A near-surface soil-gas geochemical survey was executed in conjunction with the Phase II experiment at the North Knobs, Wyoming, GR&DC-DOE UCG facility from July 22, 1981, through December 12, 1981.

The soil-gas detection method offers a unique new technique for locating potential gas leakage areas before any significant migration avenues can develop. This approach has the advantage over atmospheric measurements because the soil gases are not greatly affected by winds and other varying meteorological conditions; therefore a much lower level anomaly may be resolved.

One hundred and twenty-two, 18 foot deep permanent sites, based upon the local geology, were located and installed over the areas of both the previous Phase I and the current Phase II retorts. All sites were routinely monitored for methane, ethane, propane, i-butane, n-butane, ethylene, propylene, helium, hydrogen, carbon dioxide and carbon monoxide. Selected sites were periodically monitored for carbon disulphide, carbonyl sulphide, hydrogen sulphide and sulphur dioxide.

The survey demonstrates that residual gases from the Phase I burn are still present in the near surface and product gases generated during the Phase II burn were clearly evident. Maps are presented showing the areal distribution of these products during: (1) the pre-burn, pre-pressure period; (2) the air-pressured interval during system checkout; (3) the ignition and early burn period; and (4) the late-burn and early post-burn interval.
These maps and the tabulated data indicate a recognizable surface response occurred 3-6 days after changes in the burn retort pressures.
Product gases migrated into the near surface both vertically and laterally along well developed cross-joints and near-strike joint sets within the sandstones stratigraphically above the coal seam being burned. Additional leakage is probably associated with poor cementing of the casing in some of the deep boreholes. Casing leakage explains most of those anomalies located in the rock sequence stratigraphically below the coal. While throughout the survey abnormally high values were found in the near surface at many geochemical sites, ambient air samples routinely taken near those sites consistently had low values. It is concluded that a properly designed and operated UGG facility would not experience any adverse product gas leakage and would pose no hazard problems from either a safety or environmental point of view.

Introduction

The North Knobs UCG facility is located approximately eight miles west of Rawlins, Wyoming, in south-central Wyoming (Figure 1). It is situated on the southwest flank of the asymmetrical Rawlins uplift adjacent to the Washakie Basin. Throughout the area, the well exposed resistant sandstones all exhibit a remarkably consistent and well developed near-rectilinear joint pattern. The dominant (systematic) joint set strikes about 6º (N14W) from the strike of the beds (N20ºW) and is well exposed throughout the area. A well developed, but subordinate, non-systematic cross-joint set having a strike of approximately N49ºE is also present in all exposures. Figure 2, modified from McCurdy, schematically illustrates the rectilinear joint system and proposes a mechanism for possible migration, to the surface, of leaked gas from the coal burn. Limited trenching across some of the coal seams and adjacent sedimentary units and examination of these trench exposures by McCurdy indicate that the joint sets observed within the sandstones are pervasive in all of the rock units within the project boundaries. Detailed examination of the coals exposed by trenching also showed well developed butt and cleat faces having a close strike and dip relationship to the joint sets found in the well exposed sandstone outcrops.

The objectives for this survey were to determine if gases generated during the burn leaked into the near surface, and if so, their rates of leakage, migration paths and the composition of those gases, in order to evaluate the economic importance of such leakage and to assess their hazard potential from the standpoint of human safety and the environment.

Description of Sampling Stations

All geochemical sites were augered with a 3 inch auger to a nominal depth of 18 feet. Groundwater was not encountered in any of the sites drilled. Each site was established as a "permanent" observation site by installing a 20 foot length of 1 inch ID PVC pipe which was perforated with about 30 one-quarter inch diameter holes in the lower 1-1 1/2 feet of the pipe. During installation, sufficient pea gravel was poured in to fill the lower 2 feet. This pea gravel provides a permeable zone for collection of soil gases leaking from the adjacent formations. The remainder of the open hole was backfilled and tamped. Metal tags with the site number were then attached and a removable cap placed on the top of the PVC pipe.

After installation each site was allowed to stand for a minimum of 48 hours before sampling, thus permitting the indigenous gases remaining in the hole after drilling to come to equilibrium. All sites were then sampled at least twice before any test pressuring or other work on the facility system took place. Thus the composition and magnitudes obtained from these samples provide a set of "baseline" data, giving values at each site prior to the Phase II burn. Residual leakage products in the near surface resulting from the experimental Phase I burn cannot be excluded. A second full suite of samples (generally more than one from each site) was also taken during the Phase II pre-burn air-pressuring of the production facilities. During this period air pressures reaching 700-800 pounds per square inch were applied to the system, including the focal point in the coal seam. Geochemical near-surface soil-gas sampling during this test period permitted a preview, under maximum operating conditions, of the effective transmissability of the residual gases through the rocks surrounding the Phase II burn, including an opportunity to look for, prior to the ignition of the burn, any possible preferred migration paths through which later product gases might travel and escape to the surface. After ignition of the burn a third sequence of sampling was initiated. This sampling period extended throughout the full interval of the production burn and continued during the shutdown and post-burn period until December 12, 1981.
Sampling during the burn and post-burn phase of the UCG test included periodic resampling of all geochemical sites, with more frequent resampling of those sites that showed changes in composition and/or concentration. The post-burn sampling was carried out in order to see how long the effects of the burn could be detected in the near surface and/or to evaluate the rate of decline of values resulting from the burn.

Analytical Measurements

The analytical measurements were made with three GS&T mobile field trucks, which were equipped with a flame ionization detector (FID) gas chromatograph specially constructed by GS&T for light hydrocarbon measurements and an infrared CO2 detector. Only one of the trucks contained a dual FID-TC (thermal conductivity) gas chromatograph which can measure helium and hydrogen in addition to light hydrocarbons. This truck was also additionally equipped with an IR CO detector and a flame photometric (FPD) gas chromatograph for measurement of sulphur gases COS, H2S, CS2 and SO2. During the course of this experiment it was established that the hydrocarbons were completely adequate to define all leakage avenues.

Division of Data in Time

An examination of the change in methane flux from various sites suggests that the data can initially be divided into at least four discrete periods for mapping and discussion purposes: (1) Pre-pressure -- July 22, 1981, through August, 16, 1981; (2) pressure -- August 17, 1981, through August 23, 1981; (3) burn -- August 24, 1981, through November 10, 1981; (4) post-burn -- November 11, 1981, through December 12, 1981 (end of field survey). During field operations it was observed that it took about 3 to 5 days after the beginning of the system air-pressure test, or ignition of the coal, before any significant increases in the magnitude of the hydrocarbon gases were recognized in the near surface.

The locations of the sample sites are shown in Figure 3 and the surface geology in Figure 4. Plots of selected methane magnitudes with time are shown in Figures 5-9 for sites (27, 22), (1, 2), (5, 6), (13, 20), and (17, 19), respectively. These sites were selected to represent the typical changes noted in time response of gas migration. Note that sites 22 and 27 exhibit a very quick response to the retort pressuring which is almost identical even though they are about 50 feet apart. Site pairs 1 and 2, 5 and 6, 13 and 20, and 17 and 19 also show similar response to their close neighbor, but quite different response from other pairs which are several hundred feet apart. The leakage patterns which emerge over time are clearly not random, but are systematically changing in relation to subsurface controls.

Propane leakage is probably more related to migration from the retort of the initial products in ignition because propane is used to achieve ignition and is not a major retort gas generated during the burn. Figure 10 best illustrates this since site 22 is directly updip from the retort and clearly shows a sharp rise in propane upon pressuring and a fairly rapid decrease during the initial phases of the burn. Based on these gas flux data, the mapping windows were selected. These time windows are shown as vertical lines on Figures 5-13.

In addition to sites such as 22 which exhibit rapidly increasing magnitudes, followed by a decrease, two other types of flux behavior are noted. Sites 1 and 2 exhibit a large anomaly with no increase in magnitude, but only monotonically decreasing values in both methane (Figure 6) and propane (Figure 11). Small variations are caused by the sampling procedure occasionally removing enough gas (~ 2 1) to disturb the equilibrium. The other type of behavior is exhibited by sites 84 and 85 (Figure 12) which also show no sign of leakage from either burn. These nonanomalous sites are slowly pumped down by the sampling procedure.

Sites 86 and 88 (Figure 13) exhibit a very interesting response behavior. Both of these sites are located about 790 feet north of the Phase II retort. Site 86 was augered into a friable, very fine, very silty and slightly clayey sandstone stratigraphically only a few feet above the "G" coal seam. Site 88 is located about 70 feet north of site 86 and, while in the same gross lithologic unit, was augered into a somewhat harder and coarser facies. Comparison of the flux plots for these sites shows that both sites eventually reached comparable magnitudes. However, it is interesting to note that the response buildup at site 86 took place nearly 4 to 5 days before comparable buildup at site 88. The reason for this difference is not clear, but may be due to the presence of more open fractures in the harder and coarser sandstones at site 86 as compared with the fine, silty and clayey sandstone at site 88.

Discussion of the Data

In making an analysis of all of the near- surface hydrocarbon data, it was thought best to prepare a series of maps based upon the "arrival" times in the near surface which occurred after each change in the burn system. A statistical analysis was made to determine when the effects of these changes were best recognized and the previously defined time intervals chosen. Comparison of these dates with the dates when each activity began, with the exception of the post-burn interval, indicates a realistic figure of ± 5 days for the response to be observed. In the case of the beginning of the post-burn period no profoundly observable cutoff date could be defined. From the data it was noticed that on, or about, September 26 most sample sites showed somewhat decreasing soil-gas values. This date is well in advance of burn shutdown, but because there was no large decrease in values after shutdown was initiated, it was thought best to include a map showing the more or less gradual decrease in soil-gas values observed over the period of September 30 through December 12. It is believed that the gradual decrease as observed is
due to a general lessening of air pressure on the retort after the burn was fully established. As an explanation for the lack of a recognizable decrease shortly after shutdown was accomplished, it is believed that sufficient gases were still in the retort area and the migration paths so completely saturated that "bleeding" of the product gases would continue for an extended period at the relatively low final retort pressure (~ 30 psi). It is planned to resample all sites again in July of 1982 to better ascertain the decline rate of the residual gases from the Phase II burn.

All data for this survey were obtained on a real time basis and coded for keypunching and subsequent computer manipulation and mapping as work progressed. The coded analytical data were checked daily for accuracy, and field maps prepared for guidance in determining those areas in which abnormally high values were encountered so as to increase monitoring time density for possible hazardous conditions. The field maps were also used to make preliminary studies of possible migration paths and to be sure that sites or areas of gradually increasing product gas concentrations would be recognized quickly.

Tables giving complete listings by site number for all hydrocarbon and non-hydrocarbon data are available upon request. This includes the mean values for the four time-windows used in construction of the interpretive contour maps included in this report.

Figures 14 through 33 are contoured magnitude maps of methane, propane, carbon monoxide, carbon dioxide, and hydrogen for the four time periods discussed above. In all cases it can be clearly seen that in the area surrounding the original Phase I burn that relatively high values of these gases were still present in the near surface. Detailed examination of these maps together with the geologic map (Figure 4) indicates that most of the higher value areas are located stratigraphically above the "G" coal, which was gasified. Also comparing the pre-burn maps with the facility installation map (Figure 34) high values seem to be clustered around or near the principal injection and product wells. This clustering around these wells may be due to leakage resulting from poor cement jobs on the wells. Of special interest is the obvious "streaming" in a northerly direction along the strike of, and mainly within, the friable sandstones overlying the "G" coal. This "streaming" within this horizon is a well developed feature recognizable on all of the geochemical maps. On the methane pressure period map, Figure 15, the high values are mainly restricted to the area surrounding the Phase I burn. Pressuring of the "G" coal was done through one of the Phase II injection wells. Increases in methane values largely took place in the area surrounding the Phase I burn, and to a lesser extent to the north, but again mainly in the overlying sandstones. This would indicate that there is a well developed migration path probably associated with the well developed fracture pattern discussed earlier. Of special interest is site 86 near the north boundary of the facility.
The noticeable increase would indicate that, even during the Phase I burn, product gases preferentially migrated updip to the east and then along strike and were still present in those rocks nearly a year after shutdown of that burn. Again while diffusion through the sandstone overlying the "G" coal cannot be ruled out, it is believed that migration northward along the well developed dominant "strike" joint set and upward along the subordinate cross-joint set provides the major migration paths from the product source. With reference to Figure 16, showing methane values after ignition of the Phase II burn, there is a real increase in the methane magnitudes in the area surrounding the Phase I burn. This increase supports the conclusion that there are well developed avenues of communication within the overlying sandstones (probably fracture zones). The northerly trending "streaming" through this sandstone is obvious, and a study of the unusually high methane values obtained at sites 86 and 88 again indicate that well developed jointing provides the major avenues for product gas migration from the gasification retort. On this map the slight increase in methane values vertically over the retort as compared to the increases seen along the outcrop of the sandstone overlying the "G" coal strengthens the argument that diffusion as a transport mechanism is of only minor importance as compared to migration associated with fracturing. This is particularly so because of the apparent lack of "streaming" of the leaked products to the north along the beds vertically over the burn. The "Wally" coal and its associated bone and shales, while probably fractured, do not have the open fractures present in the sandstones immediately above the "G" coal. The somewhat higher values observed in the line of sites 33 through 39 and in the vicinity of sites 96 and 97 are thought to be most likely due to leakage of product gases into more porous and fractured sandstones above the "Wally" coal bed resulting from migration associated with possible imperfect cementing of the casings from any one, or a combination of, Injection Wells IW-2-2, IW-2-3 or the slant drilled Injection Well IW-2-1 or the Production Gas Well PGW 2-1.

The anomalous values observed to the east of the "G" coal and stratigraphically below it, particularly in the vicinity of sites 25 and 26 as well as sites 102 through 105, are also possibly due to leakage associated with imperfect cementing of Injection Wells IW-2-1 and AIW-1. Except for these two more or less restricted anomalies, very little leakage was observed in any of the rocks stratigraphically below the "G" coal.

The "post-burn" methane map (Figure 17) is in general similar to the burn map (Figure 16). As mentioned earlier, this map includes some late burn data as well as the actual post-burn values through December 12, 1981. There was no clear-cut break in the values observed during this period. After the burn was well established and the working pressures on the gasification system could be lowered, there appeared to be a slight decline in the magnitudes of leaked gases in the near surface. The lack of a clear-cut decline after burn shutdown can be explained by the fact that
the affected sediments carrying the product gases were already near saturation. When the pressures on the retort were removed, the drive "forcing" the gases to the surface no longer existed, and the gases remaining within the reservoir continued to migrate, but at a slower rate than when under pressure. The presence of relatively high values of product hydrocarbon gases remaining in the area surrounding the Phase I burn nearly a year after its shutdown, as shown on Figure 14, supports this conclusion.

Maps for propane, CO, CO2. and H2 (Figures 18 through 33) are included with this report and show fundamentally the same behavior characteristics as methane.

Conclusions

The near-surface geochemical survey conducted at the North Knobs, Wyoming, UCG facility has shown: (1) There was clear evidence that there was leakage of gases, generated during the in situ combustion of "G" coal, into the near surface and into the atmosphere. (2) Migration of these product gases took place mainly within the friable and fractured sandstones overlying the burned "G" coal seam. Migration took place laterally mainly through the dominant "strike" joint set and vertically along the subordinate "cross-joint" set. Transport by diffusion cannot be ruled out; however, it is believed to be of secondary importance. Studies of probable diffusion rates may provide more definitive data concerning the relative importance of this transport mechanism. There is evidence based upon localized high value "hot spots" indicating that some leaked products reached the surface through pathways associated with imperfect cementing of the casings in various facility boreholes. (3) The geochemical survey shows that while methane was the main product generated as a result of the burn, there were also associated with it significant amounts of ethane, propane, iso-butane, normal butane, CO, CO2. and H2. Only relatively small amounts of ethylene and propylene were observed. The compounds, methane, hydrogen, carbon dioxide, and carbon monoxide, were the principal products. Lesser, but measurable, values for carbon disulphide, carbonyl sulphide, hydrogen sulphide and sulphur dioxide were observed. Only minor amounts of helium were found and were a result of helium experiments used to measure the residence time of gases injected into the retort. (4) The survey showed that the leaked gases migrated to the surface in from 3 to 5 days after the burn was ignited. This rate of movement indicates that migration took place along more or less well developed pathways (jointing). (5) While the observed magnitudes of product gases in the near surface were fairly large in some zones and localized areas, their values were not unexpected and fall within the range of natural seeps (Jones and Drozd, 1982). Figure 35 illustrates a natural seep observed in a pristine area just south of Glacier National Park (Jones and Pirkle, 1981). The total amount of product lost through leakage must be considered small when related to the total volume generated in the retort. From an economic point of view the loss through leakage is probably well within acceptable limits. With respect to the non-hydrocarbon gases, preliminary study of the data indicates abnormal values of these gases in some sites; however, it should be pointed out that ambient air analyses taken throughout the survey never showed values of these toxic gases approaching significant or dangerous levels. For these reasons it is believed that the minor amounts of toxic gases generated in a well engineered and operated in situ coal gasification facility would not have any adverse effect from either a safety or environmental point of view.

References

Jones, V. T., and Drozd, R. J., "Predictions of Oil or Gas Potential by Near-Surface Geochemistry," manuscript submitted to AAPG, 1982.

Jones, V. T., and Pirkle, R. J., "Helium and Hydrogen Anomalies Associated with Deep or Active Faults," paper presented ACS, March, 1981, Atlanta, Georgia.

McCurdy, R., "Site Qualification Studies of the UCG-SDB Site, North Knobs, Wyoming," prepared by GS&T under DOE contract DE-ACO3-77-ET13108, May 23, 1979, C&M Division Report 64RK076.