Clays and Clay Minerals, Vol.46. No. 5. 528-536, 1998.
U S E A N D LIMITATIONS OF S E C O N D - D E R I V A T I V E D I F F U S E
R E F L E C T A N C E S P E C T R O S C O P Y IN T H E V I S I B L E TO
N E A R - I N F R A R E D R A N G E TO I D E N T I F Y A N D
Q U A N T I F Y Fe O X I D E M I N E R A L S IN SOILS
A . C. SCHEINOST,]~ " A. CHAVERNAS, 2 V. BARRON 2 AND J. TORRENT 2
~Agronomy Department, Purdue University, West Lafayette, Indiana 47907, USA
2Departamento de Ciencias y Recursos Agrfcolas y Forestales, Universidad de C6rdoba,
Apdo. 3048, 14080 C6rdoba, Spain
Abstract--We measured the visible to near-infrared (IR) spectra of 176 synthetic and natural samples of
Fe oxides, oxyhydroxides and an oxyhydroxysulfate (here collectively called "Fe oxides"), and of 56
soil samples ranging widely in goethite/hematite and goethite/lepidocrocite ratios. The positions of the
second-derivative minima, corresponding to crystal-field bands, varied substantially within each group of
the Fe oxide minerals. Because of overlapping band positions, goethite, maghemite and schwertmannite
could not be discriminated. Using the positions of the 4Tl<----6AI, 4T2<----6AI, (4E;4AI)4---6AI and the electron
pair transition (4T~h-4Ti)<----(6Ai q-6ml), at least 80% of the pure akaganeite, feroxyhite, ferrihydrite, hematite
and lepidocrocite samples could be correctly classified by discriminant functions. In soils containing
mixtures of Fe oxides, however, only hematite and magnetite could be unequivocally discriminated from
other Fe oxides. The characteristic features of hematite are the lower wavelengths of the 4"171transition
(848-906 nm) and the higher wavelengths of the electron pair transition (521-565 nm) as compared to
the other Fe oxides (909-1022 nm and 479-499 nm, resp.). Magnetite could be identified by a unique
band at 1500 nm due to Fe(II) to Fe(III) intervalence charge transfer. As the bands of goethite and hematite
are well separated, the goethite/hematite ratio of soils not containing other Fe oxides could be reasonably
predicted from the amplitude of the second-derivative bands. The detection limit of these 2 minerals in
soils was below 5 g kg t, which is about 1 order of magnitude lower than the detection limit for routine
X-ray diffraction (XRD) analysis. This low detection limit, and the little time and effort involved in the
measurements, make second-derivative diffuse reflectance spectroscopy a practical means of routinely
determining goethite and hematite contents in soils. The identification of other accessory Fe oxide minerals in soils is, however, very restricted.
Key Words---Crystal Field Bands, Diffuse Reflectance Spectroscopy, Goethite, Hematite, Intervalence
Charge Transfer, Iron Oxides, Magnetite.
INTRODUCTION
D i f f u s e r e f l e c t a n c e s p e c t r o s c o p y in t h e v i s i b l e
( 4 0 0 - 7 0 0 n m ) a n d e x t e n d e d visible r a n g e ( 4 0 0 - 1 2 0 0
n m ) h a s b e e n used as a n a n c i l l a r y m e t h o d to identify
a n d s e m i q u a n t i t a t i v e l y e s t i m a t e F e oxides in clays,
soils a n d s e d i m e n t s ( C o r n e l l a n d S c h w e r t m a n n 1996,
a n d r e f e r e n c e s therein). A l t h o u g h spectra o f s a m p l e s
with unknown mineralogy may simply be compared
w i t h spectra o f r e f e r e n c e m i n e r a l s ( S i n g e r 1982; Morris et al. 1989), the p a r a m e t r i z a t i o n o f t h e s e spectra
a l l o w s for a q u a n t i t a t i v e a p p r o a c h (Torrent a n d B a r r 6 n
1993). T h e m o s t f r e q u e n t l y u s e d p r o c e d u r e s for the
p a r a m e t r i z a t i o n are:
1) C a l c u l a t i o n o f the tristimulus values (X, Y, Z) o f
the C I E c o l o r s y s t e m ( C I E 1978). T h e tristimulus values are c o m p u t e d f r o m the spectral reflectance a n d energy o f the light source for e a c h w a v e l e n g t h , a n d c a n
b e c o n v e r t e d , b y m e a n s o f a p p r o p r i a t e f o r m u l a s or
graphs, to the p a r a m e t e r s o f o t h e r color systems, such
I Present address: Department of Plant and Soil Science,
University of Delaware, Newark, Delaware 19717-1303
USA.
Copyright 9 1998, The Clay Minerals Society
as M u n s e l l or L * a * b * ( W y s z e c k i a n d Stiles 1982).
Several indices b a s e d o n s u c h p a r a m e t e r s h a v e b e e n
correlated, for instance, with the h e m a t i t e content, prov i d i n g a s e m i q u a n t i t a t i v e e s t i m a t i o n o f this m i n e r a l in
soils (Torrent et al. 1980; B a r r 6 n a n d Torrent 1986;
N a g a n o et al. 1992, 1994).
2) A p p l i c a t i o n o f the K u b e l k a - M u n k theory. W i t h
this m e t h o d , the reflectance values, R, are u s e d to obtain the r e m i s s i o n f u n c t i o n F ( R ) = (1 - R ) V ( 2 R ) ,
w h i c h allows, in turn, c a l c u l a t i o n o f the coefficients
o f a b s o r p t i o n (K) a n d scattering (S) c o r r e s p o n d i n g to
the 2 p h e n o m e n a i n v o l v e d in the s a m p l e - l i g h t interaction process. It h a s b e e n s h o w n that K a n d S are
additive in a m i x t u r e o f p i g m e n t s . Therefore, o n e c a n
e s t i m a t e the c o n c e n t r a t i o n o f an F e o x i d e in the m i x ture b y k n o w i n g the color characteristics o f e a c h o f
the c o m p o n e n t s o f the mixture, a n d a s s u m i n g a stand a r ar color for the Fe oxides i n v e s t i g a t e d ( B a r r 6 n a n d
Torrent 1986; Fernfmdez a n d S c h u l z e 1992).
3) C a l c u l a t i o n o f d e r i v a t i v e s o f the reflectance or
the r e m i s s i o n f u n c t i o n to derive the p o s i t i o n a n d intensity o f the a b s o r p t i o n b a n d s . T h e s e b a n d s are prod u c e d b y crystal-field transitions o f Fe(III) in an oc528
Vol. 46, No. 5, 1998
Use and limitations of 2nd-derivative diffuse reflectance spectroscopy
529
Table 1. References, origin (syn = synthetic, nat = natural), number (N) and A1 substitution (Ai-subst.) of the Fe oxide
samples.
A|-subst.
(mole-fraction)
Fe oxide (abbreviation)
Origin (N)
Akaganeite (Akag)
Feroxyhyte (Feroxy)
Ferrihydrite
(Ferrih)
syn (7)
syn (5)
syn (11)
nat (11)
0
0
0
Goethite (Goeth)
syn (25)
nat (36)
syn (26)
nat (20)
0-0.32
Lepidocrocite (Lepido)
syn (14)
nat (3)
0-0.14
Maghemite (Magh)
Magnetite (Magn)
Schwertmannite (Schwt)
nat (3)
syn (5)
syn (4)
nat (6)
0-0.7
0-0.17
0
Hematite (Hem)
0-0.16
tahedral ligand field (Sherman and Waite 1985). Due
to the combination of several polarization planes and
particle-dependent scattering, the diffuse reflectance
spectra of powders show broad, strongly overlapping
bands (Kortiim 1969; Morris et al. 1982). To determine the positions of these bands, the resolution may
be mathematically enhanced by calculating the derivatives of the measured spectra (Huguenin and Jones
1986). Deaton and Balsan (1991) used the first derivative, and Kosmas et al. (1984, 1986) and Malengreau
et al. (1994, 1996) used the second derivative to detect
and quantify different Fe oxides in soils or mineral
mixtures. In a similar approach, Coyne et al. (1990)
found that second-derivative spectra were superior to
first derivative or untransformed spectra in predicting
the amount of structural or interlayer Fe(HI) in montmorillonite.
Recent studies have suggested very low detection
limits for Fe oxides and the possibility of unequivocally differentiating between the various Fe oxides
based on crystal field band positions (Malengreau et
al. 1994, 1996). Since these studies were carried out
for relatively simple mineral mixtures and with a limited number of Fe oxide samples, a closer examination
of the potential of second-derivative reflectance spectroscopy seemed necessary. Thus, the objective of this
work was to examine the ability of the second-derivative spectra to: 1) discriminate among different Fe
oxides, and 2) help semiquantitatively estimate Fe oxide contents in soils, using a substantially larger collection of samples than those used in previous studies.
We restricted this study to the evaluation of crystal
field bands. The OH and H20 vibrational bands in the
IR constitute, however, another possibility for characterizing Fe oxides (Bishop et al. 1995; Bishop and
Murad 1996).
References
Bigham et al. (1990, 1996); Schwertmann et al. (1995)
Carlson and Schwertmann (1980)
Schwertmann and Fischer (1973); Schwertmann et al. (1982);
Schwertmann and Kiimpf (1983); Carlson and Schwertmann (1981, 1987)
Schulze and Schwertmann (1984); Schwertmann et al.
(1985)
Stanjek (1991); Stanjek and Schwertmann (1992); Murad
and Schwertmann (1986); Torrent and Schwertmann
(1987)
Schwertmann and Fitzpatrick (1977); Carlson and Schwertmann (1990); Scbwertmann and Wolska (1990); Murad
and Schwertmann (1984)
Taylor and Schwertmann (1974)
Schwertmann and Murad (1990)
Bigham et al. (1990, 1996); Schwertmann et al. (1995)
MATERIALS A N D M E T H O D S
Synthetic and Natural Fe Oxides, and Soils
The origins of the Fe oxide samples are given in
Table 1. The samples from natural environments include precipitates in spring waters, lake ores, concretions, NaOH-enriched soil fractions and whole soils.
Forty soil samples with variable hematite-to-goethite
ratios come from A or B horizons of soils from several
countries of the European Union and Brazil, including
Vertisols, Luvisols, Nitosols, Acrisols and Ferralsols.
Determination of citrate/bicarbonate/dithionite-extractable iron (Fed) (Mehra and Jackson 1960), and acid
oxalate-extractable iron (Feo) (Schwertmann 1964) and
differential X RD (Schulze 1981) were carried out to
estimate their hematite and goethite contents, which
ranged between 0 and 150 g kg -~. Sixteen samples
with different lepidocrocite-to-goethite ratios are from
crusts, pipestems and other Fe-oxide-enriched soil matrices from South Africa (Schwertmann and Fitzpatrick 1977).
Spectral Measurements
Diffuse reflectance spectra were obtained from dry
and ground powders. For the hematitic soils, spectra
were taken from 380 to 750 nm at 1-nm increments,
using a Perkin-Elmer Lambda 3 spectrophotometer fitted with a reflectance attachment (Perkin Elmer, Norwalk, Connecticut, USA). BaSO4 was used as a white
standard. All other samples were measured between
200 and 2500 nm at 1-nm increments with a 20-cm
integrating sphere (Labsphere, Inc., USA) attached to
a Perkin-Elmer Lambda 19 (f3berlingen, Germany). A
Spectralon standard (Labsphere, Inc., USA) was used
as reflectance reference. The wavelength-dependent
reflectance functions were transformed into remission
530
Scheinost, Chavernas, Barr6n and Torrent
Clays and Clay Minerals
Electronic Transition
Wavelength / nrn
1100
1000
900
.
. . . . . . .
9 . . . . . . .
9 . . . . . . .
.
.
.
.
.i . . . . . . .
. i . . _ i . . J
. . . . . . .
,a . . . . . . .
.i . . . . . . .
.
800
i
700 . . . . . . . , . . = . . ~ . . , ~ . . ,
....... , .....
=
600
Z
t
~. . . . . . . , . . .
:: = :: .[ ..... I
i
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i
9 . . . . . . .
,
- . . . . . . . .
!. . . . . . . .
. . . . . . . . . . . . . -~zi"":" "",~,'" ".""",~.""" EPT
500
"------,------,--.--,--,
400
_.,...i...=....,...I...:....z. ..... _z... ._.~...
300
I
.
.
.
~
.
.
.
.
.
q.
.
.
.
.
.
.
q
4-". 4 A
6A
i
"0
~,
t-
9
O
w.-
E
s
o
=
~
o
E
--
9
m
E
9
f-
~o
Figure 1.
Median and range of the crystal field band positions determined by second-derivative minima.
functions. The first and the second derivatives were
calculated using a cubic spline fitting procedure (Press
et al. 1992). Briefly, a cubic spline consists of a number o f cubic polynomial segments j o i n e d end to end
with continuity in first and second derivatives in the
joints. A crucial input parameter in the program is the
n u m b e r of points taken for each segment, which determines the goodness of fitting (r 2) and the degree of
smoothing of the curve. Several tests showed 15 to be
the o p t i m u m number of points to obtain smooth curves
with high r 2 values; therefore, this number was adopted for the systematic calculation of the second-derivative curves. The selected smoothing parameters are
not valid in general, but must be adjusted when the
spectra are taken with different scan parameters.
RESULTS AND DISCUSSION
Discrimination between Single-Phase Fe Oxides
The second-derivative spectra revealed 4 bands in
the range from 350 to 1050 nm. These bands were
assigned, according to Sherman and Waite (1985), to
the 3 single-electron transitions 4Tle---6AI, 4T2e--6A I and
(4E;aA1)e'6A
l, and the e l e c t r o n pair transition
(41", +4T~)e--(6A~ +6A~), in the following abbreviated by
" E P T " (Figure 1).
The EPT, as the most intense absorption band, determines the position of the absorption edge which, in
turn, is closely related to the hue. The EPTs for the
hematites occur at 521 to 565 nm, clearly separated
from the EPTs of the other, m o r e yellowish Fe oxides
(479 to 499 nm) (Table 2). The difference in E P T and
color between hematite and the other Fe oxides has
been explained by the presence o f face-sharing octahedra in the hematite structure leading to an increase
of the magnetic coupling between neighboring Fe(III)
centers (Sherman and Waite 1985). In contrast to the
results of Malengreau et al. (1994), however, the ranges of the E P T of all other Fe oxides generally overlapped, that is, the discrimination o f these Fe oxides
could not be performed with the E P T only.
A similar separation of hematite f r o m the other Fe
oxides is given by the 4Tle-'-6A I transition (Figure 1).
Vol. 46, No. 5, 1998
531
Use and limitations of 2nd-derivative diffuse reflectance spectroscopy
Table 2. Median and range of crystal field band positions (nm).
4T l
Akaganeite
(N = 7)
Feroxyhyte
(N = 5)
Ferrihydrite
(N = 22)
Goethite
(N = 61)
Hematite
(N = 46)
Lepidocrocite
(N = 17)
Maghemite
(N = 3)
Schwertmannite
(N = 10)
946
932-975
933
926-940
972
947-990
953
929-1022
877
848-906
973
943-997
948
909-963
952
942-962
4T2
EPT
(4E; 4Al)
659
652-671
707
700-710
7161"
698-734
665
654-691
682
664-695
687
673-692
675
648-675
655
648-688
492
490-493
493
492-495
492
484-499
488
479-493
531
521-565
488
485-490
489
486-490
489
488-491
410
408-411
417
413-421
410t
386-410
413
409-427
423t
416-442
413
408-418
418
417-423
416~410-416
t Band not always determinable (no minimum of the second-derivative spectrum).
T h e significantly l o w e r w a v e l e n g t h o f this transition
for h e m a t i t e ( 8 4 8 - 9 0 6 n m ) is the result o f the l o w e r
crystal field splitting e n e r g y o f h e m a t i t e (14,000 c m t)
as c o m p a r e d to that o f g o e t h i t e (15,320 c m t), m a g h e m i t e (15,410 c m -1) and l e p i d o c r o c i t e (15,950 c m 1)
(Burns 1993). Again, strongly o v e r l a p p i n g b a n d positions p r e v e n t a p r o p e r d i s c r i m i n a t i o n o f the n o n - h e matite F e o x i d e s b y only this band.
With the 4T2~---6A1 transition at 698 to 734 nm, the
ferrihydrite and f e r o x y h i t e s a m p l e s are distinct f r o m
the o t h e r F e oxides, w h e r e this transition occurs at
l o w e r w a v e l e n g t h s (648 to 695 n m ) (Figure 1, Table
2). Sixteen out o f 22 ferrihydrite s a m p l e s did not
s h o w , however, a s e c o n d - d e r i v a t i v e m i n i m u m , so that
the b a n d position could not b e determined. The
(4E;4A1) <'--6A 1 transition o f all F e oxides ranged f r o m
386 to 442 nm. This b a n d was not visible in the derivative spectra o f 25 hematites.
To evaluate the d i s c r i m i n a t i v e p o w e r o f the positions o f all 4 b a n d s c o m b i n e d , a s t e p w i s e d i s c r i m i n a n t
analysis was p e r f o r m e d (StatSoft Inc. 1996). T h e a
priori probabilities w e r e a s s u m e d to b e equal for all
F e o x i d e minerals, in o r d e r to avoid the p r e d o m i n a n t
classification o f minerals with the larger n u m b e r s o f
s a m p l e s which, in turn, w o u l d d e c r e a s e the sensitivity
o f classification for the small mineral groups. To allow
for the inclusion o f all samples, w e split the groups o f
b o t h ferrihydrite and h e m a t i t e into s u b g r o u p s containing all 4 bands, and into subgroups w h e r e 1 b a n d w a s
m i s s i n g and set to zero. Table 3 s h o w s that all feroxyhite, ferrihydrite and h e m a t i t e s a m p l e s w e r e correctly
classified b y using the 4 bands, and m o r e than 80% o f
the akaganeite and l e p i d o c r o c i t e s a m p l e s w e r e correctly discriminated. The b a n d positions o f goethite,
m a g h e m i t e and s c h w e r t m a n n i t e are, however, too s i m ilar to allow for a reliable d i s c r i m i n a t i o n (Table 3).
Table 3. Classification of Fe oxides by 4T~, 4T2, EPT and (4E; 4Al). Rows: observed classifications; columns: predicted
classifications.
Akaganeite
Feroxyhite
Ferrihdyrite -4T2-~
Ferrihdyrite +4T2-~
Goethite
Hematite +(4E;4A07
Hematite -(4E;aA~)t
Lepidocrocite
Maghemite
Schwertmannite
Total
Percent
correct
Akag
Feroxy
Ferrih
- 4T2
Ferrih
+ 4T2
Goeth
Hem
+ (E;A)
Hem
- (E;A)
Lepido
Magh
Schwt
83
100
100
100
23
100
100
81
33
50
66
5
0
0
0
8
0
0
0
0
1
14
0
5
0
0
0
0
0
0
0
0
5
0
0
16
0
0
0
0
0
0
0
16
0
0
0
6
0
0
0
0
0
0
6
0
0
0
0
14
0
0
0
0
2
16
0
0
0
0
0
21
0
0
0
0
21
0
0
0
0
0
0
22
0
0
0
22
0
0
0
0
3
0
0
13
1
1
18
1
0
0
0
13
0
0
3
1
0
18
0
0
0
0
22
0
0
0
1
4
27
t Ferrihydrite -4T2: ferrihydrite samples without the 4T2 band, Ferrihydrite +4T2: ferrihydrite samples with the 4T2 band.
Hematites resp. with and without the (4E;aA1) band.
532
Clays and Clay Minerals
Scheinost, Chavernas, Barr6n and Torrent
Table 4. Coefficients of the classification function (see text).
a~
bx
c,
d~
e,
Akag
Feroxy
Ferrih +4T2
Goeth
Hem +(E;A)
Lepid
Magh
Schwt
14.92
26.21
0.26
8.62
-15,504
15.80
26.43
0.25
8.57
-16,158
16.02
26.53
0.22
8.86
-16,640
14.98
26.09
0.26
8.67
-15,527
15.43
27.61
0.26
8.33
-16,295
15.36
26.24
0.26
8.78
-15,961
15.09
26.09
0.26
8.59
-15,529
14.92
26.16
0.26
8.68
-15,535
A set of classification functions to determine the Fe
oxide mineralogy of further samples is given by:
Sx = ax. 4T2 + b x . E P T + c . , . ( 4 E ; 4A1) + dx. aTl -b e~
[1]
with the scores for each group o f minerals, Sx, and the
coefficients for each group, ax, b~, cx, d~ and ex as
defined in Table 4. Within the limitations shown before, the highest number of Sx gives the Fe oxide mineral o f the e x a m i n e d sample. As second-derivative
spectra without a m i n i m u m in the 4T2<---6A 1 range unequivocally indicated ferribydrite, and those without a
m i n i m u m in the ( h E ; h A 1)<---6A 1 range unequivocally indicated hematite, no classification functions are necessary for these subgroups. As the band positions derived from the second derivatives depend on the x and
y units of the spectra, and on the smoothing of the
spectra, it is indispensable to follow the procedure given under Materials and Methods before calculating the
scores.
Identification of Mixtures o f Fe Oxides
Because o f the well-separated EPTs of goethite and
hematite, both phases are clearly detectable from the
derivative spectra o f soils (Figure 2). The intensity of
the m i n i m a depends on the goethite-to-hematite ratio
(see below).
Although lepidocrocite and goethite could be discriminated in pure phases (Table 2), the derivative
spectra of soils containing both Fe oxides did not resolve the slightly different band positions (Figure 3).
Using the classification functions given in Table 4,
soils with l e p i d o c r o c i t e / ( l e p i d o c r o c i t e + g o e t h i t e )
~ 0 . 3 7 were classified as lepidocrocites, while those
with a ratio -<0.37 were classified as goethite or maghemite. In the same way, the derivative spectra of ferrihydrite/goethite
and
ferrihydrite/lepidocrocite
mixtures did not resolve the bands of both phases (data
not shown). That is, second-derivative spectroscopy is
not suited to detect small amounts o f ferrihydrite or
lepidocrocite in addition to a major goethite p h a s e - - a
situation which should frequently occur in soils due to
t h e r m o d y n a m i c inequilibria (Cornell and Schwertmann 1996). A l t h o u g h not tested experimentally, it
may be concluded that spectra o f other combinations
of Fe oxides with band positions in the same or strongly overlapping regions w o u l d not allow the single contributions to be resolved, either.
Besides the band wavelengths, absolute and relative
band intensities and bandwidths m a y be used to dis2e-6
4e-6
lepid/(lepid+goeth)
~ 1.00
~
I.~
UJ
-->
2e-6 / ~
le-6
hematite/(hematite+oo~
~
.~
,~
A
ll~
gt
m----
~
0.72
0.37
0.00
Y0,,j
o"'
~-
o.1
-2e-6
-4e-6
0,0
goethite EPThematite
450
500
550
-2e-6
. . . . . . . . . . . . . . . . .
600
WAVELENGTH / nm
Figure 2. Second-derivative spectra of soils with various hematite/(hernatite + goethite) ratios.
400
500
600
700
WAVELENGTH
800
900
I000
[nm]
Figure 3. Second-derivative spectra of soils with various
lepidocrocite/(lepidocrocite + goethite) ratios.
Vol. 46, No. 5, 1998
Use and limitations of 2nd-derivative diffuse reflectance spectroscopy
533
100
10
8oll w/o mtgnetlte
m-
80
IJJ
O
z
60
0
UJ
ua
>
5
-->
tr
lit
CI
0
Y
--J
u.
w
4o
I1:
A
t r
-10
20
9
.
,
400
.
.
.
.
,
500
.
.
.
.
,
.
.
.
.
600
,
.
.
.
700
WAVELENGTH (nm)
500
1000
1500
2000
WAVELENGTH / nm
Figure 4. Reflectance spectra of pure magnetite, magnetite
+ maghemite and soils with various concentrations of magnetite.
'l/,+~/
10
a
>o
-10
40O
420
2:,2
44O
460
480
5O0
WAVELENGTH (rim)
b
~,,
5
I%
z.o
-10
540
560
580
600
820
640
WAVELENGTH (nm)
Figure 5. Second-derivative spectra of the remission function f(R) in selected spectrum bands for mixtures of a deferrated soil material with different quantities of (a) goethite and
(b) hematite.
Figure 6. Position and amplitude of the bands selected for
the quantification of goethite (YI) and hematite (Y2) in soil
samples.
criminate Fe oxides. Although such differences were
evident with synthetic, single-phase samples, for example, sharper bands for goethite than for ferrihydrite,
these differences were generally less expressed for natural Fe oxides, and could not be used to differentiate
the origin of overlapping bands in mixtures and soils.
Band amplitudes of the second-derivative spectra were
higher for hematite, but no significant differences were
found between the amplitudes of the other Fe oxides.
The reflectance of the magnetite samples was very
low and, because of that, the spectra were too noisy
to allow for the determination of their crystal field
bands. This low reflectivity is caused by a strong absorption band which has its maximum in the near-IR
at about 1500 nm, but extends across the visible range
(Figure 4). It is caused by the presence of Fe(II) and
Fe(III) in neighboring octahedra in the magnetite
structure, leading to the delocalization of electrons due
to intervalence charge transfer (IVCT) (Sherman and
Waite 1985; Burns 1981). Strens and Wood (1979) determined the position of the IVTC of magnetite from
diffuse reflectance spectra at 1400 nm, and Sherman
(1987) derived an energy of 1475 nm from molecular
orbital calculations assuming equal sites for Fe(II) and
Fe(III) (q = 0). Both values correspond with ours, considering the low precision with which the position of
such broad bands could be determined. The shape of
this IVCT band is more pronounced in samples of
maghemite-magnetite mixtures, or in soils containing
magnetite, because of the generally higher reflectivity
of these samples (Figure 4). Even trace amounts of
magnetite are detectable.
Quantification of Hematite and Goethite in Soils
The well-separated EPTs of goethite and hematite
allow one to identify very small amounts of goethite
and hematite. Figure 5 shows selected parts of the sec-
534
Scheinost, Chavernas, Barr6n and Torrent
,,,~
1
y = -0.068 + 1.325 x
I.~-r
=
Clays and Clay Minerals
goethite) ratio, b y a s s i g n i n g the d i f f e r e n c e b e t w e e n cit r a t e / b i c a r b o n a t e / d i t h i o n i t e - e x t r a c t a b l e Fe a n d oxalatee x t r a c t a b l e Fe to h e m a t i t e a n d goethite.
9 J
9
0
0.8
CONCLUSIONS
§
0.6
tu
"r" 0.4
0.2
"I"
0
0
0.2
'
0,4
0
.6
0.8
Y~(Y2+Y1)
Figure 7. The relationship between the hematite/(hematite
+ goethite) ratio as determined by differential XRD, and the
112/(I"1 + Y2) ratio as determined by diffuse reflectance spectroscopy (see Figure 6 for explanation of Yj and 112).
o n d - d e r i v a t i v e c u r v e o f the reflectivity f u n c t i o n for
m i x t u r e s o f a d e f e r r a t e d soil m a t r i x (treated with cit r a t e / b i c a r b o n a t e / d i t h i o n i t e to r e m o v e Fe oxides) a n d
s y n t h e t i c h e m a t i t e a n d goethite; c o n c e n t r a t i o n s as low
as 0.5 g k g - t p r o d u c e a significant c h a n g e in the curve.
In Vertisol s a m p l e s c o n t a i n i n g only goethite as F e oxide, the d e t e c t i o n limit for this m i n e r a l was a b o u t 3 g
k g -~ (results not s h o w n ) . In s u m m a r y , the detection
limit s e e m s to b e at least 1 o r d e r o f m a g n i t u d e smaller
t h a n that o f o r d i n a r y X R D .
To o b t a i n an i n d e x o f the b a n d intensity in the seco n d - d e r i v a t i v e c u r v e s for goethite a n d hematite, we
u s e d the a m p l i t u d e b e t w e e n the - 4 1 5 - n m m i n i m u m
((4E;aAI)~---6At) a n d the - - 4 4 5 - n m m a x i m u m for goethite ( d e n o t e d as Y1), a n d b e t w e e n the - - 5 3 5 - n m m i n i m u m (EPT) a n d the - 5 8 0 - n m m a x i m u m for h e m a t i t e
( d e n o t e d as I12) (Figure 6). F o r the 4 0 soils used, w h i c h
g e n e r a l l y c o n t a i n e d less t h a n 150 g k g -~ h e m a t i t e a n d
goethite, a h i g h l y significant c o r r e l a t i o n was f o u n d bet w e e n g o e t h i t e c o n t e n t a n d Y~ ( G o e t h i t e [g kg -~] =
- 0 . 0 6 + 268111; r 2 = 0.86; P < 0.001), a n d h e m a t i t e
c o n t e n t a n d I12 ( H e m a t i t e [g k g ~] = - 0 . 0 9 + 402Y2;
r: = 0.85; P < 0.001). M u l t i p l e r e g r e s s i o n u s i n g b o t h
a m p l i t u d e s did not i m p r o v e the c o r r e l a t i o n coefficient.
F o r the s a m e soils, the r e d n e s s rating, (ROE), p r o p o s e d
b y B a r r 6 n a n d Torrent (1986) p r o v i d e d a h i g h e r corr e l a t i o n with the h e m a t i t e c o n t e n t (r 2 = 0.94). H o w ever, the p r o c e d u r e s b a s e d on the s e c o n d - d e r i v a t i v e
c u r v e h a v e the a d v a n t a g e o v e r r e d n e s s ratings in that
they help s i m u l t a n e o u s l y predict h e m a t i t e a n d goethite
contents. In practice, o n e c a n use the r e l a t i o n s h i p bet w e e n the h e m a t i t e / ( h e m a t i t e + goethite) ratio (derived
f r o m X - r a y or differential X R D m e a s u r e m e n t s ) and
the Y2/(Y~ + }'2) a m p l i t u d e ratio for that p u r p o s e (Figure 7). T h e a b s o l u t e quantities o f h e m a t i t e a n d g o e t h i t e
are c a l c u l a t e d f r o m the p r e d i c t e d h e m a t i t e / ( h e m a t i t e +
M o d e r n s p e c t r o p h o t o m e t e r s allow the fast acquisition o f reflectance spectra (typically less than 10 min,
only a few s e c o n d s with n e w diode array detectors),
and calculations of derivatives and band amplitudes
c a n b e d o n e easily w i t h the a p p r o p r i a t e software. In
addition, s a m p l e p r e p a r a t i o n for s p e c t r o p h o t o m e t r i c
m e a s u r e m e n t s is fast ( < 5 m i n ) a n d n o n d e s t r u c t i v e .
T h e s e a d v a n t a g e s m a k e s e c o n d - d e r i v a t i v e diffuse reflectance s p e c t r o s c o p y well suited for routine meas u r e m e n t s o f goethite a n d h e m a t i t e in soils a n d m i n eral mixtures. W i t h the e x c e p t i o n o f h e m a t i t e a n d
magnetite, however, the u n e q u i v o c a l d i s c r i m i n a t i o n o f
Fe o x i d e m i n e r a l s c a n n o t b e p e r f o r m e d b y u s i n g the
positions o f the crystal field b a n d s only. A b e t t e r result
m a y b e a c h i e v e d b y i n c l u d i n g the v i b r a t i o n a l b a n d s
into the d i s c r i m i n a n t analysis ( C o y n e et al. 1990; B i s h op et al. 1995; B i s h o p a n d M u r a d 1996).
ACKNOWLEDGMENTS
ACS acknowledges the research grant by the Deutsche Forschungsgemeinschaft, and thanks D. G. Schulze (Purdue University) who gave the visiting scientist a warm welcome and
provided the equipment, R. V. Morris (NASA, Houston, Texas) for his introduction into the secrets of diffuse reflectance
spectroscopy and U. Schwertmann (Technische Universit~it
Mtinchen, Germany) who generously provided his large collection of Fe oxides.
REFERENCES
Barr6n V, Torrent J. 1986. Use of the Kubelka-Munk theory
to study the influence of iron oxides on soil colour. J Soil
Sci 37:499-510.
Bigham JM, Schwertmann U, Carlson L, Murad E. 1990. A
poorly crystallized oxyhydroxysulfate of iron formed by
bacterial oxidation of Fe(II) in acid mine waters. Geochim
Cosmochim Acta 54:2743-2758.
Bigham JM, Schwertmann U, Pfab G. 1996. Influence of pH
on mineral speciation in a bioreactor simulating acid mine
drainage. Appl Geochem 11:845-852.
Bishop JL, Murad E. 1996. Schwertmannite on Mars? Spectroscopic analyses of schwertmannite, its relationship to
other ferric minerals, and its possible presence in the surface material on Mars. In: Dyar MD, McCammon C,
Schaefer MW, editors. Mineral spectroscopy: A tribute to
Roger G. Burns. Spec Publ: Geochem Soc.
Bishop JL, Pieters CM, Burns RG," Edwards JO, Mancinelli
RL, Froeschl H. 1995. Reflectance spectroscopy of ferric
sulfate-bearing montmorillonites as Mars soil analog materials. Icarus 117:101-119.
Burns RG. 1981. Intervalence transitions in mixed-valence
minerals of iron and titanium. Ann Rev Earth Planet Sci 9:
345-383.
Burns RG. 1993. Mineralogical applications of crystal field
theory. Cambridge topics in mineral physics and chemistry,
5. Cambridge Univ Pr. 551 p.
Carlson L, Schwertmann U. 1980. Natural occurrence of feroxyhite (~-FeOOH). Clays Clay Miner 28:272-280.
Vol. 46, No. 5, 1998
U s e and limitations of 2nd-derivative diffuse reflectance spectroscopy
Carlson L, S c h w e r t m a n n U. 1981. Natural ferrihydrites in
surface deposits from Finland and their association with
silica. G e o c h i m C o s m o c h i m Acta 45:421-429.
Carlson L, S c h w e r t m a n n U. 1987. Iron and m a n g a n e s e oxides
in Finnish ground water treatment plants. Wat Res 2 1 : 1 6 5 170.
Carlson L, S c h w e r t m a n n U. 1990. The effect of C O 2 and
oxidation rate on the formation of goethite versus lepidocrocite from an Fe(II) s y s t e m at p H 6 and 7. Clay Miner
25:65-71.
CIE. 1978. R e c o m m e n d a t i o n s on uniform color spaces, color
difference and psychometric color terms. Colorimetry,
Suppl. 2 to Publ 15. Paris: CIE.
Cornell RM, S c h w e r t m a n n U. 1996. The iron oxides: Structure, properties, reactions, occurrence and uses. Weinheim:
V C H Verlagsgesellschaft. 573 p.
C o y n e LM, Bishop JL, Scattergood T, Banin A, Carle G,
Orenberg J. 1990. Near-infrared correlation spectroscopy:
Quantifying iron and surface water in a series of variably
cation-exchanged montmorillonite clays. In: Coyne LM,
M c K e e v e r W S , Blake DE editors. Spectroscopic characterization of minerals and their surfaces. A m C h e m Soc. p
407-429.
Deaton BC, Balsam W L . 1991. Visible spectroscopy. A rapid
method for determining hematite and goethite concentration in geological materials. J Sediment Petrol 61:628-632.
Fern~indez RN, Schulze DG. 1992. Munsell colors o f soils
simulated by mixtures of goethite and hematite with kaolinite. Z Pflanzenernahr Bodenk 155:473-478.
H u g u e n i n RL, Jones JL. 1986. Intelligent information extraction from reflectance spectra: Absorption band positions. J
Geophys Res 91:9585-9598.
Korttim G. 1969. Reflectance spectroscopy. N e w York:
Springer-Verlag. 366 p.
K o s m a s CS, Curl N, Bryant RB, Franzmeier DP. 1984. Characterization of iron oxide minerals by second-derivative
visible spectroscopy. Soil Sci Soc A m J 48:401-405.
K o s m a s CS, Franzmeier DE Schulze DG. 1986. Relationship
a m o n g derivative spectroscopy, color, crystallite dimensions, and Al substitution of synthetic goethites and hematites. Clays Clay Miner 34:625-634.
Malengreau N, Bedidi A, Muller J-P, Herbillon AJ. 1996.
Spectroscopic control of iron oxide dissolution in two ferralitic soils. Eur J Soil Sci 47:13-20.
Malengreau N, Muller J-P, Calas G. 1994. Fe-speciation in
kaolins: A diffuse reflectance study. Clays Clay Miner 42:
137-147.
M e h r a OP, Jackson ML. 1960. Iron oxide removal from soils
and clays by a dithionite-citrate system buffered with sod i u m bicarbonate. Clays Clay Miner 7:317-327.
Morris RV, Agresti DG, Lauer HV, N e w c o m b JA, Shelfer TD,
Murali AV. 1989. Evidence for pigmentary hematite on
Mars based on optical, magnetic, and Mtissbauer studies of
superparamagnetic (nanocrystalline) hematite. J Geophys
Res 94:2760-2778.
Morris RV, Neely SC, Mendell WW. 1982. Application of
K u b e l k a - M u n k theory of diffuse reflectance to geologic
problems: T h e role of scattering. G e o p h y s Res Lett 9 : 1 1 3 116.
Murad E, S c h w e r t m a n n U. 1984. The influence o f crystallinity on the Mtissbauer spectrum of lepidocrocite. Mineral
M a g 48:507-511.
M u r a d E, Schwertmann U. 1986. Influence of A1 substitution
and crystal size on the room-temperature M 6 s s b a u e r spectrum of hematite. Clays Clay Miner 34:1-6.
Nagano T, N a k a s h i m a S, N a k a y a m a S, O s a d a K, Senoo M.
1992. Color variations associated with rapid formation of
goethite from proto-ferrihydrite at pH i 3 and 40 ~ C. Clays
Clay M i n e r 4 0 : 6 0 0 - 6 0 7 .
535
Nagano T, N a k a s h i m a S, N a k a y a m a S, Senoo M. 1994. The
use of color to quantify the effects of pH and temperature
on the crystallization kinetics o f goethite under highly alkaline conditions. Clays Clay M i n e r 42:226-234.
Press W H , Teukolsky SA, Vetterling WT, Flannery BE 1992.
Numerical recipes in Fortran. The art of scientific computing. Cambridge Univ Pr. 963 p.
Schulze DG. 1981. Identification of soil iron oxide minerals
by differential X-ray diffraction. Soil Sci Soc A m J 45:
437-440.
Schuize DG, S c h w e r t m a n n U. 1984. The influence of alum i n i u m on iron oxides: X. Properties of Al-substituted goethites. Clay Miner 19:521-529.
Schwertmann U. 1964. Differenzierung der Eisenoxide des
Bodens durch Extraktion mit A m m o n i u m o x a l a t - L 6 s u n g .
Zeitschrift fiir Pflanzenern~thr D 6 g u n g and B o d e n k u n d e
105:194-202.
Schwertmann U, B i g h a m JM, Murad E. 1995. The first occurence o f Schwertmannite in a natural stream environment. Eur J Mineral 7:547-552.
S c h w e r t m a n n U, Cambier P, M u r a d E. 1985. Properties o f
goethites of varying crystallinity. Clays Clay Miner 33:
369-378.
Schwertmann U, Fischer W R . 1973. Natural " a m o r p h o u s "
ferric hydroxide. G e o d e r m a 10:237-247.
Schwertmann U, Fitzpatrick RW. 1977. Occurrence of lepidocrocite and its association with goethite in Natal soils.
Soil Sci Soc A m J 41:1013-1018.
S c h w e r t m a n n U, K~impf N. 1983. 0 x i d o s de ferro j o v e n s e m
ambientes pedogen6ticos brasileiros. R bras Ci Solo 7 : 2 5 1 255.
S c h w e r t m a n n U, Murad E. 1990. The influence of a l u m i n u m
on iron oxides: XIV. Al-substituted magnetite synthesized
at ambient temperatures. Clays Clay M i n e r 38:196-202.
Schwertmann U, Schulze DG, Murad E. 1982. Identification
o f ferrihydrite in soils by dissolution kinetics, differential
X-ray diffraction, and Moessbauer spectroscopy. Soil Science Soc A m J 46:869-875.
Schwertmann U, Wolska E. 1990. The influence of a l u m i n u m
on iron oxides: XV. Al-for-Fe substitution in synthetic lepidocrocite. Clays Clay Miner 38:209-212.
S h e r m a n DM. 1987. Molecular orbital (SCF-Xa-SW) theory
of metal-metal charge transfer processes in minerals: I. Application to Fe 2§ ~ Fe 3§ charge transfer and "electron delocalization" in mixed-valence iron oxides and silicates.
Phys C h e m Miner 14:355-363.
Sherman DM, Waite TD. 1985. Electronic spectra o f Fe 3+
oxides and oxide hydroxides in the near IR to near UV.
A m Mineral 70:1262-1269.
Singer RB. 1982. Spectral evidence for the mineralogy of
high-albedo soils and dust on Mars. J G e o p h y s Res 87:
10159-10168.
Stanjek H. 1991. A l u m i n i u m - und Hydroxylsubstitution in
synthetischen und nattirlichen H~imatiten [PhD thesis].
Technische Universit~it Mtinchen, Freising. 194 p.
Stanjek H, Schwertmann U. 1992. The influence of aluminum
on iron oxides. Part XVI: Hydroxyl and aluminum substitution
in synthetic hematites. Clays Clay Miner 40:347-354.
StatSoft Inc. 1996. Statistica for Windows. Tulsa, OK.
Strens RGJ, Wood BJ. 1979. Diffuse reflectance spectra and
optical properties of s o m e iron and titanium oxides and
oxyhydroxides. Mineral M a g 43:347-354.
Taylor RM, Schwertmann U. 1974. M a g h e m i t e in soils and
its origin. I. Properties and observations on soil m a g h e m ites. Clay Miner 10:289-298.
Torrent J, Barr6n V. 1993. Laboratory m e a s u r e m e n t s o f soil
color: T h e o r y and practice. In: B i g h a m JM, Ciolkosz El,
editors. Soil color. SSSA Spec Publ. Madison, WI: Soil Sci
Soc A m . p 21-33.
536
Scheinost, Chavernas, Barr6n and Torrent
Torrent J, Schwertmann U. 1987. Influence of hematite on the
color of red beds. J Sediment Petrol 57:682-686.
Torrent J, Schwertmann U, Schulze DG. 1980. Iron oxide
mineralogy of some soils of two river terrace sequences in
Spain. Geoderma 23:191-208.
Clays and Clay Minerals
Wyszecki G, Stiles WS. 1982. Color science: Concepts and
methods, quantitative data and formulae. New York: J. Wiley. 950 p.
(Received 28 August 1997; accepted 23 January 1998; Ms.
97-084)