Coal Science - ACS Publications

---

13 Physicochemical Properties of Certain Minor Elements as Controlling Factors in their Distribution in Coal

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

PETER ZUBOVIC U. S. Geological

Survey, Washington,

D . C.

Analyses of sink-float separates of coal show the following order for the organic affinity of the following elements: Ge>Be>(Ga, Ti, Β, V) Ni>(Co, Y)>Mo>Cu>Sn>La>Zn. Studies in chelate chem­ istry show that generally the chelate stabilities of the

bivalent metals is Be>Cu>Ni>Co>Zn>Fe.

Except for the copper this is the same order for the organic affinity of these metals in coal. Copper could be in a reduced state in the coal depositional environment or precipitated as a sulfide.

The

trivalent metals—gallium, yttrium, and lanthanum —have the same order for their organic affinity and chelate stability. Most of the metals show a good relation between their organic affinity, ionic potential, and chelate stability.

|n a study of the geochemistry of metallic elements i n coal or other carbona­ ceous geological materials, several problems are encountered. Although m u c h data o n the minor element content of such materials exists, data on the distribution of elements among the organic matter, syngenetically formed minerals, a n d detrital minerals are very scarce. Furthermore, except for several general postulations first proposed b y Coldschmidt (6) the genesis of most minor elements i n carbonaceous rocks is almost completely u n k n o w n . T h i s paper attempts to correlate the available data on distribution of minor elements i n organic a n d inorganic matter of coal w i t h some of the k n o w n chemical prop­ erties of these elements and to determine patterns of their accumulation a n d distribution i n organic a n d inorganic phases of coal. This paper is primarily concerned w i t h the minor elements—beryllium, boron, titanium, vanadium, chromium, nickel, cobalt, copper, zinc, gallium, germanium, molybdenum, t i n , yttrium, a n d lanthanum—and to a lesser extent w i t h the major elements— iron, aluminum, a n d silicon. 221 Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

C O M SCIENCE

222

Goldschmidt (6) proposed three general methods for the accumulation of elements f o u n d i n organic matter of coal:

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

(1) Accumulation b y plants during growth. (2) Accumulation b y complex formation during decay of plant matter. (3) Accumulation after burial. These postulations generally have been accepted b y most investigators. T h e author feels that i n large coal deposits the first t w o are the dominant methods of accumulation a n d that the last process takes place only on a very limited scale, such as i n isolated fragments of carbonized wood a n d i n the tops of some coal beds, a n d therefore, that the accumulation of these elements is essentially a syngenetic process. C h e m i c a l data o n the relation of minor elements a n d the accompanying organic matter are relatively sparse. Breger a n d others (3) suggested that uranium as w e l l as nickel, cobalt, beryllium, molybdenum, titanium, vanadium, chromium, a n d t i n present i n a démineraii z e d lignite are held essentially as metal organic complexes. T h e y detected germanium i n the mineral fraction but not i n the original or the mineral-free part of the lignite, whereas Manskaya a n d others ( I I ) reported that about 8 0 % of the germanium present i n a peat was bound to organic matter w h i c h is soluble i n dilute alkali solutions a n d that 3 0 % of this was bound to strongly polymerized h u m i c acids. Table I.

Average Organic Affinity of Some Metals Determined by Float-Sink Methods Element

Germanium Beryllium Gallium Titanium Boron Vanadium Nickel Chromium Cobalt Yttrium Molybdenum Copper Tin Lanthanum Zinc

Percent organic affinity (18) 87 82 79 78 77 7β 59 55 53 53 40 34 27 3 0

Percent organic association (9) 100 75-100 75-100 75-100 75-100 100 0-75 0-100 25-50 N.D.* 50-75 25-50 0 N.D.* 50

• N.D.—not determined.

Physical separations of coal b y flotation (9, 17, 18) have produced data showing the organic affinity of a number of minor elements. Horton a n d A u b r e y (9) studied the distribution of minor elements i n five different density fractions of each of three vitrain samples. B y comparing the distribution of the elements w i t h that of the organic a n d inorganic matter, they calculated the percent association of the elements w i t h the organic matter. T h e i r data are summarized i n Table I. Zubovic a n d others (17,18) reported on the separation of 13 samples of Eastern Interior Region coals into sink-and-float fractions. T h e average organic affinity for the elements as derived from the 13 samples

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

13.

ZUBOVIC

Minor Elomont Proporth*

223

is also shown i n Table I. There is general agreement between the two sets of data i n Table I for those elements forming small highly charged ions. D a t a for cobalt, t i n , a n d zinc show less agreement. These data are plotted i n F i g u r e 1 a n d c o m p a r a i w i t h the ionic potentials of the metals.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

cr

Ge Be GoTi θ V Ni Cr Co Y MoCuSn LoZn Decreasing Affinity for the Organic Matter of Coal a 10, 15 2

>

3+

ο

^ - ^ s ^

^

°

2+

0~-i

Gt Bt Go Ti θ V Ni Cr Co Y Mo Cu SA Lo Zn Decreasing Affinity for the Organic Matter of Coal b Figure 1. Relation of organic affinity ionic potential of the elements

and

W h e n the data for vanadium, nickel, cobalt, copper, and iron i n petroleum of the Western Interior Region (15) shown below are divided b y the average crustal abundance of these elements, the relation, V > N i > C o > C u > F e is derived.

Vanadium Nickel Iron Copper Cobalt

/

//

P.p.m. in crude-oil

Average crustal abundance in p.p.m.

110 55 β 2 1

100-150 80 50,000 70 23

Ratio 1/II

1.0-0.73 0.69 0.0001 0.03 0.04

This can be considered to be the relation i n w h i c h these metals have been preferably complexed b y the progenitors of petroleum. Except for iron, w h i c h

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

224

COAL SCIENCE

was not determined i n these coals, the relation, V > N i > C o > C u , is similar to that derived from coal shown i n Table I and Figure 1. F r o m these three sets of data this author believes that the affinity of the elements for naturally occurring organic matter generally follows the sequence shown i n Figure l a . It should be pointed out that for any individual element, considerable variability has been found among the samples, and the data presented i n Table I and Figure 1 are the averages for 13 samples.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

Relation

of Metal-Organic

Affinity

to Complex

Formation

A proof of the validity of the organic affinity series could best be made by studying the complexes, if present, of these metals i n coal and other carbonaceous geological materials. However, except for the porphyrin chelates of some of the transition metals, no complexes have been extracted from such materials. Consequently, the only other approach is to assume that this is a stability series for these metals i n naturally occurring organic matter and to compare this stability sequence w i t h the stability sequences determined i n laboratory studies made on their organic complexes. It must be further assumed that the elements organically associated i n geologically o l d carbonaceous materials are present i n the most stable complex possible to survive through geologic time. T h e most stable metal-organic complexes are those wherein ring structures are formed; these complexes are called chelates. The most stable chelates form five- and six-membered rings w i t h the metals ( 1 3 ) . One property of the metals conducive to the formation of stable chelates is the ionic potential (charge/ radius) of the metals. Figure l b shows the organic affinity series plotted against the ionic potentials of the metals. There is general agreement i n that the metals w i t h high ionic potentials have high organic affinities i n coal, and those w i t h lower ionic potentials have lower organic affinities. A further indication that the metals associated w i t h the organic matter of coal may be chelated is the correlation between their organic affinity a n d the stability constants of their chelates as determined i n laboratories. Basolo and Pearson ( I ) suggest that for some of the tervalent ions of this report the general order for the stability of their chelates is G a > Y > L a . T h e same order for these three elements is shown i n Figure 1. Furthermore, data from M e l l o r and M a l e y ( 1 4 ) , Irving and Rossotti ( 1 0 ) , and summaries of these data by Basolo and Pearson ( I ) suggest that the stability of chelates of bivalent ions is B e > C u > N i > C o > Z n > F e . Except for copper, this is the same order of these metals i n their organic affinity for coals as shown i n Figure 1 and i n the subsequent discussion on petroleum. F r o m these observations it appears to this author that that portion of gallium, yttrium, lanthanum, beryllium, nickel, cobalt, and zinc that is bonded to organic matter i n coal is held as chelated complexes. The titanium, however, chelates. complex.

lack of adequate experimental data on chelates of germanium, boron, and vanadium prevents a similar comparison for these elements; their high ionic potentials do indicate that they should form stable V a n a d i u m does occur naturally i n the very stable vanadium porphyrin F o r similar reasons, molybdenum and tin are not discussed further.

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

13.

ZUBOVIC

Minor Element Properties

225

In coal, copper shows an organic affinity different from the M e l l o r a n d M a l e y (14) series, and zinc appears to be completely associated w i t h the inorganic matter. Copper i n the physicochemical environment of coal deposition can be reduced to the univalent state ( 5 ) . Univalent copper cannot be expected to behave like the bivalent metals. T h e ionic potential of univalent copper fits reasonably well into the organic affinity series (Figure l a ) . In the presence of hydrogen sulfide produced by anaerobic bacterial activity, particularly sulfate reducers, conditions are created whereby sulfides of copper and zinc could be formed. T h e partition of these metals between the sulfide phase and the organic phase depends on the relation between the stability constants of the complexes and the solubility product of the sulfides of these metals. Elements w i t h small solubility products of their sulfides and low stability constants of their chelates w o u l d be expected to go into the sulfide phase when hydrogen sulfide is present. Copper is typical of such elements. Chalcocite has a solubility product of about 10"·° and covellite about 10" , whereas the most stable chelates of copper have stability constants of about 10"*. Consequently, copper could be expected to be accumulated as the sulfide. Zinc sulfide has a m u c h larger solubility product; however, the stability of its chelates is lower. F r o m the fact that zinc appears to be completely associated w i t h the inorganic fraction of coal, it can be assumed that the relation between the solubility product of any of its sulfides and its chelates favors formation of the sulfide. Iron could be expected to follow a similar pattern. T h e data i n Table I suggest that copper may be associated to some extent w i t h organic matter. This could occur if hydrogen sulfide were not available, owing either to cessation of sulfate-reducing bacterial activity or the paucity of sulfate w h i c h could be reduced. It is also possible that a stronger chelating agent is available or that the relation of the concentrations of the chelating agent to hydrogen sulfide favors formation of the chelate. This approach can also be used for the elements that form insoluble hydroxides i n a slightly acid meduim, such as beryllium, aluminum, gallium, and chromium or for any other insoluble inorganic phase w h i c h could be formed A l l the elements studied show a degree of inorganic association. Unquestionably, a certain, i n most cases small, percentage of the elements is incorporated i n the detrital mineral matter. Some of these elements generally are found i n very resistant minerals: for example, boron i n tourmaline, yttrium a n d lanthanum i n rare earth minerals such as xenotime and monazite, and beryllium in beryl. Other elements, such as copper and zinc mentioned previously and perhaps vanadium, gallium, molybdenum, and tin (among the elements studied here), could be syngenetically formed as sulfides under certain conditions. In addition to these elements, iron, as the sulfides pyrite and marcasite, makes u p a large portion of the mineral matter of many coals. 44

Relation of the Metals to Donor

Elements

T h e donor element is an important factor i n chelate stability. Generally, certain metals form more stable complexes when bonded to a preferred donor

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

226

COM

SCIENCE

atom. M a r t e l and C a l v i n (13) presented a series of preferences of metals for donor elements. F o r the metals discussed i n this report, these are:

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

Ο > Ν 0 = N N > 0

Ga, Ge, T i , Sn, V ( IV ), V ( V ), Co Ge, Cr, Fe Cu(I),Cu(II),V(III),Ni

W h e r e the valency of a metal is not indicated, the normal valency of the metal is assumed. Beryllium probably is placed i n the 0 = N group because of the stability of its phthalocyanine chelate. Most often B e forms very stable bonds w i t h oxygen as the donor element. V a n a d i u m , nickel, a n d copper from the Ν > Ο group and iron from the O N group are the elements most frequendy found i n petroleum, chelated w i t h porphyrin ligands. T h e porphyrin chelate contains four nitrogens as donor elements. Possible Complexing

Agents in Decaying

Plant

Materials

In the presence of decaying plant material a large variety of organic ligand molecules are made available to the metals released from decomposing soil and rock. It is believed that i n such a heterogeneous environment the pre­ ceding preferences b y the metals for certain donor atoms are followed. This author believes that some of the most stable complex-forming ligands present i n decomposing plant remains w h i c h w o u l d produce stable chelates are chlorophyll, amino acids, and lignin derivatives. T h e first two could be the chelating agents for trivalent vanadium, nickel, copper, a n d to some extent iron—i.e., the elements w h i c h prefer nitrogen as the donor element. Beryllium, germanium, gallium, titanium, cobalt, a l u m i n u m , and silicon w o u l d be bonded to the oxygens of lignin derivatives. Other complex formers m a y be present; however the above are k n o w n to have bonding positions whereby stable com­ plexes are formed. T h e magnesium of chlorophyll could be replaced b y vanadium a n d the other elements found i n metal-porphyrins. A m i n o acid chelates of the transition metals are quite soluble and could be responsible for long distance transport of some of these elements. Furthermore, this author believes that the preservation

(a) Figure 2.

(b)

(c)

Structures of (a) coniferal alcohol, (b) sinapyl (c) β-isopropyltropolone, and (d) catechol.

(d) alcohol,

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

13.

ZUBOVIC

Minor Moment Properties

227

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

of amino acids could, i n coal, peat, lignite, and other carbonaceous materials, be the result of the formation of stable metal-amino acid chelates. Protected amino acids preserved i n fossil bones and shells do not necessarily fall into this category. T h e most persistent and stable amino acids appear to be glycine, aspartic acid, and glutamic acid, which were reported i n Paleozoic anthracite from Great Britain by Heijhenskjold and Mollerberg ( 8 ) . A m o n g the most stable amino acid chelates are those formed w i t h copper; however, the stability con­ stants for the α-amino acids of copper do not differ to the point where they indicate that the above mentioned three acids w o u l d be preferentially preserved. In the chelates of glycine, aspartic acid, and glutamic a c i d a l l the carbon and nitrogen atoms form part of the ring structure. This probably is the reason w h y these three acids appear to outlast all others i n high rank coal. It should be pointed out that i n the presence of hydrogen sulfide, copper is precipitated as the sulfide from copper-amino acid chelates. Thus, copper and perhaps other elements could be transported as amino acid chelates until sufficient hydrogen sulfide was encountered to cause precipitation. L i g n i n is considered to be the principal contributor to humic materials w h i c h form coal. L i g n i n is made u p of a large number of monomers such as a and b shown i n Figure 2. T h e 4-hydroxy-3-methoxyphenyl group is called the guaiacyl group. There are a large number of such groups i n a lignin molecule. U p o n its degradation to humic-like substances, the methoxy content of the lignin is greatly reduced. It is possible that during this degradation process a considerable amount of chelation w i t h metals through the guaiacyl oxygens could take place. V e r y stable chelates are formed w i t h substances such as catechol (d) and the tropolones ( c ) , w h i c h have bonding positions Table II.

Stability Constants for Some Metal Complexes of the Tropolones*' 6

Tropolone 0-Methyltropolone /9-Isopropyltropolone 3,4-Benzotropolone Catechol

Be 15.4 17.1 Ιβ.β 17.1

—

• Data shown are log B which is log K K%. » Data taken from BJerrum (2). « Estimated, (log K» - 7.9) t

Co 12.9 14.1 14.2

— —

Cu ie.e

e

— — — 30.7

Ni 13.8 15.0 15.0 15.0

—

x

similar to those of the guaiacyl groups. T h e data for the tropolones (Table II) show that attached structures on the benzene ring further increase the stability of such chelates. As a result, chelation w i t h large humic-like products of lignin should have an unusual stability. Furthermore, since a number of such sites could be available on any one of these partially degraded lignin polymers, these could, through metal linkages, form the very large highly insoluble h u m i c substances found i n coal, such as the humins or ulmins. This author does not i m p l y that the insolubility is solely caused by the formation of complexes; loss of polar groups from the degrading lignin plays a dominant role i n this phenomenon.

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

228

COAL SCIENCE

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

Manskaya (12) suggests that lignin initially degraded to phenylpropane monomers a n d later condensed structures accumulate i n lignite, and that germanium, vanadium, and uranium become fixed w i t h them. This author does not feel that complete degradation of lignin to its monomers has to take place, but rather that large fragments of the degrading lignin can be chelated and subsequently further alteration can take place through loss of easily detached groups to form the humic-like substances. Another indication that the metals could be bonded to large rather than to small structures is the generally low inherent ash of vitrains and of isolated pieces of coalified wood. V e r y clean vitrains generally contain about 1 % ash. If the metals such as copper were complexed b y monomers such as hydroxylconiferyl alcohol, then the copper content of the resulting complex w o u l d be about 1 6 % or 2 0 % copper oxide as the ash. However, if metal w i t h similar atomic weight (average about 60) were bonded to degraded lignin molecules (molecular weight about 6 0 0 0 - 8 0 0 0 ) , whereby polymers were formed through the metals, the resulting ash w o u l d be about 1 % . T h e fact that germanium and vanadium frequently are reported i n large amounts i n isolated fragments of coalified w o o d (4, 7, 16) suggests that these two elements form the most stable complexes w i t h degraded woody materials. This author believes that vanadium and germanium replace other metals i n existing complexes. Germanium was shown to have the highest organic affinity of the metals i n Figure 1, whereas vanadium ranked below some of the other, smaller, highly charged ions such as beryllium, boron, gallium, and titanium. Furthermore, trivalent vanadium forms stronger bonds w i t h nitrogen as the donor element than with oxygen. Tetravalent vanadium, however, forms more stable chelates w i t h oxygen as the donor element. In order to be chelated w i t h the oxygens of lignins, vanadium may be tetravalent. Since the accumulation of large amounts of germanium i n these coalified woody materials seems to be a replacement reaction, its emplacement must have occurred after burial and initial degradation of the woody material. T h e oxidation-reduction potential could have been higher than i n the initial phases of degradation, particularly if bacterial activity ceased when the replacement b y vanadium occurred. T h e other alternative is that trivalent vanadium is replacing elements chelated to the nitrogens of the humic matter. If such is the case, then it w o u l d appear that the germanium forms the most stable chelates of those elements bonded to oxygen and trivalent vanadium to those bonded, completely or partially, to nitrogen. W h e n the availability of these two elements were sufficient they w o u l d be the two elements complexing a large part of the organic matter that exists as metal chelates i n the isolated fragments of coalified w o o d . Replacement reactions w i t h pre-existing chelates is suggested largely because this author feels that additional chelation cannot take place after compaction and solidification of the organic mass. Reorientation of large organic ligands into positions to provide strain-free chelates w o u l d be difficult i n a solid. If some additional enrichment took place, the metals w o u l d be held as simple complexes, not as chelates. In large deposits of coal, the availability of the best complex-forming metals is quite low. Consequently other metals assume a dominant role i n

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

13.

ZUBOVIC

Minor Element Properties

229

complex formation. A m o n g the minor elements listed i n Table I and Figure 1, titanium is found i n largest amounts. This may result from the fact that its crustal abundance is the highest of these metals. T h e other abundant elements i n the earth's crust, silicon and especially aluminum, probably account for the largest share of the metal-organic complexes i n such coals. Solubility of the minerals i n w h i c h these metals occur is also a controlling factor i n their avail­ ability. T h e position of iron i n this scheme is uncertain. U n d e r most conditions, if hydrogen sulfide is being produced b y bacterial activity, most or all of the iron w o u l d be expected to form sulfides. A s the availability of the metallic elements increases i n relation to the amount of organic matter able to form stable complexes, replacement reactions cause retention of those metals w h i c h form the more stable complexes. Such conditions generally are found i n small isolated lenticular bodies of coal, i n thin-bedded coals, a n d also i n strata representing the initial and last phases of coal-bed deposition. This is probably w h y tops and bottoms of coal beds and thin beds frequently are rich i n ger­ manium. T h e end product of increasing metal and decreasing organic availa­ bility is found i n the germanium, vanadium, a n d at times i n uranium-rich, isolated coalified fragments of trees. Summary Analyses of float-sink separates of coal reveal a systematic variation of the minor elements with the organic matter w h i c h can be arranged into an organic affinity series. This series appears to be related to the chelating prop­ erties of the metals. Deviations i n this series may be explained b y the chemical nature of the depositional environment. It is believed that most of the transition metals are complexed to nitrogen donors, such as are found i n amino acids or derivatives of chlorophyll, a n d that the metals w i t h high ionic potentials, such as beryllium, boron, germanium, titanium, gallium, and major elements such as aluminum and silicon, may be bonded to oxygen donors of degraded lignin. Complexing of amino acids b y some metals of the transition series may be the reason w h y some amino acids are preserved i n coal through geologic time. Furthermore, the complexing of the humic substance derived from lignin b y metals may play a role i n forming the highly insoluble fractions of coal. T h e amount of any given element that is complexed depends upon its availability w i t h respect to the amount of available organic matter and to its position i n a complex stability series of the competing elements. Literature Cited (1) Basolo, F., Pearson, R. G., "Mechanisms of Inorganic Reactions," p. 16, John Wiley and Sons, Inc., New York, 1958. (2) Bjerrum, J., Schwarzenbach, G., and Sillen, L . G., "Stability Constants of Metal­ -Ion Complexes Part I. Organic Ligands," Chemical Society, London, 1957. (3) Breger, I. Α., Deu, M., Rubenstein, S., Econ. Geol. 50, 2 (1955). (4) Breger, I. Α., Schopf, J. M., Geochim Cosmochim. Acta 7, 290 (1955). (5) Garrels, R. M . , "Mineral Equilibria at Low Temperatures and Pressures," pp. 173-200, Harper, New York, 1960. (6) Goldschmidt, V. M., Ind. Eng. Chem. 27, 1101 (1935).

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

COM

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

230

SCIENCE

(7) Hallam, Α., Payne, K . W . , Nature 171, 1008 (1958). (8) Heijkenskjold, F., Mollerberg, H., Nature 181, 335 (1958). (9) Horton, L., Aubrey, Κ. V . , J. Soc. Chem. Ind. 69, Suppl. 1, 541 (1950). (10) Irving, H., Rossotti, H., Acta Chem. Scand. 10, no. 1, 72-93. (11) Manskaya, S. M., Drozdova, T . V . , Kravtsova, R. P., Tobelko, Κ. I., Geokhimiya 1964, 455. (12) Manskaya, S. M., Kodina, L. Α., Geokhimiya 1963, 389. (13) Martell, A . E., Calvin, M., "Chemistry of Metal Chelate Compounds," p. 169, Prentice-Hall, Inc., New York, 1952. (14) Mellor, D . P., Maley, L. E., Nature 161, 436 (1948). (15) Moore, J. W . , Dunning, Η. N., U. S. Bur. Mines Rept. Invest. 5370, 13 (1957). (16) Stadnichenko, T . M., Murata, R. J., Zubovic, P., Hufschmidt, E. L., U. S. Geol. Surv. Circ. 272, 15 (1953). (17) Zubovic, P, Sheffey, Ν. B., and Stadnichenko, T . M., U. S. Geol. Surv. Profess. Papers No. 400-B, B84-B87 (1960). (18) Zubovic, P., Sheffey, Ν. B., Stadnichenko, T . M., U. S. Geol. Surv. Profess. Papers No. 424-D, D345-D348 (1961). R E C E I V E D January 25, 1965. Publication authorized by the Director, U.S. Geological Survey.

Discussion M.-Th. M a c k o w s k y . H a v e y o u found any characteristic associations of macérais with any particular elements? Peter Z u b o v i c . W e have very little such data. W e have separated some macroscopic vitrain a n d fusain bands from blocks of coal a n d analyzed these. Generally germanium, beryllium, boron, a n d gallium are more concentrated i n the vitrain samples. Fusain samples generally contain less of most of the elements than is found i n the whole block of coal. Jacques Jedwab. W e have studied i n our laboratory the ratios C a / M g / B e . W e have observed a positive correlation between C a / M g a n d B e , indicating a relationship between the enrichment of beryllium i n the coal seams (particularly i n the roof and walls) and the circulation of ground water. G . R . H i l l . C o m p a r i n g the ratio of ionic charge to ionic radius is good for metals w h i c h form simple ions. F o r elements like boron, w h i c h exist always as complex anions, the ratio m a y not be meaningful. This may explain the anomalous position of boron i n Figure 1. D o w e have any information about the nature of the complexes formed b y borates i n plants? M r . Z u b o v i c . I do not have any data o n the nature of complexes formed b y boron i n plants. I do believe that the boron w e find associated w i t h organic matter i n coal is the result of its accumulation b y plants. Bjerrum does not list any boron complexes i n his compilation of stability constants of metal-ion complexes. M a r t e l a n d C a l v i n suggest that boron complexes resemble esters. I do not believe that the high organic affinity a n d high ionic potential shown by boron i n Figure 1 is any indication that it forms stable chelates. Rather its high organic affinity just shows that it is accumulated b y plants and is retained in the organic fraction during coalification. H. J . Gluskoter. W h a t are the differences, if any, i n analytical results between samples ashed at lower a n d higher temperatures?

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.

13.

ZUBOVIC

Minor Element Properties

231

M r . Zubovic. W e ashed several samples at higher temperatures—i.e., about 7 5 0 ° C . A l l of our coals were normally ashed at a maximum of 4 5 0 C . Comparing those samples ashed at 7 5 0 ° C . w i t h the same sample ashed at 4 5 0 C . showed no loss of any of the elements. However, I must point out that both ashings were done at a slow rate of temperature increase. I suspect that if a sample were placed i n a furnace preheated to 7 5 0 ° C . a loss of some of the elements may take place. Bhupendra M a z u m d a r . T h e author s results show that a sample of weathered coal was found to contain appreciably more of metal complexes than the unweathered coal. I w o u l d be glad to know if M r . Zubovic can explain this phenomenon. H a s he studied any correlation between extra metal retention a n d additional acidic groups ( i n particular C O O H groups) usually found i n weathered coals? M r . Zubovic. W e have not made any additional studies of these weathered samples. I do not know if there is any correlation of the elements w i t h a larger number of acidic groups i n this coal. I do believe that the additional amounts of elements found i n the weathered coals are not h e l d there as chelates but rather as simple complexes such as are formed i n i o n exchange reactions. e

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on March 18, 2017 | http://pubs.acs.org Publication Date: January 1, 1966 | doi: 10.1021/ba-1966-0055.ch013

e

Given; Coal Science Advances in Chemistry; American Chemical Society: Washington, DC, 1966.