2, r 2 . awn-«ch-Iw.‘ Vnmmvm \‘K " "“ . , :2 --~,\;*:.r279m“‘2'e‘1‘ *‘g WK}: ‘2 '— 3‘7." ‘ low fly?" r“ ' Jr 1 I! fgr '- I (I I r v a ‘l n V: r "'1" ~.:: V (v a )3" I . %; W“ . ‘ V . 2 P;- r‘ . - . ”- if, :.I\)_€_\.:.:-V‘Z’ A?“ ui" ' ' v . . . ‘ ,2 ’1‘ . ' .. -‘ . 2 ‘ ~ 3 -, V _ . . ,J . .. a- q. {rm ”mg 1; 1‘“? 1‘. "of“; {w} * 2:53. szqwé " ' EEC-‘9‘!” ‘u‘ : ‘ \ i if; . V x .u e.. A Ava-n; .t. . . . .3, 312-. r .. ‘39 .‘L {6 )‘ ' {u'v‘ y. . :74 2 v“ _ .9 2‘, ’9‘? V ‘ ”'15: ’53:») :3‘51‘ ‘15};qu f; as , KN “4‘— X r . $4“ v'e my he. ' ' ' . r . :1- , . - -:* _ . . >' 2' . . 421. _ . . 9: 91’1”” ‘ J “ 2‘? ' 5‘ ,i’3' ' 65“" /‘ I‘V- -' . 112‘ ' Va?" ’5’“: 2 ' . >3! *1 #3:: 'iwflfl‘r- ’ "I. («42): ~ ,. . ah“ . ~ ; r > . , a‘w‘ 3, ,, up? '. You 9"“;‘9'9’3 \- {mixl-v. f . n ,2”; a’m-N' ‘ " p r ‘2 ’E'nfii‘fwx.» :23): if? 3111?")- ;- .‘?..’*'_',' gum, )3‘ 7 " - 2:... 2}}: WC 513' A“) V‘- 3 [QM/r? . .39. 31%;} {if}, 2"" I . d _ W»,)Q'€., '4‘?) ,7 .. ‘2 n' if 2 ;L7u ”Wfif’é‘”%, f * #35; § A5!” 2,; It? a: v- w: , Fr .2 '12 '2” 9‘2 $23!, n ,X 'D u- !I:r"I‘o;:-:’ 83%“ ya? 3" ”6:95 “If“; n’ “2'. w’fl'," "mug” “rig” 2'1”.» n'i'igylfi, if; "r2 HUI I "21‘"??? 31);}; r we}?! - - p~.>. .-’-'A' r' u,- ”f? 5.24:; :7 " WV” 3‘” 9535i" 7?}??? 24"”5‘m'fég ‘m '9: ,. "" ) ‘ ’3‘ $4.)”. - ‘ ; ~ ' 2-. g CW"?! ’2 F' - 22329251 ’ ; ’5’" ’. f ’iazzéia’r" ‘53 z? fié’fwfi, 2995: 2*: ,Jpgyrgnfigéyfifi, ,, :97?" . . 29 I'm? .45 ‘3 :EEXC ." §‘ 214%“ Eris-r ’3! fag]: {5:5, I y. I. Mr"? {9/ r ’r’e’i’, '.‘ ,2 ~,A:, ‘r’r [iiw g 13 ‘ \ q ‘x .. , ,9)’ ,1: " nu Irma-’0 tr ‘3‘: {tyre-v77:- $55375 , 7:933 g'r’ £1“; , :9 .2”? "f a (’3' ml}; ‘ :1‘ 2‘ .1 ‘ I: N” 3: if V yifixh’gw) 5:302 ,,, A457» IRE: Vibr’r "2:. $6.. 2.; «r 1'” y“ 0.1:, :V'” r r r r '"' ‘ ' ‘ . u ‘ If ,, _ ‘- F Cfirx’z [’253: 7‘ a n ,. e" . 1.3!”, .. A 2 ; . “r!" V, . r - AW - ,. a ’. h":'}'l"r{'”l""’nr 7“ Hr; , ,_ r . ;. . . W} '1 . 'H t r 2 v , J' rrr r envy; I, V mi'e'gr: . n“ ",2. ,., 2 J «r I’l’lv-‘f ;{Inlnm » .,, . '3 r , \ mi" 0 (‘1‘? wet-gr; . r 3.5.»! III: 5""? ,{réf’ u' w. . r ’7 . ’2’: ,9), .1 .1; . ; PIX-:39: [£4,192 ’;:;f1’r;;/u Lu)”: zn , - f g . 2-" , 59.720!” ”#50:: .5; 5&7; ,3; 72 . 3x43 r . r 7 ‘5’ f u" ’o ' '1 ‘5” . ->’; 1.1%”! 8,4779% 1;- I" a” ’4'} fl 2’ V15}; ; 9’; 1;" ‘ P_ ”If, fl’lp'li.’ in, ”K :33, e w‘ wan: 5;. - I. r!" ' . «W» 1.; .91. 7:7. :79 ' ”I .17; :9}! :1. x' 257/! ,5. It .> > :17? f:fafi"’ 7’3?! . ‘1 V a. ‘ V #2,, , 9: xx" {in 2 ." ,n'.’ {r t - f r J) ’A':’ . n- 4' 7”,” :n’} ; r" H I; '5'!" ”-9, a 7 ’1‘; 2 I "f ; ’3'. {-1_ if]. ’{;:; it if], fi/m ’I/fll‘rflrr 7'}, , a. n' ra-fif" . £23.". ‘ff': 1:43., if?” 9. ,.-., r» '99" 2y“; 9;»; ' H) ’v" u..vp.u-l I, r, fg>:7f~’fy'( h ‘ I" f ,. . Aef'r’r /{ V'r'fl'afni’Er'fif V'IJ'I” {mars-g, xvi; ”1-,”. wt. 9-29-29 ’~ - '* ' ,2! ', '44-; 29: ,2" 9 ., , ' :9’" , . w, ., 9 9 {yr/.1::". ii“;- 53y I, _- I it! I" {’0‘; .r, l// . I; rig/:5" a" £1}. géf': A"; n '7’9‘5’rvfift‘lrnex’gar ' :‘ffié n’ :- 1192A"; o’fl'. n.Vl4.tl.>~a—‘.u2 . . I a “F“ C. 3-: , a. lIBRARY Michigan State University This is to certify that the thesis entitled Contact Nucleation From Aqueous Dextrose Solutions presented by Ponnampalam Elankovan has been accepted towards fulfillment of the requirements for Agricultural Engineering M' S ’ degree in Major professor Date W/l{ flay; 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution bV1ESI_} RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from w your record. FINES will - be charged if book is returned after the date stamped below. CONTACT NUCLEATION FROM AQUEOUS DEXTROSE SOLUTIONS BY Ponnampalam Elankovan A THESIS Submitted to Michigan State University in partial fulfillement of the requirements for the degree of MASTER OF SCIENCE in Agricultural Engineering Department of Agricultural Engineering 1987 Abstract CONTACT NUCLEATION FROM AQUEOUS DEXTROSE SOLUTIONS by Ponnampalam Elankovan This work is presented as two papers. The first, "Technique for Obtaining Raman Spectra of Contact Nuclei lg gitg" was reviewed and published as a spectroscopic technique in Applied Spectroscopy, volume 40, number 5, 1986. The second, "Contact Nucleation from Aqueous Dextrose Solutions," has been peer-reviewed and accepted for publication in the American Institute of Chemical Engineers Journal. The first paper demonstrates that the crystallographic phase of contact nuclei can be identified in sign with the use of Raman Spectroscopy. These results indicate that the laser Raman miocroprobe technique can be an important tool in the future study of contact nucleation. The second paper is focussed on the contact nucleation process in dextrose crystallization. Although the precise mechanism_is not known, contact nucleation is the primary source of new particles in many important industrial crystallization. Centact nuclei of both alpha monohydrate and alpha anhydrous dextrose were formed from the contact of an anhydrous parent crystals growing in pure solution indicating that simple breakage does not cause contact nucleation. Since a small amount of data is available in the literature for this important sugar, and the Raman microprobe technique is a new technique to characterize the growing crystal phase in sit , this work provides the background for subsequent studies. A summary of results, recommendations for future work, and appendix containing the complete raw data obtained during this study are presented after the second paper. DEDICATION This work is dedicated to my loving wife Sugendrini Ponnampalam who has been a great source of inspiration and encouragement through my academic career. ACKNOWLEDGEMENTS I would like to thank Dr. Kris Berglund, my major advisor, for his guidance and valuable advice during this Istudy. Acknowledgement is given to the author's co-workers Michel Cerreta, Lie Ding Shiau and Yi Ding Chu for their cooperation and discussion on the field of crystallization. The assistance of Dr. David Bruce of the department of Physics and Astronomy at Michigan State University is appreciated. Finally, the financial support from A.E. Staley Manufacturing Company and the laser Raman microprobe provided through the National Science Foundation under grant number CPE - 8409458 and the Michigan Agricultural Experiment Station are greatfully acknowledged. ii TABLE OF CONTENTS Page LISTOFTABLESOO ....... .OOOOOOOOO...OOOOOOOO..O..OOOOOOOO v LIST OF FIGURES ......................................... vi PAPER ONE. "Technique for Obtaining Raman Spectra of Contact Nuc1eimSitu".OOOOOOOOOOOOOOOOO00......OOOOOOOOOOOOOOO l PAPER TWO . "Contact Nucleation From Aqueous Dextrose Solution"..... 13 CONCLUSIONSOOOO...OOOOOOOOOOOOOOOOOO0.000......0.0.0.... 43 SUGGESTIONS FOR FUTURE WORK .................. .... ....... 44 APPENDICES Appendix A. Raw data: crystal size versus time........ 46 Appendix B. Apparent initial size and growth rate of contact nuclei of dextrose................ 57 Appendix C1. Physical properties of D-glucose........ 61 C2. Conditions of experimental runs and . number of nuclei analyzed............... 62 C3. Mean growth rate, variance of growth rate distribution and relative supersaturation at different conditions. 63 C4. Linear regression of growth rate versus apparent initial size................... 64 Appendix D1. Examples of size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 2 degrees of supercooling at 319K..................... 65 D2. Examples of size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 3 degrees of supercooling at 319K ..... .. ......... ....... ...... ........ 66 D3. Examples of size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 4 degrees of supercooling at 319K..................................... 67 iii D4. Examples of size versus time for alpha Appendix E. anhydrous dextrose contact nuclei fermed and grown at 3 degrees of supercooloing at 323K000.0.00000000000000000000000000000000 68 Materials and methods... .................. 69 iv LIST OF TABLES Table 1. Physical properties of D-glucose .......... ...... 2. Conditions of experimental runs and number of nuClei analyzedOOOOOOO0000......00.000.000.000... 3. Mean growth rate, variance of growth rate distribution and relative supersaturation of dextrose contact nuclei at various experimental conditions....................................... 4. Linear regression of growth rate versus apparent initial SizeOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.000... Page 61 62 63 64 LIST OF FIGURES Figure Page 1. Schematic diagram of nucleation cell............. 9, 32 2. Typical examples of photomicrographs of contact 10. nuclei of dextrose formed from pure solution at 314K and supercooling of 4K. The elapsed time between photographs is 72 min. Note the two different crystal forms present.................. 10, 33 Raman shift of alpha monohydrate dextrose. (A) pure reagent grade, (B) parent (needle) crystal, (C) contact nuclei in SLED at experimental conditions favorable to form two different phases, and (D) solution phase spectrum.................. 11, 34 Raman shift of alpha anhydrous dextrose. (A) pure reagent grade, (B) parent (prism) crystal, (C) contact nuclei in sign at experimental conditions favorable to form to different phases, and (D) solution phase spectrum.......................... 12, 35 Size versus time for alpha monohydrate dextrose contact nuclei in the dextrose-water system formed and grown at 3 degrees supercooling at 314K. Each line for an individual crystal............... 36 Size versus time for alpha monohydrate dextrose contact nuclei in the dextrose formed and grown at 3 degrees supercooling at 323K. Each line for an individual crystal............................. 37 Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 314K.. 38 Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 319K.. 39 Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 323K.. 40 Mean growth rate versus relative supersaturation for anhydrous dextrose contact nuclei formed and grown atvarious temperatureSOOOOooo0.0.0.000.0.0.000... 41 vi ll. 12. 13. 14. 15. Variance of growth rate distribution versus mean growth rate for dextrose contact nuclei formed and grown at various temperatures and supercoolings... ..... ....... ..... ................. 42 Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 2 degrees supercooling at 319K. Each line is for an individual crystal ........ . ....................... 65 Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 3 degrees supercooling at 319K. Each line is for an indiVidua-l Crystal-00......OOOOOOOOOOOOOOOOOOOOOOOO 66 Size versus time for alpha anhydrous dextrose contact nuclein formed and grown at 4 degrees supercooling at 319K. Each line is for an individual crystal......... ...... ....... .......... 67 Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 3 degrees supercooling at 323K. Each line is for an indiVidual Crystal.’ ..... OOOOOOOOOOOOOOOOOOOOOOOOO 68 vii Technique for Obtaining Raman Spectra of Contact Nuclei Ln Situ BY P.Elankovan and K.A.Berglund* Departments of Agricultural Engineering and Chemical Engineering Michigan State University East Lansing, MI 48824 , *Auther to whom correspondence should be addressed Published in Applied Spectroscopy, Volume 40, Number 5, pages 712-714, 1986 Introduction Secondary nucleation, the formation of new crystals due to the prior presence of other growing crystals, is the primary source of new particles in most industrial crystallizers (Larson, 1984). Of the various types of secondary nucleation possible, contact nucleation, wherein a disturbance of growing crystal surface results in nuclei,is thought to be the most prevalent (Larson, 1984). The source of contact nuclei has not been determined; however, two possibilities has been presented in the literature. In the first it is proposed that nuclei come from microattrition of the growing crystal surface. Conversely, in the second the source of the nuclei is thought to be a semiordered layer of solute adjacent to a growing crystal surface. In order to distinguish between the two models it is necessary to produce labeled contact nuclei and to analyze them as to their source. To this end a contact nucleation cell has been developed and used in the dextrOSe water system. Apparatus The cell used in the present study is shown in Figure 1. The upper chamber of the cell has temperature control provided by water circulated in the lower chamber of the cell. Since a temperature gradient is probably present in the cell, a glass cover slip was mounted in the upper chamber to hold the contact nuclei at a precise distance from the water bath chamber. A thermistor allows careful monitoring of the temperature of the solution in this position and variations are are typically less than 0.1K. Contact nuclei were analyzed using a laser Raman microprobe i_ situ. The instrument used was a Spex Model 1406 double monochromator with radiation (1545A and 1W ) provided by a Coherent Radiation CR-S argon ion laser. The sample was illuminated by a Spex 1482 Micramate illuminator using a Ealing Corporation Model 25-0506 15x long working distance objective. Spectra were recorded using a strip chart recorder and the scan speed was 0.05A/seconds. Experimental The dextrose-water system was the subject of the study. An interesting feature of this system is that at approximately 308-323K it undergoes a phase transition (Dean, 1974 and Edwards, 1982). That is, above this range alpha monohydrate dextrose is formed. This feature was exploited in the present study. Contact nucleation experiments were performemd using the same photomicroscopic technique as applied to the citric acid monohydrate system (Berglund and Larson, 1982). Parent crystals of the both alpha anhydrous dextrose and alpha monohydrate dextrose were grown to approximately 1-2mm. The alpha monohydrate parent crystals were not used in contact experiments due to their fragile nature. In each experiment an alpha anhydrous parent crystal was mounted to the moveable rod shown in Figure 1. After parent crystal was prepared and mounted, a dextrose solution saturated at 318K was added to the preheated cell. The solution was then heated to redissolve any primary nuclei which may have been generated during transfer. Heating also slightly dissolved the parent crystal to insure that no surface irregularities formed by washing and drying remained. The solution was then cooled to 314K; then the parent crystal was growing in a regular manner, it was contacted by gently sliding across the glass cover slip. ReSults and Discussion Two different crystal habits were formed by contact as evidenced by the photomicrographs in Figure 2. For convenience of discussion the well formed nuclei are designated prisms and the elongated nuclei are designated needles. An anhydros crystal (as verified by Raman I spectroscopy) was contacted under conditions in whichboth phases may form. The laser Raman microprobe was used to identify the two types of crystals i_ situ. A summary of the Raman spectra are presented in Figures 3 and 4. Figures 3A and 4A show the spectra for reagent grade alpha monohydrate and alpha anhydrous dextrose, respectively. Figures 3B and 4B are spectra of parent crystals of alpha monohydrate and alpha anhydrous, respectively, and Closely conform to the reagent grade spectra in Figures 3A and 4B. (It should be remembered that the alpha monohydrate parent crystal was not used in any contact experiment and is only presented here for the sake of completeness.) The spectra shown in Figures 30 and 4D are of the supersaturated solution in the cell far from any contact nuclei. Figures 3C and 4C are the in git; spectra of the needle and prism contact nuclei, respectively. Clearly large differences are evident in the Raman spectrum of each type of nuclei. The most proniunced 1 differences occur in the Raman shifts at about 918cm- and 1078cm-l which are assigned to the C -H strech and the C-O-H 1 deformation, respectively (Vasko gt a1, 1972). Upon comparison of the contact nuclei spectra with the reagent grade dextrose and parent crystal spectra, it is clear that the prism contact nuclei are alpha anhydrous dextrose. On the other hand, the identifiCation of the needle crystals is slightly less definitive. The predominant phase present in alpha monohydrate dextrose, but the additional band at 942cm.1 suggests another phase may also be present (Dean, 1974). At any rate, the two habits present corresponds to different crystalline phases. The present results support the notion that the semiordered solution adjacent to a growing crystal is the source of contact nuclei. Since an anhydrous parent crystal was used in the contact and two different phases were formed, microattrition of the parent crystal does not explain the formation of the monohydrate phase. Summary It has been demonstrated that contact nuclei can be identified in sign using Raman spectroscopy. The present results suggest that the laser raman microprobe can be an important tool in the future of contact nucleation. Acknowledgments This work was supported in part by the A. E. Staley Manufacturing Company. The laser Raman microprobe was provided jointly by the National Science Foundation under grant number CPR—8409458 and the Michigan Agricultural Experiment Station. The assistance of Dr. David Bruce of the department of Physics and Astronomy at Michigan State University is appreciated. Figure Figure Figure Figure Schematic diagram of nucleation cell (1) chamber containing solution; (2) parent crystal; (3) glass cover slip where parent crystal is slid: (4) support rods for glass cover slip; (5) thermistor; (6) movable rod holding parent crystal; (7) chamber containing constant , temperature water; and (8) water inlet and outlet: Typical examples of photomicrographs of contact nuclei of dextrose formed from pure solution at 314K and a supercooling of 4K. The elapsed time between photographs is 72 min. Note the two different crystals forms present. Raman shift of alpha monohydrate dextrose. (A) pure reagent grade, (B) parent (needle) crystal, (C) contact nuclei in git; at experimental conditions favorable to form two different phases, and (D) solution phase spectrum. Raman shift of alpha anhydrous dextrose. (A) pure reagent grade, (B) parent (prism) crystal, (C) contact nuclei in sign at experimental conditions favorable to form two different phases, and (D) solution phase spectrum. References, Berglund, K.A., and M.A.Larson, "Growth of Contact Nuclei of Citric Acid Monohydrate ," AIChE Sym. Ser., No. 215,78 (1982): 9-13. Dean, G.R., "An Unstable Crystalline Phase in the D-glucose Water System," Carbohyd. Res., 34 (1974): 315é22. Edwards, L.w., "Process for Continuous Crystallization of Alpha Monohydrate Dextrose Utilizing High Agitation," U.S.Pat. 4,357,172 (Nov. 2, 1982). Larson, M. A., "Advances in the characterization of crystal nucleation," AIChE Sym. Ser., No. 240,80 (1984). Vasko, P.D., J.Blackwell, and J.L.Koenig, "Infrared and Raman Spectroscopy of Carbohydrates ," Carbohyd. Res., 23 (1972): 407-16. TOP VIEW tioaég‘ g (6) r ' m f {77277.1 —* l§<\\\\“ SIDE VIEH Figure 1. Schematic diagram of nucleation cell with the features: (1) chamber containing solution; (2) parent crystal; (3) glass cover slip whore parent crystnl is slid: (4) support rods for glass cover slip; (S) thermistor; (6) movable rod holding parent crystal;,(7) chamber containing constant temperature water; and (8) water inlet and outlet; Figure 2. Typical examples of photomicrographs of contact nuclei of dextrose formed from pure solution at 314K and supercooling of 4K. The elapsed time between photographs is 72 min. Note the two different crystal forms present. .11 . RAMAN SHIFT (i) 5360 5425 5490 l I W” «Wm/A WWW. dim/m w , )9 WW 1 -Figure 3. l 780 .892 1603 1120 1£26 RAMAN SHIFT (.m" ). Raman shift of alpha anhydrous dextrose. (A) pure reagent grade, (8) parent (prism) crystal, (C) contact nuclei in situ at experimental conditions favorable to form to different phases, and (D) solution phase spectrum. 12 RAMAN SHIFT (ix) 5425 '5490 [#fiiflh ‘N%$¥HNNI L, Min/h 780 892 10.03. 11.20 1226 RAMAN SHIFT (cm") Figure 4. Raman shift of alpha monohydrate dextrose. (A) pure reagent grade, (B) parent (needle) crystal, (C) contact nuclei in situ at experimental conditions favorable to form two different phases, and (D) solution phase spectrum. 13 Contact Nucleation From Aqueous Dextrose Solutions by P.Elankovan and K.A.Berglund Departments of Agricultural and Chemical Engineering Michigan State University, East Lansing, MI 48824 Accepted by AIChE Journal 14 Abstract Contact nucleation studies were performed with the dextrose-water system. Under some conditions two crystallographic phases were formed, which were identified in 3133 using a laser Raman microprobe. In addition, the growth of the contact nuclei was monitored in situ using photomicroscopy. The results are discussed in the context _of contact nucleation and growth models. Introduction Dextrose is the common name for D-glucose, the pure crystalline solid recovered from almost completely hydrolyzed starch. Dextrose (corn sugar, starch sugar, blood sugar, grape sugar) is by far the most abundant sugar in nature and occurs either in the free state. . (monosaccharide form) or chemically linked with other sugar moieties. It can exist in three different crystalline forms: alpha monohydrate, alpha anhydrous, and beta anhydrous. Contact nucleation is the primary source of new particles in many important industrial crystallizations. The exact source of these nuclei is not precisely known. This study focused on such behavior in the important industrial process of dextrose crystallization. It was found by laser Raman microprobe that contact nuclei of both alpha monohydrate and alpha anhydrous dextrose result from the contact of an anhydrous parent crystal growing in pure solution. Under similar conditions using an industrial 15 syrup, only one phase formed. These studies suggest that simple breakage from the parent crystal does not explain the source of contact nuclei. The contact nuclei of both the anhydrous and monohydrate forms of dextrose exhibit growth rate dispersion, wherein, under the same microscopic conditions crystals grow at different rates. Review of literature Several researchers have attempted to prepare pure dextrose from starch conversion liquor either in syrup form or as a solid mass of crystals containing the mother liquor. The solubility curves of ice, alpha monohydrate, alpha anhydrous, and beta anhydrous dextrose were determined for the temperature range from -32 to 62°C by Young (1957). The shapes of these three crystals were reported by Becke (Newkirk,l924) and the optical-crystallographic properties were determined by Dean (1974). The transitional temperature range between alpha monohydrate and alpha anhydrous in various concentrated solutions is 38 to 500C (Dean, 1974, Edwards and Hillsdale, 1982). " Behr (1881) crystallized individual crystals of anhydrous dextrose by seeding concentrated corn starch with a small amount of very pure anhydrous dextrose crystals. The seeded mother liquor was allowed to stand without agitation in heated rooms and care was taken to exclude any traces of dextrose hydrate crystals. This process was quite 16 similar to that described by Soxhlet (1886). In the early 1920's a successful commercial-scale method for production of pure dextrose monohydrate dextrose was invented by Newkirk (1936). Dextrose monohydrate crystals were produced by batch crystallization, by addition of pure dextrose monohydrate seed crystals to a starch solution atua suitable temperature. A mixture of crystals and mother liquor was formed and the crystals were separated by centrifugation. At high supersaturation, nucleation of monohydrate dextrose was increased and the formation of many small crystals increased the surface area per unit weight of crystals. These small crystals required a large amount of wash water to remove the mother liquor and decreased the yields because of dissolution of crystals during centrifugation (Edwards and Hillsdale, 1982). The crystallization of anhydrous glucose required Successive crystalline crops (Dean and Gottfried, 1950). Newkirk (1936) also invented methods for manufacturing alpha anhydrous and beta anhydrous dextrose in a crystalline form of high purity. However, Dean and Gottfried found that the concentration of impurities in the mother liquor was too great for successful operation. On the other hand, crystallization of dextrose in the monohydrate form proceeded satisfactorily despite the presence of impurities. Another advantage of the monohydrate was stronger temperature dependence of solubility as contrasted with that for the anhydrous form (Newkirk, 1936). 17 Crystallization Nucleation is the single most important phenomenon that controls the crystal size distribution. In most cases heavy suspensions do not support "high" supersaturation: therefore, homogeneous nucleation is probably insignificant. Nucleation that occurs at low supersaturations and requires the presence of growing crystals is called secondary nucleation. This has been observed to occur in carbohydrate systems.and is usually referred to as "false grain"(Dean and Gottfried, 1950). In investigating the variables affecting the growth of dextrose crystals, Newkirk (1925) observed that great care was needed in cooling the mass in order to grow crystals properly without excessive formation of secondary nuclei. . Botsaris and Dank (1976) presented a comprehensive review of secondary nucleation and de Jong (1978) reviewed the progress in understanding the phenomenon of secondary .nucleation. Secondary nucleation itself has been confirmed by several investigators including, Timm and Larson (1968),- Bennett g; a1 (1973), and Youngquist and Randolph (1972) when they noted the dependency of nucleation on the suspension density. The work of Randolph and Sikdar (1974) and Bauer et a1(1974) also indicated that crystals were being born into a measurable (= 1-10 m*10'6) size range. Contact nucleation is the most important type of secondary nucleation in many industrial crystallizations. 18 In the work of Garside (1978), Garside (1979), and Rusli g; :1; (1980), the contact nucleation process was observed both with photomicroscopy and a Coulter counter. They used a hcontacting device in a batch crystallizer and found that a large number of microscopically visible nuclei were immediately formed with a distribution of sizes as well as some anomalous growth behavior. Gwynn g; glg(1980) reported contact nucleation studies with the sucrose-water system using the Coulter counter technique. These studies confirmed that contact nucleation is the primary source of new particles and contact nuclei of sucrose have a measurable initial size distribution, not zero size as assumed in the MSMPR crystallizer model. In addition, by observing the size distribution at subsequent times they inferred size dependent growth. Using the photomicroscopic techinque, Berglund and Larson (1982) and Shanks and Berglund (1985) studied contact (nucleation in the citric acid monohydrate-water and sucrose- water systems, respectively. These studies confirmed size- independent growth rate, growth rate dispersion, and initial nuclei size distribution. A linear regression analysis of time versus size showed a constant growth rate for each crystal. Based on this premise, Ramanarayanan g; gig (1982) proposed the Constant Crystal Growth (CCG) model. The occurrence of growth rate dispersion or size dependent growth are important in modeling of growth rate dispersion. The objective of this research was to study the 19 growth rate and the initial size distribution of dextrose crystals to discriminate between these two phenomena. In addition, it was hoped that the study of contact nucleation would allow inferences to be made as to its mechanism. Raman Spectroscopy Raman spectroscopy is a form of vibrational spectroscopy analogous to infrared (IR) spectroscopy. When visible radiation is incident on a sample, it is absorbed, transmitted, reflected, or scattered. The scattered radiation may be elastically scattered (Rayleigh scattering) at the same frequency of the incident radiation or inelastically scattered (Raman scattering) at some shifted ‘frequency. Raman scattered light can be used to determine chemical structures by the study of molecular vibrations at characteristic frequencies.‘ Several features make Raman spectroscopy uniquely suited to the study of crystallization. The Raman spectrum of water does not obscure the spectrum of the dissolved species. The integrated intensity of a given Raman band is to a very good approximation linear with molar concentration of the species in solution giving rise to the band. Micro-Raman spectroscopy permits the in gigg analysis of very small regions by focusing laser light through the Tobjective lens of a microscope. The laser Raman spectra of D-glucose, D-fructose, sucrose, cellobiose, maltose, and dextran for the 300 to 20 1500 cm"1 region were reported by Vasko gt g1 (1971,1972), Cael gt 31‘ (1974), Mathlouthi and Luu(1980 ), and She gt git.(1974 ), respectively. D-glucose, cellobise, maltose and dextran consist of one or more D-glucose residues. Their vibrational spectra are very similar in this region (Vasko gt gl, 1971). Tu gt g1 (1978) used the advantages of laser Raman spectroscopy to study the mute-rotation of D- glucose in aqueous solutions. Recent studies using the Raman effect provided information permitting reliable characterization of the anomers of D-glucose. In the- analysis of alpha-D-glucose, both as a crystalline solid and in aqueous solution, Vasko gt git (1972) obtained reasonable agreement between the calculated frequencies and those observed in the Raman spectrum. Experimental Apparatus Contact Nucletion In the present study, contact nucleation experiments were performed using the same photomicroscopic technique applied to the citric acid monohydrate system by Berglund and Larson (1982). The photomicroscopic contact nucleation cell is shown in Figure 1. Experiments were conducted by first gluing a parent prism dextrose crystal (1-2 mm) to the moveable rod. Growth of large crystals of dextrose was accomplished by carefully cooling a solution of dextrose to induce nucleation, then transferring a few nuclei to a slightly undercooled solution for growth. 21 After the parent crystal was obtained and mounted, a dextrose solution saturated at a known temperature was added to the cell. The solution was heated to destroy any nuclei that may have been generated by transferring the solution. Heating also slightly dissolved the parent crystal, insuring that no surface irregularities formed by washing and drying remained. The solution was cooled and the parent crystal allowed to grow. When the growth of the parent crystal was proceeding in a regular manner as evidenced by well defined faces, it was slid approximately 0.5 cm. across the glass plate shown in Figure 1 to generate contact nuclei. Contact nuclei were observed after a few minutes and the nuclei were photographed at time intervals using 100X magnification by a camera-equipped microscope. Dextrose solutions saturated at three different experimental temperatures and a standard industrial syrup saturated at 318K were used in these studies. DataAnalysis An image analyzer was used to determine the area of each crystal in the photographs. The characteristic size of each crystal was taken as the square root of the area, i.e., the geometric mean size. This is the value referred to as "size" in the subsequent discussion. Laser Raman Spectroscopy A laser Raman microprobe was used to identify different crystalline phases during contact nucleation. Dextrose solution was saturated at 318K, filtered through a 0.45 m 22 membrane, and the contact nucleation experiment was performed at 314K as described above. The solution spectrum and the contact nuclei spectra were taken at a few locations after several hours. The Raman spectra of reagent alpha monohydrate dextrose, reagent alpha anhydrous dextrose, and the parent crystals (needle and prism) were also recorded. The spectrometer used was a Spex model 1406 equipped with a photomultiplier. All samples were illuminated with the 5145 A line of a Coherent Radiation Model CR-5 Ar+ laser using a SPEX Micramate. The recorded region was from 772 to 1221 cm-1 (5360 to 5490 A) at a scan speed of 0.05 A/second. Results and Discussion For the purpose of discussion, the well—formed crystals are designated prisms and the elongated crystals designated needles. The parent crystals (prisms) were grown at 323K with 3 degrees undercooling which promoted the formation of the anhydrous form. An anhydrous crystal (as verified by Raman spectroscopy) was contacted in all experiments which were favorable to both alpha—monohydrate and alpha—anhydrous formation. At 314K with 4 degrees undercooling two different polymorfhs, neeiles and arisms, were observed in the pure solution and the photographs taken at two different time are shown in figure 2. However, these two different kinds of nuclei were not obtained simultaneously in the other contact nucleation experiments, or in comparable conditions with industrial syrup. Apparently the impurities present in the industrial syrup 23 are important in inhibiting formation of needle crystals. The laser Raman microprobe technique was used to identify the two different phases in pure solution at 314K and 4 degrees undercooling. The Raman shift obtained at two different locations it gttg are given in Figures 3 and 4, respectively. The important differences in these figures are the Raman shifts and splits at about 918 and 1078 cm-1. The band assignments for these two wavenumbers are Cl-H stretch and C-O-H deformation (Vasko gt 31, 1972). The additional water molecule present in alpha monohydrate forms pa bond (probably a hydrogen bond) with the C-0. This bond causes splitting and shifting of the bands at these wave numbers. Upon comparison of these spectra with the others recorded for alpha monohydrate and alpha anhydrous crystals, it was concluded that there were two different growing phases present. One phase was primarily alpha monohydrate dextrose (needles) and the other was entirely alpha anhydrous dextrose (prisms). It appears that some other phase may also have been present in the needle, as suggested by Dean (1974) and evidenced by the additional band at 942 cm-1. Since an anhydrous parent crystal was used in the contact and two different phaSes were formed, simple breakage cannot explain formation of the monohydrate phase. Theseresults support the possibility that a semiordered layer is removed from the surface of a growing crystal in contact nucleation (Berglund and Larson (1982)). .However, another possibility in the epitaxial growth of the 24 monohydrate phase on an anhydrous nucleus. This should not be discounted since the Raman spectra indicate that the needles are not pure monohydrate. Clearly, the mechanism is still in question. Figures 5 and 6 show examples of size versus time plots for individual alpha monohydrate and alpha anhydrous crystals at different temperatures. The most important feature is their linearity, which indicates that each crystal grew at a constant but different rate. Each line has a correlation coefficient of at least 0.98. The intercepts of these plots also suggest that an initial size distribution may have been present. Unfortunately, it was not possible to study the genesis of the initial size distribution phenomenon due to the time resolution of this experiment. The growth rate was plotted against the apparent initial size and is presented in Figures 7, 8 and 9. The large amount of scatter in these data demonstrates little correlation between initial size and growth rate. These results verify that the proper method to analyze the growth rate dispersion of this system is the "Constant Crystal Growth" (CCG) model. Berglund and Larson (1984) showed in their analysis of citric acid monohydrate with the CCG model that the curvature in the semilogarithmic population density versus size plot can be attributed to growth rate dispersion; the initial size distribution had a smaller effect. Figure 10 shows an increase in supersaturation causes an increase in mean growth rate of 25 anhydrous dextrose. It was not the purpose of this work to establish the correlation between growth rate and supersaturation. Figure 11 shows the trend between mean growth rate and variance of growth rate distribution. Faster growing crystals exhibit a larger variance of growth rate distribution. Further, these two figures indicate that growth rate dispersion may be correlated to supersaturation. To better understand or to further correleate this phenomenon, more experiments are needed. 26 Conclusions 1. Needle crystals are primarily alpha monohydrate and prism crystals are alpha anhydrous dextrose. 2. It is possible to form alpha monohydrate crystals by contacting alpha anhydrous parent crystals. 3. The laser Raman microprobe is a useful technique for it situ identification of the phases of alpha monohydrate and alpha anhydrous glucose. 4. Dextrose contact nuclei (anhydrous and monohydrate) appear to grow at a size independent rate. 5. Initial size and growth rate distributions are observed for contact nuclei of both anhydrous and monohydrate phases of dextrose. 6. The "Constant Crystal Growth" model should be used to model growth rate dispersion of contact nuclei in the- dextrose water system. Acknowledgements This work was supported in part by the A.E. Staley Manufacturing Company. The laser Raman microprobe was provided through the National Science Foundation under grant number CPE - 8409458 and the Michigan Agricultural Experiment Station. The assistance of Dr. David Bruce of the Department of Physics and Astronomy at Michigan State University is appreciated. Figure Figure Figure Figure Figure Figure Figure Figure Figure 27 Schematic diagram of nucleation cell (1) chamber containing solution; (2) parent crystal; (3) glass cover slip where parent crystal is slid; (4) support rods for glass cover slip; (5) thermistor; (6) movable rod holding parent crystal; (7) chamber containing constant temperature water; and (8) water inlet and outlet: Typical examples of photomicrographs of contact nuclei of dextrose formed from pure solution at 314K and a supercooling of 4K. The elapsed time between photographs is 72 min. Note the two different crystals forms present. (Raman shift of alpha monohydrate dextrose. (A) pure reagent grade, (B) parent (needle) crystal, (C) contact nuclei it gitu at experimental conditions favorable to form two different phases, and (D) solution phase spectrum. Raman shift of alpha anhydrous dextrose. (A) pure reagent grade, (B) parent (prism) crystal, (C) contact nuclei in; sitg at experimental conditions favorable to form two different phases, and (D) solution phase spectrum. Size versus time for alpha monohydrate dextrose contact nuclei in the dextrose-water system formed and grown at 3 degrees supercooling at 314K. Each line is for an individual crystal. Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 2 degrees supercooling at 323K. Each line is for an individual crystal. Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 314K. Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 319K. Growth rate versus initial size for anhydrous dextrose dextrose contact nuclei formed and grown at 323K. Figure 10. Figure 11. 28 Mean growth rate versus relative supersaturation for anhydrous dextrose contact nuclei formed and grown at various temperatures. Variance of the growth rate distribution versus mean growth rate for dextrose contact nuclei formed and grown at various temperatures and supercoolings. . 29 References Bauer, L.G., M.A.Larson, and V.J.Dallous, "Contact Nucleation of MgSO .7H 0 in a Continuous MSMPR Crystallizer," cued. sag. Sci.,29(1974): 1253-61. ‘Behr,A., "Process of Manufacturing Crystallized Anhydride of Grape-sugar from a Watery Solution of Grape-sugar from a Watery solution of Grape-sugar," U.S.Pat. 250,333 (Dec. 6, 1881), 256,622 (April 18,1882). Bennett, R.C., H.Fieldeman,and A.D.Randolph, "Crystallizer Influenced Nucleation," Chem. Eng. Prog., 69 (1973): 86-91. Berglund, K.A., "Growth and size Dispersion Kinetics for Sucrose Crystals in the Sucrose Water System" M.S.Thesis, Colorade State Univ., Ft. Collins Co. (1980). Berglund, K.A., "Formation and Growth of Contact Nuclei", Ph.D. Disertation, Iowa State Univ., Ames IA (1981). Berglund, K.A., and M.A.Larson, "Growth of Contact Nuclei of Citric Acid Monohydrate ," AIChE Sym. Ser., No. 215,78 (1982): 9—13. Berglund, K.A.,and M.A.Larson "Modelling of Growth 'dispersion of Citric Acid Monohydrate in Continuous Crystallizers", AIChE Journel, 30, 280 (Mar., 1984). Botsaris, G.D., "Effects of Secondary Nucleation in Crystallizing Systems ," 5th Symp. Inst. Cryst., 1972: 1-23. Cael, J.J., J.L.Koenig, and J.Blackwell, "Infrared and Raman Spectroscopy of Carbohydrates," Carbohyd. Res., 32 (1974): 79-91. Dean, G.R., and J.B.Gottfried, "The Commercial Production of Crystalline Dextrose ," Adv. Carbohyd. Chem., 5 (1950): 127. Dean, G.R., "An Unstable Crystalline Phase in the D-glucose Water System," Carbohyd. Res., 34 (1974): 315-22. De Jong, E.J., in Industrial Crystallization 78, E. J. de- Jong and S. J. Jancic, Eds. (North Holland, Amsterdam, 1979): 3-17. Edwards, L.W., "Process for Continuous Crystallization of Alpha Monohydrate Dextrose Utilizing High Agitation," U.S.Pat. 4,357,172 (Nov. 2, 1982). Garside, J. and M.A.Larson, "Direct Observation of Secondary Nuclei Production ," J. Cryst. Growth, 43 (1978): 694-701. 30 . Garside, J., I.T.Rusli and M.A.Larson, "Orgin and Size Distribution of Secondary Nuclei," AIChE J. 25, 1 (1979): 57-64. Gwynn, S. M., R. W. Hartel, and V. G. Murphy, "Contact Nucleation in the Crystallization of Sucrose," AIChE Mtg., Chicago (Nov. 16 - 20,1980). Kirchoff, G.S.C., Academic imperiale des sciences de St. Petersbourg, Memoires, 4, 27 (1811). [Cited by Dean, G.R. (1950)]. Mathlouthi, M., A.M.Meffroy-Biget, and D.V.Luu, "Laser Raman Study of solute-solvent Interactions in aqueous solutions of D-fructose, D-glucose, and Sucrose," Carbohyd. Res., 81 (1980) 213-23. Newkirk, W.B., "Manufacture and Uses of Refined Dextrose," Ind. Eng. Che., 16 (1924):1173-75. Newkirk, W.B., "Development and Production of Anhydrous Dextrose," Ind. Eng. Che., 28 (1936):?60-65. Newkirk, W.B., "Manufacture of Dextrose," U. 8. Pat., 1,521,830 (Jan. 6, 1925). Ramanarayanan, K.A., K.A.Berglund, and M.A.Larson, "Growth Kinetics in the Presence of Growth Rate Dispersion from Batch Crystallizers," AIChE 75th Annual Meeting Los Angales ( Nov. 14-16, 1982). Randolph, A. D., and S. K. Sikdar, "Effect of soft impeller coating on the net formation of secondary nuclei, " AIChE J. 20 (1974) 410- 12. Rusli, I.T., M.A.Larson, and J.Garside, "Initial Growth of Secondary Nuclei," CEP Symposium Series, 193, 76 (1980): 50- 58. _ She, C.Y., N.D.Dinh, and A.T.Tu, "Laser Raman scattering of glucosamine, N-acetylglucosamine, glucuronic acid," Biochem. Biophys. Acta, 372(2) (1974):345-57. Soxhlet, "Process of ManufaCturing Grape-sugar ," U. S. Pat. 335,044 (Jan. 26, 1886). Timm, D.C., and M.A.Larson, "Effect of Nucleation on the Dynamic Behavior of Continuous Crystallizer," AIChE. J. 14, 3(1968): 452-57. Tu, A.T., N.D.Dinh, and J.Maxwell, "Laser Raman scattering of alpha- and beta-methyl-D-glucosides and hyaluronic acid," Stud. Biophys., 63(2) (1978):115-31. 31 Vasko, P.D., J.Blackwell, and J.L.Koenig, "Infrared and Raman Spectroscopy of Carbohydrates ," Carbohyd. Res., 19 (1971): 291-310. Vasko, P.D., J.Blackwell, and J.L.Koenig,'"Infrared and Raman Spectroscopy of Carbohydrates ," Carbohyd. Res., 23 (1972): 407-16. White, E.T.and P.G.Wright, "Magnitude of Size Dispersion Effects in Crystallizer," CEP Symposium Series No. 110,67 (1971): 81-87. Wichelhaus,H., "Der Starkezucker" Akademische Verlagsgesellschaft, Leipzing (1913). [cited by Dean, G.R.(l950)]. Young, F.E., "D—glucose-water phase diagram ," J. Phy. Che. 61 (1957): 616-19. Youngquist, G.R. and A.D.Randolph, "Secondary nucleation in a class II system. Ammonium sulfate-water," AIChE J. 18 (1972): 421-9. 32 Figure l. [‘7‘" fiq- 1 lsx\\\\\ ‘ SIDE VlEH Schematic diagram of nucleation coil with the features: (1) chamber containing solution; (2) parent crystal; (3) glass cover slip where parent crystal is slid: (4) support rods for glass cover slip; (S) thermistor; (6) movable rod holding parent crystal:’(7) chamber containing constant temperature water; and (8) water inlet and outlet; 33 Figure 2. Typical examples of photomicrographs of contact nuclei of dextrose formed from pure solution at 314K and supercooling of 4K. The elapsed time between photographs is 72 min. Note the two different crystal forms present. '34 RAMAN SHIFT (A) 5360 S425 5490 I I W11 mum/“A mils/MM «AM/in K. W 780 . 892 1003 1120 1226 RAMAN SHIFT (.m°‘) Figure 3. Raman shift of alpha anhydrous dextrose. (A) pure reagent grade, (B) parent (prism) crystal, (C) contact nuclei in situ at experimental conditions favorable to form to different phases, and (D) solution phase spectrum. ' 35 RAMAN SHIFT (A) 5360 5425 '5490 L. M.,MKW 780 892 ‘-1003_ 11,20 I226 RAMAN SHIFT (cm.1 ) Figure 4, Raman shift of alpha monohydrate dextrose. (A) pure reagent grade, (B) parent (needle) crystal, (C) contact nuclei in situ at experimental conditions favorable to form two different phases, and (D) solution phase spectrum. SIZE, METERS . 10‘ 12 36 11- 10-1 T=314K AT=3K Figure 5. I U i 600 1200 1800 2400 3000 TIME, SECONDS Size versus time for alpha monohydrate¢dextrose contact nuclei in the dextrose-water system formed and grown at 3 degrees supercooling at 314K. Each line for an individual crystal. SIZE, METERS x106. 37 50 0.0. 1200 2400 3500 4800 6000 7200 TIME, SECONDS Figure 6. Size versus time for alpha monohydrate dextrose contact nuclei in the dextrose formed and grown at 3 degrees supercooling at 323K. Each line for an individual crystal. 38 contact nuclei formed and grown at 314K. 4-0 i t I I I v I I a I V V D D x v V 0 fl 0 V CE}: 0 E O .. V U U Id 0 ‘7 F- 3’ ' I no» - 6:: 2.0 - o 0 v0 0 g 0 q, ePURE AT=2K 8 O 0 ’ opune AT=3K ‘3 ' .sInous AT=2K DINDUS AT=3K VINDUS AT=4K I 1 I I I I 0.0 5.0 10.0 15.0 20.0 25-0 30-0 6 INITIAL SIZE, m x10 Figure 7. Growth rate versus initial size for anhydrous dextrose 39 61) | I l I TEMP-319K u “PURE AT=4K D. OPUIIE AT=2K ,_ ~ : ' . x o _ UI'UIII. AIle( 9‘ I. - E 4.0 D ui I— D D 3.5 o I E; 0 I" 0 o 2.0 ‘" 0 D " O '6 ' °“ : ° o o . O ' o o 1 l 1 l ' 0.0 5.0 10.0 15.0 , 20.0 25.0 , 6 IN" IAL SIZE, 111 x 10 Figure 8. Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 319K. 40 6.0 v T I T ‘l v- . g 4.0 -' o 00 o - E. o O m 0 o 1- e g o ... E5 .. . . g 2.0 - . 0. - a: ' TEMP- 323K \9 .PURE T=2K OPURE T=3K L _l 1 1 I 0.0 5.0 10.0 15.0 20.0 25.0 30.0 . , 6 INITIAL SIZE, 111 x 10 Figure 9. Growth rate versus initial size for anhydrous dextrose contact nuclei formed and grown at 323K. 41 5.00 h . —. . . I v 4.00 - .. C3 F x O W .2 I- m- 3.00 L- .1 p. <1 a: I C E 2.00 - - e 3.23 K g s ' s 319 K 0 v 314 K 1.00 - A 0.00 l l 1 1 1 0.000 _.005 .010 .015 . .020 .025 - .030 SUPERSATURATION [ a] Figure 10. Mean growth rate versus relative supersaturation for anhydrous dextrose contact nuclei formed and grown at various temperatures. VARIANCE 0F GROWTH RATE 42 2.000 1.750. 1.500 - 1.2504 1.000— 0750 - 0.500 - 0.250 - 0.000 . ‘ a NEEDLES . ' s INDUST. ‘ , PRISMS Figure'll. — I I 2 3 4 GROWTH RATE muo’ Variance of growth rate distribution versus mean growth rate for dextrose contact nuclei formed and grown at ' various temperatures and supercoolings. 43 CONCLUSIONS Needle crystals are primarily alpha monohydrate and the prism crystals are alpha anhydrous dextrose. It is possible to form alpha monohydrate crystals by contacting alpha anhydrous parent crystals. The laser Raman microprobe technique is a useful technique for in gitg identification of the phases of alpha monohydrate and alpha anhydrous dextrose Dextrose contact nucleiappear to grow at a size independent rate. Both initial size and growth rate distributions are observed for contact nuclei of dextrose in different forms. The "Constant Crystal Growth" model should be used to model growth rate dispersion in the dextrose water system. The variance of the growth rate distribution correlates directly to mean growth rate. Contact nuclei may be formed from disruption of some "intermediate phase" rather than some sort of breakage. 44 SUGGESTIONS FOR FUTURE WORK A photomicrographic cell should be designed to study crystal growth over longer periods of time without depletion of supersaturation. A thorough study of the relationship between supersaturation and growth rate should be undertaken. More contact nucleation experiments should be done to correlate variance of growth rate versus growth rate. A through study of contact nucleation of dextrose with different impurities should be performed to confirm that impurities inhibit formation of needle type of dextrose crystals. Use of immage analysis equipment should be automated to determine the nuclei size and growth rate of nuclei from photographs or slides. The laser Raman microprobe technique should be used to quantify the polymorphs present in contact nucleation of dextrose. In particular the, possibility of mixed phase crystal formation should be studied. The laser Raman microprobe technique should be used to study the formation of different polymers in other sugar systems. Contact studies of dextrose monohydrate should be undertaken to see if any anhydrous dextrose can be formed. 10. 45 More detailed studies of the surface of the growing crystal needed to detect the presence of exponential growth. More studies needed to find relative growth rate of anhydrous versus monohydrate dextrose crystals. APPENDICES APPENDIX A. RAW DATA CRYSTAL SIZE VERSUS TIME 46 Appendix A: Raw data size(n*106) and time Run #1 Pure Solution - Prisms T - 314x Supersat. a 4K Cryst. No 1800 S. 2580 S. 3120 S. 4200 S. 5220 S. 1 17.1 17.8 24.6 27.0 27.4 2 27.2 18.5 22.2 24.8 25.8 3 28.4 36.1 40.3 46.5 48.8 4 48.9 55.4 --- 66.7 70.9 5 16.2 18.8 23.2 26.6 31.6 6 20.6 24.7 32.5 35.4 37.7 7 13.4 16.0 --- 24.6 525.8 8 21.2 24.1 27.1 28.1 31.5 9 23.4 34.4 37.4 41.2 45.4 10 31.3 38.8- 43.4 48.2 52.1 11 33.6 42.1 47.3 52.2 54.7 Run #1 Pure Solution - Needles T - 314x .Supersat. - 4K Cryst. NO- 1800 S. 2580 S. 3120 S. 4200 S. 5220 S. 12 41.3 53.7 56.2 65.0 65.0 13 59.4 63.8 94.9 106 108 14 43.7 52.4 57.9 59.7 61.7 15 63.5 72.6 74.7 88.4 91.8 16 70.0 87.1 88.4 94.0 108 17 51.6 57.0 63.8 80.2 92.2 18 50.7 60.6 66.4 69.2 73.8 47 Run #2 Industrial Solution Sat. Temp. -- 318K, Supersat. = 4x Cryst. 900 1800 2700 3600 4500 5400 Sec. 19 12.9 19.9 21.6 24.8 26.9 29.6 20 15.8 ---- 19.7 22.3 25.8 26.6 21 14.9 16.0 20.6 21.6 25.0 27.0 22 14.0 19.9 ---- ---- 27.5 27.8 23 15.9 19.1 21.4 ---- 26.8 27.5 24. 15.5 18.5 21.8 22.9 26.1 27.3 25 16.2 21.2 22.5 24.1 25.5 ---- 26 13.6 15.8 18.6 19.8 20.6 22.5 27 13.0 14.1 15.7 18.9 22.5 24.8 28 12.3 17.1 22.6 ---- 25.7 19.8 48 Run #3 Pure solution Expt. Temp. = 314K, Supersat. = 2K Cryst. 1200 2400 3600 4800 6000 29 19.2 20.2 22.9 ---- 26.4 30 18.4 18.8 21.0 21.9 23.8 31 17.7 19.4 20.0 25.5 28.4 32 12.6 13.5 15.8 18.3 22.3 33 18.0 20.9 23.0 ---- 25.9 34 15.3 18.5 18.8 19.5 20.6 35 14.1 19.0 20.9 22.7 24.2 36 10.4 13.6 15.7 16.4 21.5 ---- 17.9 18.1 19.6 21.4 37 49 Run #4 Industrial soln. Sat. Temp. = 318K, Supersat. = 2K Cryst. 1200 1800 2400 3000 3690 4200 4800 38 10.8 ---- 11.7 13.8 16.8 17.3 19.4- 39 9.33 10.3 11.3 ---- ---- 13.2 ---- 40 8.71 11.0 ---- 11.7 12.5 13.5 14.0 41 8.78 10.4 11.6 ---- 15.3 17.6 19.2 42 14.0 15.2 18.7 20.9 ---- ---- 25.5 43 9.92 10.8 12.6 ---- 14.7 16.2 ---- 44 8.75 10.1 11.7 --- 14.8 ---- 16.4 45 7.79 8.58 9.58 11.0 15.6 17.3 ---- 46 7.08 10.3 11.1 ---- 12.7 14.8 15.8 47 8.59 11.3 12.0 ---- 14.9 15.9 18.2 50 Run #5 Pure Solution Expt. temp. = 314K Supersat. = 3K Cyrst. 300 900 1500 2100 2700 S 48 10.3 11.3 12.3 12.4 13.0 49 5.82 6.96 7.63 7.33 9.83 50 4.88 7.69 8.49 9.07 11.9 51 4.31 5.45 6.38 7.50 9.86 52 5.79 6.20 7.44 8.26 10.5 53 3.93 5.24 6.13 7.32 9.02 54 6.77 7.22 7.58 8.95 10.5 55 6.36 7.40 8.14 8.76 9.37 56 8.35 8.82 9.63 10.6 11.5 57 4.53 6.28 7.07 8.85 10.6 51 Run #6 Industrial Soln., Sat. Temp. 318K, Supersat. = 3K Cryst. 900 1800 2700 3600 4500 5400 6300 58 45.7 49.3 54.5 56.3 57.6 59.3 --- 59 38.3 39.3 42.7 46.2 50.3 52.1 55.3 60 31.4 33.6 35.2 37.3 38.6 43.8 42.5 61 27.9 30.5 35.2 38.1 40.2 ---- 41.6 62 23.0 25.7 27.9 30.4 32.2 34.4 36.1 63 21.8 24.3 28.2 30.6 34.4 36.2 37.8 64 25.7 28.0 31.2 33.8 36.5 37.9 ---- 65 25.7 27.5 30.4 32.8 33.6 35.9 39.7 66 26.3 28.0 30.6 32.1 33.6 34.4 ---- 67 28.7 31.9 32.5 33.9 36.6 42.5 ---- 68 23.1 26.8 29.6 32.0 34.6 36.2 38.3 69 21.2 22.7 26.1 30.3 ---- 32.8 35.4 70 21.7 24.7 26.5 29.8 32.2 33.0 36.7 71 24.2 26.6 28.3 32.1 -—-- 35.1 ---- 72 18.3 19.2 23.6 24.5 29.0 31.9 32.7 73 23.3 25.6 28.5 31.5 33.8 35.7 36.4 52 Run #7 Pure Solution, Expt. Temp. = 2319!, Supersat. - 2K Cryst. 900 2400 3600 4800 6000 S 74 7.79 11.6 13.7 14.2 16.9 75 9.03 10.7 12.5 14.3 14.7 76 12.1 14.1 16.6 17.6 18.7 77 12.2 14.7 16.2 17.5 18.7 78 9.10 11.4 12.8 14.1 15.1 79 8.06 8.79 11.8 12.9 14.5 80 12.3 14.4 16.0 19.7 20.7 81 6.48 7.99 10.5 12.5 14.1 82 13.4 15.7 17.2 19.5 20.6 83 13.1 14.6 16.0 17.4 21.4 84 7.79 10.5 12.4 15.0 16.1 85 8.30 10.6 13.6 15.4 16.4 86 8.13 10.2 12.2 14.7 15.7 53 :Rnn #8 Pure Solution, Expt. Temp. = 319K, Supersat. = 3K Cryst. 900 1800 2700 3600 4500 5400 87 3.94 4.66 6.57 8.13 8.88 11.8 88 15.2 16.7 19.0 19.8 21.1 22.2 89 13.0 16.0 17.3 17.9 18.5 18.7 90 3.66 6.98 7.99 10.2 12.2 13.3 91 22.6 24.5 26.0 26.7 29.5 30.5 92 12.9 14.9 16.1 16.6 17.6 20.1 93 9.44 11.4 13.3 14.6 16.0 17.3 94 14.4 15.0 19.4 20.5 22.0 23.7 95 14.4 15.8 1715 19.2 19.8 22.1 96 6.77 7.69 9.41 10.5 11.6 12.8 Run #9 Pure Solution, Expt. Temp. = 319K, Supersat. = 4K 54 Cryst. 100 820 1540 2260 2980 97 11.3 13.5 15.7 19.2 20.7 98 5.83 7.87 8.91 9.85 11.1 99 11.7 15.3 17.4 19.6 23.3 100 5.65 6.09 7.62 8.73 9.15 101 5.80 7.15 8.39 9.56 12.1 102 6.84 9.34 11.2 13.0 15.7 103 4.76 7.10 8.87 9.95 12.1 104 5.94 7.05 8.27 10.1 12.2 105 8.72 11.7 13.8 16.2 17.9 106 7.08 12.3 14.3 17.2 23.6 107 5.88 7.89 12.1 13.4 18.3 S Run #10 Pure Solution, 55 Expt. Temp. = 323K, Supersat. = 2K Cryst. 900 1800 2700 3600 4500 5400 6300 72008 108 23.2 27.9 31.5 34.6 37.8 44.0 44.6 52.2 109 13.9 15.1 17.5 19.3 20.0 22.4 25.1 26.8 110 9.89 10.3 11.3 13.9 16.1 ---- 19.1 ---- 111 11.1 14.0 18.3 20.1 ---- 27.6 33.2 35.5 112 12.0 ---- 15.6 16.7 17.9 19.3 21.2 23.2 113 10.8 13.4 17.3 19.4 ---- 22.3 24.0 25.7 114 13.4 16.2 17.7 ---- 20.9 23.1 27.3 ---- 115 12.8 13.9 17.8 19.1 21.9 23.6 7.2 ---- 116 8.4 10.4 15.1 — 17.8 19.5 22.9 24.5 28.1 117 18.5 21.0 28.2 30.3 34.3 36.5 40.0 42.5 118 12.1 17.1 --- 18.4 21.4 --- 24.0 25.2 119 ---- 13.6 17.0 ---- 20.8 23.9 24.6 ---- 120 12.0 13.8 15.6 16.9 20.1 ---- 23.6 ---- 121 13.4 16.8 21.7 23.4 29.1 31.1 34.3 36.1 56 Run #11 Pure Solution, Expt. Temp. = 323K, Supersat. = 3K Cryst. 900 1800 2700 3600 4500 5400 6300 72008 122 14.9 19.5 20.6 ---- 25.3 ---- 30.1 33.8 123 15.2 19.2 24.2 29.1 31.6 34.2 36.1 40.8 124 16.6 19.8 22.3 25.8 29.5 32.2 36.4 38.2 125 23.8 25.4 28.6 32.1 34.7 36.9 37.3 42.6 126 29.5 32.6 35.1 37.9 ---- 47.3 49.2 52.9 127 16.8 22.7 26.3 29.6 33.2 35.5 39.5 43.4 128 12.3 16.5 20.3 ---- 22.3 23.5 24.4 ---- 129 15.5 18.8 22.7 26.8 27.4 ---- 30.9 33.0 130 16.1 19.3 20.2‘ 24.5 27.2 29.5 31.5 33.4 131 17.3 20.4 24.5 29.8 31.0 33.9 37.7 41.3 132 14.1 19.5 20.9 24.0 27.9 31.4 32.2 ---— 133 16.2 20.6 24.2 28.3 31.8 33.7 34.7 36.3 134 20.6 23.1 26.8 29.0 34.1 ---- 39.0 43.1 135 18.9 23.5 27.3 33.9 38.8 ---- ---- 46.8 APPENDIX B. APPARENT INITIAL SIZE AND GROWTH RATE OF CONTACT NUCLEI OF DEXTROSE. 57 Appendix B. Apparent initial size and growth rate of contact nuclei of dextrose. Run # Under Crystal Initial Growth , Correlation cooling size6 rage coefficient K m*10 m*10 /s 1 4 1 11 3.4 0.92 2 12 2.7 0.96 3 20 5.9 0.97 4 38 6.5 0.99 5 7.9 4.5 0.99 6 13 5.1 0.95 7 6.5 ‘3.9 0.98 8 17 2.8 0.98 9 17 5.8 0.94 10 23 5.9 0.98 11 26 6.0 0.96 12 34 6.3 0.94 13 29 -17 0.94 14 37 4.8 0.90 15 49 8.5 0.98 16 56 9.9 0.96 17 27 12 0.99 18 43 6.3 0.95 2 4 19 12 3.4 0.97 20 13 2.5 0.99 21 12 2.8 0.99 22 13 3.0 0.97 23 16 1.6 0.97 24 14 2.6 0.99 25 16 2.4 0.95 26 12 1.9 0.98 27 9.4 2.8 0.98 28 11 2.9 0.97 3 2 29 17 1.6 0.99 30 16 1.2 0.98 31 14 2.3 0.96 32 9.3 2.0 0.97 33 17 1.6 0.98 34 15 2.0 '0.92 35 13 1.9 0.96 36 8.0 2.1 0.97 37 15 1.0 0.95 58 Appendix B Continued Run # Under Crystal Initial Growth Correlation cooling size6 rate 9 coefficient K m*10 m*10 /s 4 2 38 6.8 2.6 0.97 39 8.0 1.3 0.99 40 7.8 1.3 0.97 41 5.1 2.9 0.99 42 10 3.3 0.99 43 7.3 2.1 0.99 44 6.3 2.2 0.99 45 2.7 3.3 0.96 46 5.4 2.0 0.97 47 6.1 2.4 0.99 5 3 48 10 1.1 0.97 49 5.3 1.5 0.94 50 4.5 2.6 0.96 51 3.4 2.2 0.98 52 4.7 1.9 0.96 53 3.2 2.0 0.99 54 5.9 1.5 0.96 55 6.1 1.2 0.99 56 7.7 1.4 0.99 57 3.8 2.5 0.99 6 3 58 44 3.0 0.97 59 34 3.5 0.99 60 29 2.4 0.98 61 26 2.8 0.97 62 21 2.5 0.99 63 19 3.2 0.99 64 23 2.8 0.99 65 23 2.5 0.98 .66 25 1.9 0.99 67 26 2.4 0.97 68 21 2.8 0.99 69 19 2.8 0.99 70 19 2.8 0.99 71 22 2.5 0.99 72 15 3.0 0.99 73 21 2.7 0.99 59 Appendix B. Continued Run # Under Crystal Initial Growth Correlation cooling size6 rate9 coefficient K m*10 m*10 /s 7 2 74 6.9 1.7 0.98 75 8.0 1.2 0.99 76 11 1.3 0.99 77 11 1.3 0.99 78 8.3 1.2 0.99 79 6.5 1.3 0.98 80 10 1.9 0.99 81 4.8 1.5 0.99 82 12 1.5 0.99 83 11 1.5 0.96 84 6.4 1.7 0.99 85 6.7 1.6 0.99 86 12 1.0 0.99 8 3 87 1.2 1.7 0.98 88 14 1.5 0.99 89 13 1.2 0.91 90 2.5 2.1 0.99 91 21 1.2 0.99 92 12 1.4 0.98 93 8.2 1.7 0.99 94 12 2.2 0.98 95 13 1.7 0.99 96 5.5 1.4 0.99 9 4 97 11 3.3 0.99 98 6.1 1.7 0.99 99 11 3.8 0.99 100 5.4 1.3 0.99 101 5.4 2.1 0.99 102 6.6 3.0 0.99 103 4.8 2.4 0.99 104 5.4 2.2 0.99 105 8.8 3.2 0.99 106 6.8 5.3 0.99 107 5.0 4.3 0.99 Appendix B. Continued 6O Run # Under Crystal Initial Growth Correlation cooling size6 rate coefficient K m*10 m*10 /s 10 2 108 19 4.3 0.99 109 12 2.0 0.99 110 7.4 1.8 0.98 111 7.0 4.0 0.99 112 10 1.7 0.99 113 9.9 2.3 0.98 114 11 2.3 0.98 115 10 2.6 0.99 116 5.8 3.1 0.99 117 15 3.9 0.99 118 12 1.9 0.97 119 9.7 2.4 0.99 120 9.9 2.2 0.99 121 10 3.7 0.99 11 3 122 13 2.8 0.99 123 12 3.9 0.99 124 13 3.5 0.99 125 21 2.9 0.99 126 , 25 3.8 0.99 127 15 4.0 0.99 128 12 2.1 0.95 129 15 2.7 0.98 130 14 2.8 0.99 131 14 3.7 0.99 132 12 3.4 0.99 133 15 3.2 0.98 134 17 3.6 0.99 135 16 4.5 0.98 APPENDIX C1. C2. C3. C4. PHYSICAL PROPERTIES OF D-GLUCOSE. CONDITIONS OF EXPERIMENTAL RUNS WITH NUMBER OF NUCLEI ANALYZED. MEAN GROWTH RATE, VARIANCE OF GROWTH RATE DISTRIBUTION AND RELATIVE SUPERSATURATION AT DIFFERENT CONDITIONS. LINEAR REGRESSION OF GROWTH RATE VERSUS APPARENT INITIAL SIZE. 61 Table 1, znygigal groperties of Q-glugose. Property Alpha monohydrate Alpha anhydrous Beta anhydrous glucose glucose glucose molecular C6H1206’H20 C6H1206 C6H12°6 formula melting pt 356 K 419 K 423 K solubility 32.2-51.2a'b 62,30.2-51.2a 72-51.2a (g/lOOg soln.) [GJZOD 112.2-52.7a'b 112.2-52.7a 18.7-52.7a heat of soln. -105.4 -59.4 -25.9 (J/g-) a- Equilibrium value b- Anhydrous basis [Kirk - Othmer, 1985] 62 Conditions of experimental runs and number of Table 2. nuclei analyzed Run Super cooling Temperature Number of nuclei K K analyzed 1 4 314 18 2 4I 314 10 3 2 314‘ 9 4 ZI 316 10 5 3 314 10 6 3I 315 16 7 2 319 13 8 3 319 10 9 4 319 11 10 2 323 14 11 3 323 14 I - Industrial solution 63 Table 3. Mean growth rate, Variance of growth rate distribution and Relative supers Tempt. Super Mean growth Variance Relative ' cooling rate9 9 supersat. K K m*10 /s m* /s *10 Prisms 314 4 4.8 1.723 2.15 319 2 1.4 0.049 1.07 319 3 1.6 0.106 1.59 319 4 3.1 1.081 2.11 323 2 2.7 0.724 1.04 323 3 3.3 0.394 1.57 Needles 314 2 1.7 0.168 1.07 314 3 '1.8 0.241 1.61 Idustrial solution 316 2 2.3 0.442 1.60 315 3 2.7 0.132 1.61 314 4 2.6 0.494 2.15 64 Table 4. Linear regression of growth rate versus apparent initial size Temp. Super Slope Intercept Correlation cooling 6 6 coefficient K K m*10 /s m*10 /s Prisms 314 4 0.10 3.00 0.66 319 2 0.03 1.72 0.31 3 0.02 1.89 0.48 4 0.18 1.71 0.37 323 2 0.10 1.65 0.39 3 0.04 2.76 0.21 Needles 314 2 0.11 3.07 0.56 3 0.19 2.53 0.18 Industrial solution 316 2 0.11 3.07 0.32 315 3 0.01 2.53 0.18 314 4 0.15 4.62 0.59 APPENDIX D1. DZ. D3. D4. EXAMPLES OF SIZE VERSUS TIME FOR ALPHA ANHYDROUS DEXTROSE CONTACT NUCLEI FORMED AND GROWN AT 2 DEGREES OF SUPERCOOLING AT 319K. EXAMPLES OF SIZE VERSUS TIME FOR ALPHA ANHYDROUS DEXTROSE CONTACT NUCLEI FORMED AND GROWN AT 3 DEGREES OF SUPERCOOLING AT 319K. EXAMPLES OF SIZE VERSUS TIME FOR ALPHA ANHYDROUS DEXTROSE CONTACT NUCLEI FORMED AND GROWN AT 4 DEGREES OF SUPERCOOLING AT 319K. EXAMPLES 0? SIZE VERSUS TIME FOR ALPHA ANHYDROUS DEXTROSE CONTACT NUCLEI FORMED AND GROWN AT 3 DEGREES 0F SUPERCOOLOING AT 323K. SIZE, METERS x 10‘ 65 40.0 35.0 r- 30.0 t 20.0 15.0 0.0 T=319K AT=2 ’ l l I l 0 Figure 12. 1000 2000 3000 4000 5000 6000 TIME, SECONDS Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 2 degrees supercooling at 319K. Each line is for an individual crystal. METERS x 10‘ SIZE, 66 40 T ‘ u) 7: r] o-] 001.3 9:14 4 n U: I l ’0 ’/ -"D 15 - 1’ /°/ ,4.» .a- ‘ ’d o/ .4” 51"'/" V", .a/" 3”“ "’ /.?./ /A/ / 0’. 2.0/.. 7' / 10 - /w’fi” .r’" ;/’ ‘ w./. ./ ‘5' / O /’1/' / / a” C ”_," zf” 5 - ”/‘ _ . I’D, ‘ . o ,z/5 . . 1 o o o I o 0.0 900 1800 2700 3600 4500 5400 m L M M . SECOND Figure 13. Size versus time for alpha anhydrous dextrose ' contact nuclei formed and grown at 3 degrees supercooling at 319K. Each line is for an individual crystal. SIZE, METERS x 10‘ 20. 67 T=319K AT=4K ,/’ 1 J_. L L n 0.0 300 1000 1500 2000 2500 3000 TIME, SECONDS Figure 14. Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 4 degrees supercooling at 319K. Each line is for an individual crystal. 7 SIZE, METERS x 10‘ 68 50.0 1 . . T=323K AT=3A 45.0 - 0.0 0 .0 1200 2400 3600 4800 6000 7200 TIME, SECONDS Figure 15. Size versus time for alpha anhydrous dextrose contact nuclei formed and grown at 3 degrees supercooling at 323K. Each-line is for an individual crystal. ' APPENDIX E. DETAIL DESCRIPTION OF EXPERIMENTAL PROCEDURE. 69 Appendix E: materials and Methods Cgptagt nucleation cell In the present study, contact nucleation experiments were performed using the same photomicroscopic technique applied to the citric acid monohydrate system by Berglund and Larson (1982). The photomicroscopic contact nucleation cell used in this study is shown in Figure 1 and is similar to the design presented by Berglund (1981). The cell has two different chambers: an upper 5 ml chamber for solution, and a lower 8 ml chamber for water circulation to control temperature. The cell was enclosed with three rounded quartz glass plates which allow the transmitted light to pass through the solution and also to view the nuclei. ITwo stainless steel rods which held the glass cover slip, a movable rodehich held the parent crystal, and a thermistor at the same level of the cover slip were mounted in the upper chamber. An American Optical model 110 microscope equiped. with lenses for 100X magnification was used. The microscope was equiped with a 35 mm Olympus camera in which Kodak EL - 135 color slide film was used. The thermistor for measurement of the solution temperature was connected to an Omega digital temperature indicator, model 4108. The accuracy of the thermistor is 0.27OF and of the digital indicater is 0.40°F. Temperature measurements were precise to 0.01°F. Water was 70 circulated to the lower chamber from a Fisher water bath, model 80, which allowed temperature control in the upper Chamber of the cell precise to 0.05°F. Solution preparation Experiments were conducted at 308, 314, 319 and 322K, and saturated solutions were prepared at different temperatures to provide 0, 2, 3 and 4 degrees of supercoolings. These solutions were saturated by placing a large excess of glucose monohydrate from A.E. Staley manufacturing company in distilled - deionized water while stirring continuously. The solutions were placed in a 35 liter water bath equipped with Haake E2 circulators to control the temperature. The solutions were kept in this manner for at least 72 hours. Approximately 10 ml of this saturated solution was withdrawn from the flask and filtered through a Gelman Acrodise disposable filter No 4184, 0.45 microns, into the preheated nucleation cell for each run. t s e ' n A new parent crystal was needed for each experimental run. Typically these parent crystals were 1 to 2 mm in size. A major obstacle that was overcome was the ability to grow such large crystals of dextrose. This was accomplished by carefully cooling a solution of dextrose to induce nucleation, then transforming a few nuclei to a slightly undercooled solution for growth. Parent crystals 71 were prepared at 308 and 323 K and alpha monohydrate (needle) and alpha anhydrous (prism) parent crystals were obtained, respectively. When they were grown to the desired size, the crystalswere removed from the solution, washed with distilled water and dried. From these crystals only those with good habit, as shown in Figure 2, were chosen for contact nucleation. Since needle type crystals were very thin and fragile, only prism type parent crystals were used for experiments. W The experiments were conducted by gluing a parent crystal to the movable rod with epoxy. Always a well grown face of the parent crystal was positioned to be in contact with the cover slip . In each experiment the nucleation cell was preheated 5°C above the saturation temperature of solution and the filtered saturated solution was added to the cell. The solution was heated again to destroy any nuclei that may have been generated by transforming the solution. This heating also slightly dissolved the parent crystal to insure that no surface irregularties formed by washing and drying remained. After about five minutes of heating the solution was cooled and the parent crystal was allowed to grow. When the parent crystal's growth was proceeding in a regular manner, it was slid across the glass plate to generate contact nuclei. The length of the slide was approximately 0.5 cm. Contact nuclei were observed after a few minutes and they were 72 photographed at timed intervals to follow their growth. Each run took about two to three hours. A 50X2 micron scale was also photographed in each run to calibrate the experiment. Data Analysis .The raw*data obtained from the contacting experiments consisted of a series of slides of the same sample of crystals from the general population. The slides were projected on a screen for enlargement without loss of resolution. The outline of each crystal was carefully traced on a piece of paper. This trace was analyzed by an image analyzer to determine the area of each crystal. The characteristic size of each crystal was taken as the square root of the area, or the geometric mean size. This is the value refered to as size or as equivalent size in the subsequent work. Laser Raman spectroscopy / Raman Microprobe Raman spectra were recorded with a Spex model 1406 double monochromator with radiation provided by Coherent Radiation model CR-S Ion laser. The laser beam was directed from the laser by two 450 prisms and two mirrors to the microprobe. The light was passed through the microscope objective to illuminate the sample. Inelastically scattered (Raman scattering) light from the sample was again reflected from a 450 prism and entered the spectrometer. A Model #R928/115 photomultiplier counted the photons and transmitted the signal to a NIM 73 adaptor. From this adaptor the signal was sent to a 7118 Strip chart recorder to record the spectrum. All samples! were illuminated with the 5145 2 line of the Ar+ laser. Laser power was approximately 1W at the laser. Bansn_snsstra_ef_slnsese Initially, the Raman spectra of of pure crystalline products alpha monohydrate (a product from A.E.Staley Manufacturing company) and alpha anhydrous (Sigma Chemical company product, No G5000) were recorded from 300 to 1500 cm-1_wavenumbers. It was found that the region from 772 to 1221 cm"1 (5360 to 5490 g) was suitable to identify these anomers. Several specta were recorded in this region to determine optimum operational conditions. Always a Ammonium Paramolybdate spectrum was recorded to align the ' instrument. Optimal conditions determined were a scan- speed of 0.05‘i/sec., a photon count rate 2* 106 cOunts per second, and a chart speed of 0.1 inch./sec. The following Raman spectra were recorded: 1. The pure crystalline product of alpha monohydrate and alpha anhydrous glucose. 2. The parent crystals (prism and needle). 3. A solution spectrum in situ at 314K at 4 degrees supersaturation. 4. Contact nuclei spectra at the same experimental conditions as in 3 at several locations in the cell after several hours.