UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies of the nitration of cellulose - application in new membrane materials Cheung, Clement 2014

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

Item Metadata

Download

Media
24-ubc_2014_spring_cheung_clement.pdf [ 1.63MB ]
Metadata
JSON: 24-1.0072160.json
JSON-LD: 24-1.0072160-ld.json
RDF/XML (Pretty): 24-1.0072160-rdf.xml
RDF/JSON: 24-1.0072160-rdf.json
Turtle: 24-1.0072160-turtle.txt
N-Triples: 24-1.0072160-rdf-ntriples.txt
Original Record: 24-1.0072160-source.json
Full Text
24-1.0072160-fulltext.txt
Citation
24-1.0072160.ris

Full Text

STUDIES OF THE NITRATION OF CELLULOSE ? APPLICATION IN NEW MEMBRANE MATERIALS    by    Clement Cheung  B. Sc., University of British Columbia, 2011      A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE  in  The Faculty of Graduate and Postdoctoral Studies  (Chemistry)     THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  February 2014          ?Clement Cheung, 2014 ii   Abstract   Since the discovery of the production of nanocrystalline cellulose (NCC) from acid hydrolysis of bulk cellulose, this material has been used as a template for imparting photonic properties on materials without innate structural ordering and as an additive in polymer composites to increase structural strength.  However, the potential of NCC as a feedstock for highly nitrated cellulose has not been investigated.   It is postulated that the smaller chain lengths in NCC may permit a higher degree of nitration in comparison with conventional nitrocellulose.  In this thesis, various nitration methods have been investigated for their ability to produce highly nitrated NCC.  Traditional nitration methods using HNO3/H2SO4 and less common nitration methods using HNO3/P2O5 and HNO3/Ac2O were used and the differences between methods were studied.  There was an effect in using H2SO4 as the desiccant on the hindrance of the degree of nitration.  Titration of the nitrated material with base shows the presence of more sulfate groups in samples nitrated in the presence of H2SO4 than in the absence of it, demonstrating that H2SO4 is a less ideal desiccant.  In addition, pretreatment of NCC with a desulfation procedure improved the degree of nitration.  The investigation of the other desiccants showed the importance of using a miscible, non-degrading desiccant to obtain a high degree of nitration.  The synthesized nitrated nanocrystalline cellulose, hereby abbreviated as NNC, were analyzed by elemental analysis for the nitrogen content, powder X-ray diffraction for crystallinity, Fourier-transform infrared spectroscopy and thermal gravimetric analysis for the stability of the compound.  iii   Preface   All of the results presented in this thesis were obtained while I worked under the supervision of Professor Mark J. MacLachlan.  All experiments in Chapter 2 and Chapter 3 were conducted by me.                    iv   Table of Contents  Abstract ........................................................................................................................................................ ii Preface ......................................................................................................................................................... iii Table of Contents ....................................................................................................................................... iv List of Tables .............................................................................................................................................. vi List of Figures ............................................................................................................................................ vii List of Symbols ........................................................................................................................................... xi List of Abbreviations ................................................................................................................................ xii Acknowledgements .................................................................................................................................. xiii Chapter 1 ? Introduction ............................................................................................................................ 1 1.1 Modifications of cellulose ........................................................................................................... 1 1.2 History and uses of nitrocellulose .............................................................................................. 3 1.3 Methods of nitration of cellulose ................................................................................................ 8 1.4 Optimization, manufacturing and stability of nitrocellulose ................................................ 12 1.5 Choosing nanocrystalline cellulose .......................................................................................... 16 1.6 Overview of this thesis .............................................................................................................. 20 Chapter 2 ? Methods of nitration of NCC .............................................................................................. 21 2.1 Introduction ..................................................................................................................................... 21 2.2 Experimental ................................................................................................................................... 22 Materials. ........................................................................................................................................... 22 Aqueous desulfation of NCC. ........................................................................................................... 22 Heterogeneous desulfation of NCC. ................................................................................................ 23 HNO3/H2SO4 method. ....................................................................................................................... 23 HNO3/P2O5 method. .......................................................................................................................... 24 HNO3/Ac2O method. ......................................................................................................................... 24 Characterization techniques. ........................................................................................................... 25 2.3 Results and discussion .................................................................................................................... 25 The need for desulfation. .................................................................................................................. 25 Optimization of HNO3/H2SO4 method. ........................................................................................... 31 Optimization of HNO3/P2O5 method. .............................................................................................. 41 Optimization of HNO3/Ac2O method. ............................................................................................. 51 v   Comparison of the NNC products. .................................................................................................. 61 2.4 Conclusion ....................................................................................................................................... 66 Chapter 3 ? Stability studies of NNC ...................................................................................................... 67 3.1 Introduction ..................................................................................................................................... 67 3.2 Experimental conditions ................................................................................................................. 68 Characterization techniques. ........................................................................................................... 68 Stabilization of NNC. ........................................................................................................................ 68 3.3 Results and discussion .................................................................................................................... 68 Aqueous stabilization of NNC. ......................................................................................................... 68 Stability of NNC synthesized with different methods. ................................................................... 75 3.4 Conclusion ....................................................................................................................................... 81 Chapter 4 ? Conclusions and final remarks ........................................................................................... 83 References .................................................................................................................................................. 85              vi   List of Tables  Table 2-1 ? %N comparisons of the effect of aqueous and methanolic desulfation of NCC.  Results show the effectiveness of the latter method for achieving a higher degree of nitration. ...................................... 28  Table 2-2 ? %N comparison of the effect of water on the nitration efficiency.  The H2SO4 removes ~20% of the water in the reaction and has little effect when a more concentrated HNO3 is used. ....................... 34  Table 2-3 ? Effect of the readdition of P2O5 into the nitration mixture in 0.5 g portions on %N in NNC.  The lack of an effect demonstrates the initial amount of desiccant is sufficient in the removal of initial water content and also the water generated during the nitration. ................................................................ 48  Table 2-4 ? %N obtained from EA showing difference and reproducibility between normal and scaled-up nitration reactions.  The inability for HNO3 and P2O5 to quickly disperse NCC leads to the formation of aggregates of lesser nitrated material. ......................................................................................................... 49  Table 2-5 ? Nitrogen content of NNC following multiple nitration steps.  The initial renitration is effective in boosting the %N, however, further reaction does not provide a means of obtaining a result comparable with the smaller-scaled reactions. ........................................................................................... 50  Table 2-6 ? Visual comparison of HNO3/Ac2O preparation at different temperatures and the respective %N measured for the NNC.  Results suggest a low temperature is needed to stabilize the more reactive nitrating agent to avoid decomposition of the mixture. .............................................................................. 53  Table 2-7 ? Repetitions of nitration showing the differences and reproducibility of the normal and scaled-up HNO3/Ac2O nitration method on NCC and MGC.  There is little difference in the product between different reaction scales and also between materials of different crystallinity. .......................................... 60  Table 3-1 ? Physical and visual characteristics of NNC synthesized by HNO3/H2SO4 and HNO3/Ac2O methods after one month and six months of storage.  The sample from the latter method show no signs of degradation within this time frame. ............................................................................................................ 77        vii   List of Figures  Figure 1-1 ? Chemical structures of a) cellulose; b) starch; c) hemicellulose; d) lignin.  The hemicelluloses and lignin act as binding agents to hold together cellulose in plant cell walls. .................... 2  Figure 1-2 ? Structure of cellulose esters a) acetate; b) sulfate; c) phosphate. ............................................ 3  Figure 1-3 ? Structure of cellulose a) mononitrate; b) dinitrate; c) trinitrate.  The 1? -OH on C6 is the most reactive, followed by the 2? -OH on C2 and C3. ................................................................................. 4  Figure 1-4 ? General nitration scheme of the HNO3/H2SO4 mixed acid method.  The generation of the nitrating agent ? nitronium ion ? can be enhanced most notably by the removal of water. ......................... 9  Figure 1-5 ? Effect of H2SO4/HNO3 ratio and the water content on the nitration of cellulose.  Figure reproduced with permission from Taylor & Francis Group.53 .................................................................... 13  Figure 1-6 ? Effect of the esterification coefficient (ratio of cellulose to HNO3) and reaction time on the nitration of cellulose.  Figure reproduced with permission from Taylor & Francis Group.53 ..................... 14  Figure 1-7 ? a) Primary and b) Secondary radical decomposition pathways of nitrocellulose in the presence of moisture.  The generation of HNO3 as a degradation product leads to depolymerization and further denitration. ...................................................................................................................................... 16  Figure 1-8 ? a) Structure of bulk cellulose; b) NCC rods with surface sulfation; c) Stacking of aligned layers of NCC to form CN* phase by EISA; d) POM image of fingerprint texture and resulting films.  Figure reproduced with permission from Nature Publishing Group.68 ....................................................... 17  Figure 2-1 ? FTIR spectra of NCC and DSNCC showing no significant differences in the starting material.  Most importantly, the crystalline ?OH stretches remain intact and no broadening was observed. .................................................................................................................................................................... 29  Figure 2-2 ? FTIR spectra of NNC and DSNNC showing no significant differences except a slight reduction in ?OH band intensity due to a higher degree of nitration. ......................................................... 30  Figure 2-3 ? PXRD patterns of NCC and DSNCC showing no crystallinity changes.  Diffraction peaks can be assigned to cellulose I?.  Broader peaks are observed due to the nanocrystalline nature of the material. ...................................................................................................................................................... 31  Figure 2-4 ? Plot of %N measured in the NNC product as a function of vol. % of HNO3 in the nitrating solution.  The optimal concentration of HNO3 lies at 40%.  The decrease in nitrating ability of higher proportions of HNO3 is attributed to the inefficient removal of water from the reaction. .......................... 33  viii   Figure 2-5 ? Plot of %N versus ratio of NCC loading with respect to HNO3, which shows the independence of the degree of nitration to the amount of material added. ................................................. 37  Figure 2-6 ? Plot of %N versus nitration time (HNO3/H2SO4 method) showing a small increase in %N as the reaction duration was increased past 30 mins.  The %N stops increasing after 3 hours of nitration. ... 38  Figure 2-7 ? Plot of %N versus reaction temperature showing a small increase in %N in the isolated NNC as nitration temperature is increased.  However, significant loss in yield and the increased instability of the product at higher temperatures limits the reaction temperature to 0 ?C. ............................................... 39  Figure 2-8 ? PXRD patterns showing the loss of crystallinity after the nitration reaction.  The nitration fully disrupts the hydrogen bonding network and forms an amorphous product. ....................................... 40  Figure 2-9 ? Plot of %N versus the ratio of HNO3 to NCC showing no dependence of the DS in the synthesized NNC on the amount of NCC added to the reaction. ................................................................ 44  Figure 2-10 ? Plot of %N versus nitration duration showing the shorter time needed when nitrating filter paper for the %N to plateau.  The amorphous regions in filter paper are easier to nitrate due to the lack of extensive hydrogen bonding between cellulose chains. .............................................................................. 45  Figure 2-11 ? Plot of %N versus reaction temperature showing no effect on increasing the DS of NNC as temperature was increased.  The increase in temperature does not increase the effectiveness of the removal of water by P2O5, however, the loss in yield and instability of the product at higher temperatures favors nitration at 0 ?C. ............................................................................................................................... 47  Figure 2-12 ? Plot of %N versus the v% of HNO3 in the nitration showing an optimal mixture at 40 v% HNO3 and yielding a NNC product with 13.4 %N.  The nitrating mixture is ineffective in nitrating NCC when the amount of Ac2O was increased from 60% to 80%. ..................................................................... 55  Figure 2-13 ? Plot of %N versus ratio of HNO3 and NCC showing independence of the %N on the NCC loading.  The insensitivity is due to the compact nature of NCC compared to bulk cellulose fibers. ........ 56  Figure 2-14 ? Plot of %N versus nitration duration showing maximum nitration can be reached in one hour of reaction and remains stable for upwards of six hours.  The induction time is similar to that of the HNO3/H2SO4 and HNO3/P2O5 methods, but a much higher degree of nitration can be achieved with this method. ....................................................................................................................................................... 57  Figure 2-15 ? Plot of %N versus reaction temperature showing the gradual decrease of %N as reaction temperature is increased.  The instability of the nitrating agent at temperatures higher than 0 ?C limits the use of increased temperatures to alter the rate of nitration. ........................................................................ 58  Figure 2-16 ? FTIR spectra showing the gradual reduction of the ?OH band by increasing nitration time in HNO3/H2SO4.  However, the complete removal of the band is not observed. ........................................ 62 ix   Figure 2-17 ? FTIR spectra showing the gradual reduction of the ?OH band by increasing nitration time in HNO3/P2O5.  Most of ?OH groups have disappeared after six hours of reaction, but some traces remain. .................................................................................................................................................................... 63  Figure 2-18 ? FTIR spectra showing the gradual reduction of the ?OH band by increasing nitration time in HNO3/P2O5.  The complete loss of the band is observed after one hour of nitration. ............................. 63  Figure 2-19 ? TGA comparison of the deflagration temperature for the NNC synthesized by the three methods.  A much lower temperature was observed for the HNO3/H2SO4 method caused by the presence of ?OSO3H groups; a stable temperature of >200 ?C was observed for the other two methods. ............... 65  Figure 3-1 ? Deflagration temperatures obtained from TGA showing the effect of water rinses on the deflagration temperature of NNC synthesized by HNO3/H2SO4 method.  The deflagration temperature can be increased significantly after rinsing with D. I. H2O three times. ........................................................... 70  Figure 3-2 ? Deflagration temperatures obtained from TGA showing the effect of water rinses on the deflagration temperature of NNC synthesized by HNO3/P2O5 method.  Even though the product began with a deflagration temperature beneath the stability limit, the temperature can be increased to >200 ?C after water washes. ...................................................................................................................................... 71  Figure 3-3 ? Deflagration temperatures obtained from TGA showing the effect of water rinses on the deflagration temperature of NNC synthesized by HNO3/Ac2O method.  The increased stability of this NNC despite being more nitrated than the other two methods is due to the lack of ?OSO3H groups. ....... 72  Figure 3-4 ? TGA of NCC synthesized by HNO3/H2SO4 subjected to stabilization, aging, and restabilization showing the effects of aqueous stabilization.  This process has a more pronounced effect on this sample as it degrades much more rapidly than the other two methods.  This can be repeated multiple times to prolong the life of this material. .................................................................................................... 75  Figure 3-5 ? Representative appearances of HNO3/H2SO4 synthesized NNC from initial conditions to total failure (left to right) after six months of storage at ambient conditions in darkness.  The production of NO2 gas after two months triggered the following decomposition of the material. ............................... 78  Figure 3-6 ? Representative appearances of HNO3/Ac2O synthesized NNC after two to eight months of storage (left to right in two month increments) at ambient conditions in darkness.  No degassing was observed in stark comparison with NNC from the HNO3/H2SO4 method. ................................................. 79  Figure 3-7 ? Comparison of the %N of NNC produced by the two methods over eight months on the bench top.  The material from the HNO3/H2SO4 method remained stable for two months then rapidly dropped in stability.  The HNO3/Ac2O method, however, produced a material that is stable for storage at ambient conditions. ..................................................................................................................................... 80   x   Figure 3-8 ? Comparison of the %N of nitrated MGC produced by the two methods over eight months on the bench top.  There is a similar effect on this starting material and the stability trends correspond to that of NNC.  Again, the material from the HNO3/H2SO4 method failed to remain stable under ambient conditions. ................................................................................................................................................... 81                     xi   List of Symbols  ?C  degrees Celsius 2?  2 theta cm-1  wavenumber M  molarity v%  volume percent wt%  weight percent              xii   List of Abbreviations  1?  primary 2?  secondary CHN EA carbon/hydrogen/nitrogen elemental analysis CN*  chiral nematic D. I. H2O deionized water DCM  dichloromethane DMA  dimethylacetamide DS  degree of substitution EISA  evaporation induced self-assembly FTIR  Fourier transform infrared GPC  gel permeation chromatography LC  liquid crystal MGC  microgranular cellulose MPa  megapascals NC  nitrocellulose NCC  nanocrystalline cellulose NNC  nitrated nanocrystalline cellulose PTSA  para-toluene sulfonic acid PXRD  powder X-ray diffraction TGA  thermal gravimetric analysis TMOS  tetramethylorthosilicate  xiii   Acknowledgements   I would like to thank Dr. Mark MacLachlan for always being enthusiastic and being there to answer my questions and to provide guidance on this project.  I would also like to thank Ira Wolff for being highly supportive of this project and providing the much needed NCC-Na as starting material.  A big thank you to the members of the MacLachlan research group for being such a great presence during my time here.  In addition, I need to thank Derek Smith from the UBC Mass Spectrometry Centre for running the tremendous amounts of CHN EA and Anita Lam from the X-ray Crystallography facility for her help on PXRD.  Most importantly, a giant thank you to my family and Ivy who were more than supportive during the time I spent completing my project and provided me with care and an environment to relax in.  Thank you all for being patient with me for the past two years.             1   Chapter 1 ? Introduction  1.1 Modifications of cellulose  As the world?s perishable resources dwindle, notably fossil fuels, natural gases and minerals, people are targeting more sustainable options to produce commonly used materials such as gasoline and plastics.  Their attention quickly shifted to natural products, where numerous naturally occurring structures show remarkable stability1 and mechanical properties2; one of the most versatile and abundant biopolymers available is cellulose, which can easily be isolated from both plants, i.e. coniferous trees, and animals, i.e. tunicates, alike.  Cellulose is a linear polysaccharide consisting of D-glucose monomer units connected together by ?(1,4) glycosidic linkages with alternating orientation as opposed to the ?(1,4) linkages in the structural isomer starch.  It can be easily isolated from the cell walls of plants along with a mixture of hemicelluloses, a form of amorphous branched polysaccharide, and lignin, an ill-defined three-dimensional polyaromatic material served as filler between cellulose and hemicelluloses, using mechanical pulping (Figure 1-1).3  Then the undesirable hemicellulose and lignin are removed through chemical treatment, which renders both water soluble.  At 80% by weight in the cell wall, cellulose, as a fundamental contributor to the structure, demonstrates remarkable strength on the order of tens of GPa due to extensive hydrogen bonding between polymer chains.4  In fact, bulk cellulose is comprised of amorphous and crystalline regions that together synergistically contribute to its unique mechanical properties ? regions of high crystallinity imparts strength and regions of disorder give flexibility.5   Currently, the annual production of cellulose is estimated to be greater than 80 gigatons. Yet interestingly, the majority of the research on cellulose has not been the isolation of the 2   material, but rather the incorporation of cellulose derivatives into everyday objects.  In fact, functionalization of cellulose has comprised of most of the use of the polysaccharide, with the exception of making paper, for the last 180 years.6  A primary (1?) alcohol at the C6 position and two secondary (2?) alcohols at the C2 and C3 positions on each glucopyranose repeating unit serve as reactive handles for a multitude of modifications to the polymer (Figure 1-1a).7  These are known as cellulose esters; they are classified either into organic or inorganic esters depending on the type of substituents on the cellulose fibers.    As an example of an organic ester, cellulose acetate, the reaction of cellulose fibers with a mixture of acetic acid (AcOH) and acetic anhydride (Ac2O) quickly substitutes the alcohol groups with an acetyl group with high efficiency.8  This was first discovered in 1865 by the French chemist Paul Sch?tzenberger at the University of Strasbourg and this material was quickly adapted for use for its flexibility and clarity.  By varying the proportions of the two liquids, for instance two to one v/v AcOH to Ac2O, and the reaction time, one can control the degree of substitution (DS) on each of the Figure 1-1 ? Chemical structures of a) cellulose; b) starch; c) hemicellulose; d) lignin.  The hemicelluloses and lignin act as binding agents to hold together cellulose in plant cell walls.   3   glucose repeating unit.9  The applications of the product vary from thin films, for instance overhead projector sheets and film stock, to synthetic fibers.  Similarly, the reaction of cellulose with sulfating agents, for instance sulfonyl chloride, and phosphating agents, for instance H3PO4, yield the inorganic esters cellulose sulfate10 and cellulose phosphate11, respectively (Figure 1-2).   The ease of substitution on the ?OH groups makes cellulose a good feedstock for derivatized materials.    1.2 History and uses of nitrocellulose  Within the large selection of inorganic cellulosic esters, cellulose nitrate (more commonly known as nitrocellulose (NC) or guncotton) is one of the oldest, most researched and sought after cellulose derivatives as it has a wide range of applications.  The discovery of NC dates back to 1832 when the French chemist Henri Braconnot investigated the synthesis of a combustible material by treating wood fibers with nitric acid.  However, not until 14 years later, when the German-Swiss chemist Christian Friedrich Sch?nbein accidentally synthesized highly nitrated cotton fibers, did NC find its practical use in propellants and explosives.  The chemist decided to contain some spilled concentrated nitric acid using the nearest object, which happened to be a cotton apron.  After absorbing the acid and being hung to dry in the air, the apron spontaneously ignited and disintegrated to the amazement of Sch?nbein, leading to the discovery Figure 1-2 ? Structure of cellulose esters a) acetate; b) sulfate; c) phosphate.   4   of guncotton.  Interestingly, three chemists including Sch?nbein independently and simultaneously uncovered this new material in 1846 and a German chemist from Braunschweig, F. J. Otto, was first to publish the general method12 using a mixture of sulfuric acid (H2SO4) and nitric acid (HNO3) for producing this new material where each of the three available alcohol groups (?OH) on the glucose repeating unit is substituted with a nitrate group, ?ONO2.  However, the instability of the nitrated compound led to multiple fatal incidents during the manufacturing process; therefore, a water rinse was added at the end of the reaction to stabilize the product.13  What could have resulted in the difference in properties between the material discovered in 1832 and that discovered in 1846?  It was discovered that the careful choice of the nitration method can produce a nitrated material within a specific range of DS, where properties differ depending on the %N in the material.14  Mechanistically, the reaction rate on the 1? alcohol at the C6 position is 3 times as facile as the 2? alcohol on the C2 position and 6 times as rapid as the 2? alcohol on the C3 position due to considerable steric hindrance.15  As there are three possible positions for the first order electrophilic substitution, each subsequent reaction of the ?OH group with the nitrating agent yields cellulose mononitrate, cellulose dinitrate, and cellulose trinitrate as an indication of the DS of the nitrate (?ONO2) group (Figure 1-3).  However, the degree of Figure 1-3 ? Structure of cellulose a) mononitrate; b) dinitrate; c) trinitrate.  The 1? -OH on C6 is the most reactive, followed by the 2? -OH on C2 and C3.   5   substitution often occurs as a non-integer value, thus a more expansive naming system was introduced that takes into consideration two glucose monomer units instead of a single unit.  Therefore, the possible names for CN with different DS spans from cellulose mononitrate to cellulose hexanitrate.16  Yet, this naming system was soon phased out since it arose much confusion between the two systems, and simply the DS values between zero and three were adapted.  The correlation of DS to %N is calculated using the relationship DS = (1.62 x %N)/(14 - 0.45 x %N).17  Categorization and uses have been found for all ranges of DS: DS < 1.5 (<9.5 %N) are currently used as lacquers for billiard balls and countertops; 1.5 < DS < 2.4 (9.5-12.5 %N) are thermoplastics with good tear resistance and currently used as filter membranes; DS > 2.4 (12.5-14.1 %N) are currently utilized in smokeless powder, rocket propellants and explosives.  Moreover, the category of medium DS is often separated into three grades, A-grade, AM-grade and E-grade, depending on the solubility of the compound in alcohols and thus was marketed by Dow Wolff Cellulosics.18  It was discovered, in addition to neat NC, that materials of NC can be modified and enhanced by the use of organic solvents or the combination of various organic compounds.  Collodion, for example, is the formulation of a thick, gelatinous solution of NC in an ethanol/ether mixture.19  This found use as surgical dressing to hold dressings in place or use as a liquid sealant as it forms a flexible clear film when dried.  Another formulation is celluloid, which was synthesized by Alexander Parkes but later improved by John Wesley Hyatt in 1862 by blending NC and camphor, and found use as a hard transparent coating.  Celluloid is still currently used on table tennis balls and billiard balls.20  Typically the formulation consists of 70% NC, 30% camphor and additives to increase stability.   The ability for camphor to act as a plasticizer resulted in the recognition of celluloid as the first documented thermoplastic.21  However, despite the usefulness of collodion and celluloid, the most infamous 6   use of NC was as a film base dating back to the 1890s.22  The clarity and flexibility of the films cast was much desired in the film industry and thus were the most popular choice at that time.  Yet because of this widespread use, several tragedies occurred over the years between the time of discovery and the 1950s due to the flammability of the film base that prompted the replacement of the NC films with acetate film despite proper precautionary measures such as using asbestos, a flame retardant, in walls of rooms containing NC films.  Eventually, this drawback led to a major shift in the use of NC as film substrates to its sole use in explosives.   NC containing 12.5% or more nitrogen is considered propellant grade or explosive grade, where the higher the nitrogen content, the less residue is left behind when the material is combusted.23  Generally, 0.5% to 2% of the mass of NC is left behind even in highly nitrated samples due to the inhomogeneity of the distribution of ?NO2 groups.24  It was found that narrowing the distribution can drastically improve the material?s physical and chemical properties.  One property is the tendency to deflagrate in its finely powdered form when heated past its decomposition temperature.  NC does not detonate since the propagation velocities never reach that of supersonic flame propagation; typically, the velocity does not exceed 100 m/s even in highly nitrated NC.25  In thin film form, the NC is more spread out so as a result, the wave of heat is more efficiently transferred in powdered NC and a subsonic combustion propagates through the material.  However, in enclosed containers, the deflagration of NC would result in an explosion as the decomposition products under combustion conditions are gases: 2NC + heat ? 7H2O + 9CO + 3CO2 + 3N2.26  The production of 15 moles of gas per two cellulose trinitrate (assuming full nitration) is equivalent to 0.565 L of gas per gram of NC.  Typically, gradual pressure buildup would not be a huge hazard if the container was of an appropriate volume, but a rapid decomposition is enough to cause critical failure of the container.  Ettre and Varadi 7   investigated the decomposition of NC between 300 ?C and 950 ?C and found CO2, CO, NOx, CH4, and H2O in large quantities in the gaseous product.27  Interestingly, nitrogen gas N2 was found with increasing pyrolysis temperature, whereas ethylene, acetaldehyde, methanol, ethanol and methyl acetate were found in the mid-range decomposition temperatures.  This proves the thermal decomposition of NC can also proceed through a radical pathway.  There are also several niche applications of NC investigated in the literature that have not found widespread use ? e.g. liquid crystals and solid state nuclear track detection.  It is known that long chain cellulosic polymers, for example ethylcellulose28 and hydroxypropylcellulose29, will exhibit a LC phase above a critical concentration.  It was documented by Warren that highly nitrated NC, in this instance %N = 12.6, forms a LC phase, similar to other non-ionic cellulose derivatives, in dimethylacetamide (DMA) at the critical concentration of approximately 45% and above.30  Prior to this, Willcox also observed a LC phase for lacquer grade dinitrated NC in several different solvents.31  The high concentration required is due to the fact that interchain interaction of NC in solution is very minimal.  Only when the polymer chains are in close proximity to each other do they have a tendency to align and form the lyotropic LC phase.  The complete transition from the lyotropic phase to the isotropic phase occurs within the temperatures 27 to 67 ?C independent of the concentration of NC and this broad transition is the result of increased viscosity, which prevents the rapid switch between the two phases.  Due to the high concentration required for the LC phase to be observed, applications using this LC are limited.     For the use as solid state nuclear track detectors, NC thin films were first commercialized by Lupica in 1975 and have excellent sensitivity to protons and alpha particles.33  With exposure to bombardment with radioactive particles, the NC films can then be etched with a strongly 8   alkaline solution to expose tracks where particles penetrated the material.  Some efforts such as by Samant were made to find a simple polymer-supported NC solid state nuclear track detector that is more uniform and have high reproducibility.34  1.3 Methods of nitration of cellulose  Generally, cellulose is nitrated in the presence of the mixture of H2SO4 and HNO3 at a low temperature to prevent degradative reactions that occur from the exothermic nitration reaction.35  In the presence of a stronger acid, HNO3 is readily protonated and the resulting hydronium-like species eliminates a molecule of H2O and a molecular species called a nitronium cation (NO2+) is generated.36  As the positive nitrogen is made more reactive by the two electronegative oxygens, this is the nitrating agent and is extremely susceptible to nucleophilic attack.  This short-lived species is often thought to exist as an ion pair with in situ nitrate ions as [NO2+][NO3-] or dinitrogen pentoxide (N2O5) depending on the solvent environment.37  Without the presence of a stronger acid, it is proposed the formation of NO2+ by the disproportionation of two HNO3 molecules yields the protonated nitric acid and a NO3- anion, followed by generation of the NO2+ via elimination of water, but in a much lower quantity in comparison with the mixed acid system.38  The nitrating agent then reacts with the ?OH group through electrophilic substitution and the subsequent deprotonation by a proximal NO3- or HSO4- group yields the nitrate ester of cellulose (Figure 1-4).  Sakata and Komatsu used two techniques, X-ray diffraction and IR spectroscopy, to find that in the presence of sulfuric acid, the nitration proceeds on the surface of the cellulose fiber and gradually penetrates deeper into the material.39  Therefore, depending on the crystallinity of the cellulose, one can envision differences in the rates of nitration.  Evidently, the nitration of one ?OH group generates one molecule of H2O as a 9   side product and thus a dehydrating agent, in this instance H2SO4, is required to drive the reaction by removing water from the equilibrium.40  This mixed acid system with H2SO4 and HNO3 remains the most economical method to produce NC.   Besides the usage of H2SO4 and HNO3 as the nitration mixture, a wide range of other methods for the nitration have been explored, differing between homogenous and heterogeneous methods.  Heterogeneous reactions are generally preferred in the nitration of cellulose since the product is easily isolated either by centrifugation or filtration,41 and also the overall morphology of cellulose remains unchanged if it did not precipitate out of solution. However, heterogeneous methods often run into problems with aggregation and a large distribution of different DS.42  Notably, nitration methods such as HNO3/Mg(NO3)2,43 and HNO3/dichloromethane (DCM),44 have found use on both aromatic nitrations and cellulose nitrations; each method, depending on the acidity of the reaction, yields a product with different degrees of nitration.  A mixture of Figure 1-4 ? General nitration scheme of the HNO3/H2SO4 mixed acid method.  The generation of the nitrating agent ? nitronium ion ? can be enhanced most notably by the removal of water.  10   fuming nitric acid (> 90%) and P2O5 was reported in 1951 by Heuser and Jorgensen for the nitration of bulk cellulose to directly achieve complete nitration of the three available ?OH groups.45  P2O5, or more accurately P4O10, is the anhydride of phosphoric acid produced by oxidation of elemental phosphorus in the presence of excess oxygen.  The P2O5 was added to the nitric acid as a solid to remove the water that was present in the fuming nitric acid as the compound is an excellent desiccant ? the hydrolysis of P2O5 to H3PO4 is exothermic (approximately 180 kJ/mol downhill).  The reaction P4O10 + 6H2O ? 4H3PO4 generates phosphoric acid by the reaction of the drying agent with H2O; therefore, the H2O is actually removed from the reaction in the generation of a side product in comparison with the sulfuric acid-water hydrates (H2SO4 ? xH2O).   Sakata and Komatsu found that the ?OH groups in the cellulose fiber nearly simultaneously reacted with the nitrating agent in the presence of phosphoric acid.39  The instantaneous penetration of the acid into the cellulose fiber causes the nitration reaction to proceed at a much higher rate.  In addition, a small equilibrium of P2O5 and N2O5 in the presence of the nitric acid and water also facilitates the nitration process by generation of dinitrogen pentoxide through the reaction P2O5 + 2HNO3 + 2H2O ? N2O5 + 2H3PO4.46  Thus, the formation of the hydrolyzed desiccant, rapid penetration of the nitration acids and N2O5 as the nitrating agent are the driving forces for the facile nitration with the HNO3/P2O5 mixture.  This method of nitration is extensively used in the production of high explosives, for instance the nitroamine high explosive HMX.47 Another less common nitration method is the HNO3/acetic anhydride (Ac2O) method.48  Ac2O is the anhydride of acetic acid AcOH, which can be envisioned to be a condensate of 2 molecules of AcOH while eliminating 1 molecule of water; however it is industrially synthesized by carbonylation of methyl acetate in the presence of lithium chloride as catalyst.  Similar to the 11   previous acid mixture, Ac2O acts as a dehydrating agent as the reaction of Ac2O with H2O in the equation Ac2O + H2O ? 2AcOH, which draws the equilibrium of the nitration towards the products by removing H2O.  The marked difference between the two methods before and this method is the production of a highly reactive intermediate acetyl nitrate, which forms through the reaction Ac2O + HNO3 ? AcONO2 + AcOH.49  The AcONO2 formed and its protonated form is a much stronger nitrating agent when compared to N2O5 since its reactivity is enhanced by the electron withdrawing ability of the acetate moiety and the solubility is increased by the acetyl group.49  Essentially, the protonated AcOHNO2+ can be seen as a NO2+ molecule solvated by AcOH.  However, due to the instability of this intermediate in solution, the reaction must be conducted at a lower temperature in comparison with the HNO3/H2SO4 method and the HNO3/P2O5 method.50  The chemical equilibrium for the formation of AcONO2 depends on the initial concentration of Ac2O and HNO3 as the presence of more nitric acid promotes the formation of N2O5 instead.51  Several major advantages for this method are the handling of the two liquids is facile as opposed to the hygroscopic solid P2O5 and also the reaction does not become increasingly viscous as more dehydrating agent is added.  A disadvantage is the possibility of a runaway reaction above 40 ?C as AcONO2 decomposes at higher temperatures.  A group at Warsaw University of Life Sciences, Andrzej et al, investigated the minimization of the degradation of cellulose during nitration.35  The HNO3/AcOH/Ac2O method was utilized over the HNO3/H3PO4/P2O5 method since in the latter, thickening of the reaction occurs due to formation of polyphosphoric acid gels, thus diminishing reproducibility.  It was found that temperature had a huge effect on the quality of the product where low temperatures favored high %N and also high weight average degree of polymerization, indicating the stability of cellulose 12   inside the HNO3/AcOH/Ac2O system varies greatly by temperature.  Currently, this system is employed in the N-nitration of aromatic amines, particularly in the production of explosives.52 Between the different methods of nitration, one condition seems to be satisfied ? the removal of water from the reaction.  The generation of equimolar quantities of water per ?OH group nitrated drastically slows down the reaction rate if it is not sequestered by a dehydrating agent.  In light of this, Sun et al. investigated the effect of water content on the nitration of bacterial cellulose.53  It was found that the %N in the final product decreased linearly with increasing water content.  Hence, it is highly possible, by extrapolation, to nitrate cellulose using 100% HNO3 freshly distilled under vacuum from a mixture of HNO3 and H2SO4.  Hughes et al. investigated the nitration of cellulose using white fuming HNO3 and found that in the absence of water the reaction proceeded at a similar rate as the nitration reactions catalyzed by H2SO4, P2O5 or Ac2O.54  However, industrial scale implementation did not proceed as large scale anhydrous nitric acid distillation is extremely hazardous, storage requires specialized tanks and the process generates of an equivalent or more of spent acid that requires treatment or disposal.  1.4 Optimization, manufacturing and stability of nitrocellulose  In regards to the variability of the product yielded from nitration, Sun et al. investigated the effect of the ratio of H2SO4 to HNO3,53 the water content of the acid mixture, the reaction temperature, and the reaction time on the DS of the NC produced.  By using bacterial cellulose generated from Acetobacter xylinum through alkaline treatment to solubilize organic impurities, the nitrated product was optimized through a series of experiments.  The maximum efficiency of the HNO3/H2SO4 mixture was centered around 3:1 sulfuric/nitric acid, however, this value deviated from the normally used 2:1 sulfuric/nitric acid.  The most remarkable effect was seen in 13   the water content of the reaction.  It was shown that the DS of the final product decreases linearly from 2.6 to 2.0 by increasing the water content from 8% to 24%.  This was attributed to the consumption of the nitronium cation by surrounding water through the equilibrium H2NO3+ ? H2O + NO2+ thus reducing the overall nitrating propensity (Figure 1-5).  Furthermore, the ratio of the bacterial cellulose to the nitration mixed acid was varied; it was found that a ratio of 1:56 and above had no effect on the DS of the nitrated bacterial cellulose, but the DS rapidly drops when the ratio of starting material to acid becomes smaller.  This is inferred as the maximum viscosity the nitration can reach before diffusion of reagents limits the nitrating ability.  The reaction temperature ranging from 20 ?C to 40 ?C demonstrated little or no change in DS, but there was notable degradation of the starting bacterial cellulose when the nitration temperature was increased; a similar effect was found when the reaction time was increased from 0 minutes to 45 minutes where the degradation and denitration of the nitrated product was found (Figure 1-6).   Ultimately, the effect on the nitration reaction was ranked as water content >> acid ratio > starting material loading > temperature > time.  Figure 1-5 ? Effect of H2SO4/HNO3 ratio and the water content on the nitration of cellulose.  Figure reproduced with permission from Taylor & Francis Group.53 14   Even though all NC is manufactured by the HNO3/H2SO4 method, research has gone into increasing the degree of nitration by a multi-step process,55 where isolated lower substitution NC was subjected to the same nitration process again prior to isolation.  It is thought the repetition of the nitration step is sufficient to convert all of the unsubstituted ?OH groups to the ?ONO2 group after two renitration steps.  However, the multi-step nitration was never adapted for the production of NC because of the increase in cost of labor and the generation of increasing amounts of waste acid.  Instead, a semi-continuous method was developed and adapted at Bofors-Nobel-Chematur in Sweden in the 1960s56 that continuously nitrates, centrifuges and stabilizes the nitrocellulose product and is still used industrially today.   The process begins with mechanical disintegration and drying to increase the surface area available for nitration.  The cellulose is then mixed with the nitrating acid, defibrillated, and centrifuged.  Depending on the desired DS of the NC, the product is either reconstituted with fresh nitrating acid or moved through to the wash and stabilization stages.  The stabilization is typically done by boiling the nitrated material in water or MeOH several times after each filtration/centrifugation cycle to remove the adsorbed acids.  The addition of sodium bicarbonate, NaHCO3, acts to neutralize the Figure 1-6 ? Effect of the esterification coefficient (ratio of cellulose to HNO3) and reaction time on the nitration of cellulose.  Figure reproduced with permission from Taylor & Francis Group.53 15   desorbed acid and is added to speed up the stabilization process.  In the semi-continuous method, the most interesting feature of the process is the regeneration of the nitration acid after each centrifugation stage.  By reconstituting the dilute HNO3/H2SO4 mixed acid with fresh concentrated H2SO4, the nitration mixture can be reused in several additional cycles before it is deemed inefficient for the nitration of cellulose.    As mentioned, NC films are incredibly flammable and must be kept separate for long term storage.  However, NC films are unstable as seen in the yellowing and loss of flexibility in aged samples.57  The discoloration is a result of the release and readsorption of NO2 gas caused by the hemolytic or heterolytic cleavage of the ethereal O-N bond.58  Jones et al. investigated the effect of UV light exposure on NC films.59  It was found that the light source of lower wavelength provided enough energy to cleave the O-N bond homolytically, thus eliminating a NO2 radical as a yellow gas and generating an oxygen radical on the NC film.  Then in the presence of inherent moisture on the NC film, the NO2 gas redissolves to yield HNO3.  The generated acid can protonate the primary oxygen, exactly the reverse of the nitration process, and heterolytically, NO2 is driven off and the original cellulosic ?OH group is left behind.  The rearrangement in the glucopyranose and release of formaldehyde is also possible if the radical is centered on the primary alcohol.  Another elimination of NO2 can occur on the C3 position adjacent to the radical generated on the C2 position as two aldehyde moieties form.  Since the presence of atmospheric moisture is inevitable, the degradation is autocatalyzed by the initial liberation of NO2 gas (Figure 1-7).60  On one hand, residual acid remaining adsorbed on the polymer chains from the synthesis of NC can also initiate the degradation without homolytically cleaving the O-N bond.  On the other hand, the increasing concentration of HNO3 in the material will catalyze the hydrolysis of the ?(1?4) linkages between cellulose monomer units, thus 16   lowering the average molecular weight of the NC polymer chains; flexibility is lost by the formation of shorter NC chains.15  Evidently, the need for NC to be kept away from UV light, heat and moisture somewhat limits its applications.   1.5 Choosing nanocrystalline cellulose  In the early 1950s, it was discovered that by careful control of the hydrolysis conditions, one can obtain a micron-sized or a nano-sized cellulosic material, which were named microcrystalline cellulose61 and nanocrystalline cellulose (NCC),62 respectively.  The amorphous regions in bulk cellulose have a much greater propensity to be hydrolyzed by acid than the Figure 1-7 ? a) Primary and b) Secondary radical decomposition pathways of nitrocellulose in the presence of moisture.  The generation of HNO3 as a degradation product leads to depolymerization and further denitration. 17   crystalline regions.  The rod-shaped NCC typically have dimensions of approximately 100-200 nm in length and 5-10 nm in width depending on the duration of hydrolysis.63  Also, when H2SO4 is used during the hydrolysis step, one can obtain NCC with a high degree of surface sulfation at approximately every other alcohol group exposed on the surface.64  Similarly with H3PO4, the surface can be functionalized with phosphoric acid groups.65  These surface functionalizations impart an astounding chemical property to these nanocrystals ? the ability to spontaneously self-assemble into a chiral nematic phase (CN*) in aqueous conditions through evaporation-induced self-assembly (EISA) (Figure 1-8).66  This property was first reported by Revol et al. and it was postulated that the stabilization of the NCC in solution is by electrostatic repulsion between negatively charged sulfate groups and the organization of NCC when dried minimizes the energy of these interactions.67  Furthermore, Shopsowitz et al. synthesized NCC- templated mesoporous silica using the inverse opal method by condensing tetramethylorthosilicate (TMOS) in the presence of an aqueous dispersion of NCC, anchoring the CN* phase in the interior of the silica material.68  These combined efforts stemmed a burst of research in the field of NCC as a template for the synthesis and discovery of novel chiral nematic Figure 1-8 ? a) Structure of bulk cellulose; b) NCC rods with surface sulfation; c) Stacking of aligned layers of NCC to form CN* phase by EISA; d) POM image of fingerprint texture and resulting films.  Figure reproduced with permission from Nature Publishing Group.68 18   materials.69-71   Besides the use as photonic materials, there is also tremendous interest in the incorporation of NCC into polymer matrices as reinforcement since the strength to weight ratio of NCC is nearly unrivaled.  The Young?s modulus of NCC is 100-200 GPa, comparable to that of stainless steel.72   With NCC as the starting material for the nitration, it may be possible to improve upon the already existing material traditionally made from bulk cellulose.  The usage of NCC poses several advantages towards nitration.  Firstly, the purification process of NCC involves the delamination of and the selective removal of hemicellulose and lignin.  Hemicellulose cannot be nitrated to the same extent as cellulose as branching of the polymer removes reactive ?OH groups.73  Similarly, the polyaromatic lignin, when nitrated, does not contribute positively to the amount of nitrogen in the overall product.  The presence of such non-cellulosic nitrated products brings variation and uncertainty in the determination of the quality of the nitrated product.  Secondly, the increased crystallinity and order of nanocrystalline cellulose74 may improve the homogeneity of the nitration.  Without the amorphous regions, which contain a large variance in cellulose polymer chain lengths, the highly crystalline regions represent a narrower range of molecular weights, and thus the distribution of the nitrogen content of the final product may be better controlled to give a nitrated product with better chemical properties.  Thirdly, due to the shorter polymer chain lengths, there is an increased amount of glucosidic terminating groups in NCC.75  Whilst the average monomer in cellulose has a maximum of three reactive ?OH moieties for electrophilic substitution, the terminal groups contain four to five reactive ?OH groups, depending on whether the pyranose form or the open linear form of glucose were considered, which are also available for nitration.  As a result, there is a possibility that a nitrated product of greater nitrogen content in comparison with the theoretical maximum of 14.1% 19   nitrogen can be obtained.  This material may have excellent properties as it surpasses the explosive/propellant grade NC.  Fourthly, the smaller chain lengths of NCC allow for more concentrated solutions to be prepared without a large increase in viscosity.  In application with the nitration process, a lowered viscosity in the reaction corresponds to better diffusion of NO2+ (or AcONO2) through the acidic media to the cellulose surface.  The extent of nitration is greatly dependent on the viscosity ? poor heat transfer, which can cause localized runaway reactions and lowered mobility, results in less nitrated cellulose.  Moreover, it is more cost effective to increase output of NC without the drawback of a lower quality product.  Fifthly, due to the nature of the starting material, densification of the nitrated product is not necessary, removing the need for extra solvents and labor.76  Typically, densified NC is made through solvent emulsion techniques where the fibrous material is swollen, sheared and precipitated to generate small particles of NC.  Sixth and finally, the smaller dimensions of NCC allow clear films to be cast, which may be reflected similarly in the nitrated product.  However, research has shown the sulfate content induces instability in NC,77 leading to a lower deflagration temperature of the material, possibly due to the hygroscopic sulfate groups promoting water and acid adsorption.  Dawoud et al. have shown by Thorin titration, using BaClO4 to precipitate available SO42- ions as BaSO4, that moderately nitrated samples (12 %N), contain 0.7% sulfates whereas highly nitrated samples (13 %N) contain half the amount of sulfates at 0.35%.78  This research showed the importance of having a highly nitrated NC to prevent excess sulfate esters from forming.  Similarly, Castirubam et al. investigated the effect of stabilization of NC by boiling it in dilute acid and concluded 95% reduction of the sulfate content by leaching out adsorbed H2SO4.13  The rest of the sulfate content seems to correspond to residual bound sulfate esters on the NC.  20   1.6 Overview of this thesis  In this thesis, three methods of nitration, HNO3/H2SO4, HNO3/P2O5, HNO3/Ac2O, were investigated for their ability to produce highly nitrated samples of NCC.  Multiple aspects of the method have been investigated, including reaction time, reaction temperature, NCC ratio to HNO3, and the ratio between HNO3 and the desiccant, in order to draw conclusions on the advantages and disadvantages of each method.  In addition, the physical properties, chemical properties and stability of the nitrated NCC materials were characterized by IR, TGA, PXRD, and EA.                  21   Chapter 2 ? Methods of nitration of NCC  2.1 Introduction   Nitrocellulose (NC), production has been optimized industrially to yield a consistent nitrated material with good performance.  The typical reaction uses a mixture of HNO3 and H2SO4, where the stronger sulfuric acid simultaneously acts as a desiccant to remove H2O from the reaction and provides H+ to HNO3 to generate the active nitrating agent, NO2+.  This method has proved to be the most economical since both of the acids in question are produced in megaton scales around the globe and the acid waste generated can be treated with aqueous base to generate harmless salts as byproducts.  In addition, the HNO3/H2SO4 method can readily be tuned to produce nitrocellulose of specific grades depending on the amount of H2O present in the reaction.  The overall reaction is relatively robust and the use of the resulting products span coating to explosives.  However, in dealing with a new material, such as NCC, it is not entirely reasonable to presume the same conditions would apply for the nitration of this compound to still yield a material with comparable properties.  The much more crystalline material in NCC may impart interesting problems in the synthesis of nitrated nanocrystalline cellulose (NNC).  Through the analysis of the acid composition, the NCC to HNO3 ratio, the reaction duration and the reaction temperature, an optimized method could be developed for the nitration of NCC.  In addition, the need to desulfate the NCC before the nitration reaction is investigated since the presence of ?OSO3H groups may impart undesirable properties on the resulting NNC product.  By varying the different nitration methods, the various parameters affecting the nitration of NCC were investigated and the best nitration method was determined.  In this chapter, three different nitration methods, the HNO3/H2SO4 method, the HNO3/P2O5 method, and the HNO3/Ac2O 22   method were investigated to determine the maximum nitration achievable for NCC.  The change in the desiccant used in each method may provide an insight into the sensitivity of the nitration of NCC and how the nitration can theoretically be tuned to yield a product with maximum nitration of 14.1% nitrogen.    2.2 Experimental  Materials.  The NCC was obtained as a spray-dried sodium salt form from NORAM (pH 7.0 in water).  HNO3 (68%-70%) was obtained from Fisher Scientific, HNO3 ACS Reagent Grade (> 90%) and H2SO4 (95%-98%) were obtained from Sigma-Aldrich and all were used without further purification; NaOH pellets, KOH pellets, P2O5 ACS Grade (> 99%) and Ac2O ACS Grade (>97%) were obtained from Fisher Scientific and were used without further purification.  Acetone and MeOH were obtained from Fisher Scientific and were used without further purification.  The saturated bicarbonate solution was prepared from solid NaHCO3 (99.9%) obtained from Fisher Scientific.    Aqueous desulfation of NCC.  A 10 M NaOH solution was prepared in 5 mL of water and 1.6 g of NCC was suspended in 20 mL of water to yield an 8 wt% suspension.  The two aqueous solutions were combined in a 50 mL round bottom flask and heated to 65 ?C for 5 h.  The suspension was then transferred to a dialysis bag (cutoff MW 12400) and dialyzed against running tap water for 2 d and subsequently 23   against deionized water until effluent tested pH neutral.  The resulting suspension was lyophilized to yield a fluffy light-brown solid as DSNCC.  Yield: 1.5 g.  Heterogeneous desulfation of NCC.  In a 250 mL round bottom flask fitted with a condenser 16.8 g of KOH and 15 g of NCC were combined. After the flask was evacuated and purged with N2, 150 mL of dry MeOH was added through a syringe and the mixture was stirred to dissolve the base at room temperature.  The mixture was heated to reflux (90 ?C) for 5 h under N2.  The suspension, after cooling to room temperature, was filtered through a medium glass frit and washed with MeOH until effluent was colorless.  The beige solids were subsequently washed with water until the filtrate tested neutral by pH paper.  The product was then dried with acetone until a free-flowing light brown DSNCC powder was obtained.  Yield: 10.1 g.  HNO3/H2SO4 method.  In a 250 mL round bottom flask, 49 mL conc. HNO3 was charged and 91 mL conc. H2SO4 was slowly added to the mixture kept at 0 ?C by an ice-water bath.  1.75 g of NCC was slowly added to the rapidly stirring acid mixture whilst maintaining the temperature at 0 ?C.  The reaction proceeded at the same temperature for 2 h and then was quenched by pouring the NNC/acid mixture into 1.5 L of ice-water.  The resulting white precipitate was filtered through a medium glass frit and washed multiple times with deionized water until the effluent tested neutral by pH paper.  The dried NNC is a free-flowing white powder.  Yield: 2.33 g.  24   HNO3/P2O5 method.  In a 100 mL Erlenmeyer flask 25 mL of 90% fuming HNO3 was prechilled to 0 ?C and slowly added to 15 g of P2O5 over the course of 10 mins.  Full dissolution of the desiccant was achieved after stirring at 0 ?C for 2 to 3 h.  The mixture was filtered through glass wool to remove large agglomerates of polyphosphoric acid.  The resulting dried HNO3/P2O5 mixture was transferred into a 100 mL round bottom flask and 1.25 g of NCC was subsequently added slowly under rapid stirring.  The temperature was maintained at 0 ?C for 3 h and the reaction was then quenched by pouring the mixture into 250 mL of ice water.  The resulting white precipitate was filtered through a medium glass frit and washed with copious amounts of water until effluent tested neutral by pH paper.  The dried NNC was a free-flowing white powder.  Yield: 1.95 g.  HNO3/Ac2O method.   A 250 mL round bottom flask chilled to -30 ?C was charged with 50 mL of 90% fuming HNO3.  The addition of 75 mL Ac2O was slowly conducted through an addition funnel to achieve a colorless mixture.  The HNO3/Ac2O solution is then warmed to 0 ?C in an ice-water bath and 2.5 g of NCC was subsequently added slowly to achieve a uniform dispersion.  The reaction proceeded at 0 ?C for 3 h and was quenched by pouring the mixture into 1.5 L of ice-water.  The resulting white precipitate was filtered through a medium glass frit and was washed with copious amounts of water until effluent tested neutral by pH paper.  The dried NNC was a free-flowing white powder.  Yield: 4.2 g.     25   Characterization techniques.   Fourier-transform infrared spectroscopy was performed with a PerkinElmer Frontier ATR FTIR.  Thermal gravimetric analysis (10 ?C/min under air) was performed with a PerkinElmer STA 6000.  Powder X-ray diffraction was performed with a Bruker D8 Advance diffractometer.  CHN elemental analysis was performed with a Fisons EA 1108 CHN-O.  The instrumental errors associated with CHN EA is ? 0.3% as shown in the error bars.  All samples were dried under vacuum overnight prior to analysis and were handled with extreme caution to prevent accidental deflagration.      2.3 Results and discussion  The need for desulfation.  It is known that the presence of sulfate groups on NC impart instability to the material due to the possibility of alternate degradation pathways than the traditional ones for NC.  The sulfate groups, which are protonated and acidic, can act as a source of H+ to either protonate the oxygen attached to the NO2 group or the ethereal linkage between the glucopyranose repeating units thus initializing the autocatalyzed degradation of the NC.  Starting with NCC, which already contains surface sulfate groups from the acid hydrolysis with H2SO4, it might be counterintuitive to use an ?unstable? starting material for the production of NNC.  However, since the nitration conditions in the HNO3/H2SO4 mixed acid method produces a highly reactive nitrating agent and not a sulfating agent (treating NCC with H2SO4 only leads to further hydrolysis), the reactive ?OH groups on NCC would preferentially react to form ?ONO2 groups instead of ?OSO3H groups.  Also, it was rationalized that the preformed sulfate groups on the 26   NCC would be replaced by ?ONO2 groups due to chemical equilibrium, so the overall product should contain little or no sulfate groups, whose quantity is solely dependent on the ratio of the rate of nitration and the rate of sulfation.  Hence, the nitration of NCC and Whatman filter paper was conducted using the HNO3/H2SO4 mixed acid method to evaluate and compare the effect of sulfate groups on the degree of nitration. With the standard nitration method, the resulting DS of the NNC and NC produced from filter paper shows that there was indeed an effect from the sulfate groups present on NCC.  The nitration of the two materials was conducted using an acid ratio of 35:65 HNO3 to H2SO4, starting material loading of 1:70 NCC to HNO3, and nitration time of two hours at 0 ?C.  The change in morphology is less apparent in the NCC after nitration, which remained a white powder, than the obvious ?fluffing? and loss of flexibility of the nitrated filter paper.  EA shows that NNC and nitrated filter paper have 10.5 and 11.4 %N respectively.  Assuming both starting materials are infinite chains of glucopyranose repeating units, thus ignoring the effect of the terminal glucose, the differences between the %N from nitrating NCC and nitrating filter paper can be directly correlated to the amount of sulfate groups on the surface of NCC.  Using conductometric titration, it is possible to accurately quantify the amount of surface sulfate groups on NCC by neutralization with aqueous NaOH as the sulfate groups are highly acidic and will readily react with strong bases to yield the neutral salt form of NCC.  The conductivity of the NCC suspension reaches a minimum at the equivalence point as the readily dissociating acidic protons are neutralized and the excess of base is absent from the mixture.  Through three replicates of the titration of 1% NCC in water with 0.01 M NaOH, the amount of sulfate groups accounts for 2.0% of the weight of NCC.  Accounting for the increase in formula weight of nitrated cellulosic materials, the resulting sulfate groups account for 1.1% of the weight of NNC.  27   Comparing with the nitration of filter paper, which does not contain surface sulfate groups, the difference in the amount of sulfate groups can account for the initial difference in the degree of nitration.   To further prove the hindrance of the nitration process by the presence of sulfate groups on the NCC, desulfation of the NCC was investigated.  It is well reported in the literature the use of either acidic desulfation by treating NCC suspensions with HCl,79 solvolytic desulfation with pyridine and DMSO,80 or alkaline desulfation by treating NCC suspensions with strong base81 such as NaOH, to effectively displace the sulfate groups on the cellulose. The aqueous base treatment of NCC was selected for desulfation.  Following the literature procedure, the addition of aqueous NaOH to the NCC suspension caused the NCC to immediately gel due to the destabilization of the suspension by increasing ionic strength and aggregation of NCC by reducing intermolecular repulsion from surface sulfate groups.82  After one cycle of desulfation, however, it was obvious that due to the increase in viscosity during the reaction, the complete desulfation of NCC would most likely require multiple repetitions as the freeze-dried post-dialysis DSNCC powder was still somewhat redispersible in water.  This is an indication of the presence of residual sulfate groups that impart a surface charge on the crystallite.  Subjecting the DSNCC from the first cycle to the same alkaline conditions results in a fully non-dispersible DSNCC powder.  The nitration of the DSNCC shows remarkable results, where the %N of the resulting product (11.1%) is 0.4% higher than that of the material synthesized from NCC after only one desulfation step, and 1.0% higher than the nitrated NCC prior to desulfation.  Titration of the freeze-dried DSNCC shows that the amount of sulfate groups is diminished, but some sulfate groups remain in the DSNCC.  By extrapolation, a better desulfation method will generate DSNCC without sulfate groups, which in turn would be nitrated to a further extent; 28   thus, a harsher desulfation method was adapted.  Using a concentrated solution of KOH in MeOH, and then applying heat to reflux for 5 hours, the desulfation of NCC proceeded smoothly without the increase in viscosity observed in aqueous conditions.  The color of the methanolic solution quickly darkened from colorless to yellow, and then intensified to a brown color.  Despite changing the reaction conditions from reflux under air to reflux under an atmosphere of N2 to prevent oxidation, analysis shows that the formation of the brown byproduct is not affected by the presence of oxygen.  It likely contains a complex mixture of ring-opened glucosidic products and their polymeric counterparts.  Isolation by filtration and reprotonation with several aqueous washes generates DSNCC that is not redispersible in water.  The possible mechanism for the desulfation includes both the nucleophilic attack on the sulfate group with the hydroxide to generate sodium sulfate, or through degradation of the surface layers of the NCC by hydrolysis of the ?(1,4) linkage to generate glucosidic oligomers and other related products.  The latter mechanism is supported by the sharp decrease in yield of DSNCC as the desulfation time was varied from 3 hours to 12 hours in MeOH.  The nitration of the MeOH desulfated NCC yielded NNC of 11.1 %N, in comparison with NCC, which has a %N of 9.8% (Table 2-1). Titration of the DSNCC with NaOH further reinforces the difference in the nitrated product as the amount of sulfate remaining is 0.1%; therefore, there is a correlation of the initial sulfate content to the %N in the final nitrated product.   Table 2-1 ? %N comparisons of the effect of aqueous and methanolic desulfation of NCC.  Results show the effectiveness of the latter method for achieving a higher degree of nitration.  Aqueous (%N) MeOH (%N) NCC 9.8 9.8 1st desulfation 10.7 11.2 2nd desulfation 11.1 11.1 29    The FTIR spectra of NCC and the DSNCC from both methods shows no major changes to the ?OH stretching region at 3330 cm-1, which further implies the difference in the degree of nitration is not caused by the differences in the reaction on the ?OH groups but the presence of residual ?OSO3H groups not observed by IR spectroscopy (Figure 2-1).  Similarly, the three       ?ONO2 stretches for NNC and DSNNC located at 1637 cm-1, 1274 cm-1 and 823 cm-1 do not have a significant increase in intensity after desulfation since the amount of ?ONO2 groups is extremely large in comparison with the amount of ?OSO3H groups available.  In addition, the disappearance of any S=O stretches are not observed since they are embedded in the fingerprint region between 1000 to 750 cm-1, which directly coincides with the pyranose C-O stretches from the glucose repeating units (Figure 2-2).  Overall, FTIR spectra yielded no new data with the   Figure 2-1 ? FTIR spectra of NCC and DSNCC showing no significant differences in the starting material.  Most importantly, the crystalline ?OH stretches remain intact and no broadening was observed. 20304050607080901001101205001000150020002500300035004000%TWavenumbers (cm-1)NCCDSNCC30   exception that the retention of the multiple-peaked ?OH stretch resulting from the 1? alcohol and the 2? alcohols on the glucopyranose repeating unit indicates the preservation of crystallinity of the starting material.  PXRD patterns of NCC and DSNCC both show the characteristic broad peaks at the 2? of 15.1?, 16.8?, 22.0? and 34.9? corresponding to (0,1,0), (0,0,1), (0,1,1) and (0,2,1) hkl diffraction planes, respectively, for cellulose I? (Figure 2-3).83  This shows that both the aqueous alkaline treatment and the harsher methanolic alkaline treatment did not affect the morphology of the cellulose nanocrystals.  Using Topas Rietveld refinement, there was a clear reduction in crystallite radius from NCC to DSNCC indicating the desulfation of NCC primarily proceeds through the cleavage of the ethereal linkage between glucose repeating units leading to crystallites of smaller dimensions.  From here on, all experiments are conducted with DSNCC unless otherwise specified.  Figure 2-2 ? FTIR spectra of NNC and DSNNC showing no significant differences except a slight reduction in ?OH band intensity due to a higher degree of nitration.   20304050607080901001105001000150020002500300035004000%TWavenumbers (cm-1)NNCDSNNC31    Figure 2-3 ? PXRD patterns of NCC and DSNCC showing no crystallinity changes.  Diffraction peaks can be assigned to cellulose I?.  Broader peaks are observed due to the nanocrystalline nature of the material.  Optimization of HNO3/H2SO4 method.  The traditional method of using HNO3 and H2SO4 for the nitration of cellulosic materials and aromatic organics has been widely used owing to the low cost of HNO3 and H2SO4, and also the tunability of the method by increasing or decreasing the water content through different concentrations of H2SO4.  However, the nitration of cellulose must be carefully optimized due to the heterogeneous nature of cellulose in the HNO3/H2SO4 method.  In order to ensure proper nitration with a narrow range of DS, one must ascertain that all the ?OH groups on the starting material are available for reaction simultaneously; dispersibility of cellulose in the acid is thus an issue to address in the following optimization experiments.  With NCC as the starting material, due to the increased density of the crystalline regions of cellulose, the ability for the solid to 0501001502002503003504004500 10 20 30 40 50 60 70 80Cps2 theta (o)NCCDSNCC32   swell and disperse in the acid is ever more important.  A series of experiments was conducted at 0 ?C to provide a qualitative comparison of the swelling of NCC in the two acids.  Through a series of 100% HNO3 to 100% H2SO4 in 20 v/v% increments, it was apparent that NCC swelled much more noticeably in H2SO4 than in HNO3.  An attempt to disperse NCC in HNO3 yielded an extremely aggregated suspension where rapid stirring was inadequate to break up the large undispersed NCC.  On the other hand, NCC quickly formed a nearly uniform dispersion with increasing amounts of H2SO4.  It can be inferred that the presence of surface sulfate groups on NCC causes the material to interact more strongly with the surrounding H2SO4 solvent in contrast with HNO3, which does not have favorable interactions with the surface of NCC.  DSNCC also disperses better in H2SO4, but the differences between the two acids are less remarkable than with sulfated NCC.  H2SO4 remains the better acid to disperse NCC and DSNCC possibly due to the greater acidity and larger dipole moment of the acid.   Reactions conditions for the nitration of cellulose have already been optimized, though some details are unpublished or buried in patents.  Because of the increased crystallinity and surface area of NCC, it was thought that different conditions may be necessary to achieve the same degree of nitration as commercially available materials.  In order to find the optimal conditions, a systematic investigation on the effects of HNO3/H2SO4 ratio, NCC/HNO3 ratio, reaction time, and reaction temperature was conducted.   In order to generate the reactive nitrating agent NO2+ in situ, the equilibrium dictates that the removal of water from the reaction favors the formation of the nitronium cation.  However, it is worth pointing out that using H2SO4 as the dehydration agent poses a major drawback ? the hydrolysis and decomposition of the NCC starting material.  The nitration of NCC for 4 hours at 0 ?C in a 1:70 NCC to HNO3 ratio was used to investigate the acid ratio dependence of the 33   nitration of NCC.  By varying the HNO3 to H2SO4 ratio from 20% to 80% in 10% increments, it is clear that the maximum degree of nitration occurs around 30 to 40% HNO3 in the reaction mixture.  This mixture yields a somewhat free-flowing white NNC product with ~11.1 %N (Figure 2-4).  However, deviation from this optimal nitration lead to a dramatic drop in the DS of the nitrated NCC (~8.3 %N).    Figure 2-4 ? Plot of %N measured in the NNC product as a function of vol. % of HNO3 in the nitrating solution.  The optimal concentration of HNO3 lies at 40%.  The decrease in nitrating ability of higher proportions of HNO3 is attributed to the inefficient removal of water from the reaction. Increased proportion of sulfuric acid in the reaction mixture caused several side reactions to occur during nitration: 1) the hydrolysis of the glucosidic linkages between repeating units evident by the gelatinous nature of the translucent isolated product after quenching, which is in stark contrast with the white solid product isolated for other HNO3/H2SO4 mixtures; 2) denitration of the nitrated product.  On the other hand, by increasing the amount of HNO3 in the system from 40 v/v% to 80 v/v%, a gradual decrease in the %N of the isolated product from the 678910111213140 10 20 30 40 50 60 70 80 90%Nvol. % HNO334   optimized 11.1% down to 6.3% was observed.  In the extremes, nitration of NCC in 100 v% of 69% HNO3 led to product with very low degree of nitration (~1.2 %N) after three hours of reacting at 0 ?C.  This is attributed to the increased presence of H2O in the reaction mixture when no desiccant is used.  By starting with conc. (69%) HNO3 for the nitration, nearly 30% of the excess volume needs to be dehydrated to generate the nitronium ion.  In addition, the resulting product NNC easily redisperses in H2O, which indicates the presence of surface sulfate groups that remained unaffected in the neat nitric acid.  A control experiment using fuming (90%) HNO3 as the nitration media was conducted on NCC.  As expected, the 10% H2O in the fuming nitric acid as compared to the 31% H2O in the commercial grade nitric acid has a drastic effect on the degree of nitration on NCC; the resulting product contained 11.1 %N, similar to that obtained using a 30:70 or 40:60 HNO3 to H2SO4 nitration mixture under optimized conditions (Table 2-2).  Table 2-2 ? %N comparison of the effect of water on the nitration efficiency.  The H2SO4 removes ~20% of the water in the reaction and has little effect when a more concentrated HNO3 is used.  %N 69% HNO3 + H2SO4 11.1 90% HNO3 11.1 90% HNO3 + H2SO4 11.3  By calculation, 20% of the 30% of residual H2O in the commercial concentrated HNO3 is effectively removed from the HNO3/H2SO4 mixture by having H2SO4 in two-fold excess; each H2SO4 molecule can effectively immobilize two molecules of H2O.  In theory, the presence of more H2SO4 should further increase the efficacy of nitration, but too much H2SO4 leads to degradation of NCC.  The mixture of 90% HNO3 and H2SO4 (50:50 to minimize the amount of H2SO4 present in the reaction) was also investigated for its nitration ability.  The result is 35   surprising since the addition of H2SO4 had little effect on increasing the %N of the nitrated product, which on average improved upon the neat 90% HNO3 experiment by only ~0.2 %N.  It is postulated that besides the involvement of H2SO4 in the degradation of NCC, the presence of the acid also affects the equilibrium of acid-catalyzed desulfation of NCC ? the presence of excess H2SO4 in the reaction allowed the surface of NCC to remain sulfated after the nitration has occurred.  Thus, conductometric titrations of the NNC product from the 40:60 HNO3/H2SO4 procedure, the neat 90% HNO3 procedure, and the 50:50 90% HNO3/H2SO4 procedure were conducted to quantify the amount of sulfate groups remaining.  The data shows that the methods had varying sulfate content of 1.2%, 0.1%, and 0.4%, respectively.  This trend indicates that the presence of sulfuric acid causes more retention of sulfate groups on the product NNC then in the absence of the acid.  In addition, using a titration of the starting material NCC, a clear increase in the amount of sulfate groups was seen in the samples reacted in the presence of H2SO4, which demonstrates that the presence of H2SO4 in the reaction causes some sulfation of the ?OH groups despite being in harsh nitrating conditions.   With the nitration conducted in a mixture of a non-swelling (HNO3) and swelling acid (H2SO4), there must be a limit to which nitration can occur homogeneously before the lack of dispersibility or aggregation have an apparent effect on the %N achieved.  A lower loading of NCC per volume of the mixed acid would avoid such effects, however, the cost of producing the same quantity of NNC would increase dramatically.  On the other hand, a higher loading of NCC per volume of the mixed acid would lower the overall cost, but reproducibility may be an issue.  Industrially, the nitration of cotton linters is conducted at a ratio of 1:60 to 1:70, due to the respectively lower density of these glucosidic polymers when processed ? more solution is needed in order to contain and submerge the starting material.  In my experiments, I varied the 36   molar ratio between NCC and HNO3 from 1:30 to 1:90 to determine whether the nitration of NCC requires the same low loading of NCC in the mixed acid of HNO3/H2SO4 or whether the increased density of NCC favors nitration at higher concentrations.  The reaction conditions were specified as follows: a nitration time of two hours, acid mixture of 40:60 HNO3 to H2SO4, and the temperature maintained at 0 ?C.  Visually, the nitration reaction proceeds similarly between samples, with a slight difference in the rapidness of the breaking up of NCC aggregates by mechanical stirring.  As the nitration reaction proceeded, the samples containing a high loading of NCC visibly increased in viscosity whereas nitrations with lower loadings of NCC were less viscous, but all samples showed adequate mixing.  The results demonstrate an interesting property of using NCC as the starting material for NNC ? by EA, the %N of all the solid product obtained from each starting material mixture centered around 10.7%, with no clear difference between the isolated products from the 1:30 and the 1:70 ratio of NCC to HNO3 ( Figure 2-5).  This result reveals that NCC causes it to be less viscous in the nitration reaction, therefore the nitration can proceed efficiently even though the starting material loading is close to twice that of the industrial process.  Moreover, the amount of H2SO4 in the reaction had to be scaled to HNO3 to achieve the same acid ratio throughout the experiment, for example 1:90 NCC to HNO3.  No consumption of the starting material to form side products are seen as the isolated yields for all ratios are approximately the same.  Therefore, the ratio of NCC to HNO3 has little effect on the DS of the NNC within the specific range tested in the experiment.         37    Figure 2-5 ? Plot of %N versus ratio of NCC loading with respect to HNO3, which shows the independence of the degree of nitration to the amount of material added. It is known that by varying the nitration time, the degree of substitution on nitrocellulose can be controlled.  However, the initial nitration of the cellulose polymer chains occurs so rapidly that within the first five minutes, at least one out of three ?OH groups on cellulose is nitrated to a ?ONO2 group.  However, due to the heterogeneity of cellulose in the acidic media, the reaction of the second and third ?OH groups with the nitrating agent occurs at a much slower rate.  In the investigation of the optimal reaction time to achieve maximum nitration using the HNO3/H2SO4 method, the following conditions were used: the ratio of 40:60 HNO3 to H2SO4, a reagent ratio of 1:70 NCC to HNO3, reaction temperature of 0 ?C while the nitration time was varied between 30 mins to 3 hours in 30 min increments.  In 30 mins, the NCC is already homogeneously dispersed in the mixed acid and yields a %N of 10% by CHN EA (Figure 2-6).  The change in %N of the product slowly increases to 11 %N by extending the nitration time to 3 hours.  It is notable, however, that the viscosity of the nitration mixture visibly increased somewhat as the reaction progressed, possibly due to the better dispersity of nitrocellulose in nitric acid as more of the cellulose strands are reacted to form ?ONO2 groups.  In addition, a 24 hour nitration using the same conditions was conducted to determine whether there remains a slow increase in %N by 91011120 10 20 30 40 50 60 70 80 90 100%Nratio of HNO3 to NCC38   prolonging the nitration duration.  Disappointingly, the product obtained after 24 hours has exactly the same DS (11 %N) as that of the 3 hour nitration.    Figure 2-6 ? Plot of %N versus nitration time (HNO3/H2SO4 method) showing a small increase in %N as the reaction duration was increased past 30 mins.  The %N stops increasing after 3 hours of nitration. By varying the HNO3 to H2SO4 ratio, the NCC to HNO3 ratio, and the reaction time, the product reached a maximum of 11 %N.  The remaining parameter that can easily be investigated is nitration temperature.  For each 10 ?C increase in temperature, the reaction rate, including both the forward and backwards reaction, approximately increases two-fold.  This fact is applied to the nitration of aromatic systems, for example the production of nitrotoluene, where a higher reaction temperature accelerates the nitration reaction to completion in several hours; however, the selectivity of the nitration is often decreased as a trade-off.  Most often, the selectivity could be scavenged by using milder nitration methods, such as the mixture of Cu(NO3)2 and Montmorillonite, favoring mono-nitration in comparison with the harsher nitration methods, such as the HNO3/H2SO4 mixture.  In the application to the nitration of NCC, however, the same increase in the nitration rate could not be taken as the sole process benefiting from the increase in the energy input.  The NCC readily degrades and depolymerizes in the presence of acid and thus the rate of irreversible chain scission is also increased.  In light of the delicate balance between increasing the nitration rate and the degradation rate, a nitration experiment was conducted using 891011120 20 40 60 80 100 120 140 160 180 200%NNitration time (min)39   the following set of conditions: ratio of 40:60 HNO3 to H2SO4, ratio of 1:50 NCC to HNO3, nitration duration of 2 hours while varying the nitration temperature between 0 ?C to 60 ?C in 20 ?C increments.  As expected, the increase in temperature also increased the %N found in the nitrated cellulose.  The values of which increased from 10.9%, 11.1% and 11.5% for 0 ?C, 20 ?C and 40 ?C, respectively (Figure 2-7).   However, the isolation of the nitrated product became more difficult as the nitration temperature increased since the product resembled a soft gel instead of a solid.  In the case of nitration at 60 ?C, the isolated mixture did not pass through a medium glass frit even after hours on vacuum suction.  It is highly probable that it contains nitrated oligomers of glucose instead of intact strands of nitrocellulose.  Further increasing the   Figure 2-7 ? Plot of %N versus reaction temperature showing a small increase in %N in the isolated NNC as nitration temperature is increased.  However, significant loss in yield and the increased instability of the product at higher temperatures limits the reaction temperature to 0 ?C. temperature of nitration leads to the complete digestion of the cellulosic material and the reaction color darkens to a brown color indicating degradation of the NCC.  Moreover, the presence of NO2 gas, which is emitted from the nitration mixture upon warming past 30 ?C, has a discoloring effect on the isolated nitrated product.  The isolated solids became increasingly more yellow as either the temperature increased or the nitration time increased and could not be decolorized by washing with water.  As it is known that nitrocellulose is an excellent adsorbent of NO2 gas, the 910111213-5 0 5 10 15 20 25 30 35 40 45%NReaction temperature (oC)40   increase in %N observed in this experiment may be a result of the presence of adsorbed NO2 instead of chemically bonded ?ONO2 groups.   Powder X-ray diffraction was conducted on the starting material NCC and several of the NNC nitrated using the HNO3/H2SO4 method (Figure 2-8).  The NCC is, without surprise, nanocrystalline with diffraction peaks centered around 2? of 15.1?, 16.8?, 19.8?, 22.0?, and 34.9?, which corresponds to the diffraction pattern for cellulose I?.83  The crystallinity was still present after the reaction has proceeded for 30 mins at 0 ?C.  As the nitration duration increased, the gradual decrease in peak intensity of the largest diffraction peaks centered at 19.8? and 22.1? and the replacement of a broad peak with no well-defined maximum was seen.  After 2 hours of nitration, the complete disappearance of the crystalline NCC was confirmed and the large amorphous peak for NNC remains unchanged for longer nitration durations.  The PXRD pattern of the NNC with 11 %N shows only two broad peaks at a 2? of 13.1o and 20.5o; therefore, the  Figure 2-8 ? PXRD patterns showing the loss of crystallinity after the nitration reaction.  The nitration fully disrupts the hydrogen bonding network and forms an amorphous product. 0501001502002503003504004500 10 20 30 40 50 60 70 80Cps2 theta (o)NCCNNC41   crystallinity of NCC has little impact on the nitration process except prolonging the nitration duration as diffusion of the nitration mixture into the crystallite is hindered.   Thus, the nitration of NCC using the HNO3/H2SO4 method is similar to the conventional methods of nitration of cellulose, where a low temperature (0 ?C), relatively short duration (2-4 hours), and a 30-40:70-60 mixture of HNO3 to H2SO4 as the nitrating acid, to produce NNC with a 11 %N.  However, it is seen that more NNC can be produced per volume of the nitrating acid up to a loading of 1:30 NCC to HNO3 without noticeable changes in the %N in the final product due to the compact nature of the starting material.  In addition, the product is a compact free-flowing white powder that can be contained easier than the less dense nitrocellulose derived from cotton linters.  The apparent limit of the %N achievable using the mixture of HNO3 and H2SO4 leads to unreacted ?OH groups or ?OSO3H groups, as determined to be present in the NNC produced by conductometric titration.  Nevertheless, the HNO3/H2SO4 method can reproducibly nitrate NCC to a DS = 2, however, in order to obtain a material that is more energetic, i.e. DS > 2, an alternate method needs to be found.  Optimization of HNO3/P2O5 method.   P2O5 is a well-known solid desiccant that can be used in the places of H2SO4, however, it is much harder to handle as it readily absorbs atmospheric moisture and becomes translucent and tacky.  Yet, an advantage over the method of desiccation of H2SO4 is that P2O5 chemically reacts with H2O instead of forming adducts or hydrates.  P2O5 reacts with H2O to produce poly-phosphoric acids and ultimately H3PO4, which is much more thermodynamically stable in comparison with P2O5.  This effectively eliminates H2O from the reaction completely and thus should provide an improved nitration condition for reacting with NCC.  A drawback, however, is 42   that the mixture of HNO3 and H3PO4 does not swell cellulose as well as the HNO3/H2SO4 mixtures.  This is seen in the extreme aggregation of the starting material NCC when placed in the HNO3/P2O5 mixture; visually, the NCC amasses into globules, which prevents the proper dispersion as the center of the mass remains untouched by acid.  Often, mechanical disaggregation is required to facilitate the nitration process.  A comparison of nitration between a sample that was left to stir and a sample that was mechanically broken-up to facilitate dispersion show that a higher degree of nitration can be achieved using mechanical agitation to allow for better homogeneity.  Despite this, the validity of the nitration method with HNO3/P2O5 was evaluated and optimized according to the NCC to HNO3 ratio, reaction time, and reaction temperature.  The HNO3 to P2O5 ratio was not investigated, however, since the method of preparation of the nitrating acid calls for complete saturation of the HNO3 by drying the 90% fuming nitric acid in excess P2O5, then filtering off the remaining solids.  The resulting mixture is proposed to have a composition of 64% HNO3, 26% H3PO4 and 10% dissolved P2O5.84  In addition, due to the nature of the desiccant, an experiment was conducted where additional P2O5 was added half-way through the nitration process in order to remove the H2O that is produced in the nitration process.    The preparation of the HNO3/H3PO4/P2O5 mixture is a tedious process and proves to be extremely dangerous due to potential of a run-away reaction from the heat generated from the dehydration by P2O5.  Tremendous amounts of NO2 fumes are given off if the reaction is not immediately cooled and the final acid mixture contains less active nitrating agent due to HNO3 decomposition.  In order to prevent the highly exothermic drying of the 90% HNO3 by solid P2O5 from overheating the reaction, the solid P2O5 must be added in small portions to ensure proper dissolution and to ensure the temperature does not rise above room temperature.  The 43   preparation of the acid mixture at 0 ?C proves to be facile, although complete saturation and drying of the fuming nitric acid occurs only after 3 to 4 hours of vigorous stirring.  After filtration through glass wool to remove excess solids, the resulting acid mixture of HNO3 and H3PO4 is a clear light yellow solution at 0 ?C, indicating that HNO3 and H3PO4 are completely miscible at this temperature.  When the mixture is warmed to room temperature, precipitation occurs where dissolved P2O5 and some polyphosphoric acids becomes less soluble.  This process is completely reversible ? the homogeneous solution reforms when the mixture is cooled again to 0 ?C.    Similar to the addition of P2O5, the addition of NCC to the nitration mixture also needs to be done cautiously as the NCC aggregates severely in the non-swelling mixture of HNO3 and H3PO4.  The sudden breaking up of the NCC aggregates can cause localized overheating.  However, once the nitration process in underway, the partially substituted NCC becomes better dispersed, which eventually leads to a homogeneously nitrated product.  Preliminary results using the HNO3/P2O5 nitration method showed that the non-degrading and non-sulfating desiccant serves to promote further nitration of NCC than the HNO3/H2SO4 method ? EA revealed that the NNC contained 12.5 %N.  In investigation, the NCC to HNO3 ratio was varied between 1:75, 1:50 and 1:35 to determine whether the increased loading of NCC will ultimately result in a lower nitrated product.  Meanwhile, the acid ratio between HNO3 and H3PO4 was unmodified and the nitration was conducted at 0 ?C for 24 hours. Visually, it is clear that dispersion of NCC at a large concentration in the 1:35 sample is slow and only after 30 mins of stirring did all the solids evenly distribute in the reaction.  However, the three different NCC loadings do not show significant differences in their ability to nitrate since the products contained 12.5 %N, 12.8 %N and 12.7 %N, respectively, for 1:75, 1:50, and 1:35 ratios of NCC 44   to HNO3 (Figure 2-9).  This result is surprising since without H2SO4 as desiccant, the volume of the nitrating acid is nearly half that of the HNO3/H2SO4 method, so one might expect that the increase in viscosity at high concentrations of NCC would impact the efficiency of the nitration.  Moreover, because of the increased NCC loading in the sample with 1:35 ratio of NCC to HNO3, after overnight stirring there was phase separation between the product NNC and the rest of the nitration mixture as evident by the gelatinous substance that is pushed towards the sides where the magnetic stir bar does not stir adequately.  Overall, because of the crystalline nature of the starting material, more NCC can be loaded into a set volume of nitrating acid without much effect on the %N in the product produced.  Figure 2-9 ? Plot of %N versus the ratio of HNO3 to NCC showing no dependence of the DS in the synthesized NNC on the amount of NCC added to the reaction.     Due to the inability of the HNO3/P2O5 mixture to swell the NCC, a much larger range of time dependence of nitration was conducted.  Based on the slower dispersion of NCC in the acidic media, it was expected that the initial induction period to reach maximum nitration would be longer than that of the HNO3/H2SO4 method, which was approximately 1.5 hours after NCC addition.  The nitration of NCC was conducted under the following conditions: a 1:50 NCC to 1112131425 35 45 55 65 75 85%Nratio to HNO3 to NCC45   HNO3 ratio nitrated for the specific time at 0 ?C.  In addition, a control experiment utilizing Whatman filter paper was conducted to see the difference between the nitration of a purely nanocrystalline material, in the case of NCC, and the nitration of a partially crystalline material, in the case of the filter paper.  The nitration duration was varied from 2 hours to 10 hours in 2 hour increments in the first set of experiments and a broader time range was conducted from 5 mins to 24 hours for the second experiment in order to determine when the maximum nitration could be achieved.  For NCC, the %N determined by CHN EA after 2 hours of nitration in the HNO3/P2O5 mixture was 11.8% (Figure 2-10).  Figure 2-10 ? Plot of %N versus nitration duration showing the shorter time needed when nitrating filter paper for the %N to plateau.  The amorphous regions in filter paper are easier to nitrate due to the lack of extensive hydrogen bonding between cellulose chains. Subsequent increases in nitration duration do not increase the %N, which varies between 11.5% to 12%.  All the samples appeared well-dispersed in the nitration mixture after 30 mins of stirring and remain homogeneous throughout the nitration procedure.  In comparison, the 678910111213140 5 10 15 20 25%NNitration time (hr)Filter paper NCC 46   nitration of the filter paper also shows that the maximum nitration of the material can be achieved within 1 hour of reaction, however the EA shows that the filter paper can be readily nitrated to %N of 12.8%, which is nearly 1% more than what is achievable through the nitration of NCC.  Analysis of the 2 hour, 4 hour, and 9 hour samples of nitrated filter paper shows that the value obtained at 1 hour of nitration was already the maximum %N obtainable.  The resulting nitrated product is a white sheet-like material, resembling that of the starting filter paper, with added fluffing due to the intrusion of nitric acid between the fibers.  The structural differences, for example surface ?OSO3H, and morphological differences between filter paper and NCC contributes the differences in %N observed.  Unexpectedly, filter paper nitrated for only 5 mins contained 12.7 %N, and this is maintained throughout the 10 min, 15 min, 20 min, 30 min, and 45 min time points (Figure 2-10).  Similarly, the shorter nitration times test on NCC also revealed a very quick increase in the nitrogen content beginning around 30 mins into the nitration reaction. This slower response is attributed to the increased crystallinity of NCC compared to filter paper, which require more effort to react completely.  This result further reinforces the importance of chemically removing water from the reaction as the rate of reaction is increased in comparison with the HNO3/H2SO4 nitration method.    In addition, the effect of reaction temperature was investigated to determine whether a higher temperature could be used for the nitration of NCC without degrading the starting material.  Samples of NCC were nitrated using the following conditions: a 1:50 NCC to HNO3 ratio, a reaction duration of 2 hours, and the temperature was varied between 0 ?C to 60 ?C in 20 ?C increments.  Again, at the lower temperatures between 0 ?C and 20 ?C, the resulting product is not discolored by the increasing presence of NO2 gas, whereas the high temperatures of 40 ?C and above causes the decomposition of the nitrating acid and NO2 gas is adsorbed onto the NNC 47   product.  In total, CHN EA analysis resulted in a %N of ~12.5, shows that the nitration temperature has little effect on the DS of the final product (Figure 2-11).  However, nitrations conducted at higher temperatures led to lower isolated yield.  Therefore, a lower nitration temperature is recommended for preserving the maximum nitrogen content and also maximizing the yield.  In addition, the presence of the adsorbed NO2 gas seems to have a large effect on the stability of the isolated dry NNC powder as yellow fumes are emitted from the solids during storage.    Figure 2-11 ? Plot of %N versus reaction temperature showing no effect on increasing the DS of NNC as temperature was increased.  The increase in temperature does not increase the effectiveness of the removal of water by P2O5, however, the loss in yield and instability of the product at higher temperatures favors nitration at 0 ?C.  Due to the mode of water removal by the desiccant, it is thought that having slight excess of P2O5 in the reaction would help more efficiently remove the water generated from the nitration reaction and prevent atmospheric moisture from interfering with the reaction.  A set of experiments was conducted to compare the effects on the %N of the final NNC product by the addition of excess P2O5.  The samples were treated with the following conditions: a 1:50 NCC to HNO3 ratio, a nitration duration of 4 hours, and a reaction temperature of 0 ?C.  In addition, the nitrating acid was kept at the same composition with 64% HNO3, 26% H3PO4 and 10% dissolved 11121314-10 0 10 20 30 40 50 60 70%NReaction temperature (oC)48   P2O5, but in the experiment, 0.5 grams or 1.0 grams of solid P2O5 was added to the nitration reaction after 30 mins of reacting.  The resulting P2O5 does not readily dissolve in the HNO3/H3PO4/P2O5 mixture, but settled on the bottom of the reaction.  Nevertheless, re-addition of P2O5 into the reaction after some of the P2O5 from the initial mixture has been consumed does not have a pronounced effect on the %N found in the final product ? the original nitration yields a product of 12%, whereas the two replicates of re-addition yields a %N of 12.1% and 12.2% (Table 2-3).  This shows that the 10-15% suspended P2O5 in the initial mixture is sufficient to sequester the water produced from the nitration reaction but also that the presence of excess P2O5 does not interfere with nor have much effect on the general nitration with HNO3/P2O5.   Table 2-3 ? Effect of the readdition of P2O5 into the nitration mixture in 0.5 g portions on %N in NNC.  The lack of an effect demonstrates the initial amount of desiccant is sufficient in the removal of initial water content and also the water generated during the nitration.  %N Initial 12.0 + 0.5 g P2O5 12.1 + 1.0 g P2O5 12.2   In light of making the nitration process viable for industrial applications, reproducibility at higher scale is crucial.  A comparison between the production of 1 g and 7.5 g of NNC was conducted.   On scaling up reactions, one needs to worry about effective mixing and temperature control throughout the mixture.  Thus, the bulk of the reaction relies on both adequate diffusion of heat by mechanical stirring and the even distribution of reagents.  In the case of nitrating NCC, heat dissipation is important in this reaction to prevent localized heating, which could reduce product yield or lead to an explosive reaction.  In addition, adequate stirring is required to ensure relatively homogeneous nitrating conditions.  The comparison was thus conducted under 49   the following conditions: a 1:50 NCC to HNO3 ratio, a nitration duration of 4 hours, and the nitration temperature maintained at a constant 0 ?C.  At the lower scale of nitration, a %N of 12.7% to 12.8% by CHN EA is consistently produced over five replicates, whereas at the higher scale of nitration, a %N of 11.7% to 12.1% is consistently achieved over five replicates (Table 2-4).  Table 2-4 ? %N obtained from EA showing difference and reproducibility between normal and scaled-up nitration reactions.  The inability for HNO3 and P2O5 to quickly disperse NCC leads to the formation of aggregates of lesser nitrated material.  Normal scale (%N) Scale-up (%N) Replicate 1 12.8 11.8 Replicate 2 12.8 12.0 Replicate 3 12.7 11.8 Replicate 4 12.8 11.7 Replicate 5 12.7 12.1  Clearly, there are two properties immediately noticeable when the reaction is scaled up.  First, there is the difference between the %N achievable, where the lower scale had a higher %N.  Second, there is a greater variance in the nitrogen content of the large-scale nitrations.  The difficulty of stirring the large-scale reaction led to aggregates that were not broken up and thus a less homogeneous nitration occurred in comparison with the small-scale reactions.  Therefore, care must be taken in choosing an efficient mechanical stirrer, for example a Teflon-coated stainless steel overhead stirrer, in order to ensure the NCC is completely submerged in the nitrating acid and also supply enough shear force to break up any aggregates that are formed when NCC is first added.   50    Industrially, it is possible for the nitrated material to be subjected to a second or even a third nitration cycle before it is isolated and dried.  As seen in the semi-continuous cellulose nitration process that Bofers-Nobel-Chematur uses, either the spent acid is reconstituted with new desiccant and then reused, or fresh nitrating acid is used after the spent acid is centrifuged off.  In order to mimic the renitration conditions, the larger scale NNC product with a %N of ~12% was subjected to fresh HNO3/H3PO4/P2O5 nitration mixture after the spent acid was filtered off.  The conditions used are identical to that of the scale-up experiment described previously.  It is found by EA that the %N of the starting NNC can be increased to 12.4% after one nitration cycle in fresh nitration mixture (Table 2-5).  Disappointingly, doubling the amount acid used in total only serves to slightly increase the %N of the nitrated NCC product.  Furthermore, a third renitration step only increases the %N by a further 0.1%, showing there is a limit to which this method will work to increase the quality of the product.  Therefore for the case of replenishing the desiccant in the reaction, the addition of more P2O5, as mentioned before, has little effect on increasing the %N found in the nitrated product and the renitration is too costly for such minor improvements.  Table 2-5 ? Nitrogen content of NNC following multiple nitration steps.  The initial renitration is effective in boosting the %N, however, further reaction does not provide a means of obtaining a result comparable with the smaller-scaled reactions.  %N Initial 12.0 Renitrate 1 12.4 Renitrate 2 12.5 Renitrate 3 12.6  51    Overall, the HNO3/P2O5 method can be used to nitrate NCC at a relatively high loading of starting material due to the denser nature of NCC.  The reaction conditions are similar in that a 1:50 NCC to HNO3 ratio is used to ensure proper dispersion, but since H2SO4 was not used as the desiccant, the total volume of the nitration mixture makes it less likely for large amounts of NCC to be rapidly dispersed if the reactions were scaled up.  A reaction duration of 4 hours and a temperature of 0 ?C is optimal since it ensures proper dispersion of NCC into the nitrating acid and that the reaction does not overheat and generate unwanted NO2 gas, which has a detrimental effect on the stability of the NNC product.  In addition, the heterogeneous nature of the desiccant causes it to have little effect when added in excess since saturation of the initial nitrating acid is achieved when the mixture is made.  Therefore, there is a limit to which the %N can be obtained in the NNC produced as the parameter that has the most effect ? the HNO3 to desiccant ratio ? could not be fine-tuned in the HNO3/P2O5 method.    Optimization of HNO3/Ac2O method.   In the search of replacing P2O5 as desiccant with one that is freely miscible with HNO3 in order to determine whether the ability to vary the desiccant plays a crucial role in obtaining a nitrated NNC containing a high %N, liquid desiccants are reconsidered.  The mode of action of P2O5 is the reaction with H2O to generate polyphosphoric acid when the quantity of water is low and ultimately phosphoric acid, H3PO4, when the quantity of water is high.  One analogous compound that is commonly used in synthesis is acetic anhydride (Ac2O), which reacts with water to give acetic acid.  This would serve as good desiccant comparable with H2SO4, which has issues with the sulfating of the surface of NCC, and with P2O5, whose potential is held back by its limited solubility, as the amount of water in the reaction can be precisely controlled.  In 52   addition, the ability of Ac2O to react with HNO3 to form a more reactive nitration agent AcONO2 and also the protonated analogue AcHONO2+, serves to facilitate the nitration process on NCC.    It is known that cellulose fibers swell rapidly in AcOH85 and it was observed that when NCC was added to Ac2O and AcOH, it is quickly and easily dispersed into a homogeneous suspension.  During the variation of 20% to 100% Ac2O in the nitration reaction, the dispersibility of NCC remained good, although differences in the ease of dispersion were observed.  As expected, acid mixtures containing more Ac2O swell NCC better than those containing more HNO3.  In light of the better dispersion of NCC in HNO3/Ac2O than in HNO3/P2O5, the generation of a more reactive nitrating agent, and the ability to control the amount of desiccant, the HNO3/Ac2O method was optimized according to the ratio of HNO3 to Ac2O, the ratio of NCC to HNO3, the reaction duration, and the reaction temperature.  However, first an investigation of the stability of the nitration mixture was conducted to ensure safe usage.  To investigate the stability of the nitration mixture of HNO3 and Ac2O, 2:3 volume ratio, at various temperatures during preparation, the addition of Ac2O through an addition funnel to a prechilled or preheated HNO3 was conducted at room temperature (~20 ?C), 0 ?C and -30 ?C to determine whether the preparation of the nitration mixture is sensitive to changes in temperature (Table 2-6).  The preparation of the HNO3/Ac2O nitration mixture at room temperature is highly hazardous as the initial addition of Ac2O to HNO3 caused bubbling and immediate formation of a yellow solution and orange fumes.  It is clear that the high temperature in addition to the exothermic nature of the dehydration reaction causes extreme localized heating and boiling of the HNO3 was seen, which indicates that the reaction temperature has reached higher than 85 ?C internally.  Further addition of Ac2O caused a more vigorous reaction and, as a result, a clear yellow solution was obtained with a large amount of NO2 fumes.  Cooling of the reaction 53   mixture to 0 ?C did not serve to redissolve the NO2 fumes.  This nitration mixture is postulated to have less nitrating ability as the high temperature lowers the solubility of NO2 gas in the HNO3 and also simultaneously decomposes some of the acid.  On the other hand, preparation of the HNO3/Ac2O mixture at 0 ?C yields a yellow solution immediately upon addition of Ac2O to HNO3.  By maintaining the temperature at 0 ?C through proper mechanical stirring and the slow addition of Ac2O, a clear yellow nitration mixture is obtained, which was free of NO2 fumes.  However, at -30 ?C, which was maintained by a 30:70 EtOH:ethylene glycol dry ice bath, the slow addition of Ac2O to the reaction results in a colorless mixture.  However, increasing the addition rate causes a yellow discoloration to appear in the HNO3/Ac2O mixture, indicating the formation of NO2 gas.  This is caused by the localized increase in temperature as the 90% HNO3 is immediately dehydrated by the Ac2O in a highly exothermic process.  The discoloration subsides after the first portion of Ac2O was added.  If the initial addition was kept to a minimal rate for the first 20% of the Ac2O addition, and a more rapid rate for the remaining 80%, a clear and colorless HNO3/Ac2O mixture can be achieved at -30 ?C.  This HNO3/Ac2O mixture is postulated to have better nitrating ability than those prepared at higher temperatures.   Table 2-6 ? Visual comparison of HNO3/Ac2O preparation at different temperatures and the respective %N measured for the NNC.  Results suggest a low temperature is needed to stabilize the more reactive nitrating agent to avoid decomposition of the mixture.  Color NO2 %N -30 ?C Colorless None 13.2 0 ?C Yellow None 12.9 20 ?C Yellow Orange 7.8       The HNO3/Ac2O nitrating acids from the -30 ?C, 0 ?C and 20 ?C preparations were used to nitrate NCC at 0 ?C to compare their ability to produce NNC.  It was found that the nitrating 54   acid prepared at -30 ?C yields NNC with the highest %N at 13.2% by CHN EA, followed by the acid mixture prepared at 0 ?C, which yields NNC with a %N of 12.9%.  As expected, the nitrating acid prepared at room temperature yields NNC with the lowest %N of 7.8%.  Since within the acidic media, both nitration and acetylation can occur, with a reduction of the nitrating ability of the acid medium, the less reactive acetylation reaction can compete and affect the degree of nitration.  Since there is an effect of the amount of HNO3 contained in the nitration mixture on the %N achievable in the NNC product, a series of experiments was conducted to determine the optimal amount of HNO3 required to reach maximum nitration.  The HNO3 was varied from 20 v% to 100 v% in 20% increments and the following conditions were used: a 1:70 NCC to HNO3 ratio, a nitration time of 4 hours, and a reaction temperature at 0 ?C.  The following figure shows the %N measure for samples of NNC prepared with different HNO3/Ac2O ratios (Figure 2-12).  It is seen that the lack of HNO3 present in the reaction has a greater effect on the degree of nitration than when HNO3 is present in excess from the optimal ratio.  In comparison, the NNC produced using HNO3/Ac2O prepared at room temperature, which has 7.8 %N, falls in between that resulted from the 20 v% and the 40 v% samples.  Hence, it is clear that the appearance of the yellow color and also the degassing of NO2 fumes is detrimental to the efficiency of nitration.  From extrapolation, approximately 15 to 17% of the initial HNO3 is lost to decomposition; it is evident that a low temperature, preferably -30 ?C, is required for maximizing the nitration potential of the HNO3/Ac2O mixture.  Nevertheless, the %N obtained from CHN EA of the different HNO3/Ac2O mixtures show a clear trend ? a gradual decrease in the nitrating ability of the mixed acid is seen when the HNO3 is increased from 40 v% to 80 v%.   55    Figure 2-12 ? Plot of %N versus the v% of HNO3 in the nitration showing an optimal mixture at 40 v% HNO3 and yielding a NNC product with 13.4 %N.  The nitrating mixture is ineffective in nitrating NCC when the amount of Ac2O was increased from 60% to 80%. The generally linear decrease of 0.6% nitrogen per 20 v% increase in HNO3 content can be correlated to the amount of H2O remaining in the reaction that is not dehydrated from the limited Ac2O.  The effect of the residual H2O and the production of H2O from the nitration reaction is reflected on the lowering of the %N obtained for the nitration of NCC.  However, a large drop in the nitrogen content, i.e. 1%, attained from using 100 v% HNO3 as the nitrating acid is due to the non-linear effect of having water in the nitrating acid.  Without the desiccant, the ability of HNO3 to self-ionize and generate the nitrating agent, NO2+, is extremely low.  As for the 20 v% HNO3 sample, an interesting effect is seen in that the quantity of HNO3 in the reaction is not enough to nitrate the cellulosic material completely.  Also, there is competition between the nitration reaction and the acetylation reaction due to the overwhelming presence of AcOH and Ac2O together.  The maximum degree of nitration is obtained by mixing 60 v% of Ac2O into 40 v% HNO3 at -30 ?C over a period of 30 mins, then nitrating NCC for 4 hours.   468101214160 10 20 30 40 50 60 70 80 90 100%Nv% HNO356    The loading of NCC into the HNO3/Ac2O nitrating mixture was varied between 250 mg to 1 g in 125 mg increments, which corresponds to 1:21 to 1:80 ratio of NCC to HNO3.  As a result of the swellability of NCC in Ac2O and AcOH mixtures, which quickly forms a homogeneous mixture upon addition of the starting material, the %N in the NNC product should be independent of the amount of NCC initially added.  Within the different NCC loadings tested, only the sample containing 1 g of NCC showed an obvious, though slight, increase in viscosity when all the solids are in the reaction.  As the HNO3/Ac2O method uses a nitrating mixture that is more voluminous than that of the HNO3/P2O5 method, no observable increase in viscosity resulting from the swelling of the nitrated cellulose fibers was observed.  This, in addition to the rapid dispersibility, led to a NNC product that shows the same degree of nitration (13.3-13.4 %N) throughout the entire series of NCC loadings (Figure 2-13).  The low variability in nitrogen content seen in the NNC samples as the NCC starting material loading was changed is very encouraging.    Figure 2-13 ? Plot of %N versus ratio of HNO3 and NCC showing independence of the %N on the NCC loading.  The insensitivity is due to the compact nature of NCC compared to bulk cellulose fibers.  The time dependence of the nitration using the HNO3/Ac2O method was determined by nitrating NCC with the following conditions: a ratio of 2:3 HNO3 to Ac2O and a ratio of 1:50 NCC to HNO3 at 0 ?C.  The nitration time was varied from 5 mins up to 6 hours.  The results 12.51313.5140 10 20 30 40 50 60 70 80 90%Nratio of HNO3 to NCC57   found are interesting since one would expect the presence of a more reactive nitrating agent, AcONO2 or AcHONO2+, would decrease the amount of time required before the NCC is fully nitrated in comparison with the HNO3/P2O5 procedure.  However, using HNO3/Ac2O required a little longer (~1 h) to reach maximum nitration when compared to the other methods (~30 mins).  The following figure shows the nitrogen content of the NCC as a function of time in HNO3/Ac2O (Figure 2-14).  Figure 2-14 ? Plot of %N versus nitration duration showing maximum nitration can be reached in one hour of reaction and remains stable for upwards of six hours.  The induction time is similar to that of the HNO3/H2SO4 and HNO3/P2O5 methods, but a much higher degree of nitration can be achieved with this method. The %N of the NNC synthesized from the HNO3/Ac2O method continues to lag behind that of the HNO3/P2O5 method until 20 mins into the reaction, where the NCC continues to be nitrated further up to an hour until the %N stabilized at 13.3%.  In addition, a 24 h nitrated sample using HNO3/Ac2O as nitrating acid was investigated, but there was no significant change in nitrogen content even though the exposure to the nitrating mixture was prolonged.  It appears that the 678910111213140 1 2 3 4 5 6 7%NNitration time (hr)58   mixture of HNO3/Ac2O generates a more reactive nitrating agent but in lower quantity than the HNO3/H2SO4 method, leading to a slower overall reaction rate.  To investigate temperature effects, the nitration temperature was conducted at 0 ?C, 20 ?C, 40 ?C and 60 ?C while the other parameters used were the following: a 2:3 ratio of HNO3 to Ac2O, a 1:50 ratio of NCC to HNO3, and a nitration time of 3 hours.  Similar to what was found for the HNO3/H2SO4 and HNO3/P2O5 mixtures, the nitrations of NCC at 20 ?C and 0 ?C yield the same extent of nitration (13.2 %N), but at 20 ?C there is noticeable yellowing of the reaction mixture, which is reversible by the cooling with an ice-water bath (Figure 2-15).  Therefore, it   Figure 2-15 ? Plot of %N versus reaction temperature showing the gradual decrease of %N as reaction temperature is increased.  The instability of the nitrating agent at temperatures higher than 0 ?C limits the use of increased temperatures to alter the rate of nitration. seems that the nitrating agent, AcONO2, is stable to upwards of 20 ?C after it has been formed and that significant decomposition of the nitrating acid is not observed.  However, when the temperature of the reaction was maintained at 40 ?C, the yellowing of the acid mixture was not reversible with cooling as some of the NO2 fumes escaped the solution.  The %N by EA also reflects the loss of some nitrating ability as the %N of the NNC synthesized at 40 oC is 12.9%, however, the loss is less significant than the effect obtained from increasing the temperature 1212.51313.514-10 0 10 20 30 40 50 60 70%NReaction temperature (oC)59   during acid preparation.  Similarly, the sample nitrated at 60 ?C shows an even lower %N of 12.6%.  Overall, the increase in temperature has a negative effect on the nitration of NCC using HNO3/Ac2O.    Nitration of NCC produces one molecule of water per ?ONO2 group installed on the material, subsequently consuming Ac2O in the process.  In order to keep the Ac2O concentration approximately constant through the reaction to determine whether the gradual consumption of Ac2O prevents complete nitration, another desiccant was added to the reaction prior to nitration of NCC.  P2O5 was the ideal choice as it is near insoluble in the acid mixture and that it is a better desiccant than Ac2O as demonstrated in the rapidness to reach the maximum %N in the time dependent nitration experiments in HNO3/P2O5 mixtures.  There is no reaction upon addition of P2O5 and the solid is a free-flowing white powder within the HNO3/Ac2O mixture, indicating the effectiveness of the initial reaction of Ac2O and the water from the 90% HNO3 and that water is not present in the reaction before NCC was added.  However, only a 13.1% nitrogen was obtained in the NNC product.  There seems to be a detrimental effect of having P2O5 in the reaction mixture, which may explain the seemingly less effective nitration seen with the HNO3/P2O5 method that produces a maximum 12.8 %N.  It is postulated that the H3PO4 produced in the reaction may interfere with nitration by increasing the viscosity of the reaction, which slows the diffusion of heat, nitrating agent, and the removal of water.   In addition, by preferentially occupying the space from the swelled fibers, H3PO4 decreases the rate of which the nitrating agent and water can diffuse into and out of NCC.    All in all, the nitration of NCC using the HNO3/Ac2O method was optimized to a 2:3 ratio of HNO3 to Ac2O, a 1:50 ratio of NCC to HNO3, and a nitration time of 3 h at 0 ?C.  The initial acid mixture should to be prepared at -30 ?C for best results due to the instability of the 60   nitrating agent above 0 ?C.  Although the initial rate of nitration in the HNO3/Ac2O method is slower than that of the HNO3/H2SO4 and the HNO3/P2O5 methods, a much higher %N is achievable with HNO3/Ac2O.  This may be due to a combination of factors including better removal of water due to miscibility of the desiccant, better heat transfer due to lower viscosity, and the in situ generation of a more reactive nitrating agent.  A representative scale-up of the nitration reaction with HNO3/Ac2O shows that this method is readily able to increase in scale without affecting the %N in the synthesized product (Table 2-7).  The addition of NCC to a HNO3/Ac2O at five times the volume does not show any differences in the rate of dispersion and homogeneity of the mixture afterwards.  By EA, the %N achievable averages 13.3% over four replicates, which is the same as the %N obtainable by the smaller scale nitration.  Nitration of NCC using the HNO3/Ac2O method at larger scales in reproducible and may be good for industrial applications.  As a comparison, the nitration of micro-granulated cellulose (MGC) using the HNO3/Ac2O method proves to be just as facile.  The maximum %N obtained for the nitration of MGC is 13.4%, the same as that of NNC.   Table 2-7 ? Repetitions of nitration showing the differences and reproducibility of the normal and scaled-up HNO3/Ac2O nitration method on NCC and MGC.  There is little difference in the product between different reaction scales and also between materials of different crystallinity.    NCC MGC  Normal scale (%N) Scale-up (%N) Scale-up (%N) Replicate 1 13.1 13.4 13.3 Replicate 2 13.4 13.3 13.4 Replicate 3 13.5 13.1 13.2 Replicate 4 13.2 13.3 13.3  61   Comparison of the NNC products.   Based on the optimized methods, NNC can be synthesized to give a %N of ~11.0% using the HNO3/H2SO4 method (DS = 2.0), a %N of 12.0-12.8% using the HNO3/P2O5 method (DS = 2.3-2.5), and a %N of 13.2-13.4% using the HNO3/Ac2O method (DS = 2.7).  Progressively, the usage of a non-sulfating miscible desiccant, in this instance Ac2O, provided a facile route to nitrate NCC in one step to yield a highly energetic material in good yield; however, the difference in properties are not immediately apparent by looking at the values obtained from CHN EA.  Typically, a good indication for the presence of ?OH moieties in the compound is to look for a band situated at around 3200-3600 cm-1 by FTIR spectroscopy.  The shape of the peak at this region can provide information about the nature of the ?OH group ? a broad band without the presence of fine structure indicate an alcohol moiety, such as that in EtOH or H2O, and a sharp band with a clear maximum indicate a hydroxide moiety, such as that in NaOH.  In the instance for NCC, 2-3 overlapping peaks represent the three different ?OH environments on each glucose repeating unit.  The peaks are not as broadened as that in ethanol due to the inherent crystallinity of NCC.  Therefore, the change in shape of and the reduction of the ?OH band in NCC is a good indication that modification on either the surface or the entirety of the NCC is successful.  In addition, ?ONO2 groups should show three additional bands at 1640 cm-1, 1274 cm-1 and 823 cm-1 assigned to the asymmetric vibrational modes of the nitro group.86  However, it is hard to distinguish samples that are highly nitrated and moderately nitrated using the ?ONO2 bands since they are intense.  Therefore, the weaker ?OH band is a better indication of the degree of nitration.    In the HNO3/H2SO4 method, it is evident by FTIR that after 30 mins of reacting in nitration mixture, the ?OH band has not fully diminished, indicating the persistence of bound 62   alcohol groups (Figure 2-16).  Further nitration to 3 h serves to reduce the ?OH band slightly more, but ?OH groups are still present in the NNC product.  This observation is reflected in the CNH EA, where it suggests that one in three alcohol groups on each glucose repeating unit is still present.  As for the products isolated from the HNO3/P2O5 method, it is evident that the reduction of the ?OH band occurs much more dramatically as the nitration mixture is approximately 1.25 times more efficient than that of the HNO3/H2SO4 method to yield a product of DS = ~2.4 (Figure 2-17).  Moreover, as the material loses its crystallinity, the individual peaks seen in NCC begin to broaden and become indistinguishable from each other.  This is seen as the shift in wavenumbers from the 3330 cm-1 to 3475 cm-1.     Figure 2-16 ? FTIR spectra showing the gradual reduction of the ?OH band by increasing nitration time in HNO3/H2SO4.  However, the complete removal of the band is not observed.  2030405060708090100110120500100015002000250030003500%TWavenumbers (cm-1)30 min60 min90 min120 min150 min180 min?OH 180 min 150 min 120 min 90 i  60 i  30 in 63    Figure 2-17 ? FTIR spectra showing the gradual reduction of the ?OH band by increasing nitration time in HNO3/P2O5.  Most of ?OH groups have disappeared after six hours of reaction, but some traces remain.  Figure 2-18 ? FTIR spectra showing the gradual reduction of the ?OH band by increasing nitration time in HNO3/P2O5.  The complete loss of the band is observed after one hour of nitration.   0204060801001205001000150020002500300035004000%TWavenumbers (cm-1)2 h4 h6 h8 h10 h20304050607080901001101205001000150020002500300035004000%TWavenumbers (cm-1)5 min10 min20 min35 min60 min120 min180 min240 min300 min360 min?OH 10 h 8 h 6 h 4 h 2 h 360 min 300 min 240 min 180 min 120 min 60 min 35 min 20 min 10 min 5 min ?OH 64   With the HNO3/Ac2O method, where a DS of 2.7 was achieved, the absence of the ?OH band is clear in the FTIR analysis.  A gradual reduction of the ?OH band at 3330 cm-1 was observed as the nitration time was extended (Figure 2-18).  However, one would expect that the ?OH signature of the remaining un-nitrated alcohol moieties to still be present by FTIR spectroscopy and the disappearance of the band can be rationalized by the presence of ?OSO3H groups, which take the place of ?OH groups and are not seen in the FTIR analysis.  Another possible explanation is that the sensitivity of the FTIR is not adequate in detecting the small amounts of ?OH groups remaining since it is overshadowed by the intense ?ONO2 bands.  Nevertheless, even though FTIR could not be used to quantify the degree of nitration of NCC, this analysis can be used for monitoring the progression of the reaction by the presence or the absence of the ?OH band at 3330 cm-1.  NC is known to be an energetic material due to its ability to deflagrate, thus releasing heat and energy rapidly upon degradation without the velocity exceeding that of the speed of sound.  Mono-nitrated cellulose, di-nitrated cellulose, and tri-nitrated cellulose should in theory demonstrate different deflagration temperatures. Thermal gravimetric analysis, or TGA, is commonly used to monitor the weight loss of a sample while heating at a constant rate and is ideal for determination of the deflagration temperature of NNC.  Usually, industrial grade nitrocellulose has a deflagration temperature of >180 ?C when heated in air.  The deflagration of such materials is extremely fast.  When the NNC materials produced using the HNO3/H2SO4, HNO3/P2O5, and HNO3/Ac2O methods are analyzed by TGA, the results show that the product made from the first method has a deflagration temperature of 140-150 ?C and is the lowest in comparison with the other two methods (Figure 2-19).  This value is substantially lower than the temperature recommended for stable NC.  Also, residual mono-nitrated NNC deflagrated at its 65   own temperature 40-50 ?C above that of the first deflagration temperature.  This difference is a result of the higher thermal stability of mono-nitrated cellulose.  A gradual decrease in the quantity of mono-nitrated NNC is seen as the nitration duration was lengthened.  The second method, using HNO3/P2O5 as the nitration mixture, yields a product that has a deflagration temperature of 213-215 ?C, which is significantly higher than that of the HNO3/H2SO4 method.    Figure 2-19 ? TGA comparison of the deflagration temperature for the NNC synthesized by the three methods.  A much lower temperature was observed for the HNO3/H2SO4 method caused by the presence of ?OSO3H groups; a stable temperature of >200 ?C was observed for the other two methods. Although by EA there is more nitrogen in the compound, a higher deflagration temperature is observed.  This can be rationalized by the presence of fewer sulfate groups in the isolated product as sulfate groups are known to destabilize NC.77  As for the third method using HNO3/Ac2O, a deflagration temperature of 213-219 ?C is obtained, which is similar to that of the HNO3/P2O5 method although the %N in the third method is on average 1% higher than the 0204060801001200 50 100 150 200 250 300Weight %Temperature (oC)H2SO4P2O5Ac2OH2SO4 P2O5 Ac2O 66   second method.  The di-nitrate and tri-nitrate have similar thermal stability, so the same deflagration temperature was observed for both.    2.4 Conclusion   Generally, NCC is found to behave similarly as bulk cellulose when nitrated using the three methods of nitration ? HNO3/H2SO4, HNO3/P2O5 and HNO3/Ac2O.  The main difference was that the cellulose loading can be varied quite dramatically, from 1:70 to 1:30 ratio of NCC to HNO3, without affecting the overall efficiency of the nitration mixture.  This property is highly desirable since more NNC can be produced within the same reaction volume in comparison with the traditional cellulose nitration and thus is more economical.  The optimized conditions are similar between the three methods in that a nitration duration of 3 to 4 hours at a temperature of 0 ?C is desired for maximizing the nitrogen content and yield of the nitrated product.  The acid composition varied slightly between each method, where the optimized conditions used 30-40:70-60 HNO3 to H2SO4, saturated P2O5 in HNO3 and 40:60 HNO3 to Ac2O to yield NNC with %N of 11.0%, 12.0-12.8% and 13.2-13.4%, respectively.  The key to a proper nitration of NCC is a desiccant that is miscible with HNO3, reacts to remove water, and does not additionally yield destabilizing moieties during the reaction.  Evidently, the HNO3/Ac2O method is ideal in the nitration of NCC as it readily produces a highly nitrated product of a DS of 2.7; however, the theoretical maximum %N of 14% could not be achieved.      67   Chapter 3 ? Stability studies of NNC  3.1 Introduction   For practical application of nitrocellulose, the stability of nitrocellulose is important for handling and storage.  Typically, nitrocellulose is not stored as a dried powder as it is highly energetic and deflagration of the material will cause failure of the container.  Instead, it is usually stored under an alcohol to both remove less nitrated cellulose, which is alcohol soluble, and to keep the autocatalytic degradation to a minimum by diluting the acid produced.  However, the usage of alcohol to stabilize nitrocellulose has several drawbacks.  Firstly, it adds weight to the material so transportation of the substance is more costly than just transporting the solids alone.  Secondly, the compound needs to be isolated, either by centrifugation or filtration, and dried prior to use in explosives and other applications.  In this chapter, the effect of multiple rinses during isolation and post-stabilization in aqueous NaHCO3 on the stability of the NNC product was investigated.  Also, a comparison of the stability of the NNC synthesized by the HNO3/H2SO4, and the HNO3/Ac2O methods was conducted at room temperature to simulate storage at ambient conditions in order to determine whether storage as a dried powder is viable for these materials.    68   3.2 Experimental conditions  Characterization techniques.   Thermal gravimetric analysis (10 ?C/min under air) was performed with a PerkinElmer STA 6000.  CNH elemental analysis was performed with a Fisons EA 1108 CHN-O.  The instrumental errors associated with CHN EA is ? 0.3% as shown in the error bars.  All samples were dried under vacuum overnight prior to analysis and were handled with extreme caution to prevent accidental deflagration.  Stabilization of NNC.  2 to 3 grams of freshly prepared NNC was added to 750 mL of deionized water, rapidly stirred and then heated to a rolling boil for 1 hour.  The mixture was cooled to room temperature and neutralized with saturated NaHCO3 solution until the supernatant liquid tested neutral with pH paper.  The supernatant liquid was decanted and the slurry of stabilized NNC was filtered through a medium glass frit and washed with copious amounts of water.  The resulting white powder was dried in vacuo to yield stabilized NNC in quantitative yield.    3.3 Results and discussion  Aqueous stabilization of NNC.   Typically, NNC is isolated in the laboratory by an ice-water quench of the nitrating solution then subsequent filtration through a glass rit.  Industrial processes normally use centrifugation and water rinses to purify the nitrated samples, but for the purpose of research 69   filtration was adapted.  The solid product was then washed with copious amounts of deionized water to eliminate acid that absorbed into the cellulose fibers, however, it is immediately clear that several D. I. H2O washes are not adequate to completely remove acid that is adsorbed onto the surface of NNC.  In a specific nitrated sample, immediately after the ice-water quench, the crude solids were subjected to one, two or three rinses with large amounts of D. I. H2O.  The resulting samples were then dried under vacuum overnight at room temperature until a constant weight was achieved.  There is a remarkable difference in the stability of the NNC product as the sample subjected to only a single wash with H2O quickly showed signs of degradation through the release of yellow fumes, whereas the material is stabilized by additional washes.  By TGA, a clear trend of increasing deflagration temperature is seen as the amount of D. I. H2O washes is increased.  The deflagration temperature can be used as a gauge for NNC stability since the presence of residual acid and trace metals destabilizes the material and causes it to fail prematurely.  In the samples made using the HNO3/H2SO4 method, the NNC that is only washed once has a deflagration temperature of 118-120 ?C.  Typically, NC has a deflagration temperature of 180 ?C and above to be considered stable. With an additional D. I. H2O wash, the deflagration temperature was increased to 136 ?C.  Subsequent washes with water yielded products with increased deflagration temperature, 144 ?C and 148 ?C, but with a less dramatic change when compared to the single wash sample.  No increase in deflagration temperature was seen after three washes, which indicated the near complete removal of absorbed acid.  However, despite three washes and the stabilization of the deflagration temperature, degradation of the sample was observed after one month of storage at ambient conditions.  The resulting TGA shows that the deflagration temperature decreased to 124 ?C ( Figure 3-1).   70    Figure 3-1 ? Deflagration temperatures obtained from TGA showing the effect of water rinses on the deflagration temperature of NNC synthesized by HNO3/H2SO4 method.  The deflagration temperature can be increased significantly after rinsing with D. I. H2O three times.  Testing the pH of the solution after resuspension of the nitrated NNC into water showed that there is a significant amount of acid present in the sample.  As both the photodecomposition and acid catalyzed decompositions produce additional acid, it is unclear which pathway dominated in the degradation of NNC nitrated by the HNO3/H2SO4 method.  The stability of NNC nitrated with HNO3/P2O5 method was also evaluated.  After rinsing one, two or three times with D. I. H2O after the ice-water quench, samples were analyzed by TGA for differences in deflagration temperature.  Similar to the experiments with NNC produced by the HNO3/H2SO4 method, the deflagration temperature increased as the number of rinses increased; the deflagration temperatures are 190 ?C, 201 ?C, 207 ?C respectively.  No increase in the deflagration temperature was observed after three D. I. H2O rinses.  Unlike the NNC prepared by HNO3/H2SO4, the material prepared in HNO3/P2O5 is stable up to four months at ambient 02040608010012050 70 90 110 130 150 170 190Weight %Temperature (oC)Unstabilized1 rinse2 rinse3 rinseDestabilized71   conditions where little formation of NO2 gas is observed.  TGA analysis of the destabilized product after 4 months of storage also shows a deflagration temperature (183 ?C) that is significantly lower than that obtained immediately after filtration and rinsing ( Figure 3-2).  The difference in the stability of the rinsed NNC product between the HNO3/H2SO4 method and the HNO3/P2O5 method is remarkable.  With similar acidity, the nitration mixtures were not expected to have such a huge effect on the stability of the material nitrated.  However, as proven by conductometric titration, there are more sulfate groups on the NNC fibers in the samples nitrated by the HNO3/H2SO4 method.  The sulfate groups, although not chemically labile, provide a source of protons, H+, which can recombine with NO2 gas produced from photodecomposition to yield nitric acid, which is autocatalytic towards the degradation of NNC.    Figure 3-2 ? Deflagration temperatures obtained from TGA showing the effect of water rinses on the deflagration temperature of NNC synthesized by HNO3/P2O5 method.  Even though the product began with a deflagration temperature beneath the stability limit, the temperature can be increased to >200 ?C after water washes. 02040608010012050 70 90 110 130 150 170 190 210 230 250Weight %Temperature (oC)Unstabilized1 rinse2 rinse3 rinseDestabilized72   Moreover, the NNC synthesized using the HNO3/Ac2O method was also investigated for stability at ambient conditions.  It was observed that the NNC product is stable up to 4-5 months at room temperature after rinsing the crude NNC with copious amounts of D. I. H2O.  No obvious signs of degradation, including discoloration or degassing, were recorded.  Again, TGA shows the increase in stability from the inference of increased deflagration temperature as the number of water rinses increased, with 203 ?C, 213 ?C, and 220 ?C as a result of the increase number of rinses.  A decrease in deflagration temperature after several months? storage is similarly observed in the NNC synthesized by the HNO3/Ac2O method ( Figure 3-3).  Figure 3-3 ? Deflagration temperatures obtained from TGA showing the effect of water rinses on the deflagration temperature of NNC synthesized by HNO3/Ac2O method.  The increased stability of this NNC despite being more nitrated than the other two methods is due to the lack of ?OSO3H groups.   To improve the stability of the nitrated cellulose, it is common practice to boil the isolated solids in water for a short period to time to facilitate the removal of adsorbed acids, that 02040608010012050 70 90 110 130 150 170 190 210 230 250Weight %Temperature (oC)Unstabilized1 rinse2 rinse3 rinseDestabilized73   were not removed by filtration and water rinse.  There are two main methods for the stabilization of NC, including repetition of the boiling process, and the neutralization by aqueous base after a single boiling process.  By heating NNC in a large volume of D. I. H2O, energy is provided to the material so that adsorbed acids are removed from the surface of the cellulosic material and dissolved into the surrounding media.  However, if the neutralization method were used, care must be taken to reach neutral pH at room temperature as excess base degrades NNC.  This is the reason that NaHCO3, a weak base, is preferred over NaOH or NH4OH, so much more control can be exerted on the stabilization process.  Taking a NNC product synthesized using the HNO3/H2SO4 method, which is rinsed three times with copious amounts of D. I. H2O, and subjecting the aqueous suspension to boiling, the pH, theoretically testing neutral, is demonstrated to be slightly acidic at approximately pH 5-6 by pH paper.  The result shows that the NNC product contained large amounts of adsorbed residual acid from the nitration mixture and the process of heating the compound in water releases the adsorbed acid.  Neutralization of the aqueous suspension after cooling to room temperature yields no changes to the color of the NNC product nor the solution, but excess base causes yellowing of the solution as the NNC degrades.  The isolated material, after overnight drying under vacuum, was left at ambient conditions to determine its stability.  Initially, CHN EA revealed that the stabilized material made using the HNO3/H2SO4 method has 10.8 %N.  After one month of storage, where the non-stabilized product shows degradation, the %N of the stabilized product is 10.9%.  TGA shows only a small decrease in the deflagration temperature after one month of storage at ambient conditions.  Further testing however, shows that the material begins degradation after two months in storage as indicated by NO2 fume release and a color change within the material.  On the other hand, taking the nitrated product from the HNO3/Ac2O method and treating the material 74   with similar conditions as the experiment above, a slightly acidic solution approximately pH = 5-6 is obtained after boiling the NNC solids in water.  Neutralization and subsequent vacuum filtration and drying yields a white powder with the exact same appearance as that of the non-stabilized material.  Initially, the product shows a %N of 13.5% by EA and a deflagration temperature of 215 ?C by TGA.  After storage at ambient conditions for five months, upon which the non-stabilized NNC starts showing signs of degradation, the aqueously stabilized NNC synthesized by the HNO3/Ac2O method remained stable (13.5 %N) and a deflagration temperature that is unchanged from the initial analysis.  The sample remains stable at ambient conditions, showing no signs of NO2 fumes or discoloration months after the analysis was conducted.  This demonstrates NNC is most stable when adsorbed acids are removed by a heat treatment in D. I. H2O, however, the samples remain affected by the presence of sulfate groups in the material that can autocatalyze the decomposition of NNC.  This results in the large difference in stability of NNC produced using the HNO3/H2SO4 method compared to the HNO3/Ac2O method.    The aqueous stabilization of NNC can be used multiple times throughout the storage lifetime due to the mild nature of the treatment.  The repetition of the stabilization consecutively shows little effect on adding more stability to the material, as demonstrated on samples synthesized with the HNO3/H2SO4 method.  The samples, treated once or twice with aqueous stabilization, continues to degrade at the same rate demonstrated as similar deflagration temperatures on TGA.  However, the aqueous stabilization can be used on aged samples of NNC that showed signs of degradation (Figure 3-4).  Although the yellowing of the NNC material cannot be reversed in the aqueous stabilization and that the amount of nitrogen loss cannot be undone, it helps to reestablish stability of the material.  This is demonstrated in a sample of NNC 75   made using the HNO3/H2SO4 method after aging at ambient conditions for two months since stabilization, showing discoloration and the release of NO2 fumes.  The off-white solid was restabilized in boiling D. I. H2O and the resulting dried solids were again aged at room temperature.  The first signs of degradation only appeared after two months of storage, which is similar to that of freshly prepared and stabilized NNC prepared by the HNO3/H2SO4 method.    Figure 3-4 ? TGA of NCC synthesized by HNO3/H2SO4 subjected to stabilization, aging, and restabilization showing the effects of aqueous stabilization.  This process has a more pronounced effect on this sample as it degrades much more rapidly than the other two methods.  This can be repeated multiple times to prolong the life of this material.  Stability of NNC synthesized with different methods.   The stability of the synthesized NNC must be considered for long term storage.  It is common practice in industry to store NC as a solid suspended in ethanol as wetted NC is much more stable than the dried form.  Also, slight degradation leading to denitration causes the NC to 02040608010012050 70 90 110 130 150 170 190Weight %Temperature (oC)UnstabilizedStabilizedAged destabilizedAged restabilized76   dissolve into the alcohol; alcohols can only dissolve NC that has a DS less than 2.  This method is doubly advantageous since portions of NC that have a lower nitrogen content, either from the synthesis or from degradation, will be soluble in the alcohol.  A simple filtration to remove the ethanol solution and subsequent wash will leave behind the desired highly nitrated nitrocellulose for use.  Since NC is such a reactive material, factors such as storage temperature, exposure to light, and most importantly the method of purification are of utmost importance in the storage of nitrocellulose for long term use.  As shown, NC is easily destabilized by acid, base or trace metals, all of which catalyze the denitration and ultimately failure of the nitrated material.    Some of the most common techniques to evaluate the stability of the nitrated cellulose is through accelerated decomposition by heating at a constant temperature and then the degradation products are analyzed by CHN EA, GPC, or NOx chemiluminescence.  Typically, NC are tested by the Bermann-Junk test, which heats the product at 132 ?C for 2 hours and by the Abel Heat test, (76.6 ?C for 10 mins), which simulate storage of NC at 1 month and 2 days respectively.  Essentially, the Bergmann-Junk test works by the release of NO2 by ?ONO2 bond cleavage, showing overall stability, whereas the Abel Heat test investigates the amount of NOx adsorbed in the material to show the aging of the NC.  For our purposes, stability of nitrated NCC was done under ambient conditions, where samples are subjected to fluctuations in temperature, variations in humidity, and material confinement.  Exposure to indirect sunlight and artificial lighting versus darkness was investigated for the effect of photocatalyzed degradation of NNC.  It was found that the method of preparation of NNC drastically effects the stability of the product at room temperature.  Even though all the products isolated have different degrees of nitration, with the product resulting from the HNO3/H2SO4 method being the lowest at %N = 11%, and the product resulting from the 77   HNO3/Ac2O method being the highest at %N = 13.4%, one would expect that the product with more nitro groups to be less stable as the material is much more energetic.  However, this is not observed.  Over a period of 6 months, the NNC samples in 1 g aliquots were stored in 20 mL glass vials in darkness to determine whether ambient conditions can impart visual changes to the material as it slowly degrades over time.  Detailed recordings of the color, texture, and volume of the solid NNC were documented as the stability experiment proceeded (Table 3-1). Table 3-1 ? Physical and visual characteristics of NNC synthesized by HNO3/H2SO4 and HNO3/Ac2O methods after one month and six months of storage.  The sample from the latter method show no signs of degradation within this time frame.    HNO3/H2SO4 HNO3/Ac2O  1 month 6 months 1 month 6 months Color White Dark yellow White White Gas Yellow NO2 None None None Texture Fine powder Gel/oil Fine powder Fine powder Volume As initial 0.5x initial As initial As initial   For the NNC produced by the HNO3/H2SO4 method, the degradation occurs quite rapidly after aqueous stabilization.  There are several crucial steps in the life of the NNC product before complete failure occurs (Figure 3-5).  In samples that showed initial signs of degradation, the release of NO2 gas was seen, yet the solids remained as white as the original color of the isolated product.  The NO2 fumes given off is of an intense orange color that permeates through the container in which the NNC product is contained.  Following gas release, some reabsorption of the NO2 gas occurs as indicated in the staining of a yellow color on the product.   78    Figure 3-5 ? Representative appearances of HNO3/H2SO4 synthesized NNC from initial conditions to total failure (left to right) after six months of storage at ambient conditions in darkness.  The production of NO2 gas after two months triggered the following decomposition of the material.   At this point, the degradation can either stop due to the presence of decreased amounts of NO2 gas or accelerate due to increased absorption of NO2, which depends on the rate of gas escape through the container.  After the color change, if NO2 is continually adsorbed, the NNC product softens and intensifies in yellow color as the acid that is produced by reabsorption of NO2 gas and moisture from the air degrades and liquefies the material.  As this reaction is autocatalytic, the process accelerates causing bubbles of NO2 gas to form within the gel-like material.  The rapid gas release splatters material within the container.  Typically, due to the release of NO2 gas and the consumption of the energetic material, the degradation process stops at this stage and yields a yellow translucent oily material.  However, if a significant amount of destabilized NNC remains during the degradation process, the material can self-ignite leading to total degradation of the material as the supply of oxygen is limited in the container.  The resulting catastrophic failure of the NNC yields a mixture of carbonaceous materials.  Hence, the NNC synthesized with the HNO3/H2SO4 has low stability and cannot be stored safely under ambient conditions, even in darkness.  A sample subjected to these conditions was determined to have 8.7 %N after six months.  In comparison, the NNC made by the HNO3/Ac2O method shows amazing stability at ambient conditions without the presence of light, where no release of NO2 gas is observed six 79   months after the compound is isolated and stabilized; the EA shows a of 13.4 %N  (Figure 3-6).  The addition of light exposure into the stability experiment was expected to accelerate the degradation of NNC.  Astoundingly, the experiment shows that the NNC made by the HNO3/Ac2O method is stable in light and still showed 13.4 %N 6 months after synthesis (Figure 3-7).  Yet the NNC made by the HNO3/H2SO4 method (initially 12.2 %N) remained stable for ~2 months after synthesis, but shows a rapid deterioration to 6.6 %N after four months of storage.  The degradation continues at a rate of approximately 0.8% nitrogen per month of additional storage to a result of 5 %N after six months at ambient conditions.  The differences in stability of the NNC made using the two method is remarkable, and it shows that the HNO3/Ac2O method is the ideal way to synthesize a highly energetic yet stable nitrocellulose material.  In addition, it also shows that the exposure to light destabilizes NNC made using the HNO3/H2SO4 method, but has little effect on the HNO3/Ac2O NNC within the time span investigated.  In comparison, the products synthesized by the two methods but with MGC as starting material also followed a similar trend (Figure 3-8).   Figure 3-6 ? Representative appearances of HNO3/Ac2O synthesized NNC after two to eight months of storage (left to right in two month increments) at ambient conditions in darkness.  No degassing was observed in stark comparison with NNC from the HNO3/H2SO4 method. 80    Figure 3-7 ? Comparison of the %N of NNC produced by the two methods over eight months on the bench top.  The material from the HNO3/H2SO4 method remained stable for two months then rapidly dropped in stability.  The HNO3/Ac2O method, however, produced a material that is stable for storage at ambient conditions.  The %N by CHN EA decreased from a %N of 12.4% initially to 6.7% after six months of storage in ambient conditions for the material produced by the HNO3/H2SO4 method.  For the MGC material nitrated with HNO3/Ac2O, a %N of 13.4% was obtained initially and remained unchanged after storage for six months.  The remarkable similarity of the stability of nitrated MGC and NNC demonstrates that the HNO3/Ac2O method produces a much more stable material than the HNO3/H2SO4 method.  02468101214160 1 2 3 4 5 6 7 8 9%NStorage time (months)NCC 29% HNO3NCC Ac2OH2SO4 Ac2O 81    Figure 3-8 ? Comparison of the %N of nitrated MGC produced by the two methods over eight months on the bench top.  There is a similar effect on this starting material and the stability trends correspond to that of NNC.  Again, the material from the HNO3/H2SO4 method failed to remain stable under ambient conditions.  3.4 Conclusion   A simple filtration and wash with copious amounts of D. I. H2O is insufficient to stabilize the nitrated NCC due to the presence of adsorbed acid that is not readily removed by this means of isolation.  The usage of a relatively mild aqueous stabilization followed by neutralization of desorbed acid imparts greater stability in samples produced using the HNO3/H2SO4, HNO3/P2O5 and HNO3/Ac2O methods with the most remarkable impact on the NNC synthesized by HNO3/Ac2O, which shows no signs of degradation after six months of storage at ambient conditions.  On the other hand, NNC made by the HNO3/H2SO4 quickly degrades despite being processed through the aqueous stabilization, indicating that the inherent instability of the compound is not only a result of adsorbed acids, even though the treatment does double the 02468101214160 1 2 3 4 5 6 7 8 9%NStorage time (months)MGC 29% HNO3MGC Ac2OH2SO4 Ac2O 82   storage duration before degradation set in.  Furthermore, the method of aqueous stabilization can be used for freshly prepared NNC samples and aged samples repeatedly to prolong storage of the energetic material.  Nevertheless, despite containing much higher %N than the material synthesized by the HNO3/H2SO4 method, NNC made by the HNO3/Ac2O method is relatively stable as indicated by both CHN EA and TGA.                 83   Chapter 4 ? Conclusions and final remarks   A major drawback of the standard HNO3/H2SO4 nitration method of NCC is the additional sulfation it adds to the product, which is undesirable due its contribution to the instability of the compound.  The HNO3/P2O5 and the HNO3/Ac2O methods are vast improvements over the HNO3/H2SO4 method for the nitration of NCC as the two methods can produce a much more stable product with higher %N with the HNO3/Ac2O method being the most robust and most applicable.  The methanolic desulfation was crucial in producing a NNC product with high %N and also contributes to improving the stability of the compound.  From the optimization of each of the methods, several key factors influence the degree of nitration of the resulting NNC.  The first major factor is the amount of H2O within the reaction and the method of removing H2O.  This is directly controlled by the amount of the desiccant, where adequate amounts need to be within the reaction to initially remove water from the HNO3 and also to remove H2O that is produced within the reaction.  The second factor is the chemical nature of the desiccant, in that it should be miscible in all proportions with HNO3 for tunability and should not be strong enough to impart unwanted side reactions on the ?OH groups or degrade NCC when nitrated.  The third factor is the influence of sulfate groups on the surface of the NCC, where a methanolic desulfation pretreatment of NCC proves to increase the amount of available ?OH groups for the nitration.  The fourth and final factor is the requirement of an aqueous stabilization of the NNC product prior to storage as adsorbed acids not removed by rinsing with D. I. H2O causes product decomposition.  The HNO3/Ac2O NNC products, after stabilization, are stable under storage at ambient conditions for six months or more.  Predictably, the said NNC product should be stable at room temperature for a much longer time span.   84   However, the ultimate goal of the nitration of NCC is to determine whether NNC can outperform traditional nitrocellulose.  The increased number of glucose end groups should theoretically permit the nitration to surpass the maximum of %N = 14% in bulk nitrocellulose as more ?OH groups are available for nitration.  This result was not obtained, however, and a maximum of ~13.4 %N could be achieved, which is closer to the practical maximum for NC.  NCC is slower than filter paper to react in the nitration mixture and only reaches equilibrium after ~30-60 mins, whereas normal cellulose will reach equilibrium after submersion for only several minutes in the nitration mixture.  The resulting NNC has a maximum DS of 2.7, significantly lower than the theoretical DS of 3.  As for the thermal stability of the NNC, stabilized NNC has an average deflagration temperature of 212 ?C by TGA.  Typically, industrially produced NC has a deflagration temperature of above 180 ?C in order for the compound to be considered safe for transport.  In other words, the thermal stability of NNC far exceeds that of the thermal stability limit.  Nevertheless, more investigation is required to find the optimal nitration conditions for producing NNC with DS > 2.7.         85   References  1 Xu, L.-P.; Peng, J.; Liu, Y.; Wen, Y.; Zhang, X.; Jiang, L.; Wang, S. ACS Nano, 2013, 7, 5077. 2 Vollrath, F.; Knight, D. P. Nature, 2001, 410, 541. 3 Chakar, F. S.; Ragauskas, A. J. Ind. Crop. Prod., 2004, 20, 131. 4 Lahiji, R. R.; Xu, X.; Reifenberger, R.; Raman, A.; Rudie, A.; Moon, R. J. Langmuir, 2010, 26, 4480. 5 O?Sullivan, A. C. Cellulose, 1997, 4, 173. 6 Edgar, K. J.; Buchanan, C. M.; Debenham, J. S.; Rundquist, P. A.; Seiler, B. D.; Shelton, M. C.; Tindall, D. Prog. Polym. Sci., 2001, 26, 1605. 7 Roy, D.; Semsarilar, M.; Guthie, J. T.; Perrier, S. Chem. Soc. Rev., 2009, 38, 2046. 8 Cheng, H. N.; Dowd, M. K.; Selling, G. W.; Biswas, A. Carbohyd. Polym., 2010, 80, 449. 9 Klemm, D.; Heublein, B.; Fink, H.-P.; Bohn, A. Angew. Chem. Int. Ed., 2005, 44, 3358. 10 Gohdes, M.; Mischnick, P. Carbohyd. Res., 1998, 309, 109. 11 Granja, P. L.; Pouys?gu, L.; P?traud, M.; De J?so, B.; Baquey, C.; Barbosa, M. A. J. Appl. Poly. Sci., 2001, 82, 3341. 12 Otto, F. J. Itzehoer Wochenblatt, 1846, Columns 1626f. 13 Castirubam T. C.; Helf, S.; Aaronson, H. A.; Kaufman, J. V. R., according to Helf, S. in Encyclopedia of Explosives by Haye, S. M., 1978, 8, N217: ARRADCOM, Dover, New Jersey. 86   14 Lunge, G. J. Am. Chem. Soc., 1901, XXIII, 527. 15 Selwitz, C. Cellulose Nitrate in Conservation, 1988, 2: J. Paul Getty Trust, USA. 16 Eder, J. M. Ber. d. Chem. Ges., 1880, 13, 169. 17 Yamamoto, H.; Horii, F.; Hirai, A. Cellulose, 2006, 13, 327. 18 Dow Wolff Cellulosics, Walsroder Nitrocellulose. 19 Michel, J.-M. Actual. Chimique, 2006, 300, 7. 20 Boyd, J. E. Chem. Heritage, 2011, 29, 20.  21 Fenichell, S. Plastic: The Making of a Synthetic Century, 1996, 356, 4: Harper Collins, New York. 22 Theisen, E. J. SMPTE, 1933, 20, 259. 23 Wilks, E. S. Industrial Polymers Handbook, 2001, 1507: Wiley-VCH. 24 Kavalenko, V. I. Russ. Chem. Rev., 1995, 64, 753. 25 Chue, R. S.; Clarke, J. F.; Lee, J. H. Proc. R. Soc. Lond. A, 1993, 441, 607. 26 Dauerman, L.; Tajima, Y. A. AIAA J., 1968, 6, 1468. 27 Ettre, K.; Varadi, P. F. Anal. Chem., 1963, 35, 69. 28 Shimamoto, S.; Gray, D. G. Chem. Mater., 1998, 10, 1720. 29 Charlet, G.; Gray, D. G. Macromolecules, 1987, 20, 33. 30 Warren, R. C. Rheol. Acta., 1984, 23, 544. 87   31 Panar, M.; Willcox, D. B. Fr. Pat., 1977, 770, 3473. 32 Thomas, A.; Antonietti, M. Adv. Funct. Mater., 2003, 13, 763. 33 Lupica, S. B. Radiochem. Radioanal. Lett., 1975, 21, 31. 34 Nadkarni, V. S.; Samant, S. D. Radiat. Meas., 1997, 27, 505. 35 Andrzej, R.; Janusz, Z.; Michal, D.; Jan, S. For. And Wood Technol., 2010, 72, 213. 36 Whistler, C. Methods in Carbohydrate Chemistry ? ?Cellulose?, 1963, Vol. III: Academia Press, New York. 37 Holleman, A. F.; Wiberg, E. Inorganic Chemistry, 2001: Academia Press, San Diego. 38 Stern, S. A.; Mullhaupt, J. T.; Kay, W. B. Chem. Rev., 1960, 60, 185. 39 Sakata, H.; Komatsu, N. J. Soc. Textile Cellulose Ind. Japan, 1963, 19, 337. 40 Urbanski, T. Chemistry and Technology of Explosives, 1964: PERGAMON PRESS, London. 41 Hercules Powder Co. Inc. according to J. Roth, Encyclopedia of Explosives, 1978, 8: ARRADCOM, Dover, New Jersey. 42 Miles, F. D. Trans. Faraday Soc., 1933, 29, 110. 43 Goring, D.; Timell, T. Tappi, 1962, 45, 454. 44 Thinius, K.; Tummler, W. Makromol. Chem., 1960, 99, 117. 45 Heuser, E.; Jorgensen, L. Tappi, 1951, 34, 443. 46 Berl, E.; Rueff, G. Ber. Dtsch. Chem. Ges., 1930, 63, 3212. 88   47 Das, S.; Raut, V. D.; Gawande, N. M.; Khopade, R. S.; Narasimhan, V. L. Ind. J. Chem. Technol., 2006, 13, 404. 48 Whistler, R. Methods in Carbohydrate Chmeistry, 1963, 3: Cellulose Academic Press, New York, London. 49 Bordwell, F. G.; Garbisch, E. W. J. J. Am. Chem. Soc., 1960, 82, 3588. 50 Andreozzi, R.; Marotta, R.; Sanchirico, R. J. Hazard. Mater., 2002, A90, 111. 51 Novikov, S. S.; Khmel?nitskil, L. I.; Novikova, T. S. Russ. Chem. Bull., 1965, 14, 90. 52 Orton, K. J. P. J. Chem. Soc., 1902, 81, 806. 53 Sun, D.-P.; Ma, B.; Zhu, C.-L.; Liu, C.-S.; Yang, J.-Z. J. Energ. Mater., 2010, 28, 85. 54 Hughes, E. D.; Ingold, C. K.; Reed, R. I. J. Chem. Soc., 1950, 2400. 55 Brooks, R. M. Canadian Patent 557 671, 1958. 56 Bofors-Nobel-Chematur, Nitrocellulose Plant. 57 Pamfilor, A. J. Appl. Chem.-USSR+, 1953, 26, 227. 58 Li, L.; Frey, M. Polymer, 2010, 51, 3774. 59 Miles, F. D. Cellulose Nitrate, 1955: Interscience, New York. 60 Shashoua, Y.; Bradley, S. M.; Daniels, V. D. Stud. Conserv., 1992, 37, 113. 61 Battista, O. A. Ind. Eng. Chem., 1950, 42, 502. 62 R?nby, B. C. Discuss. Faraday Soc., 1951, 11, 158. 89   63 Bondeson, D.; Mathew, A.; Oksman, K. Cellulose, 2006, 13, 171. 64 Abitbol, T.; Kloser, E. Cellulose, 2013, 20, 785. 65 Espinosa, S. C.; Kuhnt, T.; Foster, J.; Weder, C. Biomacromolecules, 2013, 14, 1223. 66 Revol, J.-F.; Bradford, H.; Giasson, J.; Marchessault, R. H.; Gray, D. G. Int. J. Biol. Macromol., 1992, 14, 170. 67 Orts, W. J.; Godbout, L.; Marchessault, R. H.; Revol, J.-F. Macromolecules, 1998, 31, 5717. 68 Shopsowitz, K. E.; Qi, H.; Hamad, W. Y.; MacLachlan, M. J. Nature, 2010, 468, 422. 69 Kelly, J. A.; Shukaliak, A. M.; Cheung, C. C. Y.; Shopsowitz, K. E.; Hamad, W. Y.; MacLachlan, M. J. Angew. Chem. Int. Ed., 2013, 52, 8912. 70 Khan, M. K.; Giese, M.; Yu, M.; Kelly, J. A.; Hamad, W. Y.; MacLachlan, M. J. Angew. Chem. Int. Ed., 2013, 52, 8921. 71 Shopsowitz, K. E.; Stahl, A.; Hamad, W. Y.; MacLachlan, M. J. Angew. Chem. Int. Ed., 2012, 51, 6886. 72 Marcovich, N. E.; Auad, M. L.; Bellesi, N. E.; Nutt, S. R.; Aranguren, M. I. J. Mater. Res., 2006, 21, 870. 73 Lewin, M.; Epstein, J. A. Text. Res. J., 1960, 30, 520. 74 Bates, S.; Zografi, G.; Engers, D.; Morris, K.; Crowley, K.; Newman, A. Pharm. Res., 2006, 23, 2333. 75 Habibi, Y.; Lucia, L. A.; Rojas, O. J. Chem. Rev., 2010, 110, 3479. 90   76 Cook, R. L.; Andrew, E. A., U.S. Patent 2,888,713, 1959. 77 Wang, N.; Ding, E.; Cheng, R. Polymer, 2007, 48, 3486. 78 Dawoud, A. F.; Saad, H. A.; Attia, M. E. Explosivstoffe, 1972, 20, 181. 79 Pomin, V. H.; Valente, A. P.; Pereira, M. S.; Maur?o, P. A. S. Glycobiology, 2005, 15, 1376. 80 Nagasawa, K.; Inoue, Y.; Kamata, T. Carbohyd. Res., 1977, 58, 47. 81 Hasani, M.; Cranston, E. D.; Westman, G.; Gray, D. G. Soft Matter, 2008, 4, 2238. 82 Dong, X. M.; Kimura, T.; Revol, J.-F.; Gray, D. G. Langmuir, 1996, 12, 2076. 83 Nishiyama, Y.; Sugiyama, J.; Chanzy, H.; Langan, P. J. Am. Chem. Soc., 2003, 125, 14300. 84 Mitchell, R. L. Ind. Eng. Chem., 1953, 45, 2526. 85 Burger, H.; Koine, A.; Maron, R.; Mieck, K.-P. Inter. Poly. Sci. Technol., 1995, 22, 25. 86 Kovalenko, V. I.; Mukhamadeeva, R. M.; Maklakova, L. N.; Gustova, N. G. Zhurnal Strukumoi Khimii, 1993, 34, 59. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072160/manifest

Comment

Related Items