Reliable and Repeatable Evaluation of Kinetic Hydrate Inhibitors

Nov 1, 2016 - These regions range from complete inhibition (CIR), to slow growth region (SGR), and ultimately to rapid growth region (RGR) as subcooli...
2 downloads 6 Views 1MB Size
Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Reliable and Repeatable Evaluation of Kinetic Hydrate Inhibitors Using a Method Based on Crystal Growth Inhibition Houra Mozaffar, Ross Anderson, and Bahman Tohidi Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.6b00382 • Publication Date (Web): 01 Nov 2016 Downloaded from http://pubs.acs.org on November 4, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Reliable and Repeatable Evaluation of Kinetic Hydrate Inhibitors Using a Method Based on Crystal Growth Inhibition Houra Mozaffar1,*, Ross Anderson1,2 and Bahman Tohidi1,2 1. Hydrafact Ltd., Heriot-Watt University Research Park, Edinburgh, EH14 4AP, UNITED KINGDOM 2. Centre for Gas Hydrate Research, Institute of Petroleum Engineering, Heriot-Watt University, Edinburgh, EH14 4AS, UNITED KINGDOM KEYWORDS: gas hydrates, KHI, methane, natural gas, experimental data, crystal growth, inhibition ABSTRACT: Low dosage hydrate inhibitors (LDHIs) - anti-agglomerants (AAs) and kinetic hydrate inhibitors (KHIs) - are increasingly used as an alternative to the traditional thermodynamic hydrate inhibitors for controlling gas hydrate problems in the Petroleum industry. As nucleation inhibitors, KHIs induce an extended ‘induction time (ti)’ at particular sub-coolings before nucleation of hydrate can proceed to growth. However, due to the stochastic nature of nucleation, gaining repeatable and transferrable induction time results can be very challenging and time-consuming. To overcome this problem, study on KHI crystal growth

ACS Paragon Plus Environment

1

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

inhibition (CGI) has yielded a KHI evaluation technique which is considerably simpler than induction time measurements. Moreover, this study shows that KHIs create a number of welldefined hydrate CGI regions as a function of sub-cooling which are completely repeatable and transferrable between different set-ups. These regions range from complete inhibition (CIR), to slow growth region (SGR), and ultimately to rapid growth region (RGR) as subcooling increases, making KHI assessment more rapid and reliable. Furthermore, results revealed that a true induction time can only be measured for a relatively short range of subcooling within the SGR and RGR. Likewise, hydrate nucleation induction time is impossible to measure in the CIR, supporting CGI studies which show that hydrate growth is completely/indefinitely inhibited in this subcooling range.

INTRODUCTION

As a result of offering significant CAPEX/OPEX advantages, low dosage Kinetic Hydrate Inhibitors (KHIs) which are typically low molecular weight polymers that are dissolved in a carrier solvent have seen increasing popularity and use as an alternative chemical to the traditional thermodynamic hydrate inhibitors (e.g. methanol, glycols) for the control of gas hydrates in the oil and gas industry [1-3]. It is generally understood that nuclei surface adsorption is the primary method through which kinetic hydrate inhibitors (e.g. PVCap, poly-nvinylcaprolactam) delay/slow down/interfere with the process of hydrate nucleation [4-7]. This effect will in turn lead to an increase in ‘induction time’, ti; the period between the moment the system enters the hydrate stability zone (HSZ) and the moment critical nuclei are achieved and hydrate formation and growth takes place. In theory, if the KHI-induced induction time, ti, at

ACS Paragon Plus Environment

2

Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ΔT, P is greater than the pipeline fluid residence time at that condition, then the KHI should be able to prevent hydrate nucleation / growth, whereby avoiding plugging. KHI ability to delay and interfere with the nucleation of hydrate crystals has resulted in induction/hold time studies to remain the main technique employed for developing and testing of most KHI and formulations [1-3]. Laboratory induction time studies is also used for evaluating the effect of various parameters such as pressure, presence of synergists, salts, liquid hydrocarbons (condensate, oil) and other oilfield chemicals (e.g. corrosion and scale inhibitors) on the performance of KHIs. However, since induction time in based on nucleation which is a very sensitive and probabilistic property by nature, results from this method are inevitably quite stochastic, poorly repeatable and commonly non-transferable [1-3,8-10]. Moreover, focusing on the nucleation inhibition property of KHIs could result in falsely concluding that KHI has failed inhibition as soon as a hydrate crystal has appeared in the system. Due to the mentioned issues, although KHIs are currently being used successfully in many fields, operators have yet not gained sufficient confidence in using these inhibitors [1,2]. A new technique has been developed and used by Total to resolve the non-repeatability of results in a semi-industrial hydrate flow loop called the “second termination” (SG) method [8]. In this method hydrate is initially formed at a high subcooling. System is then heated up to dissociate most of the hydrate while some ‘nuclei’ of hydrates or ‘hydrate history’ is preserved. Once again system is cooled down to the temperature of interest inside the hydrate phase boundary for ti measurements or for measuring the subcooling at which hydrate starts to form at a constant cooling rate. Presence of hydrate nuclei or history helps nucleation/growth patterns become a lot more consistent when re-cooled to form hydrate. In 2006, the developed SG technique was

ACS Paragon Plus Environment

3

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

applied to autoclave cells to eliminate the stochasticity of kinetic hydrate inhibitor evaluation results [9]. This method has resulted in significant improvement of repeatability and transferability of results [10,11]. As reported in Anderson et al. (2011) a new crystal growth inhibition (CGI) technique has been developed based on the principles of the SG technique at Heriot-Watt University, UK. The aim of this method is to overcome the problem of stochastic induction time results and provide a more rapid and reliable mean for evaluating KHI performance offering a new insights into KHI inhibition mechanisms [12]. A brief summary of the successful newly developed CGI approach is reported in the first part of this communication. Moreover, repeatability and transportability of the CGI technique results and CGI patterns have been examined and compared between two different set-ups (high pressure autoclave and mixing ball rocking cells) in the second part of this work. Finally, results of the conventional induction time technique for evaluating KHIs within different crystal growth inhibition regions are presented. Furthermore, the CGI regions are closely correlated with the conventional induction time trends such that once the CGI regions are mapped; the expected ti trends can be approximated to aid ti measurements.

BACKGROUND TO THE CGI METHOD As mentioned, the main aim of developing the new CGI technique was to focus on KHI inhibition properties which were both consistent and transferable between different runs and set ups and thus avoid problems associated with stochasticity of nucleation which is in turn reflected in induction time measurements. Consistent inhibition features in KHI systems are strongly suggested in literature with three set of findings being particularly significant. In the first set of

ACS Paragon Plus Environment

4

Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

findings, single crystal studies showed that in KHI systems hydrate crystals are completely inhibited at specific degrees of subcooling from the hydrate phase boundary [14-16]. Subsequently, this indicates the likelihood of a similar behaviour in hydrate forming gaseous hydrocarbon (e.g. methane, natural gas) systems. The second set of studies report hydrate dissociation rate reduction outside the hydrate phase boundary [17,18]. This behaviour suggests a structural/compositional differences between hydrate formed in KHI and KHI-free systems which is supported by the more stable solid clathrate hydrate formed in the presence of KHI compared to a KHI free system. This pattern appears perfectly possible through KHI polymers adsorbing on to the hydrate surface in a systematic style. Finally, the third set of findings demonstrate that KHIs can further inhibit hydrate formation and growth in an already nuclei seeded system [8-11]. This pattern backs up the first set of findings in that KHIs can inhibit hydrate growth up to a certain subcooling. When all these findings are put together, it is strongly suggested that KHIs can interact with hydrate crystals during the entire process of hydrate nucleation, growth and dissociation by means of polymer crystal surface adsorption. Subsequently, since surface adsorption is a thermodynamic phenomenon, it is strongly suggested that KHI inhibition properties are largely controlled by thermodynamics and therefore have a repeatable behaviour. Knowing that in addition to inhibiting hydrate formation, KHIs can prevent already present hydrate crystals from growing any further, is of particular importance in this study. If KHIs show similar performance and thermodynamic properties in real systems, there is strong hope that they can be used much more widely in the industry [19]. Therefore as part of method development, more detailed and systematic examinations were conducted on the ability of KHIs to inhibit formation or growth of already formed clathrate

ACS Paragon Plus Environment

5

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

hydrates up to specific subcoolings inside the hydrate stability zone. For this purpose, similar to the previously explained SG method, system is initially cooled to temperatures inside the hydrate phase boundary to form hydrates. The system is then heated to dissociate these hydrate crystals but rather than keeping only hydrate history, a very small fraction of hydrate (typically around circa 0.5% of aqueous phase as hydrate) is kept in the system before re-cooling it to temperatures inside the hydrate stability region for hydrate growth pattern observations. In this method, the presence of hydrate in the system means that the nucleation process and uncertainties related to that stage is eliminated and measurements are solely representing KHI influence on growth patterns of already present hydrate crystals. Application of the above procedure revealed that, although small amount of crystals were present in the system, no failure was observed in the presence of KHI polymers (PVCap, PVP, commercial polymers). Moreover, it has been found that similar to findings of THF systems, at atmospheric pressure, KHIs can completely inhibit hydrate formation and growth up to significant subcoolings inside hydrate phase boundary [12, 19]. Furthermore, KHI inhibition properties extend to higher subcoolings at which hydrate growth rates are reduced significantly compared to a KHI-free system. Different CGI boundaries and regions can be defined and distinguished based on distinct changes of hydrate growth rates (order of magnitude changes) at different subcoolings. It is yet unclear whether the exact origins of these crystal growth inhibition ‘regions’ are related to variable polymer interference or functions of subcooling/driving force which have been found to be related to polymer adsorption on specific crystal faces by single crystal literature studies [14-16]. Regardless of this fact, these regions are completely apparent and have been defined in the CGI technique as follows:

ACS Paragon Plus Environment

6

Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Complete Inhibition Region (CIR) Hydrate growth inhibited indefinitely (completely at least up to very long test periods of more than 21 days, although evidence show continuation of this pattern up to indefinite periods) even when small fractions of crystals are present.

Slow Growth rate Region (SGR) Hydrate crystals can grow in this region, but at a rate measurably slower than what would be in the absence of a KHI. Hydrate inhibition can vary from significantly at smaller subcoolings to limited at higher subcoolings. This region can be subdivided into smaller regions based on hydrate growth rates (e.g. very slow, slow, moderate).

Rapid Growth Region (RGR) Presence of polymer does not appear to affect hydrate growth rate and these rates are close to or the same as they would be in the absence of a KHI, which at high subcoolings (e.g. for well performing inhibitors) is generally rapid.

Slow Dissociation rate Region (SDR) Similar to previous findings reported in the literature [17,18], in the presence of polymer, hydrate dissociation rate is an order of magnitude slower than the dissociation rate in a KHI-free system up to a couple of degrees outside the hydrate phase boundary. This region has previously been reported as the ‘history’ region by other researchers [8-11].

ACS Paragon Plus Environment

7

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

Table 1 summarizes the classification of crystal growth inhibition (CGI) regions based on orders of magnitude change in hydrate growth rates (% water converted to hydrate per hour).

Table 1. Classification of crystal growth inhibition (CGI) regions based on orders of magnitude change in hydrate growth rates, as commonly observed across region boundaries. Defining characteristics of the hydrate slow dissociation region (SDR), which corresponds to the ‘history’ region are also shown [19].

Typical growth rates order of Growth magnitude (% water* / hr) description

*

rate

CIR

0.00

No growth

SGR (VS)

0.01 (< 0.05)

Very slow

(S)

0.1 (≥ 0.05 to < 0.5 )

Slow

(M)

1 (≥ 0.5 to < 5)

Medium

RGR

10 (≥ 5)

Rapid

SDR

Dissociation rate one order of (Abnormally) Slow magnitude less than for no KHI dissociation

% water converted into hydrates

To sum up, in the Complete Inhibition Region (CIR) the growth is indefinitely inhibited (completely inhibited at least up to the incredibly long test periods in addition to evidence showing continuation of this pattern indefinitely with even hydrate dissociation detected in some cases, with potential hydrate dissociation). Conversely, in the Rapid Growth Region (RGR) the growth rate is somewhat equal to the hydrate growth rate of the KHI-free system. Similar to the CIR region, in the presence of KHI, at the same pressure and temperature, hydrate growth rate in the Slow Growth Region (SGR), is slower than that in a system where there is no KHI present.

ACS Paragon Plus Environment

8

Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

This region is subdivided into smaller regions classified based on a growth rate order of magnitude.

EXPERIMENTAL Equipment and Materials While no specialized equipment is required for undertaking the experiments, particular care in the design and layout of the equipment should be taken into account to maintain repeatable, reliable and transferrable data while determining crystal growth inhibition regions for the development of the CGI technique. Earlier literature studies which were done based on induction time methods have reported that results of the KHI performance evaluations can be different when tested in different set-ups and therefore are not transportable between set-ups [2]. To investigate the repeatability and transferability of the CGI technique, experiments have been conducted and compared in two different types of autoclaves (high pressure stirred autoclave and rocking cells). In both these setups constant volume method has been used (although constant pressure gas consumption tests have also yielded identical results [19]) in which CGI regions are detected by monitoring hydrate growth rate based on changes of pressure and temperature. Figure 1 demonstrates the high pressure stirrer autoclave in which most of the experiments have been conducted in this work. As illustrated in this figure, the set-up is a high pressure 280ml titanium or stainless steel cell which can operate up to 410 bar between 30 and +100 C. Cell temperature is controlled by circulating coolant from a programmable cryostat, which can maintain the cell temperature to within 0.1 °C, through a jacket surrounding the cell. A platinum resistance thermometer (PRT, ± 0.1 °C), connected to a computer for direct acquisition, is used

ACS Paragon Plus Environment

9

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

for measuring the inside temperature of the cell. Measuring the cell pressure is done by either a strain gauge pressure transducer (± 0.4 bar) or a precision Quartzdyne (± 0.07 bar) transducer directly mounted to on the cell; these being regularly calibrated against a dead weight tester. The same data acquisition unit used for the temperature probe is connected to the pressure transducer to allow real time monitoring and recording of cell temperature and pressure throughout different temperature cycles. To shorten the time required for achieving a thermodynamic equilibrium, a stirrer with a magnetic motor was used to agitate the test fluids. Accordingly, to aid further mass transfer and maximize reaction rates, the impeller speed was normally set at ~750 rpm, giving good shearing/co-mingling of aqueous and gaseous phases. Cell aqueous liquid volume fraction was typically 0.80-0.85. It is important that the system is well mixed throughout the test (i.e. no dead volume in the cell) to avoid hydrate formation from the water condensed in the vapor phase where there is no KHI. The hydrates formed from condensed water (which has not been in contact with the KHI inhibited aqueous phase) can cause confusion/error in the evaluation of the KHI. Figure 2 illustrates the 300 ml, high pressure (up to 690 bar), titanium rocking cell used in some of the tests. Temperature control and pressure measurements are done using very similar apparatuses and techniques to high pressure stirred autoclaves explained above. Similar to an autoclave system, temperature measurements are determined by a platinum resistance thermometer (PRT, ± 0.1 °C) which is connected to a computer for direct acquisition, but this time due to the equipment design cannot be in direct contact with cell fluid and is placed inside the cooling jacket surrounding the cell. Similarly, cell pressure is measured by either strain gauge pressure transducer (± 0.4 bar) or precision Quartzdyne (± 0.07 bar) transducer.

ACS Paragon Plus Environment

10

Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

The main difference between the high pressure autoclave and the rocking cell is the mixing method employed in the two systems. In a rocking cell, mixing is done using mixing balls with the cell rocked fully through 180°, typically once every 30 seconds or more.

Magnetic motor Pressure transducer

PRT Impeller Coolant jacket Inlet/outlet

Figure 1. Schematic illustration of the 280 ml high pressure (max 410 bar) autoclave cells used in some experiments.

Figure 2. Schematic illustration of the high pressure (max 690 bar), 300 ml (when piston fully retracted) high pressure rocking cell used in some experiments.

ACS Paragon Plus Environment

11

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

The KHI polymer used in the tests was poly-n-vinylcaprolactam (PVCap) which was LuvicapEG base polymer (average molecular weight / AMW =~7000) supplied by BASF with the ethylene glycol solvent removed by vacuum oven drying. Ultra high purity grade methane gas (99.995% pure) supplied by BOC was used in some tests. Also a natural gas, supplied by BOC, has been used in some tests for which the composition is given in Table 2. Solutions were prepared using deionized water throughout the experimental work with aqueous PVCap solutions prepared gravimetrically.

Table 2. Composition of the standard North Sea natural gas (NG) supplied by BOC used in experiments. Component

Mole%

Methane

89.41

Ethane

5.08

Propane

1.45

i-Butane

0.18

n-Butane

0.26

i-Pentane

0.06

CO2

1.55

Nitrogen

1.93

n-Pentane

0.06

n-Hexane

0.02

Total

100.0

ACS Paragon Plus Environment

12

Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Crystal Growth Inhibition Method Procedure The CGI method developed by Anderson et al (2011) has been used in most of the experiments. As previously explained the method is summarized as: a) System is initially cooled down to high subcoolings for hydrate formation. b) Hydrated system is heated up to a couple of degrees above the hydrate stability region until most of the hydrate has dissociated (only around < 0.5% remained). The amount of hydrate is calculated using the HydraFLASH® thermodynamic model based on system composition and volumetric data [20]. c) System is slowly cooled down (typically ~1.0 °C / hr) to lower temperatures for detection of hydrate growth rate and determining the different CGI regions. d) A further “step cooling run” inside the hydrate phase boundary (24hr steps at each temperature) with a small amount of hydrate is also done to determine the extent of the CIR and the SGR(VS) [12].

RESULTS AND DISCUSSION As already mentioned, development of the crystal growth inhibition technique and CGI studies have been undertaken and previously reported in Anderson et al. (2011). Here we initially present a summary of these findings from a wide range of tested systems. Thereafter, results of tests carried out in a second experimental set-up (rocking cell) is presented and compared to results of similar tests done in an autoclave set-up to investigate transferability of results using this technique. In the final part of this section, results of the developed CGI method and the conventional induction time method are compared and correlations between them are presented.

ACS Paragon Plus Environment

13

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

Summary of PVCap Performance and Concentration Evaluation Based on CGI Technique As detailed in our previous work, PVCap induces hydrate inhibition regions which are named the crystal growth inhibition (CGI) regions. In this study, PVCap crystal growth inhibition patterns have been measured for a number of concentrations, ranging from 0.1 to 5.0 mass% aqueous in the presence of different hydrocarbon gaseous systems. Results of this study have shown that at all of these different concentrations four KHI-induced growth/dissociation regions have been identified: (1) a region where hydrate formation and growth is completely inhibited (CIR) (2) a region where hydrate formation does occur but at a slow rate (SGR(S)), (3) a region where hydrate formation rate is moderate (SGR(M)) and (4) a region where slow dissociation is observed (SDR) [12]. Figure 3 demonstrates summary of the extent of the mentioned CGI regions for a range of PVCap concentrations in the presence methane. It should be noted that in this work subcooling is defined as temperature degrees inside the hydrate phase boundary and therefore is a positive number at temperatures within this boundary (T). It should also be noted that ΔT is the temperature difference from the hydrate phase boundary (T–Thydrate phase boundary) and therefore is a negative number at temperatures within the hydrate stability region. Results and studies such as Mozaffar et al. (2016) have shown that in PVCap systems with methane, boundaries are very consistent, commonly being present at a generally fixed subcooling from the phase boundary for a wide range of pressures [22]. However, studies on a wide variety of systems have shown that depending on the host gas (methane, ethane, CO2, NG, etc.) polymer CGI performance in terms of subcooling extent can vary slightly with increasing pressure. As evident from this figure, evaluating the performance of PVCap shows that although very consistent, a distinct PVCap concentration dependency for the degree of hydrate inhibition is

ACS Paragon Plus Environment

14

Page 15 of 31

offered. In these systems, hydrate inhibition properties clearly improve as the concentration of PVCap is increased up to 0.5 mass% PVCap. As illustrated, almost all CGI regions are relatively similar for 0.5 and 1.0 mass% PVCap systems. On the other hand, for the 0.25 mass% PVCap both CIR and SGR boundaries are smaller than the tested systems with higher PVCap concentrations. As evident from these results, while the CIR extends to a subcooling of ~5.2 °C for both 0.5 and 1.0 mass% PVCap, it is at a smaller ΔTsub of ~−3-−4 °C for the 0.25 mass% PVCap system. Data thus suggest a concentration dependence in terms of the extent of the CIR, with an optimal concentration of >0.25 and ≤ 0.5 mass% aqueous. Furthermore, so far results have demonstrated that all concentrations share the common SGR boundary line at ΔTsub = ~−5.2 °C suggesting a possible crystal growth morphology controlled feature at this subcooling. Furthermore, figure 3 shows that increasing the PVCap concentration from 0.5 to 1.0 mass% can reduce the hydrate growth rate in the SGR region (ΔTsub of ~−7.3 to −9.5°C) up to 4 times slower in the first 50 hours [18]. -10 ~9.8 RGR

SGR(M)

-8

~7.2

DTs-I (°C)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

-6 ~5.2 SGR(S)

-4

~2.8

-2

CIR

0 0

0.2

0.4

0.6

0.8

1

1.2

Mass% PVCap

Figure 3. Methane hydrate CGI regions (ΔTs-I from s-I phase boundary) as a function of mass% PVCap aqueous.

ACS Paragon Plus Environment

15

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

Tohidi et al. (2001) have described the morphology of hydrate crystals formed at high subcoolings as dendrites or crystals with very high surface to volume ratio which is due to the high driving force at such conditions. However, it has been shown that when these crystals are heated to lower subcoolings there is a tendency to reduce the surface free energy or in other words the surface area and therefore their morphology reforms to a more euhedral shape [23]. For hydrate to undergo this geometry change it is required to dissociate from the long axis and re-grow from the short axis. As Anderson et al. (2011) has described the KHI polymer will adsorb to the hydrate crystal surface which will prevent hydrate growth from the latter within the CIR. This will results in net volume reduction of hydrate crystal and hence hydrate dissociation at lower subcoolings inside the complete inhibition region [12]. Other studies have shown that similar CGI patterns and regions for PVCap have been measured for more complex multicomponent gas systems which reveal that a close relationship lies between the CGI regions and the s-I boundary [13]. This in turn suggests that the KHI failure in these systems is potentially due to the formation of s-I hydrate crystal and therefore support the fact that PVCap is mainly a hydrate structure II inhibitor [13,19]. However, understanding the actual behaviour of these KHIs and the CGI patterns require further investigation and this cannot yet be concluded from current studies.

Evaluation of Repeatability and Transportability of the CGI Technique in Different Setups: Rocking Cell & Standard Autoclave Studies have shown that the new CGI approach is repeatable in different autoclaves stirrer blade mixers [12] but the transportability of this technique to a set-up with a different mixing facility is also of importance. In this section result of CGI studies in a rocking cell set-up is

ACS Paragon Plus Environment

16

Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

presented to show the transportability of the technique when the methodology is correctly applied. Example heating and cooling curves for CGI technique and measured CGI regions for 0.5 mass% PVCap aqueous with methane in the autoclave set-up and the rocking cell, described in the Equipment and Materials section of this paper, are presented in Figure 4 for comparison. Results of the measured CGI regions for this system at ~85 bar are also presented in Table 3. Experimental results show that all he measured CGI regions are very much identical in both setups within experimental error (±0.5 °C). The slight difference observed in the results is namely caused by the different placement of the temperature probe in the two different set-ups; temperature probe is placed inside the cell in the autoclave set-up whereas it is mounted in the cell jacked in the rocking cell set-up. The placement of the probe inside the jacket rather than the rocking cell itself will result in a slight temperature deviation from the true fluid temperature during the fast heating and cooling runs in this set-up; the temperature of the jacket is slightly higher than the cell temperature at tested conditions. Also the slight temperature difference can cause a non-linear convex liquid + gas relationship which is again equipment related and can be eliminated if cooling/heating rates are slowed down.

Table 3. Comparison of experimentally determined CGI region ΔT extents for methane with 0.5 mass% PVCap aqueous at ~85 bar in autoclave and rocking cell apparatus. Regions are indistinguishable within experimental error.

Rocking Cell Autoclave

ΔTsub SGR(M) ‐RGR / °C −9.5 −9.6

ΔTsub SGR(S-M) / °C −7.9 −7.3

ΔTsub CIR ‐ SGR(S) / °C −5.2 −5.2

ΔTsub SDR / °C 4.2 3.9

ACS Paragon Plus Environment

17

Energy & Fuels

110 100

RGR

CIR

SGR SGR(S) (M)

SDR

90 80

P / bar

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

70 60

50

Autoclave CGI cooling/heating runs

40

Rocking cell CGI cooling/heating runs

30 -4

-2

0

2

4

6

8

10

12

14

16

18

T / °C

Figure 4. Comparison of example CGI cooling and heating runs for 0.5 mass% PVCap aqueous with methane in the rocking cell and standard autoclave. Similarities of these results support the theory that CGI regions are primarily thermodynamically controlled and therefore if the CGI technique is followed correctly results will be completely transferable and repeatable between different equipment.

CGI Region Relationship to Induction Times As previously mentioned KHIs have traditionally been evaluated based on induction time measurements. Therefore, for a successful replacement of the commercial but unrepeatable ti method with the new CGI technique in the industry, is it important to determine the relationship between ti and CGI regions. Literature has shown that while hydrate crystals can form almost instinctively inside the hydrate region, even at small subcoolings, this process can be delayed for very long periods due to the delay in the formation of the hydrate nuclei even at high subcoolings. However, the higher driving force for the formation of the nuclei at high subcoolings results in a higher chance of nuclei formation at such conditions leading to shorter induction times [3,10]. A similar trend is

ACS Paragon Plus Environment

18

Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

expected for a system with KHI but with a longer induction time as a result of the inhibitor in the system. A simplified illustration of this trend is demonstrated in Figure 8. Hydrate phase boundary Subcooling

Figure 8. General/conventional understanding of induction time behaviour in relation to subcooling from the hydrate phase boundary with/without KHI. As explained in the first section of this work, hydrate nucleation and growth is indefinitely inhibited in the complete inhibition region (CIR) in a well-mixed and equilibrated system regardless of having hydrate/history in the system or not. As a result, induction time is not a measurable property, ti is infinite, for a wide range of subcoolings within the CIR in this system. In this region, critical nuclei can possibly form however they are inhibited from growing into larger more stable crystals due to the presence of the polymer. In a system with hydrate/ hydrate history, growth starts almost immediately, though may be very slow, after the system enters the slow growth region (SGR). Consequently, induction time is effectively zero or incredibly small from an academic point of view in this system. Conversely, a true induction can be measured inside the SGR and the RGR when there is no hydrate history in the system. However at such conditions the measured induction time can be very stochastic due to the stochastic nature of nuclei formation.

ACS Paragon Plus Environment

19

Energy & Fuels

Example induction time measurements, demonstrating the above mentioned stochastic behaviour, are shown in Figures 9 and 10. These figure illustrates pressure changes due to hydrate formation (ΔPhyd) as a function of time for ‘no history’ induction time measurement runs compared to a ‘history present’ run for a 0.5 mass% PVCap aqueous with methane at ~70 bar and ~−6.2 °C and ~−7.6 °C subcooling ,inside the SGR(S) and SGR (M) respectively. Figures 9 and 10 show that while in both systems “history/hydrate present” and “no history” runs differ in the fact that there is no induction time in the former but a measurable one in the latter, hydrate growth patterns are essentially identical regardless of whether there is hydrate/history present at the start of the test or not. Furthermore, these figures show that hydrate growth rates in these systems agree with the current inhibition regions of very slow growth rates (Dp/Dt) in Figure 9 and slow growth rates (Dp/Dt) in Figure 10 which is typical for these regions. 2 0 No history

-2

DP / bar

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

ti = >66.3 hrs

ti = 43.7 hrs

With hydrate history

-4

0.5 mass% PVCap SGR(S), T = 3.6 C DTsub = 6.2 C Pi = 70.3 bar

-6 -8 -10 0

10

20

30

40

50

60

70

t / hours

Figure 9. Plot of change in pressure due to hydrate formation (ΔP) as a function of time for ‘no history’ induction time measurement runs (red points) compared to a ‘history present’ run (yellow points) for a 0.5 mass% PVCap aqueous with methane system at 6.2 °C subcooling (SGR(S) ) and 70.3 bar. As shown in the figure without hydrate history the induction time is controlled by nucleation, hence stochastic.

ACS Paragon Plus Environment

20

Page 21 of 31

0 No history

With history -10

DP / bar

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

ti = 2.5 hrs

ti = 23.9 hrs

ti = 8.5 hrs

-20 0.5 mass% PVCap SGR(M), T = 2.2 C DTsub = 7.6 C Pi = 69.8 bar

-30

Equilibrium Pressure -40 0

5

10

15

20

25

30

t / hours

Figure 10. Plot of change in pressure due to hydrate formation (ΔP) as a function of time for ‘no history’ induction time measurement runs (orange/red points) compared to a ‘history present’ run (blue points) for a 0.5 mass% PVCap aqueous with methane system at 7.6 °C subcooling (SGR(M)) and 69.8 bar. As shown in the figure without hydrate history the induction time is controlled by nucleation, hence stochastic. The above results support the general belief that “the higher the subcooling, the smaller the induction time”. However, since in a KHI present system induction time is not a measurable property within the CIR, it can further be concluded that the trend illustrated in Figure 8 is no longer valid for these systems. Alternatively, in KHI present systems there is a ΔTCIR shift in the induction time versus subcooling curve to demonstrate the true system property which is displayed in Figure 11.

ACS Paragon Plus Environment

21

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

Hydrate phase boundary Subcooling

Figure 11. Real behaviour of induction time in relation to subcooling from the phase boundary in KHI free and KHI present systems. Measurement of ti data for the 0.5 mass% PVCap aqueous with methane and natural gas are illustrated in Figures 12 and 13 respectively, with the CGI region boundaries also illustrated for comparison. In both these figures induction time measurements were performed after CGI studies. For a good ti measurement there should be no hydrate history in the system and for this purpose before each measurement the system was heated to at least 15 °C higher than the phase boundary and kept there for 2 hours after which the system was cooled rapidly to the aimed set point temperature. Induction time was measured as the time between cell temperature reaching within 0.2 °C of the set point temperature and the time when hydrate growth was first observed from pressure drop. The above finding on the correlation between CGI regions and induction time data are summarized in Table 3.

ACS Paragon Plus Environment

22

Page 23 of 31

Table 3. Results of investigations into whether an induction/hold time (ti) can be observed in identified crystal growth regions depending on system properties for PVCap systems [19]. Initial condition

Zone ti?

Reason

Hydrate

CIR

No

Crystal growth inhibited indefinitely

History (no hydrate)

CIR

No

Crystal growth inhibited indefinitely

No history or hydrate

CIR

No

Crystal growth inhibited indefinitely

Hydrate

SGR (All)

No

Growth begins immediately but rates slow due to polymer inhibition

History (no hydrate)

SGR

No

Growth begins immediately but rates slow due to polymer inhibition

Yes

But stochastic and behaves as for history/hydrate present when growth does begin

(All) No history or hydrate

SGR (All)

Hydrate

RGR No

Growth occurs immediately, typically rapidly

History (no hydrate)

RGR No

Growth occurs immediately, typically rapidly

200 RFR RGR

RGR SGR M-R

SGR RGR S

CIR

150

t i (hrs)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

100

50

ti data Open symbols = nucleation not observed

0 -12

-10

-8

-6

-4

-2

0

DT (C)

Figure 12. Plot of induction time versus subcooling for the 0.5 mass % PVCap aqueous with methane example system at ~80 bar. Determined CGI region boundaries are shown for

ACS Paragon Plus Environment

23

Energy & Fuels

comparison. Points illustrated with open circles have been left at that condition for the shown periods during which no sign of hydrate formation was observed. Tests were then stopped and system was heated up to higher temperatures so induction time would have been measured equal or more than the observed times [19].

60 P = 207 bar

s-I

SGR RGR

CIR VS

S

s-II

s-I+s-II

40

ti (hrs)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 31

20

0 -16

-14

-12

-10

-8

-6

-4

-2

0

DTs-II (C)

Figure 13. Plot of induction time versus subcooling for the 0.5 mass% PVCap aqueous with NG example system at ~207 bar. Determined CGI region boundaries are shown for comparison [19]. As Figures 12 and 13 show, for the tested systems, induction times approach zero a few degrees inside the RGR and rise exponentially towards infinity over a very short temperature range inside the SGR region. This behaviour is illustrated by the drawn approximate exponential lines through ti points on these figures. Moreover results illustrate that there is a lot more scatter in the measured ti at conditions closer to the SGR boundary. Therefore although due to the scatter in induction time measurements not all point will be placed exactly on this curve, an approximate trend of this property can be traced, rising from zero towards infinity as the CIR is approached.

ACS Paragon Plus Environment

24

Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Conclusively, ti measurements show that, by taking CGI regions boundaries into account, the large scatter observed for ti values over a narrow range of subcoolings (over 23 °C) can be explained very clearly. These results show that the exponential changes in the induction time starting from infinite in the CIR (no nucleation/growth) to very long in the SGR, and finally to very short in the RGR can be very closely related to changes in the growth rates which increase in an order of magnitude over a relatively small subcooling range moving from zero growth rate in the CIR to very rapid growth rate in the RGR. Thus all evidence supports the fact that induction times/nucleation inhibition patterns are primary related to underlying crystal growth inhibition patterns, meaning the widespread assumption that ti is primarily related to subcooling (e.g. through the size of the critical radius) from the thermodynamic phase boundary is not necessarily correct. Conclusively, to increase the efficiency of induction time measurements and eliminate the time loss associated with these measurements (e.g. measuring infinite ti inside the CIR), ti measurements can be followed on from initial CGI studies which will provide likely ti patterns ahead of making such measurements.

CONCLUSIONS This study had previously revealed that in addition to inhibiting the hydrate nucleation process, kinetic hydrate inhibitors can interfere/inhibit with the process of hydrate growth especially up to specific subcoolings.

This property has resulted in defining a number of repeatable and

transferrable crystal growth inhibition (CGI) regions as a function of subcooling, ranging from complete inhibition region (CIR) to slow growth region (SGR) and finally rapid growth region

ACS Paragon Plus Environment

25

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 31

(RGR), which can be used as a much more rapid, reliable and quantifiable tool for evaluating the performance of KHIs. These CGI regions can be measured promptly and simply using conventional laboratory equipment such as high pressure stirrer autoclaves and high pressure rocking cells. Experiments in both these set-ups have proven that as long as good mixing is provided, results of CGI studies are completely transferrable and repeatable between different set-ups. So far tests on different PVCap concentrations have illustrated that there is a concentration dependence in terms of the extent of the CIR, with an optimal concentration of >0.25 and ≤ 0.5 mass% aqueous. Furthermore, results have shown that all tested PVCap concentrations share a common SGR boundary suggesting this is a possible crystal growth morphology controlled feature. This further supports the general understanding that KHI hydrate inhibition properties are a result of polymer surface adsorption on to specific hydrate crystal faces that can change in morphology at different subcoolings. The final section of this study revealed that there is a very close relationship between the crystal growth inhibition regions of a KHI and induction time/nucleation inhibition patterns related to that KHI in the tested system. Correlating the two methods has shown that induction time can only be measured for a relatively short range of subcooling within the SGR and the RGR. Furthermore, it has been revealed that it is not possible to measure hydrate nucleation induction time inside the CIR region which supports CGI studies in that hydrate growth is completely/indefinitely inhibited within this subcooling range. Finally, induction time measurements show that an approximate exponential trend of this property can be traced, rising from very short/almost zero ti inside the RGR region towards infinity as the CIR is approached.

ACS Paragon Plus Environment

26

Page 27 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

AUTHOR INFORMATION Corresponding Author * Phone: +44 (0)131 449 7472 E-mail: [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ‡These authors contributed equally ‡ Bahman Tohidi Email: [email protected] ‡ Ross Anderson Email: [email protected]

ACKNOWLEDGMENT This work was undertaken as part of a Joint Industry Project (JIP) at Heriot-Watt University, Scotland, UK, supported by Baker Hughes, Champion Technologies, Clariant Oil Services, DONG Energy, OMV, Petronas, Shell/NAM, Statoil and Total, whose support is gratefully acknowledged. Authors would like to thank Rod Burgass and Antonin Chapoy for their valuable contributions to the project, and Jim Allison for manufacture and maintenance of experimental equipment. ABBREVIATIONS ΔP, Change in pressure due to hydrate formation [bar]; ΔT, Temperature difference [°C]; ΔTs-I, Temperature difference from s-I hydrate phase boundary [°C]; ΔTs-II, Temperature difference from s-II hydrate phase boundary [°C];

ACS Paragon Plus Environment

27

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 31

AA, Anti-Agglomerant; AMW, Average Molecular Weight; CAPEX, Capital Expenditure; CIR, Complete (hydrate) Inhibition Region; CGI, Crystal Growth Inhibition; HSZ, Hydrate Stability Zone; JIP, Joint Industry Project; KHI, Kinetic Hydrate Inhibitor; LDHI, Low Dosage Hydrate Inhibitors; M, Medium; NG, Natural Gas; OPEX, Operational Expenditure; PRT, Platinum Resistance Thermometer; PT, Pressure-Temperature; PVCap, Poly-n-vinylcaprolactam; PVP, Poly-n-vinylpyrrolidone; RGR, Rapid (KHI) Growth (rate) Region; S, Slow; SDR, Slow (abnormally, hydrate) Dissociation Region; SG, Second Germination (KHI test method); SGR, Slow (hydrate) Growth (rate) Region; SI, Structure I; SII, Structure II; t, time; ti, Hydrate nucleation induction time [hrs];

ACS Paragon Plus Environment

28

Page 29 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

VS, Very slow

REFERENCES (1) Kelland MA. History of the development of low dosage hydrate inhibitors. Energy and Fuels 2006;20:825-847. (2) Klomp U. The world of LDHI: From conception to development to implementation. Proceedings of the 6th International Conference on Gas hydrates, Vancouver, Canada, 2008:5409. (3) Sloan ED, Koh CA. Clathrate Hydrates of Natural Gases (3rd ed.). Taylor & Francis / CRC Press, Boca Raton, FL, 2008. (4) Freer EM, Sloan ED. An engineering approach to kinetic inhibitor design using molecular dynamics simulations. Annals of the New York Academy of Sciences, 2000;912:651-657. (5) Makogon YM, Sloan ED. Mechanism of kinetic hydrate inhibitors. Proceedings of the 4th International Conference on Gas Hydrates, Yokohama, Japan 2002:498-503. (6) Anderson BJ, Tester JW, Borghi GP, Trout B L. Properties of inhibitors of methane hydrate formation via molecular dynamics simulations. Journal of the American Chemical Society, 2005; 127:17852-17862. (7) Kvamme B, Kuznetsova T, Aasoldsen K. Molecular dynamics simulations for selection of kinetic hydrate inhibitors. Journal of Molecular Graphics and Modelling 2005;23:524-536. (8) Peytavy J-L, Glénat P, Bourg P. Qualification of low dose hydrate inhibitors (LDHIs): field cases studies demonstrate the good reproducibility of the results obtained from flow loops.

ACS Paragon Plus Environment

29

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 31

Proceedings of the 6th International Conference on Gas hydrates, Vancouver, Canada, 2008:5499 (9) Duchateau C, Dicharry C, Peytavy J-L, Glénat P, Pou T-E, Hidaldo E. Laboratory evaluation of kinetic hydrate inhibitors: a new procedure for improving the reproducibility of measurements. Proceedings of the 6th International Conference on Gas hydrates, Vancouver, Canada, 2008:5582 (10) Duchateau C, Peytavy J-L, Glénat P, Pou T-E, Hidalgo M, Dicharry C. Laboratory evaluation of kinetic hydrate inhibitors: A procedure for enhancing the repeatability of test results. Energy and Fuels 2009;23(2):962-966. (11) Duchateau C, Glénat P, Pou, T-E, Hidalgo M, Dicharry C. Hydrate precursor test method for the laboratory evaluation of kinetic hydrate inhibitors. Energy Fuels 2010;24:616–623. (12) Anderson, R., Mozaffar, H., Tohidi, B., Development of a Crystal Growth Inhibition Based Method for the Evaluation of Kinetic Hydrate Inhibitors, Proceedings of the 7th International Conference on Gas hydrates, Scotland, United Kingdom, 2011. (13) Glénat P, Anderson R, Mozaffar H, Tohidi B. Application of a new crystal growth inhibition based KHI evaluation method to commercial formulation assessment. Proceedings of the 7th International Conference on Gas hydrates, Edinburgh, Scotland, UK, 2011. (14) Makogon TY, Larsen R, Knight CA, Sloan ED. Melt growth of tetrahydrofuran clathrate hydrate and its inhibition: Method and first results. Journal of Crystal Growth 1997;179:258-262. (15) Larsen R, Knight CA, Sloan ED. Clathrate hydrate growth and inhibition. Fluid Phase Equilibria 1998;150-151:353-360.

ACS Paragon Plus Environment

30

Page 31 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(16) Larsen R, Knight CA, Rider KT, Sloan ED. Melt growth and inhibition of ethylene oxide clathrate hydrate. Journal of Crystal Growth 1999;204:376-381. (17) Habetinova E, Lund A, Larsen R. Hydrate dissociation under the influence of low-dosage kinetic inhibitors. Proceedings of the 4th International Conference on Gas Hydrates, Yokohama, Japan, 2002. (18) Svartaas TM, Gulbrandsen AC, Huseboe SBR, Sandved O. An experimental study on “un-normal” dissociation properties of structure II hydrates formed in presence of PVCAP at pressures in the region 30 to 175 bars – dissociation by temperature increase. Proceedings of the 6th International Conference on Gas Hydrates, Vancouver, Canada, 2008:5696. (19) Mozaffar, H., Development and Application of a Novel Crystal Growth Inhibition (CGI) Method for Evaluation of Kinetic Hydrate Inhibitors, PhD thesis, Heriot-Watt University, 2013. (20) HydraFLASH® Model V2.2. Hydrafact Ltd: Edinburgh, UK, 2011. www.hydrafact.com. (21) Tohidi, B., Anderson, R., Mozaffar, H., Tohidi, F., The Return of Kinetic Hydrate Inhibitors, Energy and Fuels, 2015, 29:8254−8260 (22) Mozaffar H, Anderson R, Tohidi B. Effect of alcohols and diols on PVCap-induced hydrate crystal growth patterns in methane systems. Fluid Phase Equilibria, 2016;425:1-8. (23) Tohidi B, Anderson R, Clennell MB, Burgass RW, Biderkab AB. Visual observation of gas-hydrate formation and dissociation in synthetic porous media by means of glass micromodels. Geology 2001; 29(9):867-870

ACS Paragon Plus Environment

31