You are on page 1of 6

International Journal of Heat and Mass Transfer 102 (2016) 631–636

Contents lists available at ScienceDirect

International Journal of Heat and Mass Transfer


journal homepage: www.elsevier.com/locate/ijhmt

Diffusion model of gas hydrate dissociation into ice and gas: Simulation
of the self-preservation effect
Valeriy A. Vlasov
Institute of the Earth Cryosphere, Siberian Branch of the Russian Academy of Sciences, P.O. Box 1230, 625000 Tyumen, Russian Federation

a r t i c l e i n f o a b s t r a c t

Article history: The diffusion model of gas hydrate dissociation into ice and gas is presented. This model takes into
Received 22 April 2016 account the possible porous structure of the formed layer of ice as well as the intrinsic kinetics of the
Received in revised form 31 May 2016 process of gas hydrate dissociation into ice and gas. Specifically, the problem of the dissociation of a
Accepted 20 June 2016
spherical gas hydrate particle into ice and gas was considered. In the framework of a quasi-stationary
Available online 5 July 2016
approximation, a simplified solution to this problem was obtained. From the comparison between the
calculated data obtained in the framework of a quasi-stationary approximation and the available
Keywords:
experimental data, the parameters of the developed model responsible for the kinetics of methane
Gas hydrate
Diffusion
hydrate dissociation into ice and gas were estimated. In the framework of the developed diffusion model,
Chemical kinetics an explanation for the anomalous preservation thermal regime of gas hydrates is given. From this
Kinetic model explanation it follows that for some hydrate-forming gases such a regime may not occur.
Self-preservation effect Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction which in turn may appear during the process of gas hydrate disso-
ciation at temperatures below 273 K [34–38].
In the study of gas hydrates dissociation at temperatures below In works [39–42], methane hydrate dissociation into ice and gas
273 K, their slow decomposition into ice and gas was detected at different fixed temperatures and a pressure of 0.1 MPa was inves-
[1,2]. This phenomenon is called the self-preservation effect. The tigated. These investigations revealed one feature: the rate of
self-preservation effect is attracting increasing interest because it methane hydrate dissociation into ice and gas increases monoton-
can be the basis for the gas hydrate technologies for the long- ically with increasing temperature in the range of 193–240 K. In
term storage and transportation of various gases [3–5]. the temperature range of 242–271 K, the anomalous preservation
The self-preservation effect is being actively investigated thermal regime (anomalous preservation effect) is observed: the
[6–25]. This effect is observed during gas hydrates dissociation into rate of methane hydrate dissociation into ice and gas decreases
ice and gas in the whole region of their thermodynamic instability noticeably and non-monotonically. According to the authors of
at temperatures below the ice melting point. The self-preservation works [39–41], for an explanation of the anomalous preservation
effect manifests itself not only in massive gas hydrates but also in thermal regime, it is not enough to assume that the rate of gas
disperse gas hydrates in some carrier medium [26–30]. The hydrate dissociation into ice and gas is limited only by the diffusion
self-preservation effect is generally attributed to the formation of of the gas through the ice layer. In this case, the molecular processes
the ice layer on the gas hydrate surface. Depending on the temper- occurring at the ice–hydrate interface may be important [16,19].
ature and pressure conditions, the ice layer may have a different Presently, there is no full kinetic model of gas hydrate dissocia-
degree of homogeneity. Herewith, the rate of gas hydrate dissocia- tion into ice and gas. In this paper, an attempt is made to create
tion into ice and gas is limited by the diffusion of the gas through such a model. The presented model is based on the use of the
this ever-increasing ice layer. It is obvious that the size effect will diffusion equation, and when describing the processes occurring
be the consequence of this dissociation mechanism: the larger the at the ice–hydrate interface, the general theory of chemical kinet-
gas hydrate sample is, the longer it can exist in the thermodynamic ics is used. Previously, such an approach has been used to create a
instability region due to the self-preservation effect. The size effect kinetic model of gas hydrate formation from ice [43,44].
is really manifested in experiments [31–33]. Moreover, the protec-
tive ice layer may appear after crystallisation of supercooled water, 2. Theory

Assume that gas hydrate dissociation into ice and gas proceeds
E-mail address: vlasov.ikz@gmail.com by the diffusion mechanism: the gas molecules G diffuse through

http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.06.057
0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.
632 V.A. Vlasov / International Journal of Heat and Mass Transfer 102 (2016) 631–636

the formed ice layer from the ice–hydrate interface on which the Taking into account the fact that the molar concentration of the
chemical reaction occurs gas outside the particle under consideration is determined from its
thermal equation of state, the external boundary condition is writ-
G þ nH2 OðsolidÞ $ G  nH2 O; ð1Þ
ten as follows:
where n is the hydration number. In the case under consideration, p
the rate of the forward reaction is lower than the rate of the reverse cðr; tÞjr¼RðtÞ ¼ ; t > 0; ð5Þ
ZRT
reaction. The thickness of the ice layer increases due to gas hydrate
dissociation. Let us consider the situation in which the temperature where Z is the compressibility factor of the gas under current ther-
at the front of the reaction (1) does not change during the process of mobaric conditions and R is the gas constant.
gas hydrate dissociation. This situation is implemented under Gas molecules appear on the ice–hydrate interface due to reac-
isothermal conditions in the case where the process of gas hydrate tion (1). Next, these molecules diffuse through the ice layer to the
dissociation into ice and gas occurs rather slowly (even at the initial surface of contact of the particle under consideration with the gas-
stage) or in the case where good thermal stabilisation exists. In both eous phase. Thus, the internal boundary condition can be written
of these cases, the temperature at the front of the reaction (1) does as
not have time to change significantly, so we can assume that the rg ¼ jjr¼nðtÞ ; t > 0; ð6Þ
process of gas hydrate dissociation into ice and gas occurs in the dif-
fusion regime and should be considered only a diffusion problem. where rg is the rate of change in the moles of the gas on the ice–hy-
Consider the problem of the dissociation into ice and gas of a drate interface and j is the magnitude of the molar flux of the gas in
single spherical gas hydrate particle with an initial radius R0 the ice layer. In the framework of the general theory of chemical
(Fig. 1). During the process of gas hydrate dissociation, the radius kinetics, it can be shown [45] that the following relation is satisfied
of the gas hydrate core nðtÞ decreases. We assume that during for the quantity r g :
the process of gas hydrate dissociation, the outside radius of the !
ice layer RðtÞ also decreases. Such a decrease can be due to the dif- Z eq RT
rg ¼ kdis v 1 cðr; tÞjr¼nðtÞ ; t > 0; ð7Þ
ferent densities of ice and the gas hydrate. Moreover, we assume peq
that the particle under consideration is found in the atmosphere
where kdis is the rate constant for the gas hydrate dissociation reac-
of the hydrate-forming gas at constant pressure p and constant
tion at the ice–hydrate interface, v is the molar density of the gas
temperature T. Herewith, the pressure of the hydrate-forming
hydrate, and Z eq is the compressibility factor of the gas under equi-
gas is less than the pressure of the ice–gas hydrate–gas equilibrium
peq . Note that if gas hydrate dissociation into ice and gas occurs at a librium conditions. The rate constant kdis obeys the Arrhenius equa-
tion and characterises the intensity of the gas hydrate dissociation
fixed temperature in the atmosphere of the non-hydrate-forming
reaction on the ice–hydrate interface. The molar flux of the gas in
gas (as often occurs in real experiments), then it must be assumed
the ice layer is defined as
that p ¼ 0.
We formulate the problem in the spherical coordinate system @cðr; tÞ
j ¼ Deff er ; t > 0; nðtÞ < r < RðtÞ; ð8Þ
by choosing its origin at the centre O of the particle under consid- @r
eration. We take into account the possible presence of pores in ice
where er is the radial unit vector. Taking into account Eqs. (7)
by introducing the effective diffusion coefficient of the gas in ice:
and (8) and given that r g > 0 and @cðr; tÞ=@r < 0, the internal bound-
Deff ¼ D þ Dpor ; ð2Þ ary condition (6) is written in the form
! 
where D is the diffusion coefficient of the gas in ice and Dpor is the Z eq RT @cðr; tÞ
kdis v 1  cðr; tÞjr¼nðtÞ ¼ Deff ; t > 0: ð9Þ
coefficient characterising the diffusion of the gas through the sys- peq @r r¼nðtÞ
tem of pores in ice. In this case, the diffusion equation is written as
! Appendix A shows that there is the following interrelation
@cðr; tÞ @ 2 cðr; tÞ 2 @cðr; tÞ
¼ Deff þ ; t > 0; nðtÞ < r < RðtÞ; between the quantities RðtÞ and nðtÞ:
@t @r 2 r @r sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 1  
3 vM h ð1  eh Þ Mg
ð3Þ RðtÞ ¼ 1þ R30  n3 ðtÞ þ n3 ðtÞ; t P 0;
xMw ð1  ei Þ nMw
where cðr; tÞ is the molar concentration of the gas in the ice layer, r
is the radial coordinate, and t is the time. ð10Þ
The initial condition is written in the form where x is the molar density of ice; Mw , M h and Mg are the molar
cðr; tÞjt¼0 ¼ 0; 0 6 r < R0 : ð4Þ masses of water, the gas hydrate and the gas, respectively; and ei
and eh are the porosities of ice and the gas hydrate, respectively.

Fig. 1. Geometry of the problem at (a) the initial time and (b) later times.
V.A. Vlasov / International Journal of Heat and Mass Transfer 102 (2016) 631–636 633

The equation of motion for the front of the reaction (1) can be From Eq. (3), it follows that, in the framework of a quasi-
obtained from the molar balance ratio for the gas hydrate, which stationary approximation, the concentration profile cðr; tÞ should
has the form satisfy the following equation:

dRðtÞ @ 2 cðr; tÞ 2 @cðr; tÞ
jr h j ¼ vð1  eh Þ ; t > 0; ð11Þ þ ¼ 0; t > 0; nðtÞ < r < RðtÞ: ð19Þ
dt r¼nðtÞ @r 2 r @r
where rh is the rate of change in the moles of the gas hydrate on the It is easy to verify that the solution to Eq. (19) is the function of the
ice–hydrate interface and RðtÞ is the thickness of the ice layer form
(RðtÞ ¼ RðtÞ  nðtÞ). Note that a similar molar balance ratio can also A
be written for the water phase. However, as the analysis shows, the cðr; tÞ ¼ þ B; t > 0; nðtÞ 6 r 6 RðtÞ; ð20Þ
r
choice of one of the two molar balance ratios is equivalent because
where the coefficients A and B do not depend on the radial coordinate
this choice has little effect on the results of the simulation for the
r. Determining the coefficients A and B from the boundary conditions
kinetics of gas hydrate dissociation into ice and gas in the frame-
(5) and (9) and substituting them into Eq. (20), we obtain that
work developed by the model. The molar balance ratio (11)
 
accounts for (by introducing the factor ð1  eh Þ) the possible pore peq Deff
p Zeq RT þ kdis vnðtÞ 1  r
nðtÞ

structure of the dissociating gas hydrate. Given the geometry of cðr; tÞ ¼  


the problem under consideration, Eq. (11) can be written as follows: ZRT peq Deff þ k vnðtÞ 1  nðtÞ
Z eq RT dis RðtÞ
 
dnðtÞ jr h j kdis vn2 ðtÞ 1r  RðtÞ
1
¼ ; t > 0: ð12Þ peq
dt vð1  eh Þ þ   ; t > 0; nðtÞ 6 r 6 RðtÞ:
Z eq RT peq Deff þ k vnðtÞ 1  nðtÞ
Z eq RT dis RðtÞ
From work [45], it follows that the quantity rh is represented as
! ð21Þ
Z eq RT
r h ¼ kdis v cðr; tÞjr¼nðtÞ  1 ; t > 0: ð13Þ After substituting Eq. (21) into Eq. (14), we obtain the differen-
peq
tial equation
Taking into account that, in the case of the problem under consid- dnðtÞ kdis Deff ddis
eration, r h < 0, Eq. (12) can be written in the form ¼ h  i ; t > 0; ð22Þ
dt peq Deff
vð1  eh Þ Zeq RT v þ kdis nðtÞ 1  RðtÞ
nðtÞ
!
dnðtÞ kdis Z eq RT
¼ 1 cðr; tÞjr¼nðtÞ ; t > 0: ð14Þ where
dt ð1  eh Þ peq
peq p
Under the terms of the problem, there is no ice layer at the ini- ddis ¼  ; ð23Þ
Z eq RT ZRT
tial (zero) time. Thus, the initial condition for the quantity nðtÞ is as
follows: and the quantity RðtÞ is given by Eq. (10). Integration of the differ-
ential Eq. (22) with the initial condition (15) leads to the following
nðtÞjt¼0 ¼ R0 : ð15Þ
transcendental equation:
Eqs. (3)-(5), (9), (10), (14) and (15) form a closed system of peq 1
equations describing gas hydrate dissociation into ice and gas. This ðR0  nðtÞÞ þ ðR2  n2 ðtÞÞ
Z eq RTkdis v 2Deff 0
system of equations was obtained under the assumption that the  h i2=3 
outside radius of the ice layer changes during gas hydrate dissoci- 1
 R20  R30 k þ n3 ðtÞð1  kÞ
ation. The case in which this radius does not change is considered 2Deff ð1  kÞ
in Appendix B. ddis t
¼ ; t P 0; ð24Þ
vð1  eh Þ
3. Quasi-stationary approximation
where
 1
The presented system of equations can only be solved numer- vMh ð1  eh Þ Mg
k¼ 1þ : ð25Þ
ically. However, in the framework of a quasi-stationary approxi- xMw ð1  ei Þ nMw
mation, this system of equations can be solved analytically. The
quasi-stationary approximation can be used when the characteris- Using Eq. (24), we can determine the value of the quantity nðtÞ at
tic time for an increase in the ice layer thickness each time t.

RðtÞ 4. Comparison with experimental data


sth ¼ ð16Þ
dRðtÞ=dt
is much larger than the characteristic time for the diffusion of the The process of gas hydrate dissociation can be described by
gas through the ice layer such a quantity as the degree of gas hydrate dissociation gdis ðtÞ.
This quantity is the mass fraction of the dissociated gas hydrate:
R2 ðtÞ
sdif ¼ : ð17Þ m0h ðtÞ
Deff gdis ðtÞ ¼ ; t P 0; ð26Þ
mh0
Thus, if the condition
where m0h ðtÞ is the mass of the dissociated gas hydrate and mh0 is
sth Deff the initial mass of the gas hydrate. It is easy to show that in the case
¼ >> 1; t > 0 ð18Þ
sdif RðtÞdRðtÞ=dt of the dissociation of a spherical gas hydrate particle into ice and
gas, we have
is satisfied, then the concentration profile cðr; tÞ can be considered
steady in the ice layer at each time. This approach allows us to n3 ðtÞ
neglect the partial derivative with respect to time in the diffusion gdis ðtÞ ¼ 1  ; t P 0: ð27Þ
R30
Eq. (3). Further, we shall assume that the condition (18) holds.
634 V.A. Vlasov / International Journal of Heat and Mass Transfer 102 (2016) 631–636

Note that real experiments studied the dissociation of a powdery Table 1


gas hydrate into ice and gas, i.e., the dissociation into ice and gas The values of the quantities Deff and kdis at which there is good agreement between
the experimental [7] and calculated data for the dissociation of a powdery methane
of a large set of gas hydrate particles whose shape is close to spher- hydrate into ice and gas.
ical. In this case, if the quantity R0 is the characteristic average
radius of the gas hydrate particles forming the powder, then T; K Deff ; m2/s kdis ; m/s

Eq. (27) can also be applied. This makes it possible to compare 158 2:9  1014 > 2  109
14
the results of numerical calculations and the results of experiments 168 2:8  10 > 3  109
designed to study the kinetics of the dissociation of a powdery gas 178 2:5  1014 > 5  109
hydrate into ice and gas.
In the diffusion model presented in this work, there are three
main parameters that determine the kinetics of the process of Eq. (24). At such variation was considered (following the authors
gas hydrate dissociation into ice and gas: the extrinsic parameter of work [7]) that the rate of methane hydrate dissociation in the
ddis and the intrinsic parameters Deff and kdis . The parameter ddis selected temperature range (158–178 K) is limited only by the
determines the driving force of the process of gas hydrate dissoci- diffusion of methane through the formed ice layer. Table 1 shows
ation into ice and gas, whereas the intrinsic parameters Deff and kdis the values of the quantities Deff and kdis at which the calculated
determine the intensity of this process at the molecular level. Thus, curves were constructed in Fig. 2.
in the framework of the developed model, on the kinetics of gas As mentioned above, when receiving the calculated data shown
hydrate dissociation into ice and gas in the general case affects in Fig. 2, we assumed that the characteristic radius of the methane
both the intensity of the diffusion process of the gas through the hydrate particles in the experiments [7] was 17.5 lm, whereas, in
formed ice layer (the parameter Deff ) and the intensity of the gas reality, this radius varied over a wide range (from 10 to 25 lm). As
hydrate dissociation reaction on the ice–hydrate interface (the a result, the data presented in Table 1 should be regarded as esti-
parameter kdis ). In the particular case, as the analysis shows, one mates. In particular, this is evidenced by the fact that, according
of the parameters Deff or kdis has considerably more influence on to Table 1, the value of the diffusion coefficient Deff slightly
the kinetics of gas hydrate dissociation into ice and gas than the decreases with increasing temperature. However, this decrease
other. If the influence of the parameter Deff is much greater than may be attributed to the morphology of the formed ice layer. Note
the influence of the parameter kdis , then we can assume that the that the values of the diffusion coefficient Deff presented in Table 1
rate of gas hydrate dissociation into ice and gas is limited only close in the order of magnitude to the values of the diffusion coef-
by the diffusion of the gas through the formed ice layer. If the influ- ficient of methane in ice, which were calculated in the framework
ence of the parameter kdis greatly predominates over the influence of the method of molecular dynamics [48]. Also note that, as the
of the parameter Deff , then we can assume that the rate of gas analysis shows, the ratio sth =sdif decreases with increasing gdis ðtÞ.
hydrate dissociation into ice and gas is limited only by the gas Herewith, however, the condition (18) is not satisfied only for large
hydrate dissociation reaction on the ice–hydrate interface. values of the quantity gdis ðtÞ (at least this is true in the case of the
In work [7], the dissociation of powdery methane hydrate sam- experimental conditions [7] and for the values of the quantities Deff
ples into ice and gas at different temperatures and the Earth’s and kdis given in Table 1). Thus, we can often use Eq. (24) to simu-
atmosphere (p ¼ 0) was studied using the energy-dispersive late the kinetics of the dissociation of spherical gas hydrate parti-
X-ray diffraction method. The characteristic radius of the methane cles into ice and gas, thereby avoiding the numerical solution of
hydrate particles in a single sample ranged from 10 to 25 lm, and the system of Eqs. (3)-(5), (9), (10), (14) and (15).
these particles themselves were initially prepared from ice parti- The values of the quantities Deff and kdis depend on the temper-
cles. The latter fact suggests that the methane hydrate in the ature. This fact allows us to explain the anomalous preservation
experiments [7] had a porous structure [46,47]. Fig. 2 shows the thermal regime of methane hydrate, which was reported in works
experimental data [7]. Furthermore, Fig. 2 shows the calculated [39–42]. The monotonic increase of the rate of methane hydrate
data obtained for each fixed temperature using Eqs. (24) and (27) dissociation into ice and gas in the temperature range of
by assuming that R0 ¼ 17:5 lm and eh ¼ ei ¼ 0:15. These calcu- 193–240 K observed in these works is due to the fact that the rate
lated data, which are in good agreement with the experimental of methane hydrate dissociation into ice and gas in this tempera-
data [7], were obtained by varying the parameters Deff and kdis in ture range is limited only by the diffusion of methane through
the formed ice layer: the parameter Deff has a much greater
influence on the kinetics of methane hydrate dissociation into ice
178 K 168 K and gas than the parameter kdis . In the temperature range of
158 K
1.0 242–271 K, the intensity of the methane hydrate dissociation reac-
tion on the ice–hydrate interface significantly affects the kinetics of
0.8 methane hydrate dissociation into ice and gas. In this temperature
range, the rate of methane hydrate dissociation into ice and gas is
0.6
ηdis (t )

limited both by the diffusion of methane through the formed ice


0.4
layer and by the methane hydrate dissociation reaction on the
ice–hydrate interface: the influence of the parameter Deff on the
0.2 kinetics of methane hydrate dissociation into ice and gas is not
overwhelming compared to the influence of the parameter kdis . A
0
0 significant influence of the parameter kdis on the kinetics of
50 100 150 200 250 300
methane hydrate dissociation into ice and gas leads to a decrease
t, h
in the rate of methane hydrate dissociation into ice and gas in
Fig. 2. Experimental and calculated data for the dissociation of a powdery methane the temperature range of 242–271 K, which was observed in the
hydrate into ice and gas at atmospheric pressure (p ¼ 0). The symbols are works [39–42]. If we assume that the temperature of 241 K is the
experimental data [7] obtained at different temperatures (the characteristic radius temperature above which the anomalous preservation thermal
of the methane hydrate particles in a single experiment ranged from 10 to 25 lm);
regime of methane hydrate begins to occur, then the parameter
the solid lines are calculated data obtained using Eqs. (24) and (27), where
R0 ¼ 17:5 lm and eh ¼ ei ¼ 0:15. The values for the quantities Deff and kdis used in kdis can be explicitly calculated at this temperature in the frame-
the calculations are presented in Table 1. work of the developed diffusion model without an analysis of the
V.A. Vlasov / International Journal of Heat and Mass Transfer 102 (2016) 631–636 635

specific kinetic curve of methane hydrate dissociation into ice and where mi ðtÞ and mg ðtÞ are the masses of ice and the gas, respec-
gas. Using in the calculations the value of the parameter Deff equal tively, and nw ðtÞ and ng ðtÞ are the moles of water and the gas,
to the value of the molecular diffusion coefficient of methane in respectively. From Eqs. (A1)-(A3), it follows that
ice, taken from the work [48], we obtain by Eq. (24) that for  
Mg
methane hydrate kdis ¼ ð8  3Þ  108 m=s at a temperature of m0h ðtÞ ¼ mi ðtÞ 1 þ ; t P 0: ðA4Þ
nM w
241 K.
In the framework of the developed diffusion model, the anoma- Substituting Eq. (A4) into Eq. (26), we can obtain the relation
lous preservation thermal regime of gas hydrates may potentially  
mi ðtÞ Mg
be observed for all hydrate-forming gases, not only for methane. gdis ðtÞ ¼ 1þ ; t P 0: ðA5Þ
Herewith, the explanation for this regime will be similar to the mh0 nM w
above explanation for methane. However, in the framework of Taking into account the geometry of the problem under consid-
the developed diffusion model, the anomalous preservation ther- eration (Fig. 1), the quantities mi ðtÞ and mh0 can be represented as
mal regime of gas hydrates may not occur for some hydrate- follows:
forming gases because, for these gases at temperatures below
the ice melting point, the rate of gas hydrate dissociation into ice 4  3 
mi ðtÞ ¼ qi ð1  ei Þ p R ðtÞ  n3 ðtÞ ; t P 0; ðA6Þ
and gas is limited only by the diffusion of the gas through the 3
formed ice layer.
4 3
mh0 ¼ qh ð1  eh Þ pR ; ðA7Þ
3 0
5. Conclusions
where qi and qh are the mass densities of ice and the gas hydrate,
In the formulation of the diffusion model of gas hydrate disso- respectively.
ciation into ice and gas, it was not specified which kind of the From Eqs. (A5)–(A7), it follows that
hydrate-forming gas is considered. Thus, the diffusion model pre- sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 1
sented in this work is applicable for all hydrate-forming gases. In 3 q ð1  eh Þ Mg
RðtÞ ¼ h
1þ R30 gdis ðtÞ þ n3 ðtÞ; t P 0: ðA8Þ
this work, the general theory of chemical kinetics was used in qi ð1  ei Þ nM w
the development of the diffusion model of gas hydrate dissociation
into ice and gas. This allowed to take into account the intrinsic Taking into account Eq. (27) and given that qi ¼ xMw and
kinetics of the process of gas hydrate dissociation into ice and qh ¼ vMh , Eq. (A8) becomes Eq. (10).
gas. The developed diffusion model allows for simulation of the
gas hydrates self-preservation effect. In the framework of this Appendix B. Case in which RðtÞ ¼ const
model, the anomalous preservation thermal regime of gas hydrates
can be explained. It is important to note that the diffusion model If we can assume that RðtÞ ¼ R0 ¼ const, then it is necessary to
presented in this work is universal, i.e., it can be generalised and solve the system of Eqs. (3)-(5), (9), (14) and (15), where, in
applied to any geometry of the problem under consideration. Eqs. (3) and (5), RðtÞ is replaced by R0 . Solving this system of
Furthermore, the model can be generalised to the case in which equations in the framework of a quasi-stationary approximation,
the pressure of the hydrate-forming gas is not constant but we obtain that
increases as a result of the process of gas hydrate dissociation.
peq 1  2 
When creating the diffusion model of gas hydrate dissociation into ðR0  nðtÞÞ þ R  n2 ðtÞ
ice and gas, we assumed that the temperature at the front of the Z eq RTkdis v 2Deff 0
gas hydrate formation/dissociation reaction is unchanged. This 1  3 
 R  n3 ðtÞ
assumption is not always valid when considering the initial stage 3Deff R0 0
of the process of gas hydrate dissociation into ice and gas at high ddis t
temperatures (in this case, the dissociation occurs intensively). If ¼ ; t P 0: ðB1Þ
vð1  eh Þ
the assumption regarding the constancy of the temperature at
the front of the gas hydrate formation/dissociation reaction is not Solving Eq. (B1), it is possible to determine the value of the quantity
valid, then it is necessary to solve a diffusion problem involving nðtÞ at each time t .
the heat equation.
References

Acknowledgments [1] D.W. Davidson, S.K. Garg, S.R. Gough, Y.P. Handa, C.I. Ratclife, J.A. Ripmeester, J.
S. Tse, W.F. Lawson, Laboratory analysis of naturally occurring gas hydrate
from sediment of the Gulf Mexico, Geochim. Cosmochim. Acta 50 (1986) 619–
The author is grateful to Dr. A.N. Nesterov for a useful discus-
623.
sion of the gas hydrates self-preservation effect. [2] V.S. Yakushev, V.A. Istomin, Gas hydrate self-preservation effect, in: N. Maeno,
T. Hondoh (Eds.), Physics and Chemistry of Ice, Hokkaido University Press,
Sapporo, 1992, pp. 136–140.
Appendix A. Derivation of Eq. (10) [3] J.S. Gudmundsson, M. Parlaktuna, A.A. Khokhar, Storing natural gas as frozen
hydrate, SPE Prod. Facil. 9 (1994) 69–73.
During the process of gas hydrate dissociation into ice and gas, [4] S. Watanabe, S. Takahashi, H. Mizubayashi, S. Murata, H. Murakami, A
demonstration project of NGH land transportation system, in: Proceedings of
the following relations are satisfied: the 6th International Conference on Gas Hydrates, Vancouver, Canada, July 6–
m0h ðtÞ ¼ mi ðtÞ þ mg ðtÞ; t P 0; ðA1Þ 10, 2008.
[5] G. Rehder, R. Eckl, M. Elfgen, A. Falenty, R. Hamann, N. Kähler, W.F. Kuhs, H.
Osterkamp, C. Windmeier, Methane hydrate pellet transport using the self-
mg ðtÞ preservation effect: a techno-economic analysis, Energies 5 (2012) 2499–2523.
ng ðtÞ ¼ ; t P 0; ðA2Þ [6] S. Takeya, W. Shimada, Y. Kamata, T. Ebinuma, T. Uchida, J. Nagao, H. Narita, In
Mg situ X-ray diffraction measurements of the self-preservation effect of CH4
hydrate, J. Phys. Chem. A 105 (2001) 9756–9759.
mi ðtÞ [7] S. Takeya, T. Ebinuma, T. Uchida, J. Nagao, H. Narita, Self-preservation effect
nw ðtÞ ¼ nng ðtÞ ¼ ; t P 0; ðA3Þ and dissociation rates of CH4 hydrate, J. Cryst. Growth 237–239 (2002) 379–
Mw 382.
636 V.A. Vlasov / International Journal of Heat and Mass Transfer 102 (2016) 631–636

[8] T. Komai, S.-P. Kang, J.-H. Yoon, Y. Yamamoto, T. Kawamura, M. Ohtake, In situ [29] Y.-N. Lv, M.-L. Jia, J. Chen, C.-Y. Sun, J. Gong, G.-J. Chen, B. Liu, N. Ren, S.-D. Guo,
Raman spectroscopy investigation of the dissociation of methane hydrate at Q.-P. Li, Self-preservation effect for hydrate dissociation in water + diesel oil
temperatures just below the ice point, J. Phys. Chem. B 108 (2004) 8062–8068. dispersion systems, Energy Fuels 29 (2015) 5563–5572.
[9] C. Giavarini, F. Maccioni, Self-preservation at low pressures of methane [30] V.P. Mel’nikov, L.S. Podenko, A.N. Nesterov, A.O. Drachuk, N.S. Molokitina, A.M.
hydrates with various gas contents, Ind. Eng. Chem. Res. 43 (2004) 6616–6621. Reshetnikov, Self-preservation of methane hydrates produced in ‘‘dry water”,
[10] W. Shimada, S. Takeya, Y. Kamata, T. Uchida, J. Nagao, T. Ebinuma, H. Narita, Dokl. Chem. 466 (2016) 53–56.
Texture change of ice on anomalously preserved methane clathrate hydrate, J. [31] S. Takeya, T. Uchida, J. Nagao, R. Ohmura, W. Shimada, Y. Kamata, T. Ebinuma,
Phys. Chem. B 109 (2005) 5802–5807. H. Narita, Particle size effect of CH4 hydrate for self-preservation, Chem. Eng.
[11] A.G. Ogienko, A.V. Kurnosov, A.Y. Manakov, E.G. Larionov, A.I. Ancharov, M.A. Sci. 60 (2005) 1383–1387.
Sheromov, A.N. Nesterov, Gas hydrates of argon and methane synthesized at [32] D. Sun, Y. Shimono, S. Takeya, R. Ohmura, Preservation of carbon dioxide
high pressures: composition, thermal expansion, and self-preservation, J. Phys. clathrate hydrate at temperatures below the water freezing point under
Chem. B 110 (2006) 2840–2846. atmospheric pressure, Ind. Eng. Chem. Res. 50 (2011) 13854–13858.
[12] O.S. Subbotin, V.R. Belosludov, T. Ikeshoji, E.N. Brodskaya, E.M. Piotrovskaya, V. [33] H. Mimachi, S. Takeya, A. Yoneyama, K. Hyodo, T. Takeda, Y. Gotoh, T.
Sizov, R.V. Belosludov, Y. Kawazoe, Modeling the self-preservation effect in gas Murayama, Natural gas storage and transportation within gas hydrate of
hydrate/ice systems, Mater. Trans. 48 (2007) 2114–2118. smaller particle: size dependence of self-preservation phenomenon of natural
[13] S. Takeya, J.A. Ripmeester, Dissociation behavior of clathrate hydrates to ice gas hydrate, Chem. Eng. Sci. 118 (2014) 208–213.
and dependence on guest molecules, Angew. Chem. Int. Ed. 47 (2008) 1276– [34] V.P. Melnikov, A.N. Nesterov, A.M. Reshetnikov, A.G. Zavodovsky, Evidence of
1279. liquid water formation during methane hydrates dissociation below the ice
[14] G. Zhang, R.E. Rogers, Ultra-stability of gas hydrates at 1 atm and 268.2 K, point, Chem. Eng. Sci. 64 (2009) 1160–1166.
Chem. Eng. Sci. 63 (2008) 2066–2074. [35] H. Ohno, I. Oyabu, Y. Iizuka, T. Hondoh, H. Narita, J. Nagao, Dissociation
[15] A. Falenty, W.F. Kuhs, ‘‘Self-preservation” of CO2 gas hydrates–surface behavior of C2H6 hydrate at temperatures below the ice point: melting to
microstructure and ice perfection, J. Phys. Chem. B 113 (2009) 15975–15988. liquid water followed by ice nucleation, J. Phys. Chem. A 115 (2011) 8889–
[16] S. Takeya, J.A. Ripmeester, Anomalous preservation of CH4 hydrate and its 8894.
dependence on the morphology of hexagonal ice, ChemPhysChem 11 (2010) [36] V.A. Vlasov, A.G. Zavodovsky, M.Sh. Madygulov, A.N. Nesterov, A.M.
70–73. Reshetnikov, Pulsed NMR investigation of the supercooled water–gas
[17] H. Ohno, H. Narita, J. Nagao, Different modes of gas hydrate dissociation to ice hydrate–gas metastable equilibrium, Russ. J. Phys. Chem. A 87 (2013) 1789–
observed by microfocus X-ray computed tomography, J. Phys. Chem. Lett. 2 1792.
(2011) 201–205. [37] A.O. Drachuk, V.P. Melnikov, N.S. Molokitina, A.N. Nesterov, L.S. Podenko, A.M.
[18] H. Ohno, O. Nishimura, K. Suzuki, H. Narita, J. Nagao, Morphological and Reshetnikov, A.Yu. Manakov, Dissociation behavior of ‘‘dry water” C3H8
compositional characterization of self-preserved gas hydrates by low-vacuum hydrate below ice point: effect of phase state of unreacted residual water on
scanning electron microscopy, ChemPhysChem 12 (2011) 1661–1665. a mechanism of gas hydrates dissociation, J. Energy Chem. 24 (2015) 309–314.
[19] S. Takeya, A. Yoneyama, K. Ueda, H. Mimachi, M. Takahashi, K. Sano, K. Hyodo, [38] M.Sh. Madygulov, A.N. Nesterov, A.M. Reshetnikov, V.A. Vlasov, A.G.
T. Takeda, Y. Gotoh, Anomalously preserved clathrate hydrate of natural gas in Zavodovsky, Study of gas hydrate metastability and its decay for hydrate
pellet form at 253 K, J. Phys. Chem. C 116 (2012) 13842–13848. samples containing unreacted supercooled liquid water below the ice melting
[20] V.E. Nakoryakov, S.Ya. Misyura, The features of self-preservation for hydrate point using pulse NMR, Chem. Eng. Sci. 137 (2015) 287–292.
systems with methane, Chem. Eng. Sci. 104 (2013) 1–9. [39] L.A. Stern, S. Circone, S.H. Kirby, W.B. Durham, Preservation of methane
[21] A. Falenty, W.F. Kuhs, M. Glockzin, G. Rehder, ‘‘Self-preservation” of CH4 hydrate at 1 atm, Energy Fuels 15 (2001) 499–501.
hydrates for gas transport technology: pressure–temperature dependence and [40] L.A. Stern, S. Circone, S.H. Kirby, W.B. Durham, Anomalous preservation of pure
ice microstructures, Energy Fuels 28 (2014) 6275–6283. methane hydrate at 1 atm, J. Phys. Chem. B 105 (2001) 1756–1762.
[22] H. Mimachi, M. Takahashi, S. Takeya, Y. Gotoh, A. Yoneyama, K. Hyodo, T. [41] L.A. Stern, S. Circone, S.H. Kirby, W.B. Durham, Temperature, pressure, and
Takeda, T. Murayama, Effect of long-term storage and thermal history on the compositional effects on anomalous or ‘‘self” preservation of gas hydrates, Can.
gas content of natural gas hydrate pellets under ambient pressure, Energy J. Phys. 81 (2003) 271–283.
Fuels 29 (2015) 4827–4834. [42] W.F. Kuhs, G. Genov, D.K. Staykova, T. Hansen, Ice perfection and onset of
[23] D. Bai, D. Zhang, X. Zhang, G. Chen, Origin of self-preservation effect for anomalous preservation of gas hydrates, Phys. Chem. Chem. Phys. 6 (2004)
hydrate decomposition: coupling of mass and heat transfer resistances, Sci. 4917–4920.
Rep. 5 (2015) 14599. [43] V.A. Vlasov, Phenomenological diffusion theory of formation of gas hydrate
[24] S. Takeya, S. Muromachi, Y. Yamamoto, H. Umeda, S. Matsuo, Preservation of from ice powder, Theor. Found. Chem. Eng. 46 (2012) 576–582.
CO2 hydrate under different atmospheric conditions, Fluid Phase Equilib. 413 [44] V.A. Vlasov, Diffusion model of gas hydrate formation from ice, Heat Mass
(2016) 137–141. Transfer 52 (2016) 531–537.
[25] S.Y. Misyura, I.G. Donskoy, Dissociation of natural and artificial gas hydrate, [45] V.A. Vlasov, Formation and dissociation of gas hydrate in terms of chemical
Chem. Eng. Sci. 148 (2016) 65–77. kinetics, React. Kinet. Mech. Catal. 110 (2013) 5–13.
[26] A. Hachikubo, S. Takeya, E. Chuvilin, V. Istomin, Preservation phenomena of [46] D.K. Staykova, W.F. Kuhs, A.N. Salamatin, T. Hansen, Formation of porous gas
methane hydrate in pore spaces, Phys. Chem. Chem. Phys. 13 (2011) 17449– hydrates from ice powders: diffraction experiments and multistage model, J.
17452. Phys. Chem. B 107 (2003) 10299–10311.
[27] H. Sato, H. Sakamoto, S. Ogino, H. Mimachi, T. Kinoshita, T. Iwasaki, K. Sano, K. [47] W.F. Kuhs, D.K. Staykova, A.N. Salamatin, Formation of methane hydrate from
Ohgaki, Self-preservation of methane hydrate revealed immediately below the polydisperse ice powders, J. Phys. Chem. B 110 (2006) 13283–13295.
eutectic temperature of the mother electrolyte solution, Chem. Eng. Sci. 91 [48] T. Ikeda-Fukazawa, K. Kawamura, T. Hondoh, Mechanism of molecular
(2013) 86–89. diffusion in ice crystals, Mol. Simul. 30 (2004) 973–979.
[28] A.S. Stoporev, A.Yu. Manakov, L.K. Altunina, A.V. Bogoslovsky, L.A. Strelets, E.
Ya. Aladko, Unusual self-preservation of methane hydrate in oil suspensions,
Energy Fuels 28 (2014) 794–802.