Clays and Clay Minerals,
Wol.44, No. 6, 835 842, 1996.
ILLITE P O L Y T Y P E Q U A N T I F I C A T I O N U S I N G W I L D F I R E 9
C A L C U L A T E D X-RAY D I F F R A C T I O N PATTERNS
GEORG H. GRATHOFF1 AND D. M. MOORE2
' Department of Geology, University of Illinois, Urbana, Illinois 61801
2 Illinois State Geological Survey, 615 E. Peabody Dr., Champaign, Illinois 61820
Abstract--Illite polytype quantification allows the differentiation of diagenetic and detrital illite components. In Paleozoic shales from the Illinois Basin, we observe 3 polytypes: 1Md, 1M and 2M~. 1Md
and 1M are of diagenetic origin and 2M1 is of detrital origin. In this paper, we compare experimental
X-ray diffraction (XRD) traces with traces calculated using WILDFIRE9 and quantify mixtures of all 3
polytypes, adjusting the effects of preferred orientation and overlapping peaks. The broad intensity ("illite
hump") around the illite 003, which is very common in illite from shales, is caused by the presence of
1Md illite and mixing of illite polytypes and is not an artifact of sample preparation or other impurities
in the sample. Illite polytype quantification provides a tool to extrapolate the K/Ar age and chemistry of
the detrital and diagenetic end-members by analysis of different size fractions containing different proportions of diagenetic and detrital illite polytypes.
Key Words--Cis-vacant, Illite, Polytypes, Quantification, Trans-vacant, X-ray Diffraction.
INTRODUCTION
Illite is the major component of modern muds and
ancient shaley rocks and an important diagenetic and
detrital mineral in low-temperature sedimentary systems. Weaver and Broekstra (1984) and Hunziker et
al. (1986) used the presence and quantity of diagenetic
2M, illite as a geothermometer. However, most 2M1
illite in shales is not diagenetic but detrital in origin,
while the IMd and 1M polytypes are commonly diagenetic (Bailey 1966). We quantify illite polytypes to
measure the relative proportions of detrital (2M0 and
diagenetic (1M + 1Md) illite.
Polytypism, a special case of polymorphism, is defined in Bailey et al. (1977) and Guinier et al. (1984).
Smith and Yoder (1956) derived the 6 possible mica
polytypes. For illite we find in nature the 2M 1, the 1M,
the disordered 1M, the 1M d and the 3T polytypes
(Levinson 1955). Although the 3T polytype has been
reported in few cases (Horton 1983), Reynolds and
Thomson (1993) showed that the 3T polytype can be
easily mistaken for the cis-vacant 1M polytype. Therefore, the 3T polytype is not considered in our quantification.
Various methods have been used by other workers
to identify and quantify illite polytypes. All of the
methods divide the peak area or peak height of a polytype-specific 2M 1 peak by either a peak that is common to all illite polytypes or a peak that is unique to
the 1M polytype. Maxwell and Hower (1967), Velde
and Hower (1963), and Reynolds (1963) quantified the
amount of 2M 1 and 1Mj illite by dividing the area of
a peak unique to the 2M 1 polytype by the area of the
2.58-.~ band, common to all illite polytypes. Maxwell
and Hower (1967) used the 2M~ polytype-specific
Copyright 9 1996, The Clay Minerals Society
2.80-A peak, Velde and Hower (1963) the 3.74-A
peak, and Reynolds (1963) the 3.00-,~ peak.
Caill~re et al. (1982), described in Dalla Torre et al.
(1994), quantified the amount of 2M1 and 1M illite by
dividing the area of the 3.00-A peak, which is unique
to the 2M1 polytype, by the area of the 3.06-.~ peak,
which is unique to the 1M polytype. 1Md illite cannot
be quantified using their method. Tettenhorst and Cotbat6 (1993) used the ratio of 3 peaks (5.0 A, 2.58
and 2.80 ]k) and computer modeling to adjust for preferred orientation due to sample preparation. None of
the methods cited above quantifies all 3 polytypes,
corrects for small amounts of expandable layers and
interference minerals (for example, apatite 2.80 .A) or
can deal with rotational disorder in the 1M d polytype.
Illite XRD patterns of random powders often show
an enigmatic broad intensity centered around the illite
003 position. For simplicity, we will use the term "illite hump" for this diffraction phenomenon. We will
show that this illite hump can be explained either by
the presence of the 1M d polytype or by mixing of illite
polytypes.
The method presented in this paper compares experimental XRD patterns with calculated patterns
modeled using WILDFIRE 9 (Reynolds 1993, 1994).
These comparisons can adjust for effects of preferred
orientation, overlapping peaks, different cation occupancies, % expandability, crystallite size and chemical
composition. Even mixtures of 3 different polytypes
and mixtures of 1Md and 1M illite can be quantified.
We use illite polytype quantification to extrapolate
the diagenetic age and the detrital age of illite, similar
to the method of Pevear (1992), and to extrapolate the
chemistry of the end-members. The diagenetic age of
illite is important in determining the thermal history
835
836
Grathoff and Moore
of a basin, and the detrital age can determine the provenance. The chemical composition of the diagenetic
illite can be used to characterize the fluid chemistry
that formed the illite.
METHODS
Sample Preparation and XRD Data Collection
The quantification described in this paper compares
the experimental XRD patterns of random powders
with those calculated using WILDFIRE 9 Calculated
patterns are good models for natural samples because
the physical basis for modeling XRD patterns has been
clearly demonstrated (James 1965) and because
WILDFIRE 9 is based on first principles of XRD.
However, we acknowledge the limitations of modeling
in the sense of Oreskes et al. (1994), that models demonstrate a probable solution but are not proof in an
absolute sense.
When comparing calculated patterns with experimental XRD patterns, the resolution and reproducibility of the experimental XRD pattern are crucial. To
achieve this, both the methods of sample preparation
and the XRD data collection are very important.
We first size-separated the samples, then flocculated
the suspension using a few drops of 0.1 M CaC12, and
poured off the clear supernate, Next, we dialyzed the
samples to remove any ions in the suspension (for example, CI, Na and K). The suspensions were then
evaporated. Abundant chlorite, carbonate and organic
matter were removed. The organics were removed using 33%-strength household bleach (sodium hypochlorite). The chlorite and the carbonates were removed by
dissolution in a nearly boiling 1 M HNO 3 solution
overnight. Afterwards we added 0.1 M Na2CO~ solution to remove the silica in solution formed by the
dissolution of chlorite and rinsed the supernate with
deionized water several times. All treatments were followed by dialysis.
The best resolution of the illite polytypes was
achieved by: 1) maximizing the random orientation of
the powder pack; 2) removing the interlayer water
from smectite; and 3) step-scanning the powder pack
with a long counting time.
RANDOM ORIENTATION.N O sample preparation can entirely eliminate preferred orientation. To make the
powder packs as randomly oriented as possible, we
side-packed our samples (Brown and Brindley 1980).
A good measure of the randomness was the intensity
ratio of am illite 001 d-spacing with the illite 020 (4.5
,~). As a rule of thumb, acceptable randomness was
achieved if the 020 was significantly larger than the
002. This ratio does not apply for Fe-rich illite, which
has a very weak 002 reflection.
Austin et al. (1989)
showed that the interlayer water associated with ex-
REMOVAL OF INTERLAYER WATER.
Clays and Clay Minerals
pandable interlayers obscures the polytype analysis by
XRD. Moore and Hower (1986) showed that expandable interlayers rehydrate very quickly. Therefore, we
dehydrated the sample at 250 ~ for 1 h and prevented
rehydration by running the powder pack in a dry (Nz),
controlled atmosphere chamber. Illite with very few
expandable interlayers (<<5%) were not heated.
SCANNING RATE. lllite hkl reflections are not very intense and are often difficult to distinguish from the
background. Our goal was to achieve good counting
statistics and increase the signal-to-noise ratio. Therefore, we used the step-scanning mode with a step size
of 0.05 degrees and a counting time of at least 30 s
per step.
XRD data were obtained using a Scintag| 0-0- diffractometer with a DMS | operating system and operating conditions of 40 kV, 30 mA. This diffractometer
uses CuKtx radiation, a N2-cooled germanium detector,
2 Soller slits, a 2-ram divergence slit and a 0.5-ram
slit at the detector.
WILDFIRE 9 Description
The algorithm used in WILDFIRE 9 (Reynolds
1993) is based on basic optical principles (James
1965). WILDFIRE 9 can calculate the XRD tracings
of 1M, 2M1, 2M2 and 3T mica polytypes. In addition,
it can calculate disordered 1M XRD patterns. Disorder
is a function of the amount of 60 ~ 120 ~ 180 ~ 240 ~
and 300 ~ rotations between the 2:1 layers, the proportion of cis-vacant (cv) to trans-vacant (tv) sites in the
octahedral sheet and the % of expandable layers. 1M
illite can be cv, as discovered by Drits and coworkers
(Drits et al. 1984; Tsipursky and Drits 1984; Reynolds
and Thomson 1993), but 1M and 1Md illite are most
commonly tv. A subroutine of WILDFIRE 9 (Mixer 9
was used to mix the individual calculated and experimental patterns for our iUite polytype quantification.
The peak areas and heights of our experimental and
calculated patterns were measured using Plotmod 9
within WILDFIRE 9
RESULTS
Illite polytype quantification compares experimental
XRD traces with WILDFIRE 9 calculated traces. Figure 1 shows such a comparison. The example discussed in detail here is the 0.5-1 ~m size fraction of
an Upper Ordovician Maquoketa Group shale from the
Illinois Basin, The calculated pattern represents a
physical mixture of the following polytypes: 30% 2M~,
5% 1M tv and 65% 1M d illite. The 3 polytypes that
were mixed are shown in Figure 2. The 1M~ illite is
60% cv and 40% tv, the fraction of 0~ rotations (P0)
is 0.6 and the fraction of 60 ~, 180~ and 300 ~ rotations
(P6o) is 0,6. This I M d illite was used because Reynolds
(1993) modeled the Ordovician Deicke K-bentonite
from the Illinois Basin and concluded that it is 1Md
illite (60% cv, 40% tv, P0 = 0.6, P60 = 0.6).
Vol. 44, No_ 6, 1996
Illite polytype quantification
837
0.5-11~m: 65% 1M d, 30% 2M~, 5% 1M tv
4,~A
2M
1
2.55A
O
,
AI
2.58Aband
'I2/L / L
112
O!3A Q
3.33A
experimental
2.58A
388A 349]~I3"20A2'98A
2";
I
16
20
24
28
32
36
40
'
'
'
16
I
'
'
'
20
I
~
'
24
'
I
'
~
'
I
'
'
I
'
'
'
I
28
32
36
degrees 2-THETA
44
'
'
'
40
I
44
degree,s 2-'fHETA
Figure 1. Experimental and calculated trace of the 0,5-1 Ixm
size fraction of G 26-1. The dashed lines show the position
of 1M tv illite polytype-specific peaks; the triangles indicate
peaks specific to the 2M1 polytype; Q = quartz; F = K-feldspar, A = anatase. G 26-1 is an Upper Ordovician Maquoketa
Group core sample (Parrish core, Illinois State Geological
Survey, core # 7608) taken in Fulton County, Illinois, at a
depth of 236 m at the contact to the Galena Group, in the
western part of the Illinois Basin.
T h r e e p o l y t y p e s were u s e d to m o d e l s a m p l e G 26-1
(Figure 1) b e c a u s e the 1-12 ( 3 . 6 3 - * ) and 112 (3.07-.&)
1M illite p e a k s are too s h a r p to b e 1Md illite a n d the
illite h u m p a r o u n d the illite 003 is too large to b e
solely 1M illite. W e c o m p a r e d 3 features to find the
best fit b e t w e e n the calculated a n d e x p e r i m e n t a l tracings: 1) the intensity and b r e a d t h o f the polytype-specific reflections; 2) the intensity o f the 2 . 5 8 - A b a n d ;
a n d 3) the d e g r e e o f preferred orientation. Table 1
c o m p a r e s our results u s i n g W I L D F I R E 9 w i t h q u a n titative m o d e l s used in the literature. T h e m e t h o d used
b y Caill~re et al. (1982) quantifies 1M a n d 2M1 but
c a n n o t quantify 1Md; therefore, their v a l u e for 2 M l is
too high.
T h e quantification can also b e p e r f o r m e d b y c o m p a r i n g the m e a s u r e d p e a k h e i g h t s and areas in the exp e r i m e n t a l trace w i t h those in the calculated trace.
First we m i x e d the 3 e n d - m e m b e r s s h o w n in F i g u r e 2.
A n e x a m p l e , m i x i n g 2 M l w i t h IMo, is s h o w n in Figure
3. T h e n we m e a s u r e d the polytype-specific a n d the
2 . 5 8 - A p e a k h e i g h t s and areas o f these i n d i v i d u a l
mixtures. T h e data f r o m these m i x t u r e s were t h e n plotted a n d s h o w e d the f o l l o w i n g results for e a c h polytype:
4.51A
1 M tv
3.63A
4.34A
2.56A
3.o7A
2.4oA
5.ooA
I
'
'
2.1sA
'
I
16
'
20
'
'
I
24
'
'
'
I
'
'
'
I
'
t
,
I
28
32
36
degrees2-THETA
1 M
4.49A
d
'
'
'
1
'
40
'
'
I
44
2.57A
3.32A
A
Jl
4.98A
,
/I
\
3.elAj ~ 3.,OA
i
24
36
.2AOA .
--)
[-
Figure 2. Calculated traces of the 3 end-members used for
the quantification: 2M1, 1M tv and 1Md [60% tv layers; 40%
cv layers; fraction of 0 ~ rotations (P0) = 0.6; fraction of 60 ~
180~ 300 ~ rotations ( P 6 0 ) = 0 . 6 ; 10% expandable layers].
16
i
,
i
20
28
32
degrees2-THETA
40
44
838
Grathoff and Moore
Table 1. Illite polytype quantifications of sample G 26-1
(0.5-1 p~m) comparing techniques used in the literature with
the best fit using WILDFIRE 9 Two calculations were made
for each method, one based on peak area and one on peak
height. The equations used for the calculations are listed below the table.
70
12
20
85
5
21
38
85
12
0
[~.~,,
15
, I , ,
%2M1
, I , ,
60
80
,
100
proportioned to a reflection that is a measure of the
total illite. This is necessary because 1Md illite has
only weak to no polytype-specific reflections. Figure
1 shows that a mixture containing up to 65% of 1M d
illite shows no polytype-specific 1Md peaks. A reflection c o m m o n to all 3 polytypes and least affected by
preferred orientation is the 2.58-,~ band, which has
been used previously (Velde and H o w e r 1963; Reynolds 1963; M a x w e l l and H o w e r 1967).
Plots of the peak height and area of any polytypespecific reflections divided by the area of the 2.58 -0
band also show a linear trend. Figure 5 shows one
example. The resulting linear equation can be used to
calculate the % 2M~ illite. This linear trend holds for
all 2Ml polytype-specific peaks. E x a m p l e s of resulting
linear equations for calculating the % 2M1 illite are
listed in Table 2.
1M Illite
The peak height and area of the polytype-specific
1M peaks also increase linearly with increasing
0.14
2.58A band
'
'
'
t
'
'
'
I ,~1
I '
'
'
0.32
~
~.IM
,
d tv/cv
0.1
Y=
,
t
L~---.--~-~A~
~20%
~
~
40o/o 2M,
Area(2"BS"t~J,
06
~
60% 2M 1
0.04
L.JN 80% 2M 1
0.02
~
~2M
20
40
Figure 4. Area of the polytype-specific 32.1 ~ (2.80-,~, 1 li~)
peak for calculated patterns of 2M~ + 1M and 2M~ + IM a
mixtures.
OO3
A
20
mixture
020
,
, I , ,
, ,
0
Both peak heights and areas of 2Ml polytype-specific peaks increase linearly with increasing amount of
2M 1, independent of which illite polytype or polytypes
was m i x e d with the 2M~ polytype. Figure 4 shows the
peak area of the 2M1 polytype-specific 32.1 ~ (2.80-/k,
116) peak plotted against % 2M~ illite. The 2M, polytype pattern was m i x e d with 1M and 1M d illite patterns. Both mixtures, 2M~ + 1M and 2M1 + 1Md, lie
on the same line, which implies that the relative proportion of 2M~ illite is independent of the polytype
with which it is mixed. A reason for this is that the
mass adsorption coefficients o f the different polytypes
are similar, except for Fe-rich illite. All polytype-specific reflections o f 2M~ lie on lines similar to Figure
4, but with different slopes.
To quantify 2M~ illite by measuring the peak height
or area of 2Mj polytype-specific peaks is not sufficient. A polytype-specific 2M1 reflection needs to be
,
.
10
2M1 Illite
~
O
32.1~
peakarea4 0
30
2 M t - 1M d tv/cv
,
50
I
I
20
t % 2M, = 476 • (2.80 A/2.58 ,~).
:~ % 2M, = 100 x (3.00 A/2.58 A).
w % 2M1 = 0.23 + 242X - 398XA2 + 432XA3 -- 176XA4.
X = (3.00 ,~/(3.00 /~ + 3.06 ,~).
002
A
30
Height
WILDFIRE 9 (best fit)
2M1 + 1Md
2M 1 +IM
O
60
% 2M~
Area
Reynolds (1963)t
Maxwell and Hower (1967)$
Caill6re et al. (1982)w
Tettenhorst and Corbato (1993)
Clays and Clay Minerals
25
30
degrees 2-THETA
2M 1
1 illite
35
Figure 3. Stacked tracings showing effects of mixing 2M1
illite with 1Md illite. The peak diagnostics for the 2M1 polytype are marked by triangles and dashed lines.
Area(32.1~
~',,,~-,~,,
o
0
20
~,,
40
%2M,
60
i~__~
80
100
Figure 5. The area of the 32.1 ~ peak (2M] polytype-specific)
divided by the area of the 2.58-A band is plotted against %
2M~ illite. There are 2 points for every 10% 2M 1 because IM
and 1M a are mixed separately with 2M~.
Vol. 44, No. 6, 1996
Illite polytype quantification
Table 2. Equations to calculate % 2Ma using peak area and
height of polytype-specific 2Ml peaks.
204r
A
hkt
% 2M~(A~:)
23.8 ~
25.5 ~
27.8 ~
29.8 ~
32.1 ~
3.74
3.50
3.21
3.00
2.80
023
114
114
025
116
1.40+543 •
3.19+386•
2.05+360•
3.25 + 3 3 5 MA
1.88 + 7 0 2 •
839
0
% 2M~(Hw
O
300
-0.51 + 188 • 1 6 7
1.06+ 141 •
0.34+ 132•
0.83 + 1 2 6 •
-0.05 + 2 5 2 •
250
24.3
~
200
1Md Illite
1Md illite is difficult to quantify directly. T h e 1Md
p o l y t y p e can b e quantified b y difference:
[1]
or b y visual c o m p a r i s o n . T h e first way is easy and
q u i c k with a large error, w h i l e the s e c o n d is m o r e precise b u t m u c h m o r e t i m e - c o n s u m i n g . T h e size o f the
illite h u m p c a n b e quantified best b y visual c o m p a r i son.
0
9
0
0
150
0
100
0
0
50
0
,
,
I
,
0
a m o u n t o f 1M present. F i g u r e 6 s h o w s the 24.3 ~ (3.66.~, -/-12) p e a k area plotted against % 1M illite. T h e
p e a k areas o f 1M m i x e d w i t h 2 M 1 a n d 1M m i x e d with
1Ma increase linearly but lie o n different lines. T h e
r e a s o n is that 1M polytype-specific peaks overlap w i t h
the w e a k a n d b r o a d 1Ma polytype-specific peaks.
T h e r e are 2 m e t h o d s to quantify the 1M polytype.
T h e first m e t h o d , i n t r o d u c e d b y Caill~re et al. (1982),
d e s c r i b e d in D a l l a Torte et al. (1994), ratios the intensity o f a 1M p e a k w i t h the intensity o f a 2M1 peak.
T h e 4th order e q u a t i o n w e calculated u s i n g the data
f r o m D a l l a Torre et al. (1994) is listed in Table 1. T h e i r
m e t h o d c a n n o t quantify 1M d illite, w h i c h m a k e s it difficult to quantify illites in shales a n d bentonites.
T h e s e c o n d m e t h o d ratios a 1M p e a k w i t h the 2.58.~ b a n d , w h i c h is the same m e t h o d that was used to
calculate 2M1 illite. T h e e q u a t i o n s used to calculate %
1M illite are listed in Table 3.
I f the s u m o f 2M~ and 1M illite is smaller than
100%, either 1M d illite is p r e s e n t or the p o w d e r p a c k
is preferentially oriented. W h e n 1M d is m i x e d w i t h 1M
illite, difficulties arise b e c a u s e the 1Ma polytype-specific peaks o v e r l a p with the 1 M peaks. T h e s e overlapp i n g p e a k s are v e r y difficult to detect in the X R D trace
b e c a u s e the 1Md illite is c o m m o n l y e x p r e s s e d as v e r y
w e a k b r o a d p e a k s with a large illite h u m p , similar to
the 1Ma illite in Figure 2. I f the 1M illite polytypespecific p e a k s c a n b e m e a s u r e d i n d e p e n d e n t l y , the
e q u a t i o n in Table 3 c a n b e u s e d to calculate the perc e n t a g e o f 1M. I f p e a k overlap occurs, the best option
is to visually c o m p a r e the e x p e r i m e n t a l w i t h m i x e d
W I L D F I R E 9 calculated traces.
0
9
0
peak area
-t 20 values are for CuKcr radiation.
5;A = Area of 2M~ polytype-specific peak divided by the
area of the 2.58-~ band.
w H = Height of 2Ml polytype-specific peak divided by the
area of the 2.58-,~ band.
% 1Md = 100 -- (% 1M + % 2 M j )
1M+ 1MdI
1M+2M 1
350
20
,
,
,
I
40
,
,
% 1M
,
I
60
,
,
,
I
,
,
80
,
100
Figure 6. Area of the polytype-specific 24.3 ~ (3.66-,~, ]12)
peak for calculated patterns of 1M + 2M~ and 1M + 1M d
mixtures.
T h e r e are 2 ways to detect the p r e s e n c e o f 1Md illite: 1) b y the p r e s e n c e of the illite h u m p a r o u n d the
illite 003, or 2) b y the p r e s e n c e o f b r o a d polytypespecific I M tv or c v peaks with low intensities. T h e
size o f the illite h u m p a n d the b r o a d polytype-specific
1M p e a k s are a f u n c t i o n o f the d e g r e e o f 1Ma disorder.
1Ma c a n r a n g e f r o m c o m p l e t e l y disordered to v e r y ordered, t r a n s f o r m i n g f r o m 1Ma to I M . D i s o r d e r c a n b e
d e s c r i b e d as a f u n c t i o n o f P0, the p r o p o r t i o n o f n.60 ~
a n d n. 120 ~ rotations, p r o p o r t i o n o f cv to tv, a n d perc e n t a g e o f e x p a n d a b l e layers ( R e y n o l d s 1993). W e are
able to quantify the a m o u n t o f rotational disorder in
the 1M structure b y m e a n s of the v a r i a b l e P0. T h e r e
is yet n o a g r e e m e n t o n the value o f P0 that is n e c e s s a r y
for the t e r m 1Ma. In the a b s e n c e o f a p p r o v e d n o m e n clature, we will use the t e r m 1M for structures that
s h o w n o rotational d i s o r d e r b a s e d o n b r o a d e n i n g and
w e a k e n i n g o f the polytype-specific reflections.
A w o r d o f c a u t i o n w h e n d e s c r i b i n g the illite h u m p :
p u r e 2Ml illite also has an e l e v a t e d b a c k g r o u n d bet w e e n 21 and 34 ~
The elevated background comes
f r o m the a b u n d a n c e o f 2M1 polytype-specific p e a k s in
this range. A d d i n g 1M tv illite will add to the e l e v a t e d
b a c k g r o u n d b y e l e v a t i n g the 24.3 and 29.1 ~
area.
However, the e l e v a t e d b a c k g r o u n d due to m i x i n g o f
the ordered p o l y t y p e s is only a small c o m p o n e n t o f
Table 3. Equations to calculate % 1M peaks using area and
peak height of polytype-specific 1M peaks. Peak area and
height were measured without overlapping 1M d peaks.
20t
A
hkl
24.3 ~
29.1 ~
3.66
3.07
]12
112
% IM (A~:)=
% IM(Hw
4.98 + 136 • A~ 3.54 + 59.1 • Hw
3.40 + 132 x A
2.43 + 66.8 x H
t 20 values for CuKo~ radiation.
$ A = Area of 1M polytype-specific peak divided by the
area of the 2.58-,~ band.
w H = Height of 1M polytype-specific peak divided by the
area of the 2.58-A band.
840
Grathoff and Moore
Q
Clays and Clay Minerals
DISCUSSION
The following section includes a discussion of the
illite hump, the 2.58-,~ band and the potential uses for
polytype quantification.
Illite Hump
15
20
25
30
35
40
45
~ 2-THETA
Figure 7. Comparison of a natural experimental (G 26-1;
0.5-1 ixm size fraction) XRD trace, mixed experimental [65%
1Md (Tioga K-bentonite), 30% 2M~ and 5% 1M tv (RM-30)]
XRD trace, and mixed calculated (65% 1Md, 30% 2Ml and
5% 1M tv) trace. The dashed lines show the position of 1M
tv illite polytype-specific peaks; the triangles indicate peaks
specific to the 2M~ polytype; Q = quartz; F = K-feldspar, A
= anatase.
the illite hump. In most cases, it is due to the presence
of 1Md illite.
Test for Accuracy
We physically mixed different known polytypes to
test the accuracy of our method and to test if natural
samples can be simulated by mixing pure polytypes.
The following standards were used as models for pure
end-members: 1) the Tioga K-bentonite, a 1Md illite,
2) the RM-30 illite from Dennis Eberl (USGS, Boulder, Colorado), a 1M tv illite and 3) a 2MI illite of
hydrothermal origin from the Yangsan area, Korea
(courtesy of Yeongkyoo Kim, University of Illinois,
Urbana, Illinois). The Tioga K-bentonite was used instead of the Deicke K-bentonite because the Tioga
K-bentonite in our lab has fewer smectite interlayers
and the XRD trace is similar to the 1Mo illite modeled
in Figure 2. We mixed 65% 1Mo (Tioga K-bentonite),
30% 2M1, and 5% 1M tv (RM-30). These are the same
proportions as the calculated pattern simulating G 26-1
(0.5-1 ixm), shown in Figure 1. Figure 7 shows the
resulting mixed XRD trace together with the G 26-1
(0.5-1 p,m) experimental trace and the calculated
trace. All 3 patterns are very similar, indicating that
the mineral intensity factors of each of the 3 polytypes
are similar and that mixed calculated patterns are a
good model for naturally occurring mixtures. We calculated the percentage of 2M~ and 1M of the mixed
experimental XRD trace using the equations in Tables
2 and 3. The precision ranges from 2.5-5% absolute.
The better the individual polytypes are modeled, the
smaller the error.
The illite hump is an area of elevated intensity between 21 and 34 ~ centered around the illite 003.
There are several explanations for the illite hump.
Austin et al. (1989) showed that the illite hump decreased upon glycolation or heating, stressing that
sample preparation is crucial in studying the illite
hump. They explained the illite hump as either a
broadening effect due to fine particle size or an artifact
from noncrystalline inorganic and organic material.
We tested their conclusion on a pure illite, the Waukesha Illite (Grathoff et al. 1995). H202 and HNO3
were reacted with the iUite. Neither treatment changed
the nature of the XRD pattern and the calculated proportions of polytypes.
Another possible explanation is that the illite hump
is the summation of a large number of polytypes.
However, mixing 3 polytypes using WILDFIRE 9
matches the experimental patterns as well. Therefore,
we see no need for additional polytypes.
Calculated mixed illite polytype traces indicate that
the illite hump is caused by 2 factors: 1) mixing of
polytypes, and 2) the presence of IMd illite. In the case
of mixing 1M and 2M~ polytypes, the illite hump is
very small and can be called elevated background because it is a mixing effect. However, most illite humps
are much larger than the elevated background and are
caused by the presence of 1Ma illite. Reynolds' (1993)
modeling of 1Md illites showed that inherent with 1Md
illite is a large hump caused by rotational disorder,
different proportions of cv and tv sites and the presence of expandable layers, which increases turbostratic
disorder. Small amounts of 2M1 or 1M illite mixed
together with 1M~ illite enlarge the illite hump and
cause the broad 1Mu peaks to disappear.
Our conclusion is that the illite hump is not an artifact of 1) sample preparation; 2) a large number of
polytypes; or 3) inorganic X-ray amorphous material.
It is primarily due to the presence of 1Ma illite and
mixing of polytypes.
2.58-~k Band
1Md illite has different degrees of disorder, expressed in the presence or absence of polytype-specific
peaks. This complicates attempts to quantify the polytypes, especially if only polytype-specific peaks are
used. Common to all illite polytype XRD patterns are
the 001 reflections, the 020 and the 2.58-~k band.
The 2.58-,~ band consists of a number of peaks,
depending on the polytype. Using Bailey (1980) tabulation of peaks for 1M tv and 2M~ illite and Reynolds
Vol. 44, No. 6, 1996
Illite polytype quantification
84l
K/Ar vs % 2 M
Table 4. Decomposition of the 2.58-/~ band for the 2M~, 1M
tv and 1M cv polytypes.
hkl
IM tv
d-spacing
hkl
2M~
d-spacing
hkl
of
1
G 26-1, M a q u o k e t a
550
IM cv
d-spacing
,
,
,
,
,
Group
I , , , I , , ,
y = 321 + 1.97x R= 0.999
I
,
I
,
,
:fetrital
500
130
13T
200
2.58 ,&
2.56 ,&
2.55 A
13T
116
202
2.59 ,&
2.58 A
2.56 A
701
130
200
131
2.59
2.59
2.56
2.56
,&
A
A
A
K/Ar
age
45O
age
4O0
,,,,,,,,~. 5 g m
and Thomson (1993) for 1M cv, the individual peaks
in Table 4 make up the 2.58-]k, - 3 5 ~ (CuKe0 band.
The 2.58-A band contains k = 3n peaks, which are
unaffected by 120 ~ rotations. The 1M polytype can be
thought of as having no rotations, the 2M] as having
_+ 120 ~ rotations, and the 1M d as having n.60 ~ rotations, where n is both even and odd. If n is an odd
integer, the k = 3n peaks will be affected. But half of
the n.60 rotations will have an even integer, not affecting the k = 3n peaks. That is the reason why 1Md
illite XRD patterns contain the 2.58-A band. In addition, the 2.58-A band is affected less by preferred orientation than the 00l and the 020 reflections and is of
similar intensity between the different polytypes.
Therefore, the 2.58-]k band is a good proxy for the
total amount of illite.
Uses for Polytype Quantification
In most shales, illite consists of a mixture of at least
2 different polytypes of potentially different origins.
Usually, the larger the grain size, the higher the percentage of detrital 2M~ illite (Hower et al. 1963). Our
studies of the Maquoketa Group shales in the Illinois
Basin show that neither the <0.2 ~m size fraction nor
the < 2 ~m size fractions contain solely 1 polytype.
We find that the <0.2 p~m size fraction contains about
10% detrital 2M~ illite. Therefore, if illite polytype
quantification, chemistry and K / A r age are analyzed
from at least 3 different size fractions, the chemistry
and K/Ar age of the detrital and diagenetic end-members can be extrapolated. Figure 8 shows the extrapolated diagenetic and detrital ages of sample G 26-1,
an Upper Ordovician Maquoketa Group shale. The
diagenetic age is 320 Ma and the detrital age 520 Ma.
The stratigraphic age of the Maquoketa Group is about
440 Ma. This proves that the different polytypes are
of different origin, which is reasonable because 2Ml
illite, the high-temperature illite polytype, is mixed
with 1Md, the low-temperature polytype. We use a Silurian clay in Figure 9 to illustrate that the K content
of the 2 end-members may be extrapolated, indicating
that the 2 end-members have different K contents.
Other analyses, for example, stable isotopic composition, can also be extrapolated, providing that the individual size fractions do not contain impurities.
Pevear (1992) used a similar method, quantifying
the detrital and diagenetic components of shales by
350
dlagenetlc
age
300
,.,~<0.2
~m
,
,
,
I
0
,
,
,
20
~ ,
40
%
,
,
i
,
,
,
60
2M 1
I
,
,
,
80
100
Figure 8. Extrapolation of detrital and diagenetic K/Ar age
of a Maquoketa Group shale sample G 26-1 from a Fulton
County core (IL); gee Figure 1 for location using illite polytype quantification and K/Ar dating of different size fractions.
decomposing XRD patterns of oriented mounts using
N E W M O D 9 (Reynolds 1985) to extrapolate the diagenetic and detrital end-members.
CONCLUSION
Illite polytype quantification using WILDFIRE 9
can accurately calculate and model naturally occurring
mixtures of illite polytypes. It can be used to quantify
different polytypes of diagenetic and detrital origin,
allowing the extrapolation to end-member K/Ar ages
and chemical compositions. The illite hump, common
to illites from shales, is not an artifact of sample preparation, a large number of polytypes or inorganic
X-ray amorphous material, but is due to the presence
Change
in K c o n t e n t p e r h a l f u n i t c e l l
with increasing
% 2M
1
0.76
'
'
,
'
'
'
,
'
'
i
'
i
'
/
/
0.74
R= 0.965
KO.72
j
_
J
O
~0.2-0.5
0.7
0.68
0.5-1 ~m
gm
7<0.2,urn
0.66
J , ,
0
20
~ , ,
, [
40
60
%2M
,
, I
80
,
100
1
Figure 9. Extrapolation of detrital and diagenetic K content
per O~0 (OH)2 of an illite from a Silurian clay layer at the
base of the Brandon Bridge strata, near Waukesha, Wl, using
illite polytype quantification and X-ray fluorescence data of
different size fractions.
842
Grathoff and Moore
o f 1Md illite and mixing o f illite polytypes. Mixtures o f
illite polytypes are very c o m m o n in shales and often
include the 1Md polytype, the presence o f w h i c h has
been difficult to detect and quantify due to the absence
o f polytype-specific reflections. Studying and quantifying
illite polytypes give information about illite that cannot
be obtained using oriented patterns. R.C. Reynolds put
it best (personal communication, 1995 CMS meeting in
Baltimore, Maryland): " i f you ignore illite polytypes
w h e n studying illite and illite/smectite in shales and bentonites, you are ignoring half the information."
ACKNOWLEDGMENTS
This research was supported in part by a research grant
from the Clay Minerals Society. We thank K. Wemmer from
the Institut ffir Geologie und Dynamik der Lithosph~ire, Universit/it G6ttingen, Germany, for collecting the K/Ar data. We
would also like to thank S. Altaner, R. Hay, R. Hughes, and
R, Ylagan for their helpful comments and discussions concerning our manuscript as well as D. Pevear and R. Reynolds,
Jr, for their comments in their review of our manuscript.
REFERENCES
Austin GS, Glass HD, Hughes RE. 1989. Resolution of the
polytype structure of some illitic clay minerals that appear
to be 1Md. Clays Clay Miner 37:128-134.
Bailey SW. 1966. The status of clay mineral structures. Proceedings of the 14th National Conference on Clays and
Clay Minerals. New York: Pergamon Pr. p 1-23.
Bailey SW. 1980. Structures of layer silicates. In: Brindley
GW, Brown G, editors. Crystal structures of clay minerals
and their X-ray identification, Monograph No. 5. London:
Mineralogical Society. p 1-123.
Bailey SW, Frank-Kamenetskii VA, Goldstaub S, Kato A,
Pabst A, Schulz H, Taylor HFW, Fleischer M, Wilson AJM.
1977. International Union of Crystallography, report of the
International Mineralogical Association (IMA)-International Union of Crystallography (IUCr) Joint Committee
on Nomenclature. Acta Crystallogr Sect A 33:681-684.
Brown G, Brindley GW. 1980. X-ray diffraction procedures
for clay mineral identification. In: Brindley GW, Brown G,
editors. Crystal structures of clay minerals and their X-ray
identification. Monograph No. 5. London: Mineralogical
Society. p 305-359.
Caill~re S, Henin S, Rautureau M. 1982. Min6ralogie des
Argiles. Paris: Masson. 421 p.
Dalla Torte M, Stern WB, Frey M. 1994. Determination of
white K-mica polytype ratios: comparison of different XRD
methods. Clay Miner 29:717-726.
Drits VA, Planqon BA, Sakharov BA, Besson G, Tsipursky
SI, Tchoubar C. 1984. Diffraction effects calculated for
structural models of K-saturated montmorillonite containing different types of defects. Clay Miner 19:541-561.
Grathoff GH, Moore DM, Kluessendorf J, Mikulic DG. 1995.
The Waukesha illite, a Silurian residuum from karstification,
proposed as a candidate for the Source Clay Repository. Program with abstracts, 32nd annual Clay Minerals Society
Meeting; Baltimore, Maryland; 1995 June 3-8. p 54.
Guinier A, Bokij GB, Boll-Dornberger K, Cowley JM, Durovic S, Jagodzinski H, Krishna P, De Wolff PM, Zvyagin
BB, Cox DE, Goodman P, Hahn Th, Kuchitsu K, Abrahams
SC. 1984. Nomenclature of polytype structures: Report of
the International Union of Crystallography Ad-Hoc Committee on the nomenclature of disordered, modulated and
polytype structures. Acta Crystallogr Sect A 40:399-404.
Clays and Clay Minerals
Horton D. 1983. Argillic alteration associated with the amethyst vein system, Creede Mining District, Colorado [dissertation]. Champaign/Urbana, IL: Univ of Illinois. 337 p.
Hower J, Hurley PM, Pinson WH, Fairbairn HW. 1963. The
dependence of K-Ar age on the mineralogy of various particle size ranges in a shale. Geochim Cosmochim Acta 27:
405-410.
Hunziker JC, Frey M, Clauer N, Dallmeyer RD, Friedrichsen
H, Flehmig W, Hochstrasser K, Roggwiler P, Schwander H.
1986. The evolution of illite to muscovite: mineralogical
and isotopic data from the Glarus Alps, Switzerland. Contrib Mineral Petrol 92:157-180.
James RW. 1965. The optical principles of the diffraction of
X-rays. Vol. II of The crystalline state. Bragg, Sir L, editor.
Ithaca, NY: Cornell Univ Pr. 664 p.
Levinson AA. 1955. Studies in the mica group: polytypism
among illites and hydrous micas. Am Mineral 40:41-49.
Maxwell DT, Hower J. 1967. High-grade diagenesis and
low-grade metamorphism of illite in the Precambrian Belt
Series. Am Mineral 52:843-857.
Moore DM, Hower J. 1986. Ordered interstratification of dehydrated and hydrated Na-smectite. Clays Clay Miner 34:
379-384.
Oreskes N, Shrader-Fechette K, Belitz K. 1994. Verification,
validation, and confirmation of numerical models in the
earth sciences. Science 263:641-646.
Pevear DR. 1992. Illite age analysis, a new tool for basin
thermal history analysis. In: Kharaka YK, Maest AS, editors. Water-rock interaction. Rotterdam, The Netherlands:
AA Balkema. p 1251-1254.
Reynolds RC, Jr. 1963. Potassium~rubidium ratios and polytypism in illites and microclines from the clay size fractions of proterozoic carbonate rocks. Geochim Cosmochim
Acta 27:1097-1112.
Reynolds RC, Jr. 1985. NEWMOD 9 A computer program
for the calculation of one-dimensional diffraction patterns
of mixed-layered clays. Hanover, NH: RC Reynolds, Jr, 8
Brook Rd.
Reynolds RC, Jr. 1993. Three-dimensional X-ray powder diffraction from disordered illite: Simulation and interpretation of the diffraction patterns. In: Reynolds RC, Jr, Walker
JR, editors. Clay Minerals Society workshop lectures, Vol
5. Computer applications to X-ray powder diffraction analysis of clay minerals. Boulder, CO: Clay Minerals Society.
p 43-78.
Reynolds RC, Jr, Thomson CH. 1993. Illite from the Potsdam sandstone of New York: A probable noncentrosymetric mica structure. Clays Clay Miner 41:66-72.
Reynolds RC, Jr. 1994. WILDFIRE9 A computer program
for the calculation of three-dimensional X-ray diffraction
patterns for mica polytypes and their disordered variations.
Hanover, NH: RC Reynolds, Jr, 8 Brook Rd.
Smith JV, Yoder HS, Jr. 1956. Experimental and theoretical
studies of the mica polymorphs. Miner Mag 31:209-231.
Tettenhorst RT, Corbat6 CE. 1993. Quantitative analysis of
mixtures of 1M and 2M1 dioctahedral micas by X-ray diffraction. Clays Clay Miner 41:45-55.
Tsipursky SI, Drits VA. 1984. The distribution of octahedral
cations in the 2:1 layers of dioctahedral smectites studied
by oblique-texture electron diffraction. Clay Miner 9:177-193.
Velde B, Hower J. 1963. Petrological significance of illite
polymorphism in Paleozoic sedimentary rocks. Am Mineral
48:1239-1254.
Weaver CE, Broekstra BR. 1984. Illite-mica. In: Weaver CE
et al., editors. Shale slate metamorphism in the Southern
Appalachians. Developments in Petrology 10. Amsterdam:
Elsevier Science. p 67-97.
(Received 6 November 1995; accepted 29 February 1996;
Ms. 2709)