Thermal conductivity of hydrates

Stoll and Bryan (1979) first measured the thermal conductivity of propane hydrates (0.393 Wm-1K-1 at T = 215.15 K) to be a factor of 5 less than that of ice (2.23 Wm-1K-1). The low thermal conductivity of hydrates, as well as similarities of the values for each structure (shown in Table 2.8) have been confirmed from numerous studies (Cook and Leaist, 1983 [0.45 Wm-1K-1 for methane hydrate at 216.2 K] Cook andLaubitz, 1981 Ross et al., 1981 Ross and Andersson, 1982 Asher et al., 1986 Huang and Fan, 2004 Waite et al., 2005). The thermal conductivity of the solid hydrate (0.50-0.58 W m-1 K-1) more closely resembles that of liquid water (0.605 W m-1 K-1). 

O Pure Ice, Waite etal., 2005 NIST, pure water 

X Propane hydrate, Stoll etal., 1979 CH4 hydrate+SDS (unimpacted), Huang etal., 2004 

X CH4 hydrate (10% porosity), Waite et al 2005 CH4 hydrate (inverse-modeling), Gupta et al., 2006 

FIGURE 2.17 Thermal conductivity of gas, water, ice, and hydrates (a) without and (b) with unconsolidated sediment (Gupta, 2007). 

---

The determination of in situ hydrates spawned a wave of research to measure hydrate properties needed for geological research and gas recovery. Several measurements were made of sonic velocity and thermal conductivity of hydrates in sediments (e.g., Stoll and Bryan, 1979 Pearson et al., 1984 Asher, 1987 Waite et al., 2005), while others measured the calorimetric properties (e.g., Rueff, 1985 Handa, 1986a,b,c,d Rueff et al 1988) needed to estimate dissociation energy. Davidson (1983) summarized hydrate properties as being similar to ice, with a few notable exceptions. Chapter 2 presents comparisons of physical property measurements of ice and hydrate. 

Several models have been proposed to estimate the thermal conductivity of hydrate/gas/water or hydrate/gas/water/sediment systems. The most common are the classical mixing law models, which assume that the effective properties of multicomponent systems can be determined as the average value of the properties of the components and their saturation (volumetric fraction) of the bulk sample composition. The parallel (arithmetic), series (harmonic), or random (geometric) mixing law models (Beck and Mesiner, 1960) that can be used to calculate the composite thermal conductivity (kg) of a sample are given in Equations 2.1 through 2.3. 

Uddin et al. (2008b) conducted several depressurization simulations for the Mallik 5L-38 well. Their results showed that the Mallik gas hydrate layer with its underlying aquifer could yield significant amounts of gas originating entirely from gas hydrates, the volumes of which increased with the production rate. However, large amounts of water were also produced. Sensitivity studies indicated that the methane release from the hydrate accumulations increased with the decomposition surface area, the initial hydrate stability field (P-T conditions), and the thermal conductivity of the formation. Methane production appears to be less sensitive to the specific heat of the rock and of the gas hydrate. 

Consider the ten year cumulative gas production prediction of the JOE model shown in Figure 7.46 (note the logarithmic scale of both axes). From the figure it is clear that hot water circulation alone will not be productive for a period after 0.02 years, due to the low thermal conductivity of the hydrates and sediments. However, depressurization does appear to be a favorable production mechanism, comparing favorably to hot water circulation with reduce bottom hole pressure, or partial hot water injection. 

R. Inoue, H. Tanaka, and K. Nakanishi, "Molecular dynamics simulation study on the anomalous thermal conductivity of clathrate hydrates , J. Chem. Phys. 104 (1996) 9577. 

Physical Mechanisms. The simplest interpretation of these results is that the transport coefficients, other than the thermal conductivity, of the water are decreased by the hydration interaction. The changes in these transport properties are correlated the microemulsion with compositional phase volume 0.4 (i.e. 60% water) exhibits a mean dielectric relaxation frequency one-half that of the pure liquid water, and ionic conductivity and water selfdiffusion coefficient one half that of the bulk liquid. In bulk solutions, the dielectric relaxation frequency, ionic conductivity, and self-diffusion coefficient are all inversely proportional to the viscosity there is no such relation for the thermal conductivity. The transport properties of the microemulsions thus vary as expected from simple changes in "viscosity" of the aqueous phase. (This is quite different from the bulk viscosity of the microemulsion.)... 

While water content of the stratum corneum affects permeability of the tissue, hydration also impacts various physical properties of the membrane such as tensile strength and elasticity, modihes the microenvironment for microorganisms on the tissue surface, alters the thermal conductivity of the tissue and also affects skin appearance. Further, increasing hydration also alters the thickness of the stratum corneum as shown in Figure 4, again taken from data provided by Blank et al. (1984). 

The measured power decreases steadily with tintie for the first 400 days after the beginning of the automatic operation. This is consistent with the progressive drying of the inner annulus of the barrier and the associated decrease of thermal conductivity of the bentonite. The slight increase in power in the second part of the period represented is attributed to the progressive hydration of the barrier due to the incoming water. Some predictions reproduce accurately the observed behavior. 

Aihara, Y, Sonai, A., Hattori, M., and Hayamizu, K. 2006. Ion conduction mechanisms and thermal properties of hydrated and anhydrous phosphoric acids studied with H, H, and P NMR. J. Phys. Chem. B 110 24999-25006. 

The thermal conductivity of gas hydrates is dependent on temperature, but has no pressure dependence. Table 10.4 shows the thermal conductivities of ice, water, CO2 hydrates, and methane hydrates. 

It is clear that the thermal conductivity of gas hydrates is much less than that of ice, but similar to hquid water. Furthermore, when it comes to hydrate/gas/water or hydrate/gas/water/sediment systems, the thermal properties are usually determined as the average values of the properties of the components by considering their saturation (volumetric fraction) in the sample. Because of the paucity of data of CO2 hydrates, the heat capacities of ice, methane and ethane hydrates are shown in Table 10.5. Considering the similarity between CO2 hydrates and other gas hydrates, the heat capacity of CO2 hydrates is certainly less than that of liquid water and may be similar to that of ice. 

Table 1 Experimental Thermal Conductivity of Ice Ih and Selected Clathrate Hydrates...

---