E第
Sci. Rep. Kanazawa Univ., Vol. 12, No. 1, pp. 21-66 June 1967
Studies on the Dating Methods for Quaternary Samples
by Natural Alpha-Radioactive Nuc1ides
I. 1ntroduction
Kazuhisa KOMURA * and Masanobu SAKANOUE
Radiochm伺:icalLaboratory, Deρar,初伽t01 Chemistry
Faculか01Science, Kanazawa University
(Received. 4 May 1967)
CONTENTS
1-1 1ntroduction …・・… H ・H ・...・ H ・....・ H ・...・ H ・....・ H ・・・・ H ・H ・H ・.....・ H ・.....・ H ・.....・ H ・.....・ H ・..,….231-2 History of the Excess 230Th and 231Pa Methods for Pleistocene Dating ・H ・H ・-….251-3 History of the Deficient 230Th and 231Pa Methods for Pleistocene Dating ・H ・H ・..251-4 Other Possible Methods for Pleistoc叩 eDating ....・ H ・....・ H ・....・ H ・-…....・ H ・.....・ H ・...261-5 Purposes of This Work ....・ H ・....・ H ・....・ H ・-….....・ H ・........・ H ・....・ H ・........・ H ・....・ H ・-…27
II. Theory of the Deficient 230Th and 231Pa Methods for Pleistocene Dating II-1 General ・......・ H ・...・ H ・....・ H ・...・ H ・.....・ H ・..........・ H ・.....・ H ・.....・ H ・....・ H ・....…...・ H ・....・ H ・.27II-2 Significance of the Uranium 1sotopic Ratio 234U / 238U・H ・H ・.....・ H ・H ・H ・..…...・ H ・..28II-3 Relation Between 230Th and Uranium 1sotopes 234U and 238U・H ・H ・H ・H ・.....・ H ・..31II-4 Relation Between 231Pa and Its Parent 235U・H ・H ・..…...・ H ・.....・ H ・H ・H ・.....・ H ・...…..33II-5 Relation Between 231Pa and 230Th・H ・H ・.....・ H ・.....・ H ・.....・ H ・H ・H ・.....・ H ・H ・H ・'"….'34II-6 Other Useful Methods for Age Calculation ...・ H ・.......・ H ・H ・H ・...・ H ・......・ H ・....・ H ・…-・35
III. Apparatuses and Chemicals 1II-1 Apparatuses...・H ・H ・H ・.....・ H ・.....・ H ・.....・ H ・…・…… H ・H ・.....・ H ・...・ H ・....・ H ・.....・ H ・.....・ H ・.361II-2 Chemicals …...・ H ・.....・ H ・.....・ H ・.....・ H ・-…...・ H ・-…...・ H ・.....・ H ・.....・ H ・.....・ H ・.....…...・ H ・.37
fV. Samples 1V-1 Purposes of Analysis ...・ H ・H ・H ・..…...・ H ・....・ H ・..,・・ H ・H ・..…...・ H ・....・ H ・...・ H ・.....・ H ・....371V -2 Sample Descriptions …・・… H ・H ・......・ H ・..… H ・H ・...… H ・H ・...0'........・ H ・..…...・ H ・..…・・…・・37
V. Analytical Procedures V-1 Sample Decomposition …...・ H ・.....・ H ・..…...・ H ・.....・ H ・-…… H ・H ・...・ H ・.....・ H ・.....・ H ・..…・40V-2 1nitial Concentration of Radioactive Elements with Hydroxide Precipitate ...・H ・・…・40V-3 1nitial Separation of Radium, Thorium, Protactinium and Uranium by Anion
Exchange Method .......・ H ・...・ H ・..........・ H ・.........・ H ・............・ H ・......・ H ・........・ H ・...・ H ・41V -4 Purification and Electrodeposition of Thorium 1sotopes …...・ H ・..…… H ・H ・...・ H ・..…42V-5 Purification and Electrodeposition of Protactinium 1sotopes...・H ・.....・ H ・.....・ H ・....….43V -6 Purification and Electrodeposition of Uranium 1sotopes...・H ・.....・ H ・.....・ H ・-… H ・H ・-…43
VI. Measurements....・H ・.....・ H ・..…...・ H ・..…...・ H ・..…...・ H ・..…・・…...・ H ・.....・ H ・-…....・ H ・...・ H ・..…・・45
V1I. Resu1ts and Discussion VII-1 Uranium Assay ・H ・H ・.....・
* Present address Otozai Laboratory, Department of Chemistry, Facu1ty of Science, Osaka University.
- 21-
22 Kazuhisa KOMURA and Masanobu SAKANOUE
VIII. DiscussIons on the Causes of Error VIII-1 Pipetting Error and th巴 Absolute Activity of 232U …""…..,0・H ・-ローH ・H ・..・..・ o・...…・57VIII-2 Impurities in Tracer Solution ..…・........日……"....・H ・-ー……口一……ー…・…h ・..........58VIII-3 Counting Errors.......園……........,・..,.....,.....,...・回目 …口ー・ …......園園 a・......…・ 8・0"_・H ・H ・.59VIII-4 Contamination.....…・・…ー…・・・……日....・H ・....,......……日…・……日…・…...・・……・………・61
IX. Comparison of the Results with Other Data ・H ・H ・..・・・H ・H ・.......,・H ・e・e・....…...・H ・.......…,,63
X. Acknowledgement....…・…日困層.....……u ・H ・H ・...…・・・…・・…………ー...・H ・-…圃…u ・…..,…..・H ・H ・.64
LIST OF TABLES
L List of the Coral Samples 2. List of the Trida例 aShell Samples
3. Analytical Results of the Uranium Content in Tridac飽aShells
(Comparison of the Analytical Methods)
4. Corrected 230Th and 231Pa Ages of Coral and Tridacna Shell Samples 5. Ex制時leof the Data Sheet (Coral Sample CK 5)
LIST OF FIGURES
1. a Three Natural Radioactive Decay Series
b Alpha Particle En日rgiesof th日 DecayS岳riesMembers
2. Radioadive Dating Methods Applicabl邑 forQuaternary Period
3. Model of the Introduction and the Removal of the Uranium Isotope活 fora System
4. Cha暗 記 in(r-1)/(ro-1) Values for the Sp巴cialCases of (i ) Closed System, (ii) Constant Inflow and (iii) Stationary State
5. Theoretical Growth Curves of 230Th and 231Pa
6. Contribution of fo Values to the Growth of 230Thj238U Ratio
7. Deviation from the Simplest Equation 230Thj234U=1-exp(-Ao t)
8. Change in 231Pa/230Th Ratio vefSUS Sample Age for Various fo Values
弘 IonExchange C口lumn
10. Geographic Locations of the Analytical Samples
11. Stratigraphic Positions of the CoraJ aud Trzdacna Samples 12. SampJe Location in Kikai司Jima
13. Elution Diagram of Ra, Th, Pa, U and Fe from Anion Exchange Resin Column
14. Electrodeposition Cell
15. Analytical Scheme of the Carbonate Sample
a lnitial Concentr旦tionof Radioactive Elements with Hydroxid巴 Precipitate
b Initial Separation of Ra, Th, Pa, and U from Iron Carrier by Anion Exchange
c Purification Method of Thorium Fraction
d Purification Method of Protactinium Fraction
邑 PurificationMethod of Uranium Fraction
16. Alpha Counting System
17. Alpha Spectrum of Uranium Fraction (Stalagmite Sample SC)
18. Alpha Spectrum of Thorium Fraction (Coral Sample CK 10)
19. Alpha Spectr日m of Protactinium Fraction (Coral Sample CK 11)
20. Analytical Results of Coral Samples
21. Analytical Results of Tridacna Shell Samples 22. Correlation between Pa-231 age and Th-230 age
23. Influence of th巴 UraniumLeaching to the Growth and Decay Prop位 tiesof 231Pa and 230Th
24. Analytical Results of Stalagmite and Water Samples
25. AvailabJe Limits of the Deficient 230Th and 231Pa Methods
26. Comparison of the Analytical Data with Other Data
a Results of Kikai・JimaSamples obtained by This Work and 14C Method (55)
b Results of Other Parts by 230Th Method c Solar Insulation Curve (62)
d 180/160 Terr明 rature(63) e Paleontologic Climate (64)
'27. Presumed Ancient S邑aL巴velnear Kikai-Jima
Studics 0珂 theDαting l¥IIethods for Quaternary Sanψles 23
1 Intr・oduction
1-1 Introduction
1n the history of the Earth, the creation of the human civilization may be the
most interesting and il11portant event. It l11ay afford the il11portant contribution
to our understandings both on the Earth's and man's history to know the circu-
mstances and the absolute date of the evolution of the hUl11an race.
The discovery of radioactivity, which appeared as one of the leading actor of
sci巴nceat the end of nineteenth century, not only upset the conception of the elements,
but a1so by efforts of a gr巴atmany scientist realized the dream of utilizing atomic
en色rgy. It was soon after the discovery of the radioactivity that the constancy of
the radioactive decay drew the attention of scientist as a natura1 clock.
The exsistence of the r巴lationshipbetween the leacl content ancl the geological
ages of the uranium minerals (Boltwood, 1907) had become the first cognition of the
usefulness of the radioactivity as a too1 for the dating of geological formation.
Though in following year Strutt propos巴dthe He/U method for clating of uranium
minera1s, the theoretica1 founcl丘tionwas not given until cliscovery of the displacement
law by Soclcly ancl Russell (1922). According to this law, the uranium isotopes 238U
and 235U and the thorium isotope 232Th decay through series of short-lived daughters
to stab1e leacl isotopes 20白Pb,207Pb ancl 208Pb, resp巴ctively.Fig. 1 shows these three
U-238 SERIES Th叩 232SERIES U同 235SERIES
20世Tl
Fig. 1 a. Three Natural Radioactive Decay Series
24 Kazuhisa KOMURA and Ma邑anobuSAKANOUE
AcU押$愚行告5
必、 i。l
"" CJ
<0
"" 4.39
亡コ勧司
Th-SerI告s U-Seri密事
お2ThI I 4.20 238U
一一
Fig. 1 b. Alpha Particle Energies of the D巴cayChain Members
n,atural radioactive decay series (Fig.
togeth色rwith the alpha particle
energies of the members in th自己
chain s巴ries(Fig. b)固
By the invention and the develop-
m日ntsof mass spectromεtric techni同
ques, the dating method using long
lived radioactive isotopes other than
uraniurn加 dthorium had made great
advances. They are 87Rb咽 87Srmethod
(Ahrence, 1947), 40K-40A1' method
CAldrich and Nier, 1948), 187Re-1870s
method (Herr and j¥在erz,1958), 40I{開
制 Camethod (Polev註yaet al., 1958)
etc.・Inspite 01: the wide
of these 日 theiravail呂田
bilitiεs w告relimitted to the sample
older th旦n several million y邑ars
because of the half寸ivesof the
parent nudidεs. 80, the dating mト
thod applicable for Quat邑maryage
was sought
The problem of Quaternary age
determination was solved by an
unexpected mεthod based on the
cosmogenic radioactive carbon isotope 14C(1), which is at constant rate in
the upper atmosphεre 14N p) 14C reaction. Radiocarbon produced from the
nitrogen in the atmosphere 1S soon oxidized to 14C02 and mixed w1th norトradioactive
CO2 in air, and then the circulation of the radiocarbon among the atmosphere,
and begins, Once 14C is fixed in the organisms or geological
formations containing carbon, the change in spεcific activity of 14C will indicate
the time from the end of the exchange of carbon b色tweenthe sample and its environ-
ments, unless the carbon exchange occurs during the sample As the
問 rbonis one of the most important constituent of the organic m乳terialsand has
high abundance in nature, 14C dating method is applicable nearly a11 of th邑 sampl己S
containing c乱rbon圃 However,the r邑lativelyshort half国 lifeof :J.4C (5730 limits
the available range of this method back to about 40,000 years, and the gap b色tween
million years determined by mass spεctrometric methods and seγ巴ra1ten thousand
years radiocarbon method still remained to be filled with other useful method
for dating.
Stlf.dies 01'1 the Dating且ifethods 101' (Ju,af,ωnar.v Sa狩ltles 25
[-2 History of the Excess 230Th (Ionium) and 231Pa (Protactinium) 1¥1ethoc1s for
PleistocenεDating
From th告 viewpoint of half-lives, 230Th (75,200 yr) and 231Pa ,480 yr) are
very aUractive for Pleistocenピ ag己 determination. This was pointed out first by
Khlopin (2) for the age determination of seconclary ur旦niumminera1s, but had not
b巴己11re且lizedfo1' a long time because of the difficulty of the di1'ect determination of
230Th or 231Pa. The fact discovered by Joly (1908) that the radillm content in sea
sedimεnt cor巴 ismuch greater than that in terrestrial materials, was reaHirmed by
Piggot ancl Urry (3) in 1942, and this bec呂methe initiation of the excess 230Th
method (th巴 useof 230Th cl己C旦yin samplε) for Pleistocene clating of sea sediments.
This m色thodhad been improvecl greatly by the direct clet巴rmm乱tionof 230Th (4) and
by the aivances of the ap;J品ratus己sfor alpha spectrometric measurement. Though
日.1anyworks have b吉正~n carried out in this direction, the accumulation of the analy-
tical clata showecl th色 possibilityof th色 migrationof 2301、hitself in and out the
sample and the exc巴ss230T11 method was sometimes proved to be erroneous for th巴
dating of the seclim巴ntationrate of thεse且 seclim己nts.
On the other hand, the use of 231p旦 asa too1 for agεcletermination was put
into practice only recent1y by Rosholt (5) in 1961, bec乱us記 ofthe extremely low
concentration of 231Pa in natural samples ancl of its troubl邑 behavioursin chemical
treatment. Recently several investigations as 1:0 the excess 231Pa methocl use
of 231Pa decay in the sample) ¥vere carried out by the sciεntists mainly in U. S. A.
ancl lJSSR. Sarma (6) investigatecl the accumulation rate of the cleep sea sεcliment
by llsing excess 231Pa and 2:>OTh methocls and confirmed thεusefulness of the ratio
231Pa/230Th for clating. Furth匂司 investigationby Sackett et al. (7) showed that the
excεss protactinium method afforcls much mor巴 reliabledata than th巴 εxcess230Th
method, if the clata obtained by thes日 twomethods were compared with radiocarbon
data.
1 -3 History of the Defici巴nt230Th and 231.pa Methods for Pleisto田 neDating
Since Barns et al. (8) hacl shown the possibility of using the degree of the
d邑ficiencyof 230Th with resp巴ctto its parent 234U as an age indicator for coral
cuttings of P品cificatoll, many works have been made. Sackett (9) analyzed many
marine carbonates and obtained the good agreement of thε230Th ages with radiocaト
bOI1 ages. Furthεr studies by Tatsumoto and Golclberg (10) emphasizecl the contriblい
tion of il1itial 230Th ancl proposed the 232Th content as an inclicator for correction.
1n 1955, Cherdyntsev (11) made an import乱ntcliscovery that the activity ratio
of 234Uj238U is not llnity and 234U anomaly occurs rather commonly in natural
condi tions. Th巴 authorpointecl out the need of correction for age calculation by the
230Th growth method. 1n this respect Cherdynts巴v(12) mad巴 anage determin呂tion
of thεfossi1 bon巴scoUectecl frol11 th巴間mainsof anciel1t civilization (Kostenki, 1¥10闘
lodovo, cavεKuclaro, etc.) and obtained very useful r巴su1tsfor archaeological stuclies.
26 Kazuhisa KOMURA and Masanobu SAKANOUE
They had used the simple thorium isotope m巴thodusing the ratio 230Th/234Th
instead of direct det色rminationof 234U activity and the othεrm巴thodusing the ratio
227Thf234Th instead of 231Paf235U. They a1so suggested the possibility of the
dating method using th邑 decayof 234LJ (248,000
Thurber (13) confirmed the exsistence of 234U anoma1y a1so in sea water and
marine carbonates and proposed the 234LJf238LJ dating m巴thod,which may be ap-
p1icable back to about one million years. The ratio 234Uf238U obtained from many sea water samples showed a11 good agreement within the range of statistical error
(r= 1.15土0.02),which may be taken as th色 initial234U f238U ratio in a11 sort of
marin巴 carbonatesamples.
Though Rosholt et aL (14), Sackett and Potratz (15) had suggested the use of
231Pa/235U ratio for cross checking the reliability of 230Th dating method多 the
activity of 231Pa was too small to be measured accurately without developing the
detector with very low background activity and with high counting efficiency. Ac-
cordingly, other criteria had to be sought to confirm the assumption of the closed
system Ior 230Th d呂tingmethod. Broecker (16) and Blanchard (1'7) studied many
corals and molluskan shells collected from various regions in the world. Based on
th告 analyticalresults of U /Ca, Ra/Ca and uraniurn series dis巴quilibrium in the
sample, th色Y 巴mphasizedthat the 226Raf230Th ratio was very useful criterion for
confirmation of the c10sed system. Recεntly, deficient 230Th dating have been
applied to various rnaterials as follows;
Fossil bones (18血 20),Travertine (19, , Peat bogs (21,23),
Marine and non -marine mollusks
(18齢 20,22,24同 30).
E曙 4Other Possible Mεthods for
Pleistocene Dating
1n addition to the methods merト
tioned above, uraniull1回heriummethod,
which was advanced greatly by F anale
and Schaeffer (31), the methods using
cosmogenic r旦dionuclidεssuch as 36 Cl
(3.1 x.l05yr) (32), 26Al (7.4 ><105 yr)
(33), and lOB巴 (2.5xl0自 yr)(34), and
fission track method developed by Price
and Walker (35) have the possibility
for the dating of Pleistoc巴neepoch.
However, all other methods except
uranium-herium and fission track meth-
ods seem not so promissing, because
the extremely low abundances of these
102 103 104 105 106 107 108 109y -了ー叩一戸ー下一ーーー,---.----r---y---,r-~!~一一
1ゐCdec日y
mpqF;31h 23'Th ~;~司Y
自民附ih
234U decay
4 He growth
40Ar growth
238U 草p.fiss.
36Cl growth
10Se d曹cay
18705、吋 r~growth 40Cal Lead 詞010
一園田 Alr,開 'dyApplied 圃岨圃圃 Possible
This外匂rk
This Work
Fig. 2 Radioactive Dating Methods Applicable for QuaternarγPeriod
Studies on tJze Dating M ethods for Ql~aternary Sam戸les 27
cosmog巴nicradionuclides in nature makc it difficull to obiain the reliable analytical
resu1ts of these nuclides.
In Fig. 2, all the methods cliscussed above are summalized.
I句 5 Purposes of This Work
The principal purposes of this work are followings :
i. To establish the simple and reliable analytical method for uranium,
thorium ancl p1'otactinium il1 ca1'bonate sample.
ii. To develope for the fi1'st time the deficient protactinium method as an
age inclicato1', ancl to confi1'm the limit ancl the availability of this methocl
fo1' Pleistocene clating.
iii. To apply 230Th ancl 231Pa methocls fo1' clating of ma1'ine ca1'bonate sarrト
ples collectecl from Ryukyu Islancls, in o1'cle1' to obtain some infonnations
abollt the climatic ci1'Cllmstances in Pleistocene epoch.
iv. The prelimina1'Y evalllation of the valiclity of the stalagmite fo1' clating
by 230Thj234U ancl 234U/23SU methocls.
II-1 General
n. Theory of the Deficient 2301、hand 231Pa
Methods for Pleistocene Dating
There a1'e three fllnclamental reqllisites to be satisfiecl for the sake of obtaining a
reliable age by cleficient 230Th ancl 231Pa methocls. They are as fo11ows.
i. The1'e exsists measurable amollnt of uranillm in the sample.
ii. It is clesirable that 110 initial 230Th ancl 231Pa exsist in the sample at the
time of living, fossi1izing or mine1'alizing of the sample. If these nllclicles
exsist initially in the sample, the amollnts of them ShOlllcl be negligibly
smal1 comparing with that accllmlllatecl by the raclioactive clecay in the
sample cluring the exsistence, 01' the initial vallles can be correctecl by
approp1'iat色 means.
iii. There a1'e no migration of parent u1'anium isotopes ancl/or their clallghters
after the cleposition of the sample. In other worcls, dllring the exsistence of
the sample, there occurecl no action to alter the isotopic composition exc色pt
by the processes of 1'aclioactive clecay ancl growth.
The first conclition was satisfiecl in any samples analyzecl as shown afterwarcl,
becalls己 of 1'elatively high abllndance of ll1'anium in ma1'ine or terrestrial mate1'ials.
The o1'cle1' of the u1'anium contents we1'e O. x to 3 ppm fo1' co1'al samples, O. Ox to 2
ppm for Tridacna shell samples and O. Ox to O. x ppm fo1' stalagmite sample, the1'e圃
fo1'e, the app1'op1'iate quantities of uranium we1'e assured by taking sufficient amounts
of sample.
28 Kazuhisa KOMURA and Masanobu SAKANOUE
The secondむondition is the most important one for the deficient 230Th and
231Pa methoお Howev巴r,it sεems as a mattεr of course considering thεdifferences
of thorium and protactinium in the chemical beh旦viours. Wh色reasuranium
lS Vεry soluble forming c品rbonat巴 orsulfate ([U02(C03)ZJ2-,
(C03 ,or [U02(S04hJ2-, in the pr己sence of carbon丘t邑 or
slllfatεions in sea or water, thorillm and protactinium are scarcely soluble in
such conditions. As they arεhydrolyzed to forming hydroxid邑 colloidsor have large
tendencies to bεadsorbed to the colloidal materials, thes邑 elementswill soon dεposit
with matter in the water. Hence, the separation of parent uranium from
its daughter nuclides may occur in natur叫 conditions. This was confirmed by stu圃
the contents of llranium, thorium 叩 dprotactinium in sea water and
natural waters Furth巴rmore,the exp巴rimentson the relative 1εョchabilitiesof
these己lements(38) proved the sεparation of uranium from its daught巴rsin vεry
mild which might exsIst in natural conditions.
Ti1e third condition is a1so for because either出巴
of parent uranium isotopes or that of their daughter nuc1ides distorts the ageξstima帽
tior: the ratios of 230Thj23壬U and and sometimes may make it imω
possible to estim乱防白色 ageof th巴呂如 Thecrit邑rionsfor ascert引ningthe good
。fthe sample are listed as follows:
i. S乱mplehas 100タti structure (meta-stable state of ca1cium
Exsistencεof calcite structure st呂teof calcium carbonate) indicates
the of altεration, except for th色 samplewhich s己cretes
calcium carbonate as calcite structure. This can be proved X-ray diffrac四
tion analysis or by the chemical examination using Co as testing
rεagent.
iL There are no observational proof of weath己ringsuch as the
。rcolour changes事
structure
iii. Homogεn己ousdistribution of the micro同 compon邑ntssuch as uranium and its
daughters, strontium, magnesium, etc. in carbonate sample. This can be exa耐
mined emission spectroscopy, electron micro analysis, activation analysis,
actiγation autoradiography, etc..
iv. of the ages estimated several independent methods such as
230Thj234U, 234Uj238U, 231Paj230Th, HejU, 14C
methods etc.
If th己seconditions are satisfied in the sample analyz邑d,age calclllation becomes
possible using the r記lationshipsamong parent uranium isotopes and their daughter
nuclides as describεd b色low.
U-2 Significan(万 ofthe Uraniurn Isotopic Ratio 234Uj238U
As pointed Ollt in I-3, the uranium isotopic ratio 234Uj238U has important
meanings in uraniu111 series disequilibrium method of dating, therefore, it s邑巴111Sverγ
Studies on the Dαtin!l 1v1 ethods for Quaternary Sa例 ples 29
llseful to considεr th色乱nc色 ofthc raしioof ~234Uj2 381J in stuclies
。fthe uranium isotopes. Let us consicler at first the general aspect of uranium
isotopic ratio 234Ul'l38U, and then discu錦 町V邑ral
model case illustratec1 in Fig. :3 that the uran.ium
stantly inflU'ivs into a system a
戸
U
M
τnn
fし
O:
β
U
3一
日
。U
2
s
i
n
h
u
-
-
γ
L
a
o山
う
〉
-L
y
t
(
1
7
a
ウ
111A
Y
A
t
V
3
4
1
2
、
1
司
馬
正
E
r--u
TeAA
U
己
B
S
う
Q
L
S
M
U
ρhnMM11
8
/
/
t
4
a
T
J
「し
γh、JiA
H
閣
のM
r
¢
decay 01' are remOi、/日clout from the system (for example, dεpositional removal) acconl時
ing to their half-lives 01' to th色 r邑sid巴ncetime , respecti vely, thεrelation民 twε:en
parent 238U and its daughte1' 234U can be expr色ssedas rollows:
dlVs/dt ~= aー(As十 t)Ns
=b →. AsNs -04 +β) ( 2 )
¥3ノ
ー/¥
whεre lVs, N4: number of atoms of 238U and 234U,
a, b: cO!1stants, which denote the amounts of 238U and 2:31U (in atom
number) introducecl in a given system,
118, A4,: d己cayconstants 01 il38U and 234U,
長':removal constant of uranium (depositional const呂 whichls equal
to the reciprocal of th己 residencεtime(τof uranium in a
system.
By integrating the Eqs. (1) al1d (2), the solutiol1s of (3) and (1) can be
obtail1ed if we give
tem consider忠dhere;
ancl N2 as the initial values of Ns and N4 in乳 sys-
『曲お
F勾 14生噸
,l,sNa
Fig. 3 Model of th色 Introductionand the Removal of th邑 UraniumIsotop日S
for a System
Ns = {a/(何十ρ)}{1-e一(VF)t}
トトlJ)t ( 3 )
={品/(九十ρ)}{1 -e-(以内t}
1-' {{lふ/(A"ート Jう)(九イ
{i-t一(J~+P) t'}
よ呂寸 ρ)(礼-
(,+P)t -e-C"一川 t}
ート {N~ ん/(À,~ - ,;呂、)}一(、門戸
---e-(;;t叩 )t}
→ e一(けnt (4)
Eqs. and (4) indicate the atom
numbers of >l38U and 234U at the
given timε t after the 史 ofthe
uranIum introc1uction. Th巴
general above can bε
simplified by assuming the spεcial
cases, which may b色 probab1ein natural
conditions. Th己y are the iollowing
cases.
30 Kazuhisa KOMURA and Masanobu SAKANOUE
也旦(1) I Both introduction and removal of uranium isotop田町ptfor the
radioactive decay do not occur, namely the requisite condition for the closed system.
This may be held in wel1 prεserved marine c昌rbonates(such as coral, molluskan
shells, oolitεs, secondary uranium minerals, peat bogs, fossil bonεs (団内叩iti圃
zed), etc..
Now, a, b and p in Eqs. (3) ancl (4) are put to be zero, (and '¥8' .0), simplified
equations of (5) and (6) can be obtained.
Ns =N~
N甚ニ (N~ ,18/ ,14) (1 - + N~ e-},t
( 5 )
( 6 )
Then, th己 activityratio r=234Uj238U at time t is expressed as follows if we put
the initial 234Uj238U ratio as rO=N~À4/N~À8
r = 1 -(ro -1) e-,',t
or (r - 1)/(1'0 - 1) =e-;,t
( 7 )
( 8 )
日五五万I了 Thede伊 sitionalremoval of ura山~n is small compared
with the amounts of the introduction. This case was assumed to be true for the
dating of the lake formation of Issyk狸 Kul.and Chatyト kulin Kirgizskaya USS衣 (39),
which had b,:oen believed to be formed in late Pleistocene εpoch, and the re乱sonable
clata were obtained. As N g , N~ ancl ρin Eqs. (3) and (4) can b色町glected,simpli-
fi己dsolutions of Ns ancl N4 are given as follows.
lず8 = at
N4 = aJd/九イ(る -aAs/ ,14) (1 -e-l/)/ A4
( 9 )
(10)
Since bJ'4/aAg can b色 substituteclby ro, the value r is expressecl as Eq. (11) (the
activity ratio of uranium isotopes in the cas巴 ofconstant introduction),
f ニ 1十 (ro-1) (1 -e-;,t
or (r - 1 向 -1)ニ(1-e-' 九t
(11)
(12)
[Case (III) I The deqosi出 nalremoval of uranium is g告ochemically bala悶 d
with t.hεamounts of th邑 uraniumintroducecl into a given system (a basin), namely,
the case of station乱rystate. This is perhaps held in ocean water system. In this
case, the solutiol1 yielcls the residence time of uranium in a given basin. Accorcling
to the assumption macle just above, w巴 canconsider dNs/dt and dN4/dt in original
equations (1) and (2) as zero, therefore, a and b are expressecl as follows.
a=(As+p)N:呂
b = 04十 P)l叫んNs
(13),
(14).
If the resiclence time of uranium has the value comparable to the half幽 lifeof 234 U,
namely, the order of 105 years, we have the simplifiecl r value of
r = 1 + (ro -1) PI (九十Jう)
or (r - 1)/(1'0 - 1) =P/(A4→p)
(15),
(16)巴
Studies on the Dating Methods for Quatcrnary Sa目ψles 31
Consequently, the residence time (τ) of uranium can be ca1culated from the next
equation:
τニ 1/ρ=(ro -r)/(T -1) 1/ A4 = (ro -r)/(r - 1) X 1'1/2/0.693 (17),
where T 1/2 denotes the half-life of 234U.
If activity ratio 234U j238U introclllc巴dby ri vers into the ocean is assumecl to be
YO = 1.30, the resicl巴ncetime of about 3.6 105 yr can be obtainecl. This result doεs
not contradict the vallle recognizecl today ( 5 >( 105 yr).
Variations of the (r-l)/Cro-1) values given by Eqs. (8), (12) and (16) are
shown in Fig. 4 respectively, the sample age t or the residence time being tak己主1 011
abscissa.
{詰)1.0
回 ~ 岡 崎 町 旬句勾瞳碕也-亀『
7トー一一一一+-.1 ト叶、、
下両 ト9 It"..
601
フ ドキミ 刊~dt 馬要ぷ代 CQ。
可性議忌百
ト代d去'[jf'十A
内¥ トぜ ARP O
¥ i¥
3
70
50
d
F、k トーートー ドー
¥ ¥ 3
。 |¥ 、¥¥トーー
h 唱え「句、2
「¥ -.......;ミiト、予1五言
~ 10
」一一」ーL.!- :三近出回」んJ0.00 1巧ー 2 づ一子 6 7 s 910y 5 6 7 d 910を 2 3 5 6 7 6 910子
Sample Age or Residence Time Fig.4 Change in (r-l)j(ro-l) Values for the Special Cases of (i) Closed System,
(ii) Constant Inflow a叫 (iii) Stationary State
日告 Relation Between 230Th ancl Uranium Isotopes 234U and 238U
As the clifference of half四 livesbetween 230Th and its direct precursor 234U is
relatively small, the ratios of 230Th to parent uranium isotopes change with time
somewhat more complicateclly than th巴 ratio231Pa to 235U cloes. By assuming ro
as th巴 initialvalue of 234Uj238U ancl r as its value at time t, th色目tios230Th/
238U and 230Thj234U are expressecl as Eqs. (1 8)~(21):
230Thj238U = 1 -e-iot + Cro -1) (e-/.t - e-"ot) >(Ao/(Ao-A4) (18)
= 1 -e-,'ot + (r -1) {1 -e-C.-},)t} X Ao/(,-¥o - A4) (19),
230Thj234U = (1 -e-,'ot)/r -1-(1-1/r) {l-e-Cトケ)t}X AO/OO-A4) (20),
=l-eーよJ寸 (1一一 1/1')[{1 -e-O.-},)t} >( Ao/Oo一九)
(1 -e-'ot) J (21),
where A4 and Ao denote the decay constants of 234U ancl 230Th.
32 Kazuhisa KOMURA and Masanobu SAKANOUE
100/.. 90 80 70 60
50
40
3
7. eQUIl.)
一 ---と二玄l7ue:p¥v、、
,〆
一/----'7'
/ ¥ / レノ/
Tii旨ロシ
L〆lぷトipe
レ/1/
¥ ¥ ¥
J I'n311 2 3 5 6 7 8 9 fn4v 2 3 5 6 7 8 9,,,5,. 2 3 56789
30
20
nunsnu
1
nukJJaマ
2
Sample Age
Fig. 5 Theoretical Growth Curves of 230Th and 231Pa
(% equil.J
1銘80 70 60
50
40
2
g80 g 70 60
ど50
40
ど4 30
L〆~ 20 a
~ 長司、、
/ 司、“卜¥
長ミ/ ~ レ〆、、、
y / レ〆 ー、‘4
/ぐ レ〆レ〆
〆--------890 0ー
顔____.L
'b燐LlL /〆//
70-ユ戸,ィvこ、l4b'他州60-
〆d50 リ・
40
30
,~ jシ F会〆20
~ v
r
200
30
20
10
8
/00
Rvkiv
,a守
3
!:fシ 正面5 y 03 Sample Age
Fig. 6 Contribution of ro Values to the Growth of 230Th/238U Ratio
,st!{.dies OJ1. the Dating M cthods fo1' (Juaterna1'Y Sm日βles 33
Fig. 5 shows th色 changεof2301、hj238Uratio with sample age in the case of
1'0 = 1. 00 or 1.15, which is most probable value for the marine carbon呂tesamplεs 乱S
analyz 任1in this work. In Fig. 6, th廷evarious gro仰1市vパγ叱叶thcurves of 230Thj238U 1'a抗J土tiめoa訂r‘e
i凶llu出乙st廿ra批t台吋dfor 仁dω1日if行fp仕r司er批1討1.tl'均凸 val卸¥.!閃 tωoε伊合悦the町rWl抗tht出hegrov杭fλ川正
1'0工=ニ3.00. Furthermore, for convenienc色, the d巴viationsfrom the sil11plest邑quationof
230Thj234U = l-e-ヘtare calculatεd in two ways and shown in Fig. '7: (i) the rela司
tive∞rrection factors by vvhich the measured 230Thj234U of a given sample should
be devided to obtain a corrected 230Thj2341J value to be used in the above mentioned
sil11ple equation, (ii) the absolute excess in p色1・ cent, れアhichshould be substracted
from the measured 230Th/234U (浴日午lil.)value to olコtain a corrεct巴d230Thj234U
value.
Gorrectlon Fac!or (.......-一柏崎J1.20 一一寸ー
1.18
1.16
1./4
1.04
1.02
n同4
2
Fig. 7
し一一一一一一一
つ
Age
-u Deviation. from the Simplest Eq1.1ation. 230Th/234U=1-e
Relation Bet,耳7een231Pa and Its Parent 235U
16 I
14ι
" 12 " lu
-110 ~ 弘J
K
18 i!;
ー 1--16 告ν3 4 司
3 F d
4
2
。
Differでntfrom the case of 230Th, the relation between 231Pa and its parent 235U
is very simple, because the half冊 lifeof the intermediate nudide 231 Th (UY) (T li 2
=25.6h), is short enough to consider its activity to be always inεquilibriul11 with its
parent's. So, the rel乱tionis expressed as following equation without considering the
activity of the intermediat邑 nuc1ide,
231Paj235U = 1 -e-},t (22),
where ,11 denotes the decay constant of 231Pa.
As the direct determination of 235U is very difficult, its activity is usualJy
ca1culated from the 238U activity by assumIng the cOl1stancy of the ratio 235U/238U
34 Kazuhisa KOMURA and Masanobu SAKANOUE
in natural uranium, From their atomic abundancεratio 1/138 and e乳chhalf-lifε,
the value of .7 Is genεrally accepted for this ca1culation. Hεnce. the activity
ratio 231Paj235U is calculated by the formula given below:
二 21>7 >< (231Pα = 231pα
In Fig. 5, the growth curve of 231Pa Is illustrated together with the growth設nd
dec呂ycurves of 230Th and 14C,
H-5 Relation Between 231Pa and 230Th
The ratio 231Pa/230Th hぉ th己 interestingmeaning for the age determination
using the uranium加 dactinIum s色riεsdis叫 sincethis ratio offers the
important informations about the secondary alterations of the sample as discussed
later. Furthermore, thεratio 231Paj230Th becomεs one of the indicators to examine
the reliabi1ity of the巴stimatedage by comparing the agεs obtained from independent
three methods, i.ε230Th, 231Pa and 231Paj230Th methods. However, this ratio
on th巴 ratio234Uj238U 01' on the巴xsIst巴nceof initiaI 231Pa 01'
230Th in original sample, and shows very complicated feature in 五 deficient
231Pa and 230Th methods.
Whεr巴asthe 231Pa/230Th ratio in excess 230Th and 231Pa methods is expressed
by a simple equation such as Eq. this ratio in deficient methods has complica同
ted form as expressed a general formula of Eq, (25):
(A,R, in .quil, unit)
2~ 長¥ ~ ヘ
コ 1"0 ~ LOO 同電量....
内丸一22
。 ¥; ~ I"o~ 1,15
町晒勘当町
対同口
ヲ同--- 五五
ro ~ 1,30 内》
砲、埼碕崎町同 ¥ ¥
1"0 ~ i,50 十町-- α ト、 K r--.... NTi" 貯又 竺区
ru ~ 1,80 、‘診、、、 ¥ フ
『口:2,00 -、、 h 凡¥¥ ¥、、 ト~
慢-歯周
通 Z適'-~ 、同、、 、、h --語調ro~ 2,30 砲、..,.‘- 、、 『¥ ----同司h 円、同
-明 -『口~ 3,00
同トh ¥よ戸戸町同h 豆』同 --相
v-J〆- ,..が/'
2
d
0,6
0,/,
Q
F 2 3 4 2 5 6 7 d 916会 IO~ 10シSampie Age
Fig. 8 Change in 231Pa/230Th Ratio versus Sample Age for Various 1'0 Values
Sfudies on the Datnig j1,;[ethods for Q仰 ternarySamPles 35
i. excess 2alPa and 230Th method,
(231Pa)ex/(2301、h)ω =(231PaO)口/(2301‘hO),x e--()'づけ (24),
ii. deficient 231Pa and 230Th method,
231Paj230Th = (231Pa/235U)/(230Thj238U) = (Eq.22)/CEq. 18) (25).
1n Fig. 8, 231Pa/230Th to time relationship (Eq. 25) are illustrated for various
ro values. As seen from this figure, the ratio 231Paj230Th defined by Eq. (25) falls
from 2. 31 to 1. 00 at ro = 1. 00 However, when the initial 23品Uj238Uratio CrO)
increases, this ratio shows somewhat such strange feature that the value initially
decrεases and then increases to th日 equilibriumvalue 1.00. Besides the complicated
feature of the ratio 231Paj230Th, the difficulty of precise de臼rminationof 231Pa
activity limits the application of this method, but the importan田 ofthis r呂tiofor
cross開 checkingthe estimated ages must be still emphasized h巴re.
n明 6 Other Useful Relations for Age Calculation
In addition to the relations discussed above, the isotopic ratios 230Thj234Th,
227Thj23品Th,and 226Raj238U are a1so very useful for the age calculation. They are
related to their parent isotopcs as follows:
230Thj234Th = 230Thj238U (26),
227Th/234Th = 231Paj238Uニ=(231Paj235U) (εquivalent unit) (27)
and 226Raj238U = 230Thj238Uー (e-.'ot-e-l-ot)AO/(}.6-AO)
一 (r_.1) {AoAoe-i.,t/Oo - A4)(A6 九)
AOA4e-A,t/().o - ,14) (;16 - A4)
A6AOe-Aot/06 九)06 -Ao)} (28),
where九,Ao and ;{6 denote the decay constant of 23盛U,230Th and 226Ra, respecti“
vely. When sample age is greater than 104 years, Eq. (28) must practically agree
with 230Th/238U given by Eq. (18), unless the migration of radium occur during the
tim巴 t,because the half圃 lifeof 226Ra (1,622 yr) is relatively short in comparison
with that of 230Th.
This ratio becomes one of the useful criteria to examine the validity of the age
estimation (16, 17) as pointed before, although the ratio 226Raj238U cannot afford
any information about the secondary alteration that might hav巴 occuredmore than
several thousand years ago.
III Apparatuses and Chemicals
111-1 J¥pparatuses
Electric Centrifuge: ModeトFT9S,Kubota Seisakusho Co., Ltd ..
Radiation Detectors:
i. Gamm
36 Kazuhisa KOMURA and Ma邑anobuSAKANOUE
Scaler TEN 1000 scaler, Model Kobe Kogyo CoけLtd.
iL GJVI Counter: End叶νindowtype GM tube with 2グ diam.eter町 Aloka,Model
window thickness 2, 4~-2 ⑬ 6 BG 28 30 cpm.
iiL D巴tector:ZnS with 2グ
Scaler: Alokaム!J:odelTDC~l , Nihon Musen Co., Ltd.
iv. Double Gridded Ionization Chamber: IYlodel No. 4, Osaka Denpa Co., Ltd.
R色solutionPower: 1.8% (FWHM) for
BG: O.02~, O.OLi .4-0. MeV
Counting Efficiency: 2π
v. 100 Chann巴1PulseAnalyzer: Model A.N園100,KobeKogyo Ltd.
HI同:2 Chemicals
All chεmicals are the guaranted reagent except for the organic rεagent Dトiso∞
Butyl Ketone and Tri-iso・ Amine(TIOA)。
Acids: HN03, HF, H2SO"" HZC204 and their
8a1t So1utions: 2JI;[ HCOONI-14,
A1C13 solution in 8Ll¥lHCl,
10 Fe3+ carrier S乳turated
for Solvent Extraction: DIBK pre四巴 εdwith 8NHCl, TIOA
Kao Soap Co., xylene solution.
10n C2cu"au'6忠良esms:
10n
Anion Exchange R,ε8in: DO'Ne叉ω1x 8, 200~400 Oxalate Form
As th己 resinobtained is not pure enough to be used as it
the resin was swelled up in hydrochloric acid 801ution and washed in a glass
column sufficiεwith 8NHCl until the effluent After
with distil1ed water, th色 resinwas converted to oxalate form by
passing sufficient amount of O. 51V1 HZC204 s01批 ion.
Cation Resi.n: Dowex-50 x 8, 100~~200
Thεcation resil1 wasξd
in the same manner
as the case of D.n anionεL
Column:
The ion column vifas made of hard
glass乳ndshaped as shown in 9. 1~ cor司
下osivεeffectof acid on glass is
negligible at the con白 htr叫 ionof fluoric acid
us邑din this work (0. rfhou疋hf10w rate
was not controlled in any case, it
was nεconst:mt (0.1~, O<2 ml/min in
anio孔巴 0.4-0.5mljmin in cation
exchang巴).
H-form
ト斗5句自 rnm
Resin
GIα5S Wool
Fig. 9 Ion Exchange Coiun且日
Stttdies OIZ the Dating Methods /0γQua的?拍rySm対ltles 37
IV Samples
IV -1 Purposes of Analysis
The Quaternary coral 1imestones that remain in Ryukyu Islands can be classified
into three or four m色mbersaccording to the g色ologicalview points. Particularly in
Kikaiづima,these members have b日eninvestigated in detail by Schlanger and Konishi
(40) and tentative discrimination was made as (i) Raised Coral Reefs, (ii) Ar叫 :1
Limestones, (iii) Upper Member of Ryukyu Limεstones and (iv) Lower Membe;r of
Ryukyu Limestones. It is w巴11known that the growth and the propagation of coral
depend greatly on thεtemper司turein sea water, 1n other words, on the climatic
change of the Earth (warm or cold). Neighboring Ryukyu Islands is the present
north boundary of reef forming coral dwelling, therefore, the climatic conditions seem
to have influenced more CI・iticallyon the growth and the propagation of the coral in
this region than that in tropical area and the coral limestones are the foot四 printsof
the warm periods in ancient ages. Hence, an age determination of these four members
of coral limestones will make it possibl邑 toobtain many important informations about
the ch呂ngeof thεtemperature in sea water in this region of Pacific Ocean d1.1ring
Pleistocene epoch. And this may afford the lock to s01 ve the ages and th己 periodsof
the glacial証dvancesand withdr唱awalsin late Pleistocene.
Besides the coral samples, Tridacna shells and a stalagmite were analyzed, too,
in ordεr to examin巴 theirvalidities as the radioactive dating sample by deficient
230Th and 231Pa m色thods.
Wat巴r s呂mples were
a1so ana1yzed for 234U/
238U ratio and their 1.1ra-
mum contents.
Fig. 10 shows the geo..
graphic locations of the
sample analyzed in this
work.
IV嗣 2Sample Descriptions
Stratigraphic positions
and the det丘iled sample
locations of the Kikai圃 jima
samples are shown in Figs.
11 and 12, respectively.
Coral samples except
CK 7, CK 9, CK 10 and
CO 1, have a11 aragonitic
¥28、
1一一一一一
30
b
Kume-jima匂句, !~OkÎnawa-jima ¥ ν • CjI件。
健一百折、
ITAIムN も Miyako-j ima 〓らザυ 凸二百 ベ 1らしトー
レ げもγ o 100 Y.,I",",,,o
Fig. 10 G巴ographicLocations of the Analytical Samples
stru.cture. Abs巴nc色 ofcalcite structur色 inthe sample was confiγmed by means of X国
38 Kazuhisa KOMU宜Aand Masanobu SAKANOUE
Kikai-jima Miyako-jima Okinawa幽 jima
~_TM-l r一一一""1XTG-l
ocωuo一日前問。一色
Shimajiri Group -Doan et al.,
1960
Shimajiri Group
MacNeil,
1960
Izena-jima
X CK-l CK・2I~ !~・ 2X CK・3 _一一一
Fig. 11 Stratigraphic Positions of the Coral and Tridacna Samples
Konishi &. lshibashi 1966
Somachi Formationl C:
一一一よhermatypicKo耐 hi&. coral Schlanger
1966 T; Tridacna
order to estimate the absolute ages of
these coral formations. They are listed
in Table 1.
Simi1ar situations as the case of
coral can be considered in the growth
and propagation of Tridacna, however,
the difficulty in finding out the fossil
Tridacna shel1s appropriate for dating
makes it hard to utilize them as an age
indicator of the limβstone formation.
The Tridacna shel1 samples are listed
in Table 2. TM 1 and TG 1 were ana-、丸ιg
lyzed to obtain the initial values of Kamikatetsu 91 平司
230Thj234U and 231Paj235U in Modern ~
sample. Among the samples, TU 1 and
TM 2 contain some calcite structure in
the outer bulb of the shel1. Therefore, both aragonite and calciぉ partwere analyzed
in order to investigate the inf1uence of the phase change on isotopic ratios among
parents and daughter nuc1ides.
ray diffraction analysis and no visual al-
terations such as weathering or colour
changes could not be observed in these
samples. Samples were col1ected from
the each member of these limestones in
Present Dune
Tohiyo Cape 斜掠お凡
田皿伽…向m(3m eme四回)
霊童NakazatoForm [げ7耐 eme唱e剖d)
区3針P問 s副聞制t刷 山問
盤警額塁 A削n、山叫刊叫山D血加u叩JO、
尽::S!!~問陥l白I R陶yuk句yuG酌r山 plTertiary Somachi Formj
Fig. 12 Sample 1心cationin Kikai-Jima (55)
Studies 011. the Dating Metlwds for QHatern.a1コISamPles 39
T乱ble1 List of Coral Samp!es
Code Sample I Eleva-出向 |NherjQographlcUni川叫 Sample Loc山 13 Species
CU 1 I 60-12-9 I R官cent(1ived) I - 25m I Between Naka-bishi and I Favia I Uku-bishi, Kume steciosa
CU 2 1示三 吋 Cora可証下一瓦市iムス孟一一一一一[孟示arustica
Rais邑dCoral Reef 2m W. of Araki, S. Kikai CK 2 I 65-12-22-1-2 Rais己dCoral Reef 3,....,41n Komiya, S. of shrine, I~山f1仰
Shiomichi, C. Kikai retiformis
CK 3 1 65--12-24-6-2 Raised Coral Reef 4~5m I SWW. of Nak呂zato,羽T.Kiaki
CK 4 I 66-5-10-2-1 I Araki Limestones 27m I 150m NW. of Araki, S. I Kikai
CK 5 I 65-12-23-6-2 I Upper Limestone 43m I 750m N. of Kamikatetsn, I ilca押thastr,仰Member I S. Kikai echi向。ta
CK 6 I 65--12-23-6-1 I Upper Limestone Member
CK 7 I 64-7-20b Upp巴rLimestone Member
CK 8 I 66-5-12-3 I U即位 LimestoneMemb巴r
CK 9 I 65-12司 22-8-4I Upper Limestone I Member
CK 10 I品目山[Older山 one
CK 11 I 66--5-10-14-1 I Top of the Upper I Limestone Member I
CK 12 I 66-5-10-14-2 I Top of th巴 UpperLimestone Member I
CK 13 I 66-5-10-12-1 I Upper Limestone Memb巴r
CO 1 1 64-12--4-5 1 Machinato Limes-tone
1 (N叩 moQnarry) I
43m I 750m. N. of Kamikatetsu, I Favites S. Kikai fi仰 osa
40m I 500m N. of Kamikatetsn, S. Kikai
40m I 500111 N. of Kamikatetsu, S. Kikai
50m I 500m NW. of Shiomishi, C.Kikai
150m I 750m E. of Shiramidzu, I Favia iC. KikaI magnistella
18m I 700111 SW. of Nakazato, W. Kikai
18m I 750m SW. of Nakazato, 羽T.Kikai
28m I 800m N. of Araki, W. Kikai
Table 2 List of Tridacna Samole
Code I Eleva帽 1N削除rI Geographic Uni1 ~'~'Ution I Sa町 1e1>ocations
TM 1 I Recent (lived) Om I Drift on beach, Ogami-jim呂
TG 1 I R日cenomI Drift 0n b巴ach,Miyagi-jima
TK 1 I Recent 2m I Colle氾tedin Raised Coral Reefs, W. of Araki, S.Kikai
T 1 1 I Yanoshita Limestone I high tide i Yanoshita寸ima
TO 1 I Yontan Limestone (Sobe Quarry) I I Nakagami, Okinawa-jima
TU 1 I Ohara Limestone 40m I Kitahara, W. Kume
T M 2 I Irabu Limestone 4m I Kogan印刷isaki,1.2Km NVv of I Hisamatsu, C. Miyako
40 Kazuhisa KOMURA and Masanobu SAKANOUE
sample having diameter of 18 cm near the root was collected in the
cav色 atIke-jima very ne品rby Okinawa-jima. Surface of it is brown and cocεntric
many rings are observed in cross section of the sample.
1'he wat色rsw巴resampled at the following places: cave “Yuhiヘmouthof Taibo
River and Shioya Bay づima)乱ndTsukumo (Noto陶 Peninsula)。
V旬 AnalyticalProcedures
v寸 SampleDecomposition
Except for several sampl記s,sample blocks of about 20-40 grams w日recarefully
selected from the sample of coral and Tridacna shell still in aragonitic structure and
were dissolved in 200~300 ml of 6N HN03 + O. 05N HF mixed acid in a 500 ml
beaker二1'hetracεr solutions of 234Th, 233Pa and 232U may
contain small amount of 2281'h and its daughter nuclides in the course of storing)
w邑readded to the sample solution to know the chemical in the and
The solution was th巴日 heatεd undεT an infra r日diamp and
left for several d品ysto attain the isotopic between tracer and
the in sample. If this solution contained significant amounts of undissolved
matter, then it was filtered off by passing through a filter paper Toyo Roshi
Ltd.), and the residue was decomposed with the filtεr paper, initiaHy
hεating with conc. H2SO品 andthereafter with conc. HCI04十 conc.HN03. The
resultant s01ution was to taken in 8NHCいO.lNHF mixture and
put together with the solution.
In the case of stalagmi t<色 sample,about 100 grams of blockεs was taken
and decomposed in the same manner as the case of coral samples,恥causeof its very
low uranium oontεnt and of the exsistenc宕 ofphosphate ion that interfere the recovery
of the thorium isotopes.
v -2 Initial Concentration of Radioactive Elements with Hydroxide Precipitation
After an attainment of the isotopic exchange equilibrium, th巴 solutlonwas trans-
ferred to conical beaker made of glass, and boiled to expel dissolved carbonate ions
in the solution and the hydroxide of iron was precipitated passing ammonia gas
free from COz. As the high ionic str色 ofthe solution makes the efficiency of
the copr邑cipitationworse, especially the coprecipitation yi色ldof uranium isotopεs,
bioled water was added to dilute the solution, and large excess of ammonia gas was
blowεd into th色 solutionfor the sake of bεtter coprecipitation of the trace elements.
1'he hydroxid巴 precipitatecontaining uranium, thorium, protactinium and many
other radioactive decay products was s色paratedby centrifugation in a polyethylεne
centrifuge tub己 andthen dissolved i11 20 ml of 8N HCl solution. The resultant solu同
tion which contains organic matter and small amount of calcium ion as well拠出合
radioactive elem開 ts,was transfεrred to the conical b悶 kerused previously for the
8tudies on th8 Dating lVfethods for Quatornary Sal1ψles LH
initial hydroxide precipitation, and evaporated to dryness on the hot plate. About 5
rnl of conc. H2S04 w乱sadded to the residue and heating was continuecl to conv色rtaIl
the s必tsto sulfate until the white fume of S03 ceased to come out. If the blackish
thick rnatter註ppeareclcluring the heating, heating was suspended and then O 1111 of
conc. HCI04 and conc. HN03 were added careful1y along with 出合研rallof the bε呂ker,
and the heating was resumεcl to clecompose the organic ll1atter until the solution dried
up. The residue was dissolved in呂bout150 1111 of dilute HCl by boiling the solution
long tim邑. If thεinsoluble mattεr still remain己din the solution, the suspended matter
was filt色redoff passing the solution through the filter paper (No 5C) and washed
sufficient1y ¥円th8NHCl卜o.1NHF solution. This residue was usually discarcled, b← cause the quantity of the radioactive elell1ents adsorbed on t11巴 fi1terpaper was neg同
ligibly small excεpt in t11色 caseof dealing with stalagmite sample.
In the case of stalagmite sample, thorium isotopes werεadsorbed strongly to the
unclissolved residue, therefOl占ふ carbonate fusion was carriεd out to cl日composethe
resiclue completely. Aftεr cooling the fusion products, it was tak巴nup with 3NHCl
and呂dd色dto the original solution.
Th巴 solutionthus processecl w品s boiled and the hyclroxide precipitation was
repeated twice and finally washed with the boiled water containing ammonia gas to
remove chloride ion.
All the proceclures were controlled by means of gamma countings.
V ・3 Initial Separation of Raclium, Thorium, Protactiniul11 and Uraniul11 by Anion
Exchange Method
The hydroxicle precipitate thus obtained was clissolved in O. 5M I-hC204 solution
as smal1 quantity as possiblε(usually about 5 1l11), and the resultant solution of
yellovvish gr巴encol.oured was passed through an oxalate form anion exchange column
(41), which retains uranium, thorium, protactinium, iron and polonium and pennits
radium, actinium, lead and alkaline巳arthelements to pass through. After washing
the column with 5 column
volumes of O. 5Jl!f H2C20品
solution, thorium, protac-
tinium, and uraniu111 wer巴
eluted stepwisely from the
column by passing 6 column
volull1es of 8N HCl, 4 coト
umn volumes of 8N HCl-:
o .1N HF and finally 10
column volumes of 6N HCl
イlNHCI04 mixture. Elu-
tion diagram of these proce句
dures is shown in Fig. 13,
Ra Pa U
一Cコ
7 12 16 23
I Unit in Free Column Volume)
Fig. 13 Elution Diagram of Ra, Th, Pa, U and Fe from Anion Exchange R巴sinColumn
42 Kazuhisa KO!VlURA and Masanobu SAKANOUE
which was obt乱inedby a model experim色ntusing the radioactive tracer of
233Pa, 237U and 59Fe.
In thξfirst step of thorium an yellowish green band of oxalate complex
of iron (III), [Fe(CZ04,)3]3-, is converted to brown chloride complex and the iron
band expands twice thεinitial width. Though widening of iron band occurs also in
the last step of uranium巴lution,elution of iron does not take place. If the hydroxide
PIでcipitatewas dissolved in 8N HCl solution, because of the exsistence of impurities
that could not be dissolved in oxalic acid solution, the column operation was begun
from the step 01 thor.ium elution.
All th己 procedureswerecontrolled by gamma countings and gamma spectrometry
of each fractiono Dεtection of the mutual contamin拭 ionor thorium and protactinium
is possible means of g乱mmaspectrometry even in the case of 300: 1 and 1 : 4 in
bet乳 activityratios betweεn them (42).
Although the ch己micalbεhaviours of polonium in the anion exchang色 pr目。cedur色
was not studied in detail, preliminary experiments indicated that polonium would be
still retained on the anion E己sm with the carrier smce po凶
lonium could not bεfound in品nyfraction oI eluate. If this expεctation is corrεct,
polonium will be recovered from the column separately from iron the fractional
elution using appropriate conc巴ntrationof nitric acid.
Packing( Polyethyl告ne)
Holder (8r口55)
Fig, 14 Electrod邑positionCe!l
Each fraction thus obtained were sub国
mitted to the final purification from the
interf邑rringelements Ior the electrode回
position process by, m巴ans of cation
exchange or solvent extraction tech由
mques.
V-4 Purification and Electrod巴
of Thorium Isotones
The thorium fraction which contains
oxalic acid driven out by hydrochloric
acid and small quantity of ca1cium, was
transferred to 50 ml beaker and treated
with conc. HN03ートconc.HCI04 solution
to decompose oxalic acid. White residue
containing calcium salt was taken up
with 5 ml of 3N HCl solution and passed through a cation exchange column (43)
(H-form) .which adsorbs selectively thorium, and other impurities pass through this
column. Thorium was eluted from the column with 10 ml of 0.5λ;J HZC20品 solution.
Oxalic acid in thorium f問 ctionwas decomposed by h色atingwith conc. HCI04+conc.
HNOa mixture as befor巴, and thεsolution was evaporated to dryness.
Slzdies on the Dating 1VI ethods for Quaternary S骨 nρles 43
Radiochemically puri:fied thorium was taken up v,ァith5 1111 of O. 5N HCl solution
by heating ca!川 lyon th,~ hot plate, and th己n 1 m1 of 0.5β![ J-hC20恥 4m1of 4N
NaCl and 0.8 ml of 2N HCOONH生 solutionswere added (44). The solution thus
prepared was transferred to a poly巴thylene 色lectrodepositioncell ((3 cm height of
polyethylene tubing with out-side diamet色rof a. 4 cm ancl insicle d iamet巴r01' 2.8
cm). A stainless steel disc ha ving cliameter of 3. 5 cm serv巴das the cathod, wh巴reas
a platinum wire as an 司node. The v己rticals己ctionof the electrodeposition cell is
illustrated in Fig. 14. Thorium was electrodeposited for 2~-, 3 hours at ccnstant cuト
r巴ntof 0.5 A. After an electrodeposition司 stainlesssteel cathocle was rinsed suffici崎
ently with distil1ed water, and clried under an infra red lamp. The electrodepositecl
plate was counted under a 2'1 encl叫日パnclowtype GM tube to determine the chenlical
yield of thorium isotopes. Th邑 overallchemical yield variecl from 10 to 85 per cent
(m色anvalue 45匁;).
V悶[5 Pllrification and EIectn
Pllrification 01" protactil1iul11 was carried out by a solvel1t extraction technique
l!sing DIBK-HCl system (45). To the protactinium fraction obtained by al1 anion
exchange procedur巴, 2 m1 of AICh saturated solution in 8N HCl was acldecl to mask
the fluoridεion, and protactinium was extracted with 4 m1 of DIBK by shaking
vigorously for 2 minut己s. After the cεntrifugation, an urganic phase was transferrεd
to another polyethylen日 tube,and th己 extractionwith 4 and 2 m1 of DIBK were
rep色ateclto assur巴 thecompletE‘ transfer of protactinium to organic phase. Protac同
tinium in organic phasεvvas washed ¥llith 4 m1 of 8ムrHCl and finally stripped to
aqueous phase with 4 ml of O. 5N HCl solution.
An electrodepositivn of protactinium fractiol1 was carried out by entirely the same
manner as th巴 C呂S色 ofthorium, Il1 ch己micalcomposition of electrolyt巴sand also
in conditions of electrodeposition, inst巴aclof the m巴thodrecommencled before (46).
Overall chemical yields of protactinium varied from 10 to 75 per cent (rnean value
4096).
v -8 Purification and Electrodeposition of Uranium Isotopes
The uranium fraction usually clid not need the further purification. However, in
the叩 sethat the amount of the residu巴 afterthe evaporation of the uranium fractiol1
was consicler旦blylarge, solvent ext日 ctionp1'o田 durewas carried out by using the
T10A時日Clsys民111as follows. The r己siduewas dissolved in 5 ml of 8N HCl solu-
tion and uranium was extracted with 3 ml of TIOA. Extraction was repeatecl twice
with 3 ml of TIOA and the uranium in organic phase was washed several times with
8N HCl solutiυll. Finally uranium was back extracted to aqueous phase with 5 ml 01
o .1N HCl solution and the solution was evaporated to clryn巴sswith conc. H2SOι-
conc. HN03 to decompose any organic matter mixed with.
44 I王azuhisaKOMURA and Masanobu SAKANOUE
一一よ竺39出
回目 cenfrifuge)
(… (wt同
a
-延
長-C, HC10l, ., C トIN03
(evap, )
イ-3N HCI v
t iOi7
(washing)
↓3N HCl
( e!ution)
I Q,5M H2C204
ITh I i←C, HC104 ←C, HN03
(evap, )
↓ (Elecfrodep.l
c
(Anion
持勢
!←AICI3(sCltd, in 8N HC1)
!← DIBK ( shake)
r-8N HC! )( 2
I 0,5N HCI
摂 自由
(Elecfrod牢
Fig. 15 Analytical Scherne of the Carbon岨teSarnple
exchan,CJe J
イ;ご世会3, 8上!HC1" Q,lN HF
t由州
4, 6,ti HC1,. lN l:iCLQ且唱団側持
b
持紅
白恥ーよ…
IOA
10rg.1
~
F也由 e
Studies on the Dating Alethods for Qz.atcrnar:v Samtles 45
The uranium thus purifiεd was taken up wlth 2 ml of 2N HCIO;t undεr the
h己ating,and 2.5 ml of 2lYI Hcoor可H,;, ω!utionand 10 ml of distilled ¥'1旦tcr-¥.:vere
add己dto make up th色合lectrolyticsolution, instead of the method配 velopecl Hashi四
moto and Sakanou日 (47). The solution thus p了色paredwas transferred lnc.o the elεC岨
trodεposition cell叩 duranium was electrod己 ona stainless st出 platεfo1'
about .3 hou1's at a constant current of 0.4 A. Ov邑rallchemical yield oI l11・ani.un
isotopes varied f1'om 20 to 70 per cent (mean value め.
All the purification methods clescribed above are shown schematicaIly :in 15.
VI MeaiJUl'elrUe11的
Chemical yields of thorium and protactinium isotopes through a11 the proc日dur巴
including the coprεcipitation, anion exchange多 purificationand electrodeposition w邑re
determined by beta countings using a 2'/ enふwindowtype GM counter, whereas that
of uranium was determined by m日ansof sptctrometry. Alpha spectrometr.ies
wer巴 carriedout by using Double Gidded Ionization Chamber (48) with very io¥v
back ground activity and with high counting efficiency. In Fig. 16, the apparatm;
used for alpha spectrom巴tryis shown. Enεrgy resolution of this chambεr was mεas同
ul・edby using uranium isotope 232U (E=5.318 MeV 6896, 5.261 MeV 32必)and
apparent value of 1.8-2.0 per cent in thεunit of FW.HM (full width of half
was obtained,
HTD-106S (e)
Fig. 16 'Alpha Counting System
司Jl
m
nu qJ
ぷ↑由民
Kazuhisa KOMURA and Masanobu SAKAN心UE46
ωFH江口「一
υ
的判
C30U
90
Numb唖「
Alpha Spectrum of U悶niumFraction (Stalagmite Sample SC)
Cho.rme!
Fig. 17
270 (1..68 M(:!V)
600
O ;:
55oof-4二u
100
(Derived from
22&rh 224投錨
叫~λNumber
40 ぷ息必③:
30
nu
円
U
円
υ
円
υ
n
u
円
unuw
円
U
門
U
'h叫内
JJ
《
d
t
e
』
剛
山
乱
闘
相
戸
Eh山
JOU
Chαnne!
Alpha Spectrum of Thorium Fraction (Coral Sample CK 10) Fig. 18
47 Sttldies on the Dating Methods for Quaternary Samρles
1330 rnin 「トし
{印5.05MeV) 231Pa
(4.72 MeVJ
50
30
20
10
Q.; 40 0.
mwccuzu
的判ロコ
ou
Channe! Number
Alpha Spectrum of Protactinium Fr呂ction(Coral Sample CK 11)
O
Fig. 19
Activiti四 of234U and 238U in thεsample were calculated from the sums of
countings in 20 and 18 channel r且ng色s,respectively. The ab心vementioned channel
ranges were chosen so as to give the 234Uj238U ratio of the old uraninite sample to
beし00土0.01,at the condition of 20 keVjchannel, and absolute quantities of 234U
and 238U were calculated using recov巴red232U activity. Th巴 activitiesof thorium
and pro凶ctiniumisotopes were calculated frorn the S1.lms of countings in ~22 and
~27 channel rang色s,respectively. The typical alpha spectra of the manium, thorium
and protactinium fractions are shown in Figs. 17, 18 and 19. Counting time ranged
from several homs to several days according to the 1巴velof the activities of the
counting sources. Countings wer己 contin1.ledso that the statistical errors of counting
were sεtt1ed within for 234Uj238U ratio and 230Th, and 55ぢfor231Pa.
Results and Discussion vn
Uranium Assay
In order toγerify the validity of the alpha spectrometric analysis of uranium
using 232U as uranium tracer, the uranIum concentration in Tridacna shell samples
was analyzed also by fluorimetric analysis for comparison. The sample sclution which
contains 10-15 grams of sample p巴r100 ml of acid solution (HNOs+HF) was diviふ
ed into two fractions of 300~500 ml and 5 ml for alph呂 spectrom己tryand fluorIme-
tric analysis, respectively. The analytical results are given in Table 3 together with
vn周 1
48 Kazuhisa KOMURA and Masanobu SAKANOUE
the data obtained from the diff色町ntblocks of the same sample for the sake of exa-
mining the homegeneity of the uranium distribution in a shell.
Table 3 i¥n旦lytica!:R!号su1tsof Uranium Cont邑ntin Tridacna Shells
(Comparison of the Analytical Methods)
知 ple三主;竺!と1:ZJ|| Other Blocks of a S旦mple
Acth九Anal.料水
TM 2
TU 1
1.21土 0,03 0,70 1司 11土 0.03料料
(053土 0.02
2.06土 0.06判'料
ppm 0.15
0.73
1.38
1.48
(023014s 0.26, 0.79 1,63判咋キ
* Alpha Spectrometric Analysis Using 232U as Uranium Tracer 料 FJ.uorim.etricAnalysis by Ohashi, Atomic Fue! Co.
料*: Activation Analysis by Osawa and A加, Kanazawa Univ. 料*ド UraniumContents of Recrystallized Part (Ca!cite Part)
As kn::Jwn from this table, the results of alpha spectrom巴tricanalysis agree quite
wε11 with that of fluorimetry except the case of TM 2. This fact indicates the
of the alpha method using 232U as uranium tracer. It is
also known from this table including the results of activation analysis that the
ur拠出m concentration in a Tridacna shell is f品rfrom homogeneity even if the blocks
were taken from the very near part in the same sal11ple. This fact gives an interεst-
biological problem as to the mechanism of th己 traceelement introduction in a
shell. Though th巴 alphasp巴ctrometricanalysis cannot discuss the detailed variety of
the uraniunユ concentrationin l11icro scale b巴causeof its low sensitivity (at least
several microgram of uraniul11 is needed), it is evident that the distribution oI urani働
um in a shell is heterogeneous also in fairly macro scale.
In this r色spect,a very interesting study was reported rec色ntlyby Lahoud et al.
. They the uranium distribution in the shell sal11ples collected from
variousεnvironment (fr色shwater, brackish and marine), by using newly developed
teclmique of fission track methodヲ呂ndobtained some interesting regularities about
the uranium content and distribution: (i) ur品niumconcentration in molluscan shell
differs conspicuously by the environment of the mollusc dwelling, (ii) ur・anium
concentration in a shell differs very widely the position of the same sample, i.e.
巴xternalvalve surface contains more uranium than internal valve surface, and the
fructured surface contains by far little amount of uranium than extemal and internal
valve surfaces.
Studies開 theDating Jii[ethods for QuatemaηS印刷es 49
It is wε11 lmovm that the migration of the trace e1ements may occur during the
change of crystal structure. In th巴 caseof fossil carbonate samples, this will be
recognizecl as the change from th己 metastab1earagonite to stable calcite Iorm. The
results of TM 2 and TU 1 indicate th邑 enrichm己ntof uranium at the tim日 ofrecrvs四
tal1ization, that is, calcite part contains more ur乱niumthan aragonitic par仁
As sugg巴sted alreacly abov巴 itmust be emphasized that the study of the
clistribution of the trace elements in micro scale is veIγimportant to recognize the
correlation between biologica1 action and the trace e1ements introduction during the
growth or after the death of organisl11s. Activation analysis, fission track techniqu日
and the activation autoracliography, which is cleveloped by Sakanoue et a1. (50), seem
very promissing to inv巴stigatethe clistribution of the trace elem色ntsin very small
scale, because of their high sensiti.vities and/or easiness for perfonnance.
Besicles the diffrences of the uranium distribution in a shell, the variety of
uranium content among th巴呂amespecies of mollusca must be pointecl out. It 1S very
interesting that the uranium content in Trulacna shε11 tends to incr巴asefrom geo-
logically younger sample (TM 1, 0.04 ppm) to olcler one (TM 2, 1.21 ppm). Similar
results have b巴enreported by Blanchard (51) and Stearns et a1. (28) for uranium
and by Lowenstam (52) for strontium in fossil molluscs. These facts may be ex司
plainecl by assuming that some traceεlements are intr吋 ucednot at the time dur加 g
the grQwth of th芭 organisms,but at the time oI degradation of the organic m呂tter
in their structures.
vn帽 2AnalyticaJ Results of Coral Samples
Analytical results giv邑nin Figs. 20, 21 and 24 are expressed as the activity ratios of 234U /238U, 230Thj234U (労巴quil.)and 231Paj235U(タbequilふ Here,the activity ratio 231Paj235U was calcu!ated from the activity ratio of 231Paj238U by呂ssumingthe activity ratio of 235U 知 238Uas 1 :21. 7. 1n the right hand two columns, the ages calculated from 230Thj234U and 231Paj235U ratios ar官邸時n,although these data are 110t sulヲmittedto any correction as to initial 230Th aud 231Pa as discussed below。Analytical resnlts of different blocks of the same sample afe distinguished by the subscript a, b and c.
Fig. 20 shows the analytical results of coral samples. All the samples lis民d
above CK 10 are remained still in a悶 goniticstructure. As known from this figure,
231Pa ages show considerably good agreεment with the ages of 230Th method, and
the 234U/238U ratios do not contradict to their initial ratios of 1. 15~ 1. 14 recognized
as their ratio in sea water. Particular1y, the sample having the ages of 40~70
thousand years (CK 4, CK 8, CK 5 and CK 13) give completely the same ages
betwe巴nthese two indepεncl色ntmethocls within the statistical errors, and other inde-
pendεnt method using 231Pa/230Th can a1so be applied by assuming the initial 234UI
238U ratio as 1.15.
However, 231Pa methocl seems to give systematically somewhat olcler ages than
230Th method for relativεly younger samples of less than ten thousand years, for
50 Kazuhisa KOMURA and Mas旦nobuSAKANOUE
Ages (1i 〉くP1a0-E231 y) Th-230 1 Pa-231
2.85島一'.3エ 0.'
methodl method
CU1 I.5士0,2 7.4士,.5 同 (1.6::1::0.2) ( 3ωヰ
CU2 3. 12 暗 J. 3:t 0-" 最I/Ji
a 2呂6 臨 ト唱:;}4 5.9士0.3 ,8土3 関 6.6土 0.3 9.4士2.0CK 1~ 2.74 醐 O よ2 土0.' 16+ I t,'i< 5.8土 0.2 8.1土0.6
C K2 2.54 嗣 崎)-l 6.3土O.J '1:t J 聖母 7.1土0.3 8.8士1.5
C K3 2.58 括協 ゆ47.'士O J "斗.,拶場
8.0士0.3 11.0士1.0
CK1'i 2.92 3f. c/::t t 醐 ト。→68.o:t" 事唱 44土 2 54土 6
CK12 3.01 3J.3 -::'1 t刷 最電 44土 2
CK4 3.06 J;土r 陣{ 時3叶 67ニダ 輔 45士2 日2土6
CK8 37主 f
湖 ベト63土J 岡 49土 2 46土53.70
C K 5: I~:: lJ土 噛 寸ト月七4 輔 60士3 62土7b 13.34 5 7土 5
ckli:; 83 1土 5 8',土1439土事 JU:!:4 5 士4 91士17
13. 52 fロー H~Þ{ 81土J 5 6土 2 79士15
C~,13 12.99 … 臨 75.9士山~:I7 J 糊66土 3 67士6
CK10all~:
川+ 叫 Fハ6>¥-0→坤1←O39《」争をE打毛什dトLトー@ー吋
> 200 [>500] b 11 . 35 uo土 108+
一恰H日33K 0K 一g1 r ~ I m
一一一工醐 9r:!: 80土 3 1 10土 40
106-1 0.75 73土 H齢寸 140士 10 > 120
0.77 !f6::: -1 f/6ト -1 [ナー
0
3
ny』LH
「,EU1
、もR32
4
Aつ49」L
(2Em)% り-2:3.叫 Act (号議)Ag llt琵昔Age
Fig. 20 Analytical Results of Coral Samples
U contf>nts (ppm) Activitv ratio 0.2 0.4 0.6 0.8 1.0 1.15 1.3 1. UU; リ l " v
串 0.04TMl 土0.004 l骨'"土 J ,,;土 0]
b O 土g8003111'争 曲 p 士ロ, ι'..上位 0' 腕
TG 1 0.294 土0.009 "士土" 1. fS.z a.Of5
TK1 0.32 土0.01 申 S.,士06 1.15土,0<
トH
τI 1 0.2'8 '"土 J /.06土且口2
土0.01 蹄 HlH
T01 0.37 士0.02 ,,8土 ,60士" /.02土日 0< f蝿ト即日一一寸
Arag Oラ8 '"土土"
H最~TU'l
土0.02 ト@ト トー屯←-<
2.06 Col 十日.0白
,.,土]., 99土d 0.99と aOf4
>-<
I-¥rag!1.21 "土手'"土 J I.Oj:tQ.02 噌 H 桝 1J 下関:2 土0.03 12i士 J
~al 11土1l0 03 '50:!:. ., s7土7 ι""土匂 ".トー〈トー, ,を ト唱
倍弱).Zt倖委主)ぷ (提言動Ao'
Fig. 21 Analytical:t主的ultsof Tridacna Shell Samples
Studies on tl.回 Dιrtingljllethods for Qu:7fernary Samjうles 51
cxamples CU 2, CK 1, CK 2 and CK 3. This discr己pancyis considered to be derlvァed
from the follmving two causes:
( i) Contamination with external 231 Pa, most probable source being the im司
perfection of 231 Pa removal from 232U tracer,
( ii) Exsistence of initial 231Pa introduced into the coral body during its growth.
But, the formεr caus巴 (i)was found to be negligibly small by the blank experiment
and by the experimental determination of 231Pa in the great amounts of 232U tracer
(about 2xl04 dis/min). Therεfore, the latt色rcause (ii) that initial 231Pa e交ceedth色
amount of initial 230Th become more probable. The analytical results of modern
coral sample (CU 1) seems to prove this assumption to be correct.
Contεnts of initial 2301'h and 231Pa are (l .3~ 1. 5) ::t: O. 叫uil. ancl 7.4土1.596
eqllil. with respect to their parent ur呂nillmisotopes 234U and 235U, respεctively, and
these values may be taken as the initial ratios of 2301'h/234U and 231Pa/235U to
correct the ages obtained directly Irom the ratios of 230Th/234U and 231Pa/235U in
the sample.
If the coral cannot distinguish thorium and protactinium from parent uranium in
the proc己ssesof introduction of these heavy elements, the initial isotopic ratios of
230Thj234U ancl 281Pa/235U in coral sample will have the same value with the ra司
tios of these isotopes in sea water, in which the coral had grown.
Concerning to the uranIum an仁1thorium series inequilibrium in sea water, Moore
and Sackett (53) have reported thεvaluεof 0.055説明日il.for 230Th and O. equil.
for 23lPa against the expectation of the 3,uthor of this paper, while Kuznetsov et a1.
(54) obtained the values of 1~49'ó equil. for 230Th ancl 5,-7巧 equil.for 231Pa from
many water sam.pl色scollectec1 from near equatorial region in Pacific Ocean. Th巴
latter values just corresponds to the ratios expected and agree with the analytical
results of moc1ern coral sample.
The discrepancy between the results of Moore et al. and Kuznetsov et a1. may
b巴 explainεdby the clifference of analytical method, namεly, th色 formεranalyzed the
purely clissolvec1 230Th ancl 231Pa in sea water after centrifugation away the partide
matter, whereas the latter determined th巴 totalamounts of 230Th and 231Pa in sea
water without any treatm告ntbefore analysis.
If the assumption as to the exsistence of initial 231Pa ancl 230Th are valid, not
only th色 dis
1n Table 4, corrected 231Pa and 230Th ages arεgiven by assuming the initial
ratios of 231Paj235U and 2301、h/234U as 7.4土1.5%equil. anc1 1.4土0.2;;¥づequil.,
respectively.
52 Kazuhisa KOMURA and Masanobu SAKANOUE
Tadle 4 Corrected 230Th and Z31Pa Ag告Sof Coral and Tridacna
伽 伽叫l叩帆P尚le Iト捌叩呂卸む
CωU 一 0 叩 2 0.0ω土叫0ω 8 -CU 2 2,1土 0.3
5.8土 2.1
4.5土1.0
5.2土1.7
7.4土1.3
CK la 5.0土0.4
lb 4.2土0.3
CK 2 5.5土0.4
CK 3 6.4土0.4
CK 11 39土 2
CK 12
CK 4
CK 8
CK 5a
5b
CK 6a
6b
59士5
56*5 79土 5
52土 4
(231Pa Age)/(230Th Age)
1.16士0.43
1.07土 0.25
0.94土 0.21
1.16土 0.21
50土 6 1.27土 0.15
48土 5
42土 5
1.0ヲ土0.12
0.89土 0.14
59土 7 1.00土 0.13
80土 14
37士 17
1.01土日。19
1.67ゴ=0回 83
ヲG1 ト下両06 0.0=ぷ7 下一一一; → ー
Assumed Initial γalues (in % equiL) Coral: 230Th=1.5土0.2,231Pa=7.4土1.5Tridac何回 230Th=1.5ヲ土0.05,231p旦=6.3土 1.4
As known from the Table 4 and
Fig 22, agr色巴l1'.ent of 281Pa ages
with 230Th ages bocomes beUer than
before.
Rac1iocarbon data reported by Mii
and Kigoshi for the sample taken
from the same str日比igraphicposition
as CK 1, CK 2, and CK :3 gives
the consistent ages of 6,630士
150 yr for coral (GaK and 4,370
土 130 yr for Tridacna shell (GaK
On the other hand, although
too young巴rage of 27,200士1,200yr
for echinoid (GaK is
reported for thεag色 ofthe similar
stratigraphic member with the sarrト
ples CK 8, CK 5 and CK 6, the ages
IOr一一
uod叫例
日一NB窃色
JL~ Th-230 AGE
1o" Y
Y
5
0
Fig. 22 Correlation betwe君ncorr巴cledPa-231 ag巴sand Tlト 230ages
Studies on the Dating Methods for Qωte円wrySam.ρles 53
obtained by 230Th and 231Pa methods (55-65 x 103 yr) are considered to be more
reliab1e, becauヨeth巴 contaminationwith recent 14C is very probab1e for such an old
carbonate sample.
The old巴stcora1 samp1e CK 10, which was col1ected from the lower members of
Ryukyu Limestones known as Hyakunodai Limestone, contains some ca1cite structure
and the analytica1 data shows the impossibi1ity of the age ca1cu1ation both by 230Th
and 231Pa methods, because the ∞ntents of daughter nucIides exceed their equilibrium
va1ues with their parents. On1y the巴quivoca1age of more than 500 thousand years
is estimated from the 234U/238U ratio. Uranium content of CK 10 is considerably
smal1 (1. 83 and 1. 35 ppm) compared with the content of non-recrystal1ized coral
samp1es (about 3 ppm). This fact suggests the preferentia1 1eaching of uranium
during the phase change from aragonite to ca1cite. Ana1ytica1 resu1ts of CK 7, CO 1
and CK 9, which are a1most in ca1cite structure, a1so show such circumstances. For
these recrysta1lized samples, it is characteristic that the uranium content is consト
derab1y 10wer than that in the normal samp1e and that the daughter nucIides general-
1y exceed their equi1ibrium va1ues.
Analyses for the samp1es indicated by the sign a, b and c, were carried out to
examine the reproducibility of th ana1ytical method and to know the homogeneity of
the radioactive e1ements in coral. Though good agreement was obtained for the
samp1es CK 1 and CK 5 not on1y in uranium content but in ag.es, the resu1ts of CK
6 and CK 10 showed 1arge difference in these values. This might not be exp1ained
by the ana1ytica1 error accompanied with this method.
These disagreements, particu1arly in age of the sample CK 6, may be caused by
the heterogenous distribution of radioactive e1ements including the possibi1ity of the
migration of parent uranium, even if any change of crystal1ine form cou1d not be
confirmed by the Xィaydiffraction ana1ysis. Probab1e age of CK 6 is considered to
be the same that of CK 5 (55-59 x 103 yr).
VII-3 Analytica1 Resu1ts of Tridacna Shel1 Samp1es
In Fig. 21, the ana1ytical resu1ts of Tridacna shell samples are summarized.
Differing from the case of cora1 samples, age ca1culation by 230Th and 231Pa cou1d
not be applied except for very young samp1es of TM 1 (Modern), TG 1 (Modern)
and TK 1 collected from the Raised Cora1 Reefs. A1though any alteration in their
crystal structure is obs巴rved,activities of daughter nuclides most1y exceed their equi-
librium values with the parent uranium isotopes, and on1y the re1ative age assign-
ments are possib1e from the ratio 234Uj238U by assuming their initial ratio as 1.15
and no differentia1 leaching or addition of two uranium isotopes occurred.
As discussed above, 1arge amounts of recent samples of TM 1 (30 gram and 190
gram) and TG 1 (552 gram) were analyzed to obtain the reliable initia1 activity
ratios of ,230Th/234U and 231Pa/235U. Among these three samples, TG 1 gives the
resu1ts which agree with the resu1ts obtained from the recent coral samp1e CU 1,
54 Kazuhisa KOMURA and Masanobu SAKANOUE
supporting the validity of the assul11ption l11adεabove. Th色yar巴.1.59士 0.05ク1s'equiL
for initial 230Th/234 U ratio and 6.3土1.4% equil. for ini ti且1231p旦/235Uratio. On
the other hand, TM 1 gives邑xtraordin旦lyhighεr ratio of initial 230Th/234 U (TM
la, 17士ぢ, TM lb, 4.2士o. 49o than expect巴dvalues of 1 equil.
This discrepancy can be色xplainedif the absolute value of 230Th content (dpm)
per unit weight of sample乱recomparec1, i. e.,
TM 1a 0.007 土 0.002 dis!l11in/g,
TM lb 0.0038士0.0003dis/min/g,
TG 1 0.0040土 0.0001dis/l11in/ g.
From the sam巴 ordervalues of 230Th content among thεsamples, it is known
that the large 230Th/234U ratios in modern Tridacna sh邑11(TM 1乱 anc1TM lb) are
caused their extr巴mely10w uraniul11 content (TM 0.040 ppm; TM lb, 0.082
ppm) compared with the content of uranium in TG 1 (0.3 . Therefore, it is
more preferable to discuss using th巴 absolutedisintegration rate per unit
of sample instead of the ratio 230Th/234U solely. This may perhaps be held in the 〆
問 問。f231Pa, though any information could not be obtained in this work.
According to the reasons discuss巴dabov巴 itmay b(う morerelieble to consider the
initial 230Th and Z31Pa contents as 1.59士。胴05and 6.3士1.4タ6巴qual.for thε0ト
dinary samples containing uranium about 0.3 ppm. Th巳refore,the age of the
sample (TK 1) that could be determined by 230Th/234 U method, is considered 乱S
7,750士 750yr inste旦dof the apparent age of 9,500 yr. Thεcorr巴ctedage does not
contradict to the age of the coral sample (CK 1) taken from the same stratigraphic
position.
Analyses w巴白 carried out for both the aragonitic and ca1cite parts of the oldεr
sampl出 TU1 and TM 1, so as to investigate the influences of diagenesis (aragonite
to calcite)ー Asdiscussed in VII開 1,the most noticable fact may perhaps be the increase
of the uranium content in older samples. The v乱luesof 0.58 and 2.06 ppm for the
sample TU 1 and 1.21 and 1. 11 ppm for TM 2 are larger than the ordinary uranium
cont巴ntin Tridacna shell samples. This fact indicates the secondary introduction of
uranium into thεshelL Abnormal ratios betwεen par百 ltm加lIumand their daughters
ar巴 obtainedas昌 matt色rof course for these altered samples. For these samplεs, age
should not be calculated, even if the 230Thj234U ratio would not attain itsεquili闘
brium valu色。 Forexample, though the sample TM 2 arag. gave the value of 91 ここ 4
労 equilfor 230Thj234U,昌geestimation could not b邑問rriedout, since the activity
of 231Pa exceeding about 30 タ1s' over the equilibrium value indicated the secondary
alteration of the sample.
Ther巴fore,it must be emphasized ag旦inh巴rethat the 230Th to 231Pa ratio serves
as one of the most reliable criterion for judging the validity of the agεestimation,
because the 3nomalous changes in isotopic composition occuring more than several
thousand years ago could not have any print on the ratio 226Raj230Th of the sample.
55
へ町にhhhUむも『『J-Kめ』h
b
~
、コbιwJYU」
200
150
50
100
0 35
Studies OJZ thc Datuzij I¥1Iethods fm・Qz.t!lternarySamples
斗_L1
E
LL 斗」い5
e
1004洋平
50
40
30
70
60
90
80
Ag.e
Influence of theUranium Leachig to the Growth and Decay Properti日S of 231Pa and 230'{、h
5
Fig.23
A mechanism that causes the excess 231Pa or 230Th, is shown schematically in
Fig. 23, by assuming the preferential leaching of uranium during the secondary
alteration. 1n making th巴 figure,assumption was made that the 50 96 of uranium
was preferentially leached away from thεsample without disturbing the contents of
230Th and 231Pa of the sample, and that the time needed for leaching was negligibly
shorter than the timεof exsistence of th巴 sample,since, for exampleフ thelarge excess
of daughter nuclidεs was observed in the case of the sample CK 9. Abnormal cases
may be approximately expl乱inedby this simple mechanism of uranium leaching. The
actual cases found in coral and Tridacna shell samples seem to correspond to the
following groups: (case numbers correspond to the numbers in Fig. 23)
CK 7 ・・・・・・・・・・ ・・・・ ・・ ・・ ・・・ ....................................... Case 1,
CO 1, TM 2 ara[l. .....................目・・・・・・ 0 ・・・ ・・・・・・・・田・・・・・・・・・・・.Case 3,
CK 9, CK 10, TO 1, TU 1 前川・ ・・ a ・・・・・ ・・・・・・・・・・・・・・・ ・・・Case5 or 6
Though, th色 ageand the time of leaching can be estimated, theoretically, on the
basis of these assumption by using the 234Uj238U ratio and 231Paj230Th ratio as an
age indicator, the probability of fractional leaching of 234U and 238U, and the lack
of the meth.od to determine the accurate ratio of 234Uj238U, limit the application of
this method. Furthermore, the leaching of uranium may not occur in natural condi回
tions such a simple mechanism as assumed here.
Analytical Results of Stalagmite and Water Samples.
111 Fig. 24, the analytical results of stalagmite and the water samples are illus同
VH-4
56 Kazuhisa KOMURA and Masanobu SAKANOUE
trated by the same manner as the results of coral and Tridacna shell samples.
Stalagmite 問 wasdivided into four s巴ctionsof SA 1 (Central p旦rt),SA 2
part, just uPI=色rpart of SA 1), SB (Middle part) and SC (Outside part).
Thesεfour blocks were an旦lyzed for their 234U/238U and 230Th/234U ratios to
examine their for s旦 Asknown from this figure‘contents of
uranium and thorium vary considerably wァiderange of factor and th巴 ratios
234UF"38U and 230Th戸34U do not the theoretical changes of these values, which
may be expected assuming the constant growth rat巴 ofstalagmitεwithout the
migration of radioactive elεments in it圃
These abnormal ratios between par巴ntsand 正laughternuclides, probably, due to
the mechanism of the stalagmite fonnation. 1n the repeatεd procεss of the
recrγstallization, parent uraniul11 will rnigrate fonning the carbonate complex with
carbonate ions that exsist in its environment. Another cause may be the
exsistence of the which are introduced抗 thetime of the formation of
brownish colour‘巴d And th芭 relativelylarg巴 contentof 232Th (0.02 ppm)
wil1 be ぷinednot the introduction chemical pr仁cessesof disso~
lution and but thεsimultaneous introduction with the impurities in
detrital form.
the exact age estimation could not be made for this s乳mple,apparent
lnaxIlnUTIl 乳geof 15¥103 Y巴arswas obtained from the ratio of the
central part SA 1 without the correction by initial 230Th introduced with 232Th.
{ゅ 2珂n/'刊(0;.母quiL l。 刊)
2認Th(ppm) 238U (ppm) ~~ Lsotopis: _ R口
1.2 1.4 0.4 口6 O.d
O.025~O o.o.2 0..13ωω5 i-: 13.2 ",0.6 出 土 問O0GGLC3ブ:j1~
E 一一-,--ーSム!Cei1ter)白 i
0..0.70.ω∞'2 0..16止。0.1 ⑮ 7.8土Cδ 0..96.ょ
58 (トdidcHe)* 0..0.24土0.0.0.5 0..13占 o..Q7 。10 土2 7.0.1土
SGイOutsid母) ぐ 0..0.0.7 。抑制ω2 I ~ 8.9土ο6 1J23".
関 荏 ち
応花 WATER!Okina記2 2羽U(}-'g/l) 。 0.2 0.4 0.6 0.8 1.0 1.15 1.4 c. "Yuhi"-Entrance 2.29"0.70 1 1.31,;σ0.3f ぺ〉
)) 由 Exit 2.27"0.70. 1.35:l003 I -o-
iSEt¥ W.ムTEB]
Mouth of Taii:Jむ Riv.(Okinawa) 2 .20~ 0.70 U7士0.03
Shioya Bay ( )) 2.86士0.71 7J8Io.白
Tsu!,umo 8(.'1)1 (Noto防 n.l 3.2土 0..7 UI, I003 I 寸
Fig. 24 Analytical Results of Sta!agmite and Water Samples
Stlldiesοn tJze Dating iiJethods f01~ Quaternary Sω}ltles 57
Ground wat色rof cave“Yuhi' and s色V記ralcoastal sea ¥vater of Okinawa and Noto
Peninsula w邑realso analyzed f.yr their uranium con七色ntand r品tios. As
known from Fig. 22胃 cavewater・samplesgive considerably large amounts of uranium
(2.3 microgramiliter) and high 2'l4U p38U r丘tioof し 33~1. :34. Since the
source of uranium in cave water sample is considεred to be the 1i mes torぬ bed,tlll"ough
which t11色 groundwater has pas2εd, high 23生U/23呂lJratio ¥7/i11 "be cauE;eci bv the
preferential leaching of daughter nudide 23,t U with parent 238U, because
the :!3'~UÎ238U ratio in the ]imestone b吋 cor:npos釘 ofmarine depo~ヨited calcium
carbonates is considered to ha vεat 1110St the Sanl己 vaJueof sea water (1圃
The preferential leaching oi nuclide may be due to thεr巴coi1 E f.fect of
alpha ancl beta decay processes, which makεS 234U moreεachable than parent 238U,
as旦ssumedfirstly by Cherclyntsev (1 j) to explain the anomalous ratio in
natural uranium 巴S.
Three samples of co乳stalsea water the activity ratios of 234U(l38U 1.17と
0.03. 1.18土0.03and 1.14士O.OB.These vahws do not contradlct to th巴 clatareported
by many authors (13,
VIU Discl1ssions on the Causes of El"l'Ol'
Errors expected to be brought into this dating method are classiiied into following four groups. Those品1・e(i) pipetting error or the unc巴τtaintyof the concentratiol1 of 232U spiked, (ii) i111purities in the tracer solution, (iii) col1nting statistics and (iv) contamination or the sample with foreign materials. Those fadors are discnssed below.
VIH-1 PipeUing Erro1' and the Absolutθ of 2112U
Pip日ttingof thorium and protactinium tracer was carriecl out thεultra
micro pip巴ttes (Shibata Rikagaku Kikai Co., Ltd. 01' Beckr刊誌11nCo・ ¥vhose
capacities a1'e 0.05 and 0.20 m1, ancl theεr1'o1' cle1'iveヨfromthe reproducibility of
pipetting is assumecl to be within .3 pεl' cent induding thεg色ometricalclifferences of
the mounting of thεstandard sources, whereas the uranium tracer 282U was added
to the sample by using the measuring pip色tteso as to add the ad己qu呂tεamountsof
232U according to the uranium C'Ontent in the analytical solutions.
An absolute counting rate of 232U spikecl was d色terminerlfrom th巴 activityof
thεstandard source prepared from an aliquot of 232U stock solution. 1n this case, the
232U stock solution had been purified in advance from daughter muclides which may
grow up with the ha1卜life0:1: 1.9 year (228Th) owing to an alpha d日cayof 232U.
Considering the tailing of the alpha spectrum obtained the evaporat邑dstand丘rd
source, 232U activity was ca1culaほdfrom the sum of the counts in the range Irom
3.5 to 5.32 MeV.
Besicles the direct measurement of 232U activity mentiO!1ed abov色 itwas deter咽
min巴dby using other uranium tracer 237U (half-life, 6.75 clays, beta and gamma
rayemitter) (47). In this case, 237U (inevitably contains target uranium with the ratio
58 Kazuhisa KOMURA and Masanobu SAKANOUE
of 234U(238U=O.20土0.01)was mixed with an aliquot of 232U stock solution and
the purification of uranium was carried out by means of an anion exchange method
after th邑 attainm巴ntof isotopic exchang巴 equilibrium. Then, the purified uranium
was巴lectrodepositedon a stainless steel plate as the same manner described before.
Aft訂 thedetermination of the chemical yield of uranium by the beta counting of the
recovered 237U, alpha spectrometry was carried OUt. The peak area of 232U (from
4.80 to 5.32 s1e¥T rεgion) was comput巴dand normalized by the chemical yield. The
232U activity thus determined agreed v色rywell with the value obtained from the
evaporated source within 2 per cent. Hence, the errors derived from pipetting,
geometrical difference of mounting and the counting statistics were assumed not to
exceed 3 per cent for uranium tracer 232U。
As the tailing ef fect cannot be neglected both for the electrodeposited and eva同
porated sources, an absolute quantity of uranium may be somewhat 10wer than that
of rea1 one. However, such effect will not interfere the caluculation of th色 ratiosof
230Th/234U and 231Pa/235U, since tailing effect may a180 e叉sistin the alpha spectra
oI a11 nuclide studied this work晶
VHH2ηTracer Solution
231Pa in 232U
As the uranium tracer 232U was prepared by the neutron irradiation of 231Pa by
2呂1Pa (何口町川,rけ)2幻呂2仇Paτ古告Id+232U r陀ε抗制刷t仕ionα叩n(州6回叫O的), t枇hecoα似I
231Pa is most pr叩obコable ε and c訂n此ti比C昌叫1problem, because the 231Pa activity is extremely
small in natural sample. That is, this activity n巴V色r色xceedthe value 1/2107 of the
238U activity unless the geochemical processes that色l1richprotactinium or leach out
par巴nturanium might occur. This sort of contamination is more critica1 for the
dating of younger sampJes than for that of older ones. If the 司ctivityratio of
231Paj232U in tracer solution and that of 232U/238U 乳 児 assumed as 10-3 and 1.0,
日 spectively,activity of protactinium in trac巴rsolution corresponds to about 2.296 of
the εquilibrium v乱1uewith p旦rent 235U (this value corresponds to about 1,000 year
background) and causes a serious error for the age determin以ion of the sample
younger than ten thousand years. Th巴 uraniumtracer used in this work was purified
by three cycles of solvent extraction and two cycles oI anion exchange乳ndabsence of
2S1Pa in 232U tracer solution was ascertained by l11eans of alpha spectrometry using
233p呂田 protactiniumtracer, therefore, the possibi1ity of 231Pa contamination frol11
the uraniul11 tracer spiked could pεrfectly avoided.
ii) 230Th and Ur呂niumin 234Th
234Th was prepared by the milking technique from its parent 238U by means of
cation exchange (43) and solvent extraction methods. In this case, a complete removal
of uraniul11 is the most critical problem and, furth巴rmore,the contribution of 230Th
growing up simultaneously with 234Th must also be consider巴d,because the activity of
Stu.dies on tho Dating 11/1 ethods. for Quafernary Sat1'iψles 59
230Th increases linearly with time, whereas that of 234Th has attained the satur就 ion.
The relation between 230Th and 234'Th ls described as Eq. (29):
230Thj234Th = (234Uj238U)AotjC 1 ~e-)4t)eーんT (29)フ (20),
whεre ,10, ,t1: decay constants of 23llTh and 23品Th,
t: the period between the time of the uranium (cow) purification and
the time of milking,
T: the time elapsecl after the preparation of thorium tracer 23壬Th.
By assuming 234Uj238U=1.0, t ancl T as ten and three half田 livesof 234Th (24.1
clays) , the value of 4.8 x 10-5 is obtained as th色 activity ratio of 230Thj234Th.
A.ccordingly, th巴 contributionof 230Th in thorium tracer cannot be neglected when
宮34This prepared from aged uranium or aged 234Th is used as thorium tracer.
These troubles can be avoided by using the depleted uranium (with low 234Uj238U
ratio) for cow and by taking t and T in Eq. (29) as small旦spossible.
iii). 233U, 231p丘町ldThorium Isotopes 232Th and 230Th in 233Pa
233Pa was produced from neutron irradiatec1 232Th by 232Th (n, r) 233Th
J寸 233Pa reaction, and purified by so伽 n凶t収捌t削 ionω叩n u附s討111昭gD凹I悶B.l:
(付45町) 0町r by ad白so仇r緬下ption and de出so町rpμtiめon t白echn副1廿ique using S臼iOz0町rglass powd 巴ras
adso仇r司b玩en此1此t(付44釘). 1n thεcase of 233Pa tracer, the contaminating nuclides which must
be considεred are 232Th as thεtarget material, 231Pa proc1uced together 羽一ith233Pa
from 2301:、h in 232Th, and 233U accumulated by the decay of 233Pa. A.mong these
nuc1ides, 232Th and 231Pa are consid巴redas the same manner as the case of 234Th
and can be proved not to cause serious errors.
However, 233U has somewhat critical problem that th巴 alphaparticle energy of
233U (4.82 MeV) is nearly the same as that of 234U (4.77 MeV) and may interfere
the determination of the 234U j238U ratio. The incr巴as日 of233U activity and the
dεcrease of 233Pa activity are related as follows:
233Uj233Pa =作、"pt~ 1)九ujA3P (30),
where A3u and A3P d巴notethe decay constant of 233U and 233Pa, respectively, and
t--the time after thεpreparation of 233Pa tracer.
If t is assumec1 to be five half闇 livesof 233Pa (27,4. c1ays) , the value of 1.4 x 10-5
is obtained as the activity ratio of 233Uj233Pa. This value increases exponentially
with time and may cause a little error in determination of 234Uj238U ratio when
very small quantity of uranium is analyzed. Hence, 233Pa mLlst be purified at times
to remove 233U accumulated in the 233Pa stock solution.
vm圃 3 Counting Errors
In the radioactive dating, counting errors are generally the most important deter姐
minant in the error of the estimated ag巴. A.nd the factors that contribute to these
errors may be c1assified into next three groups.
60 I三azuhlsaKOl¥lIURA and Masanobu SAKANOUE
i) The le'1el of radioactivity of the source (it is beUer to have as
much
Th巳
as possible in respect to
of the sample that det巴rminesthe amounts of parent
特 Theage of the
nuclides.
各 Th己 0'1色r乳11chemical yields of
that determInes the amounts of th巴
thorium and
ii)
ラ令
apparatus巴sfor spectrmτletry.
efficiεncy: Ionization chamb色r is better than Semiconcluctor
(く2π).
data are
Semiconductor .OOX
くDouble~gridd吋 Ionization Chamber (0.0)[
く IonizationChamber (O.X
発 Re弓olutionpowεr: S巴miconductor(0.3 -0 ② 8俗 FWHM)is bettεr than
匂"
Gridded Ioniz;uion Ch2rnber (1 ~2%
tim邑 (Thoughthe time is desirable to obtain the
total counts w1th less relative statistical error, the driff of con~
dition and b8.ckQ:round乳ctiγity must be also taken care in this case)
to these the theoretical limit of the of the de蝿
ficient 230Th and 231Pa dating methods are examined. Because the background activity
of the system limits the determinable amounts of 230Th and 231Pa 011 the
it seems most useful to c011sider at first the lowest 1imit of the age
obtainablεin this when the background activity and the measuring time are
If thε the time and the lowest limit of detection
be b cp立1,t minutes and four times of th色 stanc1arddeviation (a) of the background
tllεnext Is obtained f10m the law oI the
statistics :
。A
A
A
d
佐
一一一一
、、,ノn
I
υ「pu
〆{¥n
0
.l
lL Puu
ρレム4be
d
zr-0
十L-
qu十-
aI
L
12 = 4 12 (31 ).
As b a11d t are 0.02 ,~~O.04 cpm and 3,000 minutes in this
the lowest limit of detection becomes 0.010-0.014 cpm. By using 0.014 cpm as the
むletection1imits of 230Th and 2S1Pa the relation between total amounts
of uranium in the sample and thεlowest 1i111its of th己 ag色 determination
deficient 230Th and 2S1Pa methods can be expressed by the Iollowing εquations:
0.735 For 280Th,.,…………0.014 =ー 2一一 x(UJ x 0.5 x ( 1 ~ (32),
0.'735 For 231Pa."・u ・."-"一-0014z-EYET:下 x(U] x 0.5 x ( 1
St1tdies on the Dating ]}lethods 101' Quatemary Sam戸les
wh日reCUl: total amount of uranium in ihe analytical sampl (μg),
,10, '¥1: decay constants 01: 230Th and 231Pa,
61
To, T1: the lowest limits of the age obtainable by 230Th and 281Pa methods,
respεctively,
0.014: limit of detection (cPlll),
0.735: counting rate f01" 1μg uraniu日.1by 2;r geometry counting,
21.7 the activity ratio of 238U/235U in natural uranium sample,
0.5 assumeピias the chemic旦1yields of thorium and protactinium isotopes,
in practic色、 thisvalue varies in the range from 0.8 to O. 1.
lf AoTo and A1T1 are sufficiently small comparecl with 1 (To, 1'1<:103 yr), Eqs.
(32) and (33) havεvery simple forms and the lowest limits of age are estimatecl by
Toニ=0.0763/Ao[UJニ O.110 (1' 11 2) 0 / CUJ = 8. 3 >く 103/[U:I(year) (34),
T1ニ 0.110 21.'7 j< (T1/2h/CUJ = 7.8¥104,/[UJ (y巴ar) (35).
where (T%)o and (T%)l denote the halfω1ives of 2301、hancl 231Pa, respectively If
the sample contains 100 μg りfuranium, ab心ut80 ancl 800 years are obtained as the
lowest limits of ages in defic.ient 23JTh and 231Pa methods, respectively. The mini回
mU111 statistical errors in th色白timatedage may he th己 sameorder with these valu-
es, if the above mentioned
conditions ar巴 assumed.
Relation between totai
uramum cont巴ntand the low-
邑st limits of the age de-
t忘れninableis shown graphト
cally in Fig. 25. Thollgh
the device to lower the
background activity of the
measunng apparatus and to
prolong the measuring time
wiU enable to lower the age
limit, effect of them are
only proportional to thε
squar巴 root of l/b and t.
80, in order to obtain more
reliabl巴 ageof the younger
sample, it is moreεffective to
analyz巴 the larger amounts
of the sample with high
chemical yields.
VHI-4 Contamination
10d
0.1μg 10 100 Img 10 Ig
Tofal Amount of Urcmium
Fig. 25 Available Limits of the Deficient 230Th and 231Pa Methods
Thrεe main causes are consid色redas the source of the contamination. Those are
62 Kazuhisa KOMURA and Masanobu SAKANOUE
(i) exsistence of initial 230Th and 231Pa as discussed before in VII回 2,(ii) contam・
ination with th色 foreign materials such as clay that will contain 230Th and 231Pa
in equilibrium amounts with parent uranium, (iii)εxpεrimental contaminations from
reag巴nts,glasswares, ion exchange resins and elεctro同 depositioncell, etc..
The source (i) can be correctεd by the an旦lysisof large amounts and numbers of
modern sample, or by the appropriate correction methods such as 232Th contεnt as
indicator (10).
As the sources (ii) and (iii) can also be avoidεd by a careful experiment, these
factors are not discuss己dhere.
For exampl巴, a typical data sheet of the coral sample CK 5 is given in Table 5.
Table 5 Exampe of the Data Sheet
Radioactive Dating by 230Th and 231Pa Methods No. CK 5
Sample Classific. Upper Limeston己 Member
Location Kamikatetsu, S. Kik旦i
Dat己 65-12-23-6-2
Analytical Data
Date 66-1-28
Weight 37句 6g
Dissoln. HNOa十HF
Tracer UXl (O.10ml), Pa (0.40ml), U-232 (15.5cpm)
Sepn. Method. Fe(III) coppt. Ionex
Electrodeposition
Date
Chem. Yield
GIC M問 sur右ment
Date 66-2-9
Meas. Time 1,320
U Th Pa
66-2-2 66-2-2
45.7土 0.6 36.8土 0.5(%)
66-2--9 66-2-11
321 1,410 (min)
66-2-2
Counts/Ch 238U 7,021119 230Th 2,787/24 231Pa 806/28
234U 7,860/21 232Th 178/22
Results
232U 2,809/23
U-238 39.6土1.1(cpm) 234U/238U 1.12士 0.02(AR国)
U-234 44.3土1.1(cpm)
Th-230 19.0士0.4(cpm)
Pa--231 1圃34土 0.06(cpm)
Th-232 1.05土 0.08(cpm)
230Th/234U 42.9土 L7 (%equil.)
231Paj235U 73.5土 3.8(%equil.)
230Thj232Th 18 (A.R.)
U ω即日 2.87土 0.06(ppm)
Th concn. O.23::!=0.02(ppm)
む
U
U
}
l
U
H
M
K
U
n
u
r
必
見
3
3
Pもげ
4
り
臼
の
白
E
J
l
l
l
b
i
1
UTRuu
a告の
U
4
ム
44
q
O
A
1
0
q
d
q
u
QU9huηゐ
り
白
ou
ou
β
。A
唱刊
ue
t
a
m
t
Q E
(80土 60)Xl03(y)
(59土 3)グ
(5ヲ土7) グ
(56土 13) 11
Studies on the Dati持g 且~ethods for Quaternary SamPles 63
IX Comparison of Results with Other Data
We can consicler that th巴 agesobtain巴dfrom Ryukyu Lim色ston巴sindicate thε
warm periods in late Pieistoc色neepoch in this region of Pacific Ocean. From th巴
230Th data, they are presum巴dto be 5~7 103 yr for Raised Coral Reefs, 40)く103yr
for the beginning 01 th巴 warmperiod and 40~45 /, 103 yr for Araki Limestonεs and
55~何人 103 yr for Upper Limestone Membere, resp巴ctively. However, rellable data
was not obtain色clfor Lower Member of Ryukyu Limestones.
1n Fig. 26, the data ob幽
(a) 噌
hu
(c)
Other向rts 700 600
75
100
125
150
175
200 xlえ一一一一-
ABCDEFGH
2∞I xl0'y
図 230Th
o 11.C Fig.26
Comparison of the Analytical Data with Other Data
(d) (e) tained by this work are shown
togεthεr with other data based
on fossil carbonate samples
(17, 24~30 , 53, 61) (Fig. 26b),
solar insulation curve (62)
(Fig. 26 180j160 tem幽
perature (63) (Fig. 26 d) and
pal色ontologic c1imate (64)
(Fig. 26 e) 巴stima民d from
the data by deep sea sediment
∞re sampl色s. As seen from
this figure, Ryukyu Limestonε
S巴emsto give somewhat younger agεs than expected from 180j160 temperature or
paleontologic c1imate, and to correspond rather to solar insulation curve with mini句
mum value at about 50¥103 yr before present.
The comparison of the
sample age with the eleva-
tion above present sea level
gives very interesting infor蝿
mation th呂t the sea level
stands high己rin ancient sea.
Sea levels are becoming higher
in the order, that is, 0 m for
the present one, 2~5 m for
5~7x 103 yr B.P., 20~30 m
for 40~45)< 103 yr B. P. and
40~45m for 55~65 x 103 yr
B. P. as shown in Fig. 27.
Since the pεriod of 1O~
m(αbove present seu levell SO
40
jCOld (Wurm
301-ョ
20
10
ア亡二bdJ仁コ CK 5.6 ¥
/CK 8¥ れ/¥
、、/¥----烏~
_r;;ケ「 亡ごコプ /CK13b.' CK4 ..----。ぷ______ .~ø\f: 句/〆 l¥S<
よE了 1μへ-$勺tK11 CK 12 .AI~OI v~", IP
//宇/ \Jp\'1 e~O.4 rf¥'"
50 目。
Fig. 27 Presumed Ancient Sea Level n巴ar Kikai-Jima without any consideration of tectonic movement other than simpl邑 upheaval.
40 >く 103yr range was consi-
d色redto be in the last glacial age (Wurm) and the sea level would have the mini-
1τmm value at about 20~30 x 103 yr B. P. Therefore, the sea level nεar Kikai四 jima
might have varied乳longthe dotted line curves in Fig.27 during last seventy thousand
64 Kazuhisa KOMURA and Masanobu SAKANOUE
years unless the geological upheaval of Kikai四 jimahad occurred during this period.
If the simple upheaval of Kikai田 jimahad b色εnoccurring during this period without
any other tectonic movement, itwill b巴 givenby the chain formed line (ー.-.)
which indicates approximately the mean upheaval rate of 0.4 mm/yr. Even in this
C丘seth邑 changeof sea 1邑velmust bεconsidered to explain the ages of some samples
4, CK 8 CK 5 and CK 6).
X
The authors express gratitudes to Dr. Konishi for providing the fossil cor必a1
and Tridacna shell samples and his constant int巴restand encouragemen仁 Thanksare
also du号 toDr二 Osお入1aand Mr. Abe for uranium an昌lysisby activation method and
for the preparation of 233Pa tracer and to Mr. Omori for technical assistance il1
expετimental wor1<.
R,efexences
(1) Anderson, E.C吋Libby, W!人@♂.F、¥@吋, Weinhous叫, S包.勺, Reid, A.F.勺, Kir臼呂henb邑!1i但百ヲ A
Natural radiocarbon from cosmic rねad出i思討tion叫, Phy内7渇s匂 Rev. 7'1 931 (1 ヲ47).
(2) Khlopin, V.G" DつkLAkad. Nauk SSSR (1926) 178.
(3) Piggot, C.S. and Urry, ¥1¥人D.,Radioactivity of ocean sediments. IV, 1'he radium content of sedim告nt8of the Cayman trough, Am. Jom. Sci. 2l!0 1 (1942).
仏 Issac,N, and Piciotto, E., Ionium det巴rminationin d邑ep園seasedim邑nt8,Natuf巴171742 (1953).
(5) R口shoIt, J.N., Emi!iani, C., G巴iss,J., Koczy, F.F. 昌ndWangersky, P.J., Absolute dating of deep-s邑acores by the 231Pa/2301'h m邑thod,J. Geol. 69 162 (1961)。
(6) Sarma, 1'.P吋Datingof marine sediments by ionium and protactinium methods, Ph. D thesis, C呂r問 gieInst. 1'ecl1. (1964).
(7) Sackett, W.M 吋Broecker, W ふ and 1'hurber, D.L., The g巴ochemistryof Z31Pa and the dωat出irむい白I日口一,"".1
of p色lagics詑ed出imentおs,Annual R巴pt.AEC Contact A1' (30-1) (1965).
(8) Barns, J爪T.,Lang, E.J. and Potratz, H.A., Ratio of ionium to urar:出1min coral lhnestones)
Scienc巴 124175 (1ヲ56).
(9) S呂ckett,W.M., Ioniumωuranium ratios in marin巴 dep口sitedcalcium carbonates and related
materials, Ph. D th邑Si8,Washington Univ. (1958). (10) 1'atsumoto, M. and Goldberg, E.D., Some aspects of th巴 marinegeocheluistry of uranium~
Gεochim. Cosmochim. Acta 17 201 (1959)。
直方 Cherdyntsev, V.V., 1'r. III sessii komissii po opredeleniyu absolyutnogo vozrasta, Izv. Akad, IすaukSSSR (1955)ー
0同争 Ch巴r吋dyrぱ1詑tおts詑e、V巧fら, γ
(1961)
帥 Thurber,D.L., Anomalous 23品U123SU in natur邑, J. Geophys. Res. 67 4518 (1962)。
M Rosholt, J.N. and Antal, P.S吋Evaluationof th邑231p呂IU,230Th/U methods for dating Pleis凶
tocene carbonate rocks, USGS Profess. Paper 450-E, Article 209 El01-Ell1 (1963)。
M S呂ckett,W.M. and Potratz, H.A., Subsurface geology of Eniw邑tokatoll, Dating of carbonate
rocks by ionium and uranium ratios, USGS Profess. Paper 260-BB (1963). 伸 Bro色cker,Wふ, A preliminary evaJuation of uranium serIes ineql日ilibriumas a tool for absoJute
ag色 n1easureru告nton marine carbonat己s,J. Geophys. Res. 68 2817 (1963)。
同 Blanchard,R.L吋 Uraniumseries disequilibrium in age determination of marine calcium caト
bonates, Ph. D thesis, Wおhi時 tonUniv吋St.Louis, Mo. (1963). 同 Cherdyntsev,V.V., Kazachevsky, LV. and Kuzmina, E.A., Isotopic composition of uranium
and thorium in superg邑nezone. Studies on the
Studiesωz the Dating }¥{ elhods for QwIter持arySarntles 65
4骨 Cコhe釘r釘吋ピdyntsev,V
foωrn日ma討tionby isotopes of thoぽriu町1mand uran凶ium,Geokhimiya (1965) 1085. 。0) Alokseev, V.A., Kazachevsky, I.V. and Cherdyntsevヲ V.V.,Izd. Akad. Nauk SSSR, Absolute geochronology of Quaternary period (1964).
凶 Cherdyntsev,V.V., Malyshev, V.I. and Kazachevsky, I.V., Isotopic composition of uranium and thorium il1 supergene zone. Study of peat bog substances, Geokhimiya (1964) 399.
(22) Cherdyntsev, V.V吋 Kazachevsky,I.V, Ki.slitsina, G.I., Kuzmina, E.A.. and Kindヲ N.V.,Non融
eql1ilibril1m uraniClm in carbonate deposits and their age determination, Geokhimiya (1966) 139. (23) Titaeva, N.A., On the possibility of the absolute age det巴rmination()f organic sediments by
the ionium method, Geokhimiya (1966) 1183. 自4) Kaufman, A. and Broecker, W.S., Comparison of 230Th and 14C旦gesfor carbonate materials
from lakes Lahontan and Bonneville, J. Geophys, Res. 70 4039 (1965). (2日 Thl1rber,D.L., Broecker, W.S・, Potratz, H.A. and Blanchard, R.L吋 Oraniumseries ages of
Pacific atoll coral, Science 149 55 (1965). (2目 Broecker,W.S. and Thl1rber, D.L., Uranil1m series dating of corals and oolites from Bahaman
and Florida Key limestones, Scienc巴 14958 (1965).
ロ百 Osmond, J.K., Carpenter, J.P. and Windom, H.L., 230Th/234U ages of the Pleistocene corals
and oolites of Florida, J. Geophys. Res. 70 1843 (1965ト白日 8tearns, C.E. and Thurber, D.L., 230Th/234U dates of late Pleistocene marine fossils from
the Mediterranean and Moroccan littorals, Quaternaria 7 29 (1965).
凶 Richards,H.G. ancl Thurber, D.L., Pleistocene age cletermination from California and Oregon, Science 152 1091 (1966).
側 Veeh,H.H., 230Th(238U and 234U /238U ages of Pleistocene high sea level sta吋, ]. Geophys.
Res. 71 3379 (1966ト(31) Fanale, F. and Shaeffer, 0., The heril1m幽uranil1m ratios for Pleistocene and Tertiary fossil
aragonites, Science 149 312 (1965). 問 8haeffer,O.A. and Davis, R., 36Cl in nature, New York Acad. Sci. Ann. 62 105 (1955). 。3) Rightmire, R.A., Kohman, T.Pψand Hinterb巴rger,H., Cマberdie Halt制wertzeitcles langlebigen
26Al, Z. Naturforschung 13a 847 (1958). (胡 Merrill,J.R., Lyclen, E.F.X.,日onda,M. and Arnold, J.R., The sedimentary geochemistry of
the beryllil1111 isotopes, Geochim. Cosmoshim. Acta 18 108 (1960). (お) Price. P.忠B.ancl 羽7匂al肱ke町r,R ぷ4ι.,0一凶t
732 (1962).
(36) Sackett, Wふ1.,Protactinium-231 content of ocean wat
66 Kazuhisa KOMURA and Masanobu SAKANOUE
品目i) Sakanoue, M., Takagi, T. and Maeda, M., Studies of the adsorption and electrodeposition of
protactinium, R且diochimicaActa 5 607 (1966).
間 Hashimoto,T. and Sakanoue, M., Basic studies on the production of enriched uranium-237 and the measurement of its specific activity, Sci. Rept. Kanazawa Univ. 11 49 (1966).
(4日 Doke,T., A new m巴thodfor the reduction of alpha-ray background in a gridded ionization
chamber, Can. J. Phys. 40 607 (1962). 帥 Lahoud,J.A., Mi11er, D.S. and Friedman, G.M., Relationship between depositional environment
and uranium concentration of molluscan shells, J. Sed. Petro1. 36 541 (1966).
間1) Sakanoue, M., Konishi, K. and Komura, K., Stepwise determination of thorium, protactinium and uranium isotopes and their application for geochronological studies, IAEA symposium on Radioactive Dating and Methods of Low-level Counting at Monaco, SM 87/28 (1967).
(51) Blanchard, L., Relationship between uranium and radium in coastal marine shells and their
environmcont, J. Geophys. Res. 70 2911 (1965).
師団 Lowenstam, H.A., “Sr ICa ratio of skele句1aragc>nites from the recent marine biota at palau
and from fossi1 GastropodsヘIsotopicand Cosmic Chemistry (Craig, H., Miller, S.L. and Wasserburg, G.Jふ NorthHolland Pub, Co., Amsterdam, 114 (1964).
(臼i) Moore, W.S. and Sackett, W.M., Uranium and thorium series inequi1ibrium in sea water, J. Geophys. Res. 69 5401 (1964).
倒 Kuznetsov,Yu. V., Simonyak, Z.N., Eleizarova, A.N. and Lisitsin, A.P., Content of protac-tinium and thorium isotopes in ocean water, Radiokhimiya 8 455 (1966).
日,) Mii, H. and Kigoshi, K., Radiocarbon dating of the ancient beach and coastal dune deposits, Kikai Island, Amami. 14C age of Quaternary deposits in Japan XXXVIII, Earth Sci. 82 43 (1966) .
臨時 Blanchard, R.L., 234U 1238U ratio in coastal marine waters and calcium carbonates, J. Geo-
phys. Res. 70 4055 (1965).
(57) Koide, M, and Goldberg, E.D.,“Uranium-234/uranium-238 ratios in sea water", Progress in
Oceanography vo1. 3 (Sears, Mふ Pergamon,London, 173 (1965). 側 Umemoto,S., 234U 1238U in sea water from Kuroshio region, J. Geophys. Res. 70 5326 (1965).
捌 Miyake,Y., Sugimura, Y. and Uchida, T., Ratio 234U 1238U and the uranium concentration
in sea water in the western north Pacific, J. Geophys. Res. 71 3083 (1966).
(印1) Sakanoue, M. and Komura, K., The preparation of carrier-free U-232 and the estimation of
various nuc1ides produce