SEPARATION AND CHARACTERIZATION OF CLOSED FUNCTIONALIZED DOUBLE - DECKER SHAPED SILSESQUIOXANES By David Felipe Vogelsang Suarez A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree o f Chemical Engineering Doctor of Philosophy 201 9 ABSTRACT SEPARATION AND CHARACTERIZATION OF CLOSED FUNCTIONALIZED DOUBLE - DECKER SHAPED SILSESQUIOXANES By David Felipe Vogelsang Suarez Stoichiometric reaction of t etrasilanol octaphenyl double - decker sha ped silsesquioxanes (DDSQ - (Ph) 8 (OH) 4 ) with difunctional dichlorosilanes (R 1 R 2 SiCl 2 ) provided a new class of model hybrid organic - inorganic compounds. These model compounds (DDSQ - 2(R 1 R 2 )) have a dimensionally well - def ined closed inorganic Si - O core, eight p henyl groups for thermo - oxidative stability and compatibility to aromatic organics, and well - specified R 1 and R 2 for possible further chemical reactions. However, these compounds contain cis and trans conformations a bout the DDSQ if R 1 and R 2 are different . Separation of cis and trans isomers by liquid chromatography was proposed as the alternative technique to the more tedious fractional crystallization for isolation of pure compounds. It was found polar nature of the R group enables the separation by adso rption chromatography. In con trast, partition chromatography allows separation of isomers of a larger solubility difference. Results of HPLC also provided quantitative measure of the isomer ratio in a cis and trans DDSQ - 2(R 1 R 2 ) mixture with deviations bett er than ±5%. In contrast, the current quantification of the isomer ratio by 29 Si - NMR has been reported to have ±10% deviation from the real value. HPLC separation was extended to analyze resolution of the elution for DDSQ mixtures with polarity differences . Reaction of DDSQ - (Ph) 8 (OH) 4 with R 1 R 2 SiCl 2 and (CH 3 )SiCl 3 mixture followed by hydrolysis, lead to a mixture of DDSQ compounds with zero, one, and two hydroxyl groups. Good separation of the three expected products was made by LC. Characterization by NMR and mass - spectroscopy allowed identification of each separated product. An asymmetric structure about the DDSQ core with one hydroxyl group (DDSQ - (R 1 R 2 )((CH 3 )(OH)) was obtained. Scale - up of HPLC separation for the mixture containing zero, one, and two hyd roxyl groups was simulated in ASPEN chromatography. Linear adsorption isotherm parameters were obtained by frontal analysis. A correlation between HPLC stationary phase adsorption parameters and preparative stationary phase LC adsorption parameters was obt ained. These parameters permi tted satisfactory prediction of the column efficiency, the resolution of the elution, and total collection time in a column separation verified with a 5g - scale. In the scale of preparative - LC, fractions collected were of high p urity as verified by 29 Si NMR . Additionally, for DDSQ - 2(R 1 OH) a highly cis concentrated fraction and a nearly - pure trans fraction were also successfully isolated. DSC experiments were performed for nearly - pure cis and trans DDSQ - 2(R 1 R 2 ), and for DDSQ - 2(R 1 R 2 ) mixtures of varying cis - to - trans compositions. R 1 was fixed as methyl, and for R 2 aryl groups were selected. It was found trans isomer had a higher melting temperature than cis isomer and as the size of R 2 increases the melting temperatures of nearly - pu re isomers decrease. Interesting ly, cis and trans structures are not miscible in the solid state and form binary eutectic. Binary cis and trans eutectic temperature and composition can be predicted using the ideal binary assumption. Copyright by DAVID FELIPE VOGELSANG SUAREZ 2019 v To my beautiful wife Andrea Garcia . You have been the most important person through my life and along thi s study. It was because of you that I applied to the scholarship, and it is because of you that I am a better person now. I expect to fulfill you r expectations , and I am willing to keep providing more love to our growing family. I am very happy for our pro jects, for Tommy, for Beyoncé, and for all the good things that are coming for us. I lo ve you. vi ACKNOWLEDGMENTS I would like to acknowledge the institutions that fund my research. First, I want to thank Fulbright - Colombia, and COLCIENCIAS - Colombia for choose me to develop doctoral studies in USA and their economic support. I would like to thank the Office of Naval R esearch for funding this project through the grant N00014 - 16 - 2109 . Finally, I want to thank Michigan State University trough the CHEMS department for the unconditional financial support. I want to acknowledge the work of the prof essors in my research commi ttee. I h ighlight the guidance and patience of my advisor Dr. Andre Lee. He accepted me as his student in Fall 2013 and every day he challenges me to become a better researcher teaching me how to think, how to proceed experimenta lly, and how to formulate q uestions to find solutions to any scientific problem. His contribution was beyond his duty and he help me to become a better writer and a better speaker. Dr. Robert Maleczka for his guidance through the organic chemistry problems and his disposition to ans wer my questions every time his door was open. His contributions to this project were invaluable . H is advice solved every single issue I had related with the chemistry component of this dissertation. Dr. Carl Lira for his explana tions about how to analyze adsorption isotherms and made suggestions to my project that allowed me to generate the chapter 4 of this dissertation. Besides his contribution he was always willing to solve any question I had. Besides his contribution he is an excellent professor and I enjoyed his thermo and distillation column modelling class . Finally, Dr. Dennis Miller for his help when I was his student in the distillation column modelling by ASPEN. Also, his comments were very helpful to enhance the quality of this research. I apprec iate he was always willing to help with any question. vii Beyond the committee members I would like to express my gratitude to my friends in chemistry. Besides their contributions, their kindness and solidarity make very special my w ork in the laboratory. From them I would like to highlight the contributions of Jonathan Dannatt, his ideas and explanations help me in such a way that we even wrote research papers together. From this group I also want to remark the contributions of Gayan thi Attanayake, she taught me almost every single organic chemistry technique to develop this research. All faculties in the c hemistry department were also very kind offering their help when requested. I also like to thank to my research group friends Dr. Yuelin Wu , and Dr. Yang Lu because they were a great support in those times where we did not have idea of how to present or read properly papers. Parker Dunk was key person for production of material used in chapter 2 of this work. More recently Aditya Pat el join our group and I tha nk him to be willing to learn and work as a team trying to develop all sort of ideas. I would like to do a mention to the CHEMS department staff, all of them made a great job with paperwork, orders, solving visa issues, and much more. Also, the CHEMS depar tment faculties from whom I learn several new ideas and concepts along seminars, conferences, and classes. Next, I want to make a special mention to my Latin - American friends specially my parchecito friends. They were my family i n USA and these fun moments we enjoyed together make me keep going to reach this goal. This group of people was one of the best parts of my l ife and I expect we keep being friends forever. I want to thank my family in Colombia who understood my wish to com e to USA to develop PhD studies. They were always supportive thinking in my needs. I expect they keep feeling proud of me and specially from my mother Claudia Suarez who has been always an inspiration to me. She is the toughest person I have known in my en tire life. viii TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ....................... xi LIST OF FIGURES ................................ ................................ ................................ ................... xiii LIST OF SCH EME S ................................ ................................ ................................ ................. xxi KEY TO SYMBOLS AND ABBREVIATIONS ................................ ................................ ... xxiii CHAPTE R 1. INTRODUCTION ................................ ................................ ................................ 1 1 I ntrodu ction ................................ ................................ ................................ ........................... 2 1.1 Hybrid organic - inorganic silsesquioxanes: model molecules ................................ ........ 2 1.2 Synthesis of octaphenyl double - decker shaped s ilsesqui oxane ................................ ..... 3 1.3 Side - capping of DDSQ - (Ph) 8 (OH) 4 ................................ ................................ .................. 4 1.4 Separation techniques used for separation of DDSQ - 2(R 1 R 2 ) systems ........................ 6 1.4.1 Fractional crystallization ................................ ................................ .............................. 6 1.4.2 Liquid chromatography ................................ ................................ ................................ 7 1.4.3 Sc ale of separation ................................ ................................ ................................ ....... 9 1.4.4 Modeling and separation efficiency ................................ ................................ ........... 10 1.1 Motivation ................................ ................................ ................................ ........................ 12 1.2 Research goal ................................ ................................ ................................ ................... 12 1.3 Research idea ................................ ................................ ................................ ................... 13 1.4 Dissertation summary ................................ ................................ ................................ ..... 13 CHAPTER 2. HPLC CHARACTERIZATION OF CIS AND TRANS MIXTURES OF DOUBLE - DECKER SHAPED SILSESQUIOXANES ................................ ........................... 15 2 HPLC characterization of cis and trans mixtures of doub le - decker shaped sil sesquioxanes ................................ ................................ ................................ ..................... 16 2.1 Introduction ................................ ................................ ................................ ..................... 16 2.2 Experimental ................................ ................................ ................................ ................... 18 2.2.1 Mat erials ................................ ................................ ................................ ..................... 18 2.2.2 Synthesis of (methyl)(meta - Bis(trimethylsilyl)amino]phenyl)dichlorosilane ........... 18 2.2.3 Ge neral synthetic procedure ................................ ................................ ....................... 19 2.2.4 Analytical methods ................................ ................................ ................................ .... 20 2.3 Results and Discussion ................................ ................................ ................................ .... 22 2 .3.1 Separation of cis and trans isomers ................................ ................................ ............ 22 2.3.2 UV absorbance intensity of positional isomers of phenylamine ................................ 24 2.3.3 Analys is by HPLC of isomeric mixtures and individual isomers. ............................. 25 2.3.4 Separation by adsorption chromatography ................................ ................................ 28 2.3.5 Effect of R 2 groups in retention times ................................ ................................ ........ 30 2.3.6 Effect of polar groups (R 1 ) in t r ................................ ................................ .................. 31 2.4 Conclusions ................................ ................................ ................................ ...................... 32 CHAPTER 3. SEPARAT ION OF ASYMMETRICALLY CAPPED DOUBLE - DECKER SILSESQUIOXANES MIXTURES ................................ ................................ .......................... 34 ix 3 Separation of asymmetrically capped double - de cker silsesquioxanes mixtures ........... 35 3.1 Introduction ................................ ................................ ................................ ..................... 35 3.2 Experimental ................................ ................................ ................................ ................... 37 3.2.1 General information ................................ ................................ ................................ ... 37 3.2.2 General procedures ................................ ................................ ................................ .... 38 3.2.3 Separation of DDSQ mixtures by LC ................................ ................................ ........ 39 3.2.4 Characterization of DDSQ materials. ................................ ................................ ........ 39 3.3 Results and Discussion ................................ ................................ ................................ .... 40 3.3.1 Separation by LC ................................ ................................ ................................ ....... 40 3.3.2 HPLC Identification ................................ ................................ ................................ ... 41 3.4 Conclusions ................................ ................................ ................................ ...................... 45 CHAPTER 4. PREDICTIVE LIQUID CHROMATO GRAPHY SEPARATION FOR MIXTURES OF FUNCTIONALIZED DOUBLE DECKER SILSESQUIOXANES BASED ON HPLC CHROMATOGRAMS ................................ ................................ ............................ 46 4 Predictive liquid chromatography se paration for mixtures of functionalized double d ecker silsesquioxanes based on HPLC chromatograms ................................ ................ 47 4.1 Introduction ................................ ................................ ................................ ..................... 47 4.2 Exp erimental ................................ ................................ ................................ ................... 49 4.2.1 Materials ................................ ................................ ................................ ..................... 49 4.2.2 Methods ................................ ................................ ................................ ...................... 49 4.2.3 HPL C and preparative LC chromatograms ................................ ................................ 53 4.2.4 Chromatogram simulation ................................ ................................ .......................... 55 4.3 Results and Discussion ................................ ................................ ................................ .... 56 4.3.1 Breakthro ugh curves and calculated adsorption isotherms ................................ ........ 56 4.3.2 Simulation results and parameter fitting ................................ ................................ .... 58 4.3.3 Extrapolation of HPL C parameters to preparative column ................................ ........ 60 4.3.4 Efficiency for preparative columns ................................ ................................ ............ 63 4.4 Conclusion ................................ ................................ ................................ ........................ 65 CHAPTER 5. PHASE BEHAVIOR OF CIS - TRANS MIXTURES OF DOUBLE - DECKER SHAPED SILSESQUIOXANES FOR PROCESSABILITY ENHANCEMENT ................. 67 5 Phase behavior of cis - trans mixtures of double - decker shaped silsesquioxanes for processability enhancement ................................ ................................ ................................ 68 5.1 Introduction ................................ ................................ ................................ ..................... 68 5.2 Experimental ................................ ................................ ................................ ................... 71 5.2.1 Materials ................................ ................................ ................................ ..................... 71 5.2.2 Synthetic procedures ................................ ................................ ................................ .. 71 5.2.3 Capping of DDSQ - (Ph) 8 (OH) 4 ................................ ................................ ................... 74 5.2.4 Analytical methods ................................ ................................ ................................ .... 76 5.3 Results and discussion ................................ ................................ ................................ .... 76 5.3.1 Separation and identification of nearly - pure isomers ................................ ................ 76 5.3.2 Thermal behavior of nearly - pure isomers ................................ ................................ .. 78 5.3.3 Phase beh avior of cis - trans binary mixtures ................................ .............................. 79 5.3.4 Solid - liquid phase equilibrium of DDSQ - 2(( Me)(Ph)) 4 ................................ ........... 83 5.4 Conclusions ................................ ................................ ................................ ...................... 84 x CHAPTER 6. SIGNIFICANCE AND PERSPECTIVES ................................ ........................ 86 6 Significance and perspe ctives ................................ ................................ ............................. 87 6.1 Significance ................................ ................................ ................................ ...................... 87 6.2 Perspectives ................................ ................................ ................................ ...................... 88 APPENDICES ................................ ................................ ................................ ............................. 91 APPENDIX A. SYNTHETIC PR OCEDURES ................................ ................................ ........ 92 APPEND IX B. PERCENTAGES AFTER SEPARATION OF MIXTURES WITH ZERO, ONE, AND TWO HYDROXYL GROUPS ................................ ................. 105 APPENDIX C. TLC FOR SEPARATIO N OF DDSQ MIXTU RE WITH EACH SEPARATED FRACTION ................................ ................................ ........... 107 APPENDIX D. STRUCTURAL ANALYSIS OF AB1 - D BY 29 Si - NMR AND MASS S PECTROSCOPY ................................ ................................ ......................... 109 APPENDIX E. STRUCTURAL ANALYSIS OF A NON - POLAR MIXTURE BY 29 Si NMR ................................ ................................ ................................ ................ 112 APPENDIX F. KINETIC ANALYSIS OF DDSQ - (Ph) 8 (OH) 4 SIDE - CAPPED WITH DICHLOROSILANES WITH DIFFERENT STERIC GROUPS ............ 114 APPENDIX G. LC AND FC SEPARATIONS ANALYZED BY HPLC ............................. 144 APPENDIX H. SU MMARIZED EUTECTIC AND LIQUIDUS COMPOSITIONS ......... 155 APPENDIX I. NMR SPECTRA FOR COMPONENTS SYNTHESIZED AND SEPARATED IN THIS WORK ................................ ................................ ... 158 APPEND IX J. CRYSTALLOGRAPHIC INFORMATION ................................ ................ 206 REFERENCES ................................ ................................ ................................ .......................... 217 xi LIST OF TABLES Table 1 - 1 . Differences in melting temper ature for cis and trans isomers for selected systems. 15,27 ................................ ................................ ................................ ................................ ......................... 5 Table 1 - 2 . Chromatography classification, adapted from Snyder et al. 2013. 36,37 ......................... 8 Table 2 - 1 . TLC retardation factors, R f , with dichloromethane as mobile phase for 2 DDS Q - 2(( p - aniline)(methyl)), 3 DDSQ - 2(( m - aniline)(methyl)), 4 DDSQ - 2(( m - aniline)(isobutyl)), 5 DDSQ - 2(( m - aniline)(cyclohexyl)), and 11 DDSQ - 2((cyanopropyl)(methyl)). ................................ ........ 23 Table 2 - 2 . Comparison of the re solution obtained using equations 2 and 3 in cis - trans of 2 ...... 27 Table 2 - 3 . Comparison between cis and t rans percenta ges of 2 calculated by weighting nearly - pure cis and nearly - pure trans and calcula ted from the area under the peaks in the chromatograms presented in Figure 2 - 5 . The standard deviation was calculated based on the known percentage and the area percentage fo r each isomer. ................................ ........................ 28 Table 2 - 4 . Retention time (t r ), peak width (W), and plate number (N) after separation by adsorption HPLC with DCM as the mobile phase; Retention time (t rAcN ) after s eparation by adsorption HPLC with a mobile phase composed by DCM:acetonitrile in the volumetric ratio 98:2. ................................ ................................ ................................ ................................ .............. 29 Table 2 - 5 . Retention time for DDSQ functionalized with methyl and different polar gro ups. R stands for para - aniline 2 , hydroxyl 6 , and cyanopropyl 11 . ................................ ......................... 31 Table 3 - 1 . The calculated ratio of products in DDSQ mixtures after separation by HPLC . ........ 43 Table 3 - 2 . Mass fraction analysis of products (in precent) after DDSQ mixtur es synthesis with different ratios of methyldichlorosilane and methyltrichlorosilane. *HPLC column is calculated based on HPLC peak analysis and Mass column is calculated based on analytic balance measurement after separation by preparatory liquid chromato graphy Mass. ............................... 45 Table 4 - 1 . Solutions prepared for obtention of breakthrough curves. Concentration s of 4a are for a 1:1 mixture of cis and trans isomers. ................................ ................................ ......................... 52 Table 4 - 2 . Preparative column dimensions and operational parameters ................................ ...... 54 Table 4 - 3 . Linear adsorption i sotherm parameter (IP1) values obtained experimentally and calculated values. ................................ ................................ ................................ .......................... 60 Table 4 - 4 . Column efficiency (N) calculated from Equation 3 for HPLC and preparative column s. * Value calculated as an individual component. ................................ ........................... 64 Table 4 - 5 . Resolution of the elution bet ween analytes in each mixture after separation by HPLC and preparative LC. ................................ ................................ ................................ ....................... 65 Table 5 - 1 . Crystallization of individual isomers ................................ ................................ .......... 78 xii Table 5 - 2 . Experimental values obtained from nearly - pure cis and trans DDSQ - 2((Me)(R)) by m m /T m ................................ ................................ ................................ ..................... 79 Table B - 1. Isolated yield after column for com ponents in the ternary mixture ......................... 106 Table H - 1 . Eutectic temperature (T E ) and liquidus temp eratures (T L ) for binary cis - to - trans mixtures of compound 2 . ................................ ................................ ................................ ............ 156 Table H - 2 . Eutectic temperature (T E ) and liquidus temperatures (T L ) for binary cis - to - trans mixtures of compound 3 . ................................ ................................ ................................ ............ 156 Table H - 3 . Eutectic temperature (T E ) and liquidus temperatur es (T L ) for binary cis - to - trans mixtures of compound 4 . ................................ ................................ ................................ ............ 156 xiii LIST OF FIGURES Figure 1 - 1 . Partially condensed oligomeric silsesquioxanes used as model molecules. ............... 3 Figure 1 - 2 . Fractional crystallization diagram ................................ ................................ ............... 6 Figure 1 - 3 . Separation by liquid chromatography. t 0 column with stationary phase and wet with m obile phase; t 1 mixture injected on the packed bed; t 2 migration of analytes with different elution rates; t 3 , t 4 and t 5 elut ion of separated c omponents in different times. ............................... 7 Figure 1 - 4 . Peak shape associated to the adsorption isotherm a) type I, b) type II, an d c) type III, adapted from Fornstedt et al. , 2013. 57 ................................ ................................ ........................... 10 Figure 1 - 5 . Retention time and peak width calculation in the baseline of a g aussian - like elution peak. ................................ ................................ ................................ ................................ .............. 11 Figure 2 - 1 . DDSQ - 2(R 1 R 2 ) isomers studied in this wo rk. (2) DDSQ - 2((para - aniline)(methyl)), (3) DDSQ - 2((meta - aniline)(methyl)), (4) DDSQ - 2((meta - aniline)(isobutyl)), (5 ) DDSQ - 2((meta - aniline)(cyclohexyl)), (6) DDSQ - 2((hydroxyl)(methyl)), (7) DDSQ - 2((hydroxyl)(vinyl)), (8) DDSQ - 2((hydroxyl)(isopropyl)), (9) DDSQ - 2((hydroxyl)(isobutyl)), (10) DDSQ - 2((hydroxyl)(phenyl)), (11) DDSQ - 2((cyan opropyl)(methyl)). ................................ ................... 20 Figure 2 - 2 . 29 Si NMR for cis and trans isomers after separation of 3 . (a) trans isomer, corres ponding to the first spot in TLC; (b) cis isomer, corresponding to the second spot in TLC. ................................ ................................ ................................ ................................ ....................... 24 Figure 2 - 3 . trans isomers, and filled symbols are for cis - 2(( p - aniline)(methyl)) 2 ; DD SQ - 2(( m - aniline)(methyl)) 3 DDSQ - 2(( m - aniline)(isobutyl)) 4; DDSQ - 2(( m - aniline)(cyclohexyl)) 5 . The red dashed line represents the linear trend of all the para samples, and the straight blue line is the linear trend for the meta samples. The slope of para was determined to be two times the slope for meta. ................................ ................................ ................................ ................................ ....................... 25 Figure 2 - 4 . Effect of dichloromethane/hexane mobile phase ratios in normal phase pHPLC for cis - trans of 2 and 3 . The percentual value indicated represents the volume percentage of hexanes in the mobile phase. ................................ ................................ ................................ ...................... 26 Figure 2 - 5 . Quantitative analysis of cis - trans mixtures of DDSQ - 2(( p - aniline)(methyl)) by pHPLC. The individual isomers were first isolated and then mixed to a known ratio for comparison against the area under each peak. The w ei ghted precent of isolated isomers in mixtures was indicated next to each curve. ................................ ................................ .................. 27 Figure 2 - 6 . Retention times for cis and trans DDSQ - 2((hydroxyl)(R 2 )). (a) R 2 : methyl 6 ; (b) R 2 : vinyl 7 ; (c) R 2 : i sopropyl 8 ; (d) R 2 : isobutyl 9 ; and (e) R 2 : phenyl 10 . ................................ ........ 31 Figure 3 - 1 . Chromatograms for products of pure (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R) 2 obtained following Scheme 3 - 1 (a - e) . Chromatograms for mixtures of ( C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R) 2 , xiv (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R)(OH) , and (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (OH) 2 following synthesis proposed in Scheme 3 - 2 . The absorbance in the region between 15 and 30 minutes is zoomed for reader convenience (f - i) . The second fraction separated by LC corresponding to (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R)(OH) (j - m) . ................................ ................................ ............................ 42 Figure 4 - 1 . Breakthrough curves for a) 2 in low concentrations, b) 2 in high concentrations, c) 3a in low concentrations, d) 3a in hi gh co ncentrations, e) 4a in low concentrations, and f) 4a in high concentrations. t s was calculated from the half - concentration in the curve front. ................ 57 Figure 4 - 2 . Experimental adsorption isotherms and l inear fitting for: 2 black squares , 3a green circles trans - 4a blue triangles , and cis - 4a red diamonds . a) full experimental points; b) low concentration points. Color graph can be obtained in the digital version of this document. . 58 Figure 4 - 3 . Linear behavior between the retention times (t r ) in HPLC and the linear isotherm parameter IP1 obtained from FA for mixture A . ................................ ................................ .......... 59 Fi gure 4 - 4 . Curves in red represent the chromatograms obtained by HPLC. Curves in dotted blue lines represent the simulated chromatograms for a) mixture A , b) mixture B , c) mixture C . ...... 60 Figure 4 - 5 . Curves i n red represent the chromatograms obtained by preparative liquid chromatography. Curves in dotted blue lines represent the simulated chromatograms for a) mixture A , b) mixture B , c) mixture C . ................................ ................................ ........................ 61 F igure 4 - 6. The relation between IP1 obtained from HPLC (IP1 HPLC ) and IP1 obtained from Preparative column (IP1 prep ) ................................ ................................ ................................ ......... 62 Figure 5 - 1. Structure of cis/trans - DDSQ - 2(R 1 R 2 ) and POSS - R 1 where R are inert organic moieties and R 1 and R 2 are active functi onal groups. ................................ ................................ ... 69 Figure 5 - 2 . Structures synthesized and studied in this work. ................................ ....................... 75 Figure 5 - 3 . 29 Si NMR peaks representing the nearly - pure isomers after separation; a) cis - 2 , b) trans - 2 , c) cis - 3 , d) trans - 3 , e) 75% cis - 4 , and f) trans - 4 . ................................ ............................ 77 Figure 5 - 4 . a) DSC curves for comp ound 2 . Every curve was normalized for better identification of peaks. The reported x trans was estimated using NMR. b) Binary phase diagram for structure 2 . sq uares ( ) repr esent the peak temperature of the highest endothermic transition. The solid line represents the ideal eutectic as calculated using Equation 5 - 1 ; dashed line ( --- ) represents the calculated eutectic temperature T E which matches the transition temperature ob served using a sample with the predicted eutectic composition. Phase I: L ( ci s + trans ) ; Phase II: L ( cis + trans ) + S cis ; Phase III: L ( cis + trans ) + S trans ; Phase IV: S cis + S trans . ................................ ................................ ...... 81 Figure 5 - 5 . a) DSC curves for compound 3 . Every curve was normalized for better identification of peaks. b) Binary phase diagram for structure 3 . the first endothermic transition in DSC trace. Blue squares ( ) represent the peak temperature of the highest endothermic transition. The solid line represents the ideal eutectic as calculated using xv Equation 5 - 1 ; dashed line ( --- ) represents the calculated eutectic temperature T E. . Phase I: L ( cis + trans ) ; Phase II: L ( cis + trans ) + S cis ; Phase III: L ( cis + trans ) + S trans ; Phase IV: S cis + S trans . ................... 82 Fi gure 5 - 6 . a ) DSC curves for compound 4 . Every curve was normalized for better identification of peaks. b) Partial binary ph ase diagram for structure 4 . Green do temperatures of the first endothermic transition in DSC trace. Blue squares ( ) represent the peak temperature of the highest endothermic transition. The solid line represents the ideal solid - liquid equilibrium, light blue dashed line ( --- ) represents the experimental eutectic temperature T E . For nearly - pure trans - 4 are depicted the nearly - onset temperature from the second endo peak( ), and the pe ak temperature from the second e ndo p eak ( ). Phase I: L ( cis + trans ) ; Phase II: L ( cis + trans ) + S cis ; Phase III: L ( cis + trans ) + S trans ; Phase IV: S cis + S trans . ................................ ................................ ................................ ................................ .... 84 Figure C - 1 . TLC after separat ion of ternary DDSQ mixture ................................ ..................... 108 Figure D - 1 . 29 Si - NMR and mass spectrums obtain ed after characterization of DDSQ - (methyl)(R)(methyl)(hydroxyl) products obtained as the second fraction of the separat ion by LC of DDSQ mixtures. a) R = hydrogen, b) R = methyl, c) R = vinyl, d) R = 3 - propyl chloride. .. 111 Figure E - 1 . 1 H (a), 13 C (b), and 29 Si - NMR (c), for non - polar mixture of DDSQ - 2((methyl)(hydro )) AA1, DDSQ - 2(methyl)(3 - propyl ch loride) AA4, and DDSQ - (methyl)(hydro)(met hyl)(3 - propyl chloride) A1A4 synthesized from Cl 2 Si(H)(CH 3 ) and Cl 2 Si(CH 2 CH 2 CH 2 Cl)(CH 3 ). 29 Si - NMR is compared against pure AA1 and pure AA4 obtained following general procedure B. T his mixture is not separable by th e methods employed in this work due to lack of polar moieties in the functionalized DDSQ. ................................ ............... 113 Figure F - 1 . Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2(Ph) 2 in different times analyzed by 1 H - NMR (500 MH z, CDCl 3 ) after quench with MeOH: a = DDSQ - (Ph) 8 (OH) 4 , b = Ph 2 SiCl 2 , c = 0.67 min , d = 1.35 min, e = 2.03 min, f = 4.63 min, g = 6.85 min, h = 10.83 min, i = 20 min, j = 36 min, k = 100 min ................................ ................................ ................................ .................... 115 F igure F - 2 . Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2(Ph) 2 at different times analyzed by 29 Si - NMR (99 MHz, CDCl 3 ) after quench with MeOH and solvents evaporation (times in minutes): a=DDSQ - (Ph) 8 (OH) 4 , b = 0.67 min, c = 2.03 mi n, d = 6.85 min, e = 100 min, f = DDSQ - 2(Ph) 2 completed after 4 hours reaction ................................ ................................ ......... 116 Figure F - 3 . Conversion of DDSQ - (Ph) 8 (OH) 4 to AA1 or DDSQ - 2(Me) 2 in different times analyzed by 1 H - NMR (500 MHz, CD Cl 3 ) after quench with MeOH. Ti mes in minutes: a = 0 only DDSQ - (Ph) 8 (OH) 4 , b = 0.98 min, c = 100 min. ................................ ................................ .......... 117 Figure F - 4 . Time conversion of DDSQ - (Ph) 8 (OH) 4 to AA1 or DDSQ - 2(Me) 2 analyzed by 29 Si - NMR ( 99 MHz, CDCl 3 ) after quench wi th MeOH and solvents evaporation: a=DDSQ - (Ph) 8 (OH) 4 , b = 0.98 min, c = 100 min. ................................ ................................ ...................... 118 Figure F - 5 . Fractions of DDSQ(OH) 4 (grey line), DDSQ - 2(R 1 R 2 ) (blue line), and DDSQ( OH) 2 (ora nge line) obtained f rom model 1 after evaluation with different equivalents of chlorosilane. ................................ ................................ ................................ ................................ ..................... 122 xvi Figure F - 6 . Fractions of DDSQ - (B) 4 or BB (grey line), DDSQ - (A) 4 or AA (blue line), and DDSQ - (A 2 B 2 ) or AB (orange line) obtained from model 2 after evaluation with different equivalents of A and completion to two equivalents with B assuming A and B are equally reactive. ................................ ................................ ................................ ................................ ....... 128 Figure F - 7 . Fr actions of DDSQ - (B) 4 or BB (grey line), DDSQ - (A) 4 or AA (blue line), and DDSQ - (A 2 B 2 ) or AB (orange line) obtained from model 2 after evaluation with different equivalents of A and completion to two equivalents with B assuming A is ten times faster than B ( p = 10). ................................ ................................ ................................ ................................ ....... 129 Figure F - 8 . Modelling of functionalization of DDSQ(OH) 4 with two different chlorosilanes including the formation of byproducts. The kinetic constants are not real values but related to k1 based on experimental observations. ................................ ................................ .......................... 142 Figure F - 9 . Evolution of triethylamine, triethylamine complexes, and water production in the DDSQ - (OH) 4 functionalization. ................................ ................................ ................................ . 143 Figure G - 1 . Flow rate ramps for preparative column under non - constant flow rate. ................ 147 Figure G - 2 . 50% tran s and 50% cis mixture of DDSQ - 2((methyl)(para - phenylethynyl phenyl))after hypercar b column ................................ ................................ ................................ .. 152 Figure G - 3 . Mostly trans isomer of DDSQ - 2((methyl)(para - phenylethynyl phenyl))after hypercarb column ................................ ................................ ................................ ........................ 152 Figure G - 4 . Mostly cis isomer of DDSQ - 2((methyl)(para - phenylethynyl phenyl))after hypercarb column ................................ ................................ ................................ ................................ ......... 152 Figure G - 5 . Possible separation of the ternary mixture described in Scheme G - 1 . Synthesis o f a ternary non - pola r mixture with diphenyldichlorosilane and dimethyldichlorosilane. Scheme G - 1 using 98:2 DCM:Acetonitrile as mobile phase. Blue line represents the chromatogram with a DCM injection; yellow line is the chromatogram for a mixture of tetram ethyl DDSQ and tetra phenyl DDSQ. Red line represents the chromatogram for the ternary mixture. ................. 153 Figure I - 1 . 29 Si - NMR (CDCl 3 , 99 MHz) for (isobutyl)(para - aniline(trimethylsilyl))dichlorosilane ................................ ................................ ......................... 159 Figure I - 2 . 1 H - NMR (CDCl 3 , 500 MHz) for (isobutyl)(para - aniline(trimethylsilyl))dichlorosilane ................................ ................................ ......................... 159 Figure I - 3 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((methy l)(para - aniline)) ........................ 160 Figure I - 4 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((methyl)(para - aniline)) ........................ 160 Figure I - 5 . 29 Si - NMR (CD Cl 3 , 99 MHz) DD SQ - 2((methyl)(meta - aniline)) ............................. 161 Figure I - 6 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((methyl)(meta - aniline)) ....................... 161 xvii Figure I - 7. 29 S i - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((isobutyl)(meta - aniline)) ...................... 162 Figure I - 8 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((isobutyl)(meta - aniline)) ..................... 162 Figure I - 9 . 29 Si - NMR ( CDCl 3 , 99 MHz) for DDQS - ((cyclohexyl)(meta - aniline)) ................... 163 Figure I - 10 . 1 H - NMR (CDCl 3 , 500 MHz) for DDQS - ((cyclohexyl)(meta - aniline)) ................. 163 Figure I - 11 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((methyl)(hydroxyl)) ........................... 164 Figure I - 12 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((methyl)(hydroxyl)) ........................... 164 Figure I - 13 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((vinyl)(hydroxyl)) .............................. 165 Figure I - 14 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((vinyl)(hydroxyl)) .............................. 165 Figure I - 15 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((isopropyl)(hydroxyl)) ....................... 166 Figure I - 16 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((isopropyl)(hydroxyl)) ....................... 166 Figure I - 17 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((isobutyl)(hydroxyl)) ......................... 167 Figure I - 18 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((isobutyl)(hydroxyl)) ......................... 167 Figure I - 19 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((phenyl)(hydroxyl)) ........................... 168 Figure I - 20 . 1 H - NMR (CDCl 3 , 50 0 MHz) for DDSQ - 2((phenyl)(hydroxyl) ) ........................... 168 Figure I - 21 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((methyl)(propyl - cyanide)) ................. 169 Figure I - 22 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((me th yl)(propyl - cyanide)) ................. 169 Figure I - 23 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydro)) ................................ 170 Figure I - 24 . 13 C - NMR (125 MHz, CDCl 3 ) fo r DD SQ - 2((methyl)(hydro)) ............................... 171 Figure I - 25 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydro)) ................................ 172 Figure I - 26 . 1 H - NMR (500 MHz, CDCl 3 ) for D DSQ - 2(methyl) 2 ................................ ............. 173 Figure I - 27 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2(methyl) 2 ................................ ............ 174 Figure I - 28 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2( methyl) 2 ................................ ............. 175 Figure I - 29 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(vinyl)) ................................ . 176 Figure I - 30 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2((meth yl)(vinyl) ) ................................ 177 Figure I - 31 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(vinyl)) ................................ . 178 xviii Figure I - 32 . 29 Si - NMR (99 MHz, Acetone - D6) for DDS Q - 2((methyl) (vinyl)) ........................ 179 Figure I - 33 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2(methyl)(3 - propyl - chloride) ............... 180 Figure I - 34 . 13 C - NMR (125 MHz, C DCl 3 ) for DDSQ - 2(methyl)(3 - propyl - chloride) .............. 181 Figure I - 35 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2(methyl)(3 - propyl - chloride) ................ 182 Figure I - 36 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) ........................... 183 Figure I - 37 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) .......................... 184 Figure I - 38 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) hydrolyzed by column chromatography ................................ ................................ ................................ ............. 185 Figure I - 39 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) hydrolyzed with a cidified H 2 O ................................ ................................ ................................ .............................. 185 Figure I - 40 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl)(hydro)(methyl)(hydroxyl) ..... 186 Figure I - 41 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl)( hydro)(methyl)(hydroxyl) .... 187 Figure I - 42 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl)(hydro)(methyl)(hydroxyl) ...... 188 F igure I - 43 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (methyl)(hydroxyl) ................ 189 Figure I - 44 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (methyl)(hydroxyl) ............... 190 Figur e I - 45 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (methyl)(hydroxyl) ................ 191 Figure I - 46 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) ....... 192 Figure I - 47 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) ...... 193 Figure I - 48 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl)(vinyl)(m ethyl)(hydroxyl) ....... 194 Figure I - 49 . 29 Si - NMR (99 MHz, Acetone - D6) for DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) ................................ ................................ ................................ ................................ ..................... 195 Figure I - 50 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl)( 3 - propyl - chloride)(methyl)(hydroxyl) ................................ ................................ ................................ ....... 196 Figure I - 51 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl)(3 - propyl - chloride)(methyl)(hydroxyl) ................................ ................................ ................................ ....... 197 Figure I - 52 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl)(3 - propyl - chloride)(methyl)(hydroxyl) ................................ ................................ ................................ ....... 198 Figure I - 53 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (vinyl)(hydr oxyl) ................... 199 xix Figure I - 54 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (vinyl)(hydroxyl) .................. 199 Figure I - 55 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - ( methyl) 2 (vinyl)(hydroxyl) ................... 199 Figure I - 56 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (isobutyl)(hydroxyl) ............... 200 Figure I - 57 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (isobut yl)(hydroxyl) ............. 200 Figure I - 58 . 29 Si - NMR ( 99 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (isobutyl)(hydroxyl) ............... 200 Figure I - 59 . 1 H NMR ( 500 MHz, CDCl 3 ) fo r (methyl)(para - aniline(trimethylsilyl))dichlorosilane ................................ ................................ ......................... 201 Figure I - 60 . 13 C NMR ( 125 MHz, CDCl 3 ) for (methyl)(para - aniline(trimethylsilyl))dichlorosilane ................................ ................................ ......................... 201 Figure I - 61 . 29 Si NMR (99 MHz, CDCl 3 ) for (methyl)(para - aniline(trimethylsilyl))dichlorosilane ................................ ................................ ................................ ................................ ..................... 201 Figure I - 62 . 1 H NMR (500 MHz, CDCl 3 ) for 1,4 - Bromophenylethynylbenzene ..................... 202 Figure I - 63 . 13 C NMR (125 MHz, CDCl 3 ) for 1,4 - Bromophenylethynylbenzene .................... 202 Figure I - 64 . 1 H NMR ( 500 MHz, CDCl 3 ) for 1,4 - (Phenylethynyl)pheny l methyldichlorosilane ................................ ................................ ................................ ................................ ..................... 203 Figure I - 65 . 13 C NMR (125 MHz, CDCl 3 ) for 1,4 - (Phenylethynyl)phenyl methyldichlorosilane ................................ ................................ ................................ ................................ ..................... 2 03 Figure I - 66 . 29 Si NMR (99 MHz, CDCl 3 ) for 1,4 - (Phenylethynyl)phenyl methyldichlorosilane ................................ ................................ ................................ ................................ ..................... 203 Figure I - 67 . 1 H NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(para - phenylethynyl phenyl)) ................................ ................................ ................................ ................................ ..................... 204 Figure I - 68 . 13 C NMR (125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(para - phenylethynyl phenyl)) ................................ ................................ ................................ ................................ ..................... 204 Figure I - 69 . 29 Si NMR (99 MHz, C DCl 3 ) for DDSQ - 2((methyl)(para - phenylethynyl phen yl)) 204 Figure I - 70 . 1 H NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(phenyl)) ............................... 205 Figure I - 71 . 13 C NMR (125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(p henyl)) ............................. 205 Figure I - 72 . 29 Si NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(phenyl)) ............................... 205 Fi gure J - 1 . Molecular structure of cis - DDSQ - 2((methyl)( para - phenylamine)). White = H, Red = 0, Gray = C, Yellow = Si, Blue = N ................................ ................................ ............................ 207 xx Figure J - 2 . Packing structure for cis - DDSQ - 2((methyl)(para - phenylam ine)) .......................... 208 Figure J - 3 . Molecular structure of trans - DDSQ - 2((methyl)(para - phenylamine)). White = H, Red = 0, Gray = C, Yellow = Si, Blue = N ................................ ................................ ........................ 209 Figure J - 4 . Packing structure for trans - DDSQ - 2((m ethyl)(para - phenylamine)) ....................... 210 Figure J - 5 . Molecular structure of cis - DDSQ - 2((methyl)(para - phenylethynyl phenyl)) White = H, Red = 0, Gray = C, Yellow = Si, Blue = N ................................ ................................ ............ 211 Figure J - 6 . Packing structure for cis - DDSQ - 2((methyl)(para - phenylethynyl phenyl)) ............ 212 Figure J - 7 . Molecular structure of trans - DDSQ - 2((methyl)(para - phenylethynyl phen yl)) White = H, Red = 0, Gray = C, Yellow = Si, Blue = N ................................ ................................ ......... 213 Figure J - 8 . Packing structure for trans - DDSQ - 2((methyl)(para - phenylethynyl p henyl)) ........ 214 Figure J - 9 . Molecular structure of trans - DDSQ - 2((methyl)(phenyl)) White = H, Red = 0, Gray = C, Yellow = Si, Blue = N ................................ ................................ ................................ ......... 215 Fi gure J - 10 . Packing structure for trans - DDSQ - 2((methyl)(phenyl) ) ................................ ....... 216 xxi LIST OF SCHEMES Scheme 1 - 1 . Synthesis of DDSQ - (Ph) 8 (ONa) 4 from phenytrimethoxysilane ................................ 3 Scheme 1 - 2 . Cle aving of POSS to synt hesize DDSQ - (Ph) 8 (ONa) 4 ................................ ............... 4 Scheme 1 - 3 . Synthesis of closed DDSQ or DDSQ - 2(R 1 R 2 ) ................................ .......................... 5 Scheme 2 - 1 . Condensation reaction of DDSQ - (Ph) 8 (OH) 4 (1) with 2 molar equivalent of R 1 R 2 SiCl 2 . The resultant product contains cis and trans isomers DDSQ - 2((R 1 )(R 2 )). ................ 16 Scheme 3 - 1 . Side capping of DDSQ - (Ph) 8 (OH) 4 ( 1 ) with a dichloros ilane. ................................ 35 Scheme 3 - 2 . Side capping of DDSQ - (Ph) 8 (OH) 4 ( 1 ) with two chlorosilanes. ............................. 36 Scheme 3 - 3 . Proposed synthesis to obtain a mixture of AA, AB, and BB. ................................ . 41 Scheme 3 - 4 . Side - capping of DDSQ - (Ph) 8 (OH) 4 with chlorosilanes having moieties with different sterics, 100 % conversion time for capping with (CH 3 ) 2 SiCl 2 less than 1 minute, 100% conversion time for capping with (C 6 H 5 )SiCl 2 higher than 36 minutes. ................................ ...... 44 Scheme 4 - 1 . Capping of DDSQ - (Ph) 8 (OH) 4 with two diff e rent chlorosilanes proposed in chapter 3 ................................ ................................ ................................ ................................ ..................... 48 Scheme 4 - 2 . Synthesis of functionalized DDSQ - 2(R 1 )(R 2 ). For compound 2 : R 1 and R 2 are CH 3 ; for compound 4a : R 1 is CH 3 , and R 2 is Cl which is hydrolyzed after the reaction to form hydroxyl (OH) ................................ ................................ ................................ ............................... 50 Scheme 4 - 3 . Synthesis of mixtures containing DDSQ - 2(CH 3 ) 2 2 , DDSQ - (CH 3 ) 2 (R)(OH) 3 , and DDSQ - 2((R)(OH)) 4 , where R is methyl (CH 3 ), vinyl (CHCH 2 ), or is o butyl (CH 2 CH(CH 3 ) 2 ). .. 51 Scheme 5 - 1. Condensation of 1 with two equivalents of organo - dichlorosilanes ....................... 70 Scheme 5 - 2 . Synthesis of dichlo r o(methyl)(4 - (phenylamine(bis(trimethylsilyl))))silane ........... 71 Scheme 5 - 3 . Synthe sis of 1 - bromo - 4 - (phenylethynyl)benzene ................................ ................... 72 Scheme 5 - 4 . Synth e sis of dichloro(methyl)(4 - (phenylethynyl)phenyl)silane. ............................. 73 Scheme A - 1 . Synthesis of DDSQ - 2((methyl)(hydro)) ................................ ................................ . 94 Scheme A - 2 . Synthesis of DDSQ - 2(methyl) 2 ................................ ................................ .............. 95 Scheme A - 3 . Synthesis of DDSQ - 2((methyl)(vinyl)) ................................ ................................ .. 96 Scheme A - 4 . Synthesis of DDSQ - 2((methyl)(3 - propyl chloride)) ................................ .............. 97 Scheme A - 5 . Synthesis of DDSQ - 2((methyl)(hydroxyl)) ................................ ............................ 98 xxii Scheme A - 6 . Synthesis of DDSQ - (methyl)(hydro)(methyl)(hydroxyl) ................................ ..... 100 Scheme A - 7 . DDSQ - (methyl) 2 (methyl)(h ydroxyl) ................................ ................................ .... 101 Scheme A - 8 . DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) ................................ .......................... 102 Scheme A - 9 . DDSQ - (methyl)(3 - propyl chloride)(methyl)(hydroxyl) ................................ ....... 103 Scheme A - 10 . DDSQ - (methyl)(hydro)(methyl)(hydroxyl) varying equivalents ....................... 104 Scheme F - 1 . DDSQ(OH) 4 capping reactions generated for model 1 ................................ ......... 119 Scheme F - 2 . DDSQ(OH) 4 Capping reactions generated for model 2 ................................ ........ 123 Scheme F - 3 . Formation of triethylamine complex with chlorinated species ............................. 130 Scheme F - 4 . Capping of DDSQ(OH) 4 with chlorosilane triethylamine complex. .................... 131 Scheme F - 5 . Formation of byproducts from condensation r eactions in DDSQ(OH) 4 catalyzed by triethylamine. ................................ ................................ ................................ .............................. 132 Scheme F - 6 . Functionalization of POSS(OH) 2 with dic h lorosilanes as other possible side reactions. ................................ ................................ ................................ ................................ ..... 132 Scheme F - 7 . Production of polysiloxanes promoted by water production in DDSQ(OH)4 condensation. These reactions were not yet included in the mode l but they are highly likely based on multiple peaks observed by 29 Si - NMR in the D - Si region ................................ .................... 133 Scheme G - 1 . Synthesis of a ternary non - polar mixture with diphenyldichlorosilane and dimethyldichlo r osilane. ................................ ................................ ................................ ............... 153 xxiii KEY TO SYMB OLS AND ABBREVIATIONS Acronyms AA No n - polar symmetric double - decker shaped silsesquioxane AB Asymmetric double - decker shaped silsesquioxane BB Polar symetric double - decker shaped silsesquioxane C DCl 3 Deuterated chloroform DCM Dichloromethane DDSQ Double - decker shaped silsesquioxane DDSQ - (Ph) 8 (OH) 4 Tetrasilanol double - decker shaped silsesquioxane DDSQ - (Ph) 8 (ONa) 4 Tetrasodium cholate double - decker shaped silsesquioxane DDSQ - 2 (R1R2) Double - d ecker shaped silsesquioxane capped with difunctional dichlorosilanes DSC Differential scann ing calorimetry Et 3 N Triethylamine FC Fractional crystallization GC - MS Gas chromatography Mass spectroscopy HPLC High performance liquid chromatography I R Infrared xxiv L Liquid (in italic) LC Liquid chromatography Me Methyl NMR Nuclear magn etic resonance NP Normal phase PA para - phenylamine PEP para - phenylethynyl phenyl Ph Phenyl pHPLC Partition normal phase high performance chromatography POSS Polyhedral oligomeric silsesquioxane R Functional moiety attached to functional chlor osilanes or to functionalized double - decker shaped silsesquioxane RP Reverse phase R.T. Room temperature R 1 R 2 SiCl 2 Difunctional dichlorosilane S Solid (in i t alic ) THF Tetahydrofurane TLC Thin layer chromatography TMS Trimethyl silane xxv UPLC U ltra - high performance liquid chromatography UV Ultraviolet UV - VIS Ultraviolet - visible Symbols C Concentration S m Change of entropy at melting H m Change of enthalpy at melting i Interparticle void fraction p Intraparticle void fraction t Total void fraction E z Dispersion coefficient F v Flow rate Activity coefficient H Height of theoretical plate Hb Height of packed bed IP1 Linear adsorption isotherm parameter 1 IP2 Linear adsorption isotherm parameter 2 J Mass transfer flux xxvi L Column length MTC Mass transfer coefficient N Number of theoretical plates or col umn efficiency q v Co ncentration of analyte adsorbed in a determinate time and column location t 0 Hold - up time R Universal gas constant (in italic) R f Retardation factor R s Resolution of the elution t ext Extra column time T Temperature t Time T E Eutectic temper ature T L Liquidus temperature T m Melting temperature t r Retention time t s Shock time or front time W Peak width x Isomer fraction 1 CHAPTER 1 . INTRODUCTI ON 2 1. Introduction 1.1 Hybrid organic - inorganic silsesquioxanes: model molecules Polyhedral oligo meric silsesquioxanes (POSS) are organic - inorganic hybrid materials with a Si - O core surrounded by an organic corona having the common structure (RSiO 1.5 ) n . Partially co ndensed POSS has been functionalized to study properties of hybrid structures in divers e fields including chemistry, engineering, materials science, or medicine . 1 5 The most studied partially condensed POSS has the structure (RSiO 1.5 ) 7 (OH) 3 ( Figure 1 - 1 a ), Condensation of this structure by corner - capping with func tionalized trichlorosilanes or alkoxysilanes allow the use of these sructures as model molecules with specific functionalization in vicinal silanols . 6 8 Recently, tetrasilanol octaphenyl double - decker shaped silsesquioxanes DDSQ - (Ph) 8 (OH) 4 ( Figure 1 - 1 b ) was developed by Yoshida et al . , and further improved by Kawakami et al . 9 12 These are oligomeric structures with the formula ( PhSiO 1.5 ) 8 (O 0.5 H) 4 . The molecule has an inner Si - O core conformed by two silsesquioxane rings interconnected by two oxygens in oppo site edges. The remaining two edges are open, leaving four hydroxyl groups . The core is surrounded by eight phenyl rings bonded to each Si atom . The listed characteristics make DDSQ - (Ph) 8 (OH) 4 a model molecule to understand hybrid organic - inorganic systems with the benefit of specific number of functionalities in specific sites of the nanostructure. 3 Figure 1 - 1 . Partially condensed oligomeric silsesquioxanes used as model molecules. 1.2 Synthesis of octaphenyl do uble - decker shaped silsesquioxane Precursor of DDSQ - (Ph) 8 (OH) 4 is a tetrasodium salt (DDSQ - (Ph) 8 (ONa) 4 ). This salt has been synthesized by reac tion of phenyl(trimethoxy)silane with 0.7 equivalents of sodium hydroxide and 1.1 equivalents of water using isob utanol as solvent under reflux for 4h followed by stirring for 15h at room temperature a s observed in Scheme 1 - 1 . 9,11 Alternative production of DDSQ - (Ph) 8 (ON a) 4 has been achieved by cleaving fully condensed octaphenyl POSS . 12 This process was performed with 4 equivalents of sodium hydroxide and 2 equivalents of water using isopropanol under reflux or isobutan ol at 90°C as seen in Scheme 1 - 2 . Scheme 1 - 1 . Synthesis of DDSQ - (Ph) 8 (ONa) 4 from phenytrimethoxysilane 4 Scheme 1 - 2 . Cleaving of POSS to synthesize DDSQ - (Ph) 8 (ONa) 4 DDSQ - (Ph) 8 (ONa) 4 m ay be very unstable and further acid treatment is done to achieve the more stable DDSQ - (Ph) 8 (OH) 4 ( 1 ). This molecule is commercially available and different reactions can be performed with the silanol groups to add specific functionalities . 4 28 1.3 Side - capping of DDSQ - (Ph) 8 (OH) 4 The reaction between 1 and difunc tional dichlorosilanes produced a closed DDSQ structure that will be label in this document as DDSQ - 2(R 1 R 2 ). This reaction, described in Scheme 1 - 3 , results in unavoidable production of cis and trans isomers when R 1 is different than R 2 . Examples of DDSQ - 2(R 1 R 2 ) have been widely reported in current literature . 4,11,13 17 Applications for DDSQ - 2(R 1 R 2 ) have been mainly explored in polymer synthesis because the nanostruc tures become part of the polymer backbone. Polymers containing DDSQ - 2(R 1 R 2 ) are known for the ir low dielectric constants, hydrophobic properties and high degradation temperatures . 18 28 Few works have isolate d cis and trans isomers. S eparation ha s been performed by fractional crystallization (FC) . 4,13,14,17 FC could be lim ited by small diff erences in solubility limits between cis and trans DDSQ - 2(R 1 R 2 ) . As an example, the work from Walczak et al . was able to isolate only a fraction of the trans isomer of DDSQ - 2((CH 3 )(H)). Separation of 5 the individual isomers is a very interesting topic becau se structural differences between cis and trans DDSQ - 2(R 1 R 2 ) is omers can affect other properties like melting temperatures between isomers as reported by Schoen, and Moore et al . ( Table 1 - 1 ) . 15,27 In the study developed by Hoque et al . , melting temperature differences were reported in polymers synthesized with pure cis , pure trans , and a 1:1 cis - to - trans mixture of DDSQ - 2(( isobutyl )(OH)). Separation in this wo rk was performed by FC obtaining cis - to - trans ratios of 2:8 in the precipitated crystals and 8:2 in the solution. This particular study as well as the work developed by Schoen purified the enriched fractions by liquid chromatog raphy . 14,17 Scheme 1 - 3 . Synthesis of closed DDSQ or DDSQ - 2 (R 1 R 2 ) Table 1 - 1 . Differences in melting te mperature for cis and trans i somers for selected systems . 15,27 Compound cis melting temperature °C trans melting temperature °C DDSQ - 2(( p - C 6 H 4 NH 2 )(CH 3 )) 275 311 DDSQ - 2(( p - C 6 H 4 C 2 C 6 H 5 )(CH 3 )) 261 304 6 1.4 Separation techniques used for separation of DDSQ - 2(R 1 R 2 ) systems 1.4.1 Fractional crysta llization FC is a separation technique that is based on differences in the solubility limit between molecules . 29 FC can be performed by dissolving the analyte in a co mmon good solvent and then adding an a mount of a poor solvent to crystallize the analyte with the lower solubility limit. An alternative FC is performed by solubilizing the analyte in the minimum amount possible of good solvent and further crystallization by temperature reduction. 14,29 33 FC is usually considered a cost - effective separation. The main inconvenient s of this technique are long operation times, requirement of high pur ity in the sample, as well as impossib ility for separation in binary mixtures that may co - crystallize at some specific compositions . 30,33 After separation the solvent mixture must be evaporated and , if possible, the solvent can be recycl ed. A diagram describing the crystallization process can be observed in Figure 1 - 2 . Figure 1 - 2 . Fractional crystallization diagram 7 1.4.2 Liquid chromatography Based on t he definition provided by Snyder , l iquid chromatography or L C is a process in which a liquid mobile phase has contact with a stationary phase. 34 The stationary phase is all the non - moving material in the packed - bed. A mixture of analytes has a specific eq uilibrium distribution between the two phases. This equilibr ium determines the migration rate through the packed - bed. Differences in rates between analytes separate the mixture between their analytes as described in Figure 1 - 3 . 34 37 Different classifications for liquid chromatography have been formulated. Modern classification is listed in Table 1 - 2 . For this work, adsorption and partition chromatography will be discussed. Figure 1 - 3 . Separation by liquid chromatography. t 0 column with stationary phase and wet with mobile phase; t 1 mixture injected on the packe d bed; t 2 migration of analytes with different elution rates; t 3 , t 4 and t 5 elution of separated components in different times. 8 Table 1 - 2 . Chromatography classification, adapted from Snyder et al . 2013 . 36,37 Class Example Adsorption chromatography Liquid chromatography (Columnar method) Gas - solid chromatography Partition chromatography Liquid - liquid partition chromatography Paper chromatography Thin lay er chromatography Reverse phase partition chromatography Ion exchange chromatography Cation exchange Anion exchange Inorganic exchange Liquid ion chromatography Exclusion chromatography Gel - permeation Ion exclusion Molecular sieve Electrochromatography Zone electrophoresis Boundary layer method Curtain chromatography Capillary electrophoresis 1.4.2.1 Partition and a dsorption chromatography In partition chromatography the separation process is produced by differences between the solubilities of the components i n the mobile and stationary phases . 38 In this process a stagnant layer of solvent is located in the packed bed surface. Depending the chemical composition in the surface, the stagnant layer has different properties compared with the mobile phase. The separation is given by the partition coefficient for an analyte between the two liquid ph ases . 37,39,40 Usually partition chromatography is subdivid ed between normal phase and reverse phase, these are defined as a high polar stationary phase and low polar mobile phase, and a low pola r stationary phase and high polar mobile phase respectively. For liquid adsorption chromatography the separation is give n by adsorption of an analyte in the stationary phase displacing a previously adsorbed molecule from the mobile 9 phase. Then desorption o f the analyte happens when another molecule of the mobile phase or other analyte displace the already adsorbed analyte . 35,41 Differences in the adsorption isotherms for each analyte in a mixture defi ne the elution rate for each component. Liquid chromatography (LC) has been used for separations in every imaginable fie ld in science requiring a separation step . 42 52 1.4.3 Scale of separation Separation can be performed for analytical or processing purposes . 35 Analytical chromatography is usually developed in small densely packed columns with injections of dil uted samples usually measured in microliters. high performance liquid chromatography (HPLC) was an improvement over regu lar liquid chromatography mainly because the use of high pressures to elute the analytes through densely packed columns, allowing reduct ion in particle sizes and enhancing the resolution of the elution between analytes. Recently ultra - high performance liqu id chromatography (UPLC) devices were developed to withstand larger pressures than HPLC allowing better resolution in analytes with simi lar elution rates. Several detectors have been implemented in these systems including UV, light scattering, mass spectro scopy, IR, flame ionization among others. These detectors allowed elucidation and quantification of the analytes . 35,36 Other analytical separation technique highly used only for elucidation of analyt es is thin layer chromatography. In this, an adsorbent material is located as a layer over a supporting layer. The adsorbent is usually coated with an indicator that emits gree n at 254 nm under UV light. The mixture is spotted close to the bottom of the la yer and the latter one is placed in a container with low amount of solvent that is going to migrate by capillary effect to the top of the layer. Along this migration, the analyte is separated in spots along the layer. These spots covered the indicator 10 look ing as a black spot when evaluated under UV light. If the analy te does not have chromophores, staining solutions allow finding the location of the analyte in the layer . 35 Large scale chromatography separa tion is usually done in simulated moving beds for continuous se paration of binary mixtures . 53 55 For mixtures containing three or more analytes, a second separation cycle or larger number of col umns in the simulated bed may be required. However, large batch es of complex mixtures can be separated in large scale using a single packed column working in batches. This process is referred as preparative liquid chromatography . 36,55 57 1.4.4 Modeling and separation efficiency Modeling of separation in liquid chromatography columns requires acquisition or estimation of the following parameters: adsorption isotherms, mass transfer coefficient, f low rate, and column geometry 57,58 . Other parameters may be required depending the complexity required in the separation. Adsorption isotherms can be linear or the types I, II, or III as seen in Figure 1 - 4 . Further details about modelling are provided in chapter 4 in side this document. Figure 1 - 4 . Peak shape associated to the adsorption isotherm a) type I, b) t ype II, and c) type III, adapted from Fornstedt et al . , 2013 . 57 11 Analysis of separation is usually performed calculating the column efficie ncy. This model assumes that a column with a length (L) is divided in a certain number of theoretical plates (N). Each theoretical plate has the exact same height (H). Value of H can be theoretically calculated by the Van - Deemter equation which is a contri butive model based on column path, axial dispersion, and mass tran sfer terms. Experimentally, the elution profile in the chromatogram can be interpreted as a Gaussian - shaped peak. With the previous analogy, expressions based on the standard deviation and t he variance of a normal distribution plot were proposed to calcula te N ( Equation 1 - 1 ). Similar analogy between two normal distribution was developed to calculate the resolution of the elution (R s ) ( Equation 1 - 2 ) in which R s =1.5 is the threshold value to fu lly separate two analytes. In the listed equations, t r is the rete ntion time for an analyte and W is the peak width at the baseline. This is measured between the tangent lines as seen in Figure 1 - 5 . 36,37 Equation 1 - 1 Equation 1 - 2 Figure 1 - 5 . Retention time and peak width calculation in the baseline of a gaussian - like elution pe ak. 12 1.1 Motivation DDSQ - 2(R 1 R 2 ) molecules, defined previously, have been used as model molecules and building blocks for synthesis of linear hybrid polymers . 17 19,21 25,28,59 62 Resultant materials have enhanced hydrophobicity, reduced dielectric constant, and usually increase in thermal properties without sacrifice in mechanical properties. DDSQ have been applied in minor proportion to other appl ications such as amphiphilic molecules , 63 support for heterogenous catalysts , 64,65 and nanostructure in composites . 66 But many others are not yet explored. Most of the studies reported, use the product as synthesized knowing that this is a mixture of cis and trans isomers . 18,21 25,59,60 Production of these isomers is unavoidable. It has been reported that the use of one of the isomers in synthesis of polymers results in properties with different ma gnitudes respect to the polymer synthesized with the produced mixture or synthesized with the other isomer . 17 So far, the only method used for separation of these mixtures is fractional crystallization, t his tool allowed separation of cis and trans mixtures with dif ferent solubility limits . 4,14,17 However, when the solubility limits are very close, separation is no longer achievable. Besides, to isolate nearly - pure isomers three or more fractional crystall ization cycles are required. 1.2 Research goal This dissertation has the objective of present liquid chromatography as an alternative technique for separation of mixtures containing different variet ies of DDSQ - 2(R 1 R 2 ). This study was developed by shuffling fun ctional groups in DDSQ - 2(R 1 R 2 ) and analyzing the effect of these changes in elution times. This approach allowed production of a variety of unique molecules with potential use for engineering ap plications not reported previously. 13 1.3 Research idea Liquid chro matography is a liquid adsorption separation technique which takes advantage of differences in polarity or other interactions with a stationary phase to elute in different times the analytes in a mixture . 36,55 Knowing that the R 1 and R 2 groups depends of the bridging chlorosilane, polarity can be added to the molecule making feasible the separation of mixtures based on differences in adsorp tion energies. This idea was evaluated for separation of cis a nd trans DDSQ - 2(R 1 R 2 ) isomers containing polar groups, and by separating DDSQ - 2(R 1 R 2 ) mixtures with different number of a particular polar group. Simulation of separation process was studied by development of adsorption isotherms based on frontal analysis by high performance liquid chromatography (HPLC). These results allowed scaling up the process to separate mixtures by preparative chromatography column. Thermal behavior of selected isomers of interest was analyzed after isolation by liquid chromatography or by fractional crystallization and quantification by HPLC. This research highlights the use of liquid chromatography as a chemical engineering unitary operation to achieve high ana lyte purities. 1.4 Dissertation summary in Chapter 2 is described the use of HPLC as a tool for separate and characterize ratios between cis and trans isomers of DDSQ functionalized with polar moieties. Along Chapter 3, HPLC was tested for characterization of DDSQ mixtures with different polarities, synthesis of the mixtures, separ ation and quantification by HPLC, and isolation of compounds including an asymmetric structure by preparative LC. Chapter 4 discussed the simulation for separation of DDSQ mixtures with different number of hydroxyl groups by ASPEN chromatography in order t o scale - up to preparative LC. Finally, the analysis of melting behavior f or cis and trans mixtures 14 was developed to highlight the benefits of separation achieving eutectic compositions. Findings of this study are described in Chapter 5. 15 CHAPTER 2 . HPLC CHARACTERIZATION OF CIS AND TRANS MIXTURES OF DOUBLE - DECKER SHAPED SILSES QUIOXANES Key w ords : Double - decker shaped silsesquioxane (DDSQ), Polarity, cis and trans isomers, separation, HPLC. This chapter was submitted as a research paper to Silicon journal 16 2 HPLC characterization of cis and trans mixtures of double - decker shaped silsesquioxanes 2.1 Introduction Functionalized double - decker shaped silsesquioxanes (DDSQ) ha ve been used as the building block in polymerization to obtain inorganic - organic hybrid poly mers with enhanced dielectric constant, glass transit ion temperature s , melting temperature s , among other properties of engineering interest . 1 7 3 DDSQ are usually synthesiz e d from the condensation reaction between the commercially available DDSQ - (Ph) 8 (OH) 4 (1) and (R 1 )(R 2 ) - dichlorosilanes in the presence of triethylamine as seen in Scheme 2 - 1 . If R 1 is different than R 2 , the condense d DDSQ structures have unavoidable cis and trans isomerism 67 , 71 , 74 81 . These isomers represent as the product, cis and trans , show in Scheme 2 - 1 . Scheme 2 - 1 . Condensation reaction of DDSQ - (Ph) 8 (OH) 4 (1) with 2 molar equivalent of R 1 R 2 SiCl 2 . The resultant product contains cis and trans isomers DDSQ - 2( (R 1 )(R 2 ) ) . These isomers are different in their physical properties such as crystal structure, melting temperature, recrystallization behavior, solubility, etc . 71 It was also reported that a polymer 17 made from all cis i somers have a significantly different melting temperature as c ompared with the same polymer composed of all trans isomer 74 . Quantification of the cis - trans ratio in a mixture is often based on 29 Si - NMR. However, the 29 Si - NMR requires a large amount of functionalized DDSQ and requires a long scan time to reduce the si gnal - to - noise ratio needed for the analysis . To avoid these complications, 2D NMR was used to identify distinctive peaks in the 1 H - NM R between cis and trans isomers in DDSQ samples with aniline and methyl in the R 1 and R 2 positions . 75 The drawback of this procedure is the need to identify characteristic peaks for each moiety, which will be different depending on the R 1 and R 2 used. Fractional crystallization is the most common method used to obtain nearly - pure cis , and nearly - pure trans DDSQ isomers . 71 - 72 , 74 - 75 The separation is based on the solubility differences of cis and trans isome rs in a specific solvent . 76 After the fractional crystallization, a liquid chromatography rectification process is oft en needed to yield a higher isomeric purity . 71 ,7 3 77 Recently, DDSQ - (Ph) 8 (OH) 4 was condensed with methyltrichlorosilane followed by a hydrolysis reaction producing a mixt ure of DDSQ cages including cis and trans isomer s functionalized with hydroxyl groups. It was found t hat cis and trans isomers may be separated by a preparatory silica column and fraction s quantified by HPLC . 74 , 78 , 8 1 In this work, HPLC was reported as an alternative technique to quantify the cis - to - tr ans ratio of DDSQ isomers after the capping reaction and/or separations. The wide availability of HPLC as compared to NMR facility makes this quantificati on method more readily adoptable. Furthermore, the principle of HPLC separation is mainly based on pol arity differences between components. Hence, different pola r groups may be bonded to DDSQ which increases the number of molecules that can be quantified. In the following, several different mixtures of cis and trans compounds are presented and demonstrated a broad applicability of HPLC for the quantification analy sis . 18 2.2 Experimental 2.2.1 Materials All commercially available chemicals were used as received unless o therwise indicated. (C 6 H 5 ) 8 Si 8 O 10 (OH) 4 5,11,14,17 - Tetra(hydro)octaphenyltetracyclo[7.3.3.( 3,7 )] octasi lsesquioxane DDSQ - (Ph) 8 (OH) 4 was purchased from Hybrid Plastics (Hattiesburg, MS). 3 - [Bis(trimethylsilyl)amino]phenylmagnesium chloride [(CH 3 ) 3 Si] 2 NC 6 H 4 M gCl 1M in THF solution ( m - PhA(TMS) 2 - MgCl); Vinyltrichlorosilane (C 2 H 3 )SiCl 3 ; isopropyltrichlorosilane (C 3 H 7 )SiCl 3 ; isobutyltrichlorosilane (C 4 H 9 ) SiCl 3 ; and 3 - cyanopropylmethyldichlorosilane (C 3 H 6 CN)(CH 3 )SiCl 2 were purchased from Gelest. Methyltrichlorosila ne (CH 3 )SiCl 3 , phenyltrichlorosilane (C 6 H 5 )SiCl 3 , deuterated chloroform with 1 vol % tetramethylsilane (CDCl 3 - 1%TMS) were purchased from Sigma - Al drich. Triethylamine (Et 3 N) was purchased from Avantor and distilled over calcium hydride before use. DDSQ brid ged with (methyl)( para - aniline)dichlorosilane, (methyl)( meta - aniline)dichlorosilane, (isobutyl)( meta - a niline)dichlorosilane, and (cyclohexyl)(met a - aniline)dichlorosilane moiety were synthesized for our research group and reported in a previous study . 71 Tet rahydrofuran (THF) was refluxed over sodium/ben zophenone and distilled. Reagent grade dichloromethane (DCM) and n - hexanes were degassed with helium for HPLC experiments. The previously listed solvents were purchased from Sigma. Si - gel P - 60 was obtained fro m Silicycle. 1 H, 13 C, and 29 Si were recorded on 500 MHz NMR spectrometers. 2.2.2 Synthesis of (methyl)(meta - Bis(trimethylsilyl)amino]phenyl)dichlorosilane A 250 mL round bottom flask containing a magnet stirrer under N 2 was sealed with a rubber septum and su bmer ged in an acetone - dry ice bath. 50 mL of THF an d 24.0 mmol (4.0 mL) of isobutyltrichlorosilane were injected respectively to the setup. 20 mmol (20.0 mL) of 19 m - PhA(TMS) 2 - MgCl was added dropwise for 10 minutes. The ice bath was removed upon completion of the addition, and the solution was stirred overnig ht until it became a clear yellowish liquid. Volatiles were distilled under N 2 in an oil bath at 90 °C ; a second distillation was done under vacuum to collect the expected product as a clear pale - yellow liquid . Spectral information are provided in the Appe ndix I . 2.2.3 General synthetic procedure DDSQ - 2 (R 1 R 2 ) was synthesized following a previously reported method . 75 , 80 , 8 1 In a 250 m L flask purged with dry N 2 for 15 minutes, DDSQ - (Ph) 8 (OH) 4 (1) (2 g, 1.8 7 mmol, 1 equiv) was dissolved in THF (60 mL) at room temperature. (R 1 )(R 2 )SiCl 2 (3.74 mmol, 2 equiv) was added to the solution followed by Et 3 N (1.04 mL, 7.48 mmol, 4 equiv) under vigorous stirring. The addi tion of triethylamine took about 5 minutes in to tal , a cl oudy suspension was formed and continue stirred for 4 additional hours. The solution was then filtered through a fine fritted - funnel - filter to remove the solid triethylamine hydrochloride. The soluti on was dried in a rotary evaporator and then pas sed throu gh a silica - gel column using DCM as a solvent. These cleaning step allowed hydrolysis of Si - Cl bond in cages synthesized with trichlorosilanes. The volatiles w ere removed from the resulting solution and further dried at 0.4 mbar and 50 °C for 12 h ours to a fford DDSQ - 2 (R 1 R 2 ) as a white powder. The structures studied in this work are listed in Figure 2 - 1 . NMR spectra for cis and trans isomer mixtures are provide d in the Appendix I . 20 Figure 2 - 1 . DDSQ - 2 (R 1 R 2 ) isomers studied in this work. (2) DDSQ - 2( (para - aniline)(methyl) ) , (3) DDSQ - 2( (meta - aniline)(methyl) ) , (4) DDSQ - 2( (meta - aniline)(isobutyl) ) , (5) DDSQ - 2( (meta - ani line)(cyclohexyl ) ), (6) DDSQ - 2( (hydroxyl)(methyl ) ) , (7) DDSQ - 2( (hydroxyl)(vinyl) ) , (8) DDSQ - 2( (hydroxyl)(isopropyl ) ), (9) DDSQ - 2( (hydroxyl)(isobutyl) ) , (10) DDSQ - 2 ( ( hydroxyl)(phenyl) ) , (11) DDSQ - 2( (cyanopropyl)(methyl ) ). 2.2.4 Analytical methods 2.2.4.1 Preparatory sepa ration of cis and trans isomers Liquid chromatog raphy was performed to separate cis and trans isomers in 2 , 3 , and 11 , and those hydrolyzed structures synthesized with trichlorosilanes in 6 to 10 with a procedure 21 previously described . 76 , 78 , 8 1 A glass preparatory chromatography column, 60 cm in length and 4 cm internal diameter, with 500 m L round top reservoir was packed with 60 grams of Si - gel resulting with packing height of about 40 cm. DCM was then flushed through the packed bed under pressu re generated by a dry N 2 stream. Wetting of the packed bed was complete until no air bubbles, or d ry space was observed . A concentrated solution of DDSQ isomeric mixture in DCM (5 m L , 0.2 g/m L ) was gently inj ected from the top of the wet Si - gel bed and mov ed into the packed bed until no solution was observed above the packed bed. The column was then ge ntly charged with an additional 500 m L of DCM and flushed using the N 2 stream with an average flow rate of 10 m L /min. Fractions of 5 m L were collected at the bottom of the column until the DCM reached the top of the bed. Each fraction was injected in 5 cm TLC plates of Si - gel supported in aluminum. TLC was evaluated with DCM and then analyzed under a 245 nm UV - lam p. Similar fractions were combined and dried for further experiments. The retardation factor ( R f ) in Eq uation 2 - 1 was used as a measure of the separation efficiency after the preparatory LC experiments. ( Equation 2 - 1 ) 2.2.4.2 UV - VIS spectroscopy Wavelength sweep readings were developed for i solated isomers 2 to 5 with DCM as the solvent. Individual isomers were solu bilized forming master batches of 0.2 mg/mL. Then, the solutions were diluted reaching lower concent rations in progressive steps until a value close to 0.02 mg/mL. All readings wer e contrasted against a DCM blank 2.2.4.3 Analysis and quantification of isomers by H PLC All HPLC experiments were performed using an Agilent 1100 HPLC equipped with a UV detector. The columns selected for this work were Supelco LiChrospher ® Si - 60 for adsorption chromatography and ZORBAX ® CN column for partition normal phase chromatography 22 (pH PLC). DCM was used as the mobile phase for the adsorption chromatography . Different ratios of DC M :hexanes ranging from pure dichloromethane to 70% hexane were used as mobile phase for pHPLC. Solvents were degassed using Helium for a minimum of 15 minut es p rior to all HPLC experiments. A flow rate of 1 mL/min at a giving a constant pressure of 32 bars was used . The temperature was set in 25 o C, and the injection volume was 5 detector was emitting at 254 nm. Once the elution was finished, a bla n k sample containing mobile phase was injected and flushed trough the column for verification of the baseline and to confirm complete elution of the previous injection. Standa rd plate theory of chromatography was used for quantitative analysis of separation . 82 Two different methods were used to eval uate the separation resolution based on peak width (W n ) and full width at h alf height of the peak (W n@0.5 ) show in E quation 2 - 2 and Equation 2 - 3 , respectively. The retention time (t r ) is the elution time at the peak maximum. Values of t r , W n , and W n@0.5 w ere calculated using the Agilent chemstation software. The theoretical plat e number (N) or column efficiency based on Gaussian distribution was calculated using Eq uation 2 - 4 . ( Equation 2 - 2 ) ( Equation 2 - 3 ) ( Equation 2 - 4 ) 2.3 Results and Discussion 2.3.1 Separation of cis and trans isomers Two spots were observed by TLC for structures 2 to 5 and for 11 . These two spots indicate the presence of cis and trans isomers ; their corresponding retardation fact ors are listed in Table 23 2 - 1 . The cis and trans isomers were separated by a preparatory LC, and the isolated fraction was evaluated using 29 Si - NMR show in Figure 2 - 2 . The sp ectra obtained for the second fraction was assigned to trans isomers; the spectra obtained for the second fraction was assigned to cis isomers. A previous report from our research g roup describes the peak assignments in detail. 76 Mass balance for the mater ial injected in the column resulted in 75% recovery in two main fractions after elution. This result is comparable with a previous rep ort for separation of DDSQ mixtures by LC. 81 Th e difference in elution times between cis and trans isomers is related with the orientation of the polar moieties. For cis isomers, both polar groups are pointing at the same direction. This configuration slow s the elution rate due to possible stronger ads orption. For trans isomers, polar groups are pointing at the opposite direc tion. Here, only one of the polar groups is attracted to the stationary phase surface increasing the elution rate. A similar relation between positions isomers was reported for smal l molecules. 83 Table 2 - 1 . TL C retardation factors, R f , with dichloromethane as mobile phase for 2 DDSQ - 2( ( p - aniline ) (methyl) ) , 3 DDSQ - 2( ( m - aniline)(methyl) ) , 4 DDSQ - 2( ( m - aniline)(isobutyl ) ), 5 DDSQ - 2( ( m - anilin e)(cyclohexyl) ) , and 11 DDSQ - 2( (cyanopropyl)(methyl) ) . Compounds R f trans R f c is 2 0.28 0.14 3 0.43 0.28 4 0.66 0.44 5 0.77 0.51 11 0.86 0.73 24 Figure 2 - 2 . 29 Si NMR for cis and trans isomers after separ ation of 3 . (a) trans isomer, corresponding to the first spot in TLC; (b) c is isomer, corresponding to the second spot in TLC. 2.3.2 UV absorbance intensity of positional isomers of phenylamine UV detectors are highly employed in organic chemistry due to its lo wer concentration limit compared with other class of detection devices. In this work, HPLC detection and quantification was dev eloped by UV. However, other class of detectors with less definition such as refractive index (RI), or with higher resolution lik e mass spectroscopy can also be used to detect and quantify ratios between cis and trans DDSQ molecules. Cis and trans isomers for molecules 2 to 5 show a maximum absorbance at 254 nm which is associated with the chromophores in the phenyl rings . 8 4 However, at the same concen trations, absorbance values for meta molecules are l ower when compare to the absorbance values for para isomers. It was determined that the slope of absorbance versus concentration for all para isomers is twice as that of meta isomers show in Figure 2 - 3 . This result agrees with previous reports for isomers of positional phenylamine . 82 Ho wever, there are no differences in the slope of th e absorbance 25 versus conce ntration curve between cis and trans isomer. This result implies for the case of separation by HPLC with a UV detector, the UV absorbance for cis and trans isomers can be directly c orrelated to their ratio without concentration cor rection. In addition , the re is no effect on the absorbance values when different R groups were evaluated, as the main contribution to the UV intensity is the eight phenyl groups surrounding the DDSQ, and no t the anilines attached to the Si - O core. Figure 2 - 3 . trans isomers, and filled symbols are for cis - 2( ( p - aniline)(methyl ) ) 2 ; DDSQ - 2( ( m - aniline)(methyl) ) 3 DDSQ - 2( ( m - aniline)(isobutyl ) ) 4; DDSQ - 2( ( m - a niline)(cyclohexyl) ) 5 . The red dashed line represents the linear trend of all the para samples, and the straight blue line is the linear trend for the meta samples. The slope of pa ra was determined to be two times the slope for meta. 2.3.3 Analysis by HPLC of isomeric mixtures and individual isomers. 2.3.3.1 pHPLC of 2 and 3 HPLC evaluation by pHPLC in silica column bonded with cyano (CN) moieties show a single broad peak for the cis - trans mixtu re of 3 while two peaks were clearly observed for the cis - trans mixture of 2 using DCM as the mobile phase . By modifying the mobile phase with the addition of hexanes to DCM, the polarity of the mobile phase was reduced; this change increased the retention 26 time in each peak. The observed effect can be interpreted as a preference o f the molecules to be adsorbed in the polar stagnant layer, causing a reduction in the mass transfer between the stagnant layer and the mobile phase. An additional effect of additi on of hexanes is the enhancement of R s as seen in Figure 2 - 4 . The R S value is based on peak analogy to normal distribution and it is an indicative of the separation efficiency. Less than 1% overlapping between two normal S = 1.5. R S calculated with Eq. 2 was improved from 0.77 when only DCM was flushed as the mobile phase to 1.93 when the mobile phase was modified to a 7:3 volumetric ratio of DCM : h exanes . For mobile phase con taining less than 10% of hexane, evaluation of peak widths (W) was difficul t due to a pronounced overlap. This resulted in a significant error on the calculated value of R s . Alternatively, resultion of elution based on Eq uation 3 may be used for these high ly over lapped peaks. Results of the resolution were tabulated in Table 2 - 2 . Figure 2 - 4 . Effect of dichloromethane/hexane mobile phase ratios in normal phase pHPLC for cis - trans of 2 and 3 . The percentual value indicated represents the vol ume percentage of hexanes in the mobile phase. 27 Table 2 - 2 . Comparison of the resolution obtained using equations 2 and 3 in cis - trans of 2 Hexane % v/v in DCM:Hexane solution 0 2 5 10 20 30 50 70 R s 0.77 1.73 1.63 1.64 1.58 1.68 1.81 1.93 R s0.5 0.36 0.99 0.90 1.03 1.02 1.09 1.04 1.04 2.3.3.2 pHPLC accuracy verification HPLC experiments were performed for quantification of ratios between cis 2 and trans 2 isomers mixed from isolated samples. Five m ixtures with diff erent cis and trans ratios described in Table 2 - 3 w ere diluted in DCM. The solutions were injected to pHPLC and eluted with a mobile phase with 1:1 volumetric ratio of DCM:hexanes to achieve optimal R s . The mixture s with known ratios of isom ers were evaluated with the use of a UV detector attached to the HPLC. Eluted chromatograms observed in Figure 2 - 5 match with the known ratio of isomers in the mixtures. This result indic ates that quantification by HPCL - UV is possible and the standard deviations are lower than 5% as refer in Table 2 - 3 . Figure 2 - 5 . Quantitative analysis of cis - trans mixtures of DDSQ - 2( ( p - aniline)(methyl ) ) by pHPLC. The individual isomers were first isolated and then mixed to a known ratio for comparison against the area under each peak. The wei ghted precent of isolated isomers in mixtures was indicated next to each cu rve. 28 Table 2 - 3 . Comparison between cis and trans percentages of 2 calculated by weighting nearly - pure cis and nearly - pure trans and ca lculated from the area under the peaks in the chromatograms presented in Figure 2 - 5 . The standard deviation was calculated based on the known p ercentage and the area percentage for each isomer. % Weighted cis % Wei ghted trans % Area cis % Area trans Standard deviation cis (%) Standard dev iation trans (%) 27 73 25 75 1.5 1.5 41 59 44 56 2.3 2.3 55 45 58 41 2.1 2.1 69 31 74 26 3.3 3.4 83 17 86 14 1.9 1.9 2.3.4 Separation by adsorption chromatography For adsorption chr omatography, pure DCM was chosen to activate the Si - OH surface in the stati onary phase of the column 83 Different than pHPLC, ad sorption chromatography allowed the separation of meta - aniline isomers. R esolution of elution between cis 2 and trans 2 isomers and between cis 3 an d trans 3 isomers are much higher in adsorption chromatography than pHPLC. Using DCM as the mobile phase, resolution of elution has a value of 6 or higher. In addition, as shown in Table 2 - 4 , the retention time is high. To reduce t r , 2% acetonitrile was added to the mobile phase . This change also decreases R s to an optimal value of 1.5. Like pHPLC, it was observed that in adsorption chromatog raphy trans isomers migrate faster than cis isomers as could be seen from t he retention times presented in Table 2 - 4 . 29 Table 2 - 4 . Retention time (t r ), peak width (W), and plate number (N) after separation by adsorption HPLC with DCM as the m obile phase; Retention time (t rAcN ) after separation by adsorption HPLC with a mobile phase composed by DCM: a cetonitrile in the volumetric ratio 98:2. Compound t r (min) W (min) N t r AcN (min) trans 2 32.1 3.0 1831.8 7.6 cis 2 60.5 5.0 2342.5 11.4 trans 3 43.7 4.4 1578.2 8.5 cis 3 66.4 5.6 2249.4 14.7 trans 4 23.3 2.3 1642.0 - cis 4 47.5 4.7 1634.2 - trans 5 20.0 3.0 711.1 - cis 5 41.9 6.3 707.7 - DDSQ with meta - aniline moiet ies has higher retention times compared with para - aniline. The reason for t his retention time difference is not apparent . However, from crystallographic data, amine groups in DDSQ with meta aniline are pointing to the same direction, and it is possible the amine moiety is more exposed than in the para position 71 , 85 . Th e number o f theoretical plates (N) is a measure of the peak broadening in HPLC. in the column for cis 3 and trans 3 isomers, resulted i n higher values compared with the N value for cis 4 and trans 4 and for cis 5 and trans 5 as seen in Table 2 - 3 . It is remarkable that the non - polar group attached to the D - Si affects the column efficiency. From the data collected in Table 2 - 3 , the column was more efficient resolving 2 and 3 containing methyl group as R 2 ; followed by 4 with isobutyl as R 2 moiety; and lastly, the lower efficiency was attributed to 5 which has the bulkier cyclohexane group in the R 2 position. Comparison of N between cis and trans in evaluated molecules does not show a rec ognizable trend. N for cis of 2 and 3 is larger than N for trans of 2 and 3 . In this case, N of cis was favored by t r and not by W. However, when t r was lower for 4 and 30 5 , W had a bigger effect in the calculation of N ending with slightly better N values f or trans compared against cis . 2.3.5 Effect of R 2 groups in retention times It was observed that the size of organic groups attac h ed to the DDSQ core influences the retention times. For R 1 is meta - aniline, the retention time increases as the size of R 2 decreas es, t r (R 2 =methyl) > t r (R 2 =isobutyl) > t r (R 2 =cyclohexyl). When the polar group was changed from meta - aniline to hydroxyl, the r etention time was also observed to be affected by the size of R 2 group. Similar to meta - anilines, the bulkier group was related to a lower retention time show in Figure 2 - 6 . However, for R 2 was phenyl, the retentio n time was higher as compared to R 2 was isobutyl . This result suggests the isobutyl moiety has a higher steric effect than phenyl, which reduces the overall adsorption between the adsorption site and the hydroxyl group attached to the DDSQ core. Integratio n of DDSQ - 2((OH)(R)) after synthesis by 29 Si - NMR results in a mixture with approximately 50% cis and 50% trans isomers. After sepa ration by HPLC the calculation of the area of each peak in the chromatogram resulted in 50% ± 1.6% of the total area verifying the results obtained previously for DDSQ bonded to para - aniline. Integrations can be observed in the Ap pendix I. 31 Figure 2 - 6 . Retention times for cis and trans DDSQ - 2( (hydroxyl)(R 2 ) ) . (a) R 2 : methyl 6 ; (b) R 2 : vinyl 7 ; (c) R 2 : isopropyl 8 ; (d) R 2 : isobutyl 9 ; and (e) R 2 : phenyl 10 . 2.3.6 Effect of polar groups (R 1 ) in t r DDSQ - (Ph) 8 (OH) 4 was functionalized with different polar groups, para - aniline; hydroxyl; or cyanoproproyl, and their separation characteristics ev aluated using adsorption chromatography. It was found that the retention time for trans 2 ( p - aniline) was five times longer than t he retention times for trans 6 (hydroxyl) and trans 11 (cyanopropyl). The retention time was ten times higher for cis moeity o f p - aniline than hydroxyl ( cis 6 ) and cyanopropyl ( cis 11 ) as seen in Tabl e 2 - 5 . Tabl e 2 - 5 . Retention time for DDSQ functionalized with methyl and different polar g roups. R stands for para - aniline 2 , hydroxyl 6 , and cyanopropyl 11 . DDSQ - 2( (R)(methyl) ) t r trans (min) t r cis (min) R: para - anili ne ( 2 ) 32.19 61.08 R: hydroxyl ( 6 ) 6.31 7.03 R: cyanopropyl ( 11 ) 5.31 6.23 32 The retention time differences between 6 and 2 could be related to the location of the polar groups. For DDSQ - 2( ( para - aniline)(methyl ) ) , the amine moiety is extended away from the core of the DDSQ by a phenyl ring to avoid the steric hindrance from the eight phenyl groups surrounding the core. The ste ric effect weakens the absorption to the stationary phase, which resulted in a significant decrease in retention time. Even it is generally recognized OH has stronger adsorption than NH 2 . For both 2 and 11 , the polar group is not directly attached to the core. However, a difference in the retention time was still observed. This result can be justified based on the hydrogen bonding. For 2 , the oxygen atom from the silanol at the stationary phase and the nitr ogen atom from 2 can act as hydrogen acceptors; an d both have hydrogen donors. For 11 , the cyano group can act only as the acceptor and forms a weaker hydrogen bonding with the sta tionary phase as compared with 2 . 2.4 Conclusions This work confirmed that sepa ration of DDSQ cis/trans isomer mixtures by liquid chromatography and quantification by HPLC is possible as the primary separation technique. Evaluation of isomers by NMR and HPLC showed that the isomer expected after separation was present in percentages better than 90% purity. It was observed that the p osition of the amine group in the aniline affected the UV absorbance values at the same concentration. However, no concentration e ffect was found in absorbances between cis and trans isomers. This enables one to determine the cis/trans ratio in a mixture d irectly from the UV detector signal of an HPLC experiment. Separation by normal phase partition chromatography was distinguished a s a technique for identification and quantification of cis and trans DDSQ - 2( ( p - aniline)(methyl) ) . This 33 chromatography mode doe s not permit separation of the meta isomers. Adsorption chromatography using silica as stationary phase was a better separation te chnique allowing optimal resolution of the elution f or every case studied in this work. The moiety next to the polar group has a steric effect that affects the adsorptio n . Bulkier groups reduced the adsorption energy and in consequence the retention time. However, the elution order is not changed by the adj acent groups it means that for DDSQ functionalized cages, the trans isomer s will always elute first. Polarity in the DDSQ is a crucial factor for allowance the separation process. In addition, the strengt h of hydrogen bonding affected by the sterics of the surrounding groups significantly influnced the elution time. 34 CHAPTER 3 . SEPARATION OF ASYMMETRICALLY CAPPED DOUBLE - DECKER SILSESQUIOXANES MIXTURES Key Words: Double - decker shaped silsesquioxane (DD SQ), Asymmetric, Separation, HPLC This chapter was published as a research paper in Polyhedron journal under the following ci tation: Vogelsang, D. F.; Dannatt, J. E.; Maleczka, R. E.; Lee, A. Separation of Asymmetrically Capped Double - Decker Silsesquioxan es Mixtures. Polyhedron 2018, 155, 189 193. 35 3 Separation of asymmetrically capped double - decker silsesquioxanes mixtu res 3.1 Introduction Functionalized double - decker silsesquioxanes, DDSQs, are building blocks for diverse applications such as polymer chain modifiers . 86 90 Polymer chains modified with DDSQ can obt ain low dielectric constant, enhanced hydrophobicity, and elevated degradation temperature . 88 , 89 , 91 , 92 These materials are derived from the tetrasilanol double - decker si lsesquioxane or D DSQ - (Ph) 8 (OH) 4 ( 1 ). This compound is composed of two cyclic syn - cis - 1,3,5,7 - tetraoltetraphenyltetrasilsesqui oxane bridged by two oxygens and can be functionalized by a side - capping reaction with dichlorosilanes, Scheme 3 - 1 , to produce DDSQs that are the building blocks in many polymer applications . 89 , 92 9 4 Scheme 3 - 1 . Side cappin g of DDSQ - (Ph) 8 (OH) 4 ( 1 ) with a dichlorosilane. Side - capping with a dichlorosilane that has two distinct R - groups results in symmetric cis and trans isomers . 93 , 9 4 The products are symmetric becau se both sides are capped with the same capping ag ent. Separation of these cis and trans isomers using fractional crystal lization (FC) has been reported . 90 , 93 ,9 4 FC is a technique that allows separation based on solubility differences betwee n two or more analytes in a solution . 95 For example, cis and trans - DDSQ side - capped with 4 - (dichloro(methyl)silyl)aniline were separated with high puri ty when the initial ratio between cis and trans i somers was different than 50 % to 50 % cis : trans using THF as the go od solvent and hexane as the poor solvent . 8 DDSQ side - capped with isobutyl trichlor osilane was first separated 36 using FC, and then subsequently polymerized using mostly a single isomer type was studied . 90 Also, separation of mostly trans isomer from the mixture obtained after side - cap ping of DDSQ - (Ph) 8 (OH) 4 with methyldichlorosilane was done using FC . 9 4 However, Schoen et al. and Hoque et al. repor ted that further purification via liquid chr omatography (LC) was needed, which for po lar DDSQ isomers is a separation technique with superior resolution in comparison to FC . 90 , 93 LC is a techniq ue that allows isolation of individual molecul ar compounds even with a partial overlap in retention times . 96 98 For many applications, this technique is faster than FC ; 99 further more, industrially LC is less expensive d espite the differences in initial cost. LC is based mainly on interactions between analytes, stationary phase, and mobile phase . 96 ,1 0 0 Among the many LC operation modes, normal phase (NP) and rev erse phase (RP) have been widely used . 1 01 1 04 Separation for both NP and RP depend on polarity differences between the analytes. However, in N P analytes interact with a polar stationar y phase and a low polar mobile phase . 96 Despite the advantages of LC, there are no reports for the separation of DDSQ compounds using LC except for further pu rificati on after FC. Scheme 3 - 2 . Side capping of DDSQ - (Ph) 8 (OH) 4 ( 1 ) with two chlorosilanes. 37 The use of LC to isolate a DDSQ compound with asymmetric functionalities, ie the AB compound in Scheme 3 - 2 , is pr oposed as an engineering solution to expand the field of application for functionalized DDSQs. In this work DDSQ - (Ph) 8 (OH) 4 was side - capped with two different chlorosilanes which would produce a mixture 3 products, AA , AB , and BB . These DDSQ products have zero, one, and two hydroxyl groups after hydrolysis, and are separable by NP due to their polarity differences. Synthetic procedures, separation technique, and identification of the new AB compounds will be described in th e follo wing sections. 3.2 Experimental 3.2.1 General i nformation All commercially available chemicals were used as received unless otherwise indicated. (C 6 H 5 ) 8 Si 8 O 10 (OH) 4 5,11,14,17 - Tetra(hydro)octaphenyltetracyclo[7.3.3. - 3 3,7 ]octasilsesquioxan e or DDSQ - (Ph) 8 (OH) 4 w as purc hased from Hybrid Plastics. Dimethyldichlorosilane (CH 3 ) 2 SiCl 2 , vinylmethyldichlorosilane (CH 3 )(C 2 H 3 )SiCl 2 , methyldichlorosilane (CH 3 )HSiCl 2 , and 3 - chloropropylmethyldichlorosilane (CH 3 )(C 3 H 6 Cl)SiCl 2 were purchased from Gelest. Methyltrichlorosilane (CH 3 )SiCl 3 was purchased from Sig ma - Aldrich. Triethylamine (Et 3 N) was purchased from Avantor and distilled over calcium hydride before use. Deuterated chloroform with 1vol.% of tetramethylsilane CDCl 3 - TMS and deuterated acetone acetone - D 6 were purchased from Cambridge isotopes laboratori es. Tetrahydrofuran (THF) was refluxed over sodium/benzophenone ketyl and distilled. Reagent grade dichloromethane (DCM) and n - hexanes were degassed with helium for HPLC ex periments. The previously listed solvents were pur chased from Avantor. Si - gel P - 60 w as obtained from Silicycle. 1 H, 13 C, and 29 Si were recorded on 500 MHz NMR spectrometers. 38 3.2.2 General p rocedures 3.2.2.1 General p rocedure A: s ynthesis of s ymmetric DDSQ DDSQ - 2( (CH 3 ) (R) ) was synthesized following the reaction 1 - 1 sh own in Scheme 3 - 1 . In a 250 m L flask purged with dry N 2 for 15 minutes, DDSQ - (Ph) 8 (OH) 4 ( 1 ) (2 g, 1.87 mmol, 1 equiv) was dissolved in THF (60 mL) at room temperature. Dichlorosilane or trichlorosilane (3.74 m mol, 2 equiv.) was added to the DDSQ - (P h) 8 (OH) 4 solution. Et 3 N (1.04 mL, 7.48 mmol, 4 equiv.) was added dropwise to the stirring solution. The addition of triethylamine took 5 minutes in total; a cloudy white suspension was formed and stirred for 4 hours. The s olution was then filtered through a fine fritted - funnel - filter to remove the solid triethylamine hydrochloride. The solution was dried in a rotary evaporator and then passed through a silica - gel column using DCM as the solvent. The volatiles was remov ed fr om the resulting solution and furt her dried at 0.4 mbar and 50 °C for 12 hours to afford DDSQ - 2( (CH 3 )(R) ) as a white powder. It should be noted the reported spectra are of the cis / trans mixtures. Full experimental details and product characterization are i n the Appendix A and I respectivel ly . 3.2.2.2 General p rocedure B: s ynthesis of DDSQ s ymmetric and a symmetric m ixtures The synthesis of DDSQ mixture was done following the Scheme 3 - 2 . In a 250 m L flask purged with dry N 2 for 15 minutes, DDSQ - (Ph) 8 (OH) 4 ( 1 ) (2g, 1.87 mmol) was dissolved in THF (60 m L ) at room temperature. (CH 3 )(R)SiCl 2 (1.87 mmol, 1 equiv.) and (CH 3 )SiCl 3 (1.87 mmol, 0.24 m L , 1 equiv.) were added to the DDSQ - (Ph) 8 (OH) 4 solution and stirred for 5 minutes.Et 3 N (1.04 mL, 7.48 mmol, 4 equiv.) w as added dropwise to the stirring solution. The addition of triethylamin e took 5 minutes in total which created a cloudy white suspension which was stirred continuously for additional four hours. After, 39 the solution was fi ltered through a fine fritted - funn el - filter to remove the solid triethylamine hydrochloride. Volatiles was removed from the resulting solution which produced a white powder. This powder was a mixture of three products as shown in Scheme 3 - 2 . The po wder was dissolved in a minimum amount of DCM and passed through a silica - gel column using DCM as the solvent. The silica column hydrolyzed the chlorosilanes into silanols which were i solated. The three separated products were dried at 0.4 mbar and 50 °C f or 12 hours. Full experimental details and product characterization are in the Appendix A and I respectivelly . 3.2.3 Separation of DDSQ m ixtures by LC A glass preparatory chromatography colu mn with 500 m L round top reservoir (L = 60 cm, D = 4 cm) was packed with Si - gel (60 g, H = 40cm). DCM was flushed through the packed bed under pressure generated by a dry N 2 stream. The packed bed wetting was stopped until no air bubbles, and dry spaces we re observed. A concentrated solution o f DDSQ mixture in DCM (5 m L , 0.2 g /m L ) was gently injected from the top of the wetted Si - gel bed and moved into the packed bed until no solution was above the packed bed. The column was then gently charged with 500 m L of DCM and then flushed using the N 2 s tream with an average flow rate of 10 m L /min. Fractions of 5 m L were collected in the bottom of the column until the DCM reached the top of the bed. Each fraction was injected in 5 cm TLC plates of Si - gel supported in Aluminum. TLC was evaluated with DCM a nd then analyzed under a 245 nm UV - lamp. The graphic description of the TLC separation is shown in the Appendix C . 3.2.4 Characterization of DDSQ materials. Identification of components and their ratios in the mixture, as well as the fractions separated by LC, w ere done using an Agilent 1110 HPL C system. The samples were first dissolved in DCM and 5 m L injected into a Supelcosil column (L = 250 mm, ID = 4 mm) and separated at 1 40 m L /min with DCM as the mobile phas e. The components were detected in a UV detector at 245 nm. Mixtures and separated sam ples were evaluated by NMR taking advantage of the three nuclei of the molecules: 1 H, 13 C, and 29 Si. Dry samples (0.2 g) were dissolved in 0.6 m L of CDCl 3 - 1%TMS and place d in a Variant UNITY Innova 600 at 589 MHz for 1 H, a nd 119.16 MHz for 29 Si. Fractions were also characterized by mass spectroscopy in a Waters Xevo G2 - XS. The ionization was performed by atmospheric pressure chemical ionization (APCI) using acetonitrile as solvent. 3.3 Results and Discussion 3.3.1 Separation by LC To obtain asymmetric DDSQ, a dichlor osilane (R(CH 3 )SiCl 2 , A ) and methyltrichlorosilane (CH 3 SiCl 3 , B ) were added to DDSQ - (Ph) 8 (OH) 4 ( 1 ) as is shown in Scheme 3 - 3 . In addition to the desired AB product, two symmetric b yproducts are obtained ( AA and BB ) . This is because the two reactive sides of DDSQ - (Ph) 8 (OH) 4 are 7 Å apart; thus after capping one side , there is no additional selectivity toward either capping agent A or B . Assuming the capping rate is independent of chl orosilane used, the probability to bond any chlorosilane to one of the sides is ½ and the probability of the same type of chlorosilane capping the second side is ½. This indicates that probability to synthes ize AA or BB is 25% for each, and the probability to synthesize AB is ½ or 50% when equimolar amounts of A and B react with DDSQ - (Ph) 8 (OH) 4 which provides an expected isomer ratio of 25:50:25 AA : AB : BB . However, by using methyltrichlorosilane as the B cappi ng agent, the mixture of products will have zero, one and two chloro moieties. Duri ng the workup, hydrolysis of the resulting chlor ine atom occurs readily leaving zero, one, and two silanols; thus, the final products contain a varying number of silanol gro ups and separation can be 41 achieved based on polar ity. With silica - gel used as the s tationary phase, compound BB, with two silanol groups, migrates slower than compounds AB1 - 4 which in turn migrate slower than compounds AA1 - 4 . This technique allows separati on of the product mixtures as long as R on the A capping agent is non - polar. This i s demonstrated by the retardation factor (R f ) on TLC. A R f of 0.74 for BB , 0.83 for AB1 - 4 and 0.93 for AA1 - 4 was calculated. These R f values were independent of the R moiety used in the synthesis. Scheme 3 - 3 . Proposed synthesis to obtain a mixture of AA, AB, and BB. 3.3.2 HPLC Identification Chromatograms for independently synthesized AA1 - 4 exhibit a single peak at t r = 2.6 min as shown in Fi gure 3 - 1 ( a - d ) . This finding a grees with TLC results, where R does not have an apparent effect on retention times. Two peaks (t r = 18 min, t r = 23 min) were observed for BB as shown in Fi gure 3 - 1 e . These peaks correspond to the trans/cis c onfig urations. This behavior was reported previously in the separation of DDSQ cages side - capped with phenylamines . 93 Chromatograms for reactions following Scheme 3 - 3 have four peaks (t r = 2.6 min, t r = 6.5 min, t r = 18 min, t r = 23 min) as seen in Fi gure 3 - 1 f - i . Products AA1 - 4 and BB are present in the reaction mixtures as confirmed by the retention times for independent ly synthesized compounds (see SI for experimental details). Excitingly, the peak at 6.5 min indicates the presence of asymmetric DDSQs AB1 - 4 . To date, no literature in the separation of systems with similar DDSQs has been reported . Nevertheless, previous w ork in the separation of relatively large or bulky molecules by HPLC 42 using a Si - gel stationary phase were developed for polyols and polymers with a varying number of h ydroxyl groups . 105 , 106 Fi gure 3 - 1 . Chromatograms for products of pure (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R) 2 obtained following Scheme 3 - 1 (a - e) . Chromatograms for mixtures of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R) 2 , (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R)(OH ) , and (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (OH) 2 following synthesis proposed in Scheme 3 - 2 . The absorbance in the region between 15 and 30 minutes is zoomed for reader convenience (f - i) . The second fraction separated by LC corres ponding to (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R)(OH) (j - m) . The peaks observed in HPLC chromatograms were sufficiently resolved , so the relative ratio of the three compounds was evaluated . Since each compound has the same number of chromophores, UV was used for quantifi cation. The results are shown in Table 3 - 1 . As discussed above, we expected a statistical ratio of 25:50:25 for AA : AB : BB ; however, significant variations were observed in the HPLC ratios. Thes e variations were usua lly favoring capping of the chlorosilane with lower steric hindrance. The smallest c hlorosilane in this work was methyldichlorosilane, when this competed with 43 methyltrichlorosilane the highest yield of AA was produced. On the other ha nd, the bulkiest chlor osilane explored in this work was chloropropyl - methyldichlorosilane, when this compe ted against methyltrichlorosilane the lowest yield of AA was obtained . Overall, the results in Table 3 - 1 sho w that as the sterics of the A dichlorosilane increase the amount of AA product decreases. This indicates that the rate of side - capping is significantly affected by the sterics of the capping agent. Table 3 - 1 . The calculated rati o of pro ducts in DDSQ mixtures after separation by HPLC . R AA (%) AB (%) BB (%) H 42.0 38.8 19.2 CH 3 32.3 48.1 19.6 CHCH 2 27.6 50.9 21.5 CH 2 CH 2 CH 2 Cl 15.3 39.7 45.0 To verify this , two capping agents, (CH 3 ) 2 SiCl 2 and (Ph) 2 SiCl 2 , with distinct steric p rofiles were selected and reacted with DDSQ - (Ph) 8 (OH) 4 , Scheme 3 - 4 . These reactions were followed by 1 H - NMR and 29 Si - NMR . It was found that DDSQ - (Ph) 8 (OH) 4 was completely converted to DDSQ - 2 (C H 3 ) 2 before 59 seconds when the first 1 H NMR was collected . However, full capping of DDSQ - (Ph) 8 (OH) 4 with diphenyldichlorosilane took between 36 minutes and 100 minutes to complete; thus, conclusive ly demonstrating that the sterics of the chlorosilanes sig nificantly affect the rate of the capping reaction. 44 Scheme 3 - 4 . Side - capping of DDSQ - (Ph) 8 (OH) 4 with chlorosilanes having moieties with different sterics, 100% conversion time for capping with (CH 3 ) 2 SiCl 2 less than 1 minute, 10 0% conversion time for capping with (C 6 H 5 )SiCl 2 higher than 36 minutes. An unfortunate byproduct of the non - statistical rati os is that in cases where there is a significant difference in sterics between the A and B capping agent, the ratio of desired AB pr oduct suffered. For example, in the first and last entry of Table 3 - 1 , ~10% loss of AB product was observed. To see if this shortcoming could be overcome , we varied the ratios of H(CH 3 )SiCl 2 :( CH 3 )SiCl 3 from 1.5:0.5 to 0.5:1.5. The HPLC ratios and the mass perce ntages after separation by liquid chromatography are summarized in Table 3 - 2 . In every case, HPLC ratios match closely with the isolated masses f rom LC. The data in Table 3 - 2 indicates a preference towards the formation of DDS Q capped in both sides with methyldichlorosilane just as in Table 3 - 1 . However, this is mit igated when an excess of methyltrichlorosilane is used . Exci tingly, close to the expected probability of 25:50:25 is obtained when a ratio of 0.8:1.2 was used . 45 Table 3 - 2 . Mass fraction analysis of products (i n precent) after DDSQ mixtures synthesis with different ratios of methyldichlorosilane and methyltrichlorosilane. *HPLC column is calculated based on HPLC peak analysis and Mass column is calculated based on analytic b alance measurement after separation by preparatory liquid ch romatography Mass. H(CH 3 )SiCl 2 : (CH 3 )SiCl 3 AA1 (%) AB1 (%) BB (%) * HPLC * Mass * HPLC * Mass * HPLC * Mass 1.5 : 0.5 64.1 65.1 26.7 27.9 9.2 7.0 1.2 : 0.8 49.5 42.3 37.1 45.4 13.4 12.3 1 : 1 42.0 45.2 38.8 32.5 19.2 22.2 0.8 : 1.2 36.6 24.3 44.9 50.7 18 .5 25 0.5 : 1.5 6.4 6.0 28.9 30.0 64.7 64.0 3.4 Conclusions When two different chlorosilanes compete for capping sites on DDSQ - (Ph) 8 (OH) 4 , mixtures of AA , AB , and BB are produced. The separation of these products was possible because a varied number of hydr oxyl groups are present in each molecule. HPLC, a valid method to quantify the ratios of products, showed deviations from the expected statistical ratio of 25:50:25 AA : AB : BB . These deviations are due to the rate of side - capping being significantly affected by the sterics of the capping agent. In some cases, this deviation adversely affected the amount of desired AB product produced. This was overcome by changing the ratios of the capping agents such that the more sterically hindered ca pping agent was in exc ess. Overall, our technique provides a route to AB DDSQ systems with an average yield of 30%. This can be further optimized by modifying the ratio between chlorosil anes or modifying the reaction conditions. Isolation of AB DDSQ is a s tep forward to reach b lock copolymers linked by a single asymmetric DDSQ. Present research efforts are focused on finding a direct synthesis of difunctional DDSQ and exploring the proper ties of DDSQ based block copolymers. 46 CHAPTER 4 . PREDICTIVE LIQUID CH ROMATOGRAPHY SEPARATIO N FOR MIXTURES OF FUNCTIONALIZED DOUBLE DECKER SILSESQUIOXANES BASED ON HPLC CHROMATOGRAMS Keywords : Asymmetric double - decker silsesquioxane, Adsorption isotherms, HPLC, Preparatory liquid chromatography, Chromatogram simulation. 47 4 Predictive liquid chr omatography separation for mixtures of functionalized double decker silsesquioxanes based on HPLC chromatograms 4.1 Introduction Functionalized DDSQ structures can be obtained by reacting tetrasilanol octaphenyl double - decker shaped silse squioxane (DDSQ - (Ph) 8 ( OH) 4 ) 1 with functional chlorosilanes . 1 07 1 20 When two equivalent of difunctional dichlorosilanes (R 1 R 2 SiCl 2 ) are used, generally a closed structure (DDSQ - 2(R 1 R 2 )) is obtained . 109 , 111 ,1 20 122 DDSQ - 2(R 1 R 2 ) has been explored for differen t applications such as amphiphilic molecules in Langmuir - Blodgett films , 1 23 ionic liquids , 111 or support for heterogeneous catalyst . 1 24 ,1 25 In addition, there are numerous studies using DDSQ - 2(R 1 R 2 ) in hybrid inorganic - organic polymers as the re sultant hybrid polymers will contain the nanostructure as a part of the chain backbone . 111 , 116 ,1 18 , 119 , 126 1 2 9 These hybrid polymers have shown enhanced hydrophobic prop erties , 1 16 ,1 18 , 1 2 7 low dielectric constants , 1 18 , 1 2 8 , 130 and improved thermo - oxidative stability without sacrifice of their mechanica l performance . 11 7 1 19 , 1 2 6 From chapter 3 , a mixture of DDSQ structures functionalized with zero, one, and two hydroxyl groups was produced . 131 Th e s e mixture s w ere made by side - capping DDSQ - (Ph) 8 (OH) 4 with 1 mol equivalent of (CH 3 )(R)SiCl 2 and 1 mol equivalent of methyl - trichlorosilane (CH 3 )SiCl 3 followed by hydrolysis as depicted in Scheme 4 - 1 . 131 48 Scheme 4 - 1 . Capping of DDSQ - (Ph) 8 (OH) 4 with two different chlorosilanes proposed in chapter 3 The structure wi th a single hydroxyl group (DDSQ - ( CH 3 R 2 )( CH 3 OH)) was recognized as an asymmetric structure . Separ ation of this asymmetric structure from the mixture was achieved with optimal resolution of the elution by the use of adsorption HPLC. 131 In this chapter the as ymmetric structure is denoted with the general formu la (DDSQ - (R 1 R 2 )(R 3 OH)). To further explore engineering applications of asymmetric DDSQ structures, development of large - scale separation methodology to remove the symmetric byproducts is required. Scaling up adsorption HPLC separations to preparative liqui d chromatography (LC) requires the use of a different stationary phase. This change usually represents large void volumes, high injection volume, and high concentrated injections, among other factors whic h may end - up decreasing the resolution of the elutio n . 132 , 1 33 The scope of this work was to simulate the separation of DDSQ - (R 1 R 2 )(R 3 OH) from the mixture by HPLC. In modeling the HPLC column, breakthro ugh curves were performed by HPLC and analysis of th e elution front permit obtention of the a dsorption isotherm parameters for each compound in the mixture. Once the simulation converges with experimental 49 chromatograms, extrapolation of the adsorption para meters can then be used to predict the collection ti me needed in a large - scale preparative ch romatography separation. This prediction allows a proper column configuration to achieve separations with resolutions of the elution large enough to isolate nearly - pure asymmetrical compound. 4.2 Experimental 4.2.1 Materials All commercially available chemicals were used as received unless otherwise indicated. (C 6 H 5 ) 8 Si 8 O 10 (OH) 4 5,11,14,17 - Tetra(hydro)octaphenyltetracyclo[7.3.3. - 3 3,7 ]octasilsesquioxane or DDSQ - (Ph) 8 (OH) 4 w as purchased from Hybrid Plastics. Dimethyldichlorosilane (CH 3 ) 2 SiCl 2 , methyl trichlorosilane (CH 3 )SiCl 3 , vinyl trichlorosilane (CH 2 CH)SiCl 3 , isobutyl trichlorosilane ((CH 3 ) 2 CHCH 2 )SiCl 3 , and deuterated chloroform with 1%vol tetramethylsilane (CDCl 3 - 1%TM S) were purchased from Sigma - Aldrich and from Gelest. Triethylamine (Et 3 N) was purchased from Av antor. Tetrahydrofuran (THF) was dried passing through the alumina adsorbent column. Reagent grade dichloromethane (DCM) was degassed with helium for HPLC exper ime nts. The previously listed solvents were purchased from Sigma. Si - gel P - 60 was obtained from Silicycle. 1 H, 13 C, and 29 Si were recorded on 500 MHz NMR spectrometers. 4.2.2 Methods 4.2.2.1 Synthesis of DDSQ individual products DDSQ - 2((CH 3 ) 2 ) 2 and DDSQ - 2((CH 3 )(OH)) 4a we re synthesized following Scheme 4 - 2 . In a 500 mL fl ask purged with dry N 2 for fifteen minutes, DDSQ - (Ph) 8 (OH) 4 1 (10 g, 9.35 mmol, 1 equiv.) was dissolved in THF (200 mL) at room temperature. Dimethyldichlorosilane or methyl trichlorosilane (18.7 mmol, 2 equiv.) was added to the DDSQ - (Ph) 8 (OH) 4 1 solution. Et 3 N (37.4 mmol, 4 equiv.) was added dropwise for a period o f about 5 minutes to the stirring 50 solution; a cloudy white suspension was formed and continue to stir for additional four hours at room temperature. The solution was then filtered through a fine fritted - funnel - filter to remove the solid triethylamine hydro chloride. The volatiles were removed using a rotary evaporator. The sample was then solubilized in a m inimum volume of DCM and then passed through a short silica - gel plug using DCM as a solvent t o clean the sample and hydrolyze when methyl trichlorosilane was used. The volatiles were removed from the resulting solution and further dried at 0.4 mbar and 50 °C for 12 hours to afford 2 or 4a as white powders. Spectral information can be found in the Appendix I . Scheme 4 - 2 . Synthesis of functionalized DDSQ - 2(R 1 )(R 2 ). For compound 2 : R 1 and R 2 are CH 3 ; for compo und 4a : R 1 is CH 3 , and R 2 is Cl which is hydrolyzed after the reaction to form hydroxyl (OH) 4.2.2.2 Synthesis of DDSQ mixtures The synthesis of DDSQ mixtures with zero, one, and two hydroxyl groups was performed following Sche me 4 - 3 using a procedure previously described . 114 Synthesis of a mixture containing 2 , 3 a , and 4a (or mixture A ) is described as an example. To a 500 mL flask purged with dry N 2 for 15 minutes, 1 (1 0 g, 9.34 mmol) was added and dissolved in THF (200 mL) at room temperature. Dimethyldichlorosilane (9.34 mmol, 1.14 mL, 1 equiv.) and methyl tric hlorosilane (9.34 mmol, 1.10 mL, 1 equiv.) were added to the solution c ontaining 1 and stirred for five minute s. Et 3 N (37.4 mmol, 5.22 mL, 4 equiv.) was added in 51 1 mL installments to the stirring solution. A cloudy white suspension was formed and stirred c ontinuously for additional four hours. The solution was filtered throug h a fine fritted - funnel - filter to remov e the solid triethylamine hydrochloride. Volatiles were removed from the resulting solution which produced a white powder. The powder was dissolv ed in a minimum amount of DCM and hydrolyzed after addition of 50 mL of water and one hour of stirring. The or ganic phase was isolated, washed with brine, and then dried with magnesium sulfate. The volatiles were evaporated ending with the mixture A as a w hite powder. Nearly - pure 3a was then isolated from the mixture A using LC. The same procedure was repeated rep lacing methyl trichlorosilane for vinyl trichlorosilane to produce the mixture B ( 2 , 3b , and 4b ), and for isobutyl trichlorosilane to produce the mixture C ( 2 , 3c , and 4c ). Spectral information for 3a , 3b , and 3c cages can be seen in the Appendix I . Sche me 4 - 3 . Synthesis of mixtures containing DDSQ - 2 (CH 3 ) 2 2 , DDSQ - (CH 3 ) 2 (R)(OH) 3 , and DDSQ - 2((R)(OH)) 4 , where R is methyl (CH 3 ), vinyl (CHCH 2 ), or isobutyl (CH 2 CH(CH 3 ) 2 ). 52 4.2.2.3 Breakthrough curves Breakthrough curves were generated in an Agilent 1100 HPLC system with a dual pump using a Lichrospher - Si60 column with an inner radius (r) of 0.23 cm and a length ( H b ) of 25 cm. Analytes were detected with an UV - detector at 245 nm. Each compound was first solubilized in DCM with varying concentrations listed in Table 4 - 1 . The solution was pumped through pump A at a flow r ate (F v ) of 1 mL/min for 10 min. ; pure DCM was then pumped from pump B for 40 min with the same F v for complete el ution. Table 4 - 1 . Solutions prepared for obtention of breakthrough curves. Concentrations of 4a are for a 1:1 mixture of cis and trans isomers. Compound Concentration (C) (g/L) 1 2 3 4 5 6 7 8 2 0.10 0.18 0 .30 0.40 1.06 1.60 2.20 - 3a 0.09 0.19 0.33 0.57 0.98 1.83 2.82 6.25 4a 0.14 0.25 0.50 1.20 2.40 3.90 7.00 - Adsorption isotherms were calcula ted using Equation 4 - 1 , 134 where the hold - up time (t 0 ) in t he column used was found experimentally as 2.0 min. The extra column time (t ext ) was calculated based on the chan nels from the HPLC redirection valve to the beginning of the column, and from the en d of the column to the detector. For our HPLC, the value of t ext was found to be 0.7 min. experimentally . The shock time (t s ) was picked from the chromatogram as the half - t ime of the front located between the baseline and the beginning of the plateau after the detection of DDSQ compounds. Equation 4 - 1 53 The e xperimental adsorption isotherm was fitted to the linear isotherm model, Equation 4 - 2 . Equation 4 - 2 w here IP1 and IP2 are isotherm parameters and IP2 wa s fitted to be zero for all cases here. 4.2.3 HPLC and preparative LC chromatograms HPLC experiments were configured using DCM as the mobile phase. The solvent was deg assed with H elium before pumping to the col umn. The flow rate was set to be 1 mL/mi n for all samples, which gave a constant pressure of 34 bars, the temperature was set as 25 ° C, the an d the UV detector was set a t 254 nm. The number of theoretical plates (N) or column efficiency was calculated with the tangent line method in Equation 4 - 3 . For HPLC, retention time (t r ) and peak width at the baseline ( W) w ere calculated using the Agilent Chemstation software . 135 For preparative LC t r and W were calculated manually from the plots. Resolution of the elution (R s ) was obtained using Equation 4 - 4 . Equation 4 - 3 Equation 4 - 4 Preparative LC was performed to scale - up the results obtained from HPLC experiments. Two different columns were prepared in the following general meth odo logy. DCM was added to dry silica bed inside a glass column. The formed slurry was shaken to remov e most air bubbles. Then, the wet bed was flushed with DCM under pressure generated by a dry N 2 stream. The flushing process was stopped until the flow was co nstant in the column outlet, meaning that air was mostly removed from the column. Void fraction wa s calculated based on the volume required to elute an injection of toluene through the column at the average flow rate. A 54 concentrated solution of DDSQ mix tur e dissolved in DCM was gently injected in the top of the wet Si - gel bed. The injection was flushed until no solution was observed above the packed bed. The column was then gently charged with DCM and the column pressurized by the N 2 stream to elute the ana lytes. Fractions were collected in the bottom of the column until remaining DCM reached the top of the bed. The c oncentration in each fraction was obtained by evaporation of the solvent and weighing the solid mass or by HPLC peak areas compared against a s tandard reference. Elution times in the preparative separation for mixtures A and B were calculated based on the relation between the volume of each fraction and the average flow rate. For the separ ation of mixture C , the flow rate for every collected f rac tion was recorded to analyze the effect of flow rate gradients in the simulation results. Dimensions of the columns and parameters for the separations performed can be observed in Table 4 - 2 . Table 4 - 2 . Preparative column dimensions and operational parameters Column Preparative 1 Preparative 2 Preparat ive 3 Mixture A ( 2 , 3a , 4a ) B ( 2 , 3b , 4b ) C ( 2 , 3c , 4c ) Composition ( 2 : 3 : 4 ) 32:49:19 42:47:11 40:45:14 Height (Hb) cm 34 43 43 Radius (r) cm 1.7 3.4 3.4 Void fraction 0.7 0.6 0.6 Flow rate (mL/min) 25 54 19 to 87 for 10.4 min 87 to 103 for 13.0 min 54 to finish Mixture weight (g) 2 4 4 Volume injected (mL) 15 46 46 55 4.2.4 Chromatogram simulation ASPEN chromatography V1 0 w as used to simulate the separation process in HPLC as well as in preparative LC. A batch process was drawn starting with an inlet controlling the mobile phase and the injection. This stream was connected to a packed column, and th e column was further co nne cted to an outlet stream. To start the modeling , trace - liquid was assumed for the column operation because the density of the liquid along the column is considered to be a constant. Other two assumptions in the mass balance are: ( 1) no changes in flow r ate (F v ); and (2) no additional reaction occur during the separation process. After these assumptions, the resultant mass balance can be observed in Equation 4 - 5. Equat ion 4 - 5 t is the total void fraction which is a function i and the p which is neglected here. C i is the concentration of i molecule at a given time and position, and J i is the mass transfer flux. E zi stands for the axial dispersion coefficient. E zi was calculated based on the theoretical column number of plates (N) described in Equation 4 - 6 , where H b is the packed bed height. Eq uat ion 4 - 6 Linear lumped resistance assumption was made to obtain an expression for J i ( Equation 4 - 7 ). Under this assumption, the mass transfer driving force for component i is a linear function of the liquid phase concentration . 136 In Equation 4 - 7, q i * represents a reference value for component i adsorbed. The mass transfer coefficient MTC was estimated by ASPEN assuming 0.005 cm 2 /min as standard value for molecular diffusivity in liquids. 56 Equation 4 - 7 The system was divided into 500 nodes and further solved by finite differences using the quadratic upwind differencing scheme. 4.3 Results and Discussion 4.3.1 Breakthrough curves and calculated adsorpt ion isotherms Breakthrough curves were obtai ned for 2 , 3a , and 4a as observed in Figure 4 - 1 . For 2 , the shock times were the same for all concentrations studied. This occurs as 2 is a molecule without silanol group s and its adsorption to the stationary phase is limited. Breakthrough curves for 3a showed a slight decrease in t s when the concentrations were increased. The highly polar 4a exhibited a larger reduction in t s as the concentration increases. Two fronts wer e observed for 4a corresponding to trans - 4a and cis - 4a . t s was identi fied for each one of the fronts in 4a breakthrough curves. This observation enables the calculation for the adsorption behavior of each isomer and the prediction of cis and trans separati on in the chromatogram simulation. The plateau region for concentrati ons higher than 10 g/L was above the detection limit. As a result , there was a breakdown of linearity between absorbance and concentration. This implies possible errors in the identificat ion of t s in the half concentration of the front. In addition, sample s at a concentration above 10 g/L had a tendency to form crystals within the HPLC pump pistons and block the flow in the column. 57 Figure 4 - 1 . Breakthrough curves for a) 2 in low concentrations, b) 2 in high concentrations, c) 3a in low concentrations, d) 3a in high concentrations, e) 4a in low concentrations, and f) 4a in high concentrations. t s was calculated from the half - concentration in the curve front. Experimental re sults for adsorption equilibrium followed a linear trend. Therefore, the isotherms were fitted with high accuracy into a linear adsorption isotherm model. Adsorption equilibrium data calculated using Equation 4 - 1 and fit ting of the data in the linear Equat ion 4 - 2 is observed in Figure 4 - 2 . The experimental IP1 parameters calculated for HPLC separation for 58 the components of mixture A were tabulated and sh own in Table 4 - 3 . Some of the effects for high concentrations in the chromatogram such as peak tailing or maximum adsorption capacity were not studied as DDSQ breakthrough curves cannot be obtained at high concentrations. Figure 4 - 2 . Experimental adsorption isotherms and linear fitting for: 2 black squares , 3a green circles trans - 4a blue triangles , and cis - 4a red diamonds . a) full experimental points; b) low concentration points. Color graph can be obtained in the digital version of this document. 4.3.2 Simulation results and parameter fitting Based on th e linear behavior of the adsorption isotherms for the components in mixture A . It was hypothesized that isotherms for mixtures B and C could also be linear. The isotherm parameter IP1 for each component in the mixture A was plotted against the retention ti me correspondent to each peak seen in Figure 4 - 3 . 59 Figure 4 - 3 . Linear behavior between the retention times (t r ) in HPLC and the linear isotherm parameter IP1 obtaine d from FA for mixture A . The relation between IP1 and t r for mixture A was linear. For instance, Equation 4 - 8 was calculated making a linear fitting. Equation 4 - 8 was used for calculation of the HPLC IP1 parameters (IP1 HPLC ) for the components in the mixtu re B and in the mixture C . These parameters can be observed in Table 3 . This approximation was made taking into account the similarity of components in the mixtur es, as well as the fact that differences in retention times between 4a , 4b , and 4c are mainly affected by the organic group adjacent to the hydroxyl group . 1 37 The IP1 HPLC values obtained were feed into the model, and the HPLC chromatograms for mixture A , mi xture B and for mixture C were simulated as seen in Figure 4 - 4 . Equation 4 - 8 60 Table 4 - 3 . Linear adsorption isotherm parameter (IP1) values obtained experimentally and calculated values. Compound IP1 Experimental IP1 for HPLC column calc ulated from Equation 8 IP1 for preparative column calculated from Equation 9 2 0.3753 0.3767 0.0645 3a 0.7940 0.7968 1.6288 trans - 4a 1.9850 1.9816 6.0937 cis - 4a 2.1822 2.1844 6.8329 3b - 0.6899 1.2385 trans - 4b - 1.3026 3.5354 cis - 4b - 1.5248 4.368 4 3c - 0.5986 0.8963 trans - 4c - 0.9373 2.1660 cis - 4c - 1.1351 2.9077 Figure 4 - 4 . Curves in red represent the chromatograms obtained by HPLC. Curves in dotted blue lines represent the simulated chromat ograms for a) mixture A , b) mixture B , c) mixture C . 4.3.3 Extrapolation of HPLC parameters to preparative column - porous spherical silica particles densely packed in the column with void fraction ar ound 0.1. On the other hand, the preparative column has silica as a stationary phase with particle size aver aging 60 61 void fraction can be higher than 0.5. Consequently, HPLC and preparative LC have different adsorption isotherm parameters. A glass column packed with Silicycle P - 60 silica was prepared for separation of 2 g of the mixture A t o obtain the preparatory LC parameters. Fractions of 9.5 mL were collected and eve ry fraction dried and weighed to elaborate the experimental chromatogram observed in Figure 4 - 5 a . Figure 4 - 5 . Curves in red represent the chromatograms obtained by preparative liquid chromatography. Curves in dotted blue lines represent the simulated chromatograms for a) mixture A , b) mixture B , c) mixture C . Separation for mixture A was simulated by guessing the IP1values in the model. The graph of IP1 against t r in the preparative column followed a linear behavior like the one observed in the HPLC column. IP1 parameters from the HPLC column were correlated with IP1 parameters from the preparative column. Obtainin g the linear correlation observed in Figure 4 - 6 as well as in Equation 4 - 9 . 62 Figure 4 - 6 . The relation between IP1 obtained from HPLC (IP1 HPLC ) and IP1 obtained from Preparative column (IP1 prep ) Equation 4 - 9 . A preparative column was packed for separation of 5 g of mixtu re B , and 5 g of mixture C to test the predictive model for columns with Silicycle P - 60 silica. Simulation for separation of mixture B resulted in an accurat e prediction for collection times for every compound. However, retention times were just acceptably predicted. The experimental preparative chromatogram and the simulated chromatogram were compared and shown in Figure 4 - 5 b . Analysis of each fraction was performed by HPLC and can be observed in the Appendix G . Separation of mixture C was performed as followed: first, the flow rate was increased from 19 mL/min to 87 mL/min in a lapse of 10.4 mi n.; from 87 mL/min to 10 3 mL/min for 13 min.; and then suddenly dropped to 54 mL/min for the remaining experiment. Experimental ramps were repos ted in the Appendix G . The separation using flow rate gradients was simulated and the results obtained are in ac ceptable fitting with th e 63 experiment as seen in Figure 4 - 5 c . In average, 85% of the initial weight was recovered after the separation process. Prediction of separation between cis and trans molecules of 4 indicates that is possible to col lect fractions enriched with the cis isomers and fractio ns enriched with the trans isomers. The predicted results agree with the experimental results showing that effectively cis and trans isomers of the DDSQ with two hydroxyl group s can be partially separ ated using DCM as the mobile phase. Previously separatio n of trans and cis DDSQ cages capped with isobutyl trichlorosilane was achieved in a silica column with toluene as mobile phase . 17 Enriched fract ions of cis - 4b trans - 4b cis - 4c and trans - 4c can be observed in the HPLC analysis for preparative separations presented in the Appendix G . 4.3.4 Efficiency for preparative columns Analysis of the column efficiency by Equation 4 - 3 was used to estimate the dispersi on coefficient by ASPEN chromatography. Table 4 - 4 contains the calculated column efficiencies for preparative - 1, preparative - 2, and HPLC. For compound 2, the difference in N values between p rep arative - 1 and prepara tive - 2 columns is mainly attributed to the larger diameter and length of preparative - 2, larger column diameters generally reduce the injection height, decreasing the initial dispersion in the column. On the other hand, longer column s h ave a higher number o f theoretical plates. The calculated efficiency for separation of 4b was half of 4a and 4c in HPLC due to overlapping of elution peaks. As seen in Table 4 - 4 , the values of N for the preparat ive column are one order of magnitude smaller than the N for HPLC. Calculation of N for preparative columns has the same trend as the N calculated by HPLC. 64 Table 4 - 4 . Column efficiency (N) calculated from Eq uation 3 for HPLC and pr eparative columns. * Value calculated as an individual component. Compound N preparative - 1 N preparative - 2 N HPLC 2 261 603 7178 3a 191 - 2235 3b - 430 2359 3c - - 2635 trans - 4a 105* - 1936 cis - 4a - 1694 trans - 4b - 212 697 cis - 4b - 217 853 trans - 4c - - 1427 cis - 4c - - 1117 In Table 4 - 5 , values of resolution of the elution (R s ) calculated with Equation 4 - 4 are tabulated. R s between 2 and 3 , and between 3 and 4 were higher than the optimal value, R s - optima l =1.5, in HPLC as well as in preparative LC. These R s values indicate that every analyte was eluted with high purity. In addition, further improvement may be made to reduce the separation time. Calculations of R s for HPLC between ci s and trans isomers of 4 resulted in values larger or close to the optimal value for 4a , and for 4c . However, for the mixture B separation of trans - 4b and cis - 4b has R s value enough to recognize the components but very low to achieve an optimal resolution between the peaks. This behavior indicates that polarity difference between cis and trans isomers of 4b is lower than the difference of polarity for isomers of 4a and 4c . For preparative LC separations, R s between cis and trans isomer s of 4 was not possibl e to measure because bot h isomers were overlapped. Different solvents and column lengths may be explored to achieve R s values close to 1.5 between trans - 4 - and cis - 4 peaks. 65 Table 4 - 5 . Resolution of the elution between analytes in each mixture after separation by HPLC and preparative LC. R s between compounds R s by HPLC R s by Preparative LC 2 and 3a 7.1 2.1 2 and 3b 5.5 3.1 2 and 3c 4.3 1.8 3a and trans - 4a 7.4 2.0 3b and trans - 4b 3.9 2.9 3c and trans - 4c 3.4 1.7 trans - 4a and cis - 4 1.5 - trans - 4b and cis - 4b 0.8 0.2 trans - 4c and cis - 4c 1.3 - 4.4 Conclusion Separation by LC was performed for a mixture of double - decker shaped silsesquioxanes functionalized with zero, one, and two hydroxyl groups. The presence of the polar group allowed different retention times for the mixtures evaluated included the trans and the cis isomers for the nanostructure functionalized with two hydroxyl groups. It was determined that adsorption of the non - polar compound was neglectable b ecause it migrates throu gh the column without being retained. Linear adsorption isotherms were iden tified for the structures dissolved in DCM. Highly concentrated solutions were not studied due to problems with the pump pistons and proximity to the detecto r limit. However, evalua tion of larger concentrations may allow visualization of Langmuir type of i sotherms. Nevertheless, linearity is an advantage for scale up the separation process into a preparative column working in batch operations. HPLC was used to predict linear adsorpti on isotherm parameters that were further extrapolated to find adsorption is otherm 66 parameters for commercially available silicas. These parameters allowed the prediction of collection intervals in large - scale columns. Separations per formed experimentally wi th the simulated parameters are accurate in the baseline of the chromatogra m. But the peak shapes may require an other type of isotherms predictive of tailing or perhaps competition. Other types of separation using this commercial st ationary phase can be si mulated with the adsorption parameters found in this work. In summary, this work proposes a simple predictive methodology for scale - up HPLC separations of side - capped octaphenyl double - decker shaped silsesquioxanes by the use of com mercially available sili ca . 67 CHAPTER 5 . PHASE BEHAVIOR OF CIS - TRANS MIXTURES OF DOUBLE - DECKER SHAPED SILSESQUIOXANES FOR PROCESSABILITY ENHANCEMENT Keywords : Double - decker shaped silsesquioxanes, melting temperature suppression, binary cis and trans e utectic composition, pha se diagram . 68 5 Phase behavior of cis - trans mixtures of double - decker shaped silsesquioxanes for processability enhancement 5.1 Introduction Functionalized double - decker shaped silsesquioxane (DDSQ - 2(R 1 R 2 )), and corner capped cubic shap ed silsesquioxanes (POSS - R 1 ) ( Figure 5 - 1 ) are cage - like silsesquioxanes with a dimensionally well - defined inorganic core, inert organic groups around the core which provides compatibility with the surround organic matter of intere st, and exact sp ecified reactive organic sites. 1 38 , 139 These structures have become model compounds to investigate effects of nanostructured inorganic additives on polymer properties. 140 - 144 Applications of both DDSQ - 2(R 1 R 2 ) and POSS - R 1 nanostructures have been ex plored extensive ly. 1 38 , 140 - 142 , 145 , 146 - 167 When used in organic polymers, the hybrid characteristics provide enhanced oxidation temperature, improved hydrophobicity, and low dielectric constant. 1 47 - 1 51 ,1 56 , 166 , 168 - 173 Recent research extended the use of DDSQ - 2(R 1 R 2 ) and P O SS - R 1 as support s to reduce the amount of packed bed required in heterogenous catalysts. 142 ,1 47 , 174 - 177 Other applications for POSS - R 1 include improved ionic liquid performance and its thermal stability, 171 , 178 - 180 microstructure modifier to improve mecha nical per formance of meta llic alloys, 162 superhydrophobicity in coatings, 164 , 181 - 183 monomer for increased thermal and mechanical performance of thermosetting polymers, 184 - 187 and use in pharmaceutical applications. 182 , 188 - 191 69 Figure 5 - 1 . Structure of cis/trans - DDSQ - 2(R 1 R 2 ) and POSS - R 1 where R are inert organic moieties and R 1 and R 2 are active functional groups. Incorporation of POSS - R 1 or DDSQ - 2(R 1 R 2 ) without the use of sol vents is an e nviro nmentally friendly method to integrate these cage - like silsesquioxanes into organic polymers and has been achieved by melting the nanostructures. 165 , 169 , 192 , 193 In our prior work on POSS - (phenylethynylphthalimide) where the surrounding R moieties were phe nyl o r isobutyl, it was observed that hepta - phenyl POSS - (phenylethynylphthalimide) did not melt; however, the isobutyl system could be melt - processed with organic oligomeric phenylethynylphthalimide to form thermosets. 169 Interestingly, it was found that oc ta - ph enyl DDSQ - 2((phenylethynylphthalimide)(methyl)), which is structurally similar to hepta - phenyl POSS - (phenylethynylphthalimide), resulted in a material that exhibited melt characteristics. 169 , 194 It was hypothesized that melting of DD SQ - 2(R 1 R 2 ) was cause d by higher entropy as compared to POSS - R 1 because the as - synthesized DDSQ - 2(R 1 R 2 ) products contain cis and trans isomers about the inorganic core as shown in Scheme 5 - 1 . 144 , 145 , 192 , 195 - 197 , 198 - 201 70 Scheme 5 - 1 . Condensation of 1 with two equivalents of organo - dichlorosilanes Separation of cis and trans DDSQ - 2(R 1 R 2 ) isomers using fractional crystallization (FC), and/or liquid chroma tography (LC) has been achieved, 145 , 199 and it has been demonstra ted that polymers synthesized from mostly cis isomers exhibit different thermal characteristics than the same polymer made with mostly trans isomers. 145 In this work we investigate the effect of bulkiness of an aryl gro up in the R 1 position in DDSQ - 2((methyl )(R 1 )) on the melting temperature of the cis or trans isomers. In addition, this work focused on a more systematic understanding of the melting behavior of DDSQ compounds with varying cis t o trans ratios. Results from the differential scanning calorimetry (DSC) were used to construct the upper portion of the cis - trans binary phase diagram. The resulting binary phase diagram can be used to tailor a specified cis to trans ratio for optimizing the condition needed for mel t mixing with organic polymers. 71 5.2 Exper imental 5.2.1 Materials All commercially available chemicals were used as received unless otherwise indicated. (C 6 H 5 ) 8 Si 8 O 10 (OH) 4 5,11,14,17 - tetra(hydro)octaphenyltetracyclo[7.3.3. - 3 3,7 ]octasilse squioxane (DDSQ - (Ph) 8 (OH) 4 ) 1 was purchased from Hybrid Plastics. Methyltrichlorosilane, 4 - [bis(trimethylsilyl)amino]phenyl( bromo )magnesium, 4 - bromoiodobenzene , phenylacetylene, phenylmethyldichlorosilane, Pd(PPh 3 ) 2 Cl 2 , CuI, activated magnesium turnings, CDCl 3 with 1% TMS, dichloromethane (DCM), hexanes, ethyl acetate, a nd THF were all purchased from commercial sources and used directly unless specified . THF was refluxed over sodium/benzophenone ketyl and distilled. 5.2.2 Synthetic procedures 5.2.2.1 Synthesis of dichl oro(m ethyl)(4 - (phenylamine(bis(trimethylsilyl))))silane Scheme 5 - 2 . Synthesis of dichloro(methyl)(4 - (phenylamine(bis(trimethylsilyl))))silane Under an N 2 atmosphere, a solut ion o f 0.5 M 4 - [bis(trimethylsilyl)amino]phenyl(bromo) - magnesium in THF (30 mL, 15 mmol) was injected dropwise into a flask containing a solution of freshly distilled THF (10 mL) and 18 mmol of methyltrichlorosilane (1.2 equiv, 2.11 mL) that was cooled by an ic e bath ( Scheme 5 - 2 ) . This mixture wa s allowed to warm to room - temperature and then stir for 24 h after which a clear yellow - pale solution was observed . At the end of the reaction, the excess THF and MeSiCl 3 were removed by di stillation under N 2 in an oil bath at 90 72 °C. The product was then d istilled at 120 ° C under 0.1 mmHg vacuum produc ing a clear yellow liquid (3.95 g, 11.26 mmol, 75 % yield) . The spectra match previously reported data. 195 , 199 1 H NMR (500 MHz, CDCl 3 (m, 2H), 7.01 (m, 2H), 1.02 (s, 3H, CH 3 ), 0.10 (s, 18H, TMS). 29 Si NMR (99 MHz, CDCl 3 5.2.2.2 S ynthesis of 1 - bromo - 4 - (phenylethynyl)benzene Scheme 5 - 3 . S ynthesis of 1 - bromo - 4 - (phenylethynyl)benzene This procedure was adapted from previous literature and described in Scheme 5 - 3 . 198 , 201 To a 500 mL round bottom flask was added 4 - bromoiodobenzene (100 g, 353 mo l, 1 equiv.), Pd(PPh 3 ) 2 Cl 2 (0.2481 g, 0.353 mmol, 0.1 mol %), CuI (0.0673 g, 0.353 mmol, 0.1 mol %), and a stir bar. The flask was sealed with a rubber septum and flushed with nitrogen then 300 mL freshly distilled THF was added . Triethylamine (54 .3 mL, 0. 389 mol, 1.1 equiv.) was distilled over CaH 2 and added to the reaction mixture. Finally, phenylacetylene (42.7 mL, 0.389 mol, 1.1 equiv.) was added dropwise to the reaction mixture. The reaction solution eventually turned a brown color with a whit e preci pi tate. The white precipitate is likely Et 3 NI. The reaction was allowed to stir for 12 hours at room temperature. At the end of the reaction, the solvent was removed , and the resulting solid was extracted with DCM/H 2 O. The solid from the organic lay er was th en purified by silica column chromatography 73 (hexanes) to afford 87.25 g of a white flaky solid (96% yield) with NMR matching previous literature. 202 1 H NMR (500 MHz, CDCl 3 7.53 (m, 2 H), 7.52 7.48 (m, 2 H), 7.44 7.39 (m, 2 H), 7.39 7.35 (m, 3 H) . 13 C NMR (126 MHz, CDCl 3 128.53, 128.42, 122.92, 122.49, 122.25, 90.53, 88.34. 5.2.2.3 Synthesis of dichloro(methyl)(4 - (phenylethynyl)phenyl)silane Scheme 5 - 4 . Synthesis of dichloro(methyl)(4 - (phenylethynyl)phenyl)silane. Following Scheme 5 - 4 , to a 250 mL round bottom flask were added Mg 0 turnings (2.08 g, 0.085 mol, 1.1 equiv.) and a stir bar. T he Mg 0 turnings were stirred under vacuum for 2 hours after which the flask was put under an N 2 atmosphere. 1 - bromo - 4 - (phenylethynyl)benzene (20 g, 0.077 mol, 1 equiv.) was dissolved in 80 mL freshly distilled THF and injected into the flask containing the Mg 0 turnings. This mixture was allowed to stir for 12 h after which a green solution was observed . An aliquot of t he solution was dissolved in methanol, and a GC/MS showed only one peak with a mass of 178 suggesting full Grignard formation was achieved . I n a 500 mL round bottom flask equipped with a stir bar under an N 2 atmosphere, MeSiCl 3 (10 mL, 0.085 mol, 1.1 equiv .) was placed in 40 mL THF. The MeSiCl 3 was freshly distilled over CaH 2 . The Grignard solution was cannula transferred dropwise into the 500 mL flask containing MeSiCl 3 . The cannula transfer took approximately 45 minutes. The reaction mixture was allowed t o stir for 24 h at which time GC - MS showed full conversion. It should be noted that the initial 74 color of the solution was clear colorless whi ch turned first yellow, then green and finally a yellow - orange color. At the end of the reaction, the excess THF an d MeSiCl 3 were removed leaving behind a yellow powder. Fresh hexanes (~300 mL) were added to the powder creating a slurry. This slurry was fi lte red through a medium fritted funnel with the use of hexanes (~200 mL) to aid transfer and wash the solid on the frit. The solvent from the filtrate was removed producing a yellow solid which was dried under vacuum overnight at room temperature. Once dry , t he solid was subjected to sublimation at 70 °C under 0.01 torr . It should be noted after the first batch a cryst alline product has collected the temperature of the sublimation was raised to 95 °C. Dichloro(methyl)(4 - (phenylethynyl)phenyl)silane was coll ect ed as white crystals (12.1 g, 53% yield) with a melting point of 78 °C. 1 H NMR (500 MHz, CDCl 3 .2 Hz, 2H), 7.65 7.58 (m, 2H), 7.58 7.53 (m, 2H), 7.37 (dd, J = 4.8, 1.9 Hz, 3H), 1.05 (s, 3H). 13 C NMR (126 MHz, CDCl 3 132.98, 13 1.75, 131.24, 128.70, 128.43, 126.72, 122.78, 91.64, 88.69, 5.52. 29 Si NMR (99 MHz, CDCl 3 5.2.3 Capping of D DSQ - (Ph) 8 (OH) 4 DDSQ - 2((Me)(R)) where R = para - phenylamine (PA) 2 , para - (phenylethynyl)phenyl (PEP) 3 , and for phenyl (Ph) 4 were synthesized fol lowing Scheme 5 - 1 and are shown in Figure 5 - 2 . A 250 m L flask was charged with DDSQ - (Ph) 8 (OH) 4 ( 1 ) (2 .0 g, 1.87 mmol, 1 equiv) and a stir bar then purged with dry N 2 for 15 minutes followed by the addi tio n of THF (60 m L ) at room - temperature. To the solution was added Cl 2 Si(Me)(R) (3.74 mmol, 2 equiv) followed by dropwise addition of Et 3 N (1.04 mL, 7.48 mmol, 4 equiv). The addition of triethylamine took 5 minutes in total, and a cloudy white suspension wa s formed. The reaction mixture was stirred for 4 hours. The solution was then filtered thro ugh a fine fritted - 75 funnel - filter to remove the solid triethylamine hydrochloride. The solution was dried by rotary evapora tion to afford cis - to - trans mixtures of DDS Q - 2((Me)(R)) as a white powder. Compound 2 was obtained after deprotection of the trimethyl silyl groups using previously reported methods of methanol acidified with acetic acid and further evaporation of the volatiles . 195 , 199 These reactions generated t wo ge ometrical arrangements for Me and R. When both moieties are facing in the same directions, they are referred to cis isomers, and when they are facing in opposite directions, they are identified as trans isomers. The structures synthesized under this p roced ure are listed in Figure 5 - 2 . Figure 5 - 2 . Structures synthesized and studied in this work. 76 5.2.4 Analytical methods 1 H, 13 C, and 29 Si NMR spectroscop y wer e used to identify synthesized products. The cis - to - trans ratio of all samples investigated was first estimated using 29 Si NMR. Detailed quantification for the cis - to - trans ratio in 2 was performed based on 1 H NMR using a 2 - D NMR tec hnique previously descr ibed 195 . In general, separation of isomers was by fractional crystallization (FC) with THF as a good solvent and hexanes as a poor solvent using a procedure specified elsewhere 14 . Several cycles of fractional crystallization were need ed for required i somer purity. Isolated isomers were crystallized by slow evaporation of THF and crystals obtained were mounted on a nylon loop with paratone oil and analyzed on a Bruker APEX - II CCD diffractometer. The crystal was kept at a constant temperature of 173 K du ring d ata collection. Thermal behavior of isomers with purities superior to 95%, or nearly pure isomers, and mixtures of cis and trans isomers was studied by DSC Q2000 equipped with a mechanical cooling accessory. Typically, the temperature range of invest igatio n was from 50 °C to 350 °C with a constant heating rate of 10 °C/min. 5.3 Results and discussion 5.3.1 Separation and identification of nearly - pure isomers Nearly - pure trans and enriched cis DDSQ - 2((Me)(R)) isomers can be obtai ned by fractional crystallizatio n; 199 h owever, the ease of isolating nearly - pure cis or trans isomers varied depending on the R - group. In general, the as - synthesized DDSQ - 2 ((Me)(R)) mixtures were dissolved in THF, and addition of hexanes resulted in crystal lization and precipitation of t he tran s isomers. This in turn enriched the solution with the cis isomer. The isolation process was repeated until sufficiently isomerically pure compounds were obtained. 77 Isolation of nearly - pure cis and trans isomers by FC was easily achieved for 3 . Howe ver, se paration of the cis and trans isomers of 4 required multiple recrystallizations. The nearly - pure trans - 4 was obtained in the first cycle with this technique, but further removal of trans from the mother liquor could not be achieved after a threshold of 75 wt% cis and 25 wt% trans was reached. Unfortunately, further purification via LC was unable to increase the cis isomer ratio due to non - polar R groups in 4 . 197 , 203 Trans isomers were isolated by FC for 2 . Due to the polar nature of phenylamines in 2 , liquid chromatography proved effective for the isolation of nearly - pure cis fraction. 1 47 ,1 51 The isolated isomers were analyzed by 29 Si NMR, as seen in Figure 3 . Analysis by HPLC can be observed in the Appendix G . Figure 5 - 3 . 29 Si NMR peaks representing the nearly - pure isomers after separation; a) cis - 2 , b) trans - 2 , c) cis - 3 , d) trans - 3 , e) 75% cis - 4 , and f) trans - 4 . 78 Isolated isomers were crystallized by slow evaporation of THF to form single crystals need ed for crys tallographic analysis. We obtained crystallographic data for each compound; however, it should be noted that cis/trans - 2 and cis/trans - 3 were previously reported. 201 , 204 Data from these crystal structures are shown in Table 5 - 1 . Cry stalline packing density for the cis isomer is less than for the trans isomer and packing density was reduced as the size of R group in DDSQ - 2((Me)(R)) increases. Table 5 - 1 . Crystallization of ind ividual isome rs Compound Density (g/cm 3 ) Crystal system Space group Unit c ell a xes d im en sion ( Å) Unit Cell inclination angles (°) cis - 2 1.366 Triclinic P 1 (2) a 13.77 87.11 b 17.50 79.16 c 27.50 87.39 trans - 2 1.390 Monoclinic P 2 1 /n (14) a 10.05 90.00 b 43.57 91.66 c 14.59 90.00 cis - 3 1.321 Triclinic P 1 (2) a 14.44 85.92 b 14.90 74.70 c 18.54 79.89 trans - 3 1 .341 Triclinic P 1 (2) a 10.77 91.39 b 13.45 108.68 c 13.61 91.69 trans - 4 1.382 Triclinic P 1 (2) a 9.90 65.65 b 13.51 71.77 c 14.03 69.54 5.3.2 Thermal behavior of nearly - pure isomers Melting behavior as expressed in DSC trace for pure compounds is usually observed as a single sharp endothermic peak in which the onset temperature (T onset ) is very similar to the peak temperature (T peak ). This was observed for cis - 2 , trans - 2 , and trans - 3 indicating the purit y of these samples was >95% based on the difference between T onset and T peak and the 29 Si NMR spectra shown in Figure 5 - 3 . However, for cis - 3 and trans - 4 approximately 6 °C difference in T onset and T peak was large enough to indica te these compounds may not be as 79 pure as suggested by 29 Si NMR. In fact a first sma ll endo peak with an onset temperature near 267 °C was observed for DSC trace of nearly - pure trans - 4 . Melting temperatures (T m ) calculated from the T onset in the endo peak w ere higher for trans isomers than for cis isomers as seen in Table 5 - 2 . m ) for trans isomers is higher than cis isomers. This suggests the solid state of the trans isomer is more ordered than that of the cis isomer. From Table 5 - 2 , it was observed that the size of the R group could affect the value of T m . Larger R has a lower value of T m . The T m reduction is most likely related to the packing density as reported in Table 5 - 1 . Table 5 - 2 . Experim ental values obtained from nearly - pure cis and trans DDSQ - 2 ((Me)(R)) by m m /T m Compound T peak (°C) T m (°C) m (kJ/mol) m (J/mol K) cis - 2 27 8 27 6 37.8 69 trans - 2 313 31 1 54.6 94 cis - 3 2 70 263 46.7 87 trans - 3 30 3 300 63.7 128 trans - 4 32 1 31 4 58.9 99 5.3.3 Phase behavior of cis - trans binary mixtures Kno wing that evaluated isomers have different crystal structures we hypothesized that they may be immiscible in the solid state. Hence, it is possible for a mixture with a specified cis - to - trans ratio to exhibit eutectic transition. From the thermal analysis results for individual isomers, solid - liquid phase diagrams were calculated from the Schröder - van Laar equation ( Equation 5 - 1 ), 205 where x i is the mole fraction of the isomer i in the mixture, mi is the heat of fusion of the pure compound i , R is the gas constant, T is variable temperature, and T mi is the melting 80 temperature of compound i. Ideal behavior in the phase equilibrium was assumed, such that i ) value was s et to 1. Mixtures with near - eutectic compositions based on the calculated phase diagrams were prepared from the previously isolated isomers. Equation 5 - 1. Analysis of DSC curves for different cis to trans ratios was done based on the standard methodology used in the analysis of binary metallic alloys 206 . Here, the onset temperature of the lowest endo peak is denoted as the eutectic temperature (T E ), and the peak temperature of the highest endo peak is the l iquidus temperature (T L ). 5.3.3.1 Solid - liquid phase equilibrium of DDSQ - 2((Me)(PA)) 2 Binary mixtures of 2 were prepared using the FC method by gradual removal of trans isomers and their compositions reported within were estimated by NMR. Most DSC traces of 2 hav e two endo peaks corresponding to eutectic and liquidus transition ( Figure 5 - 4 a ). It was observed the value of T L decreases as the amount of cis isome r increases in the mixture up to the eutectic compositi o n. When x trans in the mixture reaches a value of 0.33, only one peak was resolved . The onset temperature of this DSC trace is very similar to the onset of the first endo peak in DSC traces for up 0.7 of x trans as shown in Figure 5 - 4 a . This T onset represents a near - eutectic composition. Below the eutectic point, where x trans = 0.20, the melting endotherm appears to be very broad and is representative of two overlapping me lting endotherms. T onset in the first endo pe a k agrees with the previously described eutectic temperature. For x trans = 0.50 three endothermic transitions can be observed ; it is hypothesized that cocrystal behavior was formed possibly by interactions betwe en amine moieties. Experimental onset tempera t ures and liquidus temperatures were plotted , and a 81 solid - liquid binary phase diagram of varying cis - to - trans ratios was constructed as shown in Error! Reference source not found. b . The experimental results are rela tively close to the ideal eutectic as described by Equation 5 - 1 . Figure 5 - 4 . a) DSC curves for compound 2 . Every curve was normalized for better identification of peaks. The reported x trans was estimated using NMR. b) Binary phase diagram for structure 2 . Green dots ( ) are the onset temperatures of the first endothermic transition in DSC trace. Blue squares ( ) represent the peak temperature of the highest endothermic transition. The solid line represent s the ideal eutectic as calculated using Equation 5 - 1 ; dashed line ( --- ) represents the calculated eutectic temperature T E which matches the transition temperature observed using a sample with the predicted eutectic composition. Phase I: L ( cis + trans ) ; Ph ase II: L ( cis + trans ) + S cis ; Phase III: L ( cis + trans ) + S trans ; Phase IV: S cis + S tra ns . 5.3.3.2 Solid - liquid phase equilibrium of DDSQ - 2((Me)(PEP)) 3 Phenylamine was replace with a larger non - polar PEP group to reduce possible polar interactions . Nearly - pur e cis - 3 and nearly - pure trans - 3 were separated and their purities analyzed by 29 Si - NMR and HPLC using a hypercarb ® column and ethyl acetate as mobile phase. The nearly - pure isomers were physically mixed, solubilized in THF and dried using a dynamic vacuum. Three mixtures were prepared w ith x trans = 0.3, 0.5, and 0.7. DSC traces for nearly - pure 82 isomers and their binary mixtures can be observed in F igure 5 - 5 a . The trace for x trans = 0. 3 has a single peak relatively sh arp. In contrast, traces for x trans = 0.5 and 0.7 have two endo peaks. For these traces, T liquidus was decreased as the fraction of cis isomer increased in the sample. T onset in the first endotherm transition for x trans = 0.5 and 0.7 is similar to that of x trans = 0.3. This result is distinguishing for eutectic temperature. T onset of nearly - pure cis is higher than T onset of x trans = 0.3 confirming the existence of a eutectic composition close to x trans = 0.3. The ca lculated phase diagram is close to the col lected data indicating proximity to ideal eutectic behavior for compound 3 . I n F igure 5 - 5 b are plotted the eutectic temperatu re and liquidus temperature for each mixture as well as the solid - liquid phase diagram. DSC traces for ne arly - pure cis and nearly - pure trans in F igure 5 - 5 a are similar to the same traces reported in a prior study 201 . However, the DSC trace for x trans = 0.5 is inconsistent between this work and the work reported by Moo r e et al. as they overlook ed the existence of the eutectic reaction. F igure 5 - 5 . a) DSC curves for compound 3 . Every curve was normalized for better identification of peaks. b) Binary phase diagram for st r ucture 3 . Green dots ( ) are the onset temperatures of the first endothermic transition in DSC trace. Blue squares ( ) represent the peak temperature of the hig hest endothermic transition. The solid line represents the ideal eutectic as calculated using Equation 5 - 1 ; dashed lin e ( --- ) represents the calculated eutectic temperature T E. . Phase I: L ( cis + trans ) ; Phase II: L ( cis + trans ) + S cis ; Phase III: L ( cis + trans ) + S trans ; Phase IV: S cis + S trans . 83 5.3.4 Solid - liquid phase equilibrium of DDSQ - 2((Me)(Ph)) 4 Structure 4 was analyzed to reveal the phase behavior of a DDSQ structure with the smallest aryl group possible. After synthesis of 4 , trans isomers were prog ressively removed by FC. DSC trace for x trans = 0.33 resulted in a single peak. Two endotherms were identified in all othe rs including nearly - pure trans - 4 . T liquidus steadily decreased as the fraction of cis isomer increased up to x trans = 0.33. T onset of the first endo peak has similar value as the T onset for the single peak observed in x trans = 0.33, which represents the eu tectic temperature ( Figure 5 - 6 a ). For x trans = 0.25, its DSC trace has a main peak and a shoulder. The first main peak has a T onset similar to the eutectic temperature. The shoulder located between 275 °C and 285 ° C is the T liquidus for the excess cis isomer - rich phase in the mixture. Pure cis - 4 was n ever isolated despite several crystallization attempts. It is believed that samples with compositions of cis isomers larger than 75% may be forming co - crystals making f urther r emoval of trans - 4 very difficult. The phase diagram in Figure 5 - 6 b is near to the ideal for the hypereutectic trans - rich portion. The experimentally near - eutectic composition (x trans =0.33) is extracted by fitting with the calculated trace for trans - 4 . The e utectic composition can not be calculated due to the lack of available nearly - pure cis - 4 . 84 Figure 5 - 6 . a ) DSC curves for compound 4 . Every curve was normalized for better identification of peaks. b) Partial binary phase diagram for structure 4 . Green dots ( ) are the onset temperatures of the first endothermic transition in DSC trace. Blue squares ( ) represent the peak temperature of the highest endothermic transition. The solid line represents the ideal solid - liquid equilibrium, light blue dashed line ( --- ) represents the experimental eutectic temperature T E . For nearly - pure trans - 4 are depicted the nearly - eutectic temperature from the first endo peak ( ) , the onset temperature from the second endo peak ( ) , and the peak temperature from the second endo peak ( ). Phase I: L ( cis + trans ) ; Phase II: L ( cis + trans ) + S cis ; Phase III: L ( ci s + trans ) + S trans ; Phase IV: S cis + S trans . 5.4 Conclusion s Eutectic temperatures were observed for the DDSQ - 2((Me)(R)) mixtures studied in this work. The eutectic composition was generally located near x trans = 0.3. Excitingly, the eutectic temperature was o n average 50 °C lower than the melting temperature of nea rly - pure trans isomers. DSC traces for the composition obtained after reaction ( x trans = 0.5) have liquidus temperature s 20 °C larger than the temperatures in the eutectic composition. This result indicates that liquid processing of materials containing DD SQ - 2((Me)(R)) is milder from a temperature perspective if a near - eutectic composition is used for such process. Onset t emperature in the endothermic transition for nearly - pure and for nearly - eutectic composition is inversely proportional to the size of the R group. This result indicates that selection of high steric R groups will result in lower melting temperatures for th e nearly - 85 pure isomers, as well as at the eutectic transitio n. These data allow users to tune the melting temperature of the materials by either adjusting the sterics in the system or by adjusting the ratio of cis - to - trans isomers. 86 CHAPTER 6 . SIGNIFICA NCE AND PERSPECTIVE S 87 6 Significance and perspective s 6.1 Significance This research provides a new technique for separation of DDSQ mixtures. The most significant findings are listed in the following bullets. Using adsorption HPLC , cis and trans DDSQ isomer s were resolved with optimal resolution of the elution by adjusting the solvent quality . Chromatograms obtained leaded to we ll defined peaks allowing quantitative analysis . Mixtures of isomers prepared from nearly - pure cis and nearly - pure trans were analyz ed by HPLC; the ratio between the peak areas matched with the weighted ratio. Therefore, for the first time we were able to use HPLC for q uantification of ratios in DDSQ isomeric mixtures . Mixtures containing DDSQ bonded with zero, one, and two hydroxyl gr oups were separated by HPLC using adsorption chromatography . It is believed that silanol groups in DDSQ structures are interacting with the silanol groups in the stationary phase . In consequence, DDSQ without hydroxyl groups migrate s faster than DDSQ with a single hydroxyl group. Also, DDSQ with two hydroxyl groups eluted in longer times. Also, this was separated between c is and trans isomers confirming our previous finding. Exact q uantification of the mixtures was performed by HPLC . The resultant ratio fav ors functionalization of DDSQ capped with the less bulk chlorosilane. A major finding in this research was the isolatio n for the first time of an asymmetric DDSQ. Linear a dsorption isotherm parameters for the model mixture containing different number of hy droxyl groups in an analytical Si - HPLC column were obtained experimentally by frontal analysis of breakthrough curves. It was observed that the relation between the retention times and the adsorption isotherm parameters for the DDSQ in the mixture was line ar. Thanks to this 88 linearity, adsorption parameters were extrapolated to preparative - LC packed with a commercial type o f silica . Simulation in ASPEN chromatography was performed with t he obtained parameters to estimate collection times in large - scale separ ations. Simulations predicted accurately the collection times to isolate highly - pure asymmetric DDSQ. The simulation wa s then validated by preparative - LC . This is the first time that DDSQ mixtures with different functionalities are separated in preparative - scale. DDSQ functionalized with para - phenylethynyl phenyl was separated easily between cis and trans by fractional crystallization. This was not the case for DDSQ functionalized with para - p henylamine or with p henyl. However, LC offered a solution for isol ation of nearly - pure cis and trans fr actions in the polar para - phenylamine DDSQ. Crystal structures showed that cis and trans isomers of DDSQ bonded to large groups have larger configuration differences and for instance larger solubility differences betwee n them . M elting temperature differenc es were identified between nearly - pure cis and nearly - pure trans isomers and these differences are proportional to the structure density . Evaluation of mixtures with compositions about x trans = 0.3 resulted in a single endotherm with low onset temperature compare with the nearly - pure isomers . This was recognized as the eutectic composition. In conclusion, cis and trans isomers are not miscible in the solid state and they obey the ideal mixing rule in the liquid state sho wing a eutectic composition. Besides, t he size of the aryl moiety is inversely proportional to the melting temperature of pure isomers, and the eutectic temperature of binary isomeric mixtures . 6.2 Perspectives From a separation perspective it is proposed to evaluate other stationary phases for separation of isomer systems containing non - polar molecules. Separation of isomers and ternary mixtures of non - polar DDSQ compounds was achieved along this research with the use of a hypercarb 89 column when and aryl group was present in the structure. I think the use of partition chromatography using columns bonded with C 30 could allow the separation of non - polar DDSQ isomers bonded with alkyl groups. Beyond separation, i t is possible the synthesis of asymmetric DDSQ struc tures by a direct ch emistry route. Currently, e fforts to synthesize this material has been made in our research group using boron - siloxane bonds following a route consisting in protection, capping, deprotection , and second capping . This route has showed po sitive results but t here are challenges related with premature deprotection caused by the capping solvents and reagents that should be perfected. I consider that other routes s uch as immobilization in solid - carbon substrates should be explored to achieve i mproved yields and r educe loss of starting material . Better understanding of DDSQ chemistry such as the e ffects of metals and solvents in the stability of the structure s is suggested. Several reports in reported washed with methanol, but it is possible th at byproducts of tha t washing could be harmful for the structures. Condensation of DDSQ - (Ph) 8 (OH) 4 could be done with other solvents as we are currently exploring with DCM, as well as with other bases different than E t 3 N . Along this research SOP for functi onalization using TH F and Et 3 N was elaborated. Nevertheless, a paper describing the possible outcomes for DDSQ structures under different treatments will be highly appreciated for synthetic and training purposes. Finally , the identification of rate constan ts for DDSQ capping will b e helpful to reduce reaction times in functionalization under THF considering the reactivity differences between chlorosilanes. Siloxane bonds to condense DDSQ structures can be done with chlorosilanes and methylallyloxysilanes. S ilanes and other alkoxysilanes could be potentially used for capping. 90 These groups are less reactive and for instance more cleaning techniques could be do ne to avoid several issues related with the synthesis and isolation of chlorosilanes. DDSQ structures surrounded by other groups different than phenyl will allow better compatibility with organic compounds that are not constituted by aryl groups. From a ma terial science perspective, synthesis of these DDSQ can broad the application spectrum for new materia ls and composites . Perspectives for a symmetric DDSQ structures are related with configuration of the functional groups to develop diverse grafting chemis try . I see these structures as modular building blocks with up to four connections for addition of dif ferent modules. Finally, solid - liquid behavior between binary or ternary mixtures of DDSQ structures closed with different groups could generate better un derstanding about miscibility in the liquid or in the solid phase. This knowledge is helpful for synth esis of DDSQ molecules that could serve as melting point depressors based on their chemical structure. Also, studies of ternary mixtures including an asym metric structure allow understanding the effect of DDSQ as compatibilizers between two otherwise immis cible structures. 91 APPENDI CES 92 AP P ENDIX A. SYNTHETIC PROCEDURES 93 A. Synthetic procedures General Procedure A. In a 250 m L flask purged with dry N 2 for 15 minutes, DDSQ - (Ph) 8 (OH) 4 ( 1 ) (2 g, 1.87 mmol, 1 equiv) was dissolved in THF (60 mL) at roo m temperature. Dichlorosilane or trichlorosilane (3.74 mmol, 2 equiv.) was added to the DDSQ - (Ph) 8 (OH) 4 solution. Et 3 N (1.04 mL, 7.48 mmol, 4 equiv.) was added dropwise to the stirring solution. The addition of triethylamine took 5 minutes in total; a clou dy white suspension was formed and stirred for 4 hours. The solution was then filtered through a fine fritted - funnel - filter to remove the solid triethylamine hydrochloride. The solution was dried in a rotary evaporator and then passed through a silica - gel column using DCM as the solvent. The volatiles was removed from the resulting solution and further dried at 0.4 mbar and 50 °C for 12 hours to afford DDSQ - 2( (CH 3 )(R) ) as a white powder. It should be noted the reported spectra are of the cis / trans mixtures. 94 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (H) 2 7,17 - dimethyl - 7,17 - dihydro - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AA1 ) Scheme A - 1 . Synthesis of DDSQ - 2((met hyl)(hydro)) Compound AA1 was synthesized using the general procedure A with (CH 3 )(H)SiCl 2 (3.74 mmol, 0.39 mL, 2 equiv). This provided 1.78 g (82% yield) of compound AA1 as a white powder. 1 H NMR (500 MHz, CDCl 3 ): H 0.37 (6H), 5.03 (2H), 7.16 - 7.56 (40 H). 13 C NMR (125 MHz, CDCl 3 ): C 134.27, 134.11, 134.07, 134.04, 133.94, 131.60, 130.83, 130.76, 130.70, 130.52, 130.43, 130.41, 127.86, 127.70, 127.67, 127.64, 0.67. 29 Si NMR (99 MHz, CDCl 3 ): Si - 32.77 (2Si), - 77.81 (4Si), - 79.09 ( cis, 1Si ) , - 79.30 ( tr ans, 2Si ) , - 79.50 ( cis, 1Si). 95 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (CH 3 ) 2 7,7,17,17 - tetramethyl - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AA2 ) Scheme A - 2 . Synthesis of DDSQ - 2(methyl) 2 Compound AA2 was synthesized using the general procedure A with (CH 3 ) 2 SiCl 2 (3.74 mmol, 0.46 mL, 2 equiv). This provided 1.72 g (78% yield) of compound AA2 as a white powder. 1 H NMR (500 MHz, CDCl 3 ): H 0.39 (12H), 7.16 - 7.56 (40H). 13 C NMR (125 MHz, CDCl 3 ): C 134.30, 134.10, 133.98, 132.17, 131.13, 130.37, 127.83, 127.68, 0.61. 29 Si NMR (99 MHz, CDCl 3 ): Si - 16.52 (2Si), - 78.57 (4Si), - 79.56 (4Si ) . 96 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (CHCH 2 ) 2 7,17 - dimethyl - 7,17 - divinyl - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AA3 ) Scheme A - 3 . Synthesis of DDSQ - 2((methyl)(vinyl)) Compound AA3 was synthesized using the general procedure A with (CH 3 )(CHCH 2 )SiCl 2 (3.74 mmol, 0.48 mL, 2 equiv). This provided 2.00 g (88% yield) of compound AA3 as a white powder. 1 H NMR (500 MHz, CDCl 3 ): H 0.37 (6H), 5.91 - 6.01 (4H), 6.12 - 6.19 (2H) 7.16 - 7.56 (40H). 13 C NMR (125 MHz, CDCl 3 ): C 135.07, 134.33, 133.96, 133.95, 133. 94, 133.84, 131.80, 130.90, 130.87, 130.84, 130.25, 130.21, 130.17, - 1.26. 29 Si NMR (99 MHz, CDCl 3 ): Si - 31.39 (2Si), - 78.37 ( 4Si), - 79.55 (4Si ) 29 Si NMR (99 MHz, Acetone - D 6 ): Si - 31.08 (2Si), - 78.07 (4Si), - 79.26 ( cis, 1Si ) , - 79.30 ( trans, 2Si), - 79.35 ( cis, 1Si). 97 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (CH 2 CH 2 CH 2 Cl) 2 7,17 - dimethyl - 7,17 - chloropropyl - 1,3,5,9,11,13,15,1 - octaphenylh exacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AA4 ) Scheme A - 4 . Synthesis of DDSQ - 2((methyl)(3 - propyl chloride)) Compound AA4 was synthesized using the general procedure A with (CH 3 )(C 3 H 6 C l)SiCl 2 (3.74 mmol, 0.59 mL, 2 equiv). This provided 1.86 g (76% yield) of compound AA4 as a white powder. 1 H NMR (500 MHz, CDCl 3 ): H 0.31 (6H), 0.83 - 0.86 (4H), 1.83 - 1.89 (4H) 3.37 - 3.40 (4H), 7.18 - 7.54 (40H). 13 C NMR (125 MHz, CDCl 3 ): C 134.29, 134.13, 134.11, 134.09, 134.07, 134.05, 133.92, 131.82, 130.94, 130.85, 130.77, 130.52, 130.50, 130.47, 127.92, 127.76, 47.50, 26.41, 14.42 , - 0.84 (d, J = 1.4 Hz). 29 Si NMR (99 MHz, CDCl 3 ): Si - 18.29 (2Si), - 78.46 (4Si), - 79.37 ( cis, 1Si ) , - 79.41 ( trans, 2Si), - 79.46 ( cis, 1Si). 98 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (OH) 2 7,17 - dimethyl - 7,17 - diol - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1, 9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound BB ) Scheme A - 5 . Synthesis of DDSQ - 2((methyl)(hydroxyl)) Compound BB was synthesized using the general procedure A with (CH 3 )SiCl 3 (3.74 mmol, 0.49 mL, 2 equiv ). This provided 1.58 g (71% yield) of compound BB as a white powder. 1 H NMR (500 MHz, CDCl 3 ): H 0.35 (6H), 7.16 - 7.57 (40H). 13 C NMR (125 MHz, CDCl 3 ): C 134.09, 134 .05, 134.01, 133.96, 131.37, 130.67, 130.61, 130.59, 130.55, 130.53, 127.89, 127.76, 127.73, 127.71, - 3.84. 29 Si NMR (99 MHz, CDCl 3 ): Si - 53.99 (2Si), - 78.59 (4Si), - 79.03 ( cis, 1Si ) , - 79.14 ( trans, 2Si), - 79.24 ( cis, 1Si). 99 G eneral procedure B In a 250 m L flask purged with dry N 2 for 15 minutes, DDSQ - (Ph) 8 (OH) 4 ( 1 ) (2g, 1.87 mmol) was dissolved in THF (60 m L ) at room temperature. (CH 3 )(R)SiCl 2 (1.87 mmol, 1 equiv.) and (CH 3 )SiCl 3 ( 1.87 mmol, 0.24 m L , 1 equiv.) were added to the DDSQ - (Ph) 8 (OH) 4 soluti on and stirred for 5 minutes.Et 3 N (1.04 mL, 7.48 mmol, 4 equiv.) was added dropwise to the stirring solution. The addition of triethylamine took 5 minutes in total which created a cloudy white suspension which was stirred continuously for additional four h ours. After, the solution was filtered through a fine fritted - funnel - filter to remove the solid triethylamine hydrochloride. Volatiles was removed from the resulting solution which produ ced a white powder. The powder was dissolved in a minimum amount of DC M and passed through a silica - gel column using DCM as the solvent. The silica column hydrolyzed the chlorosilanes into silanols which were isolated. The three separated products were dri ed at 0.4 mbar and 50 °C for 12 hours. 100 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 )(OH)(CH 3 )(H) 7,17 - dimethyl - 7 - hydro - 17 - ol - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AB1 ) Scheme A - 6 . Synthesis of DDSQ - (methyl)(hydro)(methyl)(hydroxyl) Compound AB1 was synthesized using the general procedure B with (CH 3 )(H)SiCl 2 (1.87 mmol, 0.19 mL, 1 equiv). This procedure produced compounds AA1 , AB1 , and BB (1.66 g, 80% yield) in a ratio of 42:39:19 respectively. Compound AB1 was isolated as a white powder (0.65 g, 31% yield). 1 H NMR (500 MHz, CDCl 3 ): H 0.35 (3H), 0.37 (3H), 4.99 (1H) 7.16 - 7.58 (40H) 13 C NMR (125 MHz, CDCl 3 ): C 134.08, 134.05, 134.04, 134.01, 133.95, 133.92, 131.52, 131.42, 130.73, 130.68, 13 0.67, 130.62, 130.55, 130.52, 130.47, 130.45, 127.86, 127.85, 127.71, 127.69, 127.68, 127.66, 0.64, - 3.86 (d, J = 1.4 Hz). 29 Si NMR (99 MHz, CDCl 3 ): Si - 32.71 (1Si), - 54.02 (1Si), - 77.60 (2Si ) , - 78.78 (2Si), - 79.06 (1Si) , - 79.15 (1Si), - 79.26 (1Si), - 79.36 (1Si). MS (APCI) m/z [M+H] + : calculated 1169.07 found 1169.06 101 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 )(OH)(CH 3 )(CH 3 ) 7,7,17 - dimethyl - 17 - ol - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AB2 ) Scheme A - 7 . DDSQ - (methyl) 2 (methyl)(hydroxyl) Compound AB2 was synthesized using the general procedure for DDSQ symmetric and asymmetric mixtures using (CH 3 )(CH 3 )SiCl 2 (1.87 mmol, 0.23 mL, 1 equiv). This procedure produced compounds AA2 , AB2 , and BB (1.67 g, 79% yield) in a ratio of 32:48:20 respectively. Compound AB2 was isolated as a white powder (0.80 g, 38% yield). 1 H NMR (500 MHz, CDCl 3 ): H 0.32 (6H), 0.35 (3H), 7.14 - 7.46 (40H) 13 C NMR (125 MHz, CDCl 3 ): C 134.17, 134.13, 134.08, 134.02, 132 .11, 131.60, 130.97, 130.92, 130.63, 130.54, 130.51, 127.95, 127.93, 127.80, 127.78, 0.67, - 3.72. 29 Si NMR (99 MHz, CDCl 3 ): Si - 16.35 (1Si), - 54.07 (1Si), - 78.45 (2Si ) , - 78.64 (2Si), - 79.26 (2Si), - 79.36 (2Si). MS (APCI) m/z [M+H] + : calculated 1183.08 found 1183.07 102 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 )(OH)(CH 3 )(CHCH 2 ) 7,17 - dimethyl - 7 - vinyl - 17 - ol - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AB3 ) Scheme A - 8 . DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) Compound AB3 was synthesized using the general procedure for DDSQ symmetric and asymmetric mixtures using (CH 3 )(CHCH 2 )SiCl 2 (1.87 mmol, 0.24 mL, 1 equiv). Th is procedure produced compo unds AA3 , AB3 , and BB (1.43 g, 64% yield) in a ratio of 28:51:21 respectively. Compound AB3 was isolated as a white powder (0.73 g, 33% yield). 1 H NMR (500 MHz, CDCl 3 ): H 0.35 (3H), 0.37 (3H), 5.91 - 6.02 (2H), 6.12 - 6.19 (1H), 7 .16 - 7.58 (40H) 13 C NMR (125 MHz, CDCl 3 ): C 135.12, 134.49, 134.06 (d, J = 1.4 Hz), 134.03 (d, J = 1.6 Hz), 133.95, 133.94, 131.80, 131.46, 130.83, 130.79, 130.73, 130.71, 130.51, 130.42, 127.83, 127.79, 127.69, 127.67, 127.64, 127.62, - 1.16, - 3.88. 29 Si NMR (99 MHz, CDCl 3 ): HSi - 31.29 (1Si), - 54.03 (1Si), - 78.32 (2Si ) , - 78.67 (2Si), - 79.30 (2Si), - 79.42 (2Si). 29 Si NMR (99 MHz, Acetone - D 6 ): Si - 31.10 (1Si), - 55.65 (1Si), - 78.04 (2Si), - 78.86 (2Si), - 79.19 (1Si), - 79.24 (1Si), - 79.35 (1Si), - 79.40 (1Si ). MS (APCI) m/z [M+H] + : calculated 1195.08 found 1195.09 103 Synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 )(OH)(CH 2 CH 2 CH 2 Cl)(CH 3 ) 7,17 - dimethyl - 7 - chloropropyl - 17 - ol - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AB4 ) S cheme A - 9 . DDSQ - (methyl)(3 - propyl chloride)(methyl)(hydroxyl) Compound AB4 was synthesized using the general procedure for DDSQ symmetric and asymmetric mixtures using (CH 3 )(C 3 H 6 Cl)SiCl 2 (1.87 mmol, 0.29 mL, 1 equiv). This procedure produced compounds AA4 , AB4 , and BB (1.23 g, 62% yield) in a ratio of 15:40:45 respectively. Comp ound AB4 was isolated as a white powder (0.47 g 20% yield). 1 H NMR (500 MHz, CDCl 3 ): H 0.33 (3H), 0.35 (3H), 0.83 - 0.86 (2H), 1.84 - 1 .89 (2H), 3.37 - 3.39 (2H), 7.16 - 7.58 (40H) 13 C NMR (125 MHz, CDCl 3 ): C 134.16, 134.14, 134.12, 134.10, 134.06, 133.97, 13 1.84, 131.50, 130.88, 130.82, 130.78, 130.73, 130.66, 130.61, 130.58, 127.99, 127.97, 127.84, 127.83, 127.82, 127.80, 47.54, 26.47, 14 .47, - 0.78, - 3.74. 29 Si NMR (99 MHz, CDCl 3 ): Si - 18.14 (1Si), - 54.08 (1Si), - 78.36 (2Si ) , - 78.65 (2Si), - 79.16 (1Si), - 79.22 (1Si), - 79.27 (1Si), - 79.32 (1Si). MS (APCI) m/z [M+H] + : calculated 1245.08 found 1245.08 104 General procedure for synthesis of (C 6 H 5 ) 8 Si 10 O 14 (CH 3 )(OH)(CH 3 )(H) 7,17 - dimethyl - 7 - hydro - 17 - ol - 1,3,5,9,11,13,15,1 - octaphenylhexacyclo[9.13.1 1,9 .1 3,15 .1 5,13 .1 11,19 ]decasiloxane (Compound AB1 ) varying the chlorosilanes equivalents Scheme A - 10 . DDSQ - (methyl)(hydro)(methyl)(hydroxyl) varying equivalents Compound AB1 was synthesized using the general procedure B varying the mol equivalents for (CH 3 )(H)SiCl 2 between 0.5, 0.8, 1.0, 1.2, and 1.5 equiv. (CH 3 )SiCl 3 was added to complete 2 equiv . of chlo rosilanes. This procedure produced compounds AA1 , AB1 , and BB in different ratios. Compound AB1 was isolated in the four cases as a white powder. 1 H NMR (500 MHz, CDCl 3 ): H 0.35 (3H), 0.37 (3H), 4.99 (1H) 7.16 - 7.58 (40H) 13 C NMR (125 MHz, CDCl 3 ): C 134.08, 134.05, 134.04, 134.01, 133.95, 133.92, 131.52, 131.42, 130.73, 130.68, 130.67, 130.62, 130.55, 130.52, 130.47, 130.45, 127.86, 127.85, 127.71, 127.69, 127.68, 127.66, 0 .64, - 3.86 (d, J = 1.4 Hz). 29 Si NMR (99 MHz, CDCl 3 ): Si - 32.71 (1Si) , - 54.02 (1Si), - 77.60 (2Si ) , - 78.78 (2Si), - 79.06 (1Si), - 79.15 (1Si), - 79.26 (1Si), - 79.36 (1Si). MS (APCI) m/z [M+H] + : calculated 1169.07 found 1169.06 105 APPENDIX B. PERCENTAGES AFTER SEPARATION OF MIXTURES WITH ZERO, ONE, AND TWO HYDROXYL GROUPS 106 B. P ercentages after separation of mixtures with zero, one, and two hydroxyl groups Table B - 1 . Isolated yield after column for c omponents in the ternary mixture H(CH 3 )SiCl 2 : (CH 3 )SiCl 3 AA1 (%) AB1 (%) BB (%) H PLC Mass HPLC Mass HPLC Mass 1.5 : 0.5 64.1 65.1 26.7 27.9 9.2 7.0 1.2 : 0.8 49.5 42.3 37.1 45.4 13.4 12.3 1 : 1 42.0 45.2 38.8 32.5 19.2 22.2 0.8 : 1.2 36.6 24.3 44.9 50.7 18.5 25 0.5 : 1.5 6.4 6.0 28.9 30.0 64.7 64.0 107 APPENDIX C. TLC FOR SEPARA TION OF DDSQ MIXTURE WITH EACH SEPARATED FRACTION 108 C. T LC for separation of DDSQ mixture with each separated fraction Figure C - 1 . TLC after separation of ternary DDSQ mixture 109 APPENDIX D. STRUCTURAL ANAL YSIS OF AB1 - D BY 29 S i - NMR AND MASS SPECTROSCOPY 110 D. S tructural analysis of AB1 - D by 29 Si - NMR and mass spectroscopy Structural Analysis for asymmetric compounds The 29 Si - NMR for 3 (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R)(OH) have eight peaks. Each Si atom was numbered in Figu re 2 for reader clarity. Si#1 bonded to R was expected to be between - 10 and - 40 ppm this peak is typical for silicas linked to two oxyge ns and two carbons. Si#2 bonded to hydroxyl group was expected to be between - 50 and - 60 ppm. This is a common value fo r silicon atoms connected to one carbon and three oxygens, being one of them a hydroxyl group. Assuming there is a cis and a trans config uration, six peaks between - 70 and - 80 ppm were expected for the eight Si atoms forming the DDSQ corners. One of this p eaks was assigned to two silicon atoms Si#3 in proximity to the edge capped with (CH 3 )(R)SiCl 2 . The other peak was assigned to two silicon atoms Si#4 in proximity to the edge capped with (CH 3 )SiCl 3 . The remaining four peaks were assigned to the four silica s Si#5a, Si#5b, Si#6a, and Si#6b in the corners exposed to four different environments. First peak for the environment containing R and OH (Si#5a, cis ), a second peak for the environment having two CH 3 groups (Si#5b cis ), a third peak for the environment i nvolving R and CH 3 (Si#6a trans ) and a fourth peak for OH and CH 3 (Si#6b trans ) in the same environment. Exact molecular mass was d etermined using the mass spectroscopy for each (C 6 H 5 ) 8 Si 10 O 14 (CH 3 ) 2 (R)(OH) . The theoretical mass matched with the exact mass found in the spectrum when one unit was subtracted from the experimental value due to the presence of a proton ionizing the molecul e. LC could not separate a mixture of cages synthesized following the general procedure B with R as H and 3 - chloropropylmethy ldichlorosilane instead of the methyltrichlorosilane. The resultant product after the column resulted in one single spot by TLC. 1 H - NMR, 13 C - NMR, and 29 Si - NMR were used to analyze the product. 29 Si - NMR spectra showed extra peaks in the 111 region between - 77 p pm and - 80 ppm plus the peaks for products AA1 and AA4 . The region between 0 ppm and - 40 ppm have two peaks near the - 18 ppm region and two peaks near the - 32 ppm region. Two of the four mentioned peaks are the characteristic peaks for the Si atom capped a t the edges of products AA1 and BB . The two other peaks suggest the presence of an extra product possibly a DDSQ cage with hydrogen and methyl to one side, and 3 - chloropropyl and methyl to the other side. Figure D - 1 . 29 Si - NMR and mass spectrums obtained after characterization of DDSQ - (methyl)(R)(methyl)(hydroxyl) products obtained as the second fraction of the separation by LC of DDSQ mixtures. a ) R = hydrogen , b ) R = methyl , c ) R = vinyl , d ) R = 3 - propyl ch loride . 112 AP PENDIX E. STRUCTURAL ANALYSIS OF A NON - POLAR MIXTURE BY 29 S i NMR 113 E. Structural analysis of a non - polar mixture by 29 Si NMR Analysis of a non - polar ternary mixture Figure E - 1 . 1 H (a), 13 C (b), and 29 Si - NMR (c), for non - polar mixture of DDSQ - 2((methyl)(hydro)) AA1, DDSQ - 2(methyl)(3 - propyl chloride) AA4, and DDSQ - (methyl)(hydro)(methyl)(3 - propyl chloride) A1A4 synthesized from Cl 2 Si(H)(CH 3 ) and Cl 2 Si(CH 2 CH 2 CH 2 Cl)(CH 3 ). 29 Si - NMR is compared ag ainst pure AA1 and pure AA4 obtained following general procedure B. This mixture is not separable by the methods employed in this work due to lack of polar moieties in the functionalized DDSQ. 114 APPENDIX F. KINETIC ANALYSIS OF DDSQ - (PH) 8 (OH) 4 SIDE - CAPP E D WITH DICHLOROSILANES WITH DIFFERENT STERIC GROUPS 115 F. K inetic analysis of DDSQ - ( P h) 8 ( OH ) 4 side - capp ed with dichlorosilanes with different steric groups Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2 (Ph) 2 analyzed by 1 H NMR at different times Figure F - 1 . Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2(Ph) 2 in different times analyzed by 1 H - NMR (500 MHz, CDCl 3 ) after quench with MeOH: a = DDSQ - (Ph) 8 (OH) 4 , b = Ph 2 SiCl 2 , c = 0.67 min , d = 1.35 min, e = 2.03 min, f = 4.63 min , g = 6.85 min, h = 10.83 min, i = 20 min, j = 36 min, k = 100 min 116 Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2 (Ph) 2 analyzed by 29 Si NMR at different times Figure F - 2 . Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2 (Ph) 2 at different times analyzed by 29 Si - NMR (99 MHz, CDCl 3 ) after quench with MeOH and solvents evaporation (times in minutes): a=DDSQ - (Ph) 8 (OH) 4 , b = 0.67 min, c = 2.03 min, d = 6.8 5 min, e = 100 min, f = DDSQ - 2(Ph) 2 completed after 4 hours reaction 117 Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2 (Me) 2 analyzed by 1 H NMR at different times Figure F - 3 . Conversion of DDSQ - (Ph) 8 (OH) 4 to AA1 or DDSQ - 2(Me) 2 in different times analyzed by 1 H - NMR (500 MHz, CDCl 3 ) afte r quench with MeOH. Times in minutes: a = 0 only DDSQ - (Ph) 8 (OH) 4 , b = 0.98 min, c = 100 min. 118 Conversion of DDSQ - (Ph) 8 (OH) 4 to DDSQ - 2 (Me) 2 analyzed by 29 Si NMR at different times Figure F - 4 . Time conversi on of DDSQ - (Ph) 8 (OH) 4 to AA1 or DDSQ - 2(Me) 2 analyzed by 29 Si - NMR (99 MHz, CDCl 3 ) after quench with MeOH and solvents evaporation: a=DDSQ - (Ph) 8 (OH) 4 , b = 0.98 min, c = 100 min. 119 Reaction kinetics for the reaction DDSQ - (Ph) 8 (OH) 4 + A 2 SiCl 2 + B 2 SiCl 2 DDSQ - A 4 + DDSQ - A 2 B 2 + DDSQ - B 4 Kinetics were modeled to analyze how much deviated were the experimental results from the theoretical results. Experiments run with less than 2 equivalents of a single chlorosilane behaves as theoretically predicted. However, (CH 3 )SiCl 3 2 resulted in an unexpected ratio. Model 1: single chlorosilane Model develop ed for prediction of ratios when less than two equivalents of chlorosilanes react with tetrahydroxyl octaphenyl double decker - shaped silsesquioxane Scheme F - 1 . DDSQ(OH) 4 capping r eactions generated for model 1 Rates: Rate laws 120 Material balance for batch reactor Rate constant simplification (this step is required because the constants are unknown) Replacing rates in the mol balance Initial conditions for the case using 1 equivalent of chlorosilane 121 MATLAB code specifying model equations function f = moles(t, C) %Derivative system for DDSQ functionalization % Concentrationn (mol) balance for Reactions DDSQ(OH)4 + R2SiCl2 % -- >DDSQR2(OH)2 and DDSQR2(OH)2 + R2SiCl2 -- > DDSQR4. Reaction rates % we re replaced in the balance. % C(1) stands for Concentration of OH4 % C(2) stands for Concentration of SiCl % C(3) stands for Concentration of OH2 % C(4) stands for Concentra tion of FC k1=1; %ml/(mmol*t) arbitrarily selected n=1; % n=1 means that f irst capping is equally fast as second capping f(1,1)= - k1*C(1)*C(2); f(2,1)= - k1*C(1)*C(2) - n*k1*C(3)*C(2); f(3,1)=k1*C(1)*C(2) - n*k1*C(3)*C(2); f(4,1)=n*k1*C(3)*C(2); end MATLAB code solving the model [tv, Cv]=ode45( 'moles' ,[0 100],[0.3 6 0.3 6 0 0]); %Derivat ive system for DDSQ functionalization % Concentrationn (mol) balance for Reactions (OH)4 + SiCl % -- >(OH)2 and (OH)2 + SiCl2 -- > FC . React ion rates % were replaced in the balance. % C(1) stands for Concentration of OH4 % C(2) stands for Concentr ation of SiCl % C(3) stands for Concentration of OH2 % C(4) stands for Concentration of FC plot(tv,Cv(:,1), 'r' ) hold on plot(tv,Cv(:,2), 'g' ) plot(tv,Cv(:,3), 'b' ) plot(tv,Cv(:,4), 'c' ) hold off title ( 'DDSQ(OH)2 kinetic model' ) ylabel( 'Concentration (mmo l/ml)' ) xlabel( 'time (s)' ) legend( 'DDSQ(OH)4' , 'A2SiCl2' , 'DDSQ(OH)2' , 'DDSQA4' ) 122 Figure F - 5 . Fractions of DDSQ( OH ) 4 (grey line), DDSQ - 2(R 1 R 2 ) (blue line), and DDSQ(OH) 2 (orange line) obtained from model 1 aft er evaluation with different equivalents of chlorosilane . 123 Model 2: two chlorosilanes Model develop ed for prediction of ratios when a mixture of two different chlorosilanes react with tetrahydroxyl octaphenyl double decker - shaped silsesquioxane Scheme F - 2 . DDSQ(OH) 4 Capping r eactions generated for model 2 124 Rates: Rate laws Material balance for batch reactor 125 Replacing rates in material balances 126 Constants assumptions First capping and second capping are similar Reactions for A are faster than reactions for B Capping rate of A in BOH2 is equal to second capping rate of A in AOH2 Capping rate of B in AOH2 is equal to second capping rate of B in BOH2 Reorganizing all constants in function of k 1 If first capping with A is no different than second capping with A, and if first capping with B is n o different than second capping with B, then: 127 Capping rate depend of A and B species, the factor p indicates that reaction of A is p times faster than reaction of B Initial conditions for the case using 1 equivalent of A and 1 equivalent of B MATLAB code specifying model equations function f = AB(t, C) %Derivative system for DDSQ functionalization % Reaction rates were replaced in the balance. % C(1) stands for Concentration of OH4 % C(2) stands for Concentration of A2OH2 % C(3) stands for Concentration of B2OH2 % C(4) stands for Concentration of A A % C(5) stands for Concentration of B B % C(6) stands for Concentration of AB % C(7) stands for Concentration of A % C(8) stands for Concentration of B k1=1000; %ml/(mmol*t) n=1; m=1; p=100; % Assuming A is 100 times faster than B k2=k1*n; k3=k1/p; k4=k3*m; k5=k4; k6=k2; f(1,1)= - k1*C(1)*C( 7) - k3*C(1)*C(8); f(2,1)=k1*C(1)*C(7) - k2*C(2)*C(7) - k5*C(2)*C(8); f(3,1)=k3*C(1)*C(8) - k4*C(3)*C(8) - k6*C(3)*C(7); f(4,1)=k2*C(2)*C(7); f(5,1)=k4*C(3)*C(8); f(6,1) =k5*C(2)*C(8)+k6*C(3)*C(7); f(7,1)= - k1*C(1)*C(7) - k2*C(2)*C(7) - k6*C(3)*C(7); 128 f(8,1)= - k3*C(1)*C(8) - k4*C(3)*C(8) - k5*C(2)*C(8); end MATLAB code solving the model [tv, Cv]=ode45( 'AB' ,[0 50],[1 0 0 0 0 0 1 1]); % DDSQ AOH2 BOH2 A A B B AB A B plot(tv,Cv(:,1), 'r' ) hold on plot(tv,Cv(:,4), 'g' ) plot(tv,Cv(:,5), 'b' ) plot(tv,Cv(:,6), 'c' ) plot(tv,Cv(:,7), 'm' ) plot( tv,Cv(:,8), 'p' ) plot(tv,Cv(:,2), 'o' ) plot(tv,Cv(:,3), 'k' ) hold off title ( 'AB kinetic model' ) ylabel( 'Concentration (mmol/ml)' ) xlabel( 'time (s)' ) legend( 'OH4' , 'A A ' , 'B B ' , 'AB' , 'A' , 'B' , 'AOH2' , 'BOH2' ) Figure F - 6 . Fractions of DDSQ - ( B ) 4 or BB (grey line), DDSQ - (A) 4 or A A (blue line), and DDSQ - (A 2 B 2 ) or AB (orange line) obtained from model 2 after evaluation with different equivalents of A and completion to two equivalents with B assuming A and B are equally rea ctive. 129 Figure F - 7 . Fractions of DDSQ - (B) 4 or BB (grey line), DDSQ - (A) 4 or AA (blue line), and DDSQ - (A 2 B 2 ) or AB (orange line) obtained from model 2 after evaluation with different equivalents of A and comp letion to two equivalents with B assuming A is ten times faster than B (p = 10). 130 Model 3 : two chlorosilanes with byproducts condensed by triethylamine Model develop ed for prediction of ratios when a mixture of two different chlorosilanes react with tetra hydroxyl octaphenyl double decker - shaped silsesquioxane . Also, condensation generated by triethylamine was included in this model. Reactions proposed and brief explanation: Formation of a complex between chlorosilanes and triethylamine. Formation of this c omplex has been described in literature and only happen between one molecule of chlorosil ane and one molecule of triethylamine. Triethylamine can be also consumed by HCl product of the rection between chlorosilanes and DDSQ - (Ph) 8 (OH) 4 . It is assumed that t hese reactions proceed extremely fast compared with the other reactions in this model. Al so it is assumed that k 16 = k 17 = k 18 Scheme F - 3 . Formation of triethylamine complex with chlorinated species Reactio n between the chlorosilane complex with DDSQ tetraol or diol, this condensation produces triethylamine hydrochloride. In here, it is assumed that the remaining Cl in the chlorosilane reacts instantaneously with the remaining hydroxyl forming HCl. In here, the capping made by A is faster than the capping made by B. k 1 > k 2 , k 1 = k 3 = k 6 , and k 2 = k 4 = k 5 131 Scheme F - 4 . Capping of DDSQ(OH) 4 with chlorosilane triethylamine complex. So far, the reactions observed d o not differ from model 2. Now a series of complexities involving byproducts will be discussed and at some extend allow understanding of yields and ratios obtained after reaction. It is known that triethylamine may act as a catalyst for condensation of sil anols. Remarkably, from the published paper it was recog nized that the reaction rate between chlorosilane and DDSQ tetrasilanol depends from the chlorosilane bulkiness. Also, it was described that a chlorosilane form complex with a single triethylamine. Ba sed on the previous descriptions it may be inferred that if the capping reaction rate is low, free triethylamine can remain in solution generating condensation between silanols. These condensation rates are slower than the chlorosilanes capping rates. k 7 < < k 1 , k 7 < k 2 , and k 7 = k 8 = k 9 = k 10 132 Scheme F - 5 . Formation of byproducts from condensation reactions in DDSQ(OH) 4 catalyzed by triethylamine. It is highly possible that a chlorosilane complex may react wit h structures condensed in one side by triethylamine. For these k 11 = k 1 and k 12 = k 2 Scheme F - 6 . Functionalization of POSS(OH) 2 with dichlorosilanes as other possible side reactions. Other reactions such as polycondensation of chlorosilanes generated by the produced water were considered but not used due to extensive simulation times. For these extra reactions it was assumed that the products of the condensation between two chlorosilan es end up as a chlorina ted polymer chain (Cl - Poly - Cl) and these chains are assumed as a single specie that is consuming chlorosilanes if there is water present. Based on these reactions it is 133 believed that the amount of DDSQ - B 4 and the amount of DDSQ - A 2 B 2 should be lower than th e value reported for the simulation without this set of byproducts. Scheme F - 7 . Production of polysiloxanes promoted by water production in DDSQ(OH)4 condensation. These reactions wer e not yet included in t he model but they are highly likely based on multiple peaks observed by 29 Si - NMR in the D - Si region Rates: 134 Rate laws 135 Material balance for batch reactor 136 Replacing rates in material balances 137 138 Reorganizing all constants to write th em in function of k 1 139 These constants were described previously. To satisfy the assumptions made for this model the values of m, n, o, p, and q must be higher than 1 Initial conditions for the case using 1 equivalent of A chlorosilane and 1 equivalent of B chlorosilane Concentrations for intermediate and final products were set to zero. MATLAB code specifying model equations function f = AB_w_Et3N(t, C) %Derivative system for DD SQ functionalization % Concentrationn (mol) balance for Reactions DDSQ(OH)4 + R2SiCl2 % -- >DDSQR2(OH)2 and DDSQR2(OH)2 + R2SiCl2 -- > DDSQR4. Reaction rates % were replaced in the balance. % C(1) stands for Concentration of DDSQ(OH)4 % C(2) stand s for Concentration of DDSQA2(OH)2 % C(3) stands for Concentration of DDSQB2(OH)2 % C(4) stands for Concentration of A2SiCl2Et3N % C(5) stands for Concentration of B2SiCl2Et3N % C(6) stands for Concentration of DDSQA4 % C(7) stands for Concentrat ion of DDSQB4 % C(8) stands for Concentration of DDSQA2B2 140 % C(9) stands for Concentration of HCl % C(10) stands for Concentration of HClEt3N % C(11) stands for Concentration of H2O % C(12) stands for Concentration of POSS(OH)2 % C(13) stands fo r Concentration of POSSA % C(14) stands for Concentration of POSSB % C(15) stands for Concentration of POSS % C(16) stands for Concentration of A2SiCl2 % C(17) stands for Concentration of B2SiCl2 % C( 18) stands for Concentration of Et3N k1=100; %ml/(mmol*t) n=100; %if n>1, A capping faster than B capping m=10; %Et3N to Chlorosilane A complex formation o=10; %Et3N to Chlorosilane B complex formation p=1000; %Condensation promoted by Et3N in DDSQ(OH)n species q=1000; %Formation of the Et3NHCl sal t, assumed to be super - fast k2=k1/n; k3=k1; k4=k2; k5=k1; k6=k2; k7=k1/p; k8=k7; k9=k7; k10=k7; k11=k1; k12=k2; k16=m*k1; k17=o*k1; k18=q*k1; f(1,1)= - k1*C(1)*C(4) - k2*C(1)*C(5) - k7*C(1)*C(18); f(2,1)=k1*C(1)*C(4) - k3*C(2)*C(4) - k6*C(2)*C(5) - k8*C(2)*C(18); f( 3,1)=k2*C(1)*C(5) - k4*C(3)*C(5) - k5*C(3)*C(4) - k9*C(3)*C(18); f(4,1)=k16*C(18)*C(16) - k1*C(1)*C(4) - k3*C(2)*C(4) - k5*C(3)*C(4) - k11*C(12)*C(4); f(5,1)= k17*C(18)*C(17) - k2*C(1)*C(5) - k4*C(3)*C(5) - k6*C(2)*C(5) - k12*C(12)*C(5); f(6,1)=k3*C(2)*C(4); f(7,1)=k4*C(3)*C(5); f(8,1)=k5*C(3)*C(4)+k6*C(2)*C(5); f(9,1)= - k18*C(18)*C(9)+k1*C(1)*C(4)+k2*C(1)*C(5)+k3*C(2)*C(4)+k4*C(3)*C(5)+k5*C(3) *C(4)+k6*C(2)*C(5)+k11*C(12 )*C(4)+k12*C(12)*C(5); f(10,1)=k18*C(18)*C(9)+k1*C(1)*C(4)+k2*C(1)*C(5)+k3*C(2)*C(4)+k4*C(3)* C(5)+k5*C(3)*C(4)+k 6*C(2)*C(5)+k11*C(12)*C(4)+k12*C(12)*C(5); f(11,1)=k7*C(1)*C(18)+k8*C(2)*C(18)+k9*C(3)*C(18)+k10*C(12)*C(18); f(12,1)=k7*C(1)*C(18) - k10*C(12)*C( 18) - k11*C(12)*C(4) - k12*C(12)*C(5); f(13,1)=k8*C(2)*C(18)+k11*C(12)*C(4); f(14,1)=k9*C(3)*C(18)+k12*C(12)*C(5); f (15,1)=k10*C(12)*C(18); f(16,1)= - k16*C(18)*C(16); f(17,1)= - k17*C(18)*C(17); f(18,1)= - k16*C(18)*C(16) - k17*C(18)*C(17) - k18*C(18)*C(9); end 141 Solver for model 3 [tv, Cv]=ode45( 'AB_w_Et3N' ,[0 100],[1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 4]); % [OH4 AOH2 BOH2 AEt3N B Et3N A4 B4 AB HCl Et3NHCl H20 POSSOH2 POSSA POSSB POSS ASi BSi Et3N] plot(tv,Cv(:,1), ' - s' , 'color' ,1/255*[255 0 0], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[255 0 0], 'MarkerFaceColor' ,1/255*[255 0 0], 'MarkerIndices' ,1:50000:length(Cv)) hold on plot(tv,Cv(:,6) , ' - *' , 'color' ,1/255*[255 0 255], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[255 0 255], 'MarkerFaceColor' ,1/255*[255 0 255], 'MarkerInd ices' ,1:50000:length(Cv)) plot(tv,Cv(:,7), ' - x' , 'color' ,1/255*[154 0 255], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[154 0 255], 'Ma rkerFaceColor' ,1/255*[154 0 255], 'MarkerIndices' ,1:50000:length(Cv)) plot(tv,Cv(:,8), ' - d' , 'color' ,1/255*[0 0 255], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[0 0 255], 'MarkerFaceColor' ,1/255*[0 0 255], 'MarkerIndices' ,1:50000:length(Cv)) plot(tv,Cv(:,13), ' - p' , ' color' ,1/255*[0 255 255], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[0 255 255], 'MarkerFaceColor' ,1/255*[0 255 255], 'MarkerIndices' ,1:50000:length(Cv)) plot(tv,Cv(:,14), ' - h' , 'color' ,1/255*[0 255 0], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[0 255 0], 'MarkerFaceCo lor' ,1/255*[0 255 0], 'MarkerIndices' ,1:50000:length(Cv)) plot(tv,Cv(:,15), ' - +' , 'color' ,1/255*[0 0 0], 'MarkerSize ' ,4, 'MarkerEdgeColor' ,1/255*[0 0 0], 'MarkerFaceColor' ,1/255*[0 0 0], 'MarkerIndices' ,1:50000:length(Cv)) plot(tv,Cv(:,11), ' - o' , 'color' ,1/255*[67 142 58], 'MarkerSize' ,4, 'MarkerEdgeColor' ,1/255*[67 142 58], 'MarkerFaceColor' ,1/255*[67 142 58], 'MarkerIndices' ,1 :50000:length(Cv)) hold off title ( 'DDSQA2B2 kinetic model' ) ylabel( 'Concentration (mmol/ml)' ) xlabel( 'time (s)' ) legend( 'DDSQ(OH)4' , 'DDSQA4' , 'D DSQB4' , 'DDSQA2B2' , 'POSSA' , 'POSSB' , 'POSS' , 'H2 0' ) Results for the proposed model For the case in which the reaction rate of chlorosilane A is higher than the rate of the B chlorosilane and using the m, n, o, p, and q constants as described in the Matlab cod e. 142 Figure F - 8 . Modelling of functionalization of DDSQ(OH ) 4 with two different chlorosilanes including the formation of byproducts. The kinetic constants are not real values but related to k1 based on experi mental observations. From this plot it can be observed how production of DDSQ - A 4 is favored compared to that of DDSQ - B 4 . It is also remarkable that under the conditions modeled byproducts add a total of 0.1 mmol/ml. It should be noticed that this result was obtained assuming that condensation caused by triethylamine is 1000 times slow er than capping of the A chlorosilane and 10 times slower than capping of the B chlorosilane. These assumptions were made based on experimental observations. 143 Figure F - 9 . Evolution of triethylamine, triethylamine complexes, and water production in the DDSQ - (OH) 4 functionalization. The behavior of chlorosilanes and triethylamine showed fast consumption of both pure chlorosilanes to form the complex with triethylamine. Howe ver, triethylamine and the complex from the chlorosilane B cannot react anymore and end up as byproducts after synthesis. It is believed that these must react with water to produced siloxane polymers or cycles. Experimentally once the reaction is finished, filtered, and dried it is possible to watch HCl smoke once DCM is added to dissolve the material. It is possible to infer that the model may be predicting the reaction byproducts, and at some extent explaining the reaction yield as well as the ratios bet ween DDSQ - A 4 , DDSQ - A 2 B 2 , and DDSQ - B 4 quantified experimentally after column. The products condensed by Et 3 N were not identified experimentally because these may be stuck in the column stationary phase. 144 APPENDI X G. LC AND FC SEPARATIONS ANALYZED BY HPLC 145 G. LC and FC separations analyzed by HPLC Fraction analysis for separation of mixture B by HPLC Chromatograms processed by HPLC for each fraction collected along the preparatory column for mixture B containin g 2 DDSQ - 2((CH 3 ) 2 ), 3b DDSQ - 2((CHCH 2 )(CH 3 )), and 4b DDSQ - 2((CHCH 2 )(OH)) . The flow rate was variable, and fractions of 19 mL were collected. At 59 min fractions were rich in trans - 4b . Final fractions were cis - 4b rich fraction. 0.0E+0 5.0E+1 0 5 10 14.0 min 0.0E+0 5.0E+1 0 5 10 14.4 min 0.0E+0 2.0E+2 0 2 4 6 8 10 14.8 min 0.0E+0 1.0E+3 0 2 4 6 8 10 15.1 min 0.0E+0 2.0E+3 0 2 4 6 8 10 15.5 min 0.0E+0 2.0E+3 0 2 4 6 8 10 15.8 min 0.0E+0 1.0E+3 0 2 4 6 8 10 16.2 min 0.0E+0 5.0E+2 0 2 4 6 8 10 16.5 min 0.0E+0 2.0E+2 0 2 4 6 8 10 16.9 min 0.0E+0 1.0E+2 0 2 4 6 8 10 17.2 min 0.0E+0 1.0E+2 0 2 4 6 8 10 17.6 min 0.0E+0 5.0E+1 0 2 4 6 8 10 26.7 min 0.0E+0 1.0E+2 0 2 4 6 8 10 27.1 min 0.0E+0 2.0E+2 0 2 4 6 8 10 27.4 min 0.0E+0 2.0E+2 0 2 4 6 8 10 27.8 min 0.0E+0 5.0E+2 0 2 4 6 8 10 28.1 min 0.0E+0 5.0E+2 0 2 4 6 8 10 28.5 min 0.0E+0 2.0E+2 0 2 4 6 8 10 28.9 min 0.0E+0 2.0E+2 0 2 4 6 8 10 29.2 min 0.0E+0 2.0E+2 0 2 4 6 8 10 29.6 min 0.0E+0 2.0E+2 0 2 4 6 8 10 29.9 min 0.0E+0 1.0E+2 0 2 4 6 8 10 30.2 min 0.0E+0 1.0E+2 0 2 4 6 8 10 30.6 min 0.0E+0 1.0E+2 0 2 4 6 8 10 31.0 min 146 0.0E+0 5.0E+1 0 2 4 6 8 10 31.3 min 0.0E+0 5.0E+1 0 2 4 6 8 10 31.7 min 0.0E+0 5.0E+0 0 2 4 6 8 10 59.1 min 0.0E+0 5.0E+0 0 2 4 6 8 10 60.9 min 0.0E+0 2.0E+0 0 2 4 6 8 10 62.6 min 0.0E+0 2.0E+0 0 2 4 6 8 10 64.4 min 0.0E+0 2.0E+0 0 2 4 6 8 10 66.1 min 0.0E+0 2.0E+0 0 2 4 6 8 10 67.9 min 0.0E+0 2.0E+0 0 2 4 6 8 10 70.0 min 0.0E+0 5.0E+2 0 2 4 6 8 10 Remain after 1L flush 147 HPLC flow rate profile for preparatory column separating mixture C Figure G - 1 . Flow rate ramps for preparative column under non - constant flow rate . Dots in this graph are experimentally measured flow rates. Each point is defined as the lapse of time for completion of 19 mL. Points below 25 mL/min represent reservoir refilling times when the pressure was relieved to add more solvent. Linear flow rate ranging 52 to 56 mL/min was obtained only when t he pressure was reduced. 0 20 40 60 80 100 120 140 160 180 200 0 10 20 30 40 50 Flow rate (Fv) mL/min time (min) 148 Fraction analysis for s eparation of mixture C by HPLC Chromatograms processed by HPLC for each fraction collected along the preparatory column for mixture IB containing the molecules 2 DDSQ - 2((CH 3 ) 2 ), 3c DDSQ - 2((CH 2 CH(CH 3 ) 2 )(CH 3 ) ), and 4c DDSQ - 2((CH 2 CH(CH 3 ) 2 )(OH)) . The flow rate was variable, and fractions of 19 mL were collected. From 32 min to 37 min fractions were rich in trans - 4c . Final fraction was a fraction rich in cis - 4c . 0.0E+0 5.0E+1 0 1 2 3 4 5 6 15.3 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 15.6 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 15.8 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 16.0 min 0.0E+0 1.0E+3 0 1 2 3 4 5 6 16.3 min 0.0E+0 1.0E+3 0 1 2 3 4 5 6 16.5 min 0.0E+0 1.0E+3 0 1 2 3 4 5 6 16.7 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 16.9 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 17.1 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 17.3 min 0.0E+0 1.0E+2 0 1 2 3 4 5 6 17.5 min 0.0E+0 1.0E+2 0 1 2 3 4 5 6 17.7 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 17.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 18.1 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 18.3 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 18.5 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 18.6 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 18.9 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 19.0 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 19.2 min 0.0E+0 1.0E+1 0 1 2 3 4 5 6 19.4 min 0.0E+0 1.0E+1 0 1 2 3 4 5 6 19.6 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 19.8 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 20.0 min 149 0.0E+0 1.0E+2 0 1 2 3 4 5 6 20.2 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 20.4 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 20.5 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 20.7 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 20.9 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 21.1 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 21,3 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 21.5 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 21.7 min 0.0E+0 5.0E+2 0 1 2 3 4 5 6 21.9 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 22.1 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 22.3 min 0.0E+0 2.0E+2 0 1 2 3 4 5 6 22.7 min 0.0E+0 1.0E+2 0 1 2 3 4 5 6 23.1 min 0.0E+0 1.0E+2 0 1 2 3 4 5 6 23.3 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 23.5 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 23.7 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 24.2 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 24.6 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 24.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 25.2 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 25.5 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 25.9 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 26.2 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 26.5 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 26.8 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 27.1 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 27.5 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 27.8 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 28.1 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 28.5 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 28.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 29.1 min 150 0.0E+0 5.0E+1 0 1 2 3 4 5 6 29.5 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 29.8 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 30.1 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 30.4 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 30.8 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 31.1 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 31.5 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 31.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 32.2 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 32.6 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 32.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 33.3 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 33.6 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 34.0 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 34.3 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 34.4 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 34.7 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 34.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 35.2 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 35.4 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 35.6 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 35.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 36.1 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 36.4 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 36.6 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 36.9 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 37.1 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 37.4 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 37.7 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 38 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 39 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 40 min 0.0E+0 5.0E+1 0 1 2 3 4 5 6 41.3 min 151 0.0E+0 5.0E+1 0 1 2 3 4 5 6 41.5 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 41.9 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 42.2 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 42.4 min 0.0E+0 2.0E+1 0 1 2 3 4 5 6 42.7 min 0.0E+0 1.0E+1 0 1 2 3 4 5 6 43.0 min 0.0E+0 1.0E+1 0 1 2 3 4 5 6 43.2 min 0.0E+0 5.0E+0 0 1 2 3 4 5 6 43.5 min 152 Purity analysis after FC of cis and trans DDSQ - 2((Me)(PEP)) 3 by HPLC Samples evaluated in a hypercarb ® column of 10 cm length and 0.46 cm internal diameter. Ethyl acetate was used as mobile phase wi th a flow rate of 0.4 ml/min. The eluent was analyzed by a UV detector Figure G - 2 . 50% trans and 50% cis mixtu re of DDSQ - 2((methyl)(para - phenylethynyl phenyl))after hypercarb column Figure G - 3 . Mostly trans isomer of DDSQ - 2((methyl)(para - phenylethynyl phenyl))after hypercarb column Figure G - 4 . Mostly cis isomer of DDSQ - 2((methyl)(para - phenylethynyl phenyl))a fter hypercarb column 0 5 10 15 20 Retention time (min) 0 5 10 15 20 Retention time (min) trans 0 5 10 15 20 Retention time (min) cis 153 Separation of a non - polar ternary mixture Samples evaluated in a hypercarb ® column of 10 cm length and 0.46 cm internal diameter. Ethyl acetate was used as mobile phase with a flow rate of 0.4 ml/min. The eluent was analyzed by a UV detector Scheme G - 1 . Synthesis of a ternary non - polar mixture with diphenyldichlorosilane and dimethyldichlorosilane. Figure G - 5 . Possible separation of the ternary mixture described in Scheme G - 1 . Synthesis of a ternary non - polar mixture with diphenyldichlorosilane and dimethyldichlorosilane. Scheme G - 1 using 98:2 DCM:Acetonit rile as mobile phase. Blue line represents the chromatogram with a 154 DCM injection; yellow line is the chromatogram for a mixture of tetramethyl DDSQ and tetraphenyl DDSQ . Red line represent s the chromatogram for the ternary mixture. 155 APPENDIX H. SUMMARIZE D EUTECTIC AND LIQUIDUS COMPOSITIONS 156 H. Summarized eutectic and liquidus compositions Summarized onset and liquidus temperatures in DSC traces Table H - 1 . Eutectic temperature (T E ) and liquidus temperatures (T L ) for binary cis - to - trans mixtures of compound 2 . x trans T E (°C) T L (°C) 0 - 275 0.1 - 272 0.2 247 263 0.3 256 - 0.5 248 272 0.6 245 296 0.7 246 302 0.8 - 305 0.9 - 309 - 310 Table H - 2 . Eutectic temperature (T E ) and liquidus temperatures (T L ) for binary cis - to - trans mixtures of compound 3 . x trans T E (°C) T L (°C) - 263 0.3 256 - 0.5 257 270 0.7 245 28 4 - 301 Table H - 3 . Eutectic temperat ure (T E ) and liquidus temperatures (T L ) for binary cis - to - trans mixtures of compound 4 . x trans T E (°C) T L (°C) 0.25 269 280 ± 5 0.33 269 - 0.5 259 295 0.6 262 312 0.9 269 316 269 320 Analysis of the experimental eutectic temperatures shows tha t TE for 2 and 3 is 14 °C lower than TE for 4. However, the eutectic composition is similar for the three cases studied 157 in this work. It was observed that the T E for 3 is 26 °C lower than the liquidus temperature of the as synthesized mixture (xtrans = 0.5 ), for 2 the difference is 20 °C, and for 4 the difference is 14 °C. These results suggest that the use of large aryl groups decreased the melting temperature o f pure components and had a pronounced effect decreasing the eutectic temperature. 158 APPENDIX I. NMR SPECTRA FOR COMPONENTS SYNTHESIZED AND SEPARATED IN THIS WORK 159 I. NMR spectra for components synthesized and separated in this work (Isobutyl)( par a - aniline(trimethylsilyl))dichlorosilane Figure I - 1 . 29 Si - NMR (CDCl 3 , 99 MHz) for (i sobutyl)(para - aniline(trimethylsilyl))dichlorosilane Figure I - 2 . 1 H - NMR (CDCl 3 , 500 MHz) fo r (i sobutyl)(para - aniline(trimethylsilyl))dichlorosilane 160 DDSQ - 2( (methyl)(para - aniline) ) Figure I - 3 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((methyl)(p ara - aniline)) Figure I - 4 . 1 H - NMR (CDCl 3 , 50 0 MHz) for DDSQ - 2((methyl)(para - aniline)) 161 DDSQ - 2( (methyl)(meta - aniline) ) Fi gure I - 5 . 29 Si - NMR (CDCl 3 , 99 MHz) DDSQ - 2((methyl)(meta - aniline)) Figure I - 6 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2(( methyl )(meta - ani line)) 162 DDSQ - 2((isobutyl)(meta - aniline)) Figure I - 7 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((isobutyl)(meta - aniline)) Fi gure I - 8 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((isobutyl)(meta - aniline)) 163 DDQS - ((cyclohexyl)(meta - aniline)) Figure I - 9 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDQS - ((cyclohexyl)(meta - aniline)) Figure I - 10 . 1 H - NMR (CDCl 3 , 500 MHz) for DDQS - ((cyclohexyl)(meta - aniline)) 164 DDSQ - 2( (methyl)(hydroxy l ) ) Figure I - 11 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((methyl)(hydroxyl)) Figure I - 12 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((methyl)(hydroxyl)) 165 DDSQ - 2( (vinyl)(hydroxyl) ) 29 Si - NMR (CDCl 3 , 99 MHz) Figure I - 13 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((vinyl)(hydroxyl)) Figure I - 14 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((vinyl)(hydroxyl)) 166 DDSQ - 2( (isopropyl)(hydroxyl) ) Figure I - 15 . 29 Si - NMR (CDCl 3 , 99 MHz ) for DDSQ - 2((isopropyl)(hydroxyl)) Figure I - 16 . 1 H - NMR (CDCl 3 , 50 0 MHz) for DDSQ - 2((isopropyl)(hydroxyl)) 167 DDSQ - 2( (isobutyl)(hydroxyl ) ) Figure I - 17 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((isobutyl)(hydroxyl)) Figure I - 18 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((isobutyl)(hydroxyl)) 168 DDSQ - 2( (phenyl)(hydroxyl ) ) Figure I - 19 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((phenyl)(hydroxyl)) Figure I - 20 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((phenyl)(hydroxyl)) 169 DDSQ - 2( (meth yl)(propyl - cyanide) ) Figure I - 21 . 29 Si - NMR (CDCl 3 , 99 MHz) for DDSQ - 2((methyl)(propyl - cyanide)) Figure I - 22 . 1 H - NMR (CDCl 3 , 500 MHz) for DDSQ - 2((methyl)(propyl - cyanide)) 170 DDSQ - 2( (methyl)( hydro ) ) Figure I - 23 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydro)) 171 Fig ure I - 24 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydro)) 172 Figure I - 25 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydro)) 173 DDSQ - 2(methyl ) 2 Figure I - 26 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2(methyl) 2 174 Figure I - 27 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2(methyl) 2 175 Figure I - 28 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2(methyl) 2 176 DDSQ - 2((methyl)(vinyl)) Figure I - 29 . 1 H - NMR (500 MHz, C DCl 3 ) for DDSQ - 2((methyl)(vinyl)) 177 Figure I - 30 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(vinyl)) 178 Figure I - 31 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(vinyl)) 179 Figure I - 32 . 29 Si - NMR (99 MHz, Acetone - D6) for DDSQ - 2((methyl)(vinyl)) 180 DDSQ - 2(methyl)(3 - propyl - chloride) Figure I - 33 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2(methyl)(3 - propyl - chloride) 181 F igure I - 34 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - 2(methyl)(3 - propyl - chloride) 182 Figure I - 35 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2(methyl)(3 - propyl - chloride) 183 DDSQ - 2((methyl)(hydroxyl)) Figure I - 36 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) 184 Figure I - 37 . 13 C - NMR (125 MHz, CDCl 3 ) for D DSQ - 2((methyl)(hydroxyl)) 185 Figure I - 38 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) h ydrolyzed by column chromatography Figure I - 39 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(hydroxyl)) h ydrolyzed with acidified H 2 O 186 DDSQ - (methyl)(hydro)(methyl)(hydroxyl) Figure I - 40 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl)(hydr o)(methyl)(hydroxyl) 187 Figure I - 41 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl)(hydro)(methyl)(hydroxyl) 188 Figure I - 42 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl)(hydro)(methyl)(hydroxyl) 189 DDSQ - (methyl) 2 (methyl)(hydroxyl) Figure I - 43 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (methyl)(hydroxyl) 190 Figure I - 44 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (methyl)(hydroxyl) 191 Figure I - 45 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (methyl)(hydroxyl) 192 DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) Figure I - 46 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) 193 Figure I - 47 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl)(vinyl)(methyl)(h ydroxyl) 194 Figure I - 48 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl)(vinyl)(methyl)(hydroxyl) 195 Figure I - 49 . 29 Si - NMR (99 MHz, Acetone - D6) for DDSQ - (methyl)( vinyl)(methyl)(hydroxyl) 196 DDSQ - (methyl)(3 - propyl - chloride)(methyl)(hydroxyl) Figure I - 50 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl)(3 - propyl - chloride)(methyl)(hydroxyl) 197 Figure I - 51 . 13 C - NMR (125 MHz, CDCl 3 ) for DDSQ - (methyl)(3 - propyl - chloride)(methyl)(hydroxyl) 198 Figure I - 52 . 29 Si - NMR (99 MHz, CDCl 3 ) for DDSQ - (methyl)(3 - propyl - chloride)(methyl)(hydr oxyl) 199 DDSQ - (methyl) 2 (vinyl)(hydroxyl) Figure I - 53 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (vinyl)(hydroxyl) Figure I - 54 . 13 C - NMR ( 125 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (vinyl)(hydroxyl) Figure I - 55 . 29 Si - NMR ( 99 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (vinyl)(hydroxyl) 200 DDSQ - (methyl) 2 (isobutyl)(hydroxyl) Figure I - 56 . 1 H - NMR (500 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (isobutyl )(hydroxyl) Figure I - 57 . 13 C - NMR ( 125 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (isobutyl)(hydroxyl) Figure I - 58 . 29 Si - NM R ( 99 MHz, CDCl 3 ) for DDSQ - (methyl) 2 (isobutyl)(hydroxyl) 201 (Methyl)(para - aniline(trimethylsilyl))dichlorosilane Figure I - 59 . 1 H NMR ( 500 MHz, CDCl 3 ) for ( m ethyl)(para - aniline(trimethylsilyl))dichlorosilane Figure I - 60 . 13 C NMR ( 125 MHz, CDCl 3 ) for ( m ethyl)(para - aniline(trimethylsilyl))dichlorosilane Figure I - 61 . 29 Si NMR ( 99 MHz, CDCl 3 ) for ( m ethyl)(para - aniline(trimethylsilyl))dichlorosilane 202 1,4 - Bromophenylethynylbenzene Figure I - 62 . 1 H NMR ( 500 MHz, CDCl 3 ) for 1,4 - Bromophenylethynylbenzene Figure I - 63 . 13 C NMR ( 125 MHz, CDCl 3 ) for 1,4 - Bromophenylethynylbenzene 203 1,4 - (Phenylethynyl)phenyl methyldichlorosilane Figure I - 64 . 1 H NMR ( 500 MHz, CDCl 3 ) for 1,4 - (Phenylethynyl)phenyl methyldichlorosilane Figure I - 65 . 13 C NMR ( 125 MHz, CDCl 3 ) for 1,4 - (Phenylethynyl)phenyl methyldichlorosilane Figure I - 66 . 29 Si NMR ( 99 MHz, CDCl 3 ) for 1,4 - (Phenylethynyl)phenyl methyldichlorosilane 204 DDSQ - 2(( m e thyl )( para - phenylethynyl phenyl )) Figure I - 67 . 1 H NMR ( 500 MHz, CDCl 3 ) for DDSQ - 2((methyl)(par a - phenylethynyl phenyl)) Figure I - 6 8 . 13 C NMR ( 125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(para - phenylethynyl phenyl)) Figure I - 69 . 29 Si NMR ( 99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(para - phenylethynyl phenyl)) 205 DDSQ - 2(( m e thyl )( phenyl ) ) Figure I - 70 . 1 H NMR ( 500 MHz, CDCl 3 ) for DDSQ - 2((methy l)(phenyl)) Figure I - 71 . 13 C NMR ( 125 MHz, CDCl 3 ) for DDSQ - 2((methyl)(phenyl)) Figure I - 72 . 29 Si NMR ( 99 MHz, CDCl 3 ) for DDSQ - 2((methyl)(phenyl)) 206 APPENDIX J. CRYSTALLOGRAPHIC INFORMATION 207 J. C rystallographic information cis - DDSQ - 2(( m e thyl )( para - phenylamine )) Figure J - 1 . Molecular structure of cis - DDSQ - 2((methyl)(para - phenylamine)) . White = H, Red = 0, Gray = C, Yellow = Si, Blue = N 208 Figure J - 2 . Packing structure for cis - DDSQ - 2((methyl)(para - phenylamine)) 209 trans - DDSQ - 2(( m e thyl )( para - phenylamine )) Figure J - 3 . Molecular structure of trans - DDSQ - 2((methyl)(para - phenylamine)). White = H, Red = 0, Gray = C, Yellow = Si, Blue = N 210 Figure J - 4 . Packing str ucture for trans - DDSQ - 2((methyl)(para - phenylamine)) 211 cis - DDSQ - 2(( m e thyl )( para - phenylethynyl phenyl )) Figure J - 5 . Molecular structure of cis - DDSQ - 2((methyl)(para - phenylethynyl phenyl)) White = H, Red = 0 , Gray = C, Yellow = Si, Blue = N 212 Figure J - 6 . Packing structure for cis - DDSQ - 2((methyl)(para - phenylethynyl phenyl)) 213 trans - DDSQ - 2(( m e thyl )( p ara - phenylethynyl phenyl )) Figure J - 7 . Molecular structure of trans - DDSQ - 2((methyl)(para - phenylethynyl phenyl)) White = H, Red = 0, Gray = C, Yellow = Si, Blue = N 214 Figure J - 8 . Packing structure for trans - DDSQ - 2((methyl)(para - phenylethynyl phenyl)) 215 trans - DDSQ - 2(( m e thyl )( phenyl ) ) Figure J - 9 . Molecular structure o f trans - DDSQ - 2((methyl)(phenyl)) White = H, Red = 0, Gray = C, Yellow = Si, Blue = N 216 Figure J - 10 . Packing structure for trans - DDSQ - 2((methyl)(phenyl)) 217 REFERENCES 218 REFERENCES (1) Hartmann - Thompson, C. Applications of Polyhedral O ligomeric Silsesquioxanes ; Springer Science & Business Media, 2011 . (2) Zhang, X.; Huang, Y.; Wang, T.; Liu, L. Influence of Fibre Surface Oxidation reduction Followed by Silsesquioxane Coating Treatment on In terfacial Mechanical Properties of Carbon Fibre/polyarylacetylene Comp osites. Compos. Part A Appl. Sci. Manuf. 2007 , 38 , 936 944. (3) Cordes, D. B.; Lickiss, P. D.; Rataboul, F. Recent Developments in the Chem istry of Cubic Polyhedral Oligosilsesquioxanes. Chem. Rev. 2010 , 110 , 2081 2173. (4) Walczak, M.; Januszewski, R.; Majchrzak, M.; Kubicki, M.; Dudziec, B.; Marciniec, B. Unusual Cis and Trans Architecture of D ihydrofunctional Double - Decker Shaped - Conjugated Arene Derivatives. New J. Chem. 2017 , 41 , 3290 3296. (5) Ghanbari, H.; Cousins, B. G.; Seifalian, A. M. A Na nocage for Nanomedicine: Polyhedral Oligomeric Silsesquioxane (POSS). Macromol. Rapid Commun. 2011 , 32 , 1032 1046. (6) Feher, F. J.; Newman, D. A.; Walzer, J. F. Silsesquioxanes as Models for Silica Surfaces. J. Am. Chem. Soc. 1989 , 111 , 1741 1748. (7) Feher, F. J.; Phillips, S. H.; Ziller, J. W. Facile and Remarkably Selective Substitution Reactions Involving Framework Silicon Atoms in Silsesquioxane Frameworks. J . Am. Chem. Soc. 1997 , 119 , 3397 3398. (8) Phillips, S. H.; Haddad, T. S.; Tomczak, S . J. Developments in Nanoscience: Polyhedral Oligomeric Silsesquioxane (POSS) - Polymers. Curr. Opin. Solid State Mater. Sci. 2004 , 8 , 21 29. (9) Morimoto, Y.; Watanabe, K.; Ootake, N.; Inagaki, J. - I.; Yoshida, K.; Ohguma, K. Silsesquioxane Derivatives and Process for Production Thereof. 20040249103A1, December 9, 2004 . (10) Ootake, N.; Yoshida, K. Organic Silicon Compound and Method for Producing Same, and Polysiloxa ne and Method for Producing Same. 20060155091:A1, July 13, 2006 . (11) Lee, D. W.; Kawakami, Y. Incompletely Condensed Silsesquioxanes: Formation and Reactivity. Polym. J. 2007 , 39 , 230 238. (12) Li, Z.; Kawakami, Y. Formation of Incompletely Condensed Oligosilsesquioxanes by Hydrolysis o f Completely Condensed POSS via Reshuffling. Chem. Lett. 2008 , 37 , 804 805. 219 (13) Schoe n, B. W.; Holmes, D.; Lee, A. Identification and Quantification of Cis and Trans Isomers in Aminophenyl Double - Decker Silses quioxanes Using 1H -- 29Si gHMBC NMR. Magn. Reson. Chem. 2013 , 51 , 490 496. (14) Schoen, B. W.; Lira, C. T.; Lee, A. Separation and Solubility of Cis and Trans Isomers in Nanostructured Double - Decker Silsequioxan es. J. Chem. Eng. Data 2014 , 59 , 1483 1493. (15) Schoen, B. W. Aminophenyl Double Deck er Silsesquioxanes: Spectroscopic Elucidation, Physical and Thermal Characterization, and Their Applications; 2013 . (16) Ervithayasuporn, V.; Sodkhomkhum, R.; Teerawatananond, T.; Phurat, C.; Phinyocheep, P.; S omsook, E.; Osotchan, T. Unprecedented Formation of Cis - and Trans - Di [(3 - Chloropropyl) Isopropoxysilyl] - Bridged Double - Decke r Octaphenylsilsesquioxanes. Eur. J. Inorg. Chem. 2013 , 2013 , 3292 3296. (17) Hoque, M. A.; Kakihana, Y.; Shinke, S.; Kawakami, Y. Polysiloxanes with Periodically Distributed Isomeric Double - Decker Silsesquioxa ne in the Main Chain. Macromolecules 2009 , 42 , 3309 3315. (18) Cao, J.; Fan, H .; Li, B. - G.; Zhu, S. Synthesis and Evaluation of Double - Decker Silsesquioxanes as Modifying Agent for Epoxy Resin. Polymer 2017 , 1 2 4 , 157 167. (19) Xu, S.; Zhao, B.; Wei, K.; Zheng, S. Organic - Inorganic Polyu rethanes with Double Decker Silsesquioxanes in the Main Chains: Morphologies, Surface Hydrophobicity, and Shape Memory Properties. J . Polym. Sci. B Polym. Phys. 2018 , 56 , 893 906. (20) Laberty - Robert, C.; Vallé , K.; Pereira, F.; Sanchez, C. Design and Properties of Functional Hybrid Organic inorganic Membranes for Fuel Cells. Chem. Soc. Rev . 2011 , 40 , 961 1005. (21) Liu, N.; Wei, K.; Wang, L.; Zheng, S. Organic inorg anic Polyimides with Double Decker Silsesquioxane in the Main Chains. Polym. Chem. 2016 , 7 , 1158 1167. (22) Liu, N.; Li, L.; Wang, L.; Zheng, S. Organic - Inorganic Polybenzoxazine Copolymers with Double Decker S ilsesquioxanes in the Main Chains: Synthesis and Thermally Activated Ring - Opening Polymerization Behavior. Polymer 2017 , 109 , 254 2 65 . (23) Wei, K.; Wang, L.; Zheng, S. Organic inorganic Polyurethanes with 3,1 3 - Dihydroxypropyloctaphenyl Double - Decker Silsesquioxane Chain Extender. Polym. Chem. 2013 , 4 , 1491 1501. (24) Wu, S.; Hayakawa, T.; Kakimoto, M. - A.; Oikawa, H. Synthesis and Characterization of Organosoluble A romatic Polyimides Containing POSS in Main Chain Derived from Double - Decker - Shaped Silsesquioxane. Macromolecules 2008 , 41 , 3481 348 7. (25) Wu, S.; Hayakawa, T.; Kikuchi, R.; Grunzinger, S. J.; Kakimoto, M. - A.; Oikawa, H. Synthesis and Characterization of Semiaromatic Polyimides Containing POSS in Main 220 Chain Derived from Double - Decker - Shape d Silsesquioxane. Macromolecules 2007 , 40 , 5698 5705. (26) Jiang, Q.; Zhang, W Hydrophobic Fluoropolymer Film Derived from a Novel Double - Decker S tructural POSS with a Low Dielectric Constant. J. Mater. Chem. 2015 , 3 , 11729 11734. (27) Moore, L. M. J.; Zavala, J. J.; Lamb, J. T.; Reams, J. T.; Yandek, G. R.; Guenthner, A. J.; Haddad, T. S.; Ghiassi, K. B . Bis - Phenylethynyl Polyhedral Oligomeric Silsesquioxanes: New High - Temperature, Processable Thermosetting Materials. RSC Adv. 2018 , 8 , 27400 27405. (28) Seurer, B.; Vij, V.; Haddad, T.; Mabry, J. M.; Lee, A. T hermal Transitions and Reaction Kinetics of Polyhederal Silsesquioxane Containing Phenylethynylphthalimides. Macromolecule s 2010 , 43 , 9337 9347. (29) Crystallization : Basic Concepts and Industrial Applicatio ns; Beckmann, W., Ed.; 1st ed.; Wiley - VCH: Weinheim, 2011 . (30) Cisternas, L. A.; Rudd, D. F. Process Designs for Fractional Crystallization from Solution. Ind. Eng. Chem. Res. 1993 , 32 , 1993 2005. (31) Walas, S. M. 16 - CRYS TALLIZATION FROM SOLUTIONS AND MELTS. In Chemical Process Equipment ; Walas, S. M., Ed.; Butterworth - Heinemann: Boston, 1990 ; pp. 523 548. (32) Smolik , M . Chemical, Physicochemical and Crystal - Chemical Aspects o f Crystallization from Aqueous Solutions as a Method of Purification. In Crystallization - Science and Technology ; Andreeta, M., Ed.; InTech, 2012 . (33) Ahn, H. W.; Clarson, S. J. Synthesis and Characterization of Cis - and Trans - Trimethyltriphenylcyclotrisiloxane. J. Inorg. Organomet. Polym. 2001 , 11 , 203 216. (34) Snyder, L. R. Linear Elution Adsorption Ch rom atography: XII. Functional Grou Adsorption Energies on the Metal Oxide Adsorbents. J. Chromatogr. A 1966 , 23 , 388 402. (35) Snyder, L. R.; Dolan, J. W. Chapter 7 - Liquid Solid Chromatography. In Liquid Chro mat ography ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Schoenma kers, P.; Lloyd, D., Eds.; Elsevier: Amsterdam, 2013 ; pp. 143 156. (36) Snyder, L. R.; Kirkland, J. J.; Dolan, J. W. Introduction to Modern Liquid Ch rom atography ; 3rd ed.; Wiley: Hoboken, N.J., 2010 . (37) Khopkar, S. M. Basic Concepts of Analytical Chemistry ; New Academic Science: Kent, UNITED KINGDOM, 2011 . (38) I nternational Union of Pure; Chemistry, A. IUPAC Gold Book - partition chromatography http://dx.doi.org/10.1351/goldbook.P04436 (accessed Oct 13, 20 18 ). 221 (39) Ah uja, S. Chromatography and Separation Science ; Elsevier, 2003 . (40) Wingren, C.; Hansson, U. - B. CHROMATOGRAPHY: LIQUID | Partition Chromatography (Li qui d Liquid). In Encyclopedia of Separation Science ; Wilson, I. D., Ed.; Academic Press: Oxford, 2000 ; pp. 760 770. (41) Scott, R. P. W. The Role of Molecular Interactions in Chromatography. J. Chromatogr. A 19 76 , 122 , 35 53. (42) Szucs, R.; Brunelli, C.; Lestremau, F.; Hanna - Brown, M. Chapter 17 - Liquid Chromatography in the Pharmaceutical Industry. In Liquid Chromatography (Second Edition) ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 515 537. (43) - Environmental Analysis: Emerging Pollutants. In Liquid Chromat ography (Second Edition) ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 451 477. (44) Gika, H.; Theodoridis, G.; Rainville, P.; Plumb, R. S.; Wilson, I. D . C hapter 9 - Meta bolic Phenotyping (metabonomics/metabolomics) by Liquid Chromatography - Mass Spectrometry. In Liquid Chromatography (Second Edition) ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 245 265. (45) Gilbert - López, B.; Valdés, A.; Acunha, T.; García - Cañas, V.; Simó, C.; Cifuentes, A. Chapter 10 - Foodomics: LC and LC - MS - Based Omics Strategies in Food Science and Nutrition. In Liquid Chromatography (Second Edition) ; F anali, S.; Hadd ad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 267 299. (46) Donato, P.; Dugo, P.; Mondello, L. Chapter 8 - Separation of Lipids. In Liquid Chromatography (Second Edition) ; F anali, S.; Hadd ad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 201 243. (47) Royle, L. Chapter 7 - Separation of Glycans and Monosaccharides. In Liquid Chromatography (Second Edition) ; Fa nal i, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 183 200. (48) Koga, R.; Miyoshi, Y.; Todoroki, K.; Hamase, K. Chapter 4 - Amino Acid and Bioamine Separations. In Liquid Chromat ogr aphy (Second Ed ition) ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 87 106. (49) - Protein and Peptide Separations. In Liquid Chromat ography (Second Edition) ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 107 157. (50) Chankvetadze, B. Chapter 3 - Liquid Chromatographic Separation of Enant iomers. In Liqu id Chromatography (Second Edition) ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Riekkola, M. - L., Eds.; Elsevier, 2017 ; pp. 69 86. 222 (51) Panfili, G.; Fratianni, A.; Irano, M. Normal Phase High - Perform ance Liquid Chr omatography Method for the Determination of Tocopherols and Tocotrienols in Cereals. J. Agric. Food Chem. 2003 , 51 , 3940 3944. (52) Anumula, K. R.; Dhume, S. T. High Resolution and High Sensitivi ty Methods for Oligosaccharide Mapping and Characterization by Normal Phase High Performance Liquid Chromatography Following Derivatization with Highly Fluorescent An thranilic Acid. Glycobiology 1998 , 8 , 685 694. (53) Grossma nn, C.; Langel, C.; Mazzotti, M.; Morbidelli, M.; Morari, M. Multi - Rate Optimizing Control of Simulated Moving Beds. J. Process Control 2010 , 20 , 490 5 05. (54) Migliorini, C.; Mazzotti, M.; Morbi delli, M. Conti nuous Chromatographic Separation through Simulated Moving Beds under Linear and Nonlinear Conditions. J. Chromatogr. A 1998 , 827 , 161 173. (55) Wankat, P. C. Separation Process Engineering: Inclu des Mass Transfer Analysis, 3rd Edition ; Hall, P., Ed.; 2007 . (56) Still, W. C.; Kahn, M.; Mitra, A. Rapid Chromatograph ic Technique for Preparative Separations with Moderate Resolution. J. Org. Chem. 1978 , 43 , 2923 292 5. (57) Fornstedt, T.; Forssén, P.; Samuelsson, J. Chapter 18 - Modeling of Preparative Liquid Chromatography. In Liquid Chromatography ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Schoenmakers, P.; Lloyd , D., Eds .; Elsevier: Amsterdam, 2013 ; pp. 407 425. (58) ASPEN. Help File in ASPEN Chromatography V 10 ; 2017 . (59) Wang, Z.; Yonggang, L.; Huijuan, M.; Wenpeng, S. ; Tao, L.; Cuifen, L.; Junqi, N.; Guichun, Y.; Zuxing, C. Novel Phosphorus - Nitrogen - Silicon Copolymers with Double - Decker Silsesquioxane in the Main Chain and Their F lame Retardancy Application in PC/ABS. Fir e Mater. 2018 , 2 , 16230. (60) Hoque, M. A.; Kawakami, Y. Synthesis of Polysilsesquioxanes with Double - Decker Silsesquioxane Repeating Units. Journal of Scientific Research 2016 , 8 , 217 227. (61) c, B.; Dutkiewicz, M.; Ludwiczak, M.; Marciniec, B.; Nowicki, M. A New Class of Stereoregular Vinylene - Arylene Copolymers with Double - Decker Silsesquioxane in the Main Chain. J. Polym. Sci. Part A: Polym. Che m. 2016 , 54 , 1044 1055. (62) Seino, M.; Hayakawa, T.; Ishida, Y.; Kakimoto, M. - A.; Watanabe, K.; Oikawa, H. Hydrosilylation Polymerization of Double - Decker - Shaped Silsesquioxane Having Hydrosilane with Diynes. Macromolecules 2006 , 39 , 3473 3475. (63) Kucuk, A. C.; Matsui, J.; Miyashita, T. Langmuir - Blodgett Films Composed of Amphiphilic Double - Deck er Shaped Polyhedral Oligomeric Silsesquioxanes. J. Colloid Interfac e Sci. 2011 , 355 , 106 114. 223 (64) Díaz, U.; Brunel, D.; Corma, A. Catalysis Using Multifunctional Organosiliceous Hybrid Materials. Chem. Soc. Rev. 2013 , 42 , 4083 4097. (65) Espinas, J.; Pelletier, J. D. A.; Abou - Hamad, E.; Emsley, L.; Basset, J. - M. A Silica - Supported Double - Decker Silsesquioxane Provides a Second Skin for the Selective Generation of Bipodal Surface Organometallic Complexes. Organometallics 2012 , 31 , 7610 7617. (66) Laine, R. M.; Choi, J.; Lee, I. Organic -- Inorganic Nanocomposites with Completely Defined Interfacial Interactions. Adv. Mat er. 2001 , 13 , 800 803. (67) Wu, S.; Hayakawa, T.; Kikuchi, R.; Grunzi nger, S. J.; Kakimoto, M. - A.; Oikawa, H. Synthesis and Characterization of Semiaromatic Polyimides Containing POSS in Main Chain Derived from Double - Decker - Shaped Si lsesquioxane. Macromolecules 2007 , 40 (16), 5698 5705. (68) Synthesis and Characterization of Functionalized Molecular and Macr omolecular Double - Decker Silsesquioxane Systems. RSC Adv. 2016 , 6 (12), 10054 10063. (69) Liu, N.; Wei, K.; Wang, L.; Zheng, S. Organic inorganic Polyimides with Double Decker Silsesquioxane in the Main Chains . Polym. Chem. 2016 , 7 (5), 1158 1167. (70) Liu, N.; Li, L.; Wang, L.; Zheng, S. Organic - Inorganic Polybenzoxazine Copolymers with Double Decker Silsesquioxanes in the Main Chains: Synthesis and Thermally Acti vated Ring - Opening Polymerization Behavior. Polymer 2017 , 109 (Supplement C), 254 265. (71) Schoen, B. W. Aminophenyl Double Decker Silsesquioxanes: Spectroscopic Elucidation, Physical and Thermal Characteriz ation, and Their Applications ; 2013. (72) We i, K.; Wang, L.; Zheng, S. Organic inorganic Polyurethanes with 3,13 - Dihydroxypropyloctaphenyl Double - Decker Silsesquioxane Chain Extender. Polym. Chem. 2013 , 4 (5), 1491 1501. (73) Hoque, M. A.; Kawakami, Y. Synthesis of Pol ysilsesquioxanes with Double - Decker Silsesquioxane Repeating Units. Journal of Scientific Research 2016 , 8 (2), 217 227. (74) Hoque, M. A.; Kakihana, Y.; Shinke, S.; Kawakami, Y. Polysiloxanes with Periodicall y Distributed Isomeric Double - Decker Silsesquioxane in the Main Chain. Macromolecules 2009 , 42 (9), 3309 3315. (75) Schoen, B. W.; Lira, C. T.; Lee, A. Separation and Solubility of Cis and Tran s Isomers in Nan ostructured Double - Decker Silsequioxanes. J. Chem. Eng. Data 2014 , 59 (5), 1483 1493. (76) Schoen, B. W.; Holmes, D.; Lee, A. Identification and Quantification of Cis and Trans Isomers in Amino phenyl Double - De cker Silsesquioxanes Using 1H -- 29Si gHMBC NMR. Magn. Reson. Chem. 2013 , 51 (8), 490 496. 224 (77) Walczak, M.; Januszewski, R.; Majchrzak, M.; Kubicki, M.; Dudziec, B.; Marciniec, B. Unusual Cis an d Trans Architec ture of Dihydrofunctional Double - Decker Shaped - Conjugated Are ne Derivatives. New J. Chem. 2017 , 41 (9), 3290 3296. (78) Ervithayasuporn, V.; S odkhomkhum, R.; Teerawatananond, T.; Phurat, C.; Phinyocheep, P.; Somsook, E.; Osotchan, T. Unprecedented Formation of Cis - and Tr ans - Di [(3 - Chloropropyl) Isopropoxysilyl] - Bridged Double - Decker Octaphenylsilsesquioxanes. Eur. J. Inorg. Chem. 2013 , 2013 (19) , 3292 3296. (79) S e urer, B.; Vij, V.; Haddad, T.;, Mabry J.M.; Lee, A. Thermal transitions and reaction kinetics of Polyhedral Silse squioxanes containing Phenylethynylphtalimides. Macromolecules 2010 , 43, 9337 - 9347. (80) Lee, A. Separation and Phase Behavior of Double - Decker Silsesquioxane Isomers. In Proceedings of the 3rd World Congress o n Recent Advances in Nanotechnology ; World Congress on Recent Advances in Nanotechnology; Avestia Publishing, 2018. (81) Vogelsang, D. F.; Dannatt, J. E.; Maleczka, R. E.; Lee, A. Separation of Asymmetrically Capped Double - Decker Silsesquioxanes Mixtures. Polyhedron 2018 , 155 , 189 193. (82) Snyder, L. R.; Kirkland, J. J.; Dolan, J. W. Introduction to Modern Liquid Chromatography , 3rd ed.; Wiley: Hoboken, N.J., 2010. (83) Ghasemi, J.; Niazi, A. Spectrophotometric Simultaneous Determination of Ni troaniline Isome rs by Orthogonal Signal Correction - Partial Least Squares. Talanta 2005 , 65 (5), 1168 1173. (84) Howell, J. A.; Sutton, R . E. Ultraviolet and Absorption Light Spectrometry. Anal. Chem. 1998 , 70 ( 12), 107 118. (85) Scott, R. P. W. The Role of Molecular Interactions in Chromatography. J. Chromatogr. A 1976 , 122 , 35 53. (86) Liu, N.; Wei, K.; Wang, L.; Zheng, S. Organic inor ganic Polyimides with Double Decker Silsesquioxane in the Main Chains. Polym. Chem. 2016 , 7 (5), 1158 1167. (87) Liu, N.; Li, L.; Wang, L.; Zheng, S. Organic - Inorganic Polybenzoxazine Copolymers with Double Dec ker Silsesquioxanes in the Main Chains: Synthesis and Thermally Activated Ring - Opening Polymerization Behavior. Polymer 2017 , 109 ( Supplement C), 254 265. (88) Jiang, Q.; Zhang, W.; Hao, J.; We i, Y.; Mu, J.; J Hydrophobic Fluoropolymer Film Derived from a Novel Double - Decker Structural POSS with a Low Di electric Constant. J. Mater. Chem. 2015 , 3 (44), 11729 11734. (89) Wei, K.; Wa ng, L.; Zheng, S. Organic inorganic Polyurethanes with 3,13 - Dihydroxypropyloctaphenyl Double - Decker Silsesquioxane Chain Extender. Polym. Chem. 2013 , 4 (5), 1491 1501. 225 (90) Hoque, M. A.; Kakihan a, Y.; Shinke, S .; Kawakami, Y. Polysiloxanes with Periodically Distributed Isomeric Double - Decker Silsesquioxane in the Main Chain. Macromolecules 2009 , 42 (9), 3309 3315. (91) Vij, V.; Haddad, T. S.; Yandek, G. R.; Ramirez, S. M.; Mabry, J. M. Synthesis of Aromatic Polyhedral Oligomeric SilSesquioxane (POSS) Dianilines for Use in High - Temperature Polyim ides. Silicon Chem. 2012 , 4 (4), 267 280. (92) Hoque, M. A.; Ka wakami, Y. Synth esis of Polysilsesquioxanes with Double - Decker Silsesquioxane Repeating Units. Journal of Scientific Research 2016 , 8 (2), 217 227. (93) Schoen, B. W.; Lira, C. T.; Lee, A. Separation and Solubi lity of Cis and Trans Isomers in Nanostructured Double - Decker Silsequioxanes. J. Chem. Eng. Data 2014 , 59 (5), 1483 1493. (94) Walczak, M.; Januszewski, R.; Majchrzak, M.; Kubicki, M.; Dudziec, B.; Marciniec, B . Unusual Cis an d Trans Architecture of Dihydrofunctional Double - Decker Shaped - Conjugated Arene Derivatives. New J. Chem. 2017 , 41 (9), 3290 3296. (95) Crystallization : Basi c Concepts and Industrial Applications , 1st ed.; Beckmann, W., Ed.; Wiley - VCH: Weinheim, 2011. (96) Snyder, L. R.; Kirkland, J. J.; Dolan, J. W. Introduction to Modern Liquid Chromatography , 3rd ed.; Wiley: Hob oken, N.J., 2010. (97) Migliorini, C.; Mazzotti, M.; Morbidelli, M. Continuous Chroma tographic Separation through Simulated Moving Beds under Linear and Nonlinear Conditions. J. Chromatogr. A 1998 , 827 (2), 161 173. (98) Grossmann, C.; Langel, C.; Mazzotti, M.; Morbidelli, M.; Morari, M. Multi - Rate Optimizing Control of Simulated Moving Beds. J. Process Control 2010 , 20 (4), 490 505. (99) Ahuja, S. Chromatography and Separation Science ; Elsevier, 2003. (100) Kamarei, F.; Gritti, F.; Guiochon, G.; Burchell, J. Accurate Measurements of Frontal Analysis for the Determination of Ads orption Isotherms in Supercritical Fluid Chromatography. J. Chromatogr. A 2014 , 1329 , 71 77. (101) Panfili, G.; Fratianni, A.; Irano, M. Normal Phase High - Performance Liquid Chromatography Method for the Determi nation of Tocopherols and Tocotrienols in Cereals. J. Agric. Food Chem. 2003 , 51 (14) , 3940 3944. (102) Anumula, K. R.; Dhume, S. T. High Resolution and High Sensitivity Methods for Oligosaccharide Mapping and C haracterization by Normal Phase High Performance Liquid Chromatography Following Deri vatization with Highly Fluorescent Anthranilic Acid. Glycobiology 1998 , 8 (7), 685 694. 226 (103) Glinski, J. A.; Davey, M. H.; Ka ne, R. E.; Glinski, V. B. Quantification of Activin® Grape Seed Extracts for Procyani din Composition with Reverse Phase and Normal Phase HPLC. Planta Med. 2008 , 74 (09), SL39. (104) Rao, D. B. S. METHOD DEVELOPM ENT AND VALIDATION OF REVERSE PHASE HPLC FOR PREGABALIN IN PHARMACEUTICAL DOSAGE BY REVERSE PHASE HPLC. WJPPS 2017 , 2001 2012. (105) Narine, S. S.; Yue, J.; Kong, X. Production of Po lyols from Canola Oil and The ir Chemical Identification and Physical Properties. J. Am. Oil Chem. Soc. 2007 , 84 (2), 173 179. (106) Andrews, G. D.; Vatvars, A. HPLC Analysis of Polymers. Separation of Hydroxy - Fu nctional Poly(methyl Methacry late) by Number of Hydroxyls and Separation of Poly(methyl Methacrylate) Oligomers. Macromolecules 1981 , 14 (5), 1603 1605. (107) Ootake, N.; Yoshida, K. Organic Silicon Compound and Method for Producing Same, a nd Polysiloxane and Method for Producing Same. 20060155091:A1, July 13, 2006. (108) Morimoto, Y.; Watanabe, K.; Ootake, N.; Inagaki, J. - I.; Yoshida, K.; Ohguma, K. Silsesquioxane Derivatives and Process for Prod uction Thereof. 20040249103A1, December 9, 2004. (109) Lee, D. W.; Kawakami, Y. I ncompletely Condensed Silsesquioxanes: Formation and Reactivity. Polym. J. 2007 , 39 , 230 238. (110) - Functional Double - Decke r Silsesquioxanes as Anchors for Reactive Functional Groups and Potential Synthons for Hybrid Materials. Chem. Commun. 2017 , 53 , 10370 10373. (111) Schoen, B. W. Aminophenyl Double Decker Silsesquioxanes: Spect roscopic Elucidation, Physical and Thermal Characterization, and Their Applications ; 2013. (112) Sch oen, B. W.; Holmes, D.; Lee, A. Identification and Quantification of Cis and Trans Isomers in Aminophenyl Doubl e - Decker Silsesquioxanes Using 1H -- 29Si gHMBC NMR. Magn. Reson. Chem. 2013 , 51 , 490 496. (113) Schoe n, B. W.; Lira, C. T.; Lee, A. Separation and Solubility of Cis and Trans Isomers in Nanostructured Double - Deck er Silsesquioxanes. J. Chem. Eng. Data 2014, 59, 1483 1493. (114) Seurer, B.; Vij, V.; Haddad, T.; M abry, J. M.; Lee, A. Thermal Transitions and Reaction Kinetics of Polyhedral Silsesquioxane Containing Phenylet hynylphthalimides. Macromolecules 2010, 43, 9337 9347. (115) Kohri, M.; Matsui, J.; Watanabe, A.; Mi yashita, T. Synthesis and Optoelectronic Properties of Completely Carbazole - Substituted Double - Decker - Shaped Si lsesquioxane. Chem. Lett. 2010 , 39 , 1162 1163. 227 (116) Xu, S.; Zhao, B.; Wei, K.; Zheng, S. Organic - In organic Polyurethanes with Double Decker Silsesquioxanes in the Main Chains: Morphologies, Surface Hydrophobicity, and Shape Memory Properties. J. Polym. Sci. B Polym. Phys. 2018 , 56 , 893 906. (117) Wang, Z.; Yo nggang, L.; Huijuan, M.; Wenpeng, S.; Tao, L.; Cuifen, L.; Junqi, N.; Guichun, Y.; Zuxing, C. Novel Phosphorus - Nitrogen - Silicon Copolymers with Double - Decker Silsesquioxane in the Main Chain and Their Flame Re tardancy Application in PC/ABS. Fire Mater. 201 8 , 2 , 16230. (118) Liu, N.; Wei, K.; Wang, L.; Zheng, S. Organic - inorganic Polyimides with Double Decker Silsesquioxane in the Main Chains. Polym. Chem. 2016, 7, 11 58 1167. (119) Wei, K.; Wang, L.; Zheng, S. Organic - inorganic Polyurethanes with 3,13 - Dihydroxypropyloctaphenyl Double - Decker Silsesquioxane Chain Extender. Polym. Chem. 2013, 4, 1491 1501. (120) Hoque, M. A.; Kak ihana, Y.; Shinke , S.; Kawakami, Y. Polysiloxanes with Periodically Distributed Isomeric Double - Decker Silsesquioxane in the Main Chain. Macromolecules 2009 , 42 , 3309 3315. (121) Ervithayasuporn, V.; Sodkhomkhum , R.; Teerawatananond, T.; Phurat, C.; Phinyocheep, P.; Somsook, E.; Osotchan, T. Unprecedented Formation of Cis - and Trans - Di [(3 - Chloropropyl) Isopropoxysilyl] - Bridged Double - Decker Octaphenylsilsesquioxanes. Eur. J. Inorg. Chem. 2013 , 2013 , 3292 3296. (122) Walczak, M.; Januszewski, R.; Majchrzak, M.; Kubicki, M.; Dudziec, B.; Marciniec, B. Unusual Cis and Trans Architecture of Dihydrofunctional Double - Decker Shape d Silsesquioxane and Synthesis of Its Ethyl Bri - Conjugated Arene Derivatives. New J. Chem. 2017 , 41 , 3290 3296. (123) Kucuk, A. C.; Matsui, J.; Miyashita, T. Langmuir - Blodgett Films Composed of Amphiphilic Double - Decker Shaped Polyhedral Oligomeric Sil sesquioxanes. J. Colloid Interface Sci. 2011 , 355 , 106 114. (124) Kucuk, A. C .; Matsui, J.; Miyashita, T. Synthesis and Photochemical Response of Ru( Ii ) - Coordinated Double - Decker Silsesquioxane. RSC Adv. 2018 , 8 , 2148 2156. (125) Espinas, J.; Pelletier, J. D. A.; Abou - Hamad, E.; Emsley , L.; Basset, J. - M. A Silica - Supported Double - Decker Silsesquioxane Provides a Second Skin for the Selective Generation of Bipodal Sur face Organometallic Complexes. Organometallics 2012 , 31 , 7610 7617. (126) , P.; Dudziec, B.; Dutkiewicz, M.; Ludwiczak, M.; Marciniec, B.; Nowicki, M. A New Class of Stereoregular Vinylene - Arylene Copolymers with Double - Decker Silsesquioxane in the Main Chain. J. Polym. Sci. Part A: Polym. Chem. 2016 , 54 , 1044 1055. 228 (127) Liu, N.; Li, L.; Wang, L.; Zheng, S. Organic - Inorganic Polybenzoxazine Copolymers with Double D ecker Silsesquioxanes in the Main Chains: Synthesis and Thermally Activated Ring - Opening Polymerization Behavior . Polymer 2017 , 109 , 254 265. (128) Wu, S.; Hayakawa, T.; Kikuchi, R.; Grunzinger, S. J.; Kakimoto , M. - A.; Oikawa, H. Synthesis and Characterization of Semiaromatic Polyimides Containing POSS in Main Chain Deri ved from Double - Decker - Shaped Silsesquioxane. Macromolecules 2007 , 40 , 5698 5705. (129) Hoque, M. A .; Kawakami, Y. Synthesis of Polysilsesquioxanes with Double - Decker Silsesquioxane Repeating Units. Journal of Scientific Research 2016 , 8 , 217 227. (130) Wu, S.; Hayakawa, T.; Kakimoto, M. - A.; Oikawa , H. Synthe sis and Characterization of Organosoluble Aromatic Polyimides Containing POSS in Main Chain Derived from Double - Decker - Shaped Silsesquioxane. Macromolecules 2008 , 41 , 3481 3487. (131) Vogelsang, D. F. ; Dannatt, J. E.; Maleczka, R. E.; Lee, A. Separation of Asymmetrically Capped Double - Decker Silsesquioxanes Mixtures. Polyhedron 2018 , 155 , 189 193. (132) Fornstedt, T.; Forssén, P.; Samuelsson, J. Chapter 18 - Modeling o f Preparative Liquid Chromatography. In Liquid Chromatography ; Fanali, S.; Haddad, P. R.; Poole, C. F.; Schoenmakers, P.; Lloyd, D., Eds.; Elsevier: Amsterdam, 2013; pp. 407 425. (133) Snyder, L. R.; Kirkland, J . J.; Dolan, J. W. Introduction to Modern Liquid Chromatography ; 3rd ed.; Wiley: Hoboken, N.J., 2010. (134) Gritti , F.; Guiochon, G. Systematic Errors in the Measurement of Adsorption Isotherms by Fro ntal Analys is Impact of the Choice of Column Hold - up Volume, Range and Density of the Data Points. J. Chromatogr. A 2005 , 1097 , 98 115. (135) Manual, I. Agilent ChemStation for LC and CE Systems. (136) ASPEN. Help File in ASPEN Chromatography V 10 ; 2017 . (137) Vogelsang, D.; Maleczka, R.; Lee, A. HPLC Characterization of cis and trans mixtures of Double - Decker Shaped Silsesquioxanes. Silicon J. Submitted 2018 . (138) Phillips, S. H.; Haddad, T. S.; Tomczak, S. J. Developments in Nanoscience: Polyhedral Oligomeric Silsesquioxane (POSS) - Polymers. Curr. Opin. Solid State Mater. Sci. 2004 , 8 (1), 21 29. https://doi.org/10.1016/j.cossms.2004.03.002. (139) L i, G.; Wang, L.; Ni, H.; Pittman, C. U. Polyhed ral Oligomeric Silsesquioxane (POSS) Polymers and Copolymers: A Review. J. Inorg. Organomet. Polym. 2001 , 11 (3), 123 154. https://doi.org/10.1023/A: 1015287910502. (140) Feher, F. J.; Phillips, S. H.; Ziller, J. W. Facile and Remarkably Selective Substitution Re actions Involving Framework Silicon Atoms in Silsesquioxane Frameworks. J. Am. Chem. Soc. 1997 , 119 (14), 3397 3398. https://doi.org/10.1021/ja963904 m. 229 (141) Feher, F. J.; Newman, D. A.; Walzer, J. F. Silsesquioxanes as Models for Silica Surfaces. J. Am. Chem. S oc. 1989 , 111 (5), 1741 1748. https://doi.org/10.1021/ja00187a028. (142) Feher, F. J.; Walzer, J. F. Synthesis and Characterization of Vanadium - Containing Silsesquioxanes. Inorg. Chem. 1991 , 30 (8), 1689 1694. https://doi.org/10.1021/ic00008a005. (143) Morimoto, Y.; W atanabe, K.; Ootake, N.; Inagaki, J. - I.; Yoshida, K.; Ohguma, K. Silsesquioxane Derivatives and Process for Production Thereof. 20040249103A1, Decemb er 9, 2004. (144) Lee, D. W.; Kawakami, Y. Incompletely Condensed Silsesquioxanes: Formation and Reactivity. Poly m. J. 2007 , 39 (3), 230 238. https://doi.org/10.1295/polymj.PJ2006169. (145) Hoque, M. A.; Kakihana, Y.; Shinke, S.; Kawakami, Y. Polysiloxanes with Period ically Distributed Isomeric Double - Decker Silsesquioxane in the Main Chain. Macromolecules 2009 , 42 (9), 33 09 3315. https://doi.org/10.1021/ma900124x. (146) Kawabata, K.; Tajima, A.; Matsuo, T.; Watanabe, Y.; Ayama, K. Compound Including Organopolysiloxane or Si lsesquioxane Skeleton Having Isocyanuric Skeleton, Epoxy Group and SiH Group, Thermosetting Resin Compositi on Containing the Compound as Adhesion - Imparting Agent, Hardened Material and Sealing Agent for Optical Semiconductor. U.S. Patent 20140148536:A1, Ma y 29, 2014. (147) Au - Yeung, H. - L.; Leung, S. Y. - L.; Yam, V. W. - W. Supramolecular Assemblies of Dinuclear Alkynylp latinum(II) Terpyridine Complexes with Double - Decker Silsesquioxane Nano - Cores: The Role of Isomerism in Constructing Nano - Structures. Chem. Commun. 2018 , 54 (3 3), 4128 4131. https://doi.org/ 10.1039/c8cc00557e. (148) Wu, S.; Hayakawa, T.; Kakimoto, M. - A.; Oikaw a, H. Synthesis and Characterization of Organosoluble Aromatic Polyimides Containing POSS in Main Chain Derived from Double - Decker - Shaped Silsesquioxane. Macromolecules 2008 , 41 (10), 3481 3487. https://doi.org/10.1021/ma7027227. (149) Kucuk, A. C.; Matsui, J.; Miyashita, T. Langmuir - Blodgett Films Composed of Amphiphilic Double - Decker Shaped Polyhedral Oligomeric Silsesquioxanes. J. Colloid Interface Sci. 2011 , 355 (1), 106 114. https://doi.org/10.1016/j.jcis.2010.12.033. (150) Wu, S.; Hayakawa, T.; Kikuchi, R.; Grunz inger, S. J.; Kakimoto, M. - A.; Oikawa, H. Synthesis and Characterization of Semiaromatic Polyimides Containing POSS in Main Chain Derived from Double - Deck er - Shaped Silsesquioxane. Macromolecules 2007 , 40 (16), 5698 5705. https://doi.org/10.1021/ma070547z. (151) Hydrophobic Fluoropolymer Film Derived from a Novel Double - Decker St ructural POSS with a Low Dielectric Constant. J. Mater. Chem. 2015 , 3 (44), 11729 11734. https://doi.o rg/10.1039/C5TC02432C. 230 (152) Liu, N.; Li, L.; Wang, L.; Zheng, S. Organic - Inorganic Polybenzoxazine Copolymers with Double Decker Silsesquioxanes in the Main Ch ains: Synthesis and Thermally Activated Ring - Opening Polymerization Behavior. Polymer 2017 , 109 (Suppl ement C), 254 265. https://doi.org/10.1016/j.polymer.2016.12.049. (153) Hoque, M. A.; Kawakami, Y. Synthesis of Polysilsesquioxanes with Double - Decker Silsesqui oxane Repeating Units. Journal of Scientific Research 2016 , 8 (2), 217 227. https://doi.org/10.3329/js r.v8i2.26791. (154) Synthesis and Characterization of Functionalized Molecular and Macromolecular Double - Decker Silsesquioxane Systems. RSC Adv. 2016 , 6 (12), 10054 10063. https://doi.org/10.103 9/C5RA20848C. (155) Haddad, T. S.; Viers, B. D.; Phillips, S. H. Polyhedral Oligomeric Silsesquioxane (POSS) - Styrene Macromers. J. Inorg. Organome t. Polym. 2001 , 11 (3), 155 164. https://doi.org/10.1023/A:1015237627340. (156) Leu, C. - M.; Chang, Y. - T.; Wei, K. - H. Polyi mide - Side - Chain Tethered Polyhedral Oligomeric Silsesquioxane Nanocomposites for Low - Dielectric Film Applications. Chem. Mater. 2003 , 15 (19 ), 3721 3727. https://doi.org/10.1021/cm030393b. (157) Lichtenhan, J. D.; Schwab, J. J.; An, Y.; Liu, Q.; Haddad, T. S. Pr ocess for the Functionalization of Polyhedral Oligomeric Silsesquioxanes. U.S. Patent 6927270, August 9, 2005. (158) Li, Y.; Dong, X. - H.; Zou, Y.; Wang, Z.; Yue, K.; Huang, M.; Liu, H.; Feng, X.; Lin, Z.; Zhang, W.; et al. Polyhedral Oligomeric Silsesquioxane Me Design and Facile Preparation of Functional Hybrid Materials. Polymer 2017 , 125 , 303 329. https://doi.org/10.1016/j.polymer.2017.08.008. (159) Cao, J.; Fan, H.; Li, B. - G.; Zhu, S. Synthesis and Evaluation of Double - Decker Silsesqu ioxanes as Modifying Agent for Epoxy Resin. Polymer 2017 , 124 , 157 167. https://doi.org/10.1016/j.polymer.2017.07.056. (160) Xu, S.; Zhao, B.; Wei, K.; Zheng, S. Organic - Inorganic Polyurethanes with Double Decker Silsesquioxanes in the Main Chains: Morphologies, Surface Hydrophobicity, and Shape Memory Properties. J. Polym. Sci. B Polym. Phys. 2018 , 56 (12), 893 906. https://doi.or g/10.1002/polb.24603. (161) Haddad, T. S.; Mabry, J. M.; Ramirez, S. Synthesis of Functional Fluorinated Polyhedral Oligomeric Silsesquioxan e (F - POSS). U.S. Patent 9249313, February 2, 2016. (162) Lee, A.; Lu, Y.; Roche, A.; Pan, T. - Y. Influence of Nano - Structured Sil anols on the Microstructure and Mechanical Properties of A4047 and A359 Aluminum Casting Alloys. Int. J. Metalcast 2016 , 10 (3), 338 3 41. https://doi.org/10.1007/s40962 - 016 - 0044 - 4. 231 (163) Blanco, I.; Abate, L.; Bottino, F. A.; Bottino, P. Thermal Degradation of H epta Cyclopentyl, Mono Phenyl - Polyhedral Oligomeric Silsesquioxane (hcp - POSS)/polystyrene (PS) Nanocomposites. Polym. Degrad. Stab. 20 12 , 97 (6), 849 855. https://doi.org/10.1016/j.polymdegradstab.2012.03.041. (164) Jin, L.; Ishida, H. New Thermo - Oxidative Prote ctive Coating for Carbon/carbon Composites in Mid - Temperature Range Using a Combination of Nano - Filler and Macro - Filler with Polybenzo xazine as a Carbon - Forming Matrix. Polym. Compos. 2011 , 32 (8), 1164 1173. https://doi.org/10.1002/pc.21135. (165) Vahabi, H.; E terradossi, O.; Ferry, L.; Longuet, C.; Sonnier, R.; Lopez - Cuesta, J. - M. Polycarbonate Nanocomposite with Improved Fire Behavior, Phys ical and Psychophysical Transparency. Eur. Polym. J. 2013 , 49 (2), 319 327. https://doi.org/10.1016/j.eurpolymj.2012.10.03 1. (166) Ke, F.; Zhang, C.; Guang, S.; Xu, H. POSS Core Star - Shape Molecular Hybrid Materials: Effect of the Chain Length and POSS Content o n Dielectric Properties. J. Appl. Polym. Sci. 2013 , 127 (4), 2628 2634. (167) Ramirez, N. V.; Sanchez - Soto, M. Effects of POSS N anoparticles on ABS - G - Ma Thermo Oxidation Resistance. Polym. Compos. 2012 , 33 (10), 1707 1718. (168) Wang, Z.; Yonggang, L.; Huijuan, M.; W enpeng, S.; Tao, L.; Cuifen, L.; Junqi, N.; Guichun, Y.; Zuxing, C. Novel Phosphorus - Nitrogen - Silicon Copolymers with Doub le - Decker Silsesquioxane in the Main Chain and Their Flame Retardancy Application in PC/ABS. Fire Mater. 2018 , 2 , 16230. https://doi.o rg/10.1002/fam.2649. (169) Seurer, B.; Vij, V.; Haddad, T.; Mabry, J. M.; Lee, A. Thermal Transitions and Reaction Kinetics of Polyhederal Silsesquioxane Containing Phenylethynylphthalimides. Macromolecules 2010 , 43 (22), 9337 9347. https://doi.org/10.1021/ma101 640q. (170) Liu, N.; Wei, K .; Wang, L.; Zheng, S. Organic inorganic Polyimides with Double Decker Silsesquioxane in the Main Chains. Polym. Chem. 2016 , 7 (5), 1158 1167. https://doi.org/10.1039/C5PY01827G. (171) Cardiano, P.; Lazzara, G.; Manickam, S.; Mineo, P.; Mili oto, S.; Lo Schiavo, S. POSS Tetraalkylammonium Salts: A New Class of Ionic Liquids. Eur. J. Inorg. Chem. 2012 , 2012 (34), 5668 5676. https://doi.org/10.1002/ejic.201200874. (172) Li, H.; Zhao, X.; Chu, G.; Zhang, S.; Yuan, X. One - Step Fabrication of a Superhydr ophobic Polymer Surfa ce from an Acrylic Copolymer Containing POSS by Spraying. RSC Adv. 2014 , 4 (107), 62694 62697. https://doi.org/10.1039/C4RA07113A. (173) Xue, Y.; Wang, H.; Yu, D.; Feng, L.; Dai, L.; Wang, X.; Lin, T. Superhydrophobic Electrospun POSS - PMMA C opolymer Fibres with Highly Ordered Nanofibrillar and Surface Structures. Chem. Commun. 2009 , (42), 6418 6420. https://doi.org/10.1039/b911509a. 232 (174) Kucuk, A. C.; Matsui, J.; Miyashita, T. Synthesis and Photochemical Response of Ru(II) - Coordinated Double - Dec ker Silsesquioxane. RSC Adv. 2018 , 8 (4), 2148 2156. https://doi.org/ 10.1039/C7RA12290J. (175) Espinas, J.; Pelletier, J. D. A.; Abou - Hamad, E.; Emsley, L.; Basset, J. - M. A Silica - Supported Double - Decker Silsesquioxane Provides a Second Skin for the Selective Ge neration of Bipodal Surface Organometallic Complexes. Organometallics 2012 , 31 (21), 7610 7617. https://doi.org/10.1021/om300918v. (176) Hartmann - Thompso n, C. Applications of Polyhedral Oligomeric Silsesquioxanes ; Springer Science & Business Media, 2011. (177) He, F. - A.; Zhang, L. - M. Using Inorganic POSS - Modified Laponite Clay to Support a - Diimine Catalyst for in Situ Formation of High Performance Polye thylene Nanocomposites. Nanotechnology 2006 , 17 (24), 5941. https://doi.org/10.1088/0957 - 4484/17/24/007. (178) Tana ka, K.; Ishiguro, F.; Chujo, Y. POSS Ionic Liquid. J. Am. Chem. Soc. 2010 , 132 (50), 17649 17651. https://doi.org/10.1021/ja105631j. (179) Tanaka, K.; Ishiguro, F.; Jeon, J. - H.; Hiraoka, T.; Chujo, Y. POSS Ionic Liquid Crystals. Npg Asia Materials 2015 , 7 , e174. https://doi.org/10.1038/am.2015.28. (180) Tanaka, K.; Ishiguro, F.; Chujo, Y. Therm odynamic Study of POSS - Based Ionic Liquids with Various Numbers of Ion Pairs. Polym. J. 2011 , 43 , 708. https://doi.org/10.1038/pj.2011.54. (181) Zhao, R.; Chen, Y.; Liu, G.; Jiang, Y.; Chen, K. Fabrication of Self - Healing Waterbased Superhydrophobic Coatings fro m POSS Modified Silica Nanoparticles. Mater. Lett. 2018 , 229 , 281 285. https://doi.org/10.1016/j.matlet.2018.07.040. (182) Rizvi, S. B.; Yildirimer, L.; Ghaderi, S.; Ramesh, B.; Seifal ian, A. M.; Keshtgar, M. A Novel POSS - Coated Quantum Dot for Biological Applic ation. Int. J. Nanomedicine 2012 , 7 , 3915 3927. https://doi.org/10.2147/IJN.S28577. (183) Ganesh, V. A.; Nair, A. S.; Raut, H. K.; Tan, T. T. Y.; He, C.; Ramakrishna, S.; Xu, J. Superh ydrophobic Fluorinated POSS - PVDF - HFP Nanocomposite Coating on Glass by Electro spinning. J. Mater. Chem. 2012 , 22 (35), 18479 18485. (184) Wang, D. K.; Varanasi, S.; Strounina, E.; Hill, D. J. T.; Symons, A. L.; Whittaker, A. K.; Rasoul, F. Synthesis and Charact erization of a POSS - PEG Macromonomer and POSS - PEG - PLA Hydrogels for Periodonta l Applications. Biomacromolecules 2014 , 15 (2), 666 679. https://doi.org/10.1021/bm401728p. (185) Tanaka, K.; Ohashi, W.; Kitamura, N.; Chujo, Y. Reductive Glutathione - Responsive Molec ular Release Using Water - Soluble POSS Network Polymers. BCSJ 2011 , 84 (6), 612 616. https://doi.org/10.1246/bcsj.20110032. 233 (186) Kolel - Veetil, M. K.; Fears, K. P.; Qadri, S. B.; Klug, C. A.; Keller, T. M. Formation of a Crosslinked POSS Network by an Unusual Hyd rosilylation: Thermo - Oxidative - Cristobalite Phase in It s Amorphous Regions. J. Polym. Sci. A Polym. Chem. 2012 , 50 (15), 3158 3170. (187) Bagheri, H.; Soofi, G.; Javanmardi, H.; Karimi, M. A 3D Nanoscale Polyhedral Oligomeric Silsesquioxan es Network for Microextraction of Polycyclic Aromatic Hydrocarbons. Mikrochim. Acta 2018 , 185 (9), 418. https://doi.org/10.1007/s00604 - 018 - 2950 - z. (188) Ghanbari, H.; Cousins, B. G.; Seifalian, A. M. A Nanocage for Nanomedicine: Polyhedral Oligomeric Silsesquiox ane (POSS). Macromol. Rapid Commun. 2011 , 32 (14), 1032 1046. https://doi.org/10.1002/marc.201100126. (189) Tan, A.; Farhatnia, Y.; Seifalian, A. M. Polyhedral Oligomeric Silsesquioxane Poly (carbonate - Urea) Urethane (POSS - PCU): Applications in Nanotechnology an d Regenerative Medicine. Crit. Rev. Biomed. Eng. 2013 , 41 (6), 495 513. (190) Olivero, F.; Renò, F.; Carniato, F.; Rizzi, M.; Cannas, M.; Marchese, L. A Novel Luminescent Bifunctional POSS as a Molecular Platform for Biomedical Applications. Dalton Trans. 2012 , 41 (25), 7467 7473. https://doi.org/10.1039/ C2DT30218G. (191) Pu, Y.; Zhang, L.; Zheng, H.; He, B.; Gu, Z. Drug Release of pH - Sensitive poly(L - Aspartate) - B - Poly(ethylene Glycol) Micelles with POSS Cores. Polym. Chem. 2013 , 5 (2), 463 470. https://doi.org/10.1039 /C3PY00965C. (192) Schoen, B. W. Aminophenyl Doubl e Decker Silsesquioxanes: Spectroscopic Elucidation, Physical and Thermal Characterization, and Their Applications ; Ph.D. Thesis, Michigan State University, East Lansing, MI, 2013. (193) Guo, T.; Wang, B. Isothermal Co ld Crystallization and Melting Behaviors of Poly(lactic acid)/Epoxy Vinyl Polyhedral Oligomeric Silsesquioxanes Nanocomposites. Polym. Plast. Technol. Eng. 2014 , 53 (9), 917 926. https://doi.org/10.1080/03602559.2014.8 86061. (194) Li, Y.; Morgan, R. J. Thermal Cure of Phenylethynyl - Terminated AFR - PEPA - 4 Imide Oligomer and a Model Compound. J. Appl. Polym. Sci. 2006 , 101 (6), 4446 4453. https://doi.org/10.1002/app.24047. (195) Schoen, B. W.; Holmes, D.; Lee, A. Identification and Q uantific ation of Cis and Trans Isomers in Aminophenyl Double - Decker Silsesquioxane s Using 1H -- 29Si gHMBC NMR. Magn. Reson. Chem. 2013 , 51 (8), 490 496. (196) Walczak, M.; Januszewski, R.; Majchrzak, M.; Kubicki, M.; Dudziec, B.; Marciniec, B. Unusual Cis and Tr ans Arch itecture of Dihydrofunctional Double - Decker Shaped Silsesquioxane and Synt - Conjugated Arene Derivatives. New J. Chem. 2017 , 41 (9), 3290 3296. https://doi.org/10.1039/C7NJ00255F. 234 (197) Vogelsang, D. F.; Maleczka, R. E.; Lee, A . HPLC C haracterization of Cis and Trans Mixtures of Double - Decker Shaped Silsesqu ioxanes. Silicon 2018 . https://doi.org/10.1007/s12633 - 018 - 0045 - 4. (198) Attanayake, G. K. Study of Different Routes to Develop Asymmetric Double Decker Silsesquioxane (DDSQ) ; Maste is, Michigan State University, East Lansing, MI, 2015. (199) Schoen, B. W.; Lira, C. T.; Lee, A. Separation and Solubility of Cis and Trans Isomers in Nanostructured Double - Decker Silsequioxanes. J. Chem. Eng. Data 2014 , 59 (5), 1483 1493. https://doi.or g/10.102 1/je4010245. (200) Ervithayasuporn, V.; Sodkhomkhum, R.; Teer awatananond, T.; Phurat, C.; Phinyocheep, P.; Somsook, E.; Osotchan, T. Unprecedented Formation of Cis - and Trans - Di [(3 - Chloropropyl) Isopropoxysilyl] - Bridged Double - Decker Octaphenylsilsesquio xanes. E ur. J. Inorg. Chem. 2013 , 2013 (19), 3292 3296. (201) Moore, L. M. J.; Zavala, J. J.; Lamb, J. T.; Reams, J. T.; Yandek, G. R.; Guenthner, A. J.; Haddad, T. S.; Ghiassi, K. B. Bis - Phenylethynyl Polyhedral Oligomeric Silsesquioxanes: New High - Temperature, Process able Thermosetting Materials. RSC Adv. 2018 , 8 ( 48), 27400 27405. https://doi.org/10.1039/C8RA05954C. (202) Thogiti, S.; Parvathaneni, S. P.; Keesara, S. Polymer Anchored 3 - Benzoyl - 1 - (1 - Benzylpiperidin - 4 - Yl) - 2 - Thiopseudourea - Pd(II) Complex: An Efficient Catalyst for the Copper and Solvent Free Sonogashira Cro ss - Coupling Reaction. J. Organomet. Chem. 2016 , 822 , 165 172. https://doi.org/10.1016/j.jorganchem.2016.08.029. (203) Vogelsang, D. F.; Dannatt, J. E.; Maleczka, R. E., Jr.; Lee, A. Separation of Asymmetric ally Cap ped Double - Decker Silsesquioxanes Mixtures. Poly hedron 2018 , 155 , 189 193. https://doi.org/10.1016/j.poly.2018.08.016. (204) Vij, V.; Haddad, T. S.; Yandek, G. R.; Ramirez, S. M.; Mabry, J. M. Synthesis of Aromatic Polyhedral Oligomeric SilSesquioxane (P OSS) Dia nilines for Use in High - Temperature Polyimides S ilicon 2012 , 4 , 267 280. https://doi.org/10.1007/s12633 - 012 - 9130 - 2. (205) van Laar, J. J. Die Schmelz - Oder Erstarrungskurven Bei Binären Systemen, Wenn Die Feste Phase Ein Gemisch (amorphe Feste Lösung Ode r Mischk ristalle) Der Beiden Komponenten Ist. Zeitschrif t für Physikalische Chemie 1908 , 63U (1). https://doi.org/10.1515/zpch - 1908 - 6314 . (206) Boettinger, W. J.; Kattner, U. R.; Moon, K. - W.; Perepezko, J. N IST Reco mmended Practice Guide: DTA and Heat - Flux DSC Measurements of Alloy Melting and Freezing. 2006 .