Cyclopentane hydrate cohesion measurements and phase equilibrium predictions

Accepted Manuscript Cyclopentane hydrate cohesion measurements and phase equilibrium predictions Erika Brown, M. Naveed Khan, Davi Salmin, Jonathan Wells, Shenglong Wang, Cornelis J. Peters, Carolyn A. Koh PII:

S1875-5100(16)30315-8

DOI:

10.1016/j.jngse.2016.05.016

Reference:

JNGSE 1493

To appear in:

Journal of Natural Gas Science and Engineering

Received Date: 15 February 2016 Revised Date:

26 April 2016

Accepted Date: 4 May 2016

Please cite this article as: Brown, E., Khan, M.N., Salmin, D., Wells, J., Wang, S., Peters, C.J., Koh, C.A., Cyclopentane hydrate cohesion measurements and phase equilibrium predictions, Journal of Natural Gas Science & Engineering (2016), doi: 10.1016/j.jngse.2016.05.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

CYCLOPENTANE HYDRATE COHESION MEASUREMENTS AND PHASE EQUILIBRIUM PREDICTIONS Erika Brown1, M. Naveed Khan1,2, Davi Salmin1, Jonathan Wells1, Shenglong Wang1, Cornelis J. Peters2, Carolyn A. Koh1*

RI PT

1

Colorado School of Mines, Chemical & Biological Engineering Dept., Center for Hydrate Research, Golden, CO 80401, USA 2

Chemical Engineering Department, Petroleum Institute, Abu Dhabi, U.A.E

SC

*Corresponding author: [email protected], Ph: 303-273-3237

Abstract

M AN U

Clathrate hydrates can form in oil and gas flowlines at high pressures and low temperatures. These solids frequently form under flowing conditions, where they can encounter other hydrate particles due to turbulent mixing. Hydrates may also form in locations where they can anneal significantly before coming in contact with other hydrate particles, such as during transient shut-in conditions in a flow line. Understanding the effect of different shut-in/annealing time periods on hydrate cohesive forces can be important in providing further insight into effective strategies to prevent hydrate plug formation.

EP

TE D

Cyclopentane hydrates were tested using a micromechanical force (MMF) apparatus to investigate the effect of annealing time on the cohesive forces between hydrate particles. Annealing time was found to reduce the cohesion force between particles at various temperatures tested. This was attributed to the reduction of micropores that connect the unconverted water center of the hydrate to the exterior shell, reducing the water layer available for cohesion.

AC C

In order to further understand these systems, an understanding of the thermodynamic phase equilibria for systems containing cyclopentane is needed. The prediction capability of the in-house hydrate phase equilibria prediction tool (CSMGem) was therefore improved by optimizing the Kihara potential parameters for cyclopentane using pure and mixed cyclopentane hydrate phase equilibria experimental data. The optimized set of Kihara parameters was then used to predict the gas hydrate phase boundary as well as the fractional cage occupancy of guest molecules, where the larger cavities of structure II were confirmed, as expected, to be filled only by cyclopentane molecules.

Keywords: Clathrate hydrate, particle force, cyclopentane, phase equilibria

1

ACCEPTED MANUSCRIPT

Introduction

RI PT

Clathrate hydrates are inclusion compounds that typically form at high pressures and low temperatures. They are formed when water molecules form cages around suitably-sized guest molecules, enclathrating them [1]. The most common clathrate hydrates are structures I, II and H.

M AN U

SC

These structures vary based on the size of the guest molecule (or molecules) that stabilize the differently-sized cages. Structure I forms with smaller molecules such as methane, while larger molecules such as propane or cyclopentane are incorporated into Structure II. Structure H utilizes two different guests in order to properly stabilize the cages. Structure II is the most studied type of hydrate due to its prevalence in flow assurance. Flowlines offer ideal conditions for hydrate formation; therefore, significant research has been conducted to understand their properties. This includes studies on thermodynamics and modeling, as well as the examination of physical variables such as agglomeration and plugging. However, flow assurance is only one facet of clathrate hydrates; significant research is being conducted into the beneficial uses of hydrates, such as seawater desalination and gas storage.

AC C

EP

TE D

Cyclopentane is considered to be a model hydrate former due to the fact that it forms Structure II hydrate, is immiscible with water, and is thermodynamically stable above the ice point at atmospheric pressure (equilibrium temperature = 7.7˚C). Cyclopentane hydrates have been used to study hydrate cohesion forces using a Micromechanical Force measurement (MMF) apparatus operating at atmospheric pressure. Aman et al. performed a detailed study of many of the physical variables that affect the cohesion force, such as the temperature, pre-load force and contact time [2]. However, one important parameter in these cyclopentane hydrate experiments has not yet been extensively investigated: annealing time. Taylor proposed a mechanism for hydrate formation (using tetrahydrofuran hydrate) that is shown in Figure 1 [3].

Water Droplet

Thin Hydrate Shell Forms

Annealing/ Shell Thickening

Fully Converted Hydrate

Figure 1. Conceptual picture showing an inward-growing shell model for hydrate formation. Redrawn from [3]. Based on this model, it is assumed that hydrate first forms at the water/gas or water/hydrocarbon interface, where the concentrations of the hydrate formers are the highest. After the hydrate shell has completely covered the interface, the hydrate shell continues to thicken, which is limited by mass transfer across the shell, as well as the 2

ACCEPTED MANUSCRIPT

availability of micro-pores in the shell that allow material to cross the shell more easily [4]. Finally, if the particle is small enough or the annealing time long enough, the particle can convert fully to hydrate.

M AN U

SC

RI PT

Cohesion between hydrate particles at short contact times is primarily governed by capillary bridging [2, 5]. Bridges consist of a fluid immiscible with the bulk fluid, in this case water in a hydrocarbon bulk phase. This water layer may exist for a variety of reasons, including surface melting [6], formation of a thermodynamic water layer to minimize surface energy [2], or communication of water from the center of the particle to the surface through pores [4]. Most likely, the water layer exists due to a combination of several of these factors. The size of this liquid layer is a determining factor in the magnitude of the cohesion force. Understanding the influence of annealing time on the cohesion force can give insight into the behavior of the water layer, allowing for better understanding of the mechanism by which these capillary bridges operate. For this study, cohesion force was measured at a variety of temperatures and annealing times for cyclopentane hydrate systems containing no surfactants.

TE D

Accurate cyclopentane hydrate phase equilibria calculations are necessary for MMF studies, as well as technological applications using cyclopentane to enhance hydrate formation. The hydrate phase equilibria predictions performed in this work utilized the classical van der Waals and Platteeuw (vdWP) model (which is based on statistical thermodynamics) with some modifications (i.e. CSMGem, with Gibbs Energy Minimization), and Kihara potential optimization for cyclopentane and its mixtures [1,7,8].

1. Methods and Materials

AC C

EP

Cohesion force measurements were performed using an MMF apparatus [2]. This apparatus consists of an inverted light microscope equipped with recording equipment. A hand-operated micromanipulator holds a glass capillary tube connected to a calibrated glass fiber. A remotely-operated mechanical micromanipulator holds a second capillary tube and glass fiber. The microscope is equipped with a jacketed cooling cell capable of maintaining an aluminum cell filled with cyclopentane at any point between the ice point and the cyclopentane hydrate equilibrium temperature (0˚C to 7.7˚C). In order to perform an experiment, hydrate particles were created at the end of two glass fibers. Water droplets were deposited onto the ends of the fibers, then submerged in liquid nitrogen until frozen. The ice particles were transferred into the cyclopentane bath. The temperature is maintained above the ice point, which causes the ice particles to melt, providing a template/seed for hydrate formation. The hydrates are left to anneal for different periods of time in this study: 30 minutes, 60 minutes, and 120 minutes.

3

ACCEPTED MANUSCRIPT

RI PT

After the annealing period, 40 pull-off measurements are performed as shown in Figure 2 consisting of four steps: (a) beginning from a separated state, (b) the top particle is pushed into the bottom at a known preload force, ∆P. After a ten second waiting period, (c) the top particle is pulled away at constant velocity, until (d) the particles break apart. The distance at which they break apart, ∆D, is used with Hooke’s Law (Equation 1) and the calibration constant of the bottom cantilever, k, is used to determine the force necessary to separate the hydrate particles. ‫∆݇ = ܨ‬஽

(1)

M AN U

SC

Where F is the cohesion force, k is the calibrated spring constant of the glass fiber, and D is the displacement between the particles after they break apart.

TE D

Figure 2. Schematic showing the procedure for a pull-off measurement in the MMF apparatus. Reproduced from Aman et al. [2] with permission from the PCCP Owner Societies.

EP

Each experiment is repeated with at least three particle pairs, resulting in a minimum of 120 individual pull-off measurements for each data point. Error bars are reported as 95% confidence intervals based on these measurements.

AC C

2. Theory/calculation

The van der Waals and Platteeuw (vdWP) model was developed to predict hydrate phase equilibria [9]. In the vdWP model, a statistical thermodynamic approach is used to derive the thermodynamic properties of gas hydrates. This statistical thermodynamic model for gas hydrates was constructed using an analogy to the Langmuir isotherm, which has multiple adsorption sites for various hydrate former species. The vdWP model is based on the equality of the chemical potential in each phase. In this work, equality of fugacity for all the phases is utilized as a fundamental criterion for the phase equilibria calculations (Equation 2) [7, 8, 10, 11]. (2) fik = fi H

4

ACCEPTED MANUSCRIPT

where fiH is the fugacity of component i in the hydrate phase and fik is the fugacity of component i in any phase k. The fugacity of component i in any phase is calculated using Equation 3.

 µ k − g wo  f i k = f wo  w   RT 

RI PT

(3)

f wo : Fugacity of water at 298.15 K and 1 bar

µ wk : Chemical potential of water in any phase k g wo : Standard molar Gibbs free energy

m

i

M AN U

SC

R : Gas constant T : Temperature The chemical potential of water in the hydrate phase is calculated using the vdWP model as shown in Equation 4. (4) µ wH = g wβ + RT ∑ vm ln(1 − ∑ θ im ) + RT ln γ wH The cage occupancy of component i in cavity m ( θ im ) in a particular cavity is calculated by Equation (4a). Cim θ im = 1 + ∑ C jm f j

where 4π Cim = kT

R1 − ai

∫ 0

TE D

j

 ∑ win ( r )  r 2 dr exp − n  kT   

(4a)

(4b)

Equation 5.

EP

The interaction potential win (r ) between a gas molecule and a shell is calculated by

AC C

 σ 12 a a σ6 10 11 4 5  win (r ) = 2ε i z n  i11 (δ in + i δ in ) − i5 (δ in + i δ in ) Rn Rn Rn r  Rn r  where Rn : Radius of shell n

zn : Coordination number

ai : Hard core radius

σ i : Soft core radius ε i : Depth of intermolecular potential

5

(5)

ACCEPTED MANUSCRIPT

δ in N =

1 N

 a N a N r r − i ) − − (1 + − i )−  (1 − Rn Rn Rn Rn  

(5a)

The activity of water in the hydrate phase ( γ wH ) is calculated using Equation 6. β H P ∆v ∆g wo ∆h β 1 1 + wo ( − ) + ∫ dP Po RT RTo R T To

(6)

RI PT

ln γ wH =

g wβ : Standard molar Gibbs free energy of empty hydrate

γ wH : Activity of water in hydrate phase

θ im : Fractional cage occupancy of component i in cage m

SC

f i : Fugacity of component i (calculated using SRK-EoS) C jm : Langmuir constant of component j in cage m [1]

β

∆hwo = b∆v ∆v

H o

H o

=v −v H

(6a)

M AN U

where β ∆g wo = a∆voH β

[

v H = vo exp α 1 (T − To ) + α 2 (T − To ) 2 + α 3 (T − To ) 3 − k ( P − Po )

(6b) (6c)

]

(6d)

The volume parameters for sII are listed in Table 1

TE D

Table 1. Volume parameters for sII hydrate Hydrate Phase

A (J/cm3)

B (J/cm3)

α1(K-1)

α2(K-2)

α3(K-2)

k (bar-1)

sII

260

-68.64

2.02E-04

1.85E-07

-1.89E-07

3.00E-06

The Gibbs free energy of the empty hydrate phase is calculated using Equation 7.

EP

β β β T h P v g wβ g wo w w = − dT + ∫ dP Po RT RT RT ∫To RT 2 Enthalpy of the empty hydrate is calculated by Equation 8.

AC C

(7)

T

β hwβ = hwo + ∫ CPβ w

(8)

To

β : Enthalpy of formation at 298.15 K and 1 bar (Table 2) hwo β : Gibbs energy of formation of empty hydrate (Table 2) g wo

Table 2. Formation properties of standard sII hydrate state Type of hydrate phase

Enthalpy of formation at To and Po β

sII-(empty hydrate)

hwo (J/mol)

β (J/mol) g wo

-292044.10

-235627.53

The heat capacity is given by Equation 9.

6

Gibbs free energy at To and Po

ACCEPTED MANUSCRIPT

CPβ w = a + bT + cT 2 + dT 3

(9)

a, b, c and d are the heat capacity parameters of the phases (Table 3).

Table 3. Heat capacity parameters for sII hydrate a/R

b/R (102)K-1

c/R (105)K-2

d/R (109)K-3

sII

0.7354

1.418

-1.727

63.510

RI PT

Type of hydrate phase

The g wo (Standard molar Gibbs free energy of the water) is given by Equation 10.

(10)

SC

o β T h g wo g wo = − ∫ wo2 dT RT RTo To RT

M AN U

The ideal gas hydrate formation enthalpy, Gibbs energy properties, and heat capacity parameters are tabulated in Table 4. Table 4. Ideal gas hydrate formation properties. Phase Pure water

β (J/mol) hwo

β (J/mol) g wo

a/R

-242000

-228700

3.874

b/R (102)K-

c/R (105)K-2

d/R (109)K-3

1

0.0231

0.126

0

The chemical potential of water in the aqueous phase is calculated using Equation 11. RT

L

L

=

L

T h pure P v pure g wo pure − ∫ w 2 dT + ∫ w dP + ln(γ w x w ) To RT Po RT RTo

TE D

µwL

(11)

The formation properties for pure liquid water are given in Table 5. Table 5. Ideal gas hydrate formation properties for pure water. β hwo

(J/mol) -237129

EP

Phase Pure water

β (J/mol) g wo

-285830

a/R

b/R K-

c/R K-2

d/R K-3

1

8.71

0.125

-0.018

0

AC C

3. Results and Discussion

3.1. Cohesion Force Measurements

Cohesion forces were measured at three different annealing times as a function of subcooling. Subcooling (Tsub) is defined as the difference between the experimental temperature (Texp) and the equilibrium temperature (Teqm) as shown in Equation 12. ܶ௦௨௕ = ܶ௘௤௠ − ܶ௘௫௣

(12)

The driving forces for hydrate formation are typically considered to be the subcooling, as well as the over-pressure (additional pressure above the equilibrium pressure needed to form hydrates). As the MMF apparatus operates at atmospheric pressure, subcooling is 7

ACCEPTED MANUSCRIPT

RI PT

considered the vital driving force. It has been previously observed that at higher subcooling, the growth rate of the hydrate shell increases, as well as producing smaller crystallites that may correspond to a higher number of nucleation sites [12]. Higher subcooling has been also found to reduce hydrate formation induction times, e.g. in rocking cell tests [13]. It has been also shown previously that the subcooling has a significant effect on the cohesion force. Figure 3 shows the dependence of the cohesion force on the subcooling. It should be noted that the annealing time in all these previous tests was 30 mins. Therefore, it is likely that the subcooling and extent of annealing are coupled in these tests.

SC

12

M AN U

8 6 4 2 0 0

TE D

Cohesion Force (mN/m)

10

2

4 Subcooling (K)

6

8

EP

Figure 3. Relationship between cohesion force and subcooling for cyclopentane hydrates annealed for 30 minutes. Adapted from Aman et al. [2] with permission from the PCCP Owner Societies.

AC C

Figure 3 shows that the force increases significantly as it approaches the equilibrium temperature. This corresponds to Atomic Force Microscopy measurements carried out on an analogous system, ice, by Doppenschmidt et al. [6], which indicated that the water layer was thicker at temperatures closer to the ice point. Figure 4 shows the effect of annealing time at three different subcoolings,

8

9

∆T=6.7˚C

8

∆T=3˚C

7

∆T=1.7˚C

6 5

RI PT

Force (mN/m)

ACCEPTED MANUSCRIPT

4 3 2

SC

1 0 20

40

60 80 100 Annealing time (min)

M AN U

0

120

140

Figure 4. Force as a function of annealing time for three different subcoolings.

AC C

EP

TE D

Three mechanisms were proposed for the existence of a liquid layer: surface melting, transport of unconverted water through the hydrate structure, or a thermodynamic layer. For each temperature (subcooling) tested, a decrease in the cohesion force was observed with an increase in annealing time. Surface melting and transport of water through pores may have temporal components, while it is unlikely that the thermodynamic layer would change with annealing time. The annealing time dependence, therefore, may be explained by increased conversion of water at the surface of the particle into hydrate and/or by transfer of unconverted water from the center of the hydrate particle to migrate outwards to the hydrate shell, which was observed experimentally by Davies [4]. As the particles anneal, micropores in the hydrate shell which facilitate the transport of water from the shell’s interior may be reduced in size or eliminated, leading to a smaller water layer available to form capillary bridges with adjacent particles. Furthermore, a hydrate structure that allows water to traverse the hydrate shell could result in further water conversion and hence a denser hydrate shell over the annealing period, decreasing the mass transfer of further water through the hydrate shell over time. However, this data also shows that, for the time periods tested, the force was reduced by similar amounts for each of the temperatures. Since hydrate growth is accelerated at higher subcooling, this indicates that the process is likely to be mass transfer limited. It is also important to note that, for annealing periods of up to 120 minutes, the different subcoolings did not reach a common value, indicating that for these annealing times, the cohesion force remained dependent on temperature. Further studies at extended annealing times would be necessary to determine whether this effect remains for completely annealed particles; 9

ACCEPTED MANUSCRIPT

although it is expected that with a long enough annealing time the subcooling would have a smaller effect, thereby enabling ‘dry hydrate’ interactions. It should be also noted that a direct measure of the extent of water conversion to hydrate for the cyclopentane system is difficult, unlike the gas hydrate systems that could use gas consumption as a metric.

M AN U

SC

RI PT

This phenomenon may be important in cases where hydrates can exist for a period of time without encountering other hydrates. Such scenarios may occur in shut-in conditions, where pockets of water form hydrate films while the flowline production is suspended, or for deposits that form off the main line, such as in stagnant regions of a flowline. If, upon restart, unconverted water is released as the hydrates are dislodged, the forces would be expected to be significantly higher. It is also important to note that the concept of force decreasing with annealing time applies only to particles that come into contact after they have annealed significantly. Particles that are in contact while annealing occurs have been shown to sinter together and display orders of magnitude higher forces. 3.2. Cyclopentane Hydrate Phase Equilibria Predictions

TE D

The MMF measurements represent data collected for pure cyclopentane under atmospheric conditions. As indicated above, the parameters of subcooling and annealing time are important when assessing hydrate growth and interparticle forces. Therefore, hydrate phase equilibria behavior is important to assess the driving forces that lead to hydrate formation. In order to accurately predict the thermodynamic equilibria, the fitting parameters for the van der Waals and Plattteeuw model (vdWP) needed to be updated for cyclopentane-containing systems.

AC C

EP

The vdWP model was used to predict hydrate phase equilibria for sII pure cyclopentane. In order to capture the phase equilibria of pure cyclopentane, the Kihara parameters were optimized using the vdWP model and phase equilibria experimental data for pure cyclopentane [8,14]. As a first approximation, it was assumed that the activity of water in hydrate is approximately one, which eventually reduces the vdWP model to Equation 13.

µ wH = g wβ + RT ∑ vm ln(1 − ∑ θ im ) m

(13)

i

The combination of Kihara parameters of sigma = 2.82 and epsilon = 330 gave 6.12% in AADP (Absolute average deviation in pressure) for predicting three phase equilibria (HV-LW). In order to improve the accuracy of the pure cyclopentane hydrate phase equilibria predictions, repulsive constants were optimized for the calculation of activity of water in the hydrate phase by using the experimental volume of the hydrate (volume = 23.11 and optimized repulsive constant = 0.03011). By accounting for the activity of water in the hydrate phase (Equation 14), the accuracy of the pure cyclopentane hydrate 10

ACCEPTED MANUSCRIPT

phase equilibria prediction was significantly improved, as shown in Table 6 where sigma = 2.928 and epsilon = 290, gave an AADP=3.592 %.

µ wH = g wβ + RT ∑ vm ln(1 − ∑ θ im ) + RT ln γ wH m

(14)

i

TE D

Epsilon (K)

AADP (%)

250 254 280 300 330 320 270 330 330 310 290 280 270 290 290 290

11.116 11.242 12.418 11.41 8.141 11.965 11.632 7.737 22.622 21.965 23.153 23.326 22.928 19.089 15.575 3.592

SC

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Sigma (Angstrom) 3.0912 3.0735 2.9735 2.9086 2.8238 2.8507 3.0093 2.8097 2.8321 2.8879 2.949 2.983 3.0205 2.917 2.942 2.928

M AN U

Sr.No

RI PT

Table 6. Absolute average deviation in pressure (AADP) for pure cyclopentane hydrate for various optimized Kihara parameters (by assuming the activity of water in the hydrate phase is about 1).

AC C

EP

The Kihara parameters were optimized using the modified vdWP model (Equation 4) and pure cyclopentane experimental data (see Figure 5), and the resultant phase equilibrium prediction is given in Figure 6. The combination of optimized Kihara parameters with sigma = 2.928 and epsilon = 290 gives the best result in predicting pure cyclopentane hydrate phase equilibria.

11

RI PT

ACCEPTED MANUSCRIPT

TE D

M AN U

data (temperature in K, pressure is in bar).

SC

Figure 5. Optimized Kihara parameters from pure cyclopentane hydrate phase equilibria

EP

Figure 6. Phase equilibria prediction for pure cyclopentane hydrate.

AC C

Figure 6 shows the comparison of the phase equilibria prediction using optimized Kihara parameters (blue line) and Multiflash(R) (red line). Both the models show close agreement with experimental data, which demonstrates the reliability of the optimized parameters. These parameters were also validated by checking the cage occupancy of cyclopentane in the larger cavity, as shown in Figure 7. The larger cavity of structure II was completely filled by cyclopentane, as expected.

12

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 7. Fractional cage occupancy of cyclopentane as a function of pressure.

M AN U

4. Conclusions

TE D

Annealing time is a physical variable that plays an important role in particle cohesion. Longer annealing times were found to result in lower cohesion forces, regardless of the temperature at which the test was conducted. One hypothesis to explain this phenomenon is that micropores in the hydrate structure allow the flow of some water from the unconverted center of the hydrate particles to the exterior, increasing the water layer available for cohesion. Over time, these pores can be minimized, and the water layer is subsequently reduced.

AC C

EP

Pure cyclopentane hydrate phase equilibria were predicted using the statistical thermodynamic van der Waals and Platteeuw model with Kihara parameter optimization. Various combinations of Kihara parameters were obtained from the experimental data. Predictions of the sII binary (CP+H2O) hydrate system was shown to be in good agreement with experimental phase equilibria measurements. The optimized model was also able to predict cage occupancies of hydrate formers, showing occupancies of cyclopentane increase with increasing pressure.

5. Acknowledgements We would like to thank the current and past CSM Hydrate Consortium members for their support: BP, Chevron, ConocoPhillips, ENI, ExxonMobil, Halliburton, MultiChem, Nalco Champion, OneSubsea, Petrobras, Schlumberger, Shell, Statoil and Total. MNK acknowledges the Petroleum Institute, Abu Dhabi for support.

6. References 1

13

E.D. Sloan, and C.A. Koh, Clathrate Hydrates of Natural Gases CRC Press, Taylor & Francis Group, Boca Raton, FL. (2007)

ACCEPTED MANUSCRIPT

7

8

9

10

11

12

13

14

14

RI PT

SC

6

M AN U

5

TE D

4

EP

3

Z. M. Aman, E. P. Brown, E. D. Sloan, A. K. Sum, & C. A. Koh, Interfacial mechanisms governing cyclopentane clathrate hydrate adhesion/cohesion, Physical Chemistry Chemical Physics, 13 (2011) 19796-19806. C. Taylor, Adhesion force between hydrate particles and macroscopic investigation of hydrate film growth at the hydrocarbon/water interface, MS Thesis, Colorado School of Mines, (2006) S. R. Davies, E. D. Sloan, A. K. Sum, & C. A. Koh, In Situ Studies of the Mass Transfer Mechanism across a Methane Hydrate Film Using High-Resolution Confocal Raman Spectroscopy. The Journal of Physical Chemistry C, 114(2010) 1173–1180 Z. M. Aman, & C. A. Koh, Interfacial phenomena in gas hydrate systems. Chemical Society Reviews. 45(2016) 1678-1690 A. Döppenschmidt, & H. Butt, Measuring the thickness of the liquid-like layer on ice surfaces with atomic force microscopy. Langmuir, 21(2000) 6709–6714 M. N. Khan, L. J. Rovetto, C. J. Peters, E. D. Sloan, A. K. Sum, and C. A. Koh, Effect of Hydrogen-to-Methane Concentration Ratio on the Phase Equilibria of Quaternary Hydrate Systems. J. Chem. Eng. Data, 60 (2014) 418–423 Y. Matsumoto, R. G. Grim, N. M. Khan, T. Sugahara, K. Ohgaki, E. D. Sloan, C. A. Koh, and A. K. Sum, Investigating the thermodynamic stabilities of hydrogen and methane binary gas hydrates. The Journal of Physical Chemistry C, 118 (2014) 37833788 J. Platteeuw, and J. Van der Waals, Thermodynamic properties of gas hydrates. Molecular Physics, 1 (1958) 91-96 Á. Martín, A simplified van der Waals-Platteeuw model of clathrate hydrates with multiple occupancy of cavities. The Journal of Physical Chemistry B, 114 (2010) 9602-9607 J. Platteeuw, and J. Van der Waals, Thermodynamic properties of gas hydrates II: Phase equilibria in the system H2S‐C3H3‐H2O AT− 3° C. Recueil des Travaux Chimiques des Pays-Bas, 78 (1959) 126-133 R. Sakemoto, H. Sakamoto, K. Shiraiwa, R. Ohmura, & T. Uchida, Clathrate Hydrate Crystal Growth at the Seawater/Hydrophobic−Guest−Liquid Interface. Crystal Growth & Design, 10 (2010) 1296–1300 P. Skovborg, H. J. Ng, P. Rasmussen, & U. Mohn, Measurement of Induction Times for the Formation of Methane and Ethane Gas Hydrates. Chemical Engineering Science, 48 (1993) 445–453 T. A. Strobel, C. A. Koh, and E. D. Sloan, Thermodynamic predictions of various tetrahydrofuran and hydrogen clathrate hydrates. Fluid Phase Equilibria, 280 (2009) 61-67

AC C

2

ACCEPTED MANUSCRIPT

Highlights

RI PT

SC M AN U TE D EP

• •

Hydrate particle annealing time can be significant during shut-in condition in subsea oil/gas flowlines. Cyclopentane hydrate particle annealing effects on hydrate cohesion force is presented. Thermodynamic predictions of cyclopentane hydrate have been performed, with optimization of the Kihara parameters.

AC C

•