Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Understanding the action of gas hydrate kinetic inhibitors Daraboina, Nagu 2012-07-09

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
24-ubc_2012_fall_daraboina_nagu.pdf [ 17MB ]
Metadata
JSON: 24-1.0059278.json
JSON-LD: 24-1.0059278-ld.json
RDF/XML (Pretty): 24-1.0059278-rdf.xml
RDF/JSON: 24-1.0059278-rdf.json
Turtle: 24-1.0059278-turtle.txt
N-Triples: 24-1.0059278-rdf-ntriples.txt
Original Record: 24-1.0059278-source.json
Full Text
24-1.0059278-fulltext.txt
Citation
24-1.0059278.ris

Full Text

  UNDERSTANDING THE ACTION OF GAS HYDRATE   KINETIC INHIBITORS     by      Nagu Daraboina  Master of Engineering (Chem. Eng.), Indian Institute of Science, Bangalore, 2008     A THESIS SUBMITTED IN PARTIAL FULFILLMENT THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES  (Chemical and Biological Engineering)    THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  July 2012   ©  Nagu Daraboina 2012    ii  ABSTRACT   The presence of inhibitors delayed hydrate nucleation and decreased the overall formation of methane/ethane/propane hydrate compared to pure water system. However, the two classes of inhibitors: chemical (Polyvinylpyrrolidone (PVP) and industrial inhibitor: H1W85281) and a biological (Type I and III antifreeze protein (AFP)) were distinguished by the formation of hydrates with different stabilities. A single hydrate-melting peak was seen with the AFP-III and this was consistent after re-crystallization. In contrast, multiple hydrate melting events were observed in the presence of the chemical inhibitors.  In stirred reactor, onset of hydrate decomposition occurred earlier in the presence of the inhibitors compared to water controls. However, depending on the type of inhibitor present during crystallization, hydrate decomposition profiles were distinct, with a longer, two-stage decomposition profile in the presence of the chemical inhibitors. The fastest, single-stage decompositions were characteristic of hydrates in experiments with either of the AFPs.  Powder X-ray diffraction and nuclear magnetic resonance spectroscopy showed that structure II hydrates dominated, as expected, but in the presence of the chemical inhibitors structure I was also present. Raman spectroscopy confirmed the complexity and the heterogeneity of the guest composition within these hydrates. However, in the presence of AFP- III, hydrates appeared to be relatively homogeneous structure II hydrates, with weaker evidence of structure I. When individual gas cage occupancies were calculated, in contrast to the near full occupancy of large cages with these inhibitors, almost 10% of the large cages were not filled when hydrates were formed in the presence of AFP-III, likely contributing to the easy decomposition of such hydrates seen in DSC and stirred reactor experiments.  iii These results argue that thought must be given to inhibitor-mediated decomposition kinetics when designing and screening of new kinetic inhibitors. This is a necessary practical consideration for industry in cases when due to long shut in periods; hydrate formation may be unavoidable even when inhibitors are utilized. This heterogeneity suggests that using these chemical inhibitors (PVP and H1W85281) may present a special challenge to operators depending upon the gas mixture and environmental conditions, and that AFPs may offer a more predictable, efficacious solution in these cases.                   iv PREFACE   The work of this thesis consists of three different manuscripts, which correspond to chapters two to four. The authors include Daraboina, N., Ripmeester, J.A., Virginia.K.Walker and Englezos P. Professor Peter Englezos is my principal research supervisor at the University of British Columbia. During the course of my research, I was fortunate enough to hold meaningful discussions with Dr. John A. Ripmeester who is a principal research officer at SIMS, National Research Council Ottawa and Professor Virginia K. Walker who is a professor at Queen’s university. The work of this thesis presented in several conferences. The literature review, experimental design, performing experiments and data analysis were done extensively by Daraboina, N under supervision of Professor P. Englezos. Finally, I did the final preparation for each manuscript after careful revision and approval of my supervisory committee. The list of published articles and conference presentations are given below:  Published articles  1. Daraboina, N., Linga, P., Ripmeester, J., Walker, V. K., and Englezos, P. (2011a). "Natural Gas Hydrate Formation and Decomposition in the Presence of Kinetic Inhibitors. 2. Stirred Reactor Experiments." Energy & Fuels, 25(10), 4384-4391.  2. Daraboina, N., Ripmeester, J., Walker, V. K., and Englezos, P. (2011b). "Natural Gas Hydrate Formation and Decomposition in the Presence of Kinetic Inhibitors. 1. High Pressure Calorimetry." Energy & Fuels, 25(10), 4392-4397.  3. Daraboina, N., Ripmeester, J., Walker, V. K., and Englezos, P. (2011c). "Natural Gas Hydrate Formation and Decomposition in the Presence of Kinetic Inhibitors. 3. Structural and Compositional Changes." Energy & Fuels, 25(10), 4398-4404.    v Conference presentations (with published abstracts)  1. Daraboina, N; Ripmeester, J. A.; "Englezos, P.  “New insights into gas hydrate decomposition in the presence of synthetic and biological inhibitors". Presented, 61st Canadian Society Chemical Engineering conference. London, Ontario, Canada. October 23-26, 2011.  2. Walker, V.K; Daraboina, N; Gordienko, R., Ohno, H.; Ripmeester, J. A.; Englezos, P., Green gas hydrate inhibitors?” Ice -Binding Protein conference. Kingston, Ontario, Canada, August 3-6, 2011.  3. Daraboina, N; Ripmeester, J. A.; Walker, V.K.; Englezos, P., "Multi-scale assessment of the performance of kinetic hydrate inhibitors" Submitted, 7th International Conference on Gas Hydrates Edinburgh, Scotland, United Kingdom, July 17-21, 2011.  4. Daraboina, N; Ripmeester, J. A.; Linga P.; Englezos, P., "Experimental investigation of the effect of poly-N-vinyl pyrrolidone on methane/propane clathrate in the presence of silica sand" submitted, 7th International Conference on Gas Hydrates Edinburgh, Scotland, United Kingdom, July 17-21, 2011.  5. Daraboina, N; Ripmeester, J. A.; Walker, V.K.; Englezos, P., "Inhomogeneous hydrate formation induced by kinetic hydrate inhibitors". Presented American Chemical Society conference, Anaheim, USA, March 27, 2011.  6. Daraboina, N; Ripmeester, J. A.; "Englezos, P. Understanding the Action of kinetic hydrate inhibitors". Presented, 60th Canadian Society Chemical Engineering conference. Saskatoon, Saskatchewan, Canada. October 26,2010.    vi TABLE OF CONTENTS   vii    viii    ix LIST OF TABLES           x  LIST OF FIGURES   xi          xii NOMENCLATURE   Abbreviation  Full form  KHI Kinetic hydrate inhibitor Tm, hydrate Hydrate melting temperature Tm, ice Ice melting temperature Hm, ice Area of ice melting peak  Hm, hydrate Area of hydrate melting peak  Hm, total Total area of melting peaks CR Crystallizer R Reservoir CV Control valve DAQ Data acquisition system ER External refrigerator ¯tind Mean induction time ¯n Mean gas consumption  ¯tind, Fresh Mean fresh induction time ¯tind, Memory Mean memory induction time ¯n Fresh Mean fresh experiment gas consumption ¯nMemory Mean memory experiment gas consumption θS Small cage occupancy θL Large cage occupancy ΘT Total large cage occupancy  xiii N Hydration number T Temperature P Pressure IT Induction time                                        xiv ACKNOWLEDGEMENTS   It is my pleasure to thank many people and organizations who supported me throughout the duration of this thesis.  I am wholeheartedly thankful to my supervisor Prof. Peter Englezos for his constant support, encouragement and guidance from initial to final level of this research project. I owe my deepest gratitude to Dr. John A. Ripmeester and Prof. Virginia K. Walker for their advice, scientific discussions and encouragement, which improved my thought process as a researcher.  I would also like to show my gratitude to Four Year Fellowship (FYF), Natural Science and Engineering Research Council of Canada (NSERC) and Shell Global Solutions for providing Financial support for this project. I would also like to extend this acknowledgement to my colleagues and friends: Praveen, Nayef, Cef, Jeffry, Alireza, Iwan, Babak, Yizhou, Negar, Sima, Dongliang, Hassan, Venkat, Rakib, and Jaishankar through group meetings and personal discussions. A special thanks to Praveen and Rajneesh for all the support, discussion and fun I had with them while working together in the lab. I would also like to thank everyone at Material Structure and Function (MSF) group at NRC Ottawa: Chris, Gary, Steve, Igor, Hailong, Hiroshi, Jeff, Olga and Robin for their valuable advice.  I would like to thank CHBE well wishers and friends: Helsa, Dhanesh, Dean, Amber, Ann, Lori, Richard, David, Alf, Ryan, James, Masita, Jidon, Paula, Jaime, Maziar, Monali, CHBE camping team, CHBE skiing team, CHBE drinking team, CHBE Hawks and Koerner’s pub team for their support and fun throughout my program.  Finally, I wish to thank my family for their love and care.   xv                      DEDICATED TO MY FATHER  AND TO MY FAMILY      1  1. INTRODUCTION   Natural gas hydrates are solid crystalline compounds composed of hydrocarbon guest molecules enclathrated in a hydrogen bonded water molecule framework that form under suitable temperature and pressure conditions (Davidson, 1973b; Makogon, 1981; Sloan, 1998a).  Structure I (sI), Structure II (sII) and Structure H (sH) are three well known structures of gas hydrates (Davidson, 1973a; Ripmeester et al., 1987; Englezos, 1993; Ripmeester, 2000; Sloan, 2003), distinguished by the size and geometry of the water framework: structure I (sI) generally enclathrates small hydrocarbons such as methane or ethane and structure II (sII) housing larger molecules such as propane  (Sloan, 2003; Uchida et al., 2007); These two structures are cubic but the third type, structure H (sH), has a hexagonal crystal structure, which can enclathrate molecules as large as methylcyclopentane or two guest molecules such as methane and dimethylbutane (Mehta and Sloan, 1993b).  sH was discovered at National Research Council in Canada in 1987 by Ripmeester and co-workers (1987). Each structure is a different combination of cages formed by water molecules.  sI consists of 2 small pentagonal dodecahedron cages and 6 large tetrakaidecahedron cages which are formed by 46 water molecules. sII consists of 16 small pentagonal-dodecahedron cages and 8 larger hexakaidecahedron cages formed by 136 water molecules. Like sI and sII, sH has the basic pentagonal dodecahedron (12 pentagons -512) cage and two other cavities: a medium cage consisting of three tetragonal, six pentagonal and three hexagonal faces (435663) and a large cage made up of twelve pentagonal and eight hexagonal faces (51268).  Because this structure can accommodate large molecules found in crude oils it is also industrially important. A new trigonal hydrate structure with three large cavities  2 (51263, 51262 and 4151063) was also reported by Udachin et al. (2001a).  The three common gas hydrate cavity structures are shown in Figure 1.1 and structural properties are given in Table 1.1.  Figure 1.1. Three common gas hydrate cavity structure  Reprinted from Storbel et al., 2009, with permission from Elsevier.   Table 1.1. Structural properties of hydrates (Sloan, 1998a).   Structure I Structure II Structure H Cavity types 512, 51262 512, 51264 512, 435663, 51268 Cages/unit cell 2, 6 16, 8 3, 2, 1 Co-ordination number 20, 24 20, 28 20, 20, 36 Crystal type Cubic Cubic Hexagonal  3  The formation of hydrate crystals is a multiphase crystallization process. The hydrate formation process is subdivided into hydrate nucleation and hydrate growth. During nucleation stable nuclei are generated which will then grow to form crystals (crystal growth period). Hydrate nucleation refers to the process where small hydrate crystals called nuclei develop until they attain a critical size, thereafter the continued growth will start. The growing of clusters of water around gas molecules is the first step of hydrate nucleation (Bishnoi and Natarajan, 1996). The time period required to reach the state where the first stable nuclei appear is called the induction time. The induction time can be obtained through a visual observation or through pressure/temperature measurements. The induction time is believed to be a stochastic phenomenon that cannot be predicted (Englezos, 1996; Klomp, 2008). Based on the result of several studies done by different research groups, it can be concluded that, the nucleation time depends on following parameters; (a) Nature and thermal history of water (Vysniauskas and Bishnoi, 1983b); (b) stirring rate (Englezos et al., 1987a); (c) temperature and pressure (Sloan, 1998a); (d) degree of super saturation (Englezos, 1987a); and (e) molecular diameter to cavity size ratio (Sloan and Fleyfel, 1991).  Miller and Smythe (1970) quantified gas hydrate growth. A systematic study on the kinetics of gas hydrate formation and decomposition was started by Bishnoi in early 1980s; his group subsequently presented an empirical model that correlated the growth rate with the degree of super cooling, temperature, pressure, and interfacial area. Englezos et al. (1987a) developed a model with three steps based on theory of crystallization and mass transfer to describe the hydrate growth kinetics of methane and ethane. The model was extended to the formation of  4 hydrates from methane and ethane mixtures of various compositions (Englezos et al., 1987d). Skovborg and Rasmussen (1994) simplified the model of Englezos et al. (1987a) by assuming that mass transfer at the gas/liquid interface was the limiting step in hydrate growth. Herri et al. (1999) proposed a kinetic model including both nucleation and growth steps. Gas uptake measurement is a typical method to study the kinetics of hydrate formation (Vysniauskas and Bishnoi, 1983a; Englezos et al., 1987c; Koh et al., 2002; Sloan and Koh, 2008). Even though there is vast amounts of kinetic data available on hydrate formation and decomposition, there is still data needed to improve or control the kinetics of hydrate formation in the presence of additives.  The different instrumental techniques were applied to investigate the kinetics of hydrate formation and decomposition in addition to the conventional gas uptake experiments. Thermal analysis instruments like high pressure-micro differential scanning calorimetry (HP-µDSC) have been widely used as for the study of hydrate formation (Dalmazzone et al., 2002a; Koh, 2002; Dalmazzone et al., 2004; Le Parlouer et al., 2004; Ripmeester et al., 2007; Lachance et al., 2009), decomposition (Rueff et al., 1988; Dalmazzone et al., 2002b; Kawamura et al., 2006; Hughes, 2008; Koh et al., 2008a; Koh et al., 2008b; Nakagawa et al., 2008) and inhibition (Koh et al., 2002; Koh et al., 2009; Lachance et al., 2009; Nihous et al., 2010).  Using molecular level techniques like diffraction, Raman spectroscopy, optical microscopy and NMR spectroscopy etc., information about hydration number, structural identification, cage occupancy and meta- stable phase identification is obtained. Such information is not accessible with macroscopic measurement techniques such as gas uptake measurements.  Powder x-ray diffraction (PXRD) is a technique widely used to identify the crystalline structures of compounds by using a beam of X-rays.  The details of the working principle of XRD were given elsewhere (Grieken and  5 Markowicz, 2002; Hughes, 2008). PXRD is a powerful and robust technique to determine the structure of gas hydrate (sI, sII and sH) formation (Davidson, 1986; Koh et al., 1996; Udachin et al., 2001b; Takeya et al., 2002; Uchida et al., 2002; Uchida et al., 2004; Susilo et al., 2005; Kumar et al., 2008b). Raman spectroscopy has also been identified as a powerful technique to characterize gas hydrates and has been used to detect the structure of the hydrate (Morita et al., 2000; Schicks et al., 2005; Ripmeester et al., 2007; Kumar et al., 2008b), transition of hydrate structures (Subramanian, 2000; Subramanian et al., 2000a; Subramanian et al., 2000b; Schicks et al., 2006) mixed gas hydrate formation kinetics (Uchida et al., 2002; Uchida et al., 2007), meta- stability of hydrates (Ohno et al., 2009) and effect of inhibitors on the kinetics and structure of hydrates (Sloan et al., 1998; Carstensen et al., 2002; Koh, 2002; Sloan and Koh, 2008). The details of the working principle of Raman spectroscopy has been given in literature (Skoog, 1982; Hughes, 2008). 13C Nuclear magnetic resonance (NMR) spectroscopy has also proved useful for obtaining quantifiable hydrate compositions and cage occupancies of mixed gas hydrate (Ripmeester and Ratcliffe, 1988; Ripmeester and Ratcliffe, 1999; Kida et al., 2007; Kumar et al., 2008b; Kida et al., 2009). Ripmeester (2000) reviewed the early correlations, instrumental methods and computational modeling used for the calculation of hydration numbers, and the characterization of hydrate structures. Moudrakovski et al., (2001) studied the nucleation and growth of xenon hydrates on the surface of frozen heavy water by NMR spectroscopy using  xenon. The results suggested that the average hydrate composition varies during the nucleation step, but remains constant during the growth step. Koh (2002) discussed the industrial significance and technological importance of gas hydrates. She suggested that the fundamental understanding on gas hydrate formation and decomposition using macro and microscopic tools is necessary and  6 have wide implications on the exploitation of sediments and managing flow assurance in oil and gas pipelines. Moudrakovski et al (2004) showed through micro imaging that hydrate shells around the water droplets could be detected. Susilo et al. (2007) demonstrated the use of PXRD, DSC, NMR and Raman spectroscopy to characterize sI and sH hydrates and obtain consistent results for hydrate structure, degree of conversion and gas content in the hydrate phase with the different techniques. Susilo et al. (2006) also monitored the kinetics of structure I and H methane hydrate growth by employing NMR spectroscopy and imaging (MRI) and found out that the rates of hydrate formation agree with the results from gas uptake measurements obtained by Lee et al. (2005). Kumar et al. (2008) reported the kinetics and structure of gas hydrates from methane/ethane/propane mixtures relevant to the design of operations involving natural gas hydrate storage and transport. The results (gas phase composition and fractionation effect) obtained from kinetics at macroscopic level (gas uptake) were found agreed with the results (hydrate phase composition and cage occupancies of individual gases) obtained at molecular level (powder x-ray diffraction, NMR, and Raman spectroscopy). The composition of the gas phase and the hydrate phase were found to evolve over time, suggesting that kinetic and transport factors contribute in addition to thermodynamics. The information obtained with these molecular techniques combined with traditional gas uptake measurements can provide useful insights to understanding the mechanism of hydrate formation and decomposition.  Humphrey Davy, who observed that a solution of chlorine gas in water freezes faster than pure water, first found the existence of gas hydrates in 1811 (Davy, 1811). From 1811 until the 1930s gas hydrates were a scientific curiosity. Gas hydrates gained industrial interest when Hammerschmidt reported the formation of gas hydrates in oil and gas production and  7 transportation pipelines (Hammerschmidt, 1934). Currently, gas hydrates are important to study for variety of applications: a potential energy source, natural gas hydrate transportation, carbon dioxide capture and sequestration and separation of gases. Hydrates are also a serious safety concern in offshore hydrocarbon drilling (Hammerschmidt, 1934; van der Waals and Platteeuw, 1959a; Davidson, 1973a; Makogon, 1981; Ripmeester et al., 1987; Englezos, 1993; Englezos, 1996; Herri et al., 1999; Koh, 2002; Koh et al., 2002; Kvamme et al., 2005; Kelland, 2006; Hughes, 2008; Ribeiro and Lage, 2008; Sloan and Koh, 2008; Walker et al., 2008). This thesis focuses on kinetic inhibition of gas hydrate formation in pipelines of the oil and gas industry. The unexpected formation of gas hydrates in hydrocarbon production facilities and transportation pipelines can lead to blockages and shutdowns therefore is a serious economic and safety issue (Sloan, 1998a; Sloan and Koh, 2008). Significant research has been carried out in order to manage the risk arising from the formation of solid gas hydrate crystals in the oil and gas industry. There are four methods based on thermodynamic considerations to prevent the hydrate formation in pipelines, a) removing water (dehydration); b) increasing the temperature of the system beyond the hydrate formation temperature at a constant pressure; c) decreasing the pressure of the system below hydrate stability at a constant temperature, and d) injecting a thermodynamic inhibitor (methanol) to change the hydrate forming conditions. This allows the pipeline to operate outside the thermodynamic stability conditions to prevent hydrate formation (Long et al., 1994).   However, the use of thermodynamic inhibitors is both uneconomical (high concentrations are needed >40 vol %) and not environmentally friendly because methanol is flammable, corrosive and has a negative effect on processing catalysts in downstream operations (refiners). Currently the oil and gas industry is searching for hydrocarbons in deeper waters, which have  8 more favorable conditions for hydrate formation. High concentrations (>60 vol %) of methanol would be required to prevent the hydrate formation under these conditions. It is estimated that  the operating costs also can be greater than $500,000,000 per year (Koh, 2002).  In the last two decades economic and environmental factors have motivated research and development to identify alternative low dosage (less than 1 wt %) hydrate inhibitors (LDHIs). The main idea behind these LDHIs is to prolong induction time for hydrate nucleation, reduce growth and agglomeration instead of shifting the thermodynamic equilibrium conditions. Thus, LDHIs are divided into two basic categories: kinetic inhibitors (KIs) which prolong induction time for hydrate formation and affect growth of hydrate crystals, and anti agglomerates (AAs) which permit hydrates to form but prevent their agglomeration and accumulation in the pipelines. Commercially, kinetic inhibitors are synthetic polymers like polyvinylpyrrolidone (PVP) and polyvinylcaprolactam (PVCap) etc, and anti agglomerates are usually quaternary ammonium salts (QAS). The search for KI is ongoing in order to improve their performance and reduce cost. Newly discovered KIs are first used in the laboratory and if successful they are then tested with field fluids and finally under field conditions (Mehta and Sloan, 1993a). A large number of synthetic chemicals, mostly polymers, have been explored as potential KHIs (Freer and Sloan, 2000; Kelland, 2006). These inhibitors are water-soluble polymers and are known as low dosage hydrate inhibitors (LDHIs) because they are used at low concentrations to control hydrate formation, as opposed to the high concentrations required of thermodynamic hydrate inhibitors. Table 1.2 shows some of the known KI’s from Kelland’s review(2006). Several molecular dynamic simulation studies attempted to show how the kinetic inhibitors effect hydrate formation (Kvamme et al., 1997; Freer and Sloan, 2000; Storr and Rodger, 2000; Moon et al., 2002; Anderson et al., 2005; Kvamme et al., 2005; Kelland, 2006; Klomp, 2008;  9 Ribeiro and Lage, 2008). Kvamme et al., (1997) suggested that inhibitor structure and free energy of active groups in the inhibitor as well as diffusivity of these groups towards hydrate crystals are important characteristics of kinetic inhibitor. Table 1.2. Kinetic hydrate inhibitors (Kelland, 2006)            The molecular structures of these compounds are given below:                          Polyvinylpyrrolidone                       polyvinylcaprolactam                                 polyethylacrylamide                       polyvinyl-N-methylacetamide Kinetic Hydrate Inhibitor (KHI) Polyvinylpyrrolidone (PVP) Polyvinylcaprolactam (PVCap) Polyethylacrylamide Polyvinyl-N-methyl acetamide Polyethyloxazoline Polyisobutylacrylamide Polyacryloylpyrrolidine polydiethylacrylamide polyisopropylacrylamide polyethylmaleimide polyethyloxazoline polyisobutylacrylamide  10                         polyethyloxazoline                                  N-methyl-N-vinylacetamide:                                                          vinylcaprolactam  copolymer                                           polyacryloylpyrrolidine                               polydiethylacrylamide                                      polyisopropylacrylamide                             polyethylmaleimide                           polyethyloxazoline                                          polyisobutylacrylamide    polyisopropylmethacrylamide  Figure 1.2.  Molecular structures of chemical kinetic inhibitors (Kelland, 2006).   11 Anderson et al., (2005) proposed a two-step mechanism for gas hydrate inhibition by kinetic inhibitors. The hypothesis is that the inhibitor molecules first disrupt the local organization of water and guest molecules, increasing the barrier to nucleation and nucleation propagation. In the second step the degree of inhibition is related to the strength of the binding of the inhibitor molecule to the surface of the hydrate crystal. Kvamme et al., (2005) reported that the presence of inhibitor molecules decreases the hydrate-water interactions by increasing the hydrate-inhibitor interactions. Moon et al., (2007) also reported that inhibitor molecules (PVP) are effective in destabilizing hydrate crystals prior to nucleation and delay the nucleation. However, the mechanism of inhibition was not clear. Kumar et al.,(2008b) studied the effect of PVP on the morphology of methane/propane hydrate formation and found that induction time increases with the decrease in under cooling and increase in PVP concentration. The growth after nucleation was catastrophic at higher concentrations of PVP. Makogon and Holditch (1981) reported that the presence of kinetic inhibitors increased the temperature at which hydrate decomposed completely. As well, Bruusgaard et al., (2009) visually noted that methane hydrate formed in the presence of the copolymer poly(VP/VC) took longer to decompose completely as compared to  the hydrate formed in the absence of inhibitor.  Of course, visual observations do not reveal the actual complexities of melting behavior, and this remains unknown in these cases. Although the hydrate inhibition mechanism is unknown, from these results it is evident that PVP and PVCap are effective nucleation and growth inhibitors.  It is interesting to note that polymers exist that even though they are unable to exhibit kinetic inhibition by themselves but improve the performance of inhibitors (King et al., 2000). Cohen et al. (1998) investigated that the addition of butoxyethanol to a kinetic inhibitor delayed the induction time by thirty fold. Lee and Englezos (2005) found that addition of poly ethylene oxide  12 (PEO) to starches enhances the performance of the inhibitor by an order of magnitude. The presence of small amount of PEO in the kinetic inhibitor (GHI 101 from ISP technologies and Luvicap EG from BASF Corporation) solutions reduced the memory effect on induction time of hydrate crystals. Lee et al. (2007) observed that a starch solution showed weak inhibition effect,  but the addition of PEO to this starch solution enhanced the performance of the inhibitor. Unfortunately, since some of the synthetic polymeric inhibitors are non biodegradable, there has been some interest in assessing the utility of biological inhibitors (Zeng et al., 2003; Kelland, 2006; Zeng et al., 2006a; Zeng et al., 2006b; Al-Adel et al., 2008; Walker et al., 2008; Ohno et al., 2010). These biological inhibitors are labeled green inhibitors. They are natural inhibitors that have been engineered by evolutionary processes in organisms over a very long period of time. An example of a green inhibitor is antifreeze protein (AFP), which enable certain organisms to survive freezing winters by lowering freezing point (Raymond, 1977). Antifreeze proteins (AFPs) are best known from ocean fish that have evolved at high latitudes where these proteins adsorb to embryonic ice crystals and prevent serum freezing in the equilibrium crystallization gap (Yeh and Feeney, 1996; Davies et al., 2002). AFPs lower the freezing point relative to the melting point by adsorbing to the ice surface (Davies et al., 2002). AFPs can inhibit tetrahydrofuran (THF), propane, methane and CO2 hydrates (Zeng et al., 2003; Zeng et al., 2006a; Zeng et al., 2006b; Uchida et al., 2007; Al-Adel et al., 2008). Gordienko et al.(2010) observed that AFP adsorption could modify THF hydrate morphology, therefore indicating that AFPs could inhibit hydrate growth likely by an adsorption-inhibition mechanism.  The structures of AFP’s were shown in Figure 1.3. To our knowledge, only two reports have documented the utility of AFPs to inhibit hydrates formed from a natural gas mixture. Recently, Ohno et al., (2010) used HP-µDSC with a silica gel medium and reported biological  13 inhibitors can inhibit the natural gas hydrate formation. In addition, Jensen et al.,(2011) reported that an ice structuring protein was found to outperform PVCap for both structure I and structure II hydrate inhibition.                    Figure 1.3. The structures of AFP's with mutated residues Ala (blue) and Thr (yellow).                          Reprinted from ( Steffen et al., 2001), with permission from Elsevier.  Thus, it is evident that these biological inhibitors can inhibit natural gas hydrate formation but how well they perform compared to commercial KHIs in an environment that approximates field conditions remains unknown. Despite this progress, however, the reasons behind the activities of either chemical or biological inhibitors towards hydrate crystal nucleation and growth are not well understood. The knowledge of kinetics and molecular level information of hydrate formation in the presence of kinetic inhibitors is important to ensure that these hydrates will not have enough time to form and block flow in pipelines at hydrate formation conditions.  In addition to these restrictions, when a gas mixture is present, the formed hydrates are more complex than with a single component. For example, a mixture of methane and ethane can form sI if the methane concentration is below 75% or above 99%, but sII at intermediate  14 compositions (Ballard and Sloan, 2000; Subramanian et al., 2000a). Uchida et al.,(2004) reported that mixed methane and propane gases form sI and sII in two steps, with propane favored for enclathration, resulting in the an enriched methane gas phase, or a fractionation effect. In natural gas mixtures with three components, methane, ethane and propane, both sI and sII hydrates can form simultaneously depending on the gas composition (Uchida et al., 2006). Although the structures, cage occupancies and compositions of mixed gas hydrates have been reported (Subramanian, 2000; Uchida et al., 2004; Kida et al., 2007; Kumar et al., 2008b; Kida et al., 2009), the hydrate preparation protocol may impact these (Susilo et al., 2005; Seo et al., 2009), strongly suggesting that kinetics can play a significant role. Although the complexity associated with natural gas hydrates is daunting, it is nevertheless crucial information for industry in their efforts to manage hydrate formation.  The majority of inhibition studies use single hydrate formers rather than gas mixtures, because of the inherit complexity of natural gas blends with regards to the different hydrate equilibria, structural characteristics, and diffusion constants of each of the components (Uchida et al., 2007; Kumar et al., 2008b; Seo et al., 2009). Makogon et al. (1997) studied tetrahydrofuran (THF) hydrate analogue of sII natural gas hydrate and their kinetic inhibition. Larsen et al. (1999) reported results for ethylene oxide (EO) hydrate as an analogue of methane hydrate (sI). Since both THF and EO are liquids, and are miscible in water, thereby eliminating mass transfer limitations. However, these liquid hydrates may not be suitable analogs to the kinetic studies of the inhibition process of gas hydrates. There is evidence that hydrates crystallized in the presence of gas mixtures do not behave like those formed from single gases (Rydzy et al., 2007; Nakagawa et al., 2008; Bruusgaard et al., 2009; Nihous et al., 2010). Even less understood is the decomposition kinetics of mixed gas hydrates in the presence of KHIs.  15 Nevertheless, understanding hydrate decomposition kinetics and predicting hydrate decomposition rates in presence of KHIs is important for efficient hydrate plug removal in pipelines.   In spite of the fact that KHIs are deployed in the field we still do not completely understand how these molecules work at the molecular level. This lack of knowledge does not allow us to assess how a kinetic hydrate inhibitor would perform in the field (Klomp et al., 2008).   In view of the fact that new synthetic and green inhibitors are being developed this thesis aims to employ several macroscopic and molecular-level methods to assess the performance of (a) synthetic inhibitors provided by industry and (b) newly discovered green inhibitors by Professor Virginia K. Walker at Queens University (Biology). The idea is to subject the inhibitors to a variety of testing conditions and thus understand their performance and the mechanism by which nucleation and crystal growth is affected. The assessment will be conducted with the use of a model methane/ethane/propane gas mixture as the hydrate forming substance.  The specific objectives of this thesis and the anticipated outcomes are as follows; (a) To determine the performance of chemical and biological kinetic inhibitors on multi component hydrate formation and decomposition by isothermal and temperature ramping using high-pressure differential scanning calorimetry. This will reveal the extent of hydrate formation, hydrate nucleation temperature, hydrate nucleation time and hydrate stability in presence of inhibitors. (b)  To fabricate a small-scale apparatus (crystallizer volume of 58 cm3), which is suitable to test inhibitors (many biological inhibitors are available in limited quantities).  The possible  16 outcome is direct comparison of the formation and dissociation kinetics of mixed gas hydrates in the presence of chemical and biological kinetic hydrate inhibitors (KHIs).     (c) To determine the effect of chemical and biological kinetic hydrate inhibitors on the structure (powder x-ray diffraction) and composition of hydrate (Raman and NMR spectroscopy). This will provide information about the hydrate structure and degree of filling of the hydrate small and large cavities in the presence of inhibitors.    The fundamental information and importance of gas hydrates is described in Chapter 1. This chapter also discusses the safety and flow assurance challenges of oil and gas industry due to gas hydrate formation in pipelines. The state of the art on hydrate research relevant to hydrate inhibition is also discussed.  Chapter 2 discusses the effect of chemical and biological inhibitors on methane/ ethane/propane gas hydrate formation and decomposition in a custom small reactor using HP-DSC. Two types of temperature programming were used to investigate the performance of these inhibitors on hydrate nucleation, growth, and decomposition. Measurements of nucleation time of hydrate formation, amount of hydrate formed and hydrate decomposition temperature in the presence of inhibitors are presented and compared with pure water (control). In addition this chapter illustrates the differences in complex melting behavior of hydrate formed in two classes of inhibitors.  Chapter 3 provides the details of the procedure developed to observe the decomposition kinetics in the presence of kinetic inhibitors using a newly fabricated stirred reactor (crystallizer volume of 58 cm3).  The hydrate nucleation times, growth and decomposition kinetics in the presence of chemical and biological inhibitors reported and compared with a  17 control system. The inhibition strength of biological inhibitors was validated and compared with chemical inhibitors. The differences in onset and complete decomposition in the presence of inhibitors compared to control were presented. The importance, possible reasons and differences in decomposition behavior in the presence of inhibitors were discussed.  Chapter 4 describes the observed differences in structure and composition of mixed hydrates formed in the presence of chemical and biological inhibitors using gas chromatography, powder x-ray diffraction, Raman and NMR spectroscopy. Powder X-ray diffraction and NMR spectroscopy showed that structure II hydrates dominated, as expected, but in the presence of the chemical inhibitors structure I was also present.  The localized compositional complexity and heterogeneity were also observed in hydrate samples prepared in the presence of chemical inhibitor using Raman spectroscopy. The detailed procedure to calculate individual gas cage occupancy and calculated values were given in this chapter. The differences in cage occupancies and possible reasons were explained and correlated with complex melting and decomposition behavior observed in chapter 3 and 4.  The conclusions of this work and recommendations for potential future work are given in chapter 5.          18 2. NATURAL GAS HYDRATE FORMATION AND DECOMPOSITION IN THE PRESENCE OF KINETIC INHIBITORS. I: HIGH PRESSURE CALORIMETRY   Differential scanning calorimetry (DSC) is a thermo analytical technique that is used to measure the differential heat flow between the sample and a reference. The sample and the reference are cooled or heated at an identical rate. Gas hydrate formation and decomposition are phase transition processes. Hence, heat is released or absorbed in the sample cell compared to the reference under similar temperature environment. This differential heat flow signal is recorded and plotted versus temperature to identify the phase changes.   High pressure-micro differential scanning calorimetry (HP-µDSC) has proved useful for the study of hydrate formation (Dalmazzone et al., 2002a; Koh, 2002; Dalmazzone et al., 2004; Le Parlouer et al., 2004; Ripmeester et al., 2007; Lachance et al., 2009), decomposition (Rueff et al., 1988; Dalmazzone et al., 2002b; Kawamura et al., 2006; Hughes, 2008; Koh et al., 2008a; Koh et al., 2008b; Nakagawa et al., 2008) and inhibition (Koh et al., 2002; Koh et al., 2009; Lachance et al., 2009; Nihous et al., 2010). Recently, Ohno et al., (2010) used HP-µDSC with a silica gel medium to examine the effect of chemical and biological kinetic inhibitors on natural gas hydrates formation. The strength of this work was that it allowed the collection of sufficient data for statistical analysis without any serious effects of pore size on the thermodynamics of hydrate formation (pore size > 1 micron).  The formation equilibrium of mixed hydrates is complex and has been observed to shift (Uchida et al., 2007; Kumar et al., 2008b). Indeed, in the presence of small (< 1 micron) silica  19 pores the shift is to higher pressure and lower temperatures (Seo et al., 2009). Here we have used HP-µDSC to analyze the effect of chemical and biological kinetic hydrate inhibitors on a methane/ethane/propane hydrate formation, but without the use of silica. Therefore, the matrix does not influence the hydrate compositional changes, and we believe that this will provide insight into the mechanism of inhibition for the two types of inhibitors.    Two commercial hydrate kinetic inhibitors polyvinylpyrrolidone (average molecular weight ~10 kDa, from Sigma Aldrich), H1W85281 (average molecular weight ~3 kDa, a proprietary commercial product of unknown composition) and Type-III AFP (Swiss Prot Database accession number P19414; average molecular weight: 7 kDa) purchased from A/F Protein Canada Inc. This AFP was obtained by fermentation and secretion from recombinant Saccharomyces cerevisiae yeast cells, purified by differential diafiltration and kept at -20 ºC until used. Deionised water was used for the water controls and to dilute each of the inhibitors to 0.1 mM. The methane (93%)/ethane (5%)/propane (2%) gas mixture (UHP grade) was supplied by Praxair Technology Inc.  The HP–µ-DSC (DSC) used in this work was model VIIa; Setaram Inc., located at the National Research council, Ottawa, Canada and can be operated between 0.1 and 40 MPa and - 45 and +1000C. A ‘droplet insert’ (Figure 2.1) was fabricated using Torlon ® (polyamide-imide; Boedeker Plastics, Texas). Briefly, a base (dia-5.4 mm) with four depressions (dia-1.4 mm, depth-1.3 mm) was supported by rod (dia-1.68 mm, length-9 mm).   20  Figure 2.1. Droplet insert used for differential scanning calorimeter experiments  Samples of water or inhibitor soutions (1 µL) were placed into the depressions using a micro-syringe and then transferred to a high-pressure cell, and then pressurized (9 MPa) with the methane/ethane/propane gas mixture. When the pressure in the sample cells reached the desired value, one of two types of temperature programs were used (Figure 2.2). For isothermal experiments, the temperature remained constant at -8 °C for 10 h and the pressure was kept at 9.0 MPa. For temperature ramping experiments, the temperature was decreased from +25 to -45 0C at a rate of 0.25°C per min and then increased to 25 0C at the same rate.   Figure 2.2. Isothermal (left) and temperature ramping (right) programs used in DSC experiments.   21 In some experiments several temperature ramping protocols were followed sequentially so that the samples were effectively cycled through several freeze-thaw events with 5 min at 25 0C between cycles. The first re-crystallization was designated as cycle two, the subsequent re- crystallization designated as cycle three, and so on. Temperature ramping and isothermal experiments were repeated with each sample analyzed in quadruplicate a total of four times (n=16).  The amount of hydrate formed depends on the rate of the hydrate growth and the duration of the run, thus it was important that these parameters be well controlled in all experiments. The amount of hydrate formed was estimated by subtracting the ice to hydrate transformation (calculated from the observed exothermic peak areas) from the amount of total hydrate formed (calculated from the hydrate melting peak areas). The mean hydrate was then expressed as a percent relative to the hydrate formed from the water samples. The average nucleation time is defined as the average of all the nucleation times associated with the observed nucleation events in a particular isothermal experiment.   2.5.1. Isothermal experiments Isothermal experiments provided information on the induction time for hydrate nucleation, determined by the appearance of exothermic peaks (Figure 2.3).   22   Figure 2.3. Typical isothermal experiments showing hydrate nucleation events for (a) 11 h (b) 3 h  The first nucleation event observed in three experiments without hydrate inhibitors was at ~ 48 min with the first nucleation event delayed almost 10-20 min in the presence of PVP and AFP, and more than 1.5 h with H1W85281 (Table 1). Overall, average hydrate nucleation times were: water (114 min) <AFP-III (186 min) < PVP (204 min) < H1W85281 (252 min). Generally,  23 nucleation in the absence of inhibitors occurred within a narrow range of earlier times compared to the more dispersed times observed with inhibitors.  Table 2.1. DSC experiments with isothermal formation and temperature scanning decomposition in the presence of chemical and biological inhibitors Inhibitor State of solution First nucleation event (t/min) ± (3-5) Hydrate melting Temperature, Tm, hydrate (0C) ± 0.1 Area of ice melting peak  Hm, ice (J) ± 0.1 Area of hydrate melting peak  Hm, hydrate (J) ± 0.1 Fresh 65 19.6 2.4 4.5 Cycle 1 38 19.3 2.2 4.8 Water Cycle 2 42 19.4 2.2 4.6 Fresh 83 18.8, 20.3 2.1 3.4 Cycle 1 56 18.5, 20.1 2.2 3.7 Water/PVP Cycle 2 67 18.6, 20.5 2.0 3.2 Fresh 153 17.1, 19.2, 21.9 1.8 2.6 Cycle 1 137 16.9, 19.1, 22.1 2.0 3.1 Water/H1W85281 Cycle 2 162 16.9, 19.2, 22.0 1.9 2.7 Fresh 51 19.2 2.2 3.8 Cycle 1 67 18.9 2.1 3.3 Water/AFP Cycle 2 63 19.1 2.2 3.6  These isothermal experiments also allowed an examination of the endothermic peaks obtained after melting the hydrates (Figure 2.4 and Table 2.1). The peak areas representing the amount of hydrate formed with PVP and AFP samples showed a decrease in formed hydrate with 74% and 71% percent, respectively, of that seen with water. The hydrate formed with H1W85281 was only 43% of that found in water controls.  Strikingly, the melting curves were ‘sharper’ with water and with AFP, as compared to the broader, less regular peaks seen in the presence of the H1W85281 and PVP, possibly suggesting some heterogeneity in the hydrate formed with the latter.   24    Figure 2.4. Typical isothermal experiments showing hydrate melting peaks in water controls or with kinetic inhibitors (a) Heat flow Vs Time and (b) Heat Flow Vs Temperature        25 2.5.2. Ramping experiments In the DSC temperature ramping experiments, when water samples were cooled from 25 0C to -45 0C, several exothermic peaks were observed between -10 0C and -24 0C  (Figure 2.5 circled in blue), corresponding to hydrate and ice nucleation events, respectively. The peaks were initially small and then increased in size due to ice/hydrate formation. When the samples were subsequently reheated to 25 °C, two endothermic peaks were observed (Figure 2.5, circled in red). The peak at ~0 0C represents melting ice and was followed by a hydrate melting peak at ~19 0C.   Figure 2.5. Typical DSC cooling and heating curves obtained using the temperature ramping protocol in which the temperature of the samples is dropped from +25 0C to -45 0C and then reheated, all at 9 MPa.  A small exothermic peak was also observed at temperatures below the hydrate melting peak (~160C to 19 0C) suggesting that some of the water derived from melted ice was transformed into hydrate. Table 2.2 summarizes the results obtained in the ramping experiments  26 in the presence of inhibitors. In the presence of any of the kinetic inhibitors the hydrate melting peaks were smaller than those seen with the water controls (Figure 2.6), indicating that the amount of formed hydrate was reduced. With only 29% of the total calculated hydrate formed relative to water samples however, H1W85281 was the most effective, compared to PVP and AFP with 65% and 53%, respectively.   Table 2.2. DSC formation/decomposition experiments with temperature scanning in the presence of chemical and biological kinetic inhibitors Inhibitor State of solution Ice melting temperat ure Tm, ice (0C) ± 0.01 Area of ice melting peak Hm, ice (J) ± 0.1 Hydrate melting temperature Tm, hydrate (0C) ± 0.1  Area of hydrate melting peak Hm, hydrate (J)  ± 0.1 Total area of melting peaks Hm, total (J)  ± 0.1 Fresh 0.18 3.66 18.8 0.48 4.14 Cycle 1 0.18 3.59 18.3 0.51 4.10 Cycle 2 0.19 3.47 18.1 0.55 4.02 Water Cycle 3 0.21 3.38 17.9 0.63 4.01 Fresh 0.16 4.10 20.1, 21.4 0.35 4.45 Cycle 1 0.13 4.06 19.9, 21.6 0.46 4.52 Cycle 2 0.21 4.01 19.8, 21.4 0.57 4.58 Water/PVP Cycle 3 0.19 3.98 19.8, 21.9 0.59 4.57 Fresh 0.59 4.41 15.1, 19.0, 22.5 0.18 4.59 Cycle 1 0.61 4.41 15.3, 19.1, 22.3 0.17 4.58 Cycle 2 0.15 4.46 15.7, 19.3, 22.3 0.18 4.64 Water/ H1W85281 Cycle 3 0.22 4.47 19.3, 22.4 0.15 4.62 Fresh 0.62 4.03 17.0 0.22 4.25 Cycle 1 0.59 4.00 16.9 0.20 4.20 Cycle 2 0.22 3.97 17.0 0.23 4.20 Water/AFP Cycle 3 0.22 3.94 16.9 0.25 4.19    Similar to the observations made with the isothermal experiments, more than one endothermic hydrate peak (Table 2.2) was detected in the presence of either chemical inhibitor: two peaks (~20 0C and ~22 0C) were observed in the presence of PVP and three peaks (~15 0C,  27 19 0C and 22 0C) were seen with H1W85281. Lachance et al.,(Lachance et al., 2009) also reported that methane hydrate formed in the presence of PVCap showed different peaks upon melting that varied with inhibitor concentration. Such multiple peaks suggest hydrate with different compositions, and thus non-uniform crystals are formed in the presence of PVP and H1W85281. In contrast to these results, a single endothermic peak was seen at ~17 0C in the presence of AFP (Figure 2.6) as well as a single peak at ~19 0C in the water controls (Figure 2.5 and Figure 2.6), again echoing the observations made with the isothermal experiments.   Figure 2.6. Typical hydrate melting curves in the presence of kinetic inhibitors at 9 MPa using the temperature ramping protocol, in the presence or absence of kinetic inhibitors  When water samples from the ramping experiments were subjected to an additional freeze-thaw cycle, less ice and 10-15% more hydrate formed, compared to fresh samples (Table 2.2). It would be surprising if this was due to residual hydrate template since the samples were incubated for 15 min at 250C between the initial temperature ramping run and the subsequent cycle. Alternatively, it may be due to the ‘memory effect’, perhaps mediated by ‘imprinted’  28 contaminants (Zeng et al., 2006a). More hydrate (25%) was formed in subsequent freeze-thaw cycles in the presence of chemical inhibitor PVP. Notably, the amount of hydrate remained about the same in the presence of AFP and H1W85281, no matter the number of re-crystallization cycles (Table 2.2).  When water controls were cycled through several ramping profiles allowing ice and hydrate to reform more than once, the hydrate melting temperature in subsequent cycles appeared to shift only marginally to slightly lower temperatures (from 18.8 0C to 17.90C) (Figure 2.7a; Table 2.2) In addition, AFP samples with a single peak at ~17 0C showed little variance over subsequent cycles (Figure 2.7b). In striking contrast, multiple hydrate melting peaks were observed with samples containing the PVP and H1W85281. In the presence of PVP, the hydrate melting temperature shifted from an initial single peak at 20.1 0C to two peaks (19.9 0C and 21.6 0C) in cycle two (Figure 2.7c). In cycles three and four, the lower temperature peak decreased relative to previous cycles and concomitant with a gradual increase in the temperature of the second peak. Similarly, multiple hydrate melting peaks (15.1 0C, 19 0C and 22.5 0C) were observed in the presence of H1W85281 in fresh samples but in re-crystallization cycles there were clear temperature shifts (Figure 2.7d). Each flanking peak moved toward the 19 0C peak, so that by cycle four, the predominant hydrate melting temperature was at 19 0C with a broader peak at ~22 0C. The amount of hydrate formed varies in the sequential cycles in the presence/absence of inhibitors (Table 2.2), so there is possibility the hydrate composition can change and affect the stability (shift in melting temperatures). It is curious that multiple hydrate melting peaks were observed only with PVP and H1W85281, suggesting that hydrate formed in the presence of these inhibitors was heterogeneous in contrast to the seemingly homogenous melting observed with water and AFP samples. Significantly, evidence for the complex  29 (composition change with time) nature of hydrate formed in the presence of PVP and H1W85281 has been observed using Raman and NMR spectroscopy (Daraboina et al., 2011c) and gas uptake experiments (Daraboina et al., 2011a), but not for biological inhibitors.    30   Figure 2.7. Hydrate melting curves formed during ramping runs in (a) pure water (b) AFP (c) PVP              (d) HIW85281 2.5.3. Hydrate heterogeneity  As indicated, hydrates formed in the presence of any of the inhibitors started melting at a lower temperature than hydrate formed in water (Figure 2.4; Table 2.1). Melting points were slightly lower and multiple hydrate melting peaks were observed in the presence of PVP and  31 H1W85281, arguing for hydrate structures of greater complexity compared to water or AFP samples (Table 2.2). Delayed melting has been previously observed for hydrates prepared in the presence of inhibitors. For example, Makogon and Holditch (2001a; 2001b) reported that the presence of kinetic inhibitors increased the temperature at which hydrate decomposed completely. As well, Bruusgaard et al. (2009) visually noted that methane hydrate formed in the presence of the copolymer poly(VP/VC) took longer to decompose completely as compared to  hydrate formed in the absence of inhibitor.  Of course, visual observations do not reveal the actual complexities of melting behavior, and this remains unknown in these cases.   Increased hydrate stability, as reflected by decomposition at elevated temperatures, depends on hydrate composition. ‘Pure’ materials melt in a narrow temperature peak, and the presence of impurities can broaden the melting temperature range (Mccullough and Waddington, 1957). It is not known if a hydrate of one composition can act as an ‘impurity’ in the hydrate of another composition, or if inhibitors can act as impurities. Since we have evidence that AFPs adsorb to hydrates using an adsorption-inhibition mechanism(Gordienko et al., 2010; Ohno et al., 2010), and the sharp melting peak was 2 °C lower than seen in the water controls (Figure 2.7a and b), we speculate that the AFP participates in hydrate formation possibly by initiating a clathrate-water complex and subsequently becomes incorporated into the growing hydrate. Strikingly, the melting point remained consistent and predictable after multiple re- crystallizations. Of all the inhibitors tested, the fact that all the hydrates formed in the presence of the AFP melted at the lowest temperatures confirms their utility in pipelines, especially at high latitudes where their dual role as hydrate and ice inhibitors may be useful.   In contrast to the single, presumably homogenous melting peak observed with AFP, multiple peaks observed in the hydrate formed in the presence of the PVP and H1W85281,  32 suggest a more complex interaction. The fact that in H1W85281 samples, multiple peaks melted over an astonishing 6 °C range (Figure 2.7d), indicates that hydrates of different compositions were formed. The natural gas mixture used in these experiments was expected to generate sII hydrate, especially since the pressure was kept below 15 MPa (Jager and Sloan, 2002). However, others have reported that even at pressures approximating those used here, mixtures of sI and sII hydrates can form (Schicks et al., 2006). The latter study also showed that in heterogeneous hydrates formed from four different natural gas-like mixtures of methane, ethane and propane, there was increased stability after formation in the presence of higher proportions of the heavier gases. Thus, we propose that H1W85281 acts as more than a simple kinetic inhibitor. We speculate that in addition to this role it acts as such an effective inhibitor so that complex mixtures of sII and I are formed with possible partial exclusion of methane from the large cages into sI methane hydrates. Although PVP and H1W85281 were effective, the heterogeneity of the formed hydrates and the changing complexity with each crystallization cycle as well as the changing thermal stability of hydrate (Figure 2.7) will surely make their utilization in the field more challenging. Unlike the AFP samples that showed consistent and predictable behavior in multiple cycles, the practical use of H1W85281 and PVP may demand more operator experience.       33 3. NATURAL GAS HYDRATE FORMATION AND DECOMPOSITION IN THE PRESENCE OF KINETIC INHIBITORS:  II STIRRED REACTOR EXPERIMENTS  It is a challenge to model pipeline conditions in the laboratory. Stirred reactors are predominantly used for hydrate studies. Stirred vessel, originally designed by Vysniauskas and Bishnoi (1983a), with their utility demonstrated by Bishnoi and his colleagues (Englezos et al., 1987b; Englezos et al., 1987c), subsequently modified and used in multiple studies (Cohen et al., 1998; McCallum et al., 2007; Al-Adel et al., 2008). Since many biological inhibitors are available in limited quantities, we have fabricated a small-scale crystallizer (volume of 58 cm3), based on industry-favored reactors.  Using this new equipment, we have successfully compared two different commercial synthetic inhibitors and two different AFPs for their impact on mixed gas hydrate nucleation, growth and decomposition.   Deionized, distilled water was used to prepare all solutions. Two commercial KHIs were used: polyvinylpyrrolidone (PVP; ~10 kDa, from Sigma Aldrich) and H1W85281 (~3 kDa, a proprietary commercial product of unknown composition). Two fish AFPs purchased from A/F Protein Inc. were used: Type I AFP (AFP-I; average molecular weight: 3.3-4.5 kDa), purified from fish serum, and Type III AFP (AFP-III; Swiss Prot Database accession number P19414; average molecular weight: 7 kDa), purified after fermentation and secretion from recombinant Saccharomyces cerevisiae yeast cells. The methane (93%)/ethane (5%)/propane (2%) gas mixture (UHP grade) was supplied by Praxair Technology Inc.   34  Figure 3.1shows the schematic of the apparatus. It consists of a crystallizer (CR) which is a cylindrical vessel (ID = 3.00 cm, height = 7.07 cm) made of 316 stainless steel, with a volume of 58 cm3. A 150 cm3reservoir (R) supplied gas during hydrate formation in a semi-batch operation.   Figure 3.1. Schematic of the apparatus for hydrate induction and decomposition   The crystallizer and the reservoir were immersed in a temperature-controlled water bath, regulated by an external refrigerator (VWR Scientific). Two Rosemount smart pressure transmitters (model 3051, Norpac controls, Vancouver, BC) with a maximum uncertainty of 0.075% of span 0-15,000 kPa (i.e. 11 kPa) were employed. The temperature of the hydrate phase  35 and the gas phase of the crystallizer were measured using Omega (Omega Engineering, Stamford, CT) copper-constantan thermocouples with an uncertainty of 0.1 K. A valve (Fisher- Baumann) coupled to a PID controller and connected between the reservoir and the crystallizer regulated the flow of gas from the reservoir to the crystallizer and vice versa. The data acquisition system (National Instruments) was coupled with a computer to record the data as well as to communicate with the control valve, and used LabView 8.0 (National Instruments) software. Figure 2.4 shows the top view and the cross sectional view of the designed crystallizer with the design specifications.   Figure 3.2. Reactor top view (left) cross sectional view (right) of the reactor    3.4.1. Hydrate formation The crystallizer was loaded with 0.5 wt% KHI solution (10 mL), and pressurized with the gas mixture and then depressurized (at a pressure below the equilibrium hydrate formation pressure) three times in order to remove air from the system. Subsequently, the crystallizer temperature  36 and pressure were set to the desired level, and when this was achieved (approximately 5 min), this was set as time zero for the experiments. Experiments were routinely conducted at 275.15 K and 8.1 MPa. The equilibrium hydrate formation (sII) pressure for the gas mixture at 275.15 K is 1.06 MPa (Sloan and Koh, 2008).  The aqueous solution in the crystallizer was stirred using a magnetic stirrer at a constant speed 400 rpm. All hydrate formation experiments with and without inhibitors were carried out in a semi- batch manner (constant pressure and temperature, with a fixed amount of aqueous solution and continuous supply of gas). The nucleation point or induction time was identified based on a sudden temperature rise or increased gas consumption. Hydrate formation is associated with the incorporation of gas and a consequent drop in crystallizer pressure. Here constant pressure was maintained with the PID controller. Pressure (P) and temperature (T) measurements were used to calculate the number of moles of gas consumed (gas uptake) by the following equation (Linga et al., 2007).     Where nH is the number of moles consumed to form hydrate (H) or dissolved in water at time t and zero, z is the compressibility factor calculated by Pitzer’s correlation, and V is the volume of the crystallizer. The memory experiments were conducted using the same procedure explained above except the fact that the memory experiments were started four hours after complete decomposition of hydrates formed in fresh solutions.  37 3.4.2. Hydrate decomposition After hydrate formation, crystals were decomposed by heating the water bath (275 to 295 K) and at 8.1 MPa at the start of the decomposition experiment, with similar heating profiles for each experiment. Briefly, the procedure is as follows. After the end of the formation experiment the heater was turned on (time zero for the decomposition experiment) to heat the reactor from 275 to 295 K. The stirrer speed of 400 rpm used for the formation experiment was continued and the data was recorded every 20 seconds. The hydrates start to decompose once the temperature crosses the equilibrium phase boundary, concomitant with an increase in crystallizer pressure. The expansion of gas due to the temperature-mediated increase was calculated by conducting a control experiment with no hydrate formation. The procedure for control experiment is as follows, water (10 mL) was introduced into the crystallizer and the pressure was set to 8.1 MPa, and the temperature was increased from 275 to 295 K without any mixing. The temperature and pressure was monitored for the control experiments. The difference between the hydrate experiments (gas expansion due to temperature rise and gas released due to hydrate decomposition) and the no-hydrate experiment corresponded to gas release attributed to hydrate decomposition.  The normalized gas release is calculated as follows, Normalized gas release = n/nt Where, n is the number of moles of gas released at any given time of the experiment and nt is the total number of moles of gas recovered at the end of the experiment.   38  3.5.1. Hydrate nucleation and growth in the presence of inhibitors  As expected, there was a significant delay in the onset of hydrate nucleation, as determined by the longer induction time in the presence of any of the KHIs (Table 3.1). When all experiments were compared, the commercial inhibitor, H1W85281 was the most effective in prolonging the period before nucleation. AFP-I was modestly more effective than PVP and AFP III (Figure 3.3). When compared to water controls, hydrate nucleation was delayed by a factor of 1.4 in the presence of AFP-III, 2.2 in the presence of PVP, 3.2 by AFP-I, and 33.6 by H1W85281. Overall, average hydrate nucleation times were: water (6.2 min) <AFP-III (8.9 min) < PVP (13.5 min) < H1W85281 (208.3 min).  It is noted that the average hydrate nucleation times in our HP-DSC work (Daraboina et al., 2011b), were water (114 min) <AFP-III (188 min) < PVP (204 min) < H1W85281 (252). The order of delay of induction in the presence of inhibitors correlates well between HP_DSC (1µL) and stirred reactor (10 mL) results.  It is not yet clear what processes govern the hydrate nucleation rate. It has been suggested that KHIs act by adsorbing to heterogeneous nucleators such as impurities in water phase and walls of crystallizer (Zeng et al., 2006a; Ohno et al., 2010). If correct, then H1W85281 is more likely, by an order of magnitude, to interact with such foreign materials thereby minimizing nucleation sites in order to delay the induction time.  39  Figure 3.3. Induction time for hydrate formation in the presence of inhibitors at 8.1 MPa.  The scale for the induction time is logarithmic due to large differences in the induction times between the inhibitors.     All the inhibitors reduced overall hydrate growth, or the moles of gas consumed 10 h after the initial nucleation (Table 3.1). In the presence of AFP-III and PVP, gas consumption was reduced by 13-14% compared to that seen in water controls. AFP-I reduced gas uptake by 30% and H1W8528 reduced it by 54%. Thus, the commercial KHI, H1W85281 was also the most effective in reducing hydrate growth concomitant with the superior delay in the onset of hydrate formation.    40  Table 3.1. Experimental conditions, induction times and mean gas consumption for methane/ethane/propane gas hydrate formation at 275.15 K and 8.1 MPa. Gas mixture CH4 (93%) C2H6 (5%) C3H8 (2%) # Sample State Induction time (min) Moles consumed after 10 h Mean induction time ¯tind (min) Mean  gas consumption  ¯n (moles) 1 Fresh 8.0 0.0167 2 Memory 4.3 0.0166 3 Fresh 6.7 0.0180 Water  4 Memory 5.7 0.0178 ¯tind, = 6.2  ¯tind, Fresh= 7.3  ¯tind, Memory= 5 ¯n= 0.0173  ¯n Fresh= 0.0174  ¯nMemory= 0.0172 5 Fresh 19 0.0157 6 Memory 12 0.0151 7 Fresh 15.7 0.0149 Water+ 0.5 wt% PVP 8 Memory 7.4 0.0146 ¯tind, = 13.5  ¯tind, Fresh= 17.3  ¯tind, Memory= 9.7 ¯n= 0.0151  ¯n Fresh= 0.0153  ¯nMemory= 0.0149 9 Fresh 292 0.0093 10 Memory 346 0.0088 11 Fresh 73.7 0.0076 Water+ 0.5 wt% H1W85281 12 Memory 121.4 0.0073 ¯tind, = 208.3  ¯tind, Fresh= 182.8  ¯tind, Memory= 233.8 ¯n= 0.0079  ¯n Fresh= 0.0085  ¯nMemory=0.0081 13 Fresh 32.0 0.0117 14 Memory 15.3 0.0109 15 Fresh 14.0 0.0101 Water+ 0.5 wt% AFP I 16 Memory 17.0 0.0104 ¯tind, = 19.6  ¯tind, Fresh= 23.0  ¯tind, Memory= 16.2 ¯n= 0.0121  ¯n Fresh= 0.0109  ¯nMemory=0.0107 17 Fresh 8.7 0.0156 18 Memory 8.0 0.0151 19 Fresh 11.7 0.0147 Water+ 0.5 wt% AFP III 20 Memory 7.3 0.0143 ¯tind, = 8.9  ¯tind, Fresh= 10.2  ¯tind, Memory= 7.7 ¯n= 0.0149  ¯n Fresh= 0.0152  ¯nMemory=0.0147  The impact of the various inhibitors on hydrate formation can be best visualized on a time profile (Figure 3.4). During the first 5 min, the rate of hydrate growth in the presence of any of the inhibitors was similar, and slower than the water controls. Subsequently, the profile  41 changed, and after 120 min the slowest growth rates were seen with H1W85281 and AFP I. Hydrate formation in the absence of KHIs reached a plateau in 5-6 h as expected in a stirred vessel where the nucleation and growth are followed by the state of crystal agglomeration (Englezos, 1996). When inhibitors were present, hydrate formation continued to increase slowly for the duration of the experiment, except for AFP-I, which similar to the water samples appeared to plateau, although at a significantly lower level than controls.   Figure 3.4. Effect of inhibitor on hydrate growth for experiments conducted at 8.1 MPa and 275.15 K for experiments 1, 5, 9, 13 and 17 from Table 1.  Time zero in the graph corresponds to the nucleation point (induction time given in Table 1) for the experiments.   KHIs are believed to reduce the transport of guest molecules to the hydrate surface (Moon et al., 2002; Moon et al., 2003).This may be mediated by their adsorption onto the  42 hydrate surface, which effectively decreases the crystal growth rate(King et al., 2000; Carstensen et al., 2002; Anderson et al., 2005; Kumar et al., 2008a).Gordienko et al., (2010) similarly demonstrated that AFPs adsorb to THF hydrate, suggesting that both commercial and biological inhibitors likely inhibit hydrate growth by an adsorption-inhibition mechanism. Although Kvamme et al., (2005) correlated hydrate-inhibitor interactions to binding strengths, it is not clear how these could differ since molecules should either bind, and be incorporated into the growing hydrate crystal at higher driving forces or over longer periods of time, or not bind. Alternatively, we suggest that very effective KHIs, once close to the hydrate surface have a high probability of being ‘anchored’ using a chemical group that fills an empty cage, resulting not only in the inhibition of further crystal growth but also partially explaining the reduction in gas uptake. Such a model has been developed in silico for PVP on hydrates (Wathen et al., 2010). In such a model, H1W85281 would have the highest probability of incorporation into the hydrate crystal.   3.5.2. Hydrate decomposition in the presence of inhibitors Hydrates formed in the presence of inhibitors were decomposed by heating, which resulted in consistent melting profiles (Figure 3.5). The heating profiles for the other experiments are not shown (Daraboina et al., 2011b).   43  Figure 3.5. Temperature profiles for decomposition experiments (experiments 1, 2 and 3 from Table 1).  The increase in pressure generated by decomposition was also consistent, as demonstrated by Figure 3.6, which shows the pressure profile of two experiments conducted under the same experimental conditions. In the absence of formed hydrate, increased pressure was solely due to gas expansion as temperature increased, with a much higher pressure associated with decomposition of hydrates.   44  Figure 3.6. Pressure profiles for No-hydrate experiment (NHE, gas expansion due to temperature increase) and hydrate experiment (HE, gas expansion due to temperature increase and gas release due to decomposition of hydrate) and difference (HE-NHE, gas release due to hydrate decomposition)  A typical decomposition curve obtained after accounting for the gas expansion is shown in Figure 3.7.   45  Figure 3.7. Typical gas hydrate decomposition along with heating profile for experiment 1 in Table 3.1  The normalized hydrate decomposition profiles for all the systems investigated are shown in Figure 3.8. It is clearly evident from the figure that decomposition started later in the absence of inhibitors, and was consistent between experiments (Figure 3.9). Complete decomposition of hydrates formed in the presence of the commercial KHIs, PVP and H1W85281, appeared to be in two stages, and was so protracted that maximum pressures were not achieved until after the control samples (water only) had decomposed (Figure 3.9b and Figure 3.9c).  These observations are not surprising since it has been previously reported that kinetic inhibitors can increase the temperature (Makogon and Holditch, 2001a; 2001b) or the time, (Lee et al., 2006; Bruusgaard et al., 2009) required for complete hydrate decomposition.    46  Figure 3.8. Normalized gas release profiles in the presence of inhibitors (experiments 1, 7, 9, 13 and 17).  A two-stage hydrate decomposition profile may be due to inhomogeneous hydrate formation, composed of both structure I and II hydrate as has been previously reported in mixed gases of methane and ethane, as well as in more complex natural gas mixtures (Subramanian, 2000; Schicks et al., 2006; Uchida et al., 2007; Kumar et al., 2008b). Makogon and Holditch (2001a; 2001b) reported that the presence of KHIs increased the temperature at which hydrates completely decomposed. It is also noteworthy too that Schicks et al. (2006) reported that the hydrate decomposition line shifted towards higher temperatures with increasing concentrations of ethane and propane in a gas mixture. Thus, it is possible that heavier hydrocarbons participate to a greater extent in hydrate formation, which likely has a heterogeneous crystal structure, in the presence of the commercial KHIs. In an alternative but related explanation, new hydrate,  47 possibly of a different composition, could form subsequent to the temperature drop induced by the ongoing dissociation of the existing hydrate and/or the changing gas mixture composition. This newly formed hydrate may even act as an impurity in hydrates of a different composition. As well, since melting temperatures can vary depending on hydrate dimension (< 1 micron), commercial KHIs could also have an impact on crystal size.     48     49   Figure 3.9. Normalized gas release profiles in the presence of (a) Water (b) PVP (c) H1W85281 (d) AFP-I (e) AFP-III  50  In contrast to the two-stage decomposition observed with the commercial KHIs, the hydrate melting profiles with the two biological inhibitors were strikingly simpler. Single-stage decompositions, more similar to the water controls albeit at lower temperatures, were seen in the presence of AFP–I and III (Figure 3.9d and Figure 3.9e). As well, unlike the case with PVP or H1W85281, there was such a rapid decomposition that the overall melting period was even less than that required for hydrates formed in the absence of any inhibitors (Figure 3.8). These observations were consistent with our previous study on hydrate formation and decomposition using high pressure differential scanning calorimetry (DSC), in that multiple endothermic peaks were observed with the commercial KHIs in contrast to the less complex profiles seen with AFP- III (Daraboina et al., 2011b) . Although the reasons for this distinct behavior in the two types of inhibitors are unknown, we speculate that crystals formed in the AFP-containing solutions were more homogeneous, consistent with a sharp melting profile, and that the incorporation of the protein resulted in a less stable crystal, which decomposed at lower temperatures compared to the commercial KHIs. In order to have a better understanding of these complex observations, the compositional and structural analysis using GC, XRD, Raman and NMR spectroscopy are presented as Part III (Daraboina et al., 2011c).          51 4. NATURAL GAS HYDRATE FORMATION AND DECOMPOSITION IN THE PRESENCE OF KINETIC INHIBITORS: III STRUCTURAL AND COMPOSITIONAL CHANGES  Despite the challenge there can be little doubt that understanding the enclathration of natural gas mixtures in the presence of KHIs would greatly assist in the development of future inhibitors to manage hydrate formation in the field. Traditionally, powder X-ray diffraction (PXRD) has been used to determine the type of hydrate (Uchida et al., 2004; Ripmeester et al., 2007; Kumar et al., 2008b), with Raman spectroscopy used as a tool for guest identification and in certain circumstances, hydrate composition and cage occupancy (Sum et al., 1997; Subramanian, 2000; Tulk et al., 2000; Kumar et al., 2008b). A third technique, 13C NMR spectroscopy, has also proved useful for obtaining quantifiable cage occupancies of mixed gas hydrate (Ripmeester and Ratcliffe, 1988; Ripmeester and Ratcliffe, 1999; Kida et al., 2007; Kumar et al., 2008b; Kida et al., 2009).   As indicated, only a few molecular descriptions are available for natural gas hydrates in the presence of inhibitors (Koh, 2002; Kelland, 2006; Ohno et al., 2009; Seo et al., 2009; Ohno et al., 2010) but, they suggest that inhibitors can have an impact on the structure and composition of mixed hydrates. Exothermic and endothermic peak analysis reflecting hydrate formation and disassociation, respectively, and monitored by differential scanning calorimetry (DSC) were more complex in the presence of two chemical KHIs, PVP and H1W85281 (Daraboina et al., 2011a).  In addition, hydrates formed in a stirred reactor with these KHIs decomposed in two stages, again suggestive of mixed structures (Daraboina et al., 2011c).  Here we observed the  52 effect of chemical and biological KHIs on natural gas hydrate formation using three solid state analytical techniques, in addition to gas chromatography to measure the vapour phase composition changes during hydrate formation. It is our belief that knowledge of the guest distribution and homogeneity of the hydrate products is important for an understanding of inhibition mechanisms of different KHIs.     Two commercial, chemical hydrate kinetic inhibitors polyvinylpyrrolidone (PVP; ~10 kDa, from Sigma Aldrich) and H1W85281 (~3 kDa, a proprietary commercial product of unknown composition) were used. The active ingredient concentration of the H1W85281 was 40 wt%. A biological inhibitor, Type-III AFP (~7 kDa, purchased from A/F Protein Canada Inc., Swiss Prot Database accession number P19414), was obtained by fermentation and secretion from recombinant Saccharomycescerevisiae yeast cells, purified by differential diafiltration and kept at -20 ºC until used. Deionised water was used to dilute the inhibitors to 0.1 mM. A synthetic natural gas mixture consisting of methane (93%)/ ethane (5%)/ propane (2%) was used in all experiments. It was at extra high purity, and purchased from Praxair Canada.  To prepare hydrate samples water (3 mL) and inhibitor (0.1 mM PVP, H1W85281 or AFP-III) solutions were frozen (253 K), ground, and then loaded into a 50 mL pressure vessel. The loading was performed in the freezer (~253 K) to prevent the melting of ice. The vessel was pressurized with the methane/ethane/ propane mixture until 9 MPa and kept in the cooler for about 1 h until the solutions nucleated. The vessel was then transferred to a water bath  53 maintained at 274 K. After 48 h the hydrate samples were collected from the vessel under liquid nitrogen and stored in liquid nitrogen to avoid hydrate decomposition.  4.4.1. Gas chromatography (Gas phase analysis)                  A Varian CP-3800 gas chromatograph (GC) with a thermal conductivity detector and flame ionization detector along with a CP-PoraPLOT U capillary column were used to analyze the hydrate samples. Ultra high purity He was used as carrier gas. The gas sample was transferred from the crystallizer to a 3.1 mm stainless steel sampling tube, with a volume of 300 µL. Considering that the volume of the crystallizer is 323 mL, sampling did not affect the mass balance calculations (Linga et al., 2007). The sampling tube was flushed three times with He before samples were collected for analysis. Subsequently, the gas from the sampling tube was injected into the GC through a six-port valve with a sampling loop (100 µL).  4.4.2. Powder X-ray diffraction (PXRD) In this work PXRD was used to detect the structural transition of methane/ethane/propane hydrate in the presence of synthetic and biological inhibitors. To obtain crystal structure information on hydrates PXRD measurements were obtained on an instrument (40 KV, 40 mA, BRUKER axs model D8 advance) using a θ/2θ step scan mode, a step size of 0.01, and a counting time of 0.4 s/step using CuKα radiation (λ= 1.5406). The measurements were performed at atmospheric pressure and 125 K to prevent hydrate dissociation. The PXRD patterns were indexed with available sI and sII patterns in literature.  54 4.4.3. Raman spectroscopy  To obtain localized structural and compositional analysis, hydrate samples were analyzed with an Acton Raman spectrometer equipped with fiber optics, a grating (1200 grooves/mm), and a charged couple detector (CCD).  An Ar–ion laser was used as the excitation source (514.53 nm). The laser was focused on the sample using a 10x microscope objective. The spectrograph was computer-controlled and spectra were recorded with a 1 s integration time over 200 scans.  4.4.4. NMR spectroscopy A DSX 400 MHz NMR spectrometer was used to record 13C spectra for the hydrate samples. Guest 13C resonance intensities were used to determine the cage occupancy of individual gases and the gas composition of the hydrate samples. A Zirconium rotor charged with powdered hydrate at low temperatures was loaded into a low temperature probe maintained at 173 K.  Single pulse excitation (900 of 5 µs) and pulse repetition delay of 300 s under proton decoupling and magic angle spinning at 2.5 kHz was employed to record spectra. The detailed procedure has been described by (Kumar et al., 2008b). Cage occupancies were calculated using the assumptions and descriptions as provided in literature (Ripmeester and Ratcliffe, 1988; Kida et al., 2007; Kumar et al., 2008b; Kida et al., 2009).  4.5.1. Gas phase analysis and powder XRD When hydrates were formed from the mixed gases, gas phase analysis showed that the concentration of methane increased from 93% to 94.5% over 2.5 h (Table 4.1). The increase was mirrored with a gas phase decrease in the ethane and propane concentrations of 5% to 4.1% and 2% to 1.4%, respectively. This is evidence of the fractionation that takes place during gas  55 hydrate formation with pure water (Linga et al., 2006; Linga et al., 2007), and is likely due to different rates of incorporation of the three gas components in the hydrate lattice.    Table 4.1. Vapour phase composition during hydrate formation in the absence (water) or the presence of various inhibitors at 9.0MPa and 274 K.  Water PVP H1W85281 AFP Time (min) CH4 C2H6 C3H8 CH4 C2H6 C3H8 CH4 C2H6 C3H8 CH4 C2H6 C3H8 0 0.930 0.050 0.020 0.930 0.050 0.020 0.930 0.050 0.020 0.930 0.050 0.020 15 0.932 0.047 0.021 0.933 0.049 0.018 0.932 0.049 0.019 0.931 0.048 0.021 30 0.933 0.048 0.019 0.936 0.047 0.017 0.937 0.047 0.016 0.934 0.046 0.020 45 0.938 0.048 0.014 0.941 0.045 0.014 0.944 0.044 0.012 0.937 0.044 0.019 60 0.94 0.047 0.013 0.944 0.042 0.014 0.949 0.04 0.011 0.942 0.045 0.013 90 0.943 0.044 0.013 0.948 0.039 0.013 0.953 0.037 0.01 0.945 0.041 0.014 120 0.944 0.042 0.014 0.952 0.036 0.012 0.956 0.034 0.01 0.947 0.039 0.014 150 0.945 0.041 0.014 0.956 0.033 0.011 0.961 0.030 0.009 0.948 0.037 0.015  The addition of chemical KHIs (PVP and H1W85281) changed the proportion of the gases in the vapor phase. In the presence of PVP the concentration of methane was 1% higher than in water controls after 2.5 h, and reached 95.6% after 2.5 h (Table 4.1). Similarly, compared to controls, concentrations of ethane and propane decreased (20%), and reached 3.3% and 1.1%, respectively. Likewise, in the presence of H1W85281 the concentration of methane after 2.5 h was 96.1% and those of ethane and propane decreased to 3.0% and 0.9%. The biological  56 inhibitor, AFP-III, did have some influence on the gas phase but it was not as notable as the commercial KHIs. The concentration of methane in the gas phase reached 94.8% and those of ethane and propane decreased to 3.7% and 1.5%. From these results it is evident that at least in the presence of the chemical inhibitors, and perhaps with AFP-III, there is an impact on gas consumption, with the heavier hydrocarbons (ethane and propane) appearing to preferentially incorporate into hydrates during the early stages.   57 .         58                       Figure 4.1. A PXRD pattern obtained at 125 K for a hydrate sample synthesized from a methane/ethane/propane/water mixture at 9.0 MPa and 274 K.(The structure I peak is indicated by a delta and ice peaks are indicated by an asterisk).  (a) Control experiments with no additives (b) PVP (c) H1W85281 and (d) AFP-III   Since gas phase analysis indicated that the different classes of KHIs could have an impact on guest gas incorporation, PXRD was used to determine the type of hydrate formed.  When hydrates were prepared at 274 K and 9 MPa without any inhibitors, sII hydrates were present (Figure 4.1a) as expected(Uchida et al., 2007). PXRD patterns also showed that although sII hydrates were predominantly formed in the presence of the KHIs, some sI was also detected (Figure 4.1; indicated by the peaks marked with a delta). The sI peak was most obvious in hydrates formed with PVP and H1W85281 (Figure 4.1b, c), and was so small that it was difficult to detect in the AFP-containing samples (Figure 4.1d). This result confirms our previous suspicions based on DSC analysis (Daraboina et al., 2011a) and stirred reactor experiments (Daraboina et al., 2011c) that more complex hydrates are formed in the presence of chemical  59 inhibitors. The PXRD patterns also revealed the presence of ice (Figure 4.1; indicated by peaks marked with an asterisk). Because the ice peak intensities were higher in the presence of all of the tested KHIs compared to controls prepared without inhibitors, it is apparent that the conversion of ice to hydrate was impeded with the additives, which has never been reported previously with PXRD. A slower ice to hydrate conversion is consistent with the inhibition activities of KHIs.  4.5.2. Raman spectroscopy  Once the presence of hydrate was confirmed in the solid phase, the samples were analyzed using Raman and NMR spectroscopy in order to obtain cage occupancies, hydration numbers and localized as well as overall hydrate composition. Raman spectra obtained from hydrates formed without inhibitors clearly showed the C-H region from 2800-3000 cm-1and the O-H region from 3000-3400 cm-1 (Figure 4.2a). Two peaks at ~2904 cm-1 and 2914 cm-1 represented the C-H stretch vibration from methane molecules encaged in large and small cages, respectively. Other small peaks in the C-H region were derived from ethane and propane in the large cages. Peak positions around 2870, 2878 and 2984 cm-1 represented C-H stretching vibrations of propane in large cages, with peak positions at 2884 and 2940 cm-1 corresponding to ethane. The broad peak at 3090 cm-1 represents the vibrational mode of water molecules in the host lattice of the hydrate structure (Sloan, 1998b; Tulk et al., 2000; Kumar et al., 2008b). When spectra were obtained for four different locations in the same sample, no significant differences were observed, demonstrating consistency in the spectra obtained in the water controls (Figure 4.2a).   60     61   Figure 4.2. Raman spectra of methane/ethane/propane hydrate formed with (a) Control experiments with no additives (b) PVP (c) H1W85281 and (d) AFP-III    62  Contrary to the consistent spectra obtained without inhibitors, the presence of chemical KHIs resulted in substantially different profiles even though they were obtained from the same samples (Figure 4.2b and c). For example, the peak at 2904 cm-1 for methane in large cages varied considerably in the profiles for PVP (Figure 4.2b). Although it is difficult to distinguish between peaks for sI and sII in the C-H stretch region because of strongly overlapping peaks (Subramanian et al., 2000a; Subramanian and Sloan, 2002), the ratio of the integrated intensity of the peak for methane in large cages relative to that of small cages could still be used to obtain structural information. Using these calculations it became evident that at least for some positions in the hydrate, methane occupied more large cages compared to water controls and was consistent with the heterogeneity of hydrate composition in the presence of PVP and H1W85281 (Figure 4.2b and c). With PVP samples, peaks corresponding to ethane (2884 and 2940) and propane (2870, 2878 and 2984) were much bigger at locations 2 and 3 (Figure 4.2b). As well, the peak at 2904 (corresponding to methane in large cages) was as large as the peak at 2914 (methane peak corresponding to small cages). Similarly, in the presence of H1W85281 the contribution from ethane and propane was greater in some crystals and also methane was seen to occupy more large cages at some locations. Again, these results strongly indicate that hydrate formed in presence of PVP and H1W85281 was not homogeneous in composition.   In striking contrast to the observations made with PVP and H1W85281, hydrate spectra obtained from samples formed in the presence of AFP-III resembled those obtained from hydrate formed without KHIs. Both water and AFP-III spectra showed within-sample consistency and little evidence of structural heterogeneity (Figure 4.2 a, d). Significantly, evidence for the inhomogeneous nature of hydrate formed in the presence of PVP and H1W85281 has been  63 previously observed in DSC (Daraboina et al., 2011a) and gas uptake experiments (Daraboina et al., 2011b), but not for biological inhibitors . 4.5.3. NMR spectroscopy Hydrate composition revealed by Raman analysis was further characterized by 13C magic angle spinning (MAS) NMR spectra. In hydrates formed in water controls, 5 peaks were observed (Figure 4.3a). All were associated with mixed gas sII hydrates including peaks at -4.36 and -8.25 ppm representing methane in small and large sII cages, the 6.15 ppm peak representing ethane in sII large cages, and peaks at 16.81 (methylene carbons) and 17.52 ppm (methyl carbons) representing propane in the sII large cages (Subramanian, 2000; Kida et al., 2007; Kumar et al., 2008b). These same predominant peaks were identified in the 13C MAS NMR spectrum of hydrate samples synthesized in the presence of KHIs, except that an additional small peak at -6.61 ppm was observed in the PVP and H1W85281 samples (Figure 4.3b, c; see arrow). This peak represented methane in sI cages. Thus, NMR analysis on these samples supports the sI hydrate evidence derived from the gas phase analysis, PXRD patterns and Raman spectroscopy (Table 4.1; Figure 4.1 and Figure 4.2). It must be noted that in contrast, no distinct peak at -6.61 was identified in the NMR spectrum obtained from the AFP-containing hydrates (Figure 4.3d), again consistent with the more homogeneous Raman spectral analysis in such samples.    64        65        Figure 4.3. 13C NMR spectrum of methane/ethane/propane hydrate formed in the presence of (a) Control experiments with no additives (b) PVP (c) H1W85281 and (d) AFP-III  The average hydrate gas composition was also calculated from the 13C NMR data (Table 4.2). Although the propane concentration was approximately the same in all samples, there was less enclathrated methane found in the presence of any of the inhibitors compared to controls, consistent with the gas phase analysis. Interestingly, although enclathrated ethane increased in the presence of KHIs, the increase over hydrates formed without inhibitors was doubled in  66 samples containing the two chemical inhibitors (~30% increase) compared to those with AFP-III (~15% increase).  Table 4.2. Hydrate phase composition of the hydrate synthesized in the presence of water or various inhibitors. Inhibitor  Methane % Ethane% Propane% None 62.8 13.3 23.9 PVP 58.5 17.9 23.6 H1W85281 57.4 17.1 25.5 AFP 60.0 15.2 24.8  In order to fully describe the distinct differences in guest composition, the 13C NMR data coupled with a statistical thermodynamic model (van der Waals and Platteeuw, 1959b; van der Waals and Platteeuw, 1959a) was used to calculate cage occupancies of mixed gas hydrates (Ripmeester and Ratcliffe, 1988; Kida et al., 2007; Kumar et al., 2008b; Kida et al., 2009). These cage occupancies were then used to calculate the average hydration number, defined as the number of host molecules per guest molecule. As seen in the gas composition analysis, propane occupancy of large cages was similar in all experiments (Table 4.3). Small cages were 93-97% occupied by methane, but in the presence of any of the inhibitors, the ratio of small to large cage occupancy by methane changed from 4.1 to 5.2-5.7. This was mainly due to the ~25% reduction of methane in large cages in the presence of any of the inhibitors (Table 4.3), consistent with hydrate gas composition (Table 4.2). Large cages were almost fully occupied (97-99%) by methane, ethane or propane when formed in the presence of water or the chemical KHIs (PVP and H1W85281). However, the overall occupancy of large cages was somewhat lower (92%) in the AFP-containing hydrates, reflecting the smaller increase in enclathrated ethane seen in the gas occupancy and composition analysis (Table 4.2 and Table 4.3). Nevertheless, in the presence of any of the KHIs, substantially more ethane occupied large cages than methane, likely as a  67 result of the preferential enclathration of the heavier hydrocarbon. When this occurs, the methane concentration in the gas phase rises, leading to an increase in the equilibrium pressure and that in turn, reduces the driving force for hydrate formation. Thus, it is to be expected that all of the studied KHIs reduced hydrate growth, which they do (Ohno et al., 2010; Daraboina et al., 2011a; Daraboina et al., 2011c). It should also be noted that there was little change in the average hydration number, but even the small variations were likely due to the observed compositional changes.  Table 4.3. Cage occupancy values obtained with 13C MAS NMR spectra in the presence of inhibitors. Methane Ethane  Propane  Inhibitor Small cage occupancy θS Large cage occupancy θL Large cage occupancy θL Large cage occupancy θL Total  large cage occupancy ΘT Hydration number N None 0.96 0.23 0.26 0.48 0.97 5.88 PVP 0.95 0.18 0.35 0.46 0.99 5.88 H1W85281 0.93 0.17 0.33 0.49 0.99 5.96 AFP 0.97 0.17 0.28 0.47 0.92 5.94  Preferential enclathration of individual gases in mixed gas systems has been previously reported (Uchida et al., 2007; Kumar et al., 2008b). However, here we showed that in the presence of chemical KHIs, the heavier gas, ethane, appeared to encage more readily than when hydrate was formed in the absence of inhibitors (Table 4.2 and Table 4.3). The result of reducing the partial pressure of these components and the consequent increase in methane concentration in the vapor phase (Table 4.1), could lead to the formation of pure methane sI hydrate or methane- ethane sI hydrate, as was detected using a variety of techniques. We suspect that it was methane- 68 ethane sI hydrate that was formed because the ethane composition increased substantially in the chemical KHI experiments, which showed strong support for sI hydrate formation. In the presence of AFP-III, however, significantly less ethane was enclathrated (Table 4.3), and the methane concentration in the vapor phase was only modestly increased. Therefore, there would be a considerable reduction in the driving force for sI hydrate. Indeed, in contrast to the chemical KHI data, we saw no strong evidence for sI hydrate in the presence of the biological inhibitor. The striking decrease in the overall large cage occupancy with the AFP-III samples is curious, and we speculate that protein groups, such as methyl residues could occupy a gas position in the hydrate, anchoring the inhibitor to the crystal and thus reducing the overall number of ethane guests. The adsorption of AFPs to model hydrates has been demonstrated visually (Gordienko et al., 2010). These biological inhibitors then likely inhibit gas hydrate formation by an absorption/inhibition mechanism, as has been previously suggested (Zeng et al., 2003; Zeng et al., 2006a). As well since the fewer larger cages are fully occupied, this may explain the quicker dissociation of hydrates formed in the presence of AFPs than with chemical KHIs (Daraboina et al., 2011c). In support of this argument, it is well known that when the fractional occupancy of cages is increased, hydrates are more stable due to the lower chemical potential of water (Dec et al., 2007). Chemical KHIs may inhibit gas hydrates by increasing the heterogeneity of the crystals, resulting in not only a longer induction time but a longer decomposition time, which we and others have previously observed (Makogon and Holditch, 2001a). Taken together, these experiments unequivocally demonstrate that the chemical and biological KHIs are distinct in their inhibition mechanisms, and thus will likely find practical utility under different conditions.       69  5. CONCLUSIONS AND RECOMMENDATIONS   The performance of chemical (PVP and H1W85281) and biological (antifreeze protein III) inhibitors on multi component (methane/ethane/ propane) hydrate formation and decomposition was tested in a custom small reactor using HP-DSC and newly designed and fabricated autoclave. Gas hydrates made in a batch reactor were monitored with gas chromatography as the synthesis progressed and extensively analyzed with a variety of molecular techniques after formation. Isothermal experiments in DSC showed that both chemical and biological inhibitors significantly delayed the onset of hydrate nucleation, with the new H1W85281 most effective at delaying nucleation as well as decreasing the total amount of hydrate formed. Complex and multiple hydrate melting peaks, were observed in the presence of the chemical inhibitors, PVP and H1W85281, which continued to vary with subsequent reformation cycles. In contrast, AFP addition resulted in a single hydrate peak, which was less stable as indicated by the lower melting temperature and which remained consistent and reproducible through multiple cycles. For field operators, the newer H1W85281 will be more effective at inhibiting hydrate formation, but under conditions where reformation is likely, its unpredictability and the elevated stability of the formed hydrates may make its use more challenging.  A newly-fabricated stirred reactor operated at a constant pressure and temperature allowed a direct comparison of the formation and dissociation of mixed gas hydrates in the presence of commercial and biologically-based kinetic hydrate inhibitors (KHIs). All inhibitors  70 significantly delayed hydrate nucleation and reduced the hydrate growth. A new commercial inhibitor, H1W85281, was the most effective in prolonging the induction time (by a factor of 33.6) and reducing growth (by a factor of 2.2) compared to the other tested inhibitors. These analyses confirm that KHIs perform both as nucleation inhibitors and growth inhibitors, and such properties may reflect different mechanisms of action. A two-stage decomposition of hydrates formed in the presence of the two commercial inhibitors is suggestive of heterogeneous hydrate crystals, consistent with our previous study using calorimetry, as well as the observations of other researchers. In contrast, hydrates formed in the presence of either AFP type decomposed in a manner similar to that observed for no KHI controls, but faster. Although the probability of hydrate formation can be reduced with KHIs, unusual circumstances such as long shut-in periods can nevertheless result in hydrate formation and plugging of pipelines. In these cases, decomposition kinetics in the presence of inhibitors is an important factor for consideration. Use of biological inhibitors not only delays nucleation and inhibits hydrate growth, but when conditions change, hydrates formed in the presence of AFPs show complete decomposition at an earlier time, an advantageous and valuable attribute for any KHI.  As predicted, and demonstrated by PXRD and NMR spectroscopy, sII hydrates predominated, but a peak representing sI hydrate was also seen in the presence of the chemical inhibitors. A sI peak was difficult to detect in AFP-III studies. Raman spectroscopy confirmed that hydrates in the chemical KHI experiments were heterogeneous in contrast to the seemingly homogenous hydrates formed in water controls or AFP-III experiments. As has been previously observed with gas mixtures, the gas phase methane concentration increased during synthesis, but this fractionation effect, confirmed by NMR and Raman spectroscopy, was rather more dominant in the presence of the chemical KHIs. Large hydrate cages formed in the presence of all the  71 inhibitors showed a reduction in methane. With the commercial inhibitors, these large cage methane guests appeared to be substituted by ethane, resulting in a decreased driving force for hydrate production. Concomitant with this change, we speculated that the formed sI hydrate was likely methane-ethane sI. In contrast to the near full occupancy of total (methane+ethane+propane) large cages in chemical kinetic inhibitor experiments, almost 7% of the total large cages were not filled when hydrates were formed in the presence of AFP-III, possibly supporting an adsorption-inhibition mechanism. This is the first time mechanism for kinetic inhibition of natural gas hydrates by two very different classes of KHIs of natural gas hydrates have been investigated.   Based on the insights gained from the current work, the following future recommendations are proposed. 1. Molecular scale experiments showed the structural and compositional heterogeneity when hydrates formed in the presence of inhibitors. Further work is required to investigate the time evolution of this behavior using PXRD and Raman spectroscopy. The possible outcome is to observe the time evolution of structural and compositional changes in the presence of inhibitors, which can reveal how these molecules effect the hydrate formation or decomposition with time. In addition, the differences in the action between chemical and biological inhibitors on hydrate formation can be observed with respect to time.  2.    Use of biological inhibitors not only delays nucleation and inhibits hydrate growth, but enhances complete decomposition at an earlier time, an advantageous and valuable attribute for any KHI. However, some chemical inhibitors have better inhibition strength than biological inhibitors. It will be interesting to see how the combination of chemical  72 and biological inhibitors can improve the efficacy in inhibiting strength and decomposition kinetics. 3. Visual observations of hydrate formation in the presence of these inhibitors can provide useful information such as how the hydrate crystal morphology changes in the presence of additives.  By doing so, we can try to identify the differences in inhibition mechanisms of these chemical and biological inhibitors by observing the shape of hydrate crystals through morphology changes (i.e. are chemical and biological inhibitors working in similar fashion?).  4.  In-situ magnetic resonance imaging will allow us to observe the hydrate formation in the presence of the inhibitors simultaneously and provide localized kinetic information in small droplets. This should provide insights into how these localized kinetics are different from average kinetics obtained in stirred reactor in the presence of inhibitor molecules. 5.  Since many biological inhibitors are available in limited quantities, molecular dynamic studies can be helpful to understand the action of biological inhibitors on hydrate formation and decomposition. It will be useful to do simulations with different possible side chain groups (like threonine), to determine specific molecular sources of hydrate inhibition, which is difficult to do using experimental procedures and will allow us to investigate how these groups are affecting the inhibition mechanism (e.g. adsorption).            73   REFERENCES  Al-Adel, S., Dick, J. A. G., El-Ghafari, R., and Servio, P., 2008. The effect of biological and polymeric inhibitors on methane gas hydrate growth kinetics. Fluid Phase Equilibria 267(1), 92-98. Anderson, B. J., Tester, J. W., Borghi, G. P., and Trout, B. L., 2005. Properties of Inhibitors of Methane Hydrate Formation via Molecular Dynamics Simulations. J. AM. CHEM. SOC. 127, 17852-17862. Ballard, A. L., and Sloan, E. D., 2000. Optimizing thermodynamic parameters to match methane and ethane structural transition in natural gas hydrate equilibria. Gas Hydrates: Challenges for the Future 912, 702-712. Bruusgaard, H., Lessard, L. D., and Servio, P., 2009. Morphology Study of Structure I Methane Hydrate Formation and Decomposition of Water Droplets in the Presence of Biological and Polymeric Kinetic Inhibitors. Crystal Growth & Design 9(7), 3014-3023. Carstensen, A., Zugik, M., Creek, J., and Koh, C. A., 2002. Clathrate hydrate formation and inhibition. Internet Journal of Vibrational Spectroscopy [online computer file] 6(1), No pp. given. Cohen, J. M., Wolf, P. F., and Young, W. D., 1998. Enhanced hydrate inhibitors: Powerful synergism with glycol ethers. Energy & Fuels 12(2), 216-218. Dalmazzone, D., Clausse, D., Dalmazzone, C., and Herzhaft, B., 2004. The stability of methane hydrates in highly concentrated electrolyte solutions by differential scanning calorimetry and theoretical computation. American Mineralogist 89(8-9), 1183-1191.  74 Dalmazzone, D., Dalmazzone, C., and Herzhaft, B., 2002a. Differential scanning calorimetry: A new technique to characterize hydrate formation in drilling muds. Spe Journal 7(2), 196- 202. Dalmazzone, D., Kharrat, M., Lachet, V., Fouconnier, B., and Clausse, D., 2002b. DSC and PVT measurements - Methane and trichlorofluoromethane hydrate dissociation equilibria. Journal of Thermal Analysis and Calorimetry 70(2), 493-505. Daraboina, N., Ripmeester, J. A., Walker, V. K., and Englezos, P., 2011a. Natural Gas Hydrate formation and decomposition in the presence of kinetic inhibitors.  I: High Pressure Calorimetry  Energy & Fuels Submitted. Daraboina, N., Ripmeester, J. A., Walker, V. K., and Englezos, P., 2011b. Natural Gas Hydrate formation and decomposition in the presence of kinetic inhibitors.  I: High Pressure Calorimetry  Energy & Fuels Submitted(ef-2011-00812m). Daraboina, N., Ripmeester, J. A., Walker, V. K., and Englezos, P., 2011c. Natural Gas Hydrate formation and decomposition in the presence of kinetic inhibitors.  II: Stirred Reactor Experiments. Energy & Fuels Submitted. Daraboina, N., Ripmeester, J. A., Walker, V. K., and Englezos, P., 2011d. Natural Gas Hydrate formation and decomposition in the presence of kinetic inhibitors.  III:structural and compositional changes  Energy & Fuels Submitted(ef-2011-00814z). Daraboina, N., Ripmeester, J. A., Walker, V. K., and Englezos, P., 2011e. Natural Gas Hydrate formation and decomposition in the presence of kinetic inhibitors. II Stirred reactor experiments. Energy & Fuels Submitted(ef-2011-00813). Davidson, D. W., 1973a. Gas Hydrates, Plenum Press, New York.  75 Davidson, D. W., 1973b. Gas Hydrates. In Water: A Comprehensive Treatise, Plenum Press;, New York. Davidson, H. R. R. T. D. L. C., 1986. Crystallographic Studies of Clathrate Hydrates. Part I. Mol. Cryst. Liq. Cryst. 141, 141-149. Davies, P. L., Baardsnes, J., Kuiper, M. J., and Walker, V. K., 2002. Structure and function of antifreeze proteins. Philosophical Transactions of the Royal Society of London Series B- Biological Sciences 357(1423), 927-933. Davy, H., 1811. The Bakerian Lecture. On Some of the Combinations of Oxymuriatic Gasand Oxygene, and on the Chemical Relations of these Principles to Infl. Phil Trans Roy Soc (London) 101, 1-35. Dec, S. F., Bowler, K. E., Stadterman, L. L., Koh, C. A., and Sloan, E. D., 2007. NMR study of methane plus ethane structure I hydrate decomposition. Journal of Physical Chemistry A 111(20), 4297-4303. Englezos, P., 1993. Clathrate Hydrates. I&EC Research 32, 1251-1274. Englezos, P., 1996. Nucleation and Growth of Gas Hydrate Crystals in Relation to "Kinetic Inhibition". Rev. IFP 51`(6), 789. Englezos, P., Kalogerakis, N., Dholabhai, P. D., and Bishnoi, P. R., 1987a. Kinetics of Formation of Methane and Ethane Gas Hydrates. Chem. Eng. Sci. 42(11), 2647. Englezos, P., Kalogerakis, N., Dholabhai, P. D., and Bishnoi, P. R., 1987b. Kinetics of Formation of Methane and Ethane Gas Hydrates. Chemical Engineering Science 42(11), 2647-2658.  76 Englezos, P., Kalogerakis, N., Dholabhai, P. D., and Bishnoi, P. R., 1987c. Kinetics of Gas Hydrate Formation from Mixtures of Methane and Ethane. Chemical Engineering Science 42(11), 2659-2666. Englezos, P., Kalogerakis, N., Dholabhai, P. D., and Bishnoi, P. R., 1987d. Kinetics of Gas Hydrate Formation from Mixtures of Methane and Ethane. Chem. Eng. Sci. 42(11), 2659. Freer, E., and Sloan, E. D., 2000. An engineering approach to kinetic inhibitor design using moleculardynamics simulations. Gas hydrates, challenges for the future NYAS 912, 651. Gordienko, R., Ohno, H., Singh, V. K., Jia, Z. C., Ripmeester, J. A., and Walker, V. K., 2010. Towards a Green Hydrate Inhibitor: Imaging Antifreeze Proteins on Clathrates. Plos One 5(2), -. Grieken, R. E. V., and Markowicz, A. A., 2002. Handbook of X-ray spectrometry, Marcel Dekker, New York, Basel. Hammerschmidt, E. G., 1934. Formation of Gas Hydrates in Natural Gas Transmission Lines. Industrial and Engineering Chemistry 26, No. 8, 851-855. Herri, J. M., Pic, J. S., Gruy, F., and Cournil, M., 1999. Methane hydrate crystallization mechanism from in-situ particle sizing. AIChE Journal 45(3), 590-602. Hughes, T. J., 2008. Plug Formation and dissociation of Mixed Gas Hydrates and Mehane Semi- Clathrate Hydrate stability, University of Canterbury, Christchurch. Jager, M. D., and Sloan, E. D., 2002. Structural Transition of Clathrate Hydrates Formed from a Natural Gas. Fourth International conference on gas hydrates. Jensen, L., Thomsen, K., and von Solms, N., 2011. Inhibition of Structure I and II Gas Hydrates using Synthetic and Biological Kinetic Inhibitors. . Energy & Fuels 25, 17-23.  77 Kawamura, T., Sakamoto, Y., Ohtake, M., Yamamoto, Y., Komai, T., Haneda, H., Yoon, J. H., and Ohga, K., 2006. Dissociation behavior of pellet shaped mixed gas hydrate samples that contain propane as a guest. Energy Conversion and Management 47(15-16), 2491- 2498. Kelland, M. A., 2006. History of the development of low dosage hydrate inhibitors. Energy & Fuels 20(3), 825-847. Kida, M., Hachikubo, A., Sakagami, H., Minami, H., Krylov, A., Yamashita, S., Takahashi, N., Shoji, H., Khlystov, O., Poort, J., and Narita, H., 2009. Natural gas hydrates with locally different cage occupancies and hydration numbers in Lake Baikal. Geochemistry Geophysics Geosystems 10, -. Kida, M., Sakagami, H., Takahashi, N., Hachikubo, A., Shoji, H., Kamata, Y., Ebinuma, T., Narita, H., and Takeya, S., 2007. Estimation of gas composition and cage occupancies in CH4-C2H6 hydrates by CP-MAS C-13 NMR technique. Journal of the Japan Petroleum Institute 50(3), 132-138. King, H. E., Jr., Hutter, J. L., Lin, M. Y., and Sun, T., 2000. Polymer conformations of gas- hydrate kinetic inhibitors: A small-angle neutron scattering study. Journal of Chemical Physics 112(5), 2523-2532. Klomp, U., 2008. The world of LDHI: From conception to development to implementation. 6th international conference on gas hydrates, vancouver, British columbia. Koh, C. A., 2002. Towards a fundamental understanding of natural gas hydrates. Chemical Society Reviews 31(3), 157-167.  78 Koh, C. A., Gupta, A., Lachance, J., and Sloan, E. D., 2008a. Measurements of methane hydrate heat of dissociation using high pressure differential scanning calorimetry. Chemical Engineering Science 63(24), 5848-5853. Koh, C. A., Lachance, J. W., and Sloan, E. D., 2008b. Effect of hydrate formation/dissociation on emulsion stability using DSC and visual techniques. Chemical Engineering Science 63(15), 3942-3947. Koh, C. A., Lachance, J. W., and Sloan, E. D., 2009. Determining gas hydrate kinetic inhibitor effectiveness using emulsions. Chemical Engineering Science 64(1), 180-184. Koh, C. A., Savidge, J. L., and Tang, C. C., 1996. Time-Resolved inSitu Experiments on the Crystallization of Natural Gas Hydrates. J. Phys. Chem. 100, 6412-6414. Koh, C. A., Westacott, R. E., Zhang, W., Hirachand, K., Creek, J. L., and Soper, A. K., 2002. Mechanisms of gas hydrate formation and inhibition. Fluid Phase Equilibria 194, 143- 151. Kumar, R., Lee, J. D., Song, M., and Englezos, P., 2008a. Kinetic inhibitor effects on methane/propane clathrate hydrate-crystal growth at the gas/water and water/n-heptane interfaces. Journal of Crystal Growth 310(6), 1154-1166. Kumar, R., Linga, P., Moudrakovski, I., Ripmeester, J. A., and Englezos, P., 2008b. Structure and kinetics of gas hydrates from methane/ethane/propane mixtures relevant to the design of natural gas hydrate storage and transport facilities. Aiche Journal 54(8), 2132-2144. Kvamme, B., Huseby, G., and Forrisdahl, O. K., 1997. Molecular dynamics simulations of PVP kinetic inhibitor in liquid water and hydrate/liquid water systems. Molecular Physics 90(6), 979-991.  79 Kvamme, B., Kuznetsova, T., and Aasoldsen, K., 2005. Molecular dynamics simulations for selection of kinetic hydrate inhibitors. Journal of Molecular Graphics & Modelling 23(6), 524-536. Lachance, J. W., Sloan, E. D., and Koh, C. A., 2009. Determining gas hydrate kinetic inhibitor effectiveness using emulsions. Chemical Engineering Science 64(1), 180-184. Le Parlouer, P., Dalmazzone, C., Herzhaft, B., Rousseau, L., and Mathonat, C., 2004. Characterisation of gas hydrates formation using a new high pressure MICRO-DSC. Journal of Thermal Analysis and Calorimetry 78(1), 165-172. Lee, J. D., Song, M., Susilo, R., and Englezos, P., 2006. Dynamics of methane-propane clathrate hydrate crystal growth from liquid water with or without the presence of n-heptane. Crystal Growth & Design 6(6), 1428-1439. Linga, P., Kumar, R., and Englezos, P., 2006. Capture of Carbon dioxide from conventional power plants or from integrated gasification plants through gas hydrate formation/ dissociation. Journal of Energy & Climate Change 1(2), 75-82. Linga, P., Kumar, R., and Englezos, P., 2007. Gas hydrate formation from hydrogen/carbon dioxide and nitrogen/carbon dioxide gas mixtures. Chemical Engineering Science 62(16), 4268-4276. Long, J. P., Lederhos, J. P., Sum, A., Christiansen, R. L., and Sloan, E. D., 1994. Kinetic Inhibitors of Natural Gas Hydrates. Proc 73rd Gas Processors Association Annl Conv. New Orleans 3/7-9/94, 85-93. Lu, H. L., Seo, Y. T., Lee, J. W., Moudrakovski, I., Ripmeester, J. A., Chapman, N. R., Coffin, R. B., Gardner, G., and Pohlman, J., 2007. Complex gas hydrate from the Cascadia margin. Nature 445(7125), 303-306.  80 Makogon, Y. F., 1981. Hydrates of Natural Gas. Pennwell books, Tulsa, Okla, 176. Makogon, Y. F., and Holditch, S. A., 2001a. Gas hydrates - conclusion: experiments illustrate hydrate: morphology, kinetics. Oil & Gas Journal 99(7), 45-50. Makogon, Y. F., and Holditch, S. A., 2001b. Lab work clarifies gas hydrate formation, dissociation. Oil & Gas Journal 99(6), 47-48,50-52. McCallum, S. D., Riestenberg, D. E., Zatsepina, O. Y., and Phelps, T. J., 2007. Effect of pressure vessel size on the formation of gas hydrates. Journal of Petroleum Science and Engineering 56(1-3), 54-64. Mccullough, J. P., and Waddington, G., 1957. Melting-Point Purity Determinations - Limitations as Evidenced by Calorimetric Studies in the Melting Region. Analytica Chimica Acta 17(1), 80-96. Mehta, A. P., and Sloan, E. D., 1993a. Structure-H Hydrate Phase-Equilibria of Methane Plus Liquid-Hydrocarbon Mixtures. Journal of Chemical and Engineering Data 38(4), 580- 582. Mehta, A. P., and Sloan, E. D., 1993b. Structure H Hydrate Phase Equilibria of Methane+Liquid Hydrocarbon Mixtures. J. Chem. Eng. Data 38, 580-582. Miller, S. L., and Smythe, W. D., 1970. Carbon Dioxide Clathrate in the Martian Ice Cap. Science 170, 591-594. Moon, C., Taylor, P. C., and Rodger, P. M., 2002. Computer Modelling of Gas Hydrate Formation. Proc. 4th International Conference on Gas Hydrates, Yokohama May 19-23, 2002, 665-668. Moon, C., Taylor, P. C., and Rodger, P. M., 2003. Molecular dynamics study of gas hydrate formation. Journal of the American Chemical Society 125(16), 4706-4707.  81 Morita, K., Nakano, S., and Ohgaki, K., 2000. Structure and stability of ethane hydrate crystal. Fluid Phase Equilibria 169(2), 167-175. Moudrakovski, I. L., Ratcliffe, C. I., and Ripmeester, J. A., 2001. Probing transient hydrate structure with hyperpolarized 129Xe NMR spectroscopy: a metastable structure II hydrate of Xe. Angewandte Chemie, International Edition 40(20), 3890-3892. Nakagawa, R., Akihiro, H., and Shoji, H., 2008. Dissociation and specific heats of gas hydrates under submarine and sublacustrine environments Proceedings of 6th International Conference on Gas Hydrates. Nihous, G. C., Kuroda, K., Lobos-Gonzalez, J. R., Kurasaki, R. J., and Masutani, S. M., 2010. An analysis of gas hydrate dissociation in the presence of thermodynamic inhibitors. Chemical Engineering Science 65(5), 1748-1761. Ohno, H., Strobel, T. A., Dec, S. F., Sloan, E. D., and Koh, C. A., 2009. Raman Studies of Methane-Ethane Hydrate Metastability. Journal of Physical Chemistry A 113(9), 1711- 1716. Ohno, H., Susilo, R., Gordienko, R., Ripmeester, J., and Walker, V. K., 2010. Interaction of Antifreeze Proteins with Hydrocarbon Hydrates. Chemistry-a European Journal 16(34), 10409-10417. Ribeiro, C. P., and Lage, P. L. C., 2008. Modelling of hydrate formation kinetics: State-of-the-art and future directions. Chemical Engineering Science 63(8), 2007-2034. Ripmeester, J., and Ratcliffe, C., 1999. On the contributions of NMR spectroscopy to clathrate science. Journal of Structural Chemistry 40, 654. Ripmeester, J. A., 2000. Hydrate research - From correlations to a knowledge-based discipline - The importance of structure. Gas Hydrates: Challenges for the Future, 1-16.  82 Ripmeester, J. A., and Ratcliffe, C. I., 1988. Low-Temperature Cross polarization /mAGIC aNGLE sPINNING 13c nmr OF Solid CH4 Hydrates; Structure, Cage Occupancy, and Hydrate Numnber. J. Phys. Chem. 92(2), 337. Ripmeester, J. A., Susilo, R., and Englezos, P., 2007. Characterization of gas hydrates with PXRD, DSC, NMR, and Raman spectroscopy. Chemical Engineering Science 62(15), 3930-3939. Ripmeester, J. A., Tse, J. S., Ratcliffe, C. I., and Powell, B. M., 1987. A New Clathrate Hydrate Stucture. Nature,  325, 135-136. Rueff, R. M., Sloan, E. D., and Yesavage, V. F., 1988. Heat Capacity and Heat of Dissociation of Methane Hydrates. AIChE Journal,  34(9), 1468. Rydzy, M. B., Schicks, J. M., Naumann, R., and Erzinger, J., 2007. Dissociation enthalpies of synthesized multicomponent gas hydrates with respect to the guest composition and cage occupancy. Journal of Physical Chemistry B 111(32), 9539-9545. Schicks, J. M., Erzinger, J., and Ziemann, M. A., 2005. Raman spectra of gas hydrates - differences and analogies to ice 1h and (gas saturated) water. Spectrochimica Acta Part a- Molecular and Biomolecular Spectroscopy 61(10), 2399-2403. Schicks, J. M., Naumann, R., Erzinger, J., Hester, K. C., Koh, C. A., and Sloan, E. D., 2006. Phase transitions in mixed gas hydrates: Experimental observations versus calculated data. Journal of Physical Chemistry B 110(23), 11468-11474. Seo, Y., Kang, S. P., and Jang, W., 2009. Structure and Composition Analysis of Natural Gas Hydrates: C-13 NMR Spectroscopic and Gas Uptake Measurements of Mixed Gas Hydrates. Journal of Physical Chemistry A 113(35), 9641-9649. Skoog. 1982. Raman Spectroscopy. Instrumental Analysis, 362.  83 Skovborg, P., and Rasmussen, P., 1994. A Mass Transport Limited Model for the Growth of Methane and Ethane Gas Hydrates. Chem. Eng. Sci. 49(8), 1131-1143. Sloan, E. D., 1998a. Clathrate Hydrates of Natural Gases. Marcel-Dekker, Inc, New York, 664. Sloan, E. D., 2003. Fundamental principles and applications of natural gas hydrates. Nature 426(6964), 353-359. Sloan, E. D., and Fleyfel, f., 1991. A Molecular Mechanism for Gas Hydrate Nucleation from Ice. AIChE Journal 37(9), 1281. Sloan, E. D., Jr., 1998b. Clathrate Hydrates of Natural Gases, Second Edition, Revised and Expanded, Marcel Dekker, NY. Sloan, E. D., and Koh, C. A., 2008. Clathrate Hydrates of Natural Gases, CRC press, Taylor & Francis Group, New york. Sloan, E. D., Subramanian, S., Matthews, P., Lederhos, J., and Khokhar, A., 1998. Quantifying hydrate formation and kinetic inhibition. Industrial and Engineering Chemistry Research 37, 3124. Storr, M. T., and Rodger, P. M., 2000. A molecular dynamics study of the mechanism of kinetic inhibition. Annals of the New York Academy of Sciences 912(Gas Hydrates), 669-677. Subramanian, S., 2000. Measurements of clathrate hydrates containing methane and ethane using Raman spectroscopy. Subramanian, S., Kini, R. A., Dec, S. F., and Sloan, E. D., 2000a. Evidence of structure II hydrate formation from methane plus ethane mixtures. Chemical Engineering Science 55(11), 1981-1999.  84 Subramanian, S., Kini, R. A., Dec, S. F., and Sloan, E. D., 2000b. Structural transition studies in methane plus ethane hydrates using Raman and NMR. Gas Hydrates: Challenges for the Future 912, 873-886. Subramanian, S., and Sloan, E. D., 2002. Trends in vibrational frequencies of guests trapped in clathrate hydrate cages. Journal of Physical Chemistry B 106(17), 4348-4355. Sum, A. K., Burruss, R. C., and Sloan, E. D., 1997. Measurement of clathrate hydrates via Raman spectroscopy. Journal of Physical Chemistry B 101(38), 7371-7377. Susilo, R., Lee, J. D., and Englezos, P., 2005. Liquid-liquid equilibrium data of water with neohexane, methylcyclohexane, tert-butyl methyl ether, n-heptane and vapor-liquid- liquid equilibrium with methane. Fluid Phase Equilibria 231(1), 20-26. Takeya, S., Kamata, Y., Uchida, T., Nagao, J., Oyama, H., Shimada, W., Ebinuma, T., and Narita, H., 2002. Crystal Structure of CH4+CO2 Mixed Gas Hydrate. Proc. 4th International Conference on Gas Hydrates, Yokohama, May 19-23,2002, 586-589. Tulk, C. A., Ripmeester, J. A., and Klug, D. D., 2000. The application of Raman spectroscopy to the study of gas hydrates. Gas Hydrates: Challenges for the Future 912, 859-872. Uchida, T., Moriwaki, M., Takeya, S., Ikeda, I. Y., Ohmura, R., Nagao, J., Minagawa, H., Ebinuma, T., Narita, H., Gohara, K., and Mae, S., 2004. Two-step formation of methane- propane mixed gas hydrates in a batch-type reactor. Aiche Journal 50(2), 518-523. Uchida, T., Takeya, S., Kamata, Y., Ikeda, I. Y., Nagao, J., Ebinuma, T., Narita, H., Zatsepina, O., and Buffett, B. A., 2002. Spectroscopic observations and thermodynamic calculations on clathrate hydrates of mixed gas containing methane and ethane: Determination of structure, composition and cage occupancy. Journal of Physical Chemistry B 106(48), 12426-12431.  85 Uchida, T., Takeya, S., Kamata, Y., Ohmura, R., and Narita, H., 2007. Spectroscopic measurements on binary, ternary, and quaternary mixed-gas molecules in clathrate structures. Industrial & Engineering Chemistry Research 46(14), 5080-5087. Udachin, K. A., Ratcliffe, C. I., and Ripmeester, J. A., 2001a. A dense and efficient clathrate hydrate structure with unusual cages. Angewandte Chemie-International Edition 40(7), 1303-+. Udachin, K. A., Ratcliffe, C. I., and Ripmeester, J. A., 2001b. Structure, composition, and thermal expansion of CO2 hydrate from single crystal X-ray diffraction measurements. Journal of Physical Chemistry B 105(19), 4200-4204. van der Waals, J. H., and Platteeuw, J. C., 1959a. Clathrate Solutions. Adv. Chem. Phys., InterScience, New York II, 1-58. van der Waals, J. H., and Platteeuw, J. C., 1959b. Validity of Clapeyron's Equation for Phase Equilibria involving Clathrates. Nature 183, 4659, 462. Vysniauskas, A., and Bishnoi, P. R., 1983a. A Kinetic-Study of Methane Hydrate Formation. Chemical Engineering Science 38(7), 1061-1072. Vysniauskas, A., and Bishnoi, P. R., 1983b. A Kinetic Study of Methane Hydrate Formation. Chem. Eng. Sci 38(7), 1061-1072. Walker, V. K., Zeng, H., Gordienko, R., Kuiper, M. J., Huva, E. I., and Ripmeester, J. A., 2008. The mysteries of memory effect and its elimination with antifreeze proteins. Proceedings of 6th International Conference on Gas Hydrates. Wathen, B., Kwan, P., Jia, Z. C., and Walker, V. K., 2010. Modeling the Interactions between Poly(N-Vinylpyrrolidone) and Gas Hydrates: Factors Involved in Suppressing and  86 Accelerating Hydrate Growth. High Performance Computing Systems and Applications 5976, 117-133, 418. Yeh, Y., and Feeney, R. E., 1996. Antifreeze proteins: Structures and mechanisms of function. Chemical Reviews 96(2), 601-617. Zeng, H., Moudrakovski, I. L., Ripmeester, J. A., and Walker, V. K., 2006a. Effect of antifreeze protein on nucleation, growth and memory of gas hydrates. Aiche Journal 52(9), 3304- 3309. Zeng, H., Wilson, L. D., Walker, V. K., and Ripmeester, J. A., 2003. The inhibition of tetrahydrofuran clathrate-hydrate formation with antifreeze protein. Canadian Journal of Physics 81(1-2), 17-24. Zeng, H., Wilson, L. D., Walker, V. K., and Ripmeester, J. A., 2006b. Effect of antifreeze proteins on the nucleaton, growth, and the memory effect during tetrahydrofuran clathrate hydrate formation. Journal of the American Chemical Society 128(9), 2844-2850.    87  APPENDIX A - REACTOR DESIGN  Since many biological inhibitors are available in limited quantities, we have fabricated a small-scale apparatus (crystallizer volume of 58 cm3), based on industry-favored reactors. The design of the reactor (Front view: Figure A.1, Top view: Figure A.2, Cross section A: Figure A.3, Cross section B: Figure A.4) and water bath (Front view: Figure A.5, Side view: Figure A.6) was shown below. Two Rosemount smart pressure transmitters (model 3051, Norpac controls, Vancouver, BC) with a maximum uncertainty of 0.075% of span 0-15,000 kPa (i.e. 11 kPa) were used to monitor the pressure in the crystallizer and reservoir (Figure A.7). The data acquisition system (National Instruments) was coupled with a computer to record the data as well as to communicate with the control valve was shown in Figure A.8.  The dimensions of crystallizer, reservoir, liquid contents, gas phase and volume of tubing are given in Table A.1. The crystallizer and reservoir are manufactured with stainless steel and water bath with GE-07 clear Lexan.   88  Figure A.1. Front view of the reactor  89   Figure A.2. Top view of the reactor.  90   Figure A.3. Cross section A of the reactor   91   Figure A.4. Cross section B of the reactor.         92   Figure A.5. Front view and top view of the waterbath.   93   Figure A.6. Side view of the water bath.    94 Table A.1. Dimensions of the reactor and supply vessel   Liquid contents in CR   Diameter 3.00 cm Height 1.41 cm Volume 10.00 cm3    CR dimensions   Diameter 3.00 cm Height 7.07 cm Volume of CR 50.00 cm3    Volume of the gas phase in CR 40.00 cm3 Volume of SV 150.00 cm3    Tubing connections to CR   1/8" Tube length 257.50 cm 1/4" tube length 24.50 cm Inner diameter (1/4" tube) 0.40 cm Inner diameter (1/8" tube) 0.16 cm Volume of tubing for CR (1/8") 5.09 cm3 Volume of tubing for CR (1/4") 3.08 cm3 Total volume of tubing added to CR 8.17 cm3 Total volume of the Crystallizer + Tubing 58.17 cm3 Total volume of the gas phase of CR+Tubing 48.17 cm3    Tubing connections to SV   1/8" Tube length 440.00 cm Inner diameter (1/4" tube) 0.40 cm Inner diameter (1/8" tube) 0.16 cm Volume of tubing for CR (1/8") 8.70 cm3 Total volume of tubing added to CR 8.70 cm3 Total volume of the Supply vessel + Tubing 158.70 cm3              95       Figure A.7. Pressure transmitter setup.    Pressure transmitter for supply vessel (SV) Differential pressure transmitter for Crystallizer Automated control valve & actuator to regulate flow of gas Water bath – unique design to perform experiments at constant T and Variable T . Eliminates the need for cooling coils and external cooling Stirrer arrangement to mix the liquid contents Valve arrangement to allow flow of gas in and out  96  Figure A.8. Experimental apparatus (Back view showing NI DAQ arrangement.   NI hardware: to record pressure and temperature data and also to regulate flow of gas through PID controller  97  APPENDIX B - RAMAN SPECTROSCOPY   The C-C stretch mode region for ethane and propane in the hydrate cages is shown in Figure. The Raman active stretching mode of propane was observed at four positions 878, 1054 and 1157 cm-1. The peak position at 992 cm-1 represents the C-C stretch mode for ethane. These peak positions are match well with the literature values (Subramanian, 2000; Uchida et al., 2007; Kumar et al., 2008b). When spectra were obtained for four different locations in the same sample, no significant differences were observed, demonstrating consistency in the spectra obtained in the water controls. Contrary to the consistent spectra obtained in control, the presence of synthetic inhibitors resulted in substantially different profiles even though they were obtained from different locations in the same samples. The peak at 878 cm-1 representing propane and the peak at 992 cm-1 for ethane varied considerably in the profiles of PVP and H1W85281. The variation in ethane peak (increasing intensity) was considerably high in the presence of H1W85281. These results suggest that the contribution from ethane and propane was greater in some crystals formed in the presence of chemical inhibitors. In contrast to chemical inhibitors, the spectra obtained in the presence of AFP-III have small evidence of variation. Again, these results strongly indicate that hydrate formed in presence of PVP and H1W85281 was not homogeneous in composition and in good agreement with Raman C-H mode profiles and NMR results in chapter 4.      98      99    FigureB.1. C-C stretch mode of Raman spectra of methane/ethane/propane hydrate formed with water and inhibitors          

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.24.1-0059278/manifest

Comment

Related Items