Ecotoxicology and Environmental Safety 112 (2015) 105–113

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv

Chemical and toxicological evaluation of underground coal gasification (UCG) effluents. The coal rank effect Krzysztof Kapusta n, Krzysztof Stańczyk Główny Instytut Górnictwa (Central Mining Institute), Plac Gwarków 1, 40-166 Katowice, Poland

art ic l e i nf o

a b s t r a c t

Article history: Received 4 August 2014 Received in revised form 28 October 2014 Accepted 30 October 2014

The effect of coal rank on the composition and toxicity of water effluents resulting from two underground coal gasification experiments with distinct coal samples (lignite and hard coal) was investigated. A broad range of organic and inorganic parameters was determined in the sampled condensates. The physicochemical tests were supplemented by toxicity bioassays based on the luminescent bacteria Vibrio fischeri as the test organism. The principal component analysis and Pearson correlation analysis were adopted to assist in the interpretation of the raw experimental data, and the multiple regression statistical method was subsequently employed to enable predictions of the toxicity based on the values of the selected parameters. Significant differences in the qualitative and quantitative description of the contamination profiles were identified for both types of coal under study. Independent of the coal rank, the most characteristic organic components of the studied condensates were phenols, naphthalene and benzene. In the inorganic array, ammonia, sulphates and selected heavy metals and metalloids were identified as the dominant constituents. Except for benzene with its alkyl homologues (BTEX), selected polycyclic aromatic hydrocarbons (PAHs), zinc and selenium, the values of the remaining parameters were considerably greater for the hard coal condensates. The studies revealed that all of the tested UCG condensates were extremely toxic to V. fischeri; however, the average toxicity level for the hard coal condensates was approximately 56% higher than that obtained for the lignite. The statistical analysis provided results supporting that the toxicity of the condensates was most positively correlated with the concentrations of free ammonia, phenols and certain heavy metals. & Elsevier Inc. All rights reserved.

Keywords: Underground coal gasification Condenser water Groundwater pollution Acute toxicity Vibrio fischeri

1. Introduction The increasing energy demand and energy prices observed today in many parts of the world in combination with environmental, political and socioeconomic concerns related to the energy sector favour research on alternative technologies for energy production and energy utilisation. One of the technologies is underground coal gasification (UCG). Coal is the most abundant fossil fuel in the world and is likely to outlast the gas and oil resources combined (IEA, 2013). Its current share in global power generation is dominant, and according to forecasts, it is expected to steadily increase in the future, mainly due to increasing energy demands in developing countries and the gradual depletion of the proven oil reserves (IEA, 2013). However, according to geological surveys, large amounts of the global coal resources, owing to technological or economic constrains, are currently beyond the range of traditional mining. These unmineable coal deposits may, in some cases, constitute suitable candidates for UCG technology. n

Corresponding author. Fax: þ 48 32 3246522.

http://dx.doi.org/10.1016/j.ecoenv.2014.10.038 0147-6513/& Elsevier Inc. All rights reserved.

The concept of UCG is not new and dates back more than one hundred years (Gregg and Edgar, 1978; Burton et al., 2006). During this process, coal is converted into a combustible gas directly in the seam (in situ). In its simple configuration, UCG involves drilling two vertical wells into the desired coal deposit, at some distance apart, enhancing the permeability of the coal between the two wells, (e.g., by directional drilling), igniting the coal seam in one well, and then injecting the gasification media (air, oxygen or steam) through the injection well. The product UCG gas, of which the primary components are hydrogen, carbon monoxide, carbon dioxide and methane, is recovered through the production well on the surface. The composition, as well as the heating value, of the produced gas depends mostly on the type of coal, gasification reagent applied and the gasification conditions (Khadse et al., 2007; Shafirovich and Varma, 2009; Kapusta and Stańczyk, 2009). The gas composition also determines its usefulness for possible downstream applications (Wang et al., 2011; Stańczyk et al., 2011). Liquid fuels by the Fischer–Tropsch process, production of substitute natural gas (SNG) and ammonia or urea synthesis are typical UCG syngas applications. Several research studies have proven that, under specific conditions, a product with a considerable

106

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

contribution of hydrogen can be obtained in the UCG process (Yang et al. 2008; Shu-gin et al., 2009; Yang and Ding, 1990; Stańczyk et al., 2010, 2012). The low-calorific UCG gas, typical for air-blown UCG operations, is usually intended for energy applications (heat or electricity). Although the UCG process has several advantages over surface coal gasification, such as lower capital investment costs, limited human labour, no coal transportation and no need for surface coal processing, the possible environmental risks associated with field UCG operations must be considered. One of the major environmental concerns addressed before UCG commercialisation is water quality. As a result of coal pyrolysis and the series of homo- and heterogeneous reactions occurring between the gasification products, a number of hazardous environmental contaminants can be produced during the UCG process. The major organic groundwater pollutants typical of UCG are phenols, benzene and its derivatives (BTEX), polycyclic aromatic compounds (PAHs) and certain heterocyclic compounds containing nitrogen, sulphur and oxygen heteroatoms. The inorganic matrix includes ionic compounds, such as ammonia, cyanides, sulphates, chlorides and wide range of metal and metalloid elements (Stuermer et al., 1982; Edgar et al., 1981; DeGraeve et al., 1980; Yang, 2009; Kapusta and Stańczyk, 2011; Liu et al., 2006a, 2006b). Operating below the hydrostatic pressure is the main preventive measure to effectively control the migration of contaminants during UCG operation (Kapusta et al., 2013; Liu et al., 2007). However, during the production phase, in many parts of the surface gas-processing facility, considerable quantities of postgasification condenser water are produced. These water effluents are formed as a result of gas–moisture condensation onto the cooler parts of the installation, and their production is strongly influenced by the overall balance of water available for the gasification process. Water in the UCG process derives from three distinct sources:

 water present in the coal seam as moisture (static resource);  groundwater infiltrating into the UCG cavity from the surrounding strata (dynamic resource); and

 water (steam) supplied as a gasifying agent. The production of such chemically complex wastewater is typical for the technologies of the thermochemical processing of coal (Jin et al., 1999; Parkhurst et al. 1981; Dong and Zhang, 2010). These effluents, if not treated appropriately on the surface, can cause serious environmental and ecological impacts. Many of the individual chemical characteristics for the UCG condensates are reported to be refractory and toxic (DeGraeve et al., 1980; Hill and Kocornik, 1986). Because chemical analysis is usually limited to a selected list of substances, it may be complemented by the application of bioassays that provide an integrated measure of toxicity. Numerous standard procedures for toxicity testing have been developed, suggesting plants, microorganisms, invertebrates and vertebrate organisms (Loibner et al., 2004). Acute and embryo-larval toxicity studies with Daphnia pulicaria, rainbow trout and fathead minnows have indicated that UCG condenser waters are extremely toxic, even when highly diluted (DeGraeve et al., 1980). Although standard acute-toxicity tests with fish and aquatic macro invertebrates have long played a major role in aquatic hazard and risk assessments (Toussaint et al., 1995), a number of alternative (rapid) tests have been proposed for screening due to their experimental simplicity, sensitivity, reproducibility and short exposure time. Rapid toxicity tests are also less expensive than standard acute-toxicity tests (Toussaint et al., 1995). The luminescent bacterial toxicity test system using V. fischeri is a reliable and widely applied aquatic bioassay that shows good sensitivity to a broad range of organic pollutants (Loibner

et al., 2004). In this method, the reduction in light emission by luminescent bacteria is attributable to the toxic effect of the tested sample (Gellert, 2000). A study of the physicochemical and ecotoxicological characteristics of condenser waters originating from two UCG experiments with coals of different ranks, was performed in the course of field and surface gasification trials performed at the Experimental Mine “Barbara” in Mikołów, Poland. The ecotoxicity was investigated with the luminescent bacterial assay based on V. fischeri test organisms. In this report, the main findings of this research are presented.

2. Materials and methods 2.1. The origin of the water-condensate samples The experiments were conducted both in natural (in situ) and surface (ex situ) conditions and among various aspects relevant to UCG: the influence of the coal rank on the composition and the efficiency of the gas production has been thoroughly investigated. The results of the experiments were described in a series of published research papers (Stańczyk et al., 2010, 2011, 2012; Wiatowski et al., 2012). Many ancillary factors related to UCG technology were also studied, including one connected to the environmental aspects of the in situ coal gasification (Kapusta and Stańczyk, 2011; Kapusta et al., 2013). The raw UCG product gas contains water vapour, originating mainly from the evaporation of coal moisture, the coal-pyrolysis process (pyrogenic water) or undesired hydrogen combustion. This gas moisture tends to condense onto the cooler parts of the installations, such as the internal surfaces of gas pipelines or in particular devices of the gas-treatment module. To prevent environmental pollution during the UCG operations, the resulting post-gasification water condensates are systematically collected and transported for off-site treatment. The UCG effluent samples used in the study originated from two gasification trials. One was a 16-day hard coal in situ gasification trial performed in the Barbara Experimental Mine in Mikołów, Poland. The second set of effluents was from a surface (ex situ) experimental simulation with a largeblock lignite sample, performed in an ex situ gasification unit. For each gasification experiment, the sampled water condensates represent four distinct periods of the overall process run time. Detailed descriptions of the installations were published in the reference papers (Wiatowski et al., 2012) and (Kapusta and Stańczyk, 2011) for hard coal and lignite trials, respectively. 2.2. Sample preparation and chemical analysis The collected raw water condensates (four samples for each gasification trial) were pretreated by filtration in a separatory funnel with a 0.45-mm membrane filter for the removal of emulsified tars and oils as well as solid particles. The obtained filtrates were stored at 4 °C until analysed. Apart from the two standard water parameters as the conductivity and pH, the following inorganic parameters were determined in the condensates: total ammonia nitrogen, nitrites, chlorides, cyanides, sulphates, and 17 metal and metalloid trace elements (Sb, As, B, Cr, Zn, Al, Cd, Co, Mn, Cu, Mo, Ni, Pb, Hg, Se, Ti, and Fe). The organic analysis of the effluents included phenolics, benzene with its three alkyl homologues toluene, ethylbenzene and xylene (BTEX), and 15 polycyclic aromatic hydrocarbons (PAHs). The conductivity, pH, biological oxygen demand (BOD5), chemical oxygen demand (CODCr) and total organic carbon (TOC) were additionally determined in the representative post-gasification effluents as typical nonspecific industrial wastewater parameters.

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

The chemical analyses were performed applying standard analytical methods. Electrochemical methods, i.e. potentiometry and conductometry, were applied to determine the reaction (pH) and the conductivity of the condensates, respectively. Flow-injection analysis (FIA) with gaseous diffusion and spectrophotometric detection (light absorption at the wavelength of 590 nm) was applied to determine the content of the total ammonia nitrogen (PN-EN ISO 11732, 2007). The nitrites were determined by a spectrophotometric method according to PN-EN 26777, 1999 (light absorption at 540 nm after reaction with 4-aminobenzenesulfonamide). The chlorides were determined by titration with a silver nitrate solution. The cyanides and the phenolics were determined by applying segmented flow analysis (SFA) with spectrophotometric detection according to PN-EN ISO 14403:2004 and PN-EN ISO 14402:2004 respectively. A gravimetric method after precipitation with barium was used for the quantitative determination of sulphates. Determinations of the 17 metals and metalloids were performed by inductively coupled plasma–optical emission spectroscopy (ICP–OES) according to PN-EN ISO 11885:2009 standard. For the analysis of benzene and its derivatives (BTEX), the gas chromatography headspace method using an Agilent Technologies 7890A chromatograph coupled with a static headspace auto sampler Agilent 7697A and FID detector was applied. The chromatographic column was DB-5MS (30 m, 0.25 mm, 0.5 mm) and helium was used as the carrier gas at 1.0 ml/min. The split ratio was 1:30. Oven programme temperature was 40 °C (hold 3 min), rate 15 °C/min up to 200 °C (hold 9 min). Temperature of FID detector was 300 °C. The determination procedure of the 15 polycyclic aromatic hydrocarbons involved preparation by solid-phase extraction (SPE) on ENVI-C18 cartridges purchased from Supelco (St Louis, MO, USA) and analysis with high-performance liquid chromatography (HPLC) using an Agilent Technologies HPLC 1200 Series. The SPE cartridges were conditioned by passing methanol followed by demineralised water. Subsequently, samples of 500 ml were passed through the cartridges at a constant flow rate. The cartridges were eluted with hexane, then the hexane was evaporated to dryness and 0.5 ml of acetonitrile was added. The obtained acetonitrile solutions were analysed by HPLC equipped with fluorescence detector. Separation of the compounds was achieved on a Agilent ZORBAX Eclipse PAH column (3.0 mm  250 mm, 5 mm). 2.3. Evaluation of unionised (free) ammonia in the water condensates The total ammonia in an aqueous solution includes two forms, i.e., both ionised (NH4 þ –N) and unionised ammonia (NH3–N), the relative concentrations of which are pH and temperature dependant (Emerson et al., 1975). Because NH3 has proven to be toxic to aquatic organisms and the ammonium ion is relatively nontoxic, the distinction between the effects of the two forms is necessary (Parkhurst et al., 1981). The fraction of unionised ammonia can be estimated from the following equation (Clement, 1995):

2.4. Toxicity-test procedure The ecotoxicological assays were performed with a LUMIStox 300 test kit, purchased from Dr. Lange GmbH, Germany. The test was performed according to the standard procedure (ISO 11348-3, 2007). Before starting the test procedure, the V. fischeri bacteria (LCK491 reagent purchased from Dr. Lange GmbH) were reactivated and kept at 5 °C, and a serial dilution of the effluent samples was prepared. V. fischeri as a marine bacterium requires an adequate water salinity, and therefore, all samples were prepared in an aqueous solution containing 2% NaCl. A 2% NaCl solution was also used as a control. The lyophilised bacterial reagent after reconstitution was added to the tested effluent samples. The organisms were exposed for 15 min at 15 °C to several concentrations of osmotically adjusted condensate samples. The bacterial luminescence intensity was measured before sample addition and after 15 min of incubation. All samples were tested in duplicate. The inhibition of the natural luminescence of the bacteria was regarded as the toxicity endpoint. The results were expressed as the 15-min median effective concentration (EC50). The EC50 is defined as the sample concentration that reduces the bacterial luminescence by 50% within a specified period of time. The EC50 values were calculated by the standard LUMIStox software package (Dr. Bruno Lange GmbH, 1999). The potential toxicities of the tested gasification condensates were described using a system of toxicity classification proposed by Persoone et al. (2003). According to this approach, the results obtained with microbiotests are transformed into toxic units (TU). The TU of a given condensate sample represents the inverse of its EC50 value:

TU = 100(EC50 )−1. Such a transformation of the obtained toxicity results makes the data more convenient for interpretation in terms of environmental relevance (Farré et al., 2008; Zgórska et al., 2011). 2.5. Statistical analysis The principal component analysis (PCA) and Pearson correlation analysis were adopted to assist in interpretation of the raw experimental data set. These tools were used in an attempt to analyse the relationships among the measured physicochemical parameters and to investigate the similarities between the particular effluent samples. Principal component analysis is a multivariate analytical tool used to reduce the dimensionality of experimental data, i.e., the set of original variables. The resulting small number of latent factors (principal components, PCs), is subsequently used to analyse the relationships among the observed variables (parameters) and objects (samples) (Golobočanin et al., 2004). A multiple-regression statistical method was then used to enable toxicity predictions based on the values of the selected physicochemical parameters. The theoretical backgrounds of these statistical methods are commonly available in the specialist literature (de Sá, 2007). The calculations were performed with the Statistica 10.0 package software (StatSoft, Inc., 2011).

NH3 − N(%) = 100/(1 + 10 pKa − pH ).

3. Results

The logarithm acidity constant pKa for ammonia, as a temperature-dependant parameter, can be expressed by the formula (Emerson et al., 1975):

3.1. Physicochemical characteristics of water condensates

pKa = 0.09018 + 2729.92/T , where T denotes the solution temperature in Kelvin.

107

The physicochemical characteristics of the condensate samples obtained during UCG of hard coal and lignite are presented in Table 1. The study conducted revealed a significant difference between the compositions of the two series of condensate samples

108

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

Table 1 Physicochemical characteristics of the UCG water condensates from the hard coal and lignite experiments. Parameter/compound

Conductivity pH BOD5 CODCr TOC Ammonia nitrogen Unionised ammonia Nitrites Chlorides Cyanides Sulphates Sb As B Cr Zn Al Cd Co Mn Cu Mo Ni Pb Hg Se Ti Fe Phenolics Total BTEX including benzene Total PAH including naphthalene

Unit

mS/cm – mg/l O2 mg/l O2 mg/l mg/l N mg NH3–N/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l mg/l

Hard coal condensate

Lignite condensate

Ha1

Ha2

Ha3

Ha4

Li1

Li2

Li3

Li4

15,900 6.15 2510 4540 551 2410 1.193 o 0.02 1206 1.37 2620 o 0.03 1.2 6.4 0.055 0.76 12 o 0.005 0.014 1.1 o 0.01 0.015 0.033 o 0.02 o 0.005 0.14 0.16 131 364 65 62 1780 1360

13,400 5.9 2540 4670 830 1830 0.509 o 0.02 1272 1.08 6680 o 0.05 5.2 9.3 0.12 1.2 34 o 0.02 0.042 3.75 o 0.05 o 0.01 0.05 o 0.05 o 0.005 o 0.05 0.58 460 626 26.7 20.6 3165 2900

13,300 7.3 3370 5310 902 1630 11.340 n.d. 3235 1.56 1070 o 0.1 o 0.1 4.7 1.8 11.4 2.6 o 0.02 0.043 13.3 o 0.02 0.25 0.86 o 0.03 o 0.02 o 0.05 0.93 1820 888 105 97 832 703

15,100 5.8 490 2710 181 1930 0.427 o 0.02 929.2 1.04 2510 o 0.03 2.4 5.5 0.05 0.75 22 o 0.02 0.025 1.5 o 0.03 o 0.01 0.029 0.044 o 0.003 0.14 0.40 188 59.1 26.5 24.7 1870 1600

7630 5.1 520 2030 200 1050 0.046 o 0.02 653 o 0.5 620 o 0.05 o 0.05 0.21 o 0.005 28 0.65 o 0.005 o 0.005 1.2 o 0.05 o 0.005 o 0.03 o 0.01 o 0.005 0.094 o 0.005 303 11 63 59 1450 586

9690 6.8 1690 2350 410 1550 3.423 o 0.02 487 o 0.5 915 o 0.005 o 0.05 0.13 o 0.005 0.3 0.44 o 0.002 o 0.005 0.48 o 0.01 o 0.01 o 0.01 o 0.005 o 0.005 0.35 0.0062 24.5 271 177 157 1140 465

4230 3.8 600 1460 170 269 0.001 o 0.02 553 o 0.5 1310 o 0.1 o 0.05 0.21 o 0.02 162 0.4 o 0.01 o 0.02 3.1 o 0.1 o 0.01 o 0.05 o 0.03 o 0.01 o 0.05 o 0.005 900 670 46.2 43.6 817 418

13,000 6.05 1690 2620 200 2030 0.798 o 0.02 606 o 0.5 1210 o 0.05 o 0.05 0.16 o 0.005 4 1.6 o 0.003 o 0.005 0.43 o 0.03 o 0.01 o 0.01 o 0.05 o 0.005 0.24 o 0.005 73 34.7 134 125 801 278

with respect to most of the measured parameters. Except for zinc, selenium, benzene and its alkyl homologues (BTEX) and certain polycyclic individuals, all of the values were significantly higher for the hard coal effluent samples. Because many of the UCG parameters change over the experimental run, e.g., the temperature distribution, coal properties (loss of volatiles) and geometry of the gasification cavity, distinct differences in the values of the particular condensates’ parameters are also observed within a group of samples originating from the same gasification experiment. 3.2. Acute toxicity testing results For each analysed condensate sample, the linear toxicity response of the test bacteria to the sample concentration was recorded. Based on the concentration–inhibition response curves obtained, the effective concentration percentages of the tested condensate solutions resulting in a 50% decrease in bioluminescence (EC50) were calculated, following the procedures outlined in the ISO 11348 standard protocol. Table 2 summarises the values of

the toxicity (TUEC50), obtained for the investigated gasification condensates. 3.3. Results of the statistical analysis 3.3.1. Principal component analysis (PCA) The PCA of the physicochemical (Table 1) and toxicological data (Table 2) obtained for the tested UCG water condensates is summarised in Fig. 1. Because the data set studied included measurements made within various magnitudes of ranges, the PCA analysis was performed for the centred and standardised data. The percent of modelled variance was used to correctly determine the number of significant components (PCs) (Wold, 1987). The results revealed that the reduction in the data dimensionality was effective because the PCA model with three PCs described 92.0% of the total data variance. The first two dimensions cover 75.73% of the total variation, with 49.77% being represented by the first principal component (PC1). The interpretation of the PCA results is based on the spatial distribution of variables and objects within the score and loading

Table 2 Estimated values of the toxicity parameter for the gasification condensates. Toxicity

Hard coal condensate

TUEC50

Ha1 323

Ha2 400

Lignite condensate Ha3 714

Ha4 244

Mean 420

Li1 217

Li2 370

Li3 270

Li4 217

Mean 269

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

109

Fig. 1. PCA loading (a) and score (b) plots for the physicochemical data of the UCG-derived. water condensates studied.

plots for the selected principal components. The loading plot (Fig. 1a) reveals the correlation structure of the analysed set of variables (physicochemical parameters). The correlation between the variables is interpreted in terms of their relative positions in the PC plane. A relatively small variation is observed with PC1, whereas a significant differentiation between the variables is introduced through the second dimension (PC2). The PC1–PC2 plot (Fig. 1a) shows three distinct clusters of highly, positively correlated parameters. The relationship structure of the analysed objects (samples) is revealed by the score plot shown in Fig. 1b. Similarities between the particular condensate samples are interpreted in terms of their relative proximity in the analysed PC plane. Two distinctly separate clusters, each representing a different coal rank used in the experimental part of the study (sample series Ha and Li), can be distinguished in the plot. Such a distribution of objects indicates that the two series of samples differ significantly qualitatively and in the description of their contaminant profiles. The score plot also reveals that one effluent sample (Ha3) profoundly differs from that of the other objects relative to their positions in the PC plane. Based on the values of factor loadings from the particular variables, the conclusion may be drawn that the differences between the samples collected in the hard coal (Ha) and lignite (Li) gasification experiments were caused mainly by the values of BOD5, COD, TOC, CN  , phenols, PAHs, free ammonia, TUEC50 and selected metal and metalloid ionic species. 3.3.2. Linear correlation analysis A linear correlation analysis (Pearson) was performed for the input data matrix based upon the physicochemical (Table 1) and toxicological data (Table 2) obtained for the tested condenser waters. The resulting correlation-coefficient matrix is presented as Table 3. The values of the Pearson coefficient, indicating that a significant correlation exists between the variables ( 40.7), were bolded and underlined. The interpretation of the analysis results was based on the absolute values and the sign of the correlation coefficient, which lies within the closed interval [  1,1].

4. Discussion 4.1. Physicochemical evaluation of the UCG-derived condensates The condensate samples analysed exhibit very high values of the standard nonspecific parameters, i.e., BOD5, COD and TOC, which is characteristic of the effluents resulting from the thermochemical processing of coal (DeGraeve et al., 1980; Hill and Kocornik, 1986). As can be inferred both from the PCA loading plot shown in Fig. 1a and from the correlation matrix (Table 3), all of these three parameters are strongly positively correlated with each other (values of correlation coefficients above 0.9). Because the COD/BOD5 ratio determines the susceptibility of wastewater contaminants to biological degradation, the determination of this ratio is of crucial importance in assessing the harmfulness of coalprocessing effluents. The higher the value of the COD/BOD5 ratio, the smaller the portion of the total load of organic contaminants that must be removed by the biological-treatment process. This poorly degradable part of the total contaminant budget can also significantly add to the toxicity level. The study conducted revealed that both the COD and BOD5 values for the hard coal condensates were approximately two times higher than those obtained for the lignite (Table 1). For the hard coal condensates, except for the case of one sample (Ha4), the COD/BOD5 ratios were similar and ranged between 1.6 and 1.8. For the sample Ha4, the ratio determined was extremely high, i.e., 5.5. The values of this parameter for the lignite condensates were more variable, ranging from 1.4 to 2.4 for the samples Li2-Li4, and with one significantly greater value of 3.9 obtained for the sample Li1. With respect to the wastewater from thermochemical coal processing, the differences in the COD and BOD5 values are frequently attributed to the content of poorly removable refractory compounds, which include polycyclic aromatics (PAHs), benzene and its homologues (BTEX) and some poorly degradable phenols. Among the target 15 PAHs, 9 individuals for hard coal and 7 for lignite were identified in concentrations above the lower detection limit (0.01 mg/l). Naphthalene (NaP) was a major compound for all effluent samples under study, contributing to, on average,

110

Table 3 Results of the linear correlation analysis (Pearson correlation matrix). pH

BOD5

COD

TOC

Phenolics NH4 þ

Cl 

CN 

SO42 

As

B

Cr

1.00 0.68 0.54 0.72 0.44  0.09 0.96 0.40 0.74 0.41 0.45 0.70 0.22  0.80 0.55 0.57 0.11 0.22 0.13 0.50  0.14  0.06 0.41 0.21 0.24

1.00 0.73 0.67 0.62 0.11 0.70 0.58 0.51 0.02 0.03 0.30 0.56  0.81 0.07 0.47 0.40 0.56 0.35 0.53 0.13 0.54 0.02 0.70 0.64

1.00 0.91 0.91 0.62 0.52 0.75 0.66 0.31 0.19 0.52 0.66  0.44 0.17 0.63 0.62 0.66  0.16 0.66 0.44 0.26 0.14 0.69 0.80

1.00 0.92 0.54 0.62 0.81 0.89 0.49 0.43 0.79 0.64  0.53 0.46 0.84 0.62 0.64  0.35 0.81 0.43  0.10 0.40 0.59 0.74

1.00 0.73 0.36 0.79 0.73 0.50 0.41 0.68 0.68  0.37 0.38 0.81 0.70 0.67  0.38 0.80 0.54 0.00 0.38 0.66 0.86

1.00  0.24 0.68 0.48 0.26 0.14 0.37 0.66 0.34 0.09 0.58 0.78 0.65  0.64 0.63 0.82  0.13 0.02 0.60 0.79

1.00 0.77 0.07 0.04 0.46 0.96  0.23 0.08 0.78 0.95 0.96  0.50 0.90 0.83  0.02  0.04 0.88 0.92

1.00 0.46 0.46 0.86 0.59  0.41 0.55 0.86 0.58 0.59  0.44 0.83 0.42  0.37 0.41 0.49 0.63

1.00 0.97 0.84  0.15  0.24 0.94 0.62  0.02  0.16  0.37 0.40  0.11  0.55 0.92  0.26 0.07

1.00 0.84  0.17  0.30 0.98 0.64  0.05  0.18  0.34 0.41  0.16  0.61 0.95  0.29 0.01

1.00 0.22  0.41 0.88 0.85 0.29 0.22  0.45 0.69 0.13  0.56 0.80 0.10 0.36

1.00  0.13 1.00  0.15  0.35 0.64  0.36 0.97 0.03 1.00  0.13  0.41  0.43 0.82  0.31 0.89 0.31 0.14  0.27  0.26  0.37 0.95  0.22 0.92  0.19

1.00 0.22 0.56 0.34 0.35 0.55 0.06  0.87 0.44 0.37  0.08 0.06 0.35 0.28  0.34 0.13 0.37 0.11 0.11

Zn

Al

Co

Mn

Ni

Se

1.00 0.67  0.05  0.15  0.32 0.45  0.17  0.64 0.93  0.27 0.01

1.00 0.70 0.64  0.57 0.96 0.54  0.39 0.53 0.51 0.71

1.00 0.97  0.59 0.85 0.96  0.02  0.17 0.88 0.92

1.00  0.41 1.00 0.81  0.55 0.88  0.69 0.15 0.74  0.27  0.27 0.95  0.15 0.92  0.35

Ti

Fe

BTEX

PAHs

Free NH3

1.00 0.71 1.00  0.25  0.10 1.00 0.29  0.28  0.59 1.00 0.70 0.77 0.40  0.36 1.00 0.83 0.81 0.20  0.07 0.93

TUEC50

1.00

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

Cond. pH BOD5 COD TOC Phenolics NH4 þ Cl  CN  SO42  As B Cr Zn Al Co Mn Ni Se Ti Fe BTEX PAHs Free NH3 TUEC50

Cond.

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

b

3200 2800

Ha1 Ha2 Ha3 Ha4

Concentration, g/l

2400 2000 1600 1200 800

200 150 100

Li1 Li2 Li3 Li4

600

500

Concentration, g/l

a

111

400

300

200

100

50

0

0 NaP

AcP

Flu

Phe

AnT

Fla

Pyr

BaA

Chr

PAHs

NaP

AcP

Flu

Phe

AnT

Fla

Pyr

PAHs

Fig. 2. Distribution of PAHs in the hard coal (a) and lignite (b) gasification condensates: NaP – naphthalene, AcP – acenaphthene, Flu– fluorene, Phe – phenanthrene, AnT – anthracene, Fla – fluoranthene, Pyr– pyrene, BaA – benzo(a)anthracene, and Chr – chrysene.

approximately 85% and 42% of the total PAH content in the hard coal and lignite condensates, respectively (Fig. 2). Whereas in the case of the hard coal condensates, two-ring naphthalene was explicitly a dominant polycyclic aromatic compound, the individual PAHs were distributed more evenly for the lignite samples (Fig. 2b). The second most abundant polycyclic hydrocarbons occurring in the lignite condensates were acenaphthene (AcP) and phenanthrene (Phe), with respective contributions of 18.3% and 25.6% to the total content of PAHs. For the hard coal effluents, the next most abundant polycyclic hydrocarbons were fluoranthene (Flu) and phenanthrene (Phe) (Fig. 2a). All of the water condensates produced during the UCG experiments with lignite were fluoranthene (Flu) free, which can be derived from Fig. 2b. The macromolecular structure of coal (two distinct ranks used) is likely to be the most significant parameter influencing the distribution of particular polyaromatic hydrocarbons in the process condensates. The second important group of UCG-related refractory materials comprises benzene and its alkyl derivatives, referred to generally as the BTEX compounds. For both coals under study, benzene dominates as the main contributor to the total BTEX contents in the tested post-gasification condensates, contributing about 91% of total BTEX for both condensates. Average concentrations of benzene were 51.1 and 96.2 mg/l for hard coal and lignite effluents respectively. BTEX is also one of the few parameters, for which higher values were obtained in the case of lignite-derived condensates, i.e., 105 mg/l compared to 56 mg/l (on average) for the hard coal condensates. For both types of condensates, the concentrations of particular individual species decline along with the length of the attached alkyl chain. No significant correlations were identified between the concentrations of aromatic compounds and other parameters, as can be observed from Fig. 1a. Phenolic compounds receive special attention during UCG operations. Due to their high water affinity, they are the most abundant organic contaminants identified in UCG-process condensates, with the concentrations exceeding by three orders of magnitude the contents of other organic compounds, such as PAHs and BTEX (Kapusta and Stańczyk, 2011; Kapusta et al., 2013). This study revealed that the average concentration of phenolic compounds for the hard coal condensate is approximately two times that of the lignite condensate, i.e., 484.3 mg/l compared to 246.7 mg/l. The value of this parameter is therefore one of the most important for differentiating the two groups of gasification condensates, which is clearly reflected by the distribution of the

objects studied (condensate samples) in the PCA score plot PC1– PC2 shown in Fig. 1b. The significant positive correlation coefficient (0.73) for the association between phenols and TOC (Table 3) indicates that the phenolic compounds make the relevant contribution to the value of this nonspecific parameter. This positive correlation is also reflected by the close vicinity of these two parameters in the PCA-loading plot presented in Fig. 1a. The study revealed that the two sets of condensates also differ significantly in the occurrence and concentration of selected metal and metalloid ions. From the 17 species under study, 12 were identified in concentrations above the lower detection limits of the analytical procedures in the hard coal condensates. As regards the lignite condenser water, the applied analytical methods allowed quantitation of only six elements (Table 1). The following five elements: Sb, Cd, Cu Pb and Hg were not detected in any condensate sample produced from the two gasification experiments. In the group of elements quantifiable in both types of condensates, the concentrations of particular ions, excluding Zn and Se, were significantly higher for the hard coal-derived samples. The differences in the chemical composition of the coals and the lower pH values of the lignite effluents can be considered as the main reasons for this. Several relevant associations can also be drawn from the correlation matrix presented in Table 3. The most profound are the Cr–Ni and Cr–Mn correlations with the respective Pearson coefficients of 1.0 and 0.97 respectively. Although the average content of ammonia nitrogen in the condensate samples is relatively high (1950 mg/l and 1225 mg/l for the hard coal and lignite, respectively), the percentage fraction of free ammonia (NH3–N) is small for the most effluent samples (Table 1). This is due to the low pH values, which, except for one hard coal effluent sample (Ha3), did not exceed 7. No relevant correlations between the free ammonia and other physicochemical parameters (apart from the reaction pH and toxicity) were identified in this study. 4.2. Toxicological evaluation of the UCG condenser waters The condensate samples tested were classified into one of the five toxicity categories, following the assessment procedure proposed by Persoone et al. (2003) for wastes discharged into the aquatic environment. The classification system is based on the absolute values of the TUEC50 parameter. Based on the values obtained for the TUEC50 parameter, all higher than 100, all tested UCG effluents can be classified into class V of toxicity according to

112

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

Persoone et al. (2003). This means that all of the samples are extremely toxic to the test organisms V. fischeri and that their EC50 values are reached in more than hundred-fold dilutions. The studies revealed, however, several differences in the levels of toxicity of the water condensates obtained for both types of coal. The toxicity of condensate from the hard coal gasification process, expressed by the value of the parameter TUEC50, ranged from 244 to 714, which corresponds to the range of effective sample concentration (EC50) between 0.41 and 0.14%. In the case of the lignite gasification effluents, the toxicity parameters determined were within a narrower range of values, i.e., from 217 to 370 for TUEC50, which corresponds to the concentration range from 0.46 to 0.27%. The set of condensate samples from the gasification of hard coal was characterised by both a higher variation in the toxicity values (standard deviation (SD) for TUEC50 ¼ 7 206) and a higher general level of toxicity, with the averaged value of the TUEC50 parameter equal to 420. The average toxicity level observed in the sample population from the lignite experiment was approximately one third less than that for the hard coal (TUEC50 value 269), with a standard deviation of SD ¼ 7 72 in this group of tested samples. 4.3. Relationship between the chemical composition and the toxicity of UCG condensates As can be derived from the PCA (Fig. 1a) and the Pearson correlation studies (Table 3), most of the toxicity of these complex UCG water condensates appears to be attributable primarily to constituents such as phenol, unionised ammonia and selected heavy metals, i.e., chromium, nickel and manganese. Due to the many interdependencies between the measured parameters, typical for solutions with such complex chemical compositions, considerable positive correlations are also observed between the toxicity parameter and certain nonspecific parameters, i.e., BOD5, COD and TOC. A comparison of the qualitative and quantitative data for the process condensates (Table 1) with results from the toxicity studies (Table 2) indicates that, for both coals, the samples with the highest toxicity level are those with the highest concentrations of free ammonia, i.e., samples Ha3 and Li2 for the hard coal and lignite, respectively. Phenolics, which are dangerous toxicants appearing in high quantities in the gasification condensates, appear to play a greater role in influencing the toxicity of the samples originating from the hard coal experiment. The greatest concentration of phenol was observed for sample Ha3, which may suggest the occurrence of synergistic interactions between phenol and free ammonia for the final toxic effect. The lignite-derived condensate with the highest value for the TUEC50 parameter (sample Li2), was not that with the highest phenol concentration. This may suggest that the contributions of phenol to the observed toxicity levels were less significant in the studied group of lignite condensates. The studies conducted showed that no considerable contributions of polycyclic aromatic hydrocarbons (PAHs) and BTEX to the toxicities of the tested samples occurred. The reason for this is that these groups of contaminants appear in the process condensates in concentrations approximately three orders of magnitude lower than those obtained for the more water-soluble phenolics. Hence, the solubility parameter appears to also be one of the major factors influencing the possible toxic effects of the coal-conversion effluents. Among the analysed parameters, phenolics and free ammonia are inherent to the gasification condensates because the two are typical of the thermochemical coal-conversion processes. The remaining parameters, although frequently positively correlated with toxicity, are usually derived from the input raw coals (exogenous constituents) and, depending on their presence in the coal,

Table 4 Parameters of the multiple regression model: TU¼ f (phenol, N–NH3). Variable

Unstandardised coefficients

Standardised coefficients

β

Error β

b

Error b

0.1169 0.1169

211.331 0.177 30.816

22.3367 0.0578 4.9945

Free term Phenol 0.358 Free ammonia (N–NH3) 0.721

p-Statistics

0.00022 0.02806 0.00163

they may or may not appear in the process condensates. Heavy metals are a good example of this. Based on these assumptions, the construction of a multiple-regression model was attempted to quantitatively assess the toxicity of these complex chemical matrices using only the two toxicants as the independent variables, i.e., phenol and free ammonia (N–NH3). The parameters of the multiple-regression model obtained are presented in Table 4. As a consequence, the toxicity-prediction equation appears as: TUEC50 ¼ 211.331 þ0.177 (phenol conc., mg/l) þ 30.816 (N–NH3 conc., mg/l). The regression model obtained is characterised by a good fit to the experimental data, with a high coefficient of determination R2 ¼0.956. The statistical significance of the regression analysis is confirmed by the low values of the p-statistics (o 0.05). As can be concluded from the calculated values of the β weights (Table 4) the proportion of free ammonia in predicting the toxicity is much higher that than of phenol (0.721–0.358).

5. Conclusions These studies revealed a considerable effect of the coal rank on the qualitative and quantitative profiles of the UCG process condensates. Except for the BTEX compounds, certain polycyclic hydrocarbons (PAHs), Zn and Se, the values of the remaining parameters were significantly greater for the hard coal condensate samples. All tested UCG water condensates were characterised by very high levels of acute toxicity (TUEC50 4100), placing them into class V toxicity. These effluents therefore pose a very high risk to the aquatic environment in the event of discharge or unintentional penetration into surface water bodies or into groundwater aquifers without prior purification. Due to the statistically significant differences both the organic and inorganic contaminant profiles, the condensates originating from the two gasification experiments exhibited considerable differences in their toxicity levels. The average value of the TUEC50 parameter for the hard coal condensates was approximately 56% higher than that obtained for the lignite samples. Principal component analysis (PCA) combined with correlation analysis showed supporting results, implying that, among all of the tested samples, the values of the TUEC50 parameter were most positively correlated with the concentrations of unionised (free) ammonia, phenols and certain heavy metals. However, the greatest Pearson correlation coefficient was observed for free ammonia (0.93). The total ammonia content in the aqueous solution consists of the two major forms, i.e., the ionic form (ammonium ion NH4 þ ) and free ammonia (NH3). Studies have shown that the toxicity of the tested effluent was only correlated with the content of ammonia in the free form. Thus, the pH value, which determines the balance between the two ammonia forms, is an intermediate parameter, but it is of crucial importance to the toxicity levels of the condensates from the UCG process.

K. Kapusta, K. Stańczyk / Ecotoxicology and Environmental Safety 112 (2015) 105–113

The multiple-regression analysis performed confirmed that the toxicity of the tested effluents can be quantitatively described by a regression model including two parameters characteristic for the UCG process condensates, namely, the concentration of free ammonia and the content of phenolic compounds. The contribution of free ammonia in the toxicity-prediction model was almost double that calculated for the phenolic compounds.

Acknowledgements This work was a part of the HUGE and HUGE2 projects jointly supported by the RFCS, under Contracts nos. RFCR-CT-2007-00006 and RFCR-CT-2011-00002, and by the Polish Ministry of Science and Higher Education.

References Burton, E., Friedmann, J., Upadhye, R., 2006. Best Practices in Underground Coal Gasification. Lawrence Livermore National Laboratory. Clement, B., Merlin, G., 1995. The contribution of ammonia and alkalinity to landfill leachate toxicity to duckweed. Sci. Total Environ. 170, 71–79. de Sá, J.P.M., 2007. Applied Statistics Using SPSS, STATISTICA, MATLAB and R, 2nd Edition Springer. DeGraeve, G.M., Overcast, R.L., Bergman, H.L., 1980. Toxicity of underground coal gasification condenser water and selected constituents to aquatic biota. Arch. Environ. Contam. Toxicol. 9, 543–555. Dong, Y., Zhang, J., 2010. Testing the genotoxicity of coking wastewater using Vicia faba and Hordeum vulgare bioassays. Ecotoxicol. Environ. Saf. 73, 944–948. Bruno Lange GmbH, Dr, 1999. Lumistox 300 Operating manual. Düsseldorf, Germany. Edgar T.F., Humenick M.J., Kaiser W.R., Charbeneau R.J., 1981. Environmental Effects of In Situ Gasification of Texas lignite. Project Summary. US EPA, EPA-600/S781-035. Emerson, K., Russo, R.C., Lund, R.E., Thurston, R.V., 1975. Aqueous ammonia equilibrium calculation: effect of pH and temperature. J. Fish. Res. Board Can. 32, 2379–2383. Farré, M., Asperger, D., Kantiani, L., González, S., Petrovic, M., Barceló, D., 2008. Assessment of the acute toxicity of triclosan and methyl triclosan in wastewater based on the bioluminescence inhibition of Vibrio fischeri. Anal. Bioanal. Chem. 390, 1999–2007. Gellert, G., 2000. Sensitivity and significance of luminescent bacteria in chronic toxicity testing based on growth and bioluminescence. Ecotoxicol. Environ. Saf. 45, 87–91. Golobočanin, D.D., Škrbić, B.D., Miljević, N.R., 2004. Principal component analysis for soil contamination with PAHs. Chemom. Intell. Lab. Syst. 72, 219–223. Gregg, D.W., Edgar, T.F., 1978. Underground coal gasification. AIChE J. 24, 753–781. Hill, S.L., Kocornik, D.J., 1986. Final Report on the Toxicological Testing of Groundwaters Affected by Underground Coal Gasification. Western Research Institute, Laramie, Wyoming. InternationalEnergyAgency (IEA), 2013. World Energy Outlook 2013. OECD/IEA, Paris. ISO 11348-3, 2007. Water quality. Determination of the inhibitory effect of water samples on the light emission of Vibrio fischeri (Luminescent bacteria test). Part 3: method using freeze-dried bacteria. Jin, H., Yang, X., Yu, H., Yin, D., 1999. Identification of ammonia and volatile phenols as primary toxicants in coal gasification effluent. Bull. Environ. Contam. Toxicol. 63, 399–406. Kapusta, K., Stańczyk, K., 2009. Development conditions and limitations of the underground coal gasification in Poland. Przem. Chem. (Eng. Chem. Rev.) 88, 331–338. Kapusta, K., Stańczyk, K., 2011. Pollution of water during underground coal gasification of hard coal and lignite. Fuel 90, 1927–1934. Kapusta, K., Stańczyk, K., Wiatowski, M., Chećko, J., 2013. Environmental aspects of a field-scale underground coal gasification trial in a shallow coal seam at the experimental mine barbara in Poland. Fuel 113, 196–208.

113

Khadse, A., Qayyumi, P., Mahajani, S., Aghalayam, P., 2007. Underground coal gasification: a new clean coal utilization technique for India. Energy 32, 2061–2071. Liu, S., Wang, Y., Yu, L., Oakey, J., 2006a. Thermodynamic equilibrium study of trace element transformation during underground coal gasification. Fuel Process. Technol. 87, 209–215. Liu, S., Wang, Y., Yu, L., Oakey, J., 2006b. Volatilization of mercury, arsenic and selenium during underground coal gasification. Fuel 85, 1550–1558. Liu, S., Li, J., Mei, M., Dong, D., 2007. Groundwater pollution from underground coal gasification. J. China Univ. Min. Technol. 17, 467–472. Loibner, A.P., Szolar, O.H.J., Braun, R., Hirmann, D., 2004. Toxicity testing of 16 priority polycyclic aromatic hydrocarbons using lumistox. Environ. Toxicol. Chem. 23, 557–564. Parkhurst, B.R., Meyer, J.S., DeGreave, G.M., Bergman, H., 1981. A reevaluation of the toxicity of coal conversion process waters. Bull. Environ. Contam. Toxicol. 26, 9–15. Persoone, G., Marsalek, B., Blinova, I., Törökne, A., Zarina, D., Manusadzianas, L., Nalecz-Jawecki, G., Tofan, L., Stepanova, N., Tothova, L., Kolar, B., 2003. A practical and user-friendly toxicity classification system with microbiotests for natural waters and wastewaters. Environ. Toxicol. 18, 395–402. PN-Water Quality, 1999. Determination of nitrite: molecular absorption spectrometric method. PN-EN ISO 11732, 2007. Water quality. Determination of ammonium nitrogen. Method by flow analysis (CFA and FIA) and spectrometric detection. PN-EN ISO 11885, 2009. Water quality. Determination of selected elements by inductively coupled plasma optical emission spectrometry (ICP-OES). PN-EN ISO 14402, 2004. Water quality. Determination of phenol index by flow analysis (FIA and CFA). PN-EN ISO 14403, 2004. Water quality. Determination of total cyanide and free cyanide by continuous flow analysis. Shafirovich, E., Varma, A., 2009. Underground coal gasification: a brief review of current status. Ind. Eng. Res. 48, 7865–7875. Shu-gin, L., Yuan-yuaon, W., Ke, Z., Ning, Y., 2009. Enhanced-hydrogen gas production through underground gasification of lignite. Min. Sci. Technol. 19, 384–394. Stańczyk, K., Smoliński, A., Kapusta, K., Wiatowski, M., Świądrowski, J., Kotyrba, A., Rogut, J., 2010. Dynamic experimental simulation of hydrogen oriented underground gasification of lignite. Fuel 89, 3307–3314. Stańczyk, K., Howaniec, N., Smoliński, A., Świądrowski, J., Kapusta, K., Wiatowski, M., Grabowski, J., Rogut, J., 2011. Gasification of lignite and hard coal with air and oxygen enriched air in a pilot scale ex situ reactor for underground gasification. Fuel 90, 1953–1962. Stańczyk, K., Kapusta, K., Wiatowski, M., Świądrowski, J., Smoliński, A., Rogut, J., Kotyrba, A., 2012. Experimental simulation of hard coal underground coal gasification for hydrogen production. Fuel 91, 40–50. StatSoft, Inc., 2011. STATISTICA (Data Analysis Software System), Version 10. Stuermer, D.H., Ng, D.J., Morris, C.J., 1982. Organic contaminants in groundwater near an underground coal gasification site in northeastern Wyoming. Environ. Sci. Technol. 16, 582–587. Toussaint, M., Shedd, T.R., van der Schalie, W.H., Leather, G.R., 1995. A comparison of standard acute toxicity tests with rapid-screening toxicity tests. Environ. Toxicol. Chem. 14, 907–915. Wang, Z., Huang, W., Zhang, P., Xin, L., 2011. A contrast study on different gasifying agents of underground coal gasification at huating coal mine. J. Coal Sci. Eng. (China) 17, 181–186. Wiatowski, M., Stańczyk, K., Świądrowski, J., Kapusta, K., Cybulski, K., Krause, E., Grabowski, J., Rogut, J., Howaniec, N., Smoliński, A., 2012. Semi-technical underground coal gasification (UCG) using the shaft method in experimental mine “Barbara”. Fuel 99, 170–179. Wold, S., 1987. Principal components analysis. Chemom. Intell. Lab. Syst. 2, 37–52. Yang, L., Ding, R., 1990. The fundamental researches on hydrogen manufacturing from underground coal gasification. Resour. Mater. 12, 27–32. Yang, L., Zhang, X., Liu, S., Yu, L., Zhang, W., 2008. Field test of large-scale hydrogen manufacturing from underground coal gasification (UCG). Int. J. Hydrog. Energy 33, 1275–1285. Yang, L., 2009. Modeling of contaminant transport in underground coal gasification. Energy Fuels 23, 193–201. Zgórska, A., Arendarczyk, A., Grabińska-Sota, E., 2011. Toxicity assessment of hospital wastewater by the use of a biotest battery. Arch. Environ. Protect. 37, 55–61.

Chemical and toxicological evaluation of underground coal gasification (UCG) effluents. The coal rank effect.

The effect of coal rank on the composition and toxicity of water effluents resulting from two underground coal gasification experiments with distinct ...
513KB Sizes 0 Downloads 7 Views