Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Computer-aided technique for blending wine : application of simplex optimization to headspace gas chromatographic… Datta, Seema 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1990_A6_7 D37.pdf [ 4.78MB ]
Metadata
JSON: 831-1.0098116.json
JSON-LD: 831-1.0098116-ld.json
RDF/XML (Pretty): 831-1.0098116-rdf.xml
RDF/JSON: 831-1.0098116-rdf.json
Turtle: 831-1.0098116-turtle.txt
N-Triples: 831-1.0098116-rdf-ntriples.txt
Original Record: 831-1.0098116-source.json
Full Text
831-1.0098116-fulltext.txt
Citation
831-1.0098116.ris

Full Text

COMPUTER-AIDED TECHNIQUE FOR BLENDING WINE: APPLICATION OF SIMPLEX OPTIMIZATION TO HEADSPACE GAS CHROMATOGRAPHIC PROFILES OF WINE By SEEMA DATTA B. Sc. (Agr.) U n i v e r s i t y of B r i t i s h Columbia, 198 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of Food Science We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1989 Seema Datta, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT An objective method to blend wines for standardizing flavor q u a l i t y was developed. Aroma v o l a t i l e s of v a r i e t a l and white stock wines were analyzed at 6<>C and 37°C by headspace gas chromatography with cryofocussing. Pattern s i m i l a r i t y constants of the chromatographic p r o f i l e s were entered into the simplex optimization program which determined the best blending rat i o s of wines to simulate the target wine. Thirteen and 23 vertices were required to give the optimum response for t r i a l s 1 and 2, respectively. For both t r i a l s the computer optimized blends could not be d i f f e r e n t i a t e d from the target wines by a sensory taste panel consisting of both untrained and expert judges. i i TABLE OF CONTENTS ABSTRACT i TABLE OF CONTENTS i i LIST OF TABLES v LIST OF FIGURES v i LIST OF APPENDICES v i i i ACKNOWLEDGEMENT i X I . INTRODUCTION 1 I I . LITERATURE REVIEW 3 A. Commercial Winemaking Operation 3 1. Harvesting 3 2. Stemming and Crushing 3 3. S u l f i t i n g 4 4. Pres s i n g 5 5. Fermentation 5 6. C l a r i f i c a t i o n , Aging and F i l t e r a t i o n 5 B. The Flavor of Wine ..6 1. Higher Alcohols (Fusel O i l s ) ...7 2. F a t t y Acids 7 3. F a t t y Acid Esters 8 4. Esters 8 5. Carbonyls 9 6. Terpenes 10 7. Hydrocarbons 10 8. V o l a t i l e Phenols 10 C. V o l a t i l e A n a l y s i s of Wine 11 i i i 1. L i q u i d - L i q u i d E x t r a c t i o n 11 2. Headspace E x t r a c t i o n Techniques 14 3. Purge and Trap Headspace A n a l y s i s 15 4. S t a t i c Headspace A n a l y s i s 17 D. Simplex O p t i m i z a t i o n 20 1. EVOP 21 . 2. Simplex EVOP 21 3. Simplex Method --24 4. Improvements i n the Method ..24 I I I . MATERIALS AND METHODS 27 A. Samples 27 1. Orange J u i c e 27 2. Grape J u i c e C o n c e n t r a t e s 27 3. Apple J u i c e s 27 4. Wines 27 B. Sample P r e p a r a t i o n 28 C. I n s t r u m e n t a l A n a l y s i s 29 1. Wine and Apple J u i c e 29 i . Gas Chromatography ....29 i i . Headspace Sampling 30 2. Orange J u i c e and Grape J u i c e C o n c e n t r a t e . . . . 30 i . Gas Chromatography 30 i i . Headspace Sampling 31 D. I n t e r n a l S tandard 31 E. O p t i m i z a t i o n 32 F. T i t r a t a b l e A c i d i t y and T o t a l S o l u b l e S o l i d s 34 1. T i t r a b l e A c i d i t y 34 i v 2. T o t a l Soluble S o l i d s 34 G. Sensory E v a l u a t i o n 35 IV. RESULTS AND DISCUSSION 36 A. Method Development. 36 1. P r e l i m i n a r y Work 37 2. Cold Trapping 42 3. C a p i l l a r y Column 51 B. Wine Headspace A n a l y s i s 53 C. P r e c i s i o n and I n t e r n a l Standard 56 1. P r e c i s i o n 56 2. I n t e r n a l Standard 59 D. Simplex Optimization 68 1. Blending Optimization 73 E. Adjustments for A c i d i t y and Sweetness 78 F. V e r i f i c a t i o n of Results 81 1. Sensory E v a l u a t i o n 81 2. S i m i l a r i t y Constants of Blends 85 V. CONCLUSIONS 88 VI. REFERENCES. 91 V I I . APPENDIX 99 v LIST OF TABLES Table Page 1. Repeatibi1ity of headspace method using apple juice samples 57 2. R e p e a t i b i l i t y of the internal standard 58 3. Blending and target wines for t r i a l s 1 and 2 .70 4. Pattern s i m i l a r i t y constants for t r i a l 1 and 2 71 5. Factors and their l i m i t s for the blending optimization of t r i a l 1 and 2 wines ..74 6. Blending optimization of t r i a l 1 wine 76 7. Blending optimization of t r i a l 2 wine 77 8. Blending ra t i o s of computer-aided blends and commercial blends for t r i a l s 1 and 2 79 9. T i t r a t a b l e a c i d i t y of the computer optimized blends commercial blends, and the target wines...... 80 10. Total soluble so l i d s for the blends and targets t r i a l s 1 and 2 .' 82 11. Results of the triangle test comparing computer optimized blends and commercially formulated blends with the target wines 84 v i LIST OF FIGURES Figure Page 1. Single-factor-at-a-time strategy on a well behaved response surface (Massert et a l . , 1988) 22 2. Single-factor-at-a-time strategy on a response surface exhibitaing a diagonal ridge (Massert et a l . , 1988) 23 3. Chromatogram of headspace v o l a t i l e s from fresh orange juice analyzed at 70°C 38 4. The eff e c t of temperature of e q u i l i b r a t i o n on peak area for orange juice 39 5. The ef f e c t of temperature of e q u i l i b r a t i o n on peak area for wine .40 6. Chromatogram of headspace v o l a t i l e s from fresh orange juice analyzed at 70°C using cryofocussing....43 7. Headspace v o l a t i l e s from grape juice concentrate at 55°C (sample 1) 45 8. Headspace v o l a t i l e s from grape juice concentrate at 55°C (sample 2) 46 9. Analysis of Winesap var i e t y apple juice at 55°C 48 10. Analysis of Sinta va r i e t y apple juice at 55°C 49 11. Headspace analysis of wine (Leibesheim) at 55°C 50 12. The ef f e c t of increasing concentration of internal standard on peak area 61 13. Representative HSGC p r o f i l e of a Chenin blanc v a r i e t a l wine of t r i a l 1 62 14. Representative HSGC p r o f i l e of a white stock wine of t r i a l 1 63 15. HSGC p r o f i l e of the target wine, Leibesheim, for t r i a l 1.... 64 16. Representative HSGC p r o f i l e of a Verdelet v a r i e t a l wine of t r i a l 2 65 17. Representative HSGC p r o f i l e of a white stock wine of t r i a l 2 66 v i i 18. HSGC p r o f i l e of the target wine, Cuvee white, for t r i a l 2 67 19. Chromatograms of headspace v o l a t i l e s from (a) the target, (b) commercial blend and (c) the computer optimized blend for t r i a l 1 86 20. Chromatograms of headspace v o l a t i l e s from (a) the target, (b) commercial blend and (c) the computer optimized blend for t r i a l 2 87 v i i i LIST OF APPENDICES Appendix Page 1. GC Data E n t r y computer program 99 2. GC Data C o r r e c t i o n computer program 102 3. S i m i l a r i t y Constant computer program 106 4. Blending O p t i m i z a t i o n computer program 107 5. S i m i l a r i t y Constant of Blend computer program 113 6. S i m i l a r i t y Constant of Blend, Data C o r r e c t i o n computer program 116 ix ACKNOWLEDGEMENTS I wish to express my appreciation to Dr. Nakai for his support and encouragement in the course of t h i s study. The assistance of Dr. W. Powrie, Dr. J. Vanderstoep and Dr. T. Durance is also acknowledged. A further word of thanks goes to the Bio-Resource Engineering Department of U.B.C. for the use of their laboratory during the early part of my thesis work. I am also grateful for the genetous donation of f a c i l i t i e s by Agriculture Canada and the assistance of the s t a f f of the Agriculture Research Station at Summerland, B.C. F i n a l l y , my sincere thanks to Lynn Stark of Brights Wines, Oliver, B.C. for a l l her cooperation with th i s project. x INTRODUCTION M a i n t a i n i n g product q u a l i t y constant i s of primary concern to food manufacturers. Blending i s a common and necessary p r a c t i c e for s t a n d a r d i z i n g product q u a l i t y i n the food i n d u s t r y f o r many foods i n c l u d i n g wine (Vine/ 1981; Peynaud, 1984; J a c k i s h , 1985; Rankine, 1988), d i s t i l l e d beverages (Lang, 1983), tea (Theobald, 1977), c i t r u s j u i c e s (Charley, 1969; Cook, 1983; Anon., 1987), c o f f e e , processed cheese and processed meat products. A large p r o p o r t i o n of the world's wines, both o r d i n a r y and f i n e , are blended ( J a c k i s h , 1985). Wines may be blended f or c o l o r , t a s t e , a l c o h o l and body, and aroma (Peynaud, 1984 ). Blending for a l l of these q u a l i t y parameters except aroma i s guided by p h y s i c a l or chemical a n a l y s i s . Blending for aroma, on the other hand, i s dependent on sensory a n a l y s i s al.one; i t i s a d e l i c a t e operation that r e q u i r e s a great deal of experience and s k i l l . I f the winemaker i s attempting to d u p l i c a t e the aroma of a wine blend from a previous year, the s t r a t e g y w i l l be to f i r s t to i d e n t i f y the strengths and weaknesses of the a v a i l a b l e stocks. Normally, one wine i s se l e c t e d as the primary stock while the others are i d e n t i f i e d as secondary blending stocks. T r i a l blends are made by the winemaker on the basis of sensory e v a l u a t i o n . These blends are then tested by experienced wine t a s t e r s p r i o r to the f i n a l blending. Many t r i a l s may be necessary to match the f l a v o r of the targe t product, e s p e c i a l l y i f i t i s a complex blend. 1 T r a d i t i o n a l methods for b l e n d i n g for the purpose of m a i n t a i n i n g u n i f o r m i t y of wine f l a v o r are d i f f i c u l t , time consuming and s u b j e c t i v e . C l e a r l y , a more r a p i d and o b j e c t i v e q u a l i t y t e s t method i s d e s i r e d . A i sh ima et a l . , (1987) deve loped an o b j e c t i v e system f o r f i n d i n g the best b l e n d i n g r a t i o s of s t r a w b e r r y essences wi th the c o n c e n t r a t e d s t r a w b e r r y j u i c e to s i m u l a t e the aroma of f r e s h j u i c e . The idea was to maximize the s i m i l a r i t y c o e f f i c i e n t c a l c u l a t e d between gas chromatograph ic p a t t e r n s of the f r e s h j u i c e and a b l end of the c o n c e n t r a t e wi th the essences u s i n g two d i f f e r e n t s implex o p t i m i z a t i o n programs, computer ized and e x p e r i m e n t a l . Both programs were s u c c e s s f u l i n f i n d i n g optimum b l e n d i n g r a t i o s . T h i s method, however, can be improved to make i t more f e a s i b l e for q u a l i t y c o n t r o l purposes by employing a l e s s t e d i o u s and c o m p l i c a t e d method of v o l a t i l e a n a l y s i s and e l i m i n a t i n g the e x p e r i m e n t a l s implex o p t i m i z a t i o n t h a t r e q u i r e s many t r i a l b lends to be a n a l y z e d . The o b j e c t i v e s of the present r e s e a r c h - w e r e : (1) To deve lop a headspace gas chromatographic method to a n a l y z e wine. (2) To use computer ized s implex o p t i m i z a t i o n to determine b l e n d i n g r a t i o s of v a r i e t a l and white s tock wines to. s i m u l a t e the aroma of the p r e v i o u s y e a r ' s wines . (3) To compare b lends of the c o m p u t e r - a i d e d t echn ique and t r a d i t i o n a l method wi th the t a r g e t product on the b a s i s of s ensory t e s t s . 2 LITERATURE REVIEW A. COMMERCIAL WINEMAKING OPERATION  H a r v e s t i n g The making of wine r e a l l y beg ins a t the time of h a r v e s t i n g ; the grapes must be p i c k e d a t the proper s tage of m a t u r i t y and they must be of sound q u a l i t y (Amerine and S i n g l e t o n , 1977). To f o l l o w the p r o g r e s s of r i p e n i n g , grape b e r r i e s sh ou l d be sampled r e g u l a r i l y for d e t e r m i n i n g the average c o n c e n t r a t i o n of sugar and a c i d i t y of the crop so t h a t the h a r v e s t date can be s e t . H a r v e s t i n g sh ou ld be done as to a v o i d damaging and b r u i s i n g b e r r i e s . O x i d a t i o n and m a c e r a t i o n of the grapes be fore they reach the winery can be d e t r i m e n t a l to the f i n a l q u a l i t y of the wine (Peynaud, 1984) . Stemming and C r u s h i n g P r o c e s s i n g begins, as soon as the f r u i t a r r i v e s a t the w i n e r y . The f i r s t s t ep i s to remove stems as they c o n t r i b u t e to the t a n n i c a c i d a s t r i n g e n c y ( V i n e , 1981). C r u s h i n g breaks open each of the b e r r i e s to a l l o w r e l e a s e of p u l p and j u i c e d u r i n g p r e s s i n g . For making red wine, t h i s o p e r a t i o n f a c i l i t a t e s c o n t a c t and f ermentaiori, by y e a s t ( V i n e , 1981; Amerine and S i n g l e t o n , 1977) . Both stemming and c r u s h i n g o p e r a t i o n s are n o r m a l l y accompl i shed t o g e t h e r i n a c r u s h e r -stemmer. T h i s machine c o n s i s t s of l a r g e h o r i z o n t a l c y l i n d e r t h a t i s p e r f o r a t e d wi th h o l e s l a r g e enough to a l l o w b e r r i e s to 3 pass through but not the stems and a r o t a t i n g a x l e f i t t e d w i t h a s e r i e s of p a d d l e s . When t h i s a x l e r o t a t e s a t h i g h speeds , the crushed b e r r i e s , now c a l l e d must, are r e l e a s e d through the h o l e s i n t o a c o l l e c t i n g b a s i n (Amerine and S i n g l e t o n , 1977). S u l f i t i n g Whi le the p r a c t i c e of employing s u l f u r d i o x i d e as an a n t i s e p t i c agent for wines i s of a n c i e n t o r i g i n (Ough and Amer ine , 1988), s u l f i t i n g the must i s a r e l a t i v e l y r e c e n t p r a c t i c e (Peynaud, 1984) . Compressed s u l f u r d i o x i d e i s commonly employed by l a r g e w i n e r i e s whi l e s m a l l e r o p e r a t i o n s add p o t a s s i u m or sodium m e t a b i s u l f i t e , sodium s u l f i t e or sodium b i s u l f i t e . S u l f u r d i o x i d e i s added to musts or wines to c o n t r o l u n d e s i r a b l e m i c r o o r g a n i s m s , to i n h i b i t browning enzymes and to serve as an a n t i o x i d a n t (Amerine and S i n g l e t o n , 1977). Pres s ing D i f f e r e n t winemaking procedures are c a r r i e d out once the s u l f i t e d must i s ready for f u r t h e r p r o c e s s i n g . I f white grapes are be ing v i n i f i e d , the crushed grapes are pres sed to s eparate the j u i c e from the s o l i d s ( s k i n s , seeds , some pulp) c a l l e d pomace, be fore f e r m e n t a t i o n . Red g r a p e s , on the other hand, are pres sed a f t e r f e r m e n t a t i o n . Rack and c l o t h p r e s s e s , basket pres se s and cont inuous pres se s are used by w i n e r i e s (Amerine and S i n g l e t o n , 1977). 4 Fermentation White grapes are not fermented on s k i n s as reds because the leucoanthocyanin pigment c o n t r i b u t e s undesirable c o l o r s and f l a v o r s to white wine (Vine, 1981). The must, f r e s h j u i c e or crushed grapes, i s in o c u l a t e d with a pure s t a r t e r c u l t u r e of Saccharomyces cezeviseae. Fermentation i s conducted at low temperatures of 18° to 20°C (Peynaud, 1984 ) but even lower temperatures (10° to 15.5°C) are recommended (Amerine and S i n g l e t o n , 1977). Higher temperatures of 26° to 30°C are s u i t e d f o r making red wine to a l l o w for thorough maceration of the grapes and r a p i d fermentaion (Peynaud, 1984). C l a r i f i c a t i o n . Aging and F i l t e r a t i o n A f t e r fermentation i s complete, the wine i s allowed to stand to c o l l e c t yeast c e l l s and other f i n e suspended m a t e r i a l c a l l e d lees at the bottom of the container. The wine i s racked by pumping i t out of the fermentation container without d i s t u r b i n g the l e e s . Racking may be c a r r i e d out s e v e r a l times p r i o r to the aging p e r i o d . Wine i s c l a r i f i e d f u r t h e r by f i n i n g . Bentonite i s a montmorilonite c l a y that has been used for wine c l a r i f i c a t i o n and s t a b i l i z a t i o n , e s p e c i a l l y for c l e a r i n g cloudiness caused by p r e c i p i t a t i n g p r o t e i n s . I t removes p r o t e i n s , m e t a l l i c hazes and adsorbs n u t r i e n t s necessary for m i c r o b i a l growth and enzymes (Vine, 1981). Now the wine i s ready to be aged i n s t a i n l e s s s t e e l tanks or i n wooden b a r r e l s . A f i n a l f i l t e r a t i o n i s c a r r i e d out j u s t before b o t t l i n g . 5 B.THE FLAVOR OF WINE A r e c e n t rev i ew by Nykanen and Soumalainen ( 1 9 8 3 ) i n d i c a t e d t h a t some 1 , 3 0 0 v o l a t i l e compounds have been i d e n t i f i e d i n a l c o h o l i c beverages . Over 550 f l a v o r components have been r e p o r t e d i n wines ( W i l l i a m s , 1 9 8 2 ) . F l a v o r subs tances occur i n wines i n a wide range of c o n c e n t r a t i o n s from nanograms to grams per l i t e r ( S c h r e i e r , 1 9 7 9 ) . A c e r t a i n p r o p o r t i o n of wine v o l a t i l e s are b e l i e v e d to o r i g i n a t e from the grape i t s e l f and are thought to remain unchanged d u r i n g the winemaking p r o c e s s . Some of these o r i g i n a l components, however, ac t as p r e c u r s o r s and are changed d u r i n g the f e r m e n t a t i o n s t ep ( W i l l i a m s , 1 9 8 2 ; S c h r e i e r , 1 9 7 9 ) . A c o n s i d e r a b l e number of new aroma p r o d u c t s r e s u l t from the a c t i v i t y of yeas t s on the sugar s u b s t r a t e . Aging too c o n t r i b u t e s to the f i n a l f l a v o r of the wine . A l t h o u g h s e v e r a l f a c t o r s may be c o n s i d e r e d important i n i m p a r t i n g f l a v o r components to wine, i t i s g e n e r a l l y assumed t h a t most of the aroma c o n s t i t u e n t s a r i s e through the a c t i o n of y e a s t s d u r i n g f erment ion (Webb and M u l l e r , 1 9 7 2 ; Soumala inen, 1 9 7 1 ) and t h a t they are r e s p o n s i b l e for the body of wine aroma (Nykanen and Soumalainen 1 9 8 3 ; Nykanen, 1 9 8 6 ; Montedoro and B e r t u c c i o l i , 1 9 8 6 ; S c h r e i e r , 1 9 7 9 ) . A l s o , u n l i k e some foods , i t i s not p o s s i b l e to r e f e r to any ' c h a r a c t e r impact ' compound t h a t i s r e s p o n s i b l e for the t y p i c a l aroma of wine (Nykanen and Soumala inen , 1 9 8 3 ) ; the f l a v o r c h a r a c t e r of wine r e s u l t s from a complex mixture of aroma components (Webb and M u l l e r , 1 9 7 2 ) . 6 Higher Alcohols (Fusel O i l s ) Higher a l c o h o l s produced during fermentation, c o n t r i b u t e to the complex f l a v o r of wine (Ough and Amerine, 1988). Q u a n t i t a t i v e l y , f u s e l a l c o h o l s are the l a r g e s t group of f l a v o r compounds i n wines (Nykanen, 1986). This group includes a l i p h a t i c a l c o h o l s such as 1-propanol, 2-methyl-propanol, 2-methyl-l-butanol and 3-methyl-l-butanol and an aromatic a l c o h o l , phenethyl a l c o h o l (Nykanen and Soumalainen, 1983; Nykanen, 1986). O r i g i n a l l y i t was thought that f u s e l a l c o h o l s r e s u l t e d from the c a t a b o l i c conversion of the amino acids to the a l c o h o l s or the E h r l i c h mechanism.. Later i n v e s t i g a t i o n s , however, showed that f u s e l a l c o h o l s can be formed by an anabolic process from sugars and that the E h r l i c h pathway could account for only a small p o r t i o n of the f u s e l o i l s formed. F a t t y Acids A large number of free f a t t y acids have been i d e n t i f i e d i n wines but r e l a t i v e l y few are s u f f i c i e n t l y v o l a t i l e to c o n t r i b u t e to odor. A c e t i c , p r o p i o n i c and butanoic acids are found i n high enough concen t r a t i o n to c o n t r i b u t e to the aroma of wine. A l l of the a l i p h a t i c acids are odorless at the l e v e l s present i n wine (Montedoro and B e r t u c c i o l i , 1986). F a t t y a c i d s y n t h e s i s by the yeast c e l l r e q uires acetyl-coA. 7 F a t t v A c i d E s t e r s L i k e the f a t t y a c i d s , f a t t y a c i d e s t e r s are formed d u r i n g f e r m e n t a t i o n and a l s o r e q u i r e a c y l - c o A for s y n t h e s i s (Montedoro and B e r t u c c i o l i , 1986) . They are the l a r g e s t group of f l a v o r compounds and are c o n s i d e r e d of much importance to the odor of a l c o h o l i c beverages . E s t e r s E s t e r s c o n t r i b u t e e x t e n s i v e l y to the aroma of wines , p a r t i c u l a r l y to wines wi th s t r o n g f r u i t y aromas ( W i l l i a m s , 1982) . Lower f e r m e n t a t i o n temperatures seem not o n l y to r a i s e the t o t a l content of e s t e r s but a l s o the content of those t h a t i n c r e a s e the f r u i t y s e n s o r y response (Ough and Amerine , 1988) . I s o a m y l - a c e t a t e , h e x y l a c e t a t e (Simpson, 1979a) and e s p e c i a l l y 2 , 6 - 6 - t r i m e t h y l - 2 v i n y l - 4 - a c e t o x y t e t r a h y d r o - p y r a n e ( W i l l i a m s , 1982) s t r o n g l y c o n t r i b u t e to wines wi th i n t e n s i v e f r u i t y aroma. E s t e r s important to the f l a v o r of wines may a l s o o r i g i n a t e from the grape i t s e l f or may be formed d u r i n g the a g i n g p e r i o d . M e t h y l a n t h r a n i l a t e , a " c h a r a c t e r impact ' compound of n a t i v e American grape v a r i e t i e s ( V i t i s labrusca) and r e l a t e d h y b r i d s , i s not found i n European V i t i s v i n e f e r a grape v a r i e t i e s ( S c h r e i e r , 1979; S h e w f e l t , 1986) . I t can be i s o l a t e d from grapes and wines wi th F r e o n 11 (Ough and Amerine , 1988) . With s t o r a g e , the v o l a t i l e e s t e r content i n c r e a s e s making wines more f l a v o r s o m e . 8 Carbonvls Although many aldehydes and ketones have been found i n wines, most of them e s p e c i a l l y ketones, have l i t t l e sensory importance compared to e s t e r s . D i a c e t y l and acetaldehyde are more important c o n t r i b u t o r s to the aroma of s h e r r i e s than other wines. Acetaldehyde, an intermediatary product of yeast metabolism from pyruvate, has a dry choking aroma that i s r e s p o n s i b l e for the o x i d i z e d note of t a b l e wine (Montedoro and B e r t u c c i o l i , 1986). I t s concentration i s an i n d i c a t o r of the o x i d a t i o n s t a t e of a wine ( S c h r e i e r , 1979). Other aldehydes, hexenal, trans-hex-2-enal and cis-hex-3-enal have a grassy odor that may be t y p i c a l of wines made from unripe grapes (Montedoro and B e r t u c c i o l i , 1986). D i a c e t y l and 3-hydroxybutan-2-one have sweet sugary aromas and occur i n wines i n concentrations up to 3 and 30 mg/L, r e s p e c t i v e l y . Many of the aldehydes found i n wines may be products of lipoxygenase a c t i v i t y . L i n o l e i c and l i n o l e n i c acids are found in the bloom of the s k i n of grapes and are considered to be p o s s i b l e precursors of these components ( S c h r e i e r , 1979). Only a few aldehydes have been detected among the wine aroma c o n s t i t u e n t s . Studies have shown that aldehydes can be reduced to t h e i r r e s p e c t i v e a l c o h o l s during fermentation. A l s o , some aldehydes are converted to b i s u l f i t e a d d i t i o n products when they react with s u l f u r d i o x i d e and are not ext r a c t e d i n i s o l a t i o n procedures because of t h e i r high water s o l u b i l i t y ( S c h r e i e r , 1979). Ketones i n wines normally have l i t t l e sensory impact 9 ( S c h r e i e r , 1979) . Terpenes The d e l i c a t e aroma of many c l a s s i c a l wines has been a t t r i b u t e d to t erpenes and t h e i r v a r i o u s o x i d a t i o n p r o d u c t s , i n c l u d i n g l i n a l o o l , g e r a n i o l and rose o x i d e , and l i n a l o o l o x i d e . Most t erpenes o r i g i n a t e i n grapes o c c u r r i n g as g l u c o s i d e s ( W i l l i a m s , 1982) whi le some may be m e t a b o l i z e d by microorganisms ( S c h r e i e r , 1979) . Hydrocarbons Many C12 and C18 a l k a n e s , s t y r e n e and terpene hydrocarbons have been i d e n t i f i e d i n wines but one t h a t i s of importance to aged wine i s 3 , 8 - 8 - t r i m e t h y l d i h y d r o n a p h t h a l e n e . I t deve lops upon s torage and at c o n c e n t r a t i o n s of 20 to 100 ppb and imparts a kerosene or b o t t l e - a g e d c h a r a c t e r to the wine. ( W i l l i a m s , 1982) . V o l a t i l e Phenols The importance of p h e n o l i c substances to the t a s t e and odor of wines has been rev iewed by S i n g l e t o n and Noble (1976). P o l y p h e n o l s ( t a n n i n s ) are r e s p o n s i b l e for the a s t r i n g e n c y of wine . They p r e c i p i t a t e p r o t e i n s of the s a l i v a and the mucous s u r f a c e s c a u s i n g a c o n t r a c t i n g d r y m o u t h f e e l . V o l a t i l e phenols i n wine range from p h e n o l , c r e s o l s wi th m e d i c i n a l odors to more p l e a s a n t odorants such as v a n i l l i n and methyl s a l i c y c l a t e . A l l odorous p h e n o l i c compounds wi th the e x c e p t i o n of a c e t o -10 v a n i l l o n e , are not found i n grapes ; i t i s l i k e l y t h a t they are m e t a b o l i c p r o d u c t s of microorganisms or c l eavage p r o d u c t s of h i g h e r phenols t h a t o r i g i n a t e i n the grapes ( S h r e i e r , 1979) . E t i e v a n t (1981) demonstrated t h a t v o l a t i l e phenols o r i g i n a t e from m i c r o b i o l o g i c a l pathways. C . VOLATILE ANALYSIS OF WINE F l a v o r compounds c o n s t i t u t e o n l y a v e r y s m a l l p a r t of foods and beverages , as they are p r e s e n t a t p a r t s per m i l l i o n to p a r t s per b i l l i o n l e v e l s ( F l a t h and Sugisawa, 1981). I s o l a t i o n of these aroma substances i s made more d i f f i c u l t f o r wines and other a l c o h o l i c beverages because they c o n t a i n l a r g e q u a n t i t i e s of water and e t h a n o l . Rapp (1981) emphasized t h a t the method s e l e c t e d for a n a l y z i n g v o l a t i l e s shou ld r e s u l t i n an e t h a n o l - f r e e aroma c o n c e n t r a t e and t h a t a l t h o u g h methods such as f r e e z i n g o u t , s a l t i n g or d i s t i l l i n g a c h i e v e e n r i c h m e n t , they a l s o c o n c e n t r a t e e t h a n o l and t h e r e f o r e are not a p p r o p r i a t e enr ichment t echn iques for t r a c e components i n a l c o h o l i c beverages . Many i s o l a t i o n and c o n c e n t r a t i o n t e c h n i q u e s d e s c r i b e d i n the l i t e r a t u r e for wines can be grouped e i t h e r as l i q u i d - l i q u i d e x t r a c t i o n or gas e x t r a c t i o n . L i q u i d - L i q u i d E x r a c t i o n One of the commonly employed methods for i s o l a t i n g f l a v o r compounds i s by e x t r a c t i o n wi th an o r g a n i c s o l v e n t . Most 11 researchers have r e l i e d on the s e l e c t i v i t y of solvents or mixtures of solv e n t s such as pentane (Williams and Tucknott, 1973; Chaudhary et a l . , 1968 ), pentane-dicholromethane (Lamikanra, 1987; Schaefer et a l . , 1983), Freon 11 (trichloromonofluoromethane) (Hardy, 1969; Williams and Tucknott, 1973; Stevens et a l . , 1969; Cobb and Bursey, 1978; and Nelson et a l . , 1978), Freon-methylene c h l o r i d e (Guntert et a l . , 1986), and Freon 113 (Nelson and Acree, 1978) to obtain an ethanol-free e x t r a c t . Methylene c h l o r i d e (Brander, 1974; Brander et a l . , 1980; Kwan and Kowlaski, 1980; van Wyk et a l . , 1967a,1967b; Slingsby et a l . , 1980) was used i n s t u d i e s which involved separating organic and n e u t r a l f r a c t i o n s from wine. Hardy (1969) i n v e s t i g a t e d the s u i t a b i l i t y of Freon 11 as a solvent for ana l y s i n g a l c o h o l i c beverages. Recoveries of a l c o h o l s , ketones and e s t e r s from 10% aqueous ethanol i n a model system using continuous e x t r a c t i o n were high for a l c o h o l s of C5 and above and low for C4 and below. E x t r a c t i o n e f f i c i e n c y of the other v o l a t i l e s was high. This demonstrated that Freon 11 i s a w e l l s u i t e d solvent for a l c o h o l i c beverages. Furthermore, Hardy (1969) found that by adding propylene g l y c o l to Freon 11, higher a l c o h o l s which were the major components of the Freon 11 e x t r a c t s were removed. This allowed e s t e r s and other minor components to be concentrated a f u r t h e r f i v e to ten t i me s . Pentane seems to behave much l i k e the e x t r a c t a n t Freon-propylene g l y c o l . In a comparative study, Williams and Tucknott (1973) examined ether, pentane and Freon 11. 12 Separatory funnel e x t r a c t i o n s of ethanol s o l u t i o n s of e s t e r s and a l c o h o l s showed p r e f e r e n t i a l removal of e s t e r s by pentane. Freon 11 and ether were l e s s s e l e c t i v e or not s e l e c t i v e at a l l . When a continuous l i q u i d - l i q u i d e x t r a c t o r was used, pentane was no longer s e l e c t i v e . Since ether i s not s e l e c t i v e against ethanol, the authors concluded that for exhaustive e x t r a c t i o n of a l l components other than ethanol, pentane or Freon 11 should be used i n a continuous e x t r a c t o r and that for s e l e c t i v e e x t r a c t i o n of e s t e r s i n the presence of a l c o h o l s pentane should be employed with l i t t l e f r a c t i o n a t i o n of the s o l v e n t . In a l a t e r comparative study, Cobb an Bursey (1978) supported the f i n d i n g of Hardy (1969) and Williams and Tucknott (1973) i n that Freon i s a s u i t a b l e solvent for i s o l a t i n g v o l a t i l e s from wines. Commonly used e x t r a c t i n g solvent for grape j u i c e s and wines i n c l u d i n g d i e t h y l ether, dichloromethane, 2-methylbutane and Freon 11 were compared. A model system c o n t a i n i n g nine f l a v o r compounds found i n Concord wine i n a 12% ethanol-water mixture were ext r a c t e d with these four s o l v e n t s using a separatory funnel. S u b s t a n t i a l losses occurred during the i s o l a t i o n procedure, i n p a r t i c u l a r for ether and isopentane. O v e r a l l , Freon 11 extracted more of the compounds at a higher e f f i c e i n c y than the other s o l v e n t s . Dichloromethane was a close second having s i m i l a r r e s u l t s while ether and isopentane f a i r e d much poorer. One innovation i n f l a v o r e x t r a c t i o n i n foods i s the use of s u p e r c r i t i c a l carbon d i o x i d e . S u p e r c r i t i c a l carbon dioxide e x t r a c t i o n has become very popular r e c e n t l y i n food f l a v o r 13 a n a l y s i s a l t h o u g h no a p p l i c a t i o n s for wine have been r e p o r t e d . One e x p l a n a t i o n for t h i s may be t h a t t h i s e x t r a c t a n t may behave l i k e l i q u i d carbon d i o x i d e i n t h a t i t w i l l s e l e c t i v e l y e x t r a c t a l c o h o l s . S c h u l t z and R a n d a l l (1970) have r e p o r t e d t h a t e t h a n o l i s f u l l y m i s c i b l e w i th l i q u i d carbon d i o x i d e . Headspace E x t r a c t i o n Techniques Headspace t e c h n i q u e s o f f e r s e v e r a l advantages over e x t r a c t i o n t e c h n i q u e s i n p a r t i c u l a r , they produce r e s u l t s more r e p r e s e n t a t i v e of the v o l a t i l e s as they would be s m e l l e d . In g e n e r a l , s o l v e n t e x t r a c t i o n methods i s o l a t e and c o n c e n t r a t e a l l of the v o l a t i l e components t h a t c o n t r i b u t e to the f l a v o r of a food wh i l e headspace methods u s u a l l y e x t r a c t the lower b o i l i n g p o i n t components tha t are present i n h i g h c o n c e n t r a t i o n . In a d d i t i o n s o l v e n t methods r e q u i r e long sample p r e p a r a t i o n t i m e s . For example, e x t r a c t i o n of v o l a t i l e s from P i n o t n o i r wines w i th p e n t a n e - d i c h l o r o m e t h a n e r e q u i r e d 8 h r s . ( S c h r e i e r et a l . 1980) whi l e the method of Hardy (1969 ) took 17 h r s . to comple te . F i n a l l y , t o t a l v o l a t i l e methods have the a d d i t i o n a l problem of p o t e n t i a l l y e n r i c h i n g t r a c e i m p u r i t i e s t h a t may be present i n the s o l v e n t . I t i s important a t t h i s p o i n t to make a d i s t i n c t i o n between the two types of gas e x t r a c t i o n t e c h n i q u e s t h a t have been used i n wine aroma a n a l y s i s namely, e q u i l i b r i u m or s t a t i c headspace and purge and t r a p or dynamic headspace . In purge and t r a p a n a l y s i s , the aqueous sample i s bubbled w i t h an i n e r t gas to remove v o l a t i l e components which are then adsorbed onto a 14 polymer t r a p . N o r m a l l y , the t r a p i s deve loped by heat and the aroma subs tances are then c o o l e d p r i o r to b e i n g i n j e c t e d i n the gas chromatograph. With s t a t i c headspace t e c h n i q u e s , on the o ther hand, a sample of gas immediate ly above the food t h a t has been e q u i l i b r a t e d a t a p a r t i c u l a r temperature and i s c o n t a i n e d i n a c l o s e d v e s s e l i s removed and a n a l y z e d by gas chromatography. Purge and T r a p Headspace A n a l y s i s R e s e a r c h e r s employing s o l v e n t e x t r a c t i o n t e c h n i q u e s had to depend on the s e l e c t i v i t y of o r g a n i c s o l v e n t s to e l i m i n a t e removal of e t h a n o l from a l c o h o l i c beverages . Workers u s i n g headspace c o n c e n t r a t i o n methods have used polymer t r a p s to e l i m i n a t e e t h a n o l from the i s o l a t e d v o l a t i l e s . Porous r e s i n s such as Porapak Q, Tenax GC and Chromosorb 105 have low a f f i n i t y f o r dominant compounds i n wine namely, water and e t h a n o l (He ide , 1985; Simpson 1979b). They p r o v i d e a means for c o n c e n t r a t i n g a l l but the low m o l e c u l a r aroma compononents (Simpson, 1979b). U s i n g Porapak Q, Jenn ings et a l . (1972) deve loped a method to f o l l o w changes i n the v o l a t i l e c o m p o s i t i o n of b e e r . P u r i f i e d n i t r o g e n was purged through the ' sample and through the polymer t r a p for a d s o r p t i o n of the r e l e a s e d aroma components. To r i d the t r a p of e t h a n o l and water , n i t r o g e n was passed through the column wi th one end of the column open to atmosphere . Aroma v o l a t i l e s were c o l l e c t e d i n a g l a s s t r a p c o o l e d w i t h d r y i c e by h e a t i n g and b a c k f l u s h i n g the Porapak Q 15 t r a p . Cordner et a l . (1978) employed the same procedure to examine the e f f e c t of c r o p l e v e l on c h e m i c a l c o m p o s i t i o n and headspace v o l a t i l e s of Z i n f a n d e l grapes and wines but used a Tenax GC t r a p i n s t e a d . Murray (1977) i n t r o d u c e d an new headspace t echn ique u s i n g a Chromosorb 105 column t h a t f i t i n t o a s o p h i s t i c a t e d i n t r o d u c e r system of a l a b o r a t o r y c o n s t r u c t e d gas chromatograph. D i l u t e d brandy was among the samples t h a t were a n a l y z e d . W i l l i a m s and S t r a u s s (1977) adapted t h i s t echn ique f o r examining wines and o ther a l c o h o l i c beverages but used the i n t r o d u c e r w i t h a c o n v e n t i o n a l commercia l gas chromatograph. Two Chromosorb 105 t r a p s were used i n s e r i e s . A 30 L volume of sweep gas was passed through the sample and the t r a p . Excess e t h a n o l and water were removed by p a s s i n g n i t r o g e n through both t r a p s . Chromatograms of the t a b l e wine r e v e a l e d t h a t the f i r s t t r a p adsorbed o n l y a few of the h i g h e r c o n c e n t r a t i o n a l c o h o l s wh i l e the second t r a p adsorbed many v o l a t i l e s i n c l u d i n g the ones adsorbed by the f i r s t but i n lower p r o p o r t i o n . The a u t h o r s c a u t i o n e d t h a t the d e s o r p t i o n c o n d i t i o n s shou ld be adequate to a v o i d s e l e c t i v e r e t e n t i o n of some v o l a t i l e s on the t r a p . Two German r e s e a r c h e r s Rapp and K n i p s e r (1980), i n t r o d u c e d a new t echn ique of headspace a n a l y s i s of wine t h a t was e s s e n t i a l l y a c o m b i n a t i o n of headspace and s o l v e n t e x t r a c t i o n methods. Ins tead of u s i n g a porous polymer t r a p to c o l l e c t v o l a t i l e s s t r i p p e d from the sample by a p u r g i n g gas , the aroma components were c o l l e c t e d i n t o 10 % aqueous e t h a n o l . 16 As t h i s c o l l e c t i o n was being c a r r i e d out 7 the ethanol s o l u t i o n was continuously extracted with Freon 11. Several advantages of t h i s technique were c i t e d i n t h i s o r i g i n a l p u b l i c a t i o n . In comparison to the polymer t r a p methods i n which the sample v o l a t i l e s can be analyzed only once, the headspace sample obtained by t h i s procedure can be examined s e v e r a l times using d i f f e r e n t d e t e c t i o n systems (flame i o n i z a t i o n detector or GC-mass spectrometer). A l s o , no s p e c i a l i n j e c t o r system i s needed as with the method of Murray (1977). No i m p u r i t i e s are contained i n the e x t r a c t and a r t e f a c t formation i s u n l i k e l y to occur since the operation i s c a r r i e d out at 28°C. Owing to the s u p e r i o r i t y and success of t h i s procedure, the method has gained p o p u l a r i t y . Drucret (1984) employed the techique described by Rapp and Knipser (1980) for comparing headspace v o l a t i l e s of carbonic maceration and t r a d i t i o n a l wine. More r e c e n t l y , Craig (1988) used t h i s technique to compare headspace v o l a t i l e s of k i w i f r u i t wine and grape wine. S t a t i c Headspace A n a l y s i s I t was mentioned e a r l i e r that headspace a n a l y s i s represents the composition of aroma compounds as they would be sensed by the nose. According to Dravnieks and O'Donnell (1971), however, t h i s i s only true for s t a t i c headspace methods and not purge and t r a p methods. E x t r a c t i o n procedures i n v o l v i n g bubbling of i n e r t gas i n the sample and subsequent tra p p i n g of v o l a t i l e s onto porous polymer columns, or c o o l i n g traps do not lead to the " n a t u r a l ' composition of the headspace 17 of the food ( B e r t u c c i o l i and Montedoro, 1974) . F i r s t l y , i n accordance w i t h H e n r y ' s law, the headspace c o n c e n t r a t i o n s of v o l a t i l e s are not determined by the vapor p r e s s u r e s of t h e i r r e s p e c t i v e v o l a t i l e s and t h e i r a n a l y t i c a l c o n c e n t r a t i o n s i n the food a l o n e ; they a l s o depend on t h e i r a c t i v i t y c o e f f i c i e n t s . The a c t i v i t y c o e f f i c i e n t i s a f f e c t e d by the content of water , l i p i d s , p r o t e i n s , c a r b o h y d r a t e s , p o l y p h e n o l s and other m a t e r i a l s i n the food and w i l l v a r y d u r i n g sampl ing as i n the case of p u r g i n g methods. H e n r y ' s law i s o n l y v a l i d when the headspace gas above the food i s a l l owed to e q u i l i b r a t e w i th i t (Dravnieks and O ' D o n n e l l , 1971) . In a d d i t i o n , when a cont inuous l o s s of v o l a t i l e s o c c u r s , d i f f u s i o n r a t e s of v o l a t i l e s from the bulk of the food to i t s i n t e r f a c e wi th a i r become f u r t h e r r a t e l i m i t i n g f a c t o r s t h a t can v a r y the r a t i o s of v o l a t i l e s i n the headspace (Dravnieks and O ' D o n n e l l , 1971) . In view of the above, B e r t u c c i o l i and Montedoro (1974) deve loped and o p t i m i z e d a method to a n a l y z e the ^ n a t u r a l ' headspace c o m p o s i t i o n of wine . Samples of wine (100 mL) were p l a c e d i n a l a r g e s y r i n g e (2 L s i z e ) and e q u i l i b r a t e d at 20<>C for 15 min . The headspace volume of the s y r i n g e was passed through a Porapak Q t r a p which was connected to the s y r i n g e . The motor d r i v e n s y r i n g e was se t to r e l e a s e the headspace gas a t a f low r a t e of 100 to 150 mL/min . V o l a t i l e s were desorbed by h e a t . Many new components i n wine not p r e v i o u s l y r e p o r t e d by headspace methods were d e t e c t e d by B e r t u c c i o l l i and V i a n i (1976) when they r e p l a c e d the Porapak Q t r a p wi th Tenax GC. 18 Noble (1978) and Noble et a l . (1979 ) employed the same p r i n c i p l e as B e r t u c c i o l i and Montedoro (1974) but used a d i f f e r e n t apparatus to i n v e s t i g a t e the r e p r o d u c i b i l i t y of the headspace method. The headspace volume over the wine sample was d i s p l a c e d by f l u i d r a ther than a motor d r i v e n s y r i n g e . Noble et a l . (1980) and Noble (1981) employed p r i n c i p a l component a n a l y s i s of wine headspace v o l a t i l e s c o l l e c t e d by displacement to c l a s s i f y wines by v a r i e t y . To study changes caused by Botrytls clnezea i n f i n i s h e d wine, F l a t h et a l . , (1972) used a precolumn c o o l i n g t r a p i n place of a polymer trap to analyze headspace v o l a t i l e s . A 10 mL volume of sample vapor was removed with a gas t i g h t syringe and i n j e c t e d i n t o the precolumn, a s t a i n l e s s s t e e l tubing immersed i n a cold bath. The condensed v o l a t i l e s were t r a n s f e r e d i n t o the gas chromatograph oven by removing the r e f r i g e r a n t and heating the tubing with a hot a i r gun. In a more recent study, Gelsomini (1985) reported a d i r e c t headspace a n a l y s i s with c a p i l l a r y columns using an automatic headspace chromatographic system. Wine samples saturated with anhydrous sodium s u l f a t e were conditioned i n the thermostatic bath for 1 hr.. at 50°C. Diff e r e n c e s i n red and white wine were apparent from the headspace chromatograms that contained a handful of peaks. Use of t h i s system was recommended for q u a l i t y c o n t r o l of wines i n the i n d u s t r y . A comment about c a p i l l a r y columns i s warranted here. In the d i s c u s s i o n of e x t r a c t i o n methods for v o l a t i l e a n a l y s i s , no mention was made of c a p i l l a r y columns. Enrichment of aroma 19 compounds, whether by s o l v e n t e x t r a c t i o n or headspace methods, r e s u l t s i n the c o n c e n t r a t i o n of v o l a t i l e s t h a t are composed of hundreds of i n d i v i d u a l components which c o n t a i n many d i f f e r e n t c l a s s e s of c o n s t i t u e n t s , and a wide range of b o i l i n g p o i n t s . S e p a r a t i o n of such complex mix tures t h a t v a r y i n c o n c e n t r a t i o n by s e v e r a l magnitude , demands ex treme ly h i g h e f f i c i e n c y of the column which c o u l d have o n l y been p o s s i b l e u s i n g c a p i l l a r y column gas chromatography (Rapp, 1981). D. SIMPLEX OPTIMIZATION O p t i m i z a t i o n has been d e s c r i b e d as f i n d i n g the best p o s s i b l e method of c a r r y i n g out some o p e r a t i o n (Bayne and R u b i n , 1986) . Even today a popu lar o p t i m i z a t i o n procedure i s the c l a s s i c a l o n e - f a c t o r - a t - a - t i m e method (Bayne and R u b i n , 1986) . T h i s s e q u e n t i a l approach r e q u i r e s t h a t a l l f a c t o r s except one be h e l d c o n s t a n t whi l e the f a c t o r be ing t e s t e d i s e v a l u a t e d a t v a r y i n g l e v e l s . U s i n g a two f a c t o r example, Massert et a l . , (1988) d e s c r i b e d i n a v e r y s imple and unders tandab le manner the s e q u e n t i a l s i n g l e f a c t o r - a t - a - t i m e s t r a t e g y . I f there i s no i n t e r a c t i o n between the two f a c t o r s , then the optimum can be found; the r i d g e on the response s u r f a c e of t h i s example would l i e p a r a l l e l to the f a c t o r axes ( F i g . 1 ) . The no i n t e r a c t i o n case i s , however, an e x c e p t i o n ; i n g e n e r a l , f a c t o r s do not operate i n d e p e n d e n t l y on the r e s p o n s e . When i n t e r a c t i o n s o c c u r , a p l o t of the response s u r f a c e r e v e a l s t h a t the r i d g e 20 (optimum) does not l i e p a r a l l e l to the axes; i t i s instead oblique with respect to the other f a c t o r axes ( F i g . 2 ) , thus making the task of f i n d i n g the optimum d i f f i c u l t . A one-fac t o r - a t - a - t i m e approach can improve the response but not optimize i t . The optimum may be found i f the f a c t o r s are changed together i n the d i r e c t i o n of the a x i s of the r i d g e . EVOP Noting the inadequacy of the one-factor-at-a-time method, Box (1957) suggested an a l t e r n a t i v e o p t i m i z a t i o n technique that was capable of f i n d i n g the optimum even when i n t e r a c t i o n s between f a c t o r s e x i s t e d . E v o l u t i o n a r y operation (EVOP) s t r a t e g y was based on f a c t o r i a l designs i n which each of the v a r i a b l e s would be a l t e r e d s l i g h t l y at the same time (Saguey, 1986). EVOP i s valuable f o r i n d u s t r i a l process o p t i m i z a t i o n s since f a c t o r s are not v a r i e d e x t e n s i v e l y but i t has l i m i t e d a p p l i c a t i o n f or research purposes where extensive s h i f t s i n experimental c o n d i t i o n s may be req u i r e d . Another drawback of t h i s method i s the large number of experiments i n each f a c t o r i a l design needed to complete the o p t i m i z a t i o n (Nakai and Arteaga, 1989; Berridge, 1985). Simplex EVOP In 1962 Spendley et a l . introduced the s e q u e n t i a l simplex method c a l l e d simplex EVOP which overcame major l i m i t a t i o n s of the f a c t o r i a l EVOP method. This f i x e d - s i z e s e q u e n t i a l simplex c o n s i s t e d simply of r e f l e c t i o n r u l e s (Massert et a l . , 1988). 21 Fig. 1. S i n g l e - f a c t o r - a t - a - t i m e s t r a t e g y on a well behaved r e s p o n s e s u r f a c e (Massert et al., 1988). 22 F i g ' 2 ' S i n g l e - f a c t o r - a t - a - t i m e s t r a t e g y on su r f ace exhibitaing a diagonal r idge (Massert et al., ^tir** 23 Nedler and Mead (1965) improved the o r i g i n a l simplex by giving the simplex the a b i l i t y to accelerate in favorable directions and to decelerate in unfavorable directions (Morgan and Deming, 1974 ) . Simplex Method A simplex is a geometric figure that has n+1 vertices where n i s the number of factors. For two factor optimization, the simplex w i l l be a tria n g l e and for a four factor case, the simplex w i l l be a tetrahedron. Once the factor ranges are determined, the simplex is established by carrying out n+1 observations. Each move of the simplex in the search for the optimum requires one observation of the response (Jurs, 1986). Because the search is without perspective of the response surface, i t has been ca l l e d a "search in the dark" (Aishima and Nakai, 1986). The movement of the simplex is determined by a set of rules for r e f l e c t i o n (move away from the worst response), contraction ( i f a step is made in a wrong direction) and expansion ( i f a move is made in a desirable d i r e c t i o n ) . These rules have been i l l u s t r a t e d with simple examples by numerous authors (Morgan and Deming, 1974; Berridge, 1985; Massert et a l . , 1988 and Jurs, 1986). Improvements in the Method Modifications of the o r i g i n a l simplex have been made by 24 several researchers (Routh et a l . , 1977; Ryan et a l . , 1980; Nakai, 1982; Nakai et a l . , 1984 and Nakai and Kaneko, 1985). Improved optimization schemes have been developed for two reasons: (1) because the simplex optimization i s an i t e r a t i v e search, the speed of the search is high at the beginning but slow near the end (2) the p o s s i b i l i t y of the simplex optimization finding the l o c a l optimum rather than the global one. Nakai et a l . , (1984) introduced mapping and simulataneous s h i f t in order to improve the e f f i c i e n c y of the optimization with a new mapping super-simplex optimization program. Response values are plotted against each factor and the data points are grouped. From the maps of a l l the factors, target values are predicted. If the d i r e c t i o n of search is evident, simultaneous s h i f t i s executed. A s h i f t in a l l factor levels from the best response ve r t i c e is made toward the target value. This procedure s i g n i f i c a n t l y expedited the optimization at the later stages (Nakai and Kaneko, 1985). Nakai (1982) advocated the use of simplex optimization for application to food product and process development. Optimization methods including f r a c t i o n a l f a c t o r i a l , one-at-a-time search, pattern search method, response surface and simplex optimization methods of Morgan and Deming (1974), Routh et a l . , (1977), Ryan et a l . , (1980) and Nakai (1982 ) were compared for e f f i c i e n c y using two mathematical models. Many food applications of simplex optimization and other optimization techniques can be found in a current publication 25 by Nakai and A r t e a g a (1989). M A T E R I A L S AND METHODS A. SAMPLES  Orange J u i c e F r e s h orange j u i c e made by an i n - s t o r e j u i c i n g machine was purchased from a l o c a l supermarket and s t o r e d a t 8°C. Samples were a n a l y z e d w i t h i n t h r e e days of the purchase. Grape J u i c e C o n c e n t r a t e s Seven d i f f e r e n t grape j u i c e c o n c e n t r a t e s r e c e i v e d from Sun-Rype P r o d u c t i o n s L t d . (Kelowna, B.C.) and s t o r e d f r o z e n a t Oc-c. Apple J u i c e s E i g h t v a r i e t a l a p p l e j u i c e s from a p p l e s grown and p r o c e s s e d a t the Summerland Research S t a t i o n were o b t a i n e d from A g r i c u l t u r e Canada and s t o r e d a t 0°C. Wines A l l wine samples were o b t a i n e d from B r i g h t s Wines ( O l i v e r , B.C.). Wine samples were o b t a i n e d d i r e c t l y from the w i n e r y ' s l a r g e s t a i n l e s s s t e e l h o l d i n g . v a t s by d i s c a r d i n g the f i r s t 400 t o 600 mL of wine and f i l l i n g s t e r i l e dark green g l a s s b o t t l e s u n t i l t h e r e was l i t t l e headspace. B o t t l e s were then screwcapped and s e a l e d w i t h p a r a f i l m b e f o r e s t o r i n g a t 5 0 c . 27 Two commercial white table wines of Brights Wines were optimized for blending using headspace gas chromatography and the simplex optimization program. T r i a l 1 involved blending eigth possible wines to formulate the housebrand wine c a l l e d Leibesheim. Four of these wines were v a r i e t a l (Chenin blanc, vats 7, 14, 20 and 25) and the rest were white stock wines from vats 61, 70, 69 and 12. For the second blending t r i a l , to formulate Cuvee White, five v a r i e t a l wines (Verdelet, vats 74, 47, 51, 71,and 58), two white stock wines (vats 70 and 12), and one premium white stock wine (vat 22) were available for the blending problem. B. SAMPLE PREPARATION Samples of 15 mL size were pipetted into 20 mL glass v i a l s , capped with butyl rubber t e f l o n faced septa (Hewlett Packard) and crimped with aluminum caps with a pressure release safety feature. The v i a l containing the sample was then placed in the heated carousel for the specified time. Wine samples analyzed at 6°C were refrigerated for more than 4 hrs. prior to sampling. A l l glassware used in t h i s work was soaked in detergent (Extran) overnight or longer, rinsed five times with tap water and d i s t i l l e d water, and oven dried. 28 C. INSTRUMENTAL ANALYSIS Wine and Apple Ju i c e i . Gas Chromatography Chromatography was performed on a Varian V i s t a 6000 gas chromatograph equipped with a flame i o n i z a t i o n detector and a Series 651 Data System. The detector s e t t i n g s were range 12 a t t e n u a t i o n 1, and temperature 325 °C. The i n j e c t i o n port was operated at 200°C. Chromatograms were p l o t t e d on a Hewlett Packard T h i n k j e t P l o t t e r at a chart speed of 1 cm/min. A l l gases connected to the gas chromatograph were pre-p u r i f i e d grade (Linde, Vancouver) and operated at the f o l l o w i n g c o n d i t i o n s : helium ( c a r r i e r ) 3 mL/min., hydrogen 36 mL/min., nitrogen (make-up) 18 mL/min., and a i r 300 mL/min. Linear gas v e l o c i t i e s were measured with a bubble flow meter at 40°c oven temperature. Moisture traps (Chemical Research Service Inc.) were i n s t a l l e d i n a l l gas l i n e s between the gas c y l i n d e r and the chromatograph. An a d d i t i o n a l hydrocarbon t r a p (Chromatographic S p e c i a l t i e s ) was i n s t a l l e d for the c a r r i e r gas. A one ramp oven temperature programming sequence was used: an i n i t i a l temperature of -20°C was held for 2 min. and then increased to 300°C at a rate of 10<>C/min. and held at 300<>c f o r 10 min. The f i n a l high temperature programming was a cl e a n i n g step between runs that e l i m i n a t e d any high b o i l i n g components remaining i n the column. 29 i i . Headspace Sampling Aroma c o n s t i t u e n t s above the samples contained i n glass v i a l s were withdrawn and i n j e c t e d i n t o the gas chromatograph by a heated t r a n s f e r l i n e that connected the headspace sampler (Dani HSS 3950 Sampling Unit and Dani HSS 3950 Programmer) to the gas chromatograph. P r e s s u r i z a t i o n and vent times were set at 3 sec. and the i n j e c t i o n time for 44 sec. The carousel bath temperature was 37<>c while the manifold temperature was set to maximum at 150°C. The pressures for the gases were: c a r r i e r 1.4 bar, a i r 3.4 bar, and a u x i a l l a r y a i r 0.6 bar. Orange and Grape Juice Concentrates i . Gas Chromatography A Hewlett Packard 5890 gas chromatograph equipped with a flame i o n i z a t i o n detector and a HP 3396A i n t e g r a t o r was used. The detector temperature, range and a t t e n u a t i o n were set at 250 °C, 0 and 0, r e s p e c t i v e l y . The i n j e c t i o n port was operated at 225°C. For orange j u i c e samples, a three step column temperature programming sequence was as f o l l o w s : i n i t i a l column temperature (-20°C) was held for 2 min. and then advanced at 20°C/min. to 4 0 O C f o r 3 min. holding time. A second temperature program was i n i t i a l i z e d at 10°C/min. to 90°C for 4 min. holding time and followed by a t h i r d programming sequence at 4°C/min. to a f i n a l temperature of 170°C for a f i n a l h olding time of 4 min. The same 30 temperature programming procedure was used for grape j u i c e concentrates except that the i n i t i a l column was lowered to -40°C. Each gas l i n e was equipped with oxygen traps while the c a r r i e r gas l i n e was f i t t e d with' an a d d i t i o n a l moisture t r a p . Gas flows were: helium ( c a r r i e r ) 2 mL/min., helium (make-up) 30 mL/min., hydrogen 30 mL/min. and a i r 400 mL/min. i i . Headspace Sampling Two 1 mL headspace samples from two d i f f e r e n t v i a l s c o n t a i n i n g the same sample, were i n j e c t e d i n t o the gas chromatograph by a m u l t i p l e headspace i n j e c t i o n technique as described by Wylie (1986). This was achieved by programming the Hewlett Packard 19395A headspace sampler. D. INTERNAL STANDARD Several a l c o h o l s were examined as p o s s i b l e standards i n c l u d i n g 3-pentanol, 2-methyl-l-propanol, 2-methyl-2-butanol and 2-methyl-2-propanol. 3-pentanol has been i d e n t i f i e d i n grapes (Stevens et a l . , 1969 ) but not i n wines. Cordner et a l . (1978) used 3-pentanol as an i n t e r n a l standard f o r purge and t r a p a n a l y s i s of wine. Since t h i s component was not a v a i l a b l e i n high p u r i t y and i t was not separated w e l l from other peaks i n a wine chromatogram, i t could not be used as an i n t e r n a l standard. Both 2-methyl-l-propanol and 2-methyl-l-31 butanol were a v a i l a b l e i n chromatography grade but were r e j e c t e d as p o s s i b l e standards because they have been i d e n t i f i e d by purge and t r a p and s t a t i c headspace a n a l y s i s of wines ( B e r t u c c i o l i and V i a n i , 1976; Noble, 1981; Cordner et a l . , 1978; C r a i g 1988). The best choice was 2-methyl-2-propanol as i t was a v a i l a b l e i n high p u r i t y (Polyscience Corp.), i t was resolved from other peaks i n the chromatogram and no headspace method ( s t a t i c or purge and trap) had reported i t s presence i n wines. Stevens et a l . (1969) i d e n t i f i e d t h i s substance i n Grenache rose wine i n trace amounts by e x t r a c t i n g the wine with Freon 11. One uL of 2-methyl-2-propanol was added to 15 mL of wine i n a 50 mL t e s t tube and throughly mixed by v o r t e x i n g . The sample was then immediately t r a n s f e r r e d to a glass v i a l which was then capped and crimped. E. OPTIMIZATION The main o b j e c t i v e of t h i s work was to determine what i s the best blending r a t i o of the v a r i e t a l wines and white stock wines to match the tar g e t wine. In blending o p t i m i z a t i o n the premise i s that only the v o l a t i l e components c o n t r i b u t e to the aroma of the product, t h e r e f o r e , gas chromatographic data together with simplex optimzation could be used for f i n d i n g the best blending r a t i o s (Nakai and Arteaga, 1989). The f i r s t step was to analyze the wine samples to be used for blending 32 by headspace gas chromatography to determine how s i m i l a r the f l a v o r p r o f i l e s of these wines were to the t a r g e t wine. This was accomplished by c a l c u l a t i n g a pat t e r n s i m i l a r i t y c o e f f i e n t shown below: (Equation 1) S(AB) = 5 X . X ' / i / 2 X 2 X . ' 2 0<S(AB)<1.0 I i i i This equation represents the s i m i l a r i t y of two chromatograms, A = ( X / X . . . . X„ ) and B = ( X ' , X ' . . . X ' ) where X. and X.' are 1 2 ' " 1 2 " 1 1 areas of peaks i i n the chromatograms of samples A and B, r e s p e c t i v e l y . As i n r e g r e s s i o n a n a l y s i s , the patt e r n s i m i l a r i t y c o e f f i c i e n t varys from 0 to 1. I f two chromatograms have i d e n t i c a l p r o f i l e s , then the s i m i l a r i t y c o e f f i c i e n t would be c a l c u l a t e d as 1 while two completely d i s s i m l a r p r o f i l e s would have a c o e f f i c i e n t of 0 (Nakai and Arteaga, 1989). Peak s e l e c t i o n of the p a t t e r n s i m i l a r i t y c o e f f i c i e n t was based on two r u l e s . The f i r s t c r i t e r i o n was that the peak i n question had to be present i n the tar g e t wine as w e l l as i n at l e a s t one of the blending wines. The second important c o n d i t i o n was that the standard d e v i a t i o n of a peak area had to be high i n order for i t to be s e l e c t e d . I f the peak areas of the t a r g e t wine and the blending wines are very c l o s e , then t h i s data i s not of value i n c a l c u l a t i o n of the s i m i l a r l y c o e f f i e n t because the idea i s to detect d i f f e r e n c e s i n the pa t t e r n s . The pattern s i m i l a r i t y c o e f f i c i e n t formula i s a 33 subprogram subroutine of the simplex program. When the data of the reference and blending wines is entered into the simplex program, i t searches for the optimal blending ratios that give the highest s i m i l a r i t y c o e f f i c i e n t between the gas chromatographic pattern of the blend and the target wine. This optimization procedure is cal l e d an automated sequential simplex technique (Nakai and Arteaga, 1989). F. TITRATABLE ACIDITY AND TOTAL SOLUBLE SOLIDS To decide i f corrections of a c i d i t y and sweetness were necessary for the blends of t r i a l one and t r i a l 2, t o t a l soluble s o l i d s and t i t r a t a b l e a c i d i t y were determined at the winery using their procedures. T i t r a t a b l e A c i d i t y A 5 mL sample of wine was pipetted into a 250 mL beaker and di l u t e d with 100 mL of d i s t i l l e d water. The diluted sample was s t i r r e d while being t i t r a t e d with 0.1 N NaOH to an end point of pH 8.2. Tit r a t a b l e a c i d i t y was calculated as mg t a r t a r i c acids per 100 mL wine. Total Soluble Solids Samples of 200 mL were adjusted for temperature to 20<>c. Total soluble s o l i d s were measured as °Balling with a hydrometer. 34 G. SENSORY EVALUATION To determine i f t h e r e were d e t e c t a b l e d i f f e r e n c e s between the b l ends and the t a r g e t wine , two s e t s of t r i a n g l e t e s t s were c a r r i e d o u t , one f o r the s u b j e c t i v e b lend and the other for the o b j e c t i v e b l e n d . For t r i a l 1, the pane l c o n s i s t e d of 12 t e c h n i c a l s t a f f form the A g r i c u l t u r e Research S t a t i o n at Summerland. Three of the p a n e l i s t s were e x p e r i e n c e d judges as they were p r e v i o u s l y t r a i n e d i n e v a l u a t i o n of wines and t e s t e d wines r e g u l a r l y . F i f t e e n members e v a l u a t e d the wines b lends of the second t r i a l . Twelve were from the Research S t a t i o n and the o ther t h r e e were wine e x p e r t s form the w i n e r y . In t o t a l , seven judges were e x p e r i e n c e d i n t a s t i n g wines for t h i s t r i a l . Samples of 40 mL s i z e were presented a t or near room temperature i n wine g l a s s e s under red l i g h t to e l i m i n a t e any p o s s i b l e b i a s due to c o l o r . 35 RESULTS AND DISCUSSION A. Method Development Headspace gas chromatography (HSGC) was s e l e c t e d as the method of v o l a t i l e a n a l y s i s m a i n l y because the procedure of Ai sh ima et a l . (1987) was not s u i t e d for q u a l i t y c o n t r o l p u r p o s e s . In t h i s method v o l a t i l e s from s t r a w b e r r y essences and c o n c e n t r a t e s were i s o l a t e d u s i n g a m o d i f i e d L i k e n s N i c k e r s o n apparatus to c a r r y out s imul taneous d i s t i l l a t i o n and e x t r a c t i o n wi th methylene c h l o r i d e for two h o u r s . The r e s u l t i n g e x t r a c t s were c o n c e n t r a t e d wi th a Kuderna Danish c o n c e n t r a t o r and f u r t h e r e n r i c h e d under a s tream of n i t r o g e n . S e p a r a t i o n of the t o t a l v o l a t i l e s by GC r e q u i r e d 90 min . to comple te . I t i s t h e r e f o r e apparent t h a t t h i s sample p r e p a r a t i o n procedure i s not o n l y c o m p l i c a t e d but f a r too time consuming, making i t i m p r a c t i c a l f o r r o u t i n e a p p l i c a t i o n i n the i n d u s t r y . In c o n t r a s t , s t a t i c headspace sampl ing permi t s d i r e c t a n a l y s i s of the sample vapor wi thout p r i o r i s o l a t i o n and c o n c e n t r a t i o n treatments (Jennings and Rapp, 1983). The f i r s t goa l of the c u r r e n t s tudy was to deve lop a r a p i d and s imple method of aroma e x t r a c t i o n u s i n g an automated headspace sampl ing sys tem. A number of d i f f e r e n t samples i n c l u d i n g orange j u i c e , - g r a p e j u i c e c o n c e n t r a t e , apple j u i c e and wine were a n a l y z e d because a r e l i a b l e source of samples from the i n d u s t r y was not a v a i l a b l e u n t i l c o n t a c t wi th the winery ( B r i g h t s Wines, O l i v e r , B . C . ) was made. 36 Preliminary Work An examination of applications of equilibrium headspace methods for food in the l i t e r a t u r e revealed that the method lacked s e n s i t i v i t y (Heath and Reineccius, 1986; McNally and Grob 1985; Issenberg and Hornstein, 1970; Reineccius and Anandaraman, 1984; Shibamoto; 1984; Hachenberg and Schmidt, 1977 and Ioffe and Vitenberg, 1984). Headspace analysis of 15 mL of fresh orange juice confirmed the l i m i t a t i o n of th i s method. The chromatogram of the orange juice in F i g . 3 shows that only about 10 major peaks eluted. In l i g h t of t h i s , preliminary work was carried out to improve the lower detection l i m i t of the concentration headspace components. Two general texts on s t a t i c headspace analysis (Hachenberg and Schmidt, 1977 and Ioffe and Vitenberg, 1984) have addressed thi s problem. Hachenberg and Schmidt (1977) discuss two methods of increasing s e n s i t i v i t y : r a i s i n g the incubation temperature and addition of el e c t r o l y t e s for aqueous samples. Increasing the thermostating temperature improved r e s u l t s . Orange juice samples were analyzed at 40, 50, 60 and 70°C and wine samples were analyzed at 6, 37 and 55°C. As expected, new peaks appeared and the peak areas of components that were detected at lower temperatures increased. The ef f e c t of r a i s i n g temperature while keeping a l l other parameters constant on t o t a l peak area is shown in F i g . 4 and 5. The overall e f f e c t is an increase in t o t a l peak area with temperature. The res u l t of r a i s i n g temperature on enhancing s e n s i t i v i t y (E ) is related by the following equation: H S 37 Ul Jj •A 11 5 10 15 20 25 Rentention Time (min.) Fig. 3. Chromatogram of headspace volatiles . from fresh orange juice analyzed at 70C. 40 48 52 I 56 I 60 —r— 64 l 68 Fig. 4. The e f f e c t of area f o r orange juice. TEMPERATURE (*C) temperature of equilibration on peak 39 E H S = c i Poi T £ (Equation 2) where c i s the concentration of component i , P 0 . i s the saturated vapor pressure of the pure substance and is i t s a c t i v i t y c o e f f i c i e n t (Hachenberg and Schmidt, 1977). The saturated vapor pressure of the components in the headspace of the food increases with temperature. F i g . 4 and 5 show interesting trends not related to s e n s i t i v i t y . The curve for orange juice i s sigmoidal; between 40 to 50oc and 60 to 70°C there i s a small change in t o t a l peak area but there is a sharp increase between 50 and 60°C. Wine, in contrast, showed a very d i f f e r e n t trend; the relat i o n s h i p between the detector response and e q u i l i b r a t i n g temperature is almost l i n e a r . Some comments about differences in the two figures can be made even though the samples were not analyzed at i d e n t i c a l temperatures. It i s possible that in orange juice there may be an abundance of v o l a t i l e flavor components with b o i l i n g points in the 50 to 60°C range. Wine, however, may contain aroma substances that are more or less equally spread out in number over the b o i l i n g point range tested. Enhancing s e n s i t i v i t y by increasing temperature has limited application because thermally induced chemical changes such as oxidation, hydrolysis and non-enzymatic browning reactions can occur. Analysis of flavor components at elevated temperatures may be appropriate when the interest 41 i s to i s o l a t e cooked v o l a t i l e s . High temperatures of 6 0 ° C or more can induce non-enzymat ic M a i l l a r d browning r e a c t i o n s (Heath and R e i n e c c i u s , 1986) . S e n s i t i v i t y of headspace a n a l y s i s may a l s o be improved by the a d d i t i o n of s a l t s . S a l t s cause the va lue of the a c t i v i t y c o e f f i c i e n t to i n c r e a s e i n e q u a t i o n 2. The s o l u b i l i t y of m a i n l y p o l a r substances i s lowered so t h a t they are f o r c e d i n t o the headspace ( I o f f e and V i t e n b e r g , 1984) . Ammonium s u l f a t e and sodium c h l o r i d e were added to orange j u i c e at s a t u r a t e d l e v e l s but the chromatograms showed l i t t l e improvement. C o l d T r a p p i n g The use of an enr ichment t echn ique c a l l e d c o l d t r a p p i n g or c r y o f o c u s s i n g appeared to the most s u c c e s s f u l procedure for u p g r a d i n g s e n s i t i v i t y of headspace s a m p l i n g . F i g . 3 and 6 i l l u s t r a t e the c o n c e n t r a t i o n e f f e c t . The f i r s t chromatogram shows headspace a n a l y s i s of f r e s h orange j u i c e wi thout any c o o l i n g wh i l e the second chromatogram r e s u l t e d from c r y o f o c u s i n g a t - 2 0 ° C . The number of major peaks doubled from 10 to 20 d e m o n s t r a t i n g the s t r i k i n g i n c r e a s e i n s e n s i t i v i t y due to e n r i c h m e n t . In a d d i t i o n , i t shou ld be mentioned t h a t the c o m b i n a t i o n of c r y o f o c u s s i n g and headspace a n a l y s i s s t i l l r e p r e s e n t s a t r u e e q u i l i b r i u m a n a l y s i s (Kolb et a l . , 1986) , u n l i k e enrichment by a d d i t i o n of s a l t s i n which not a l l of the v o l a t i l e s are a f f e c t e d e q u a l l y ; the degree of enr ichment i s d i f f e r e n t for e s t e r s , a ldehydes and a l c o h o l s 42 00 i j i ~j r~ 10 15 20 25 30 Retention Time (min.) Fig. 6. Chromatogram of headspace volat i les f rom f r e s h orange juice analyzed at 70-C using c r y o f ocussing. ( P o l l and F l i n k , 1984 ) . C r y o g e n i c f o c u s s i n g i s a r e l a t i v e l y r e c e n t development f o r s t a t i c headspace a n a l y s i s i n which a band f o c u s s i n g e f f e c t of the aroma substances occurs as a r e s u l t of c o o l i n g the f i r s t p a r t of a c a p i l l a r y column or by c o o l i n g the e n t i r e column ( K o l b , 1985) . A c o o l a n t , e i t h e r l i q u i d carbon d i o x i d e or l i q u i d n i t r o g e n i s employed to ach i eve subambient t e m p e r a t u r e s . The headspace sample i s i n t r o d u c e d i n t o a c o l d column c a u s i n g the sample components to condense i n a narrow band at the head of the co lumn. V o l a t i l e s do not a c t u a l l y f r e e z e by c r y o g e n i c t r a p p i n g ; t h e i r m i g r a t i o n r a t e s through the column s low down (Kolb et a l . , 1986) . For orange j u i c e samples , c o l d t r a p p i n g was used i n c o m b i n a t i o n wi th d e l i v e r i n g a l a r g e r headspace volume which i s another t e c h n i q u e to improve the d e t e c t i o n l i m i t of the headspace method. I n j e c t i n g l a r g e r gaseous samples i n t o the GC i s not p o s s i b l e to be c a r r i e d i n the absence of c o o l i n g because r e s o l u t i o n d e t e r i o r a t e s (Jennings and Rapp, 1983). Peaks t h a t shou ld be sharp and w e l l s e p a r a t e d appear as wide bands and are p o o r l y r e s o l v e d . I n c r e a s i n g sample volume wi thout c r y o f o c u s s i n g does l i t t l e or n o t h i n g to b e t t e r s e n s i t i v i t y . The method of Wyl ie (1986) was a p p l i e d for orange j u i c e ( F i g . 6) and grape j u i c e c o n c e n t r a t e s ( F i g . 7 and 8 ) . M u l t i p l e Headspace I n j e c t i o n , MHI, not to be confused w i t h m u l t i p l e headspace e x t r a c t i o n which i s used for q u a n t i t a t i o n when the m a t r i x e f f e c t i s of c o n c e r n , a l l o w s the headspace sampler to make s e v e r a l r a p i d i n j e c t i o n s from each 44 re 10 15 20 Retention Time (min.) 25 30 i 35 Fig. 7. Headspace volatiles f r o m grape juice c o n c e n t r a t e at 55*C (sample 1). tnn— VJ to 10 15 20 Retention Time (min.) Fig. 8. Headspace volatiles from 55-C (sample 2). 25 30 35 grape juice concentrate at of one or more v i a l s or to make one i n j e c t i o n from each of two or more v i a l s by programming the HP 19395A. In the case of orange j u i c e and grape j u i c e c o n c e n t r a t e s , two f a s t i n j e c t i o n s were made from two v i a l s c o n t a i n i n g the same sample whi l e the column temperature was - 2 0 ° C . Apple j u i c e and wine samples were a n a l y z e d w i t h o n l y one headspace i n j e c t i o n u s i n g a d i f f e r e n t but s i m i l a r automated sampl ing sys tem. In t h i s system the headspace samples were i n t r o d u c e d d i r e c t l y i n t o the c a p i l l a r y column. The needle of the heated t r a n s f e r l i n e t h a t connects the headspace sampler to the GC f i t snugg ly around the c a p i l l a r y column l i k e a s l e e v e . Sample i n t r o d u c t i o n for the headspace system used for orange j u i c e and grape j u i c e c o n c e n t r a t e s was of d i f f e r e n t d e s i g n as the needle of the heated t r a n s f e r l i n e was connected to the i n j e c t i o n p o r t , not to the column d i r e c t l y . The t r a n s f e r of the headspace sample from the i n j e c t i o n p o r t to the column r e s u l t s i n d i l u t i o n of the sample s i n c e the c a r r i e r gas i s e n t e r i n g the i n j e c t i o n p o r t as w e l l (Takeoka and J e n n i n g s , 1984) . Because no d i l u t i o n was o c c u r r i n g wi th the new sampl ing sys tem, one i n j e c t i o n of the headspace sample was s u f f i c i e n t to i n c r e a s e the needed s e n s i t i v i t y ; MHI was not r e q u i r e d . Chromatograms of app le j u i c e v a r i e t i e s and a wine sample are i l l u s t r a t e d i n F i g . 9 and 10 and 11. A l t h o u g h i t i s g e n e r a l l y accepted t h a t purge and t r a p headspace a n a l y s i s i s f a r more s e n s i t i v e i n comparison to s t a t i c headspace methods, the l a t t e r can be made e q u a l l y s e n s i t i v e i f not b e t t e r p r o v i d e d t h a t the headspace i n j e c t i o n s 47 L A r » 9 u u o*. ^  OS M © 0> if/ r> —r~ 10 —r~ 15 Rentention Time (min.) Analysis of Winesap variety apple juice — i 20 at 5 5 C — r 25 are p r o p e r l y executed . A c c o r d i n g to Takeoka and Jennings (1984), i f a headspace sample i s i n j e c t e d d i r e c t l y i n t o the i n t e r i o r of a s m a l l bore c a p i l l a r y column and i t i s c r y o f o c u s s e d , then f o r some samples , r e s u l t s of s t a t i c headspace can r i v a l and be s u p e r i o r to those o b t a i n e d by c o n v e n t i o n a l dynamic headspace a n a l y s i s . C a p i l l a r y Column A l l samples were a n a l y z e d wi th one column f o r the e n t i r e d u r a t i o n of t h i s s t u d y . A n o n - p o l a r fused s i l i c a c a p i l l a r y column c r o s s l i n k e d wi th 5% phenylmethy l s i l i c o n e l i q u i d s t a t i o n a r y phase was used . Many of the headspace s t u d i e s of wines , however, have been conducted wi th columns coated w i t h p o l y e t h y l e n e g l y c o l (PEG) , ' a l s o c a l l e d Carbowax 20M, l i q u i d s t a t i o n a r y phase ( B e r t u c c i o l i and Montedoro, 1974; G e l s o m i n i , 1985; W i l l i a m s and S t r a u s s , 1977; Simpson, 1979 and M u r r a y , 1977) or r e l a t e d s t a t i o n a r y phases such as Carbowax 400 ( B e r t u c c i o l i and V i a n i , 1976) . Carbowax 20M columns are popu lar among r e s e a r c h e r s working wi th headspace s t u d i e s because the r e t e n t i o n of low m o l e c u l a r weight p o l a r components i s s i g n i f i c a n t l y h igher (Takeoka and J e n n i n g s , 1984) . On fused s i l i c a , however, t h i s s t a t i o n a r y phase has s e v e r a l l i m i t a t i o n s (Takeoka and J e n n i n g s , 1984; J e n n i n g s , 1987) . PEG phases are s u s c e p t i b l e to damage by water . T h i s i s of concern for headspace methods because headspace samples u s u a l l y c o n t a i n a p p r e c i a b l e amounts of water u n l e s s some p r o v i s i o n has been made to remove i t . 51 Another d i s a d v a n t a g e of Carbowax columns i s t h a t they posses r e l a t i v e l y low h i g h temperature and h i g h low temperature l i m i t s . At temperatures of 50 to 6 0 ° C the s t a t i o n a r y phase s o l i d i f i e s r e s u l t i n g i n a d r a s t i c l o s s of r e s o l v i n g power of the co lumn. T h i s drawback would make i t i m p o s s i b l e to use c o l d t r a p p i n g procedures w i t h PEG columns. Other drawbacks of t h i s columnn are i t s h i g h a f f i n i t y for oxygen which causes the columns to d e t e r i o r a t e f a s t e r than other types of co lumns. Even t r a c e s of oxygen can have adverse e f f e c t s a t h i g h t e m p e r a t u r e s . A l though new PEG columns (DB-WAX) which are r e s i s t a n t to water and remain l i q u i d a t subambient temperatures of down to 0 ° C are a v a i l a b l e c o m m e r c i a l l y (Takeoka and J e n n i n g s , 1984) , they may not be e n t i r e l y a p p l i c a b l e for c r y o g e n i c methods which o f t e n r e q u i r e temperatures of -100 to -20<>C. C r a i g (1988 ) i n v e s t i g a t e d headspace v o l a t i l e s of wine wi th a DB-WAX column but no c o l d t r a p p i n g of v o l a t i l e s i n the column was used . C a r e f u l c o n s i d e r a t i o n was g i v e n to the c h o i c e of the column i n t h i s s t u d y . A r e l a t i v e l y n o n - p o l a r column was s e l e c t e d e s s e n t i a l l y because i t was n e c e s s a r y to per form c r y o f o c u s s i n g to i n c r e a s e the lower d e t e c t i o n l i m i t of the headspace method. Other advantages o f f e r e d by t h i s column were i t s t o l e r a n c e for water and i t s s e l e c t i v i t y . Water was p r e s e n t a t h igh amounts i n the headspace of the samples a n a l y z e d u n l i k e most of the headspace s t u d i e s on wine , wi th the e x c e p t i o n of G e l s o m i n i (1985), i n which some type of t r a p wi th low a f f i n i t y for water was used . In a d d i t i o n , t h i s 52 column o f f e r s i n c r e a s e d s e l e c t i v i t y towards d i f f e r e n t c l a s s e s of s o l u t e s compared to o ther n o n - p o l a r columns due to the presence of the phenyl group i n the s t a t i o n a r y phase ( J e n n i n g s , 1987) . A 1 um t h i c k f i l m was chosen to i n c r e a s e the sample c a p a c i t y of the co lumn. B . WINE HEADSPACE ANALYSIS V o l a t i l e a n a l y s i s of a l c o h o l i c beverages i s more c o m p l i c a t e d than i t i s f or o ther food p r o d u c t s because of the presence of h i g h c o n c e n t r a t i o n s of e t h a n o l . Researchers a n a l y z i n g such p r o d u c t s have, t h e r e f o r e , deve loped methods which are s e l e c t i v e a g a i n s t e t h a n o l . E t h a n o l - f r e e c o n c e n t r a t e s of v o l a t i l e s have been o b t a i n e d by employing s e l e c t i v e s o l v e n t s such, as pentane and Freon 11 or i n the case of headspace methods, by the use of polymers such as Tenax GC, Chromosorb 105 or Porapak Q. In t h i s s t u d y , however, no e f f o r t was made to r i d the heaspace sample of e t h y l a l c o h o l . F i g . 11 shows the l a r g e e t h a n o l peak e l u t i n g e a r l y i n the chromatogram which o b l i t e r a t e s a l a r g e p o r t i o n of the chromatogram, masking an unknown number of aroma components of the wine t h a t e l u t e i n t h i s a r e a . Because the focus of t h i s s tudy was d i r e c t e d towards d e v e l o p i n g a method for q u a l i t y c o n t r o l f o r the i n d u s t r y , a qu ick method wi th minimum sample h a n d l i n g was d e s i r e d . D e v e l o p i n g a headspace procedure tha t i n c l u d e d s teps to e l i m i n a t e e t h y l a l c o h o l would have made the t e c h n i q u e c o m p l i c a t e d , t e d i o u s , long and b a s i c a l l y i m p r a c t i c a l 53 f o r r o u t i n e use i n the i n d u s t r y . A l though i t i s p o s s i b l e t h a t the masked peaks may have been important i n c a l c u l a t i o n of the p a t t e r n s i m i l a r i t y c o n s t a n t of the wines , i t was hoped t h a t the o b j e c t i v e s of t h i s work c o u l d be a t t a i n e d wi thout t h i s e x t r a i n f o r m a t i o n . Bes ides e t h a n o l , water i s a l s o a dominant component of wine t h a t can cause problems i n aroma a n a l y s i s t e c h n i q u e s . Large volumes of headspace v o l a t i l e s are o f t e n p r e c o n c e n t r a t e d i n c h i l l e d t r a p s , a t e c h n i q u e s i m i l a r i n p r i n c i p l e to c r y o f o c u s s i n g , but the major v o l a t i l e r e c o v e r e d i s water (Jennings et a l . , 1972) . When 1 mL of wine headspace sample i s p r e c o n c e n t r a t e d by c o l d t r a p p i n g and a n a l y z e d by GC, the same s i t u a t i o n p r o b a b l y r e s u l t s e s p e c i a l l y a t h igher temperatures such as 5 5 0 C . Even though water may be the p r i n c i p l e component i n the headspace sample , the flame i o n i z a t i o n d e t e c t o r (FID) i s i n s e n s i t i v e to water . Aqueous samples can be a n a l y z e d by the FID wi thout a l a r g e s o l v e n t peak o b s c u r i n g the f i r s t p a r t of the chromatogram (Rowland, 1974). Flame i o n i z a t i o n d e t e c t i o n i s p a r t i c u l a r l y s u i t e d for headspace a n a l y s i s not o n l y because of i t s l a c k of response to water , but a l s o because of i t s h i g h s e n s i t i v i t y to o r g a n i c compounds and i t s l a r g e l i n e a r range (Nawar, 1966). Thermal c o n d u c t i v i t y d e t e c t o r s , i n c o n t r a s t , have u n i v e r s a l response but poor s e n s i t i v i t y (Rowland, 1974). One other aspec t of headspace a n a l y s i s of wine needs to be d i s c u s s e d . F i g . 11 shows t h a t p r i o r to the e l u t i o n of e t h a n o l , t h e r e appears to be some peak d i s t o r t i o n o c c u r r i n g . 54 K o l b et a l . (1986) r e p o r t e d t h i s phenomenon i n the headspace a n a l y s i s of cheese when a l a r g e volume of headspace gas was i n t r o d u c e d i n t o the co lumn. S ince peak d i s t o r t i o n d i s a p p e a r e d when the headspace sample volume was r e d u c e d , they a t t r i b u t e d the malformed peaks to column o v e r l o a d . S p l i t or malformed peaks have been c a l l e d the "Chris tmas t r e e e f f e c t . " A c c o r d i n g to Jenn ings (1987), t h i s r e s u l t s from the exposure of fused s i l i c a column to n o n - u n i f o r m h e a t i n g from the oven h e a t e r . Peak d i s t o r t i o n occurs when the f r o n t of the c h r o m a t o g r a t i n g band i s exposed to a h i g h e r temperature and the back of the band i s d e c e l e r a t e d by a lower t e m p e r a t u r e . I t i s u n l i k e l y t h a t the s p l i t t i n g peaks are o c c u r r i n g because of n o n - u n i f o r m h e a t i n g i n t h i s s tudy because apple j u i c e samples were a n a l y z e d u s i n g the same ins trument and under a lmost i d e n t i c a l c o n d i t i o n s but no such phenomenon was o b s e r v e d . I f sample o v e r l o a d i n g i s the cause , i t i s not c l e a r why the problem was e x c l u s i v e to wine; a l l of the o ther samples a n a l y z e d d i d not e x h i b i t peak d i s t o r t i o n . One e x p l a n a t i o n may be t h a t u n l i k e other samples , wine may c o n t a i n h i g h c o n c e n t r a t i o n s of some ex treme ly v o l a t i l e low b o i l i n g s u b s t a n c e s , l i k e a c e t a l d e h y d e , which are o v e r l o a d i n g the column and c a u s i n g peak s p l i t t i n g . Column o v e r l o a d i n g has a l s o been blamed for t h i s problem by G u n t e r t et a l . (1986 ) . - Another cause for the poor peak shape of some of the components e l u t i n g i n the f r o n t p a r t of the chromatogram may be r e l a t e d to the type of l i q u i d s t a t i o n a r y phase used . G u n t e r t et a l . (1986) r e p o r t e d a 55 s i m i l a r problem i n a n a l y s i n g for low v o l a t i l i t y components i n wine . They b e l i e v e t h a t many a c i d i c wine compounds of r e l a t i v e l y h i g h c o n c e n t r a t i o n t h a t are e l u t e d e a r l y are o n l y s l i g h t l y s o l u b l e i n the s t a t i o n a r y phases such as DB-5 (the column used i n t h i s work) and DB-1701 are r e s p o n s i b l e f o r t h i s p r o b l e m . C . PRECISION AND INTERNAL STANDARD  P r e c i s i o n R e p e a t a b i l i t y of the HSGC p r o f i l e s was determined w i t h an apple j u i c e sample . T h i r t e e n peaks of v a r y i n g areas were s e l e c t e d from the chromatogram. Tab le 1 shows the means, s t a n d a r d d e v i a t i o n s and c o e f f i c e n t s of v a r i a t i o n of the 13 peaks for 3 i n j e c t i o n s . The mean c o e f f i c i e n t of v a r i a t i o n of the peaks ranged from 1.70 to 9.28% and the average c o e f f i c e n t of v a r i a t i o n of a l l the 13 peaks was 5.26%. R e p r o d u c i b i l i t y of the i n t e r n a l s t a n d a r d wi th 18 r e p l i c a t e s of wine samples was 5.08% as shown i n T a b l e 2. P r e c i s i o n of manual headspace e x t r a c t i o n s w i th g a s - t i g h t s y r i n g e s i s u s u a l l y not adequate compared to automated headspace sampl ing u n i t s t h a t use h i g h p r e c i s i o n pneumatic sampl ing ( C l o s t a et a l . , 1983) . R o d r i q u e z and C u l b e r t s o n (1983) used gas t i g h t s y r i n g e s to q u a n t i t a t e s e l e c t e d compounds i n the headspace of orange j u i c e . R e l a t i v e s t a n d a r d d e v i a t i o n s ranged from 10 to 40%. E t t r e et a l . (1980) t e s t e d the r e p e a t a b i l i t y of a n a l y z i n g an n - a l k a n e mixture wi th an 56 Table 1. R e p e a t a b l i t y of headspace method u s i n g apple j u i c e samples (n=3). Peak Mean St. Dev. Cof. Var. 1 42848 2636 6.15 2 6746 626 9.28 3 8635 783 9.07 4 26953 459 1.70 5 276607 9878 3. 57 6 18837 1166 6.19 7 9770 655 6.70 8 103416 2356 2.28 9 28583 1691 5.92 10 19251 858 4.46 11 35713 986 2.76 12 207358 12716 6.13 13 19141 786 4.11 57 Table 2. R e p e a t a b i l i t y of the i n t e r n a l standard i n the wine samples an a l y z e d . Average Peak Area* 7632522 7552853 7409673 7286683 7434091 7245762 6435650 7057862 7780734 Mean Standard D e v i a t i o n C o e f f i c i e n t of V a r i a t i o n 7315092 371546 (%) 5.08 * average of two r e p l i c a t e s 58 automated headspace sampler and found that the r e l a t i v e standard d e v i a t i o n for the 4 compounds was l e s s than 1.0%. Using the same headspace sampling system, Geiger (1978) examined the headspace composition of beer. R e p r o d u c i b i l i t y of 6 s e l e c t e d components v a r i e d from 2.1 to 7.6% with an o v e r a l l mean c o e f f i c i e n t of v a r i a t i o n of 4.2%. Results of t h i s study should be compared to the r e s u l t s of Geiger (1978) rather than E t t r e et a l . (1980) because an a l c o h o l i c beverage was t e s t e d instead of a high p u r i t y of one c l a s s (n-alkanes) components. In view of t h i s , the r e p e a t i b i l i t y of the method i s comparable to that of Geiger (1978). I n t e r n a l Standard Even though an automated headspace sampling system was used, an i n t e r n a l standard was added because of p o s s i b l e e r r o r s a r i s i n g from the sample preparation steps, or from adsorption of v o l a t i l e s by the glassware or other parts of the sampling u n i t s . Adsorption of v o l a t i l e s has been reported on the w a l l s of the glass syringes used for sampling (Buttery et a l . , 1965) but the problem can be overcome by coating the i n s i d e w a l l s of the syringe with T e f l o n , s i l a n e or other i n e r t m a t e r i a l s (Franzen and K i n s e l l a , 1975). To determine i f any adsorption was o c c u r r i n g on the glass v i a l s or septum, or the t r a n s f e r l i n e and valves and tubings of the sampling u n i t , the e f f e c t of concentration on the peak area of the i n t e r n a l standard was t e s t e d . F i g . 12 shows that the r e l a t i o n s h i p i s l i n e a r (R = 0 .9998 ) i n d i c a t i n g that no adsorption of aroma 59 components was oc c u r r i n g i n the headspace sampling u n i t or the glassware. This i s a l s o supported by the f a c t that a blank run showed a clean chromatogram. Blank runs between samples a l s o i n d i c a t e d that there was no carry-over o c c u r r i n g . Before d i s c u s s i n g wine blending o p t i m i z a t i o n , i t i s important to mention why t r i a l 1 and 2 blends were analyzed at d i f f e r e n t e q u i l i b r a t i n g c o n d i t i o n s . T r i a l 1 wines were conditioned at 6°C while t r i a l 2 wines were incubated at 37°C for headspace a n a l y s i s . The chromatogram shown i n F i g . 11 i s the a n a l y s i s of a wine sample at 55°C. Although there are many peaks present i n d i c a t i n g good s e n s i t i v i t y , there may be adverse r e a c t i o n s t a k i n g place at t h i s temperature. In a d d i t i o n , v o l a t i l e a n a l y s i s at 55°C does not r e f l e c t the headspace composition of wine as i t would be when consumed. Since white wine i s often served at c h i l l e d or r e f r i g e r a t i o n temperature, 6°C was s e l e c t e d for t r i a l 1. A temperature of 37°C was chosen because when wine i s taken i n t o the mouth i t i s at body temperature. Representative chromatograms of t r i a l 1 and 2 wines i n c l u d i n g an example of the v a r i e t a l wine, white stock and the ta r g e t are shown i n F i g . 13 to 15 and 16 to 18. As expected, s e n s i t i v i t y dropped when wines were analyzed at 37°C and e s p e c i a l l y at 6°C. A quick examination of the chromatograms rev e a l s that at each temperature the patterns of the v a r i e t a l , white stock and targe t wines are quite s i m i l a r . A c l o s e r examination of the chromatographic r e p o r t , however, i n d i c a t e s that there are s i g n i f i c a n t d i f f e r e n c e s i n . a r e a s of i d e n t i c a l 60 1.7 0. 60 80 100 120 140 160 180 200 CONCENTRATION (PPM) Pig. 12. The Effect of increasing concentration of internal standard on peak area. 61 v> r> cscn o\ cor> o> <\JrV CO 5 —i 10 — r ~ 15 20 — T -25 —r~ 30 Rentention Time (min.) Pig. 13. Representative varietal wine of trial 1. HSGC profile of a Chenin Blanc <r«D4>» tf> <\r>*.0<» <7> ro CD L- JUL V 1 Fig. of - i r-5 10 14. Representative trial 1. 15 Retention Time (min.) HSGC profile of a 20 white stock wine 25 30 b>-4H PI VXD • r- -t I! .1 "J VI Li I 30 5 Fig. 15. trial 1. — r ~ 10 HSGC — _ —| 1 15 20 Rentention Time (min.) profile of the target wine, Leibesheim, f o r i 25 i V <SO» • <7-a -«r*t» (4 • NN • 03 — r -30 T -.0 5 Fig. wine 16. of —j-10 Representative trial 2. 1 15 Rentention Time (min.) HSGC profile of a : r 20 Verdelet varietal 1 25 I I I i i j — 5 10 15 20 25 30 Retention Time (min.) Fig. 18. HSGC profile of the target wine, Cuvee White, for trial 2. r e t e n t i o n t ime peaks . These r e s u l t s are i n agreement w i t h e a r l i e r f i n d i n g s . Brander (1974) suggested t h a t e s s e n t i a l l y the same v o l a t i l e components are presen t i n a l l wine v a r i e t i e s and t h a t aroma d i f f e r e n c e s among v a r i e t i e s are due to these components b e i n g p r e s e n t i n v a r y i n g r a t i o s . D i f f e r e n c e s are q u a n t i t a t i v e r a t h e r than q u a l i t a t i v e . Moreover , Nykanen (1986) suggested t h a t the b a s i c f l a v o r components of a l c o h o l i c beverages i n c l u d i n g wines , brandy and whiskey are the same because most of them are formed d u r i n g f e r m e n t a t i o n and t h a t d i f f e r e n c e s i n t a s t e and s m e l l are due to d i f f e r e n c e s o c c u r i n g i n the q u a n t i t i e s of compounds. A l s o C r a i g , (1988) r e p o r t e d t h a t the aromagrams of k i w i f r u i t and grape wine appeared q u i t e s i m i l a r but t h a t s i g n i f i c a n t d i f f e r e n c e s o c c u r r e d i n the q u a n t i t i e s of many peaks . S t e r n et a l . (1975), however, s t r e s s e d the importance of t r a c e q u a n t i t i e s of subs tances to the aroma of wine . These v o l a l i t e s may be presen t at too low c o n c e n t r a t i o n to be d e t e c t e d but may be s i g n i f i c a n t o r g a n o l e p t i c a l l y . D. SIMPLEX OPTIMIZATION The f i r s t s t ep of the o p t i m i z a t i o n was to c a l c u l a t e the s i m i l a r i t y c o n s t a n t s of the v a r i e t a l and white s tock wines to the t a r g e t wine . Tab le 3 shows a l l of the b l e n d i n g a n d - t a r g e t wines f o r both t r i a l s 1 and 2. Twenty peaks for t r i a l 1 and 36 peaks for t r i a l 2 w i th d i s t i n c t v a r i a t i o n were s e l e c t e d based on the c r i t e r i o n d i s c u s s e d i n the m a t e r i a l s and methods 68 s e c t i o n . The r e l a t i v e s t a n d a r d d e v i a t i o n of these peaks was g r e a t e r than 20%. Peaks i n the e a r l y p o r t i o n of the chromatogram i n c l u d i n g e t h a n o l were omi t ted from the s e l e c t i o n p r o c e d u r e . U s i n g the GC Data E n t r y program, the chosen peaks were e n t e r e d i n t o an IBM PC computer . C o n s i d e r i n g the l a r g e da ta e n t r y r e q u i r e d , any e r r o r s made i n e n t r y c o u l d be c o r r e c t e d w i t h the next program c a l l e d GC Data C o r r e c t i o n . T h i s program a l s o n o r m a l i z e d the peak areas u s i n g the area of the i n t e r n a l s t a n d a r d . Normal i zed , data was r e c a l l e d by the S i m i l a r i t y Constant program to c a l c u l a t e the s i m i l a r i t y c o e f f i c i e n t s of the v a r i e t a l and white s tock wines . T a b l e 4 show the s i m i l a r i t y c o n s t a n t s of t r i a l 1 and 2 wines . The v a l u e s for t r i a l 1 are c l o s e t o g e t h e r and q u i t e h i g h around 0.969 to 0.991 wi th the e x c e p t i o n of 0.817 and c l o s e t o g e t h e r . In c o n t r a s t , the s i m i l a r i t y c o n s t a n t s of t r i a l 2 are more v a r i e d , r a n g i n g from 0.598 to 0 .901 . Before e x p l a i n i n g what these r e s u l t s mean, i t i s important to d i s c u s s why there i s such a l a r g e v a r i a t i o n i n the s i m i l a r i t y c o n s t a n t s of t r i a l 1 and 2. Because t r i a l 1 wines were a n a l y z e d a t a v e r y low t e m p e r a t u r e , o n l y a s m a l l number of v e r y v o l a t i l e peaks c o u l d be a n a l y z e d by the headspace method. S ince fewer peaks , a lmost h a l f the number, were a v a i l a b l e , the s i m i l a r i t y c o n s t a n t s for most of these wines were v e r y h i g h . D e s p i t e t h i s s i t u a t i o n , the r e s u l t s s t i l l showed some v a l i d i t y . For example, wine sample 7 was a preb lended wine t h e r e f o r e , i t had one of the h i g h e s t match w i t h the t a r g e t . Sample 25 and 14 were white s tock wines made 69 Table 3. Blending and target wines for t r i a l s 1 and 2. T r i a l Target V a r i e t a l White Stock Premium Stock (Chenin blanc) 1 Leibesheim 7,14,20,25 61,70,69,12 (Verdelet) 2 Cuvee White 74,47,51,58,71 70,12 22 70 Table 4. P a t t e r n s i m i l a r i t y constants of t r i a l 1 and 2 wines. Sample No S i m i l a r i t y Constant T r i a l 1 7 69 14 20 25 61 70 12 0.991 0.976 0.987 0.995 0.817 0.978 0.969 0.985 T r i a l 2 71 0.901 74 0.805 22 0.648 47 0.856 51 0.598 58 0.610 70 0.738 12 0.778 71 from grapes o r i g i n a t i n g from the same v i n e y a r d except t h a t for sample 25, the grapes were p i c k e d e a r l y i n the season and for sample 14, they were h a r v e s t e d l a t e . Wine made from the l e s s r i p e grapes had a s i m i l a r i t y c o n s t a n t va lue of 0.817 whi l e the wine made from the more r i p e grapes had a s i m i l a r i t y c o n s t a n t v a l u e 0 .985 . Even though l e s s data was a v a i l a b l e to c a l c u l a t e the p a t t e r n s i m i l a r i t y v a l u e s , the program c o u l d s t i l l d i s c e r n d i f f e r e n c e s i n the wine samples . For t h i s r e a s o n , i t was d e c i d e d to c o n t i n u e c a r r y i n g out the b l e n d i n g o p t i m i z a t i o n on t r i a l 1. A c c o r d i n g to commerc ia l p r a c t i c e , the winemaker grades the v a r i e t a l wines by s ensory methods and s e l e c t s the best v a r i e t a l wine as the p r i n c i p a l wine which w i l l be used for b l e n d i n g wi th the white s tock wines . T h i s procedure i s based on the b l e n d e r ' s e x p e r i e n c e wi th v a r i e t a l wines a l o n e . S ince i n t h i s s t u d y v a r i e t a l and white s tock wines were be ing compared to the t a r g e t , which i s a b lended wine , i t was not an a p p r o p r i a t e comparison to make. I t would have been b e t t e r to grade v a r i e t a l wines by comparing them to a known best v a r i e t a l wine . Because t h i s was a f i r s t - t i m e s t u d y on b l e n d i n g o p t i m i z a t i o n of wines , i t was not p o s s i b l e to c l a s s i f y these wines i n terms of f l a v o r q u a l i t y wi thout hav ing some GC data of v a r i e t a l wines t h a t have been graded for f l a v o r q u a l i t y by wine e x p e r t s . Due to the l a c k of t h i s i n f o r m a t i o n , the v a r i e t a l wines judged as best by the winemaker who was c o o p e r a t i n g wi th t h i s s t u d y , were s e l e c t e d as the p r i m a r y b l e n d i n g s t o c k . Chenin b l a n c , sample 14, and 72 V e r d e l e t , sample 47, were, the p r i n c i p a l b l e n d i n g wines for t r i a l s 1 and 2, r e s p e c t i v e l y . B l e n d i n g O p t i m i z a t i o n Because t h i s was an i n i t i a l s tudy on wine b l e n d i n g the o p t i m i z a t i o n problem was kept r e l a t i v e l y s i m p l e . P r a c t i c a l wine b l e n d i n g takes i n t o account r e g u l a t i o n s f o r commercia l winemakers , f o r example the amount of f o r e i g n s tock a l l owed to be b lended w i t h domest ic s tock wine , a v a i l a b i l i t y of b l e n d i n g s t o c k s , consumer p r e f e r e n c e s and c o s t c o n s i d e r a t i o n s ( J a c k i s h , 1985) . However, t h i s i s not to imply t h a t the o p t i m i z a t i o n program i s i n c a p a b l e of accomodat ing such f a c t o r s . T h i s program can e a s i l y be m o d i f i e d to handle any of the above mentioned c o n s t r a i n t s . In t h i s s t u d y , however, wines were o p t i m i z e d for aroma wi th o n l y one c o n s t r a i n t s i n c e the o b j e c t i v e of the work was to determine i f a c o m p u t e r - a i d e d approach to b l e n d i n g c o u l d be s u c c e s s f u l . More complex b l e n d i n g problems c o u l d be i n v e s t i g a t e d l a t e r a f t e r the outcome of t h i s s t u d y . Once a p r i n c i p a l v a r i e t a l wine had been s e l e c t e d , f a c t o r ranges for the r e s t of the white s tock wines were en tered i n t o the B l e n d i n g O p t i m i z a t i o n program. Tab le 5 shows the ranges of the white s tock wines f o r both t r i a l s . For t r i a l 1 a l l the upper l i m i t s were se t a t 40% but for t r i a l 2 s i n c e wine sample 22 was a premium white s tock wine , more c o s t l y than r e g u l a r white s tock wines , the upper l i m i t was se t at 20%. The r e s t of the white s tock wines of t h i s t r i a l were a s s i g n e d 73 Table 5. Factors and their l i m i t s for the blending optimization of t r i a l 1 and 2 wines. Factors Lower Limit Upper Limit T r i a l 1 Sample 70 12 69 0.000 0.000 0.000 0, 0 0 400 400 400 T r i a l 2 Sample 70 12 22 0.000 0 .000 0.000 0 0 0 400 400 200 7 4 l i m i t s of 40%. The o p t i m i z a t i o n program used the p a t t e r n s i m i l a r i t y subprogram and the p r e v i o u s l y entered GC data of the p r i n c i p a l blending stock, the white stock, and the t a r g e t wine to search for the optimal blending r a t i o s of the v a r i e t a l and white stock wines which gives the highest s i m i l a r i t y constant values between the GC p r o f i l e s of the blend and the t a r g e t wine. Results of the t h e o r e t i c a l o p t i m i z a t i o n for t r i a l 1 are shown i n Table 6. A f t e r 13 v e r t i c e s , a s i m i l a r t y c o e f f i e c i e n t of 0.993 was obtained. V e r t i c e s 11 to 13 were averaged to c a l c u l a t e the f i n a l blending r a t i o of 40.0, 49.0 and 22.9% of white stock wines 70, 12 and 69 with the v a r i e t a l wine 14. For the a c t u a l blending, the t o t a l a v a i l a b l e volume of sample 14 would be taken as 100% and the the amounts of the other wines would be a percentage of the t o t a l v a r i e t a l wine as determined by the o p t i m i z a t i o n program. For example, i f there was 10.0 L of the primary stock wine a v a i l a b l e for blending, and 40.0% was the d i c t a t e d r a t i o of a white stock wine, then 4.0 L of t h i s wine would be required for the formulation. For the blending o p t i m i z a t i o n of t r i a l 2, the program i t e r a t e d 23 v e r t i c e s to reach a s i m i l a r i t y constant of 0.861 (Table 7). A r a t i o of 26.9% of wine sample 70 was needed to blend with wine sample 47, the p r i n c i p a l blending wine. As i n the case f o r t r i a l 1, the three f i n a l v e r t i c e s were averaged to obtain the f i n a l blending r a t i o s . Results of t h i s o p t i m i z a t i o n suggest that wine samples 12 and 22 do not c o n t r i b u t e favorably to the formulation of the t a r g e t wine. 75 Table 6. Blending o p t i m i z a t i o n of t r i a l 1 wine. Vertex Sample Ratios Response No. 70 No. 12 No. 69 I n i t i a l 1 0. .000 0, . 000 0, , 000 0. .988 S implex 2 0. . 377 0, .094 0. ,094 0. .993 3 0. . 094 0. . 377 0, .094 0, .991 4 0. , 094 0. .094 0. , 377 0. ,992 R e f l e c t i o n 5 0. .377 0, .377 0. ,377 0. .993 Expans ion 6 0, . 566 0. .566 0. , 566 0 . .993 R e f l e c t i o n 7 0. .471 0. .000 0. , 471 0. .993 Expans ion 8 0. , 660 - o . .189 0. ,660 0 , .992 R e f l e c t i o n 9 . 0. .723 0. .220 0. ,251 0, .993 R e f l e c t i o n 10 0. .670 0, . 304 0. ,639 0 . .993 C o n t r a c t i o n 11 0. .400 0. .147 0, ,230 0, .993 R e f l e c t i o n 12 0, . 400 0. ,000 0. ,258 0. .993 Expansion 13 0. .400 0. ,000 0, ,199 0, ,993 F i n a l Average Value 0. .400 0 . ,049 0, . 229 0. ,993 76 Table 7. Blending o p t i m i z a t i o n of t r i a l 2 wine. Vertex Sample Ratios Response No. 70 No. 12 No. 22 I n i t i a l 1 0. .000 0, .000 0. ,000 0. ,856 S implex 2 0. . 377 0. .094 0. , 047 0 . ,856 3 0 . ,094 0. , 377 0. . 377 0. .850 4 0. ,094 0 . 094 0 . ,189 0. . 854 Re f l e c t i on 5 0. .220 -0, ,2 51 0, ,110 0. ,854 Cont r a c t i o n -R 6 0. ,189 -0. ,094 0. ,094 0. , 859 R e f l e c t i o n 7 0, ,283 -0. .094 -0. ,094 0 . , 861 Expansion 8 0 . , 377 -0. ,189 -0. , 236 0. .850 R e f l e c t i o n 9 0. , 566 -0. .063 0. , 031 0. , 858 R e f l e c t ion 10 0. . 314 -0. , 262 -0 , .026 0. ,857 Cont r a c t i o n -R 11 0. , 330 0. .000 0, ,000 0 . ,861 R e f l e c t i o n 12 0. ,000 0. , 000 0. . 000 0, ,856 Contract i on -W 13 0. ,400 0 , 000 0 , 016 0 . ,860 R e f l e c t i o n 14 0 . ,400 0. .031 0 . , 000 0, ,859 Co n t r a c t i o n -W 15 0 . , 263 0 . 000 0 . 034 0 . 860 R e f l e c t i o n 16 0. ,184 0. .000 0, , 000 0 . , 860 R e f l e c t i o n 17 0. ,268 0, ,000 0 , 000 0. , 861 R e f l e c t i n 18 0. ,400 0, .000 0. , 000 0 . , 860 Co n t r a c t i o n -W 19 0. ,239 0. .000 0 , 000 0 . ,861 R e f l e c t i o n 20 0. ,197 0, .000 0. .000 0, , 861 Contraction-' W 21 0 . , 297 0 . , 000 0 , 000 0 . ,861 R e f l e c t i o n 22 0. , 230 0. .000 0, .000 0. , 861 Contract ion -W 23 0 , 280 0 , 000 0 , . 000 0 . ,861 F i n a l Average Value 0. , 269 0, , 000 0. , 000 0. , 861 77 Blending r a t i o s and the corresponding a c t u a l volumes of wines used f o r s i m u l a t i n g the t a r g e t are shown i n Table 8. This t a b l e a l s o shows the r a t i o s of the commercial blend for t h i s year. E. ADJUSTMENTS FOR ACIDITY AND SWEETNESS Once the wines were blended to match the aroma of the t a r g e t wine, the aroma optimized wines had to be matched for the other component of f l a v o r , t a s t e . Making adjustment to a blend f o r a c i d i t y and sweetness i s one of the f i n a l steps i n p r a c t i c a l wine blending i n the i n d u s t r y . Procedures i d e n t i c a l to the ones followed by the winery were used. F i r s t t i t r a t a b l e a c i d s of the blends were measured by t i t r a t i o n with 0.1 N NaOH to pH 8.2. T i t r a t a b l e a c i d i t y of the t a r g e t wine and the computer optimized blend of both t r i a l s are shown i n Table 9. For t r i a l 1, the a c i d i t y value of the blend was 0.622 g per 100 mL compared to the t a r g e t wine with a value of 0.570 g per 100 mL. To reduce the a c i d or sourness character of the wine, t h i s blend was d i l u t e d with water (2% of the t o t a l volume of wine). Canadian r e g u l a t i o n s permit a maximum of 10% d i l u t i o n of wine with water. A f t e r d i l u t i o n , the blend now had a t i t r a t a b l e a c i d value of 0.555 which was close to the t a r g e t . No adjustments for a c i d i t y were needed for the t r i a l two blend as the values were judged to be close enough to the t a r g e t . T o t a l s o l u b l e s o l i d s were determined by a B r i x hydrometer 78 Table 8. Blending ra t i o s of computer-aided blends and commericial blends for t r i a l s 1 and 2. T r i a l 1 Computer-aided blend Commercial blend Wine Percent* Volume Wine Percent Sample (mL) Sample 14 59.5 1500 7 70 70 23.8 600 69 15 69 13.7 345 12 15 12 3.0 70 T r i a l 2 47 78.7 1200 5§ 60 70 21.3 324 12 40 * percent of t o t a l volume of wine 79 Table 9. Tit r a t a b l e a c i d i t y of the computer optimized blends, commercial blends and the target wines. T r i a l 1 T r i a l 2 (g/100 mL) Target wine 0.570 0.525 Commercial blend 0.600 0.570 Computer optimized blend 0.555* 0.540 * ameliorated with d i s t i l l e d water 80 as a measure of the sweetness character of wine. For both t r i a l s the levels of sugars were too low compared to the target wine and had to be ameliorated with l i q u i d invert sugar. The l e v e l of the t o t a l soluble s o l i d s before and after adjustment are shown in Table 10. This table also shows the acid and sugar levels of the commercial blends that were made to simulate the previous years wines, Leibesheim and Cuvee White. These wines were formulated using the same available stock wines as were available for thi s study for the computer-aided optimization. F. VERIFICATION OF RESULTS To confirm the res u l t s of the computer optimization of blending wine, sensory tests were conducted with untrained consumers and expert taste panels. In addition, the t h e o r e t i c a l l y optimized blends were analyzed by HSGC to determine how close the actual s i m i l a r i t y constants of these blends would be to the predicted value. S e n s o r y Evaluation. To determine i f there was a detectable difference in the computer optimized blends and the target, t r i a n g l e tests were conducted. Commercially blended wines were also tested against the target. Samples were served at near room temperature in coded wine glasses. Twelve judges evaluated t r i a l 1 wines. Each judge received 3 coded samples: 6 judges 81 Table 10. Total soluble s o l i d s for the blends and targets of t r i a l s 1 and 2. T r i a l 1 T r i a l 2 (©Balling) Target wine + 0.30 -1.05 Subjective blend -0.60 -1.10 Computer optimized blend -1.50 -1.70 Computer optimized after adjustment blend + 0.35 -1.15 82 tested 2 samples of the computer-optimized blend and one of the t a r g e t blend, and the other s i x judges evaluated one computer-optimized blend and two t a r g e t blends. The order of the 3 samples was randomized for every p a n e l i s t . The same scheme was used for the 15 member panel that te s t e d t r i a l 2 blends. Results of the t r i a n g l e d i f f e r e n c e t e s t are shown i n Table 11. Six judges c o r r e c t l y i d e n t i f i e d the odd sample i n comparing the computer-aided blend while 4 judges s u c c e s s f u l l y picked out the odd sample for the commercially blended wine. Results of t r i a l 2 blends and the commercial blend t e s t s were s i m i l a r with 8 out of 15 judges and 5 out of 15 judges c o r r e c t l y i d e n t i f y i n g the odd sample, r e s p e c t i v e l y . Since 9 c o r r e c t judgements out of 12 and 10 c o r r e c t judgements out of 15 are necessary to e s t a b l i s h a s i g n i f i c a n t d i f f e r e n c e at the 99% l e v e l of confidence, i t may be concluded that there i s no detectable d i f f e r e n c e between the computer optimized blend and commercial blend, and the t a r g e t wines. This i m p l i e s that both the conventional and innovative blending schemes were s u c c e s s f u l i n s t a n d a r d i z i n g the f l a v o r q u a l i t y of Leibesheim and Cuvee White wines. I t i s i n t e r e s t i n g to note that i n t h e i r e v a l u a tions of these wines, n e a r l y a l l of the judges, both untrained and t r a i n e d , expressed d i f f i c u l t y i n s e l e c t i n g the odd sample. Many of them commented that they had to r e l y on t h e i r sense of smell alone to detect d i f f e r e n c e s . 83 Table 11. Results of the t r i a n g l e test.comparing the computer optimized blends and commercailly formulated blends with the t a r g e t wine. Compar ison Correct response T r i a l 1 Objective blend 6* Subjective blend 4* T r i a l 2 Objective blend 8* Subjective blend 5* * not s i g n i f i c a n t 84 S i m i l a r i t y constants of the Blends S i m i l a r i t y constants of t r i a l 1 and 2 blends and the corresponding commercial blends were determined. A l l blends were analyzed by HSGC ( F i g . 19 and 20) and the peak data was entered i n t o the S i m i l a r i t y Constant program of the o p t i m i z a t i o n program. The pr e d i c t e d s i m i l a r i t y constants for both t r i a l s were 0.993 and 0.861, r e s p e c t i v e l y while the a c t u a l values were 0.997 and 0.865, r e s p e c t i v e l y . The closeness of the a c t u a l values to the pre d i c t e d values supports the o r i g i n a l assumption that experimental o p t i m i z a t i o n was not necessary i n blending o p t i m i z a t i o n . 85 j i i i i i i i I i I i I 1 ! I i j ! i i i i i 1 1 1 I T~i T i i i—\ J 15 30 Retention Time (min.) Fig. 19. Chromatograms of headspace volatiles f r o m (a) the t a r g e t , (b) commercial blend and (c) the computer optimized blend f o r t r i a l 1. 86 ,1 JJL: 311 3 : , i i—i—i ! i—i i i i i—I i i " i I ' > ' ' ' 0 15 R e n t e n t i o n Time (min.) Fig. 20. Chromatograms o f headspace v o l a t i l e s f r o m t a r g e t , (b) commercial blend and (c) the computer optimized blend f o r t r i a l 2. • 1 30 (a) the 87 CONCLUSION A l t h o u g h s u g a r s , a c i d s and phenols are r e s p o n s i b l e for sweet, sour and b i t t e r and a s t r i n g e n t s e n s a t i o n s of wine , r e s p e c t i v e l y , odor i s the most important s e n s a t i o n i n the p e r c e p t i o n of wine q u a l i t y (Acree and C o t t r e l l , 1985) . V o l a t i l e f l a v o r components r e s p o n s i b l e for the aroma of wine have been s e p a r a t e d and i d e n t i f i e d by GC and GC-mass s p e c t r o m e t r i c methods. In t h i s s t u d y , white wine aroma c o n s t i t u e n t s were a n a l y z e d by HSGC u s i n g a c o l d t r a p p i n g t e c h n i q u e . S e n s i t i v i t y was s i g n i f i c a n t l y improved by t h i s method compared to other methods of enhancing the lower d e t e c t i o n l i m i t of headspace methods. The procedure b e i n g s imple and r a p i d ^ c o u l d be e a s i l y employed by w i n e r i e s . Based on the aroma p r o f i l e s of a number of b l e n d i n g component wines i n c l u d i n g v a r i e t a l and white s t o c k , b l e n d i n g to s t a n d a r d i z e two w i d e l y s e l l i n g commercia l w ines , L e i b e s h e i m and Cuvee W h i t e , was c a r r i e d out u s i n g s implex o p t i m i z a t i o n . The o p t i m i z a t i o n program was s u c c e s s f u l i n d e t e r m i n i n g the optimum b l e n d i n g r a t i o s of the white s tock wines and v a r i e t a l wines for s i m u l a t i n g the r e f e r e n c e or t a r g e t wine f o r both t r i a l s . R e s u l t s of the s e n s o r y a n a l y s i s i n d i c a t e d t h a t no s i g n i f i c a n t d i f f e r e n c e e x i s t e d between the computer o p t i m i z e d b lend and the p r e v i o u s y e a r ' s commercia l b l e n d . In a d d i t i o n , the s i m i l a r i t y c o e f f i c i e n t v a l u e s of the b lends were v e r y c l o s e to the p r e d i c t e d v a l u e s by the program. These r e s u l t s 88 confirmed the i n i t i a l assumption that computerized simplex optimization could be used for the blending problem instead of experimental simplex optimization. Despite the fact that both t r i a l s 1 and 2 for blending wines analyzed at 6°C and 37°C, respectively were successful in simulating the reference wine, the higher temperature of analysis is recommeneded for future work. At r e f r i g e r a t i o n temperature the chromatogram contained fewer peaks thus when the s i m i l a r i t y constant values were calculated by the optimization program, almost a l l of the wines had very high and similar values. Headspace analysis of wines at t h i s temperature did not provide enough peak data to adequately d i f f e r e n t i a t e t r i a l 1 wines as did the analysis at 37°C for t r i a l 2 wines. Sensory evaluation data provided further support to t h i s finding. Even though a l l of the triangle test t r i a l s were s t a t i s t i c a l l y not s i g n i f i c a n t , more panelists had d i f f i c u l t y s e l ecting the odd sample when presented with the commercially prepared blend and the target wine for both the Leibesheim and Cuvee White wines. Headspace anlaysis and subsequent results of the pattern s i m i l a r i t y constants of t r i a l 2 wine's were consistent with these findings as the commercial blend had a s l i g h t l y higher s i m i l a r i t y values than the computer optimized blend. For t r i a l 1, however, the sensory and s i m i l a r i t y constant results did not agree. From the above findings i t appears that headspace analysis of wines for blending optimization should be conducted at 37°C rather than at 6°C 89 for more a c c u r a t e r e s u l t s . With f u r t h e r work, a p p l i c a t i o n of t h i s r e s e a r c h on b l e n d i n g of wines for use as a q u a l i t y c o n t r o l method i n w i n e r i e s t h a t c o u l d r e p l a c e c o n v e n t i o n a l s ensory based b l e n d i n g procedures appears p r o m i s i n g . However, i t shou ld be s t r e s s e d t h a t the r e s u l t s of t h i s s tudy cannot be c o n s i d e r e d c o n c l u s i v e as the number of b l e n d i n g t r i a l s conducted were too few to make any d e f i n i t i v e c o n c l u s i o n s . The s tudy does , n e v e r t h e l e s s , p r o v i d e an important b a s i s for f u r t h e r r e s e a r c h to d e f i n e c l e a r l y the p o t e n t i a l use of HSGC and computer i zed s implex o p t i m i z a t i o n procedure for product f o r m u l a t i o n . 90 REFERENCES Acree, T. E. and T. H. E. C o t t r e l l (1985) Chemical i n d i c e s of wine q u a l i t y . In: A l c o h o l i c Beverages. G. G. B i r c h and M. G. L i n d l e y (Eds.). E l s e v i e r Applied Science Pub. L t d . , N.Y., pp. 145-159. Aishima, T. and S. Nakai (1986) Centroid maping o p t i m i z a t i o n : a new e f f i c i e n t o p t i m i z a t i o n for food research and processing. J . Food S c i . 51:1297-1310. Aishima, T., D. L. Wilson and S. Nakai (1987) A p p l i c a t i o n of . simplex a l g o r i t h m to f l a v o u r o p t i m i z a t i o n on patte r n s i m i l a r i t y of GC p r o f i l e s . In: Flavour Science and Technology: Proceedings of 5th Weurman Flavour Research Symposium. M. Martens, G. A. Dalen and J . Russwurm J r . (Eds.). John Wiley and Sons., Chichester, pp. 501-508. Amerine, M. A. and V. L. Si n g l e t o n (1977) Wine, An I n t r o d u c t i o n . U n i v e r s i t y of C a l i f o r n i a Press. Berkely, C a l i f o r n i a . Anon. (1987) Orange j u i c e . Consumer Reports. 76-80. Bayne, C. K. and I. B. Rubin (1986) P r a c t i c a l Experimental Designs and Optimization Methods for Chemists. VCH P u b l i s h e r s Inc., D e e r f i e l d Beach, F l o r i d a , pp. 105-128. Berridge, J.C. (1985) Techniques for Automated Optimization of HPLC Separation. Wiley I n t e r s c i e n c e P u b l i c a t i o n . Chichester. pp. 125-152. B e r t u c c i o l i , M. and G. Montedoro (1974) Concentration of the headspace v o l a t i l e s above wine for d i r e c t chromatographic a n a l y s i s . J . S c i . Fd. A g r i c . 25:625-687. B e r t u c c i o l i , M. and R. V i a n i (1976) Red wine aroma: I d e n t i -f i c a t i o n of headspace c o n s t i t u e n t s . J . S c i . Fd. A g r i c . 27: 1035-1038. Box, G. E. P. (1957) E v o l u t i o n a r y operation: A method of i n c r e a s i n g i n d u s t r i a l p r o d u c t i v i t y . Appl. S t a t . 6:81. Brander, F. C. (1974) V o l a t i l e composition of 'Zinfa n d e l ! t a b l e wine: some n e u t r a l components. Am. J . Enol. V i t i c . 25:13-16. Brander, C. F., R. F. Kepner, and A. D. Webb (1980) I d e n t i -f i c a t i o n of some v o l a t i l e compounds of wine of Vitis vinefera c u l t i v a r P i n o t n o i r . Am. J . Enol. V i t i c , 31:69-75. 91 B u t t e r y , R. G . , D. G . Guadagni and S. Okano (1965) J . S c i . Food A g r i c . 16:691. C h a r l e y , V . L . S. (1969) C i t r u s beverages - a s u r v e y of r e c e n t deve lopments . I n : P r o c e e d i n g s of the F i r s t I n t e r n a t i o n a l C i t r u s Symposium. V o l . 1. H . D. Champman ( E d . ) . R i v e r s i d e C o l o r P r e s s , R i v e r s i d e , C a l i f o r n i a . 249-252. Chaudhary, S. S . , A . D. Webb and R. E . Kepner (1968) GLC i n v e s t i g a t i o n of the v o l a t i l e compounds i n e x t r a c t s from Sauvignon b l a n c wines from normal and b o t r y t i s e d g r a p e s . Am. J . E n o l . V i t i c . 19:6-12 . C l o s t a , W. , H. Klemm, P . P o s p i s i l , R. R i e g g e r , G. S i e s s and B. K o l b (1983) The HS-100, an i n n o v a t i v e concept for automat i c headspace s a m p l i n g . Chromatogr . N e w s l e t t e r . 11:13-17 . Cobb, C . S. and M. Bursey (1978) Comparison of e x t r a c t i n g s o l v e n t s for t y p i c a l v o l a t i l e components of e a s t e r n wines i n model a q u e o u s - a l c o h o l i c sys tems. J . A g r i c . Food Chem. 26:197-199. Cook, R. (1983) Q u a l i t y of c i t r u s j u i c e s as r e l a t e d to c o m p o s i t i o n and p r o c e s s i n g p r o c t i c e s . Food T e c h . 37:68-77 ,133 . C o r d n e r , C . W. , C . S. Ough, A . N. K a s i m a t i s and J . J . K i s s l e r (1978) E f f e c t s of c r o p l e v e l on c h e m i c a l c o m p o s i t i o n and headspace v o l a t i l e s of L o d i Z i n f a n d e l grapes and wines . Am. J . E n o l . V i t i c . 29:247-253. C r a i g , J . T . (1988) A comparison of the headspace v o l a t i l e s of K i w i f r u i t wine and those of wine of V i t i s v i n e f e r a v a r i e t y M u l l e r T h u r g a u . Am. J . E n o l . V i t i c . 39:321-324. D r a v n i e k s , A . and A. O'Donnel (1971) P r i n c i p l e s and some t e c h n i q u e s of h i g h r e s o l u t i o n headspace a n a l y s i s . J . A g r i c . Food Chem. 19:1049-1056. D u c r u e t , V . (1984) Comparison of headspace v o l a t i l e s of c a r b o n i c m a c e r a t i o n and t r a d i t i o n a l wine. L e b e n s m . - W i s s . U . - T e c h n o l . 17:217-221. E t i e v a n t , P . X. (1981) V o l a t i l e phenol d e t e r m i n a t i o n i n wine. J . A g r i c . Food Chem. 29:65-67 . E t t r e , L . S . , J . E . P u r c e l l , J . Widomski , B. K o l b and P. P o s p i s i l (1980) I n v e s t i g a t i o n s on e q u i l i b r i u m headspace -open t u b u l a r column gas chromatography. J . Chromatogr . S c i . 18:116-125. F l a t h , R. A . , R. R. F o r r e y and A . D. K i n g J r . (1972) Changes produced by B o t r y t i s c i n e r e a p e r s . i n f i n i s h e d wines . Am. 92 J . Enol. V i t i c . 23:159-164. F l a t h , R. A. and H. Sugisawa (1981) Problems i n f l a v o r research. In: Flavor Research. Recent Advances. R. T e r a n i s h i , R. A. F l a t h and H. Sugisawa (Eds.) Marcel Dekker Inc., N. Y., pp. 1-10. Franzen K. L. and J . E. K i n s e l l a (1975) Physiochemical aspects of food f l a v o u r i n g . Chem. & Ind. 505-509. Geiger, E. (1978) Gas chromatographic headspace a n a l y s i s i n brewing. In: Applied Headspace Gas Chromatography. B. Kolb (Ed.). Heyden and Sons L t d . , Great B r i t a i n , pp. 73-79 . Gelsomini, N. (1985) Head space a n a l y s i s with c a p i l l a r y columns i n q u a l i t y c o n t r o l of wines. In: S i x t h I n t e r n a t i o n a l Symposium on C a p i l l a r y Chromatography. P. Sandra and W. Bertsch (Eds.). I t a l y , May 14-16. Huethig. Guntert, M., A. Rapp, G. Takeoka and W. Jennings (1986) HRGC and HRGC-MS a p p l i e d to wine c o n s t i t u e n t s of lower v o l a t i l i t y . Z. Lebensm. For.ch. 182:200-204. Hachenberg, H. and A. P. Schmidt (1977) Gas Chromatographic Headspace A n a l y s i s . Heyden and Sons L t d . , Rheine, Germany, pp. 10-18. Hardy, P. J . (1969) E x t r a c t i o n and concentration of v o l a t i l e s from d i l u t e aqueous and aqueous-alcohlic s o l u t i o n using t r i c h l o r o f l u o r o m e t h a n e . J . A g r i c . Food Chem. 17:656-658. Heath, H. B. and G. Reineccius (1986) Flavor Chemistry and Technology. AVI P u b l i s h i n g Co. Inc., Westport, Conn. pp. 1-3. Heide, T. R. (1985) Concentration of odorous headspace v o l a t i l e s . In: E s s e n t i a l O i l s and Aromatic P l a n t s : Proceedings of the 15th I n t e r n a t i o n a l Symposium on E s s e n t i a l O i l s . A. B. Svendsen and J . J . C. Scheffer (Eds.). K l u i v e r Academic Press, Hingham, M. A., pp. 43-60. I o f f e , B. V. and A. G. Vitenberg (1984) Headspace A n a l y s i s and Related Methods i n Gas Chromatography. John Wiley and Sons., N.Y., pp. 57-66. Issenberg, P. and I. Hornstein, (1970) A n a l y s i s of v o l a t i l e f l a v o r components i n foods. Adv. Chromatogr. 9:295-341. J a c k i s c h , P. (1985) Modern Winemaking. C o r n e l l U n i v e r s i t y Press, Ithaca, N.Y. Jennings, W. G. (1987) A n a l y t i c a l Gas Chromatography. Academic Press Inc., Orlando, F l o r i d a , pp. 59-73. 93 Jennings, W. G. and A. Rapp (1983) Sample Prep a r a t i o n for Gas Chromatographic A n a l y s i s . Dr. A l f r e d Huthig. Verlag GmbH, Heidelberg. Jennings, W. R. Wohleb and M. J . Lewis (1972) Gas chromatographic a n a l y s i s of headspace v o l a t i l e s of a l c o h o l i c beverages. J . Food S c i . 37:69-71. J u r s , P. C. (1986) Computer Software A p p l i c a t i o n s i n Chemistry. Wiley I n t e r s c i e n c e Pub., N. Y., pp. 125-133. Kolb, B. (1985) Q u a n t i t a t i v e aspects of f l a v o u r a n a l y s i s by e q u i l i b r i u m and dynamic headspace GC with c a p i l l a r y columns. In: E s s e n t i a l O i l s and Aromatic P l a n t s : Proceedings of the 15th I n t e r n a t i o n a l Symposium on E s s e n t i a l O i l s . A. Baerheim Svendsen and J . J . C. Scheffer (Eds.), pp.3-21. Kolb, B., B. Liebhardt, S. E t t r e (1986) Cryofocussing i n the combination of gas chromatography with e q u i l i b r i u m headspace sampling. Chromatographia. 21:305-311. Kwan, W. and B. R. Kowalski (1980) P a t t e r n r e c o g n i t i o n a n a l y s i s of gas chromatographic data. Geographic c l a s s i f i c a t i o n of V i t i s v e n i f e r a c v . Pinot Noir from France and the United States. J . A g r i c . Food Chem. 28:356-359. Lamikanra, 0. (1987) Aroma c o n s t i t u e n t s of Muscadine wines. J . Food Q u a l i t y . 10:57-66. Lang, J . W. (1983) Blending for sa l e and consumption. In "Flavour of D i s t i l l e d Beverages: O r i g i n and Development. Ed. J . R. P i g g o t t . E l l i s Horwood L t d . , Chichester, pp.256-263 . Massert, D. L., B. G. M. Vandeginste, S. N. Deming, Y. Michotte and L. Kaufman (1988) Chemometrics: A Textbook. E l s e v i e r Science P u b l i s h e r s . N.Y., pp. 293-304. McNalley, M. E. and R. L. Grob (1985) Current a p p l i c a t i o n s of s t a t i c and dynamic headspace a n a l y s i s : A review. Part Two: Non environmental a p p l i c a t i o n s . Am. Lab. 17:106-120. Montedoro, G. and M. B e r t u c c i o l i (1986) The f l a v o u r of wines, vermouth and f o r t i f i e d wines. In: Developments i n Food Science 3B. The Flavour of Beverages. I. D. Morton and A. J . MacLeod (Eds.). E l s e v i e r Press, N. Y. Morgan, S. L. and S. N. Deming (1974) Simplex o p t i m i z a t i o n of a n a l y t i c a l chemical methods. Anal. Chem. 46:1170-1181. Murray, K. E. (1977) Concentration of headspace, airborne and aqueous v o l a t i l e s on Chromosorb 105 for examination by gas 94 chromatography and gas chromatography-mass spectrometry. J. Chromatogr. 135:49-60. Nakai, S. (1982) Comparison of o p t i m i z a t i o n techniques for a p p l i c a t i o n to food product and process development. J . Food S c i . 47:144-152,157. Nakai, S. and G. Arteaga (1989) P r a c t i c a l Computer-aided Optim i z a t i o n for Food Research and Processing. E l s e v i e r Applied Science P u b l i s h e r , London, U. K. Nakai, S. and T. Kaneko (1985) S t a n d a r d i z a t i o n of mapping simplex o p t i m i z a t i o n . J . Food S c i . 50:845-846. Nakai, S., K. Koide and K. Eugster (1984) A new mapping super-simplex o p t i m i z a t i o n of food product and process development. J . Food S c i . 49:1143-1148. Nawar, W. W. (1966) Some co n s i d e r a t i o n s i n i n t e r p r e t a t i o n of d i r e c t headspace gas chromatographic analyses of food v o l a t i l e s . Food Tech. 20:213-215. Nedler, J . A. and R. Mead (1965) A simplex method for f u n c t i o n m inimization - e r r a t a . Comput. J . 7:308. Nelson, R. R. and T. E. Acree (1978) Concord wine composition as a f f e c t e d by maturity and processing technique. Am. J . Enol. V i t i c . 29:83-86. Nelson, R., T. E. Acree and R. M. Butts (1978) I s o l a t i o n and i d e n t i f i c a t i o n of v o l a t i l e s from Catawba wine. J . A g r i c . Food Chem. 26:1188-1190. Noble, A., R. A. F l a t h , and R. R. Forrey (1980) Wine headspace a n a l y s i s . R e p r o d u c i b i l i t y and a p p l i c a t i o n to v a r i e t a l c l a s s i f i c a t i o n . J . A g r i c . Food Chem. 28:346-353. Noble, A. C. (1978) Sensory and instrumental e v a l u a t i o n of wine aroma. In: A n a l y s i s of Foods and Beverages. Headspace Techniques. G. Charalambous (Eds.). Academic Press N. Y., pp.203-228. Noble, A. C. (1981) Use of p r i n c i p a l component a n a l y s i s of wine headspace v o l a t i l e s i n v a r i e t a l c l a s s i f i c a t i o n . V i n i d ' i t a l i a . 23:325-334. Noble, A. C , A. A. Murakami and G. F. Coope I I I (1979) R e p r o d u c i b i l i t y of headspace a n a l y s i s of wines. J . A g r i c . Food Chem. 27:450-452. Nykanen, L. (1986) Formation and occurrance of f l a v o r compounds in wine and d i s t i l l e d a l c o h o l i c beverages. Am. J . Enol. V i t i c . 37:84-96. 95 Nykanen, L. and H. Soumalainen (1983) Aroma of Beer, Wine and D i s t i l l e d A l c o h o l i c Beverages. D. R e i d e l Pub. Co., Dordrecht, Holland. Ough, C. S. and M. A. Amerine (1988) Methods of A n a l y s i s of Musts and Wines. John Wiley and Sons. N. Y. Peynaud, E. (1984) Knowing and Making Wine. John Wiley and Sons N.Y. P o l l , L. and J . M. F l i n k (1984) Aroma a n a l y s i s of apple j u i c e : Influence of s a l t a d d i t i o n on headspace v o l a t i l e compositon as measured by gas chromatography and corresponding sensory e v a l u a t i o n s . Food Chem. 13:193-207. Rankine, B. (1988) Blending: a most important aspect of winemaking. A u s t r a l i a n Grapegrower and Winemaker. 18:17-18. Rapp, A. and W. Khipser (1980) A new method for the enrichment of headspace components using wine as an example. Chromatographia 13:698-702. Rapp, A. (1981) A n a l y s i s of grapes, wines and brandies. In: A p p l i c a t i o n of Glass C a p i l l a r y Gas Chromatography. Marcel Dekker Inc. N. Y., pp. 579-621. Reineccius, G. and S. Ananadaraman (1984) Food Constituents and Food Residues: Their Chromatographic Detection. J . F. Lawerence (Ed.). Marcel Dekker, N. Y., pp.195-293. Rodriguez, P. A. and C. R. Culbertson (1983) Q u a n t i t a t i v e headspace a n a l y s i s of s e l e c t e d compounds i n e q u i l i b r i u m with orange j u i c e . In: Instrumental A n a l y s i s of Foods. Recent Progress. V o l . 2. G. Charalambous and G. I n g l e t t (Eds.). Academic Press. N. Y. pp. 187-195. Routh, M. W., P. A. Swartz and M. B. Denton (1977) Performance of the super modified simplex. Anal. Chem. 49:1422. Rowland, F. (1974) The P r a c t i c e of Gas Chromatography. 2nd Ed. Hewlett Packard Co., Avondale D i v i s i o n . P. A. Ryan, P. B., L. B. Richard and H. D. Todd (1980) Simplex techniques for non-linear o p t i m i z a t i o n . Anal. Chem. 52:1460. Saguey, I . , M. Mishkin and M. K a r e l (1986) Optimization methods and a v a i l a b l e software. Part I. CRC C r i t . Reviews Food S c i . Nutr. 20:275-299. Scheafer, J . , A. C. Tas, J . V e l i s e k , H. Maarse, M. C. ten Noever de Brauw and P. Slump (1983) A p p l i c a t i o n of patter n r e c o g n i t i o n techniques i n the d i f f e r e n t i a t i o n of wines. 96 I n : I n s t r u m e n t a l A n a l y s i s of Foods . Recent P r o g r e s s V o l . 2. E d . G. Charalambous and G. I n g l e t t . Academic P r e s s . N. Y . S c h r e i e r , P . (1979) F l a v o r c o m p o s i t i o n of wines: a r e v i e w . CRC C r i t . Reviews Food S c i . N u t r . 13:59-111. S c h r e i e r , P . , F . Drawert and K. 0. Abraham (1980) I d e n t i f i c a t i o n and d e t e r m i n a t i o n of v o l a t i l e c o n s t i t u e n t s i n burgundy P i n o t N o i r wines . Lebensm. W i s s . U . - T e c h n o l . 13:318-321. S c h u l t z , W. G. and J . M. R a n d a l l (1970) L i q u i d C02 for s e l e c t i v e aroma e x t r a c t i o n . Food T e c h . 24:1283-1286. S h e w f e l t , R. L . (1986) F l a v o r and c o l o r of f r u i t s as a f f e c t e d by p r o c e s s i n g . I n : Commercial F r u i t P r o c e s s i n g . J . G. Woodroof an B . S. L u h . ( E d s . ) AVI P u b l i s h i n g Co. I n c . , W e s t p o r t , Conn. Shibamoto, T . (1984) Gas chromatography. I n : A n a l y s i s of Foods and Beverages . Modern T e c h n i q u e s . G. Charalambous ( E d . ) . Academic P r e s s , O r l a n d o , F l o r i d a , pp . 93-115. Simpson, R. F . (1979a) Some important aroma components of white wine. Food T e c h . A u s t r a l i a . 516-522. Simpson, R. F . (1979b) I n f l u e n c e of gas volume sampled on wine headspace a n a l y s i s u s i n g p r e c o n c e n t r a t i o n on Chromosorb 105. C h r o m a t o g r a p h i a . 12:733-736. S i n g l e t o n , V . L . and A . C . Noble (1976) Wine f l a v o r and p h e n o l i c s u b s t a n c e s . I n : P h e n o l i c , S u l f u r and N i t r o g e n Compounds i n Food F l a v o r s . G. Charalambous and I . K a t z ( E d s . ) . American Chemica l S o c i e t y . Wash. D. C , ACS Symposium S e r i e s 26. S l i n g s b y , R. W. , R. E . Kepner , C . J . M u l l e r and A . D. Webb (1980) Some v o l a t i l e components of V i t i s v i n e f e r a v a r i e t y Cabernet Sauvignon wine . Am. J . E n o l . V i t i c . 31:360-363. S p e n d l y , W. , G. R. Hext and F . R. Himsworth (1962) S e q u e n t i a l a p p l i c a t i o n of s implex des igns i n o p t i m i z a t i o n and e v o l u t i o n a r y o p e r a t i o n . T e c h n o m e t r i c s . 4:441-461. S t e r n , D. J . , G. Guadagni and K. L . Stevens (1975) Aging of wine: q u a l i t a t i v e changes i n the v o l a t i l e s of Z i n f a n d e l wine d u r i n g two y e a r s . Am. J . E n o l . V i t i c . 26:208-213. S tevens , K. L . , R. A. F l a t h , A . Lee and D. J . S t e r n (1969) V o l a t i l e s from g r a p e s . Comparison of Grenache j u i c e and Grenache rose wine . J . A g r i c . Food Chem. 17:1102-1106. Suomala inen , H. (1971) Yeast and i t s e f f e c t on the f l a v o u r of 97 a l c o h o l i c beverages . J . I n s t . Brew. 77:164-177. Takeoka , G. and W. Jennings (1984) Developments i n the a n a l y s i s of headspace v o l a t i l e s : On-column i n j e c t i o n i n t o fused s i l i c a c a p i l l a r i e s and s p l i t i n j e c t i o n w i t h a low tempera-t u r e bonded PEG s t a t i o n a r y phase . J . Chromatographic S c i . 22:177-184. T h e o b a l d , D. J . (1977). Sensory e v a l u a t i o n i n t ea buy ing and b l e n d i n g . I n : Sensory Q u a l i t y C o n t r o l : P r a c t i c a l Approaches i n Food and Dr ink P r o d u c t i o n . P r o c e e d i n g s of a J o i n t Symposium h e l d i n the U n i v e r s i t y of A s t o n . The S o c i e t y of Chemica l I n d u s t r y . London, U . K. van Wyk, C . J . , R. E . Kepner , and A . D. Webb (1967a) Some v o l a t i l e components of V i t i s v i n e f e r a v a r i e t y White R e i s l i n g . 2. O r g a n i c a c i d s e x t r a c t e d from wine . J . Food S c i . 32:664:668. van Wyk, C . J . , R. E . Kepner , and A . D. Webb (1967b) Some v o l a t i l e components of V i t i s v i n e f e r a v a r i e t y White R e i s l i n g . 3. N e u t r a l components e x t r a c t e d from wine . J . Food S c i . 32:669-673. V i n e , R. P . (1981) Commercial Winemaking P r o c e s s i n g and C o n t r o l s . AVI Pub. Co . I n c . , W e s t p o r t , Conn. Webb, A . and C . J . M u l l e r (1972) V o l a t i l e aroma components of wines and other a l c o h o l i c beverages . Adv. A p p l . M i c r o b i o l . 15:75-146. W i l l i a m s , A . (1982) Recent developments i n the f i e l d of wine f l a v o u r r e s e a r c h . J . I n s t . Brew. 88:43-53 . W i l l i a m s , P . J . and C . R. S t r a u s s (1977) Apparatus and procedure for r e p r o d u c i b l e , h i g h - r e s o l u t i o n gas chromatograph ic a n a l y s i s of a l c o h o l i c beverage headspace v o l a t i l e s . J . I n s t . Brew. 83:213-219. W i l l i a m s , A . A . and D. G. Tucknot t (1973) The s e l e c t i v e e x t r a c t i o n of aroma components from a l c o h o l i c d i s t i l l a t e s . J . S c i . F d . A g r i c . 24:863-871. W y l i e , P . L . (1986) Headspace a n a l y s i s w i th c r y o g e n i c f o c u s s i n g : A procedure for i n c r e a s i n g the s e n s i t i v i t y of automated c a p i l a r y headspace a n a l y s i s . C h r o m a t o g r a p h i a . 21:251-258. 98 Appendix 1. GC Data Entry computer program. CLEAR:KEY OFF:CLS:DIM F(150),G(10,150) INPUT "Number of peaks? ",N1 INPUT "Number of samples to be blended? ",NN:PRINT DO K=0 TO NN-1 PRINT "Entry No.";K+l INPUT " Sample No.? (2 d i g i t s ) ",P(K) ENDDO PRINT:PRINT "To keep e n t e r i n g , press ENTER;" PRINT "To reenter for c o r r e c t i n g misentry, press M":PRINT PRINT "Target" INPUT " Int Stand Pk Area ",A DO B$="" B$=INKEY$ ENDDO IF B$="m" B$="M" ENDIF IF B$="M" INPUT " Int Stand Pk Area ",A ENDIF INPUT " Volume ",V DO C$="" C$=INKEY$ ENDDO IF C$="m" C$="M" ENDIF IF C$="M" INPUT " Volume ",V ENDIF PRINT DO 1=1 TO NI PRINT " Peak " ; I ; INPUT " Peak area? ",F(D DO D$="" D$=INKEY$ ENDDO IF D$="m" D$="M" ENDIF IF D$="M" INPUT " Peak area? ",F(I):D$="" ELSE D$ = "" ENDIF ENDDO DO K=0 TO NN-1 PRINT PRINT "Sample";P(K) INPUT " Int Stand Pk Area ",A(K) DO E$="" 99 E$=INKEY$ ENDDO IF E$="m" E$="M" ENDIF IF E$="M" INPUT " Int Stand Pk Area ",A(K):E$ = "" ELSE E$ = "" ENDIF INPUT 11 Volume ",V(K) DO F$="" F$=INKEY$ ENDDO IF F$="m" F$="M" ENDIF IF F$="M" INPUT " Volume ",V(K):F$="" ELSE F$ = "" ENDIF PRINT DO 1=1 TO NI PRINT " Peak " ; I ; INPUT " Peak area? ",G(K,I) DO G$="" G$=INKEY$ ENDDO IF G$="m" G$="M" ENDIF IF G$="M" INPUT " Peak area? ",G(K,I):G$="" ELSE G$ = " " ENDIF ENDDO ENDDO LPRINT TAB(5) "Target"; LPRINT TAB(12) "No."; P(0); LPRINT TAB(19) "No."; P ( l ) ; LPRINT TAB(26) "No."; P(2); LPRINT TAB(33) "No."; P(3); LPRINT TAB(40) "No."; P(4); LPRINT TAB(47) "No."; P(5); LPRINT TAB(54) "No."; P(6); LPRINT. TAB(61) "No."; P(7); LPRINT TAB(68) "No."; P(8) :LPRINT DO 1=1 TO NI LPRINT USING "###"; i ; LPRINT USING "#######";F(I); DO J=0 TO NN-2 LPRINT USING "#######";G(J,I); 100 ENDDO LPRINT USING "#######";G(NN-1,I) ENDDO KEY ONcPRINT PRINT "Store data; diskette ready in Drive B? Press F5 for storing":STOP OPEN "0", #1, "B:DATA" PRINT #1, N1,NN,A,V DO K=0 TO NN-1 PRINT #1, A(K),V(K),P(K) ENDDO DO 1=1 TO NI PRINT #1, F(I) DO J=0 TO NN-1 PRINT #1, G(J,I) ENDDO ENDDO CLOSE #1 PRINT:PRINT "END":END 101 Appendix 2. GC Data Correction computer program. CLEAR:CLS:DIM F(150),G(10,150) PRINT "Recall data; diskette ready in Drive B? Press F5 for recalling":STOP OPEN " I " , #1, "B:DATA" INPUT #1, N1,NN,A,V DO K=0 TO NN-1 INPUT #1, A(K),V(K),P(K) ENDDO DO 1=1 TO NI INPUT #1, F(I) DO J=0 TO NN-1 INPUT #1, G(J,I) ENDDO ENDDO CLOSE #1 DO CLS:PRINT TAB(29) " MENU ":PRINT PRINT TAB(15) "1: Correction 2: Deletion 3: Insertion":PRINT:PRINT INPUT "Menu No.? ",Z IF Z = l PRINT:KEY OFF INPUT "Correct data for target?(Y/N) ", A$ 'Data Correction IF A$="y" A$="Y" ENDIF IF A$="Y" INPUT " How many data to correct? ",B DO 1=1 TO B INPUT " Data No.? ",C PRINT 11 Stored data";F(C) INPUT " Correct data? ",F(C) ENDDO ELSE CLS ENDIF INPUT "How many samples to correct? ",U DO K=l TO U INPUT "Sample No.? ",W DO M=0 TO NN-1 IF W=P(M) L=M ENDIF ENDDO PRINT " Sample ";P(D INPUT " How many data to correct? ",E DO J=l TO E INPUT " Data No.? ",F PRINT " Stored data ";G(L,F) INPUT " Correct data? ",G(L,F) 102 ENDDO ENDDO CLS ELSEIF Z=2 INPUT "Delete data for target?(Y/N) ", D$ 'Data Deletion IF D$="y" D$="Y" ENDIF IF D$="Y" INPUT " How many data to delete? ",B PRINT " If more than one deletion, s t a r t from the bottom data" DO 1=1 TO B INPUT " Data No.? ",E DO I=E TO Nl-1 F(I)=F(I+1) ENDDO ENDDO ENDIF INPUT "How many samples to delete? ",U DO K=l TO U INPUT "Sample No.? ",W DO M=0 TO NN-1 IF W=P(M) L=M ENDIF ENDDO PRINT " Sample No.";P(L) INPUT " How many data to delete? ",B PRINT " If more than one deletion, s t a r t from the bottom data" DO J=l TO B INPUT " Data No.? ",E DO I=E TO Nl-1 G(L,I)=G(L,I+1) ENDDO ENDDO ENDDO CLS ELSE INPUT "Insert data for target?(Y/N) ", 1$ 'Data Insertion IF l$="y" I$="Y" ENDIF IF I$="Y" INPUT " How many data to insert? ",B PRINT " If more than one insertion, s t a r t from the bottom data" DO 1=1 TO B INPUT " Data No.? ",C DO I=N1 TO C+l STEP-1 F(I+1)=F(I) 103 ENDDO INPUT " Data to be inserted? ",F(C+1) ENDDO ENDIF INPUT "How many samples to insert? ",U DO K=l TO U INPUT "Sample No.? ",W DO M=0 TO NN-1 IF W=P(M) L=M ENDIF ENDDO PRINT " Sample No.";P(L) INPUT " How many data to insert? ",E PRINT " If more than one insertion, s t a r t from the bottom data" DO J=l TO E INPUT " Data No.? ",F DO I=N1 TO F+l STEP -1 G(L,1+1)=G(L,I) ENDDO INPUT " Data to be inserted? ",G(L,F+1) ENDDO ENDDO CLS ENDIF INPUT "Correction completed?(Y/N) ",H$ IF H$="y" H$="Y" ENDIF IF H$="Y" INPUT "New number of peaks, i f changed? ",N1 ENDIF ENDDO H$="Y" LPRINT "GC DATA" GOSUB @TTL DO 1=1 TO NI LPRINT USING "###";I; LPRINT USING "#######";F(I); DO J=0 TO NN-2 LPRINT USING " If ######"; G ( J , I ) ; ENDDO LPRINT USING "#######";G(NN-1,I) ENDDO PRINT "Store corrected GC data; diskette ready in Drive B? Press F5":STOP OPEN "O", #1, "B:DATA" PRINT #1, N1,NN,A,V DO K=0 TO NN-1 PRINT #1, A(K),V(K),P(K) ENDDO DO 1=1 TO NI PRINT #1, F(I) DO J=0 TO NN-1 104 PRINT #1, G(J,I) ENDDO ENDDO CLOSE #1 DO 1=1 TO NI F(I)=F(I)/(A*V) DO J = 0 TO NN-1 G(J,I)=G(J,I)/(A(J)*V(J)) ENDDO ENDDO LPRINT:LPRINT:LPRINT "STANDARDIZED DATA" GOSUB @TTL DO J=l TO NI LPRINT USING "###";J; LPRINT USING "####.##";F(J); DO 1=0 TO NN-2 LPRINT USING "####.##";G(I,J); ENDDO LPRINT USING "####.##";G(NN-1,J) ENDDO KEY ON:PRINT PRINT "Store standardized data; diskette ready in Drive B? Press F5":STOP OPEN "0", #2, "B:DATA1" PRINT #2, NI,NN DO K=0 TO NN-1 PRINT #2, P(K) ENDDO DO 1=1 TO NI PRINT #2, F(I) DO J=0 TO NN-1 PRINT #2, G(J,I) ENDDO ENDDO CLOSE #2 PRINT:PRINT "END":END @TTL LPRINT TAB(5) "Target"; LPRINT TAB(12) "No. it ;P(0) LPRINT TAB(19) "No. II ;P(D LPRINT TAB(26) "No. II ;P(2) LPRINT TAB(33) "No. ti ;P(3) LPRINT TAB(40) "No. II ;P(4) LPRINT TAB(47) "No. II ;P(5) LPRINT TAB(54) "No. II ;P(6) LPRINT TAB(61) "No. II ;P(7) LPRINT TAB(68) "No. II ;P(8) RETURN LPRINT TAB(47) "No. ";P(5) LPRINT T 105 Appendix 3. S i m i l a r i t y Constant computer program. CLEAR:KEY OFF:CLS:DIM F(150),G(10,150) PRINT " R e c a l l data; d i s k e t t e ready i n Drive B? Press F5 for recalling":STOP OPEN " I " , #2, "B:DATA1" INPUT #2, N1,NN DO K=0 TO NN-1 INPUT #2, P(K) ENDDO DO 1=1 TO NI INPUT #2, F( I ) DO J=0 TO NN-1 INPUT #2, G(J,I) ENDDO ENDDO CLOSE #2 CLS .-PRINT "SIMILARITY CONSTANT" :PRINT LPRINT "SIMILARITY CONSTANT":LPRINT DO 1=0 TO NN-1 A=0:B=0:C=0 DO J=l TO NI A=A+F(J)*G(I,J) B=B+F(J)*F(J) C=C+G(I /J)*G(I,J) ENDDO R=A/SQR(B*C) PRINT " Sample No. " ; P ( I ) ; PRINT TAB(30) USING "### . ###";R LPRINT " Sample No. ";P(D; LPRINT TAB(30) USING "###•. # # # " ; R ENDDO PRINT:PRINT "END":END 106 Appendix 4. Blending Optimization computer program. CLEAR:CLS DIM X(10 /100) /Y(100) /F(150),G(10 #150) /H(150) /D(10 #150) PRINT " R e c a l l data; d i s k e t t e ready i n Drive B? Press F5 f o r recalling":STOP OPEN " I " , #2, "B:DATA1" INPUT #2, N1,NN DO K=0 TO NN-1 INPUT #2, P(K) ENDDO DO 1=1 TO NI INPUT #2, F ( I ) DO J=0 TO NN-1 INPUT #2, G(J,I) ENDDO ENDDO CLOSE #2 DO 1=1 TO NI DO J=0 TO NN-1 D(J,I)=G(J,I) ENDDO ENDDO MM=NN DO CLS:INPUT "Sample No. of the p r i n c i p a l i n g r e d i e n t ? ",T DO 1=0 TO MM-1 IF T=P(I) L = I ENDIF ENDDO DO J=l TO NI G(0 /J)=D(L /J) ENDDO INPUT "How many samples f o r blending with the p r i n c i p a l i n g r e d i e n t ? ";A DO 1=1 TO A INPUT " Enter sample No. "/Ad) DO J=0 TO MM-1 IF A(I)=P(J) B(I)=J ENDIF ENDDO ENDDO DO 1=1 TO NI DO J=l TO A K=B(J) G(J/I-)=D(K/I) ENDDO ENDDO NN=A:PRINT:MV=100 INPUT "Terminating d i f f e r e n c e value? ",TERM INPUT "How many v e r t i c e s without 107 p r o h l b l t - t r e s p a s s l n g ? ",Z:PRINT DO 1=1 TO NN PRINT "Sample No. ";A(I) INPUT " Enter lower then upper l i m i t s " , L ( I ) , U ( I ) ENDDO LPRINT " V e r t i c e s without p r o h i b i t - t r e s p a s s i n g ";Z LPRINT "Terminating d i f f e r e n c e value";USING "###.####"; TERM:LPRINT LPRINT " P r i n c i p a l i ngredient i s sample ";T LPRINT "Lower and upper limits":LPRINT " LL: "; DO J=l TO NN LPRINT USING "###.###";L(J); ENDDO LPRINT:LPRINT " UL: "; DO K=l TO NN LPRINT USING "###.###";U(K); ENDDO LPRINT:LPRINT P=(1/(NN*SQR(2)))*(NN-1+SQR(NN+1)) Q=(1/(NN*SQR(2)))*(SQR(NN+1)-1) DO J=l TO NN ' I n i t i a l simplex M(l,J)=L(J) ENDDO DO 1=2 TO NN+1 DO J=l TO NN IF I-1=J M(I,J)=L(J)+P*(U(J)-L(J)) ELSE M(I /J)=L(J)+Q*(U(J)-L(J)) ENDIF ENDDO ENDDO DO M=l TO NN+1 DO J=l TO NN S(J)=M(M, J) ENDDO GOSUB @FTN B(M)=R ENDDO DO XX=1 TO NN+1 DO 1=1 TO NN X(I /XX)=M(XX /I) ENDDO ENDDO DO Y = l TO NN+1 Y(Y) = =B(Y) ENDDO LPRINT TAB(12) "No. II ; A ( l ) LPRINT TAB(19) "No. it ;A(2) LPRINT TAB(26) "No. II ;A(3) LPRINT TAB(33) "No. II ;A(4) LPRINT TAB(40) "No. II ;A(5) LPRINT TAB(47) "No . II ;A(6) LPRINT TAB(54) "No. II ;A(7) 108 'Find WORST 'Find BEST LPRINT TAB(61) "No.";A(8); LPRINT TAB(70) "Response":XX=XX-1:Y=Y-1 LPRINT:LPRINT TAB(12) " ( I n i t i a l s i mplex)" DO J=l TO XX LPRINT "Vertex ";USING "###";J; DO K=l TO NN LPRINT USING "###.###";X(K,J); ENDDO LPRINT TAB(66) USING " ###.###";Y(J) ENDDO LPRINT SEARCH WORST=B(l):WL=1 DO 1=2 TO NN + 1 IF B(I)<WORST WORST=B(I):WL=I ENDIF ENDDO BEST=B(1):BL=1 DO 3 = 2 TO NN+1 IF B(J)>BEST BEST=B(J):BL=J ENDIF ENDDO T=0 DO 1=1 TO NN+1 T=T+B(I) ENDDO NXT=(T-WORST-BEST)/(NN-1) DO K=l TO NN S = 0 DO L=l TO NN+1 S=S+M(L/K) ENDDO S=S-M(WL/K):N(K)=S/NN ENDDO C=1:C$="(Reflection)":GOSUB @SRC DO M=l TO NN R(M)=S(M) ENDDO REFL=R IF REFL>BEST C=2:C$="(Expansion)":GOSUB @SRC IF R>REFL DO N=l TO NN Q(N)=S(N) ENDDO GOSUB @WRPL ELSE DO 1=1 TO NN Q(I)=R(I) ENDDO R=REFL GOSUB @WRPL •Compute NEXT to the worst Centroid ' R e f l e c t i o n 'Expansion 109 ENDIF ELSEIF REFL>NXT DO J=l TO NN Q(J)=R(J) ENDDO R=REFL GOSUB @WRPL ELSEIF REFL>WORST C=0.5:C$="(Contraction-R)":GOSUB @SRC 'Contraction-R IF R>REFL DO 1=1 TO NN Q(I)=S(I) ENDDO GOSUB @WRPL ELSE C=0.25:C$="(Massive contrac-t i o n - ^ ) " :GOSUB @SRC 'Massive contrn.-R IF R<REFL DO K=l TO NN Q(K)=R(K) ENDDO R=REFL GOSUB @WRPL ELSE DO L=l TO NN Q(L)=S(L) ENDDO GOSUB @WRPL ENDIF ENDIF ELSE 'Contraction-W C=-0.5:C$="(Contraction-W)":GOSUB @SRC IF R>WORST DO J=l TO NN Q(J)=S(J) ENDDO GOSUB @WRPL ELSE C=-0.25:C$="(Massive contrac-tion-W)":GOSUB @SRC 'Massive contrn.-W DO K=l TO NN Q(K)=S(K) ENDDO GOSUB @WRPL ENDIF ENDIF EXITIF XX>MV ORELSE A=XX:B=XX-l:C=XX-2 'Termination IF ABS(Y(A)-Y(B))>TERM T$="N" ELSEIF ABS(Y(B)-Y(C))>TERM T$="N" ELSEIF ABS(Y(A)-Y(C))>TERM 110 T$="N" ELSE T$="Y" ENDIF ENDLOOP T$="Y" DO 1=1 TO NN 'Average of l a s t three AV(I)=(X(I,A)+X(I,B)+X(I,C))/3 ENDDO BV=(Y(A)+Y(B)+Y(C))/3:LPRINT LPRINT " F i n a l average values":LPRINT " "; DO 1=1 TO NN LPRINT USING "###.###";AV(I) ; ENDDO LPRINT TAB(66) USING "###.###";BV ENDSRCH PRINT:PRINT:PRINT INPUT "Another combination of in g r e d i e n t s for blending?(Y/N) ",H$ IF H$="n" H$="N" ENDIF ENDDO H$="N" PRINT "END":END @SRC 'New vertex DO 1=1 TO NN S(I)=N(I)+C*(N(I)-M(WL,I)) ENDDO IF XX>Z-1 DO J=l TO NN ' P r o h i b i t t r e s p a s s i n g IF S ( J X L ( J ) S(J)=L(J) ELSEIF S(J)>U(J) S(J)=U(J) ENDIF ENDDO ENDIF GOSUB @FTN Y=Y+1:XX=XX+1 Y(Y)=R DO J=l TO NN X(J,XX)=S(J) ENDDO LPRINT "Vertex ";USING ";XX; LPRINT C$:LPRINT " "; DO 1=1 TO NN LPRINT USING "###.###";X(I,XX); ENDDO LPRINT TAB(66) USING " ###.###";Y(Y) RETURN @WRPL 'W r e p l a c i n g B(WL)=R DO 1=1 TO NN M(WL /I)=Q(I) ENDDO 111 RETURN @FTN 'Function DO L=l TO NI H(L)=0 ENDDO DO 1=1 TO NI DO J=l TO NN H(I)=H(I)+S(J)*G(J,I) ENDDO H(I)=H(I)+G(0,I) ENDDO A=0:B=0:C=0 DO K=l TO NI A=A+F(K)*H(K) B=B+F(K)*F(K) C=C+H(K)*H(K) ENDDO R=A/SQR(B*C) RETURN H(I)+G(0,I) ENDDO A=0:B=0:C=0 DO K=l TO NI A=A+F(K)*H(K) B=B+F(K)*F(K) C=C+H(K)*H(K) ENDDO R=A/SQR(B*C) 112 Appendix 5. S i m i l a r i t y Constant of Blend computer program. CLEAR:KEY OFF:CLS:DIM F(150),G(10,150) INPUT " R e c a l l data (Fresh) from diskette?(Y/N) ",A$ IF A$="y" A$="Y" ENDIF IF A$="Y" PRINT:KEY ON PRINT "Diskette ready i n Drive B? Press F5 f o r recalling":STOP OPEN " I " , #1, "B:DATA" INPUT #1, N1,NN,A,V DO K=0 TO NN INPUT #1, A(K),V(K),P(K) ENDDO DO 1=1 TO NI INPUT #1, F ( I ) DO J=0 TO NN INPUT #1, G(J,I) ENDDO ENDDO CLOSE #1 ELSE CLS-.INPUT "Number of peaks? ",N1 PRINT "Fresh" INPUT " Int St Pk Area ",A DO B$ = '»" B$=INKEY$ ENDDO IF B$="m" B$="M" ENDIF IF B$="M" INPUT " Int St Pk Area ",A ENDIF INPUT " Volume ",V DO C$="" C$=INKEY$ ENDDO IF C$="m" C$="M" ENDIF IF C$="M" INPUT " Volume ",V ELSE CLS ENDIF PRINT DO 1=1 TO NI PRINT "Peak 11; I; INPUT " Peak area? ",F(I) DO D$="" 113 D$=INKEY$ ENDDO IF D$="m" D$="M" ENDIF IF D$="M" INPUT " ELSE D$ = "" ENDIF ENDDO ENDIF KEY OFF: PRINT:INPUT "Number DO K=l TO NN INPUT " Pattern t i t l e ? " INPUT " Int St Pk Area DO E$="" E$=INKEY$ ENDDO IF E$="m" E$="M" ENDIF IF E$="M" INPUT " Int St Pk Area ELSE E$ = " " ENDIF INPUT 11 Volume DO F$="" F$=INKEY$ ENDDO IF F$="m" F$="M" ENDIF IF F$="M" INPUT " Volume ELSE F$ = " " ENDIF PRINT DO 1=1 TO NI PRINT "Peak INPUT " Peak area? DO G$="" G$=INKEY$ ENDDO IF G$="m" G$="M" ENDIF IF G$="M" INPUT " ELSE G$ = "11 ENDIF Peak area? ",F(I):D$="" of GLC patterns? ",NN P$(K) ",A(K) ",A(K):E$="" ",V(K) ",V(K):F$="" ,G(K,I) Peak area? ",G(K,I):G$="" 114 ENDDO ENDDO LPRINT TAB(9) "Fresh"; LPRINT TAB(17) LPRINT TAB(27) LPRINT TAB(37) LPRINT TAB(47) LPRINT TAB(57) P$(5) DO 1=1 TO NI LPRINT USING "###" P $ ( D P$(2) P$(3) P$(4) LPRINT i ; LPRINT USING "##########";F(I); G (1,I);G( 2 ,I); G(3,I);G(4,I);G(5,I) ENDDO PRINT:PRINT "Store data; diskette ready in Drive Press F5 for storing":STOP OPEN "0", #2, "B:DATA2" PRINT #2, N1,NN,A,V DO K=l TO NN PRINT #2, A(K),V(K),P$(K) ENDDO DO 1=1 TO NI PRINT #2, F(I) DO J=l TO NN PRINT #2, G(J,I) ENDDO ENDDO CLOSE #2 PRINT:PRINT "END":END 115 Appendix 6. S i m i l a r i t y Constant of Blend, Data C o r r e c t i o n computer program. CLEAR:CLS:DIM F{150),G(10,150) PRINT " R e c a l l data; d i s k e t t e ready i n Drive B? Press F5 for r e c a l l ing"-.STOP OPEN " I " , #2, "B:DATA2" INPUT #2, N1,NN,A,V DO K=l TO NN INPUT #2, A(K),V(K),P$(K) ENDDO DO 1=1 TO NI INPUT #2, F ( I ) DO J=l TO NN INPUT #2, G ( J # I ) ENDDO ENDDO CLOSE #2 PRINT:KEY OFF:INPUT "Correct data f o r fresh?(Y/N) A$ 'Data C o r r e c t i o n IF A$="y" A$="Y" ENDIF IF A$="Y" INPUT "How many data to c o r r e c t ? ",B DO 1=1 TO B INPUT "Data No.? ",A INPUT "Correct data? ",F(A) ENDDO ELSE CLS ENDIF INPUT "Correct data for GLC pattern?(Y/N) ",E$ IF E$="y" E$ = "Y" ENDIF IF E$="Y" DO L=l TO NN PRINT "PATTERN: ";P$(D INPUT "How many data to c o r r e c t ? ",E DO K=l TO E INPUT "Data No.? ",F INPUT "Correct data? ",G(L,F) ENDDO ENDDO ELSE CLS ENDIF INPUT "Delete data for fresh?(Y/N) ", D$ 'Data D e l e t i o n IF D$="y" D$="Y" ENDIF 116 IF D$="Y" INPUT "How many data to delete? ",B PRINT "If more than one deletion, s t a r t from the bottom" DO 1=1 TO B INPUT "Data No.? ",E DO I=E TO Nl-1 F(I)=F(I+1) ENDDO ENDDO ELSE CLS ENDIF INPUT "Delete data for GLC pattern?(Y/N) ",D$ IF D$="y" D$="Y" ENDIF IF D$="Y" DO L=l TO NN PRINT "PATTERN: ";P$(L) INPUT "How many data to delete? ",B PRINT "If more than one deletion, s t a r t from the bottom" DO 1=1 TO B INPUT "Data No.? ",E DO I=E TO Nl-1 G(J,I)=G(J,I+1) ENDDO ENDDO ENDDO ELSE CLS ENDIF INPUT "Insert data for fresh?(Y/N) ", 1$ 'Data Insertion IF I$="y" I$="Y" ENDIF IF I$="Y" INPUT "How many data to insert? ",B PRINT "If more than one insertion, s t a r t from the bottom" DO 1=1 TO B INPUT "Data No.? ",C DO I=N1 TO C+l STEP-1 F(I+1)=F(I) ENDDO INPUT "Data to be inserted? ",F(C+1) ENDDO ELSE CLS ENDIF INPUT "Insert data for GLC pattern?(Y/N) ",I$ IF I$="y" I$="Y" ENDIF IF I$="Y" 117 DO L=l TO NN PRINT "PATTERN: ";P$(L) INPUT "How many data to insert? ",E PRINT "If more than one insertion, s t a r t from the DO K=l TO E INPUT "Data No.? ",F DO I=N1 TO F+l STEP -1 G(L,I+1)=G(L,I) ENDDO INPUT "Data to be inserted? ",G(L,F+1) ENDDO ENDDO ELSE CLS ENDIF INPUT "New number of peaks, i f changed? ",N1 LPRINT "GLC DATA" LPRINT TAB(9) "Fresh"; LPRINT TAB(19) P$(l); LPRINT TAB(29) P$(2); LPRINT TAB(39) P$(3); LPRINT TAB(49) P$(4); LPRINT TAB(59) P$(5):LPRINT DO 1=1 TO NI LPRINT USING "###"; I; LPRINT USING "##########";F(I);G(1,I);G(2,I);G(3,I); G(4,I);G(5,I) ENDDO DO 1=1 TO NI F(I)=100*F(I)/(A*V) DO J=l TO NN ' G(J, I)=100*G(J,I)/(A(J)*V(J)) ENDDO ENDDO LPRINT: LPRINT-.LPRINT "STANDARDIZED DATA" DO J=l TO NI LPRINT USING "###";J; LPRINT USING "######.###";F(J);G(1,J);G(2,J);G(3,J); G(4,J);G(5,J) ENDDO LPRINT:LPRINT:LPRINT "SIMILARITY CONSTANT" DO 1=1 TO NN A=0:B=0:C=0 DO K=l TO NI A=A+F(K)*G(I,K) B=B+F(K)*F(K) C=C+G(I,K)*G(I,K) ENDDO R=A/SQR(B*C) LPRINT " ";P$(I); LPRINT TAB(30) USING "###.###";R ENDDO PRINT:PRINT "END":END 118 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098116/manifest

Comment

Related Items