A Protocol for Preparing Preserved Flowerswith Natural Color and Texture

Hiroaki Ito1,5, Takahiro Hayashi2, Masaki Hashimoto3,

Katsuro Miyagawa3, Saori Nakamura3, Youichi Mizuta4,

and Susumu Yazawa4

ADDITIONAL INDEX WORDS. postharvest technique, processed flower, soakingsolvent, shrinkage, discoloration

SUMMARY. A protocol for the preparation of preserved flowers retaining naturalcolor and texture of ‘Moondust Velvet Blue’ carnations (Dianthus caryophyllus) wasdeveloped. This three-step process consists of soaking flowers in ethyl alcohol, thensoaking them in polypropylene glycol, followed by a rinse with ethyl alcohol. Somekinds of flowers processed in this manner retained their natural color and texture forat least 6 months. The physicochemical properties of appropriate solvents used forretaining natural pigmentation and texture are discussed. This protocol is applicableto 13 kinds of flowers among 30 kinds of flowers tested and adds a new dimension topostharvest techniques for cut flowers.

Preserved flowers have a long or-namental period compared withfresh flowers, and can be more

suited to flower arrangements, wed-ding bouquets, or store window deco-rations. In spite of their high processingcosts, these flowers are in high demand.Preserved flowers were first developedin 1991, and are prepared from freshflowers by replacing their internalmoisture with polyethylene glycol (deWinter-Scailteur, 1991). These pro-cessed flowers can retain their freshtexture and flexibility for several years.However, preserved flowers do notretain their natural color. Flowers areartificially stained by soaking in poly-ethylene glycol with synthetic dyes. Itis difficult to stain sepals, stems, andleaves separately from petals, thus mul-ticolor flowers are usually stainedmonotone. If the internal moisture offresh flowers can be replaced withsolvents that allow petals to retain theiroriginal color, the processed flowerswould look more natural than the cur-rently available preserved flowers.There is a need to reconsider the soak-ing solvents to address the challenge ofretaining petal color. We have studiedthe physicochemical properties of soak-ing solvents on petal tissues and have

screened solvents to determine whichsolvents allow the petals to retain theirshape and color for a long period. Thisreport introduces an established pro-tocol for preparing preserved flowersretaining natural color and texture.

Materials and methodsPLANT MATERIALS. Flowers of

‘Moondust Velvet Blue’, a geneticallyengineered carnation variety withdelphinidin-type anthocyanins (Fukuiet al., 2003), were used to establisha protocol of three combined processesbecause its pigments have already beenidentified and it shows the highestpigment content among the three‘‘Moondust’’ varieties. Cut ‘MoondustVelvet Blue’ flowers imported fromColumbia were obtained from SuntoryFlowers (Tokyo). The stem bases wereinserted in a vase with 20 mL�L–1

of a preservative containing sugars, inor-ganic ions, and antibacterial agents(Misaki; Otsuka Chemical, Osaka,Japan). Flowers after full bloom wereused within 1 week.

Thirty kinds of flowers with differ-ent pigments such as flavonoids, carot-enoids, chlorophylls, betalains, andanthocyanins were used for the appli-cability test of the established protocol.

Flowers were obtained from floristsor experimental fields and were usedimmediately.

A PROTOCOL FOR FLOWER PRO-

CESSING. Unless otherwise mentioned,ethyl alcohol was used to prevent petalshrinkage in the primary soaking pro-cess, followed by polypropylene glycol(diol type, average molecular weight:400) to prevent petal discoloration inthe secondary soaking process.

The protocol consisted of 1) soak-ing flowers in the primary solvent(ethyl alcohol) overnight; 2) soakingin the secondary solvent (100% poly-propylene glycol) overnight or 2 d; 3)rinsing the flowers for 1 min with ethylalcohol. The ‘Moondust Velvet Blue’flowers with their stems trimmed to10 cm were held upside down and thewhole flower was soaked in 500 mLplastic containers containing 300 mLof primary or secondary soaking sol-vent. Thirty other kinds of flowerswere soaked in the same way using100 to 1000 mL of solvents, depend-ing on the flower volume. The tem-perature of the solvents was 25 �C. Thecontainers were covered with plasticwrap during steps 1 and 2. After thethree treatment steps, the processedflowers were allowed to stand in a roomat 25 �C and 45% ± 25% relative hu-midity (RH) for 1 d or more. Threereplicate flowers were processed.

‘Moondust Velvet Blue’ carnationand the 30 other kinds of flowers wereevaluated and classified into one ofthree groups: 1) petals that retainedtheir color, 2) petals that retained theircolor only somewhat, and 3) petals thatalmost or completely lost their color.

Other preservation solvents suchas 1-hexanol or methyl alcohol as theprimary soaking solvents or polyethyl-ene glycol (average molecular weight:400); 2-methyl-2,4-pentanediol; or1,3-butanediol as the secondary soak-ing solvents also were tested.

ResultsFLOWER PROCESSING. A protocol

for preparing preserved ‘MoondustVelvet Blue’ carnations so that theyretain their natural color and texture

UnitsTo convert U.S. to SI,multiply by U.S. unit SI unit

To convert SI to U.S.,multiply by

29.5735 fl oz mL 0.03387.8125 fl oz/gal mL�L–1 0.1280

(�F – 32) O 1.8 �F �C (1.8 · �C) + 32

1Koshien Junior College, Nishinomiya 663-8107,Japan

2Faculty of Agriculture, Kinki University, Nara 631-8505, Japan

3Suntorymidorie Limited, Osaka 618-8503, Japan

4Graduate School of Agriculture, Kyoto University,Kyoto 606-8502, Japan

5Corresponding author. E-mail: h-itou@koshien.ac.jp.

• April 2010 20(2) 445

is detailed with bold frames in Fig. 1.Five serious quality defects were ob-served during and after the flower pre-servation process when solvents such as1-hexanol or methyl alcohol were usedinstead of ethyl alcohol as the primarysoaking solvent or polyethylene glycol;2-methyl-2,4-pentanediol; or 1,3-butanediol was used instead of poly-propyleneglycol as the secondary soak-ing solvent (Figs. 1 and 2).

The quality defects observed wereas follows: 1) Petal shrinkage occurredby soaking flowers in high-viscositysolvents (e.g., 1-hexanol) as the pri-mary soaking solvent (Fig. 2C). Theuse of polypropylene glycol, polyethyl-ene glycol, and glycerin also caused thesame defects (data not shown). 2) Petaldiscoloration occurred by soakingflowers in high-hydrophilic solvents(e.g., methyl alcohol) as the primarysoaking solvent or polyethylene glycolas the secondary soaking solvent(Fig. 2, D and E). The use of ethyleneglycol, propylene glycol, and glycerincaused the same defects (data notshown). 3) Petal shrinkage occurreddue to the volatilization of the primarysoaking solvent when the flowers werekept in air before being transferred tothe secondary soaking solvent (Fig.2F). Flowers must be soaked in thesecondary soaking solvent immediatelyafter being soaked in ethyl alcohol asthe primary soaking solvent (Fig. 2B).Severe petal shrinkage also occurredeven when a low-volatile solvent (e.g.,2-methyl-2,4-pentanediol) was used asthe secondary soaking solvent (Fig. 2J).The use of 2-(2-ethylhexyloxy)ethanol;

2,3-butanediol; and 4-hydroxy-2-butanone caused the same defects(data not shown). 4) The natural tex-ture was not obtained just after soakingtreatment with a nonvolatile stickysolvent (e.g., polypropylene glycol) be-cause it remained on the petal surface(Fig. 2I). The flower needed to berinsed with low-viscosity solvents (e.g.,ethyl alcohol) to remove the secondarysoaking solvent (Fig. 2G). The use ofpropylene glycol, glycerin, and 2-[2-(2-ethylhexyloxy)ethoxy]ethanol causedthe same defects (data not shown). 5)Petals gradually lost their color afterbeing processed using 1,3-butanediolas the secondary soaking solvent(Fig. 2K). The use of 1,3-propanediol;1,4-butanediol; and 1,5-pentanediolcaused the same defects (data notshown). Petal color was retained for atleast 6 months by using polypropyleneglycol as the secondary soaking solvent(Fig. 2L).

APPLICATION TO VARIOUS FLOWERS.Similar to ‘Moondust Velvet Blue’carnation, the flower shapes of mostspecies tested were retained when thestandard preserving protocol wasused (Fig. 3). Flower species withthin petals that are particularly easyto deform were exceptions (Fig. 3,C2, D2, K2, L2, and S2). Comparedwith the color of unprocessed flowers(Fig. 3, A1–DD1), processed petalswere grouped into those that retainedtheir color well (Fig. 3, A2–L2), thosethat somewhat retained their color(Fig. 3, M1–O2), or those that nearlyor completely lost their color (Fig. 3,P2–DD2). Some multicolor flowers

retained their color patterns (Fig. 3,B2, E2, F2, G2, and N2).

DiscussionEven under the best postharvest

handling practices, the vase life of freshflowers is relatively short and many ofthe flowers are lost during the market-ing chain (Nowak and Rudnicki,1990). Preserved flowers can retaintheir fresh texture and flexibility forseveral years but cannot retain theirnatural color. There is a need to re-consider the soaking solvents to ad-dress the challenge of retaining petalcolor. However, there are few reportson the effects of solvent treatment onpetal tissues. It has been reported thatprocessed flowers can be preparedby replacing internal moisture with2-propyl alcohol or t-butyl alcohol(Romero-Sierra and Webb, 1982).However, these solvents volatilizedfrom the petals, resulting in losingfresh texture. Replacement techniqueshave been developed for preparingmicro- and macro-specimens from or-gans or whole bodies of animals, aswell as processed flowers. Plastination(von Hagens, 1981) is an exampletechnique of replacing the internalmoisture with epoxy resin using anascending series of acetone. However,this dehydration process results in thecomplete discoloration of tissues. A re-placement technique using an ascend-ing series of ethyl alcohol also causespigments to leach out of petals duringdehydration owing to the water in thesolvent.

Fig. 1. Outline in bold frames of a protocol to preserve natural flower color, with quality defects observed during each stepif alternate solvents were used. The protocol consisted of soaking flowers in the primary solvent (ethyl alcohol) overnight,then soaking in the secondary solvent (100% polypropylene) glycol overnight or for 2 d, and then rinsing the flowers for1 min with ethyl alcohol. Figure 2 references illustrate flower appearance at different point.

446 • April 2010 20(2)

TECHNOLOGY AND PRODUCT REPORTS

Fig. 3. Photographs of fresh (1) and processed flowers (2) of various species. Processed flowers were prepared by soaking in ethylalcohol overnight, followed by overnight or 2 d soaking in polypropylene glycol (diol type, average molecular weight: 400),before rinsing with ethyl alcohol. Photographs of processed flowers were taken 1 d after treatment except for A2 and B2, whichwere taken 15 and 13 months after processing, respectively. Plant species were as follows: (A) corn flower (Centaurea cyanus),(B) dwarf delphinium (Delphinium grandiflorum), (C) spiderwort (Tradescantia ohiensis), (D) asiatic dayflower (Commelinacommunis), (E) pansy (Viola ·wittrockiana), (F) lisianthus (Eustoma grandiflorum), (G) dutch iris (Iris ·hollandica), (H)cockscomb (Celosia cristata), (I) bougainvillea (Bougainvillea spectabilis), (J) easter cactus (Rhipsalidopsis gaertneri), (K)portulaca (Portulaca grandiflora), (L) four o’clock (Mirabilis jalapa), (M) snapdragon (Antirrhinum majus), (N) cooktownorchid (Dendrobium phalaenopsis), (O) rose (Rosa ·hybrida), (P) bigleaf hydrangea (Hydrangea macrophylla), (Q) daffodil(Narcissus pseudonarcissus), (R) carnation (Dianthus caryophyllus), (S) iceland poppy (Papaver nudicaule), (T) african violet(Saintpaulia ionantha), (U) salvia (Salvia guaranitica), (V) bigleaf hydrangea, (W) camellia (Camellia japonica), (X) geranium(Pelargonium incrassatum), (Y) tulip (Tulipa gesneriana), (Z) sunflower (Helianthus annuus), (AA) pansy (V. ·wittrockiana),(BB) oil seed rape (Brassica napus), (CC) watermelon (Citrullus lanatus) and (DD), and sweet william (Dianthus barbatus).

Fig. 2. Photographs of ‘Moondust Velvet Blue’carnations at different processing stages. (A) Freshflower. (B) The flower after soaking in ethyl alcohol.(C) Petal shrinkage soon after Step 1 when 1-hexanolwas used as the primary soaking solvent. (D)Discoloration soon after Step 1 when methyl alcoholwas used as the primary soaking solvent. (E)Discoloration soon after Step 2 when polyethyleneglycol (average molecular weight: 400) was used asthe secondary soaking solvent. (F) Petal shrinkage 1 din air after Step 1 when ethyl alcohol was used asthe primary soaking solvent. (G) Flower colorimmediately after soaking in polypropylene glycol asthe secondary soaking solvent. (H) Flower color afterStep 3 rinsing with ethyl alcohol. (I) Loss ofnatural texture after �1 month of storage in air afterStep 2. (J) Petal shrinkage after�5 months storage offlowers that were prepared using 2-methyl-2,4-pentanediol as the secondary soaking solvent. (K)Discoloration after�2 months storage of flowers thatwas prepared using 1,3-butanediol as the secondarysoaking solvent. (L) Processed flower prepared usingethyl alcohol as the primary soaking solvent,polypropylene glycol as the secondary soakingsolvent, followed by rinsing with ethyl alcohol andbeing held in storage for �6 months in air.

• April 2010 20(2) 447

The moisture replacement tech-nique adopted here does not use water-containing solvents. In fact, pigmentsdiffused but remained in the petals ofdehydrated flowers. Some pigmentssuch as flavonoids may get depositedon cell walls (Markham et al., 2000).

In this protocol, ethyl alcoholand polypropylene glycol were usedas the primary and secondary soakingsolvents, respectively. Primary soak-ing solvents with a low viscosity, suchas ethyl alcohol, tended to preventpetal shrinkage (data unpublished).In addition, petal shrinkage inducedby the secondary soaking solvents wasgreatly reduced if the petals weresoaked preliminarily in such primarysoaking solvents. Polypropylene gly-col was selected as the most appropri-ate secondary soaking solvent becauseit retained the petal shape and color of‘Moondust Velvet Blue’ carnationsfor a long time. Taking the cost, thesafety, and waste disposal into ac-count, the combination of ethyl alco-hol and polypropylene glycol seemedto be a reasonable choice.

The flower color of ‘MoondustVelvet Blue’ carnation and 13 otherflower species was retained well among31 kinds of flowers tested (Figs. 2Land 3, A2–L2). Coexisting with poly-propylene glycol, most flowers con-taining yellow flavonoids did notretain their petal color, except foryellow snapdragon (Antirrhinummajus). All flowers and leaves thatincluded oil-soluble pigments, suchas carotenoids and chlorophylls, didnot retain their petal color. All flowerscontaining betalains retained theirpetal color vividly. Most flowers con-taining dephinidin-type anthocyaninsretained their petal color, except for

bigleaf hydrangea (Hydrangea macro-phylla) and salvia (Salvia guaranitica).Most flowers containing cyanidin-typeanthocyanins did not retain their petalcolor, except for corn flower (Centau-rea cyanus), which contains pigmentsthat form metal complexes with cal-cium (Ca2+), iron (Fe3+), and magne-sium (Mg2+), and copigmentationwith flavons (Kondo et al., 1998).Flowers containing peonidin-, petuni-din-, and pelargonidin-type anthocya-nins retained somewhat or did notretain their color.

Whether flower colors expressedby yellow flavonoids or anthocyaninswere retained with polypropylene gly-col depended on species. In additionto skeletons of flavonids, their concen-trations and formations may affect thestability of processed flower colors.Pigments were thought to exist in thesolvent and to be deposited in cellwalls in the processed petals. Hence,cell wall components, as well as sol-vents, may affect the formation andcolor expression of yellow flavonoidsor anthocyanins.

The study showed that the pro-cessed ‘Moondust Velvet Blue’ carna-tion retained its petal shape and colorover 6 months and the processingmethod can be applied to several kindsof flowers (Fig. 3). Some processedflowers, such as corn flower (Fig. 3A2;15 months storage after processing)and dwarf delphinium (Delphiniumgrandiflorum) (Fig. 3B2; 13 monthsstorage after processing), retained theirflower color vividly for more than 1year. The keeping quality of processedflowers seems to be influenced bymany factors, such as the cultivationmethod, harvest maturity, and storageconditions after processing. Exclusion

of direct sunlight and high humidityare essential during storage becauselight, oxygen, and humidity cause thecolor to fade and the flowers to deform.Studies on cultivation method and stor-age conditions for processed flowersmay lead to improvements in the keep-ing quality of processed flowers.

Literature citedde Winter-Scailteur, N. 1991. Long-lifecut flowers and method of treatment forobtaining same. WO91/03160. Euro-pean Patent Office, Munich, Germany.

Fukui, Y., T. Tanaka, K. Kusumi, T.Iwashita, and K. Nomoto. 2003. A ratio-nale for the shift in colour towards blue intransgenic carnation flowers expressingthe flavonoid 3#,5#-hydroxylase gene.Phytochemistry 63:15–23.

Kondo, T., M. Ueda, M. Isobe, and T.Goto. 1998. A new molecular mechanismof blue color development with protocya-nin, a supramolecular pigment from corn-flower, Centaurea cyanus. TetrahedronLett. 39:8307–8310.

Markham, K.R., K.G. Ryan, K.S. Gould,and G.K. Rickards. 2000. Cell wall sitedflavonoids in lisianthus flower petals. Phy-tochemistry 54:681–687.

Nowak, J. and R.M. Rudnicki. 1990.Postharvest handling and storage of cutflowers, florist greens, and potted plants.Timber Press, Portland, OR.

Romero-Sierra, C. and J.C. Webb. 1982.Flower preservation. United States Patent4,349,459. U.S. Patent and TrademarkOffice, Washington, DC.

von Hagens, G. 1981. Animal and vegetaltissues permanently preserved by syn-thetic resin impregnation. United StatesPatent 4,278,701. U.S. Patent and Trade-mark Office, Washington, DC.

448 • April 2010 20(2)

TECHNOLOGY AND PRODUCT REPORTS