Journal of South American Earth Sciences 14 (2001) 387±399
www.elsevier.nl/locate/jsames
Trace elements of Paleocene TaÂchira coals, southwestern Venezuela: a geochemical study M. MartõÂnez a,*, M. Escobar b, I. Esteves a, C. Lopez a, F. Galarraga a, R. GonzaÂlez a a
Instituto de Ciencias de la Tierra, Facultad de Ciencias, Universidad Central de Venezuela, Apartado Postal 3895, Caracas 1010A, Venezuela b Gerencia de InvestigacioÂn y Desarrollo, INZIT-CICASI, Km 14 La CanÄada, Maracaibo, Venezuela Received 1 January 2000; revised 1 November 2000; accepted 1 January 2001
Abstract The concentration and distribution of 35 elements in 78 coal samples from the TaÂchira State coal ®elds belonging to Los Cuervos Formation (Paleocene) of western Venezuela were obtained using atomic absorption spectrometry (AAS) and inductively coupled plasma atomic emission (ICP-AES). The aim of this work was to determine, through a multivariate statistical approach (factor analysis), if there is a correlation between the trace element content in coal beds and the provenance rocks at the time of peat deposition. Comparison with world averages and geometric means for trace element concentrations in coal-bed samples shows that B, Ba, La, V, Mn, Zn, and Pb are depleted. However, TaÂchira coal samples show a perceptible enrichment in Bi, Sb, As, Cd, and Mo, and they are highly enriched in Ag and Co. Nicholl's plot suggests that only B and Co show a distinctive organic af®nity. Enriched elements (Ni, Ag, Cd, Mo, Co) are both statistically and genetically related, and we attribute their origin to a volcaniclastic Jurassic unit. Other statistical factors (Th±V±P and Ca±Mg±Mn) reveal a different provenance, indicating a felsic plutonic source and a sedimentary limestone, respectively. A fourth factor (K±Mo±Th±S) is composed mainly of clay minerals and authigenic sul®des. q 2001 Elsevier Science Ltd. All rights reserved. Keywords: Coal; Trace elements; Geochemistry; Factor analysis; Venezuela
1. Introduction The TaÂchira±Tarra coal basin is the second largest in Venezuela in extent and reserves, more than 150 £ 106 MMTM (Escobar and MartõÂnez, 1993). Its coals vary in rank between lignite and low volatile bituminous, with a high calori®c value. Because of their low ash
,8% and sulfur content
,1%; these coal beds are of excellent quality for thermal use. Few studies have been conducted on the composition, distribution, and signi®cance of inorganics in TaÂchira coals (BricenÄo, 1992; MartõÂnez, 1996; Escobar and MartõÂnez, 1997). Variations in thickness, texture, rank, and quality suggest that variable trace element content in these coals is important for both environmental and geochemical studies. On the other hand, the concentration of trace elements at signi®cant levels offers new exploitation opportunities for this resource. In other countries, this property has permitted the development of coal®elds with secondary recovery of economic or strategic elements: * Corresponding author. Fax: 158-212-605-1201. E-mail address:
[email protected] (M. MartõÂnez).
germanium (Yudovich et al., 1972), uranium (Breger, 1958) and gallium (Zhou and Ren, 1981), among others. The objective of this work is to investigate the general chemical composition of TaÂchira coal beds with special emphasis on their trace element content in order to compare them with values obtained for other reference coal samples and with the terrestrial crust. In addition, using multivariate statistical methods (factor analysis) on the trace element data of the coal samples studied allows a better inference regarding the main source rocks at the time of peat formation. 2. Trace elements in coal The current renewed interest in coal as a major energy source and the associated environmental problems underscore the need to understand the complex nature of coal. Comprehensive geochemical information of this type is vital in designing environmentally acceptable technologies for utilizing coal as a fuel. Coal is derived from plant debris undergoing biological, chemical, and physical reactions over millions of years; it is one of two major ways of storing
0895-9811/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved. PII: S 0895-981 1(01)00035-9
388
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
Fig. 1. The study area in TaÂchira State, and location of the Tertiary coal-bearing unit, Los Cuervos Formation.
carbon in the Earth's crust. Thus, the study of coal geochemistry is also fundamental to better understanding the interaction between the biological and inorganic geochemical cycles of elements (Nicholls, 1968). Research on this topic attempts to establish the af®liation of minor and trace elements in coals with mineral and organic phases. Previous studies of minor elements in coal have usually been performed with the intent of determining their concentration in the whole coal seam (Breger, 1958; Gluskoter et al., 1977; Bouska, 1981; Valkovic, 1983; Goodarzi, 1988; Swaine, 1990; Finkelman, 1999). In the majority of those studies, the utility of trace element content and distribution in coals was demonstrated to be a geochemical tool for the establishment of relevant aspects of the origin and evolution of coal seams: sedimentary environments (Bailey, 1981; Harris et al., 1981; Hart et al., 1982; Gayer et al., 1999); paleosalinity and physicochemical conditions of formation (Bohor and Gluskoter, 1973;
Krejci-Graf, 1984; Banerjee and Goodarzi, 1990); coal seam correlation (O'Connor, 1986); provenance rocks and preferential direction of sediments supply (Zubovic, 1966; Ward et al., 1999); and organic af®nity of several trace elements (Bouska, 1981; Miller and Given, 1987; Swaine, 1990; Ren et al., 1999). Today, however, interest in this subject is focused more towards environmental impacts associated with coal use (Crowley et al., 1995; Finkelman, 1999). 3. The study area The TaÂchira±Tarra coal basin is located within the Venezuelan Andes in western Venezuela. Metamorphic and igneous rocks are present in the central core of the mountains. In the western and southeastern areas, a thick wedge of sedimentary rocks is part of the ®lling of the
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
Fig. 2. Simpli®ed stratigraphic sequence for TaÂchira (from GonzaÂlez de Juana et al., 1980). Shaded formation names are the coal-bearing units.
TaÂchira sub-basin (Fig. 1). Coal-bearing units in the sedimentary domain are of great extent and, in many cases, have lateral continuity towards eastern Colombia. Several outcrops, sub-surface galleries, and open-cut mines are present in this area. The igneous±metamorphic basement of the Andes in the TaÂchira depression is composed of the Iglesias Group (Precambrian). Metamorphosed Paleozoic sediments,
389
including carbonaceous slates, gneisses, augengneisses, and schists, often with crossed felsic intrusions (GonzaÂlez de Juana et al., 1980), overly the basement. The region received sediments in two large episodes during the Mesozoic. A Triassic±Jurassic sedimentation phase resulted in ¯uvio-lacustrine rocks rich in volcanic intercalation. Sedimentation during the Cretaceous phase consisted of highly calcareous strata (GonzaÂlez de Juana et al., 1980). The sedimentary set deposited during the Triassic±Jurassic is designated the La Quinta Formation and consists of a sequence of `red beds' of great thickness and lateral extent that crop out over all of western Venezuela and eastern Colombia (Audemard, 1982). This set has been identi®ed in the subsurface of Lake Maracaibo and in the Llanos Basin. The Cretaceous strata contain several calcareous units, representing a transgressive regional megacycle. The maximum transgressive event was reached with the La Luna Formation, which is Aptian±Albian in age. Towards the end of the Cretaceous, a regional regressive event took place that lasted until the Eocene, with the consequent formation of sedimentary environments representing shelf, coastal, delta and/or estuarine, tidal plain, and nearshore environments in general (the Barco, Los Cuervos, and and Mirador Formations). During the Oligocene, a new transgressive event took place that resulted in the LeoÂn Formation; The Andean molasse, represented by the Guayabo Group, was unconformably deposited on the LeoÂn Formation (Fig. 2). There are two coal-bearing units in the area of study: the Los Cuervos Formation, of Paleocene age, which belongs to the Orocue Group; and the Carbonera Formation, of Late
Fig. 3. Paleogeographic setting from the TaÂchira depression during the Paleocene (after MartõÂnez, 1996).
390
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
Fig. 4. (a) Location of coal sampling sites along TaÂchira depression. b) Stratigraphic position for some coal samples within the unit, showing a tentative lateral correlation.
Eocene±Early Oligocene age (Fig. 2). Both units appear to be delta or tidal sedimentary environments. The Los Cuervos Formation consists of ¯uvio-deltaic plains, with little or no marine in¯uence, within a regressive megacycle affecting northern Venezuela during the Early Tertiary. The Carbonera Formation is the result of deposition in an upper and lower delta plain in a transgressive sequence (Marquez and Mederos, 1989; MartõÂnez, 1996). Two igneous±metamorphic structural arches, the Arauca and MeÂrida uplifts, which were derived from the Guayana Shield, conditioned the paleogeographic setting of the zone during the Paleocene (Fig. 3). Paleo-current determinations indicate that the Paleocene delta system (Paleo±Magdalena) was partially controlled by both uplifts.
4. Experimental 4.1. Location and character of sampling sites The Los Cuervos Formation crops out extensively (a discontinuous belt about 180 km long) in the southern, western, and northern portions of TaÂchira State (Fig. 1). Numerous exposures of this unit occur in roadcuts and open-cut mines. Full-face channel coal samples were obtained of freshly exposed seams from 10 different sites along the entire thickness (Fig. 4a). At four sites, (San Pedro del Rio, Villa PaÂez, Santo Domingo, Las Adjuntas), coal
sampling was accompanied by constructing the respective lithostratigraphic columns for the unit. At San Pedro del Rio and Villa PaÂez, the sequence appears in well-exposed roadcuts; fresh samples were taken by chipping away about 30 cm of the weathered surface with a spade and then collecting the sample from a continuous channel about 10 cm wide and 5 cm deep in the newly opened surface. Partings in the coal bed more than 5 cm thick were discarded. In addition, the coal beds were being actively mined in the Las Adjuntas and Santo Domingo coal mines, which facilitated the collection of fresh samples, mainly from galleries. At the other six localities (La Virgen, Casigua, Zea, Las Mesas, Las Dantas, San SimoÂn), only a few coal beds were available for appropriate sampling. The number of samples for each locality is roughly proportional to outcrop area and thickness. A total of 78 incremental channel coal samples were collected; their locations and stratigraphic positions are shown in Fig. 4. It should be emphasized that all coal beds were sampled in their entirety. Lateral correlation between coal beds from the different places and mines sampled is very dif®cult, due mainly to the tectonic framework, high similarity between coal beds along the sedimentary record, and the poor knowledge of the Paleocene TaÂchira coal basin (MartõÂnez, 1996). However, a tentative lateral correlation is shown (Fig. 4b) for some coal seams, based on lithologic features of the unit (e.g. persistent thick sandstones with a high lateral continuity at the uppermost; coal beds thickness; occurrences of partings),
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
391
Table 1 Summary statistics of partial proximate and elemental analyses, including R0 of coal samples from the TaÂchira coal basin, Los Cuervos Formation
a
Moisture (%) Volatile matter a (%) Ash a (%) Carbon b (%) Hydrogen b (%) Nitrogen b (%) Sulfur (% dry) R0 (%) a b
Minimum value
Maximum value
Geometric mean
Standard deviation
No. of analyses
0.2 23.0 0.9 55.2 2.82 0.77 0.21 0.38
13.4 65.0 47.4 93.2 7.88 2.91 7.40 1.17
3.4 43.1 7.0 79.2 5.39 1.60 0.83 0.72
3.4 10.3 9.9 7.9 1.23 0.36 1.17 0.22
78 78 78 78 78 78 78 16
As received. Dry and ash free.
and physicochemical properties of coal, where such information is available (Bar and PenÄa, 1985; MartõÂnez, 1996). 4.2. Analytical methods Finely ground coal samples were submitted for proximate, elemental, sulfur, and trace element content. Proximate analyses were accomplished to the ASTM D-3172 (1985) procedures for coal and coke. The total sulfur concentration was determined in a LECO SC-432 analyzer. Quanti®cation of C, H, and N contents were carried out in a CARLO ERBA 1106 elemental analyzer. For determining Na, K, Ca, and Mg content, samples were dissolved by acid digestion (HNO3 ±HF) in a Parr bomb, following the Hartstein method (Hartstein et al., 1973). For Si, Ca, and Mg, dissolved samples were analyzed by atomic absorption spectrometry (AAS). Na and K were determined by AAS in a Varian Techtron AA6 spectrometer. B, Al, P, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Sr, Mo, Ag, Cd, Sb, Ba, La, Pb, Au, Bi, Th and U concentration analyses were accomplished by induced plasma spectrometry (ICP) in a IVON±JOVI. Organic petrography on selected samples was performed using a LEITZ ORTHOPLAN POL microscope; vitrinite re¯ectance (R0) measurements were obtained from an average of at least 100 points, always according to ASTM D-2748-79. The statistical calculations were performed with the NCSS 2000e statistical software package. Previous exploratory data analyses and descriptive statistics using the Dixon and Kronmal (1965) criteria for class interval choice indicated a lognormal distribution for trace elements. For this reason, all matrix data were transformed to the respective natural logarithmic values for a better approach to a Gaussian distribution; this adjustment does not modify the original distribution, but allows the use of parametric statistics. After this, all transformed data were standardized, for an arithmetic mean of 0 and a standard deviation of 1, applying the formula: Z
x2X s
where Z recalculated; standardized value derived from original data, x original value, X arithmetic mean, and s standard deviation. The use of standardized values allowed removal of artifacts derived from scale attributes within each set of data and equalized the in¯uence of variables with small variation as opposed to those with large variation (Crowley et al., 1995). Factor analysis was achieved after Varimax rotation, with a robust estimation of covariance, with at least 10 iterations. The minimum loading factor was 0.400. 5. Results The summary statistics of the proximate and elemental analyses, including total sulfur, are presented in Table 1. When compared with the mean% R0 measured values on selected samples, these values demonstrated that the coal samples analyzed oscillate in rank between sub-bituminous and medium-volatile bituminous. The minor and trace element concentrations of the coal samples are shown in Table 2. Results for Al, Na, K, Mg, Ca, Ti, and P are not shown in this table, but they were considered for the statistical analysis; values for Au and U were discarded because of their very low concentration values, which in most cases were below the detection limit for the analytical technique employed. Table 3 shows the statistical summary derived from Table 2 compared with average values for Illinois coal samples (Gluskoter et al., 1977), mean USA coal samples (Finkelman, 1999), Australian coal samples (Swaine, 1990), worldwide mean coal samples (Valkovic, 1983), and the Earth's crust (Mason and Moore, 1982). Upon comparing coal samples data from the Los Cuervos Formation with other coal data, the following are observed: ² The Los Cuervos coals are depleted in Al, V, Cr, Mn, Fe, Zn, Sr, Ba, Pb, La, and B. ² The TaÂchira coal samples are slightly enriched in Bi, As, Cd, Sb, and Mo. ² The coal samples are found to be strongly enriched in Co and Ag.
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
392
Table 2 Minor and trace element concentrations (ppm) in analyzed coals (Fe given in wt%) Location
Sample
B
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
As
Casigua
CASI1
12
4
1
3
0.2
5
10
15
56
1
La Virgen
THV20
6
18
4
6
0.6
6
18
41
128
4
Las Adjuntas
TLAS1 TLAS2 TLAS3 TLAS4 TLAS5 TLAS6 TLAS7 TLAS8 TLAS9 TLAS10 TLAS11 TLAS12 TLAS13 TLAS14 TLAS16
11 13 5 6 2 1 1 1 3 6 8 6 9 15 25
1 6 51 1 5 16 58 59 43 7 0.1 5 5 6 40
0.9 1 12 1 8 9 19 22 21 3 1 3 2 0.9 3
7 14 33 14 1370 469 450 1182 19 11 13 8 95 12 21
0.1 0.2 0.6 0 0.5 1.4 3 1.2 0.5 0.1 0.2 0.1 0.1 2.2 2.6
5 4 8 5 15 60 50 29 24 11 5 3 7 14 15
10 6 17 4 25 197 140 95 47 26 7 7 11 23 39
3 10 47 1 6 13 22 53 23 9 3 6 9 14 38
18 19 46 16 17 424 185 109 111 39 16 10 22 24 38
3 2 3 2 1 4 10 8 14 3 3 2 3 63 23
Las Dantas
TLD7 TLD8 TLD9
8 6 6
1 1 1
1 2 2
3 7 3
0 0.1 0
4 2 2
2 3 3
3 3 3
3 3 3
1 1 1
Las Mesas
TLM1 TLM2 TLM3 TLM4
3 2 1 1
5 1 5 9
2 0.9 2 5
15 1 6 5
0.4 0.1 0.3 0.1
17 1 6 3
9 2 6 8
13 5 2 15
42 3 20 21
13 2 2 5
San Pedro del Rio
TSP1 TSP2 TSP3 TSP4 TSP6 TSP8 TSP9 TSP10 TSP11 TSP12 TSP13 TSP14 TSP15 TSP16
3 8 7 3 28 7 6 5 5 6 10 6 4 4
5 7 26 149 1 10 17 17 40 62 20 14 54 13
2 3 4 8 1 4 2 2 4 7 2 1 15 10
70 33 47 6 7 38 6 12 28 23 24 271 6 11
0.2 0.3 0.2 0.2 0.1 0.8 0.3 0.7 0.3 0.6 1.1 0.2 0.6 0.3
11 5 9 8 7 76 13 2 3 5 3 5 26 21
15 8 12 7 2 10 5 6 11 15 8 13 24 9
6 2 9 16 3 2 8 20 34 28 15 14 26 18
51 11 34 47 0.5 9 25 7 17 69 18 33 50 20
2 1 5 3 1 11 24 1 4 13 34 7 9 1
San SimoÂn
TSS1 TSS2 TSS3 TSS4
31 11 18 37
55 180 225 99
4 12 12 6
2 1 1 1
1.5 1.5 1.3 1.9
20 16 17 37
27 15 28 48
19 67 126 70
40 57 94 96
10 10 9 13
Santo Domingo
TSD1 TSD2 TSD3 TSD4 TSD5 TSD6 TSD7 TSD8 TSD9 TSD10 TSD11 TSD12 SDTC3
42 12 14 44 55 57 65 78 81 41 45 41 43
12 20 59 8 4 5 5 4 4 33 2 21 12
8 11 12 2 3 3 2 2 2 20 2 3 2
36 5 14 20 17 24 119 60 53 107 25 70 20
0.2 2.6 0.4 0.2 0.3 0.3 5.1 0.3 3.3 0.2 0.2 0.3 0.2
3 9 13 2 12 6 12 3 5 13 5 2 11
16 20 15 6 21 21 15 7 13 26 2 9 5
5 132 31 4 7 6 5 6 7 14 4 7 4
5 9 25 3 4 0.5 13 30 18 25 2 3 3
1 9 1 2 1 1 7 1 2 1 1 1 2
Villa PaÂez
TVP1 TVP3
2 5
56 83
12 13
10 5
0.3 0.2
7 3
28 11
18 30
25 5
5 4
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
393
Table 2 (continued) Location
Sample
B
V
Cr
Mn
Fe
Co
Ni
Cu
Zn
As
TVP4 TVP5 TVP7 TVP8 TVP9 TVP10 TVP11 TVP12 TVP13 TVP14 TVP16 TVP17 TVP18 TVP19 TVP20 TVP21 TVP22 TVP23 TVP24
1 2 3 4 8 10 9 6 4 4 6 8 7 6 9 4 8 13 9
57 64 19 3 5 4 3 5 6 27 13 11 15 8 8 31 3 2 8
21 13 3 1 3 2 1 1 1 4 3 5 3 4 2 5 1 1 1
16 4 25 6 12 1 5 3 3 9 40 23 8 31 9 115 24 89 7
1.1 0.1 0.1 0 0.1 0 0 0 0.1 0.2 0.1 0.6 0.1 3.1 0 0.2 0.1 0.4 0.1
7 9 5 2 11 2 5 5 2 13 6 3 2 3 37 9 7 5 1
35 24 13 4 7 4 1 6 2 6 17 10 4 7 6 35 9 14 2
37 33 17 4 6 5 5 12 9 13 11 22 12 10 6 41 6 3 13
46 23 18 8 9 1 9 34 3 9 28 23 6 24 7 80 30 54 16
28 6 4 1 1 1 1 1 1 3 4 3 1 1 1 2 1 2 1
Zea
MZ3 MZ4
23 30
4 32
2 3
31 2
0 1.3
23 143
19 10
8 12
98 9
1 5
Location
Sample
Mo
Ag
Casigua
CASI1
1
1
0.1
8.7
0.1
La Virgen
THV20
31
8
0.1
0.4
4
Las Adjuntas
TLAS1 TLAS2 TLAS3 TLAS4 TLAS5 TLAS6 TLAS7 TLAS8 TLAS9 TLAS10 TLAS11 TLAS12 TLAS13 TLAS14 TLAS16
10 14 23 25 310 211 264 290 153 18 19 8 19 148 13
1 2 2 0.5 3 2 10 3 1 2 1 0.5 1 16 23
0.1 0.1 0.2 0.1 0.1 0.7 0.6 0.6 0.2 0.1 0.1 0.1 0.1 0.1 0.2
0.1 0.2 1 0.1 1 14 3.7 5.6 4.5 2 0.2 0.4 0.6 0.1 2.2
Las Dantas
TLD7 TLD8 TLD9
6 6 3
3 0.5 0.5
0.1 0.1 0.1
Las Mesas
TLM1 TLM2 TLM3 TLM4
6 3 3 4
1 2 0.5 8
San Pedro del Rio
TSP1 TSP2 TSP3 TSP4 TSP6 TSP8 TSP9 TSP10 TSP11 TSP12 TSP13
13 453 13 11 17 605 18 6 17 31 40
5 1 6 20 0.5 1 8 10 11 13 12
Sr
Cd
Sb
Ba
La
Pb
Bi
Th
7
4
8
2
1
33
11
13
1
2
0.1 2 3 3 2 0.1 3 5 5 3 0.1 0.1 2 0.1 0.1
12 17 113 27 79 310 1181 232 128 31 26 30 30 10 18
1 1 13 1 1 4 12 10 3 2 1 1 1 3 6
4 5 13 1 1 3 5 12 6 1 1 1 3 2 9
1 1 1 1 1 1 1 1 2 2 2 2 2 2 2
1 1 2 1 1 1 6 5 4 1 1 1 1 1 2
0.1 0.2 0.1
0.1 0.1 2
11 9 4
1 1 1
3 3 2
2 2 3
1 1 1
0.1 0.1 0.3 0.2
1.3 0.1 0.5 0.8
2 0.1 0.1 3
30 9 58 22
3 1 2 2
8 6 41 5
2 4 1 2
1 1 1 2
0.5 0.5 0.2 0.5 0.5 0.1 0.1 0.1 0.2 0.2 0.1
0.7 0.2 0.6 12 0.1 0.2 0.6 0.2 1.8 0.4 0.4
8 199 24 34 16 612 22 10 39 150 12
5 1 4 1 1 1 11 4 10 5 3
9 4 6 5 2 4 5 5 6 14 4
3 2 2 1 4 1 1 1 1 1 1
1 1 2 1 1 1 2 1 4 4 1
2 3 5 2 2 3 6 5 3 4 5
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
394 Table 2 (continued) Location
Sample
Sr
Mo
Ag
TSP14 TSP15 TSP16
311 401 16
3 11 19
0.1 0.5 0.1
1.7 1.3 4.1
3 5 4
23 68 62
1 22 2
2 14 3
1 1 1
1 4 1
San SimoÂn
TSS1 TSS2 TSS3 TSS4
6 6 11 3
8 8 7 6
0.1 0.7 0.7 0.5
0.3 1.1 2.5 1.4
2 0.1 0.1 2
25 87 75 13
3 49 55 19
10 27 26 15
1 1 2 1
2 18 19 9
Santo Domingo
TSD1 TSD2 TSD3 TSD4 TSD5 TSD6 TSD7 TSD8 TSD9 TSD10 TSD11 TSD12 SDTC3
127 25 137 234 416 461 293 382 305 340 326 294 241
0.5 14 4 1 0.5 0.5 2 0.5 1 0.5 1 1 2
0.1 0.1 0.3 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1
0.2 1.4 0.5 0.1 0.1 0.1 0.7 0.1 0.4 0.4 0.1 0.1 0.2
2 3 5 3 3 0.1 0.1 2 0.1 2 2 2 3
144 328 304 288 399 398 12 300 16 308 355 352 293
1 13 8 1 2 1 2 1 2 6 1 4 1
4 6 8 1 4 2 4 9 7 8 2 1 3
2 1 1 1 1 1 1 1 1 1 1 1 1
1 2 6 1 1 1 2 1 1 3 1 1 1
Villa PaÂez
TVP1 TVP3 TVP4 TVP5 TVP7 TVP8 TVP9 TVP10 TVP11 TVP12 TVP13 TVP14 TVP16 TVP17 TVP18 TVP19 TVP20 TVP21 TVP22 TVP23 TVP24
9 7 9 50 21 8 17 13 10 19 23 8 27 13 37 40 17 31 18 13 7
5 7 8 11 3 0.5 1 1 0.5 1 1 4 1 2 0.5 1 1 1 1 1 1
0.1 0.1 0.1 0.3 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.2 0.1 0.2 0.1
0.4 0.5 0.7 0.3 0.9 0.2 0.4 0.2 0.2 0.3 0.3 0.4 1.3 0.9 0.1 0.1 0.1 1.4 0.2 0.8 0.1
4 2 5 3 2 2 0.1 2 2 2 2 3 2 2 2 2 4 3 2 4 2
72 31 68 86 24 10 29 13 17 62 33 61 34 21 41 43 23 63 17 21 13
6 14 8 13 4 1 1 1 1 1 5 6 3 4 7 3 1 3 1 1 1
13 4 22 18 1 1 1 1 1 1 1 11 1 3 3 3 1 5 3 1 4
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
4 7 9 4 1 1 1 1 1 1 1 3 1 2 3 2 1 1 1 1 1
Zea
MZ3 MZ4
2 4
1 4
0.5 0.1
0.6 0.1
0.1 4
3 20
3 9
2 5
1 1
1 2
By plotting each element against ash% (Nicholls, 1968), the organic/inorganic af®nity can be determined. A positive trend in this plot suggests an inorganic af®nity of the element (Fig. 5a). The majority of the analyzed elements show this behavior: Mg, Al, P, K, Ti, V, Cr, Fe, Ni, Cu, Zn, As, La, Pb, and Th. However, B and Co show an opposite trend, suggesting a probable organic af®nity (Fig. 5b). Ag, Sr, Mn, Sb, Bi, Ca, Ba and Na, on the other hand, show no correlation with ash. Four factors or associations are de®ned by factor analysis. A main factor is substantially built by the elements Ni, Cd, Co and to a lesser extent by Ag, just these elements were found slightly or strongly enriched in the coal samples
Cd
Sb
Ba
La
Pb
Bi
Th
studied. These results indicate that the enriched elements have a common origin, possibly from a sediment source highly concentrated in such elements. Other statistical associations are shown in Table 4.
6. Discussion 6.1. Organic/inorganic af®nities of elements The Nicholl's plot provides evidence for an af®nity of most elements for the inorganic fraction. Elements such as V, Ni, Cu, and Zn show this behavior, contrary to coal
Table 3 Statistical results from Table 2 compared with average values in various coals from around the world; n.r. not reported Geometric mean for TaÂchira Coals (this work)
Average values for Illinois Coals a
Geometric mean for USA Coals b
Fording Mine, BC Canada c
Average for Most Coals d
Average for Terrestrial Crust e
Mo Cu Pb Zn Ag Ni Co Mn Fe f As Th Sr Cd Sb Bi V Ca f Pf La Cr Mg f Ba Ti f B Al f Na f Kf
2 11 4 17 0.11 11 7 15 0.25 3 2 28 0.46 1 1.2 10 0.086 0.0016 2.8 3.3 0.030 43 0.008 8 0.111 0.01 0.01
6.2 13 15 87 0.03 19 6 40 1.9 7.4 1.9 30 0.59 0.81 n.r. 29 0.51 0.0045 6.4 16 0.05 75 0.06 98 1.2 0.03 0.16
1.2 12 5.0 13 0.01 9 3.7 19 0.75 6.5 1.7 90 0.14 0.61 n.r. 17 0.23 0.002 3.9 10 0.07 93 0.06 30 1.1 0.04 0.10
0.9 14 n.r. 9 n.r. n.r. 1.4 81 0.68 0.2 1.3 191 n.r. 0.3 n.r. 12 0.12 n.r. 5.2 2.2 0.09 637 0.037 47 0.72 0.004 0.05
1.5 15 10 25 0.03 15 4 60 n.r. 3 4 100 0.1 0.7 0.05 25 n.r. 0.031 2.7 20 n.r. 185 n.r. 45 n.r. n.r. n.r.
1.5 25 12.5 52 n.r. 13 10 950 5 1.8 10.5 375 0.2 0.2 n.r. 135 5.37 0.105 30 35 2.09 425 0.44 10 8.1 2.83 2.59
a b c d e f
Gluskoter et al., 1977. Finkelman, 1999. Goodarzi, 1988. Swaine, 1990. Mason and Moore, 1982. Expressed as%.
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
Element
395
396
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
Fig. 5. Nicholl's plot for Cr (a) and Co (b) in the studied coals (after Nicholls, 1968).
samples from other parts of the world (Smith, 1980; Van der Flier-Keller and Fyfe, 1987; Kizil'shtein, 1997). Of the selected elements, only Co and B seem to show weakly organic af®nities; these elements have been previously reported with the same behavior in other coal samples (Zubovic, 1966; Van der Flier-Keller and Fyfe, 1987). However, data obtained here are not conclusive, because they are not supported by a rigorous statistical test. F-test results on the comparison of trace element data for Ag, Sr, Mn, Sb, Bi, Cd, Ca, Ba, and Na with ash content show no rejected null hypothesis, H0: b 0 (Canavos, 1988), supporting the conclusion that these elements do not correlate with ash content. The lack of correlation of these elements with ash may be due to a different association scheme with coal. Association through chemical or physical adsorption to the coal
surface (Na, for example, reported elsewhere), simultaneous organic/inorganic behavior (Ca, Sr), or very low values Ð near the determination limit (Ag, Bi) Ð can be responsible for the lack of a clear correlation with ash. 6.2. Enrichment or depletion of elements Strong depletion in most elements, instead of an average ash content (7%), is in contrast with a coal seam at Fording Coal Mine, British Columbia, reported by Goodarzi (1988) to be of similar ash content (7.7%; Table 3). In fact, TaÂchira coals show a systematic depletion in most elements. This result is consistent with a typical domed ombrogenous peat deposit at the time of peat accumulation, as occurs today in the West Kalimantan
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
397
Table 4 Factor analysis results and genetic relationships in trace elements studied
Contributing elements (loading factors .0.400) Other elements in factor (loading factor ,0.400) Features Geochemical inference
Factor 1
Factor 2
Factor 3
Factor 4
Co Ni Cd La
Ca Mg Mn Sr
P V Th
Zn Ba Th Fe
Ba Cu Ti
K Mo Th Cr Cu As Na Ag S La V Pb Cd La Al
Divalent transition metals
Divalent earth-alkaline metals
Source rock: Jurassic volcaniclastic red beds)
Source rock: Cretaceous limestones
peat deposits formed in a poorly drained coastal lowland (Cecil et al., 1993; Neuzil et al., 1993). However, a systematic palynological and petrographical study is needed in order to con®rm the previous statement (Eble and Grady, 1993). The detected slight enrichment of Ag, Co, Cd, Mo, Sb, As, and Bi suggests a common origin for these elements (Ren et al., 1999). This is consistent with the statistical results. 6.3. Statistical associations of elements: geochemical implications A powerful statistical tool (factor analysis) was used to obtain accurate information about trace element associations in the studied coal samples. Four factors were determined (Table 4). The ®rst, Ni±Cd±Co, involves mainly transition divalent elements. Other elements present in this factor, but in a minor weight, are Zn and Ba. This association has a geochemical signi®cance, because the meaningful enrichment that TaÂchira coals exhibit in relatively uncommon elements, such as Cd, Ag, Co and Mo, as well as their statistical association in these samples allow us to infer that these elements originated from a source enriched in these elements. Rocks with high relative concentrations of the cited elements tend be of volcanic or hydrothermal origin (Mason and Moore, 1982). Analyzing the pre-existing rocks in the lithological column of the Venezuelan Andes, it is deduced that the red beds of La Quinta Formation are responsible for enrichment in Ag and other exotic elements in coals of Los Cuervos Formation. La Quinta Formation is anomalous in chalcophile elements: the Zn and Ni tenors oscillate among 54±125 and 54±81 ppm, respectively (Audemard, 1982). La Quinta Formation crops out at various places in the Andes (Perija Mountains, Eastern Chain of the Colombian Andes) and appears in the subsurface of Lake Maracaibo and in the Llanos Basin. This unit is mainly composed of volcaniclastics, especially in the lower section of the unit where it frequently consists of rhyolitic and dacitic tuffs. These materials are rich in Cu, Pb, and Zn sul®des. Today, there are several metallic sul®de mineralizations
Mainly lithophile and chalcophile elements Mainly clays and pyrite
W As Cu La Incompatible elements Heavy minerals derived from a felsic source rock like granite, granitic gneiss, and pegmatite
and ores on Jurassic terrain (La Quinta): Seboruco, El Cobre, and Pregonero, in TaÂchira State and in several locations in PerijaÂ. The remaining statistical associations belong to other sediment sources. The second factor, Th±V±P, most likely reveals a heavy mineral association, probably with vanadinite among mineral phases. Other elements associated with this factor, although to a lesser extent (factor loading ,0.400), are Cu, W, and La. These mineralogical assemblages, together with heavy lithophilic, incompatible elements, are consistent with a felsic source (granites, granitic gneisses, pegmatites). This lithological composition is present within the Iglesias Group, which comprises the basement of Los Andes (PerijaÂ) and has the equivalent lithological designation in the Oriental Mountain Range of the Colombian Andes. A third factor consists of earth-alkaline metals: the Ca± Mg±Mn factor. Other elements included in this association are Sr, Ba, and Ti. This factor suggests the presence of a calcareous source, such the Cretaceous units exposed during the period of accumulation of the peats. These carbonates were exposed in the Paleocene while the Arauca Arch was a positive zone (MartõÂnez, 1996). Finally, the fourth factor, K±Mo±Th, is very wide; there are eleven elements within this association (loading factor . 0:400; Table 4). Detailed analysis of metals belonging to this factor reveals at least two different habitats for trace elements: (1) detrital minerals, mainly metal-adsorbed clays; and (2) together with authigenic pyrite, as in substitution of S in a pyrite crystal arrangement, and Pb, Cd and Cu replacing Fe. 7. Conclusions TaÂchira coals are depleted in most trace elements content relative to many well-studied coals worldwide and the worldwide averages. On the other hand, the TaÂchira coals present a conspicuous enrichment in Mo, Co, As, Sb, Bi, Cd, and Ag with respect to coals in the USA and worldwide. Among the elements analyzed, only Co and B have a distinctive organic af®nity with the coal matrix. Almost 15 of 21 analyzed elements exhibit inorganic af®nities; the
398
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399
species Ag, Sr, Mn, Sb, Bi, Cd, Ca, Ba, and Na show no correlation with organic or inorganic matrix. The enriched elements in the coal samples (Ag, Co, Mo, Cd) appear to have a common origin and the Triassic±Jurassic volcaniclastic rocks of La Quinta Formation as the probable initial source. Other geochemical associations indicate other source rocks at the time of peat deposition (granites, pegmatites, granitic gneisses; Cretaceous limestone) or the presence in the coal samples of clays and authigenic sul®des (mainly pyrite). Factor analysis has been shown to be a powerful tool for the geochemical analysis of trace elements in coal. Acknowledgements This work was supported by the Consejo de Desarrollo Cienti®co y Humanistico at the Universidad Central de Venezuela, CDCH Projects C-03.32.2274/93 and C03.32.2274/1999. The LECO SC-432 sulfur analyzer and the CARLO ERBA 1106 elemental analyzer were acquired through CONICIT Project S1-2453. Constructive comments from the reviewers greatly improved the manuscript. References ASTM, 1985. Annual Book of ASTM Standards Ð Petroleum Products, Lubricants and Fossil Fuels. American Society for Testing and Materials, West Conshohocken, PA, USA. Audemard, F., 1982. Geology and Copper Mineralizations of La Quinta Formation, Sierra de Perija, and Western Zulia, Venezuela. Unpublished MSc Thesis. Colorado School of Mines, Golden CO USA. Bailey, A., 1981. Chemical and mineralogical differences between Kittanning coals from marine-in¯uenced versus ¯uvial sequences. J. Sedim. Petrol. 51 (2), 383±395. Banerjee, I., Goodarzi, F., 1990. Paleoenvironment and sulfur±boron contents of the Mannville (Lower Cretaceous) coals of southern Alberta, Canada. Sedim. Geol. 67, 297±310. Bar, T., PenÄa, R., 1985. Yacimiento del carboÂn de Santo Domingo, Estado TaÂchira. VI Congreso GeoloÂgico Venezolano (Caraas). Memorias, 3799±3831. Bohor, P., Gluskoter, H., 1973. Boron in illite as an indicator of paleosalinity of Illinois coals. J. Sedim. Petrol. 43 (4), 945±956. Bouska, V., 1981. Geochemistry of Coal. Elsevier Science, Prague, Czechoslovakia (translated by H. Zarubova) 284 pp. Breger, I., 1958. Geochemistry of coal. Econ. Geol. 53, 823±841. BricenÄo, J., 1992. DistribucioÂn de Elementos Minoritarios y Traza en Carbones de la Franja Nororiental, Estados TaÂchira y MeÂrida. Unpublished Degree Thesis. Universidad Central de Venezuela, Caracas, Venezuela. Canavos, G., 1988. Probabilidad y EstadõÂstica. Aplicaciones y MeÂtodos. McGraw-Hill, New York, USA. Cecil, C., Dulong, F., Cobb, J.C., Supardi, 1993. Allogenic and authogenic controls on sedimentation in the central Sumatra basin as an analogue for Pennsylvanian coal-bearing strata in the Appalachian basin. Modern and Ancient Coal-Forming Environments, Cobb, J.C., Cecil, C.B. (Eds.). Geol. Soc. Am. 286, 3±22 Special Paper. Crowley, S., Warwick, P., Ruppert, L., Pontolillo, J., 1995. The origin and distribution of HAPs elements in relation to maceral composition of the A1 lignite bed (Paleocene, Calvert Bluff Formation, Wilcox Group), Calvert Mine area, east-central Texas. Coal Geology of the Paleocene± Eocene Calvert Bluff Formation (Wilcox Group) and the Eocene
Manning Formation (Jackson Group) in East-Central Texas, Warwick, P.D., Crowley, S.S. (Eds.). US Geol. Surv., Open-File Rep., 95±595, 71±86. Dixon, W., Kronmal, R., 1965. The choice of origin and scale for graphs. J. Assoc. Comp. Mach. 12, 259±261. Eble, C., Grady, W., 1993. Palynologic and petrographic characteristics of two Middle Pennsylvanian coal beds and a probable modern analogue. Modern and Ancient Coal-Forming Environments, Cobb, J.C., Cecil, C.B. (Eds.). Geol. Soc. Am. 286, 119±138 Special Paper. Escobar, M., MartõÂnez, M., 1993. Los depoÂsitos de carboÂn en Venezuela. Interciencia 18 (5), 224±229. Escobar, M., MartõÂnez, M., 1997. CaracterõÂsticas y aplicaciones del carboÂn en Venezuela. Interciencia 22 (1), 1±10. Finkelman, R., 1999. Trace elements in coal: Environmental and health signi®cance. Biol. Trace Elem. Res. 67, 197±204. Gayer, R., Rose, M., Dehmer, J., Shao, Y., 1999. Impact of sulphur and trace element geochemistry on the utilization of a marine-in¯uenced coal Ð Case study from the South Wales Variscan Foreland basin. Int. J. Coal Geol. 40, 151±174. Gluskoter, P., Ruch, R., Miller, W., Cahill, R., Dreher, G., Kuhn, J., 1977. Trace elements in Coal: Occurrence and Distribution. Illinois Geological Survey, Circular 449, Champaign IL USA, 154 p. GonzaÂlez de Juana, C., Iturralde de, A.J., Picard, X., 1980. GeologõÂa de Venezuela y Sus Cuencas PetrolõÂferas. Ediciones Foninves, Caracas, Venezuela. Goodarzi, F., 1988. Elemental distribution in coal seams at the Fording coal mine, British Columbia, Canada. Chem. Geol. 68, 129±154. Harris, L., Barrett, H., Kopp, O., 1981. Elemental concentrations and their distribution in two bituminous coals of different paleoenvironments. Int. J. Coal Geol. 1, 175±193. Hart, R., Leahy, R., Falcon, R., 1982. Geochemical investigation of the Witbank Coal®eld using instrumental neutron activation analysis. J. Radioanal. Chem. 71, 285±297. Hartstein, A., Freedman, R., Platte, D., 1973. A novel wet-digestion procedure for trace-metal analysis of coal by atomic absorption. Anal. Chem. 45 (3), 611±614. Kizil'shtein, L., 1997. Vanadium geochemistry of coal: an ecological aspect. Geochem. Int. 37 (1), 63±68. Krejci-Graf, K., 1984. Uber die Elemente in Kohlen. Erdol Kohle ErdgasPetrochemie 37 (10), 451±457. Marquez, X., Mederos, S., 1989. Areniscas deltaicas del Grupo OrocueÂ: Potenciales yacimientos petrolõÂferos. VII Congreso GeoloÂgico Venezolano (Barquisimeto). Memorias, 773±793. MartõÂnez, M., 1996. Estudio GeoquõÂmico Regional de los Yacimientos CarbonõÂferos del Estado TaÂchira. Unpublished PhD Thesis. Universidad Central de Venezuela, Caracas. Mason, B., Moore, C., 1982. Principles of., Smith and Wyllie Intermediate Geology Series. 4th ed. Wiley, New York, USA p. 344. Miller, R., Given, P., 1987. The association of major, minor and trace inorganic elements with lignites: II. Minerals, and major and minor element pro®les, in four seams. Geochim. Cosmochim. Acta 51 (5), 1311±1322. Neuzil, S., Supardi, Blaine, C., Kant, J., Soedjono, K., 1993. Inorganic geochemistry of domed peat in Indonesia and its implication for the origin on mineral matter in coal. Modern and Ancient Coal-Forming Environments, Cobb, J.C., Cecil, C.B. (Eds.). Geol. Soc. Am. 286, 23± 44 Special Paper. Nicholls, G., 1968. The geochemistry of coal-bearing strata. In: Murchison, D., Westoll, T.S. (Eds), Coal and Coal Bearing Strata (Papers Contributed to the 13th Inter-University Geological Congress, Newcastleupon-Tyne, 1965) pp. 269±307. American Elsevier Publishing, New York USA, 418 p. O'Connor, J., 1986. Discriminant analysis of trace elements in coal beds of Early and Middle Pennsylvanian age from the Central Appalachian Basin. In: Carter, L.M.H. (Ed), USGS Research on Energy Resources, 1986: Program and Abstracts (Second V. E. McKelvey Forum on
M. MartõÂnez et al. / Journal of South American Earth Sciences 14 (2001) 387±399 Mineral and Energy Resources, Denver, February 1986). US Geological Survey, Circular 0974, Reston VA USA, pp. 46±47. Ren, D., Zhao, F., Wang, Y., Yang, S., 1999. Distribution of minor and trace elements in Chinese coals. Int. J. Coal Geol. 40 (2/3), 109±118. Smith, R., 1980. The trace element chemistry of coal during combustion and the emissions from coal-®red plants. Prog. Energy Combust. Sci. (UK) 6, 53±119. Swaine, D.J., 1990. Trace Elements in Coal. Butterworths, London, UK p. 278. Van der Flier-Keller, E., Fyfe, W., 1987. Geochemistry of two Cretaceous coal-bearing sequences: James Bay lowlands, northern Ontario, and Peace River basin, northeast British Columbia. Can. J. Earth Sci. 24, 1038±1052. Valkovic, V., 1983. Trace Elements in Coal. CRC Press, Boca Raton, FL, USA.
399
Ward, C., Spears, D., Booth, C., Staton, I., 1999. Mineral matter and trace elements in coals of the Gunnedah Basin, New South Wales, Australia. Int. J. Coal Geol. 40, 281±308. Yudovich, Y., Korycheva, A., Obruchnikov, A., Stepanov, Y., 1972. Mean trace-element content in coals. Geochem. Int. 9, 712±720. Zhou, Y., Ren, Y., 1981. Gallium distribution in coal of Late Permian coal®elds, southwestern China, and its geochemical characteristics in the oxidized zone of coal seams. Int. J. Coal Geol. 1, 235±260. Zubovic, P., 1966. Physicochemical properties of certain minor elements as controlling factors in their distribution in coal. Coal Science (American Conference on Coal Science, University Park PA, 1964). Advances in Chemistry 55American Chemical Society, Washington DC, USA, pp. 221±231.
