KGS--Four Clays from the Dakota Formation--Chemical Properties

Prev Page--Occurrence || Next Page--Petrography

Chemical Properties

Chemical Analyses

Chemical analyses of four portions of each clay were made: the original clays, the minus 2-micron fraction, the 2- to 8-micron fraction, and the plus 8-micron fraction. The analyses consisted of quantitative determinations for major elements present. Sulfide sulfur, sulfate sulfur, and phosphorus pentoxide were determined for all the original clays and for some of the fractions. Generally, only traces were noted. The procedures used to determine silica, alumina, iron oxide, titanium dioxide, calcium oxide, magnesium oxide, sodium and potassium oxides, and loss on ignition are accepted procedures for research accuracy. A brief description of these methods is given.

Procedures

Silica--Analysis for silica was begun by fusion with sodium carbonate followed by double dehydration with hydrochloric acid. The crude silica obtained was ignited to 1200° C., then treated with hydrofluoric and sulfuric acids. The residue was ignited, weighed, and the loss of weight calculated as silica.

Alumina--The entire ammonium hydroxide group was double precipitated after removing silica with hydrofluoric and sulfuric acids. The R₂O₃ precipitate was ignited to 1200° C., cooled, and weighed. The other elements were subsequently removed, leaving alumina by difference.

Calcium oxide--Calcium was determined by double precipitation with ammonium oxalate from the R₂O₃ filtrate. The calcium oxalate was titrated with ceric sulfate and calculated to calcium oxide. The combined filtrates were used to determine magnesium.

Magnesium oxide--The combined filtrates from the calcium determination were treated with nitric and hydrochloric acids to destroy excess oxalate and ammonium salts. The magnesium was then double precipitated with diammonium phosphate. The final weighing was as magnesium pyrophosphate.

Loss on ignition--The determination of loss on ignition represents the net change of weight occurring between 140° C. and 1000° C. It generally represents a net loss of weight and includes such elements and compounds as carbon (organic matter), sulfide sulfur, carbon dioxide, and water of hydration.

Iron and titanium--The procedure used for iron and titanium was a gravimetric procedure utilizing cupferron (Runnels, Utter, and Reed, 1950). The R₂O₃ precipitate was put into sulfuric acid solution after fusion with potassium bisulfate. The iron and titanium (along with zirconium, vanadium, copper, and tin if present) were then precipitated with cupferron, leaving alumina (and phosphorus pentoxide and gallium oxide). The combined precipitate was ignited and weighed. Total iron was then determined, reported as Fe₂O₃, and subtracted from the cupferron precipitate, leaving titania (RO₂ plus R₂O₃₅) by difference.

Potassium and sodium oxide--The alkali metals were determined with a model 52-C Perkin-Elmer flame photometer using acetylene and air. Solution was accomplished by treatment with hydrofluoric and sulfuric acids.

Discussion of Data

Table 1 shows the results of the chemical analyses. Comparison of the analyses of the four clays indicates some of the basic differences observed and which are discussed in the other sections of this report. All the minus 2-micron fractions show a marked increase in alumina and a corresponding decrease in silica. The other fractions reflect the free silica. There is a sharp increase in titanium in the 2- to 8-micron fraction and iron decreases, suggesting that most of the titanium is present as a dioxide and not as ilmenite. The other elements such as magnesium and the alkalies reflect the clay mineral composition and vary according to the concentration of the clay. Tables 2, 3, 4, and 5 show the various fractions calculated on a weighted basis. Totals calculated from the various fractions show fair agreement with the analyses of the whole clays. The most deviation occurs in sample O-5-6.

Table 1--Chemical analyses of original clays and the various size fractions. (N.D. is "not determined.")

	EL-60-6				EL-69-2				O-5-6				O-38-4
	Whole clay	<2 microns	2-8 microns	>8 microns	Whole clay	<2 microns	2-8 microns	>8 microns	Whole clay	<2 microns	2-8 microns	>8 microns	Whole clay	<2 microns	2-8 microns	>8 microns
Silica (SiO₂)	73.26	45.50	65.75	96.22	66.80	49.04	73.37	93.38	62.14	51.04	83.50	95.47	64.23	49.21	82.63	94.57
Alumina¹ (Al₂O₃)	17.86	37.09	22.93⁵	1.661⁵	20.94	32.83	15.91⁵	3.27⁵	24.27	31.47	9.26⁵	1.69⁵	22.84	32.55	10.15⁵	2.19⁵
Titania² (TiO₂)	1.25	1.54	1.92	0.97	1.24	0.94	1.81	0.83	1.09	0.81	1.96	1.13	1.29	1.19	1.61	1.23
Total iron (Fe₂O₃)	0.65	1.23	0.57	0.13	1.40	2.17	0.97	0.44	2.16	2.87	0.77	0.48	1.51	2.29	0.57	0.26
Calcium oxide (CaO)	0.14	0.09	0.07	0.04	0.24	0.16	0.13	0.09	0.54	0.52	0.14	0.07	0.28	0.30	0.10	0.08
Magnesium oxide (MgO)	0.15	0.12	0.15	0.10	0.65	0.86	1.27	0.04	0.54	0.65	0.43	0.01	0.63	0.78	0.72	0.03
Potassium oxide (K₂O)	0.06	0.05	0.05	0.05	1.98	2.30	1.85	0.41	0.72	0.75	0.53	0.28	1.59	1.93	1.03	0.20
Sodium oxide (Na₂O)	0.05	0.08	0.04	0.03	0.16	0.19	0.18	0.07	0.15	0.16	0.15	0.06	0.20	0.25	0.13	0.05
Phosphorus pentoxide (P₂O₅)	tr	tr	N.D.	N.D.	tr	tr	N.D.	N.D.	nil	tr	N.D.	N.D.	tr	tr	N.D.	N.D.
Sulfate sulfur (SO₃)	tr	0.14	N.D.	N.D.	tr	0.12	N.D.	N.D.	tr	0.05	N.D.	N.D.	tr	0.09	N.D.	N.D.
Loss on ignition (L.O.I.)	6.48	13.73	8.41	0.60	6.70	10.89	5.19	0.99	8.41	11.09	3.17	0.62	7.33	10.60	3.21	0.83
Sulfide⁴ sulfur (S)	0.02	N.D	N.D.	N.D.	0.12	N.D.	N.D.	N.D.	nil	N.D.	N.D.	N.D.	nil	N.D.	N.D.	N.D.
Total	99.90	99.57	99.89	99.80	100.11	99.50	100.68	99.52	100.02	99.41	99.91	99.81	99.90	99.19	100.15	99.44
¹Contains gallium (Ga₂O₃) and MnO₂ when present									⁴Included in L.O.I. for total.
²Contains ZrO₂ and V₂O₅ when present. Gravimetric method.									⁵Contains P₂O₅ when present.
³Net loss of weight from 140° C. to 1000° C.

Discrepancies between percentages obtained by adding the various fractions and analyses of the whole sample are owing, at least in part, to not sampling the same percentage of the various fractions as determined by particle size analysis.

Table 2--Weighted composition of the sized fractions of clay EL-60-6.

	<2 microns, 35 percent	2-8 microns, 16 percent	>8 microns, 49 percent	Total of fractions	Whole clay
SiO₂	15.9	10.5	47.1	73.5	73.3
Al₂O₃	13.0	3.7	0.8	17.5	17.8
TiO₂	0.5	0.3	0.5	1.3	1.3
Fe₂O₃	0.4	0.1	0.1	0.6	0.6
CaO	0.03	0.02	0.02	0.1	0.1
MgO	0.03	0.03	0.05	0.1	0.2
K₂O	0.03	0.02	0.02	0.1	0.1
Na₂O	0.03	----	0.02	0.05	0.1
L.O.I.	4.8	1.3	0.3	6.4	6.5

Table 3--Weighted composition of the sized fractions of clay EL-69-2.

	<2 microns, 49 percent	2-8 microns, 23 percent	>8 microns, 28 percent	Total of fractions	Whole clay
SiO₂	24.0	16.9	26.2	67.1	66.8
Al₂O₃	16.1	3.6	0.9	20.6	20.9
TiO₂	0.4	0.4	0.2	1.0	1.2
Fe₂O₃	1.1	0.2	0.1	1.4	1.4
CaO	0.1	0.02	0.03	0.2	0.2
MgO	0.4	0.3	0.01	0.7	0.7
K₂O	1.1	0.4	0.1	1.6	2.0
Na₂O	0.1	0.05	0.02	0.2	0.2
L.O.I.	5.3	1.2	0.3	6.8	6.7

Table 4--Weighted composition of the sized fractions of clay O-5-6.

	<2 microns, 62 percent	2-8 microns, 23 percent	>8 microns, 15 percent	Total of fractions	Whole clay
SiO₂	31.6	19.2	14.3	65.1	62.1
Al₂O₃	19.5	2.1	0.3	21.9	24.3
TiO₂	0.5	0.5	0.2	1.2	1.1
Fe₂O₃	1.8	0.2	0.1	2.1	2.2
CaO	0.3	0.02	0.01	0.3	0.5
MgO	0.4	0.1	----	0.5	0.5
K₂O	0.5	0.1	0.04	0.6	0.7
Na₂0	0.1	0.04	0.01	0.2	0.2
L.O.I.	6.9	0.7	0.1	7.7	8.4

Table 5--Weighted composition of the sized fractions of clay O-38-4.

	<2 microns, 60 percent	2-8 microns, 29 percent	>8 microns, 11 percent	Total of fractions	Whole clay
SiO₂	29.5	24.0	10.4	63.9	64.2
Al₂O₃	19.6	3.0	0.2	22.8	23.1
TiO₂	0.7	0.5	0.1	1.3	1.3
Fe₂O₃	1.4	0.2	0.03	1.6	1.5
CaO	0.2	0.03	0.01	0.2	0.3
MgO	0.5	0.2	----	0.7	0.6
K₂O	1.1	0.3	0.02	1.4	1.6
Na₂O	0.2	0.03	----	0.2	0.2
L.O.I.	6.4	0.9	0.1	7.4	7.3

Spectrochemical Analyses

Spectrographic analyses are generally divided into three types: qualitative, semiquantitative, and quantitative. A qualitative analysis makes no attempt to give a concentration value for any constituent, except occasionally to classify the constituents present as major, minor, or trace. The definitions of these terms are not fixed. In semiquantitative analysis the concentrations are determined definitely, but owing to the fact that many of the variables inherent in spectrography are not controlled, the accuracy of the method is generally conceded to be not better than 50 percent of the amount reported. In quantitative analysis the accuracy ranges from about 20 percent in the easier and less closely controlled methods to as high as 4 percent under the most ideal conditions with variables very closely controlled. The method used in this study was the "internal standard" quantitative method, using a general internal standard. An accuracy of 7 to 10 percent should be expected from this method.

Sample Preparation

The original clays and the minus 2-micron fractions were analyzed spectrographically. The chemical analyses show that the minus 2-micron fractions differ widely in composition from the original clays. As appreciable compositional variation has a decided effect on the spectrographic sensitivities of the various elements, the samples were divided into two groups, by particle size. The elements chosen for internal standards were iron and germanium, the iron to be used for the high-melting, high-boiling elements Mn, V, Cr, Mo, Ni, and Ga; for the low-melting, low-boiling elements Pb, Zn, Sn, and Cu, germanium was used. Iron was already present in all the samples. The amount of Fe₂O₃ necessary to bring the concentration of iron in each original clay to the concentration in the sample highest in iron (clay O-5-6 which contained 2.12 percent Fe₂O₃) was dried, weighed, and added to each of the other original clays. The Fe₂O₃ necessary to bring the iron in the minus 2-micron fractions to a concentration identical with that in the minus 2-micron fraction of clay O-5-6 (2.63 percent) was added to these fractions. To 1 gram of each of the samples 1 percent germanium as the oxide GeO₂ and 1 gram of pure anhydrous Li₂SO₄ (to act as spectrographic buffer) were added. The 2 grams of each powder were then intimately ground and mixed in a mullite mortar to a particle size of about minus 150 mesh.

Standard Preparation

Separate matrices were made up for the two groups of samples. The first, as standard for the original clays, made up to the average major composition of the original clays, had a composition as follows.

SiO₂	2.6640 grams
Al₂O₃	0.8636
Fe₂O₃	0.0848
CaO	0.0400
GeO₂	0.0526
Total	3.7050

This mixture was equivalent to 4.0000 grams of the average original clay, the remainder being the loss the average original clay sustained on ignition to 1000° C.

The second clay standard, for the minus 2-micron fractions, had a composition as follows.

SiO₂	1.9480 grams
Al₂O₃	1.3316
Fe₂O₃	0.1052
CaO	0.0400
GeO₂	0.0576
Total	3.4824

This mixture was equivalent to 4.0000 grams of the average minus 2-micron fraction, the remainder being loss on ignition.

To each of these two matrices 4.0000 grams pure anhydrous Li₂SO₄ was added and the mixtures ground together. Because the four original clays varied rather widely from the average composition, as did the minus 2-micron fractions, from their average, it was decided, in the interest of greater accuracy, to reduce compositional variations by reacting the oxide mixtures. Accordingly, the two raw matrices were heated at 900° C., and the resulting sinter crushed and reground to about minus 150 mesh. To 1.9275 grams of the total clay matrix and 1.8706 grams of the minus 2-micron fraction (1.0000 gram equivalent for each) 0.0100 gram (1.00 percent) each of V, Cr, Mn, Ni, Pb, Sn, Mo, and B, and 0.0010 gram (0.10 percent) Cu was added. Portions of the original matrices were used to make successive dilutions until standards of 0.1 percent, 0.05 percent, and 0.010 percent of each of the elements V, Cr, Mn, Ni, Pb, Sn, Mo, and B and 0.01 percent, 0.005 percent, and 0.001 percent Cu were available for each of the two groups.

Spectrographic Equipment

The instrument used was a 1.5-meter Applied Research Laboratories grating-type spectrograph powered by a 250 volt D.C. arc source. The film on which the spectra were recorded was Eastman Spectrum Analysis no. 1, which after exposure was developed for 3 minutes in Eastman D-19 developer, short-stopped for 10 seconds in 3 percent acetic acid solution, and fixed in Kodak rapid liquid fixer with hardener. After fixing, the film was given a 1-minute tap-water wash, and a 30-second distilled water wash, after which it was rinsed with fresh distilled water, the excess water removed with a sponge, and the film dried on an infra-red forced-air dryer. When dry, the film was placed in an A.R.L. densitometer-comparitor and the densities of the element lines read. The line pairs used for this study were as follows.

Mn: 2576.1 A	Fe: 2584.6 A
V: 3185 A	Fe: 3205 A
Cr: 4289.7 A	Fe: 4294.3 A
Mo: 3170.3 A	Fe: 3205 A
Ni: 3414.8 A	Fe: 3413.1 A
Ga: 2943 A	Fe: 2941.6 A
Cu: 3274 A	Ge: 3269.5 A
Pb: 2833 A	Ge: 3269.5 A
Sn: 3262.3 A	Ge: 3269.5 A

Procedure

Fourteen milligrams of clay-buffer mix was packed into a standard concave, center-post platform electrode formed from quarter-inch National Carbon Company standard electrode-grade graphite rod. The counterelectrode was a similarly formed graphite rod. Three samples of each clay were arced consecutively, without changing the position of the film, thus superimposing the 3 ignitions on the film as 1 spectrum. For this study, 8 spectra for each sample were made, making a total of 24 ignitions on each clay. Each ignition was made at 8 amperes for 60 seconds, with the rotating sector at 10 percent, the grating doors set at 3.4, or about 50 percent transmission, a slit-width of 50 microns, and an electrode-counterelectrode distance of 4 mm. The same procedure was followed with each of the standards in which concentrations of the various elements were known. A curve showing the ratio of intensity of the various element lines to the internal standard line of the pair was drawn, and the intensity ratios of the line pairs in the clay samples were read on these curves. The averages of the eight determinations on each element of each clay, along with the mean deviation of the eight determinations and the percent mean deviation are shown in Table 6. Zinc, zirconium, germanium, strontium, barium, silver, cobalt, cadmium, and the lanthanides were not present. Boron could not be determined because of interference of silicon.

Table 6--Spectrochemical analyses of clays. Amounts shown are the averages of 8 groups of 3 separate samples each, or a total of 24 samples per determination per clay or clay fraction. M.D. (mean deviation) is the average deviation from the mean value shown. P (precision value) obtained from the ratio of the mean deviation to the mean value obtained, expressed as percent. N.D. is "Not detected." Trace is defined in this study as less than 0.001 percent but present in detectable amounts.

Element		Whole clay	<2 microns	Whole clay	<2 microns	Whole clay	<2 microns	Whole clay	<2 microns
		O-5-6		O-38-4		EL-60-6		EL-69-2
Mn		0.0069	0.0065	0.0070	0.0069	0.0073	0.0067	0.0072	0.0078
	M.D.	0.0006	0.0004	0.0006	0.0008	0.0006	0.0006	0.0004	0.0006
	P	8.5	6.0	8.5	11.5	8.0	9.0	5.5	7.5
V		0.0125	0.0178	0.0131	0.0183	0.0076	0.0119	0.0177	0.0231
	M.D.	0.0008	0.0030	0.0013	0.0025	0.0008	0.0016	0.0031	0.0030
	P	6.5	17.0	10.0	13.5	10.5	13.5	17.5	13.0
Cr		0.0102	0.0121	0.0105	0.0124	0.0071	0.0083	0.0105	0.0177
	M.D.	0.0004	0.0004	0.0005	0.0005	0.0002	0.0001	0.0004	0.0016
	P	4.0	3.5	5.0	4.0	3.0	1.0	4.0	9.0
Mo		0.0033	trace	ND	0.0047	0.0029	0.0046	0.0027	0.0050
	M.D.	0.0001			0.0003	0.0002	0.0003	0.0002	0.0005
	P	3.0			6.5	7.0	6.5	7.5	10.0
Ni		0.0050	0.0054	0.0038	0.0042	0.0029	0.0034	0.0033	0.0055
	M.D.	0.0003	0.0006	0.0002	0.0002	0.0004	0.0006	0.0004	0.0005
	P	6.0	11.0	5.5	5.0	14.0	17.5	12.0	9.0
Ga		0.0049	0.0038	0.0038	0.0033	0.0036	0.0037	0.0034	0.0031
	M.D.	0.0006	0.0007	0.0003	0.0005	0.0005	0.0005	0.0005	0.0003
	p	12.0	18.5	16.0	15.0	14.0	13.5	14.5	9.5
Cu		0.00135	0.0058	0.0018	0.0053	0.00037	0.0039	0.00103	0.0085
	M.D.	0.00028	0.0009	0.0010	0.0013	0.00012	0.0007	0.00032	0.0022
	P	20.5	15.5	55.5	24.5	32.5	18.0	31.0	26.0
Pb		trace	trace	trace	trace	trace	trace	trace	ND
Sn		trace	trace	trace	trace	0.002	trace	trace	trace

Discussion of Results

In a quantitative spectrochemical analysis by the internal standard method, compensation for many variables is automatic. One, however, for which it is impossible to compensate, is the amount of basic difference which exists between the physico-chemical properties of the "unknown" element and the internal standard element and the effect of variation of the matrix upon this difference. In a study which involves only one element in a number of closely similar samples, a careful and exacting selection of the internal standard element may be made. When several elements are being determined for a group of rather widely differing samples, compromise conditions must be accepted. If iron is chosen as the compromise internal standard, it cannot be the ideal choice for an element such as gallium. The behavior of the two elements in the electric arc will be dissimilar, and these differences may be intensified by the effect exerted by the major constituents on the behavior of the more minor constituents. A high-silica clay, for example, may cause the iron lines to be darker, relative to the gallium, than a high-alumina clay. Therefore, a compromise internal standard introduces inaccuracies which may be exaggerated by the other differences between samples. This was probably the case in the determination of Cu (with an average precision of about 28 percent, scarcely better than a semiquantitative analysis), Ga (with an average precision of about 14 percent), and to a lesser extent, V (with an average precision of about 12 percent). Determinations of Ni, Cr, Mo, and Mn seem to have greater precision in this general-type analysis (with an overall precision average of 7.4 percent) using iron as the internal standard.

In order to show more clearly the enrichment or depletion of the minor elements in the minus 2-micron fraction, calculations were made to show the amount of each element present based upon the percentage of minus 2-micron fractions present in the original clays. The percentage increase or decrease of the various elements was then computed. These figures are shown in Table 7.

Table 7--Comparisons of the proportions of the minor elements present in the minus 2-micron fractions to the amount (percent by weight) of minus 2-micron fraction in each of the four clays.

Elements present	Element in whole clay, percent¹	<2-micron fraction
Elements present	Element in whole clay, percent¹	Element present, percent¹	Percent of whole clay represented²	Weighted amount of element percent³	Relative percentage of element present⁴	Enrichment of element in percent⁵
EL-60-6
Mn	0.0073	0.0067	35	0.0023	31	-11
V	0.0076	0.0119	35	0.0042	55	+57
Cr	0.0071	0.0083	35	0.0029	41	+ 17
Mo	0.0029	0,0046	35	0.0016	55	+ 57
Ni	0.0029	0.0034	35	0.0012	41	+17
Ga	0.0036	0.0037	35	0.0013	36	+3
Cu	0.00037	0.0039	35	0.0014	378	+980
EL-69-2
Mn	0.0072	0.0078	49	0.0038	53	+8
V	0.0177	0.0231	49	0.0113	63	+29
Cr	0.0105	0.0177	49	0.0087	83	+69
Mo	0.0027	0.0050	49	0.0024	90	+84
Ni	0.0033	0.0055	49	0.0027	82	+67
Ga	0.0034	0.0031	49	0.0015	44	-10
Cu	0.00103	0.0085	49	0.0032	311	+739
O-5-6
Mn	0.0069	0.0065	62	0.0040	58	-6
V	0.0125	0.0178	62	0.0110	89	+44
Cr	0.0102	0.0121	62	0.0075	73	+18
Mo	0.0033	0.001	62	0.0006	18	-71
Ni	0.0050	0.0054	62	0.0033	67	+8
Ga	0.0049	0.0038	62	0.0024	49	-21
Cu	0.00135	0.0058	62	0.0036	267	-352
O-38-4
Mn	0.0070	0.0069	60	0.0041	59	-2
V	0.0131	0.0183	60	0.0110	84	+40
Cr	0.0105	0.0124	60	0.0074	70	+17
Mo	ND	0.0047	60	0.0028
Ni	0.0038	0.0042	60	0.0025	66	+10
Ga	0.0038	0.0033	60	0.0020	53	-12
Cu	0.0018	0.0053	60	0.0032	178	+197
¹From spectrographic analysis. ²From particle size data. ³Calculated: percent element present times percent whole clay represented. ⁴Calculated: weighted percent in minus 2-micron fraction divided by percent element in whole clay. ⁵Calculated: percent of whole clay represented minus relative percentage present divided by percent of whole clay represented.

Considering these results, a few general statements can be made. Vanadium, chromium, and nickel (to a lesser degree) are enriched in the minus 2-micron fractions. Manganese and gallium show depletion. This depletion is directly opposed to normal theory. Gallium, especially, is supposedly associated with aluminum and was expected to be enriched in the minus 2-micron fractions. The copper determination is known to be erroneous; the explanation is that in separating the minus 2-micron fractions (particle size analysis) a brass agitator was used in conjunction with a dilute ammonium hydroxide suspension of the clay.

Although the molybdenum concentration seemed to be similar to the other elements the calculations show it to range from minus 71 percent to plus 84 percent. This variation was too erratic to indicate any definite trend.

Prev Page--Occurrence || Next Page--Petrography

Kansas Geological Survey, Geology
Placed on web June 22, 2007; originally published Dec. 1954.
Comments to webadmin@kgs.ku.edu
The URL for this page is http://www.kgs.ku.edu/Publications/Bulletins/109_10/04_chem.html