Journal of Petroleum Science and Engineering 48 (2005) 70 80

www.elsevier.com/locate/petrol

A general correlation for predicting the suppression of

hydrate dissociation temperature in the presence of
thermodynamic inhibitors
Kasper K. stergaard1, Rahim Masoudi2, Bahman Tohidi*, Ali Danesh, Adrian C. Todd
Centre for Gas Hydrate Research, Institute of Petroleum Engineering, Heriot-Watt University, Edinburgh, EH14 4AS, UK
Received 27 August 2004; received in revised form 5 April 2005; accepted 12 April 2005

Abstract
Thermodynamic inhibitors can be used in avoiding gas hydrate problems in natural gas and oil production and transportation. A new correlation for predicting the suppression of the hydrate dissociation temperature of various petroleum fluids in
the presence of different thermodynamic inhibitors (e.g., electrolytes and organic inhibitors) is presented. The correlation, which
can be applied to drilling and reservoir fluids, requires inhibitor concentration, pressure of the system and, if known, the
dissociation pressure of the fluid in the presence of distilled water at 273.15 K.
The correlation has been developed using hydrate phase boundaries for methane and seven natural gases in the presence of
various single inhibitors generated by a well-proven comprehensive thermodynamic model. It covers nine electrolyte solutions
and six organic inhibitors, which may be available and/or added to the reservoir and drilling fluids. The developed correlation is
generic and could be easily extended to other salts and organic inhibitors.
D 2005 Elsevier B.V. All rights reserved.
Keywords: Gas hydrate; Suppression temperature; Inhibition effect; Salt; Organic inhibitor; Correlation

1. Introduction

* Corresponding author. Tel.: +44 131 451 3672; fax: +44 131
451 3127.
E-mail address: bahman.tohidi@pet.hw.ac.uk (B. Tohidi).
1
Present address: DONG E&P (Danish Oil and Natural Gas),
DONG, Agern Alle 24-26, 2970 Hoersholm, Denmark.
2
Present address: NIOC R&D, No. 27, Keshavarz Boulvar,
Valiasr Square, Tehran, Iran.
0920-4105/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.petrol.2005.04.002

Gas hydrates are a serious concern in oil and gas

exploration and production due to potential blockage
of wells, flowlines and other process facilities. To
control the risk of gas hydrate formation in drilling
and production operations, it is common to add thermodynamic inhibitors (e.g., electrolytes and organic
inhibitors) to the fluid systems. Accurate knowledge
of the thermodynamic stability of hydrates as a function of concentrations of inhibitors is crucial to the

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

success of any flow assurance strategy. Considerable

efforts are therefore put into the development of
methods and tools, capable of predicting hydrate
phase boundaries for different fluids at various temperature and pressure conditions.
Since 1945, the gas gravity method given by Katz
(1945) has been an indispensable and simple tool for
predicting the gas hydrate stability zone. Despite the
development of more sophisticated predictive tools,
such as the vapoursolid equilibrium ratio (K i value)
method (Carson and Katz, 1942; Wilcox et al., 1941)
or the solid solution theory of Van der Waals and
Platteeuw (1959), the gas gravity method has kept
its popularity among engineers in the petroleum industry. The main advantage of this technique is the
availability of input data (it only requires the specific
gravity of the mixture, i.e., the molecular mass of the
gas mixture divided by that of air) and the simplicity
of the calculation, which can be performed by using
charts or hand-held calculators. The original gas gravity method is only applicable to dry gas systems.
However, the development of offshore/arctic oil and
gas-condensate fields necessitated a robust and simple
method for predicting the hydrate stability zone for
these systems.
stergaard et al. (2000) developed a method similar to the gas gravity method, applicable to all reservoir fluids, in the presence of distilled water, from
natural gas to black oil, which only requires information on the specific gravity and the concentration of
the hydrate forming components in the system. The
method can take into account the effect of non-hydrocarbon gases (i.e., CO2 and N2) in the petroleum
fluid.3
There also exist simple tools for estimating the
hydrate inhibition effect of salts and organic inhibitors. The most famous one is the Hammerschmidt
(1934) equation, but also equations by Nielsen and
Bucklin (1983) and Yousif and Young (1993) are
available. These equations calculate the suppression
of the hydrate dissociation temperature, DT, compared to that of distilled water (i.e., the hydrate
phase boundary in the presence of distilled water
should be determined separately). Inherently, they

3
A free copy of the spreadsheet is available at the Hydrate
Groups website: www.pet.hw.ac.uk/research/hydrate.

71

have the following simplifying assumptions, reducing their accuracy:

DT is independent of the system pressure.
DT is independent of gas/oil composition and hydrate structure.
DT is independent of the inhibitor.
In addition to the above shortcomings, to our
knowledge, there is no reported general correlation
in the open literature that is capable of estimating the
hydrate inhibition effects of various salts and organic
inhibitors, which might be available and/or added to
the reservoir and drilling fluids.
In this communication, first, the effective parameters, which are important to be taken into account
in developing any reliable hydrate inhibition correlation, are discussed. A general equation, then, is presented which is capable of predicting hydrate
inhibition effect of various salts and organic inhibitors. The correlation relates the suppression of hydrate
dissociation temperature of hydrocarbon fluids to the
concentration of inhibitor, hydrate dissociation pressure of the fluid in the presence of distilled water and
pressure of the system.
The predictions of the developed correlation have
been compared with experimental data, predictions of
a comprehensive thermodynamic model and some
other correlations reported in the literature. The results
demonstrate the reliability of the developed correlation, despite its relative simplicity.

2. Effect of important parameters on the hydrate

inhibition characteristics of aqueous solutions of
salts and/or organic inhibitor
In order to develop a reliable, yet simple, correlation for predicting the gas hydrate inhibition there are
some difficulties (such as inaccurate experimental
data and thermodynamic models, pressure effect, hydrocarbon fluid composition, inhibitor effect and ionisation effect), which should be considered in a
comprehensive correlation. In the followings, the importance of the above parameters on the hydrate prevention and the shortcomings of the existing
correlations are discussed, on the basis of which the
new correlation is developed.

72

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

100
Distilled Water
Mud A
Mud B
Mud C
Mud D

P/MPa

15.5 K @ 15 MPA

Mud C

10
12 K @ 5.5 MPA

Mud C

1
268

273

278

283

288

293

T/K
Fig. 1. Effect of pressure on the inhibition effect of inhibitors in drilling fluids on methane hydrates.

2.1. Effect of pressure

Common to all the available correlations is the fact
that the suppression of the hydrate dissociation temperature is only assumed to be a function of the
inhibitor concentration. However, other parameters
may also influence the hydrate suppression, e.g., the
pressure of the system. This is illustrated in Fig. 1
where the hydrate phase boundaries of methane
hydrates are shown in the presence of pure water
and four different water based drilling fluids (muds)
with added hydrate inhibitors. The hydrate phase

boundaries for these muds were previously measured

in this laboratory. The pressure is seen to have an
important effect on the suppression of the hydrate
dissociation temperature, DT. As an example, at 15
MPa, the DT of mud C is 15.5 K, whereas at 5.5 MPa
the DT of the same mud is only 12 K.
2.2. Effect of hydrocarbon fluid composition
Fig. 2 shows the effect of a drilling fluid with added
hydrate inhibitors on the hydrate free zones of methane
(structure I hydrate) and a natural gas (structure II

Distilled water & C1

Distilled water & NG
Mud 1 & C1
Mud 1 & NG

P/MPa

C1

10

1
267

NG

272

277

282

287

292

T/K
Fig. 2. Inhibition effect of a drilling fluid on hydrate of methane and natural gas.

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

hydrate). The hydrate phase boundaries were previously measured in this laboratory. The effect of pressure, as
discussed in the previous section, is obvious. Furthermore, at a given pressure, the temperature suppression
by the drilling fluid is different for methane hydrates
compared to hydrates formed from natural gas. For
example, at 5.5 MPa, the DT in the presence of mud
1 is 13 K for the natural gas but 14.3 K for methane.
This effect has not been considered by any of the
available correlations.
2.3. Effect of inhibitor
According to the available correlations, the hydrate
inhibition characteristics of organic inhibitors (e.g.,
methanol and glycol) is independent of the type of
inhibitor at similar molar concentrations. This is a
reasonable assumption at low inhibitor concentrations
as there is a negligible difference between different
inhibitors, but as the concentration increases notable
differences in the degree of suppression by different
inhibitors become apparent. This is illustrated in
Fig. 3 where the temperature suppression of methane
hydrate formation is shown at 14.66 MPa for methanol and ethylene glycol. The constant pressure experimental data have been extracted from (Ng et al.,
1987; Robinson and Ng, 1986; Ng and Robinson,
1985). As seen in the figure, at inhibitor concentrations below about 10 mol%, the degree of inhibition

caused by the two compounds is similar. At higher

concentrations, however, a clear distinction between
the two curves is observed.
2.4. Effect of ionisation
In order to include non-associated compounds,
such as electrolytes, in the correlations available in
the literature, both Hammerschmidt (1934) and Yousif and Young (1993) considered the degree of ionisation and the number of ions formed from each
molecule of electrolyte. The degree of ionisation is
a strong function of the concentration of the salt at
the same time showing significant individual characteristics from salt to salt. This is illustrated in
Fig. 4, which shows the degree of ionisation of
NaCl and CaCl2 as function of concentration at
293.15 K as presented by Yousif and Young
(1993). Furthermore, the degree of ionisation is
also a function of the system temperature and cannot
be satisfactory represented by assuming fixed temperature (Robinson and Stokes, 1959). Fig. 5 shows
the hydrate temperature suppressions of a natural gas
by sodium chloride, calcium chloride and methanol
as a function of inhibitor mole fraction (predicted by
the in-house thermodynamic model, HWHYD). The
mole fraction of the electrolytes (X*) has been corrected for the degree of ionisation and number of
ions according to Hammerschmidt correction (1934)

25
Methanol (Ng & Robinson, 1985; Robinson & Ng, 1986)
Ethylene glycol, (Robinson & Ng, 1986)

20

dT/K

15

10

5
P = 14.66 MPa

0
0

73

10

15

20

25

mole%
Fig. 3. Inhibition of methane hydrates at 14.66 MPa by methanol and ethylene glycol.

74

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

NaCl
CaCl2

0.9

degree of ionisation

T = 293.15K

0.8
0.7
0.6
0.5
0.4
0

10

15

20

25

30

35

40

wt%
Fig. 4. Degree of ionisation of NaCl and CaCl2 at 293.15 K as function of concentration.

and Fig. 4. Although DT seems to be inhibitor

independent at lower concentrations, this is clearly
not the case at higher inhibitor concentrations.

3. Construction of the new correlation

A new correlation could be developed either based
on the available experimental data in open literature or
the results of a thermodynamic model. In this work,
the predictions of a thermodynamic model (pseudo
experimental data) are used in developing the new

correlation rather than real experimental data for the

following reasons.
Firstly, because the amounts of experimental hydrate dissociation data in the presence of electrolytes
and organic inhibitors is very limited (Amy et al.,
2002; Matthews et al., 2002). Secondly, because of
the limited experimental data, any error could easily
result in unreliable correlation. This approach has
previously been applied with success when developing correlations for hydrate phase boundaries in the
presence of water as well as ice (stergaard et al.,
2000; stergaard and Tohidi, 2002).

60
NaCl

50

CaCl2
Methanol

dT/K

40
30
20
10
P = 30 MPa

0
0

0.1

0.2

0.3

0.4

0.5

0.6

x*
Fig. 5. Predicted hydrate temperature suppression of a natural gas by sodium chloride, calcium chloride and methanol as function of inhibitor
mole fraction.

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080
Table 1
Origin of natural gases used in the optimisation of the correlation
NG1
NG2
NG3
NG4
NG5
NG6
NG7

(Tohidi et al., 1996a),

Table 1 (Natural gas)
(Katz et al., 1959),
Tables 15 (Leduc)
(Katz et al., 1959),
Tables 15 (Viking)
(Katz et al., 1959),
Tables 15 (West Cameron)
(Katz et al., 1959),
Tables 15 (Deep Lake)
(Katz et al., 1959),
Tables 58 (4-H)
(Katz et al., 1959),
Tables 58 (6-K)

North Sea
Natural Gas
Natural gas
from Alberta
Natural gas
from Alberta
Natural gas
from Louisiana
Natural gas
from Louisiana
Natural gas
Natural gas

The task for developing the new correlation can be

divided into three steps:
! Selecting appropriate systems and generating the
necessary data (pseudo experimental).
! Identifying the appropriate numerical expression and
minimising the deviation between the pseudo experimental data in the database and the predictions.
! Validation and examining the reliability of the
correlation by comparing the predictions against

75

real experimental data and the results of other

predictive tools.
In the first step, the Heriot-Watt University Hydrate
Program, HWHYD model, was used to generate the
required data on hydrate phase boundaries in the
presence of distilled water, salts and organic inhibitors
for methane and for the seven natural gas systems,
presented in Table 1. It should be noted that the
HWHYD model has been already validated against
independent hydrate phase boundary data on the range
of various inhibitor concentrations used in this study.
The HWHYD comprehensive thermodynamic
model uses the Valderrama modification of the Patel
and Teja equation of state (VPT EoS) for fugacity
calculations in all fluid phases (Valderrama, 1990).
Non-density dependent (NDD) mixing rules are applied to model polarnonpolar and polarpolar interaction (Avlonitis et al., 1994). The hydrate phase is
modelled by using the solid solution theory of Van der
Waals and Platteeuw (1959), as implemented by Parrish and Prausnitz (1972). The Kihara model for
spherical molecules is applied to calculate the potential functions for compounds forming the hydrate
phase (Kihara, 1953). A detailed description of the

Table 2
Constants in Eq. (1) for various salts
Constant

NaCl

CaCl2

NaBr

HCOONa

HCOOK

C1
C2
C3
C4
C5
C6
Max. con. (mass%)
Max. con. (mol%)
AAD%a

0.3534
1.375d 10
3
2.433d 10
4
4.056d 10
2
0.7994
2.25d 10
5
26.5
10
1.17

0.194
7.58d 10
3
1.953d 10
4
4.253d 10
2
1.023
2.8d 10
5
40.6
10
1.39

0.419
6.5d 10
3
1.098d 10
4
2.529d 10
2
0.303
2.46d 10
5
38.8
10
1.33

0.605
9.39d 10
3
1.34d 10
5
2.657d 10
2
0.353
2.47d 10
5
29.5
10
1.28

4.43
3.9d 10
2
1.766d 10
3
2.86d 10
3
4.094d 10
2
2.6d 10
5
34.2
10
1.81

Constant

HCOOCs

K2CO3

KBr

KCl

C1
C2
C3
C4
C5
C6
Max. con. (mass%)
Max. con. (mol%)
AAD%a

2.41
2.02 d 10
2
9.81 d 10
4
2.49 d 10
3
3.5524 d 10
2
2.92 d 10
5
52.3
10
1.70

0.1837

5.7 d 10
3
2.551 d 10
4
6.917 d 10
2
1.101
2.71 d 10
5
40.0
8
1.31

0.3406
7.8 d 10
4
8.22 d 10
5
3.014 d 10
2
0.3486
2.3 d 10
5
36.5
8
0.96

0.305
6.77 d 10
4
8.096 d 10
5
3.858 d 10
2
0.714
2.2 d 10
5
31.5
10
1.08

Average absolute percentage deviation, defined by Eq. (2).

76

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

numerical modelling is given elsewhere (Tohidi et al.,

1995). The HWHYD thermodynamic model has proven reliable in predicting the hydrate free zone of various
fluid systems (Tohidi et al., 1996a,b, 1997a,b; stergaard et al., 1998).
The hydrate dissociation data, which were used in
developing the correlation, was from a wide range of
pressure (up to 150 MPa), salt concentration (up to
saturation point) and organic inhibitor concentration
(up to 30 mol%). All these data were used as input
in a multi-dimension regression procedure, utilising
least square optimisation routines. The objective
function was defined as the relative error between
the pseudo experimental suppression of the hydrate
dissociation temperatures and those calculated by the
tested equations.
The resulting equation, which includes inhibitor
concentration, the pressure of the system and the
locus of the hydrocarbon fluid in the presence of
the pure water, is given by the following expression:



DT C1 W C2 W 2 C3 W 3 C4 ln P C5
 C6 P0
1000 1
1
where:
DT
P
W
P0

Ci

Suppression of the hydrate dissociation

temperature (in K or 8C)
Pressure of the system (in kPa)
Concentration of the inhibitor in liquid water
phase (in mass%)
Dissociation pressure of hydrocarbon fluid in
the presence of distilled water at 273.15 K
(in kPa)
Constants

The constants C i are given in Tables 2 and 3 for

various salts and organic inhibitors, respectively. P 0
for the above correlation may be available in form
of experimental data, or can be calculated by most
available thermodynamic models, or can be estimated using gas gravity methods such as the original
Katz (1945) method or the correlation developed
for all reservoir fluids by stergaard et al. (2000).
The last term of Eq. (1) (C 6( P 0
1000) + 1)
describes the effect of the locus of the hydrate phase
boundary of the reservoir fluid in the presence of
distilled water. If not known, however, this term can

Table 3
Constants in Eq. (1) for various organic inhibitors
Constant

Methanol

Ethanol

Glycerol

C1
0.478
1.118
0.135
C2
7.17 d 10
2
4.48 d 10
3
8.846 d 10
3

5

4
C3

1.44 d 10
6.979 d 10
1.15 d 10
5
C4
2.947 d 10
2 5.85 d 10
3
1.335 d 10
2
C5
0.596
0.225
0.378
C6
3.1 d 10
5
3.4 d 10
5
4.6 d 10
5
Max. con. (mass%) 43.3
31.2
68.6
Max. con. (mol%) 30
15
30
AAD%a
1.22
1.80
4.86
Constant

Ethylene
glycol

C1
C2
C3
C4
C5
C6
Max. con. (mass%)
Max. con. (mol%)
AAD%a

38.93
0.343
0.1964

0.522

3.47 d 10
3
5.81 d 10
3
1.767 d 10
2 2.044 d 10
4 1.393 d 10
4
3.503 d 10
4 1.8 d 10
2
2.855 d 10
2
5.083 d 10
3 0.3346
0.854
2.65 d 10
5
2.74 d 10
5
3.24 d 10
5
59.6
51
59.5
30
15
15
1.29
1.35
1.70

DEG

TEG

Average absolute percentage deviation, defined by Eq. (2).

be simply omitted with limited effect on the results, as

it has been shown for a selected system including
various concentrations of triethylene glycol (TEG) in
Fig. 6. It should be mentioned that the developed
correlation is applicable over the range of inhibitor
concentration reported in Tables 2 and 3 and pressures
up to 100150 MPa.

4. Validation of the developed correlation

The newly developed correlation enables fast and
accurate determination of the hydrate free zone of
various fluid systems in the presence of thermodynamic inhibitors. For validation purposes, the correlation was compared with the predictions of the
comprehensive thermodynamic model, HWHYD,
comparing the results of methane and the seven natural gas systems (Table 1) for each individual inhibitor used in the optimisation, as well as the available
correlations in open literature.
The average absolute percentage deviation
(AAD%), as defined below, was used to evaluate the
reliability of the developed correlation in predicting

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

77

100
40 mass% TEG

P/MPa

20 mass% TEG

Distilled water

10

Thermodynamic model, Distilled water

Developed correlation, with P0 term
Developed correlation, without P0 term

1
270

275

280

285

290

295

T/K
Fig. 6. Evaluation of the effect of P 0 term (last term) in Eq. (1) by predicting methane hydrate phase boundary in the presence of various
concentrations of TEG.

the hydrate free zone of studied hydrocarbon fluids in

the presence of various thermodynamic inhibitors:

n 

1 X
 DTi;model
DTi;corr   100
AAD%
2


n i
1
DTi;model
where n is the number of data points, and DTi,model
and DTi,corr are suppression of the hydrate dissociation temperatures calculated by the HWHYD thermodynamic model and the developed correlation,
respectively.

The values of AAD% for the studied salts and

organic inhibitors are presented in Tables 2 and 3.
As seen in the tables, the agreement between the
predictions of the thermodynamic model and the developed correlation is satisfactory, with typical AAD
values between 1% and 2%. Only, for glycerol, the
AAD is slightly higher, i.e., 4.86%.
Next, the developed correlation was compared with
two available correlations in the literature, i.e., the
Hammerschmidt correlation (1934) and the Yousif
and Young correlation (1993), for some selected sys-

100

P/MPa

Distilled water
11.8 mass% NaCl
21.5 mass% NaCl
Developed correlation
Corr., Hammerschmidt (1934)
Corr., Yousif & Young (1993)

10

Exp. data: de Roo et al. (1983)

1
260

265

270

275

280

285

T/K
Fig. 7. Comparison between available correlations for methane hydrate phase boundaries in the presence of sodium chloride aqueous solutions.

78

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

80

Exp., Ng & Robinson (1994)

Corr., Yousif & Young (1993)
Thermodynamic model
Developed correlation

70

23 wt% CaCl2

15 wt% CaCl2

60
Comp. (mole%)

P/MPa

50
40
30

C1:
C2:
C3:
C4:
CO2:
N2:

80.5
10.3
5.0
4.3
1.8
0.1

Pure water

20
10
0
270

275

280

285

290

295

300

305

T/K
Fig. 8. Experimental and predicted hydrate phase boundaries of a natural gas in the presence of CaCl2.

tems. Fig. 7 presents the comparison between the

above mentioned literature correlations and the
newly developed correlation for methane hydrate
phase boundary in the presence of various concentrations of sodium chloride. The available experimental
data (de Roo et al., 1983) for this system has been also
included in the figure. As shown in the figure, the
literature correlations show significant deviations
from experimental data. However, the results of the
developed correlation are in good agreement with
experimental hydrate data.

P/MPa

100

In Fig. 8, the developed correlation is compared

with the HWHYD thermodynamic model and the
correlation by Yousif and Young (1993) for a natural
gas hydrate phase boundary in the presence of various
aqueous solutions of calcium chloride. The experimental hydrate data (Ng and Robinson, 1994) for
this system and the composition of the natural gas
are also illustrated in the figure. Again, as can be seen,
Yousif and Young correlation has a significant deviation from the experimental data, especially at high
concentrations of salt.

Distilled water, de Roo et al. (1983)

10 mass% EG, Robinson and Ng (1986)
30 mass% EG, Robinson and Ng (1986)
50 mass% EG, Robinson and Ng (1986)
Thermodynamic model
Developed correlation

10

1
260

265

270

275

280

285

290

T/K
Fig. 9. Experimental and predicted (HWHYD thermodynamic model and the developed empirical correlation) methane hydrate phase
boundaries in the presence of ethylene glycol.

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

Fig. 9 shows the predictions of the empirical correlation and the HWHYD thermodynamic model, as
well as available experimental data (de Roo et al.,
1983; Robinson and Ng, 1986) for methane hydrate
dissociation conditions in the presence of different
concentrations of ethylene glycol (EG). The good
agreement between the predictions of the developed
empirical correlation with those of the HWHYD thermodynamic model demonstrates the reliability of the
correlation.

5. Conclusions
A generic correlation for predicting the hydrate
stability zone of reservoir and drilling fluids in the
presence of thermodynamic inhibitors (i.e., electrolytes and organic inhibitors) has been developed.
The correlation requires inhibitor concentration, pressure of the system and, if known, the dissociation
pressure of the hydrocarbon fluid in the presence of
distilled water at 273.15 K.
The developed correlation has been validated
against the comprehensive HWHYD thermodynamic
model for various hydrocarbon systems containing
salts or organic inhibitors with typical absolute average
deviation (AAD%) values between 1 and 2 and maximum AAD% of 4.86 for glycerol. In addition, the
correlation has shown superior reliability in comparison with reported correlations in the open literature.
The correlation can be used in hand-held calculators or spreadsheets4 as a tool for estimating hydrate
inhibition effect of thermodynamic inhibitors. It also
can be used to provide a reliable initial guess for more
sophisticated computer programs.
The hydrate stability zone of fluids in the presence
of distilled water, the main requirement of the developed correlation, can be estimated using the available
correlations in the open literature, such as stergaard
et al. (2000). As a result, the hydrate stability zone of
reservoir fluids in the presence or absence of thermodynamic inhibitors could be estimated by coupling the
developed correlation with the above mentioned open
literature correlation.

4
A free copy of the spreadsheet is available at the Hydrate
Groups website: www.pet.hw.ac.uk/research/hydrate.

79

Nomenclature
DT
Suppression of the hydrate dissociation
temperature (in K or 8C)
P
Pressure of the system (in kPa)
W
Concentration of the inhibitor in liquid water
phase (in mass%)
Dissociation pressure of hydrocarbon fluid in
P0
the presence of distilled water at 273.15 K
(kPa)
Ci
Constants given in Tables 2 and 3
DTi,model Suppression of the hydrate dissociation
temperature calculated by the HWHYD
DTi,corr Suppression of the hydrate dissociation
temperature calculated by correlation
AAD% Absolute average percentage deviation
n
Number of data points

Acknowledgments
This work was part of a joint industrial project
funded by ABB Offshore Systems Ltd., the UK Department of Trade and Industry, Petrobras, Shell UK
Exploration and Production, and TOTAL, whose support is gratefully acknowledged.

References
Amy, T.K., Gongmin, F., Malene, A.W., Mason, B.T., 2002. Effect
of hydrate inhibitors on oilfield scale formation and inhibition.
SPE 74657 presented at the SPE Oilfield Scale Symposium,
Aberdeen, 3031 January.
Avlonitis, D., Danesh, A., Todd, A.C., 1994. Prediction of VL and
VLL equilibria of mixtures containing petroleum reservoir
fluids and methanol with a cubic EoS. Fluid Phase Equilib.
94, 181 216.
Carson, D.B., Katz, D.L., 1942. Natural gas hydrates. Trans. AIME
146, 150.
de Roo, J.L., Peters, C.J., Lichtenthaler, R.N., Diepen, G.A.M.,
1983. Occurrence of methane hydrate in saturated and
unsaturated solutions of sodium chloride and water in dependence of temperature and pressure. AIChE J. 29 (4),
651 657.
Hammerschmidt, E.G., 1934. Formation of gas hydrates in natural
gas transmission lines. Ind. Eng. Chem. Res. 26, 851.
Katz, D.L., 1945. Prediction of conditions for hydrate formation in
natural gases. Trans. AIME 160, 140 149.
Katz, D.L., Cornell, D., Kobayashi, R., Poettmann, F.H., Vary, J.A.,
Elenbaas, J.R., Weinaug, C.E., 1959. Handbook of Natural Gas
Engineering. McGraw-Hill Inc., New York, USA.

80

K.K. stergaard et al. / Journal of Petroleum Science and Engineering 48 (2005) 7080

Kihara, T., 1953. Virial coefficient and models of molecules in

gases. Rev. Modern Phys. 25 (4), 831 843.
Matthews, P.N., Subramanian, S., Creek, J., 2002. High impact,
poorly understood issues with hydrate in flow assurance. Proceeding of the 4th International Conference on Gas Hydrates,
Yokohama, May 1923, pp. 899 905.
Nielsen, R.B., Bucklin, R.W., 1983. Why not use methanol for
hydrate control? Hydrocarbon Process. 62 (4), 71 78.
Ng, H.J., Robinson, D.B., 1985. Hydrate formation in systems
containing methane, ethane, propane, carbon-dioxide or hydrogen-sulfide in the presence of methanol. Fluid Phase Equilib. 21
(12), 145 155.
Ng, H.J., Robinson, D.B., 1994. New developments in the measurement and prediction of hydrate formation for processing
needs. Ann. N. Y. Acad. Sci. 715, 450 462.
Ng, H.J., Chen C.-J., Robinson, D.B., 1987 The influence of high
concentrations of methanol on hydrate formation and the distribution of glycol in liquidliquid mixtures. Gas Processors Association Research Report, RR-106.
stergaard, K.K., Tohidi, B., 2002. A correlation for predicting the
hydrate stability zone in the presence of ice. 4th International
Conference on Gas Hydrates, Yokohama, Japan, 1923 May.
stergaard, K.K., Burgass, R.W., Tohidi, B., Danesh, A., Todd,
A.C., 1998. Advances in experiments and computer modelling
software to simulate hydrate formation with respect to salts and
inhibitors. Practicalities of Predicting, Preventing and Controlling Waxes, Hydrates and Asphaltenes, Aberdeen, UK, 1718
June.
stergaard, K.K., Danesh, A., Tohidi, B., Todd, A.C., Burgass,
R.W., 2000. A general correlation for predicting the hydrate
free zone of reservoir fluids. SPE Prod. Facil. 15 (4), 228 233.
Parrish, W.R., Prausnitz, J.M., 1972. Dissociation pressures of gas
hydrates formed by gas mixtures. Ind. Eng. Chem. Process Des.
Dev. 11 (1), 26 35.

Robinson, D.B., Ng, H.-J., 1986. Hydrate formation and inhibition

in gas or gas condensate streams. J. Can. Pet. Technol. 25 (4),
26 30.
Robinson, R.A., Stokes, R.H., 1959. Electrolyte Solutions. Butterworths & Co., London, UK.
Tohidi, B., Danesh, A., Todd, A.C., 1995. Modelling single and
mixed electrolytesolutions and its applications to gas
hydrates. Chem. Eng. Res. Des. 73 (A4), 464 472.
Tohidi, B., Danesh, A., Burgass, R.W., Todd, A.C., 1996a. Effect
of heavy hydrate formers on the hydrate free zone of real
reservoir fluids. SPE 35568, The European Production Operations Conference and Exhibition, Stavanger, Norway, 1617
April, pp. 257 261.
Tohidi, B., Danesh, A., Todd, A.C., Burgass, R.W., 1996b. Measurement and prediction of hydrate-phase equilibria for reservoir
fluids. SPE Prod. Facil. 11 (2), 69 76.
Tohidi, B., Danesh, A., Todd, A.C., 1997a. Predicting pipeline
hydrate formation. Chem. Eng. (Lond.) 642, 32 37.
Tohidi, B., Danesh, A., Todd, A.C., Burgass, R.W., 1997b. Hydratefree zone for synthetic and real reservoir fluids in the presence
of saline water. Chem. Eng. Sci. 52 (19), 3257 3263.
Valderrama, J.O., 1990. A generalized PatelTeja equation of state
for polar and nonpolar fluids and their mixtures. J. Chem. Eng.
Jpn. 23 (1), 87 91.
Van der Waals, J.H., Platteeuw, J.C., 1959. Clathrate solutions. Adv.
Chem. Phys. 2, 1 57.
Wilcox, W.I., Carson, D.B., Katz, D.L., 1941. Natural gas hydrates.
Ind. Eng. Chem. 33 (5), 662 665.
Yousif, M.H., Young, D.B., 1993. A simple correlation to predict
the hydrate point suppression in drilling fluids (SPE/IADC
25705). The 1993 SPE/IADC Drilling Conference, Amsterdam,
The Netherlands, 2325 February, pp. 287 294.