Adsorption capacity of the Velenje lignite: methodology and equipment Adsorptivnost velenjskega lignita: metodologija in oprema Janja Žula1-*, Jože Pezdič2, Simon Zavšek1, Edi Burič1 Abstract: Laboratory study of adsorption-desorption characteristics of various lignite lithotypes is very important for prevention of coal dust and gas outbursts that represent a dangerous and unpredictable phenomenon in underground mining of the Velenje lignite. The performed study consists of adsorption investigations carried out in the geotech-nical laboratory of the Velenje Coal Mine (VCM). Lignite samples contained in the reaction cell were exposed for a limited time to a gas pressure of up to 100 bar. The measurements were performed on an updated instrument according to the improved volumetric gas sampling method. The gas was sampled from the reaction cell during the sorption simulation. Updating of the instrument affected positively the data quality. The performed three sets of measurements led to results that were crucial for further research. In the experiment, the amount of adsorbed and desorbed gas was calculated from the pressure difference in the known cell volume. The obtained results could permit the risk assessment of possible coal dust and gas outbursts. The performed tests shall enable a more exact determination and understanding of gas effects, possibly add some new knowledge to understanding of the dangerous phenomena, maybe also improve their prediction and lead to better safety in mining. Izvleček: Laboratorijske raziskave adsorpcijsko-desorpcijskih lastnosti različnih litotipov lignita v Premogovniku Velenje so zelo pomembne pri raziskovanju in preprečevanju izbruhov plina in premogovega prahu, ki so nevarni in nepredvidljivi pojavi pri rudarjenju v omenjenem premogovniku. * 1Velenje Coal Mine, Velenje, Slovenia 2GEORIS, Radovljica, Slovenia Corresponding author. E-mail: janja.zula@rlv.si Received: May 23, 2011 Accepted: June 15, 2011 Raziskovanje je zajemalo meritve sorpcijskih lastnosti lignita v ge-otehničnem laboratoriju Premogovnika Velenje, kjer so bili različni vzorci lignita zaprti v reakcijski celici in obremenjeni z do 100 bar plina za določen čas. Vzpostavili smo sistem meritev na posodobljeni napravi, izpopolnili volumetrično metodo in način vzorčevanja plina iz reakcijske celice med potekom simulacije sorpcije. Posodobitev naprave je vplivala na kvaliteto pridobljenih podatkov. Izvedli smo tri sklope meritev, ki so pokazali odločilne rezultate za nadaljnje raziskave. Po končanih meritvah sorpcije smo matematično določili količino adsorbiranega in desorbiranega plina iz razlike tlaka plina v znanem volumnu celice. S količino plina ocenimo stopnjo tveganja pri odko-pavanju različnih plasti premoga, kjer lahko pride do vdora plina. S temi poskusi bo mogoče bolj natančno opredeliti in razumeti plinske efekte v Premogovniku Velenje in morda tudi prispevati k razumevanju nevarnih dogodkov, njihovemu morebitnemu napovedovanju in s tem k večji varnosti. Key words: sorption, lignite, lithotype, laboratory experiments, Velenje Coal Mine - Slovenia Ključne besede: sorpcija, lignit, litotip, laboratorijski poskusi, Premogovnik Velenje - Slovenija Introduction Various lignite lithotypes are of micropo-rous structure in which the gas occurs compressed in the pores, cleats and capillaries. The gas in these tiny spaces can be in gaseous, liquid or even solid state. The basic lignite lithotypes are determined macroscopically, and they can be recognized with an unaided eye. The lithotypes are named according to the classification of the International Committee for Coal and Organic Petrology (ICCP, 1993) (e.g. in Taylor et al. (1998), p. 280). A gas-mix accumulated in a coal, termed as the coalbed gas, can be of various origins and migration paths and this is also an outstanding fact in the case of the Velenje lignite. This thematic was thoroughly studied by Pezdic and his co-workers for almost 15 years - from 1998 onwards - and is in details well documented in yearly-made elaborations for the VCM company as "Monitoring of gas components" in the Velenje Coal Mine as well as published in papers by Kanduc et al. (2003), Kanduc (2004), and Kanduc & Pezdic (2005). The content presented in this paper is mainly a summary from the B.Sc. thesis work of the first author (Zula, 2006), which was also mentored by professor Pezdic. Coalbed gases, together with rock-mechanical properties of the lignite in different areas of the Velenje lignite mining are highly responsible for dynamical processes during advancement of mine workings. Among them, the most dangerous events are sudden coal-falls, gas exhalations, gas and lignite outbursts, and pillar chocks. This thematic was studied and evidenced for decades. More recently, it was summarized and interpreted in representative works of Likar (1995), Zavsek (2004) and in many working reports and studies cited by these two authors. In connection with rock-mechanics, Likar and his co-workers studied in the 1990 and 2000 also sorption of gases in lignite at different strain conditions (Pezdic et al. 1999 a, b). Results and long-term investigation are also archived at the VCM in the form of annual reports and elaborations. The coalbed gas occurs in pores and cleats of coal under the lithostatic pressure of rocks above the coal seam and under the pore pressure. When pressure decreases, the gas is released from the coal, and can migrate freely. Migration of gas is especially enhanced in tectonized zones along fault systems. Gases can be accumulated in the void volumes (pores, cleats, crushed zones, chambers) of various sizes both adsorbed on the surfaces and dissolved in water. Studying gases in coals and their adsorption/desorption behavior is of great interest in the last two decades especially in connection with efforts to assess potential of coals as a geological medium for eventual storage of CO2, particularly in combination with enhanced methane recovery. In this context, at least studies of sorption on different coals worldwide should be mentioned here: Litwiniszyn (1990), Gam-son et al. (1996), Pezdic et al. (1999b), Mastalerz et al. (2004), Majewska et al. (2009), Gruszkiewicz et al. (2009), weniger et al. (2010). In 2009, a special issue of the International Journal of Coal Geology (Vol. 77/1-2) was devoted to CO2 sequestration in coals and enhanced coalbed methane recovery. The origin of gases can be inferred from the isotope composition of its gas components. Such a pioneering study for the Velenje lignite was carried out by Kanduc & Pezdic (2005). On the basis of carbon isotope composition of carbon dioxide and methane they concluded that the following types of origin of CO2 and methane can be recognized: microbial methane and CO2 methane generated by microbial CO2 reduction and/or methane affected by processes of oxidation, and endogenic CO2. The isotope composition of coal-bed gases is affected by a range of factors: the original organic matter composition, temperature, Eh-pH conditions, microbiological activity, so called secondary processes (diffusion, migration, oxidation), and by mixing of gases of various origin (Kanduc & Pezdic (2005); and references there-in). Gases in the Velenje basin are of a highly variable composition both within the lignite seam and the non-coaly basin sediments. According to data from Kanduc & Pezdic (2005), the CDMI (carbon dioxide to methane index expressed as cp(CO2 )x 100/p(CO2 + CH4) varies between 80 % and 99 % in the lignite-seam gases (with very rare exceptions of the CDMI being considerably below 80 %), whereas this index is from 1 % to ca. 55 % in the gases as coming from the subsurface wells penetrating different basin sediments and the pre-Pliocene basement rocks. Considerably variable is also content of N2, i.e. between 0 % and almost 55 % (as reported in Kanduc & Pezdic (2005)), but this range is highly dependent on the sampling approach (see samples of the A and B types in Table 1 in Kanduc & Pezdic (2005)). It is evident that the gases in the Velenje lignite (as well as in other lithologies) have various origins and are affected by possible mixing and changes owing to different phys-icochemical properties of methane and CO2 during gas migration (Kanduc et al. (2003), Kanduc & Pezdic (2005)). Amount of the gas sorbed in the coal depends also on pressure, temperature, mineral matter, moisture and the coal lithotype. Rate of achieving the adsorption equilibrium depends on the sorption rate in pores and on gas diffusion into the porous coal matrix (Pezdic (1999), Kanduc (2004)). In closed systems, as e.g. in areas of still un-excavated and relatively undisturbed parts of coal/lignite seams, accumulations of gases at high pressures may occur. In a final stage, they may attain or exceed the value of the lithostatic pressure. Such circumstances are characteristic for zones ahead of mining longwalls and mine-road faces (Kanduc et al. (2003), Kanduc & Pezdic (2005)). Opening of such a zone is often a highly dynamic process that can lead to a coal-dust and gas outburst. As the total and partial pressure of the adsorbed gases decreases, this can affect a wider area of the coal seam. In dependence of conditions and properties of the coal matter, the gas desorption can be a sudden, momentary event (e.g. Likar (1995), Zavsek (2004)). Sorption method, equipment and measurements The quantity of adsorbed gas was determined according to the sorption method, as described in Pezdic et al. (1999b). For measuring the adsorption-desorption properties of various lignite lithotypes at high pressures (up to almost 100 bar), the volumetric method was applied. For calculating the quantities of adsorbed gas the theoretical bases were used as cited in Pezdic et al. (1999b) and in Zula (2006), taken mostly after Moore (1974). We proceeded from the general gas law (1) and the real gas law (2) (see e.g. in Lazarini & Brencic, 1992): PV = nRT (1) For gases deviating from ideal conditions the gas equation for real gases is used, the so-called van der Waals equation. (P + an2/V) (V- nb) = nRT (2) We calculate the volume of the adsorbed gas V after equation (3): (P0 - Pt) Vf = V P v 0 v free a ; (3) P0 is the initial pressure, Pt is the pressure determined after a certain time period t, P is 1.013 bar and V the A 'a iree volume in autoclave. Amount of the adsorbed gas is given in three different ways (4, 5, 6): = V /m (L/ kg) adsorb a material n = Vadsorb/Vo (mmol/kg) mgas = nM/1000 (g/kg) (4) (5) (6) Equipment The apparatus for gas sorption measurements has been set in the Geotech- nical laboratory of the VCM. Measurements of sorption were performed on the third modification of the high-pressure system as initially used by Pezdič in 1998 (Figure 1). The apparatus has been meanwhile improved by adding a new measurement registering system. Readings of pressure and temperature are now directly transformed to graphic signals that are continuously monitored on a display (Figure 2). In this way, by updating the computer program (Supovec & Filipic, 2004), detection of temperature and pressure signals became more accurate, since being performed in one second time intervals. This permitted registration of the most important data during the first minute after dosing, when adsorption is at the highest. Added were also two additional temperature sensors for registration of temperature in the system and in the room. The system temperature (T „ ) is measured on A v system7 the housing of autoclave with a temperature sensor (PT1000) fixed to the housing. An additional construction improvement was done on net volume of autoclave. The robe for pressure measurements was placed closer to autoclave, and the volume reduced by valve (V). The additional valve on the autoclave permits the extraction of gas during adsorption (Figure 3). Figure 1. Modernized apparatus for sorption measurements - autoclave, probe, valves, software application Figure 2. Computer output of pressure and temperature measurements (Supovec & Filipic, 2004). Manometer System with a valve (V ) Probe Valve for gas Reaction cell -autoclave Figure 3. System with the reaction cell Sampling of lignite Lignite was sampled in borehole cores, and from mine road and longwall faces (Table 1). Six distinct samples differing in their lithotype composition were prepared and classified according to the ICCP classification (1993) (in: Taylor et al. (1998), p. 280). Sample 1 was pure xylite (X), and samples 3 and 4 were xylite-rich and xylite very rich lignite lithotypes. Samples 2, 5 and 6 were lithotypes of fine detrital lignite. Among them, sample 2, which was taken from a fault zone, and sample 5, both exhibited a pronounced degree of gelification. Generally speaking, fine detrital and gelified lignite is more characteristic for the upper and the inner part of the Velenje lignite seam, whereas more or less xylite-rich lignite predominates in the lower and the outer part of the seam (Markic & Sachsen-hofer, 1997, 2010). In Table 1, petro-graphic composition is also expressed in terms of lithotype components (in percents and by code values) as introduced by Markic et al. (2001) and Markic & Sachsenhofer (2010). For a sorption measurement around 100 g of crushed lignite material of 2-4 mm grain size was used. Measurement process The whole sorption measurement process consisted of five steps. In the first step, after dosing to ca. 60 bar CO2, adsorption was measured for 45 min. In the following step, dosing to ca. 60 bar CO2 was repeated and adsorption was measured again for 45 min. After that, the type A gas sampling was performed. As during the gas sampling some pressure in the flask was lost, the missing CO2 was replaced. Into the equilibrated sample, we added nitrogen N2 overpressure up to a total pressure around 85 bar. Adsorption of the gas mixture (CO2+N2) lasted for additional six hours. After completed adsorption of the (CO2+N2) gas mixture, we performed the B type gas sampling and then exposed the sample to atmospheric pressure (1 bar). Desorption followed, and was measured for 15 h. Table 1. Samples chosen for sorption tests Sampling of gas components from Results an autoclave For gas sampling during sorption a spe- Because gas sorption m a coal is de- cial equipment shown in Figures 4, 5 was pendent on coal's so1id matter compo- used. Gas capture from an autoclave was moisture content, pre^me time a two-step procedure. First, the gas was and kind of gas dosed, and because gas captured into a piston ampoule (1 bar), comp°sition is a1ternating due to dos- and then transferred to a laboratory am- ing of different gases and due to adsorp- poule. Two types of sampling were differ- tion/desorption rfferts three sets of entiated, the type A and type B sampling. measurements were performed. The 1st set represents measurements at varying The type A sampling was carried out at ^^ures the 2nd set comprises meas- the end of the second CO2 adsorption urements on various Hgrnte 1ithotypes, phase i.e. after 90 min from the begin- and the 3rd set represents comparison ning of the adsorption measurements. of composition of gases as sampled at varying pressures (Zula, 2006). The type B sampling was capture of gas mixture (CO2 + N2) in autoclave. It First set of measurements was carried out 6 h after N2 dosing, just The 1st set of measurements was done before desorption. on sample No 3 (see Table 1) of the following lithotype composition: fD Gas analysis was performed on a 80 % dXxD 20 %. In order to study homemade NIER mass spectrometer at influence of moisture on sorption, the J. Stefan Institute in Ljubljana. the whole sample was split into three Figures 4. (left) and 5. (right): Gas sampling into a piston ampoule (1 bar) from autoclave (at 60 bar to 100 bar), and then transferring of gas to a laboratory ampoule. parts to measure moisture contents at three stages/conditions: • at the initial stage (sample-part S-1), • after moistening with 20 mL of distilled water, before sorption (sample-part M-1), • after moistening with 10 mL distilled water, after sorption (sample part M-2). Measurements were performed by a simple experiment. As gas, only CO2 was used. The autoclave with lignite sample was dosed by CO2 and left for 45 min for the adsorption to take place. The same time duration was applied for desorption. Table 2. Measurement results for S-1 The results are given in Tables 2-4: Moisture content at the as received (a. r.) basis of the "initial" sample-part S-1 was 50.00 % (Table 2). This moisture content is very close to a real bed-moisture content (at the ash-free basis) as typical for the Velenje lignite regarding its coalification rank (Markic & Sachsenhofer (2010), p. 141-153). Additional moistening of the sample-part M-1 with 20 mL distilled water (Table 3) resulted in an increase of moisture from 50.00 % (S-1) to 83.60 % (M-1). Additional moistening of the sample-part M-2 with "only" 10 mL distilled water Sample Test m(p1) (g) m(p2) (g) P0 (bar) ^^ads" (bar) Ads. (g/kg) ^Pdes" (bar) Des. (g/ kg) Moisturel (%) Moisture2 (%) S-1 a. r. moisture co2 57.52 57.62 52.69 -1.30 2.47 1.39 1.65 50.00 42.25 Table 3. Measurement results for M-1 Sample Test m(p1) (g) m(p2) (g) P0 (bar) ^Pads" (bar) Ads. (g/kg) ^Pdes" (bar) Des. (g/kg) Moisturel (%) Moisture 2 (%) M-1 a.r. moisture + 20 mL distilled water CO2 58.71 58.80 55.05 -4.05 7.48 2.71 5.16 83.60 81.02 Table 4. Measurement results for M-2 Sample Test m(p1) (g) m(p2) (g) P0 (bar) ^Pads. (bar) Ads. (g/kg) ^Pdes. (bar) Des. (g/ kg) Moisturel (%) Moisture 2 (%) M-2 a.r. moisture + 10 mL distilled water CO2 59.98 60.12 52.98 -2.78 5.10 2.51 4.61 59.74 59.49 (Table 4) resulted in an increase of moisture from 50.00 % (S-1) to "only" 59.74 % (M-2). Both additionally moistened sample-parts showed significantly larger adsorption and desorption than the initial sample (S-1) (Tables 2-4). CO2 adsorption in additionally moistened sample-parts M-1 and M-2 was 2 to 3 times greater (more effective) than in S-1, and similarly was desorption. After sorption, moisture content of the sample-parts was measured again ("moisture 2" data in Tables 2-4). After sorption, the moistures decreased as follows: in sample S-1 for 7.75 %, in sample (M-1) for 2.58 %, and in sample M-2 for 0.25 %. Second set of measurements Sorption measurements were performed on six samples (1, 2, 3, 4, 5, 6), differing in lithotype composition (Table 1). Detailed measurements for all samples are given in Žula (2006), whereas in this paper, measurements for all samples are given only graphically (Graphs 3, 4, 5), and only exam-pling tabular data will be presented and discussed (Table 5). On each sample, two identical measurements were done. Between the two measurements on the same sample, slight deviations appeared - most probably due to heterogeneity of samples, different initial sample weights, dosage velocities, amounts of dosed gas, slight temperature differences and slight differences in moisture. As an example, results for sample No 2 are presented in Table 5. Explanation of results in Table 5 and pressure conditions shown in Figures 6, 7, 8. • After first dose (56.50 bar CO2), lignite adsorbs in 45 min 3.65 bar or 6.65 g/kg (grams of gas per kilogram of coal). • After second dose (58.15 bar CO2), in the next 45 min, lignite adsorbs 1.28 bar or 2.33 g/kg. • In total it adsorbs 4.93 bar or 8.98 g/kg. • Third dose (58.60 bar CO2) represents a maximal filling of reaction cell with CO2. • The fourth stage is dosing 87.26 bar N2 onto equilibrated sample. At the first moment, in less than 1 minute, the total pressure decreases for 2.15 bar, but then, after ca. six hours, it increases up to 95.12 bar. In total, 10.02 bar of gas or 18.25 g/ kg (grams of gas per kg of coal) became released. A momentary pressure decrease after N2 dosing is also characteristic for other samples (as measured by Zula (2006)). Subsequent slowly permanent increase in pressure for about 10 bar to 11 bar in six hours or so, indicates release of gas from coal. This is maybe the most important phenomenon in sorption investigations of the Velenje lignite. It was already clearly Table 5. Adsorption and desorption key data for sample No 2 from Table 1 based on two repeated measurements (2-1 and 2-2). Sample composition: fD 90 %, G 10 %, autoclave volume: 105 mL, coal mass: 60 g, environmental temperature: 22 °C, V0: 24.12 mL/mmol, specific mass: 1.33 g/cm3, free volume: 59.9 mL. Dots indicate gas sampling. Abbreviations W1 and W2 mean two measurements of mass of coal samples analyzed. Tsist/K is temperature of system in autoclave in Kelvin scale. Px to P4 are starting pressure (e.g. 56.50 bar) and pressure after a certain time (e.g. 52.85 bar). Px is increased pressure at adsorption. Apads is difference in pressures due to adsorption, and Apdes is difference in pressures due to desorption. Ads. and Des. are amounts of adsorbed and desorbed gas(es), respectively, expressed in grams of gas per kilogram of lignite. Sample Test W1 (s) W2 (g) Time (h) T?ist (K) P (bar) (bar) Ads (gkg) des. (bar) Des (gkg) 2-1 60.0 60.11 0 (CO;) 0.001 297. IS (PI) 56.50 -3.65 6.65 0.75 52.85 1.76 296.56 (P4) 87.05 -2.04 85.01 + 10.70 19.48 m 8 (PK) 9S.71 Drop of pressure to 1 bar and desorption of gas (mainly CO;+N:) lasting for 15 hours 6.89 12.55 Decrease then tacmwe of pressure nt ntlsorplion of COj + Ni Pi tPT Jl - ft i : m* M P(«iiar# 4SD- p. Pi r— Pi nt nl hid M i-nD i^f iiat Timr ("b> 1UP 1MQ Iti* IP&fl "1 Prt-nUfflt«1) H>P-> Figure 6. Graphic presentation of the used starting pressures (Pp P2, P3, P4) with adsorption in particulate phase and especial event (Px) with decrease and then increase of pressure during the adsorption process of CO2 - - N2 mixture. Figure 7. Pressure behaviour as measured in autoclave after 1st and 2nd dosing by CO2, and after dosing by N2 (in a linear and the logarithmic scale). Note that dosing by CO2 causes adsorption of gas into coal substance (detected as two slight decreases of pressure in the first two 45 min lasting steps), and that N2 dosing causes - after an instant time of adsorption - a release of gas from coal (detected as slight increase and then stabilizing of pressure in the time interval between 1.75 h and 8 h). Note also considerably higher pressure of N2 dosing in comparison to CO2. 1 7 i i Da 5 sorption Sample 2 71m« (Hour] - lin.sc« 6 7 a -1 lie 9 10 11 --- - ressire (tarHlofl. time scale} Tessure ihaiXfcn line st^cj 2 15 N 5 ! S * ...... 55 - fi J ï , S '5 t a 2.5 2 1 î g o.c M>1 0 il Tit e(H QU ri- to 9 3C 1 ate 0 1 DO Figure 8. Desorption of gas from sample 2-1. Measurements expressed by A pressure values are shown in a normal linear scale and the logarithmic scale, respectively. detected by Pezdic at el. (1999b) and was explained by different possible processes, which would need additional investigations. One explanation is enhanced release of gases from coal, and some other explanations are thoroughly given in Pezdic et al. (1999b), Pezdic et al. (2007) . • Comparing results for sample 2-1 and 2-2, adsorption due to CO2 dosing is slightly more effective for the first sample, whereas gas release due to N2 dosing is somewhat higher in the second sample. • Desorption is 6.67 bar or 12.09 g/ kg in the first measurement case (21), and 6.89 bar or 12.55 g/kg in the second case (2-2), respectively. Sample 1-1 and Sample 4-1: Samples 1-1 (X 100 %) and 4-1 (X 80 %, dXxD 10 %, fD 10 %) are very similar in lithotype composition, but demonstrate quite diverse sorption characteristics (Figure 10). In the first and the second CO2 loading stage together, sample 4-1 adsorbed 3.26 g/kg or 43.10 % less of CO2 than sample 1-1 with 5.73 g/kg. Additional N2 pressure caused initially a slight decrease of the total pressure in both samples, but then the pressure started to rise (as principally visible from Figure 7), and reached after six hours the value of 90.25 bar for sample 1-1 and the value of 92.57 bar for sam- ple 4-1. Sample 1-1 released 5.87 bar of gas or 10.44 g/kg, whereas sample 4-1 released 11.09 bar of gas or 19.01 g/kg i.e. for almost 80 % more than sample 1-1. Proportion of released gas depends on the proportion of adsorption in the first two steps, on CO2 compression at the time of addition of N2 overpressure, and on velocity of gas mixture release, as we decreased the pressure to 1 bar. For the same reason desorption varies too. The desorption for sample 1-1 equals to 5.30 g/kg or to 32.78 % of the wholly adsorbed gas (5.73 + 10.44 g/ kg), and for 4-1 to 3.74 g/kg or 16.80 %. Already our findings so far categorized the samples of such entirely xylite, and xylite-very-rich lignite lithotypes as the most resistant for sorption. In this sense, our present measurements published in this paper confirm this finding once again. Sample 3-1 and Sample 6-1: Samples 3-1 (dXxD 20 %, fD 80 %) and sample 6-1 (dXxD 7 %, fD 93 %) differ significantly in content of xy-lite-detrite (dXxD) component. Even though this difference (20-7 %) is numerically not very big it seems to be decisive to give the two samples quite distinctive gas adsorption characteristics. As visible from Figure 10, total CO2 adsorption in the sample 3-1 amounts to 6.06 g/kg, and in sample 6-1 to 7.62 g/kg. In the second stage, less than a half of gas than in the first stage was adsorbed. Additional N2 pressure caused initially a slight decrease of the total pressure for both samples, but afterwards the pressure started to rise and reached the value of 95.25 bar for sample 3-1, and 93.93 bar for sample 6-1. Sample 3-1 released 16.06 g/kg of gas and sample 6-1 16.69 g/kg of gas. Desorption of the sample 3-1 equals 8.09 g/kg or 36.57 %, and of the sample 6-1 7.09 g/ kg or 29.90 %. Sample 2-1 and Sample 5-1: Samples 2-1 (fD90%, G10%) and sam- ple 5-1 (X 15 %, fD 65 %, G 20 %) have the same lithotype code (9.5), although their composition is different. The sample 2-1 comes from a fault zone and the sample 5-1 from a borehole. What is common to both samples is remarkable degree of gelification which was most probably enhanced due to alkalinity evidenced by e.g. carbonate coatings of vegetal remnants (see Table 1). Their gas adsorption and desorption values are much higher than for all previous samples, and this feature is connected to our opinion to geli-fication. Total adsorption of sample 2-1 is 8.98 g/kg, and of sample 5-1 9.80 g/ kg (Figure 10). Compared to the sample 4-1 with the lowest adsorption, this is around 60 % more. Figure 9. Amount of adsorbed CO2 gas depends on lithotype code cs «.00 7,00 6.00 5.00 $ -1.00 1 3.00 1 | 2.00 1,00 Amount of adsorbed C02 gas in the first two stages 3 2D 9 10 fD9i n uoa 030 n K BD 1 ■ 1 - 1 1 n 1 0.00 SaH^1 □ Adsorption I stage ■ Adsorption 2 stage Satni)lei sa^t^^ sa^P1®* semi*®5 sa^6 Ligniit samples Figure 10. Amounts of adsorbed CO2 as achieved by 1st and 2nd step of dosing by CO2 at pressures of between 55 bar and 59 bar. Time of adsorption was 45 min for each step. (JS US. hi C! Cdinpil lMili (if ¡H c\hui r ill jdiarpliOH fc „ ^^^^ I ni r G (n n»