'. A.- .Ad x... ._.‘.' . . . . «4034-; . :~I_.4:II--‘-=;I:1§¢F~ m." I {:3 .573" '. "Q4 . ~ ( ~53v.l- $2224.22? - r ‘4 ~‘v (“"n"; 's4,I. , 91f .‘ 5 F,” . J'- 34/. I 1%;5, "l‘ . I . 1’. g «4. 'A K'.‘ _ , v- ‘ }. (i::;{“ W“): \p‘K‘IJ'A "" [3/ "I | .1, ‘- 15", 4 4 . ,,| “#5:. } 1 4.4 _I >4 gi‘.’ ... I '1’.” 2.3 9- 2- :11!“- 14‘ w. «am am} :7“. f 4 ti!” :V A Fry .25. "$3133: (1’52‘ 2:12.1' { If? 0 I" Kzzf'4-4v . "I" "J. "‘ ' M" ' ’64”? . 1.1%”4} 3.6.1:},- I I O4 6.; 421,35 44:: . -‘~I‘ “ w is)" " ' {o1 “gwii’zzm ”I “I 4. éfi'I/u 4 ’32 . I. A". I” ‘t’r ‘HV 4 >4 I ‘/“"F"I"4a’.f‘ 0.... 1... m . .14.“, 4. 7.1 {J -. .4 I'; . .. t9: . r ’15.. 1.: (3’; .444: 2,; -oflu -\ .U 1 , 4 43:7; {.0 M’Vv I "5- . 4‘" 1, ~‘<'?~‘T’*-4 . fiflgk .3: $53 '[xl'UV' \' EUNIVERSH’Y ‘LIBRAFNEsi-.fl 1 111111111111 11111 111“ 11111 1111 111 1293 005509 LIBRARY-w1 Michigan State University This is to certify that the thesis entitled DEHYDRATION OF D-GLUCITOL USING Ca Y AND Pt/Ca Y ZEOLITE CATALYSTS presented by Gregory Alan Kudiac has been accepted towards fulfillment of the requirements for M . S . degree in CHE £1-14. 9 71141309., . V Major professor Date 717“ 1/518,” 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. . I' c.- DEHYDRATION OF D-GLUCITOL USING Ca Y AND Pt/Ca Y ZEOLITE CATALYSTS BY Gregory Alan Kudlac A Thesis Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemical Engineering 1988 ABSTRACT DEHYDRATION OF D-GLUCITOL USING Ca Y AND Pt/Ca Y ZEOLITE CATALYSTS BY Gregory A. Kudlac A process has been developed for the catalytic de- hydration of D-glucitol. This dehydration is achieved using Ca Y and Pt/Ca Y zeolite catalysts. The advantage of using these catalysts is that essentially complete con- versions of D-glucitol were obtained. In addition, the shape selectivity of the catalysts limited the number of products formed. Several batch reactions were carried out with deca- hydronaphthalene as a solvent under varying conditions which included a temperature range of 180-260 °C and a H2 pressure range of 160-900 psi. The catalytic activity observed was primarily dehy- dration and dehydrogenation, with some isomerization. Main products formed were 1,4:3,6-dianhydro-D-glucitol, 1,4-an- hydro-D-glucitol and 3,5,6-trihydroxy-3-hexen-2-one. To my parents, John and Patricia Kudlac, for their undying and boundless support throughout my education, and to my fiancee, Maureen Dorsey, whose emotional support has been instrumental in my success. ii ACKNOWLEDGEMENTS The author wishes to thank Dr. Dennis J. Miller for his support and guidance during all aspects of this project; Miss Lynn Frostman for her excellent HPLC analy- ses; and especially Mr. Douglas Gage and Ms. Betty Baltzer for their immeasurable assistance interpreting the mass spectrometry data. iii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . LIST OF FIGURES . . . . . CHAPTER 1. BACKGROUND AND INTRODUCTION A. Biomass Conversion Pathways . . B. Proposed Reaction Sequence C. Direct Conversion D. Hydrogenation of Carbohydrates . E. Dehydration . F. Choice of Catalyst and Reaction Conditions for Conversion of D—Glucitol to Hydrocarbons . . . . . . . . APPARATUS . . . . . . . . . . . . . A. Ion-Exchange/Metal Loading of Catalyst B. Calcination/Reduction of Catalyst C. Batch Reactions . . . .,. . . . D. Product Analysis via Gas Chromatography and High Performance Liquid Chromatography . . . . . . . . . E. Product Analysis via Gas Chromatography- Mass Spectrometry . . . . . . . . . . . iv Page vii viii ll 23 26 26 27 30 33 34 EXPERIMENTAL CONDITIONS AND PROCEDURES A. B. C. D. Catalyst Preparation A.l. Ion-Exchange/Metal Loading A.2. Calcination/Reduction . . Dehydration/Hydrogenation Reactions Product Removal . . . . . . . . Product Analysis . . . . . RESULTS AND DISCUSSION A. Preliminary Experiments A.l. Blank Runs . A.2. Hexanol and Hexanediol Experiments Experimental Results . . . . . . B.l. Results of Hexanol Dehydration Experiment . . . . . . . . . 8.2. Results of 1,2-Hexanediol Experiment . . . . . . . . 8.3. Results of Sorbitol Experiments . B.3.1. Analysis by High Performance Liquid Chromatography B.3.2. Analysis by Gas Chroma- tography-Mass Spectroscopy . Mass Spectroscopic Analysis of Experi- ments Eight and Six . . . . . . . . . C.1. Mass Spectroscopic Analysis of Experiment Eight . . . . . . . . C.1.1. Peak 558 . . . . . . . . . . C.1.2. Peak 756 . . . . . . . . . C.1.3. Peak 870 . . . . . . . . . . 36 36 37 39 41 43 45 47 47 47 50 51 51 54 56 58 62 67 67 68 71 72 C.2. Mass Spectroscopic Analysis of Experiment Six . . . . . . . . . . . 72 C.2.1. Peak 20:07 . . . . . . . . . 75 C.2.2. Peak 20:01 . . . . . . . . . 76 C.2.3. Peak 19:27 . . . . . . . . . 79 C.2.4. Peak 19:07 . . . . . . . . . 81 C.2.5. Peak 18:54 . . . . . . . . . 83 C.2.6. Peak 18:44 . . . . . . . . . 83 C.2.7. Peak 18:20 . . . . . . . . . 84 C.2.8. Peak 18:09 . . . . . . . . . 88 C.2.9. Peak 15:20 . . . . . . . . . 89 C.2.10. Peaks 24:20, 24:12 and 23:53 . . . . . . . . . . . . 89 D. Discussion . . . . . . . . . . . . . . . . 90 5. CONCLUSIONS . . . . . . . . . . . . . . . . . 98 LIST OF REFERENCES . . . . . . . . . . . . . . . . . . 100 APPENDIX A . . . . . . . . . . . . . . . . . . . . . . 103 APPENDIX B . . . . . . . . . . . . . . . . . . . . . . 134 vi TABLE LIST OF TABLES Types of Dehydration Reactions and Corresponding Catalysts (Taken from Reference (9) . . . . . . . . . . . . Variation of Platinum Dispersion with Mode of Preparation (Taken from Reference 24) Turnover Numbers N for Hydrogenation of Ethylene (Taken from Reference 24) . Catalysts Prepared Summary of Blank Runs, Hexanol and Hexanediol Experiments Summary of Sorbitol Experiments Summary of Water-Soluble Products . vii Page 12 22 24 38 48 57 94 FIGURE 1. 10. 11. 12. 13. 14. 15. 16. LIST OF FIGURES Page Biomass Conversion Pathways (Taken from Reference 1) . . . . . . . . . . . . . . . . . 3 Location of Sodalite Cages in Type A and Type Y zeolites (Taken from Reference 23) . . l8 Plot of Atomic Ratio H/Pt vs. Calcination Temperature (Taken from Reference 21) . . . . 20 Plot of Cyclohexane Conversion and Hydrogen Chemisorption vs. Calcination Temperature (Taken from Reference 21) . . . . . . . . . . 21 Sketch of Calcination/Reduction Reactor (Taken from Reference 25) . . . . . . . . . . 28 Sketch of Calcination/Reduction Process . . . 29 Sketch of Batch Reactor Process . . . . . . . 31 Gas Chromatography Traces for Hexanol Dehydration Experiment . . . . . . . . . . . . 52 Gas Chromatography Trace for 1,2-Hexanediol Experiment . . . . . . . . . . . . . . . . . . 55 High Performance Liquid Chromatography Trace for Experiment Six . . . . . . . . . . . 59 High Performance Liquid Chromatography Trace for Experiment Ten . . . . . . . . . . 60 High Performance Liquid Chromatography Trace for Experiment Eight . . . . . . . . . . 61 General Fragmentation Scheme for TMS Carbohydrate Ethers . . . . . . . . . . . . . 64 Rearrangement Ions of TMS Carbohydrate Ethers (Taken from Reference 27) . . . . . . . 66 Gas Chromatography Trace for Experiment Eight 0 O O O O O O O O O I O O O O O O O O O 69 Partial Fragmentation Scheme for 1,4-Anhydro- D-Glucitol (Taken from Reference 34) . . . . . 73 viii 17. 18. 19. 20. A-10. A-ll. A-12. A-13. A-14. Gas Chromatography Trace for Experiment Six . . . . . . . . . . . . Cis and Trans Hydroxyl Group Configuration for 1,4 and 3,6-Anhydro-D-G1ucitol . . Pinacol Rearrangement Reaction and Fragmentation Reaction Scheme for 3,5,6-Trihydroxy-3-hexen-2-one Formation Location of Ca Ions in Type Y Zeolite Structure (Taken from Reference 37) Mass Spectrum of Hexane from EPA/NIH Mass Spectral Data Base Mass Spectrum of l-Hexene from EPA/NIH Mass Spectral Data Base Mass Spectrum of 1-Hexanol from EPA/NIH Mass Spectral Data Base . . Mass Spectrum of Hexene from Hexanol Experiment . . . . . Mass Spectrum of Hexane from Hexanol Experiment . . . . Mass Spectrum of Hexanol from Hexanol Experiment . . . Mass Spectrum of Hexane from 1,2-Hexanediol Experiment . Mass Spectrum of Hexene from 1,2-Hexanediol Experiment . Mass Spectrum of Hexanol from 1,2-Hexanediol Experiment Mass Spectrum of Peak 558 . Mass Spectrum of Isosorbide Dimethyl Ether Mass Spectrum of Peak 756 . Mass Spectrum of Peak 870 . Mass Spectrum of Peak 20:07 ix 74 78 86 87 92 103 104 105 106 107 108 109 110 111 112 113 114 115 116 A—15. Mass Spectrum of Peak 20:01 . . . . . . . . . 117 A-16. Mass Spectrum of Peak 19:27, m/z 452 . . . . . 118 A—17. Mass Spectrum of Peak 19:27, m/z 380 . . . . . 119 A-18. Original Mass Spectrum of Peak 19:07 . . . . . 120 A-19. Mass Spectrum of Peak 19:07, m/z 380 . . . . . 121 A-20. Mass Spectrum of Peak 19:07, m/z 452 . . . . . 122 A-21. Mass Spectrum of Peak 18:54 . . . . . . . . . 123 A-22. Mass Spectrum of Peak 18:44 . . . . . . . . . 124 A-23. Mass Spectrum of Peak 18:20, m/z 362 . . . . . 125 A-24. Mass Spectrum of Peak 18:20, m/z 380 . . . . . 126 A-25. Original Mass Spectrum of Peak 18:20 . . . . . 127 A-26. Mass Spectrum of Peak 18:09, m/z 290 . . . . . 128 A-27. Mass Spectrum of Peak 18:09, m/z 308 . . . . . 129 A-28. Mass Spectrum of Peak 15:20 . . . . . . . . . 130 A-29. Mass Spectrum of Peak 24:20 . . . . . . . . . 131 A-30. Mass Spectrum of Peak 24:12 . . . . . . . . . 132 A-31. Mass Spectrum of Peak 23:53 . . . . . . . . . 133 B-1. Specific Ion Traces for Hexanol Experiment . . 134 B-2. Specific Ion Traces for 1,2-Hexanediol Experiment . . . . . . . . . . . . . . . . . . 135 B—3. Specific Ion Traces for m/z 452 and m/z 437 . 136 8-4. Specific Ion Trace for m/z 380 . . . . . . . . 137 8-5. Specific Ion Trace for m/z 247 . . . . . . . . 138 8-6. Specific Ion Traces for m/z 349 and m/z 347 . 139 B-7. Specific Ion Traces for m/z 290 and m/z 308 . 140 B-8. Specific Ion Trace for m/z 512 . . . . . . . . 141 CHAPTER 1 BACKGROUND AND INTRODUCTION A. BIOMASS CONVERSION PATHWAYS Investigation into the use of biomass materials as a potential source of energy has been both widespread and varied, and can be divided into two main areas of interest: 1) the direct use of biomass as an energy source, and 2) converting or thermally upgrading the biomass to a fuel which can be used as a substitute for fossil fuels. Combustion is the primary method in which biomass is directly converted to energy and simply involves burning the biomass, using air as the oxidant. The chemistry of complete oxidation of materials containing mainly carbon, oxygen and hydrogen is: CxHyOz (biomass) + (x + y/4 - 2/2) 02 ------- > x co2 + y/2 H20 where x, y and 2 represent the mean elemental composition of the biomass [1]. Incomplete oxidation of the biomass 2 yields carbon, carbon monoxide, hydrocarbons, and other gases as by-products and the heat of the reaction is reduced. Nitrogen and other impurities present in the biomass are converted to gases and ash. Conversion or thermal upgrading of biomass by a combination of heat and partial combustion yields a variety of solids, liquids and gases with at least some of the properties of coal, oil and natural gas. Major processes in this area include pyrolysis, gasification, liquefaction and hydrolysis. These processes, with the exception of hydrolysis, are related to each other as can be seen in Figure 1. Pyrolysis involves the heating of the biomass in the absence of an oxidizing agent such as 02 or air. Above 100 0C, the biomass begins to decompose and between 250 and 600 0C the main products are charcoal and an oily acidic mixture of tar and variable quantities of methanol, acetic acid, acetone and traces of other organic molecules. An empirical example of a pyrolysis reaction scheme is given below [1]: Charring reaction (C6H 05)n ------ > 6n C + 5n H O 10 2 E 9335mm 8136250 8285 4 653m 114 Dow: 11111 emmbm 524:5: A 28556 10% 2:638:38 * fl 1+ 1:0 >>bo¢>d E_6L>d 111111 Dom: 41111. cowumaasou mmmeoflm Oil Formation (C6H1005)n ------ > 0.8n C6H8O +1.8n H20 + 1.2n CO2 (where C6H80 is assumed as a mean composition of the oil) In a "flash pyrolysis" system, the biomass is heated rapidly and produces olefins (c611 05)n ------ > 2n C H + n H O + 2n CO . [1] 10 2 4 2 2 Gasification of biomass with oxygen yields a medium energy gas containing mainly carbon monoxide and hydrogen. The chemistry of gasification involves the reaction of pyrolysis char with steam or carbon dioxide to form synthesis gas [1]: C+HO ------ >CO+H C + CO ------ > 2CO Above 1000 0C, the only stable products are carbon monoxide and hydrogen whereas at lower temperatures ethylene, methane and other low molecular weight molecules are stable. In liquefaction processes, the biomass is converted to a liquid similar to heavy fuel oil by reacting it with carbon monoxide and hydrogen (generally obtained from a 5 separate pyrolysis or gasification process) in the presence of a catalyst under high temperatures and pressures. A typical liquefaction process is given by the Pittsburgh Energy Center [1]: 0.61(C6H9.1304.33) + 0.23 C0 + 0.08 H2 ------ > biomass H o + 0.64 co + 0.53 (C6H 2 2 6.9301.22) liquefaction oil The final process, hydrolysis, differs from the previous methods in that biomass materials such as cellulose, hemicellulose and starch are reacted only to monosaccharides such as glucose, fructose and xylose. This reaction proceeds according to the following stoichiometry: 1005)n- + n H20 ------ > n C6H1206' -(C6H The ease of this hydrolysis reaction varies considerably depending on the material used. Starches and pentosans (hemicelluloses) require relatively mild conditions using dilute acids and medium temperatures, whereas cellulose needs higher temperatures, stronger acids and pressurized reactors [1]. These monomeric sugars may then be fermented to ethanol and carbon dioxide using a variety of microorganisms. 6 As will be shown in the next section, these monomeric sugars may be further reacted to form polyols, or alditols, which represent the starting material for the proposed scheme of forming hydrocarbons from biomass materials. B. PROPOSED REACTION SEQUENCE The following reactions represent the proposed pathway for converting biomass materials to hydrocarbons via inorganic heterogeneous catalysts: Sugar Formation Biomass (cellulose) ----------- > Sugars and Oligomers (1) hydrolysis Alditol Formation H 2 Sugars (e.g. D-Glucose) ------- > Alditols (D-Glucitol) (2) metal catalyst Dehydration Reaction -HZO Alditol ----------- > "Polyhydroxylated Olefin" (3) acid catalyst Hydrogenation Reaction H 2 "Polyhydroxylated Olefin" ----------- > "Lower Polyol" (4). Pt Reactions (3) and (4) are carried out successively and can be summarized as follows: dehydration "Polyhydroxylated Olefin" ------------- > Hydrocarbons (5). hydrogenation In reactions (3), (4) and (5), "polyhydroxylated olefin" simply refers to a compound with the formula C6H1205 with the departure of one water molecule generating a carbon- carbon double bond or ring structure. In reaction (4), "lower polyol" refers to a compound with the formula C6H1405 and is the saturated form of the corresponding "polyhydroxylated olefin". The next two sections review literature dealing with the direct inorganic catalytic conversion of biomass and the various reactions occurring in the proposed pathway, except for the hydrolysis reaction which has been prev- iously discussed. C. DIRECT CONVERSION The papers reviewed in this section present processes in which the starting carbohydrate material is reacted directly to hydrocarbons in the presence of a catalyst. United States Patent 4,430,253 [2] describes a process in which a hydrogenation catalyst and a sulfide-modified 8 ruthenium catalyst are used sequentially under batch conditions. The first stage involves hydrogenating the starting material, such as glucose, to the polyol corres- ponding to the same chain length, i.e. sorbitol, in the presence of the hydrogenation catalyst. The second stage is carried out in the presence of the sulfide-modified ruthenium catalyst. This hydrogenolysis reaction results in the formation of shorter chain length polyols such as ethylene glycol and 1,2-propylene glycol. It is reported that by using the sulfide-modified ruthenium catalyst a greater range of temperatures and a higher degree of selectivity is possible when compared with a traditional hydrogenation catalyst. In United States Patent 4,401,823 [3] a catalyst comprising of "a carbonaceous pyropolymer possessing recurring units containing at least carbon and hydrogen atoms" and impregnated with a transition metal is used for the hydrogenolysis of carbohydrates to a variety of prod- ucts. In this process the catalyst used is prepared by treating an inorganic support material such as alumina with an organic pyrolyzable compound at high temperatures. This results in the decomposition and polymerization of the organic material on the surface of the inorganic support. This inorganic support material is then chemically leached from the pyropolymer. The pyropolymer is then impregnated with an aqueous solution of one of several transition metals. Use of this catalyst in reacting carbohydrate 9 compounds such as sucrose, glucose and lactose under batch conditions resulted in the formation of a variety of products including alcohols, acids, ketones, ethers, and hydrocarbons. Finally, United States Patent 4,503,278 [4] presents a process in which carbohydrates such as starch, cellulose and glucose are converted into hydrocarbons using a crys- talline zeolite catalyst such as ZSM-S. These catalysts, obtained directly from the manufacturer, were used to react the carbohydrates in a fluidized bed reactor and resulted in the formation of carbon monoxide, carbon dioxide, water, and a variety of hydrocarbons, primarily short chain a1- kanes and alkenes. In each of these patents, the carbohydrate feed con- sisted of an aqueous solution of the particular carbo- hydrate, except for [4] which employed both aqueous solu- tions of the carbohydrate itself and an aqueous mixture of the carbohydrate and methanol. D. HYDROGENATION OF CARBOHYDRATES The papers reviewed in this section discuss processes in which carbohydrates such as glucose and fructose are hydrogenated to the corresponding alditol (e.g. sorbitol). In United States Patent 4,380,679 [5] the "carbo- naceous pyropolymer" discussed in [3] (in this case the impregnated metal is limited to a Group VIII metal) is used 10 to convert glucose to sorbitol and mannitol. In a series of eXperiments, this group of catalysts provided very high conversion of the glucose in addition to a high selectivity to sorbitol versus mannitol. United States Patent 4,380,680 [6] presents a method for hydrogenating carbohydrates in which ruthenium, when dispersed on a-alumina, yields a hydrothermally stable hydrogenation catalyst. The use of this catalyst in the hydrogenation of glucose not only provided high conversion and selectivity of glucose to sorbitol, but also suffered minimal leaching of either the metal or support. United States Patent 4,382,150 [7] presents a catalyst for the hydrogenation of carbohydrates similar to that in [6] except that the support material used is titanium dioxide and the Group VIII metal is nickel. This catalyst provides the same conversion, selectivity and leaching characteristics as in [6]. In this method, however, it is also shown that the reduction temperature used in the preparation of the catalyst has an affect on the product distribution. For example, when a reduction temperature of 250 °C is used, 98 percent of the glucose used in the experiment reacted to yield 95 percent sorbitol and 3 per- cent mannitol. However, when a reduction temperature of 450 °C is used, only 63 percent of the glucose reacted yielding a product distribution of 44 percent sorbitol, 4 percent mannitol and 27 percent fructose. Finally, in United States Patent 4,413,152 [8] a 11 catalyst is presented similar to that used in [6] except that the support material in this case is titanated alum- ina. This catalyst also affords the same conversion and selectivity characteristics as those in [6] and [7]. In addition, the use of ruthenium instead of nickel results in a lower metal leaching level under hydrogentation condition. As in section 1.3, all of the carbohydrates used in [5-8] are in aqueous solution and are reacted over fluid- ized beds except for [5] where the experiments are con- ducted under batch conditions. E. DEHYDRATION The next two reactions in the proposed sequence, de- hydrating the alditol and hydrogenating the resulting product(s), represents both the major step in the proposed reaction sequence and the starting point of this research. The papers reviewed in this section discuss the dehydration of alcohols over solid catalysts, and in particular, zeo- lites. Since the hydrogenating capabilities of Group VIII metals such as platinum is well-documented in the litera- ture, it will be discussed only briefly. As can be seen in Table 1 [9], numerous studies have been done concerning the dehydration alcohols and diols over solid catalysts. The general reaction scheme for the dehydration of alcohols is given below [10]: 12 TABLE 1 TYPES OF DEHYDRATION REACTIONS AND CORRESPONDING CATALYSTS [9] Reactants Catalysts Products Alcohols Mothmml A130; . A150 810: or (limcthyl other 3(:\l«-ohol‘. l'illmlml 'Illizlliul l-l'ropuool (pi-11px] alcohol) l-l’l'opuuol 3-l'1‘opuonl (l.\(!))l()||_\'l :Lll'UllHll 2-1\'lol.h'\'l-l-pi'op:1.11ol (isobul-yl :tlcoliol) 2-Mctl1yl-‘2-prop:mol (tort-liutyl :Llcoliol) l-liutzuiol l-Butnuol l-Butuuol 2-Butzuiol (soc-liutyl alcohol) Methyl hutuuols 2-Mcthyl-2-hutauol (tcrt-muyl alcohol) 3-Mcthyl-l-hutauol (isoamyl alcohol) l-l’cut:uml (:unyl :Ll- cohol) CT103 or A130; :Lc- tivulcd with oxides ofg1‘oupsl,0,7 \K’. S synthetic uwllizuiol catalyst. :\l-:(),1. Also :\l:(),1- SK); ;'l‘i()g ;Crg()v;; “EU; ;Si()~_- ; 7.11): AIL-(l; ; SiU-g. .-\l.\'o 1’._.o,_ ; 7.111., ;\\'...o,; 'l‘iU: ; l\' -_-CU1 ; [\l:()1 With CM), NiO or (31'3“, ; 1 'l‘h()—_.; .\l;,:1(l’()a);- on coho; ll,1l’(l1; mo|_\'lulilvs (\l3(l:i Al-_.(),, Al-_'()J . Also 74M); (l:i1(l'(),): ; :lS(‘.:l|l- itc clay pumice. Also A130, pumice. Also All’U. ; Al silicate; Al2(5()i): Al203 . Also Fed)J + Ale; ; SlU: + 'l‘iOg ; N33POJ + bauxite llgl’O. + A130, Al3()3 A130; A120, A1,o. Alg(); . Also A|203 + additives Al:()3 . Also All’O. liiglior primary (11‘ .\'(‘(‘- omlzu'y alcohols (21 her ('1 hylouo (llpl'nll_\'l ('llu‘r pi'opylouv isopropyl NIH-r + pro- pylmu: 2-mctl1yl-propmue (iso- hutylcuc) + l-l)lll-(‘.ll(‘ 2-mcthyl—prmicuc l-hutcnc (a-hutylcuc) 2-nicthylpromnc l-butcnc ' + cis-2-bu- tcuc + trans-2-hutcuc (pseudoliutylcncs) l-hutcnc + cis-2-hutcnc methyl butcncs 3-mcthyl-2-butcuc (B-isonmylcnc) isozunyl cthcr + :unylcuo. dipcntyl cthcr + pcu- tcnc 13 <2’I/z’ether + H20 ;:::5 2 alcohols 1L olefin + alcohol +H 0 2 dehydration In addition: The change in selectivity of the catalyst with gross structure and mode of preparation affords valuable in- sight into the mechanism of dehydration, dehydrogena- tion, ether formation, and alkene isomerization. De- hydrogenation is favored at the expense of dehydration in primary and secondary alcohols by methods that in- crease the crystal size, decrease the surface area, and reduce the catalyst porosity. For example, the dehydrating activity of alumina is decreased when it is heated to high temperatures, a process that causes healing of the irregularities in the crystal lattice. This observation suggests that dehydrogenation is a surface reaction but dehydration occurs within the pores of the catalyst [12]. As an example of this, Weisz and Frilette [13] studied the dehydration of isobutanol in the presence of 1-butanol. They found that an amorphous aluminosilicate catalyst de- hydrated both alcohols at comparable rates whereas a Ca A zeolite selectively dehydrated the 1-butanol. The above information suggests that zeolites, which are crystalline aluminosilicates, highly selective and thermally stable, would be excellent candidates for de- hydration catalysts. Rudham and Stockwell [14] investigated the dehydration of 2-propanol on HY zeolites. Using a continuous-flow reactor under a variety of temperatures, pressures, and extents of ion—exchange (the zeolites were prepared by ion- 15 exchanging NH for Na and then deammoniating to the HY 4 form), 2-propanol was dehydrated to both propene and di- isopropyl ether. Ballantine et al. [15] studied the dehydration of a variety of alcohols over sheet silicates (ion-exchanged montmorillonites) under heterogeneous conditions. In contrast with homogeneous acid catalysts, where dehydration to the corresponding alkene is the primary reaction, pri- mary alcohols were found to undergo an intermolecular de- hydration reaction to the corresponding di(alk-l-yl) ether with very little formation of the corresponding alkene. Secondary and tertiary alcohols were dehydrated by the normal intramolecular process to the corresponding alkene with very little ether formation. Chong and Curthoys [16] investigated the vapor-phase decomposition of ethylene glycol over a cation-exchanged mordenite catalyst using a constant flow microreactor system. At atmospheric pressure and using a temperature range of 230 to 500 0C, the main products of the reaction were acetaldehyde, water, and glycoaldehyde (HOCHZCHO) with dioxane, ethanol, glyoxal (OHC-CHO) and ethylene as minor products. Minachev et a1. [17] used a CaY zeolite in their study of the dehydration of alcohols. Using a flow-reactor sys- tem, they found that varying the flow conditions the extent and direction of the reaction could be varied. For ex- ample, at 250 oC and a flowrate of 9.8 moles/liter * hour, 16 ethanol was dehydrated to an extent of 60% and diethyl ether was found among the reaction products. At a flowrate of 3.8 moles/liter * hour, however, all of the ethanol was converted to ethylene at the same temperature. Similar results were found with other primary and secondary alco- hols. In a similar study using CaNaY zeolites [18], Sharf et. al. determined that 1-pentanol was dehydrated faster when the calcium content was increased. Numerous other studies involving the dehydration of alcohols over zeolites have been conducted [19]. As mentioned previously, the hydrogenating activity of Group VIII metals such as platinum is well-represented in the literature. This catalytic activity has been found to depend on two major properties: the dispersion of the Pt metal on the catalyst and the H/Pt ratio observed during the adsorption of H both of which can be affected by the 2, mode of catalyst preparation. It was originally observed by Rabo et a1. [20] that a CaY zeolite ion-exchanged with 0.5% Pt gave a H/Pt ratio of 2 which corresponded to an atomic dispersion of Pt on the zeolite surface. Several subsequent papers, however, have observed that a majority of the atomically dispersed Pt loses its ability to chemi- sorb hydrogen. Performing hydrogen Chemisorption studies on several zeolite-supported Pt catalysts, Kubo et a1. [21] observed that while for some of the catalysts the data indicated particle sizes of 1300 A, electron microscopy data revealed most of the particles were actually 15 A in 17 size. This indicated that a large portion of the Pt not detected by E. M. (scanning limit = 10 A) had lost its ability to chemisorb hydrogen. Similar observations were made by Gallezot et al. [22] using a PtNaNH Y zeolite 4 pretreated in 0 at 600°C and reduced at 300°C. A H/Pt 2 ratio of 0.25 was found for this catalyst while X-ray methods showed that the majority of the platinum was atom- ically dispersed in the sodalite cages of the zeolite (see Figure 2 [23]). As with the previous paper, it was con- cluded that these Pt atoms were not able to chemisorb hy- drogen. It was shown experimentally that as the zeolite was evacuated from 300 to 800°C, the Pt atoms migrated out of the sodalite cages and sintered into 15 to 20 A part- icles while increasing the hydrogen chemisorption ratio from 0.25 to 0.65. These data have not as yet been definitely explained, but four interpretations have been proposed [22]: l) at least 2 Pt atoms are required to dissociate molecular hy- drogen, 2) atomically dispersed platinum loses its metallic properties including chemisorption capability, 3) hydrogen chemisorption is inhibited because a partial electron transfer occurs from Pt atoms to electron acceptor sites of the support, and 4) molecular hydrogen cannot enter the sodalite cages because its kinetic diameter is 2.89 8 compared to the 2.2 A aperture of six-membered oxygen rings. 3Q moiooN > 33. pee < 33. E 8me dam—meow co :2803 .m oSwE 19 Two examples of the effect of catalyst preparation on metal dispersion and catalytic activity are discussed. A two-part study on platinum particle distribution and cat- alytic activity of zeolite-supported Pt catalysts was done by Kubo et al. [21]. In the first part, several PtNaY and PtNH4Y catalysts were prepared by varying the calcination temperature. The Pt particle size distributions and H/Pt ratios were then measured for each of the catalysts. It was found that with increasing calcination temperature, the Pt particle size also increased: the effect was more prom- inent for the PtNaY catalyst. It was also shown that the H/Pt ratio increased to a maximum at 3000C and then de- creased for the PtNaY while the ratio generally increased for the PtNH4Y (see Figure 3). For the catalyst calcined at 300°C, it was also shown that the majority of the Pt particles existed on the surface of the catalyst. The second part of the report investigated the cat- alytic activity of these catalysts in addition to that of a PtCaY zeolite for the dehydrogenation of cyclohexane and the hydrodemethylation of toluene. The catalytic activity of these zeolites was found to correspond very closely to the H/Pt ratio previously discussed (see Figure 4). Dalla Betta and Boudart [24] investigated the Pt dis- persion and catalytic activity for PtY zeolites prepared following several different procedures.~ Table 2 presents the results of the Pt dispersion study and shows that 20 OJ 0 l H/Pt value ><10~11 to O 1 10»— ‘1 l l l- L [00 2'00 300 400 500 600 Calcination tcm pcraturc (°C) —O— Pt-NaY type catalyst. "G“ Pt-NHA’ type catalyst. Figure 3. Plot of Atomic Ratio H/Pt vs. Calcination Temperature [21] 21 40 - -6 /(§ X10 6 so «3 ,3 3. i .. \O Q L.) 8) an o. 3 H o 3.. ° E O v "c L... U 0 20»- ‘ «2 “E c: {3, E > 22’ g u U 0 :13: 10»- ‘ “l o 0 i r i ICC 200 500 400 600 Calcination temperature (°C) Figure 4. Plot of Cyclohexane Conversion and Hydrogen Chemisorption vs. Calcination Temperature [21] 22 0004 3 NE rem cm of +2... 3; or; 2.0 loooeam NE ”com... am No m cool. 3 NI ”em as .o as .o 3 .o .6 omm + :3... .003 am No m 3 .o 3 .o 3. .o r83 mm Sr 066.... mm o m; .1 f .1 i .o 003. cm mm ”comm rm NO 535 I S .o a: .0 002. am Mr “comm cm .086 m mo .0 so .0 so .0 comm cm N: 4 280355 E m: 1:3 Go E. 8333 ohm MO commHonHQ Cowumhmmohnm 3.3 ZC_._./..N_ N mum/C. 23 calcination in 02 prior to H2 dispersion and H/Pt ratios. The two columns under ’Dis- reduction gives higher Pt persion of Pt’ indicate the references for the methods used in this determination. Table 3 presents the catalytic results for the hydrogenation of ethylene. As can be seen from the data, the catalysts with the higher H/Pt ratio have higher catalytic activities. Also, those catalysts that have been ion-exchanged show higher activities than those that have not (PtCaY versus PtNaY). F. CHOICE OF CATALYST AND REACTION CONDITIONS FOR CONVERSION OF D-GLUCITOL TO HYDROCARBONS All of the papers reviewed in this chapter lead to several main factors to be considered when deciding on the catalyst to be used for the conversion of D-glucitol to hydrocarbons: 1) the catalyst must be able to carry out both dehydration and hydrogenation reactions, 2) these reactions should be carried out so as to minimize the num— ber of products obtained, i.e. the catalyst should have some degree of selectivity, 3) the catalyst should be thermally stable to withstand both the reaction and prep- aration temperatures, and 4) the catalyst must be able to selectively admit reactant to the catalytic sites while in- hibiting the solvent’s admission. Analysis of these fac- tors in combination with the information presented in the above papers led to the choice of a highly Ca-exchanged Y zeolite loaded with platinum as the catalyst for the 24 10111.6 9.16 1aNo1mtn1 s8 .6 m .om v .1 3 wmmim exam .0 m .3. 1. .1 3 $1215 .3 .o o .3 1.... .1 1a sword 1.1.1.1.. .o 1..., .m 11. .1 smwazbd .1.va Tm m1: V31 1180: #1111 191188 3N. mmerwIHm n10 ZOFm MOm 2 358232 mm>OZ§ m mimicry. 25 proposed reaction scheme. The immediate goal of this research was to determine the applicability of the catalyst proposal. Because of this goal, identification of any reaction products was a primary objective. Batch conditions were chosen because higher conversions of sorbitol and thus higher concen- trations of products could be obtained. The choice of decahydronaphthalene as a solvent for the reaction system was based on three characteristics. First, the use of a hydrocarbon solvent as opposed to water favors dehydration due to the absence of water in the system. Second, its size inhibits its entrance to the interior catalytic sites of the zeolites, thus minimizing solvent degradation products. Finally, its use as a hydrogen transfer solvent in coal liquefaction indicates excellent H solubility. This results in large amounts of 2 H2 being available for hydrogenation activity. CHAPTER 2 APPARATUS This chapter describes the equipment used in each phase of the experimental research. Sections A and B de- scribe the equipment used in catalyst preparation, Sec- tion C describes the apparatus for the reactions, and Sec- tions D and E describe the equipment used in analyzing and identifying the reaction products. A. ION-EXCHANGE/METAL LOADING OF CATALYST The equipment used for the ion-exchanging and metal loading consisted of standard laboratory equipment and included a magnetic stirrer/heater, a condenser, several Erlenmeyer flasks of various sizes with ground glass fittings, an adapter to connect the flasks to the con- denser, and a vacuum filtration flask, Bfichner funnel and No. 42 filter paper for recovery and washing of the catalyst. Miscellaneous items such as a water wash bottle and tubing are used extensively in all of the experimental methods and are not listed. 26 27 B. CALCINATION/REDUCTION OF CATALYST The reactor used for calcination and reduction procedures was designed by Treptau and Miller [25]. Fig- ures 5 and 6 represent the reactor and process, respect- ively. The reactor consists of an insulated stainless steel pressure vessel equipped with a coil heater and two thermocouple probes. The gas inlet and outlet also double as entrances for the power leads and thermocouple probes, respectively. Gas enters the reactor through the inlet in the cover of the reactor and flows down the annular space between the heater and the insulation. It then flows up the inside of the reactor tube and through the outlet in the cover. The height of the heating element is approximately three inches so that to insure uniform heating the catalyst sample must be below the top of the element. Glass wool is used above and below the catalyst to minimze material loss during operation. For the process, two three-way valves are used down- 2, N2 and compressed air cylinders. These three-way valves allow for easy switching between the diff- stream of the H erent gases at different times during the experiment, while keeping the H flow separate from the N and compressed air 2 2 to minimize the risk of an explosion. Each gas cylinder is equipped with a regulator and valve; H was passed through 2 a molecular sieve filter before entering the reactor. A 28 63s lnlet and Gas Outlet and Power Leads Thermocouple Probes \ / Cover 153:1 L111"; --.LIL..1 Flange Alumina Reattor Tube Insulation Heater a 6.0.0... \ Stainless Steel Pressure Vessel Thermocouple Z L//// Probes : . Figure 5. Sketch of Calcination/Reduction Reactor [25] 29 LoEoEBoE 03215 \J mmuooi :ouoseoéeozmfioiu Co 11985 .© Emmi 111m boom 9 15> 888m comouosm 121E 26% 5111100102 coonZ e>1e> 18:50 32.1 i 5‘ 8:3 031500865. one museum 30 needle valve is used downstream of the reactor to control the gas flow rate and two other valves are further down— stream to switch the flow between the bubble flowmeter and the vent tubing. All tubing is stainless steel except for the tubing immediately before and after the reactor, which is copper. An Omega programmable temperature controller is used for controlling themperature and heating rate and an Omega multichannel temperature indicator is used to show the temperature from the two thermocouples in the reactor. C. BATCH REACTIONS All of the D-glucitol conversion reactions were carr- ied out in the reactor system illustrated in Figure 7. The reactor used is a Parr Series 4561 300 ml High Pressure Mini reactor. The reactor is fitted with a re- movable glass liner which reduces its working capacity to 225 ml. The reactor is equipped with a magnetically sealed 0 agitator drive and is rated to 3000 psig pressure at 25 C and a maximum temperature of 350 OC. The reactor agitator speed, heating rate and temperature are controlled by a single control box which features two heating rates, con- tinuously variable agitator speed and a maximum temperature set point of 400 0C. All fittings on the reactor are stainless steel with the exception of the thermocouple which is a Type J 31 0:5 Lesson wczmom 951m E158 > .a 683 o. 15> N. 388m 568% seam .10 11886 .N. 211mm . 51831 .501 1l®r £11839... meisam meU :omBPAI m cowouzz 63.5 6:6:me one; 033088.129 3:5 .83 3 wczooo 32 (iron-constantan) thermocouple. External fittings include gas inlet and outlet valves, a liquid sampling valve, a rupture disc (bursting pressure of 2037 psig at 72 oF), water connections for the cooling jacket for the magnetic drive seal, and a pressure gauge with a 0-2000 psi range for working pressures up to 1400 psig. Internal fittings include a cooling coil, four-blade agitator, and sampling tube inside the reactor. The interior cooling coil was removed for the experiments since it interfered with the insertion of the glass reactor liner. All tubing in the process is stainless steel except for the line leading to the vacuum pump which is copper attached to rubber vacuum tubing. All tubing connections are stainless steel Swagelok fittings. Regulators and valves are attached to the H2 and N2 cylinders. Tubing from the regulator valves leads to a three-way valve. High-pressure flexible hose supplied with the reactor runs from the three—way valve to the gas inlet valve on the reactor. Tubing from the gas outlet valve leads to a four- way connector. As can be seen in Figure 7, a gas sampling apparatus, vacuum pump, and an exhaust line are attached to the four-way connector. The gas sampling apparatus con- sists of a pressure gauge, a gas sampling cylinder, a sep- tum holder for syringe sampling and a valve. This appar- atus was not used in this phase of the project. A needle flow control valve used for varying the flow from the reactor is located in the exhaust line. This exhaust line 33 is vented directly into the fume hood exit. Similarly, a shutoff valve is used in the vacuum pump line. The vacuum pump is a Marvac model A20 rotary pump with a minimum attainable pressure of 10 microns. A Plexiglas shield is constructed around the reactor to contain projectiles in case of a catastrophic failure of the pressure vessel. During operation, this shield and the Plexiglas fume hood door were both closed. Also, the fume hood was operating during all phases of the experimental procedure. Other safety precautions included the rupture disc previously mentioned and the operating procedure of purging the reactor with N for 5 minutes to minimize H 2 2 and 02 contact which could result in an explosion. D. PRODUCT ANALYSIS VIA GAS CHROMATOGRAPHY AND HIGH PER- FORMANCE LIQUID CHROMATOGRAPHY The gas chromatograph used is a Varian Model 3700 outfitted with a 6 foot x 1/8 inch stainless steel 3% OV-101 packed column. All gas chromatography results, unless otherwise indicated, were obtained on this column. A Flame Ionization Detector (FID) on the GC is used for data generation. Carrier gas (He), H2, and air flow rates for the FID were 30 cc/minute, 30 cc/minute, and 300 cc/minute, respectively. The results were recorded using a Hewlett-Packard 3390A Integrator. The HPLC (High Performance Liquid Chromatograph) is a Waters Model 600 Multisolvent Delivery System Chromato- 34 graph equipped with an Aminex Ion Exclusion HPX-87H Organic Acid Analysis Column and 2 guard columns. The detector used is a Waters Model 410 Differential Refractometer. Data was collected by an IBM XT Personal Computer and an- alyzed using the WIRC data acquisition package supplied by Waters. The solvent used for the analysis was 0.01N sul- furic acid. The system was operated isochratically with a 336 psig back pressure and a flow rate of 0.5 ml/minute. E. PRODUCT ANALYSIS VIA GAS CHROMATOGRAPHY-MASS SPEC- TROMETRY Three mass spectrometers located in other laboratories were used in analyzing the reaction products. A Hewlett-Packard 5985 Quadrupole Gas Chromatograph- Mass Spectrometer located in the Biochemistry Mass Spec- trometry Department was used for the hexanol and 1,2-hex- anediol experiments. This gas chromatograph was equipped with a 21 meter DB-5 Megabore capillary column. A temp- erature program of 40-140 0C at a rate of 5 OC/minute was used for the hexanol and 1,2-hexanediol analyses. This machine was also used for analysis of Experiment Eight but was equipped with a 6 foot 3% OV-l packed column. For Ex- periment Eight, a temperature program of 55—225 0C at a rate of 5 oC/minute was used. A Finnegan 2000 Gas Chromatograph—Mass Spectrometer located in the Chemistry Mass Spectrometry Department was also used in the analysis for Experiment Eight. This gas 35 chromatograph was equipped with a 6 foot 3% OV—101 packed column. The same temperature program used for Experiment Eight on the Hewlett-Packard instrument was used for this machine, however a 3 minute initial hold was added. For Experiment Six, a JEOL JMS HXllO Double-Focusing Gas Chromatograph-Mass Spectrometer located in the Bio- chemistry Mass Spectrometry Department was used. This gas chromatograph was equipped with a 50 meter ULTRA-2 DB-5 capillary column. The same temperature program used on the Hewlett-Packard was used on this machine for Experiment Six. CHAPTER 3 EXPERIMENTAL CONDITIONS AND PROCEDURES This chapter discusses the experimental procedures used in preparing the Pt/Ca Y zeolites, running the reactions, and removing the products from the reactor in preparation for analysis. A. CATALYST PREPARATION Two forms of Y zeolite, approximately 100 grams each of Y-62 and Y-82, were obtained from Union Carbide Corpor- ation. The Y-62 form of the zeolite is a NH4-exchanged form of their primary Y zeolite, Y-52 which is a Na-Y zeolite. The Y-82 of the zeolite is a NH4-exchanged form of their steam-stabilized zeolite, Y-72. The advantage of using this zeolite is its increased thermal stability re- sulting from steam treating. Because the expected reaction temperatures were well below the decomposition temperature of the zeolite and because of the higher cation content (12.3 wt. % for the Y-62 vs. 4.2 wt. % for the Y-82) which results in a higher Ca-exchange and Pt-loading potential, the Y-62 form of the zeolite was chosen for the 36 37 experiments. As can be seen from Table 4 several different Pt/Ca Y zeolite catalysts were prepared. A.1. ION-EXCHANGE/METAL LOADING When received from Union Carbide Corporation, the Y zeolite was in the NH4 form and thus had to be modified through two ion-exchange processes in order to achieve the desired Pt/Ca Y form. The first of these processes in- volved contacting the zeolite in a calcium nitrate solu- tion. During this contacting, a majority of the NH4 ions present in the zeolite structure were exchanged with the Ca ions from the solution. Several contacts with fresh solution were used to achieve as complete an exchange as possible. The second of these processes involved loading the Ca Y zeolite with Pt to achieve the final catalyst. This loading was achieved by contacting the zeolite with a sol- ution of tetraamineplatinum (II) chloride (Alfa Products, 55.69 % Pt) under the same conditions as the Ca ion- exchange. In this case, however, only one contact was needed since the Pt loading was essentially quantitative. The ion-exchange/metal loading procedure used was as follows: The following procedure is for 10 g of Y—62 zeolite. Dissolve 135.04 9 of calcium nitrate (Baker Chemicals, 99% 38 TABLE 4 CATALYSTS PREPARED Calcination/ Calcination/ Catalyst % Pt Amount Reduction Temp. Reduction Time Ca Y-62 5.0 5.0 9 300/500 00 4 hrs./4 hrs. Ca Y-62 0.5 5.0 g 300/500 00 4 hrs./4 hrs. Ca Y-62 0.0 5.0 g 300/500 00 4 hrs./4 hrs. Ca Y-62 0.0 5.0 g not calcined/reduced Ca Y-82 5.0 5.0 g 300/500 00 4 hrs./4 hrs. Ca Y-82 0.5 5.0 g' not calcined/reduced 39 purity) in 500 ml of deionized water. The amount of cal- cium nitrate and water is based on five zeolite/solution contacts and a 500% excess of calcium nitrate. Place 10 g of zeolite and 100 ml of deionized water in an Erlenmeyer flask. Place the flask on a heater/stirrer and heat under reflux for one hour to degas the zeolite. Slowly add 100 ml of the calcium nitrate solution (over a period of two hours) to the zeolite slurry to ensure uniform distribu- tion. Continue heating under reflux for 2 hours. Filter the mixture while hot and wash with 500 ml of deionized water. Dry the zeolite at 120 0C for two hours then repeat contacting and drying procedure four times to maximize Ca loading. For the metal loading, place the zeolite in twice its weight of deionized water. To produce a 5 wt. % zeolite catalyst, dissolve 0.9155 g of the tetraamineplatinum (II) chloride mentioned earlier in 300 ml of deionized water. Add all of the salt solution and heat under reflux using the same procedure as for the ion-exchanging. Filter, wash and dry the catalyst as before. When loading Pt, only one contacting with the Pt sol- ution is needed. A.2. CALCINATION/REDUCTION To obtain the final catalyst, the zeolite was calcined and then reduced so that the catalyst would remain stable 40 at elevated temperatures. The calcination step was run at an elevated temperature under air to drive off ammonia and adsorbed water. The reduction step was run at an elevated temperature under hydrogen to reduce Pt to its zero-valent state. All calcination/reduction procedures were performed in the reactor designed by Treptau and Miller [25]. The calcination/reduction procedure used was as follows: Place plug of glass wool in bottom of inner ceramic reactor tube (0.75 in. 0.0. and 0.5 in. 1.0.) and load with catalyst to a depth of three inches (the height of the heating element). This depth corresponds to approximately five grams of catalyst. This is done to ensure even heating of the catalyst during the reactions. Place another plug of glass wool in top of tube and secure to reactor head, making sure thermocouples are properly inserted into the catalyst (thermocouples should not be touching). Seal the reactor and attach all tubing, thermocouple and power supply wires. Turn on controller and temperature indi- cator. While controller is warming up, establish an air flow rate of 30-40 cc/minute by calibrating with the bubble flowmeter. Once controller has stabilized, heat reactor to calcination temperature of 300 oC and hold for four hours. After four hours have elapsed, stop the air flow and purge reactor with N for 15 minutes at the same flowrate. 2 Purging is done to minimize contact of H2 and 02 which 41 could result in an explosion. Upon completion of the N2 purge, establish a 30-40 cc/minute H2 flowrate. Set the controller to the reduction temperature of 500 OC and run for another four hours. Turn off power to the heater, then turn off the H2 flow and purge the reactor with N2 for 15 minutes. Let cool to room temperature. Remove cat— again alyst from reactor and store in sealed sample bottles. The calcination and reduction conditions used in these ex- periments were reported by Kubo et al. [21] as giving both the most uniform Pt dispersion and highest catalytic activity. B. DEHYDRATION/HYDROGENATION REACTIONS All D-glucitol conversion reactions were carried out under batch conditions in the Parr microreactor system previously described. During the entire procedure the fume hood was operating. The reaction procedure used was as follows: Weigh out 2.00 grams of reactant and 1.00 grams of catalyst into glass reactor liner. Add 125 ml of Decalin (decahydronaphthalene: Alfa Products, 99% purity), place liner in reactor and seal. Connect the gas and cooling water tubing and the thermocouple wire. Set N2 regulator to approximately 20 psi. Referring to Figure 7 in Chapter? 2, open valves 1 and 3, and slowly open valve 4 to bring 42 reactor to the regulated pressure. Open valve 5 and slowly open valve 7 until a small amount of bubbling is heard inside of the reactor. Do not open the valve too far be- cause excessive gas flow will result in reactant mixture overflowing into the space between the liner and reactor wall. Purge the reactor for five minutes. Close valve on N2 cylinder and then close valves 1, 3, and 4. When reactor reaches atmospheric pressure, close valves 5 and 7. To evacuate the reactor, turn on the vacuum pump and open valve 6. Slowly open valve 5 and evacuate the reactor for five minutes (pressure gauge on reactor will deflect slightly in the negative direction). Close valves 5 and 6 and turn off vacuum pump. Set H2 regulator to approximately 80 psi and open valves 2 and 3. Very slowly open valve 4 to bring reactor up to 80 psi (this step must be done with care because of the large pressure difference between the reactor and the gas inlet line which can result in overflowing as before). Once reactor is pressurized, open valves 5 and 7 in same 2. Let H2 flow continue for five minutes then close valves 7 and 5 in manner as when purging reactor with N order. Next, pressurize reactor to the desired starting H2 pressure using the reactor gauge reading. Finally, close valves 4, 3, and 2 in order. Raise the heating mantle around reactor, start the cooling water flow to stirrer drive seals, and turn on the power to controller box. Close the inside Plexiglas door. 43 Turn on the stirrer motor then set speed by turning knob clockwise to the second setting. Set controller to desired reaction temperature. Turn heater on HIGH until temperature is 40 oC below desired temperature then switch heater to LOW and close hood door. Once the reactor reaches set temperature, run batch reaction for desired time. After reaction has been allowed to progress for the desired time, turn the heater to OFF, lower the heating mantle, and let the reactor cool to room temperature (the stirrer should be left on until temperature is 100 oC and the cooling water left running until temperature is around 70 0C). Once the reactor has cooled, open valves 5 and 7 and vent the reactor through the hood until atmospheric pressure is reached. C. PRODUCT REMOVAL Once the reactor was vented, the reaction products were ready to be removed. Deionized H20 was used to rinse out the reactor, wash the catalyst, and remove any water- soluble products from the reaction mixture. Rubber gloves were worn during this step to minimize contact with the reaction mixture. This step was also carried out under the hood to minimize inhaling any potentially harmful vapors. The product removal procedure used was as follows: 44 Fill a one liter volumetric flask with deionized H20. Place approximately one-half of the water in a wash bottle. Remove the glass liner from the reactor and pour the reac- tion mixture into a two liter beaker. Holding the reactor head over the beaker, rinse stirrer, liquid sampling tube, and thermocouple until clean. Next rinse the glass liner, using a spatula to dislodge solid particles from liner walls. The beaker should now contain the organic phase and approximately one-half liter of an aqueous phase. Using a piece of Tygon tubing as a siphon, remove almost all of the aqueous phase and catalyst (lower phase) to a second beaker. This is done to facilitate filtering and washing of the catalyst, since the organic phase does not pass through the filter paper very well. Decant or- ganic phase into a third beaker and pour remaining organ- ic/aqueous mixture and any remaining catalyst into a grad— uated cylinder. By using a graduated cylinder almost all of the organic phase may be removed and thus minimize the amount having to be filtered. Using a vacuum filtration flask, filter the aqueous phase to recover the catalyst, making sure to wash the sides of the beaker and graduated cylinder holding the aqueous phase into the filter funnel. Wash the catalyst with the remaining deionized H20. Pour the aqueous and organic phases back into the two liter beaker, place on a magnetic stirrer, and mix rapidly for 10 minutes to thor- oughly contact the organic phase with the aqueous phase 45 This will extract any water-soluble products. After mix- ing, remove the beaker from the magnetic stirrer, let phases separate and take samples of each. While the phases are separating, thoroughly wash catalyst with acetone to remove residual organic phase and dry at 120 oC overnight. D. PRODUCT ANALYSIS Since the water-soluble products are essentially non- volatile, trimethylsilyl derivatization was used to vola— tilize these compounds for both GC and GC-MS analysis. The derivatization procedure used was as follows [26]: Place 10 ml of the aqueous phase in a small (25 ml) Erlenmeyer flask. Attach a needle to one end of a piece of Tygon tubing and the other end to a N regulator fitted 2 with a tubing adapter. Place the tubing in a clamp on a ring stand so that the end of the needle is approximately one-quarter inch above the surface of the sample. Using a regulator pressure of 8-10 psi, establish a flowrate such that a small depression is formed in the surface of the aqueous sample by the N Dry overnight 2. until all the water is evaporated. Using a stirrer/heater on the lowest heat setting in combination with the N2 flow will expedite drying. Once dry add 1-2 ml of pyridine to the Erlenmeyer flask to act as a solvent. When residue is dissolved, place 46 solution in a small, screw-top sample bottle. Add 150 ul of BSTFA (N, O-bis(Trimethylsilyl)trifluoroacetamide, Pierce Chemical) directly to the solution. Let sample stand for 15 minutes, swirling occasionally. Sample is now ready for GC or GC-MS analysis. Typical injection size for GC and GC-MS is 0.7 11. To prevent degradation of deriva- tized products, samples should be stored in a refrigerator. CHAPTER 4 RESULTS AND DISCUSSION This chapter consists of the results and discussion of the experiments performed and the analysis of the products obtained. A. PRELIMINARY EXPERIMENTS A.l. BLANK RUNS A series of blank runs were conducted to initially in- vestigate the performance and response of the proposed re- action system. All of the blank runs were performed using the procedures outlined in Chapter 3. The conditions used for the blank runs are summarized in Table 5. The first of these, Run One, was performed in order to observe the pressure response of the reactor when loaded with decalin. This was done to determine if any unusual pressure changes, which could result in catastrophic fail- ure of the reactor, occurred when the reactor was being heated. The reactor was loaded with 125 ml of decalin, 47 48 TABLE 5 SUMMARY OF BLANK RUNS, HEXANOL AND HEXANEDIOL EXPERIMENTS BLANK RUN SUMMARY Hydrogen Run Materialts) Pressure Heating Program One decalin 160 psig. 20 OC every 5 mins utes, 100 c-260 0. Hold 10 minutes. Two decalin, catalyst 160 psig same as One Three decalin, sorbitol 160 psig same as One HEXANOL AND 1,2-HEXANEDIOL EXPERIMENTS SUMMARY Catalyst Run Hydrogen Beactant Used Temp. Pressure Run Time Hexanol 5% Pt/Ca Y-62 220 °c 160 psig 1 hour Hexanediol 5% Pt/Ca Y-62 220 °c 160 psig 1 hour 49 pressurized to 160 psig with H2 and heated. After reaching a temperature of 100 0C, the reactor was held at that temp- erature for 5 minutes to allow the reactor to stabilize. The pressure of the system was recorded; the temperature was then increased 20 oC and the process repeated until a temperature of 260 0C was reached and held for 10 minutes. The reactor was then cooled to room temperature, unloaded, and cleaned. During the entire run, an essentially linear pressure increase was observed for the system, with a maximum press— ure of approximately 280 psig occurring at 260 OC. Run Two was performed to determine whether the cat- alyst would have any effect on the decalin. This was done to both check the applicability of decalin as the solvent and to mark any decalin degradation products for comparison with the products formed during the experimental reactions. The reactor was loaded with 125 ml of decalin and 1.00 grams of the 5% Pt/Ca Y-62 catalyst, pressurized with H2 to 160 psig, and heated according to the procedure used in Run One. After reaching 260 0C, the reactor was again held at that temperature for 10 minutes before being cooled to room temperature. The reactor was then unloaded, the catalyst recovered using a filter funnel and vacuum filtration flask and the decalin analyzed via gas chromatography to determ- ine if any catalytic degradation had occurred. Gas Chromatography analysis of the decalin revealed no observable degradation; the only peak observed was that of 50 tetralin (tetrahydronaphthalene), the major impurity in decalin. Run Three was performed to determine if any thermal decomposition of sorbitol occurred in the absence of cat— alyst. This was done so that any products formed could be compared with those formed via the catalyzed reactions. This comparison would show which products were formed via thermal decomposition, which were formed catalytically, and which were formed by both reactions. The reactor was loaded with 1.00 grams of sorbitol and 125 ml of decalin, pressurized with H to 160 psig, and op- 2 erated according to the same procedure as Runs One and Two. After opening the reactor, the mixture was separated and the sorbitol and any products recovered using the same pro- cedure as outlined in Chapter 3. The aqueous phase was an- alyzed using High Performance Liquid Chromatography to de- termine if any thermal decomposition of sorbitol had occurred. Results from the HPLC analysis revealed that no therm- al decomposition occurred. A.2. HEXANOL AND HEXANEDIOL EXPERIMENTS After the blank runs were completed, the catalytic activity of the Pt/Ca-Y zeolite was tested using 1-hexanol (Alfa Products, 99+% purity) and 1,2-hexanediol (Aldrich Chemical, 98+% purity) as model compounds. This was 51 required in order to determine if the catalyst had sig- nificant dehydration and/or hydrogenation activity. For these reasons, two reaction runs were performed using the batch reaction procedure outlined in Chapter 3: one with hexanol and one with 1,2-hexanediol. The conditions and catalysts used for the hexanol and 1,2—hexanediol experi- ments are summarized in Table 5. The hexanol dehydration experiment was performed primarily to determine if the catalyst had dehydrating and hydrogenating activity for alcohols, and if these activ- ities would take place simultaneously. This experiment was performed first because it represents the "easiest" de- hydration reaction: a reaction involving a primary hydroxyl group. The 1,2-hexanediol experiment was performed to de- termine whether or not the catalyst had the capability to perform dehydration and hydrogenation on polyhydroxyl compounds. B. EXPERIMENTAL RESULTS B.1. RESULTS OF HEXANOL DEHYDRATION EXPERIMENT The primary product of the hexanol experiment was hexane with a small amount of hexene as a secondary product. Figure 8, from top to bottom, presents the gas .7. i. ll "H’I" '0’9.' 52 lll in: 1.9.11 lll 7.03 I. ll "1 Figure 8. Gas Chromatography Traces for Hexanol Dehydration Experiment 53 chromatography traces of a 1% hexanol/decalin standard, a 1% hexanes/decalin standard, and the reaction mixture from the hexanol experiment. Comparison of the reaction trace with those of the standards indicates the presence of both hexanol and hexane. The absence of a hexanol peak in the reaction GC trace would suggest complete conversion of the hexanol. However, some residual hexanol may remain in the catalyst after being washed, and thus not be recovered. To confirm this observation, a sample of the reaction mixture was sent to the Mass Spectrometry Facility in the Biochemistry Department, where mass spectra were taken of each of the standards and the reaction mixture. Figures A-l through A-3 represent the mass spectrum of hexane, the mass spectrum of hexene and the mass spectrum of hexanol, respectively. All of these were obtained from the EPA/NIH Mass Spectral Data Base and were used for com- parison with the mass spectra obtained from the reaction mixture. Figure B-l includes the TIC (Total Ion Count) of the reaction mixture taken at the Mass Spectometry Facility and Figures A-4, A-5, and A-6 represent the mass spectra of the major peaks in Figure B-l. Comparison of these spectra with those from the Mass Spectral Data Base indicate that the products of the hexanol dehydration experiment are pre- dominantly hexane and some hexene. Based on the specific ion traces given in Figure B-l for hexane (m/z 86.2) and hexene (m/z 84.2), the peak areas give approximate concen- trations of 88% hexane and 12% hexene. 54 B.2. RESULTS OF 1,2-HEXANEDIOL EXPERIMENT Two main products, hexane and hexanol, and one secon- dary product, hexene, are produced in the 1,2-hexanediol experiment. No gas chromotography trace is given for the 1,2-hexanediol standard since the hexanediol peak is covered by the decalin peak. Figure 9 is the gas chromatography trace of the re- action mixture. When compared with the GC traces of the hexane and hexanol standards given in Figure 8, the re- action mixture trace indicates the presence of hexanol and hexane. As with the hexanol experiment, a sample of the 1,2-hexanediol reaction mixture was submitted for to the Biochemistry Mass Spectrometry Department for analysis. Figure B-2 includes the TIC trace from the Mass Spec- trometry Department and Figures A-7, A-8, and A-9 are the mass spectra of the major peaks in Figure B-2. Comparison of these mass spectra with those from the Mass Spectral Data Base indicate that the products of the 1,2-hexanediol experiment are predominantly hexane and hexanol and a small amount of hexene. Because the specific ion traces in Fig— ure B-2 do not include the m/z 31 trace for hexanol, the peak areas cannot be used to determine concentrations. These peak areas, however, do give relative concentrations for the hexane (94%) and hexene (6%) peaks. Comparison of the areas for the hexane/hexene and hexanol peaks in 55 HIV. (.6. Fl senemhaxm 1omeocsxom-m.1 soc 815. 3118311188110 30 .o oSwE 56 Figure 9 gives approximate concentrations of 41% hex- ane/hexene and 59% hexanol. Combining these results gives approximate individual concentrations of 59% hexanol, 38% hexane, and 3% hexene. B.3. RESULTS OF SORBITOL EXPERIMENTS Having determined that the catalyst had at least some desired catalytic activity, the major set of reactions, with sorbitol (Aldrich Chemical, 99+% purity) as the reactant, were performed. These were performed with diff- erent catalysts at a variety of temperatures, pressures, and time duration to determine the effect of these param- eters on the conversion of sorbitol and the types and con- centrations of products formed from the reactions. The conditions and catalysts used and the conversion for all of the sorbitol experiments are summarized in Table 6. Determination of the types of products formed in the sorbitol experiments was not nearly as straightforward as with the previous two experiments. First, both water- soluble and organically-soluble products were formed. The formation of water-soluble products is shown in the next section. The formation of organically-soluble products was shown by both a yellow color and "caramel" aroma present in the organic phase. Second, upon analysis of the organic phase by gas chromatography, the only peak observed was for Pry-J 57 TABLE 6 SUMMARY OF SORBITOL EXPERIMENTS Charge Sorbitol Expt. Catalyst Temp. Pressure Time Conv. One CaY-62 5% Pt 130 °c 160 psig 1 hr. 0 Two CaY-62 5% Pt 200 °c 160 psig 1 hr. 0 Three CaY-62 5% Pt 220 °c 160 psig 1 hr. 6% Four CaY-62 5% Pt 240 °c 160 psig 1 hr. 34% Five CaY-82 5% Pt 180 °c 160 psig 1 hr. 10% six CaY-62 5% Pt 260 °c 160 psig 1 hr. 87% Seven CaY-62 5% Pt 240 °c 300 psig 1 hr. 21% Eight CaY-62 0% Pt 24o °c 160 psig 1 hr. 92% Nine CaY-62 5% Pt 240 °c 900 psig 1 hr. 43% Ten CaY-62 5% Pt 260 °c 160 psig 4 hrs. 91% Eleven CaY-62 5% Pt 260 °c 300 psig 1 hr. 80% Temperature, °c 180 - 200 220 240 260 % Increase in Pressure 37 38 50 55 75 58 decalin, indicating that the organic products were non- volatile. This led to the decision to first investigate the water-soluble products, which was done using both HPLC and GC-MS. B.3.1. ANALYSIS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY High Performance Liquid Chromatography was used as a preliminary method of determining how many different prod- ucts were formed and the relative amounts of each product. The chromatogram obtained from the HPLC analysis was also used to determine the sorbitol conversion in each experi- ment. As can be seen from the HPLC results in Figures 10-12, there is little difference in the types of products formed in the various reactions: the only difference is in the relative amounts of the products, which arises from the use of different catalysts. Figures 10 and 11 represent the chromatograms for Experiments Six and Ten, respectively, and show similar concentrations of the products. Figure 12, the chromatogram for Experiment Eight, shows a much greater concentration for the compound labeled P601 and a smaller concentration for the compound labeled P501, when compared with Experiments Six and Ten. For this reason Experiment Six, performed with the Pt Ca-Y catalyst and Experiment Eight, performed with the Ca-Y catalyst, were 59 ___:+~'Z 11m 119‘“ - =____—_'__—' _ __._\.;r - —--.)';. 111-:4 Zb‘Sl .. -._—;—_-__——-___-_f—_—" -1" o _ In 10"” z: 11 - it.” .II_-— '4 86‘“ -__-_:::::‘_-_1-__ .___ _ -..-. ICE-d Sxf‘ -_:—--— T. T.—: _ _-_-_.___.___ lotnws ‘ I" 1‘ I! .." m a" o . u :5 '~._‘ 0 C I. *- °, 1: [ —( . H O | .4 r'——’: x -- “I ll 1- .I ‘0 '1 L o 1 .‘n l _l I ‘I l 8 '1 I V ‘—r I I 1 j # 1— ' fi_— 'l— -‘ o 0 ° 0 o 0 ° 00 ‘0 v 5‘ 1 0h ., _0 l‘ ’< ”0201340 sum "HIM ”NM! (Hal lO-SIFII ‘”‘l“”V “ea-luau moors us :aotulu I “HOMO ”MIND ix S nment d Chromatography Trace for Expe iqm Figure 10. High Performance L 60 i -;.;' I, C )d :8'" - -;'—:.:;-___ _ v." ’\: .5"! a“! .. 0:- ..-_.-. _.. _ . ‘. .0‘. —-— _. LV 13! 3231-13-7 1cm :1 U “"""—"“"—""“~~--—~— —-- ; _~-. J I. .Jr; blil " —-"":;~——I . '- -— ‘ ' - _--_ 1 1 14 1, 1 1. I I .1 I. l 1 '1 I I1 ll |l 1 '1 l l I l. 1 fir fl 1 I ‘7 I 1 7—W v 1 Ifi T v T Tfi r l-‘1'-"--J O O O O O O O O O 0 (so \0 In ‘1' m SIIO“ Z_OT x ”0an son "nuns 3mm “2;; (9-5.19-1,- :unntn an 11.001 noun 1 U I 1m MOM") mutual mans -— —_.-_-_ l 1.00 1 Minutes a. K 00 (I. Figure 11. High Performance Liquid Chromatography Trace for Experiment Ten 61 1 'I .I 1094 tam - -==-—-.—__-..._-.L;_";__:_'I_-_'7 ' v- ." 'QSd - --\ ‘ 01 ‘l . _ .. _-_'_——-‘- G 1::__ -—_._...._.... IQ? ¢ INC! .-, ._ ‘ _. [0131—Ng bl'll - -=-—'——"‘—: ............ 1 I .’ J I 1 .J 12 .1 l I. I I I. 1 -""“ I"" ' '0' """ l" ’7'- V "Y'L 7" 1 T r O ‘0 o o O 0 to «3 v '3) Inn 2..()]' x 1103:1340 svn: patterns :poqiou (1:11 (5-5"..9; :paainDJv iqbia Iaorualt, I IN") ' IW'NJ gin-lulu Indus .50 l 0.00 x 101 Minutes 1 Eight nnnen d Chromatography Trace for Expe 1qu1 Figure 12. High Performance L 62 chosen as representative examples for product identi- fication. Experiment Eight was run for one hour at a temperature of 240 oC and an initial H2 charging pressure of 160 psig. One gram of the Ca-exchanged Y-62 zeolite (0% Pt) was used as the catalyst. According to Minachev et al. [17], this Ca Y zeolite is a dehydration catalyst. Experiment Six was run for one hour at a temperature of 260 oC and an initial H2 charging pressure of 160 psig. One gram of the Pt/Ca-exchanged Y-62 zeolite (5% Pt) was used as the catalyst. As mentioned in Chapter 1, this cat- alyst was expected to have both dehydration and hydrogen- ation activity. B.3.2. ANALYSIS BY GAS CHROMATOGRAPHY-MASS SPECTROMETRY Gas Chromatography-Mass Spectrometry was used as the primary method for determining the types of products formed. Since the water-soluble products are non-volatile, the TMS derivatization procedure described previously [26] was used to facilitate GC-MS characterization. This derivatization technique replaces the hydroxyl groups on a compound with a trimethylsilyl group: R----OH + (CH3)381----NCH3COCF3 ------- > R--O--Sl(CH3)3 + CF3CONCH3OH 63 This reaction yields stable, volatile products which can be separated by gas chromatography. These TMS ethers have similar fragmentation pathways to those of methyl ethers. In addition, derivatization facilitates improved GC sep- aration: mass spectrometry of TMS carbohydrates is a major analytical technique for compound characterization. The major ions obtained in the mass spectra of TMS compounds result from two major fragmentation pathways: 1) the cleavage of a C-C bond in the carbohydrate chain and 2) the loss of trimethylsilanol. A general fragmentation scheme is given in Figure 13. Cleavage of the carbon chain results in ions of mass M - (103 + n x 102) where M is the mass of the parent com- pound and n the number of secondary carbons in a straight chain with a hydroxyl group. The strongest of these is the (M-205) ion. The ion corresponding to M-103 represents the loss of a terminal carbon with a primary hydroxyl group. Loss of a methyl group from the TMS substituent may also be classified as a cleavage loss and generates an M-15 ion. Loss of trimethylsilanol ((CH SiOH) results in ions 3)3 of mass M - m x 90 where m is the number of trimethyl- silanol molecules that are lost. As can be seen in Fig- ure 13, these losses can be combined in a variety of ways. The two pathways listed above result in ions whose structure can be directly traced back to that of the parent compound. Other major ions, however, cannot be immediately 64 265m HEoEobSU mg 18 oEocom 112.858on 1861.60 .2 23$ 95% $0- shoeooom 81865 "me/Ebro - 96% E0- ESE 31865 ”mZHONEO - 91on m2 .1. Soc 3.2 1m :0 - So— - om . 2 so om . 81- 2v m2 - 2 of - 2 00. 2- 2 mom - .2 11 > > : asses so - 11052.1. - moms: . mEoro . om - E 2 - 2 no“ - 2 > > moaze - mmo . egos mo . 65 connected to the parent compound and result from structural rearrangement, common to TMS compounds [27]. Some of these ions and their proposed structures [27] are given in Figure 14. Ions m/z 73, m/z 217, and m/z 204 are all common to TMS ethers of carbohydrates. Ion m/z 147 is also common to TMS carbohydrate ethers and is characteristic for most compounds with more than one trimethylsiloxyl group [28]. Since most of the rearrangement ions listed above and in Figure 14 are common to TMS carbohydrate ethers, struc- tural determination must usually be done using the higher mass ions obtained from the chain cleavage and trimethyl- silanol loss pathways [28]. Another problem in determining structures arises when isomers are present. In almost all cases, the mass spectra of diastereoisomers differ only in the relative intensity of particular ions [29-31]. Thus, for example, it is virt- ually impossible to differentiate between the TMS mass spectra of D-glucitol and D-mannitol. Molecular formulas, however may be determined from the mass spectral data. If a molecular ion is not present, as is often the case with higher mass TMS compounds [29-31], molecular weight determinations may be made based on the presence of the M-15 (M - CH3) and M-90 (M - Si(CH3)3) ions. After the molecular weight is determined, a simple procedure may be used to determine plausible molecular formulas: 66 m/z 217 m/z 191 + + Hcl: : CHfH (CH 3) 3 SIOCH: 0S1 (CH 3) 3 (CH3)3SiO OSi(CH3)3 m/z 103 m/z 73 __ "’ . ‘*'. HZC—OSI(CH3)3 s 81(CH3)3 -CH O 2 .m/z 117 m/z 147 + + <|:H 2031(CH 3) 2 (CH 3) 3 SiOSi(CH 3) 2 HCZO m/z 305 OSi(CH 3) 3 l + ClIH —_ CCH l (CH 3 ) 3 S10 OSi(CH 3) 3 Figure 14. Rearrangement Ions 0f TMS Carbohydrate Ethers [27] 67 1. Subtract the molecular weight of the carbon chain from the total molecular weight e.g. 6 car- bon atoms yields M - 72, 2. Subtract 89 for each TMSiO group assumed to be present in the compound, 3. Subtract 16 for each oxygen atom assumed to be present and not part of a TMSiO group, 4. Use the 2n + 2 formula to determine the number of hydrogens for a saturated hydrocarbon of the chain length assumed in step 1, 5. Subtract 1 for each TMSiO group from the number obtained in step 4, 6. Subtract the number obtained in step 5 from the number in step 3 and divide by 2. This will give the number of double bonds and/or rings present in the compound. If the number obtained in step 6 before dividing is negative, too many carbon atoms, TMSiO groups, or oxygen atoms were assumed and vice versa if the number obtained is larger than the mass of an oxygen atom. Having a prelim- inary idea of the type of reaction going on (i.e. no chain cracking of reactant) greatly reduces the number of permu- tations since one of the above values (i.e. chain length) can be assumed as constant. C. MASS SPECTROSCOPIC ANALYSIS OF EXPERIMETS SIX AND EIGHT C.1. MASS SPECTROSCOPIC ANALYSIS OF EXPERIMENT EIGHT Mass spectra for Experiment Eight were taken at the Mass Spectrometry Facility in the Biochemistry Department using the Hewlett-Packard 5985 GC-MS described in 68 Chapter 2. Mass spectra were also taken in the Chemistry Department using the Finnegan 2000 GC-MS described in Chapter 2. As mentioned in Chapter 2, the Hewlett-Packard was equipped with a 6 foot 3% OV-1 packed column and the Finnegan with a 6 foot 3% OV-101 packed column. The results from both runs were comparable. Because the mass spectrometers used for the analysis are both quadrupole- type mass spectrometers, higher mass ion sensitivity is not as great as that for a double-focusing type such as the JEOL machine used for Experiment Six. Subsequently, small, low-intensity peaks such as 701 and 733 do not exhibit enough high mass ions to give an accurate description of the compound. Thus, only peaks 558, 756, and 870 will be discussed in detail. The gas chromatogram from the Finnegan 2000 GC-MS for Experiment Eight is given in Figure 15 and shows three major peaks. Each of the individual peaks can be identi- fied by either scan numbers or retention time, both of which are plotted along the bottom of the graph. In this discussion, the peaks will be identified by scan number. C,1,1, EQQE §5§ Peak 558 represented a compound with a molecular weight of 290. This was determined from both the ion at 275 (M-15) and from the more extensive analysis of Exper- iment Six. The mass spectrum of peak 558 is given in Fig- ure A-lo. Applying the procedure outlined previously, it €59 EoEtoaxm .18 comb. EmfiwossEoEU £0 .2 Semi omumm souom svsmm smsmm Canon s41w1 swim soo1 com sow com sow can see Pl b h h — lr _ P P 1P — p )1"..w...m:h.”llr.l.a|..l.lllol... .. .sl‘clIIII-Iovo on. .\I‘ll ‘IIICIIWIMWHII I'll...» .1... a Ill I I....:I..._. .I II] I a _ I any. .II ‘ as... .... — on m n. a ’44.. r. o .o._ — a" . .— — ._ _ ... .. .— ~ F P n _ .. . _. _ _ _: ___ _ _ ,_ .1 _ . __ _ _. __ _ _. _ 1 _ 1 ._ _ . _ 1.....1, _ _ t ... 1 .1 mm“ 3,. 1s.oo1 o \l n o I o o o D OW.“ 0. - o . m .om 2 .woem c 1 .c a .zq2o a 4 .s : .swaab mw11 .wuzqa m. has 115:9 sao1 as OS4 mzqom moms momm1 m panw mason; "qhqo ‘. U oeuvm 1 1 "wbazqm .xnwxmo mum ac CC ac‘ hi SE ta pr Pa th ex 5111 70 was determined that the compound had the molecular formula C 02(TMSiO)2 with 2 double bonds or rings. 6H8 An M-90 peak is present at m/z 200 and an M-15-90 peak is present at m/z 185. No peak is present at M-103 in- dicating that the TMSiO groups are not on a primary carbon. A small ion at m/z 247 could be an M-43 (M - COCH3) frag- mentation indicating a pinacol-type dehydration, however this would imply that the two TMSiO groups would be ad- jacent leading to a much larger m/z 147 than is present. The m/z 147 is present at a low intensity, indicating that there is more than one trimethylsiloxyl group present, but m/z 147 is too small to indicate adjacency. The total num- ber of oxygens present suggests that two dehydrations have taken place. According to several studies [32,33], the preferential product for dehydration of D-glucitol in acids is 1,4:3,6-dianhydro-D-g1ucitol (isosorbide). This compound matches the observed mass spectra by having no adjacent hydroxyl groups and no primary carbons with hydroxyl groups, and is thus tentatively assigned as peak 558. The mass spectrum of isosorbide dimethyl ether was taken as a check and is given in Figure A-ll. As mentioned previously, methyl ethers have some similar fragmentation pathways as TMS ethers (but with less rearrangement) so that the data obtained should provide some similar ions but exhibiting a mass difference of 58 due to the different substituents. The ion at m/z 174 corresponds to the m/z 290 71 ion of the TMS compound. The ion at m/z 133 results from a M-CH3O loss, which is comparable to the M-90 loss (m/z 200). The following comparisons can also be made (dimethyl ether ions are listed first): the group between m/z 111 and m/z 116 to the small m/z 169-174 group; the m/z 99-101 group to the m/z 157-159 group; the m/z 85-88 and the m/z 143-146 groups; the m/z 71-75 and m/z 129-133 groups; the m/z 41-45 and m/z 99-103 groups; the m/z 69 and m/z 185 and the m/z 58-59 and m/z 116-117 ions. QA;&_. Peak 756 Peak 756, whose mass spectrum is given in Figure A-12, exhibited a highest mass ion of m/z 362; however, upon com- parison with the data from the Hewlett-Packard 5985 on the same peak, an ion at m/z 437 was seen to be present. This ion, in combination with the m/z 362 (M-90) and m/z 349 (M- 103) ions, suggest a compound with a molecular weight of 452. This molecular weight gives a molecular formula of C6H80(TMSiO)4 with one double bond or ring. Comparing the mass spectral data from both runs with the general frag- mentation scheme given in Figure 13 yields all of the high mass ions expected from a compound with mass 452. Also, no ion is found at the M-103-204 site, indicating that there is a primary carbon and one secondary carbon with hydroxyl groups. A stronger 147 ion than that for peak 558 indi- cates adjacent hydroxyl groups. Ion m/z 157 comes from the decomposition of the M-205 ion, m/z 247, which loses a 72 trimethylsilanol group. In addition, an ion at m/z 335 is a result of a rearrangement and gives evidence for a 5- membered anhydro ring and is shown in the partial frag- mentation scheme given in Figure 16 [34]. From the molec- ular formula already determined, the compound is the prod- uct of a single dehydration reaction. The preferential compound formed from the dehydration of D-glucitol is 1,4- anhydro-D-glucitol [32,33]. This compound fits well with the mass spectral data, having adjacent hydroxyl groups to give the m/z 147 ion, a primary and secondary carbon with hydroxyl groups to give the M-103 and M-205 ions, and a 5- membered anhydro ring to give the m/z 335 ion. Thus it is tentatively assigned as peak 756. C. 870 The mass spectra of the final peak, 870, is given in Figure A-13. Comparing the data from both spectrometers with the mass spectrum found in [28] shows that this com- pound is the starting material, D-glucitol. C.2. MASS SPECTROSCOPIC ANALYSIS OF EXPERIMENT SIX Mass spectral for Experiment Six were taken at the Biochemistry Mass Spectroscopy Department using the JEOL JMS HX110 gas chromatograph-mass spectrometer described in Chapter 2. The gas chromatogram for Experiment Six is given in Figure 17. As with Experiment Eight, the 73 $2 121620-9062111‘4; 8.1 oEoeom couscoEmsi 135$ .2 6.13mi mmm N\E mZPO ll 13 £8 mEPO ll l— O 1 . ll 82s mzeoiou Nmo - + 1.22.1.0 ll v mEHO l EU «we NE mat/2.0 ll ll Owe/E. mEHO ll mEono 74 Sam EoEtoaxm soc 88.1. szdoomossEoEU HO .2 211mm seems. com era was soo osm sow Us 1, 1 . . .1 1 . . , 1 1 . t . 1 . . . . . - a .\.\l\. ill/ ($333.41 7... \l ll. lltlzfiltrrltlli J a .5 4...... f. I |(. 1 41\ 5.1/1 . \4/ 7 f. fi—fi/fil _ s] I‘\ l4 1 1 1 . 1.9m m u. . 1 . C ’ .mv m u. r . c . 1 .ow . U . \J . 1 mm a . .0 . 1 m h . 9 2.0. m . . . 11- . . . .1 1 . . 11. - . . . .1 1 . . . . . . 1 sea , am am am m1 ms 41 m1 e a mesm xcz “ L 0 #0 L u «_0 os.o so 1.66a11u 1 mm mm m1.mm or :mm.o1 em 1mwssssmm 64 saw scoom m1o saxu owmoxmmosos “o1aeem mm.m1 ma ah:-1mommsmwsm “@111 .3..o one pr th me} con ion an}; C0111; 75 individual peaks can be identified by either scan number or retention time. Because of the extensive manipulation performed on the raw data to obtain usable spectra, the peaks are identified here by retention time. Depending on how the background subtraction is done, the scan numbers for each peak can change. §&2.l. Peak 20:07 Peak 20:07, whose mass spectrum is given in Figure A-14, exhibits a molecular ion at m/z 452. Applying the molecular formula procedure, a molecular formula of C 0(TMSiO)4 with one double bond or ring was determined 6H8 for this compound. This is supported by the the higher mass ions found in the fragmentation scheme: m/z 437 (M-15), m/z 362 (M-90), m/z 349 (M-103), m/z 347 (M-lOS), m/z 272 (M-90-90), m/z 259 (M-103-90), and m/z 247 (M-205). The absence of a significant ion at m/z 145 indicates that the fragmentation pathway M - (103 + n x 102) stops at one sec- ondary carbon. The ion found at m/z 231 has been reported for TMS sugar derivatives and has C9H195i203 as its probable formula [27,28]. The ion found at m/z 244 may be the product of a M-103-90-15 fragmentation pathway. As mentioned previously, the ions at m/z 305 and m/z 319 are common to TMS carbohydrate derivatives. In addition, an ion at m/z 335 suggests the presence of a S-membered anhydro ring [34]. Furthermore, the data for this peak is comparable to that of peak 756 from Experiment Eight, so 76 that 1,4- or 3,6-anhydro-D-glucitol is a likely compound for this peak. C. . ea 20:01 The mass spectrum for peak 20:01 is given in Figure A-lS. This compound exhibits a molecular ion at m/z 380. This is supported by the presence of the higher mass ions for the fragmentation scheme given in Figure 13: m/z 365 (M-lS), m/z 290 (M-90), m/z 277 (M-lO3), m/z 275 (M-105), m/z 200 (M-90-90), m/z 187 (M-103-90), and m/z 175 (M-ZOS). Even though m/z 73 is a possiblity for an M-205-102 ion, indicating one primary and two secondary carbon atoms with hydroxyl groups, the fact that m/z 73 is a common TMS ion tends to preclude this conclusion. Due to the close prox- imity to peak 20:07, the presence of ions m/z 259 and m/z 362 are suspected to be carryovers, since they do not fit the major fragmentation pathways for a compound with mass 380. A mass of 380 yields a molecular formula of C6H902(TMSiO)3 with one double bond or ring. This formula is similar to that of the previous peak, except that there is one less TMSiO group, one more oxygen atom, and one more hydrogen atom. However, this compound only has one double bond or ring, suggesting that one of the oxygen atoms is not present in a ring or double bond and thus singly bound to a carbon atom. Furthermore, the mass difference between this compound and the previous one is 72, which can be 77 represented by a TMSi group minus a hydrogen, (CH Si - H. 3)3 In other words, one of the hydroxyl groups on the compound was not derivatized; a hydrogen atom is lost during the derivatizing reaction and the mass of the TMSi group is 73 (i.e. the difference in mass between a —OH substituent and a -OTMSi substituent is 72). Possibile compounds to exhibit this effect include 3,6-anhydro-D-glucitol and 1,4-anhydro-D-mannitol. As shown in Figure 18, 3,6-anhydro-D-glucitol, 1,4-anhydro-D- mannitol being similar, has hydroxyl groups in a gig con- figuration within the ring while 1,4-anhydro-D-glucitol ex- hibits a trans configuration for the hydroxyl groups within the ring. This gig configuration may result in one of the hydroxyl groups being derivatized and hindering the deriva- tization of the second. Primary OH groups are derivatized more readily than secondary or tertiary OH groups [35]. Another possibility is that of inadequate derivatizing agent although this is doubtful, as large excess is pres- ent. As will be seen, this phenomenon is not isolated to this particular peak. Another factor which supports the idea that one of the ring hydroxyl groups is hindered is the absence of the m/z 335 ion which is present when the compound contains a 5- membered fully-derivatized anhydro ring. It is probable that if one of the ring hydroxyl groups is hindered the structure that gives the m/z 335 ion should give a m/z 263 ion (335 - 72 = 263). This ion is not present, however, 78 HO ———~ HO— O < > ' I OH 1 ,4-Anhydro-D-glucitol —— OH OH OH 1— OH 3,6-Anhydro-D-g1ucitol Figure 18. Q; and Trans Hydroxyl Group Configuration for 1,4 and 3,6-Anhydro- D-Glucitol 79 and ions m/z 175 and m/z 205 are very small, suggesting that this compound contains a 6-membered anhydro ring, such as 1,5-anhydro-D-mannitol. Finally, most of the mass spec- tra which suggest a hindered hydroxyl group also exhibit a m/z 158 or m/z 159 ion, which is not seen in most TMS carbohydrates. The origin of this ion is as yet unex- plained. An ion at m/z 157, however, is commonly seen for TMS carbohydrates and is not present for the hindered hy— droxyl compounds. C.2.3. Peak 19:27 Peak 19:27 contains two compounds, one having a mol- ecular ion of m/z 452 and the other having a molecular ion of m/z 380. The mass spectra are given in Figures A-16 and A-17, respectively. The separate mass spectra were obtained by first determining where the m/z 452 and m/z 380 ions were at highest intensity. This was done by using the specific ion traces given in Figures B-3 and B-4, respect- ively. From these traces it was determined that the max- imum for the m/z 452 ion occurred at the front of the peak and the maximum for the m/z 380 ion at the back. Using this information, mass spectra scans were taken at the appropriate positions to give Figures A-16 and A-17. Even though the mass spectra given in Figure A-16 shows a highest mass ion of m/z 362, the specific ion traces in Figure B-3 both m/z 452 and m/z 437 ions at 19:27 which were evidently subtracted when Figure A-16 was 80 generated. As with the compound corresponding to peak 20:07, this compound shows a fragmentation scheme con- sistent with that from a molecular ion of m/z 452: m/z 437 (M-15), m/z 362 (M-90), m/z 349 (M-103), m/z 347 (M-lOS), m/z 272 (M-90-90), m/z 259 (M-103-90), and m/z 247 (M-205). An ion at m/z 169 results from the loss of a trimethyl- silanol group from the m/z 259 ion. As with the spectrum for the 20:07 peak, no ion is found at m/z 145 and the m/z 244 and m/z 231 ions are present. Also, the m/z 335 ion is present and indicates a 5-membered anhydro ring. The ab- sence of the m/z 145 ion suggests that the compound has one primary and one secondary hydroxyl-containing carbons attached to the ring (see Figure 16). The m/z 231 and m/z 244 are also present and have been previously discussed (peak 20:07). Due to the similarity between the fragmen- tation pathways for this peak and the 20:07 peak, it can be assumed that this peak at 19:27 is a single dehydration product, where the dehydration has formed a 5-membered an- hydro ring and does not contain a hindered hydroxyl group. As with the compound in peak 20:01, the mass spectrum presented in Figure A-17, the second compound at 19:27, ex- hibits a molecular ion of m/z 380. Due to the significant overlap of the two compounds at 19:27, m/z 349, m/z 305, and m/z 247 are thought to come from the m/z 452 compound. Also, these ions do not fit the major fragmentation path- .ways given earlier for the generation of higher mass ions from a compound with a molecular ion of m/z 380. The other 81 ions present, however, do fit the fragmentation scheme and are the same as those found in peak 20:01. In addition, the ion at m/z 263 shows a difference of 72 from the m/z 335 ion which is indicative of a 5-membered anhydro ring. Thus, this ion could result from the anhydro ring of a hindered hydroxyl compound. The loss of CZHZOZTMS (mass 131) via a rearrangement fragmentation of the anhydro ring is a possible source for the ion at m/z 249. Finally, this compound exhibits the m/z 158-159 peak which appears with the hindered hydroxyl compounds. This compound also shows some isomeric tendency with that of peak 20:01 because of the similarity in ions. Because of the m/z 263 ion, how- ever, this compound is a more likely candidate to be 1,4- anhydro-D-mannitol or 3,6-anhydro-D-glucitol, which both have hindrance potential. Q,2,4, Beak 19:07 The peak at 19:07 also contains two compounds. The original mass spectrum taken for this peak exhibited ions indicative of a fragmentation pattern for both m/z 452 and m/z 380. This spectrum is given in Figure A-18. Using the ion traces given in Figures B-3 and B-4 it was determined that the m/z 380 maximum occurred at the front of the peak and the m/z 452 at the back of the peak. The mass spectra for these compounds are given in Figures A-19 and A-20, respectively. Because of the overlap, several ions show up for both compounds. These include m/z 247, m/z 259, 82 m/z 349, and m/z 362 carrying over from the m/z 452 com- pound onto the spectrum for the m/z 380 compound. As with the other m/z 380 peaks, this compound ex- hibits the m/z 365 (M-lS), m/z 290 (M-90), m/z 277 (M-103), m/z 275 (M-105), m/z 200 (M-90-90), m/z 187 (M-103-90), and m/z 175 (M-205) ions. As stated previously, the M-205 ion indicates one primary and one secondary carbon with a hy- droxyl group coming from the carbon-chain fragmentation. The m/z 158-159 ion is also present, showing this compound to be of the hindered-hydroxyl type. Once again, this data is comparable to that for the other m/z 380 compounds, varying primarily in ion intensity, so that this compound is also likely to be a 5-membered anhydro ring compound that has hindrance potential. The m/z 452 component of this peak exhibits the same fragmentation scheme as that of the other m/z 452 peaks. As can be seen in Figure A-20 two ions, m/z 404 and m/z 479, are also present. These ions are thought to be "junk" ions since they do not yield any feasible fragmentation paths that give the other major ions present in the scan. As with the other m/z 452 compounds, the m/z 437 (M-15), m/z 362 (M-90), m/z 349 (M-103), m/z 347 (M-105), m/z 272 (M-90-90), and m/z 259 (M-103-90) ions are present. The M- 205 ion, m/z 247 is not shown in Figure A-20, but is pres- ent in both Figure A-18, the original mass spectra for the peak and Figure B-S, the specific ion scan for m/z 247, suggesting that the m/z 247 ion was subtracted during the 83 scan generation. The m/z 169 (M-103-90-90) and a small m/z 335 (5-membered ring rearrangement) ions are also present. Thus, this compound seems to be yet another isomer of the single dehydration compounds discussed earlier. C.2.5. Peak 18:54 Figure A-21 shows the mass spectrum for the peak at 18:54. The molecular ion for this compound is m/z 452 which is apparent from the ions present (same fragmentation as previous m/z 452 compounds). Also, the specific ion traces given in Figure B-3 show the presence of both the m/z 452 and m/z 437 ions. This compound, however, gener- ates no m/z 247 (M-205) ion, indicating that the compound only loses the primary carbon from the carbon chain and not the adjacent secondary carbon. Furthermore, there is no m/z 335 ion to suggest the presence of a 5-membered ring; the m/z 205 ion intensity is also very small. This suggests the formation of a 6-membered anhydro ring rather than a 5-membered ring. . e :44 The mass spectrum in Figure A-22 corresponds to the peak at 18:44. Even though no ion is present at m/z 380, this compound exhibits other high mass ions such as m/z 365 (M-lS), m/z 290 (M-90), m/z 277 (M-103), m/z 275 (M-105), and m/z 200 (M-90-90) to support m/z 380 as the molecular 84 ion. Even though this compound exhibits almost no m/z 175, suggesting that a M-205 pathway is not present, the strength of the m/z 205 ion precludes this conclusion. The ion at m/z 187 is formed by the decomposition of the m/z 277 ion (m/z 277-90). Because the m/z 349 ion is not pres- ent for this compound, as shown in Figure B-6, it is prob- able that the m/z 259 (349-90) and m/z 169 (259-90) ions are not from this compound. From this information this peak seems to be a S-membered anhydro ring compound with a hindered hydroxyl group even though this is the only com- pound that does not show the m/z 158-159 ion. 9.2.7. Peak 18:20 The next and largest peak occurs at 18:20 and also contains two compounds. One of these compounds exhibits a molecular ion at m/z 362 while the other at m/z 380. The mass spectra for these two compounds are given in Figures A-23 and A-24, respectively. Because of the absence of a m/z 452 or m/z 437 ion in both Figures A-23 and B-3, the m/z 362 ion is the molec- ular ion of this compound. The M-15 ion, m/z 347, is not present on the mass spectrum given in Figure A-23 but does show up in the specific ion trace given in Figure B-6. The m/z 272 (M-90) and m/z 257 (M-105) ions are present in small amounts but show up more in Figure A-24 due to the overlap of the two compounds. Also present is the m/z 259 (M-103) ion. The m/z 290, m/z 277, and m/z 290 ions are 85 all carryovers from the m/z 380 compound. This compound exhibits two major differences from the others that have been discussed. The first is the presence of the m/z 319 ion. As mentioned before, this is a common rearrangement ion for TMS carbohydrates. To form this ion, the compound loses a fragment of mass 43 which corresponds to a formula of CZHBO or H3CCO. A common dehydration reaction for polyols is the pinacol rearrangement shown in Figure 19 for a glycol. As can be seen in the figure, the pinacol rearrangement reaction results in a portion of the molecule whose mass is 43. In addition, the presence of a methyl group in the molecule inhibits the cleavage of the C1-C2 bond and promotes the cleavage of the CZ-C3 bond which is also shown in Figure 19. This fragmentation results in the formation of an M—43 ion. Another major difference for this compound arises when the molecular formula is obtained. Using the procedure outlined previously, this compound gives a molecular form- ula of C6H7O(TMSiO)3 with two double bonds or rings. Because there is only one oxygen atom, however, the other double bond must be between two carbon atoms. This leads to the possible reaction scheme shown in Figure 20. The proposed location of the C-C double bond in Figure 20 is a result of the presence of the m/z 205 ion, requiring the compound to have a primary and an adjacent secondary carbon with a derivatized hydroxyl group. Based on the structure in Figure 20, the m/z 362 is tentatively identified as 86 C—C—R >H3C—C—— 1 1 +120 H O OH OH Pinacol Formation fragmentation inhibited m/z 43 >> m0 >m<2$5m N. mqm 23:282. Dec. 8:83 ode 08:. 8:83 92.8 32 ass. 8.65 £me EoEtoaxm mHUDQOME mAmDAOmimP/x? m0 >m<235m .800 h mamfl—i 96 a Ca Y zeolite is an effective alcohol dehydration cat- alyst. A variety of products were formed in Experiment Six. These included 1,4-anhydro-D-glucitol, 1,4:3,6-dianhydro-D- glucitol, 3,6-anhydro-D-glucitol, and 3,5,6-trihydroxy-3- hexen-Z-one (shown in Figure 20). A number of other aldi- tans and pinacolones were formed and are listed in Table 7. As mentioned before, the positive identification of these compounds is not possible. Two major differences occur between this experiment and Experiment Eight. First, the number of products is much larger. This can be attributed to the presence of the Pt. The isomerization characteristics of Pt acid zeolite catalysts is well-documented. In addition, it is report- ed that the high-pressure hydrogenation of D-glucitol in the presence of Pt results in an equilibrium between D- glucitol, D-mannitol, and L-iditol [32]. This leads to the possibility of forming dehydration products from all three of these compounds. For the compounds shown in Figures A- 29 through A-31, the major difference is in the relative intensities of the major ions. This similarity in spectra lends credibility to the occurence of isomerization. Due to the size of the peaks, however, the positive identi- fication of the 24:12 and 23:53 peaks as D-mannitol and L- iditol is not possible. The second difference is the presence of pinacol de- hydration products. This can also be attributed to the 97 presence of the Pt. Because Pt is both an excellent hy- drogenation/dehydrogenation catalyst and an excellent hy- drogen adsorber, it facilitates the hydrogen abstraction necessary for the pinacol dehydration to occur and offers adsorption sites for the H+ ions generated from the re- action. In addition to the pinacol reaction, another dehy- dration reaction was observed. As can be seen in Figure 20, a C-C double bond is formed via a dehydration reaction. Because the compound formed from this reaction does not occur in Experiment Eight, this reaction can also be attributed to the presence of the Pt. This reaction occurs after the formation of a pinacolone compound and involves the fi-hydroxyl group [36]. Because of the keto group, the dehydration of the fi-hydroxyl group is greatly enhanced and the a-hydroxyl dehydration is greatly inhibited. Because this reaction also involves hydrogen abstraction, the cat- alytic property of Pt mentioned in the previous paragraph is also applicable here. From the above discussion, it is apparent that the Pt exhibits no significant hydrogenation activity, even at Hz pressures above 1300 psig. Instead, the catalytic activity of the Pt is primarily dehydrogenation, with isomerization as a secondary activity. CHAPTER 6 CONCLUSIONS A summary of the conclusions obtained from the re- search is given below. First, the catalytic activity of the catalysts used was primarily dehydration and dehydrogenation, with some isomerization activity. The dehydration activity was higher for the Ca Y catalyst than the Pt/Ca Y catalyst which is due to the replacement of some of the Ca ions with Pt ions during the metal loading. This dehydration activity resulted in the formation of anhydro ring com- pounds, primarily 1,4-anhydro-D-glucitol. The dehydrogenation and isomerization activities were present only for the Pt/Ca Y catalyst. The presence of the Pt resulted in pinacol rearrangement and fi-hydroxyl dehydration reactions in addition to the anhydro ring formation. The Pt activity resulted in the formation of pinacolone compounds, specifically 3,5,6-trihydroxy-3- hexen-z-one. Second, high conversions of sorbitol were obtained under relatively mild conditions. Between 87% and 92% 98 99 conversion was observed at 240-260 °C and 160 psig. This high conversion of sorbitol was much more sensitive to changes in temperature as opposed to changes in pressure. Finally, shape selectivity was apparent for the cat- alysts used due to the small number of water-soluble prod- ucts obtained. This shape selectivity resulted in good yields of isosorbide, especially in Experiment Eight. Because it is relatively volatile, the gas phase con- version of isosorbide to hydrocarbons should be considered as a possibility for further investigation. LIST OF REFERENCES 100 LIST OF REFERENCES [1] Boyles, D. T., io-ener ' e no 0 hermod nam- igs, and Costs, Ellis Howard Limited, 1984 pp. 46-60. [2] Dubeck, M. and Knapp, G. 6., United States Patent 4,430,253, 1984. [3] Arena, B. J., United States Patent 4,401,823, 1984. [4] Chen, N. Y. and Koenig, L. R., United States Patent 4,503,278, 1985. [5] Arena, B. J., United States Patent 4,380,679, 1983. [6] Ibid., United States Patent 4,380,680, 1983. [7] Ibid., United States Patent 4,382,150, 1983. [8] Ibid., United States Patent 4,413,512, 1983. [9] Winfield, M. E., Emmett, P. H. ed., gatslysis, Volume VII, Reinhold Publishing Corporation, 1960 [10] Baunford, C. H. and Tipper, C. F. H., eds., Comprehens- iys thmigal Kinstigs, Volume 20, Elsevier Scientific Publishing Company, 1978 pg. 281. [111 Szabo. z. 6.. mm. 5. .458 (1966). [12] Saunders, W. H., Jr. and Cockerill, A. F., Mechanisms gt Elimination Beggtigns, John Wiley and Sons, 1973 pg. 256. [13] Weisz, P. B. and Frilette, V. J., J, Rhys, Qhem,, __ 382 (1960). [14] Rudham, R. and Stockwell, A., Imelik, B. et al., eds., Catalysis_bx_fisglitss, Elsevier Scientific Publishing Company, 1980 pp. 113-119. [15] Ballantine, J. A., Davies, H., Patel, 1., Purnell, J. H., Rayanakorn, H., Williams, K. J., and Thomas, J- H., £&_ugl&_g§;§lLl 2§J 37 (1984)- [16] Chong, P. J. and Curthoys, G., lsglitss, 1, 41 (1966). [17] Minachev, Kh. H., Garanin, V. I., and Isakov, Ya. I., Izzl_Akadl_nauk_SSSBi_Serl_Khiml. 1635 (1964)- [18] [19] [2°] [21] [22] [23] [24] [25] [25] [27] [23] [29] [3°] [31] [32] [33] [34] 101 Sharf, V. z., Piontkovskaya, H. A., Freidlin, L. Kh., Neimark, I. E., Rastorgueva, H. N., and Shameka, 6- 8.. WW. 2196 (1971)- Jacobs, P- A., W. Elsevier Scientific Publishing Company, 1977 pg. 101. Rabo, J. A., Schomaker, V., and Pickert, P. E., 2:22; , North Holland Publishing Company, 2, 1264 (1965). Kubo, T., Arai, H., Tominaga, H., and Kunugi, T., W. 4.5. 607 (1972)- Gallezot, P., Alarcon-Diaz, A., Dalmon, J-A., Renouprez, A. J., and Imelik, B., 1t_§at§11sis, 12, 334 (1975). Reference 19, pg. 3. Dalla Betta, R. A. and Boudart, H., Qstglysis, 1, pg. 1329, North-Holland Publishing, 1973 Treptau, M. H. and Miller, D. J., Ingt_fingt_gnsmt_, asst, as, No. 10, 2007 (1987). Pierce Chemical Company Catalog, 1987, pg. 186. DeJongh, D. C., Radford, T., Hribar, J. D., Hanessian, 8., Bieber, H., Dawson, 6., and Sweeley, C. C., £1.5mi_£h§mi_§22ip 21: 1728 (1959)- Petersson, C., Istzghsgzgn, 25, 4437 (1969). Kochetkov, N. K. and Chizhov, O. S., HEEDQQE_§QIDL Qhfimi: Q: 540 (1972)- Ibid-. Wm... 21. 39 (1966)- Lbnngren, J. and Svensson, S., Adyt_§gzbt_ghsmt WI 2.2.: 42 (1974)° Stanek. J. . WM. Chapter 26. New York Academic Press, 1963 pp. 636-639. Bock, H., Pedersen, C., and Thogersen, H., Astg_§hsmt Eganl_nl. 15. 441 (1981)- Chizhov, O. s., Zolotarev, B. H., Usov, A. I., Rechter, H. A., and Kochetkov, N. H., ggtpghy§t_3sst, 1g, 29 (1971). 102 [35] Knapp, D. R., Bsggtigns, John Wiley and Sons, 1979 pg. 8. [36] Streitweiser, A., Jr. and Heathcock, C. H., Inttg; nn2tien_te_Qrganic_£hemistrx. HacMillan Publishing Company, Incorporated, 1976 pp. 678-679. [37] Vaughan. Do E- W-. Shemi_£ngi_2regl. February 1988. pg. 25. APPENDIX A 103 Sam «:5 Eboomm $2.4 E— >— b- ? 4:11 7m- .flI11 1 0 10m T3 100 row 00— 22.2130: memxmm @4010: . 2380 mm 104 Sam Sea “aboudm 3&2 SEQ/Em Eococoxom; mo 858on 83.4 .~-< Emmi 00— Ce.— omp ON— 0: oo— Om om On Ow on 0.. on ON 0— p 1 p L F F P 1.. I. I- b— ~1- I- I- p— 0 ion I wow row row I . oo— 022 35:66.: 6:881; @131me SISU 3w 105 8mm «:5 2:8on 3:2 _-:Z\- y. p- p. q -——l- ~4D- >- '_ hI-l _‘ .. PHI p. .- ‘— p— i- —_ p- _h _— P -0 .— L... tLO m IZI 21- 'T'J I : 0:0 0:! — - O -3 '-L' I. a} —— ..‘ . —I- t "L - y- " F . f ——h- E --——__.—b —f:fi' 3' __4p . . L 3. D- ._ - :— r '3] t b '2: L ,— __1: ‘F p— h- ” - :-I:I v.3: "' I". 0).! ,. . : c-d p- p- p- r- C‘ E‘ ' o 1. | W >0 t . ~ I h— F | I I I I I I I j __I v. I f' I '_ I I [3| III I I 0:. 0:! ISI ISI 0;. IZI _-.I I _-.I so I .3. 02' w '3' N GI 02' u" '3' w | Figure A-6. Mass Spectrum of Hexanol from Hexanol Experiment 109 EoEtouxm ComnocmonA; EEC 388$ Co 838on $32 .>-< oSmE Il'-l .‘I'I --I-|OII.IOII'II III'I" ' .l' |1IIIIOI I ..I J l . : CVC Q.wC mi; 9m :0 C., EEK ..C.:- - FLE ..Fr: ...FFCFCIFFI CERF" .I.FL._I.F_.. FFF+iCFFCC£ I E“? lll--.IIIIIIC'I.. . CI' 0‘. 'IIIHI I” I .I I- -I I'll II :' -l | .I . II J C. - J O I O ... 1 CC. C., .....C... 1...”; ..........I.. ......r.. ..I...I. . . ..C...C..... Jun; '1'" - -. l ' . .FF LL f. I. up ...I—Ll 9: IL. . .I.I I; rlr .55.” ..vpsllsu r5... I.l .II. b FHICIL'. OlrLI'f—‘CtlnrchLIruHL Pr-th. Li... ..C CL “FrH .I I.., ...... . . 5L. .frsof. L C :2.“ .LLFI v.0 ErcIrl—L‘P CI...” D C... . C .u ... H um .I.C .u n. .w.wm n.m .m._m m.mm .q.mw w.C .m.mm C ..II .s. ,.I3 ....I I.I: .:.Jfl -.--..I..a . : no. n Cfl 0 00 0 nu r Pu h Ln. C CCC P (b .v .:;.m. mzCC :QCszCmm w“ ::CF :um. . manm Fit—4:41.: “tF t ._ : tichrLtEtLhtth boCPCrL _l I J G ' ISI LU 110 EoEtomxm CoCcocmxom-~.C SEC 28on .Co 823on v.32 .w-< 8:ow awn .aam wmm mom @vm amm mam amC u Elttrfnrc vrttFEfitIFrrrEtrrLEFEr. Fttrtttsrrrtlrt. rFF. IF . LIE ..rLIrrLLIrEFLPrrFuFr—ttIFL. I o HwC QvC omC mmC 9m 9m .. 9? mm FEFCFFIFFCIIIL.....rIF.-.I_..VIFFCHEILFLttIrErtw tittt EL Crutwrtx C..37.?.F£.T::F.FCFCFLFPECIFCLIECLFFII __ ”z; .C . _ _ C . ..I 10 CC I r r o t- CI IIIII- ---- -.I.-I--:-..-----I...--.--I..... I; . . L. mo.C u e C moan a.» .m pm a m .m -m m.mm .m.Cm - m .m H“ ”a Cw m.mm .m.ww m.mm .m.fifl. m.mw .m.uw 9.0oC,C.nw "v Cmmm rI--I.1:L. I IgCIILCuCInwmmL I am“ Ci¥uCuu m amom. Cfimu 111 ,ll-‘I Eucioaxm _o€ocmxom-~4 80¢ ~05on go 888on $32 .o-< 2.de QNfi QDW FELL“.:L 'I‘flut ( Eli Q3“ amm : O m n 33% . pbb—»>~> >th.>h>> >>>>—»>»> >h>>—P>>> tt>s£ 3m~ . am“ a?“ Gm" Q9“ am am 97 am Err».pig—Er»:.._E.CFPPEL.CEEFCp-—pppp C»; bu» p:_— kw uppb—P: upbh—PbbbL At.» --b:~n b».._:p» #:LIPF...o ___. I ‘ HJ— . # . A w_ :1: m. _ _ I 0 lo . ’0‘.” T. CV T ram 1 .om w! flag“ Do.“ u a mem . . . . . - m.a, .m.7m m.fi .m.mm a.v “.mu o.n .a.“u .q Hug; m.om .m.mw u.um .m.mm m.mq .~.mm mo.aaa.m.ww "v FwnmaJ \w.mux wzfih ~H__t:_um 9‘ c:.rMM¢m - I omow zwu 112 wmm xwum mo 588on $32 .oT< oSwE Zr... .. 3.... gm om”. . on. v a... m w. ..z I»..:...__._,_...........-....._..........__........— x.a.om ccr c.c Ctr Cu_ 0:. :w. ma: .: ..: . o . . _ ._ — _ _ .... .IL I...» _ _ _ _. f .. . _. . I_...._.._: Livy: ... .....«4 . _ . _ —_..— - . . ..,.. .._ — .. ... .I J -.. . . . A .q4 “41141 1 .. _ fl... « .J— 21...: .4. ”3..“ TN... [LN -.n; w: .:_ A: _H _ 4 . _ m.m I I I\ .1 C. II“ b“. '5 '1’ r ___—— —- _.._-- j n_~ ~o_ p r .... fir“ ..p—H— «mm: 1 m.wm . u....— mJuEIm a E. .m ..z .02.... ... 3.1 .mmmwm "u_m m Jig "~Jzu m. m“ + 1:“,u.:_ my‘mm.mo m“ "m\: wmzm mmmx w;mm~ "qhza tampomum mmxz 56m $5085 oEpuomomH .«o 833on £32 .:-< 2:me .. .. .cow“...-...‘...owmlilammlnllafim. 0mm... 3m «ama . .‘ I a n \. .... ... K a r r c . . Q ‘ . Z .. x 0m. .- f a , I . . o. .1 Q. C v I .o . A 'o. o ‘ ~ . s c c o ..OV r l . ., 0 o . O ’ I I v I . h . . run .I ? . 0m . (a! . . 2.3 v . ; m. . . V ~ . ,. . . v. .. . . .. 0m . .c Q o o I. W...” I lure ‘ ‘ I Q I . x 113 .Cva. ppxbbpbpbnp. h..-—_..~... .p..~.... pp..— Qmu o’. o - :b- bbhb ..Gofi .Gw .La.w®m phppp .....p... .ph.-.... .phppbbp». ....— .L..Jmm..;;..: ;&.Lp,.;V ... 4‘. Hm. . ....Q8L om ..~ ..p._ .. ...._._pb ...._...p‘.p-—...._ ppb._...»xh.pb_.~n>‘»... ...... . . .. ‘ . .141 =_ . .1_._.; =__—_ . 4..-: ..___ . 0.. w.dmwx ‘ ._ . km 3...“ . . .om qu , 07 .w. row ,y. ..3. . .1 . . .o:.( 6‘1..- _. .. ,.. , . x .r. . «3..., “8.731.... . . .. 1001 . an. ‘ . ‘ . Io. .. ... ... . fl...; 4. u. . .. Q .— . . u y. ... o o... . ‘ . E. \a . . . . . .c I] I. I'll un‘fniall I ‘III II‘I.’finl’ ll m;vm H. mv m.M7d "w kmiq t ‘a m d wSHF :OHFZMFUQ m .w u¢ kmomdg > a m.mna ‘vu&fl .m.vnH m.” . H m? a. A ooom :mu own 56m Mo Esbooam $82 NT< 83E 1.]:4 1 f 9.0m r flags: .. cw... cu ...... # ow wxz ._._L.__..__._q._ 4.....4..LFF.E._E._._r._.u..q......_..__..:1:__.:..E_ -4 a... Z 4. 4w“ ‘. 4 7:1 4 4.. 5 -. .. .4 n. “a; (tr mvm _mm _mq_ mm" “M“ . : _L _ "a __4‘ mw u 7. r in a U 1 mm” up. mu" muu 1 mvfi w 1060 r 1 r... lad—4.: mmm. h-umu m p.xm uq40:x momm_ “w4uzam nu~m M: 419 "~4qu v_"mm + convuuow mmxmmxmo wm1m umna mumm— "qpqo :DLFuwum MmI: "n .1. \u If. LII (';‘ II. 5. 01:3. RL fl -. 1'1 ..ss .. 1 DHTH TF‘JJH :c:‘ “HE'S . PU ‘ .‘ ‘ c HLI: . I + 00 lv.|_|[li__p! 0:24: 5‘ "93.2.1 ‘0 o O I ‘- ’- g'x. E :f': L.» . 3‘3 3 1“ EHHFLE: “HM '.‘. 1‘ .o n:?. (‘I 0.0:: [ti .. --- - "I'm I"'| i ‘- I" ’ _ l- — 7: 4: .. t!- -. . __———————-.-——-- ___. ..‘ ,..._——-...—---~ 4! n" . - _- U" 4: $3,0-1 J l 115 EQfi v” y, C "rm 'W'V' v H/E f 3-4 C _U H/E Figure A-13. Mass Spectrum of Peak 870 116 N\Z Snow xaum .8 Esbooam $22 .3-< 22mm Gmm mam nan.“ .+UUUu QG.A >4 VHDV.NmH .ACH mamm.vam N\E "mm om.m mmnm. mmlmiqlhm mammhmwmm "mfi OZJIZC'UGCOQJ MH (3...)... D m m Amme, x “emu 0+ mmng ¢caum um A.nom. Hm =~m~mm hm :w paxu wmmwxmugeH: "mfiaemm _pwa znmpumam mmqs 117 Snow xwom mo 8.50on 832 .mH-< oSmE N\z amq 5mm -am amm new “a“ aufi am . , 1. ,l 7: ___J” e. 1. _ __.. #13. _ . ._ .w .L w 3 4%: :4: _ .n; u g _ # m U fij _ _ _ : _ mm .._m . 53 gm. . m ‘ u a. mm o . a . m2; MG.— . C .[Jfii'F—uf L v G . -J . B 1 a _:J Uh." :8? C be 7 3 . . 3 c emJ a e 1 n ufifi fimm m L T > mwfl . m . . a m U lam fl . m . , m . mam mm“ b . m wad “as.“ .euuuU name ”emu, : “mm“ av «mu. ¢cuum as.” 3; mmmm.wfifl .ch mama.mafl N\2 "am om.m Um A.nou. Hm =Hm~mm hm mam paxu mummxmmqquz “o_aeom mm ma nmlmma-nm Fmo.mamwnmv@m "mfiwu name zsmpummm mqu BEBE] BEBE Int. 75. m/z 85-9. 7 2 BF‘ 8.66 1.08] M ILLER/‘N EI 195a?" SPEC TRUM ple Scan* (731) MRS 5 Sam RT 0'1 1 A 3 '19 '27 4559 3 E: '339 V Y (\i was 191 1117’ u .1 ESQ wf 2 I' 1. L713 jJ 188 l- 1 n rl '— SE3 mummy-«>4: 2E3- CCJIZC‘GGCOID M/Z Figure A-16. Mass Spectrum of Peak 19:27, m/z 452 119 0mm ~\E KNHE xmum “0 8:28am 832 .:-< Emmi N\Z . flaw 5mm Dem Dmm Gad Ema Ema Gm n‘ Eccum CCJIZC (INHIB-Q-H>m new a .+unuu Ammee u “Mme” ¢cuum mm.u >4 mmmm.ev .ucH mama.ham Nxz "um om.m Um A.nou, Hm :mmsmd hm mum paxm muawxmmuuuz "ogaeam mm ma mmimuqrnm mammnmvmm ”mfiwu «pea 33mpuwam mmaz 120 Si x8e Mo 88525 as: @535 .E-< Baum N\Z Gav 5mm mam Gme GEN mm..— am: flma g _ . » p 44—» L . < y _ L1 7 J” L; J. m . ._ map .27 ___—11.44;. =» : . :flid . .L‘ _..I~F4_ _ d «:13- _»_.L:4: C mam : _ # =_ mm V mnum 1 km a. . m . (EN 0 . hem mam : , c 1 ._ ‘_ I“: v a . ....aL . u . mm“ -mq c r 3 r pfim V m . ......WH .m. m .H fl 1 ram m 1 v > . m r e n— . «... 1 ram A 1 fi ‘ m . mmfi H m Mm. an: new...“ Quouu 3.2L 1 Ewan; *caum .....D n... -_ Hahn ..Srm 2:: @3356... HE: "mm m.m.m HE A man: Hm 523mg hm . mum Fax mmmwxmuuuuz umfiaecm 0.-.. ..- a “mum"--m.n__£ um mamthVQm... " m; «a fiend szhummm mmdz 121 0mm NE. .8”? fian “o 828on $52 .o_-< oSmE N)... Gav . a. U1 p cam 6mm Dam Gnu... saw mm m o m.au; c . c u C 3 3 c m 3 m a m ~ m m afl flag A .+uouU “mmn ”wave 1 name my mfie, ¢cuum cm a ; mcmm mfim .ucm maaa.mh Hus "um om.m Um “.moau Hm zmmemu Hm are Haxu mmawxmmuuuz umflaeem r' U l v—4 [2:] :_;_I l £1: r' [ _ TU It; ‘0 mammfimwmm H m; w n ma F..__._mkum&m mmcs 122 N3 ~\E Song Mama “0 538on v.32 .om-< ouswE Nxz mom :5: Sam arm 59“ u p a F: L. p L . _ - Lfiqi‘: p p 4 ; JFJ _ . L 1 . M . . p ram m - - A e m . m L pm: mmw . i. mad Ham.“ .+uouU Heme ”many I mmmu 0+ amp” #:oum ”a u ,4 amme mm .ucfi omqu.mmm ~xs "um om.m Um “.moae Hm =Hfirmd hm mum Faxu emmmxmmuqu uo_aeum mo na mm-ueu-hm mamanmwam "mfiwu apes zsmkumam mmcz v2: venom mo Esbooam 322 .HN-< oSwE 123 as? . mam mam ma“ J . L 14141:? :_ Z: 5 3...; 4+ mm a m .n. mcu e m t pru , ow mvm mm; H ha , . u m nan m_a -am I: U c.au¥ mmfi , c 3.: .11., .. .1. 5 . E m v we“ mmfl V m fin . v I\[ rm 3 . m _ ram m , .2 . 4.. . a E -am A my; . m em? mam m Gm; H.“ Arar 5.. . fifiue my man. icuum mo.a p; momm.m¢ .ucH naam.mrm was new um.m um “.moau Hm =wmsmd rm E u mmawxmmuuuz "mfiaemm L H 23mpumum mmcz (F03.J (7GB) MILLER/GREG EI M955 SPECTRUM Sample: RT 18’44" Scan“ (734) 124 1». (\J' N -' 488 —-‘~— Y 2 ‘ fi 2B5 18? Y— 159 J11 fl Y 198 mm~m+~~>m CEJZDCEBCDID BBB 258 2GB SB N/Z Figure A-22. Mass Spectrum of Peak 18:44 125 mom N\E dag: xwom mo Esbooqm $52 .m~-< 95me N\Z . Gav 8mm, Dam flmm GEN am.“ . am: Gm cum-«azu-«3m nae H .*uuuu “away I gamma ¢cuum mm.H >J vmhm.vmm .ucH ammm.mh N\2 "am om.m ow A.»om. Hm =mm~ma Pm mum meu mummxmmqguz "oflaeum Jm ma mwimuq|m mamwnmqmm "mfifiu «awn znmkummm mmcz TU 126 0mm NE .02: gm Mo 833% 3“: 32¢ 2am Nxz can 3mm mam mmm amm mm“ amfi flm P p m »._‘— > p L p [L — «- >4— _ :4 F.‘ p . u r a s b — 24‘ up : — A . u J _f_—d-_ 44'»!— L— .u G d ham 4 _ mm“ mm . A m H jam m , ®.mr mpg , c .. p . >_ p p p p . mkh . a . . u - “mm .G? c l mvH :. . fl 3 . bum nqm j m - NAN -am m 1 v 3 l w . m . ....... m . p m 1 Imm H L . m A v m . H“ mm“ HT sea naa.a .+uouu. nfimm .mmmv no A ,4 mmam.mm ch mamm.amfi Nxs "um om.m um “.mouv mIL meu .ufi mmluaq:um mamunmwmm mafia gum; mwm®\quJHz I Rummy icaum Hm :hfixmd hm "mfiaemm Efimhummm mmcz II: If; h“.- H E1 29 5;. LU - l CI [1' CL L" c: __I I [k 7U W? E’ I pl [D (U El [T]. _‘_'. C r—1 E SI (SI El [-.. If) N 6' "x [I] E (I: [‘n. o. N [L ‘7 III Ci] L0 .—~. I ”3 -- LEI ca IL: . . H [D -.—I .... m h. I - L0 9- “? L.) Lu +3 I— 13 0 It: CL u C: 3-2 u L‘J ’— r_n o m U 0 II] CL [L G] LO v *1] .\ ._,' C}: H E LU LU I :i .1 01 _J A I— I—4 : 0‘: U E B CD LIJ m ID a_ on s v (.0 (I: III F! u—I :3: Ln 0. C (.0 E 0 CE «I l— u E U) IIUI 127 mm—«m-p-u>m 3 RH 2951 158 'JBFJ 58 V (II—HZC'BGCOM M ./ 2? Figure A-25. Original Mass Spectrum of Peak 18:20 com ~\E do”? xaom .8 888on $22 .om-< 8:me 128 amm mom amm mam am“ aefi amo r p . . . . ... b p :. » id > .LA r . .JL. r 4.:Ir.4.. ..4 _ LI p . _Iquuq .r: 4L1: LAIJ... _ __ g u f e u _fi = _a. I smfi mm . mm“ m .mm o s.afi§ mm“ _ c VIII 1 $.Li4. flill. v 11'“ i G _ _ ; : H d. . u mum . g i I t g c . . mam 3 w I n «I "U c m...... ram m . 3 V m u p .m IGm a m m e u meJ 5.- an cafi has H .+uouu mama “mmmu I “mm am.a )4 maufi.mhm .ucH mama.mam Nxs "um om.m Um A.nouw Hm a are Flam mummxmmuuwz ”mdaeam mm «H mmnmuq-um mammwmwam ”mfiwu new; 23mpu mm mqu 129 wOm SE .00an xaom mo Esbooam $22 .R-< Emmi Nxz can amm mom mam mam mm“ aafi Jmu _ I I L I p I I II p 4— L4— i. I J . F II.‘ I. Iazilr :F HI 4 +4 I I _ .I d—. __ 1‘ _I: _II . .I. .I _. . . J _: 3:: I... Tm ... m . W3 wan U I D . r..._ H . “h": fi . _L I mil _L I . I114. -. I c — ”urn. ~ I U 95m .0? c fiat... . 3 I . I mi? w . n. . m I raw m I amm . > I , m I v a . m . lam ~ I mm 5 m I . m I w mam I aaa “as.“ .IUUUU “Jaw Imam” I Ammo 0+ wmm, ¢cuum am.” >4 vamw.mmfi .ch mamm.mmm Nxs "um om.m mo n.nomv Hm ImfiImH hm arm Iaxu wumwxmmuuuz "mfiaewm mm ma murmmcnum mammwmwmm “mfiwu Mama EJmFUmmm mmcz 130 omnmfi flag «o 886on 332 .mm-< Eswi N\z Gem 0mm 9% amfi am; am (IJIDC'DECOIIJ (rm—amu-uzom w a N. .n naa.fi .+uouu “finmv I “mum. ¢cuum am.a I; mamm.umm .ucH mamm.uafi N\5 “am om.m ow “Inna. Hm IamImd hm _ mIm haxu wmmwxmmJJHz “oHQEIm mm ma mmImchmm mommnmwmm “muuu spam znmkummm mqu ID I; I‘- u 36 Int. 8.60 1.88] Data Fi 6‘3 GC [cocf. (P03?) (911) MILLER/"EG EXPT EI MRSS SPECTRUM Sample RT 24’83" Scan“ (914) 131 1 1 1 1 1 1 1 I 1 1 l 1 1 1 l 1 1 1 a [:1 v- ”G r If: -— m : r 10;? .1 t. 3 ‘L L -n-d 4 ‘I TS ..i A l‘J ___—1? El 7;? w c\‘ . E "- “AI I '3 7:" 3n" I. ‘3 V ‘r $——-—:: E 07 _A b) .3 Ch fi———d_" h m d C] __ E ('7‘. I. G _.: m A» -{ U: ..I_ r———= n m 1 V _ Iu % J _— rIZI—‘fl N _‘P In ‘ 1?: m _ a (\1 LE] 0': _‘ N a+———— H —t I r» —4L If) . 1"1 I‘- ‘T :1 T” I —E N I I-1—=‘ (T) H a I3.____._‘I_I_t3 fl ('1 1 m _' [\. +7- ' ' 1 fi 1 ‘ 1 ' l ' ' O ('3 C9 C9 CS] (SJ CD CD u'J V N v'l EMHIB+J~H>M CEJI‘ITJC'UGCOM Figure A-29. Mass Spectrum of Peak 24:20 132 Nfivm fium mo 838on 832 .0m-< “momma Nxz GEN 6mm mam Gm? 8mm aww Ema I I I I I I III #1} I r I I I II flFw: I IquqflflI _I #IJI‘ML I I «II I _II . i:|»_ . IJI I “HI I m m mvm 4 Odd I mg ..I , I 0mm law IGN m I m. Wm I. ma fl I w I v a I z I C I mm as? 3 I I n I I m I Iam m I . > . m I p I m 1 10m ~ I . m m . m I BIN . I am.“ maaé .Tuou”. mommy I. Amam. #cuum am.a >4 umHmImmfi .ucfi mamm.uam N\s "um ow.m Um A non. Hm ImfiIvm rm mIm Fax“ mmmwxmmuuuz “ufiaeum mm "ma. mm mt“..I+. I I I I I m ..I tam H . m . m I J I In. aaa naa.a .Iuouu Ammmv I Rummy ¢caum mm.a 3; mmmm.mm .IcH mamm.mam N\2 “am .om.m ow n.nuu, Hm ImmImm rm Iw Fax _muawxmm44Hz “Idaewm mI.IH mmImQQIIm mammwmqmm ”IA u Inga I rampummm mmcz APPENDIX B 134 1 H0 1. . 1’1 H FRH T 1 EDIT % ES—EEBRMU .TPUM DISPLgyg Eli-11,. 11:1 FEV HPl 3.5UL EI JQMFI_E '«T 5 n . a - 5‘. 1-14BWSDE' k 4 {MILLER’v .L .35; I -l.-— g ‘.I : ’_7-.. E I I l . ,—-";‘ I a “———‘ I T I'- -:r U! H I I II __C'J ‘3 C3 . 9% 0 ISO ft] 'IL'O §~ I'"- H W m ' 1:. I." ‘ ’ f'Ij U” . 7A a (1': n u 5 g I i <1 i __- j u ._.-": II I‘ -r-"~—-——‘-d_ . _..’ ‘5 .1 ———":'.‘- r-.—_ _ ___——" E __——""'_ : _--—-_ i_ : - —1.——___ ...—_— __.1. ----—'I.J I I I l - I. OJ VJ 1:. V \D v-t 0.1 CO (I) If”) I— Figure B-l. Specific Ion Traces for Hexanol Experiment 135 EoEtoaxm _omnucmxum-m.fi I8 3089. E: 058% .m-m oSwE ll. I ' I 'T'J '4 niIli-.- ill- - ....r:l... n: _J kl! - JOI< ~ : 1 I'll I 00" 0- o H... C 'I l [I —. i' I (0'. I I "I D-‘ 'n' ll--l - ' l [‘1 r III-Iii a I - lal . ..IIIII IIIoI II IIIIII.II.IIIII' II- I at: .9- T. G; __ .3....., n "L III-I III! .-.II III-0‘ uI . nI.I0. I. I a nu Iv. .0 .I _l. I , .l . . . . . ... I .. I. -..”..I u.,.. Frusyluu _.V.,h.._.uwu.wuow ... In“ oIm N4 uwxm~.m mquzqm @mgqmz -Icuflm:5_:¥w:.; ......:iI.IH;m.n;MgunHm &:HMMMJ.WJ +1 -?iI- II II .II-III III sci-...III'III ’nllc 136 nmv £8 98 va NxE I8 mound. :2 £50on .m-m oSmE anywaam amI man ... ., 9mm . saw him v III\III(!\IIII\..I.IIIII\III\.\..I/\ )Js\1\l ?/\)v2\a{|/q IfijIrIII _ m _ .. m. . v H I e .m $41,531. D ‘37....ié. >3. \J../ 1) é ..1......J\I.. m /\ w . U A I _ c _ I m.m m uhhxx.1.111111 .1. 1111 . -xx1111111111111 > a d x x / 5),, \1/5. ... \.r\ I _ _ c. _ I)... I. ..x _ u. _ __ . K _ j IVA .0 J E . J w H . V may ..u.._ N. 4 i 4 q < . J1 4 ¢ 4 4 4 1 .1 q 4 I u m .H.E an... E” @w T; u..._.. Mum—yaw .XCZ “LI a.Mm a Imv "and: a3.a :4 r:LUHm I mI.Am ma Im ac mL.uI Ia PaamemCm :I mam tacom m:o u mg. \mmJ.Iz "mfiaemm an oI mmnm;q-nm Iqa.mgmmhuwam mfific ...I; sang QI:2:m1u mm I: 137 0mm N\E I8 88p. :2 2.28% 41m Emmi cwom -;. 1 88$. . . . . I IGVI. . I . I -mmuh. . . ......me . 199m mmmJ . .... \_>.\.z ... NIL/Ir I . : RAW/E}. fi/u 23...)fo 2...... _...\- . .......<,..2 .....2 .. .. .... .. m ”I I m. m.m u . I UIIIJ 1.1-111.-.11 .1113 .1.1\11>4111111111111. c / I_ _I I2) 7,5 7f)? I1/xx - H R: _7 _ w _ .1: .. _ _ r._ : I x I i u I U I.qmmI m . .. .. mm -. . . ow -. . . m4 . m.I1 . . . I4 . .1.aI I I m.m.xaz UHI a.amm "and: 59.: :4 I.IouIIm IzmI.Im.. .mI Im II mm.mI Im ImamIQQm DI 5mm tcuom m1w Iaxm mwmwxmmJJIz umIaemm mm.mI mm1II¢IIm mommImvam "IIII III; zqmonqzoqu mqu l [11' Cg 6' I; c I Cl 1‘ \— 'I C. J‘ Bo- -_'1‘ |_. .—-4 [D ‘ (1:: c—1 |--— m f J 7304 - 1’ I'u ,3 q’qfi” Fil FE - [I D3. t. -:". - 3 EL-‘ZF’ 1' PT 1 R‘I BBBtBBQ) TIC q 0 HRDMHTOGRRH MILLER!" 247.9 '633 10 .‘ d w 03 Ht.‘ U1 QC I1 Wifdn (I It; 0:! [J EU?!Z 138 R.T. r247 ~.—-1 fr.- 7‘ "- UT“ #19 \§ru . ._‘ , \J 11:: i' 1"; 1‘ q. o 0 4 'l .0 J I I I #— fix ':' “___ .— m‘ _’——- _r- ‘ —-—" - _F-F” . ‘_ _ .... u 1 K. . fl .J. J’- \- I 4 ._’__._r ’- .\ ’J 1 -’ .. ‘ __J ‘ 3 ___- — ..— "—— _ ___—E --V \ ‘ ___-d ‘-— —\———- — - __ ’ O---I - ---- ._4 " "x— _\ a. -..-. ’4 - L .... “.——-J 88% So an "\ ~ U"! [\.. LL 0 . ' D I, p I .’ \ ‘\ L ESQ . P m- ll Mif Hax.1 ZUHC:'P'H}‘U HE+‘UC$‘I"‘+‘D’J 1,— 638 Figure B—5. Specific Ion Trace for m/z 247 139 gm Q8 was 9% N}: H8 808B :2 21:0on .o-m uSmE CHw 0m @Gm 9mm as» 9mm saw me ;a|rIIIIIIII|. _ P h r n b p P [bl n n p P h r P b P ¥ b b h h F P b b Tiffi4 n.no&va nzmfihamuzmaqfim 0: :mm.fifl Fm “Eamwmam 0+ mug Puma mumv\mmggu Wu mac; Zimmohqz mvm "and: saw *cdom : "mfiascm QEIU mqu to L, a ’t *4 III -I- ¢ 5: ; - TtJ }_ lI :4 A! If] [I] III i — I J 'T (3 If: lll 140 mom QB 98 com N): ho.“ mount. c3 oEouam .n-m Esmi .mv. P om 33m 35:- 33:. 3mm. n 33 “huh-wm l I b 5‘ L b P . p P; “\IIILIPHIIL “(\Iplh chi-(b » L b h b ulllllpllwlbllqu-l L r b b F r _. _ \. lot/...!)l. .2) __ x -/\ IJ . .1 { MK _ m ... _ x .- m.amwomm __ .. :5. 5. w .C .....\ (Mi-...} 5 28.1,. _\..<-.-..§|J¢>I)1\,\,x . .. .. .\\d..-..r5-\,?\-<.-Ir.\,-..../\.-..l.\ \(fi 85 , m . r. ./ ..x /\/\I/_\ /\).s _ ._ m a a x p . E e H a. uuhn. n - . -!1:-:5IIIII m 4 d _x xi ./ KI\)/. 5Q), \IIrllk-x-K ...-III! M. .... A W/ x #1 __ /5 «a a . . x A c 3 fi.. VWQN... ‘ 1 1 'i . < . 4 < . 1 ‘ Ji all. 3 < 4 < 4 q 4 4 1 i m .h.m am am mfi m3 m« m“ m.mm.xaz . - . UHF 3.33m.a.amm "“no: 33.: >3 n.nouva m:mfi.3m..ma.fim 03 .mm.mfi Fm Hammuaom 03 333 tacom m-m Huxu mumwxmm33uz "mfiaeem 5m 13 mm-mmq-mm pca.mammwmqmm "mfiwm Mama z¢330hqzomxu mmaz 141 NS N\E H8 8E..- E: 22025 .w-m 8sz :com E5 [0 Ch 5:25 6 CD CD 3 C0 mam 33w 3:. - ... . 5 5 i. x}. . . n b b p L F F > p P b u > . h u - . P r . b: p p _ p P » GNU m N I n. #55..- .....-5 .. 55 -l-ll ..... .5llll-155 . I r . 3 _. . 55. 5...755||-.5 75..-... -. . .57. . r. _x . X a. J. . __ ...... .__ _1... _ _T T 5.. _\%I_ _ _._ ._,_ . I ‘ . C 3 ..- m r .._ a . m . mum .....- 55-....l5 - 3 ...._—_ _312 5'1 "" +-‘ .55: I _ . 3.355%- .) -. .|-.-I- .- - . .. - --I .- 3 5.1..- I- c J. .. -_ __ ...5 .- m _ _. 1' :4" r 5'. rr: ['1 l 3(- (U H In H ‘ - ‘O(ll-o[lv}\ll . ( . ..\ P. ...-.31.... tit/.1, . .....J . ...! .- .. . — \u. 0.. I..\.0 .| I».. \‘II1 .- _ ... q...» .... __ ...: ...\...I. . .1./l\l :15 I U H t— i I .2 a j i I ll l l l _ . _ t. r —. _ J _ .I 9 .4 H _A \ _‘_ J __ J a (3% m. ...-$.- .. __ m . ._-. ...m AU-u _ TN .U .... FN H0 :0 UH. .L ¢.~mm.xcz am.5 >4 n.wo&va nzmm.mmv. m5 . . m mmuflfi mm! muI. hm mamm ivmm ”mflwu fluca ZImmD