SPE-203762-MS

Using Difference Factor Thermodynamics Models in Hydrate Formation

Prediction During Surface Production Well Testing

Igoniwari Green, NPDC; Emenike Wami and Dorcas Jimmy, Rivers State University

Copyright 2020, Society of Petroleum Engineers

This paper was prepared for presentation at the Nigeria Annual International Conference and Exhibition originally scheduled to be held in Victoria Island, Lagos, Nigeria,
11 - 13 August 2020. Due to COVID-19 the physical event was not held. The official proceedings were published online on 11 August 2020.

This paper was selected and peer reviewed for presentation by an SPE program committee following review of information contained in an abstract and paper submitted
by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material
does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of
this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than
300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
All oil fields under water-drive, including gas-cap and depletion reservoirs may produce some water along
with oil. This co-production of water causes an additional set of challenges: corrosion, scale/salts deposition,
disposal of the water itself, and gas-hydrate formation. Hydrates are ice-like solids that form when free water
and natural gas combine at high pressure and low temperature. This can exist in gas and gas/condensate
wells, as well as in oil wells. Hydrate formationis a very serious challenge and can result in a huge loss
especially during production. It's also one of the major challenges encountered during surface production
well testing and it is controlled by several factors which include thermodynamics and kinetics of the system.
This study presents a thermodynamic model under operating field conditions, which centers on changes in
the Langmuir constant of the guest molecule(s) which is an important parameter in the hydrate formation
process. In this study, real-time production data acquisition system was used to acquire surface production
well testing data such as pressure, temperature, and choke size, from a hydrate well. The acquired real-time
data were used to determine the effect of the variation in choke size on Langmuir constant and therefore
on hydrate formation. Evaluating the Langmuir constant for large cavity under field conditions using a
generalized correlation, the Difference Factor method was developed and expanded to incorporate the choke
sizes variation. The correlation shows that Langmuir constant (decreases) is inversely proportional to the
choke size (increases) causing a decrease in the rate of hydrate formation. Due to an increase in choke size,
Langmuir constant approaches zero, therefore the product of fugacity and Langmuir constant which is the
driving force for hydrate formation approaches zero, leading to zero hydrate formation.
Keywords: Hydrate Formation, Langmuir constant, well testing, surface production, host molecules,
guest molecules, choke size, cavity, thermodynamics.

Introduction
Natural gas hydrates are solids that form when liquid water and natural gas are brought in contact at ambient
temperature and high pressure, typically 5 – 15°C and 50 – 100bar [1]. Such conditions are readily found
in natural gas industry and oil and gas production operations. As exploration and production operations
now concentrate on regions of deep-water (offshore) or lower temperature, more emphasis is being laid
2 SPE-203762-MS

on hydrates formation on production facilities such as separators, pipes, manifolds, etc., due to the hazard
hydrates impose on the personnel and facilities.
Surface production well testing is a vital aspect of well intervention activity that is routinely carried out in
the oil and gas industry for well's production optimization and enhancement with a view to maximizing the
bottom line. During this activity, valves, flow lines as well as the downstream of the test separator mainly
the gas line are usually plugged by hydrates. This is more pronounced in hydrates wells. This plugging has
constantly posed a serious flow assurance challenge to the industry.
Hydrate formation, a pressure – temperature dependent process, is thermodynamic. Thus, understanding
of the thermodynamics of hydrate formation is crutial if their hazards in the oil and gas industry are to be
minimized.
This study attempts to tackle this challenge via a novel thermodynamics approach of a combined
theoretical and practical – oriented method.

Procedure
Theory of the Thermodynamics of Hydrate Formation Models. Hydrates are crystalline cage-like
structure that forms when water reacts with gas (guest) molecules. Common gas hydrates found in the oil
and gas industry are composed of water and eight molecules of gases such as: methane, ethane, propane,
iso, butene, normal butane, nitrogen, carbon dioxide, and hydrogen sulphide [2]. Three types of gas hydrates
have been identified as structure I, structure II and structure H. Also, it has been found that the structure
I is the commonly formed hydrate during surface well testing. Resolving this challenges using choke size
variation based on thermodynamic principles became of interest.
The well-known Van der Waals – Platteeuw model for the prediction of gas hydrate stability, using
thermodynamic considerations were developed based on fundamental assumptions that every single cavity
cavity can considered as a perfect sphere containing at most one guest molecule that freely rotates within
the cavity and whose interactions with water molecules can be described by a pair of potential functions.
Also, it assumes that there are no interactions between guest molecules in different cavities, while the gas
molecules interact only with the nearest water molecules which do not distort the hydrate lattice [3].
Our earlier work [4] has shown that thermodynamics phase equilibrium equations, in terms of chemical
poteintial, fugacity and activity coefficients can be successfully applied to study the effect of Langmuir
constant variations on hydrate formation during surface production well testing.
Chemical Potential of Water in the Hydrate Lattice Model. The chemical potential of water in hydrate
lattice model, , is expressed in equation (1) below as[5, 6].

(1)

Where: represents the chemical potential of the water in the hypothetical empty hydrate lattice not
occupied by guest molecules. The hydrate structure with the lowest chemical potential of water (με) will
be stable.
vm is the number of cavities of type ‘m’ per water molecule in the lattice. The summation over species j is
necessary when these are multiple hydrate forming species [7].
For a pure component structure 1 (s1) hydrate system, the complete expression using equation 1 is;
(2)
Since structure 1 and 2 has 2 small cavities (2.512) and 6 large cavities (6.512 62) Cj,1 is the Langmuir
constant for a type j guest within the pentagonal dodecahedral cavity (small cage, 512), Cj,2 is the Langmuir
SPE-203762-MS 3

constant for j guest within the tetra Kaidecahedral cavity (large cavity, 512 62), and fj is the equilibrium
fugacity of a component in the multiphase mixture.
Equation 2 holds for a two-caged structure 1 hydrate. A ‘two-caged occupancy’ refers to structure 1
clathrate system in which the guest molecule occupies both large and small cavities during for ‘one-caged
occupancy’, the guest occupies only the small or large cavity [8]. Therefore, for a one-caged occupancy for
which it is the large cavity that is occupied, equation 2 reduces to:
(3)
Equation 1 can also be rewritten as[9, 10and 11]:

(4)

Providing θj,mv = the fractional occupancy of the jth guest molecule in the mth hydrate cavity.
Chemical Potential of Water in the Aqueous Liquid Phase Model. The chemical potential of water in
the aqueous liquid phase from classical thermodynamics, assuming ideal relationship for the water and
dissolved gas phase is[9, 5].

(5)

but
(6)

(7)

is the fugacity of water in the water-rich aqueous phase and is the fugacity of water in the reference
state, which is chosen to be a pure water phase at the T and P of the mixture; therefore;
(8)
xw The mole fraction of water in the water-rich phase and yw is the activity coefficient of water.
= the difference in the chemical potential of water between the hypothetical standard hydrate and
liquid water at a reference temperature T0 (273.15K) and zero absolute pressure; ΔHw and ΔVw are the
enthalpy and volume difference.
The Langmuir Constant Model. The Langmuir constants are temperature dependent functions that
describe the potential interaction between the encaged quest molecule and the water molecule surrounding
it and can be express as[12, 13]:

(9)

Providing:
Aj,i and Bj,i are experimentally fit parameters, and are dependent on which guest molecule is present.
Fractional Occupancy Model. The fractional occupancy can be given in a Langmuir type manner as [14]:

(10)
4 SPE-203762-MS

Fugacity Model Using Peng Robinson Equation of State. The guest species fugacity can also be calculated
using a suitable equation of state, such as the Peng Robinson equation of state. The generalized form of
the Peng Robinson is [15].

(11)

Providing:

(12)

(13)

(14)

(15)
ε= acentric factor,
Tc = critical temperature,
PC = critical pressure
Equation 11 can be written as [14]:
(16)
Providing:

(17)

(18)

(19)
Z is the compressibility factor, ‘a’ is the interaction (or the energetic) parameter and ‘b’ the molar co-
volume [16].
When Peng Robinson equation of state is applied to fugacity calculation, the resulting equation is:

(20)

Providing:
Also, the fugacity, fj,l, at the various pressures and temperatures on the different choke sizes utilized
for this work were obtained using the program for thermodynamic properties calculation using the Peng
Robinson cubic equation of state developed by Sandler [15]. The various methane parameters used for the
fugacity computation were those found in Abboth and Vanness [17] and Sandler [15].

Methodology
Application of Thermodynamics Models to Well Test Data and Hydrate Phase Equilibrium. Clathrate
hydrates phase equilibrium can be described by:
(21)
Therefore,
SPE-203762-MS 5

(22)
Providing:
μwβ − L or S is the chemical potential of water in the solid phase (S) or liquid phase (L) at the given
temperature and pressure, depending on the stable phase.
From equations 5, 6 and 22, we have;

(23)

(24)

Equilibrium, ΔμH = ΔμL; hydrate can exist; that is:

(25)

For this study, assumptions such as the hydrates formed are structure 1 (sI) single hydrate of methane
species.
Using the mole fraction of water in the liquid phase, the last term of equation 25 which accounts for the
effect of dissolved gas (methane) is removed by assuming that the testing of the well is taking place in a
well of known basic sediment and water (BS&W), therefore, the concentration of water is very close to
unity (xw = 1) in water-rich solutions and because the solubility of methane in water is low; hence aw = 1.
The first two terms on the right-hand side of equation 25 represents ΔμL (T, P =0), the chemical potential
difference at a fixed temperature, and zero pressure.
From equation 24 when aw = 1 or xw =1

(26)

Hydrate is given by literature as: Jmol−1 if taking the following

thermodynamic reference state properties of structure1 (sI)
Jmol−1 and ΔVw = 3.0 × 10−6m3/Jmol. The ollowing steps were taken to develop
the excel program.
1.
Using equation 26, the various values of at the various temperatures and pressures were
obtained while testing the hydrate well that was calculated.
2.
At equilibrium, hence the various values of obtained from step 1 were
equated to equation 23
3. Since methane hydrate is a two-cage occupancy structure, the following equation holds:
(27)
4. Putting and in equation 27 and assuming θj,s =0.920 from literature, the values of θj,l
were obtained.
5. Using the Sandler's software, various fugacity values were obtained in equation 10 for field data.
6. By substituting the various values of the fugacity obtained from step 5, the values of the Langmuir
constant field data using equation 10 were obtained.
6 SPE-203762-MS

7.
Calculating the various field data Langmuir constant, , to equation 9, yeilds

(28)
8. Taking the natural log and re-arranging, yeilds

(29)
9.
Equation 29 has as the slope and as the intercept.

10. Using literature values of 0.02335 and 2653 for Aj,l and Bj,l for large cavity of methane hydrate, the
various values of the Langmuir constant for pure methane at various operating field conditions were
calculated using equation 30.

(30)

Which gives equation (31) on rearrangement as

(31)

Model Development. Plotting the values of Ln (T × Cij) T−1 for pure methane and those of field gases as
guest molecules respectively we obtain the values (Bi,j) and (Aj,i) for slope and intercept of the line. The
magnitude of these differences at various temperatures were calculated and a factor, called difference factor
method was used to obtain a general expression that could enable the prediction of the Langmuir constant
during well testing provided its values for pure methane are known under field conditions. The same linear
profiles was obtained in our previous study which developed similar model for hydrate formation using a
Ratio factor method [4].
The Difference Factor Method. Subtracting equation (29) from (31), we have;
(32)

(33)

(34)
Applying equation 32 to the various choke size with a view to incorporating the various choke size, s,
into the generalized proposed modified model (equation 34) gives,
(35)
Providing λs is the difference factor for every choke size and is defined as:

(36)
Assuming the line of best fit to be f(s) + k in a plot of difference factor versus choke size
(37)
Putting equation 37 into equation 35 gives
SPE-203762-MS 7

(38)

Results and Discussion

Slope and Intercept Determination from Field Data. From the excel program, a natural log plot of the
product of temperature and Langmuir constant for field data versus the reciprocal of temperature, Figure
1, as shown below yield as the slope and as the intercept gave Aj,l for

field data equals 0.06227.

Figure 1—Variation of Natural Log Product of Temperature and Langmuir

Constant (Large Cavity) for Field Data with Reciprocal of Temperature

The intercept and slope values found in the literature are 2653 and 0.02335 respectively. These values
are empirically determined values that depend on the nature of the gas molecules (guest) that occupies the
hydrate cavities.
The Aj,l and Bj,l derived from this study differs from the Aj,l and Bj,l for large cavity of structure 1
hydrate occupied by pure methane found in the literature. The difference in Aj,l and Bj,l shows that the
guest component(s) of the hydrate formed during testing of the well is not pure, but rather a mixture of
component(s), that is, methane and other hydrocarbons. Also, the host molecule of the hydrate formed in
the field is not pure water, but rather a water with some dissolved salts. The presence of non-hydrocarbon
gases such as CO2 also affects the values of Aj,l and Bj,l hence the Langmuir constant.
The Difference Factor Determination. From the plot of the natural log of the product of the reciprocal of
temperature obtained by using the values from equation 29 versus the Langmuir constant and temperature
for pure methane as shown in figure 2, and on comparison with the plot of the natural log of the product of
the reciprocal of temperature versus Langmuir constant field data and temperature, figure 3, the difference
factor equals 1.42 as shown in Table 1 above and confirmed by the perfect fit of the cross plot in figure
4 which has slope of 1.
8 SPE-203762-MS

Figure 2—Comparing of Ln [T × Cj,l] for Pure Methane and Field Data of Temperature Reciprocal

Figure 3—Normalization of Ln [T × Cj,l]FD (Large Cavity) by Adding 1.4201, the Average Difference (Δλ)
SPE-203762-MS 9

Figure 4—Cross plot of Ln [T × Cj,l] PM versus Ln [T × Cj,l] FD + (Δλ)

Table 1—The Difference Factor evaluation from the Plots of Methane Hydrates using
Aj,l and Bj,l for pure Methane and Field Data within the Tested Temperature Range.

S/N 1/T(k−1)
(k/ (k/bar)
(K/
bar)

bar)

1. 0.0033 4.9977 3.6310 1.3667

2. 0.00335 5.1304 3.7281 1.4023

3. 0.0034 5.2630 3.8251 1.4379

4. 0.00345 5.3957 3.9222 1.4732

Average = 1.4201

(39)

Also from Table 1 and Figure 3, it could be deduced that within the field hydrate formation temperature
range of 289K to 303K and pressure between 104bar and 235bar, the Langmuir constant for pure methane
occupying the large cavity of a structure 1 (sl) simple methane hydrate is about 4 times that of the field data.
From table 1 above, it shows that within the field hydrate formation temperature range of 289K to 303K
and pressure between 104bar and 235bar, the below equation holds.
(40)

Also, as the temperature decreases, increases.

Specific Correlation. Table 2 above and Figure 5 below, and on comparison by error, the analysis gives
a polynomial relationship as the line of best fit as shown below.
10 SPE-203762-MS

Figure 5—Plot of the Difference Factor versus Choke Size, s.

Table 2—Average Values of the Difference Factor for the Various Choke Sizes Tested

S/No. Choke Size (1/16’) λ(k/bar) Remark (s)

1 12 0.8430 Test on 12/64’ Choke

2 16 (1) 1.0403 First Test on 16/64’ Choke

3 20 1.3372 Test on 20/64’ Choke

4 24 (1) 1.5051 First Test on 24/64’ Choke

5 32 2.3399 Test on 32/64’ Choke

6 40 1.4457 Test on 40/64’ Choke

7 44 1.3698 Test on 44/64’ Choke

8 36 1.4570 Test on 36/64’ Choke

9 24 (2) 1.5330 Second Test on 24/64’ Choke

10 16 (2) 1.5200 Second Test on 16/64’ Choke

Note: Test on various choke sizes are based on operator's (Client) decision.

The line of best fit equals, Δλ = - 0.0026S2 + 0.1630S - 0.7310 from Figure 6 below. Aalysis of Figure
5 and Table 2, show that the Langmuir constant for field data from this study, when applied to the various
choke sizes based on difference factor, is;
(41)
SPE-203762-MS 11

Figure 6—Plot of Variation of Difference Factor for for every Choke Size with Polynomial as Best Fit Line

From the above model, the Langmuir constant field data for large cavity equals Langmuir constant pure
methane multiply by a negative polynomial exponential.
Model Comparison. When the values of Langmuir constants from field data using the existing equation
from literature, from Table 3, were compared with the generalized proposed model, equations 35, it gives a
better approximation based on error analysis. Also, equation 41 is the preferred specific correlation model
involving various sizes, gotten from the proposed specific correlation model.

Table 3—Values of Aj,l and Bj,l for the Various Choke Sizes Tested

Choke size (1/64’) Aj,l Bj,l

12 5.1296*10 −12 9147.8

16(1) 1.80615 1039.6

20 38638.3 -1999.7

24(1) 5999.9 -1436.7

32 1.51892 1000.9

40 0.25684 1536.5

44 0.01209 2445.6

36 4.5234*10 −9 6746.6

24(2) 0.43089 1333

16(2) 0.22773 1509.1

Variation of Pressure with Choke Sizes. On chokes 12, 16(1), 20 and 24(1), from Figure 6 below, the
pressure-choke size data are not inversely proportional, hence it can be said that the well may be undergoing
a clean-up operation as this cndition is normally associated with clean-up operations. Therefore during
clean-up operations, impurities are higher, which in turn affects Aj,l and Bj,l values. Beside the first four choke
12 SPE-203762-MS

sizes tested, that is, during the cleanup operation, from choke 32/64’, the relationship between pressure and
choke size became inverse, which is in conforms with literature.
Variation of Pressure with Temperature for for every Choke Size. When the average pressures and
temperatures for the respective choke sizes were taken as shown in Table 4 above and a plot of pressure
against temperature made, Figure 7 below, the graph has a negative slope which differs from plots of
pure hydrate forming components that have positive slopes. This may be attributed to the azeotropic effect
confirming that hydrate formation during the well testing forms from a mixture of components.

Figure 7—Variation of Well Head Pressure with Choke Sizes at Different Points

Figure 8—Variation of Average Well Pressure with Average Temperature For every Choke Size
SPE-203762-MS 13

Table 4—Average Pressure and Temperature for every Choke Size

S/N Choke size (1/64’) Pressure (bar) Temperature (K)

1 12 106.7 303.5

2 16(1) 128.4 298.3

3 20 183.7 296.2

4 24(1) 202.9 292.6

5 32 193.3 290

6 40 177.3 291

7 44 164.9 291.9

8 36 185.8 292.4

9 24(2) 221.9 294.3

10 16(2) 233.8 298.6

Conclusions
The following conclusions can be drawn from the study,:
1. Under field conditions, the componets forming hydrates are never pure methane but rather mixtures
of components.
2. Under hydrate forming conditions, the values of Langmuir constant for a simple hydrate, when the
large hydrate cavity is assumed to be occupied by pure methane, is four times greater its field data
values for hydrate cavity occupied by the guest molecule(s) that forms the hydrate.
3. Generalized correlation of the form , was developed using difference factor
method which reflexes the changes in Langmuir Constant values when the hydrate large cavity was
assumed to be occupied by pure methane as guest molecule, relative to the real field conditions in
which the guess molecules are not methane only.
4. A new correlation of the form - , was also
developed which took into account the effect of choke size variations on Langmuir constant during
sufface production well testing, using real-time pressure, temperature and choke size data.
5. Choke size is inversely proportional to Langmuir constant at a given pressure and temperature.
6. A decrease in Langmuir constant leads to a decrease in hydrate formation.
7. Variation in choke sizes during surface production well testing is an important opearating criterion in
hydrate formation studies which shows that at higher choke sizes the possibility of hydrate forming
and thus plugging testing facility is reduced. This is done, mindful of the reservoir management.

References
1. M. Monk and J. S. Gudmundson (2001): ‘Hydrate Formation Rate in a Continuous Stirred
Tank Reactor (CSTR),’ 2001 International Gas Research Conference, Amsterdam, Neverlands.
November 5 – 8. Available: http://www.ipt.ntnu.no/-ngh/library/study10/study10.htm December
18, 2004.
2. E. D. Sloan (1991): ‘Natural gas hydrates,’ SPE Colorado School of Mines, JPT, December, 1414
– 1417.
3. B. U. Haq (1999): ‘Methane hydrate research and development,’ [online] Available: http://
www.nsf.gov/about/congress/106/106-hearings-first. [September 20, 2004].
14 SPE-203762-MS

4. Green, I. S., Jimmy, D. E. and Wami, E. N. "Application of Thermodynamic Models to Hydrate

Formation Prediction during Surface Production Well Testing Using the Ratio factor Method".
International Journal of Scientific & Engineering Research, Volume 6, Issue 10, Oct., (2015)
5. J. H. Vander-Waals and J. C. Platteeuw (1965): ‘Clathrate solutions,’ Adv. Ind. Phys. Chem., 2.
1–57.
6. S. O. Yang, Y. S. Kim, K. S. Ryu and C. S. Lee (2001): ‘LW – H equilibrium measurements
and unified predictions of hydrate-containing phase equilibria for methane, ethane, propane
and their mixtures,’ [online]. Available: http://infosys.korea.ac.kr/ippage/g/ipdata/2001/04/file/
g20010420.ppt [October 8, 2005].
7. G. D. Holder, L. P. Makka and R. P. Warzinski (2002): ‘Formation of gas hydrates
from single-phase aqueous solution,’ [online]. Available: http://www.netl.doe.gov/osta/
techstudys/2002-574.pdf [June 5, 2005].
8. B. Kyamme (1994): ‘Mechanisms for initiation of hydrate from liquid water, liquid phase
cluttering, surface adsorption or what?’ Annals of the New York Academy of Science, 715,
306–310.
9. A. L. Ballard and E. D. Sloan (2002): ‘The next generation of hydrate prediction: An
overview,’ [online] Available: http://www.mines.edu/research/chs/Ballard/.pdf [August 18, 2004].
10. ‘Vijay John's Research group: ‘Gas Hydrates’ [online] Available: http://www.tulane.edu/ceng/
faculty/john/website/gashydrates.htm [September 29, 2004].
11. Z. Cao, J. W. Tester, K. A. Sparks and B. L. Trout (2001): ‘Molecular Computations using robust
hydrocarbon – water potentials for predicting hydrate phase equilibrium,’ J. Phys. Chem. B, 105,
10950–10960 [online]. Available: http://troutgroup.mit.edu/studys/jpo12292b.pdf [May 2, 2005]
12. J. Munck, S. Skjoid-Jorgensen and P. Rasmussen (1998): ‘Computations of the formation of gas
hydrates,’ Chemical Engineering Science,pp. 43, 2661 – 26672.
13. D. Smith, K. Seshadri and W. J. Wilder (2004): ‘Assessing the thermodynamic feasibility of the
conversion of methane hydrate into carbon dioxide hydrate in porous media,’ [online]. Available:
http://www.searchanddiscovery.com/documents/abstracts/2004/hedberg_vancouver/extended/
mcgrail.htm [May 20, 2004].
14. C. K. Hester and E. D. Sloan (2003): ‘sll Structural transition from binary mixtures of sample sl
formers’, A study presented at the fifteenth symposium on thermo physical properties, Boulder,
Colorado, USA, June 22 – 27, 2003.
15. S. I. Sandler (1999): ‘Chemical and Engineering thermodynamics,’ 3rd Edition, John Wiley &
Sons. USA, 772.
16. A. Chapoy (2002): ‘Phase behaviour in water/hydrocarbon mixtures involved in gas
production systems’. [Online] Available: http://pastel.parishtech.org/achieve/0000/202/01/
these_Antonin2.0pdf [September 16, 2005].
17. M. M. Abbott and C. H. Vanness (1976): ‘Schaum's outline series theory and challengess of
thermodynamics,’ SI (metric) edition, McGraw-Hill, Britain, 344.