.0'"-m--.'—“- — — “"00. 0

 

 

COURSE AND EXTENT OF ALTERATION OF SELECTED
FERROMAGNESIAN SILICATES BY AQUEOUS
SOLUTIONS OF OXALIC ACID

Thesis for the Degree of M... S.
MICHIGAN STATE UNIVERSITY
ALAN BAILEY
1967

    
 
  

LII: "'{A RY
NiiL‘I: Ian State

  
 
  

I h 2:313

 

ABSTRACT

COURSE AND EXTENT OF ALTERATION OF SELECTED
FERROMAGNESIAN SILICATES BY AQUEOUS
SOLUTIONS OF OXALIC ACID

by Alan Bailey

The extent and course of alteration of three ferro—
magnesian silicates by a simple organic acid were inves-
tigated. Sections of an olivine, a pyroxene and an
amphibole were placed in plastic beakers containing 200 ml
of a 90gm/lOOO ml solution aqueous solution of oxalic acid.
Two grams of finely ground pyroxene were also placed in a
beaker with 200 ml of the oxalic acid solution. After
six months the solutions were analyzed for calcium,
magnesium and total iron by means of atomic absorption,
and for silica colorimetrically. Both colloidal silica
and silica in true solution were determined. Portions of
the unaltered minerals were dissolved and analyzed for
calcium, magnesium, total iron and silica using the same
techniques as above. In addition, x-ray patterns were
taken of the unaltered minerals and of the residue present
after alteration by oxalic acid.

All minerals altered to some extent with the olivine
and amphibole altering appreciably. 10,577 ppm total
silica were determined in the solution over the olivine,

2&6 ppm in the solution over the ground pyroxene, and

Alan Bailey

163 ppm in the solution over the amphibole. A comparison
of the silica with the iron data indicated that the
olivine was being attacked by removal of silica while the
pyroxene was being attacked more by preferential removal
of iron from between the polymerized chains of (810”)
tetrahedra. In the case of the amphibole the prefer-
ential removal of iron was even more pronounced. Exam-
ination of the x-ray patterns indicated calcium and mag-
nesium oxalates to be present in the residues, and so
perferential removal was not directly demonstrated by this
data. Precipation, however, would tend to maintain a
concentration gradient between the mineral and the liquid
and thus promote removal of the ions in the precipitate
from the mineral.

Results of this study combined with other previous
studies indicate that simple organic acids which are
present in natural systems can alter ferromagnesian sili-
cates. Also, the type of alteration outlined above is
indicated in nature by the lack of residue of weathered
olivine and the morphology of weathered pyroxenes and
amphiboles. In addition, the effect of the organic radical
on the solubilities of the various cations may possiblyfbe
seen in the development of soil profiles, the movement of
iron in natural solutions, and in the occurrence of natural
organic salts as minerals. Finally, the probable importance

of organisms as weathering agents is indicated by this and

other studies.

COURSE AND EXTENT OF ALTERATION OF SELECTED
FERROMAGNESIAN SILICATES BY AQUEOUS

SOLUTIONS OF OXALIC ACID

By

Alan Bailey

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Geology

1967

ACKNOWLEDGMENTS

The writer wishes to thank Dr. Ehrlich for sug-
gestions and aid rendered throughout the course of the
work and for the patience and enthusiasm demonstrated
during the time this work was being carried out.

Appreciation is extended to Dr. S. B. Romberger
and Dr. M. M. Mortland for aid and constructive criti-
cism offered during the work and while reading the
manuscript.

Thanks is also extended to Dr. D. E. Scherpereel of
the Metallurgy Department and Dr. B. G. Ellis of the Soil

Science Department for the use of equipment.

11

TABLE OF CONTENTS

. ACKNOWLEDGMENTS . . . .'. . . . . . .

LIST OF TABLES . . . . . . . . . . .
LIST OF FIGURES . . .

LIST OF PLATES . . . . . . . . . .

Chapter
I. INTRODUCTION . . . . . . . .
II. PREVIOUS WORK . . . . . . . .
III. DESIGN OF EXPERIMENT . . . .
General Design . . . -.
Properties of Oxalic Acid .
Properties of Minerals Used

IV. COLLECTION OF DATA . . .

General Methods of Analysis
Analysis for Silica in True

in the Study .

Solution . . .

Analysis for Colloidal Silica Plus

Silica in Solution . . .

Analysis for Calcium, Magnesium and

Total Iron . . . . .
X- Ray Diffraction Work . .

O O O O O O

V. DISCUSSION OF RESULTS AND CONCLUSIONS . . . .

Extent of Alteration . .

Processes . . . .,. . . . .
Fe/Si Ratios . . . . . . .
Mg/Si Ratios . . . . . . .
Ca/Si Ratios . . . .

Summary of Processes Indicated by the Study
Application to Natural Weathering . . . .

REFERENCES CITED . . . . . . . . . .

APPENDIX 0 O O 0 O O O O O O O O O O

Page
ii

iv

Table

13.
1“.

LIST OF TABLES

Logarithms of stability constants of chelates
formed between oxalic acid and magnesium,
calcium, ferrous iron and ferric iron . .

Percentages of SiO , CaO, total iron as FeO

and MgO in the unaltered minerals . . . .
Silica. in true SOlUtIOI'l o o o o o o o o o I. c
Total Silica O O O O 0 O O O O O O O 0 O O O

COllOidal 811108. a o o o "o o a o lo 0 o a o 0

Ca, Mg and Fe for solutions separated in
January-~no lanthanum added . . . . . . .

Ca, Mg and Fe for solutions separated in
January-~lanthanum added . . . .>. . . . .

Ca, Mg and Fe for solutions separated in
MaPCh--lanthanum added 0 o o ' o o 7 o o o o 0

New peaks in the olivine residue pattern . .
New peaks in the amphibole residue pattern .
New peaks in the pyroxene residue pattern . .
Ratios of concentrations in solutions . . . .
Ratios of concentrations in minerals . . . .

Precipitates formed in the mineral-
solution systems . . . . . . . . . . . . .

iv

Page

ll

17
25
25
25

29

29

29
32
33
33
38
38

38

Figure

1.

LIST OF FIGURES

Oxalic, malonic and succinic acids
Structure of forsterite (after Deer, et al.)
Structure of diopside (after Fyfe)

Structure of actinolite (after Fyfe)

Curves showing disaggregation of silica sol

on standing (after Krauskopf)

Page
l8
l8
l9
19

26

LIST OF PLATES

Plate Page

I. Photomicrographs of thin sections of
material used in the study . . . . . . . . . . 16

vi

CHAPTER I

INTRODUCTION

The following study concerns the alteration of
ferromagnesian silicates at surface temperatures and
pressures by organic agents in aqueous solutions. The
results of the study should provide insight into the role
of these agents in geochemical processes taking place at
the surface of the earth. Such processes must be taken
into account when considering geochemical processes close
to the surface of the earth because water charged with
atmospheric gases percolates through what is usually an
organic-rich zone existing in the upper portion of the
soil. During this process organic matter is dissolved or
transported in colloidal form. In addition the products
secreted by plants and micro-organisms are added to the
solution. The solutions then come into contact with the
inorganic portion of the soil and react in various ways,
such as complexing of chemical constituents and acid
attack. Because of the ubiquitous nature of organic
matter throughout much of geologic time and because of
the above considerations it can be seen that there is a
need to evaluate the role of dissolved organic material as

weathering agents.

As can be seen by examining the above processes in
detail, the reactions taking place are probably very com-
plex. One way of studying such natural processes, however,
is to examine the controlling factors by means of relatively
simple laboratory experiments. This has been done exten-
sively in the case of high temperature and pressure altera-
tion of minerals and, to a lesser degree, for the alteration
of minerals at surface temperatures and pressures in inorganic
situations. Analogous work carried out with organic materials
is lacking, and it is necessary to assess the role of these
materials.

For this study an experiment was set up to evaluate the
extent and course of alteration of three common ferromagne-
sian silicates under relatively simple conditions. An
olivine, a pyroxene, and an amphibole were placed in aqueous
solutions of oxalic acid in plastic beakers. Thin films of
plastic were then taped over the tops of the beakers to keep
out the dust and the beakers allowed to stand for several
months. A more detailed description of the materials
and the reasons for using them are given in a section on
the design of the experiment. At the end of several
months the solutions were analyzed for dissolved iron,
magnesium, and calcium using an atomic absorption spectro-
meter, and for silica using a colorimeter; the solids were
analyzed primarily by x-ray diffraction. The methods of

analysis and the results are given in a section on the

collection of data. The results were then evaluated to
determine the degree of alteration and what processes
were involved in the alteration. The application of
these results to the role of such alteration in natural

weathering was then considered.

CHAPTER II

PREVIOUS WORK

As indicated in the Introduction, relatively little
work has been done on the organic weathering of rock-
forming minerals. Some data was found, however, scattered
through journals on soil science, microbiology and geology.
Summaries of the studies more closely related to the
geologic aspects of the problem and to the present type of
are included here. One of the earlier studies was one by
Gruner (1922). He suggests that unusually high concentra-
tions of iron in some natural waters may be due to the
presence of organic material in the waters. To support
this he notes a correlation of high concentrations of
iron with high amounts of organic material in natural
waters. He also cites the results of laboratory exper-
iments using peat and peat solutions. In one experiment
fourteen minerals, including oxides, sulfides, carbonates
and silicates, were allowed to come in contact with peat
and peat solutions. After 90 days 26-161 ppm silica and
9-27 ppm iron were determined in the solutions over the
silicates. After 165 days the silica concentrations were
33—106 ppm and the iron concentrations 8-31 ppm. Calcium

and magnesium were also determined as 8-31 ppm and 2—2A ppm

respectively (some samples showed only a trace of mag-
nesium). The relatively low calcium and magnesium con-
centrations were explained by Gruner as being due to
absorption by colloids of the peat, or precipitation as
compounds. The acidity of the solution was comparable to
that of carbonic acid. Gruner regards the organic material
in soil solutions as being colloidal. This work tends to
support statements made by Konova (1961) and Bunting

(1965) that water soluble organic material plays a definite
part in the decomposition of silicates.

Investigations more closely connected to the present
study have been carried out by Gallagher and Walsh (19A3).
They set up experiments in which oxalic acid was used as
an extractive agent for soils. The experiments were also
used to examine the possibility of podsolization by rela-
tively simple organic acids which the authors state should
be liberated from decomposing plant material. They state
that, despite the relatively weak acidity of oxalic acid
compared with that of mineral acids, it has been shown to
have greater extractive power. In one experiment where
20 grams of soil was treated with 200 cc of cold lN oxalic
acid, up to .55% of the silica and 7.33% of the iron from
B horizons were taken into solution. Comparable results
were obtained for alumina. In another experiment boiling
the samples was shown to increase the amount of silica

taken into solution (up to 1.13% from one C horizon). The

authors quote Tamm as having extracted 13% of the alumina
from pulverized feldspar using a solution prepared by
dissolving .2 gram—molecular wt. of acid ammonium oxalate
and .075 gram-molecular wt. of neutral ammonium oxalate
in 1 liter of water.

More recent work by Evans (196A) using nucleic
acids and nucleotides indicates the potential of organic
material and organisms in dissolving minerals. In one
experiment Evans used a l/5 molar solution of Na-adenosine
triphosphate to dissolve a number of minerals (including
augite) at pH's of 7.5 to 8.5. In another experiment a
quartzite was produced from pure quartz sand, again using
a 1/5 molar solution of Na-adenosine triphosphate.

More recent investigations by Schalscha, Appelt and
Schatz (1967) using naturally occurring chelates indicate
the effects on the release of Fe, Al and K from minerals
including the silicates augite, epidote and biotite. One
of the important results of this study was that the authors
found no correlation between the dissolving action of the
solution and the pH. It was concluded by the authors that
chelation was important, although the authors state that
both acid attack and chelation may be involved.

The above list of articles does not include those
which contain experiments involving microorganisms.
Articles by Thiel (1927), Boyle and Voigt (1967) and

Wagner (1967) stress the role of microorganisms, and these

may be important in the present study because of the fact
that many microorganisms secrete organic acids such as
oxalic acid. Comments by Keller (1957) indicate the
possible role of chelates and material secreted by plant

rootlets.

CHAPTER III

DESIGN OF THE EXPERIMENT

General Design

 

Sections of a single crystal of pyroxene, a single
crystal of amphibole and a polycrystalline sample of
olivine were placed in plastic beakers containing 200 ml
of a 90 gm/lOOO m1 aqueous solution of oxalic acid.
Portions of the minerals not placed in oxalic acid were
saved for x-ray diffraction and chemical analyses. The
results of work on the unaltered minerals are given
later in this section to indicate more specifically what
minerals were used in the study. Also, photomicrographs
of the unaltered minerals are given in Plate I (page 16)
to indicate the homogeneity of the original material.
Because reconnaisance studies had shown the pyroxene to
be relatively unreactive, 2 grams of the finely ground
mineral were placed in a fourth beaker containing 200 ml
of the oxalic acid solution. The beakers were then
covered with a thin plastic film to keep out the dust and
allowed to stand at room temperature from July, 1966

until December, 1966.

Properties of Oxalic Acid

Oxalic acid was chosen for this study because it is
simple in structure; it has chemical prOperties which may
be found in naturally occurring low—molecular weight
organic acids; and it is found in many environments where
weathering occurs.

Roberts and Caserio (1965) indicate that oxalic
acid has the most simple structure in a series of organic
compounds known as dicarboxylic acids. This series is
characterized by two carboxyl groups which are usually
bonded to another organic group. Oxalic acid, however,
consists of two carboxyl groups bonded directly to each
other. Another acid which has been identified in soils
is succini acid (Kononova, 1961). The first three acids
in this series are shown in Figure l.

The structure of oxalic acid gives it two properties
which are felt to be particularly relevant to this study.
One is that the close proximity of the carboxyl groups
causes it to be relatively acid with respect to the
other members oftfimeseries. ‘As the carboxyl groups become
separated in the larger members of the series, the effect
of one carboxyl group on the other goes down and the
acidity decreases. Another is that oxalic acid is a
chelate. According to Martell and Calvin (1952) chelates
are special types of complexes in which the groups combining

with the metal contain at least two electron-donating groups

10

so that in the chelation process rings are formed. One
of the unusual properties of chelates is that the type
of complex they form may be very stable. Stability
constants provide a measure of the degree to which the
metals are tied up and may vary with such factors as
pH, the particular metal, and the particular chelating
agent. A few of the stability constants related to this
study are given in Table 1. According to Martell and
Calvin (1952) chelation can also change the solubilities
of metals markedly in the process. The solubilities may
be raised or lowered depending on tflle properties of the
chelate and the solution. If the chelate contains hydro-
philic groups the solubility in water may increase. A
nonpolar chelate may result in a decrease in solubility.
In addition to having these chemical features
oxalic acid is ubiquitous in nature. It is produced in
the metabolic processes and is secreted by the roots of
pines. In an article by Boyle and Voigt (1967) it is
suggested that the decomposition of biotite may sometimes
be brought about by this secretion and field evidence is
cited along with laboratory results. Oxalic acid is also
secreted by bacteria and fungi and produced by the action
of molds on cellulose. According to Kaurichev, 22 a1.
(1962, 1963) oxalic acid was found to be the most abundant
low-molecular weight, nonvolatile organic acid in several

samples of forest litter. It can be produced by the

11

TABLE l.--Logarithms of stability constants of chelates
formed between oxalic acid and magnesium, calcium,ferrous
iron and ferric iron.*

 

 

Metal Ion Log K
MgII 3.u3 (K)
2.55 (K)
2.65 (K)
A.38 (klk2)
Ca 3.00 (K)
Fell A.7 (K)
u.5 (klk2)
5.2 (klk2k3)
FeIII 10 (K)
18 (klk2k3)

 

*From Martell and Calvin (1952, pp. 516-517). K is
for the 1:1 chelate formed from the chelating agent in
which all the ionizable protons have been removed. The
different values are those obtained by different experi-
mental methods. (klk2) and (klk2k3) are the logarithms

of the stability constants for steps in the formation of
a complex in whch two and three chelating molecules may~
combine with the metal ion. The equations on which the

constants are based are

 

K = (MKe)
(MIIKeI ’
k k = (MA) (MAz)
1 2 TM)(A) (MA)(A)
(MA ) (MA )
and k k k = (MA) 2 3

 

l 2 3 <M)(A7 (MA)(A5 (MA25TAT °

The constants were all obtained at temperatures around
room temperature.

l2

oxidation of many general substances in soil such as peat
and humic acids. It is slightly unstable in the presence
of light but small amounts have been separated from river
water by Mueller, Larson and Fenetti (1960).

While oxalic acid is ubiquitous in nature, and also
reactive, it is certainly not the only organic acid of
this type available in the zone of weathering. For this
reason the results of this study should be taken more as
an indication of the role of low-molecular weight organic
acids than as an indication that oxalic acid alone is a

primary factor.

Properties of the Minerals Used in the Study

The olivine, amphibole, and pyroxene were chosen for
this study because they represent common minerals; they
appear to alter easily in nature (Goldich, 1938; and
Pettijohn, 19A1); and little of this type of work has
been done with these particular minerals.

In general,the olivine, amphibole, and pyroxene groups
belong to a large class of minerals known as the ferro-
magnesian silicates and are distinguished by the arrange-
ment of their (810“) tetrahedra. The olivine group has a
structure made up of independent (SiOu) tetrahedra linked
together by cations and is described by Deer, gt a1.
(1962, vol. 1). The composition usually lies between the
(and members forsterite (MgSiOu) and fayalite (FeSiOu). A

(liagram of the orthorhombic unit cell is shown in Figure 2.

13

The results of chemical analysis (given at the end of this
section) indicate that the mineral used in this study was
a high-magnesian olivine near the forsterite end of the
forsterite-fayalite series.

The pyroxene group is characterized by a structure
made up of single chains of (SiOu) tetrahedra linked hori-
zontally by cations and is described by Deer, gt gt.
(1962), vol. 2). The chemistry of the pyroxenes is
complex but chemical analysis and x-ray diffraction work
indicate that the particular pyroxene used in this study
was diopside (CaMgSiZO6). The structure of diopside
has been determined and is shown in Figure 3.

In the amphibole group the degree of polymerization
of the (810“) tetrahedra is carried still further with a
basic structure consisting of double chains of (810“) tetra-
hedra linked together by cations of various charges.
According to Whittaker (1959) the general structure contains
talc-like strips if one views it in the (100) projection.
While minerals of this group are even more complex than
those ofthe pyroxene group, chemical analysis and general
appearance indicates that the amphibole used in this study
was a hornblende with a low percentage of silica and a
high percentage of iron. Actinolite has been worked out
structurally and is shown in Figure A, even though it is
not the mineral used in this study, to illustrate the

ggeneral features of the amphiboles. The lack of precise

14

structural data on the amphiboles made the diffraction
patterns more difficult to interpret in this study.

As indicated earlier analyses were carried out on
the unaltered minerals. Percentages of Si02, CaO, MgO
and total iron as FeO were determined in order to better
identify the unaltered minerals and to facilitate the rim
evaluation of the results obtained after alteration by
oxalic acid. The methods of analysis and percentages
obtained are included here. For the determination of 8102 A

the minerals were taken into solution by heating 0.0500

 

grams of the finely ground sample in a nickel crucible
containing NaOH. Water was then added to the crucible
and, after allowing it to stand overnight, the contents
emptied into a beaker containing A00 ml of water which
included a few ml of HCl. The volume was then brought to
1000 ml in a 1000 m1 volumetric flask using deionized
water and the silica determined colorimetrically, using
the silicomolybdate blue method. The general scheme for
determination of 8102 is that outlined by Shapiro and
Brannock (1956). A Bausch and Lomb colorimeter with a
13 mm cell was used to measure the percentage of light
transmitted at 700 mu. The percentages of 8102 obtained
are given in Table 2.

For the determination of CaO, MgO and total iron the
minerals were taken into solution by adding 0.1000 gm of

the finely ground mineral to a 25 ml platinum crucible and

15

adding 2.5 ml concentrated HF and 0.75 ml of concentrated
H280“. The sample was then digested overnight on a steam-

bath, 0.25 ml of concentrated HNO was added, and the

3
resulting solution brought 200 m1. This is essen-

tially the method used by Shapiro and Brannock, but with

5 ml of a 5% solution of lanthanum oxide included in each T‘“
25 ml of sample to prevent silica interference. All I

quantities were reduced to one fourth the size in order

to use 25 m1 platinum crucibles instead of the 100 ml .

 

size called for. The reduced sample size probably resulted iva
in reduced precision, causing the percentages for the
olivine and pyroxene to be slightly high. The samples were
then run on a Perkin Elmer Model 303 atomic absorption
unit. The data, obtained in ppm, were converted to per-
centages of CaO, MgO and total iron (as FeO) and are given
in Table 2. The total percentage of less than 100% in the
case of the amphibole is due to the fact that not all of

the major constituents were determined.

l6

 

Plate I Photomicrographs of Thin Soctiono of Meterial
Used in the Study

   

 

agnotite
Olivine - plane light showing
trace of magnetite
“Ax Magnotito
.__.Quartz

'1

I? Aux:

Amphibolo -- plane light showing

magnetite and quartz
inclusions

Scale
(for all
photographs)

ha
0 01 02 m

 

Pyroxono -- plane light showing
homogeneity

 

 

 

TABLE 2.--Percentages of SiO

17

2’

CaO, total iron

as FeO and

 

 

 

MgO in unaltered minerals. WEI
Olivine Amphibole Pyroxene .
8102 A2.A % 35.2 53.6 7
CaO 3.50 13.3 27.3 r-~
FeO 9.65 25.7 4.50
MgO 51.5 6.89 18.3
107.05 81.09 103.70

 

~18

 

 

 

203:0 O

N) ux 3 533.302 0
onoNux no 5232002 .0

 

 

 

. 203.300 5:3
32033..“ *0 9.33:5 moi

o_c_.oo:w

I
_

U-OI
U—OI

 

_
—

—_

—

o

I—O
I— 0—1
0

320.92

a
o

_
I

0:0 —OI
o:U-O I

oZoxO

CIT—O 1‘.
020—01

mEo< 25005 uco

2:322 6:85 for.

L

 

 

l9

 

 

 

 

0

E3200.
EEmmcoozo

I!
V.

I :10

9

CE :53
o:_oc:o< yo 33035vi

EEO—00 .
£33332 0

 

$31,533
mquo 5 do 9.32:5 M .9m

 

 

CHAPTER IV

COLLECTION OF DATA

General Methods of Analysis

 

As mentioned in the section on the design of the
experiment the beakers containing the minerals and oxalic
acid were allowed to stand from July, 1966 until December,
1966. Two weeks after the experiment was set up, how-
ever, visible alteration had taken place with noticeable
bleaching of the olivine, slight disintegration of the
amphibole, and with a light green color appearing in the
solution over the amphibole. No visible alteration
of the pyroxene (either ground or sectioned) was noticed.
In December, after a period of about six months, the
effects mentioned above were still more pronounced and
analytical work was begun on the solutions. The solu-
tions were first analyzed for calcium, magnesium and
total iron in January, 1967. In February and March work
was done on the determination of silica, both that in
true solution and that in colloidal form. In March the
solutions were again anayzed for calcium, magnesium and
total iron. Diffraction patterns of the residue in the
olivine and amphibole experiments were taken in June,

1967. No patterns were taken of the sectioned pyroxene

2O

21

because little alteration was apparent, but a pattern of
the ground pyroxene residue taken in the fall of 1966

is included. The methods used in collecting the data and
the results are described in detail at this point beginning

with the determination of the silica.

Analysis for Silica in True Solution

For the silica analyses 10 ml portions were taken
from each of the four mineral solutions using a 10 m1
pipette. Care was taken not to disturb the fine fraction
of solids on the bottom of some of the beakers. The
solutions were then filtered through Whatman #2 filter
paper and placed in plastic bottles.

Portions of these solutions were then analyzed
colorimetrically for silica in true solution and after
NaHCO3 treatment to destroy the colloidal silica, for
total silica. The difference between the values deter-
mined before and after treatment with NaHCO3 was taken
to determine the amount of colloidal silica present.
Colorimetric determination of silica by reduction of the
silicomolybdate complex to molybdate blue is now being
used extensively and accuracy close to that of gravimetric
determinations is claimed when the determination is
carried out under optimum conditions. The acid ammonium

molybdate, tartaric acid and reducing solutions were pre-

pared according to a scheme outlined by Shapiro and

22

Brannock (1956). A Bausch and Lomb colorimeter with a
13 mm cell was used to measure the per cent transmission
at 700 mu.

While few inorganic ions have been shown to inter-
fere with this method, iron and orthophosphates interfere
markedly if proper precautions are not taken. Bunting
(19AM) has shown that tartaric acid removes the phosphate
interference. Iron presents a problem only in the
presence of phosphates so that removal of the phophates
also causes the iron intereference to be removed. Other
investigations by Ingamells (1966) and Kahler (19Al) have
given similar results.

Tartaric acid was added according to the scheme of
Shapiro and Brannock (1956) to eliminate iron inter-
ference. The characteristic colors, however, failed to
develop when the solutions were run through the deter-
mination scheme even though silica was to be expected.
Interference due to the high concentration of oxalic
acid was suspected and a series of tests indicated this
to be the case. Four sets of stock silica solutions were
made up and oxalic acid added in varying amounts to
determine the interference. Enough oxalic acid was added
to the first set to make the final concentration of
oxalic acid 9 gm/1000 ml, the second .9 gm/1000 m1 and
.09 gm/1000 ml in the third set. No oxalic acid was added

to the fourth set. In the first two sets no color developed

23

at all. In the last two sets equal color developed indi-
cating that concentrations of oxalic acid of .09 gm/1000
ml or less cause no interference. 1000—fold dilution of
the mineral solutions produced this concentration of
oxalic acid and was therefore used in the analytical
scheme. The solutions for the working curves were made up
so as to contain .09 gm/1000 m1 oxalic acid as an added
precaution.

In addition to the above interference, the presence
of colloidal silica raised further problems in the analysis
for silica in true solution. The colorimetric method will
determine only silica in true solution or present as low-
molecular weight polymers. In the case of the olivine
solution it was found that dilution caused some of the
colloidal silica to go into solution after a time. To
eliminate the possibility of large errors in the values
determined for silica in true solution, all four solutions
‘were run immediately after dilution (within about five
Ininutes). A slight error, however, could have resulted
from the breakdown of colloidal silica. This breakdown,
liowever, is not instantaneous as indicated by Figure 5
(after Krauskopf, 1956) and probably did not cause a large
earror. Also, except for the low value for the nonreactive
Seectioned pyroxene, the values determined are approximately
tllose determined by Krauskopf (1956) for the equilibrium

COncentrations of silica in very acid solutions containing

2A

colloidal silica. The results of the analyses (done in

triplicate) are shown in Table 3.

Analysis for Colloidal Silica Plus
Silica in Solution

 

 

Having determined that colloidal silica was present
in one of the solutions, an attempt was made to determine
the total silica in all the solutions. This was done by

treating the solutions with NaHCO to destroy the colloidal

3
silica (Roy, 1935). In this treatment 0.2 gm NaHCO3 were
added to 50 m1 of the solution, and the solution heated

on a steamtable for one hour. The solutions were then run
through the scheme as above. The results of the analyses
(done in triplicate) are shown in Table A. The decrease
for the amphibole may be due to loss in the small amount
of precipitate which formed in this case upon addition

of the NaHCOB. This precipitate was filtered off and may
have contained some silica. It is assumed that there was
at least 163 ppm in this case. The difference between the
means for the total silica and the means for the silica

in true solution were calculated and are given in Table 5
to indicate the concentrations of colloidal silica.

Analysis for Calcium, Magnesium
and Total Iron

 

 

For the determination of calcium, magnesium and
total iron 10 m1 portions of each of the four mineral solu-

tions were drawn off in January, 1967 (six months after the

25

TABLE 3.--Silica in true solution, in ppm 810

2.

 

 

 

 

 

 

 

 

 

. . Pyroxene Pyroxene
Amphlbole Olivine (ground) (sectioned)
180 190 220 80
180 210 210 30
130 180 180 30
E 163122 l93i11 207114 A7122
TABLE A.—-Total Silica, in ppm 8102.
Pyroxene Pyroxene
Amphibole Olivine (ground) (sectioned)
1A0 10,200 220 220
150 11,200 250 210
130 10,900 270 110
X 140:7 10,770i377 2A6il8 180iu7
(163?)
TABLE 5.--Colloidal silica, in ppm 8102.
. . Pyroxene Pyroxene
Amphibole OliVlne (ground) (sectioned)

 

0? 10,577 39

113

 

26

 

 

 

Dissolved Silica

—--_.—. . _——.—_

Fig- 5 Curves Showing Disoggregofion
of Silica Solon Standing

(offer 'KrouskOpf)

|20
PPm

IOO

80

 

 

flmtotoIAIIica I87

PW"

 

O'

/

Aer——

D- II

 

 

I-Oc'“ 94

d

i

)-—- u

 

ISOppm

75ppm

 

"-0— " 47pipm

 

 

 

 

 

 

 

 

20 4

O

Time

60

curves I, II and III from solutions mode

80 Days

by a geing and diIuIing solutions of Na§iO,omr
neutralizingwithHCI.CurveslVond Vfrom

hot-spring water.

 

 

27

experiment was set up) using a 10 m1 pipette. Care was
taken not to disturb the fine material in the bottom of
some of the beakers. The solutions were then filtered
through Whatman #2 filter paper and analyzed for calcium,
magnesium and total iron using a Perkin Elmer Model 303
atomic absorption unit. No pretreatment outside of
dilution was used. The results are shown in Table 6.

A desire to see if the systems were at equilibrium
as well as the possibility of silica interference in the
first runs made it necessary to rerun the solutions
separated in January and to run new portions separated in
March, 1967. 5 m1 of a 5% solution of lanthanum oxide
was included in each 25 ml of the mineral solution during
these runs to eliminate silica interference.

In addition to the interference from silica, it was
found that an orange precipitate formed in the amphibole
solution after a few weeks (possibly ferrous oxalate). It
was assumed that any precipitate of Ca, Mg or iron was
probably an oxalate and so all solutions were dry-ashed
at “50°C overnight to destroy the oxalates. The residue
was taken up in 2N HCl and both the January and March
solutions run with the lanthanum oxide solution added to
prevent silica interference. The results are shown in
Tables 7 and 8.

A drastic reduction in the values for magnesium and

total iron in the solution separated from the olivine in

28

March (see Table 8) indicated that these ions might have
been tied up in a small amount of solid that could not

be brought into solution with the 2N HCl. The high con-
centration of silica in the olivine solution before dry-
ashing led to speculation that perhaps the slight amount
of solid formed was a silicate. The small amount of solid
was separated and treatment with hydrofluoric and sulfuric
acid brought it into solution. The solution was then
diluted to the same volume as the solution it was
separated from and run on the atomic absorption unit for
magnesium and iron. The concentrations were then added

to the magnesium and iron concentrations determined for
the olivine solution before the solid was digested and

the totals given in Table 8. As can be seen from Table 8
this treatment increased the value for the iron but not

the value for the magnesium.

X-Ray Diffraction Work

 

In addition to the analyses of the solutions, the
solids were examined by means of x-ray diffraction. In
Fall of 1966 a portion of the solid residue from the beaker
containing the ground pyroxene was sedimented on a glass
slide and the sample run on a Norelco x-ray diffractometer.
At the same time a portion of the unaltered pyroxene was
ground in an agate mortar, sedimented on a glass slide and

a pattern taken.

29

TABLE 6.--Ca, Mg and Fe for solutions separated in

January--no lanthanum added.

 

 

 

 

 

Pyroxene Pyroxene
Amphibole Olivine (ground) (sectioned)
Mg 446 ppm 281 183 10
Fe 1450 65 40 2
Ca - — l 2
TABLE 7.--Ca, Mg and Fe for solutions separated in
January--lanthanum added.
Pyroxene Pyroxene
Amphibole Olivine (ground) (sectioned)
Mg 493 ppm 454 213 12
Fe 1755 89 ”6 5
Ca 27 4 l3 9

 

TABLE 8.-—Ca, Mg and Fe for solutions separated in March--

lanthanum added.

 

 

Pyroxene Pyroxene
Amphibole Olivine (ground) (sectioned)
Mg 789 ppm 76 péus 8 298 15
= 4
Fe 2250 34 plus 64 64 12
=98
Ca 33 3 15 10

 

30

In June, 1967 diffraction patterns were taken of the
amphibole and olivine and their residues in the oxalic
acid solution. Portions of the solid residues and the
original olivine and amphibole were collected and ground
in a mullite mortar. As more of the sample was available
in these cases, the samples were mounted as powders
rather than sedimented on glass slides. No patterns were
taken of the sectioned pyroxene as little alteration
occurred in this case. The scanning speed on all
patterns was 1° 2 theta/minute.

Each weathering product pattern was then compared
with the pattern given by the unaltered mineral and the
changes noted. In addition to new peaks, shifts and
changes in shape of old peaks were noted. Also, some
peaks present in the patterns of the unaltered minerals
were missing in the patterns of the products. An attempt
was made to identify as many of the new peaks as possible
and it was found that most of the new peaks might be
accounted for by the presence of magnesium oxalate, calcium
oxalate, and changes in some of the interplaner spacings
of the original mineral. Lists of the new peaks found
in the olivine, amphibole and pyroxene weathering product
patterns are given in Tables 9, 10 and 11. Included in
these lists are the d-spacings, estimated intensities and
possible materials causing them. The complete patterns

are included at the end of the report and some discussion

31

as to the processes indicated by the patterns and the

solution data are included in the next section.

32

TABLE 9.--New peaks in olivine residue pattern.

 

 

Estimated Possible
d-spacing Intensities Material
4.98 A 10* MO**
3.90 4 SO***
3.77 3 SO
3.35 1 ?
3.28 2 ?
3.21 10 MO
2.788 3 SO
2.571 2 MO
2.535 2 MO
2.481 2 SO
2.398 4 MO
2.292 1 SO
2.265 2 SO
2.104 2 MO
2.049 5 MO
1.87 2 MO

 

*On a scale of 10.
**MO equals magnesium oxalate.
***SO equals shifted olivine.

33

TABLE 10.--New peaks in the amphibole residue pattern.

 

 

Estimated Possible

d-spacing Intensities Material

5.95 A 7 00*

4.80 7 SA**

4.53 8 CO?

3.65 6 CO

3.19 3 SA

2.96 6 CO plus A***

2.564 7 CO?

2.350 6 co plus A

2.027 8 CO?

 

*CO equals calcium oxalate.

**SA equals shifted amphibole.

***A equals amphibole.

TABLE ll.--New peaks in the ground pyroxene residue.

 

 

 

Estimated Possible
d‘SpaCing Intensities Material
6.51 A 2 ?
6.37 l ?
5.91 8 CO*
5.76 5 CO?
4.00 2 ?
3.75 1 ?
3.62 9 CO
3.18 4 SP**
2.96 6 CO
2.83 2 CO
2.59 1 CO?
2.483 3 CO
2.34 4 CO
2.25 1 CO

 

*CO equals calcium oxalate.
**SP equals shifted pyroxene.

CHAPTER V

DISCUSSION OF RESULTS AND CONCLUSIONS

Extent of Alteration

 

As indicated in the section on the collection of data
all the minerals altered to some extent. This work, along
with other work referred to earlier, indicates that such
organic acids can alter a number of common ferromagnesian
silicates. In the alterations the olivine released the
most material into the solution and the amphibole less than
the olivine but more than both the sectioned pyroxene and
the ground pyroxene. The sectioned pyroxene was barely
altered. If one examines Tables 7 and 8 it can be seen
that not even the finely ground pyroxene was altered as
strongly as the olivine and amphibole because not as much

material was released into solution.

Processes

 

From closer consideration of the chemical analyses
and the diffraction patterns a better picture of the alter-
ation processes was obtained. The processes indicated are
given here schematically to illustrate, in a simple

fashion, the overall processes in the systems.

34

""1

y-

. _
i

35

olivine plus oxalic acid::§>altered olivine plus

colloidal silica plus

(Ca++

-C2++Ou,Mg -
C20“, Fe+ - C 20“,
HuSiOu, C20“) plus
magnesium oxalate as a

precipitate

amphibole plus oxalic acid ::>altered amphibole plus
(CaH — c 204’ MgH - c 20,,
++ 0:
Fe - C204, HuSiOu,C O4
and other ions) plus

calcium oxalate as a

precipitate

pyroxene plus oxalic acid ::;>slight1y altered pyroxene

plus colloidal silica plus

(Ca ++

++
Fe - c 204’ HuSiOu, c 20:)

- c 204’ MgH — c 204’

plus calcium oxalate as a

precipitate

When these general processes are considered it can be seen

that the systems are complex from a chemical standpoint. In
addition, general increases in the concentrations of calcium,
magnesium and iron from Jaunary to March (see Tables 7 and 8)

indicated that the systems were not at equilibrium.

36

While the systems are complex and were not in equil-
ibrium some idea as to the processes involved in the alter-
ation can be obtained by examining the solution data and
diffraction patterns more closely. Ratios of moles* of
Fe/Si, Mg/Si and Ca/Si were determined for both the minerals
and the solutions in order to determine preferential
removal of ions where no precipitate was involved and to
examine net mobilization of material in cases where precip-
itates were involved. The values for the total silica and
the concentrations of iron, magnesium and calcium deter-
mined in March were used. These ratios are given in
Tables 12 and 13.

*The way in which these ratios were obtained will
be given here. As a first step the concentrations of CaO,
FeO, MgO and 8102 in weight percentages in the minerals
were expressed as grams per 106 grams of mineral. The
numerators were then divided by the molecular weights of
the oxides to get the number of moles of ion per 106 grams
of mineral. This was done in order to compare numbers of
particles rather than direct weights. A similar procedure
was carried out for the values obtained in ppm in the
solutions. The concentrations were expressed as grams per
106 grams of solution and the numerator divided by the
atomic weight of the element in the cases of iron and
magnesium, and the weight of the oxide in the case of

silica. This gave the concentrations in terms of moles of

37

ion per 106 grams of solution. Ratios of Fe/Si, Mg/Si,
and Ca/Si were then calculated for the solutions and
analogous ratios calculated for the minerals.

While the net relative mobilities indicated by these
ratios are of value in discussing natural systems, some
consideration must be given to the precipitated oxalates
if such ratios are used to examine the way in which the
minerals themselves were attacked. For instance, a
decrease in the Ca/Si ratios going from the mineral to the
solution probably means that calcium was being removed
from solution as calcium oxalate rather than that silica
was being removed from the mineral preferentially. Table
14 indicates whether precipitates were formed containing
iron, magnesium or calcium for each of the mineral-solu-

tion systems.

Fe/Si Ratios

 

As can be seen from examination of Tables 12 and 13
the Fe/Si ratios increased in the solutions over all
minerals except the olivine. Ferric oxalate is very
soluble in pure water while the solubility of ferrous
oxalate in pure water is 220 ppm. No iron oxalates were
detected in the residues. These results suggests prefer-
ential removal of iron in the cases of the amphibole and
pyroxene and preferential removal of silica in the case of

the olivine.

 

38

TABLE 12.--Ratios of concentrations in solutions.

 

 

Pyroxene Pyroxene
Amphibole Olivine (ground) (sectioned)
Fe/Si 14.81 .0098 .280 .0716
Mg/Si 11.85 .0193 3.00 .206
Ca/Si .304 .000418 .0914 .0830

 

TABLE l3.--Ratios of concentrations in minerals.

 

 

Pyroxene Pyroxene
Amphibole Olivine (ground) (sectioned)
Fe/Si .611 .190 .0703 .0703
Mg/Si .292 1.81 .508 .508
Ca/Si .404 .0880 .545 .545

 

TABLE 14.--Precipitates formed in the mineral-solution

 

 

system.
Pyroxene Pyroxene

Amphibole Olivine (ground) (sectioned)

Fe No No No Not
precipitate precipitate precipitate x-rayed

Mg No Magnesium No Not
precipitate oxalate precipitate x-rayed

Ca Calcium No Calcium Not

oxalate precipitate oxalate x-rayed

 

39

Mg/Si Ratios

 

Similar results were obtained for the Mg/Si ratios.
These ratios increased in the solutions over the amphibole
and ground pyroxene but decreased over the olivine. Two
differences were noted, however, between the behavior of
the iron and that of the magnesium. First, a decrease
occurred in the Mg/Si ratio in the solution over the sec-
tioned pyroxene. This decrease may be the result of error
in the determination of the magnesium concentration in
this solution since the concentration involved was only in
the range of 15 ppm. Second, magnesium oxalate was indi-
cated in the olivine residue by the diffraction patterns.
The solubility of magnesium oxalate in pure water is about
700 ppm. This is well above the value of 84 ppm obtained
for the March analysis and even above the 454 ppm obtained
in January. It should be kept in mind, however, that the
700 ppm solubility is for pure water whereas the olivine
solution contained large amounts of colloidal silica. These
results imply a preferential removal of magnesium in the
amphibole and pyroxene systems. The decrease in the Mg/Si
ratio in the solution over the olivine only implies a net
movement of silica (colloidal silica and silica in true
solution) into the liquid phase. Preferential removal of
silica cannot be maintained in this case and, in fact, the
reverse situation (preferential removal of magnesium) may

have been the case as precipitation of magnesium oxalate

40

would tend to maintain a concentration gradient for the
magnesium between the mineral and the liquid thus promoting

removal of magnesium.

Ca/Si Ratios

 

Somewhat in contrast to the behavior of the Fe/Si and
Mg/Si ratios, the Ca/Si ratios decreased in the solutions
over the minerals in all cases. The presence of calcium
oxalate peaks in patterns of the amphibole and the ground
pyroxene residues, however, indicates that this was due to
the formation of calcium oxalate which is almost insoluble
in pure water. Calcium oxalate was not detected in the
olivine residue, probably because not enough calcium was
present in the mineral for the formation of a detectable
amount of calcium oxalate. While the net result was removal
of silica (as colloidal particles and in true solution) the
calcium may have been removed from the mineral preferentially
As with the formation of magnesium oxalate in the olivine
solution, the formation of calcium oxalate would tend to
maintain a concentration gradient between the mineral and
the liquid and thus promote removal of calcium from the

mineral.

Summary of Processes Indicated by the Study

 

In general, consideration of these ratios indicates
the role played by the mineral structure and by the organic
acid in determining the relative mobilities. At one extreme
the Fe/Si ratios appear to reflect structural control. In

the olivine system more silica was released than iron while

Flu—v: ma

J-

[MTMH'h‘M‘I‘I-y ’ ‘L‘!

41

in the pyroxene system there was an increase in iron. In

the amphibole system the increase of iron in the solution
was still greater. These results seem reasonable in that
it becomes relatively easier to remove iron and more
difficult to remove silica as the degree of polymerization
of the (SiOu) tetrahedra increases. Going from the
olivine to the amphibole the type of alteration goes from
one in which the whole mineral is attacked to one in

which the cations are more easily extracted from between
the polymerized chains of (810“) tetrahedra. At the other
extreme the decreasing Ca/Si ratios going from mineral to
solution reflect the control of the organic acid. In the
formation of calcium oxalate the calcium concentrations
were held down in the solutions. The Mg/Si ratios appear
to be controlled in part by the mineral structure (in the
amphibole and pyroxene systems) and in part by the solu-
bility of the oxalate (in the olivine system).

In close connection with the alteration discussed
above is the release of silica demonstrated by the experi-
ment. Both silica in true solution and colloidal silica
were determined. These results bring up a point which has
sometimes been overlooked in discussions on the removal
of silica from minerals. This is that colloidal silica
may be involved as well as silica in true solution and
should be included in calculations having to do with
removal of material. This is well illustrated in the case

of the olivine which contained more than 10,000 ppm silica

42

even at a very low pH. The sectioned and ground pyroxene
solutions also contained some colloidal silica. The inclu-
sion of colloidal silica in the calculations may effect the
results significantly. In the natural situation the col-
loidal silica may be taken into solution after dilution or
it may be flocculated very soon after the colloid forms r
(perhaps on the mineral itself) or further down in the

soil horizon. The net result, however, is the removal from

the mineral and possible formation of new minerals. In A

 

what state the silica was released from the mineral was not
determined. The silica may have been released as molecular-
sized particles and then aggregated to colloidal-sized
particles as the amount of silica in true solution reached
the equilibrium value, or it may have been released

directly as colloidal—sized particles.

Also in close connection with the alteration discussed
above are the exchange processes involved in the removal of
charged particles. The net removal of iron and magnesium
in the amphibole and pyroxene; suggested removal of calcium
in all of the minerals; and the suggested removal of mag-
nesium from the olivine could not proceed without intro-
duction of positive charge into the structure to maintain
charge neutrality. A logical source of charge can be found
in the hydrogen ions. pH values of the solutions, measured
five months after the experiment was set up, tend to support

this. The pH of the oxalic acid solution with no mineral in

43

it was about .95. The pH of the ground pyroxene solution
was about 1.05, the pH of the amphibole solution about
1.10 and the pH of the olivine solution about 1.35. These
pH values increase in the same direction as the degree of
exposure of silica surface. However, the trend of expan-
si6n on alteration indicated by the asymmetry t6wards
lower 2-theta values of the majority of those diffraction
peaks which were shifted suggests the introduction of the
larger H3O+- ion rather than the hydrogen ion. In the
case of the iron in the olivine solution a source of
negative charge is needed. Perhaps the oxidation state
of the iron was lowered during the alteration process;

however, but no iron-bearing compound was identified in

the residue, indicating that the product was amorphous.

Application to Natural Weathering

 

The results of this study combined with Boyle and
Voigt's (1967) study on biotite show that oxalic acid can
alter several ferromagnesian silicates which are common
rock-forming constituents. This, combined with the demon-
strated presence of oxalic acid and simple organic acids
in soil solutions and the products of microorganisms, pro-
vides evidence for simple organic acids as weathering
agents in nature.

The oxalic acid-mineral interaction produced differ-
ent types of alteration that seemed to depend on the degree

of polymerization of the (SiOu) tetrahedra. This variable

44

type of alteration may possibly be seen in such natural
features as the absence of olivine residues in highly
weathered material, and is indicated by the investiga-
tions of Edelmann and Doeglas (1932) on detrital, etched
pyroxenes and amphiboles in which structural control was
indicated by the shapes of the grains. The preferential
removal of cations demonstrated in this study also implies
a possible means for formation of new minerals. In addi-
tion, the formation of soluble and insoluble complexes in
the alteration processes may occur in nature. For
instance, iron in natural waters at the concentrations
determined is difficult to explain by inorganic means and
the formation of soluble organic complexes provides a
possible explanation. Also, according to Pecora and Kerr
(1954) hydrated oxalates occur naturally and have been
reported from a wide variety of organic and inorganic
environments, indicating the formation of insoluble com-
plexes in nature. According to Kononova (1961, p. 173)
and Bunting (1965, p. 94) the formation of a soil may
involve the formation of chelates, the solubility of
which may vary in the soil profile.

Finally, an aSpect of the study which has been
implied but not emphasized is that the processes indicated
by the study provide a mechanism whereby organisms can alter
minerals. The work of Wagner (1966) and Boyle and Voigt

(1967) using microorganisms support this.

REFERENCES CITED

American Society for Testing Materials (1960) X-ray
Powder Data File. American Society for Testing
Materials, 1916 Race Street, Philadelphia 3,
Pa. card 1-0256.

 

Bunting, B. T. (1965) The Geography of Soil. Aldine
Publishing Company, Chicago, pp. 50 and 94.

 

Bunting, W. E. (1944) "The Determination of Soluble
Silica in Very Low Concentrations." Indust. and
Eng. Chem., Anal. Ed. 16, 612-615.

 

 

Boyle, J. R. and G. K. Voigt (1967) "Biotite Flakes:
Alteration by Chemical and Biological Treatment."
Science. 155, 193—195.

Deer, W. A., R. A. Howie and J. Zussman (1962) Rock-
Formigg Minerals. John Wiley and Sons, Inc.,
New York, New York, vol. 1, pp. 2-3 and 10-13,
vol. 2, p. 43 and pp. 47-54.

 

Edelman, C. H. and D. J. Doeglas (1932) "Reliktstruk-
turen Detritischer Pyroxene und Amphibole."
Min. Pet. Mitt. XLII, 482-490 (abs.) in Boswell,
P. G. H. (1933) On the Mineralogy of Sedimentary
Rocks. Van Nostrand, N. Y., p. 209.

 

Evans, W. D. (1964) "The organic Solubilization of

 

 

Minerals in Sediments." Advances in Organic
Geochemistry. Macmillan 00., N. Y., N. Y., pp.
263-270.

Fyfe, W. S. (1964) Geochemistry of Solids. McGraw—
Hill Book Company, New York, San Francisco,
Toronto, London, pp. 122-123.

 

Gallagher, P. H. and T. Walsh (1943) "The Solubility
of Soil Constituents in Oxalic Acid as an Index
of the Effects of Weathering." Proc. Royal Irish

 

Academy. XLIX, 1—25.

Goldich, S. S. (1938) "A Study in Rock Weathering."
Jour. Geol. 46, 17-58.

 

45

 

46

Gruner, J. W. (1922) "Origin of Sedimentary Iron
Formations." Jour. Ec. Geol. 17, 407-460.

 

Ingamells, C. O. (1964) "Absorptionometric Methods in
Rapid Silicate Analysis." Analytical Chemistry.
38, 1228—1234.

 

Kahler, L. H. (1941) "A Photometric Method." Ind.
and Eng. Chem. 13, 612—615.

 

Kaurichev, I. S., T. N. Ivanova and E. M. Nozdrunova
(1963) "Low Molecular Weight Organic Acids in
Water-Soluble Matter of Soils." Pochvovedenie
(3) 27-35. (ast in Chem. Abs. 59:9265f.

 

 

Kaurichev, I. S., T. N. Ivanova and E. M. Nozdrunova
(1962) "The Content of Low Molecular Organic
Acids in Forest Litter." Dokl. Mosk. Sel Skokhoz.
Akad. NO. 76, 19-25. (abs.) in Chem. Abs.
58:10681a.

Keller, W. D. (1957) The Principles of Chemical
Weathering. Lucas Brothers publishers, Columbia,
Missouri. pp. 28-34.

 

Kononova, M. M. (1961) Soil Organic Matter.
Pergamon Press, New York. p. 47 and p. 173.

Krauskopf, K. B. (1956) "Dissolution and Precipitation
of Silica at Low Temperatures." Geoch. et.
Cosmochim. 10, 1-26.

 

 

Martell, A. E. and M. Calvin (1952) Chemistgy of the
Metal Chelate Compounds. Prentice-Hall, Inc.,
Englewood Cliffs, N. J. pp. 50-55 and 516-517.

 

 

Mueller, H. F., T. E. Larson and M. Fenetti. (1960)
"Chromatographic Separation and Identification of
Organic Acids--Application to Detection of Organic
Acids in River Waters." Analyt. Chem. 32,
687—690.

 

Pecora, W. T. and J. H. Kerr (1954) "Whewellite from a
Septarian Limestone Concretion in Marine Shale
Near Havre, Montana." Am. Miner., 39, 208—214.

 

Pettijohn, F. J. (1941) "Heavy Minerals and Geologic
Age." Jour. Geol., 49, 610-625.

 

47

Roberts, J. D. and M. C. Caserio (1965) Basic Prin-
ciples of Organic Chemistry. W. A. Benjamin, Inc.,
New York, Amsterdam, p. 550.

 

 

Roy, J. C. (1945) "Silica in Natural Waters." Am. Jour.
Sci., 248, 393-403.

Schalsha, E. B., H. Appelt and A. Schatz (1967)
"Chelation as a Weathering Mechanism--I. Effect
of Complexing Agents on the Solubilization of
Iron from Minerals and Granodiorite." Geoch. et.
Cosmochim., 31, 587-596.

 

 

Shapiro, L. and W. W. Brannock (1956) "Rapid Analysis
of Silicate Rocks." U. S. G. S. Bull. 1036-0.

 

Thiel, T. A. (1927) "The Relative Effectiveness of
Bacteria as Agents of Chemical Denudation."
Jour. Geol., 35, 647-652.

 

Wagner, M. (1966) "Vorkommen und Rolle Oxalat-Verwer-
tender Mikroorganismen Bei Verwitterungsprozesses."
Zeitschrift fur Microbiologie., 6, 197-209.

(abs.) in Chem. Abs., 65:10995c.

 

 

Whittaker, E. J. W. (1960) "The Crystal Chemistry of
the Amphiboles." Acta. Cryst., 13, 291-293.

 

APPENDIX

48

49

altered pyroxene (below)--taken
at 1 degree/minute using Cu

unaltered pyroxene (above) and
radiation.

x-ray diffraction patterns of

 

 

Dashed lines indicate peaks in
the residue due to unaltered

material.
C0 equals calcium oxalate and

SP equals shifted pyroxene.

  

 

 

 

 

 

 

 

 

 

 

 

 

50

mwm

H I 1 d d J 1 1 1 .- 1 I 4 c J. - 11 u 4 1!

4L0

 

 

 

 

 

 

 

 

 

.o:fi>HHo oouhanm massao cm
can mandoxo adfimocwma macado o:

.hopafim Hosea: m can

scandaomn so wcfims ouscda\ochwoc H as :oxmull
.Azoaonv ona>aao oououam can Ao>onmv oGH>HHo
oonouamns mo mnhmppmo coauomhmmac moans

3:3 3}

0.2m0mo os.

 

O<4

 

 

 

 

 

51

 

  

 

 

 

!—-——————-—————--—-I-——-—-|:

4

 

1

—..——_.’___.-_@_‘.._--__..

 

 

 

 

 

 

 

 

 

 

 

m. .iil .. .. ..21-;.lii.ImVWaliii..-.;
_ .5 n8 H
_ 1 oo . oo . n+ _
_ OU _ H _
“ <m m H. + «00 K< ”
_ .g e 1m " _ m

h ___ _ —

weaonannsn nauseo.1 one oaonsnose
poundnm mesuo 4m .mvmudxo afifiOHmO mamsuo 00

~
‘41 J u 1 q d 1 Junk-'—

.
so woodman can :a nxmmn_caMofiU:H wo:HH.conmnm
_ _ “

.nu _.noaanu Hosea: a new
coaumHUMA :o_ .mcfim: ouacda\oonwmo w an woBMplr
oamnv oHoanwEm cohowwd_ bad Actonwv oHonHaQaw
ochopacnd no m8 893nm cospowhmmaw amulw

—_.———

 

 

 

 

 

 

 

 

OU

—————po——————-————-

    

  

"'IIIIIIIIIIIIIIIIIIIIIIIIIZIIIIIIIIIIIIIIs