Available online at www.sciencedirect.

com

J. of Supercritical Fluids 44 (2008) 115–122

Elemental sulphur deposition in natural gas transmission

and distribution networks
Pierre Cézac a , Jean-Paul Serin a,∗ , Jean-Michel Reneaume a ,
Jacques Mercadier a , Gérard Mouton b
a Laboratoire de Thermique, Energétique et Procédés, Ecole Nationale Supérieure en Génie des Technologies Industrielle, Université de Pau et des Pays de
l’Adour, BP 7511, 64075 Pau Cedex, France
b Total Infrastructures Gaz France, Direction Opérations, Services Supports Techniques-Méthodes,7 rue de la Linière, 64140 Billère, France

Received 25 April 2007; received in revised form 8 November 2007; accepted 9 November 2007

Abstract
Deposition of elemental sulphur in natural gas transmission line systems has been reported on only in recent times. The most common location
is immediately downstream of a point of significant pressure reduction. A necessary step to eliminate solid sulphur deposition is to understand
the deposit mechanism. Two mechanisms appear possible: condensate formation and desublimation of sulphur vapours. To study the plausible
paths of solid sulphur formation, we established a thermodynamic model. The constitutive equations are classically the chemical and physical
equilibriums, partial mass balances and energy balance. We chose the Peng Robinson equation of state to describe the phase behaviour. Mixing rules
are conventional: quadratic for a and linear for b with one binary interaction coefficient. Desublimation seems to be the most plausible mechanism
to explain the solid sulphur formation. Pressure reduction implies temperature reduction, which reinforces solid formation. A solution to minimize
the solid sulphur deposition seems to be heating the gas stream immediately before pressure reduction.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Modelling; Solid–gas equilibrium; Natural gas; Sulphur; Pressure drop

1. Introduction transportation and listed the most common places where this
phenomenon occurs. It appears that the most common location
Sour gas is a natural gas that contains hydrogen sulphide is immediately downstream of a pressure reduction facility. This
(H2 S). Deposition of sulphur during sour gas wells exploita- deposition can have serious consequences for pipeline operators.
tion is a problem that was recognized several decades ago. For instance, sulphur deposit can be observed around pressure
Hyne [1] showed the difficulty to establish a yardstick of gen- control valves and then affect the stem movement or plug the
eral applicability to explain sulphur deposition. He reached the valve orifice. Solid sulphur can also modify the accuracy of flow
conclusion that no single factor can be used to predict or ratio- meters, Chesnoy and Pack [2] reported that the deposit can intro-
nalize deposit. Thus, he proposed various factors, which could duce a measurement error of up to 2% and in some cases even
act on sulphur deposition in sour gas wells like the composi- more. A necessary step to eliminate solid sulphur deposition
tion or the gradient of temperature and the pressure between is to understand the deposit mechanism. The aim of this study
the bottom and the head of the well. Occurrence of sulphur is to contribute to a better comprehension of the mechanisms
deposits in natural gas transmission facilities has recently been involved.
observed. Chesnoy and Pack [2] described sulphur deposit for-
mation in a nozzle placed in a line connected to a natural gas
pipeline. Wilkes and Pareek [3] observed sulphur deposits near 2. Solid sulphur deposition theories
the control valves of a gas turbine. Pack [4] studied natural gas
Three potential deposition mechanisms have been consid-
ered: chemical reaction, condensate formation and desubli-
∗ Corresponding author. Tel.: +33 5 59 40 78 31. mation of sulphur. Each could interact with one or several
E-mail address: [email protected] (J.-P. Serin). others.

0896-8446/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.supflu.2007.11.005
116 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122

Studies on this field are far from numerous and no experi-

Nomenclature mental data is available. This theory does not suit the physical
location of the deposit: the solid sulphur deposition appears
a attractive parameter in cubic EOS (m3 mol−1 )2 immediately downstream of a pressure reduction facility.
b covolume in cubic EOS (m3 mol−1 ) Indeed, it seems surprising that this reaction does not occur ear-
cp heat capacity (J mol−1 K−1 ) lier in the pipeline. Moreover, if solid elemental sulphur is the
f fugacity (Pa) product of a chemical reaction, this reaction should be very fast.
H enthalpy (J) In fact, the gas velocity is close to 10 m s−1 in the network and
kij binary interaction coefficient for the parameter a could reach mach 1 inside pressure reduction facilities [4]. Con-
K equilibrium constant sequently the residence time is extremely short, i.e. lower than
n molar flow (mol s−1 ) 0.01 s in a gas expander. Finally, pressure reduction in natural gas
P pressure (MPa) transmission systems involves a temperature reduction, which
Pa atmospheric pressure (0.101325 MPa) generally does not favour chemical reactions.
P◦ standard pressure (0.1 MPa)
R ideal gas constant (8.314 J mol−1 K−1 ) 2.2. Condensate
T temperature (K)
v molar volume (m3 mol−1 ) Wilkes and Pareek [3] identified condensate formation as a
x liquid composition potential cause of the deposit formation. As shown in Fig. 1,
y vapour composition liquid hydrocarbons can appear with pressure reduction. The
z molar composition natural gas compositions are given in Table 1. The liquid hydro-
Z compressibility factor carbons can dissolve a part of the vapour sulphur and then
remove it from the gas. Evaporation of the condensed liquids
Greek letters might lead to the deposition of sulphur particles. This forma-
H enthalpy of phase change (J mol−1 ) tion hypothesis matches the physical location of the deposit, i.e.
Π product over all species downstream of a pressure reduction facility.
φ fugacity coefficient
ν stoichiometric coefficient 2.3. Desublimation
ω acentric factor
ξ reaction extent (mol s−1 ) The most common location for elemental sulphur deposition
in natural gas transmission pipelines is immediately downstream
Superscript of a point of pressure reduction [3,4]. Thus, the pressure drop
␣ relative to orthorhombic sulphur (solid form) appears to be an important parameter of the sulphur deposition
␤ relative to monoclinic sulphur (solid form) mechanism. Moreover, a pressure drop leads to a temperature
f fusion property reduction. So, the change of both temperature and pressure con-
GP ideal gas state property ditions can result in a supersaturated state for sulphur. Because
Inlet relative to the inflow the temperature is below the temperature of the triple point
L liquid property (95.5 ◦ C [5]) of sulphur, the excess of vapour sulphur changes
Outlet relative to the outflow directly to solid state. The sulphur desublimation hypothesis is
res residual property therefore in agreement with the physical location of the solid
S solid property deposition observed.
sat saturation property
sub sublimation property 2.4. Conclusion on solid sulphur deposition theories
V vapour property or vaporization property
tr solid/solid transition property As indicated before, the particular place of the deposit, the
◦ standard state short residence time and the decrease in temperature indicate that
* pure state
Table 1
Molar compositions (%) of the natural gases studied
Gas A Gas B Gas C Gas D Gas E Gas F
2.1. Chemical reaction CO2 -rich gas H2 S-rich gas

CH4 90.69 96.79 82.73 83.92 86.79 90.69

Chesnoy and Pack [2] and Pack [4] listed several potential
C2 H6 7.88 2.21 12.2 3.62 2.21 7.88
chemical reactions producing solid sulphur. For instance they C3 H8 0.82 0.09 2.69 0.74 0.09 0.82
proposed the following reactions: C4 H10 0.12 0.54 0.39 0.24 0.54 0.12
C5 H12 – – 0.05 0.14 – –
2H2 S(g) + O2 (g) → 2S(s) + 2H2 O(g) (1) N2 0.49 0.37 1.36 10.1 0.37 0.49
CO2 – – 0.57 1.24 10.00 –
C6 H6 (g) + 4H2 S(g) → C6 H14 (g) + 4S(s) (2) H2 S – – – – – 1.E−03
 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122 117

Fig. 1. Phases envelopes of four gases.

solid sulphur deposition is not the result of a chemical reaction. 3.1.1. Vapour–liquid equilibrium
The condensate and desublimation mechanisms seem the most The thermodynamic equilibrium conditions imply that pres-
plausible. To study them, we have established a thermodynamic sure, temperature and fugacity are identical in both phases:
model to describe the system in ranges of temperature, pressure
and composition, which correspond to natural gas transportation fiL (T, P, x) = fiV (T, P, y) (5)
conditions. Gaseous elemental sulphur is a complex compound,
which exists in several forms from S8 to S. They co-exist through 3.1.2. Vapour–solid equilibrium of sulphur
gaseous equilibriums [1,6]: In the same way, we have:
x fS∗S
8
(T, P) = fSV8 (T, P, y) (6)
S8 ⇔ Sx , x = 1, 7 (3)
8
3.1.3. Vapour chemical equilibriums
Moreover, Hyne [1] has experimentally showed the existence of
The criterion of chemical equilibrium is:
equilibriums between sulphur and polysulfanes when hydrogen
sulphide is present.   f V (T, P, y) vi,r
i
= K◦ (T ) (7)
(x − 1) P◦
i
H2 S + S8 ⇔ H2 Sx , x = 2, 9 (4)
8
As indicated before, we consider the gaseous equilibriums (3)
These two types of equilibriums (3) and (4), which involve only and (4).
gaseous species, can influence the gas composition and conse-
quently the thermodynamic properties of the system. That is the 3.1.4. Energy balance
reason why we integrated them in our model. The energy balance is classically based on the first principle
of thermodynamics. We assume that there is no energy exchange.
3. Model description So, the energy balance is expressed as:
H Inlet (T Inlet , P Inlet ) = H Outlet (T Outlet , P Outlet ) (8)
The model used is a reactive flash model. It is based on a ther-
modynamic description of the macroscopic physical–chemical
phenomena occurring in a vapour–liquid–solid-system, which is 3.1.5. Partial mass balances
assumed perfectly mixed. Solid sulphur is supposed pure and we The partial mass balances are written as follows:
assume that there is no metastable state. The pressure reduction nSi + nV A

νi,r ξr = 0 (9)
i − ni −
is isenthalpic. Assuming the above conditions, unknown factors r
are the composition of the vapour and liquid phases, and the
quantity of solid sulphur. 3.2. Numerical procedure

3.1. Governing equations The equations make up a non-linear system in which all the
thermodynamic properties are expressed as a function of com-
The inlet of the flash is always a vapour phase. The outlet position, pressure and temperature. The degree of freedom is
can be vapour, solid or (and) liquid. Thus we have three types zero. The overall system is solved in steady state using the
of equilibrium. Newton–Raphson’s method.
118 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122

3.3. Thermodynamic models form:

f ∗␣ (T, P a ) H tr (T tr , P a ) 1
 
1
3.3.1. Fugacities in the vapour phase and in the liquid ln ∗␤ =− − tr
phase f (T, P a ) R T T
The thermodynamic model used to describe the fugacity of  T T ∗α a ∗β a
T tr (Cp (T, P ) − cp (T, P )) dT
a compound in a fluid phase is the Peng Robinson equation of − dT
T tr RT 2
state [7] coupled with the van der Waals mixing rules. These (14)
latter can be written as follows:
 √ To obtain the solid fugacity of both solid forms at a pressure
am = yi yi ai aj (1 − kij (T )) (10) other than atmospheric pressure, a pointing term is added. The
i j
data sources required to express solid fugacity are reported by
 
bi + bj
 Cezac et al. [8].
bm = yi yj (11)
2 3.3.3. Enthalpies in the vapour phase and in the liquid
i i
phase
3.3.2. Solid phase fugacity Enthalpy is the sum of the residual enthalpy Hres and the
The fugacity of a pure solid specie can be written in differ- enthalpy of the perfect gas:
ent ways [8]. The most common way relates the solid fugacity
to the vapour fugacity using the sublimation pressure. In this H V or L (T, P, z) = H res (T, P, z) + H GP (T, z) (15)
case, only two physical properties are required: the solid molar with:
volume and the sublimation pressure. However, below 330 K,  ∞  
the sublimation pressure is unknown [9]. So, this expression res ∂P
H (T, P, z) = P −T dv + RT (ZV or L − 1)
of fugacity cannot be used under natural gas transportation v ∂T v
conditions. Another approach to express the pure solid fugac- (16)
ity is to link it to the pure liquid fugacity. Then the physical
properties needed are the fusion enthalpy and temperature, the 3.3.4. Pure sulphur solid phase enthalpy
heat capacities and the molar volume. In this expression, the We assume that the molar volume of both solid sulphur forms
heat capacities are considered as constant and are evaluated at is constant in the range of temperature and pressure. Thus the
the fusion temperature. Because the temperature range of gas enthalpy of the monoclinic form is written:
transportation conditions is far from the fusion temperature, we
H ∗␤ (T, P) = H ∗L (T f , P atm ) − H f (T f , P a )
established a new way to write the pure solid fugacity. The ther-
 T
modynamic path to obtain it is detailed in a previous article
[8]. + cp∗␤ (T, P a ) dT + v∗␤ (P − P a ) (17)
Tf

f (T, P a )
 ∗S 
Similarly to the fugacity expressions, the enthalpy of the
ln
f ∗L (T f , P a ) orthorhombic form is calculated from the enthalpy of the
  monoclinic form under atmospheric pressure and transition tem-
H f (T f , P a )+H V (T f , P sat (T f ))−HC

T perature.
=− 1− f
RT T
H ∗␣ (T, P) = H ∗␤ (T tr , P a ) − H tr (T tr , P a )
(12)  T
+ cp∗␣ (T, P a ) dT + v∗␣ (P − P a ) (18)
T tr
T
T cp∗S−GP (T, P a ) dT

Tf
− dT
Tf RT 2 3.3.5. Chemical equilibrium constants
where The chemical equilibrium constants are directly linked to the
standard free Gibbs energy. All the data except for H2 Sx (x = 2,
HC = H res (T f , P sat (T f )) 9) are available in the databank of the National Institute of Stan-
  ∗L   dards Technology [10]. We used for H2 Sx the data reported by
f ∂v
+ v −T ∗L
(P a − P sat (T f )) (13) Heidemann et al. [11]. Polynomial expressions of the chemical
∂T P a
equilibrium constants are given in Cezac et al. [8].
Just below the fusion temperature, the thermodynamic sta-
ble form of solid sulphur is monoclinic (␤-form). Its fugacity 3.4. Parameters of the model
is calculated according to Eq. (12). In gas transport, the
temperature ranges are below the solid transition tempera- 3.4.1. Characteristic parameters
ture. Then, the solid stable form is the orthorhombic one Calculations using the Peng Robinson equation of state
(␣-form). Its fugacity is related to that of the monoclinic require: the critical temperature, the critical pressure and
 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122 119

Table 2
Influence of the drop in pressure on solid sulphur formation—gas A and gas B
Upstream pressure (fixed) (MPa) Gas A Gas B

Downstream temperature Solid sulphur repartition Downstream temperature Solid sulphur distribution
(calculated) (K) (calculated) (mass%) (calculated) (K) (calculated) (mass%)

5 250.0 99.86 252.4 99.74

4 257.0 99.01 258.7 98.49
3 263.8 94.22 264.9 92.44
2.5 267.1 87.14 268.0 84.24
2 270.4 73.27 271.0 69.11
1.6 273.0 24.99 273.5 50.18

Upstream temperature: 280 K, downstream pressure: 0.5 MPa.

Table 3
Influence of the drop in pressure on solid sulphur formation—gas C and gas D
Upstream pressure (fixed) Gas C Gas D
(MPa)
Downstream temperature Solid sulphur repartition Downstream temperature Solid sulphur distribution
(calculated) (K) (calculated) (mass%) (calculated) (K) (calculated) (mass%)

5 246.5 99.94 253.2 99.66

4 254.4 99.47 259.2 98.17
3 262.1 96.11 265.3 91.44
2.5 265.8 90.46 268.3 82.68
2 279.5 78.34 271.2 67.01
1.6 272.3 61.03 273.6 50.64

Upstream temperature: 280 K, downstream pressure: 0.5 MPa.

the acentric factor of each compound. For all non-sulphured the other polysulfanes (H2 S6 to H2 S9 ) are calculated accord-
compounds of natural gas except hydrogen sulphide, these ing to the Prausnitz and Heidemann approach [14,15]. S8
values are available in Chemistry Data Series [12,13]. For critical parameters are also given in Cezac et al. [8]. The
polysulfanes H2 S2 to H2 S5 , we deduced these parameters characteristic parameters of the other sulphur species (Sx ) are
from the data provided by Simmrock et al. [13]. They are calculated according to the Prausnitz and Heidemann approach
given in Cezac et al. [8]. The characteristic parameters of [14,15]

Table 4
Influence of the temperature on solid sulphur formation for a typical pressure drop from 5 to 0.5 MPa—gas A and gas B
Upstream temperature Gas A Gas B
(fixed) (K)
Downstream temperature Solid sulphur repartition Downstream temperature Solid sulphur distribution
(calculated) (K) (calculated) (mass%) (calculated) (K) (calculated) (mass%)

280 250.0 99.86 252.4 99.14

290 262.4 98.55 264.5 97.57
300 274.4 88.39 276.4 81.86
310 286.3 24.33 288.0 0
320 298.0 0 299.5 0

Table 5
Influence of the temperature on solid sulphur formation for a typical pressure drop from 5 to 0.5 MPa—gas C and gas D
Upstream temperature Gas C Gas D
(fixed) (K)
Downstream temperature Solid sulphur repartition Downstream temperature Solid sulphur distribution
(calculated) (K) (calculated) (mass%) (calculated) (K) (calculated) (mass%)

280 246.5 99.94 253.2 99.66

290 259.3 99.33 265.2 96.87
300 271.7 94.08 277.0 77.02
310 283.9 58.16 288.6 0
320 295.8 0 300.1 0
120 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122

Table 6
Influence of the H2 S composition on solid sulphur formation for a typical pressure drop from 5 to 0.5 MPa—gas A/gas F
Upstream temperature Downstream temperature Solid sulphur distribution
(fixed) (K) (calculated) (K) (calculated) (mass%)
Gas A Gas F Gas A Gas F

280 250.0 250.0 99.79 99.78

290 262.4 262.4 97.91 97.88
300 274.4 274.5 83.30 83.25
310 286.3 286.3 0 0
320 298.0 298.0 0 0

3.4.2. Binary interaction coefficients networks. We plotted on Fig. 1 the gas envelope of these gases.
Binary interaction parameters are necessary to use the van der The formation of condensates could occur if the pressure and
Waals mixing rules. For all non-sulphured compounds except temperature conditions during the pressure reduction cross the
hydrogen sulphide, the binary interaction coefficients are clas- phase envelope. Typical operational conditions of a natural gas
sically available in the databank [16]. flow upstream of a gas expander are 5 MPa and 280 K. If we
The knowledge of all binary interaction parameters between impose a classical downstream pressure of 0.5 MPa, the cal-
sulphur and other components should lead to a better descrip- culated outlet temperature is near 253 K. Thus, according to
tion of the natural gas systems. Due to the lack of experimental the phase envelope, the probability of condensate formation is
data, we regressed the binary coefficients between sulphur and very low. However, the gas composition can be an important
respectively CH4 , CO2 and H2 S from sulphur solubility data factor. Fig. 1 shows that the phase envelope of a natural gas,
[17,18] in each binary and from experimental sulphur solubility which contains heavier compounds (gas C), presents a shift in
in ternary system reported by Sun and Chen [19]. We assumed the phase envelope at higher temperature. So, for this natural
the other binary interaction coefficients to be equal to zero. The gas, the probability of condensate formation increases. Never-
determination of these three binary interaction coefficients is theless, it remains low under the classical conditions of pressure
presented in Cezac et al. [8] and temperature met in natural gas transportation.

4. Application of the model 4.2. Hypothesis of desublimation

4.1. Hypothesis of condensate By definition, desublimation is the direct transformation of a

vapour compound into a solid. If the vapour phase becomes
To study this hypothesis, we considered four natural gases, supersaturated because of the change of operational condi-
whose compositions are reported in Table 1. These compositions tions during a pressure reduction, according to a desublimation
are typical of gases transported in transmission and distribution process, deposition can appear. To study this hypothesis, we con-

Table 7
Influence of the CO2 composition on solid sulphur formation for a typical pressure drop from 5 to 0.5 MPa—gas B/gas E
Upstream temperature Downstream temperature Solid Sulphur distribution
(fixed) (K) (calculated) (K) (calculated) (mass%)
Gas B Gas E Gas B Gas E

280 252.4 249.9 98.93 99.80

290 264.5 262.2 96.97 97.98
300 276.4 274.3 77.40 83.81
310 288.0 286.2 0 0
320 299.5 297.9 0 0

Table 8
Influence of the composition of the heavy compounds on solid sulphur formation for a typical pressure drop from 5 to 0.5 MPa—gas B/gas C
Upstream temperature Downstream temperature Solid sulphur distribution
(fixed) (K) (calculated) (K) (calculated) (mass%)
Gas B Gas C Gas B Gas C

280 252.4 246.5 98.93 99.90

290 264.5 259.3 96.97 98.79
300 276.4 271.7 77.40 89.27
310 288.0 283.9 0 24.09
320 299.5 295.8 0 0
 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122 121

sidered the same four gases as for the condensate hypothesis. downstream of a point of pressure reduction. Thus, drops in
The most common location of the sulphur deposit is imme- pressure and temperature are both very important parameters.
diately downstream of a pressure reduction facility. Thus, we The composition of the gas, specially in CO2 and heavy hydro-
studied this kind of configuration. The upstream conditions and carbons, also has an impact in the operational conditions of
downstream pressure are fixed. The downstream temperature gas transport. It is very important to note that these results are
and sulphur distribution between the vapour and solid phases obtained by prediction. Moreover, the hypothesis of the non-
are calculated. As no experimental value of the sulphur vapour existence of a supersaturated vapour phase can quantitatively
concentration in the temperature and pressure domain of interest affect the results. Due to the lack of experimental results, mod-
has been published yet, we chose to consider that the upstream elling and simulation are nowadays the only way to study the
flow of natural gas is saturated in sulphur. Tables 2 and 3 present solid sulphur formation observed by several gas companies.
the calculated downstream temperature and the calculated distri- Experimental validation is necessary before using the model
bution of the sulphur between the vapour and solid phases. As we for optimizing operational conditions like determinating the
assumed that the enthalpy of the system is constant, pressure and upstream temperature avoiding the sulphur deposition.
temperature are linked. At fixed upstream temperature, the larger
the decrease in pressure, the lower the downstream temperature.
For the gas A, a reduction of 1 MPa entails on average a reduc- 5. Conclusion
tion of 5.6 K. If we consider a drop in pressure of 4.5 MPa, more
than 99 mass% of the sulphur present in the upstream gas flow is Deposition of elemental sulphur in natural gas transmis-
in solid state downstream of the pressure reduction facility. The sion line systems appears around pressure control facilities and
more the pressure drop decreases, the more the solid distribution affects their operations. A necessary step to eliminate solid sul-
decreases, but in all cases studied solid sulphur appears. Thus, phur deposition is to understand the deposit mechanism. We
desublimation is the most likely sulphur deposition mechanism. established a reactive flash model. The constitutive equations
Next, we used the model to study the influence of operational are classically the chemical and physical equilibriums, partial
conditions, like upstream temperature or gas composition, on mass balances and energy balance. The condensate hypothesis
the sulphur deposit formation. The higher the pressure drop, appears the less probable. However, it cannot be totally rejected,
the lower the downstream temperature and the higher the sul- especially if the gas is rich in heavy hydrocarbons. Desublima-
phur desublimation. Tables 4 and 5 present the influence of the tion appears to be the most plausible mechanism explaining the
upstream temperature on the downstream sulphur distribution formation of a sulphur deposit. Pressure reduction implies tem-
between solid and vapour phases. For all the gases studied, a perature reduction, which reinforces solid formation. A solution
higher upstream temperature means a decrease in the solid sul- to minimize the solid sulphur deposition is to heat the gas stream
phur mass. Moreover, there is an upstream temperature above before pressure reduction.
which no precipitation occurs. Thus, increasing the upstream
gas temperature can prevent the sulphur deposition problem by
References
maintaining sulphur in vapour phase. We also used the model
to study the influence of the gas composition on the solid sul- [1] J.B. Hyne, Study aids prediction of sulphur deposition in sour-gas wells,
phur formation. Tables 6–8 compare the calculated downstream Oil Gas J. (1968) 107–113.
temperature and the downstream solid sulphur distribution of [2] A.-B. Chesnoy, D.J. Pack, S8 threatens natural gas operations, Oil Gas J.
a natural gas poor in H2 S (gas A) and a rich one (gas F), one (1997) 74–78.
poor in CO2 (gas B) and a rich one (gas E), and one poor in [3] C. Wilkes, V. Pareek, Sulfur Deposition in a Gas Turbine Natural Gas Fuel
Control System, Energy-Tech Online, 2001.
heavy hydrocarbons (gas B) and a rich one (gas C). For all these [4] D.J. Pack, Elemental sulphur formation in natural gas transmission
studies on the gas composition influence, we fixed the sulphur pipelines, Paper Presented at the 14th Biennal Joint Technical Meeting
content in upstream at the same value: 7E−4 mg N m−3 . So, all on Pipelines Research, Berlin, Paper No. 31, pp. 1–14, 2003.
the gases are not saturated before the pressure reduction. Results [5] W.N. Tuller, The Sulphur Data Book, McGraw-Hill, New York, 1954.
[6] H. Rau, T.R.N. Kutty, J.R.F. Guedes de Cavarho, thermodynamics of sul-
presented in Table 6 allow us to study the impact of H2 S con-
phur vapour, J. Chem. Thermodyn. 5 (1973) 833–844.
tent on the formation of sulphur deposit. For each pressure drop [7] D.Y. Peng, D.B. Robinson, GPA Research Report, Tulsa, 1978.
at fixed upstream temperature, calculated solid sulphur distri- [8] P. Cezac, J.-P. Serin, J. Mercadier, G. Mouton, Modelling solubility of solid
butions are quite identical. Presence of H2 S has no significant sulphur in natural gas, Chem. Eng. J. 133 (1-3) (2007) 283–291.
influence. Table 7 presents solid sulphur distribution for gas B [9] X. Shuai, A. Meisen, New correlations predict physical properties of ele-
mental sulphur, Oil Gas J. 93 (1995) 50–55.
and gas E, the last one being richer in CO2 . It appears that the
[10] P.J. Linstrom, W.G. Mallard (Eds.), NIST Chemistry WebBook, National
richer in carbon dioxide the gas, the higher the solid sulphur Institute of Standards and Technology, 2005 (NIST Standard Reference
distribution. Also the richer in CO2 the gas, the lower the down- Database Number 69).
stream temperature The CO2 presence affects the properties of [11] R.A. Heidemann, A.V. Phoenix, K. Karan, L.A. Behie, A chemical equilib-
the system. The behaviour is similar when the gas contains heavy rium equation of state model for elemental sulphur-containing fluids, Ind.
Eng. Chem. Res. 40 (2001) 2160–2167.
carbons as indicated in Table 8. The solid sulphur distribution is
[12] K. Simmrock, R. Janowsky, A. Ohnsorge, Critical Data of Pure Substances,
greater for gas C than for gas B. vol. II, Part 1 Chemistry Data SeriesPN Frankfurt/Main, 1986.
The model permits to study the influence of several parame- [13] K. Simmrock, R. Janowsky, A. Ohnsorge, Critical Data of Pure Substances,
ters on the sulphur distribution between solid and vapour phase vol. II, Part 2 Chemistry Data Series, Frankfurt/Main, 1986.
122 P. Cézac et al. / J. of Supercritical Fluids 44 (2008) 115–122

[14] R.A. Heidemann, J.M. Prausnitz, A van der Waals-type equation of state [17] M.-X. Gu, Q. Li, S.-H. Zhou, W.-D. Chen, T.-M. Guo, Exper-
for fluids with associating mixtures, Proc. Natl. Acad. Sci. U.S.A 73 (1976) imental and modeling studies on the phase behaviour of high
1773–1776. H2 S-content natural gas mixtures, Fluid Phase Equilib. 82 (1993) 173–
[15] R.A. Heidemann, J.M. Prausnitz, Phase Equilibria from an equation of state 182.
for fluids with associating molecules, in: AIChE Annual Meeting, Chicago, [18] J.J. Roof, Solubility of sulphur in hydrogen sulphide and in carbon disulfide
1976. at elevated temperature and pressure, Soc. Pet. Eng J. 11 (1971) 272–276.
[16] H. Knapp, R. Döring, L. Oellrich, U. Plöcker, J.M. Prausnitz, Vapour– [19] C.-Y. Sun, G.-J. Chen, Experimental and modelling studies on sul-
liquid equilibria for mixtures of low boiling substances, in: Chemistry Data phur solubility in sour gas, Fluid Phase Equilib. 214 (2003) 393–
Series, Frankfurt/Main, 1982. 402.