimental section (where NMR data
was collected in CD3CN due to better solubility). NMR and other physical properties are
ppm
5.1 1.00 0.04 1.00 0.04
C5H5(Dia_1)
Diastereomer 1 enriched mixture Diastereomer 2 enriched mixture
C5H5(Dia_2)
5.0 4.9 4.8 5.1 5.0 4.9 4.8
C5H5(Dia_1)
C5H5(Dia_2)
3 Preparation of Rhenium-Containing Phosphines 89
summarized in Table 3.3.
Table 3.3. Comparison of selected physical properties of diastereomers 1 and 2 of (SReSP)/(SReRP)-
[17c-H]+ PF6–.
Diastereomer 1 Diastereomer 2
PPh3 21.2 22.3
PPh(α-naph)H 23.1 29.0 31P{1H} NMR
signala 3J(P,P) 14.1 12.6
C5H5 4.98 4.64 1H NMR signala CHH' m (2.82-2.54, 2H) 2 m (3.06-2.91, 1H; 2.82-2.54, 1H)
IRb νNO 1644 1652
CH3CN good (dec. within 24 h) very good (dec. within 30 min)
acetone good (dec. within 24 h) very good (dec. within several h)
CH2Cl2 low very good solubility
CHCl3 low (dec. within 24 h) very good (dec. within several h)
a δ in ppm, CD2Cl2. b cm–1, thin film.
Notably, the strong IR νNO bands differ by 8 cm–1. However, this is not useful as a de assay,
as only one broader band is observed if a diastereomeric mixture with low de is sampled. But this
nicely shows that the electronic structures of the diastereomeric complexes are modified by the
phosphorus stereocenter. The NO ligand is a good π-acceptor as there is a vacant π* orbital that can
take up electron density. The more electron density is found in the π* orbital, the lower the bond
order. The lower the bond order, the lower the energy needed to excite the vibration. This means
that the complex with the higher electron density has the lower IR wave number. In this case it is
diastereomer 1 (1644 cm–1). This is supported by the PPh3 31P{1H} NMR signal. Diastereomer 1 is
upfield of diastereomer 2, suggesting a higher electron density at the phosphorus atom in
diastereomer 1 (21.2 ppm vs. 22.3 ppm).
90 3 Preparation of Rhenium-Containing Phosphines
As each diastereomer of (SReSP)/(SReRP)-[17c-H]+ PF6– could be obtained with a de > 90%,
each was characterized by elemental analysis, NMR (1H, 13C{1H}, 31P{1H}) and IR spectroscopy,
and mass spectrometry. The complete assignment of 13C{1H} NMR signals was not possible, due
to many overlapping aryl resonances as well as some small signals from the opposite diastereomer.
Since the samples were not diastereomerically pure, optical rotations and melting points were not
measured.
Unfortunately, the diastereomers of (SReSP)/(SReRP)-[17d-H]+ PF6– could not be as
efficiently separated and only low de values were obtained (48% and –38%). Hence, the
diastereomer separation at this stage is not detailed in the experimental section. The solubility
properties of the diastereomers were too similar to get good separation. Furthermore, decomposition
was noticed in many solvents. Even treating the crude reaction mixture with n-pentane to precipitate
the product gave a black mixture and an impure oily precipitate. Only precipitation with diethyl
ether gave the desired product as a powder, although here the supernatant also showed a very dark
coloring, indicating decomposition. However, no diastereomeric enrichment took place. When this
solid was treated with ethyl acetate, one diastereomer preferentially dissolved. Nonetheless, all
attempts to separate the diastereomers of (SReSP)/(SReRP)-[17d-H]+ PF6– were unsuccessful. Thus,
the characterization was carried out with the diastereomeric mixture. Elemental analysis, NMR (1H,
13C{1H}, 31P{1H}) and IR spectroscopy, and mass spectrometry data are summarized in the
experimental section. Since the samples were not diastereomerically pure, optical rotations and
melting points were not measured.
The deprotonation of (SReSP)/(SReRP)-[17c-H]+ PF6– was carried out analogous to that in
Scheme 3.3. The diastereoenriched mixtures were treated with t-BuOK in C6H6. After filtration
through a plug of Celite®, the solvent volume was reduced and the solution was layered with n-
pentane. The phosphine (SReSP)/(SReRP)-17c precipitated and was characterized by NMR (1H and
31P{1H}). The de remained unchanged, as assayed by integration of the cyclopentadienyl 1H NMR
signals, consistent with retention of configuration at rhenium and phosphorus atoms. The resulting
solids were then employed as catalysts (section 2.4.1.2).
A 50:50 diastereomeric mixture of (SReSP)/(SReRP)-[17d-H]+ PF6– was similarly
deprotonated with t-BuOK in C6H6. After reduction of the C6H6 volume, a layer of n-pentane was
3 Preparation of Rhenium-Containing Phosphines 91
added. A dark red precipitate was collected by filtration. The 1H NMR spectrum proved that
(SReSP)/(SReRP)-17d was obtained, and integration of the cyclopentadienyl signals indicated a 48%
de. The enriched diastereomer was arbitrarily assigned to be diastereomer 1.
The filtrate was treated with more n-pentane and the opposite diastereomer 2 of
(SReSP)/(SReRP)-17d precipitated with –92% de. Both precipitates were characterized by 1H NMR
and 31P{1H} NMR spectroscopy and employed for catalysis.
Complexes (SReSP)/(SReRP)-[17a-H]+ PF6– and (SReSP)/(SReRP)-[17b-H]+ PF6
–, which
were obtained from Dante Castillo,52 were deprotonated in the same manner as above, and the
resulting orange solids were employed for organocatalytic reactions without further characterization
or purification.
3.2.1.4 Preparation of (SReSC)-[(η5-C5H5)Re(NO)(PPh3)(CHCH3PPh2H)]+ PF6– ((SReSC)-
[15b-H]+ PF6–)
The preparation of (SReSC)-[15b-H]+ PF6–, which features both rhenium and carbon
stereocenters, was carried out analogous to the previously reported phenyl analog (SReSC)-[15a-H]+
PF6– (Scheme 3.1).51 The sequence started with the rhenium ethyl complex (S)-(η5-C5H5)Re(NO)-
(PPh3)(CH2CH3),89 which was treated as shown in Scheme 3.5.
When the rhenium ethyl complex and Ph3C+ PF6– were combined in CH2Cl2 at –78 °C, an
α hydride was regiospecifically abstracted to give the ethylidene complex (sc)-[(η5-
C5H5)Re(NO)(PPh3)(=CH2CH3)]+ PF6–. This species was not isolated, but has been extensively
studied earlier.89 It is very important to keep the temperature at –78 °C. If the mixture is warmed,
the opposite (ac) Re=C isomer of the ethylidene complex is formed and the opposite
stereospecificity is obtained. Therefore, PPh2H was added at low temperature. This reacted with the
ethylidene complex upon warming to give the phosphonium salt (SReSC)-[15b-H]+ PF6–. The crude
product was precipitated from the reaction mixture by the addition of n-pentane, and isolated by
decantation as an oil that solidified with drying by oil pump vacuum.
The solid was precipitated a second time from a CH2Cl2/t-BuOH solution by slowly
concentrating the sample under vacuum with vigorous stirring. The product (SReSC)-[15b-H]+ PF6–
92 3 Preparation of Rhenium-Containing Phosphines
precipitated, and was characterized by elemental analysis, NMR (1H, 13C{1H}, 31P{1H}) and IR
spectroscopy, mass spectrometry, and melting point. No evidence for a second diastereomer was
observed. The carbon configuration was assigned by analogy to other addition products of
ethylidene and benzylidene complexes that have been crystallographically characterized.40
Scheme 3.5. Preparation of the diastereopure rhenium-containing phosphine (SReSC)-[15b-H]+
PF6–. a) Ph3C+ PF6
–, CH2Cl2, –78 °C. b) PPh2H.
The (SReSC)-[15b-H]+ PF6– was subsequently treated with t-BuOK (1.5 equiv.) in C6H6.
After the deprotonation was finished, the mixture was filtered through a plug of Celite®. Then the
solvent volume was reduced and n-pentane was added. The deprotonated complex 15b precipitated,
and was employed without purification or characterization for catalysis (section 2.2.1.5).
3.2.2 Preparation of Secondary Phosphines
For the syntheses of the above rhenium-containing phosphines, secondary phosphines of the
formula PRR'H were needed. In the case of R = R' = Ph, or p-Tol, both compounds were
commercially available. All others had to be synthesized. Two different synthetic routes were
evaluated. One involved the reduction of a phosphine oxide and the other was the reduction of a
ON PPh3Re
CH2CH3
ON PPh3Re
C
PF6
a) b)
(SReSC)-[15b-H]+ PF6−, 55%
H
CH3HPh2P(S)
ON PPh3H
CH3
ON PPh3Re
C
H CH3
PF6
3 Preparation of Rhenium-Containing Phosphines 93
diarylchlorophosphine.
3.2.2.1 Preparation via Reduction of Secondary Phosphine Oxides
The preparation of secondary phosphines oxides, as well as their reduction to secondary
phosphines, is a well known procedure reported by Busacca.87 The usual sequence is shown in
Scheme 3.6.
Scheme 3.6. Syntheses of the secondary phosphines 28c,d (yields overall). a) Mg, THF. b)
OP(OEt)2H. c) DIBAL-H. d) NaOH/H2O.
The literature procedure was followed exactly and the therein reported secondary phosphine
oxides were purified by crystallization. The second step, the reduction with DIBAL-H was nearly
quantitative. The phosphines 28c,d were obtained with satisfactory purity and no further
manipulations after hydrolysis were necessary.
3.2.2.2 Preparation via Reduction of Diarylchlorophosphines
The above method for the preparation of phosphines is a very useful one. However, when
extended to the naphthyl and biphenyl analogs, the secondary phosphine oxides proved to be poorly
soluble in nonprotic solvents. Protic solvents on the other hand did not precipitate products for
purification. Hence, another procedure was applied.
OPR2H PR2H
28c: R = p-C6H4OCH3, 73%
28d: R = p-C6H4N(CH3)2, 67%
a)
b) 28
RBrc)
d)
94 3 Preparation of Rhenium-Containing Phosphines
Preparation of symmetric secondary phosphines. Two phosphines PR2H (R = 2-biphen
(28e), α-naph (28f)) were synthesized by the reaction of Grignard reagents with phosphorus
trichloride, followed by the reduction with LiAlH4. They were obtained via a three step synthesis
without isolation of the intermediates (Scheme 3.7).
Scheme 3.7. Synthesis of the secondary phosphines 28e,f. a) Mg, THF. b) PCl3. c) LiAlH4. d)
NaOH/H2O.
Grignard reagents were prepared from the commercially available aryl bromides RBr (R =
2-biphen, α-naph). Then the reagents (2.2 equiv.) were added dropwise to dilute solutions of
phosphorus trichloride at 0 °C. After warming to room temperature, 31P{1H} NMR spectra of
aliquots indicated full conversion of phosphorus trichloride as the signal (231 ppm) disappeared.
The 31P{1H} NMR spectra exhibited several peaks. In addition to some small unassigned signals,
the crude products exhibited two signals that differed by 5-10 ppm. These were provisionally
assigned as the target PR2Cl species and the corresponding PR2Br compound, with the bromide
derived from the Grignard reagents. For P(α-naph)2X, signals were observed at 77.7 ppm and 68.9
ppm (ca. 1:1, X = Cl and Br, respectively). For P(2-biphen)2X a dominant signal at 75.4 ppm and
another small signal at 69.8 ppm were found (ca. 10:1, X = Cl and Br, respectively). No signals
characteristic of tertiary phosphines were noted in the samples of 28e. In contrast, samples of 28f
exhibited several small signals at ca. –45 ppm, which could indicate the presence of the tertiary
phosphine P(α-naph)3.
These mixtures were treated with excesses of LiAlH4, and after a few minutes 31P{1H}
NMR spectra indicated total consumption of the PR2Cl species. In both cases one major signal for
the secondary phosphine was obtained (28e: –53.1 ppm, 28f: –61.9 ppm). The 31P{1H} NMR
spectrum of crude 28e showed one more signal at –124.7 ppm, suggesting the presence of some
primary phosphine P(2-biphen)H2. Crude 28f also showed one more but very small signal at –135.1
RBr
a)b)c)d)
PR2H28e: R = 2-biphen, 70%
28f: R = α-naph, 35%28
3 Preparation of Rhenium-Containing Phosphines 95
ppm, which was also supposed to be the corresponding primary phosphine.
The secondary phosphine 28f could be purified by Kugelrohr distillation with subsequent
crystallization, while for 28e crystallization alone was sufficient. This gave the pure phosphines in
moderate and good yields (28e: 70%, 28f: 35%). Since 28e was a new compound, it was
characterized by NMR spectroscopy (1H, 13C{1H}, 31P{1H}). The secondary phosphine 28f had
been mentioned in the literature.88 The reported procedure was different but not reproducible in the
author's hands. However, 1H NMR and 31P{1H} NMR data for 28f matched that previously
reported.
Preparation of racemic secondary phosphines. The racemic secondary phosphines PRR'H
(R/R' = Ph/α-naph (29c); Ph/β-naph (29d)) required for the rhenium-containing phosphines
(SReSP)/(SReRP)-17c and (SReSP)/(SReRP)-17d were obtained by a procedure similar to the
symmetric ones. This route is depicted in Scheme 3.8.
Scheme 3.8. Synthesis of the racemic secondary phosphines 29c,d. a) Mg, THF. b) PPhCl2. c)
LiAlH4. d) NaOH/H2O.
The Grignard reagents (1.1 equiv.) were added to commercially available PPhCl2. Also here
mixtures of the PRR'X species (X = Cl, Br) were obtained. These were treated with LiAlH4, and
subsequently with aqueous NaOH. In this case both pure phosphines were obtained by Kugelrohr
distillation in moderate yields (29c: 38%, 29d: 32%). Both phosphines were characterized by NMR
spectroscopy (1H, 13C{1H}, 31P{1H}).
3.3 Discussion
This chapter has established that many phosphines [Re]CH2PRR' with the rhenium fragment
RBr
a)b)c)d)
PRR'H29c: R/R' = Ph/α-naph, 38%
29d: R/R' = Ph/β-naph, 32%29
96 3 Preparation of Rhenium-Containing Phosphines
[Re] = (η5-C5H5)Re(NO)(PPh3) can be easily synthesized in good to excellent yields. Furthermore
many of these are readily obtained in enantiopure form.
However, with increasing numbers of reaction steps, it is more important to obtain good
yields for each step. For example, the overall yield for the known complex 14a is 44% for the
racemate (6 steps); if the enantiopure form is needed the yield decreases to 36% (9 steps). The new
catalysts 14b-f and (SReSP)/(SReRP)-17c,d have been synthesized in comparable yields. However,
excluding 14b which is prepared using commercially available P(p-Tol)2H, those catalysts require
secondary phosphines that must themselves be synthesized. As these preparations are also multistep
procedures giving low to good yields (32-70%), the economics of the catalyst preparation becomes
less attractive. However, the routes employed are very flexible with respect to the phosphorus
substituents, and as noted in the previous chapter these exert a tremendous influence upon the
enantioselectivities. In the event that a commercial catalyst of this type were developed, it is
probable that via careful optimization an economical, high-yield synthetic sequence could be
developed.
3 Preparation of Rhenium-Containing Phosphines 97
3.4 Experimental
3.4.1 General Data
All experiments were conducted under N2, except in the cases where simple distilled
solvents were employed. NMR spectra were recorded at ambient probe temperature on Bruker 300
and 400 MHz FT spectrometers and referenced to the residual solvent signal (1H: CHCl3, 7.26 ppm;
CHDCl2, 5.32 ppm; C6D5H, 7.27 ppm; 13C{1H}: CDCl3, 77.0 ppm; CD2Cl2, 53.8 ppm; C6D6,
128.0 ppm) or H3PO4 (31P{1H}, external capillary, 85%, 0.0 ppm). IR spectra were recorded on an
ASI React IR®-1000 spectrometer. Optical rotations were measured using a Perkin-Elmer model
341 polarimeter. Mass spectra were obtained using a Micromass Zabspec instrument. Elemental
analyses were determined with a Carlo Erba EA1110 CHN instrument. Melting points were
measured on an Electrothermal IA 9100 apparatus.
Solvents were treated as follows: THF, n-pentane (2 × Grüssing), distilled from sodium
/benzophenone; CH2Cl2 (Staub) and C6H6 (Grüssing), distilled from CaH2; t-BuOH (Grüssing),
distilled from Mg; diethyl ether (Hedinger), hexanes, CH3OH, and EtOH (3 × Staub), simple
distilled; CDCl3 (99.8%), CD2Cl2 (99.6%), CD3CN (99.0%), and C6D6 (99.5%, 4 × Deutero
GmbH), freeze-pump-thaw degassed (3 ×) in the case of air sensitive products.
Chemicals were treated as follows: Ph3C+ PF6– (≥ 95%, Fluka),90 stored under argon at –30
°C; PCl3 (Merck) and PPhCl2 (97%, Acros), distilled and stored under N2; LiAlH4 (95%), t-BuOK
(98+%), 2-biphenylbromide (3 × Acros), α-naphthylbromide (96%, Fluka), β-naphthylbromide
(97% Lancaster), and P(p-Tol)2H (28b, 99%, Strem), used as received.
Silica gel 60M (Macherey-Nagel) and Celite® 535 (Fluka) for filtration, used as received.
98 3 Preparation of Rhenium-Containing Phosphines
3.4.2 Preparation of Rhenium-Containing Phosphines
[(η5-C5H5)Re(NO)(PPh3)(CH2P(p-Tol)2H)]+ PF6– ([14b-H]+ PF6
–). A Schlenk flask was
charged with racemic (η5-C5H5)Re(NO)(PPh3)(CH3) (26, 0.244 g, 0.437 mmol)82 and CH2Cl2 (15
mL). The solution was cooled to –78 °C and Ph3C+ PF6– (0.187 g, 0.481 mmol) was added with
stirring. After 1 h, P(p-Tol)2H (28b, 0.112 g, 0.787 mmol) dissolved in CH2Cl2 (1 mL) was added.
After 20 min, the cold bath was removed. After 1 h, the mixture was concentrated by oil pump
vacuum (to ca. 2.5 mL) and added dropwise to stirred hexanes (25 mL). An orange powder
precipitated, which was collected by filtration and washed with n-pentane (3 × 5 mL). A
31P{1H}NMR spectrum still showed some 28b. The powder was dissolved in CH2Cl2 (3 mL) and
C6H6 (15 mL). Hexanes (15 mL) were added with stirring and again an orange powder precipitated,
which was collected by filtration, washed with hexanes (2 × 5 mL) and n-pentane (5 × 5 mL), and
dried by oil pump vacuum to give pure [14b-H]+ PF6– (0.376 g, 0.411 mmol, 94%). M.p. 179-181
°C, dec. (capillary). Elemental analysis calcd (%) for C38H37F6NOP3Re (917.2): C 49.78, H 4.07,
N 1.53; found: C 49.45, H 3.98, N 1.50.
1H NMR (400 MHz, CD2Cl2, δ in ppm): 7.73-7.65, 7.51-7.43, 7.37-7.30 (3 m, C6H5 and
C6H4, 23H), 6.91 (dd, 1J(H,P) = 484 Hz, 3J(H,H) = 13.3 Hz, PH, 1H), 4.87 (s, C5H5, 5H), 2.65-
2.54 (m, CHH', 1H), 2.50, 2.40 (2 s, CH3 and CH3', 2 × 3H), 2.32-2.18 (m, CHH', 1H); 13C{1H}
NMR (101 MHz, CD2Cl2, δ in ppm): 91.0 (s, C5H5), –34.9 (dd, 1J(C,P) = 29.8 Hz, 2J(C,P) = 4.1
Hz, CHH'); PPh3 at91 134.4 (d, 1J(C,P) = 54.0 Hz, i), 134.0 (d, 2J(C,P) = 10.6 Hz, o), 131.6 (d,
4J(C,P) = 2.1 Hz, p), 129.5 (d, 3J(C,P) = 10.4 Hz, m); P(p-Tol)(p-Tol)' at 146.1 (d, 4J(C,P) = 2.8 Hz,
p to P), 145.5 (d, 4J(C,P) = 2.8 Hz, p' to P), 132.6 (d, 2J(C,P) = 10.6 Hz, o to P), 131.9 (d, 2J(C,P) =
10.5 Hz, o' to P), 131.2 (d, 3J(C,P) = 12.9 Hz, m to P), 130.9 (d, 3J(C,P) = 12.2 Hz, m' to P), 121.5
(d, 1J(C,P) = 71.8 Hz, i to P), 119.2 (d, 1J(C,P) = 88.2 Hz, i' to P), 22.1, 22.0 (2 s, CH3 and CH3');
31P{1H} NMR (162 MHz, CD2Cl2, δ in ppm): 28.8 (d, 3J(P,P) = 10.9 Hz, PH), 21.7 (d, 3J(P,P) =
10.9 Hz, PPh3), –144.0 (sept, 1J(P,F) = 708 Hz, PF6).
IR (thin film, cm–1):71 1668 (s, νNO). MS:92 772 (90) [14b-H]+, 558 (100) [14b–P(p-
Tol)2H]+.
3 Preparation of Rhenium-Containing Phosphines 99
(S)-[(η5-C5H5)Re(NO)(PPh3)(CH2P(p-Tol)2H)]+ PF6– ((S)-[14b-H]+ PF6
–). A Schlenk
flask was charged with (S)-26 (0.250 g, 0.448 mmol)82 and CH2Cl2 (15 mL). The solution was
cooled to –78 °C and Ph3C+ PF6– (0.191 g, 0.493 mmol) was added with stirring. After 1 h, 28b
(0.115 g, 0.787 mmol) dissolved in CH2Cl2 (1 mL) was added. After 20 min, the cold bath was
removed. After 1 h, the mixture was concentrated by oil pump vacuum (to ca. 2.5 mL) and EtOH (7
mL) was added with stirring. A yellow powder precipitated, and the solvent volume was reduced by
oil pump vacuum (to ca. 6 mL). The mixture was kept at –20 °C. After 2 h, the yellow powder was
collected by filtration, washed with EtOH (1 mL) and hexanes (2 × 5 mL), and dried by oil pump
vacuum to give pure (S)-[14b-H]+ PF6– (0.353 g, 0.385 mmol, 86%). M.p. 179-180 °C, dec.
(capillary). Elemental analysis calcd (%) for C38H37F6NOP3Re (917.2): C 49.78, H 4.07, N 1.53;
found: C 49.50, H 4.12, N 1.54. [α]26589 = 257° ± 1° (c = 2.00 mg/mL, CH2Cl2). Spectroscopic
data were similar to those of the racemate.
(η5-C5H5)Re(NO)(PPh3)(CH2P(p-Tol)2) (14b). A Schlenk flask was charged with [14b-
H]+ PF6– (0.161 g, 0.191 mmol) and C6H6 (20 mL). The suspension was vigorously stirred and t-
BuOK (0.0321 g, 0.287 mmol) was added. After 1 h, the orange suspension was filtered through a
plug of Celite®. The plug was rinsed with C6H6 until the filtrate became colorless. The filtrate was
concentrated by oil pump vacuum (to ca. 2 mL), layered with n-pentane (15 mL), and kept at 4 °C.
After 48 h, the orange crystals were collected by filtration and dried by oil pump vacuum to give
14b (0.118 g, 0.153 mmol, 80%). Dec. pt. 160-162 °C (capillary). Elemental analysis calcd (%) for
C38H36NOP2Re (770.9): C 59.21 H 4.71, N 1.82; found: C 58.98, H 4.98, N 1.81.
1H NMR (300 MHz, C6D6, δ in ppm): 7.74 (apparent t, 3J(H,H) = 3J(H,P) = 7.0 Hz, C6H4,
o to P, 2H), 7.66 (apparent t, 3J(H,H) = 3J(H,P) = 7.0 Hz, C6H4', o to P, 2H), 7.53 (dd, 3J(H,P) =
11.6 Hz, 3J(H,H) = 9.6, o-C6H5, 6H), 7.12-6.93 (m, m-, p-C6H5 and C6H4, m to P, 13H), 4.55 (s,
C5H5, 5H), 2.83 (dd, 2J(H,H) = 11.6 Hz, J(H,P) = 9.7 Hz, CHH', 1H), 2.12, 2.08 (2 s, CH3 and
CH3', 2 × 3H), 2.15-2.05 (m, CHH', 1H); 13C{1H} NMR (76 MHz, C6D6, δ in ppm): 89.9 (s, C5H5),
–18.1 (dd, 1J(C,P) = 37.2 Hz, 2J(C,P) = 4.7 Hz CHH'); PPh3 at 136.6 (d, 1J(C,P) = 50.8 Hz, i),
100 3 Preparation of Rhenium-Containing Phosphines
134.0 (d, 2J(C,P) = 10.2 Hz, o), 130.1 (s, p), 128.5 (d, 3J(C,P) = 10.2 Hz, m); P(p-Tol)(p-Tol)' at
144.7 (d, 1J(C,P) = 20.6 Hz, i to P), 143.4 (d, 1J(C,P) = 19.8 Hz, i' to P), 137.0, 136.6 (2 s, p and p'
to P), 133.7 (d, 2J(C,P) = 18.6 Hz, o to P), 133.1 (d, 2J(C,P) = 17.6 Hz, o' to P), 128.9 (d, 3J(C,P) =
4.9 Hz, m to P), 128.9 (d, 3J(C,P) = 5.8 Hz, m' to P), 21.3, 21.2 (2 s, CH3 and CH3'); 31P{1H} NMR
(121 MHz, C6D6, δ in ppm): 26.8 (d, 3J(P,P) = 6.7 Hz, PPh3), 5.8 (d, 3J(P,P) = 6.7 Hz, P(p-Tol)2)).
IR (thin film, cm–1):71 1644 (s, νNO). MS:92 771 (11) [14b]+, 558 (100) [14b–P(p-Tol)2]+.
(S)-(η5-C5H5)Re(NO)(PPh3)(CH2P(p-Tol)2) ((S)-14b). (S)-[14b-H]+ PF6– (0.187 g, 0.204
mmol), t-BuOK (0.0343 g, 0.306 mmol), and C6H6 (20 mL) were combined in a procedure
analogous to that given for the racemate. An identical workup gave (S)-14b as an orange powder
(0.138 g, 0.180 mmol, 88%). Dec. pt. 125-130 °C (capillary). Spectroscopic data were similar to
those of the racemate.
[(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4OCH3)2H)]+ PF6– ([14c-H]+ PF6
–). A Schlenk
flask was charged with racemic 26 (0.246 g, 0.441 mmol)82 and CH2Cl2 (15 mL). The solution was
cooled to –78 °C and Ph3C+ PF6– (0.188 g, 0.485 mmol) was added with stirring. After 1 h, P(p-
C6H4OCH3)2H (28c, 0.141 g, 0.573 mmol)87 dissolved in CH2Cl2 (1 mL) was added dropwise.
After 20 min, the cold bath was removed. After 1 h, the sample was concentrated by oil pump
vacuum (to ca. 4 mL). A CH3OH/EtOH mixture (2.5 mL, 2:3 v/v) was added, followed by CH2Cl2
until the sample became homogeneous. Hexanes (ca. 20 mL) were added with stirring and an
orange powder precipitated, which was collected by filtration, washed with hexanes (3 × 3 mL), and
dried by oil pump vacuum to give pure [14c-H]+ PF6– (0.369 g, 0.389 mmol, 88%). M.p. 207-208
°C, dec. (capillary). Elemental analysis calcd (%) for C38H37F6NO3P3Re (949.1): C 48.10, H 3.93,
N 1.48; found: C 47.76, H 3.84, N 1.45.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.78 (dd, 3J(H,P) = 13.0, 3J(H,H) = 8.5 Hz, C6H4, o
to P, 2H), 7.31-7.49 (m, C6H5 and C6H4', o to P, 17H), 7.12 (d, 3J(H,H) = 8.8 Hz, C6H4, m to P,
2H), 6.96 (d, 3J(H,H) = 8.5 Hz, C6H4', m to P, 2H), 6.90 (dd, 1J(H,P) = 478 Hz, 3J(H,H) = 12.0 Hz,
PH, 1H), 4.90 (s, C5H5, 5H), 3.89, 3.83 (2 s, OCH3 and OCH3', 2 × 3H), 2.68 (apparent dt, 2J(H,H)
= 19.8 Hz, J(H,P) = 3J(H,H) = 14.5 Hz, CHH', 1H), 2.50 (dd, 2J(H,H) = 19.8 Hz, J(H,P) = 12.2 Hz
3 Preparation of Rhenium-Containing Phosphines 101
CHH', 1H); 13C{1H} NMR (101 MHz, CD2Cl2, δ in ppm): 90.9 (s, C5H5), –34.0 (dd, 1J(C,P) =
30.8 Hz, 2J(C,P) = 4.4 Hz, CHH'); PPh3 at 134.4 (d, 1J(C,P) = 40.0 Hz, i), 134.0 (d, 2J(C,P) = 10.4
Hz, o), 131.6 (d, 4J(C,P) = 2.4 Hz, p), 129.5 (d, 3J(C,P) = 10.4 Hz, m); P(C6H4OCH3)(C6H4OCH3)'
at 164.8 (d, 4J(C,P) = 2.8 Hz, p to P), 164.4 (d, 4J(C,P) = 2.4 Hz, p' to P), 134.71, 134.70 (2 s, m
and m' to P), 116.2 (d, 2J(C,P) = 13.6 Hz, o to P), 115.8 (d, 2J(C,P) = 12.8 Hz, o' to P), 115.5 (d,
1J(C,P) = 75.9 Hz, i to P), 113.0 (d, 1J(C,P) = 92.7 Hz, i' to P), 56.3, 55.2 (2 s, OCH3 and OCH3');
31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 29.2 (d, 3J(P,P) = 10.9 Hz, PH), 23.1 (d, 3J(P,P) =
10.9 Hz, PPh3), –142.9 (sept, 1J(P,F) = 708 Hz, PF6).
IR (thin film, cm–1):71 1640 (s, νNO). MS:92 804 (100) [14b-H]+, 558 (90) [14b–P(p-
C6H4OCH3)2H]+.
(S)-[(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4OCH3)2H)]+ PF6– ((S)-[14c-H]+ PF6
–).
Complex (S)-26 (0.399 g, 0.715 mmol),82 CH2Cl2 (20 mL), Ph3C+ PF6– (0.305 g, 0.787 mmol),
28c (0.211 g, 0.858 mmol),87 and CH2Cl2 (1 mL) were combined in a procedure analogous to that
given for the racemate. An identical workup gave (S)-[14c-H]+ PF6– as an orange powder (0.645 g,
0.680 mmol, 95%). M.p. 204-205 °C, dec. (capillary). Elemental analysis calcd (%) for
C38H37F6NO3P3Re (949.1): C 48.10, H 3.93, N 1.48; found: C 47.98, H 3.67, N 1.45. [α]25589 =
248° ± 1° (c = 2.00 mg/mL, CH2Cl2). Spectroscopic data were similar to those of the racemate.
(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4OCH3)2) (14c). A Schlenk flask was charged
with [14c-H]+ PF6– (0.154 g, 0.162 mmol) and C6H6 (10 mL). The suspension was vigorously
stirred and t-BuOK (0.0273 g, 0.243 mmol) was added. After 1 h, the orange suspension was
filtered through a plug of Celite®. The plug was rinsed with C6H6 until the filtrate became colorless.
The filtrate was concentrated by oil pump vacuum (to ca. 2 mL), layered with n-pentane (15 mL),
and kept at 4 °C. After 48 h, the orange crystals were collected by filtration and dried by oil pump
vacuum to give 14c (0.100 g, 0.125 mmol, 77%). Dec. pt. 162 °C (capillary). Elemental analysis
calcd (%) for C38H36NO3P2Re (802.9): C 56.85, H 4.52, N 1.74; found: C 56.38, H 4.46, N 1.74.
1H NMR (400 MHz, C6D6, δ in ppm): 7.74 (dd, 3J(H,H) = 8.6 Hz, 3J(H,P) = 5.8 Hz, C6H4,
o to P, 2H), 7.68 (dd, 3J(H,H) = 8.6, 3J(H,P) = 5.8 Hz, C6H4', o to P, 2H), 7.56-7.50 (m, o-C6H5,
102 3 Preparation of Rhenium-Containing Phosphines
6H), 7.07-7.01, 7.00-6.95 (2 m, m-, p-C6H5, 9H), 6.91 (d, 3J(H,H) = 8.6 Hz, C6H4, m to P, 2H),
6.79 (d, 3J(H,H) = 8.6 Hz, C6H4', m to P, 2H), 4.59 (s, C5H5, 5H), 3.30, 3.27 (2 s, OCH3 and
OCH3', 2 × 3H), 2.83 (dd, 2J(H,H) = 11.9 Hz, J(H,P) = 9.7 Hz, CHH', 1H), 2.09 (dd, 2J(H,H) =
11.9 Hz, J(H,P) = 2.5 Hz, CHH', 1H); 13C{1H} NMR (101 MHz, CD2Cl2, δ in ppm): 90.3 (s,
C5H5), –18.1 (dd, 1J(C,P) = 35.4 Hz, 2J(C,P) = 5.2 Hz, CHH'); PPh3 at 136.3 (d, 1J(C,P) = 51.6 Hz,
i), 134.1 (d, 2J(C,P) = 10.7 Hz, o), 130.5 (s, p), 128.8 (d, 3J(C,P) = 10.4 Hz, m);
P(C6H4OCH3)(C6H4OCH3)' at 159.8, 159.5 (2 s, p and p' to P), 138.4 (d, 1J(C,P) = 18.8 Hz, i to P),
137.4 (d, 1J(C,P) = 18.1 Hz, i' to P), 134.2 (d, 2J(C,P) = 20.6 Hz, o to P), 134.0 (d, 2J(C,P) = 18.4
Hz, o' to P), 113.7 (d, 3J(C,P) = 7.0 Hz, m to P), 113.6 (d, 3J(C,P) = 5.9 Hz, m' to P), 55.5, 55.4 (2 s,
OCH3 and OCH3'); 31P{1H} NMR (162 MHz, C6D6, δ in ppm): 27.6 (d, 3J(P,P) = 6.9 Hz, PPh3),
5.1 (d, 3J(P,P) = 6.9 Hz, P(C6H4OCH3)2).
IR (thin film, cm–1):71 1633 (s, νNO). MS:92 803 (10) [14c]+, 558 (100) [14c–P(p-
C6H4OCH3)2H]+.
(S)-(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4OCH3)2) ((S)-14c). (S)-[14c-H]+ PF6– (0.225
g, 0.237 mmol), t-BuOK (0.0398 g, 0.356 mmol), and C6H6 (20 mL) were combined in a procedure
analogous to that given for the racemate. An identical workup gave (S)-14c as an orange-red
powder (0.166 g, 0.206 mmol, 87%). M.p. 125-130 °C, dec. (capillary). Spectroscopic data were
similar to those of the racemate.
[(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4N(CH3)2)2H)]+ PF6– ([14d-H]+ PF6
–). A
Schlenk flask was charged with racemic 26 (0.212 g, 0.380 mmol)82 and CH2Cl2 (15 mL). The
solution was cooled to –78 °C and Ph3C+ PF6– (0.162 g, 0.418 mmol) was added with stirring.
After 1 h, P(p-C6H4N(CH3))2H (28d, 0.134 g, 0.494 mmol)87 dissolved in CH2Cl2 (1 mL) was
added dropwise. After 20 min, the cold bath was removed. After 1 h, the sample was concentrated
by oil pump vacuum (to ca. 4 mL). A CH3OH/EtOH mixture (5.5 mL, 4:5 v/v) was added, followed
by CH2Cl2 until the sample became homogeneous. Then hexanes (ca. 20 mL) were added with
stirring. The precipitate was washed with EtOH (2 × 1 mL) and hexanes (2 × 3 mL). After drying
by oil pump vacuum [14d-H]+ PF6– was obtained as an orange powder (0.345 g, 0.349 mmol, 93%).
3 Preparation of Rhenium-Containing Phosphines 103
M.p. 230-232 °C, dec. (capillary). Elemental analysis calcd (%) for C41H45F6N3OP3Re (989.2): C
49.28, H 4.45, N 4.31; found: C 48.91, H 4.39, N 4.22.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.61 (dd, 3J(H,P) = 12.6 Hz, 3J(H,H) = 8.5 Hz,
C6H4, o to P, 2H), 7.49-7.42, 7.38-7.31 (2 m, C6H5, 15H), 7.26 (dd, 1J(H,P) = 472 Hz, 3J(H,H) =
10.7 Hz, PH, 1H), 7.18 (dd, 3J(H,P) = 12.1, 3J(H,H) = 8.6 Hz, C6H4', o to P, 2H), 6.80 (d, 3J(H,H)
= 8.6 Hz, C6H4, m to P, 2H), 6.62 (d, 3J(H,H) = 8.6 Hz, C6H4', m to P, 2H), 4.87 (s, C5H5, 5H),
3.07, 2.99 (2 s, CH3 and CH3', 2 × 6H), 2.52-2.37 (m, CHH', 2H); 13C{1H} NMR (101 MHz,
CD2Cl2, δ in ppm): 90.8 (s, C5H5), –32.0 (dd, 1J(C,P) = 33.3 Hz, 2J(C,P) = 3.4 Hz, CHH'); PPh3 at
134.7 (d, 1J(C,P) = 53.7 Hz, i), 134.0 (d, 2J(C,P) = 10.5 Hz, o), 131.5 (d, 4J(C,P) = 2.0 Hz, p),
129.5 (d, 3J(C,P) = 10.4 Hz, m); P(C6H4N(CH3)2)(C6H4N(CH3)2)' at 154.2 (d, 4J(C,P) = 1.9 Hz, p
to P), 153.7 (d, 4J(C,P) = 2.0 Hz, p' to P), 134.0 (d, 3J(C,P) = 11.5 Hz, m to P), 133.3 (d, 3J(C,P) =
11.5 Hz, m' to P), 112.6 (d, 2J(C,P) = 13.3 Hz, o to P), 112.3 (d, 2J(C,P) = 12.5 Hz, o' to P), 108.2
(d, 1J(C,P) = 79.9 Hz, i to P), 104.7 (d, 1J(C,P) = 98.9 Hz, i' to P), 40.3, 40.2 (2 s, CH3 and CH3');
31P{1H} NMR (162 MHz, CD2Cl2, δ in ppm): 28.7 (d, 3J(P,P) = 11.9 Hz, PH) 23.9 (d, 3J(P,P) =
11.9 Hz, PPh3), –142.9 (sept, 1J(P,F) = 714.4 Hz, PF6).
IR (thin film, cm–1):71 1645 (s, νNO). MS:92 830 (100) [14d-H]+, 558 (62) [14d–P(p-
C6H4N(CH3)2)2H]+.
(S)-[(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4N(CH3)2H)]+ PF6– ((S)-[14d-H]+ PF6
–).
Complex (S)-26 (0.300 g, 0.538 mmol),82 CH2Cl2 (15 mL), Ph3C+ PF6– (0.234 g, 0.603 mmol),
28d (0.175 g, 0.646 mmol),87 and CH2Cl2 (1 mL) were combined in a procedure analogous to that
given for the racemate. After the reaction mixture was stirred at room temperature for 1 h, the
sample was concentrated by oil pump vacuum (to ca. 3 mL). Then EtOH (7 mL) was added, and the
mixture was kept at –20 °C. After 2 h, the precipitate was collected by filtration, washed with EtOH
(1 mL) and hexanes (2 × 3 mL), and dried by oil pump vacuum to give (S)-[14d-H]+ PF6– as a
yellow powder (0.400 g, 0.410 mmol, 76%). M.p. 180-182 °C, dec. (capillary). Elemental analysis
calcd (%) for C40H43F6N3OP3Re (975.2): C 49.28, H 4.45, N 4.31; found: C 48.79, H 4.71, N 4.21.
[α]25589 = 202° ± 1° (c = 2.00 mg/mL, CH2Cl2). Spectroscopic data were similar to those of the
racemate.
104 3 Preparation of Rhenium-Containing Phosphines
(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4N(CH3)2) (14d). A Schlenk flask was charged
with [14d-H]+ PF6– (0.212 g, 0.217 mmol) and C6H6 (20 mL). The suspension was vigorously
stirred and t-BuOK (0.0365 g, 0.326 mmol) was added. After 1 h, the orange suspension was
filtered through a plug of Celite®. The plug was rinsed with C6H6 until the filtrate became colorless.
The filtrate was concentrated by oil pump vacuum (to ca. 4 mL), layered with n-pentane (15 mL),
and kept at 4 °C. After 48 h, the orange crystals were collected by filtration and dried by oil pump
vacuum to give 14d as a red solid (0.153 g, 0.185 mmol, 85%). Dec. pt. 157-158 °C (capillary).
Elemental analysis calcd (%) for C41H44N3OP2Re (829.2): C 57.96, H 5.11, N 5.07; found: C
58.07, H 4.89, N 4.97.
1H NMR (300 MHz, C6D6, δ in ppm): 7.82 (dd, 3J(H,H) = 8.6 Hz, 3J(H,P) = 6.3 Hz, C6H4,
o to P, 2H), 7.76 (dd, 3J(H,H) = 8.6 Hz, 3J(H,P) = 6.0 Hz, C6H4', o to P, 2H), 7.65-7.55 (m, o-C6H5,
6H), 7.11-6.95 (m, m-, p-C6H5, 9H), 6.72 (d, 3J(H,H) = 8.6 Hz, C6H4, m to P, 2H), 6.60 (d, 3J(H,H)
= 8.6 Hz, C6H4', m to P, 2H), 4.63 (s, C5H5, 5H), 2.94 (dd, 2J(H,H) = 11.5 Hz, J(H,P) = 9.7 Hz,
CHH', 1H), 2.52, 2.50 (2 s, CH3 and CH3', 2 × 6H), 2.25 (dd, 2J(H,H) = 11.5 Hz, J(H,P) = 2.5 Hz,
CHH', 1H); 13C{1H} NMR (76 MHz, C6D6, δ in ppm): 89.9 (s, C5H5), –16.6 (dd, 1J(C,P) = 36.6
Hz, 2J(C,P) = 4.8 Hz CHH'); PPh3 at 136.8 (d, 1J(C,P) = 50.9 Hz, i), 134.1 (d, 2J(C,P) = 10.4 Hz, o),
130.0 (d, 4J(C,P) = 1.6 Hz, p), 128.5 (d, 3J(C,P) = 10.0 Hz, m); P(C6H4N(CH3)2)(C6H4N(CH3)2)'
at 150.5, 150.2 (2 s, p and p' to P), 134.8 (d, 1J(C,P) = 19.8 Hz, i to P), 133.9 (d, 1J(C,P) = 18.4 Hz,
i' to P), 134.4 (s (other line of expected d obscured), o to P), 133.8 (d, 2J(C,P) = 15.9 Hz, o' to P),
113.0 (d, 3J(C,P) = 6.0 Hz, m to P), 112.7 (d, 3J(C,P) = 7.1 Hz, m' to P), 40.4, 40.3 (2 s, CH3 and
CH3'); 31P{1H} NMR (121 MHz, C6D6, δ in ppm): 27.2 (d, 3J(P,P) = 6.7 Hz, PPh3), 2.8 (d, 3J(P,P)
= 6.7 Hz, P(p-C6H4N(CH3)2)2).
IR (thin film, cm–1):71 1633 (s, νNO). MS:92 830 (51) [14d]+, 558 (100) [14d–P(p-
C6H4N(CH3)2)2H]+.
(S)-(η5-C5H5)Re(NO)(PPh3)(CH2P(p-C6H4N(CH3)2)2 ((S)-14d). (S)-[14d-H]+ PF6–
(0.196 g, 0.201 mmol), t-BuOK (0.0338 g, 0.302 mmol), and C6H6 (20 mL) were combined in a
procedure analogous to that given for the racemate. An identical workup gave (S)-14d as a red
3 Preparation of Rhenium-Containing Phosphines 105
powder (0.149 g, 0.179 mmol, 89%). M.p. 130-135 °C, dec. (capillary). Spectroscopic data were
similar to those of the racemate.
[(η5-C5H5)Re(NO)(PPh3)(CH2P(2-biphen)2H)]+ PF6– ([14e-H]+ PF6
–). A Schlenk flask
was charged with racemic 26 (0.212 g, 0.380 mmol)82 and CH2Cl2 (15 mL). The solution was
cooled to –78 °C and Ph3C+ PF6– (0.162 g, 0.418 mmol) was added with stirring. After 1 h, P(2-
biphen)2H (28e, 0.167 g, 0.494 mmol) dissolved in CH2Cl2 (1.5 mL) was added dropwise. After 20
min, the cold bath was removed. After 1 h, the sample was concentrated by oil pump vacuum (to ca.
4 mL). A CH3OH/EtOH mixture (4 mL, 1:1 v/v) was added. The solution was added dropwise to
vigorously stirred hexanes (75 mL). The precipitate was collected by filtration, washed with
hexanes (3 × 3 mL), and dried by oil pump vacuum to give [14e-H]+ PF6– as an orange powder
(0.376 g, 0.361 mmol, 95%). M.p. 236 °C, dec. (capillary). Elemental analysis calcd (%) for
C48H41F6NOP3Re (1041.2): C 55.38, H 3.97, N 1.35; found: C 55.28, H 4.05, N 1.37.
1H NMR (400 MHz, CD2Cl2, δ in ppm): 7.79-7.67, 7.62-7.35, 7.26-7.17, 7.11-7.00 (4 m,
aryl-H, 33H), 6.48 (dd (other part of expected ddd obscured), 3J(H,H) = 12.4 Hz, 3J(H,H') 3.1 Hz,
PH, 0.5H), 4.46 (s, C5H5, 5H), 2.30-2.16 (m, CHH', 1H), 1.61-1.50 (m, CHH', 1H); 13C{1H} NMR
(101 MHz, CD2Cl2, δ in ppm): 90.7 (s, C5H5), –30.9 (dd, 1J(C,P) = 28.8 Hz, 2J(C,P) = 3.8 Hz,
CHH'); PPh3 at 134.4 (d, 1J(C,P) = 53.7 Hz, i), 133.8 (d, 2J(C,P) = 10.7 Hz, o), 131.6 (d, 4J(C,P) =
1.9 Hz, p), 129.4 (d, 3J(C,P) = 10.7 Hz, m); P(2-biphen)(2-biphen)' at93 147.8, (d, J(C,P) = 6.5 Hz),
146.6 (d, J(C,P) = 9.2 Hz), 139.6 (d, J(C,P) = 5.0 Hz), 139.5 (d, J(C,P) = 4.6 Hz), 134.3 (d, J(C,P)
= 2.3 Hz), 134.0 (d, J(C,P) = 2.7 Hz), 133.5 (d, J(C,P) = 10.7 Hz), 132.7 (d, J(C,P) = 10.0 Hz),
132.4 (d, J(C,P) = 8.8 Hz), 132.1 (d, J(C,P) = 8.1 Hz), 129.78, 129.76, 129.7, 129.5, 129.4 (5 s),
129.2 (d, J(C,P) = 10.7 Hz), 128.7 (d, J(C,P) = 11.9 Hz), 125.4 (d, 1J(C,P) = 65.2 Hz, i to P), 118.7
(d, 1J(C,P) = 87.4 Hz, i' to P); 31P{1H} NMR (162 MHz, CD2Cl2, δ in ppm): 24.0 (d, 3J(P,P) =
20.8 Hz, PH), 20.5 (d, 3J(P,P) = 20.8 Hz, PPh3), –142.9 (sept, 1J(P,F) = 708 Hz, PF6).
IR (thin film, cm–1):71 1660 (s, νNO). MS:92 896 (80) [14e-H]+, 558 (100) [14e–P(2-
biphen)2H]+.
(η5-C5H5)Re(NO)(PPh3)(CH2P(2-biphen)2) (14e). A Schlenk flask was charged with
106 3 Preparation of Rhenium-Containing Phosphines
racemic [14e-H]+ PF6– (0.520 g, 0.500 mmol) and C6H6 (30 mL). The suspension was vigorously
stirred and solid t-BuOK (0.0840 g, 0.749 mmol) was added. After 1 h, the orange suspension was
filtered through a plug of Celite®. The plug was rinsed with C6H6 until the filtrate became colorless.
The filtrate was concentrated by oil pump vacuum to ca. 5 mL and layered with n-pentane (35 mL).
The orange precipitate was collected by filtration and dried by oil pump vacuum to give pure 14e
(0.406 g, 0.453 mmol, 91%). Dec. pt. 205 °C (capillary). Elemental analysis calcd (%) for
C48H40NOP2Re (895.2): C 64.42, H 4.50, N 1.57; found: C 64.15, H 4.11, N 1.64.
1H NMR (400 MHz, CD2Cl2, δ in ppm): 7.41-7.33, 7.31-7.20, 7.14-6.98 (3 m, aryl-H, 32H),
4.39 (s, C5H5, 5H), 1.89 (apparent dt, 2J(H,H) = 12.6 Hz, 2J(H,P) = 3J(H,P) = 8.6 Hz, CHH', 1H),
1.32 (ddd, 2J(H,H) = 12.6 Hz, 2J(H,P) = 8.6 Hz, 3J(H,P) = 2.5 Hz, CHH', 1H); 13C{1H} NMR (76
MHz, CD2Cl2, δ in ppm): 90.9 (dd, 2J(C,P) = 4.4 Hz, 3J(C,P) = 1.2 Hz, C5H5), –16.3 (dd, 1J(C,P) =
39.6 Hz, 2J(C,P) = 5.1 Hz, CHH'); PPh3 at 136.7 (d, 1J(C,P) = 51.4 Hz, i), 134.1 (d, 2J(C,P) = 10.5
Hz, o), 130.5 (d, 4J(C,P) = 2.0 Hz, p), 128.8 (d, 3J(C,P) = 10.1 Hz, m); P(2-biphen)(2-biphen)' at93
149.2, 149.2, 148.8, 148.7, 146.5, 146.2, 144.3, 144.2, 144.0, 144.0, 139.8, 139.6, 135.8, 135.7,
131.3, 131.0, 131.0, 130.7, 130.7, 130.5, 130.4, 130.4, 130.3, 128.2, 128.1, 127.4, 127.3, 127.0,
126.9, 126.7; 31P{1H} NMR (162 MHz, CD2Cl2, δ in ppm): 24.6 (d, 3J(P,P) = 6.9 Hz, PPh3), –5.7
(d, 3J(P,P) = 6.9 Hz, P(2-biphen)2).
IR (thin film, cm–1):71 1621 (s, νNO). MS:92 m/z (%): 896 (80) [14e-H]+, 558 (100) [14e–
P(2-biphen)2]+.
[(η5-C5H5)Re(NO)(PPh3)(CH2P(α-naph)2H)]+ PF6– ([14f-H]+ PF6
–). A Schlenk flask
was charged with racemic 26 (0.500 g, 0.896 mmol)82 and CH2Cl2 (25 mL). The solution was
cooled to –78 °C and Ph3C+ PF6– (0.382 g, 0.986 mmol) was added with stirring. After 1 h, P(α-
naph)2H (28f, 0.308 g, 1.075 mmol) dissolved in CH2Cl2 (5 mL) was added. After 20 min, the cold
bath was removed. After 1 h, the sample was concentrated by oil pump vacuum (to ca. 4 mL). A
CH3OH/EtOH mixture (4 mL, 1:1 v/v) was added. Then hexanes (ca. 20 mL) were added dropwise
with vigorous stirring. The precipitate was collected by filtration, washed with hexanes (3 × 3 mL),
and dried by oil pump vacuum to give [14f-H]+ PF6– as an orange powder (0.772 g, 0.781 mmol,
87%). Dec. pt. 181-183 °C (capillary). Elemental analysis calcd (%) for C44H37F6NOP3Re (989.2):
3 Preparation of Rhenium-Containing Phosphines 107
C 53.44, H 3.77, N 1.42; found: C 53.20, H 3.72, N 1.39.
1H NMR (400 MHz, CD2Cl2, δ in ppm): 6.48 (dd (other part of expected ddd obscured),
3J(H,H) = 12.3 Hz, 3J(H,H') = 3.0 Hz, PH, 0.5H), 8.45 (dd, 3J(H,P) = 17.3 Hz, 3J(H,H) = 7.1 Hz, 2-
C10H7, 1H), 8.34 (dd, 3J(H,P) = 16.0 Hz, 3J(H,H) = 7.2 Hz, 2-C10H7', 1H), 8.25, 8.17, 8.12, 8.02,
7.95, 7.82 (6 d, 3J(H,H) = 8.2, 8.4, 8.4, 8.0, 8.0, and 8.3 Hz, 4-, 5-, 8-C10H7 and 4-, 5-, 8-C10H7', 6
× 1H), 7.80-7.71, 7.69-7.60 (2 m, 6-, 7-C10H7 and 6-, 7-C10H7', 2 × 2H), 7.55 (apparent t, 3J(H,H)
= 7.0 and 7.0 Hz, 3-C10H7, 1H), 7.49 (apparent t, 3J(H,H) = 7.2 and 7.2 Hz, 3-C10H7', 1H), 7.47-
7.31 (m, C6H5, 15H), 4.62 (s, C5H5, 5H), 3.06-2.96, 2.87-2.74 (2 m, CHH', 2 × 1H); 13C{1H}
NMR (101 MHz, CD2Cl2, δ in ppm): 90.6 (s, C5H5), –33.9 (d, 1J(C,P) = 27.4 Hz, CHH'); PPh3 at
134.1 (d, 1J(C,P) = 54.8 Hz, i), 133.8 (d, 2J(C,P) = 10.5 Hz, o), 131.4 (d, 4J(C,P) = 2.1 Hz, p),
129.4 (d, 3J(C,P) = 10.5 Hz, m); P(α-naph)(α-naph)' at93 135.8 (d, J(C,P) = 2.9), 135.6 (d, J(C,P) =
2.9 Hz), 135.5 (d, J(C,P) = 10.5 Hz), 135.2 (d, J(C,P) = 11.8 Hz), 134.0 (d, J(C,P) = 4.6 Hz), 133.9
(d, J(C,P) = 3.4 Hz), 132.8 (d, J(C,P) = 8.4 Hz), 132.3 (d, J(C,P) = 5.9 Hz), 130.4, 130.2, 129.6,
128.8, 128.1, 127.5 (6 s), 126.0 (d, J(C,P) = 11.4 Hz), 125.9 (d, J(C,P) = 11.4 Hz), 124.1 (d, J(C,P)
= 8.0 Hz), 123.5 (d, J(C,P) = 8.0 Hz), 120.5 (d, 1J(C,P) = 65.3 Hz, 1-C10H7), 119.6 (d, 1J(C,P) =
83.5 Hz, 1-C10H7'); 31P{1H} NMR (162 MHz, CD2Cl2, δ in ppm): 22.7 (d, 3J(P,P) = 12.9 Hz, PH),
20.3 (d, 3J(P,P) = 12.9 Hz, PPh3), –142.9 (sept, 1J(P,F) = 708 Hz, PF6).
IR (thin film, cm–1):71 1656 (s, νNO). MS:92 844 (45) [14f-H]+, 558 (100) [14f–P(α-
naph)2H]+.
(η5-C5H5)Re(NO)(PPh3)(CH2P(α-naph)2) (14f). A Schlenk flask was charged with [14f-
H]+ PF6– (0.361 g, 0.365 mmol) and C6H6 (22 mL). The suspension was vigorously stirred and
solid t-BuOK (0.0613 g, 0.548 mmol) was added. After 1 h, the orange suspension was filtered
through a plug of Celite®. The plug was rinsed with C6H6 until the filtrate became colorless. The
filtrate was concentrated by oil pump vacuum (to ca. 5 mL), and n-pentane (35 mL) was added
dropwise with vigorous stirring. The precipitate was collected by filtration and dried by oil pump
vacuum to give 14f as a pale yellow powder (0.197 g, 0.234 mmol, 64%). M.p. 187-189 °C, dec.
(capillary). Elemental analysis calcd (%) for C44H36NOP2Re (843.2): C 62.70, H 4.30, N 1.66;
found: C 62.80, H 4.50, N 1.76.
108 3 Preparation of Rhenium-Containing Phosphines
1H NMR (400 MHz, CD2Cl2, δ in ppm): 8.87-8.82, 8.78-8.72, 7.82-7.69, 7.52-7.28 (4 m,
aryl-H, 29H), 4.62 (s, C5H5, 5H), 2.63 (dd, 2J(H,H) = 11.5 Hz, J(H,P) = 9.6 Hz, CHH', 1H), 1.91 (d,
2J(H,H) = 11.5 Hz, CHH', 1H); 13C{1H} NMR (76 MHz, CD2Cl2, δ in ppm): 90.4 (s, C5H5), –19.1
(dd, 1J(C,P) = 36.8 Hz, 2J(C,P) = 4.8 Hz, CHH'); PPh3 at 136.3 (d, 1J(C,P) = 51.9 Hz, i), 134.2 (d,
2J(C,P) = 10.4 Hz, o), 130.7 (d, 4J(C,P) = 2.1 Hz, p), 128.9 (d, 3J(C,P) = 10.1 Hz, m); P(α-naph)(α-
naph)' at93 145.9, 145.5, 143.5, 143.2, 136.5, 136.5, 136.3, 136.2, 134.2, 133.9, 133.8, 130.8, 130.3,
129.1, 128.4, 128.4, 127.6, 127.3, 127.2, 126.9, 126.1, 126.0, 125.8, 125.8, 125.7; 31P{1H} NMR
(162 MHz, CD2Cl2, δ in ppm): 26.5 (d, 3J(P,P) = 6.9 Hz, PPh3), –21.2 (d, 3J(P,P) = 6.9 Hz, P(α-
naph)2).
IR (thin film, cm–1):71 1644 (s, νNO). MS:92 842 (30) [14f–H]+, 558 (100) [14f–P(α-
naph)2]+.
(SReSC)-[(η5-C5H5)Re(NO)(PPh3)(CHCH3PPh2H)]+ PF6– ((SReSC)-[15b-H]+ PF6
–). A
Schlenk flask was charged with (S)-(η-C5H5)Re(NO)(PPh3)(CH2CH3) (0.114 g, 0.199 mmol)89 and
CH2Cl2 (5 mL). The solution was cooled to –78°C and Ph3C+ PF6– (0.0722 g, 0.219 mmol) was
added with stirring. After 1 h, the orange mixture had turned bright yellow. Then PPh2H (0.0555 g,
0.299 mmol) dissolved in CH2Cl2 (0.6 mL) was added. After 1 h, the mixture was allowed to
slowly warm to room temperature over the course of 1 h. The solution was concentrated (to ca. 1
mL) and n-pentane was added. The oil-like precipitate was isolated by decantation and dissolved in
CH2Cl2/t-BuOH (7 mL, 2:5 v/v). The solvent was concentrated by oil pump vacuum with vigorous
stirring and a yellow powder precipitated, which was collected by filtration, washed with n-pentane
(2 × 0.5 mL), and dried by oil pump vacuum to give (SReSC)-[15b-H]+ PF6–as a pale yellow
powder (0.0920 g, 55%). Dec. pt. 125-128 °C (capillary). Elemental analysis calcd (%) for
C37H35F6NOP3Re (902.8): C 49.22, H 3.91, N 1.55; found: C 49.17, H 4.06, N 1.52. [α]25589 =
105° ± 1° (c = 1.00 mg/mL, CH2Cl2).
1H NMR (400 MHz, CDCl3, δ in ppm): 7.79-7.30 (m, C6H5, 25H), 7.48 (dd, 1J(H,P) = 493
Hz, 3J(H,H) = 9.0 Hz, PH, 1H), 5.06 (s, C5H5, 5H), 3.46-3.35 (m, CHCH3, 1H), 1.27 (dd, J(H,P) =
24.5 Hz, 3J(H,H) = 7.5 Hz, CHCH3, 3H); 13C{1H} NMR (76 MHz, CD2Cl2, δ in ppm; the ReCH
signal was not observed): 91.6 (s, C5H5); PPh3 at 133.9 (d, 1J(C,P) = 54.0 Hz, i), 133.7 (d, 2J(C,P)
3 Preparation of Rhenium-Containing Phosphines 109
= 10.6 Hz, o), 131.5 (d, 4J(C,P) = 2.2 Hz, p), 129.5 (d, 3J(C,P) = 10.4 Hz, m); PPhPh' at 134.6 (d,
4J(C,P) = 2.6 Hz, p to P), 134.1 (d, 4J(C,P) = 2.6 Hz, p' to P), 133.2 (d, 2J(C,P) = 9.1 Hz, o to P),
132.6 (d, 2J(C,P) = 9.1 Hz, o' to P), 130.5 (d, 3J(C,P) = 11.5 Hz, m to P), 130.3 (d, 3J(C,P) = 11.7
Hz, m' to P), 122.4 (d, 1J(C,P) = 57.7 Hz, i to P), 121.5 (d, 1J(C,P) = 68.1 Hz, i' to P), 20.4 (s, CH3);
31P{1H} NMR (122 MHz, CD2Cl2, δ in ppm): 22.7 (d, 3J(P,P) = 14.9 Hz, PPh3), 31.9 (d, 3J(P,P) =
14.6 Hz, PH), –142.8 (sept, 1J(P,F) = 713 Hz, PF6).
IR (thin film, cm–1):71 1668 (s, νNO). MS:92 758 (18) [15b-H]+, 572 (100) [15b–PPh2H]+.
(SReSP)/(SReRP)-[(η5-C5H5)Re(NO)(PPh3)(CH2PPh(α-naph)H)]+ PF6– ((SReSP)/
(SReRP)-[17c-H]+ PF6–). A Schlenk flask was charged with (S)-26 (0.200 g, 0.358 mmol)82 and
CH2Cl2 (20 mL). The solution was cooled to –78 °C and Ph3C+ PF6– (0.153 g, 0.394 mmol) was
added with stirring. After 1 h, PPh(α-naph)H (29c, 0.126 g, 0.537 mmol) dissolved in CH2Cl2 (1
mL) was added. After 3 h, the mixture was allowed to slowly warm to room temperature over the
course of 1 h. The mixture was concentrated by oil pump vacuum (to ca. 7 mL) and n-pentane (10
mL) was added dropwise with stirring. A first crop of (SReSP)/(SReRP)-[17c-H]+ PF6– precipitated
as a yellow powder (0.148 g, 0.152 mmol, 72% de, 43%), which was collected by filtration and
dried by oil pump vacuum. The yellow powder was dissolved in CH2Cl2 (8 mL) and n-pentane (8
mL) was slowly added dropwise with vigorous stirring. The (SReSP)/(SReRP)-[17c-H]+ PF6– was
similarly isolated. Several such cycles may be necessary to obtain (SReSP)/(SReRP)-[17c-H]+ PF6–
with a high diastereomeric excess (best result for diastereomer 1: 94% de, absolute configuration
not assigned). The supernatant from the first crop was treated with n-pentane (15 mL) with stirring.
A yellow powder precipitated, which was collected by filtration and dried by oil pump vacuum to
give diastereomer 2 (0.120 g, 0.124 mmol, –92% de, 35%).
Analytical data for diastereomer 1 (94% de): Elemental analysis calcd (%) for
C40H35F6NOP3Re.(CH2Cl2)0.33 (967.1): C 50.09, H 3.72, N 1.45; found: C 50.40, H 3.71, N 1.41.
1H NMR (400 MHz, CD3CN, δ in ppm): aryl-H at 8.28 (d, J = 8.3 Hz, 1H), 8.18-8.03 (m,
2H), 7.84-7.23 (m, 24H); 7.88 (dt, 1J(H,P) = 497 Hz, 3J(H,H) = 8.1 Hz, PH, 1H), 5.32 (CH2Cl2),
5.05 (s, C5H5, 5H), 2.80-2.47 (m, CHH', 2H); 13C{1H} NMR (76 MHz, CD3CN, δ in ppm): 91.7 (d,
2J(C,P) = 1.4 Hz, C5H5); PPh3 at 135.0 (d, 1J(C,P) = 53.5 Hz, i), 134.2 (d, 2J(C,P) = 10.4 Hz, o),
110 3 Preparation of Rhenium-Containing Phosphines
131.8 (d, 4J(C,P) = 2.5 Hz, p), 129.8 (d, 3J(C,P) = 10.4 Hz, m); P(α-naph)Ph at93,94 136.5, 136.3,
136.05, 136.01, 134.79, 134.75, 133.0 (d, J(C,P) = 10.4 Hz), 130.7, 130.6, 129.8, 129.4, 128.2,
126.5, 126.3, 125.9, 125.1, 125.0, 124.9; 31P{1H} NMR (162 MHz, CD3CN, δ in ppm): 28.3 (d,
3J(P,P) = 17.1 Hz, PH), 22.1 (d, 3J(P,P) = 17.1 Hz, PPh3), –143.2 (sept, 1J(P,F) = 707 Hz, PF6).
IR (thin film, cm–1):71 1644 (s, νNO). MS:95 792 (100) [17c]+, 558 (62) [17c–PPh(α-
naph)H]+.
Analytical data for diastereomer 2 (–92% de): Elemental analysis calcd (%) for
C40H35F6NOP3Re.(CH2Cl2)0.33 (967.1): C 50.09, H 3.72, N 1.45; found: C 50.35, H 3.67, N 1.49.
1H NMR (400 MHz, CD3CN, δ in ppm): aryl-H at 8.38-8.23 (m, 2H), 8.13 (d, J = 7.9 Hz,
1H), 7.89-7.27 (m, 24H); 7.80 (ddd, 1J(H,P) = 500 Hz, 3J(H,H) = 9.8 and 6.7 Hz, PH, 1H), 5.32
(CH2Cl2), 4.84 (s, C5H5, 5H), 2.86-2.71 (m, CHH', 1H), 2.61-2.44 (m, CHH', 1H); 13C{1H} NMR
(76 MHz, CD2Cl2, δ in ppm): 90.5 (s, C5H5), –35.8 (d, 2J(C,P) = 27.2 Hz, CH2); aryl-C at93 136.4,
136.32, 136.30, 134.6, 134.31, 134.29, 134.2, 134.0 (d, 2J(C,P) = 10.4 Hz, o-PPh3), 133.8, 132.5,
132.4, 132.1, 132.0, 131.7, 131.6, 131.5, 131.5, 130.7, 130.4, 130.2, 129.8, 129.6 (d, 3J(C,P) = 10.7
Hz, m-PPh3), 129.4, 128.4, 126.1, 125.9, 125.2, 124.7, 124.6, 124.3, 118.4, 117.3; 31P{1H} NMR
(162 MHz, CD2Cl2, δ in ppm): 29.0 (d, 3J(P,P) = 12.6 Hz, PH), 22.3 (d, 3J(P,P) = 12.6 Hz, PPh3), –
143.9 (sept, 1J(P,F) = 709 Hz, PF6).
IR (thin film, cm–1):71 1652 (s, νNO). MS:95 792 (100) [17c]+, 558 (32) [17c–PPh(α-
naph)H]+.
(SReSP)/(SReRP)-[(η5-C5H5)Re(NO)(PPh3)(CH2PPh(α-naph)) ((SReSP)/(SReRP)-17c). A
Schlenk flask was charged with (SReSP)/(SReRP)-[17c-H]+ PF6– (each 0.160 g, 0.172 mmol,
diastereomer 1: 94% de; diastereomer 2: –92% de) and C6H6 (20 mL). The suspension was stirred
and t-BuOK (0.0288 g, 0.257 mmol) was added. After 1 h, the orange suspension was filtered
through a plug of Celite®. The plug was rinsed with C6H6 until the filtrate became colorless. The
filtrate was concentrated by oil pump vacuum (to ca. 2 mL), layered with n-pentane (15 mL), and
kept at 4 °C. After 48 h, an orange solid precipitated, which was collected by filtration and dried by
oil pump vacuum to give (SReSP)/(SReRP)-17c (diastereomer 1: 0.0868 g, 0.110 mmol, 94% de,
64%; diastereomer 2: 0.0709 g, 0.0894 mmol, –92% de, 52%).
3 Preparation of Rhenium-Containing Phosphines 111
Analytical data for diastereomer 1 (94% de): 1H NMR (300 MHz, CD2Cl2, δ in ppm):
aryl-H at 8.61-8.55 (m, 1H), 7.81-7.72 (m, 2H), 7.62-7.57 (m, 1H), 7.49-7.31 (m, 20H), 7.21-7.13
(m, 3H); 4.71 (s, C5H5, 5H), 2.47 (apparent t, 2J(H,H) = J(H,P) = 10.5 Hz, CHH', 1H), 1.98 (dd,
2J(H,H) = 10.5 Hz, J(H,P) = 1.99 Hz, CHH', 1H); 31P{1H} NMR (121 MHz, CD2Cl2, δ in ppm):
26.3 (d, 3J(P,P) = 7.4 Hz, PPh3), –6.7 (d, 3J(P,P) = 7.4 Hz, PPh(α-naph)).
Analytical data for diastereomer 2 (–92% de): 1H NMR (400 MHz, CD2Cl2, δ in ppm):
aryl-H at 8.70-8.65 (m, 1H), 7.83-7.75 (m, 2H), 7.62-7.59 (m, 1H), 7.47-7.31 (m, 20H), 7.21-7.14
(m, 3H); 4.71 (s, C5H5, 5H), 2.55 (apparent t, 2J(H,H) = J(H,P) = 12.0 Hz, CHH', 1H), 1.86 (dd,
2J(H,H) = 12.0 Hz, J(H,P) = 2.1 Hz, CHH', 1H); 31P{1H} NMR (121 MHz, CD2Cl2, δ in ppm):
26.4 (d, 3J(P,P) = 6.9 Hz, PPh3), –4.0 (d, 3J(P,P) = 6.9 Hz, PPh(α-naph)).
(SReSP)/(SReRP)-[(η5-C5H5)Re(NO)(PPh3)(CH2PPh(β-naph)H)]+ PF6– ((SReSP)/
(SReRP)-[17d-H]+ PF6–). A Schlenk flask was charged with (S)-26 (0.200 g, 0.358 mmol)82 and
CH2Cl2 (20 mL). The solution was cooled to –78 °C and Ph3C+ PF6– (0.153 g, 0.394 mmol) was
added with stirring. After 1 h, PPh(β-naph)H (29d, 0.126 g, 0.537 mmol) dissolved in CH2Cl2 (1
mL) was added dropwise. After 3 h, the cold bath was removed. After 1 h, the mixture was
concentrated by oil pump vacuum (to ca. 5 mL) and diethyl ether (10 mL) was added dropwise. A
yellow powder precipitated, which was collected by filtration and dried by oil pump vacuum to give
(SReSP)/(SReRP)-[17d-H]+ PF6– (0.280 g, 0.298 mmol, 83%) as a 50:50 mixture of diastereomers.
Elemental analysis calcd (%) for C40H35F6NOP3Re.(CH2Cl2)0.33 (967.1): C 50.09, H 3.72, N 1.45;
found: C 50.12, H 3.90, N 1.37.
1H NMR (300 MHz, CD2Cl2, δ in ppm): 8.49 (d (other part of expected ddd obscured),
3J(H,H) = 16.1 Hz, ½ PH, 0.25H), 8.16-7.23 (m, aryl-H and ½ PH, 27.5H), 6.43 (d (other part of
expected ddd obscured), 3J(H,H) = 12.5 Hz, ½ PH, 0.25H), 4.91, 4.87 (2 s, 2 C5H5, 5H), 2.78-2.63,
2.58-2.40 (2 m, CHH', 2H); 13C{1H} NMR (101 MHz, CD2Cl2, δ in ppm): 91.1, 90.0 (2 d, 2J(C,P)
= 1.1 and 1.4 Hz, 2 C5H5), –35.5 (d, 1J(C,P) = 32.1 Hz, CHH'); PPh3 at 134.3 (d, 1J(C,P) = 54.1 Hz,
i), 134.0 (d, 2J(C,P) = 10.6 Hz, o), 131.6 (d, 4J(C,P) = 2.1 Hz, p), 129.6 (d, 3J(C,P) = 10.5 Hz, m),
PPh(β-naph) at93 135.8, 135.7, 134.9, 134.9, 134.8, 133.9, 133.2, 132.8, 132.6, 132.1, 132.0, 130.6,
130.4, 130.3, 130.2, 130.0, 129.4, 128.7, 128.5, 125.6, 125.5, 123.5, 122.3, 122.1, 121.2; 31P{1H}
112 3 Preparation of Rhenium-Containing Phosphines
NMR (121 MHz, CD2Cl2, δ in ppm): 30.2, 29.7 (2 d, 3J(P,P) = 12.6 and 11.9 Hz, 2 PH), 21.6, 21.5
(2 d, 3J(P,P) = 11.9 and 12.6 Hz, 2 PPh3), –143.9 (sept, 1J(P,F) = 711.4 Hz, PF6).
IR (thin film, cm–1):71 1652 (s, νNO). MS:95 793 (60) [17d-H]+, 558 (100) [17d–PPh(β-
naph)H]+.
(SReSP)/(SReRP)-(η5-C5H5)Re(NO)(PPh3)(CH2PPh(β-naph)) ((SReSP)/(SReRP)-17d). A
Schlenk flask was charged with (SReSP)/(SReRP)-[17d-H]+ PF6– (0.200 g, 0.213 mmol, 50:50
diastereomer mixture) and C6H6 (20 mL). The suspension was vigorously stirred and t-BuOK
(0.0358 g, 0.320 mmol) was added. After 1 h, the orange suspension was filtered through a plug of
Celite®. The plug was rinsed with C6H6 until the filtrate became colorless. The filtrate was
concentrated by oil pump vacuum (to ca. 4 mL), layered with n-pentane (10 mL), and kept at 4 °C.
After 168 h, a dark red solid had precipitated, which was collected by filtration and dried by oil
pump vacuum to give (SReSP)/(SReRP)-17d (0.0868 g, 0.110 mmol, 26%) as a 74:26 mixture of
diastereomers 1/2 (48% de). The C6H6/n-pentane filtrate was treated with a large excess of n-
pentane. The precipitate was collected by filtration and dried by oil pump vacuum to give
(SReSP)/(SReRP)-17d (0.0709 g, 0.0894 mmol, 21%) as a 4:96 mixture of diastereomer 1/2 (–92%
de).
Analytical data for diastereomer 1 (48% de): 1H NMR (300 MHz, CD2Cl2, δ in ppm):
aryl-H at 8.01, 7.92 (2 d, 3J(H,H) = 7.1 Hz, together 1H (27:73)), 7.82-7.64 (m, 3H), 7.26-7.19 (m,
25H); 4.78, 4.76 (2 s, C5H5, together 5H (74:26)), 2.58-2.45 (m, CHH', 1H), 2.02-1.91 (m, CHH',
1H). 31P{1H} NMR (121 MHz, CD2Cl2, δ in ppm): 27.8 (d, 3J(P,P) = 7.4 Hz, PPh3), 9.11 (d,
3J(P,P) = 7.4 Hz, PPh(β-naph)).
Analytical data for diastereomer 2 (–92% de): 1H NMR (300 MHz, CD2Cl2, δ in ppm):
aryl-H at 7.99 (d, 3J(H,H) = 7.1 Hz, 1H), 7.81-7.73 (m, 2H), 7.69 (d, 3J(H,H) = 8.3 Hz, 1H), 7.52-
7.16 (m, 25H); 4.76 (s, C5H5, 5H), 2.48 (dd, 2J(H,H) = 12.2 Hz, J(H,P) = 9.9 Hz, CHH', 1H), 1.98
(dd, 2J(H,H) = 12.2 Hz, J(H,P) = 2.4 Hz, CHH', 1H); 31P{1H} NMR (121 MHz, CD2Cl2, δ in ppm):
27.6 (d, 3J(P,P) = 6.7 Hz, PPh3), 9.51 (d, 3J(P,P) = 6.7 Hz, PPh(β-naph)).
3 Preparation of Rhenium-Containing Phosphines 113
3.4.3 Preparation of Secondary Phosphines
P(2-biphen)2H (28e). A Grignard reagent was prepared from Mg (0.3360 g, 13.99 mmol)
and 2-biphenylbromide (2.964 g, 12.72 mmol) in THF (40 mL). Then THF (40 mL) was added, but
some white precipitate remained undissolved. A solution of PCl3 (0.8314 g, 6.025 mmol) in THF
(30 mL) was cooled to –78 °C. The Grignard mixture was added dropwise with stirring over 2 h.
The solution was allowed to warm to room temperature overnight. It was then cooled to 0 °C, and
LiAlH4 (0.2290 g, 6.025 mmol) was added carefully with vigorous stirring (evolving gas). After 5
min, freshly degassed aqueous NaOH (2 N, 12 mL) was added. The organic phase was separated
and filtered through a plug of silica gel. The plug was rinsed with THF (30 mL). The solvent was
removed from the filtrate by oil pump vacuum. The residue was dissolved in a minimum of CH2Cl2.
This solution was layered with n-pentane and kept at 4 °C. After 168 h, the white precipitate was
collected by filtration and dried by oil pump vacuum to give 28e as a white solid (1.425 g, 4.218
mmol, 70%).
1H NMR (400 MHz, CDCl3, δ in ppm): 7.40-7.12 (m, aryl-H, 18H), 4.61 (br s, PH, 1H);
13C{1H} NMR (76 MHz, CDCl3, δ in ppm):93 aryl-C at 146.9 (d, J(C,P) = 15.9 Hz), 144.9 (d,
J(C,P) = 3.2 Hz), 135.5 (d, J(C,P) = 10.3 Hz), 133.9 (d, J(C,P) = 14.6 Hz), 129.8 (d, J(C,P) = 2.8
Hz), 129.1 (d, J(C,P) = 3.0 Hz), 128.3, 127.8, 127.2, 127.1 (4 s); 31P{1H} NMR (162 MHz, CDCl3,
δ in ppm): –53.3 (s).
P(α-naph)2H (28f). A Grignard reagent was prepared from Mg (1.217 g, 50.00 mmol) and
α-naphthylbromide (10.35 g, 50.00 mmol) in THF (60 mL). Then THF (30 mL) was added. A
solution of PCl3 (3.136 g, 22.70 mmol) in THF (30 mL) was prepared and cooled to 0 °C. The
Grignard solution was added dropwise with stirring over 2 h. The mixture was allowed to warm to
room temperature overnight, and was filtered through a plug of Celite®. The filtrate was then
cooled to 0 °C, and LiAlH4 (1.670g, 22.70 mmol) was added carefully with vigorous stirring
(evolving gas). After 5 min, freshly degassed aqueous NaOH (2 N, 10 mL) was added. The organic
phase was separated and filtered through a plug of silica gel. The solvent was removed from the
114 3 Preparation of Rhenium-Containing Phosphines
filtrate by oil pump vacuum. The residue was distilled twice (Kugelrohr, 2 × 10–2 mbar, first
distillation 280-300 °C, second distillation 265-270 °C). The distillate was crystallized from a very
concentrated solution in CH2Cl2. White crystals were collected by filtration and dried by oil pump
vacuum to give 28f (2.000 g, 7.840 mmol, 35%). Analytical data agreed with that in the literature.88
PPh(α-naph)H (29c). A Grignard reagent was prepared from Mg (1.217 g, 50.00 mmol)
and α-naphthylbromide (10.35 g, 50.00 mmol) in THF (50 mL). Then THF (50 mL) was added. A
solution of PPhCl2 (8.136 g, 45.45 mmol) in THF (160 mL) was cooled to –78 °C. The Grignard
solution was added dropwise with stirring over 2 h. The solution was allowed to warm to room
temperature overnight. It was then cooled to 0 °C, and LiAlH4 (2.073 g, 54.54 mmol) was carefully
added with vigorous stirring (evolving gas). After 30 min, freshly degassed aqueous NaOH (2 N, 10
mL) was added. After 1 h, the organic phase was filtered through a plug of silica gel topped with a
plug of Celite®. The plugs were rinsed with THF (50 mL). The solvent was removed from the
filtrates by oil pump vacuum. The residue was distilled (Kugelrohr, 6.6 × 10–3 mbar, 185-200 °C)
to give 29c as a colorless liquid (4.537 g, 19.23 mmol, 42%).
1H NMR (400 MHz, CDCl3, δ in ppm): aryl-H at 8.29-8.24 (m, 1H), 7.90-7.85 (m, 2H),
7.68 (t, J = 7.2 Hz, 1H), 7.53-7.46 (m, 4H), 7.43 (t, J = 7.5 Hz, 1H), 7.33-7.27 (m, 3H); 5.53 (br s,
PH, 1H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm):93 134.9, 134.8, 134.2, 134.1, 133.9, 133.9,
133.8, 133.6, 133.6, 132.1, 132.0, 129.8, 128.8, 128.6, 128.5, 128.5, 126.5, 126.4, 126.2, 126.0,
125.6, 125.5; 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): –50.0 (s).
PPh(β-naph)H (29d). A Grignard reagent was prepared from Mg (1.217 g, 50.00 mmol)
and β-naphthylbromide (10.35 g, 50.00 mmol) in THF (50 mL). Then THF (50 mL) was added. A
solution of PPhCl2 (8.136 g, 45.45 mmol) in THF (160 mL) was cooled to –78 °C. The Grignard
solution was added dropwise with stirring over 2 h. The mixture was allowed to warm to room
temperature overnight. It was then cooled to 0 °C, and LiAlH4 (2.073 g, 54.54 mmol) was carefully
added with vigorous stirring (evolving gas). After 30 min, freshly degassed aqueous NaOH (2 N, 10
mL) was added. After 1 h, the organic phase was filtered through a plug of silica gel topped with a
plug of Celite®. The plugs were rinsed with THF (50 mL). The solvent was removed from the
3 Preparation of Rhenium-Containing Phosphines 115
filtrates by oil pump vacuum. The residue was distilled (Kugelrohr, 6.4 × 10–3 mbar, 196-200 °C)
to give 29d as a viscous colorless liquid (3.874 g, 16.42 mmol, 35%).
1H NMR (400 MHz, CDCl3, δ in ppm): aryl-H at 8.06 (d, J = 9.0 Hz, 1H), 7.85-7.78 (m,
3H), 7.57-7.50 (m, 5H), 7.37-7.33 (m, 3H); 5.43 (br s, PH, 1H); 13C{1H} NMR (101 MHz, CDCl3,
δ in ppm):93 134.6, 134.5, 134.2, 134.0, 133.9, 133.8, 133.2, 133.2, 133.0, 132.0, 131.9, 130.6,
130.5, 128.5, 128.5, 128.4, 128.0, 127.9, 127.7, 127.6, 126.5, 126.3; 31P{1H} NMR (162 MHz,
CDCl3, δ in ppm): –39.6 (s).
116 3 Preparation of Rhenium-Containing Phosphines
4
Recycling of Fluorous Phosphines from Organocatalytic Reactions
4.1 Introduction
4.1.1 Introduction of Fluorous Concepts and Recycling
The "art" of fluorous syntheses and the attendant applications are quite new developments
within the chemistry community. The story began in 1994 when Horváth and Rábai introduced the
adjective fluorous within a seminal publication in Science.96 In the same edition a perspective was
given by Gladysz concerning the future of fluorous catalysis.97 Many applications were
subsequently developed98 so that it is today an accepted subarea of organic chemistry. Over the
intervening years, the following general definition of fluorous has evolved:98,99 "of, relating to, or
having the characteristics of highly fluorinated saturated organic materials, molecules or molecular
fragments." In other words, compounds with aliphatic moieties that are fully or to a high degree
fluorinated are fluorous.
The most important property of fluorous solvents is the contrasting solubility or miscibility
behavior, compared to their non-fluorinated organic counterparts. All chemists know the fact that
water is immiscible with many organic solvents. Within many workup procedures this effect is
exploited to separate the product from side products. Accordingly, the organic phase is extracted
with a water solution and a simple liquid/liquid phase separation provides a preliminary purified
compound. These two liquid phases are so called orthogonal phases.
Also fluorous phases are orthogonal to many other solvents. This includes the immiscibility
with a water phase as well as with nearly all organic solvent phases. The orthogonality of aqueous,
organic, and fluorous solvents is illustrated schematically in Figure 4.1.
118 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Figure 4.1. Common orthogonal liquid phases.
It is shown above that in principle, fluorous solvents are hydrophobic and lipophobic.
However, there are organic solvents that are miscible with fluorous solvents - depending on the
temperature. In other words, the orthogonal phases are immiscible at room temperature and after
heating the phase boundary disappears and the phases become miscible. Such behavior is known as
thermomorphic. This attribute was exploited by Horváth and Rábai to effect homogenous catalysis
followed by liquid/liquid biphase workup in their original publication. The schematic strategies for
catalyst recycling that were applied are illustrated on the following pages. The first method, termed
Concept A, is shown in Figure 4.2.
fluorous
organic aqueous
fluorophobic
hydr
opho
biclipophobic
4 Recycling of Phosphine Catalysts by Fluorous Concepts 119
Figure 4.2. Concept A: Recycling of a fluorous catalyst from biphasic liquid/liquid systems. I:
lower temperature, biphasic system, catalyst dissolved in phase B, substrate dissolved in phase A. II:
higher temperature, monophasic system, catalyst and substrate dissolved in the mixed phase AB. III:
lower temperature, biphasic system, catalyst dissolved in phase B, product dissolved in phase A.
As organic substrates are usually highly soluble in organic solvents (phase A), the catalyst
should be highly soluble in the orthogonal phase; in this case the fluorous one (phase B). Therefore,
a fluorous catalyst is employed. When the reaction mixture is heated the two liquid phases give one
homogenous phase (phase AB) and the catalyst is mixed with the substrates. After a reaction is
completed, the mixture is cooled and the orthogonal solvents segregate from each other. The
fluorous and organic phases can easily be separated. Ideally, the fluorous catalyst should be found
exclusively in the fluorous phase, while the organic reaction products should be found in the
organic phase. The fluorous catalyst can easily be recycled; a fresh charge of phase A and reactants
are simply added.
The question as to how a catalyst becomes fluorous is more or less answered above. There
heatingphase A
phase B
phase ABphase A
phase B
from phase A:
product separation
+
workup
I II III
: fluorous catalyst
: organic substrate
: organic product
cooling
phase A: organic solvent
phase B: fluorous solvent
recycle catalyst dissolved in phase B
Catalyst Recycling: Concept A
120 4 Recycling of Phosphine Catalysts by Fluorous Concepts
only has to be sufficient fluorous moieties remote from the catalytically active center. In general,
the more fluorous moieties a molecule contains, the higher the partition coefficient (relative
solubilities) between two orthogonal phases. At the same time, the longer the fluorous moieties, the
lower the absolute solubilities in all solvents.
To integrate fluorous moieties, perfluoralkyl chains are usually introduced. This method is
known as the "ponytail" concept and was also introduced in the 1994 paper.96 In practice these
ponytails usually have the formula (CH2)m(CF2)n-1CF3, which is often abbreviated as (CH2)mRfn.
The methylene chain (CH2)m can be viewed as a spacer or insulator between the
catalytically active part of the molecule and the perfluoralkyl chain Rfn. This is necessary as the
fluorine atoms are very electron withdrawing. As for all catalytic reactions, the electron density at
the active center plays a very important role; usually the electron withdrawing effect has to be
modulated by such methylene spacers. If this is achieved, the active center is decoupled from the
electron withdrawing effect, and there is furthermore the possibility to modulate the solubility of the
catalyst in organic phases. Of course, it makes no sense to synthesize a catalyst that is so fluorous,
that it is insoluble, even in the mixed phase AB. The more methylene groups are in the molecule,
the better the solubility in organic solvents.
When there is a perfect balance between the methylene spacer and Rfn moiety, the fluorous
solvent can be omitted.100,101 This second variant, Concept B, is depicted in Figure 4.3. Under
optimal conditions the catalyst remains undissolved when the substrate solution is added (I). After
heating the mixture the catalyst dissolves and the reaction takes place (II). After full conversion, it
is cooled again and the catalyst precipitates from the product solution (III). The solid catalyst and
the product solution are now separated and the catalyst is added to fresh substrate solution for the
next cycle.
4 Recycling of Phosphine Catalysts by Fluorous Concepts 121
Figure 4.3. Concept B: Recycling of a fluorous catalyst from biphasic solid/liquid systems. I: lower
temperature, biphasic system, undissolved solid catalyst, substrate dissolved in phase A. II: higher
temperature, monophasic system, catalyst and substrate dissolved in phase A. III: lower
temperature, biphasic system, solid undissolved catalyst, product dissolved in phase A.
Comparing Concept B with Concept A, there is the advantage that a fluorous solvent is no
longer needed. This saves money as well as natural resources, which is even more important.
However, in practice it often happens that the catalyst does not fully precipitate while cooling. To
improve this precipitation behavior, a fluoropolymer can be added as a solid support.
Fluorous compounds are fluorophilic and this also applies to fluorous solid supports.100-102
This leads to the third variant, Concept C, which is depicted in Figure 4.4.
heating
from phase A:
product separation
+
workup
: fluorous catalyst
: organic substrate
: organic product
cooling
Catalyst Recycling: Concept B
phase A phase A phase A
I IIIII
recycle solid catalyst
phase A: organic solvent
122 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Figure 4.4. Concept C: Recycling of a fluorous catalyst from biphasic solid/liquid systems with
fluoropolymer support. I: lower temperature, biphasic system, undissolved solid catalyst which is
coated on a fluoropolymer, substrate dissolved in phase A. II: higher temperature, monophasic
system, catalyst and substrate dissolved in phase A, uncoated solid fluoropolymer. III: lower
temperature, biphasic system, undissolved solid catalyst which is again coated on the fluoropolymer,
product dissolved in phase A.
Here the catalyst is coated onto a fluoropolymer which could be Teflon® tape, as it is
available in every building center. Of course, every other perfluoralkane based polymer can be
employed, for example later also a Gore-Rastex® fiber was tested. The coated polymer is then
heated with the substrate mixture and the catalyst becomes soluble and diffuses into solution to
catalyze the reaction. When the reaction is complete and the mixture is cooled, the catalyst
precipitates or adsorbs onto the fluoropolymer, as both (catalyst and polymer) are fluorophilic. The
polymer can then be removed manually from the mixture and here is the second advantage of the
fluoropolymer support. With Concept B it can be challenging to remove the catalyst manually in
heating
phase A phase A phase A
I II III
cooling
recycle with catalyst coated polymer
Catalyst Recycling: Concept C
fluoropolymer
from phase A:
product separation
+
workup
: fluorous catalyst
: organic substrate
: organic product
phase A: organic solvent
4 Recycling of Phosphine Catalysts by Fluorous Concepts 123
solid form from the reaction vessel, as it precipitates over the wall of the entire vessel and the
stirring bar (which is usually a magnet coated with Teflon® polymer). Within Concept C the
catalyst can be removed from the vessel very easily by taking out the coated polymer.
4.1.2 Introduction of Fluorous Phosphines
In the years since the initial paper from Horváth and Rábai,96 a variety of processes have
been developed that exploit fluorous concepts and many of them are summarized in the "Handbook
of Fluorous Chemistry".98 Within this handbook, numerous reactions are mentioned that involve a
catalytically active metal center that is coordinated by fluorous phosphines. These fluorous
phosphines render the whole catalyst fluorophilic.
However, there are many reactions that are catalyzed by phosphines themselves without
involving an active metal center. The transformations are now termed organocatalytic reactions.8,10
In most cases the organocatalysts are not recovered, as most of this research is still in an early,
fundamental stage. But nevertheless, with a look to the future, if any reaction should become
industrially important, it would be of great advantage if the catalyst would be recoverable and
recyclable.
Therefore, the fluorous concept was applied and two aliphatic phosphines, tagged with
different chain length ponytails, were utilized for the types of intramolecular Morita Baylis Hillman
reactions described in Chapter 2. The employed phosphines are depicted in Figure 4.5.
These phosphines can be viewed in the context of three components, all necessary for
applying fluorous catalysis concepts. Of course, there must be a catalytically active center, which is
segment A. Segment B acts as a spacer that modulates the electron density at the phosphorus atom.
Especially for organocatalysis, the electron density at this atom plays a very important role as
usually the phosphine acts as a nucleophile in the initial reaction step (Schemes 1.5 and 1.8). If the
electron density is decreased by the electron withdrawing effect of the fluorine atoms, the
nucleophilicity is decreased. And by this, attenuated or dramatically lower reactivity can be
expected. Furthermore, the methylene spacers provide higher solubilities in organic solvents.
Finally, the most important part for catalyst recovery is segment C, as this is the fluorophilic part,
124 4 Recycling of Phosphine Catalysts by Fluorous Concepts
acting as described above.
Figure 4.5. General structure for the phosphines P((CH2)3Rfn)3 (30, a: n = 6, b: n = 8) that were
employed for the intramolecular Morita Baylis Hillman and related reactions. Segment A:
catalytically active center. Segment B: methylene spacers for modulation of electronic and
solubility properties. Segment C: fluorous tag, also for modulation of solubility properties.
The preparation of these phosphines was done according to literature procedures. All
phosphines are well known and have been previously characterized in detail.103-106 In general there
are mainly two ways to synthesize fluorous phosphines 30, as depicted in Scheme 4.1. The first is
the three-fold addition of alkenes like 31 to PH3 by a radical mechanism.103 As PH3 is a very toxic
and pyrophoric gas,107 there is much interest to avoid its use. Therefore, PH3 can be substituted by
the primary phosphines 32, which can be obtained by a two step Arbuzov sequence.104 As 32 is
also the first non-radical intermediate during the procedure with PH3, the same conditions but with
a different stoichiometry can be employed. The tertiary fluorous phosphines are obtained as a liquid
(30a) or as a solid (30b), and their application as organocatalysts are described in the following
sections.
P
(CH2)m
(CH2)m
(CH2)m
(CF2)n-1
CF3
(CF2)n-1
CF3
(CF2)n-1
CF3
30a: m = 3, n = 6
30b: m = 3, n = 8
segment A
segment Bsegment C
P((CH2)m(CF2)n-1CF3)3
4 Recycling of Phosphine Catalysts by Fluorous Concepts 125
Scheme 4.1. General procedure for the syntheses of tertiary fluorous phosphines 30a,b.
Rfn
Rfn
PH3
Rfn
H2P
3
P
3
31
32
330
a: n = 6
b: n = 8
initiator
126 4 Recycling of Phosphine Catalysts by Fluorous Concepts
4.2 Results
4.2.1 Catalytic Reactions with P((CH2)3Rf6)3 (30a)
From mechanistic considerations, it would seem that in principle every phosphine can be
employed for the Morita Baylis Hillman reaction (Scheme 1.8). While the phosphorus center is
needed for catalytic activity, the fluorous pony tail opens up many possibilities for recycling. To
apply Concepts B and C, data on catalyst solubilities are essential. Thus, preliminary tests were
conducted with the less heavily fluorinated phosphine P((CH2)3Rf6)3 (30a).106 This phosphine is a
liquid at room temperature and solidifies at ca. –45 °C upon cooling. The solubility in different
degassed solvents was qualitatively tested by "naked eye" and some results are summarized in
Table 4.1.
Table 4.1. Qualitative solubility of 0.050 g 30a in several solvents at different temperatures.
Solvent V (mL) T (°C) soluble
CH2Cl2 3.0 20 yes
CH3CN 2.3 50 yes
THF 0.3 20 yes
CH2ClCH2Cl 2.5 70 no
CDCl3 0.5 20 yes
The solubility of 30a was a strong function of solvent. At room temperature, it was highly
soluble in THF as well as in CDCl3. It was also soluble in CH2Cl2, but much more solvent volume
was needed. In the cases of CH3CN and CH2ClCH2Cl, several mL of solvent as well as elevated
temperatures were needed. In the latter, complete dissolution could not be achieved. As mentioned
before, usually very polar and/or protic solvents are used for the Morita Baylis Hillman reactions.
4 Recycling of Phosphine Catalysts by Fluorous Concepts 127
As CH3CN is very polar (Figure 2.15) and the fluorous phosphine 30a seemed to be insoluble at
room temperature but dissolved with heating, this phosphine was tested as a catalyst. For this, 10
mol% of 30a was employed under conditions where complete dissolution would give a 0.00500 M
CH3CN solution. The substrate C1Ph (0.0500 M) was first used, as it was known to be a smoothly
reacting one (Scheme 4.2).
Scheme 4.2. Intramolecular Morita Baylis Hillman reaction of C1Ph catalyzed by 10 mol% of
fluorous phosphine 30a.
The reaction was conducted at 60-64 °C (sand bath), which gave a homogeneous solution.
After 2.5 h, the reaction mixture was cooled to room temperature and the CH3CN phase was
carefully removed from the vial. Subsequently the solvent was replaced by CDCl3 and the mixture
was assayed by 1H NMR. The product yield was calculated from the integrals of the signals at δ =
6.68 (CH=C) and 5.30 (CHOH) ppm versus all phenyl signals. The yield was 86%, and some
unreacted C1Ph could be detected.
Under the same conditions a series of three catalytic cycles was conducted. Concept B
(Figure 4.3) was utilized and of course, the catalyst did not precipitate as a solid but as a liquid. The
yields are summarized in Table 4.2.
Details of the recycling sequence were as follows. After 2.5 h at 60-64 °C, the sample was
cooled to –30 °C for several hours to facilitate phase separation of the catalyst. The catalyst was not
detectable to the "naked eye", as it was a very small amount of a liquid that was distributed over the
whole wall of the glass vial or possibly the stirring bar. The CH3CN phase was removed by a
pipette and examined by 1H NMR as above. For the next cycle, new substrate solution was added
and the sample was heated again.
CHOPh
OO
Ph OH10 mol%
P((CH2)3Rf6)3 (30a)
C1Ph C1Phprod
CH3CN
60-64 °C
128 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Table 4.2. Reaction of C1Ph with 10 mol% of 30a in CH3CN at 60-64 °C for 2.5 h. Product yield
as a function of cycle (Concept B).
However, the product yields decreased with every cycle and furthermore there was always
unconverted substrate. To get better results, the reaction setup was improved in two sets of
experiments. In the first, the reaction time was increased from 2.5 h to 6 h. In the second, Teflon®
shavings were furthermore added (Concept C, Figure 4.4). The results for each are given in Table
4.3.
Table 4.3. Reaction of C1Ph with 10 mol% of 30a in CH3CN at 60-64 °C for 6 h. Product yield as
a function of cycle. Catalyst was recovered by precipitation (Concept B) or adsorption onto Teflon®
shavings (Concept C).
Product Yield (%)
Cycle Concept B Concept C
1 96 92
2 78 80
3 74 20
It is easily seen that the yield of product increased with longer reaction times (96% for
Concept B vs. 86% in Table 4.2). Unfortunately, the next two cycles showed a yield decrease. Also
Cycle Product Yield (%)
1 86
2 78
3 73
4 Recycling of Phosphine Catalysts by Fluorous Concepts 129
the addition of Teflon® shavings gave no improvements. Furthermore, there was a big drop in
reactivity after the second cycle. Given these recycling problems, a different phosphine catalyst was
sought.
4.2.2 Catalytic Reactions with P((CH2)3Rf8)3 (30b)
The longer the fluorous chains, the less soluble are the above mentioned phosphines. As it
seemed likely that the phosphine 30a with Rf6 moieties was partially removed with the product
solution during the recycling procedures, homologs with longer fluorous chains were targeted. The
most logical candidate was P((CH2)3Rf8)3 (30b), which in comparison to 30a is not a liquid but a
white solid at room temperature. The solubility of 30b in CH3CN was tested by the "naked eye". It
seemed that it was not soluble at room temperature, while sufficient solid disappeared at elevated
temperatures (> 50 °C) to give 0.00500 M solutions. After cooling to –30 °C for several hours, the
phosphine was found by "naked eye" as a white, solid precipitate on the glass wall.
4.2.2.1 Reactions with Substrate C1Ph
A first series was carried out with substrate C1Ph in the glove box under argon atmosphere
(Scheme 4.2, but with 30b as catalyst). A Schlenk flask was charged with 10 mol% of 30b, and a
CH3CN solution of C1Ph was added. In all cases, the final substrate concentration was 0.0500 M.
The mixture was stirred at 60-64 °C for 6 h and afterwards stored at –30 °C for several hours. The
CH3CN phase was then removed with a pipette and the yield of C1Phprod was determined by 1H
NMR as described in section 4.2.1. The vial was recharged with substrate solution and the next
cycle was run. The yields are summarized in Table 4.4.
From these data, it is clear that 30b is recovered much more efficiently than 30a. Even with
the fifth cycle a yield comparable to the first cycle was obtained.
130 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Table 4.4. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C for ca. 6 h. Yield as a
function of cycle (Concept B).
Cycle Yield (%)
1 70
2 76
3 69
4 73
5 68
As a next step, the reaction of substrate C1Ph with 10 mol% of catalyst 30b was monitored
with time. To confirm the accuracy, two different methods were employed. In the first, aliquots
were assayed by 1H NMR. The yields were calculated as described in section 4.2.1. In the second,
aliquots of a smaller scale independent run were assayed by HPLC. The HPLC injections were of
constant volume, and the detector adsorption of the final data point was normalized to the yield
determined by 1H NMR. In any case, the results are presented in Figure 4.6.
The green trace (■) gives the reaction profile obtained by NMR. The red one (▲) shows the
data from HPLC measurements. Both reactions ended with a yield of 85% after 24 h.
4 Recycling of Phosphine Catalysts by Fluorous Concepts 131
Figure 4.6. Reactions of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots of one were
assayed by 1H NMR (■), and the other (smaller scale) by HPLC (▲).
4.2.2.2 Reactions with Substrates C1S(i-Pr) and C1S(i-Pr)2
The catalyst 30b was tested with two more substrates. The reactions are shown in Scheme
4.3. The first leads to a Morita Baylis Hillman cyclization and the second to a Rauhut Currier
cyclization.
Reactions were monitored by 1H NMR (■) and GC (▲). The latter were conducted
independently, on smaller scales, and in the presence of an internal standard. As can be seen in
Figures 4.7 and 4.8, the larger scale reactions (conducted in Schlenk flasks) were somewhat slower.
NMR yields were calculated from the integration of the total i-Pr methine 1H NMR signal versus
characteristic products peaks. The final NMR and GC yields were in good agreement (Yields:
NMR/GC; C1S(i-Pr)prod, 95/92%; C1S(i-Pr)2prod, 75/72%).
0 5 10 15 200
20
40
60
80
( ) NMR
( ) HPLC
Yie
ld (%
)
Time (h)
132 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Scheme 4.3. Cyclizations of substrates C1S(i-Pr) and C1S(i-Pr)2 catalyzed by 10 mol% of 30b.
Figure 4.7. Reactions of C1S(i-Pr) with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots of one
were assayed by 1H NMR (■), and the other (smaller scale) by GC (▲).
0 10 20 30 400
20
40
60
80
( ) NMR
( ) GCPro
duct
Yie
ld (%
)
Time (h)
CHOi-PrS
OO
i-PrSOH10 mol%
P((CH2)3Rf8)3 (30b)
C1S(i-Pr)prod
CH3CN
60-64 °C
C1S(i-Pr)
S(i-Pr)
O
i-PrS
O
O
i-PrS O
S(i-Pr)
C1S(i-Pr)2prodC1S(i-Pr)2
10 mol% 30b
CH3CN
60-64 °C
4 Recycling of Phosphine Catalysts by Fluorous Concepts 133
Figure 4.8. Reactions of C1S(i-Pr)2 with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots of one
were assayed by 1H NMR (■), and the other (smaller scale) by GC (▲).
4.2.2.3 Reactions with Substrate C2(p-Tol)
To extend the series of substrates, C2(p-Tol) was tested next. Cyclization was catalyzed as
depicted in Scheme 4.4. Due to the slower rate, a slightly higher temperature was employed. This
afforded complete reaction after 72 h.
Scheme 4.4. Cyclization of substrate C2(p-Tol) catalyzed by 10 mol% of 30b.
0 10 20 30 400
20
40
60
80
( ) NMR
( ) GCPro
duct
Yie
ld (%
)
Time (h)
p-Tol
O
C2(p-Tol)prodC2(p-Tol)
CHO
p-Tol
O
OH
10 mol% 30b
CH3CN
70-72 °C
134 4 Recycling of Phosphine Catalysts by Fluorous Concepts
In this case the reaction was exclusively monitored by HPLC. Since no internal standard
was present, the maximum yield (79%) was calculated from the total aryl 1H NMR signals versus
characteristic product signals. Unassigned side products were also apparent. The results are shown
in Figure 4.9.
Figure 4.9. Reaction of C2(p-Tol) with 10 mol% of 30b in CH3CN at 70-72 °C. Aliquots were
assayed by HPLC.
The reaction was again carried out in a small vial in the glove box under the same conditions
as above. The reaction time (72 h) was somewhat longer than those of C1Ph (24 h), C1S(i-Pr), and
C1S(i-Pr)2 (48 h each) at lower temperatures (60-64 °C).
4.2.2.4 Preparative Experiments
Substrates C1S(i-Pr) and C2(p-Tol) were subjected to catalytic reactions on preparative
scales (0.120 and 0.150 g). The same reaction conditions and times as above were employed. The
crude products were purified by column chromatography. The yield of C1S(i-Pr)prod was lower
0 10 20 30 40 50 60 700
20
40
60
80
( ) HPLCPro
duct
Yie
ld (%
)
Time (h)
4 Recycling of Phosphine Catalysts by Fluorous Concepts 135
than the NMR yield (81% vs. 95%). However, it was already shown in Chapter 2 that always some
product is lost by silica gel chromatography. The yield of C2(p-Tol)prod was also lower (67% vs.
79%).
4.2.3 Recycling of 30b by Precipitation
As noted above, a fluorous catalyst can be recycled by cooling the mixture to lower
temperatures after the reaction is finished. The general procedure was depicted in Figure 4.3, and
termed Concept B.
Every experiment began as a biphasic mixture of 30b, the substrate, and the CH3CN solvent.
At room temperature the substrate was dissolved, but the catalyst was apparently not (I, Figure 4.3).
After heating with stirring, the catalyst dissolved and the reaction took place (II, Figure 4.3). From
this hot mixture aliquots (0.010 mL) were periodically taken by a syringe. These aliquots were
assayed by HPLC or GC. These data were used for the reaction profiles. After the reaction was
finished, the mixture was cooled and the catalyst precipitated (III, Figure 4.3). Recycling was
facilitated when the samples were cooled slowly (3 h at room temperature followed by –30 °C
overnight). The resulting precipitate was coarser. The product solution was carefully removed by
syringe and assayed by 1H NMR. The remaining solid catalyst was washed with a small amount of
CH3CN at room temperature and dried by oil pump vacuum. The vial with the recycled catalyst was
then recharged with a CH3CN solution of substrate. With each cycle less substrate solution was
added to compensate for the catalyst removed in aliquots (i.e., to maintain the 10 mol% catalyst
loading). The next cycle was conducted similarly with respect to monitoring and recycling. After
the last cycle, the recovered catalyst was dissolved in PhCF3, and 1H NMR and 31P{1H} NMR
spectra were recorded to give information about the resting state.
4.2.3.1 Recycling from Reactions with C1Ph
The first set of recycling reactions was carried out with substrate C1Ph and 10 mol% of 30b
in CH3CN. The same procedure as given in Figure 4.3 was applied at a reaction temperature of 60-
136 4 Recycling of Phosphine Catalysts by Fluorous Concepts
64 °C. All reactions were carried out in the glove box under an argon atmosphere in a sand bath.
Photographs of a representative sequence are given in Figure 4.10.
A B C D
Figure 4.10. A: Reaction mixture prior to heating. B: Reaction mixture after cooling. C: Catalyst
after separation from the reaction mixture. D: Recovered catalyst.
The vial with the starting mixture (A) shows a clear solution with a solid residue at the
bottom that is difficult to see. The phosphine 30b is a white solid but becomes somewhat
transparent upon the addition of CH3CN. After the reaction finished a red solution was obtained (B).
As described in section 2.5.4, the product C1Phprod should be a slightly yellow oil. The color was
attributed to unassigned side products. In particular, dehydration involving the secondary alcohol
would give a conjugated cyclic diene that could be prone to further condensations. After removing
the product solution, the catalyst was obtained as a slightly transparent solid (C). This solid was
washed and dried under oil pump vacuum and a white solid was again obtained (D). This was
employed for the next reaction cycle. The results of five cycles are charted in Figure 4.11.
There was nearly no loss in reactivity within the first three cycles. The yields after the
reactions ended were comparable (85%, 83%, and 84%). Also the reaction rates were similar.
Unfortunately, with the fourth and fifth cycle the reactivity decreased a little and lower yields (both
78%) were obtained.
4 Recycling of Phosphine Catalysts by Fluorous Concepts 137
Figure 4.11. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots were
assayed by HPLC.
Formation of phosphine oxide. After the fifth cycle the recovered catalyst was assayed by
1H NMR and 31P{1H} NMR. The spectra showed that 24% of the phosphine was oxidized (ca. 2%
in a parallel experiment after one cycle). The phosphine oxide O=P((CH2)3Rf8)3 (O=30b) has been
reported in the literature106 and was for comparison purposes synthesized by the oxidation of 30b
with H2O2.103
4.2.3.2 Recycling from Reactions with C1S(i-Pr), C1S(i-Pr)2, and C2(p-Tol)
Following the procedures shown above for C1Ph, three other substrates were similarly
studied, with catalyst recycling by precipitation at low temperatures (Concept B). For substrate
C2(p-Tol), higher reaction temperatures were employed (60-64 °C vs. 70-72 °C), as the formation
of six-membered ring systems is slower than five-membered analogs. The data are presented in
Figures 4.12-4.14.
0 5 10 15 200
20
40
60
80
( ) Cycle 1 ( ) Cycle 2 ( ) Cycle 3 ( ) Cycle 4 ( ) Cycle 5P
rodu
ct Y
ield
(%)
Time (h)
138 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Figure 4.12. Reaction of C1S(i-Pr) with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots were
assayed by GC.
Figure 4.13. Reaction of C1S(i-Pr)2 with 10 mol% of 30b in CH3CN at 60-64 °C. Aliquots were
assayed by GC.
0 10 20 30 400
20
40
60
80
( ) Cycle 1 ( ) Cycle 2 ( ) Cycle 3P
rodu
ct Y
ield
(%)
Time (h)
0 10 20 30 400
20
40
60
80
( ) Cycle 1 ( ) Cycle 2 ( ) Cycle 3P
rodu
ct Y
ield
(%)
Time (h)
4 Recycling of Phosphine Catalysts by Fluorous Concepts 139
Figure 4.14. Reaction of C2(p-Tol) with 10 mol% of 30b in CH3CN at 70-72 °C. Aliquots were
assayed by HPLC.
In each case very good results were obtained. The three cycles in Figure 4.12 gave yields of
95% to 96%, which can be considered within experimental error. However, the product mixtures
were yellow, indicating the formation of unassigned side products. Indeed, yellow to orange
product mixtures were obtained for all substrates.
The third cycle with the substrate C1S(i-Pr)2 (Figure 4.13) appeared to show a little loss of
activity. However, the yield after 48 h was comparable to that in the first cycle (with 73% vs. 72%).
The first two cycles with C2Tol (Figure 4.14) are virtually identical. There is some scatter in
the third cycle, suggestive of artifactual data points. The NMR yields ranged from 79% to 81%.
Formation of phosphine oxide. As in the section before, the recovered catalysts were
examined by 1H NMR and 31P{1H} NMR. In every experiment phosphine oxide O=30b was found,
indicating the presence of oxygen or other oxidation mechanisms that were not investigated. The
data are summarized in Table 4.5.
0 10 20 30 40 50 60 700
20
40
60
80
( ) Cycle 1 ( ) Cycle 2 ( ) Cycle 3P
rodu
ct Y
ield
(%)
Time (h)
140 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Table 4.5. Phosphine oxide O=30b formed during the reactions in Figures 4.12-4.14, as assayed by
1H NMR and 31P{1H} NMR spectra. The result after the first cycle is from an independent reaction.
4.2.4 Recycling of 30b with Fluoropolymer Supports
As shown above, catalyst precipitation can be a very useful recycling method. However, it
can be problematic when the catalyst precipitates on the walls of the reaction vessel, especially if it
has to be transferred to another vessel, and if small quantities are involved. Fortunately, these
phosphines are very fluorophilic and can easily be adsorbed to fluoropolymers from solutions by
cooling from an appropriate solvent (Concept C, Figure 4.4). This method was applied to recover
catalyst from reactions with C1Ph.
In contrast to Concept C as it is depicted above, the first cycle did not involve 30b coated
onto the fluoropolymer; rather they were added separately. Subsequently, C1Ph and the solvent
were added to the vial at room temperature in a glove box. The mixture was then treated under
conditions identical to the reactions without fluoropolymer support. This meant heating to 60-64 °C
with stirring and regularly taking aliquots that were assayed by HPLC. After the reaction time (24
h), the vial was kept at room temperature for 3 h and then at –30 °C overnight. Then the product
solution was removed and the fluoropolymer, now coated with the catalyst, was washed. After
drying the coated polymer by oil pump vacuum, it was transferred to another vial that had been
precharged with substrate solution, and the next cycle begun.
Reaction with substrate
C1S(i-Pr) C1S(i-Pr)2 C2(p-Tol)
After cycle 1 4 / 96 2 / 98 7 / 93 O=30b / 30b
After cycle 3 23 / 77 12 / 88 12 / 88
4 Recycling of Phosphine Catalysts by Fluorous Concepts 141
4.2.4.1 Recycling with the Support of Teflon® Tape
The first experiments employed commercial Teflon® tape, as it is available in every
building center for sealing water pipe connections. It is a very thin material and therefore offers a
large surface for potential adsorption per unit weight. The mixture was treated under the same
conditions as shown above (Figure 4.6). Some photographs from different stages are depicted in
Figure 4.15.
A B C
Figure 4.15. Photographs from cycle 1 of Figure 4.16. A: Reaction mixture prior to heating. B:
Reaction mixture after cooling to –30 °C. C: Washed and dried Teflon® tape, coated with catalyst.
The Teflon® tape is apparent as the white solid in picture A. Again the catalyst can hardly
be seen as it is quite transparent. Picture B shows the mixture, from which the tape was removed
after the completion of the reaction and cooling. Again a red coloring occurred that was similar to
that without fluoropolymer support. After the Teflon® tape was removed from the reaction mixture
it was washed with CH3CN. Neither in the reaction mixture nor in the washing solvent could solid
catalyst be seen by the "naked eye". This suggests that the tape is firmly coated with the catalyst.
This tape shown in picture C was directly employed for the next cycle. During each cycle, aliquots
were assayed by HPLC, and the results are shown in Figure 4.16.
142 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Figure 4.16. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C in the presence of
Teflon® tape. Aliquots were assayed by HPLC.
As depicted, five cycles were conducted, showing a gradual loss in activity of the catalyst.
The fifth cycle resulted in a yield of 66%, as opposed to the 82% reached in the first one. There was
substrate left in the fifth cycle, and also some side product. With the first cycle, the substrate C1Ph
was fully consumed. However, the first cycle also started quite slowly, as illustrated by the green
trace. This was assumed to be due to poor convection in the vial when the experiment was started.
The catalyst was a solid at the bottom of the vial and the tape, which is quite voluminous, was
stacked above (A, Figure 4.15). After warming the mixture, the catalyst dissolved but the tape
hindered the generation of a homogenous catalyst solution. Additionally, though two stirring bars
were employed, stirring was very ineffective as the stirring bars were hindered from moving by the
Teflon® tape. In subsequent cycles the catalyst dissolved directly from the coated tape and therefore
gave a homogenous solution right from the start. Accordingly, these run faster at the beginning.
0 5 10 15 200
20
40
60
80
( ) Cycle 1 ( ) Cycle 2 ( ) Cycle 3 ( ) Cycle 4 ( ) Cycle 5
Pro
duct
Yie
ld (%
)
Time (h)
4 Recycling of Phosphine Catalysts by Fluorous Concepts 143
4.2.4.2 Recycling with the Support of Gore-Rastex® Fiber
Another attractive option for the fluoropolymer support would be Gore-Rastex® fiber. This
material can be viewed as Teflon® that is converted by special techniques to give a fiber that is very
porous. This affords a material with many small holes that are permeable for small molecules, such
as a water. But it is impermeable to large aggregates of molecules, such as water drops. The fiber on
one side and the porous material on the other offer a very large surface for the adsorption of
fluorous catalysts such as 30b. It was assumed that this could improve the capacity for recovery by
the fluoropolymer support, according to Concept C.
All conditions were analogous to the experiment with Teflon® tape in section 4.2.4.1.
Pictures were taken directly from the experiment from time to time and are depicted in Figure 4.17.
The observations were similar to those in the experiments with the Teflon® tape and are therefore
not repeated. The results of monitoring the reaction of C1Ph with 10 mol% of catalyst are shown in
Figure 4.18.
This figure shows that good recovery of the catalyst was obtained with Gore-Rastex® fiber
support. There is a definite improvement over the results with Teflon® tape. The biggest loss in
catalyst activity was found after the fourth cycle. The catalyst recovery in the cycles before was
quite reliable. Anyway, within any cycle good yields were obtained (82%, 80%, 81%, 78%, and
74%; cycle 1-5).
At low conversion, the first cycle appeared "normal", without the quasi induction period
behavior in Figure 4.16. It was assumed that the fiber allows near-normal fluid convection in the
sample.
144 4 Recycling of Phosphine Catalysts by Fluorous Concepts
A B C
Figure 4.17. Photographs from cycle 1 of Figure 4.18. A: Reaction mixture prior to heating. B:
Reaction mixture after cooling. C: Washed and dried Gore-Rastex® fiber, coated with catalyst.
Figure 4.18. Reaction of C1Ph with 10 mol% of 30b in CH3CN at 60-64 °C in the presence of
Gore-Rastex® fiber. Aliquots were assayed by HPLC.
Formation and reactivity of phosphine oxide. As in the above sections, phosphine oxide
was also found after the reaction was finished. This was despite the fact that the reactions were
performed in the glove box. The amounts detected by 1H NMR and 31P{1H} NMR are summarized
in Table 4.6.
0 5 10 15 200
20
40
60
80
( ) Cycle 1 ( ) Cycle 2 ( ) Cycle 3 ( ) Cycle 4 ( ) Cycle 5
Pro
duct
Yie
ld (%
)
Time (h)
4 Recycling of Phosphine Catalysts by Fluorous Concepts 145
Table 4.6. Phosphine oxide O=30b formed during the reactions in Figures 4.11, 4.16, and 4.18 as
assayed by 1H NMR and 31P{1H} NMR spectra.
Recovery method
Precipitation Teflon® tape Gore-Rastex® fiber
O=30b / 30b 24 / 76 25 / 75 37 / 63
To exclude that phosphine oxide O=30b shows reactivity towards substrate C1Ph, a control
experiment was conducted. Therefore, a reaction mixture with substrate C1Ph analogous to the
precipitation experiments (section 4.2.3.1) was prepared, but with O=30b instead of 30b. This
mixture was heated to 60-64 °C with stirring. The oxide appeared to be insoluble in CH3CN under
such conditions. After 24 h, the mixture was allowed to cool and assayed by 1H NMR. No
detectable amount of C1Phprod was present. However, very low amounts of decomposition
products of C1Ph were noted.
Horváth also reported that this phosphine oxide was very poorly soluble.106 In fact, while
conducting the recycling experiments in section 4.2.3, after several cycles small amounts of
precipitates were noted in the warm reaction mixtures. These precipitates were never isolated but
are thought to be O=30b.
4.2.5 Thermomorphic Behavior of 30b in CH3CN
The "naked eye" determinations mentioned above indicated that 30b seems to be insoluble
in CH3CN at room temperature, but to have appreciable solubility at elevated temperature. As a
check, the solubility of 30b in CH3CN was assayed by 19F NMR at different temperatures, using
Na+ BArf– (0.00188 M) as an internal standard.108 The relative areas of the CF3 signals were
integrated. The influence upon the solubility of 30b was presumed to be minimal.
The results are shown in Figure 4.19. Hysteresis was evident. The heating curve (red trace,
146 4 Recycling of Phosphine Catalysts by Fluorous Concepts
■), which was extended to just under the boiling point of CH3CN (82 °C) showed the expected
marked solubility increase. The probe was equilibrated for 5 min at every temperature, and the 384
pulses were collected. Although no 30b was detected at 25 °C and 35 °C, the signals could have
been obscured by the noise. Some very low solubility is likely.
The cooling curve (blue trace, ▲) gave higher solubilities. However, it is likely that the
hysteresis could decrease, if the samples had been equilibrated 1 h at each temperature instead of 5
min. Nonetheless, the maximum concentration of 30b attained in Figure 4.19 is only 0.00321 M, as
opposed to the 0.00500 M solutions used for the catalysis experiments that were by the "naked eye"
homogeneous. However, there is a possible rationale. With all of these observations, an excess of
the substrate C1Ph (0.0500 M) was present. During the first step of the catalysis mechanism
(Schemes 1.5 and 1.8), a zwitterion is generated. Although the equilibrium constant is unknown, to
the extent that catalyst-substrate or catalyst-product adducts are present, increased solubilities can
be anticipated.
Figure 4.19. Solubility of 30b in CH3CN as a function of temperature. The hysteresis curve was
determined from two independent 19F NMR experiments. Red trace (■): heating curve. Blue trace
(▲): cooling curve.
20 40 50 60 70 80 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Pho
sphi
ne C
once
ntra
tion
(mM
)
Temperature (°C)
30
4 Recycling of Phosphine Catalysts by Fluorous Concepts 147
4.3 Discussion
The above experiments demonstrate that the fluorous phosphines 30a and 30b are effective
catalysts for the Morita Baylis Hillman reaction of substrate C1Ph as well as others. Good yields of
the product C1Phprod were obtained. However, recycling of catalyst 30a was not as effective as for
catalyst 30b (Tables 4.2 and 4.3 vs. Table 4.4, Figures 4.11, 4.16, and 4.18). It is presumed that 30a
has a higher solubility at room temperature in CH3CN than 30b. Thus, the catalyst leaching during
the workup procedure was higher for 30a.
While a reaction mixture containing catalyst 30a and substrate C1Ph gave a 96% yield of
C1Phprod after 6 h (Table 4.3), an analogous mixture but with catalyst 30b gave only a 70% yield
after the same time (Table 4.4). Hence, although a detailed reaction profile for catalyst 30a is not
available, a higher reactivity in the former mixture can be supposed. As the structures of the
catalysts only vary in two CF2 groups (per alkyl chain) this observation might be attributed to
solubility effects.
Prior to this study, three catalyst systems had been shown to be recyclable using
fluoropolymer supports. The first involved the use of Teflon® shavings as a mechanical support for
recycling the fluorous phosphine P((CH2)2Rf8)3, which is an effective Lewis base or nucleophilic
catalyst for additions of alcohols to propiolate esters (n-octane, 65 °C).100,101 The second
demonstrated the use of Teflon® tape as an adsorbent for recycling a rhodium hydrosilylation
catalyst derived from ClRh[P((CH2)2Rf6)3]3 (dibutyl ether, 55 °C).102 The third and newest study
employed Teflon® tape for recycling of a fluorous phase transfer catalyst
(Rf8(CH2)2)(Rf6(CH2)2)3P+ X– (X = Cl/I) acting in a biphasic liquid/liquid system (H2O/PhCF3,
100 °C).109 This successful series of fluoropolymer supported catalyst recycling has been
significantly extended with this study. Furthermore, the enhanced recycling capabilities of Gore-
Rastex® fiber compared to Teflon® tape has been shown. However, precipitation of the catalyst in
the absence of a support remains superior. It has been demonstrated that catalysts can be identified
that are both effective for the Morita Baylis Hillman reaction - a process very sensitive to the exact
148 4 Recycling of Phosphine Catalysts by Fluorous Concepts
conditions - and on the other hand can be recycled very efficiently. In most studies involving Morita
Baylis Hillman reactions no catalyst recycling is reported.
However, to the author's knowledge, to date there is only one publication, wherein the
catalyst from a Morita Baylis Hillman reaction is recycled from a liquid/liquid biphase system
(Concept A). It deserves emphasis, as the catalyst was recovered with nearly no leaching. The
catalytic system, developed by Yi, is depicted in Scheme 4.5.30 It consists of two parts. One is the
ytterbium salt, which acts as an electrophilic activator for carbonyl groups. The other is a fluorous
pyridine derivative. With this system, five reaction cycles were conducted with no loss in activity.
Although this catalyst system nicely complements the one used in this study, and points the way to
even more future developments, it does suffer in the need for a fluorous solvent. Furthermore, this
system is totally different to the one used in this work.
Scheme 4.5. Reaction sequence with recyclable catalyst system, reported by Yi.
N
R
R = CH(CH2CH2Rf8)2
OCH3
O
OCH3
O
Ph
OH
PhCHO +0.5 mol%
3 mol%
Yb(OSO2Rf8)3
4 Recycling of Phosphine Catalysts by Fluorous Concepts 149
4.4 Experimental
4.4.1 General Data
NMR spectra were recorded on Bruker 300 MHz or 400 MHz FT spectrometers at ambient
probe temperature (27 °C, unless noted) and referenced to the residual solvent signal (1H: CHCl3,
7.26 ppm; 13C{1H}: CDCl3, 77.0 ppm). IR spectra were recorded on an ASI React IR®-1000
spectrometer. HPLC analyses were conducted with a ThermoQuest instrument package
(pump/autosampler/detector P4000/AS3000/UV6000 LP; Column: Macherey-Nagel, Nucleosil
100-5). GC analyses were conducted with a ThermoQuest TRACE GC 2000 instrument with FID
and an Optima-5-0.25 μm capillary column. The Gore-Rastex® fiber (30 den) was obtained as a
sample from Gore corporation (http://www.gore.com/en_xx/products/fabrics/sewing/ras-
tex_weave_datasheet.html).
Solvents were treated as follows: CH3CN (Grüssing), distilled from CaH2, and for catalysis
furthermore freeze-pump-thaw degassed (3 ×); PhCF3 (ABCR), freeze-pump-thaw degassed (3 ×);
n-Bu2O (Acros, 99%), freeze-pump-thaw degassed (3 ×); isopropanol (Roth) and hexanes (Fischer)
for HPLC, used without purification; CDCl3 (99.8%, Deutero GmbH), used without purification.
The phosphines P((CH2)3Rf6)3 (30a)110 and P((CH2)3Rf8)3 (30b)103 have been previously
reported. Both were obtained by a sequence involving an Arbuzov reaction, thereby avoiding PH3
(Scheme 4.1).103,104
4.4.2 Analytic Experiments Listed by Figures or Tables
4.4.2.1 Experiments with Catalyst 30a
Recycling Sequence A (Tables 4.2 and 4.3). A vial was charged with C1Ph (0.0188 g,
0.100 mmol), evacuated (10–2 mbar), filled with N2, and transferred into an argon glove box. The
vial was further charged with 30a (0.0114 g, 0.0100 mmol)110 and CH3CN (2.00 mL) to give a
150 4 Recycling of Phosphine Catalysts by Fluorous Concepts
0.0500 M solution of C1Ph. The mixture was heated at 60-64 °C (sand bath) with stirring. After 2.5
h (Table 4.2) or ca. 6.0 h (Table 4.3), the mixture was cooled to –30 °C (overnight). The
supernatant was carefully separated by pipette, taken to dryness, and dissolved in CDCl3. A 1H
NMR spectrum was recorded. The yield of C1Phprod was calculated by integration of the signals at
δ = 6.68 (CH=C) and 5.30 (CHOH) ppm versus all phenyl signals. The vial was recharged with a
0.0500 M CH3CN solution of C1Ph (2.00 mL) and the above procedure repeated.
Recycling Sequence B (Table 4.3). A vial was charged with C1Ph (0.0188 g, 0.100 mmol),
evacuated (10–2 mbar), filled with N2, and transferred into an argon glove box. The vial was further
charged with 30a (0.0114 g, 0.0100 mmol),110 Teflon® shavings (PFA440HPFE, ca. 0.10 g) and
CH3CN (2.00 mL) to give a 0.0500 M solution of C1Ph. The mixture was heated at 60-64 °C (sand
bath) with stirring. After 6 h, the mixture was cooled to –30 °C (overnight). The supernatant was
carefully separated from the coated Teflon® shavings by pipette, taken to dryness, and analyzed per
sequence A.
The coated Teflon® shavings were dried by oil pump vacuum and transferred to another vial
that was charged with a 0.0500 M CH3CN solution of C1Ph (2.00 mL). The above procedure was
repeated.
4.4.2.2 Experiments with Catalyst 30b
Monitoring by 1H NMR (Figures 4.6-4.8). A Schlenk flask was charged with C1Ph,
C1S(i-Pr), or C1S(i-Pr)2, evacuated (10–2 mbar), filled with N2, and transferred into an argon glove
box. The flask was further charged with 30b (10 mol%)103,104 and CH3CN (typically 5-7 mL) to
give a 0.0500 M solution of substrate. The mixture was heated at 60-64 °C (sand bath) with stirring.
Aliquots were periodically removed (0.600 mL). The solvent was replaced by CDCl3 and the
solutions analyzed by 1H NMR spectroscopy (the yields were calculated as follows: C1Phprod, see
section 4.4.2.1; C1S(i-Pr)prod, by integration of the signals at δ = 6.89 (CH=C) and 5.14 (CHOH)
ppm versus all i-Pr methine signals; C1S(i-Pr)2prod, by integration of the signal at δ = 6.80 (CH=C)
ppm versus all i-Pr methine signals). The NMR monitoring gave the following data:
4 Recycling of Phosphine Catalysts by Fluorous Concepts 151
Yield (%)
Time (h) C1Phprod C1S(i-Pr)prod C1S(i-Pr)2prod
0 0 0 0
1 10 3 4
2 22 7 8
3 31 11 13
5 50 22 21
7 62 30 27
9 72 40 34
12 79 50 44
24 85 76 64
48 92 75
Recycling Sequence A (Figures 4.6-4.9 and 4.11-4.14). A vial was charged with C1Ph,
C2(p-Tol), C1S(i-Pr), or C1S(i-Pr)2 (0.100 mmol), evacuated (10–2 mbar), filled with N2, and
transferred into an argon glove box. The vial was further charged with 30b (0.0141 g, 0.0100
mmol)103,104 and CH3CN (2.00 mL; C1S(i-Pr) or C1S(i-Pr)2, GC standard n-Bu2O, 0.0500 M) to
give a 0.0500 M solution of substrate. The mixture was heated at 60-64 °C (sand bath; C2(p-Tol),
70-72 °C) with stirring. Aliquots were periodically removed (0.010 mL), diluted with CH3CN
(C1Ph or C2(p-Tol), 1.50 mL; C1S(i-Pr) or C1S(i-Pr)2, 0.180 mL) and analyzed (C1Ph or C2(p-
Tol), HPLC; C1S(i-Pr) or C1S(i-Pr)2, GC). After reaction ceased (C1Ph, 24 h; C1S(i-Pr) or C1S(i-
Pr)2, 48 h; C2(p-Tol), 72 h), the mixture was cooled to room temperature (3 h) and then –30 °C
(overnight). The supernatant was carefully separated from the precipitated 30b by syringe, taken to
dryness, and dissolved in CDCl3. A 1H NMR spectrum was recorded and the product yield
calculated (C1Phprod, C1S(i-Pr)prod, or C1S(i-Pr)2prod, see above; C2(p-Tol)prod, by integration
of the signals at δ = 6.70 (CH=C) and 4.71 (CHOH) ppm versus all aryl signals).
The precipitated 30b was washed with CH3CN (0.5 mL) and dried by oil pump vacuum.
The vial was recharged with a 0.0500 M CH3CN solution of substrate (1.92 mL; C1S(i-Pr) or
152 4 Recycling of Phosphine Catalysts by Fluorous Concepts
C1S(i-Pr)2, GC standard n-Bu2O, 0.0500 M) and the above procedure repeated. The volume was
decreased by 0.080 mL for each cycle to compensate for the catalyst lost in the aliquots. After the
final cycle, the catalyst was washed with CH3CN (0.5 mL), dried by oil pump vacuum, and
dissolved in PhCF3. Then 1H NMR and 31P{1H} NMR spectra were recorded (data: Tables 4.5 and
4.6).
Recycling Sequence B (Figure 4.16). A vial was charged with C1Ph (0.0188 g, 0.100
mmol), evacuated (10–2 mbar), filled with N2, and transferred into an argon glove box. The vial was
further charged with 30b (0.0141 g, 0.0100 mmol), Teflon® tape (0.160 g, ca. 550 × 12 mm) and
CH3CN (2.00 mL) to give a 0.0500 M solution of C1Ph. The mixture was heated at 60-64 °C (sand
bath) with stirring. Aliquots were periodically analyzed per sequence A. After 24 h, the mixture was
cooled to room temperature (3 h) and then –30 °C (overnight). The supernatant was carefully
separated from the coated Teflon® tape by syringe, taken to dryness, and analyzed per sequence A.
The coated Teflon® tape was washed with CH3CN (4 × 1.5 mL) and dried by oil pump vacuum.
Another vial was charged with a 0.0500 M CH3CN solution of C1Ph (1.92 mL) and the coated
Teflon® tape. The above procedure was repeated. The volume was decreased by 0.080 mL for each
cycle to compensate for the catalyst lost in the aliquots. After the final cycle, the Teflon® tape was
washed with CH3CN (4 × 1.5 mL) and dried by oil pump vacuum, and the coating dissolved in
PhCF3. Then 1H NMR and 31P{1H} NMR spectra were recorded (data: Table 4.6).
Recycling Sequence C (Figure 4.18). This was conducted identically to sequence B, using
Gore-Rastex® fiber (0.160 g) in place of Teflon® tape.
4 Recycling of Phosphine Catalysts by Fluorous Concepts 153
HPLC monitoring of the reaction of C1Ph with 30b according to sequence A (Figures 4.6
and 4.11) gave the following data:
Yield C1Phprod (%)
Time (h) Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5
0 0 0 0 0 0
1 14.7 16.3 15.2 12.8 12.9
2 31.1 30.7 32.2 26.7 25.9
3 46.8 45.2 47.1 40.3 34.4
5 62.9 60.1 65.5 57.7 53.9
7 73.6 72.9 71.1 65.9 61.0
9 80.8 79.4 79.8 73.8 71.5
12 83.9 82.1 83.5 79.2 74.1
24 84.5 83.0 83.5 78.4 78.1
HPLC monitoring of the reaction of C1Ph with 30b according to sequences B and C
(Figures 4.16 and 4.18) gave the following data:
Yield C1Phprod (%); catalyst recovery by Teflon® tape
Time (h) Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5
0 0 0 0 0 0
1 5.6 14.4 9.4 7.7 6.2
2 20.3 29.3 18.6 15.9 12.3
3 36.5 39.4 25.2 23.7 16.9
5 57.6 53.1 35.6 35.5 23.9
7 68.9 64.6 46.6 45.0 33.8
9 78.6 70.0 61.4 53.1 42.5
12 79.7 75.5 69.4 57.5 52.0
24 81.6 78.4 74.9 69.8 65.5
154 4 Recycling of Phosphine Catalysts by Fluorous Concepts
Yield C1Phprod (%); catalyst recovery by Gore-Rastex® fiber
Time (h) Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5
0 0 0 0 0 0
1 17.4 16.5 13.5 10.2 7.3
2 27.7 28.4 26.1 23.0 12.9
3 -- 39.5 39.7 34.3 18.5
5 60.7 59.2 58.4 47.6 29.1
7 71.5 68.9 68.5 59.1 39.7
9 76.7 71.9 73.3 67.4 50.8
12 84.0 73.7 79.7 77.3 60.6
24 82.0 80.0 80.5 77.7 73.7
HPLC monitoring of the reaction of C2(p-Tol) with 30b according to sequence A (Figure
4.14) gave the following data:
Yield C2(p-Tol)prod (%)
Time (h) Cycle 1 Cycle 2 Cycle 3
0 0 0 0
1 0.4 1.5 3.3
2 2.5 3.8 5.6
3 4.7 5.9 7.0
5 9.7 9.8 11.0
7 13.7 15.9 17.1
9 19.8 17.1 25.7
12 26.8 25.6 32.2
24 47.6 46.0 55.6
48 69.2 71.1 62.2
72 79.3 81.1 79.1
4 Recycling of Phosphine Catalysts by Fluorous Concepts 155
GC monitoring of the reactions of C1S(i-Pr) or C1S(i-Pr)2 with 30b according to sequence
A (Figures 4.12 and 4.13) gave the following data:
Yield C1S(i-Pr)prod (%) Yield C1S(i-Pr)2
prod (%) Time (h) Cycle 1 Cycle 2 Cycle 3 Cycle 1 Cycle 2 Cycle 3
0 0 0 0 0 0 0
1 2.3 3.0 3.6 -- 3.7 5.1
2 12.1 10.2 9.6 6.8 9.0 10.2
3 19.8 22.5 16.3 -- 11.9 16.8
5 -- 39.8 39.1 24.2 17.6 26.4
7 48.1 49.3 45.7 35.2 26.7 --
9 -- 59.4 51.7 -- -- 38.7
12 -- 64.5 65.2 -- 43.4 46.0
24 79.9 80.7 84.1 69.2 66.8 64.3
48 95.0 95.0 96.0 72.0 71.0 73.0
4.4.3 Preparative Reactions
Reaction of (C1S(i-Pr) or C2(p-Tol) with 30b. A flask was charged with C1S(i-Pr) (0.120
g, 0.645 mmol) or C2(p-Tol) (0.150 g, 0.694 mmol), evacuated (10–2 mbar), filled with N2, and
transferred into an argon glove box. The vial was further charged with 30b (0.0912 g, 0.0645 mmol
or 0.0982 g, 0.0694 mmol)103,104 and CH3CN (12.9 mL or 13.9 mL) to give a 0.0500 M solution of
substrate. The mixture was heated at 60-64 °C or 70-72 °C (sand bath) with stirring.
After 48 h or 72 h the mixture was cooled to –30 °C (overnight). The supernatant was
carefully separated from the precipitated 30b by syringe, taken to dryness by rotary evaporation,
and purified by column chromatography as detailed in section 2.5.4. Yields (section 4.2.2.4): C1S(i-
Pr)prod: 0.0804 g, 0.432 mmol, 81%; C2(p-Tol)prod: 0.122 g, 0.562 mmol, 67%.
156 4 Recycling of Phosphine Catalysts by Fluorous Concepts
4.4.4 Thermomorphic Behavior of 30b
Solubility of 30b in CH3CN as a function of temperature (Figure 4.19). A NMR tube
was charged under argon with 30b (0.00490 g, 0.00347 mmol),103,104 and a CH3CN solution of
Na+ BArf– internal standard (0.00188 M; 0.693 mL).108 The tube was transferred to a NMR probe
that was warmed in stages. After 5 min at each temperature, a 19F NMR spectrum was recorded
(384 pulses). The concentration of 30b was calculated by integration of the signals at δ = –62.8 (s,
CF3, Na+ BArf–) and –82.3 (t, 3J(F,F) = 9.8 Hz, CF3, 30b) ppm. The probe temperature was
similarly cooled in stages. Results from two independent experiments were averaged to give the
following data:
Temperature (°C) Concentration (mM)
25 0
35 0
45 0.470
50 0.730
55 0.972
60 1.28
65 1.67
70 2.17
75 2.67
80 3.20
80 3.21
70 2.59
60 1.91
50 1.33
40 0.705
30 0.311
5
Preparation of Substrates
5.1 Introduction
In order to study the intramolecular Morita Baylis Hillman reaction as described in the
previous chapters, suitable dicarbonyl substrates were necessary. Hence, efficient syntheses were
sought. The existing literature gives three routes. One is from Koo, who started with 2-
cyclohexenone (Scheme 5.1).28
Scheme 5.1. General route to substrates C1R, published by Koo.
After oxidation of 2-cyclohexenone at the position α to the carbonyl group to an acetoxy
derivative, a nucleophile RLi is added. This gives a 1,2-addition product, and simultaneously the
acyl ester is attacked by RLi and cleaved. The resulting cyclic glycol is then cleaved with Pb(OAc)4
to give the desired substrate.
Another route was found earlier by Denmark (Scheme 5.2).46 He used the fact that there is
always an equilibrium between cyclic hemiacetals and their acyclic aldehyde form.
O
Pb(OAc)4
O
OAcRLi
HOOH
R
CHOR
O
Pb(OAc)4
C1R
158 5 Preparation of Substrates
Scheme 5.2. General route to substrates C1R, published by Denmark.
The procedure starts with the commercially available 2-hydroxytetrahydrofuran and its open
form, 4-hydroxybutyraldehyde. This reacts with a stable ylide as shown. The resulting alcohol is
then oxidized by pyridinium chlorochromate (PCC) to the aldehyde, giving the desired substrate
C1R.
The third method was first published by Murphy and is shown in Scheme 5.3.20
Scheme 5.3. General route to substrates C1R, published by Murphy.
In contrast to Denmark's method, a dialdehyde is employed from the beginning. This reacts
similarly in a Wittig reaction with stable ylides as above, yielding directly the desired substrates
C1R.
This last method was selected for substrate preparation within this work. There were several
OHO CH2OHOHC
R
O
PPh3
R
O
CH2OH
R
O
CHO
PCC
C1R
CHOOHC
R
O
PPh3
R
O
CHO
C1R
5 Preparation of Substrates 159
advantages compared to the others. One was that only (E) C=C isomers were obtained. Furthermore,
no heavy metals such as lead or chromium had to be used. For obtaining Rauhut Currier substrates
only the reactant stoichiometry had to be changed. Additionally this method was the most universal
one, allowing different chain lengths and moieties R. In the following section the preparation of
Morita Baylis Hillman as well as Rauhut Currier substrates is detailed, following the route of
Murphy.
160 5 Preparation of Substrates
5.2 Results
5.2.1 Preparation of the Stable Ylides
Most stable ylides employed in this study are known from the literature, but syntheses are
summarized in Scheme 5.4 anyway to show the broad applicability of these preparations. The yields
shown are those that I obtained upon synthesizing these previously reported or in some cases new
compounds.
Scheme 5.4. Preparation of the stable ylides 34.
Pred, Br2
CH2BrBr
O
EtOH, or
CH3OH, or
i-PrSHCH2BrR
O
CH3R
O
Br2
30-50%
33a: > 90%
33b: > 90%
33c: > 90%
33d: 42%
33e: 54%
33f: 36%
CH2BrR
O
PPh3
Toluene
R
O
PPh3
Br
R
O
PPh3
[34a-H]+ Br−: 70%
[34b-H]+ Br−: 80%
[34c-H]+ Br−: 91%
[34d-H]+ Br−: 75%
[34e-H]+ Br−: 92%
[34f-H]+ Br−: 60%
NaOHToluene/
H2O
34a: 76%
34b: 80%
34c: 57%
34d: 85%
34e: 52%
34f: 78%
R = Ph
p-Tol
CH3
33
CH3HO
O
33
[34-H]+ Br−
a: R = EtO
b: R = CH3O
c: R = i-PrS
d: R = Ph
e: R = p-Tol
f: R = CH3
34
5 Preparation of Substrates 161
In all cases, the first objective was the α-bromoacetyl derivative 33. Depending on the
moiety R, two routes were available. If RCO was an ester derivative (R = CH3O, EtO, i-PrS), the
sequence began with α-bromoacetyl bromide, which was treated with an alcohol or thiol to give the
corresponding α-brominated esters 33a-c in nearly quantitative yields (> 90%). If RCO was an acyl
moiety (R = Ph, p-Tol, CH3), the sequence began with the α-bromination of the corresponding
methyl ketone to give the bromomethyl ketones 33d-f. However, the yields were only moderate
(36-54%) due to losses during recrystallizations. In any case, as these are very basic reactions, large
amounts of starting material can be employed.
The α-brominated compounds 33 were subsequently treated with PPh3. As these reactions
were performed in toluene, the resulting phosphonium salts [34-H]+ Br– started to precipitate
directly after the reactants were mixed. The yields were good to excellent (60-91%). However,
recrystallization from EtOH was necessary in most cases. Only with the salt [34c-H]+ Br– with the
i-PrS moiety were a less polar solvent (hexanes) and a lower temperature (–24 °C) required. The
salts [34-H]+ Br– were characterized by NMR spectroscopy (1H, 13C{1H}, 31P{1H}). Although the
salts [34 (a,b,d-f)-H]+ Br– have been previously reported, NMR data are presented in the
experimental section, in cases where only partial characterization was given earlier. As the salt
[34c-H]+ Br– was a new compound, IR, elemental analysis, and melting point data are also
presented.
The salts [34-H]+ Br–, which were very slightly or even insoluble in toluene, were
subsequently deprotonated with aqueous NaOH using a biphasic water/toluene mixture. The
addition of the indicator phenolphthalein showed when an excess of NaOH was present, signifying
full deprotonation. The stable ylides 34 were very soluble in toluene and were crystallized from the
dried toluene phase. They were obtained in good yields (52-85%) and characterized by NMR
spectroscopy (1H, 13C{1H}, 31P{1H}). Although the stable ylides 34a,b,d-f have been previously
reported, NMR data are presented in the experimental section, in cases where only partial
characterization was given earlier. As ylide 34c was a new compound, IR, elemental analysis, and
melting point data are also presented.
162 5 Preparation of Substrates
5.2.2 Wittig Reactions of the Stable Ylides with Dialdehydes
The next and last step en route to the substrates was the Wittig reaction of the stable ylides
34 with the dialdehydes as depicted in Scheme 5.5.
Scheme 5.5. Preparation of the substrates CnR and CnR2, by reaction of ylides and dialdehydes.
First, succinaldehyde (n = 1)48,111 as well as glutaraldehyde (n = 2)112 were isolated as pure
compounds, according to the literature. These aldehydes are not stable at room temperature and
polymerize, even at –24 °C. Thus, they are not commercially available in solvent free form. Their
generation from stable acetal precursors is shown in Scheme 5.5. Moderate yields were obtained
(20-40%; literature: 30-70%).48,111,112 All subsequent Wittig reactions were carried out in dry
CH2Cl2 under N2. The yields of those reactions are summarized in Tables 5.1 and 5.2. Interestingly,
all C=C linkages were obtained as a single isomer, and were assigned to trans (E) geometries as
described below.
OHCCHO
n = 1,2
RPPh3
O
1 equiv.
2 equiv.
CHOR
O
R
O
R
O
CnR
CnR2
OOCH3H3CO
OEtO
Morita Baylis Hillman substrates
Rauhut Currier substrates
H2O/H+
H2O/H+
n = 1
n = 2
n = 1,2
n = 1,2
34
5 Preparation of Substrates 163
Table 5.1. Yields of substrates obtained from the reactions of succinaldehyde with ylides 34
(Scheme 5.5).
Morita Baylis
Hillman substrates
Yield (%) Rauhut Currier
substrates Yield (%)
C1OMe 40 C1S(i-Pr)2 71
C1OEt 39 C1Ph2 56
C1S(i-Pr) 71 C1Tol2 82
C1Ph 59 C1Me2 40
C1Tol 34
C1Me 35
Table 5.2. Yields of substrates obtained from the reactions of glutaraldehyde with ylides 34
(Scheme 5.5).
Morita Baylis
Hillman substrates
Yield (%) Rauhut Currier
substrates Yield (%)
C2OMe 66 C2Tol2 80
C2OEt 63 C2Me2 70
C2Ph 75
C2Tol 55
C2Me 71
164 5 Preparation of Substrates
5.3 Discussion
All of the reactions shown in this section are quite basic ones, of the type that can be found
in most standard textbooks. The phosphonium salts [34-H]+ Br– are very stable compounds that
never showed any sign of decomposition while stored for more than two years. They can be
deprotonated under mild conditions by NaOH to give the ylides 34.
The ylides so obtained are also quite stable. The carbanion is stabilized by the phosphonium
phosphorus atom as well as by the carbonyl group. In any event, these compounds show signs of
slow decomposition over time. The formerly white solids became yellow to orange. Nonetheless,
these slightly decomposed compounds can be successfully used for the Wittig reactions.
The subsequent Wittig reactions were all carried out in dry CH2Cl2 while warming from 0
°C to room temperature. Non dried CH2Cl2 can also be used, but the yields decreased by 5% to
10%. In any case, only (E) C=C linkages were found in the products CnR and CnR2. This was
evidenced by the values of the vinylic 1H NMR coupling constants (3J(H,H), ca. 15-16 Hz). The
products were purified by column chromatography on silica gel. As the Morita Baylis Hillman
substrates CnR are not that stable and even less stable on the very polar silica gel phase,
decomposition occurred during chromatography. The stability depends, among other factors, on the
chain length of the Morita Baylis Hillman substrates. Accordingly, the substrates with n = 1 (Table
5.1) are less stable than those with n = 2 (Table 5.2), as evidenced by the uniformly lower yields in
Table 5.1 vs. Table 5.2, and as further evidenced by their shelf lives in the refrigerator. Here, the
Morita Baylis Hillman substrates were always stored in a low temperature freezer at –60 °C.
Nevertheless, after several months some decomposition was always evident.
The Rauhut Currier substrates CnR2 are much more stable. All of these are solids that can
be stored at ambient conditions without problems. Nevertheless, they decompose during column
chromatography, with the yields ranging from acceptable to good (40-82%).
The stabilities of succinaldehyde as well as the glutaraldehyde also merit note. As pure
compounds both of them are quite unstable. The succinaldehyde, which is a liquid, gives a viscous
material after several hours at ambient conditions. After one day a glassy solid is obtained,
5 Preparation of Substrates 165
suggesting polymerization. On the other hand, glutaraldehyde is more stable but also decomposes to
a highly viscous mixture. In any case both could be stored as pure compounds at –60 °C for several
months.
In summary, the synthetic pathways optimized for the substrates CnR and CnR2 (n = 1, 2)
in Scheme 5.5 are simple, extremely versatile, and likely extendable to a wide variety of R groups
as well as higher values of n. This methodology is based upon a route pioneered by Murphy.20 As
described in Chapter 2, most - but not all - of the substrates undergo high yield Morita Baylis
Hillman and Rauhut Currier reactions in the presence of rhenium-containing phosphine catalysts.
Furthermore diastereomeric catalysts gave appreciable enantioselectivities.
166 5 Preparation of Substrates
5.4 Experimental
5.4.1 General Data
NMR spectra were recorded at ambient probe temperature on Bruker 300 and 400 MHz FT
spectrometers and referenced to the residual solvent signal (1H: CHCl3, 7.26 ppm; 13C{1H}: CDCl3,
77.0 ppm) or H3PO4 (31P{1H}, external capillary, 85%, 0.0 ppm). IR spectra were recorded on an
ASI React IR®-1000 spectrometer. Elemental analyses were determined with a Carlo Erba EA1110
CHN instrument. Melting points were measured on an Electrothermal IA 9100 apparatus.
Solvents were treated as follows: CH2Cl2 (Staub), distilled from CaH2; diethyl ether,
toluene (2 × Hedinger), n-pentane (Grüssing), ethyl acetate, hexanes, CH3OH, and EtOH (4 ×
Staub), simple distilled; CDCl3 (99.8%) and DMSO-d6 (99.0%, 2 × Deutero GmbH , used as
received.
Chemicals were treated as follows: i-PrSH (97%, Fluka) and PPh3 (99%, Acros), used as
received.
TLC was carried out with ALUGRAM® SIL G/UV254 plates (Macherey-Nagel). Column
chromatography was conducted using silica gel 60M (Macherey-Nagel).
5.4.2 Ylide Preparation
Preparation of bromomethyl ketones RCOCH2Br (33); general. A flask was charged
with α-bromoacetyl bromide (10.1 g, 50.0 mmol)113 and the alcohol/thiol (50.0 mmol) was added
dropwise at 0 °C with vigorous stirring. HBr evolved. The cold bath was removed. After 1 h, the
pure products 33a-c were obtained by distillation using a short Vigreux column (a, R = EtO, 64 °C,
30 mbar;114 b, R = CH3O, 45 °C, 20 mbar;115 c, R = i-PrS, 80 °C, 15 mbar;116 all > 90% yield).
Analytical data agreed with that in the literature.
5 Preparation of Substrates 167
Preparation of phosphonium bromides [R(CO)CH2PPh3]+ Br– ([34-H]+ Br–); general.
This procedure was adapted from one in the literature.117
A flask was charged with PPh3 (5.76 g, 22.0 mmol) and toluene (50 mL). The α-bromo
carbonyl compound (33a-c, 33d,118 33e,118 or 33f,119 each 20.0 mmol) was dissolved in toluene
(10 mL) and added with stirring. A white solid immediately precipitated. The mixture was kept at
120 °C for 10 min and cooled to room temperature. The precipitate was isolated and washed with
toluene (50 mL) to give a nearly pure product. The white solid was recrystallized from EtOH (for
33c 1:1 v/v EtOH/hexanes was used) to give the pure phosphonium bromide [34-H]+ Br– (a, 70%;
b, 80%; c, 91%; d, 75%; e, 96%; f, 60%).
[EtO(CO)CH2PPh3]+ Br– ([34a-H]+ Br–).114 1H NMR and 13C{1H} NMR agreed with
that in the literature. 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 21.1 (s).
[CH3O(CO)CH2PPh3]+ Br– ([34b-H]+ Br–).115 1H NMR and 13C{1H} NMR agreed with
that in the literature. 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 22.0 (s).
[i-PrS(CO)CH2PPh3]+ Br– ([34c-H]+ Br–). Dec. pt. 170 °C (capillary). Elemental analysis
calcd (%) for C23H24BrOPS (458.0): C 60.13, H 5.27, S 6.98; found: C 60.11, H 5.31, S 6.84.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.80 (dd, 2J(H,P) = 13.4 Hz, 3J(H,H) = 8.1 Hz, o-
C6H5, 6H), 7.71 (t, 3J(H,H) = 7.6 Hz, p-C6H5, 3H), 7.63-7.56 (m, m-C6H5, 6H), 5.64 (d, 2J(H,P) =
13.4 Hz, CH2, 2H), 3.40 (sept, 3J(H,H) = 7.0 Hz, (CH3)2CH, 1H), 1.05 (d, 3J(H,H) = 7.0 Hz,
(CH3)2CH, 6H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm):91 190.1 (d, 2J(C,P) = 6.6 Hz, CO),
134.9 (d, 4J(C,P) = 2.9 Hz, p-C6H5), 133.8 (d, 2J(C,P) = 11.0 Hz, o-C6H5), 130.0 (d, 3J(C,P) = 13.2
Hz, m-C6H5), 117.7 (d, 1J(C,P) = 88.9 Hz, i-C6H5), 40.4 (d, 1J(C,P) = 52.3 Hz, CH2), 36.4 (s,
(CH3)2CH), 22.0 (s, (CH3)2CH); 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 21.6 (s).
IR (thin film, in cm–1):71 1668 (s, νCO).
[Ph(CO)CH2PPh3]+ Br– ([34d-H]+ Br–).120 Analytical data agreed with that in the
literature.
168 5 Preparation of Substrates
[p-Tol(CO)CH2PPh3]+ Br– ([34e-H]+ Br–). 1H NMR (400 MHz, CDCl3, δ in ppm): 8.20
(d, 3J(H,H) = 8.2 Hz, C6H4, o to CO, 2H), 7.89 (dd, 2J(H,P) = 13.4 Hz, 3J(H,H) = 7.4 Hz, o-C6H5,
6H), 7.70 (dt, 3J(H,H) = 7.5 Hz, 5J(H,P) = 1.9 Hz, p-C6H5, 3H), 7.60 (ddd, 3J(H,H) = 7.7 and 7.5
Hz, 4J(H,P) = 3.5 Hz, m-C6H5, 6H), 7.23 (d, 3J(H,H) = 8.1 Hz, C6H4, m to CO, 2H), 6.25 (d,
2J(H,P) = 12.2 Hz, CH2, 2H), 2.33 (s, CH3, 3H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm):
191.5 (d, 2J(C,P) = 12.2 Hz, CO), 146.0 (s, C6H4, p to CO), 134.6 (d, 4J(C,P) = 3.2 Hz, p-C6H5),
134.0 (d, 2J(C,P) = 10.7 Hz, o-C6H5), 132.7 (d, 3J(C,P) = 5.7 Hz, C6H4, i to CO), 130.1 (d, 3J(C,P)
= 13.5 Hz, m-C6H5), 130.0 (d, 4J(C,P) = 1.6 Hz, C6H4, o to CO), 129.6 (s, C6H4, m to CO), 118.9
(d, 1J(C,P) = 89.3 Hz, i-C6H5), 38.3 (d, 1J(C,P) = 61.7 Hz, CH2), 21.7 (s, CH3); 31P{1H} NMR
(121 MHz, CDCl3, δ in ppm): 23.3 (s).
[CH3(CO)CH2PPh3]+ Br– ([34f-H]+ Br–). 1H NMR (400 MHz, CDCl3, δ in ppm): 7.80
(dd, 3J(H,P) = 13.0 Hz, 3J(H,H) = 7.3 Hz, o-C6H5, 6H), 7.68 (t, 3J(H,H) = 7.3 Hz, p-C6H5, 3H),
7.60-7.54 (m, m-C6H5, 6H), 6.98 (d, 2J(H,P) = 13.6 Hz, CH2, 2H), 2.50 (d, 4J(H,P) = 2.3 Hz, CH3,
3H); 13C{1H} NMR (100 MHz, CDCl3, δ in ppm): 200.7 (d, 2J(C,P) = 7.0 Hz, CO), 134.6 (d,
4J(C,P) = 2.9 Hz, p-C6H5), 133.8 (d, 2J(C,P) = 11.0 Hz, o-C6H5), 130.0 (d, 3J(C,P) = 13.1 Hz, m-
C6H5), 118.5 (d, 1J(C,P) = 88.9 Hz, i-C6H5), 40.7 (d, 1J(C,P) = 58.9 Hz, CH2), 32.7 (d, 3J(C,P) =
6.6 Hz, CH3); 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 21.0 (s).
Preparation of ylides RCOCHPPh3 (34); general. This procedure was adapted from one
in the literature.117
An Erlenmeyer flask was charged with the phosphonium bromide [34-H]+ Br– (12.0 mmol),
water (60 mL), and some phenolphthalein. Toluene (100 mL) was added with stirring, and the
mixture was cooled to 0 °C. Then aqueous NaOH (wt 10%) was added dropwise to the emulsion
until a purple color persisted. The phases were separated. The water phase was extracted with
toluene (3 × 50 mL). The toluene phases were combined and dried (MgSO4). The solvent was
removed by rotary evaporation to give 34 as a white solid. The samples were recrystallized from
very concentrated toluene solutions to give the pure ylides 34 (a, 76%; b, 80%; c, 57%; d, 85%; e,
5 Preparation of Substrates 169
52%; f, 78%).
EtO(CO)CHPPh3 (34a).121,122 1H NMR (400 MHz, CDCl3, δ in ppm): 7.64 (dd, 3J(H,P)
= 12.3 Hz, 3J(H,H) = 8.0 Hz, o-C6H5, 6H), 7.53 (t, 3J(H,H) = 7.2 Hz, p-C6H5, 3H), 7.47-7.40 (m,
m-C6H5, 6H), 3.96 (br s,123 CH2, 2H), 2.87 (br s, CH, 1H), 1.16 (br s,123 CH3); 13C{1H} NMR
(100 MHz, CDCl3, δ in ppm): 171.2 (br s,123 CO), 132.9 (d, 2J(C,P) = 10.3 Hz, o-C6H5), 131.8 (d,
4J(C,P) = 2.2 Hz, p-C6H5), 128.6 (d, 3J(C,P) = 12.1 Hz, m-C6H5), 127.8 (d, 1J(C,P) = 94.4 Hz, i-
C6H5), 57.7 (s, CH2), 30.0 (d, 1J(C,P) = 129.5 Hz, CH), 14.7 (s, CH3); 31P{1H} NMR (162 MHz,
CDCl3, δ in ppm): 18.3 (s).
CH3O(CO)CHPPh3 (34b).115,124 1H NMR (400 MHz, CDCl3, δ in ppm): 7.68-7.58 (m, o-
C6H5, 6H), 7.55-7.48 (m, p-C6H5, 3H), 7.47-7.38 (m, m-C6H5, 6H), 3.51 (br s,123 CH3, 3H), 2.88
(br s, CH, 1H); 13C{1H} NMR (100 MHz, CDCl3, δ in ppm): 171.7 (br s,123 CO), 132.9 (d, 2J(C,P)
= 9.9 Hz, o-C6H5), 131.8 (s, p-C6H5), 128.6 (d, 3J(C,P) = 12.1 Hz, m-C6H5), 127.8 (d, 1J(C,P) =
88.6 Hz, i-C6H5), 49.7 (s, CH3), 29.6 (d, 1J(C,P) = 128.4 Hz, CH); 31P{1H} NMR (162 MHz,
CDCl3, δ in ppm): 19.2 (s).
i-PrS(CO)CHPPh3 (34c). Dec. pt. 119 °C (capillary). Elemental analysis calcd (%) for
C23H23OPS (378.1): C 72.99, H 6.13, S 8.47; found: C 72.59, H 6.11, S 8.55.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.61 (dd, 3J(H,P) = 12.8 Hz, 3J(H,H) = 7.5 Hz, o-
C6H5, 6H), 7.53 (t, 3J(H,H) = 7.0 Hz, p-C6H5, 3H), 7.47-7.41 (m, m-C6H5, 6H), 3.63 (d, 2J(H,P) =
23.0 Hz, CH, 1H), 3.58 (sept, 3J(H,H) = 6.7 Hz, (CH3)2CH, 1H), 1.29 (d, 3J(H,H) = 6.8 Hz,
(CH3)2CH, 6H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm): 180.8 (d, 2J(C,P) = 5.1 Hz, CO),
132.9 (d, 2J(C,P) = 10.6 Hz, o-C6H5), 132.1 (d, 4J(C,P) = 2.9 Hz, p-C6H5), 128.8 (d, 3J(C,P) = 12.4
Hz, m-C6H5), 126.7 (d, 1J(C,P) = 90.4 Hz, i-C6H5), 47.1 (d, 1J(C,P) = 109.0 Hz, CH), 33.6 (d,
4J(C,P) = 1.5 Hz, (CH3)2CH), 24.1 (s, (CH3)2CH); 31P{1H} NMR (162 MHz, DMSO-d6, δ in ppm):
14.0 (s).
IR (thin film, in cm–1):71 1568 (s, νCO).
170 5 Preparation of Substrates
Ph(CO)CHPPh3 (34d).121,125,126 1H NMR (400 MHz, CDCl3, δ in ppm): 8.02-7.95 (m, o-
C6H5(CO), 2H), 7.74 (dd, 3J(H,P) = 12.3 Hz, 3J(H,H) = 7.7 Hz, o-C6H5, 6H), 7.56 (t, 3J(H,H) = 7.0
Hz, p-C6H5, 3H), 7.51-7.44 (m, m-C6H5, 6H), 7.38-7.33 (m, m-, p-C6H5(CO), 3H), 4.43 (d, 2J(H,P)
= 24.6 Hz, CH, 1H); 13C{1H} NMR (100 MHz, CDCl3, δ in ppm): 185.4 (s, CO), 141.7 (d, 3J(C,P)
= 15.4 Hz, i-C6H5(CO)), 133.6 (d, 2J(C,P) = 10.3 Hz, o-C6H5), 132.5 (s, p-C6H5), 129.8, 128.2,
127.4 (3 s, o-, m-, p-C6H5(CO)), 129.3 (d, 3J(C,P) = 12.1 Hz, m-C6H5), 127.6 (d, 1J(C,P) = 89.3 Hz,
i-C6H5), 51.0 (d, 1J(C,P) = 111.0 Hz, CH); 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 17.9 (s).
p-Tol(CO)CHPPh3 (34e).125,127 1H NMR (400 MHz, CDCl3, δ in ppm): 7.90 (d, 3J(H,H)
= 8.1 Hz, C6H4, o to CO, 2H), 7.73 (dd, 3J(H,P) = 12.5 Hz, 3J(H,H) = 7.2 Hz, o-C6H5, 6H), 7.55 (t,
3J(H,H) = 7.2 Hz, p-C6H5, 3H), 7.49-7.43 (m, m-C6H5, 6H), 7.17 (d, 3J(H,H) = 8.1 Hz, C6H4, m to
CO, 2H), 4.42 (d, 2J(H,P) = 23.7 Hz, CH, 1H), 2.36 (s, CH3, 3H); 13C{1H} NMR (100 MHz,
CDCl3, δ in ppm): 184.7 (d, 2J(C,P) = 3.3 Hz, CO), 139.2 (s, C6H4, p to CO), 138.4 (d, 3J(C,P) =
15.4 Hz, C6H4 i to CO), 133.1 (d, 2J(C,P) = 10.2 Hz, o-C6H5), 131.9 (d, 4J(C,P) = 2.8 Hz, p-C6H5),
128.8 (d, 3J(C,P) = 12.2 Hz, m-C6H5), 128.3, 126.8 (2 s, C6H4, o, m to CO), 127.1 (d, 1J(C,P) =
91.1 Hz, i-C6H5), 50.0 (d, 1J(C,P) = 112.5 Hz, CH), 21.3 (s, CH3); 31P{1H} NMR (161 MHz,
CDCl3, δ in ppm): 17.8 (s).
CH3(CO)CHPPh3 (34f).126 1H NMR (400 MHz, CDCl3, δ in ppm): 7.65 (dd, 3J(H,P) =
12.6 Hz, 3J(H,H) = 7.0 Hz, o-C6H5, 6H), 7.53 (t, 3J(H,H) = 7.4 Hz, p-C6H5, 3H), 7.47-7.41 (m, m-
C6H5, 6H), 3.74 (br s, CH, 1H), 2.09 (s, CH3, 3H); 13C{1H} NMR (100 MHz, CDCl3, δ in ppm):
190.7 (s, CO), 133.0 (d, 2J(C,P) = 9.9 Hz, o-C6H5), 131.9 (d, 4J(C,P) = 2.6 Hz, p-C6H5), 128.7 (d,
3J(C,P) = 12.1 Hz, m-C6H5), 127.1 (d, 1J(C,P) = 88.6 Hz, i-C6H5), 51.7 (d, 1J(C,P) = 110.1 Hz,
CH), 28.3 (d, 3J(C,P) = 15.4 Hz, CH3); 31P{1H} NMR (162 MHz, CDCl3, δ in ppm): 15.5 (s).
5 Preparation of Substrates 171
5.4.3 Substrate Preparation
5.4.3.1 Morita Baylis Hillman Substrates
(E)-R(CO)CH=CHCH2(CH2)nCHO; general. A flask was charged with a solution of a
dialdehyde (n = 1, succinaldehyde;48,111 2, glutaraldehyde112) in CH2Cl2 (40.0 mL, 0.300 M, 12.0
mmol) and cooled to 0 °C. Then the ylide R(CO)CHPPh3 (34a-f, 5.80 mmol) dissolved in CH2Cl2
(10 mL) was added dropwise with stirring. The mixture was allowed to warm to room temperature
overnight. The solvent was removed by rotary evaporation and the residue was extracted with
diethyl ether (3 × 20 mL). The diethyl ether was removed by rotary evaporation, and the products
were purified by column chromatography (SiO2).
(E)-Ph(CO)CH=CHCH2CH2CHO (C1Ph)25 was obtained (7:3 v/v ethyl acetate/n-pentane)
as a slightly yellow liquid (0.643 g, 3.42 mmol, 59%). Analytical data agreed with that in the
literature.
(E)-CH3O(CO)CH=CHCH2CH2CHO (C1OMe)46,47 was obtained (1:4 v/v ethyl
acetate/n-pentane) as a clear liquid (0.329 g, 2.32 mmol, 40%). Analytical data agreed with that in
the literature.
(E)-EtO(CO)CH=CHCH2CH2CHO (C1OEt)25 was obtained (2:3 v/v ethyl acetate/n-
pentane) as a clear liquid (0.353 g, 2.26 mmol, 39%). Analytical data agreed with that in the
literature.
(E)-p-Tol(CO)CH=CHCH2CH2CHO (C1(p-Tol))42,43 was obtained (3:2 v/v ethyl
acetate/n-pentane) as a slightly yellow liquid (0.398 g, 1.97 mmol, 34%). Analytical data agreed
with that in the literature.
172 5 Preparation of Substrates
(E)-CH3(CO)CH=CHCH2CH2CHO (C1Me)28 was obtained (7:13 v/v ethyl acetate/n-
pentane) as a clear liquid (0.256 g, 2.03 mmol, 35%). Analytical data agreed with that in the
literature.
(E)-i-PrS(CO)CH=CHCH2CH2CHO (C1(S-i-Pr)) was obtained (1:4 v/v ethyl acetate/n-
pentane) as a colorless oil (0.766 g, 4.11 mmol, 71%). Elemental analysis calcd (%) for C9H14O2S
(186.1): C 58.03, H 7.58, S 17.21; found: C 57.82, H 7.78, S 16.80.
1H NMR (400 MHz, CDCl3, δ in ppm): 9.78 (s, CHO, 1H), 6.82 (dt, 3J(H,H) = 15.5 Hz and
6.8 Hz, CH=CHCH2, 1H), 6.08 (d, 3J(H,H) = 15.5 Hz, CH=CHCH2, 1H), 3.70 (sept, 3J(H,H) = 7.0
Hz, (CH3)2CH, 1H), 2.63 (t, 3J(H,H) = 7.0 Hz, CH2CHO, 2H), 2.50 (dt, 3J(H,H) = 7.0 Hz and 6.9
Hz, CH=CHCH2, 2H), 1.31 (d, 3J(H,H) = 7.0 Hz, (CH3)2CH, 6H); 13C{1H} NMR (101 MHz,
CDCl3, δ in ppm): 200.6 (s, CHO), 190.2 (s, CO), 141.9 (s, CH=CHCH2), 129.7 (s, CH=CHCH2),
41.8 (s, CH2CHO), 34.5 (s, (CH3)2CH), 24.3 (s, CH=CHCH2), 23.0 (s, (CH3)2CH).
IR (thin film, in cm–1):71 1725 (s, νCO), 1664 (s, νCO), 1629 (s, νC=C).
(E)-p-Tol(CO)CH=CHCH2CH2CH2CHO (C2(p-Tol)) was obtained (1:3 v/v ethyl
acetate/n-pentane) as a slightly yellow liquid (0.689 g, 3.19 mmol, 55%). Elemental analysis calcd
(%) for C14H16O2 (216.1): C 77.75, H 7.46; found: C 77.18, H 7.53.
1H NMR (400 MHz, CDCl3, δ in ppm): 9.80 (s, CHO, 1H), 7.84 (d, 3J(H,H) = 8.1 Hz, C6H4,
o to CO, 2H), 7.27 (d, 3J(H,H) = 8.1 Hz, C6H4, m to CO, 2H), 7.00 (dt, 3J(H,H) = 15.5 Hz and 6.8
Hz, CH=CHCH2, 1H), 6.32 (d, 3J(H,H) = 15.5 Hz, CH=CHCH2, 1H), 2.53 (t, 3J(H,H) = 7.5 Hz,
CH2CHO, 2H), 2.42 (s, CH3, 3H), 2.36 (dt, 3J(H,H) = 7.5 Hz and 7.5 Hz, CH=CHCH2, 2H), 1.88
(tt, 3J(H,H) = 7.5 Hz and 7.5 Hz, CH2CH2CHO, 2H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm):
202.0 (s, CHO), 190.4 (s, CO), 147.7 (s, CH=CHCH2), 143.9 (s, C6H4, p to CO), 135.6 (s, C6H4, i
to CO), 129.7, 129.1 (2 s, C6H4, o, m to CO), 127.0 (s, CH=CHCH2), 43.5 (s, CH2CHO), 32.3 (s,
CH=CHCH2), 22.1 (s, CH2CH2CHO), 20.9 (s, CH3).
IR (thin film, in cm–1):71 1722 (s, νCO), 1668 (s, νCO), 1621 (s, νC=C).
(E)-CH3O(CO)CH=CHCH2CH2CH2CHO (C2OMe)47 was obtained (1:1 v/v ethyl
5 Preparation of Substrates 173
acetate/n-pentane) as a clear liquid (0.597 g, 3.82 mmol, 66%). Analytical data agreed with that in
the literature.
(E)-EtO(CO)CH=CHCH2CH2CH2CHO (C2OEt)25,49 was obtained (2:3 v/v ethyl
acetate/n-pentane) as a clear liquid (0.621 g, 3.65 mmol, 63%). Analytical data agreed with that in
the literature.
(E)-Ph(CO)CH=CHCH2CH2CH2CHO (C2Ph)25 was obtained (3:7 v/v ethyl acetate/n-
pentane) as a clear liquid (0.410 g, 2.03 mmol, 35%). Analytical data agreed with that in the
literature.
(E)-CH3(CO)CH=CHCH2CH2CH2CHO (C2Me)44 was obtained (3:7 v/v ethyl acetate/n-
pentane) as a clear liquid (0.577 g, 4.12 mmol, 71%). Analytical data agreed with that in the
literature.
5.4.3.2 Rauhut Currier Substrates
(E,E)-R(CO)CH=CHCH2(CH2)nCH=CH(CO)R; general procedure. A flask was
charged with a solution of a dialdehyde (n = 1, succinaldehyde;48,111 2, glutaraldehyde112) in
CH2Cl2 (20.0 mL, 0.300 M, 6.0 mmol). Then the ylide R(CO)CHPPh3 (34a-f, 12.6 mmol)
dissolved in CH2Cl2 (5 mL) was added with stirring. After 24 h, the solvent was removed by rotary
evaporation. The products were purified by column chromatography (SiO2).
(E,E)-Ph(CO)CH=CHCH2CH2CH=CH(CO)Ph (C1Ph2)45 was obtained (1:1 v/v ethyl
acetate/n-pentane) as a white solid (0.974 g, 3.36 mmol, 56%). Analytical data agreed with that in
the literature.
(E,E)-p-Tol(CO)CH=CHCH2CH2CH=CH(CO)p-Tol (C1(p-Tol)2) was obtained (1:1 v/v
ethyl acetate/n-pentane) as a yellow solid (0.286 g, 0.900 mmol, 82%). M.p. 113.0-113.5 °C
174 5 Preparation of Substrates
(capillary). Elemental analysis calcd (%) for C22H22O2 (318.4): C 82.99, H 6.96; found: C 83.00, H
6.86.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.85 (d, 3J(H,H) = 8.1 Hz, C6H4, o to CO, 4H),
7.27 (d, 3J(H,H) = 8.1 Hz, C6H4, m to CO, 4H), 7.06 (dt, 3J(H,H) = 15.3 Hz and 6.4 Hz,
CH=CHCH2, 2H), 6.96 (d, 3J(H,H) = 15.3 Hz, CH=CHCH2, 2H), 2.59-2.56 (m, CH=CHCH2, 4H),
2.43 (s, CH3, 6H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm): 189.9 (s, CO), 146.8 (s,
CH=CHCH2), 143.5 (s, C6H4, p to CO), 135.1 (s, C6H4, i to CO), 129.2, 128.6 (2 s, C6H4, o, m to
CO), 126.6 (s, CH=CHCH2), 31.2 (s, CH=CHCH2), 21.6 (s, CH3).
IR (thin film, cm–1):71 1664 (s, νCO), 1598 (s, νC=C).
(E,E)-CH3(CO)CH=CHCH2CH2CH=CH(CO)CH3 (C1Me2)50 was obtained (1:1 v/v
ethyl acetate/n-pentane) as a slightly yellow solid (0.398 g, 2.40 mmol, 40%). Analytical data
agreed with that in the literature.
(E,E)-i-PrS(CO)CH=CHCH2CH2CH=CH(CO)S(i-Pr) (C1(S-i-Pr)2) was obtained (1:9
v/v ethyl acetate/n-pentane) as a colorless solid (1.37 g, 4.80 mmol, 80%). M.p. 36.8-37.8 °C
(capillary). Elemental analysis calcd (%) for C14H22O2S2 (286.1): C 58.70, H 7.74, S 22.39; found:
C 58.35, H 7.55, S 22.25.
1H NMR (300 MHz, CDCl3, δ in ppm): 6.81 (dt, 3J(H,H) = 15.4 Hz and 6.1 Hz,
CH=CHCH2, 2H), 6.07 (d, 3J(H,H) = 15.4 Hz, CH=CHCH2, 2H), 3.69 (sept, 3J(H,H) = 7.0 Hz,
(CH3)2CH, 2H), 2.33 (d, 3J(H,H) = 6.1 Hz, CH=CHCH2, 4H), 1.31 (d, 3J(H,H) = 7.0 Hz,
(CH3)2CH, 12H); 13C{1H} NMR (101 MHz, CDCl3, δ in ppm): 189.8 (s, CO), 142.3 (s,
CH=CHCH2), 129.6 (s, CH=CHCH2), 34.5 (s, (CH3)2CH), 30.3 (s, CH=CHCH2), 23.0 (s,
(CH3)2CH).
IR (thin film, cm–1):71 1683 (s, νCO), 1652 (s, νC=C).
(E,E)-H3C(CO)CH=CHCH2CH2CH2CH=CH(CO)CH3 (C2Me2)26 was obtained (1:1
v/v ethyl acetate/n-pentane) as a clear liquid (0.756 g, 4.20 mmol, 70%). Analytical data agreed
with that in the literature.
5 Preparation of Substrates 175
(E,E)-p-Tol(CO)CH=CHCH2CH2CH2CH=CH(CO)p-Tol (C2(p-Tol)2) was obtained (1:1
v/v ethyl acetate/n-pentane) as a slightly yellow solid (1.59 g, 4.79 mmol, 80%). M.p. 49-51 °C
(capillary). Elemental analysis calcd (%) for C23H24O2 (332.4): C 83.10, H 7.28; found: C 79.96, H
7.09.
1H NMR (400 MHz, CDCl3, δ in ppm): 7.84 (d, 3J(H,H) = 8.2 Hz, C6H4, o to CO, 4H),
7.26 (d, 3J(H,H) = 8.2 Hz, C6H4, m to CO, 4H), 7.05 (dt, 3J(H,H) = 15.4 Hz and 7.1 Hz,
CH=CHCH2, 2H), 6.92 (dt, 3J(H,H) = 15.4 Hz, 4J(H,H) = 1.0 Hz, CH=CHCH2, 2H), 2.42-2.35 (m,
CH=CHCH2, 4H), 2.41 (s, CH3, 6H), 1.78 (pen, 3J(H,H) = 7.3 Hz, CH=CHCH2CH2, 2H); 13C{1H}
NMR (101 MHz, CDCl3, δ in ppm): 190.1 (s, CO), 148.0 (s, CH=CHCH2), 143.5 (s, C6H4, p to
CO), 135.2 (s, C6H4, i to CO), 129.2, 128.6 (2 s, C6H4, o, m to CO), 126.4 (s, CH=CHCH2), 32.1 (s,
CH=CHCH2), 26.7 (s, CH=CHCH2CH2), 21.6 (s, CH3).
IR (thin film, cm–1):71 1664 (s, νCO), 1602 (s, νC=C).
176 5 Preparation of Substrates
6
References and Notes
(1) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2004, 43, 5138; Angew. Chem. 2004, 116,
5248.
(2) Ahrendt, K. A.; Borths, C. J.; MacMillan, D. W. C. J. Am. Chem. Soc. 2000, 122, 4243.
(3) Pracejus, H. Liebigs Ann. Chem. 1960, 634, 9.
(4) Pracejus, H. Liebigs Ann. Chem. 1960, 634, 23.
(5) Buckley, B. R. Annu. Rep. Prog. Chem., Sect. B: Org. Chem. 2007, 103, 90.
(6) de Figueiredo, R. M.; Christmann, M. Eur. J. Org. Chem. 2007, 2575.
(7) Dondoni, A.; Massi, A. Angew. Chem., Int. Ed. 2008, 47, 4638; Angew. Chem. 2008, 120,
4716.
(8) Methot, J. L.; Roush, W. R. Adv. Synth. Catal. 2004, 346, 1035.
(9) Pellissier, H. Tetrahedron 2007, 63, 9267.
(10) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Chem. Rev. 2003, 103, 811.
(11) Rauhut, M. M.; Currier, H. US patent 3,074,999 (filed December 31, 1958); Chem. Abstr.
1963, 58, 66109.
(12) McClure, J. D. US patent 3,225,083 (filed August 15, 1963); Chem. Abstr. 1965, 64, 103655.
(13) Baizer, M. M.; Anderson, J. D. J. Org. Chem. 1965, 30, 1357.
(14) Takashina, N.; Price, C. C. J. Am. Chem. Soc. 1962, 84, 489.
(15) Oda, R.; Kawabata, T.; Tanimoto, S. Tetrahedron Lett. 1964, 1653.
(16) Morita, K.; Suzuki, Z.; Hirose, H. Bull. Chem. Soc. Jpn. 1968, 41, 2815.
(17) Baylis, A. B.; Hillman, M. E. D. German patent 2,155,113 (filed November 6, 1970); Chem.
Abstr. 1972, 77, 34174.
(18) McClure, J. D. J. Org. Chem. 1970, 35, 3045.
(19) Wang, L.-C.; Luis, A. L.; Agapiou, K.; Jang, H.-Y.; Krische, M. J. J. Am. Chem. Soc. 2002,
178 References and Notes
124, 2402.
(20) Brown, P. M.; Kappel, N.; Murphy, P. J. Tetrahedron Lett. 2002, 43, 8707.
(21) Mergott, D. J.; Frank, S. A.; Roush, W. R. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 11955.
(22) Aroyan, C. E.; Miller, S. J. J. Am. Chem. Soc. 2007, 129, 256.
(23) Roth, F.; Gygax, P.; Fráter, G. Tetrahedron Lett. 1992, 33, 1045.
(24) Dinon, F.; Richards, E.; Murphy, P. J.; Hibbs, D. E.; Hursthouse, M. B.; Abdul Malik, K. M.
Tetrahedron Lett. 1999, 40, 3279.
(25) Richards, E. L.; Murphy, P. J.; Dinon, F.; Fratucello, S.; Brown, P. M.; Gelbrich, T.;
Hursthouse, M. B. Tetrahedron 2001, 57, 7771.
(26) Brown, P. M.; Kaeppel, N.; Murphy, P. J.; Coles, S. J.; Hursthouse, M. B. Tetrahedron 2006,
63, 1100.
(27) Keck, G. E.; Welch, D. S. Org. Lett. 2002, 4, 3687.
(28) Yeo, J. E.; Yang, X.; Kim, H. J.; Koo, S. Chem. Commun. 2004, 236.
(29) Aroyan, C. E.; Vasbinder, M. M.; Miller, S. J. Org. Lett. 2005, 7, 3849.
(30) Yi, W.-B.; Cai, C.; Wang, X. J. Fluorine Chem. 2007, 128, 919.
(31) Buhro, W. E.; Zwick, B. D.; Georgiou, S.; Hutchinson, J. P.; Gladysz, J. A. J. Am. Chem.
Soc. 1988, 110, 2427.
(32) Gladysz, J. A.; Boone, B. J. Angew. Chem., Int. Ed. Engl. 1997, 36, 551; Angew. Chem.
1997, 109, 566.
(33) Zwick, B. D.; Dewey, M. A.; Knight, D. A.; Buhro, W. E.; Arif, A. M.; Gladysz, J. A.
Organometallics 1992, 11, 2673.
(34) Kahn, S. D.; Pau, C. F.; Chamberlin, A. R.; Hehre, W. J. J. Am. Chem. Soc. 1987, 109, 650.
(35) Delacroix, O.; Gladysz, J. A. Chem. Commun. 2003, 665.
(36) Eichenseher, S.; Delacroix, O.; Kromm, K.; Hampel, F.; Gladysz, J. A. Organometallics
2005, 24, 245.
(37) Friedlein, F. K.; Kromm, K.; Hampel, F.; Gladysz, J. A. Chem. Eur. J. 2006, 12, 5267.
(38) Kromm, K.; Zwick, B. D.; Meyer, O.; Hampel, F.; Gladysz, J. A. Chem. Eur. J. 2001, 7,
2015.
(39) Kromm, K.; Osburn, P. L.; Gladysz, J. A. Organometallics 2002, 21, 4275.
References and Notes 179
(40) Kromm, K.; Hampel, F.; Gladysz, J. A. Helv. Chim. Acta 2002, 85, 1778.
(41) Kromm, K.; Eichenseher, S.; Prommesberger, M.; Hampel, F.; Gladysz, J. A. Eur. J. Inorg.
Chem. 2005, 2983.
(42) Kwong, C. K.-W.; Huang, R.; Zhang, M.; Shi, M.; Toy, P. H. Chem. Eur. J. 2007, 13, 2369.
(43) Teng, W.-D.; Huang, R.; Kwong, C. K.-W.; Shi, M.; Toy, P. H. J. Org. Chem. 2006, 71, 368.
(44) Enholm, E. J.; Kinter, K. S. J. Org. Chem. 1995, 60, 4850.
(45) Yung-Son, H.; Kuan-Po, C.; Pao-Chiung, H.; Ling, L. Synth. Commun. 1992, 22, 429.
(46) Denmark, S. E.; Senanayake, C. B. W. J. Org. Chem. 1993, 58, 1853.
(47) Freiria, M.; Whitehead, A. J.; Tocher, D. A.; Motherwell, W. B. Tetrahedron 2004, 60, 2673.
(48) House, H. O.; Cronin, T. H. J. Org. Chem. 1965, 30, 1061.
(49) Trost, B. M.; Chan, D. M. T. J. Am. Chem. Soc. 1982, 104, 3733.
(50) Black, G. P.; Murphy, P. J.; Walshe, N. D. A. Tetrahedron 1998, 54, 9481.
(51) Kromm, K.; Hampel, F.; Gladysz, J. A. Organometallics 2002, 21, 4264.
(52) Castillo, D. A. Doctoral Dissertation, Universität Erlangen-Nürnberg, 2008.
(53) Bode, M. L.; Kaye, P. T. Tetrahedron Lett. 1991, 32, 5611.
(54) Price, K. E.; Broadwater, S. J.; Walker, B. J.; McQuade, D. T. J. Org. Chem. 2005, 70, 3980.
(55) Roy, D.; Sunoj, R. B. Org. Lett. 2007, 9, 4873.
(56) Handbook of Chemistry and Physics, 51. ed. (Ed.: Weast, R. C.), The Chemical Rubber Co.:
Cleveland, 1971.
(57) Scherer, A. Doctoral Dissertation, Universität Erlangen-Nürnberg, 2009.
(58) Rafel, S.; Leahy, J. W. J. Org. Chem. 1997, 62, 1521.
(59) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734.
(60) Thalji, R. K.; Roush, W. R. J. Am. Chem. Soc. 2005, 127, 16778.
(61) Eichenseher, S. Doctoral Dissertation, Universität Erlangen-Nürnberg, 2005.
(62) Simpson, A. F.; Bodkin, C. D.; Butts, C. P.; Armitage, M. A.; Gallagher, T. J. Chem. Soc.,
Perkin Trans. 1 2000, 3047.
(63) Wilt, J. W.; Niinemae, R. J. Org. Chem. 1979, 44, 2533.
(64) Castillo, D. A.; Gladysz, J. A. in Trivalent Phosphorus Compounds in Asymmetric Catalysis
(Ed.: Börner, A.), Wiley-VCH: Weinheim, 2008.
180 References and Notes
(65) (a) Pietrusiewicz, K. M.; Zablocka, M. Chem. Rev. 1994, 94, 1375. (b) Grabulosa, A.;
Granell, J.; Muller, G. Coord. Chem. Rev. 2007, 251, 25. (c) Engel, R.; Rizzo, J. I. Curr.
Org. Chem. 2006, 10, 2393. (d) Crépy, K. V. L.; Imamoto, T. Top. Curr. Chem. 2003, 229,
1. (e) Valentine, D. J. in Asymmetric Synthesis, Vol. 4, Chap. 3 (Eds.: D. J. Morrison, J. W.
Scott), Academic Press, New York, 1984, pp. 263-312.
(66) The typical pattern of cyclic Rauhut Currier products was depicted in 1H NMR spectra from
the reaction mixtures containing C2Me2prod or C2(p-Tol)2
prod.
(67) Kempf, B.; Mayr, H. Chem. Eur. J. 2005, 11, 917.
(68) Streitwieser, A.; Heathcock, C. H.; Kosower, E. M. Introduction to Oganic Chemistry; 2. ed.,
VCH: New York, 1994.
(69) Birman, V. B.; Rheingold, A. L.; Lam, K.-C. Tetrahedron: Asymmetry 1999, 10, 125.
(70) Graff, M.; Al Dilaimi, A.; Seguineau, P.; Rambaud, M.; Villieras, J. Tetrahedron Lett. 1986,
27, 1577.
(71) The most distinctive and characteristic signals are given.
(72) Cargill Thompson, A. M. W. Annu. Rep. Prog. Chem., Sect. A: Inorg. Chem. 1999, 95, 153.
(73) Fergusson, J. E. Coord. Chem. Rev. 1966, 1, 459.
(74) Friedlein, F. K.; Hampel, F.; Gladysz, J. A. Organometallics 2005, 24, 4103.
(75) Alvey, L. J.; Delacroix, O.; Wallner, C.; Meyer, O.; Hampel, F.; Szafert, S.; Lis, T.; Gladysz,
J. A. Organometallics 2001, 20, 3087.
(76) Stark, G. A.; Gladysz, J. A. Inorg. Chim. Acta 1998, 269, 167.
(77) Amoroso, A. J.; Arif, A. M.; Gladysz, J. A. Organometallics 1997, 16, 6032.
(78) Dewey, M. A.; Stark, G. A.; Gladysz, J. A. Organometallics 1996, 15, 4798.
(79) Seidel, S. N. Doctoral Dissertation, Universität Erlangen-Nürnberg, 2009.
(80) Buhro, W. E.; Georgiou, S.; Hutchinson, J. P.; Gladysz, J. A. J. Am. Chem. Soc. 1985, 107,
3346.
(81) Friedlein, F. K. Doctoral Dissertation, Universität Erlangen-Nürnberg, 2006.
(82) Agbossou, F.; O'Connor, E. J.; Garner, C. M.; Mendez, N. Q.; Fernandez, J. M.; Patton, A.
T.; Ramsden, J. A.; Gladysz, J. A. Inorg. Synth. 1992, 29, 211.
(83) Tam, W.; Lin, G. Y.; Wong, W. K.; Kiel, W. A.; Wong, V. K.; Gladysz, J. A. J. Am. Chem.
References and Notes 181
Soc. 1982, 104, 141.
(84) Kiel, W. A.; Lin, G.-Y.; Gladysz, J. A. J. Am. Chem. Soc. 1980, 102, 3299.
(85) Merrifield, J. H.; Fernandez, J. M.; Buhro, W. E.; Gladysz, J. A. Inorg. Chem. 1984, 23,
4022.
(86) Merrifield, J. H.; Strouse, C. E.; Gladysz, J. A. Organometallics 1982, 1, 1204.
(87) Busacca, C. A.; Lorenz, J. C.; Grinberg, N.; Haddad, N.; Hrapchak, M.; Latli, B.; Lee, H.;
Sabila, P.; Saha, A.; Sarvestani, M.; Shen, S.; Varsolona, R.; Wei, X.; Senanayake, C. H.
Org. Lett. 2005, 7, 4277.
(88) Hobbs, C. F.; Knowles, W. S. J. Org. Chem. 1981, 46, 4422.
(89) Kiel, W. A.; Lin, G. Y.; Bodner, G. S.; Gladysz, J. A. J. Am. Chem. Soc. 1983, 105, 4958.
(90) The quality of commercial Ph3C+ X− can vary, and crystallization from CH2Cl2/hexanes or
CH2Cl2/benzene is recommended, see: Patton, A. T.; Strouse, C. E.; Knobler, C. B.;
Gladysz, J. A. J. Am. Chem. Soc. 1983, 105, 5804.
(91) Assignments of resonances to the PPh3, carbons were made as follows. Both ipso (i) and
para (p) carbon resonances were easily distinguished on the basis of intensity and J(C,P).
The meta (m) resonance was assigned according to the proposal of Mann that it is the one
closest to the chemical shift of benzene: Mann, B. E. J. Chem. Soc., Perkin Trans. 2 1972,
30.
(92) FAB, 3-NBA, m/z (%); the peaks correspond to the most intense signal of the isotope
envelope.
(93) Due to the complexity of the recorded signals, no clear assignment of the atoms and
coupling constants can be given.
(94) Due to the poor solubility and decomposition with time, not all signals were detected.
(95) MALDI-TOF, SIN, m/z (%); the peaks correspond to the most intense signal of the isotope
envelope.
(96) Horvath, I. T.; Rabai, J. Science 1994, 266, 72.
(97) Gladysz, J. A. Science 1994, 266, 55.
(98) Handbook of Fluorous Chemistry (Eds.: J. A. Gladysz, D. P. Curran, I. T. Horváth), Wiley-
VCH: Weinheim, 2004.
182 References and Notes
(99) Gladysz, J. A.; Curran, D. P. Tetrahedron 2002, 58, 3823.
(100) Wende, M.; Gladysz, J. A. J. Am. Chem. Soc. 2003, 125, 5861.
(101) Wende, M.; Meier, R.; Gladysz, J. A. J. Am. Chem. Soc. 2001, 123, 11490.
(102) Dinh, L. V.; Gladysz, J. A. Angew. Chem., Int. Ed. 2005, 44, 4095; Angew. Chem. 2005,
117, 4164.
(103) Alvey, L. J.; Rutherford, D.; Juliette, J. J. J.; Gladysz, J. A. J. Org. Chem. 1998, 63, 6302.
(104) Emnet, C.; Gladysz, J. A. Synthesis 2005, 1012.
(105) Vlad, G.; Richter, F.; Horváth, I. T. Org. Lett. 2004, 6, 4559.
(106) Vlad, G.; Richter, F. U.; Horváth, I. T. Tetrahedron Lett. 2005, 46, 8605.
(107) Kligerman, A. D.; Bishop, J. B.; Erexson, G. L.; Price, H. C.; O'Connor, R. W.; Morgan, D.
L.; Zeiger, E. Environ. Mol. Mutagen. 1994, 24, 301.
(108) Yakelis, N. A.; Bergman, R. G. Organometallics 2005, 24, 3579.
(109) Mandal, D.; Jurisch, M.; Consorti, C. S.; Gladysz, J. A. Chemistry, an Asian Journal 2008,
3, 1772.
(110) Jurisch, M. Doctoral Dissertation, Universität Erlangen-Nürnberg, 2008.
(111) Fakstorp, J.; Raleigh, D.; Schniepp, L. E. J. Am. Chem. Soc. 1950, 72, 869.
(112) Longley, R. I., Jr.; Emerson, W. S. J. Am. Chem. Soc. 1950, 72, 3079.
(113) Goerger, M. M.; Hudson, B. S. J. Org. Chem. 1988, 53, 3148.
(114) Alvarez-Sanchez, R.; Basketter, D.; Pease, C.; Lepoittevin, J.-P. Chem. Res. Toxicol. 2003,
16, 627.
(115) Werkhoven, T. M.; Van Nispen, R.; Lugtenburg, J. Eur. J. Org. Chem. 1999, 2909.
(116) Dellaria, J. F., Jr.; Nordeen, C.; Swett, L. R. Synth. Commun. 1986, 16, 1043.
(117) Denney, D. B.; Smith, L. C.; Song, J.; Rossi, C. J.; Hall, C. D. J. Org. Chem. 1963, 28, 778.
(118) (a) Autorenkollektiv Organikum, 2. ed., Wiley-VCH: Weinheim, 1999. (b) see also Paul, S.;
Gupta, V.; Gupta, R.; Loupy, A. Tetrahedron Lett. 2003, 44, 439.
(119) Levene, P. A. "Organic Syntheses"; Wiley: New York, 1943; Collect. Vol. II, p 88.
(120) Balema, V. P.; Wiench, J. W.; Pruski, M.; Pecharsky, V. K. Chem. Commun. 2002, 724.
(121) Spencer, E. C.; Kalyanasundari, B.; Mariyatra, M. B.; Howard, J. A. K.; Panchanatheswaran,
K. Inorg. Chim. Acta 2006, 359, 35.
References and Notes 183
(122) Considine, W. J. J. Org. Chem. 1962, 27, 647.
(123) This broadened signal is believed to be due to geometric isomers about a O=C...C bond that
are not rapidly interconverting on the NMR time scale under the conditions of the
measurement. For similar observations or a rationale of the structural possibilities/isomers
see reference 121.
(124) Boers, R. B.; Randulfe, Y. P.; Van der Haas, H. N. S.; Van Rossum-Baan, M.; Lugtenburg, J.
Eur. J. Org. Chem. 2002, 2094.
(125) Bestmann, H. J.; Bomhard, A.; Dostalek, R.; Pichl, R.; Riemer, R.; Zimmermann, R.
Synthesis 1992, 787.
(126) Bestmann, H. J.; Arnason, B. Chem. Ber. 1962, 95, 1513.
(127) Bestmann, H. J.; Kisielowski, L. Chem. Ber. 1983, 116, 1320.
List of Publications
"Synthesis and Properties of Fluorous Arenes and Triaryl Phosphorus Compounds with Branched
Fluoroalkyl Moieties ("Split Pony Tails")", Wende, M.; Seidel, F.; Gladysz, J. A. J. Fluorine Chem.
2003, 124, 45.
"Enantioselective Catalysis of Intramolecular Morita Baylis-Hillman and Related Reactions by
Chiral Rhenium-Containing Phosphines of the Formula (η5-C5H5)Re(NO)(PPh2)(CH2PAr2)",
Seidel, F.; Gladysz, J. A. Synlett 2007, 6, 986.
"Catalysis of Intramolecular Morita Baylis Hillman and Rauhut Currier Reactions by Fluorous
Phosphines; Facile Recovery by Liquid/Solid Organic/Fluorous Biphase Proctocols Involving
Precipitation, Teflon® Tape and Gore-Tex® Fibers", Seidel, F. O.; Gladysz, J. A. Adv. Synth. Catal.
2008, 350, 2443.
Curriculum Vitae
Persönliche Daten Florian O. Seidel
geboren am 12.04.1978 in Nürnberg
verheiratet
Schulbildung 1984 – 1988 Grundschule Herpersdorf b. Nürnberg
1988 – 1997 Pirckheimer-Gymnasium Nürnberg
Studium Okt. 1997 – Jan. 2004 Chemie (Diplom)
An der Friedrich-Alexander-Universität
Erlangen-Nürnberg
Nov. 2000 Diplomvorprüfung
Feb. 2004 Diplomhauptprüfung
Diplomarbeit März 2004 – Dez. 2004 Institut für Organische Chemie der
Friedrich-Alexander-Universität
Erlangen-Nürnberg
bei Prof. Dr. J. A. Gladysz
"A New Class of Catalysts for the Baylis
Hillman Reaction: Chiral Rhenium-Containing
Phosphines"
Dissertation Jan. 2005 – Dez. 2008 Institut für Organische Chemie der
Friedrich-Alexander-Universität
Erlangen-Nürnberg
bei Prof. Dr. J. A. Gladysz
"New Design Strategies for Phosphine
Organocatalysts: Enantioselective Processes
involving Chiral Rhenium Fragments and
Recycling involving Perfluorinated Pony Tails"
Recommended