REFLECTANCE SPECTROSCOPY AS A REMOTE SENSING
TECHNIQUE FOR THE IDENTIFICATION OF
PORPHYRY COPPER DEPOSITS
by
KRISTINE LOUISE ANDERSEN
S.B., Massachusetts Institute of Technology
(1974)
SUBMITTED IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE
DEGREE OF
DOCTOR OF PHILOSOPHY
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
July,
1978
Signature of Author .
Department of Earth and Planetary Sciences, July, 1978
.,r
Certified by .......
Accepted by
.
... ,
s
1
4W
4
- .--
.
...............
........................
Thesis Supervisor
........
...
........
.
Chairperson, Departmental Committee on Graduate Students
Wif"93IUWN
MAS
OF
iL9,M t
NSTTUTE
akES
LIBRARIES
REFLECTANCE SPECTROSCOPY AS A REMOTE SENSING
TECHNIQUE FOR THE IDENTIFICATION OF
PORPHYRY COPPER DEPOSITS
by
KRISTINE LOUISE ANDERSEN
Submitted to the Department of Earth and Planetary Sciences,
July, 1978, in partial fulfillment of the requirements for
the Degree of Doctor of Philosophy
ABSTRACT
Reflectance spectroscopy has been used successfully
as a remote sensing technique to study the mineralogy and
composition of the surfaces of planetary objects. This thesis attempts to develop reflectance spectroscopy to locate
and identify copper bearing regions on the earth. Development of the technique required background laboratory studies
involving measurement and interpretation of reflectance
spectra of minerals--oxidized copper minerals, ferric oxides,
clays, MnO2--occurring in copper bearing soils. Absorption
features appearing in the oxidized copper phases, malachite,
azurite, and chrysocolla, were assigned on the basis of
ligand field theory and theories of molecular vibrations.
These studies showed that malachite, the most common oxidized copper phase, could be detected in surface soils from
analysis of ratio spectra generated by dividing the spectra
of areas containing different malachite proportions. In the
laboratory, reflectance spectra of simulated soils were measured, scaled to unity at 0.87 p, and divided by each other,
generating a set of ratio spectra. Intensity ratios,
1(0.75 0)/I(1.20 p) versus 1(0.65 y)/I(0.75 P), calculated
from the ratio spectra, were found to be sensitive indicators to the presence of malachite. The sensitivity of the
technique, determined from the graph of the ratioed intensity values, was found to be on the order of 2000-5000 ppm
copper. This technique was tested in the laboratory using
natural soil samples collected from the Silver Bell and
Mineral Park copper districts since there currently does not
exist a spectrometer with the required detection capabilities. Reflectance spectra of these soils were measured in
the laboratory, scaled to unity at 0.87 p, and representative samples were ratioed. Intensity ratios were calculated
from the ratio spectra and plotted on graphs similar to
those generated from the laboratory simulated soils. Relative copper concentrations contained in the natural soils,
deduced from the intensity ratio graphs, were found to be
too low (<2000 ppm as analyzed from XRF) to be detected by
this technique. The laboratory studies determined the
instrumental parameters required for the detection of
copper in the field. A spectrometer such as the one proposed for geochemical orbiting missions of planetary bodies
possesses instrumental capabilities for accurate measurement
of field spectra. Problems arising in field measurements of
spectra include vegetation, minerals which possess absorption features near the diagnostic copper wavelengths, phase
angle variations, and atmospheric effects. Vegetation not
only covers the surface, thereby masking the spectral properties of the soil, it also produces a chlorophyll absorption
feature in the 0.65-0.70 y region. The effect of this band
is to decrease the 0.65 yj/0.75 p intensity value which could
lead to erroneous interpretations at high relative amounts
of vegetation. Some clinopyroxenes (diopside, augite,
salite) possess absorption features near the bands diagnostic
of copper which could be misinterpreted. However, these
minerals are not expected to be present in abundance in surface soils characteristic of copper bearing regions. The
relative absorption band strengths in the clinopyroxene
spectra are different from those in the malachite spectrum,
so that intensity ratios calculated from the spectra of
clinopyroxenes plot outside of the region diagnostic of
malachite. Atmospheric effects are minimized by scheduling
reconnaissance missions during clear, stable weather conditions. Phase angle variations are minimized by dividing
spectra measured at similar phase angles. Reflectance
spectroscopy may be used for other earth applications, including mineral exploration. Identification of chromium
deposits is a potential candidate for application of this
technique.
iii
ACKNOWLEDGMENTS
I wish to acknowledge the Smithsonian Museum of
Natural History in Washington, D.C. for providing the many
mineral samples described in Appendix I.
Gratitude is extended to the following people who
provided their time and equipment for sample preparation and
analyses: Dr. Michael Rhodes, for XRF sample preparation;
Dr. Ramon Barnes, for the XRF machine and for helpful discussions regarding sample analyses; and Dr. Graham Hunt,
for generously measuring the infrared spectra of azurite
and malachite and for his very useful and helpful discussion regarding the assignment of the vibrational features.
I am very grateful to Dr. John Adams for providing
the time and the use of his reflectance spectrometer, and to
Messrs. Keith Ronnholm and William Thomson, who both very
kindly spent much time and energy helping me reduce and
manipulate the spectral data.
Mr. James Galey of ASARCO generously provided the
soil and rock chip samples from the Silver Bell district,
as well as many helpful comments and suggestions.
I wish
to acknowledge ASARCO for releasing the topographic map of
the North Silver Bell area for publication in this thesis.
My thanks to Mr. Clancy Wendt of the Duval Corporation for
iv
providing the Mineral Park soil samples.
My sincere thanks to Dr. Edward Solomon, who
patiently shared his knowledge and insight into the chemistry of copper complexes and whose discussions and suggestions were invaluable to this thesis.
I am grateful to Dr. Roger Burns for the use of his
X-ray diffractometer, but mostly I am deeply grateful for
his interest, discussions, and kind encouragements throughout my graduate research.
The existence of this thesis is due totally to Dr.
Thomas McCord, who had enough faith (or chutzpah) to admit
me into the graduate program.
His past concern and guidance
is most deeply appreciated.
I wish I knew how to express my thanks to Dr. Robert
Huguenin.
He suggested the idea of applying reflectance
spectroscopy to the identification of copper deposits.
Without his unbounded enthusiasm, encouragement, direct
guidance, and sense of humor, this thesis could never have
been possible.
My warmest thanks to Mr. Robert Scott and to my
family and friends for their support and enthusiasm during
my graduate career.
I also wish to thank Diane Plosia for
her graphic services, and thanks to those individuals whom
I may have inadvertently omitted.
Finally, I must acknowledge the cosmic soul of the
v
universes whose interactive forces allow the transient
existence of all things material and immaterial, including
NASA grant NSG-7405 (through the University of Massachusetts,
R.L. Huguenin, principal investigator) which supported this
research.
In fond and loving memory
of my late grandmother,
Irene K. Sladek
vii
TABLE OF CONTENTS
Page
TITLE PAGE
ABSTRACT
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
. - .- -
ACKNOWLEDGMENTS
DEDICATION
.
.
.
.
.
.
.
.
TABLE OF CONTENTS
.
.
.
-
i1
-
-
.
iv
.
.
.
vii
.
.
.
viii
xii
LIST OF FIGURES AND TABLES
LIST OF APPENDICES
-
xv
.
Chapter
I.
INTRODUCTION
The Technique of Reflectance
. . . . . . . . . . .
Spectroscopy
. . . . .
Ore Exploration Techniques
Advantages of the Technique of
. . . . .
Reflectance Spectroscopy
Reflectance Spectroscopy as a Remote
Sensing Technique to Identify
Porphyry Copper Deposits
II.
DESCRIPTION OF PORPHYRY COPPER DEPOSITS
Definition of a Porphyry Copper
. . .
Deposit
Occurrence and Distribution
.
.
Geology of Porphyry Copper Deposits
Nature of Intrusive and Host Rocks
. .
Structural Controls
. . . .
Alteration Processes
.
. . .
.
.
Mineralogy
Ore
. . . . . .
Supergene Enrichment
. . .
Leached Ore Cappings
Genesis of Porphyry Copper Deposits
viii
6
6
10
10
10
11
11
12
14
15
Page
Chapter
III.
DIFFUSE REFLECTANCE SPECTRA OF AZURITE,
MALACHITE, AND CHRYSOCOLLA
.
.
- -
. . . .
Introduction
The Crystal Structures of Malachite
. . .
. .
. .
and Azurite
Reflectance Spectra of Malachite and
. . . . . . . . - - - - Azurite
. . . . . .
Spectrum
Malachite
The
.
.
Spectrum
The Azurite
.
Features
Analysis of Vibrational
.
Reflectance Spectra of Chrysocolla
. . . . . . . .
Green Chrysocolla
. . . . . . .
Blue Chrysocolla
IV.
23
.
.
23
.
.
29
.
.
30
.
.
36
36
36
.
.
45
45
SPECTRA OF LABORATORY SIMULATED COPPER. . . . . . . . . . . .
BEARING SOILS
49
. . . . . . . . . . . . - Soil Formation
Reflectance Spectra of Soil Con. . . . . . . . . . . .
stituents
. . . . . . . . . . . . .
Ferric Oxides
. . . . . . . . . . . . . . . . .
Clays
. . .
Desert Varnish and Organic Matter
of
Mixtures
of
Reflectance Spectra
. . . . . . . .
Soil Constituents
. . . . . .
Malachite and Ferric Oxides
Malachite, Ferric Oxides, and
. . . . . . . . . . . . . .
Kaolinite
Malachite, Ferric Oxides, Kaolinite,
.
. . .
- - - - - and MnO 2
. . .
Ratio Spectra of Simulated Soils
.
Presentation of the Ratio Spectra
The Graph of Intensity Ratios
ix
.
40
40
40
41
Definition of the Term "Ferric Oxide"
. . . . . . .
Crystal Structures
Ferric Oxides
of
Spectra
Reflectance
TransiElectronic
Assignment of
.
.
. . .
.
.
tions
Ferric
of
Spectra of Mixtures
. . . . . . . .
Oxides
V.
19
20
.
THE OPTICAL PROPERTIES OF FERRIC OXIDES
19
.
.
.
49
51
52
52
54
56
56
56
57
57
58
59
Page
Chapter
VI.
VII.
SPECTRA OF NATURALLY OCCURRING COPPER. . . . . . . . . . . . .
BEARING SOILS
.
63
.
Description of Soil Samples
Mineral Park Samples
Silver Bell Samples
Reflectance Spectra of the Soil
. . . . . . . . . . . . . .
Samples
Comparison with Laboratory Spectra
Effect of Particle Size
. . . . . .
The Spectra of Large Areas
. . . .
Ratio Spectra of Representative Soil
Samples
. . . .
Presentation of Spectra
. . . . .
Predictions of Relative Copper Concentrations from Intensity Ratios
The Determination of Copper Concentrations
. . . . . . . . .
.
. . .
Graphs of Intensity Ratios
Absolute Concentration Versus
. . .
.
Relative Concentration
Soil Mineralogy versus Sample
Distance
. . . . . . . . . . . .
Conclusions from Intensity Ratio
Graphs
.
. ..
. ..
. . . . .
63
64
64
78
78
78
81
81
81
84
87
.
88
Instrumental Requirements....
.*....
..
. . . . . . . . . .
Available Instruments
88
89
INSTRUMENTATION
.
.
.
.
.
.
The Geochemical Orbiter Spectrometer
Instrumental Specifications for Copper
Reconnaissance
. . . . . . . . . .
. . . . .
Instrumental Parameters
. . . .
Calibration
Instrumental
. .
Location of the Ground Track
Data Collection
. . . . . . . . .
Obstacles and Uncertainties Involved
in Field Measurements
. . . . . .
Vegetation
. . . . . . . . . . .
Phase Angle Effects
. . . . . . .
Flight Variations
. . . . . . . .
VIII.
70
70
72
75
..
.
.
.
.
.
.
.
. .
. .
. .
. .
. .93
91
91
92
92
.
.
.
.
.
.
.
.
93
93
94
95
.
.
.
.
.
DISCUSSION
Summary
90
96
..
96
Page
Chapter
Potential Problems in Field
. . . . . . . . .
Application
of ClinoSpectra
Reflectance
. . . . . . . . .
.
. . .
pyroxenes
. . . .
Occurrence of Clinopyroxenes
Clinopyroxene Discrimination from
. . . . . . . . .
Intensity Ratios
Fe 3 +-rich Clinopyroxenes: Salite
Reflectance Spectra of Representative
. . . . .
Desert Vegetation
Effects of Vegetation of Soil
. . . . . . . . . .
Spectra
the Technique
of
Sensitivity
Advantages of the Technique
. . . . .
Potential Applications
REFERENCES
.
.
.
.
.
.
.
.
.
APPENDIX I
.
.
.
.
.
.
.
.
.
APPENDIX II
.
.
.
.
.
APPENDIX III
.
.
.
.
.
.
.
.
.
.
.
.
-
97
.
97
100
.
.
101
101
.
104
.
.
.
.
107
107
109
109
111
115
120
.
155
APPENDIX IV
204
APPENDIX V
212
LIST OF FIGURES AND TABLES
Page
I.
1.
2.
3.
4.
5.
FIGURES
Map of porphyry copper deposits occurring in
the southwestern United States (after Titley
. . . . . . . . . . . . ...
and Hicks, 1966)
8
Map of granitic rocks in the western United
States (after Titley and Hicks, 1966)
9
The process of secondary enrichment (after
. . . . . . . . .
Whitten and Brooks, 1973)
. .
Cross section of a typical porphyry copper
. . . . . . . .
system (after Sillitoe, 1973)
.
16
Porphyry copper deposits in relation to plate
boundaries (after Sillitoe, 1972) . . . . . . .
18
6.
Malachite site structures according to Sisse
.*..
. . . .21
. .
..
. . . . . . ..
(1967) .
7.
Azurite site structures according to Gattow
. . . . . . . . . . . . . .
and Zemann (1958)
8.
Reflectance spectra of malachite....
9.
Reflectance spectra of azurite.
10.
13
....
.
.....
.
22
24
. ..
25
Splitting of the one-electron energy levels of
Cu2 + in crystal fields of axial symmetry
. . . . . .
(after Hathaway and Billing, 1970)
27
spectra of malachite (from Sadtler
Labs, 1943)
spectra of azurite (from Sadtler
..
. . .
.. . . . ..
Labs, 1943)
.
31
12a. XRD pattern of azurite 5000 (CoKa radiation
used: X = 1.790A)
b. XRD pattern of azurite 5001 . . . . . . . . .
.
35
.
37
lla. Infrared
Research
b. Infrared
Research
13.
Reflectance spectra of blue and green
chrysocolla . . . . . .. . . . . . .
xii
. .
. .
Page
I.
14.
FIGURES (cont.)
Reflectance spectra of goethite and
limonite
.
o. .
.
.
.. ..
42
. . .
. . . -.-.-.
. . . .
.
. . .
15.
Reflectance spectra of hematite
16.
Reflectance spectra of maghemite and
lepidocrocite . . . . . . . . . - . . . . . . -.
17.
18.
19.
20.
21.
22.
Reflectance spectra of mixtures of hematite
. . - - . . . . . . ...
and goethite
-.
.-
Reflectance spectra of kaolinite and montmorillonite (after Adams, 1975) . . . . . . .
43
44
. 47
.
53
. .
55
Intensity ratio graph calculated from laboratory
. . . . .
simulated soils . . . . . . . . . ..
60
Soil collection grid of the Mineral Park samples
. ..
. ..
(courtesy C. Wendt of Duval Corp.)
65
Topographic map of the Mineral Park district
showing soil sample location (after Eidel
.. . . . . . . . .
. . .. . . .
et al., 1968)
66
Reflectance spectra of desert varnishes
(courtesy John Adams) . . . .. . . . .
. .
23.
Map showing the approximate location of the
North Silver Bell district (after Titley and
. . . .67
.. . . . . . . .. . ..
Hicks, 1966)
24.
Topographic map of the North Silver Bell
district depicting alteration zones and soil
and rock chip sample locations (courtesy
. .. . ....
.. .
. . . ...
ASARCO, 1978)
69
Reflectance spectra of a soil sample at four
. . ...
. ..
.. .
different particle sizes
73
25.
26.
27.
28.
Ratio spectrum of a soil sample showing the
effect of particle size . . . . . . . . ..
.
.
74
Averaged spectra representing large sample
.. .......
..
. . . . .
areas.............
76
. . . ...
77
Ratio spectra of large sample areas
xiii
Page
I.
FIGURES
(cont.)
29.
Intensity ratio graph generated from the soil
. . . . . . . .. . . . .
sample ratio spectra
79
30.
Intensity ratio graph comparing absolute
.....
versus relative copper concentrations
82
31.
Intensity ratio graph comparing soil
mineralogy versus relative sample
.- .
- -.-.- -.-.
. . . . . . . . .. .
distances
85
Plot of centers of absorption bands appearing
in the reflectance spectra of common rock. . . .
forming minerals (after Adams, 1975)
98
32.
.
33.
Reflectance spectra of clinopyroxenes (after
. . . . . . . . . . - - - - Adams, 1975)
34.
Intensity ratio graph of laboratory simulated
. . . . . . .
soils including clinopyroxenes
.
102
Absorption spectra of plant pigments (after
. . . . . . . . . . . . .
Gates et al., 1965)
.
105
36a. Spectra of desert succulent plants (after
Gates et al., 1965)
b. Spectra of typical lichens (after Gates et al.,
. .
- - - -.
- - -.
1965) . . . . . . . . ..
106
Reflectance spectra of chromite and chrome. . . . . .
diopside (after Adams, 1975).
110
35.
37.
II.
1.
2.
..
TABLES
Energies (cm~ ) of the fundamental ligand
vibrations in malachite and azurite . . . .
. .
33
. .
34
Energies (cm 1) of electronic transitions in
ferric oxides . . . . . . . . . . . . . . . . .
46
Vibrational features appearing in the reflectance spectra of malachite and azurite
3.
99
xiv
.
.
LIST OF APPENDICES
Page
.
115
.
.
.
.
115
117
117
117
118
APPENDIX II: REFLECTANCE SPECTRA OF LABORATORY
. . . . . . . . . . . .
SIMULATED SOILS
120
APPENDIX I:
SAMPLE DESCRIPTION AND ANALYSES
.
Sample Description
.
.
.
Analyses
Sample
Reflectance Spectra
.
Infrared Spectra
X-ray Fluorescence
. . .
. . .
. .
. . .
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
..
.
.
. . .
. . .
. .
. . .
. . .
.
.
.
.
.
.
.
.
. .
..
. .
. .
. .
. . . . . .
Figure 1: Malachite and ferric oxides
Figures 2-4: Malachite, ferric oxides, and
. . . . . . . . .
. . . . . . . ..
kaolinite
oxides,
ferric
Figures 5-32: Malachite,
- - - - - - - - - - - kaolinite, and MnO 2 .
APPENDIX III: RATIO SPECTRA OF LABORATORY
. . . . ..
SIMULATED SOILS
Figures 1-16: MnO2, malachite, kaolinite,
. .
. . . . . . . . . ..
and goethite
Figures 17-32: MnO 2 , malachite, kaolinite,
. . . . . . . . . . . . .
and hematite
Figures 33-48: MnO 2 , malachite, kaolinite,
. . ..
(50% hematite + 50% goethite)
Figure
7:
Silver Bell rock chip samples
.
. . . .
156
. . . .
and
.
. ..
172
.
.
o . . .
.....
.
.
.
.
APPENDIX V: RATIO SPECTRA OF NATURALLY OCCURRING
. . . o . . . . ....
..
SOIL SAMPLES
Figures
Figures
1-4: Mineral Park soil sample ratios
5-28: Silver Bell soil sample ratios
xv
125
155
.
APPENDIX IV: REFLECTANCE SPECTRA OF NATURALLY
. . . . . ..
OCCURRING SOIL SAMPLES
Figure 1: Mineral Park soil samples. . .
Figures 2- 6: Silver Bell soil samples
122
.
.
.
121
188
204
205
206
211
212
.
.
.
.
213
217
C H A P T E R
I
INTRODUCTION
The Technique of Reflectance Spectroscopy
Reflectance spectroscopy has been used successfully
as a technique for remotely sensing the mineralogy and composition of the surfaces of some solar system objects using
ground based telescopes, cf. McCord et al. (1977); Head et
al.
(1978).
This thesis attempts to develop reflectance spec-
troscopy for earth applications; in particular, to locate and
identify copper bearing regions.
Once this technique is de-
veloped for copper, it may be applied to include remote
sensing identification of other economically important ore
deposits and other applications such as pollution and crop
monitoring, geological surveying, etc.
The theory of reflectance spectroscopy has been established in the literature (Ballhausen, 1962; Burns, 1970;
Adams, 1975).
Briefly, spectral reflectance measures the
fraction of incident solar radiation that is reflected from
a surface as a function of wavelength.
Sufficient spectral
resolution and intensity'precision can resolve absorption
bands which are diagnostic of surface composition and miner1
alogy.
Mineralogical interpretation depends primarily on the
wavelength dependence of reflectance and to a lesser extent
on the albedo and the angular dependence of the reflected
radiation.
Material on the surfaces of planetary bodies including the earth occurs as randomly oriented fragments.
When
observed in reflected light, the surface particulate material
returns two components of radiation:
(1) a specular component
consisting of first surface reflections described by Fresnel's laws for absorbing dielectrics; and (2) a diffuse component which is composed of light that has entered at least
one grain and has been scattered back toward the observer.
It is the diffuse component that contains the most compositional information.
Most of the absorption features appearing in reflectance spectra arise from electronic transitions and charge
transfers by valence shell electrons in transition metal ions
2+
2+
.4+
(Cu
,
Fe
,
Ti
, etc.) and from overtones and combinations
of molecular vibrations.
The wavelength positions of the ab-
sorption band centers depend on the types of ions present and
on the dimensions and symmetry of the sites in which the ions
are situated.
To a large extent, these two factors define
the mineralogy of a sample.
Reflectance spectra are interpreted using laboratory
and theoretical techniques.
Laboratory spectra of samples
of known composition are studied as functions of particle
size, phase angle, mixing ratios, and mixing heterogeneity.
Interpretations of absorption bands are based on ligand field
theory, molecular orbital theory, and theories of molecular
vibrations.
Relative modal abundances are analytically de-
termined by comparing the relative strengths of bands contributed by the constituent minerals using assumptions about
homogeneity and relative grain sizes.
Ore Exploration Techniques
There are many techniques being used in mineral exploration, most of them involving indirect identification of
ore deposits.
Geophysical detection techniques include
studies of magnetic susceptibilities, gravitational anomalies, electromagnetic induction, and induced polarization.
Geochemical techniques include the study of surface color
anomalies, and soil, plant, and ground water sampling analyses.
Remote identification techniques have also been used.
Dykstra (1975) used ERTS imaging in an attempt to delineate
copper bearing regions according to rock facies associated.
with disseminated deposits.
In his study of the Silver Bell
porphyry copper district, he found that the spectral contrast among individual rock types was too subtle to be diagnostic -since surface weathering processes obscured the characteristics of the underlying rocks.
Advantages of the Technique of
Reflectance Spectroscopy
The advantage of reflectance spectroscopy as a remote
sensing technique compared with ERTS broad band imaging and
most of the other techniques is that it is able to directly
detect the presence of copper bearing minerals.
It is able
to scan areas that may be remote and inaccessible to in situ
exploration regimes.
With this technique, a wide region may
be surveyed during a single reconnaissance, and subtle
changes in surface composition can be detected where they may
not be discerned by visual inspection.
Reflectance Spectroscopy as a Remote Sensing
Technique to Identify Porphyry
Copper Deposits
Identification of disseminated copper deposits was
chosen as the initial earth application of the technique of
reflectance spectroscopy for the following reasons.
Most of
the porphyry type copper deposits found in the United States
occur in semi arid environments (Arizona and New Mexico)
which support minimal ground vegetation thus making surface
soils accessible for spectral measurement.
The surface ex-
pressions of copper.deposits contain oxidized copper minerals
(azurite, malachite, chrysocolla) which possess absorption
bands at spectrally diagnostic wavelengths.
In this thesis,
those wavelengths diagnostic of oxidized copper minerals
5
will be determined.
The sensitivity of the technique will
also be determined from laboratory and field studies.
C H A P T E R
I I
DESCRIPTION OF PORPHYRY COPPER DEPOSITS
Definition of a Porphyry Copper Deposit
The term "porphyry copper deposits" refers to disseminated ore of low grade usually mined by large scale/low
cost techniques such as open pit versus bulk or vein-type
mining.
It is now mostly a mining-engineering term used to
designate a large, low grade deposit regardless of genesis
and wall rock type.
However, there are features inherent
to this type of deposit as will be revealed.
It is the pur-
pose of this section to give a brief description of these
deposits and to discuss their genesis in relation to occurrence.
Most of the information in this section was obtained
from the Geology of the Porphyry Copper Deposits by Titley
and Hides (1966), and references to this book of collected
papers will be made by quoting the author and page number.
Occurrence and Distribution
The United States is the largest copper producer in
the world with about 1,411,000 tons mined in 1975 (World Mining Copper Map, 1976).
Most of the ore came from deposits
in the porphyry copper region of the southwest shown in
6
Figure 1.
The dominant physiographic feature here is the
Colorado Plateau which extends through Utah, Colorado, Arizona, and New Mexico.
It is bounded on the east by the
Southern Rocky Mountains and on the south and west by the
Basin and Range province which consists of mountain blocks
bounded by faults.
The Colorado Plateau is well known for
its rich uranium deposits, but it is the Basin and Range province surrounding it which hosts porphyry copper (Figure 1)
as well as lead, silver, gold, and molybdenum deposits.
This seemingly preferred distribution of deposits
has been related to the local crustal history (Gilluly,
1963).
The porphyry copper deposits formed before the active
uplift of the Colorado Plateau and the development of the
Basin and Range province.
They are a part of what Anderson
(p. 13) calls an orogenic-epeirogenic cycle which includes
chronologically:
(1) Cretaceous sedimentation due to a west-
ward advancing sea;
(2) Rocky Mountain orogeny from the
Cretaceous to early Tertiary resulting in intense deformation and thrust faulting;
(3) associated plutonic and vol-
canic activity; and (4) Cenozoic development of the Colorado
Plateau and the Basin and Range province.
torially summarized in Figure 2.
This cycle is pic-
Recent studies (Sillitoe,
1972 and 1973) have given strong evidence that porphyry copper deposits are indeed related to crustal history particularly in orogenic areas associated with consuming plate margins.
The genesis of these deposits will be discussed at
MILES
M
E
X
I
C
0
*CANANEA
Figure 1. Map of the southwestern
United States showing the principal
physiographic subdivisions and the
location of porphyry copper deposits.
(After Titley and Hicks, 1966).
0
200 MILES
EXPLANATION
Middle T.rtiary
x
+
Lawer Tertiary and Upper Cretaceoum
Middle Cretaceous
A
Lower Jurassic and Upper Taimc
.
Facies boundary'
Major Cretaceous thrust belt
Figure 2. Granitic rocks of Mesozoic and Cenozoic age
inmte western United States. The age assignments are
based in part on all radiometric ages published to July
1962; the ages of many small plutons are arbitrary. Facies
boundary in Washington, Idaho, and Nevada: eugeosvclinal rocks to the west, miogeosynclinal rocks to the
east. From Gilluly (21, fig. 22). Porphyry copper deposits: 1. Chino (Santa Rita); 2. Tyrone; 3. Morenci;
4. Bisbee; 5. Cananea; 6. Miami-Inspiration; 7. Ray; 8.
San Marpel; 9. Silver Bell; 10. Pima and Mission; 11.
Esperanza; 12. Bagdad; 13. Ajo; 14. Bingham Canyon;
15. Ely; 16. Yerington.
(After Titley and Hicks, 1966).
10
the end of this section.
Geology of Porphyry Copper Deposits
Although the term "porphyry copper" has become largely a mining term (Schmitt, p. 18) referring to the size and
grade of the deposit, there are geologic features which are
characteristic of this type of deposit.
be discussed briefly in this section.
These features will
In general, dissemin-
ated copper deposits are closely related to igneous intrusions (usually monzonitic porphyries), intense hydrothermal
alteration is usually found, and supergene enrichment has
operated to concentrate copper over the primary sulfide zone.
Nature of intrusive and host rocks.
Stringham (pp. 35-40)
discussed the nature of the intrusive and host rocks associated with porphyry copper deposits.
He found that the in-
trusives are of intermediate stock with a preference for
quartz monzonite-quartz latite associations, and that intrusive porphyry is an absolutely necessary feature.
The wall
rock may be of all lithologic types, ages, and thicknesses.
The most favorable host rocks for ore emplacement are intrusive or highly siliceous metamorphic and sedimentary rocks.
Structural controls.
Locally, porphyry copper districts are
found to be concentrated at orogenic and fault-zone intersections, usually triple or more complex intersections
(Schmitt, p. 30).
In the desert Southwest, most of the cop-
11
per deposits occur along the central rift of the WasatchJerome orogen, and more than half of the total U.S. copper
production comes from the intersection of this copper with
the Texas structural zone (consisting of many thrust-fault
areas) (Schmitt, p. 30).
The structural controls favoring
ore emplacement arise from the tectonics of consuming plate
margins (see Figure 5) and will be included in the discussion
of ore genesis.
Alteration processes.
Intense hydrothermal alteration is a
characteristic feature of porphyry deposits.
The alteration
is complex and varied due to differences in the prevailing
thermodynamic and compositional factors.
The principal al-
teration minerals are alunite, antigorite, micas, and clays.
Sericitization is the single most important alteration process which can attack almost every mineral in the primary
assemblage except for pyrite.
Under favorable conditions,
a porphyry deposit may end up essentially as a quartzsericite-pyrite aggregate (Schwartz, p. 44).
Ore mineralogy.
Although the mineralogy of a typical por-
phyry copper deposit is made up of a complex group of primary, hydrothermal, and supergene minerals, the ore minerals
form a relatively simple assemblage.
The primary ore
minerals are sulfides--dominantly chalcopyrite and bornite
with minor amounts of molybdenite, sphalerite, and ubiquitous pyrite.
Chalcocite and covellite are the predominate
supergene sulfides.
In the oxidized zone, malachite, azur-
ite, cuprite, and chrysocolla are more or less abundant depending on the individual deposit.
The leached ore cappings
are relatively depleted in copper minerals, and the concentration of copper naturally increases downward.
The primary sulfides occur as discrete grains
throughout the host rock and also as fillings of small veins.
Pyrite is the most abundant sulfide in the disseminated deposits; it usually is the first sulfide to be deposited, but
it may also continue to form throughout the period of metallization (Schwartz, p. 46).
The primary copper sulfides are
deposited more or less simultaneously and display mutual
boundary relations (Schwartz, p. 46).
Supergene enrichment.
The process of supergene or secondary
enrichment serves to concentrate the copper ore below the
level of the water table.
Massive blankets of chalcocite are
formed in this highly reducing environment.
Replacement of
primary sulfides is selective; chalcocite replaces bornite in
preference to chalcopyrite and then to pyrite.
Figure 3 depicts the process of supergene enrichment.
Upon oxidation, the sulfides of a porphyry copper deposit
form sulfuric acid and copper and iron sulfates.
If the con-
centration of carbonate and silicate ions is high, copper
becomes fixed as an oxidized mineral (malachite, azurite,
chrysocolla).
Otherwise, copper remains soluble and migrates
13
Figure 3
COSSAN
Secondary enrichment. Gossan is the residuum of iron oxides
(limonite) from which mobile elements have been removed, e.g.
copper, sulphur, as sulphates, etc. In the leacbed zone the sulphides
are oxidised to sulphates and transported in solution. In the zone of
oxidised enrichment, the reaction of the sulphate solution with the
original ore .in an oxygen-rich environment (carbon dioxide is also
usually present) results in the formation of carbonates and o:ides,
native metals, and (rarely) silicates. Below the water table, secondary
sulphide enrichment occurs in an oxygen-free environment.
(After Whitten and Brooks,1973).
downward.
Under reducing conditions at depth, it replaces
sulfides low or lacking in copper.
Since the porphyries are
only mildly neutralizing, they provide ideal conditions for
the process of secondary enrichment by allowing sulfuric
acid to exist long enough to oxidize the primary sulfides.
Leached ore cappings.
The leached capping of the typical
disseminated copper deposit is a highly iron-stained, porous,
siliceous mass in which the original characteristics of the
deposit are obscured.
Oxidation of pyrite and chalcopyrite
finally results in limonite which is mostly a mixture of
ferric oxides.
According to Locke (1926), the leached cap-
pings normally contain copper, though not necessarily evident
to the naked eye.
Limonitic characteristics (gossan) have
been used by geologists as a guide to the location of copper
deposits.
The outcrops of these deposits display a deep
maroon or reddish-brown color in contrast to the brick red
color over pyritic rocks.
Locke (1926) and Blanchard (1968)
both relate texture of the limonitic capping to the relative
amounts and types of primary sulfides from which it was derived.
1 Lovering
(1950) found that soils taken from a
chrysocolla-rich outcrop near the San Manuel copper deposit
contained copper contents ranging from 1,000-10,000 ppm and
that the highest Cu concentrations occurred in the finer
soil fractions. Clarke (1952) collected soil samples near
the Ray copper deposit and found copper values from 150-2,000
ppm with an average value of 650 ppm above the ore body.
Genesis of Porphyry Copper Deposits
The genesis of porphyry copper deposits has long been
the subject of much interest.
Schmitt (pp. 17-33) proposes
that deep seated tectonic activity such as that found in
orogenic zones associated with consuming plate margins remobilizes copper and associated metals.
More recent studies
(Sillitoe, 1972; 1973) confirm this "orthomagmatic" model of
ore genesis.
system.
Sillitoe defines the typical porphyry copper
It is originally emplaced 1.5 to 3 km beneath the
summit of a strato volcano and can possess a vertical extent
up to 8 km.
During ascent to higher crustal levels, bodies
of calc-alkaline magma lose fluid during retrograde boiling.
Alteration, mineralization, and brecciation in the early consolidated hood of the intrusive and the immediate wall rocks
are accomplished by these escaping fluids.
Fumarolic activ-
ity accompanies porphyry copper formation suggesting that the
magma bodies that host the porphyry copper may represent
magma chambers that had not erupted but evolved passively.
Figure 4 summarizes the typical porphyry copper system.
Porphyry copper deposits are intimately genetically
related to igneous calc-alkaline activity.
This restricts
ore formation to orogenic areas associated with subduction
zones since it is this type of magmatism that is produced
at consuming plate margins.
The temporal and spatial dis-
tribution of deposits depends upon the level of erosion and
Fig .
. Idealized cross section of a typical, simple porphyry copper deposit showing its position at the boundary between
plutonic and volcanic environtbents.
(After Sillitoe,
Vertical and horizontal dimcnsions arerneant to be only approximate.
1973).
17
the amount of copper available in the underlying subduction
zone.
Figure 5 is a powerful statement in support of this
theory, and it suggests regions for future porphyry copper
prospecting.
18
------- -
-
Accretnq platemargins
-
Active transform faults
plate margins
Consuming
-
Plate margins of indeterunate nature
Mesozoc-Cenozoic Mountain bells
Reqions with Dorphyry copper and molybdenum deposits
Plate boundories token from Dewey and bir d, 170
Fig.
5.
The western Americats. southwest Pacific and Alpide porphyry belts in relation to Mesozoic-cenozoic
orogenic belts and accreting and consuming plate boundaries.
(After Sillitoe, 1972).
C H A P T E R
I I I
REFLECTANCE SPECTRA OF MALACHITE,
AZURITE, AND CHRYSOCOLLA
Introduction
The copper carbonates azurite, Cu3 (CO3 ) 2 (OH)2 , and
malachite, Cu2 CO 3 (OH) 2 ' are formed in the oxidized zone of
copper deposits and usually occur together.
Although they
are chemically similar, the fact that their crystal structures are different is apparent from analysis of their spectra.
Hunt and Salisbury (1970b) recorded the diffuse re-
flectance spectra of both minerals, and Lakshman and Reddy
(1973) recorded the optical absorption spectrum of malachite.
It is the intention of this chapter to refine the previous
work by assigning all of the optical and near infrared features appearing in the reflectance spectra of a suite of both
minerals.
Features arising in the reflectance spectra are
electronic transitions due to Cu2+ and ligand vibrations.
Electronic transitions arising from Cu2+ in tetragonally
distorted octahedral sites are assigned on the basis of
ligand field theory.
The near infrared vibrational features
arise from combinations and overtones of ligand CO 3 and OH
19
fundamentals whose energies are determined from analysis of
their infrared spectra.
The copper silicate chrysocolla
also occurs in association with azurite and malachite in the
oxidized zone of copper deposits, and features appearing in
its reflectance spectrum are also discussed at the end of the
chapter.
The Crystal Structures of
Malachite and Azurite
The malachite crystal structure was determined by
Wells (1951) and refined by Ssse (1967). The crystal be5
longs to the space group P2 /a = C2h with four formula units
2+2
per unit cell. Cu2+ occupies two different sites as shown
in Figure 6, both of which are axially elongated octahedra
with the ligands in trans-positions.
2+
populated by Cu
Both sites are equally
Gattow and Zemann (1958) determined the azurite crystal structure as shown in Figure 7. The crystal belongs to
5
the space group P2 /c = C2 h with two formula units in the
2+2
unit cell. Cu2+ occupies two different sites, one of which
is essentially square planar with the ligands in trans-positions and the other site which is a square-based pyramid with
the ligands in cis-positions.
This square pyramid site oc-
curs twice as often as the other site in the crystal structure.
OH
03C03
OH
to
OH
'
1.915
e 1.
03
1.899
Cu
OH
003
C
2.116
Cu
2.042
C
\
\-p (0
/c'
nw
C03
SITE
CN
OH
I
SITE 2
Figure 6. Malachite site structures according to SUsse
(1967). Both sites are axially elongated octahedra (0 h +
D4h distortion).
3
003 C0
3
C03
OH
1"98
1.98
/
Cu
OH
OH
2.04
/
Cu
2.01
C03
OH
C03
SITE
I
SITE 2
Figure 7. Azurite site structures according to Gattow and
Zemann (1958). The extent of axial elongation in Site 1
is so great that i.t is essentially square planar (D4h).
Site 2 is nearly a square pyramad (C4v).
Reflectance Spectra of Malachite and Azurite
) reflectance
Visible-near infrared (4000+ 29,000 cm
spectra of suites of malachite and azurite samples are shown
in Figures 8 and 9 respectively.
The sharp, weak absorption
features in the infrared region arise from carbonate and
hydroxyl ligand vibrations, and the broad, strong absorption
2+.
electronic
features in the visible region are due to Cu
transitions.
The reflectance peaks near 20,000 cm~
give
azurite its blue color and malachite its green color.
The electron configuration of Cu2+ is 3d9 which means
that only one electronic transition,
2 Eg+
T2g
is possible
for Cu2+ in an octahedral (0 h) ligand environment. However,
2+
complexes are found in
it is well known that very few Cu
perfect octahedral coordination (CuSiF6*6H 20 is the one exception; Pappalardo, 1961) but rather they are found in
tetragonally (D4 h) distorted
0h
environments.
Such is the
case in azurite and malachite where inspection of their
2+.
occupies
cation sites (Figures 6 and 7) shows that Cu
axially elongated (tetragonally distorted (D4 h))
octahedral
sites in both minerals, although the extent of axial elongation is greater in the azurite sites.
Figures 8 and 9 show
the occurrence of more than one ligand field transition in
2+ ..
is not in
the spectra of each mineral which means that Cu
perfect octahedral coordination.
The malachite spectrum.
The malachite spectra (Figure 8)
Figure 8. Reflectance spectra of malachite.
24
0.8
1 - Malachite 6000
0.7
2 - Malachite 6001
0.6
3 - Malachite 6002
4 - Malachite 6003
5 - Malachite 6004
0.5
0.4
0.3
0.2
0.I
0.1
0.1
0.1
0.1
0.0 L
3.0
-
2.5
2.0
1.5
C M~1 x
1.0
iO~4
0.5
Figure 9. Reflectance spectra of azurite.
1.0
1 - Azurite 5000L
0.9
2 - Azurite 5001
-
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
--
3.0
2.5
2.0
1.5
CM~' x
I.0
I0-4
0.5
show three electronic transitions centered near 9200 cm
13,000 cm~
(1.08 p),
(0.77 p), and 22,400 cm~
1
(0.45 y); the
azurite spectra show two such transitions centered near
13,000 cm~
(0.77 p) and 16,100 cm
1
(0.62 y).
The bands
may be tentatively assigned on the basis of Figure 10, bearing in mind that band assignments based on spectral studies
alone are rarely definitive in the case of copper complexes
(Hathaway and Billing, 1970).
In malachite, the extent of axial elongation is not
great enough for the sites to be considered square planar
(the ratio of the average copper-ligand band length along the
z-axis/the average copper-ligand band length in the x-y plane
is approximately 5/4).
Therefore, the 2B
+
2A
transition
should be the least energetic and may be assigned to the band
at 9,200 cm 1.
(dxz' d z)
Figure 10 shows that the b 2 g (dxy) and e
orbitals lie close in energy, even when the extent
of axial elongation is great.
Thus, the band at 13,000 cm~1,
which is more broad and intense than the 9,200 cm 1 band, may
2.
2
2
2
transitions.
+ E
and B
+ B
be assigned to both the B
g
lg
2g
lg
This 13,000 cm 1 band has an asymmetric band width which may
be due to a small energy separation between the b2g and eg
orbitals.
The assignment of these transitions compares well
with band assignments of other similarly coordinated Cu 2 +
complexes (Hathaway and Billing, 1970, p. 176).
The value of 10 D equals the energy of the transi2
2
which depends on the actual site symmetry,
tion B + B
lg
2g
A&-41 (bit)
.lIS,
N Nek~~iL
-
-
4'~
~
N
,
N
c~q LbL1J
____
j
a
Y1
Il
Coepmrcaf
d~;,d1 L
N
~
z
Ctaz (01)
oc~k,'e&tc
SgtroJe
C00v
-
-1 -
?( ran
LDjy6%
Figure 10. The splitting of the one electron energy
levels of Cul* in crystal fields of axial symmetry.
(After Hathaway and Billing, 1970).
C~)
but not on the extent of axial elongation (Lever, 1968, p.
-l
Thus, in malachite, D = -1300 cm
which compares well
q
with other similarly coordinated Cu2+ complexes (Burns, 1970;
78).
Pappalardo, 1961; Hathaway and Billing, 1970).
The origin of the band at 22,400 cm
is uncertain,
but it may be due to some low-lying change transfer transition.
A metal to metal charge transfer transition from Cu 2 +
to Cul+ is probably not the reason for this band as it is
found to generally occur in the 1-2 p region (E.I. Solomon,
personal communication).
However, if a ligand has a rela-
tively high energy filled orbital, and the metal has a relatively low lying empty orbital, then a charge transfer transition will be observed either in the visible or near-ultraviolet region (Lever, 1968, p. 228).
This may be the case in
malachite where such a transition is possible from a ligand
-bonding orbital to a copper eg (7*) orbital.
Origin of
this transition may arise from the hydroxyl ligand as it is
more reducible than carbonate (E.I. Solomon, personal communication).
Few studies have been made concerning the na-
ture of charge transfer transitions arising in transition
metal complexes, and further studies are needed in order to
confirm the above hypothesis.
The origin of this transition could also be due to
3+
the presence of an impurity such as Fe . However, the
spectrum of reagent grade malachite (spectrum #5, Figure 8),
whose maximum limit of iron impurities is less than 0.01%,
exhibits this 22,400 cm
feature.
Thus, this feature must
be intrinsic to the malachite structure.
The azurite spectrum.
The ligand field transitions in
azurite may also be tentatively assigned on the basis of
Figure 10.
All three transitions occur in the 13,000-16,000
cm-1 region, and considerable overlap can occur since electronic transitions have half-widths on the order of 500-2000
cm~
(Lever, 1968).
This could account for the higher inten-
sity of these bands relative to those in malachite.
The extent of axial elongation in the azurite sites
is much greater than in the malachite sites (compare Figures
6 and 7) which means an increase in the extent of tetragonal
(D4 h) distortion.
This would account for the shift to higher
energies relative to malachite of the 2B + 2E transition
lg
g
1
(from 13,000 cm- to 16,100 cm ) with the alg orbital shifting into the e
orbital (refer to Figure 10).
The b2g or-
bital is not much affected by the extent of axial elongation
so that the 2B
+ 2B
transition energy should be approxilg
2.g
mately the same as it is in malachite (13,000 cm ).
The
ligand field transitions in azurite are therefore assigned to
2
2
the B
+ B
transition (= 10 D ) at 13,000 cm -i and the
lg
2g
qi
2
2
2
B + 2A , E transitions at 16,100 cm -l1. The occurrence
of two transitions near 16,100 cm-
would account for this
band's greater intensity and broadness relative to the single
transition at 13,000 cm-1 .
The absorption feature found at 22,400 cm
in mala-
chite may also occur in the azurite spectrum but at an energy
higher than can be recorded by reflectance measurements
( 29,000 cm
1
).
The strong fall-off in reflectivity at the
high energy end of both the azurite and the malachite spectra
(2 25,000 cm~
) most likely arises from ligand to metal
charge transfer transitions of the type discussed previously.
Analysis of Vibrational Features
The absorption features in the infrared region (7700
to 4000 cm- ) of the reflectance spectra arise from combinations and overtones of the fundamental vibrations of the
hydroxyl and carbonate ligands.
The hydroxyl ligand pos-
sesses only one fundamental vibration which is IR active.
Carbonate, governed by its intrinsic D 3 h point group symmetry, has four fundamental vibrations, two of which are
doubly degenerate (v3 and v4 ) and one IR inactive mode (v).
However, carbonate is found to occupy a general lattice position (C1 symmetry) in both azurite and malachite (Halford,
1946) thus removing all selection rules and degeneracies under its intrinsic D3 h symmetry.
IR spectra (4000-250 cm~
N
In Figure 11, representative
) of both minerals suggest that the
vibration becomes IR active, and that the v4 degeneracy
is removed (the degenerate component.s arising from v3 are not
resolvable).
The frequencies of the fundamental vibrations in a
Figure 11. Infrared spectra (4000-4 250 cm 1 ) of
(A) malachite (smithsonian sample B1044 2), and
(B) azurite (Smithsonian sample R15412)
(Source: Sadtler Research Laboratories, Inc., 1943).
100
80
60
40
20
0 L4.0
3.5
3.0
2.5
2.0
100
-
1.6
1.8
1.4
1.2
1.0
0.8
0.6
0.4
0.2
1.2
1.0
0.8
0.6
0.4
0.2
x IO- 3 CM~
B
80
604020
04.0
3.5
3.0
2.5
2.0
1.8
1.6
1.4
v x 10-3CM-1
32
suite of both minerals are listed in Table 1.
The reason for
the appearance of two sets of frequencies for each fundamental is that the ligands occupy two different positions in
both crystal structures, viz. the average copper-ligand distance in the x-y plane is much shorter than that along the
z-axis of the site (refer to Figures 6 and 7).
In azurite,
the Cu-OH bond lengths are all approximately equal (~2.OOA)
so that only one frequency for VOH is observed.
Only one
value of the o. fundamental appears in both minerals: the
frequency of this node may be invariant with respect to
ligand position or one vibration may be of very low intensity.
Also, the 1025 cm
1
V 1 frequency does not appear in
the IR spectrum of azurite 5000.
The IR and reflectance
spectra as well as the x-ray diffraction pattern of this sample (Figure 12) all possess sharper and more well defined
peaks than the other azurite sample indicating a more ordered
crystal structure.
This would affect ligand orientation to
some degree and consequently destroy the symmetry coupling
needed to produce the (IR forbidden) v
-1
cm
vibration at 1025
Table 2 lists the observed frequencies of the vibrational features appearing in the malachite and azurite reflectance spectra.
The calculated energies are found to
agree well with the observed values, but the fact that each
fundamental possesses two frequencies due to unequivalent
structural positions leads to some rather broad bands where
Table 1.
Energies (cm 1) of the fundamental liyand vibrations appearing
in the infrared spectra (4000-250 cm~ ) of malachite and azurite.
OH
CO 3
V
1030-1045
V
2
875
V3
1375-1390
4
4
700-705
810-820
Malachite
1080-1095
Azurite
(1025)
1080 1
L3395-2405.
1475-1495. _735-745]
945
1410
730
800-805
_1460]
760
815-820
This feature does not appear in sample 5000.
3300-3320
3415-3420
Table 2.
Observed energies (cm ) of ligand vibrational
features appearing in the reflectance spectra
(29,000-4000 cm- 1 ) of malachite and azurite.
Observed Energy
Malachite
(cm-1 )
Assignment
5500
5450-4720
V
+ 3v3
2v
+ 2v3
3v 1+
2v4
V1 + 2v3 +
\4
4480
3v
3
4420
Azurite
7690
3v
6755
+ 3v 3
2v OH
6290-6210
2v
+ 3v3
6175-5585
3v
+ 2v3
2v
+ 3v3
V
5375
V
4975
4855-4610
+ 2v 2 + 2vo3
+ 3v 3
2v1 + 2v3
S+
2v3 + V
3v1 + 2v4
4320
3v3
*
All vibrations are due to carbonate unless otherwise noted.
A
(D
H
B
50
45
Figure 12. Powder X-ray diffraction pattern of azurite using Co
Ka radiation (X = 1.790 A).
(A) Azurite 5000
(B) Azurite 5001.
several features overlap.
The hydroxyl overtone at 6755 cm 1
appears in the azurite but not the malachite spectrum since
the overtone position would be obscured by the low energy
electronic transition at 9200 cm 1 .
Reflectance Spectra of Chrysocolla
Green chrysocolla.
Chrysocolla, CuSiO3-nH20,
is a blue to
green mineral found in the oxidized zone of copper deposits
associated with azurite and malachite.
Blue and green ap-
pearing chrysocollas have been thought to be structurally the
same, but inspection of Figure 13, showing reflectance of
representative blue and green chrysocollas, reveals that this
is not the case.
Intense infrared features occur in the
spectrum of blue chrysocolla near 7150 cm 1 , 5260 cm 1 , and
4440 cm
1
due to water vibrations, while the infrared bands
in green chrysocolla are very similar to the carbonate and
hydroxyl bands found in the malachite spectrum.
Further,
the ligand field transitions in green chrysocolla occur at
the same energies as those in the malachite spectrum, viz.
9200 cm
and 13,000 cm
Blue chrysocolla.
Green chrysocolla, therefore, seems to be
an intimate mixture of malachite and silica.
Blue chryso-
colla, however, is apparently a hydrous copper silicate which
does possess a well-defined crystal structure (although it
has not yet been reported in the literature) since its re-
Figure 13. Reflectance spectra of chrysocolla.
0.7
A - Blue chrysocolla (7002)
B - Green chrysocolla (7005)
0.6
8
0.5
0.4
I
0.3
0. 2
0.1 -B
A
0.I
2.5
2.0
1.5
cm 1 x 10~4
1.0
0.5
flectance spectrum (Figure 13) is unique compared with the
spectra of other copper minerals.
Ligand field bands in blue
chrysocolla occur near 11,800 cm
and 14,000 cm
.
These
transitions may be tentatively assigned on the basis of
Figure 10 since the effective ligand environment in blue
chrysocolla is similar to those found in azurite and malachite.
(Indeed, it is well known that almost every Cu2+ com-
plex similar to the ones in this discussion possesses a
tetragonally distorted octahedral ligand environment (Hathaway and Billing, 1970).)
The energies of the ligand field
transitions in blue chrysocolla occur between the energies
of the corresponding transitions in azurite and malachite
(between 9000 cm
1
and 16,000 cm~).
In blue chrysocolla,
the energy separation between the bands (~2000 cm
than that found in azurite (~3000 cm~
-l
) is less
) which means that the
extent of tetragonal distortion in blue chrysocolla is probably not as great as it is in azurite where the a
is expected to occur at the same energy as the e
The low energy transition at 11,800 cm
orbital
orbital.
in blue chrysocolla
is more intense relative to the intensity of the corresponding feature in azurite and malachite, suggesting that the a g
orbital may occur near the same energy as the b2g orbital.
This analysis leads to the assignment of the 2B
+2 A
, 2B2g
transitions at 11,800 cm 1 and the 2 gB+ 2 Eg transition at
-gl
1400 cm~ , and implies that the extent of tetragonal distortion in blue chrysocolla is greater than in malachite but
39
less than in azurite.
The reason for this cannot be ascer-
tained until the chrysocolla structure is determined.
A relatively weak feature is noted at 23,800 cm1 in
blue chrysocolla similar to the one at 22,400 cm~
chite.
in mala-
The mechanism for this transition is probably similar
to the one discussed in malachite.
C H A P T E R
I V
THE OPTICAL PROPERTIES OF FERRIC OXIDES
Definition of the Term "Ferric Oxide"
Ferric (Fe +) oxides are common and widespread in occurrence since they are weathering products of metallic iron
and iron bearing minerals.
In arid regions, they impart a
characteristic brick red to ochre color to surface rocks
and soils, and they tend to be enriched in areas associated
with ore depoxits due to the weathering of iron bearing sulfides (mostly pyrite).
The term "ferric oxide" is used here to encompass
the polymorphs of Fe 2 0 3 and the polymorphs of FeO(OH).
By
far the most abundant ferric oxides are those with the hexagonal close-packed oxygen structure--hematite (a-Fe20 3) and
goethite (a-FeO(OH)).
The cubic close-packed oxygen poly-
morphs are maghemite (y-Fe 2 O 3 ) and lepidocrocite (y-FeO(OH))
which occur naturally but to a much lesser extent than the
a structures.
Crystal Structures
Bernal et al.
(1959) found the crystal structures of
these ferric oxides to be composed of different stackings of
40
41
close-packed oxygen/hydroxyl ions with various arrangements
of Fe + ions in the interstices.
Fe + occupies octahedral
sites in all of these structures, and it also occupies tetrahedral sites in maghemite which has a spinel-like structure
very similar to that of magnetite (y-Fe 3 04 ).
The degree of
cation ordering can vary greatly in the maghemite structure,
but the octahedral sites are always enriched relative to the
tetrahedral sites.
Reflectance Spectra of Ferric Oxides
The electron configuration of Fe
gives rise to a 6A
field.
is (Ar)3d5 which
ground state in a weak octahedral ligand
Since all of the electrons are unpaired, any elec-
tronic transition involves spin pairing and thus all transitions are spin forbidden.
tra (29,000 cm
1
Representative reflectance spec-
- 4000 cm-1 ) of goethite, hematite, mag-
hemite, and lepidocrocite are shown in Figures 14-16; Figure
14 contains spectra of limonite, an omnibus term implying a
mixture of x-ray amorphous iron hydroxides and hydrated iron
oxides.
X-ray diffraction analyses of these samples showed
that they contain mostly goethite.
The spectra of all of the
ferric oxides are very similar in that the electronic transitions occur at similar energies and intensities.
The mag-
hemite specimen was obtained by oxidizing magnetite, and its
electronic transitions occur at higher energies than in the
other ferric oxides.
This may be due to contributions from
Figure 14.
Reflectance spectra of limanite and goethite.
1.0
1 - Limcnite 2001L
0.9-
2 - Limonite 2000L
3 - Goethite 3004
4 - Limonite 2003
0.80.70.60.5-
4
3
0.4
0.3
u
2
4
Ld
: 0.1
O0
0
0.1 -3
0.1
-2
0.1
0.0I
2.5
2.0
1.5
c m~' x
1.0
I0- 4
0.5
0.0
1.0
Figure 15. Reflectance spectra of hematite.
43
1 - Hematite 1003
0.9
2 - Hematite 1001
3 - Hematite 1000
4 - Hematite 1005
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1 -4
-2
0.1 -3
0.1
0.1
0.0
II
I
2.5
I
2.0
I
1.5
cm x I0~4
1.0
I
0.5
0.0
44
Figure 16. Reflectance of maghemite and lepidocrocite.
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.I
0.I
0.0
2.5
2.0
1.5
cm' x i0~4
1.0
0.5
0.0
45
Fe
in the tetrahedral sites (Dq would be smaller meaning
a higher energy separation between levels in the weak field
d5 case).
Assignment of electronic transitions.
The bands are assigned
according to those given in the literature (Jorgensen, 1963,
p. 92; Ballhausen, 1962, p. 253; Mao and Bell, 1973-74; Hunt
et al.,
1971a) for other high spin ferric oxide complexes.
The lowest energy transition near 11,000 cm~1 (0.90 P) is as4
+ T
transition; the transition near
lg
lg
-16
4
16,000 cm 1 (0.63 p) is assigned to the A
+4 T
transilg
2g
tion; and the high energy transitions near 21,000 cm~
signed to the
(0.50 y)
to the
6
A
and 23,000 cm
6
A
4
46
+4 Agr 4 E
1
(0.44 p) are assigned respectively
and
A
+
4
T
(D) transitions.
lg lg
g
lg
2
3 lists the band assignments and energies for the ferric
Table
oxides studied here as well as for some others reported in
the literature.
Spectra of mixtures of ferric oxides.
When several phases
of iron oxides are present, the band positions are found to
vary linearly with relative amount.
Figure 17 shows how the
band positions change with varying amounts of hematite and
goethite.
The success of reflectance spectroscopy as a technique to identify copper deposits depends on the ability to
spectrally identify the weathering phases of copper, viz.
azurite, malachite, chrysocolla.
In arid regions and es-
Table 3. Energies (cm 1) and assignments of electronic
transitions appearing in the reflectance spectra
of ferric oxides.
(cm 1)
6A
-+
4T
T2g
IA
4E
4T 2g(D)
aFe(H0) 3+
H2 0 6
12,600
18,500
24,600
27,500
a-FeO(OH)
11,200
16,000
22,000
23,700
b(s)
10,930
15,500
b(y)
11,210
15,340
y-FeO(OH)
11,800
15,600
21,250
23,200
b(H)
10,140
13,700
15,300
b (Y)
10,890
14,950
a-Fe 2 0 3
11,900
16,000
18,850
20,000
y-Fe 2 0 3
12,100
16,300
20,000
20,650
aData from Jorgensen, 1963.
bData from Mao and Bell, 1973-74.
Figure 17. Reflectance spectra of mixtures of hematite 47
(reagent) and goethite (limonite 2001). Mixtures are in wt.%.
0.9
I
I
1 2- 2% hematite + 78% goethite
0.8
32% hematite + 68% goethite
3 - 59% hematite + 41% goethite
2
--
0.7
-
4 -
--
4
78% hematite + 22% goethite
0.6-
3
0.5 0.4-
2
LU
i
z0.3
FU
U
0.2
4
0.1
-3
0 .1I --2
0.0
0.0
2.5
2.0
2.0
1.5
cm- 1 x 10-4
..
1.0
0.5
0.0
48
pecially over mineral deposits, the surface concentration
of ferric oxides will be much higher relative to the concentration of copper phases.
However, the electronic transi-
tions in the copper phases--malachite in particular--occur
at different energies than in ferric oxides.
A spectrum
measured above a copper deposit will be dominated by ferric
oxide, so the presence of copper will be detected from analyzing the subtle changes in the ferric oxide spectrum.
C H A P T E R
V
SPECTRA OF LABORATORY SIMULATED
COPPER-BEARING SOILS
The success of reflectance spectroscopy as an exploration technique to identify copper deposits depends upon
the ability to detect the presence of copper bearing minerals
in soils above copper deposits.
Development of this tech-
nique requires that the composition of copper-bearing soils
be known so that the presence of copper may be determined
from the soil spectra.
It is the purpose of this chapter to
present laboratory spectra of samples containing representative amounts of natural soil constituents including various
amounts of malachite, the most common weathering phase of
copper.
The spectra are studied to determine how the pres-
ence of malachite can be detected in the representative soil
spectra.
Finally, the spectra of natural samples taken from
copper bearing regions are analyzed, and the results are correlated with their laboratory determined copper concentrations.
Soil Formation
The majority of porphyry copper deposits in the
United States are situated in the desert Southwestern states
49
(Arizona and New Mexico).
The semi-arid climate of this re-
gion supports little extensive ground vegetation so that the
surface is mainly composed or rocks and soils.
The soils
contain minor amounts of decayed organic matter mixed with
the weathered mineral phases derived from the surrounding
surface rocks.
The geology of porphyry copper deposits was discussed
in a previous section where it was noted that hydrothermal
alteration is an important and necessary process in the formation of a porphyry deposit.
Creasey (1966, in Geol. of the
Porphyry Copper Deposits) notes three types of hydrothermal
alteration processes characteristic of porphyry copper deposits--potassic, argillic (phyllic), and propylitic.
Al-
teration of granodiorite rocks associated with porphyry copper deposits produces clays (chiefly kaolinite and montmorillonite), micas (sericite and minor biotite), potassic
feldspar, and to a lesser extent, chlorite, carbonates, and
epidote which are formed mostly by propylitic alteration.
Surface exposure of these altered rocks results in subaerial
weathering of their surfaces--iron bearing minerals are
.
2+
oxidized to Fe
oxides (limonite, hematite) and feldspars
are altered to clays.
Soils are formed by mechanical weathering of the
rock surfaces, and thus their compositions will reflect the
composition of the rock surfaces from which derived.
Soils
from the porphyry copper region should be composed mostly of
Fe + oxides, clays, and some decayed organic matter.
Reflec-
tance spectra of soils obtained from the region of porphyry
copper deposits show that this is indeed the case (see
Figures 1- 7, Appendix IV).
These spectra will be discussed
in a later section.
Huff (1952) found that copper tends to accumulate
in soils with high concentrations of clay or organic matter,
and that leaching does not affect the concentration of copper
in soils nearly as much as it affects the rocks from the
supergene zone of a copper deposit.
A copper anomaly is
formed by the weathering of exposed rock surfaces, and the
shape and size of the anomaly are influenced by topography
which controls soil creep.
Although the extent of surface vegetation is minimal
in the porphyry copper district, decayed organic matter and
desert varnish are. present in soils and on rock surfaces.
Reflectance Spectra of Soil Constituents
In the following sections, reflectance spectra of
samples modeled after soils typical of copper anomalies are
presented, and spectral features are correlated with the
constituent minerals.
The addition of various amounts of
malachite to these representative soil samples is studied for
the effects on the spectra.
Finally, ratio spectra are
generated, and the regions of the spectrum which are diagnostic of malachite are determined.
Ferric oxides.
Soils from copper bearing regions generally
contain ferric oxides, clays, and decayed organic matter.
Understanding the spectrum of a mixture of these minerals
involves determination of the spectral contributions arising
from each constituent mineral.
Reflectance spectra of com-
monly occurring ferric oxides and hydroxides were presented
and discussed in the previous chapter (refer to Figures 1416).
Limonitic goethite and hematite, the predominate
weathering phases of iron bearing minerals, form on the surfaces of exposed rocks and are thus released to the soil as
a result of mechanical weathering.
Clays.
Clays are formed by hydrothermal alteration or
weathering of silicates, and they generally make up a large
percentage of many soils.
Kaolinite and montmorillonite are
expected to be the primary clay constituents found in copper
bearing soils.
Reflectance spectra (0.35-2.50 p) of representative
samples of kaolinite and montmorillonite are shown in Figure
18.
Salisbury and Hunt (1970) and Adams (1975) discussed the
spectra of these clays, and their findings are summarized
here.
The major spectral features exhibited by these clays
are several strong hydroxyl bands in the infrared region,
centered near 1.4 p and 2.2 p.
Compared to the montmoril-
lonite spectrum, kaolinite exhibits a weak band at 1.9 p in-
Figure 18
0.5
1.0
15
WAVELENGTH (;pm)
2.0
2.5
Diffuse reflectance spectra of micas and clays. Muscovite and the clays (kao-
linite and montmorillonite) show strong hydroxyl and water bands, respectively. Phlogopite and biotite in contrast have Fe2 + and Fe+-Fel+ absorptions.
(After Adams,
1975).
dicating the lack of appreciable bound H 2 0.
The features
near 0.95 p and 1.15 p are also due to vibrations of the
water molecule, but they are much weaker (by about a factor
of ten) than the bands at 1.4 p and 1.9 p.
Desert varnish and organic matter.
Although the extent of
surface vegetation is minimal in the porphyry copper district, decayed organic matter and desert varnish are present
in soils and on rock surfaces.
Spectral features arising
from these materials are expected to be minimal; however,
they act as darkening agents which would reduce the overall
albedo thus affecting the shape of the entire soil spectrum.
Desert varnish forms a dark coating on exposed rock
surfaces.
Engel and Sharp (1958) and Potter and Rossman
(1977) both studied desert varnishes and showed that most of
them are composed of iron and manganese oxides, and that they
tend to be enriched in trace elements, especially copper.
Reflectance spectra (0.35-2.50 p) of desert varnish
coating two different types of rocks are shown in Figure 19
along with the spectrum of finely ground MnO 2 for comparison.
Spectral features arising from the rock-forming minerals are
strongly obscured by the varnish.
The low albedo and fea-
tureless MnO 2 spectrum is probably due to intense charge
transfer bands arising from metal-metal and metal-ligand
electron transfer processes.
60
50
uJ
C
40
-
30
z
.J
u.20
0
10
\0
\0
0
0.5
1.0
1.5
WAVELENGTH
Figure 19.
2.0
(/L)
Reflectance spectra of desert varnishes.
(Courtesy J. Adams) .
2.5
Reflectance Spectra of Mixtures of
Soil Constituents
Samples containing predetermined amounts of malachite, hematite, goethite, kaolinite, and manganese dioxide
(MnO 2 ) were prepared in the laboratory, and their reflectance
spectra (0.35-2.50 y) were measured.
The minerals used were
made up of very small sized particles ( 5 50 p diameter), although they were not sorted.
Malachite and ferric oxides.
Figure 1 in Appendix II shows
shows reflectance spectra of samples containing selected
weight percentages of malachite combined with an equal
weight percent mixture of hematite and goethite.
Increasing
the amount of malachite depresses the ferric oxide reflectance peak at 0.75 p and changes the slope of the spectrum
between 1.0 p and 1.5 p.
This is due to the electronic
transitions in malachite centered at 0.77 p and 1.08 p.
(Azurite and chrysocolla (see Chapter III) would likewise
depress the ferric oxide reflectance peak due to their electronic transitions at 0.72 p and 0.76 p, respectively.)
Car-
bonate vibrational features at 2.25 p and 2.40 p gain intensity at higher malachite percentages.
Malachite, ferric oxides, and kaolinite.
Reflectance spectra
of mixtures of malachite, an equal weight percent mixture of
hematite and goethite, and kaolinite are presented in Figures
2-4 in Appendix II.
The addition of kaolinite raises the
57
overall spectral albedo and adds characteristic H2 0 vibrational bands to the spectrum at 1.4 p, 1.9 p, and 2.2 p.
In-
creasing the weight percentage of malachite relative to ferric oxides also has the effect here of flattening the ferric
oxide peak at 0.75 P and lowering the spectral albedo between
1.0 and 1.5 p.
Malachite, ferric oxides, kaolinite, and MnO
Selected
weight percentages of MnO 2 were added to mixtures of malachite, ferric oxides, and kaolinite, and their reflectance
spectra are shown in Figures 5-34 in Appendix II.
Increasing
the weight percentage of MnO 2 correspondingly lowers the
total spectral albedo by approximately the same percentage,
e.g., increasing the amount of MnO 2 from 5% to 50% lowers the
overall curve albedo by about 40%.
The spectral contrast is
also lowered at higher percentages of MnO 2*
At the same
amount of MnO 2 ' malachite affects the ferric oxide spectrum
as described in the paragraph above.
Ratio Spectra of Mixtures of Simulated Soils
In the laboratory, a reflectance spectrum is measured
against a standard reference source, usually MgO or halon.
However, in the field, a reference standard will not be
available.
To eliminate spectral contributions arising from
the atmosphere and instrumentation, an individual spectrum
may be ratioed by another.
This residual ratio spectrum will
depict any relative differences between the two spectra involved.
It is hoped that differences in copper concentra-
tions between two samples will diagnostically affect the
ratioed spectrum.
Presentation of the ratio spectra.
Reflectance spectra
(0.35 p + 2.50 y) of laboratory prepared samples involving
mixtures of malachite, ferric oxides, kaolinite, and MnO
2
were scaled to unity at 0.87 p and divided by each other to
generate a set of ratio spectra shown in Figures 1-48 in Appendix III.
Albedo differences among the spectra are mini-
mized by normalizing each spectrum by the 0.87 y intensity
value.
This wavelength was chosen because the spectral al-
bedo in this region is essentially unvarying due to characteristic ferric oxide absorptions (the 0.85-0.89 p electronic
transitions).
Since the albedos of naturally occurring
soils are not expected to vary greatly over a small range
(on the order of 1/2 km), each ratio spectrum was generated
at a constant amount of MnO 2 with either the amount of malachite or ferric oxide being varied.
This was done so that
effects arising from the variation of one constituent could
be easily analyzed from the ratio spectrum.
In the interest of conserving space, the ratio spectra presented here were chosen so that the intensity values
at wavelengths shorter than 0.5 p are less than 1.0.
In
most of the ratioed spectra, this criterion is met by placing
the sample with the greater concentration of ferric oxides
in the numerator.
This criterion breaks down at high concen-
trations (-50%) of hematite (see Figures 17, 22, and 29 in
Appendix III) because the 64
A + T
electronic transition
lg
2gelcrnctasio
at 0.63 p gains intensity relative to the 6A g 4 A g 4 E
transition at 0.53 y with increasing hematite concentrations.
Thus, a ratio spectrum containing 50% hematite in the numerator and 20% hematite in the denominator will generally have
ratioed intensity values greater than unity below 0.63 p.
The graph of intensity ratios.
The determination of relative
copper differences from natural soil sample spectra may require visual comparison with representative laboratory ratio
spectra.
However, the copper signature sought by visual in-
spection may be quantified and a method devised which will
determine the relative copper concentration and will hopefully be at least as effective as visual inspection.
Figure 20 is a graph of ratioed intensity values-1(0.75 p)/I(l.20 p) versus 1(0.65 p)/I(0.75 p)--calculated
from the normalized ratioed reflectance spectra of mixtures
containing kaolinite, ferric oxides, and malachite.
The
spectra were ratioed at a constant amount of MnO 2 ranging
from 0% to 50% by weight.
The 0.65 i/0.75 p ratio was chosen
to measure the depression of the 0.75 p ferric oxide reflectance peak by the 0.77 p electronic transition in malachite;
the value at 0.65 p was chosen because the intensity values
60
Figure 20.
Graph of intensity ratios, 1(0.75 y)/I(l.20 p)
versus 1(0.65 p)/I(O.75 -f), calculated from
ratio spectra of laboratory simulated soils.
Region I depicts relative malachite differences when there
is more malachite in the numerator spectrum.
Region II depicts relative malachite differences when there
is more malachite in the denominator spectrum.
Ferric oxide differences were calculated from ratio spectra
containing more ferric oxide in the numerator, except for
the points denoted by "Z" which contain more hematite in
the denominator. See text for full discussion.
Graph Key
Relative malachite concentrations*
Region I
Region II
5%/l%
10%/l%
10%/5%
15%/10%
20%/5%,15%/5%
20%/10%
20%/15%
50%/20%
50%/5%
Relative ferric oxide concentrations*
Graph Code
10%/5%
20%/10%
20%/5%
50%/5%
50%/20%
20%/50% (hematite)
wt. % in numerator spectrum/wt. % in denominator spectrum.
Special
Points
Points
Special
Symbol_
Represented
Symbol
Represented
M,R
h
M,0
D,H
i
D,F
N,0
U,vW
O,P
k
KY
F,G
1
v,Y
C,F,I
m
VX
L,0,S
n
I,J
nnuiluinninmuuin
II.~niionmioninnni
Fn-niininnon
Iminoigion
niinionininniginninin~~~~
I-.=i~~~an==i.
lignininnjg~mniggg-Mnin~gpli~nnnillinin
nnu.
miuimlli
m.I~nimnlllmieln
------------
Tie.m.-uinjirIiis
-v-q-n
4
Nl -4nvw
-pTmls-=I:.:=-n.I
UnonXprnnliinn
mnnnn-noi7nninni+NII
==I=.-~~
uminileninemnimiimumonen5m
I.im
====|====.====::==::::::===7=:1::::::7=:::=1-=::::=:::
iniinimininninini....M~~~~~~~InIn~numn~niiinininnimniini
iltjliiingginpnnnlmniijiinimncniliilnil
--
gmIIILun
..
IlleennuenioinininonnnninminnrS-
geliimuliinillnillumaninnemingl~I
.i~~~~~~~~
Ann^ f%4-
lin
xm it u
I:=:L
n-n-----in
..
-=ml|=-
Ininnlin
I1=:iimrm=:-en~~irIlr
mj"++lin+
ninlpnirnnnio.Nn.
ir--inn--iin+
n-jAjn-=N
- - -in-nn--n-P--
r-==
at shorter wavelengths may vary greatly due to the relatively
6
4
4
strong A g
A g
E transition occurring at 0.53 y in
hematite and 0.45 y in goethite.
The 0.75 p/1.20 p ratio was
chosen to reflect the slope change between 1.0 and 1.5 y due
to the 1.08 p electronic transition in malachite.
It can be seen from inspection of Figure 20 that
points reflecting relative malachite differences cluster
within a diagnostic area apart from the points reflecting
ferric oxide differences.
Increasing relative concentrations
of malachite systematically affect the intensity ratio values.
The ratio of two spectra containing much smaller relative
amounts of copper than are depicted in Figure 20 is expected
to show a corresponding lack of spectral contrast, i.e., the
two intensity ratios would tend to cluster closer to unity.
C H A P T E R
V I
SPECTRA OF NATURALLY OCCURRING
COPPER-BEARING SOILS
Success of the remote-sensing regime developed here
is determined from field testing.
Currently, there does
not exist a spectrometer capable of testing this method in
the field, but a simulated field test can be modeled in the
laboratory.
Spectra of surface regions are simulated by
measuring the spectra in the laboratory of soil samples collected from two copper bearing districts.
Ratio spectra,
generated by dividing the spectra of representative soil samples, simulate the spectra that would be obtained from
ratioing the spectra of adjacent surface regions.
The ratio
spectra are analyzed for their relative copper contents.
The
results are correlated with the measured copper concentrations of the soil samples, thus testing the sensitivity of
this technique in practical applications.
Description of Soil Samples
Soil samples were collected from the regions of two
porphyry copper deposits--the Mineral Park (Ithaca Peak) and
Silver Bell mining districts.
A brief description of these
deposits follows, and their locations are shown in Figure 1.
63
Mineral Park samples.
Ithaca Peak is a mining district
situated on the Mineral Park property owned by the Duval
Corporation.
Thirty-six soil samples were collected at 1.5
meter intervals in a 6x6 grid pattern as shown in Figure 21.
The samples were obtained near the Ithaca Peak stock, and
the exact location is shown in Figure 22.
Assayed copper
and molybdenum values in this area are on the order of 0.5%
and 0.03-0.05%, respectively (Clancey Wendt, personal communication).
The predominate rock type is a quartz-feldspar
gneiss, and the Ithaca Peak stock is pervasively sericitized
or argillized.
The entire Mineral Park area is surrounded
by a zone of propylitic alteration.
A complete description
of this deposit is given by Eidel et al. (1968).
Since the soils were collected at only 1.5 meter
intervals, one sample representative of a 3x3 grid portion
of the entire area was made by combining the nine samples
contained in this area into one representative sample (see
Figure 21).
Four samples, each representative of one
quadrant of the grid, were sieved to contain particle sizes
less than 125 my. in diameter.
Silver Bell samples.
Twenty soil samples and five rock chip
samples were collected from the North Silver Bell district,
owned by the American Smelting and Refining Company (ASARCO),
shown in Figure 23.
Copper mineralization occurs in hydro-
thermally altered volcanics--alaskite, dacite, and dacite and
AS6
10
86
0
D6
C6
0
0
A C46.
1
I
A4
84
F6
0
0
DF46
C5
I
E6
I
C4'
D5
0
E5
0
FS
04
14
e-
~1
----
A3
0
83
0
C3
D F 13
A C13
A2
I
At0
82
81
e
C2
F2
.0
I
I
I
I
I
CI
e
)
I
It
Dl
El
0
Ft
O.
SCALE
5 FEET
Figure 21. Soil collection grid of the Mineral Park samples (courtesy C. Wendt, Duval
Corp.). Nine soil samples were combined to
give the sample representative of a 10' x 10'
(3 x 3 m) area. The soil collection site is
shown in Fig. 22.
--
66
AA
r
AR TIN
RIDGE
U
PE AK-
in Feet
\Scale
TUROUOIsE
Figure 22. Topographic map of the Mineral
Park district (Arizona) depicting the soil
collection site (e) and the present mining
pit
(After
perimeter.
Eidel,
et
al,
1968).
67
0WW
4
CC*
-o o
W
a- w
*
*i
I
w1
a
..-
o
Lo
-
x
a
-
,J
o0-
-
-
z
.
--...
-...
Figure 23. Map of the Silver Bell district (Arizona)
showing the approximate area (shaded) of the soil and
rock chip collection sites (see Fig. 24).
(After Titley and Hicks, 1966).
68
quartz-monzonite porphyries.
Potassic, phyllic, and propy-
litic alteration zones are all present at Silver Bell, but
the copper tends to be concentrated mostly in the phyllic
zone.
Mining operations have been confined to the El Tiro
and Oxide pits which consist of tabular chalcocite blankets
formed by twofold to threefold enrichment of copper contained
in the primary mineralization.
Richard and Courtright (1966,
in Geology of the Porphyry Copper Deposits) have described
the geology of Silver Bell.
Figure 24 is a topographic map (courtesy of ASARCO)
showing the sites where the soils and rock chips were collected.
(Soil samples are prefaced by "SS"; rock chips by
"GOC.")
The map also delineates the zones of hydrothermal
alteration.
For each sample site, ten small samples at an
estimated weight of 10-15 grams were collected within a 15
meter radius and at a depth of a centimeter into the soil.
The small samples were combined to give the sample for that
site, the sites being situated at least one hundred meters
apart.
Some samples contain dark organic matter, most of it
being roots of "resurrection moss"
munication).
(J. Galey, personal com-
The sieving process separated most of this or-
ganic matter into the larger particle fractions (125 mp to
1000 mp diameters), especially in the 125-500 my size range.
Five rock chip samples were collected near the soil
sample sites indicated in Figure 24.
Each sample was ob-
tained by collecting 10-20 rock chips of approximately 50-100
p
K0
I
4
to
p
-'-Figure 24. Topograghic map of
!the North Silver Bell area
depicting alteration zones and
Athe soil and rock chip sample
sites (courtesy ASARCO). Soil
samples are denoted by "SS";
samples are denoted
by "GOC".
Location: Vaca Hills Quad.
Sections 33 & 34
Township 11 South
Range 8 East
SScale:
s(
//
jj
1/
rock chip
- .....
Air
P- ---
Nt
I
A
lk
-h
N
lip f 0
m
FT
wr
00#v JIM %A
Jv
N 1
;
Lox
f~N\
-
mat
4I~j
%,
,00
/
METT
00o
7100,
TGO
G$1GO0
'N'I
4
8
S
"N'~
--
SS5
(I1
(##3
c
P'T4
7A j
4'
>
si
.......
J
s
7
s
20
1
I
'I
\
1001,
,-
Jr
K
~
1/',,
4"T
~
£KA
-
0
P
.0000
grams within a 15 meter radius.
The combined sample was
finely pulverized for assay (performed by ASARCO).
Since the
copper concentrations in these samples were predetermined,
they were used here as standards in the XRF analyses of the
soil samples (Appendix I).
Reflectance Spectra of the Soil Samples
Reflectance spectra (0.35-2.50 p) of soil and rock
chip samples collected from the Mineral Park and Silver Bell
porphyry copper districts are presented in Figures 1-7 in
Appendix IV.
The smallest particle sizes were used to
generate the spectra: the Silver Bell samples were all less
than 45 p, and the Mineral Park samples were less than 125 y
in size.
Comparison with laboratory spectra.
Laboratory samples (de-
scribed in Chapter V) chosen to model natural soils contained
mixtures of ferric oxides, clays, and a substance to lower
the spectral albedo (MnO2 ).
Examination of the natural soil
spectra shows that the laboratory generated spectra (Figures
5-34 in Appendix II) accurately represents them.
The pres-
ence of clay is inferred from the natural soil spectra by the
intense OH and H20 vibrational features at 1.4, 1.9, and
2.2 p.
The intensity of the 1.9 p feature indicates a water
retaining clay, such as a montmorillonite.
A mixture of fer-
ric oxides is indicated by characteristic electronic transi-
tions near 0.5, 0.6, and 0.85 p." The exact positions of
these bands indicate a more goethite rich, rather than
hematite rich, mixture (refer to Figure 14),
especially in
the Mineral Park samples.
The albedos of the soil samples measured at the
highest reflectance value in each spectrum ranges between
45% and 50%, indicating the presence of at least one albedo
lowering substance such as the organic matter noted during
sample preparation, desert varnish, or an opaque mineral.
If an opaque is responsible for the lowering of the albedo,
then comparison with laboratory spectra (Figures 5-32 in
Appendix II) containing MnO 2 (an opaque) indicates that it
is present at 30-35% by weight in the natural soil samples.
The rock chip spectra (Figure 7 in Appendix
IV) show compositions similar to the soil samples, although
their albedos are generally higher.
The rock chip samples
were pulverized, and their particle sizes were estimated by
inspection to be smaller than the 0-45 p particle size fraction of the soil samples.
Smaller particle sizes enhance
scattering effects which lower the spectral contrast and
increase albedo (Adams and Filice, 1967).
The spectrum of rock chip sample G0C43 (Figure 7
in Appendix IV) is significantly lower throughout the visible region compared with the other rock chip spectra.
The
assayed value for copper in this sample (3200 ppm) is about
ten times greater on the average than the other rock samples.
Copper in this sample may occur in opaque oxides such as
cuprite, Cu 2 0, or tenorite, CuO.
Although the reflectance
spectrum of tenorite has not been measured, Salisbury and
Hunt (1971a) showed that cuprite is a trans-opaque mineral,
i.e., it is reflecting in the infrared region but highly
opaque in the visible, which could account for the spectral
behavior of QOC43.
Effect of particle size.
It was previously mentioned that
the smallest sized particles control the albedo and band
contrast of a spectrum.
The effect of particle size is
demonstrated in Figure 25, which shows reflectance spectra
of a representative soil sample (SS2) at four different
particle sizes.
Note that the albedo increases with de-
creasing particle size due to enhanced scattering.
The
spectral contrast, however, decreases with decreasing grain
size.
The H 2 0 vibrational features near 1.4 y and 1.9 p are
more intense in the spectra of the larger sized particles,
which may be due to the presence of small water inclusions.
Figure 26 represents the spectrum of a mixture containing
equal amounts of the three smallest particle sizes (0-500 y)
divided by the spectrum of the smallest particle size
(0-45 p).
Ratioed intensity values tend close to unity
across the entire wavelength region.
This is probably due
to the fact that the smallest sized particles tend to coat
the larger particles, but it shows that the spectral
50
40
30
20
10
I0
I0
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(L)
Figure 25. Reflectance spectra of a soil (SS2) showing
the effect of particle size. See text for discussion.
74
cd
1.50-
LCD
1.00-
-P
.H
0
- 050-
I
050
I
I
1.00
I
I
1.50
I
I
200
Wavelength (y-)
Figure 26. The effect of particle size as demonstrated by
the ratio spectrum of a soil sample (SS2). The numerator
spectrum contains 0-500p sized particles; the denominator
spectrum contains 0-451, sized particles.
properties are controlled by the smaller sized grains.
The spectra of large areas.
The spectra of several adjacent
soil samples taken from the Silver Bell district were averaged together to form a spectrum representative of a larger
single area containing those samples.
The averaged spectrum
should approximate the spectrum of the large area since the
soil spectra themselves are very similar.
(If they were
very different, then this would not be the case as some
spectral bands, especially electronic transitions, are
stronger than others--thus, a spectrum weighted according
to the optical depth/grain size ratio would be more accurate.)
The averaged spectra, each representing a 1/4 - 1/2
km area, were generated by averaging the spectra of the
following soil samples:
SSAV1 = SS1, SS2, SS9
SSAV2 = SS3-SS8, SS10
SSAV3 = SS12, SS14, SS15, SS17, SS18
SSAV4 = SS19, SS20.
The averaged area spectra are shown in Figure 27.
The fact
that these spectra are very similar is shown by their ratio
spectra presented in Figure 28.
Thus, over a large area,
the reflectance spectra lose individuality.
This means that
in practical application, any reconnaissance technique would
require a high spatial resolution, probably on the order of
the soil sample dimensions (-15 meters).
76
-
1 - SSAV1 (Average of SS1, SS2, SS9)
2 - SSAV2
(Average of SS3 - SS8, SS10)
3 - SSAV3 (Average of SS12, SS14, SS15, SS17, SS18)
4 -
SSAV4 (Average of SS19, SS20)
0.25b-
0.25
0.25-
0.2 5i-
,
0.50
I
0.75
I
1.00
I
I -I
1.75
1.50
1.25
2.00
I
2.25
Wavelength (y)
Figure 27. Spectra of the average of several soil samples representing the spectra of large sample areas.
7
__
~
1 -
SSAVl/SSAV2
2 -
SSAV3/SSAV2
3 -
SSAV4/SSAV3
4 -
SSAV3/ SSAV1
5 -
SSAV4/sSAV1
,00
1OOH-
2
00
3
1.OOFH
4
1.0OH-
I
0.50
I
0.75
I
1.00
I
1.25
I
1.50
1
1.75
|
2.00
2.25
Wavelength (P)
Figure 28. Ratio spectra of the averaged spectra representing
large sample areas (shown in Fig. 27).
Ratio Spectra of Representative Soil Samples
The reflectance spectra of the
Presentation of spectra.
soil samples were normalized to the intensity value at 0.87
p, and then representative samples were divided, generating
a set of ratio spectra.
Twenty-four ratio spectra, shown
in Figures 5-28 in Appendix V, were generated from the
Silver Bell soil spectra, with most of the samples involved
in each ratio being less than 1/3 km apart.
Figures 1-4
in Appendix V show four ratio spectra generated from the
Mineral Park soil spectra.
All of the samples here lie next
to each other, the mean sample distance being 5-6 meters.
Predictions of relative copper concentrations from intensity
ratios.
The ratio spectra were visually inspected, but no
copper bands were plainly evident.
1(0.75 y)/I(l.20 p)
Intensity ratios,
and 1(0.65 p)/I(O. 7 5 y),
were calculated
for each ratioed spectrum and plotted as shown in Figure 29.
This graph was compared to the same graph generated from the
laboratory ratio spectra (Figure 20) in an attempt to predict the relative copper concentrations of the soil samples
before analyzing for their copper contents.
Comparison of
the graphs show that thirteen samples plot within the laboratory designated regions showing copper differences; the other
samples may contain relatively lower amounts of copper which
would yield undiagnostic ratios.
Figure 29.
Graph of intensity ratios generated from the
soil sample ratio spectra. The outlined
regions I and II are diagnostic of malachite
differences as explained in the caption to
Figure 20.
Graph Key
Silver Bell soil sample
ratios
SS1/SS2
SS5/SS8
SS8/SS7
SS6/SS3
SS6/SS4
SS8/SS6
SS5/SS6
SSll/SS10
SS14/SS12
SSll/SS12
SS18/SS17
SS19/SS16
SS17/SS15
SS16/SS15
SS19/SS20
SS10/SS7
SS11/SS2
SS18/SS12
SS15/SS2
SS16/SS2
SS20/SS2
SS9/SS2
SS13/SS9
Graph Code
Copper
concentrations
370/285
1585/305
305/710
395/270
395/710
305/395
1585/395
290/760
175/95
290/95
135/100
145/55
100/275
55/275
145/105
760/710
290/285
145/95
275/285
55/285
105/285
245/285
125/245
Duval soil sample ratios
ACl3/DF46
ACl 3/AC4 6
DF13/DF46
DF1 3/AC4 6
930/1140
930/1250
1075/1140
1075/1250
(ppm)
Figure 29
80
gensoiimimiinllisiisilntilannislilinniiAisli
IIIIIIIII"IIIIIIIIII--
-
II----IIIIIIII
IIIIII
--
......'-- --------- ---M---IIIII-IIIII-I-i1-III-l--III--Iiii IIIIII
- ------------EEMMEEMMMMMMMEMMMMMMMMM
----- -|
i
n
n
n
n n
+Ifin
inMM
-- i----MM
MM
EM
EM
ME
M-------------MM
il~
n n---i
-------------- --ni
.
inn4n
MINIIII!I!II~JIII!IIIIIII
n li4I
MMMM
EME MMM
MMMMMMM
EMEM
MMM
-- ----------.-----------
il~
ili
H~n
MMM
EMEM
MMM
i
n lliiiiiiill
I--IIIIIIIIIIIIII
MMM
MMMLMM MM
BE MM
MM
liinn--n----
MM---MM
---------M MMM
niM ini-n---ni--------------iliiliil
inn
!!IIIIIIIIIII
IIMMIM-n------------I-I--IIIIIII
EMEErJ
4
MMMMEMMMEMIMEIMMMMEMLM
AMMEMI.,MMME
iiiiin
iinnilisinininnilig-
IViiiiiniininnsii
M~ninilinnEMIIINMI"IIIIIIIII~niIITI
MMMMMMMEMMEMMMMMEMMMMEMEMMMM-4
IMEMME^MM
IiniliniiniliinnnifiEII~liligligliglgiigigr
MqMEMM
MMEraMMMMMMMEMEMEMMMMMMEEMMEMMM---MMMMMMM-M--M--M
-----
n------------
----------
----
---
-----
l
M
M
III
n------n--------i.-------iH
i
I I ..
----
------!IIIIIIM!II
EMMMMMEEMBEEMMMEMMMMMEMESEAMMMM-MM------EMMM-------MMM
I~li-n--n----n---n---il
i+44#niiiilii-li---n-in-n-.
llil---.-- ininN~onMMIIIIIMIMMMIMIIIIII----II----g--g--
i-n
llrVnMI
--
--
-
--
The determination of copper concentrations.
Standard XRF
analysis, described in Appendix I, was used to determine
the copper content of each soil sample.
The mean copper
concentration of each sample is listed in Table 2 in Appendix I.
Most of the Silver Bell samples have copper con-
centrations onthe order of 100-500 ppm; the Mineral Park
samples have higher copper concentrations, on the order of
1000-1200 ppm.
Of the thirteen samples that showed relative
copper differences in Figure 29, seven were correctly predicted, meaning that the copper concentrations in the soil
samples are too low to be diagnostic.
Graphs of Intensity Ratios
Graphs of intensity ratios are presented in the
following sections.
In the interest of conserving space
and avoiding repetitive explanations, the format of the
graphs will be described here.
Each graph consists of ratioed intensity values,
1(0.75 p)/I(l.20 y) and 1(0.65 y)/I(0.76 p),
each taken
from the ratioed spectrum of normalized (at 0.87 y)
spectra.
soil
The 0.75 y/l.20 p intensity ratio is plotted
against the 0.65 p/1 .2 0 p ratio.
The ratioed spectra were
chosen to contain the sample with the greater copper concentration in the numerator.
Absolute concentration versus relative concentration.
Figure
30 is a graph of intensity ratios which compares the absolute
82
Figure 30.
Intensity ratio graph comparing relative and
absolute copper concentrations. The intensity
ratios are calculated from soil sample ratio
spectra which contain more copper in the
numerator sample spectrum. See text for
discussion.
Graph Key
Absolute concentration (ppm)
Relative concentration
0-200
200-500
1.0-1.5
V
1.5-2.0
w
2.0-3.0
3.0-5.0
>5.0
a
b
c
d
e
>1000
U
-1.0
Special
Symbol_
500-1000
Points
Represented
D,G,I
B,J,S
B,H,Y
B,0,Z
H,I
Figure 30
ILnoInIi
MooinnnII
8
IoI
nInnlinin
III~nlinninP"..M!H!MM!IMI+
I IIMP.MHII
Iinnin
iiilinntniininn"t"
ME!MEN!~~~~~I
I=IMPIn"".I..L"H~gnM
===n==P===on====.n==.====
EI
hnni.~innnniliinnnonMI
Ininiinlinnoiliinno~IT TUI
EIM!
!in!MMI
inlPillnninnininilnnoP.IninliMIM.Innninili
MMMMMME
EMMM.EMMMMMEIMI.M~non=
!==E!!!n!!E=!!!EP.=.i.. I
U
=
Ii!!=
WnnnI!Mni.WgiMMPIMIMIIM
!!IIMM~nP.M~inIME..noininniEEP!EE
f
M~niIM~nniiP~nliInlll=I
MINIMP.IIMI1171n11
M!IM~.9
iiliCnii
innin
iliinninniinn
rnil
1
1nnnn11iin
lPIrnii
_II~inInin
HE~iinlinininlnnNI~nPInnniP~liiilliin®ri
vInniME
M~oi.~in!MMM
concentration of copper to the relative copper concentration
in some 60 Silver Bell soil sample ratio spectra.
The samples
involved in each ratio were chosen to lie at relative distances less than 1/2 km.
The absolute concentration of a
sample ratio is defined to be the copper concentration of the
sample involved in the numerator of each ratio.
The graph
shows no correlation between the absolute and the relative concentrations at the levelof copper detectedin the soil samples.
Soil mineralogy versus sample distance.
A graph of some 70
Silver Bell soil sample intensity ratios is shown in Figure 31.
This graph compares soil mineralogy with relative sample distance, i.e., the distance between the two samples used to
generate a ratio spectrum.
The graph shows that intensity
ratios made from samples contained within the same zone of
alteration (refer to the Silver Bell topographic map, Figure
24) all cluster near unity.
Samples taken from different al-
teration zones have intensity ratios which are quite different
from unity, reflecting intrinsic differences in soil mineralogy rather than differences in copper concentrations.
Thus, in
practical application, sample areas should all be ratioed within the same zone of alteration.
Since the size of an altera-
tion zone depends on the individual deposit, a reasonable approach that would insure meaningful results would be to ratio
sample areas that lie fairly close to each other (on the
order of 1/2 km in the case of the Silver Bell samples).
85
Figure 31.
Intensity ratio graph comparing soil mineralogy
and relative distances between soil samples.
The intensity ratios are calculated from soil
sample ratio spectra which contain more copper
in the numerator spectrum. See text for discussion.
Graph Key
Sample distance (km)
Soil mineralogy
1/4
1/4-1/2
1/2-1
1-2
Potassic
A
B
C
D
Phyllic
E
F
G
H
Propylitic
I
J
K
L
Potassic-Phyllic
M
N
0
P
Potassic-Propylitic
R
S
T
U
Phyllic-Propylitic
W
X
Y
Z
Special
Symbol
a
b
c
d
e
f'
g
h
i
Points
Represented
B,J,K
F,I,J
B,E,I
N,X
BMX
B,Y
A,T
AM
A,N
Figure 31
HinlilI
MI~~nnininn~I
En inilii
== =i===I
IE
Ii L
I I I
IInJCnili"
I
IIMI
I
IM
8
lilinnlnMIIniin
I I 11in1nnn1j1
..
=P
I======.1=
MM
!III!E!M!!!EininE!I!!!!HiR!nnE.I...nn!!
innnl
..
IIMM
giiig
n..
InInIniiliiniPgni
Iiiniinl
IP.Miinnilln~nnn.nlMMnniiin
Immnneni
minenomenmiP-mm
VnnnIn!.
!IE~ninni!!!nnE~ii!!Enin.-IF,
Pli""EM
iniini~~~ininilinilliii~~LIiin
iin
EIIMMM~~nn!IIMEMME~INEEI~n!Mnn.!M
E!inEE!!~n!!!E!=in~nnia!!!!n!!+!
E!!IMEINEE!!IonininnMM~nLIn!IE!IM
MEMEMMEMIEMP.IMEIIMI~noINinEIMi
niiniinniinnnininiiiniiniioninP.niini1i
E!M~nP"EMMEM!!IEninEI~nEMIIn!Minho
====Eig==========ll'======
II IT 111M....
ME=IMTnilnienFMMI
innioinnnM!!n.!EIIn!!MIE!MMEMIIMM!5IM
inliniilliinl
iI
~I
IlIra-%Iinin
InIIIM
mem=omm~n~mnminiufill I======imm=I
==PHil===
iin
iniiini
njnM
I
Conclusions from intensity ratio graphs.
Analysis of the
intensity ratio plots (Figures 30 and 31) show that, in
practical application, samples taken from relatively far apart
should not be ratioed because the mineralogy of the soil is
not constant over a large distance.
Since there was no
correlation between the absolute and relative copper concentrations, the amount of copper analyzed in the soil
samples (<2000 ppm) is probably below the detection limit
of this technique.
Inspection of Figure 20 shows that
points diagnostic of copper cluster within a certain area,
as delineated on the graph.
A relative copper difference
of at least 2:1 is needed to produce diagnostic intensity
ratios.
The minimum absolute concentration of malachite
needed to be detectable ranges between -5 - 10%, depending
on the overall spectral albedo.
(At lower albedos, higher
concentrations of malachite are needed because of the corresponding decrease in spectral contrast.)
This means that
copper concentrations on the order of at least 2000-5000 ppm
are needed to be detectable by this technique.
C H A P T E R
V I I
INSTRUMENTATION
Instrumental Requirements
In developing the technique of reflectance spectroscopy for the detection of copper deposits, laboratory
studies are necessary to define the instrumental parameters
needed to delineate copper bearing regions.
The laboratory
studies presented in previous chapters of this thesis adequately define the important instrumental parameters which
are intensity precision, spectral range and resolution, and
spatial resolution.
Determination of these parameters sug-
gests the method of reconnaissance and the type of instrument needed to detect surface copper anomalies.
The required wavelength range is determined from
the optical properties of the weathered (oxidized) phases
of copper which most commonly occur in surface anomalies,
e.g., malachite, chrysocolla, azurite.
Electronic transi-
tions diagnostic of these phases occur in the 0.4-1.20 y
region.
(Infrared transitions due to ligand (CO3 and OH)
vibrations are non-unique.)
The amount of information regarding mineralogy and
modal abundances obtainable from a spectrum depends on the
extent to which absorption band positions, strengths, and
88
asymmetries can be resolved.
A spectral resolution of 5 to
10 nm is required between 0.4 y and 1.50 p.
The intensity
precision should be in the neighborhood of a few tenths of
a percent (relative reflectance) across the entire wavelength region.
Greater spectral resolution or intensity
precision would not much improve the interpretation, but a
degradation of either parameter would significantly limit
the interpretation.
The spatial resolution depends on the size of the
typical copper anomaly.
The ASARCO soil samples collected
from the Silver Bell area used in this thesis are representative of a circular region thirty meters in diameter.
This
value would be a reasonable upper limit to the spatial extent since spectra taken of larger areas tend to lose
spectral individuality.
Available Instruments
Possible instruments available at the moment for
remote sensing of the earth include the Skylab S-191 spectrometer and a variety of spectral imagers (ERTS-1, Mariner,
etc.).
None of these, however, may be used to detect copper
by the remote sensing regime proposed in this thesis.
The
spectral imagers all employ broad band filters (~100 nm)
which certainly do not give the required spectral resolution.
The S-191 spectrometer had problems in instrument calibration
as the data was found to be shifted along the wavelength axis.
90
Nonetheless, thereis a proposed instrument system that does
meet the required specifications.
The Geochemical Orbiter Spectrometer
The reflectance spectrometer designed for geochemical orbiting missions of planetary objects (McCord et
al., 1975 (NASA proposal) designed a spectrometer for lunar
orbiting) has instrumental specifications that meet (or
exceed) those required here for the detection of copper deposits.
The spectrometer is designed according to the above
mentioned proposal to measure the spectrum of light reflected
from the lunar surface along a ground track.
The spectrum
would then be interpreted in terms of the mineralogy and
composition of the regolith, and would be comparable in
quality to the presently available laboratory spectra of
lunar samples.
The geochemical orbiter spectrometer is designed to
obtain high resolution spectra of the surface in 220 channels between 0.35 p and 2.50 p with intensity precision of
0.1 to 0.5 percent.
The spectral resolution ranges between
5 nm at the high energy end of the spectrum (-0.35 p) to
20 nm near the low energy end of the spectrum (-2.50 y).
spatial resolution of 200:1 is attainable (0.5
tion from an altitude of 100 km).
A
km resolu-
In order to obtain a high
signal-to-noise ratio (103 ), a 0.5 second integration time
is used so that the instrument will not be photon limited at
an altitude of 100 km.
The ground track rate is 1.7 km/sec
which yields a nominal spatial element of 1.35 km x 0.1 km
at 0.5 second integration time.
At a 100 km altitude, a
circular aperture 1.0 mm in diameter situated at the focal
point is required to give 0.5 km spatial resolution.
Instrumental Specifications for Copper
Reconnaissance
Instrumental parameters. A high spatial resolution should
be used to locate small highly rich copperareas.
bearing
To attain 10 m resolution using the design specifi-
cations for the geochemical orbiter reflectance spectrometer
requires an instrument flying altitude of 2 km.
Hence,
aircraft reconnaissance is the most feasible method to attain
the required spatial resolution of 200:1.
The spectral
resolution and intensity precision of the geochemical orbiter spectrometer meets the specifications for this application.
The spectral range in this case may be limited to
0.35 p to 1.50 p (instead of 2.50 p) since the infrared end
of the spectrum does not contain information necessary for
this application.
If the average speed of a low-flying aircraft is
about 140 mph (62.5 m/sec), then an integration time of 0.16
seconds or less is obtained.
To attain an integration time
of 0.1 seconds (which is one-fifth of the integration time
for the geochemical orbiter spectrometer) means that the
92
light reflected from the earth's surface should be roughly
five times as great as that which is reflected from the moon.
The Bond albedo of the moon is on the order of 6-12% and
that of the earth ranges between 30-50%.
Thus, 0.1 second
is a reasonable integration time for use here to give a high
signal-to-noise count.
The nominal spatial element as
measured by the spectrometer, then, would be an area 16.25 m
x 10 m and would require an aperture 20 y in diameter for
10 m resolution.
Instrumental calibration.
Preflight instrument calibration
should include defining maximum (100%) reflectance on high
albedo soils at low phase angles along the noon meridian
(high sun angles).
Inflight calibration will not be neces-
sary as ratioing the spectra will eliminate instrumental as
well as most of the atmospheric contributions.
Location of the ground track.
During flight, a camera may
be used to precisely locate the ground track.
Its optic
axis should be bore sighted with that of the spectrometer.
The camera's field of view should be larger than that of the
spectrometer's (10 m) in order to put the spatial element in
perspective with the surrounding area.
A camera with a field
of view on the order of 200 meters is sufficient for this
purpose, and would require that a picture be taken every one
second to provide sufficient frame overlap at the expense of
a minimum amount of film.
Data collection.
Inflight data collection should consist
of uncalibrated reflectance spectra of contiguous 16.25 m
x 10 m segments along the ground track.
Each spectrum will
be accompanied by its time of measurement and phase angle,
and the flight altitude and ground speed may be recorded,
say, every tenth frame.
The surface location should be
recorded on photographic film by the camera attached to the
spectrometer.
The data may be ratioed immediately during
flight according to a prescribed ratio pattern, i.e., contiguous spectra may be ratioed, but ratioing every fifth or
tenth spectrum will produce greater spectral contrast.
Candidate spectra may be ratioed later on the ground after
visual inspection of the ground track segments.
Obstacles and Uncertainties Involved in
Field Measurements
Vegetation.
There are some obstacles and uncertainties in-
volved in the field measurement of spectra.
The major ob-
stacle is the effect of vegetation on reflectance.
The
porphyry copper deposits in the United States are situated
in desert environments so that the extent of surface vegetation is minimal.
However, local areas may be rather
heavily vegetated with cactus and sagebrush, and such areas
should be avoided.
For those areas not heavily covered by
cactus and brush, desert varnish ean cover a large
part of the surface of many desert floors. Desert varnish
is a dark material which can coat up to the entire surface
of an exposed rock, and it is mainly composed of manganese
and iron oxides (Engeland Sharp, 1958; Potter and Rossman,
1977).
The presence of desert varnish would lower the
overall albedo of a reflectance curve and thus decrease
spectral contrast.
The concentration of desert varnish does
not vary greatly on a local scale, and so its effects may be
minimized by surveying areas of low concentrations (525%).
Lichens and moss may also abound in certain regions, and
highly lichenous areas should be avoided.
There are several
types of lichens which are able to assimilate copper and
have been reported to change color dramatically according
to the amount of copper assimilated (Czehura, 1977).
A
study of the spectral properties of these indicator lichens
may provide a useful and worthwhile addition to this proposed remote sensing regime.
Representative spectra of
selected desert plants and lichens are presented in the
discussion chapter (refer to Figure 36).
Absorption fea-
tures appearing in the plant spectra are discussed in relation to their contributions to soil spectra.
Phase angle effects.
An uncertainty in the measurement of
spectra involves the effect of phase angle variations on reflectance.
Laboratory spectra measures the overall Bond
albedo of the reflected material (using an integrating
sphere), while inflight spectra are measurements of the
95
geometric albedo, that is, the albedo measured at a certain
phase angle.
Variations in phase angle affect the extent of
shadowing and the overall slope of the spectrum.
Band
centers and symmetries are also slightly affected by this.
The effects of phase angle variations on reflectance would
be minimized by flying at low phase angles.
A 10* phase
angle will permit observations of the least shadowed portions
of the surface while not encountering opposition (anomalously high specular reflection which occurs near 0* phase angle).
Useful data may be acquired between 100 and 60* (the albedo
may vary by a factor of 50 between 0* and 60*),
and only
those spectra taken at similar phase angles should be
ratioed.
The effect of phase angle may also be minimized by
changing the viewing angle of the spectrometer.
This would
be useful to survey strongly sloping objects such as the
sides of mountains and deep trenches.
Flight variations.
Other obstacles involved in the measure-
ment of spectra include variations in flight speed, altitude,
and pitch, and variations in atmospheric conditions.
The
effect of flight variations should be negligible over 1 to
5 seconds (10-50 measurements).
Atmospheric variations will
be minimized by flying during relatively clear conditions.
C H A P T E R
V I I I
DISCUSSION
Summary
Reflectance spectroscopy has been used successfully
to determine the composition of solar system objects as
was discussed in the Introduction.
In this thesis, the
technique is used to determine copper bearing regions on the
earth.
Development of the technique for this purpose re-
quired understanding the spectral properties of minerals
which are contained in surface soils and rocks.
This was
accomplished from background laboratory studies which not
only determined the wavelength regions diagnostic of copper
bearing minerals and other representative soil constituents
such as clays, ferric oxides, and vegetation, but also determined the limits of copper concentration needed for detection.
Since there is not a field spectrometer which
currently possesses sufficient detection capabilities, a
field test was simulated in the laboratory using naturally
occurring soil samples taken from the regions of copper
deposits.
Reflectance spectra of these samples were used to
generate representative ratio spectra simulating those that
would be taken in the field.
These laboratory ratio spectra
were analyzed for their relative copper concentrations and
96
the results defined the sensitivity limit of the technique.
Development of a spectrometer with the required instrumental parameters (such as the geochemical orbiter spectrometer described in the chapter on Instrumentation) is
needed before this technique can be applied in the field.
Potential Problems in Field Application
There are some potential problems which can arise
when reflectance spectroscopy is applied in the field.
Obstacles involved in the measurement of spectra such as the
effect of phase angle, atmospheric conditions, and flight
variations were discussed in the Instrumentation Chapter.
However, a potential problem involved in spectral interpretation may arise from the presence of other minerals possessing absorption features near the wavelengths diagnostic
of copper.
Reflectance spectra of clinopyroxenes.
Figure 32 is a plot
(fromAdams, 1975) of centers of absorption bands appearing
in the reflectance spectra of common rock-forming minerals.
Inspection of the graph shows that a few minerals possess
absorption bands near the diagnostic copper (malachite)
wavelengths at 0.77 p and 1.08 p.
These minerals are the
clinopyroxenes diopside, augite, and salite.
Representative
reflectance spectra of these minerals are shown in Figure 33,
and features appearing in the spectra have been discussed by
98
Figure 32
I
I
I
I
I
I
I
I
ALMANDINE
~dsPYROXENE
. 2 ABSORPTION
BANDS ONLY
WER
. b 3 ABSORPTION
-
-
IflO'F
BANDS
METALS,
LOWER
OXIDES
. OLIVINE SCALE
FELDSPARI
GLASS
PHLOG-SAL
O0LO,
.
SEPID
*
EPID
090
CMICA
S90-
KR
K
1P-
DOP-
-SulAL
RCDIOP-
gSAL
00
Q70
cr-iNST
F8T0-
Cr00~P
050
CraOs
* JCr-DIOP
i
.
04
I
0
iCr-EST
I
0.6
Q8
W
FlOiFt
08
I I I
LO 12
1.0
I
L4
I I
I II
I I I I
L6
L8
20
22 24
L2
L4
L6
12
WAVELENGTH, MICRONS
20
Plot of centers of absorption bands in the diffuse reflectance spectra of several
common minerals. Minerals having only one main band are plotted on the lower scale.
Minerals with two or more bands in their spectra are designated by points in the main
upper field. Hydroxyl and water bands at 1.4 and 1.9 sm are not used. Bands for minerals
species and for glasses plot in separate fields. See text for discussion. Diop-sal - disosidesalite; bio-phlog - biotite-phlogopite; aug - augite; pige - pigeonite; opx -= orthopyroxene; dolo - dolomite; epid - epidote; actin-hornb - actinolite-hornblende;
trem-ged-kaer - tremolite-gedrite-kaersutite; amph - amphibole; enst - enstatite;
chlor - chlorite; Fo - forsterite; Fa - fayalite; Ab - albite; Olig - oligoclase; And
andesine; Lab - labradorite; Byt - bytownite; An - anorthite.
(After Adams, 1975).
Figure .33
I
.1
a'
0
4I-
I
0.5
1.0
1.5
WAVELENGTH (pm)
2.0
2.5
Diffuse reflectance spectra of clinopyroxenes. Iron-rich caleie pyroxenes
typically have more complex Fe'+ and Fe'-Fes+ band structure than do the orthopyroenes and pigeonites.
(After Adams,
9
1975).
100
several authors (Burns, 1970; Adams, 1975; Hunt and Salisbury, 1970a).
Briefly, the absorption feature in the 0.75-
0.80 ypregion is attributed to Fe 2 -Fe + change transfers
whose absorption intensity varies with the amount of iron
substitution.
The absorption strength of this feature is
not greater than the 1.1 p feature except in the spectra of
salite, which is more iron rich than diopside. The presence
2+
of the 1.1 y band is due to Fe
in the pyroxene Ml site,
and the presence of a fairly weak feature near 0.95 y in the
salite-diopside spectra arises from Fe + substitution in
the M2 site (Burns, 1970).
Occurrence of clinopyroxenes.
Although clinopyroxenes have
absorption bands in the regions of the malachite bands, the
presence of these minerals in the spectra of naturally occurring soils may not be much of a problem.
Disseminated
copper deposits are formed mostly in acidic-intermediate
rock facies, usually monzonite, whereas the clinopyroxenes
are formed mostly in basic igneous rock facies and diopside
is commonly found in contact metamorphosed limestone (skarn).
Areas chosen for the reconnaissance of copper will be those
containing rocks of principally acidic-intermediate composition associated with subduction zones, and so the occurrence of mafic rock facies which are more likely to contain
the clinopyroxenes should be limited in these areas.
101
Clinopyroxene discrimination from intensity ratios.
However,
should clinopyroxenes be encountered during reconnaissance,
the fact that the 1.1 y band is generally more intense than
the one near 0.75 y in the spectra of these minerals will be
evident when the 0.65 y/0.75 y and 0.75 y/1.20 y intensity
ratios are calculated.
In malachite, the 0.75 p feature is
always more intense than the one at 1.1 y which results in
the 0.75 p/l.20 y intensity ratio generally being less than
or equal to unity.
However, in clinopyroxenes, this inten-
sity ratio will generally be greater than unity since the
1.1 y band is more intense than the 0.75 y band.
The pyrox-
ene spectra (Figure 33) also show that the 0.65 y/0.75 y intensity ratio is greater than unity.
A tentative pyroxene
field generated from estimates of the 0.65 y/0.75 y and
0.75 y/1.20 y intensity ratios is outlined on the graph of
the malachite and ferric oxide intensity ratios in Figure
34,
which shows that pyroxene will probably not be confused
with malachite.
Fe +-rich clinopyroxenes: salite.
The only possible case
when a clinopyroxene may be confused with malachite is when
the pyroxene is relatively rich in Fe3+ compared to Fe 2 +
as in salite.
The spectrum of salite (Figure 33) shows that
the 0.75 P band becomes more intense than the 1.1 p band due
to Fe + enrichment.
Olivines and pyroxenes with high rela-
tive abundances of Fe3 + show exsolved lamallic intergrowths
102
Figure 34.
Intensity ratio graph of laboratory simulated
soils (Figure 20) including ratios calculated
from the spectra of clinopyroxenes. The effect
of vegetation is also shown. See text for full
discussion.
Figure 34
103
103
44-
-T-t-
Im
I
s4$
A
44 1 w
V
IA
If
IA
All
I V
11.1
v
I
/I
F
ol%1 ri
liT
V
1611
If
I
I/
.4,
1
All
V
3111.
r
IF I
IN
I L L/L
Y I J
v
V-
-ely
I --
I
YIf
+
V
of
I
ri
or
tj I
0,1
VT
-Op
4pvv
T-T
(X,
11A
104
of hematite (a-Fe 2 03 ) within the crystal structure (Deer,
Howie, and Zussman, 1962a,b).
exposed at the surface,
When these minerals are
the presence of Fe 3+ tends to break
up the crystal structure in favor of formation of ferric
oxides (Deer, Howie, Zussman, 1962a).
enriched in Fe
Thus, pyroxenes
are not expected to be common at the sur-
face since weathering processes would probably have destroyed them.
Reflectance spectra of representative desert vegetation.
The
presence of vegetation may also lead to erroneous spectral
interpretation.
Although highly vegetated areas should be
avoided during field measurements, some vegetation will invariably be encountered.
UV-visible-IR absorption spectra
of chlorophyll and other essential plant pigments are shown
in Figure 35.
The predominant plant pigments absorb in the
0.44 p region (22,500 cm1 ),
but only chlorophyll absorbs
in the vicinity of 0.65 y (15,000 cm~ )
(Gates et al., 1965).
Vibrational bands due to water occur throughout the infrared region.
The materials comprising leaves are moderately
reflecting in the green, near 0.54 y (18,500 cm 1)
highly reflecting in the 0.70-2.0 y region.
and are
The reflectance
and absorptance spectra of typical desert succulent plants
and typical lichens are shown in Figures 36a and 36b respectively.
The prominent feature in the spectra of both types
of plants is the chlorophyll absorption in the 0.65-0.70 y
region.
105
55
44
is
2- ir re'0 3
s)
40r
VW^J41.MICRO
me1MMI ars.1
Figure 35. Specific absorption
versus frequency of the principal plant pigments and liquid
water.
(After Gates, et al.,
1965).
106
94
#1
-ULTRAVO.ET
VIStOLE
a
-
------
05
as it
sl
WAVELENGTH,
MICRONS
-
is
0
(G400
INRARE0
-0
CEREUS
GIGANTEUS
AGAVAMERICANA
(LOWR
M A
E £AMvi
SWRACE)
O"MTI LAEVIS
''-'''
WA/EMAER% cm
1
&w9 I
A
SAVELENGTH.
MILLlNCA
B
Fig re16.
(A) The spectral absorptance and reflectance of the
stems of desert succulent plantt.
(B) Reflectance spectra of
selected lichens.
(After Gates, et al., 1965).
107
Effects of vegetation on soil spectra.
This chlorophyll
feature does not occur at the wavelengths diagnostic of
malachite (0.77 y and 1.08 y),
but it will affect the
0.65 p/0.75 p intensity ratio calculated from the spectra
of soils containing vegetation.
The effect of plants will
be to lower the reflectance at 0.65 p relative to the reflectance at 0.75 p which will cause a decrease in the
0.65 p/0.75 p intensity ratio.
Intensity ratios calculated
from spectra containing a relatively high proportion of vegetation will fall shortward of those values calculated from
spectra containing little vegetation on the 0.65 p/0.75 p
axis.
Assuming that malachite is present in diagnostic
amounts, a high proportion of vegetation in the spectra
would yield a 0.65 p/0.75 p ratio outside of the field
diagnostic of malachite.
The relative proportion of vege-
tation required to produce an undiagnostic 0.65 p/0.75 p intensity ratio has not been studied, and further laboratory
work is needed in order to determine the effects of vegetation.
Sensitivity of the Technique
The detection limit of the technique was determined
from Figure 20 which was discussed in Chapter VI.
It was
found that at least a 2:1 relative copper difference
needed to produce diagnostic intensity ratios.
is
The minimum
absolute concentration of malachite needed to be detected was
'108
found to be on the order ,of 5-10% (2,000-5,000 ppm Cu) depending on the overall spectral albedo.
In application, ad-
jacent spectra are ratioed in order to minimize instrumental
and atmospheric variations.
The ratio spectrum of two areas
both containing a high absolute malachite concentration, but
with less than a 2:1 relative malachite concentration would
yield undiagnostic
to unity).
intensity ratios (they would both be close
Therefore, there is a trade-off between the sen-
sitivity of the technique and minimizing the atmospheric and
instrumental effects.
These effects, however, may not vary
greatly during a reconnaissance mission so that spectra taken
from rather widely separated areas which contain different
relative malachite concentrations would produce diagnostic
intensity ratios thus increasing the sensitivity of the technique.
Higher concentrations of malachite may be needed if
a relatively high proportion of vegetation is present in the
spectra.
Although this technique is not sensitive to soils containing approximately less than 2,000 ppm Cu such as were
found in the natural soil samples studied here, it should
still be useful to identify copper deposits.
Small, localized
surface areas can contain malachite and other oxidized copper
minerals at levels high enough to be detected.
al.,
(Lovering et
1950, found that some soils collected around
the San
Manual porphyry deposit contained copper concentrations on the
order of 10,000 ppm.)
A 10 m spatial resolution for the
109
spectrometer proposed in the Instrumentation chapter should
be high enough to detect small areas that are enriched in
malachite.
Advantages of the Technique
Reflectance spectroscopy has advantages as an exploration technique over other geophysical, geochemical, and remote
sensing techniques.
It is capable of directly detecting the
presence of copper in surface soils, it can survey a wide area
during a single reconnaissance, and it can scan areas that may
be inaccessible to in situ measurements.
Potential Applications
The technique of reflectance spectroscopy may be used for
other applications.
The remote sensing regime developed in
this thesis for copper exploration may be extended to other
economically important transition metals.
Chromium is a good
candidate since it possesses absorption features at diagnostic
wavelengths.
Figure 37 shows reflectance spectra of a chrome-
diopside containing 1.2% Cr2 0 3 and chromite (Cr2
3
).
Strong,
relatively sharp absorption features arising from electronic
transitions in Cr 3 + occur near 0.45p and 0.60p.
Inspection of
Figure 32 shows that the 0.6Op feature is unique only to Cr 3 +bearing minerals, and hence chromium would be a good choice
towards application of this technique.
110
Cr DDPMDE
U.
0*
1.0
.
~
Figure 37. Reflectance spectra of
chrome bearing minerals.
(After Adams, 1975).
REFERENCES
Adams, J. (1975). Interpretation of visible and near-infrared
diffuse reflectance spectra of pyroxenes and other
rock-forming minerals. In: Karr, C. (Editor), Infrared and Raman Spectroscopy of Lunar and Terrestrial
Minerals, Academic Press, New York, 91-116.
Argall, G. and Wyllie, R. (1976). World Mining Copper Map.
Miller Freeman Publications, Belgium.
Ballhausen, C. (1962). Introduction to Ligand Field Theory.
McGraw-Hill, New York.
Bernal, J. et al. (1959). The oxides and hydroxides of iron
and their structural inter-relationships. Clay Min.
Bull. 4, 15-31.
Blanchard, R. (1968). Interpretation of Leached Outcrops.
Nevada Bureau of Mines, Reno.
Burns, R. (1970). Mineralogical Applications of Crystal Field
Theory. Cambridge University Press, Cambridge.
Chaffee, M. (1976). The zonal distribution of selected elements above the Kalamazoo porphyry copper deposit.
J. Geochem. Exploration 5, 145-165.
Clarke, 0. (1952). Geochemical prospecting for copper at Ray,
Arizona. Econ. Geol. 48, 39-45.
Creasey, S. (1966). Hydrothermal alteration. In: Titley, S.
and Hicks, C. (Editors), Geology of the Porphyry
Copper Deposits, Univ. of Arizona Press, 51-74.
Czehura, S. (1977). A lichen indicator of copper mineralization, Lights Creek District, Plumas County,
California. Econ. Geol. 42, 796-803.
Deer, W., Howie, R. and Zussman, J. (1962a). Rock Forming
Minerals: Vol. I: Ortho and Ring Silicates.
Longmans, Green and Co., London.
.
(1962b). Rock Forming Minerals: Vol. 2: Chain
Silicates. Longmans, Green and Co., London.
111
112
Dykstra, J. (1975). Detection of Porphyry Copper Alteration
by Computer Processing of ERTS-1 Digital Data.
M.A. Thesis, Dartmouth College, Hanover, New Hampshire.
Eidel, J., Frost, J. and Clippinger, D. (1968). Coppermolybdenum mineralization at Mineral Park, Mohave
County, Arizona. In: J.D. Ridge (Editor), Ore
Deposits of the United States, 1933-1967 (GratonSales Vol.), SPE/AIMIE, New York, 1258-1281.
Engel, C. and Sharp, R. (1958). Chemical data on desert
varnish. Geol. Soc. Am. Bull. 69, 487-518.
Galey, J. (personal communication).
Gates, D. et al. (1965). Spectral properties of plants.
Applied Optics 4, 11-20.
Gattow, V. and Zemann, J. (1958). Neubestimung der Kristallstruktur von Azurit. Acta Cryst. 11, 866-872.
Gilluly, J. (1963). The tectonic evolution of the western
United States. Geol. Soc. London Quart. Jour. 119,
133-174.
Halford, R. (1946). Motions of molecules in condensed systems:
I. Selection rules, relative intensities and orientation effects for Raman and infrared spectra. J. Chem.
Phys. 14, 8-15.
Hathaway, B. and Billing, D. (1970). The electronic properties and stereochemistry of mono-nuclear complexes
of the copper (II)
ion. Coordin. Chem. Rev. 5,
143-207.
Head, J. et al. (1978). Definition and detailed characterization of lunar surface units using remote observations. Icarus 33, 145-172.
Huff, L. (1952). Abnormal copper, lead, and zinc content of
soils near metalliferous veins. Econ. Geol. 47, 517542.
Hunt, G. and Salisbury, J. (1970a). Visible and near infrared spectra of minerals and rocks: I. Silicate
minerals. Mod. Geol. 1, 283-300.
.
(1970b). Visible and near infrared spectra of
minerals and rocks: II. Carbonates. Mod. Geol. 2,
23-30.
113
,
and Lenhoff, C. (1971a). Visible and near infra-
red spectra of minerals and rocks: III. Oxides and
hydroxides. Mod. Geol. 2, 195-205.
.
(1973a). Visible and near infrared spectra of
minerals and rocks: VI. Additional silicates. Mod.
Geol. 4, 85-106.
Jorgensen, C. (1963). Spectroscopy of transition group
complexes. Advances in Chemical Physics 5, 33-147.
Lakshman, S. and Reddy, B. (1973). Optical absorption spectra of Cu 2 + in chalcanthite and malachite. Can. Min.
12, 207-210.
Lever, A. (1968). Inorganic Electronic Spectroscopy. Elsevier
Publishing Co., New York.
Liehr, A. (1969). Reciprocation of electrostatic and electromagnetic forces in ligand field theory. J. Phys. Chem.
64, 43-51.
Locke, A. (1926). Leached Outcrops as Guides to Copper Ore.
Baltimore, Williams, and Wilkins Co.
Lovering, T., Huff, L., and Almond, H. (1950). Dispersion of
copper from the San Manual copper deposit, Pinal
County, Arizona. Econ. Geol. 45, 493-514.
Lowell, J. and Guilbert, J. (1970). Lateral and vertical alteration-mineralization zoning in porphyry ore
deposits. Econ. Geol. 65, 373-408.
Mao, H. and Bell, P. (1973-74). Crystal field effects of
ferric iron in goethite and lepidocrocite: band
assignment and geochemical applications at high
pressure. Annual Report of the Director, Geophysical
Laboratory, Carnegie Institution, Washington, D.C.
McCord, T. (principal investigator)(1975). Surface Mineralogy
Experiment for the Lunar Polar Orbiter. NASA proposal.
. et al.
(1977). Spectral reflectance of Martian
areas during the 1973 opposition: photoelectric filter
photometry 0.33-1.10 p. Icarus 31, 25-39.
Pappalardo, R. (1961). Absorption spectra of Cu2+ in different
crystal coordinations. J. Mol. Spectroscopy 6, 554573.
114
Potter, R. and Rossman, G. (1977). Desert varnish: the
importance of clay minerals. Science 196, 1446-1448.
Sillitoe, R. (1972). A plate tectonic model for the origin
of porphyry copper deposits. Econ. Geol. 67, 184197.
.
(1973).
The tops and bottoms of porphyry copper
deposits. Econ. Geol. 68, 799-815.
Silman, J. (1958). The stabilities of some oxidized copper
minerals in aqueous solutions at 25*C and 1 atmosphere pressure. Ph.D. Thesis, Harvard University,
Cambridge, Massachusetts.
Solomon,
E.I.
(personal communication).
Ssse, V. (1967). Verfeinerung der Kristallstruktur des Malachits. Acta Cryst. 22, 146-151.
Titley, S. and Hicks, C. (1966). Geology of the Porphyry
Copper Deposits. Univ. of Arizona Press, Tucson.
Wells, A. (1951). Malachite: re-examination of the crystal
structure. Acta Cryst. 4, 200-204.
Whitten, P. and Brooks, J. (1972). A Dictionary of Geology.
Penguin Books, England.
A P P E N D I X
I
SAMPLE DESCRIPTIONS AND ANALYSES
Sample Description
Mineral samples used to generate the laboratory reflectance spectra presented in this study were obtained from
the Smithsonian Museum of Natural History in Washington, D.C.
as well as from mineral and chemical supply companies.
The
origins of the samples are listed in Table 1 along with their
respective Smithsonian identification numbers, where applicable.
The samples used to generate the laboratory mixtures of
ferric oxides, clays, malachite, and MnO 2 are the following:
Limonite 2001 (Cartersville)
Hematite,
reagent
Malachite 6004 (reagent)
Kaolinite, England
MnO 2 , reagent
Samples that needed to be ground were done so with a porcelain mortar and pestle.
Most samples contain unsorted par-
ticles sized less than 90 p, and a few samples noted in the
text were dry sieved using standard Tyler sieves according
to the particle sizes: J = 63-90 p; K = 45-63 p; L = <45 p.
Standard x-ray diffraction analysis was used to identify the
azurite, malachite, and ferric oxide samples.
115
116
Table 1.
Sample descriptions and identification numbers.
Mineral
M.I.T.
Sample #
Smithsonian
Sample #
Sample Origin
Azurite
5000
5001
118913
117707
Tsumeb, S.W. Africa
Hermosillo, Mexico
Malachite
6000
6001
6002
6003
6004
R13881
10323
122455
C5823
Bisbee, Arizona
Kennecott, Alaska
Broken Hill, Rhodesia
Apex Mine, Utah
Reagent (Mallinckrodt)
Chrysocolla
(blue)
(blue)
(blue)
(green)
(green)
(green)
7000
7001
7002
7003
7004
7005
104956
120272
R4844
B17050
R16369
R4859
Bisbee, Arizona
Arizona
Gila County, Arizona
Herrengrund, Hungary
Tagilsk, U.S.S.R.
Tagilsk, U.S.S.R.
Goethite
3004
C1736
Minas Gerais, Brazil
Limonite
2000
2001
2003
118532
1000
1001
1003
1005
46422
128254
64655
2002
104895-1
Hematite
Lepidocrocite
Tuscaloosa, Ga. (Wards)
Cartersville, Ga. (Wards)
Bisbee, Arizona
Ironton, Minn. (Wards)
Camerata, Africa
Mt. Newman, Australia
Missouri
Reagent (Matheson,
Coleman, and Bell)
(Limonite) Bisbee, Az.
Maghemite
Steam oxidized magnetite
Kaolinite
England (Mallinckrodt)
Manganese
dioxide
8002
Reagent (Baker)
117
Sample Analyses
Reflectance spectra.
The theoretical basis for interpreta-
tionof visible-near infrared spectra of minerals has been
established in the literature
1962).
(Burns,
1970; Ballhausen,
Adams (1975) presents references describing the
application of crystal field theory to the interpretation
of optical spectra of natural minerals.
Laboratory diffuse reflectance spectra of powdered
samples were measured in the range 0.35-2.50 y with a Beckman DK-2A ratio-recording spectrophotometer belonging to
John Adams at the University of Washington.
MgO was used
as a reference surface except for those spectra measured of
the natural soil and rock chip samples, where halon was
used as a reference.
Spectra were recorded simultaneously
on chart paper and on magnetic tape except for the azurite
and malachite spectra, which were recorded on digital papertape.
Infrared spectra.
Infrared spectra of a suite of azurite
and malachite samples were measured by Graham Hunt at the
Air Force Cambridge Research Laboratories in the 2.50-40 y
range with a Perkin-Elmer Model 180 spectrophotometer.
conventional KBr pellet technique was used as well as a
reflection technique,
microreflectance
which employed a modified standard
attachment.
The
118
X-ray fluorescence.
Soil samples collected from the Silver
Bell and Mineral Park districts were analyzed for their copper concentrations using X-ray fluorescence.
The smallest
sized particle samples (<45 y for the Silver Bell samples and
<125 y for the Mineral Park samples) were pressed into pellets
and then measured on a Diano (General Electric XRD-5) X-ray
fluorescent instrument operated by the Department of Chemistry at the University of Massachusetts.
A tungsten tube pro-
vided the X-ray source operating at 40 kV and 35 mA.
The op-
tical path was air,and the X-rays were dispersed using a LiF
crystal.
A proportional detector operating at 1.4 kV was
used to convert the X-ray photons to electron pulses.
Copper
was analyzed at the CuKa peak at 45.03*, and the background
was measured at 44.20* and 48.200 for each sample.
Counts
were recorded at a fixed time of 10 seconds per measurement,
and 10 measurements were used to calculate the average number
of counts at each angle setting.
Since the copper concentra-
tions of the Silver Bell rock chip samples were previously
determined by ASARCO, they were used as reference standards
with copper concentrations ranging between 70 and 3200 ppm.
The estimated error in the copper concentrations of the soil
samples ranged between 15 and 45 ppm, with the average error
on the order of 25 ppm, although the error in the rock chip
analyses could not be included, since it was not given by
ASARCO.
The average copper concentration of the soil and
rock chip samples are given in Table 2.
119
Table 2.
Copper concentrations of the Silver Bell and
Mineral Park samples as determined from XRF.
The concentrations of the rock chip samples
was provided through the courtesy of ASARCO.
Silver Bell
soil sample
Cu (ppm)
Silver bell
rock chip
Cu (ppm)
SS1
370
GOC43
3200
SS2
285
GOC44
310
SS3
270
GOC46
100
SS4
465
GOC49
SS5
1585
GOC78
750
SS6
395
SS7
710
SS8
305
Mineral Park
soil sample
Cu (ppm)
SS9
245
ACl 3
930
SS10
760
DFl3
1075
SSll
290
AC4 6
1250
DF46
1140
SS12
SS13
125
SS14
175
SS15
275
SS16
SSl7
100
SS18
135
SS19
145
SS20
105
A P P E N D I X
I I
REFLECTANCE SPECTRA OF LABORATORY SIMULATED SOILS
Figure 1: Malachite and ferric oxide
Figures 2-4: Malachite, ferric oxides, and
kaolinite
Figures 5-32: Malachite, ferric oxides, kaolinite,
and MnO
2
120
121
80
70
A
60-
B
W 50 Z
40
W
30
-
D
I
20 -
od
10
A -
5% malachite
+ 95% ferric oxides
10
B - 25% malachite + 75% ferric oxides
C - 50% malachite + 50% ferric oxides
10
D - 75% malachite + 25% ferric oxides
10
0
0.5
1.0
1.5
2.0
WAVELENGTH
2.5
(p)
Figure 1. Reflectance spectra of laboratory prepared mixtures of
malachite and an equal mixture (wt.%) of hematite and goethite.
3.0
122
100
9,0
80-
10
z
U <60
C
z
~
L, 50-~
;- 4,0
A - 5%malachite + 5% ferric oxides +
90% kaolinite
B - 5% malachite + 20% ferric oxides +
75% kaolinite
30~
A
C - 5% malachite + 50% ferric oxides +
45% kaolinite
B
10
C
0.5
10
1.5
2.5
Z.0
WAVELENGTH
3.0
(/.)
Figure 2. Reflectance spectra of laboratory prepared mixtures of
malachite, and equal mixture (wt.%) of hematite and goethite, and
kaolinite.
123
90
80
70
W
Q
Z
W
-j
LW
60
A
50
50
B
C
40
30
20
10
A - 20% malachite + 5% ferric oxides + 75% kaolinite
B -
20% malachite + 20% ferric oxides + 60% kaolinite
C - 20% malachite + 50% ferric oxides + 30% kaolinite
10
10
t
0 L
0.5
1.0
1.5
2.5
2.0
WAVELENGTH
3.0
(p4
Figure 3. Reflectance spectra of laboratory prepared mixtures of
malachite, and equal mixture (wt.%) of hematite and goethite, and
kaolinite.
124
80
70
60
A
50
B
40
30
20
A - 50% malachite +
5% ferric oxides + 45% kaolinite
B - 50% malachite + 20% ferric oxides + 30% kaolinite
10 -
0
0.5
1.0
1.5
2.0
2,5
WAVELENGTH
Figure 4. Reflectance spectra of laboratory prepared mixtures of
malachite, an equal mixture (wt.%) of hematite and goethite, and
kaolinite.
3.0
125
90
80
70
w
A
60
50
B
W 40
I
C
w 30
0
$
20
A - 5% malachite
+ 5% goethite +
85% kaolinite + 5% MnO
B -5% malachite + 20% goethite +
70% kaollinite + 5% MnO2
10
C-
5%malachite + 50% goethite +
40% kaolinite + 5% MnO
2
-
10
0
0.5
1 .0
2.0
1.5
WAVELENGTH
(,L)
Figure 5.Reflectance spectra of laboratory prepared
mixtures (wt.%).
2.5
126
70
60
W
U
Z
50
A
40
B
C
w 03 0
-J
LI-
W
20
o
10
A - 5% MnO 2 + 20% malachite +
5% goethite + 70% kaolinite
B - 5% MnO 2 + 20% malachite + 20% goethite + 55% kaolinits.
C - 5% Mn0 2 + 20% malachite + 50% goethite + 25% kaolinite
10
10
~
0
0.5
1.0
1.5
2.0
2.5
3.0
WAVELENGTH (p.)
Figure 6.
Reflectance spebtra of laboratory prepared mixtures (wt.%).
127
80
70
w
60
U
Z
50
W
40
40
W
30
A
B
20
A - 5% MnO2 + 50% malachite + 5% goethite + 40% kaolinite
10
B - 5% MnO 2 + 50% malachite + 20% goethite + 25% kaolinite
10
0
0.5
1.0
1.5
2.5
2.0
WAVELENGTH
3.0
(
Figure 7. Reflectance spectra of laboratory prepared mixtures (wt.%).
128
70
60
A
50
Z
U
40
B
C
W 30
30
-j
IL
W
20
O
IO
A - 15% MnO + 5% malachite + 5% goethite +
75% kaoiinite
B - 15% MnO
5% malachite + 20% goethite +
60% kaoiinite
C - 15% Mno + 5% malachite + 50% goethite +
30% kaoIinite
10
10
0
0.5
1.0
1.5
2.0
WAVELENGTH (p)
Figure 8. Reflectance spectra of laboratory prepared
mixtures (wt.%).
2.5
129
70
60
A
50
0
z
40
W30
C
-
20
$
\0
10
10
10
-+
A - 15% Mno + 20% malachite +
+ 60% k olinite
5% goethite
B - 15% MnO, + 20% malachite + 20% goethite
+ 45% kaolinite
C - 15% Mno + 20% malachite + 50% goethite
15% k olinite
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
A
(p.
Figure 9. Reflectance spectra of laboratory prepared
mixtures (wt.%).
130
80
70
W60
z 5,0
I-
40
W4
LL
Cr
30
0
20
-.
A
-
-0
10
10
A-
15% Mato
5% goethite +
B -
15% MnO
+ 50% malachite + 20% goethite +
50% .malachite +
30% kao~inite
2
15% kaolinite
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 10. Reflectance spectra of laboratory prepared
mixtures (wt.%).
131
70
60
w 50
0
z 4
A
W30
C
W 20
o*
O
5% malachite + 5% goethite +
65% kaoiinite
A
B
-
10
+
25% MnO
+
5% malachite + 20% goethite +
50% kaoiinite
C
10
10
25% MnO
-
25% MnO
+
5% malachite + 50% goethite +
20% kao~inite
-
II
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p)
Figure 11. Reflectance spectra of laboratory prepared
mixtures (wt.%).
132
70
60
W
50
z
p.
4 0
130
C
Li
X:
20
$
10
A - 215% Mno + 20% malachite + 5% goethite
+ 50% k olinite
B - 25% MnO + 20% malachite + 20% goethite
+ 35% k alinite
C - 25% MnO + 20% malachite + 50% goethite
+ 5% ka1inite
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 12. Reflectance spectra of laboratory prepared
mixtures (wt.%).
133
70
60
w
050
zA
40
0
w
-3
20
0
A
25% MnO + 50% malachite + 5% goethite +
20% kaoiinite
B - 25% MnO + 50% malachite + 20% goethite +
5% ka linite
10
0
0.5
-
1.0
1.5
WAVELENGTH
2.0
2.5
(p)
Figure 13. Reflectance spectra of laboratory prepared
mixtures (wt.%).
134
70
A -
50% MnO
+
5% malachite + 20% goethite +
25% kaoiinite
60
w
0
Z
5% goethite +
B -
50% MnO
+ 20% malachite +
C -
50% MnO
+ 20% malachite + 20% goethite +
-
25% kaolinite
10% kaolinite
50
40
[
A
LL
B
30
-J
IL
W
20
o
10
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(.L)
Figure 14. Reflectance spectra of laboratory prepared
mixtures (wt.%).
135
80
70
A
60
z
50
w
40
U.
30
w
B
C
20
0
10
A - 5% MnO + 5% malachite + 5% hematite
+ 85% Naolinite
B - 5% MnO + 5% malachite + 20% hematite
+ 70% iaolinite
10
C-
5% MnO + 5% malachite + 50% hematite
+ 40% aolinite
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(l.)
Figure 15. Reflectance spectra of laboratory prepared
mixtures (wt.%).
136
80
70
60
W 50
0
z
H
0
40
w
LL 3 0
w
r.20
0
10
A a 5% MnO + 20% malachite +
+ 70% aolinite
10
B
C
10
-
-
5% MnO
5% hematite
+ 20% malachite + 20% hematite
-
+ 55%
5% MnO + 20% malachite + 50% hematite
+ 25% aolinite
aolinite
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p)
Figure 16. Reflectance spectra of laboratory prepared
mixtures (wt.%).
137
80
70
W 60
0
z
4
B
50
0
IL
Oc
0
W40
30
20
A - 5% MnO + 50% malachite +
+ 40% Raolinite
5% hematite
B - 5% MnO + 50% malachite + 20% hematite
+ 25% aolinite
10
-
10~
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 17. Reflectance spectra of laboratory prepared
mixtures (wt.%).
138
70
60
Li
0B
z
50
40
C
1-
W 30
W 20
0*
10
Q
A -
15% MnO
+ 5% malachite +
5% hematite
+ 75% k.olinite
B -
10
10
15% MnO
+ 5% malachite + 20% hematite
+ 60% k linite
C - 15% MnO + 5 % malachite + 50% hematite
+ 30% k olinite
-
-
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(piL)
Figure 18. Reflectance spectra of laboratory prepared
mixtures (wt.%).
139
70
60
50
B
U
Z 40
A
0
C
w
w
30
LL
W
a: 20
9
10
A - 15% MnO + 20% malachite + 5% hematite
+ 60% kaolinite
B - 15% MnO + 20% malachite + 20% hematite
+45% kaglinite
10
C -
1O
15% MnO
+ 20% malachite + 50% hematite
+15% kalinite
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(.)
Figure 19. Reflectance spectra of laboratory prepared
mixtures (wt.%).
140
80
+ 30% kiolinite
B
+ 50% malachite + 20% hematite
15% MnO
+ 15% kiolinite
70
w
-
60
0A
Z
50
W 40
IL
W 30
$
20
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 20. Reflectance spectra of laboratory prepared
mixtures (wt.%).
141
70
60
A
50
w
Z
40
W30
-
C
w20
0
10
10
0.5
A
-
25% MnO + 5% malachite +
+ 65% kaolinite
B
-
C
-
25% MnO + 5% malachite + 20% hematite
+ 50% kaolinite
25% MnO + 5% malachite + 50% hematite
+ 20% k olinite
1.0
1.5
WAVELENGTH
5% hematite
2.0
2.5
(L)
Figure 21. Reflectance spectra of laboratory prepared
mixtures (wt.%).
142
70
60
W
z
50
0
40
A
-B
W 30
-J
LLC
X
20
%%
0
10
A
-
B -
t0
25% MnO + 20% malachite +
+ 50% k olinite
25% MnO
5% hematite
+ 20% malachite + 20% hematite
+ 35% kaolinite
C - 25% MnO + 20% malachite + 50% hematite
+ 5F kaglinite
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(/p)
Figure 22. Reflectance spectra of laboratory prepared
mixtures (wt.%).
143
80
70
A - 25% MnO + 50% malachite +
+ 20% k~olinite
B -
5% hematite
+ 50% malachite + 20% hematite
25% MnO
+ 5% kaglinite
w60
0
z
50
(U
laa
-J
40-
L
w
30
o
20
B
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p. )
Figure 23. Reflectance spectra of laboratory prepared
mixtures (wt.%).
144
70
60
w
Q
z
50
-
+ 25% kaolinite
B - 50% MnO + 20% malachite + 5% hematite
+ 25% kiolinite
C - 50% MnO + 20% malachite + 20% hematite
+ 10% k~olinite
-
-
40
A
-
W 30
..J
LL
B
X20
IS 10
-
10
10
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 24. Reflectance spectra of laboratory prepared
mixtures (wt.%).
145
80
V
70
A
60
B
050
z
-
40
--J30
w0
20
A - 5% MnO
+ 5% malachite + 5% (hematite +
goethite) + 85% kaolinite
B - 5% MnO + 5% malachite + 20% (hematite +
goethi~e) + 70% kaolinite
I0
C -
+ 5% malachite + 50%
(hematite +
goethife) + 40% kaolinite
10
10
5% MnO
-
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p)
Figure 25. Reflectance spectra of laboratory prepared
mixtures (wt.%).
146
70
A
60
w
0
z
50
40
C
I
W30 LiL
W 20
- 5% MnO + 20% malachite + 5% (hematite
+ goet ite) + 70% kaolinite
10
B - 5% MnO + 20%
+ goet ite) +
C - 5% MnO + 20%
+ goethite) +
10
malachite + 20% (hematite
55% kaolinite
malachite + 50% (hematite
25% kaolinite
10
0.5
1.0
1 .5
WAVELENGTH
2.0
2.5
(p.)
Figure 26. Reflectance spectra of laboratory prepared
mixtures (wt.%).
147
80
70
w
60
z
50
-
A
0
-
W 40
-J
w
30
0f%20
--
A - 5% MnO + 50% malachite + 5% (hematite
+ goet~ite) + 40% kaolinite
B - 5% MnO + 50% malachite + 20% (hematite
+ goet ite) + 25% kaolinite
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 27. Reflectance spectra of laboratory prepared
mixtures (wt.%).
148
70
60
A
50
B
z40
FUC
W30
LiL
w
20
A - 15% MnO + 5% malachite + 5% (hematite
+ goethite) + 75% kaolinite
B - 15% MnO + 5% malachite + 20% (hematite
+ goethite) + 60% kaolinite
C - 15% MnO + 5% malachite + 50% (hematite
+ goethite) + 30% kaolinite
10
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(/p.)
Figure 28. Reflectance spectra of laboratory prepared
mixtures (wt.%).
149
70
60
A
w
50
C.
z<40
B
0
C
.
w
0
o
20
10
A - 15% MnO + 20% malachite + 5% (hematite~
+ goethite) + 60% kaolinite
10
B -
15% MnO
C
15% MnO + 20% malachite + 20% (hematite
+ goethite) + 15% kaolinite
10
0
--
0.5
-
+ 20% malachite + 20% (hematite
+ goethite) + 45% kaolinite
1.0
1 .5
WAVELENGTH
2.0
2.5
(p4
Figure 29. Reflectance spectra of laboratory prepared
mixtures (wt.%).
150
80
B
-
goethit6) + 30% kaolinite
15% MnO + 50% malachite + 20% (hematite +
goethit ) + 15% kaolinite
70
60
0A
z
50
W 040
-J
IL
LLJ3
W 30o
20
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.)
Figure 30. Reflectance spectra of laboratory prepared
mixtures (wt.%).
151
70
60
50
A
z 040
B
W30
1
U
C
20
A -
10
B
-
5% MnO
+ 5% malachite +
5% (hematite
+ goethite) + 65% kaolinite
25% MnO + 5% malachite + 20% (hematite
+ goethite) + 50% kaolinite
10
C -
25% MnO
+ 5% malachite + 50%
(hematite
+ goethite) + 20% kaolinite
10
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p.
Figure 31. Reflectance spectra of laboratory prepared
mixtures (wt.%).
152
70
60
50
Z
-
B
40
W30
-
0 -oJ
o1010
B
-
+ goethite) + 50% kaolinite
25% MnO + 20% malachite + 20% (hematite
+ goethite) + 35% kaolinite
C - 25% MnO + 20% malachite + 50% (hematite _
+ goethite) + 5% kaolinite
10
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
( p.)
Figure 32. Reflectance spectra of laboratory prepared
mixtures (wt.).
153
70
goethit6) + 20% kaolinite
B
60
--
-
25% MnO
+ 50% malachite + 20%
goethit ) +
(hematite +
5% kaolinite
W 50
z
-40
B
0w
LL 3 0
w
20
10
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(p)
Figure 33. Reflectance spectra of laboratory prepared
mixtures (wt.%).
154
60
50
w
0
40
-
z
H30
0
w
L.L20
w
10
10
A -
50% MnO
+
5% malachite + 20%
(hematite
+ goethite) + 25% kaolinite
B - 50% MnO + 20% malachite + 5% (hematite
+ goethite) + 25% kaolinite
C - 50% MnO + 20% malachite + 20% (hematite
kaolinite
+ go th~te) + 10f
10
L
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(.)
Figure 34. Reflectance spectra of laboratory prepared
mixtures (wt.%).
A P P E N D I X
I I I
RATIO SPECTRA OF LABORATORY SIMULATED SOILS
Figures 1-16: MnO , malachite, kaolinite, and
goethite ra io spectra
Figures 17-32: MnO 2 , malachite, kaolinite, and
hematite ratio spectra
Figures 33-48: MnO 2 , malachite, kaolinite, and
(50% hematite + 50% goethite) ratio spectra
Note: Each spectrum was scaled to unity at 0.87 p
before ratioing.
155
5% MnO2 + 5% malachite + 40% kaolinite + 50% goethite
5% MnO 2 + 5% malachite + 70% kaolinite + 20% goethite
09
-I
-
(D
~*
Os
WAVELENGTH -
5% MnO 2 + 20% malachite + 25% kaolinite + 50% goethite
5% MnO 2 + 20% malachite + 55% ksolinite + 20% goethite
-
L1
-JJ
(1
(D
-GT
In
.-
. -. . 5
2
3
05
-7 M
S
WAVELEN-
H"
e1
5% MnO2 + 5% malachite + 70% kaolinit.e. + 20% goethite
5% MnO 2 + 20% malachite + 55% kaolinite + 20% goethite
Ln
FJ.
~1
09
(D
LA
Is(A
.150
CJ75
1A
15
WAVELENGTH -
1w
MICRONS
Ij7T5
eJoW
"2
5% MnO 2 + 5% malachite + 40% kaolinite + 50% goethite
5% MnO 2 + 20% malachite + 25% kaolinite + 50% goethite
r2
-4
Ufl
.
fia
.r
W..
WAVELENGTH -
MICRONS
e
's
'
lse
1A
15% MnO 2 + 5% malachite + 60% kaolinite + 20% goethite
15% MnO2 + 5% malachite + 75% kaolinite + 5% goethite
0
(D
Lu
-5
'ise
so0
()7
ti75
ii5s
iles
WAVELENGTH - MICRONS
2)00
2)25
2250
15% MnO 2 + 5% malachite + 30% kaolinite + 50% goethite
15% MnO 2 + 5% malachite + 75% kaolinite + 5% goethite
IC
I
. Z
s
Q1$0
t2TJ0
1i2
WAVELENGTH -
ij
MICRONS
E)
1175~r
a
50
us
Mls
N
ps-
15% MnO 2 + 5% malachite + 30 % kaolinite + 50% goethite
15% MnO 2 + 5% malachite + 60 % kaolinite + 20% goethite
70
n
VS7S
o2
O
tICe
WAVELENGTH - MICRONS
I7
Qles
Las
.'
50n
15% MnO 2 + 20% malachite + 15% kaolinite + 50% goethite
15% MnO 2 + 20% malachite + 45% kaolinite + 20% goethite
IU,
(ft
M
ast
I'-S
(D
00
-I.
0
-1
:~i )~s
9
9
se
075
0AE
LE
-
I50
WAVELENGTH - MICRONS
1ji~J
W
~ceCc.
'50
15% MnO 2 + 5% malachite + 60% kaolinite + 20% goethite
15% MnO 2 + 20% malachite + 45% kaolinite + 20% goethite
Ln
P-e
I4
om
--
UjSe
Q'75
tow0
ije5
WAVELENGTH -
1jbw
MICRONS
1
a? I5:D
Cifw
I;e a.
2
15% MnO 2 + 20% malachite + 15% kaolinite + 50% goethite
15% MnO 2 + 5% malachite + 30% kaolinite + 50% goethite
vi
Zn
(/
"
S
e
Oise
U7/5
1000
25
WAVELENGTH
I.5
-
MICRONS
c)
e
w
7
es
c5oA
50
ii
25% MnO 2 + 5% malachite + 50% kaolinite + 20% goethite
25% MnO 2 + 5% malachite + 65% kaolinite + 5% goethite
U.'
/1
afse
es
G)-75
0
u.25
WAVELENGTH
-
i3S
J
MICRONS
- 2)J0
27e
2ise
25% MnO 2 + 5% malachite + 20% kaolinite. + 50% goethite
-J
25% MnO 2 + 5% malachite + 65% kaolinite + 5% goethite
ru
UI
u0
eI-
sU,
is
am
-1.1
61e
U350
0175
100
125
WAVELENGTH -
1.150
MICRONS
iJ75
270e
j2es
2J50
25% MnO 2 + 5% malachite + 20% kaolinite + 50% goethite
25% MnO 2 + 5% malachite. + 50% kaolinite + 20% goethite
Lfl
U'
Iljjjjrl
111
-4'
''111111till
PolII!
M6
MJ
U'
L,
Lfl4
sS
61
Gi
Qf'75
t 00
U bw
U25
WAVELENGTH - MICRONS
c
caca
ewe
1
J
to
.o
150
SI
S
-J
25% MnO 2 + 20% malachite + 5% kaolinite + 50% goethite
I
25% MnO 2 + 20% malachite + 35% kaolinite + 20% goethite
'Ji I
U1
Ul
I
U725
iWo
ie25
150
WAVELENGTH - MICRONS
.--
1
~ ~
~ec
~
'53
c
C fAewe
25% MnO + 5% malachite + 50% kaolinite + 20% goethite
2*
25% MnO + 20% malachite + 35% kaolinite + 20% goethite
2
-I.
-J
~DISW
Djso
0)75
1,J75
1)50
1)00
iNTHW7 i)~5
ies
iCR0ON
0)7s
WAVELENGTH - MICRONS
-
-
a)0s~eia~~a~50
ejoe
ej25
215so
25% MnO 2 + 20% malachite + 5% kaolinite + 50% goethite
25% MnO 2 + 5%
Lfl
malachite + 20% kaolinite + 50% goethite
Lfl
M
00
I(I)
H
a'
t
% Mno 2 + 5% malachite + 40% kaolinite + 50% hematite
5% MnO 2 + 5% malachite + 70% kaolinite + 20% hematite
-4..
In
L'I
$ZWAVELENGTH
-
MICRONS
1w
6fQ
5% MnO 2 + 20% malachite + 55% kaolinite + 2G% hematite
5% MnO 2 + 50% malachite + 25% kaolinite + 50% hematite
LI!
-I.
CD
co
L1
ga
(Gj7S
t00
1)25
WAVELENGTH -
L50
MICRONS
J.75
LJWW
c)
ears
50
owl
5% MnO 2 + 5% malachite + 70% kaolinite + 20% hematite
5% MnO2 + 20% malachite + 55% kaolinite + 20% hematite
!1
4-
ru
Ln
- aI
do
UYse
0P7S
t0
tie5
I050
WAVELENGTH - MICRONS
i7s
2Je
25es
21s
5% MnO 2 + 5% malachite + 40% kaolinite + 50% hematite
5% MnO 2 +20% malachite. + 25% kaolinite + 50% hematite
~J1
-4.
S
dl
S
-
U'
S
S
S
)
-m
U)s@
0. 75
'
1)00
1)2J
WAVELENGTH -
1e5
MICRONS
1 /5
eI w4eD
22b
2.5S
15% MnO 2 + 5% malachite + 60f kaolinite + 20% hematite
15% MnO 2 + 5% malachite + 75% kaolinite + 5% hematite
Lfr
Lu
&I
0
50
0 -75
i)
1MCt!S
1N25
WAIVELENG'fH
-
M'ICIRN3
I1
/b
2
2J25
1
15% MnO 2 + 5% malachite + 30% kaolinite + 50% hematite
15% MnO 2 + 5% malachite + 75% kaolinite + 5% hematite
Q~q
-4.
09
)S
O2-7S
00
T.ON
7n
)
)I$~
11w
11
15% MnO 2 + 5% malachite + 60% kaolinite + 20% hematite
15% MnO 2 + 5% malachit-e + 30% kaolinite + 50% hematite
n
Pu
L'
4.
a&
Sn
- W25
~
i:~~~~~
WAVELENGTH -
MICRONS
irAW.L
iJ
-
15% MnO
+ 20% malachite + 15% kaolinite + 50% hematite
15% MnO
+ 20% malachite + 45% kaolinite + 20% hematite
I-s
CD
R)
4:::-
C-4WAVELENG'TH
Wres
0[15L
-
02~7C
--MICRONS
Z
500
15% MnO, + 5% malachite + 60% kaolinite + 20% hematite
15% MnO 2 + 20% malachite + 45% kaolinite + 20% hematite
LU
U
(b) so
0)-75
10
L25
MICbN
5
WAVELENGTH - MICRONS
IZ
.we
eC
s
eCI
1f/i
se
15% MnO2 + 20% malachite 15% kaolinite + 50% hematite
-J
15% MnO2 + 5% malachite + 30% kaolinite + 50% hematite
In
C!'
CD
R)
0)SG
&'7S
1)75
I)~0
I)~S
1)00 WA VEL5
t15
s o
t. 25
i j 75
WAIVELENGTH - MICRONS
2)00
2joe
2)25
2j25
2.~50
2.'50
25% MnO 2 + 5% malachite + 50% kaolinite + 20% hematite
's
Lfl2
25% MnO 2 + 5% malachite + 65% kaolinite + 5% hematite
ru
l'!
~wv~'V~
IWAVE0LEN7GS
WAJVELENG'TH
EwJ
-
M)IoN
-j~s
MICRONS
Id~~~~~~~
.W
;..J
L.
A.
25% MnO 2 + 5% malachite + 20% kaolinite + 50% hematite
'
25% MnO 2 + 5% malachit.e + 65% kaolinite + 5% hematite
00
r\M
WAVE.LENGTH
-
MIICRONS
25% MnO 2 + 5% malachite + 20% kaolinite + 50% hematite
25% MnO 2 + 5% malachite + 50% kaolinite + 20% hematite
rx)
-j
)sO
no
i)es
WAVELENGTH
-
-Jsbi
MICRONS
ls
i
l
e
e
--
-
'es
e
-
se
25% MnO 2 + 20% malachite + 35% kaolinite + 20% hematite
25% Mn O 2 + 20% malachite + 5% kaolinite + 50% hematite
W
ru
U1
l
U1
ru
Ul
25
25
0.150
/b
e025
ow
1V
.TH
- 1ICRON
WAVELENGTH - MICRONS
1 1i.1
53
C i
2
0
W
2
!
2-C
2In
W
50
25% MnO 2 + 20% malachite + 35% kaolinite + 20% hematite
25% MnO 2 + 5 % malachite + 50% kaolirite + 20% hematite
,
.
lv"V
,
v 111,
0 ,JFA l,'111,1111,
,,
111111
111"111,11,11111
CD
w
Le*i
,_j
eYse
e2s
i)00
tjes
i0se
WAVELENGTH - MICRONS
13/b
i.,ws
e1
e-e Z>
.s
'
W
25% MnO 2 + 5% malachite + 20% kaolinite + 50% hematite
25% MnO2 + 20% malachi-te + 5% kaolinite + 50% hematite
en
Ln
(n
IS-
Lfl
-A
0.So
0)25
1j00
iies
WAVELENGTH -
iise
MICRONS
t 12/b0
e2es5
els.,e
i.,b
5% MnO2 + 5% malachite + 40% kaolinite + 50%
'
(hematite + goethite)
5% MnOn2 + 5% malachite + 70% kaolinite + 20% (hematite + goethite)
U9
11
0
ru
-4
Lt
U,1
WAVELENGTH
-
MICRONS
i)7S'
2je
eis
2
5% MnO2 + 20% malachite + 25% kaolinite + 50% (hematite + goethite)
5% MnO2 + 20% malachite + 55% kaolinite + 20%
(hematite + goethite)
-4
LI
E;
U,
49
V S2
1
Hes t 150
i
AV0L0(DO
Sj25
WAVELENGTH - MICRONS
2joo
2ej)2?5
250
5% MnO2 + 20% malachite + 55% kaolinite + 20% (hematite + goethite)
5% MnO2 +
-J
5% malachite + 70% kaolinite + 20%
(hematite + goethite)
(-1
Ul
.......
.......................
............
LI.
QI
U1
S
,I
f-
0)54
0)2 S1i25
1*7
MICo
5
i j275
WAVELENGTH - MICRONS
23
2j25
2150
5% MnO2 + 20% malachite + 25% kaolinite + 50% (hematite + goethite)
5% MnO 2 +
5% malachite + 40% kaolinite + 50% (hematite + goethite)
Ul
A
(A
-J
S in
U1
S
I
S
Tys
7T3W5
i25
WAVELENGTH -
i5s
MICRONS
'
ee
i~~~~
3
5
5
M
2
53
15% MnO2 +
2
15% NnO 2+
5% malachite + 60% kaolinite + 20%
5% malachite + 75% kaolinite +
(hematite + goethite)
5% (hematite + goettehite)
Lfl
LI,
1j7s
WAVELENG'fH -
NICRONS
2) 00
2j2s
2.150
15% MnO2 + 5% malachite + 30% kaolinite + 50% (hematite + goethite)
15% MnO 2 + 5% malachite + 75% kaolinite +
5% (hematite + goethite)
rk)
1-'09
(D
w
4.'
co
L9~
(YR
)$
s -)2S
i)Vo
5
1,IIis
I25
WAnVELENGTrH - MICRONS
in2
27o
2j25
2.'s
15% MnO 2 + 5% malachite + 30% kaolinite + 50% (hematite + goeth te)
15% MnO2 + 5% malachite + 60% kaolinite + 20% (hematite + goethite)
EQ
4
Li,
*1
2-S--5-
5
1"S
12
50
I j15O
e
WAVELENGTH
-
MICRONS
i j75
e1o
2)2s
2.50
15% MnO 2 + 20% malachite + 15% kaolinite + 50%
(hematite + goethite)
15% MnO 2 + 20% malachite + 45% kaolinite + 20% (hematite + goethite)
Ui
C)
U'
S
0 -~7
t)SO
12S
WAVELENGrH
-
MICRONS
i7S
e)2 o
2j2S
2si
(n
15% MnO 2
15% NnO2
15% MnO 2
5% malachite + 60% kaolinite + 20% (hematite + goethite)
5% malachite + 60% kaolinite + 20% (hematite + goethite)
+ 20% malachite + 45% kaolinite + 20% (hematite + goethite)
r-
LI
Lii
,) so
25
1.100
1
WAVELENGTH
1 .5
J25 i
- MICRONS
1-'
1aw
/1
C
ei
c.
C50
15% MnO2 +
5% malachite + 30% kaolinite + 50%
(hematite + goethite)
i
15% MnO 2 + 20% malachite + 15% kaolinite + 50%
(hematite + goethite)
Lin
Ul
Ln
H.
CD
PU
-
1
w
WAVELENGTH - MICRONS
~ ~
~
aas
~
E5 Z vi5E~e
25% MnO 2 + 5% malachite + 50% kaolinite + 20% (hematite + goethite)
25% 'MnO2 + 5% malachite + 65% kaolinite +
5% (hematite + goethite)
4
i
0,
250
Q
WAVELENGTH -
M
25% MnO2 + 5% malachite + 20% kaolinite + 50%
25% MnO 2 + 5% malachite + 65% kaolinite +
(hematite + goethite)
5% (hematite + goethite)
L9
ro)
09
(D
L-4
Ln
F-Y75s
1
EG(H125 r-1IC
WAVELENGTH - MICRONS
I.
I
/s
w.
C *m
'
eww
c
CA
25% MnO
+ 5% malachite + 20% kaolinite + 50% (hematite + goethite)
25% MnO 2 + 5% malachite + 50% kaolinite + 20% (hematite + goethite)
L-e
~:ij
I-i.
CD
'31
-- j
V1
1/2
.e::
0t25
0)5
iL
115C O
Ei25
WAVELENGTH - MICRONS
'
i
El j
m5
i:JeCA
0
ful
25% MnO
+ 20% malachite +
2
5% kaolinite + 50% (hematite + goet ite)
25% MnO
+ 20% malachite + 35% kaolinite + 20%
2
(hematite + goethite)
CD
0~\
Ln
)nSO
s
r62
fb '7S
Ges
tiso
WfVELENGTH -,MICRONS
if2s
2J)O
ej2S
2.'so
25% MnO2 +
-
5% malachite + 50% kaolinite + 20%
25% MnO2 + 20% malachite +
5% kaolinite + 20%
(hematite + goethite)
(hematite + goethite)
(.n
U51
-4
CD
U5ln
:D
...
...
~
Gy)5
Q)-75
ij00
tj25
L!)5Q
WfWELENGTH - MICRONS
-
-L
I75
e
wa
a
--
50
25% MnO2 + 20% malachite +
25% MnO 2 +
5% kaolinite + 50%
(hematite + goethite)
5% malachite + 20% kaolinite + 50%
(henatite + goethite)
, "111",
ICD
X=:
Co
(il
-
is
(21b)
IiV(bbHjes
jICo
i)2S
WAVELENGTH - MICRONS
2JO0
2)ls
2lse
A P P E N D I X
I V
REFLECTANCE SPECTRA OF NATURALLY OCCURRING SOIL SAMPLES
Figure
1
: Mineral Park soil sample spectra
Figures 2-6: Silver Bell soil sample spectra
Figure
7
: Silver Bell rock chip sample spectra
204
205
50
40
LU 30
Z
20
LL.
I0
I0
\0
|0
0
0.5
1.0
1.5
2.0
2.5
WAVELENGTH (L)
Figure 1.
samples.
Reflectance spectra of the MIneral Park soil
206
50
40
2
LU 3 0
3
20
4
uio
-J 10
LL
-Ljj
1 - SS 1
10
2 -
SS 2
3 - SS 3
0*
4
10
10
-
SS
4
-
0
0.5
1.0
1 .5
WAVELENGTH
Figure 2. Reflectance of Silver Bell
2.0
(a)
soils.
2.5
207
50
5
40
6
zS30|020
-J
LL
8
1
O'
10
5 - SS 5
- SS 6
_6
- SS 7
-07
8 - ss 8
10
-
10
0
0.5
1 .0
1.5
WAVELENGTH
Figure 3.
2.0
(p)
Reflectance spectra of Silver Bell soils.
2.5
208
50
40
z
030
20
uL
10
10
9---9
1-20
10
10 -
SS 10
11 -
SS 11
12
SS 12
-
200- -0.5
1.0
1.5
WAVELENGTH
Fifgure
4.
2.0
(p)
Reflectance spectra of Silver Bell soils.
2.5
209
50
40
U
Z
30
20
U_j)
10
10
13 - SS 13
0
14 - SS 14
10
15 - SS 15
16 - SS 16
10
0
0.5
1.0
1.5
WAVELENGTH
2.0
2.5
(pL)
Figure_5. Reflectance spectra of Silver Bell
soils.
210
50 -40
18
-
UJ 30
2
<.. 20
-9
-
20
LU
LU
10
17 - SS 17
10
18 -
SS 18
19 -
SS 19
20
SS
--
20
I0
-
0
0.5
1.0
1.5
WAVELENGTH
Figure
6.
2.0
2.5
(u)
Reflectance spectra of Silver Bell
soils.
211
60
50
40
LUJ
Q
z
30
-2
LU
LOft
10
A - GOC 43
B - GOC 78
C - GOC 44
IN0
E - GOC 49
10
10
0
0.5
1 .0
O.5
WAVELENGTH
Figr
2.0
2.5
(L)
. Reflectance spectra of Silver Bell rock chips.
A P P E N D I X
V
RATIO SPECTRA OF NATURALLY OCCURRING SOIL SAMPLES
Figures 1-4: Mineral Park soil sample ratio spectra
Figures 5-28: Silver Bell soil sample ratio spectra
Note: Each spectrum was scaled to unity at 0.87 p
before ratioing.
212
AC13 / AC46
I'll -
Q0
0)25
1)1010
i25
WAVELENGTH
i )50
-
MICRONS
i )-7S
2)00
2)25
2.150
DF13 /
DF46
1
1,10
111,
0
ru
Gaes
t)7S
WAVELENGTH - MICRONS
2
0
ees
2Jso
DF13 /
12
0.Se
s
AC46
U)-75
l)eo
0)~7)
12
WAVELENGTH -
MICRONS
200
2)27
2j50
AC13 / DF46
UDE
S-
f3
'/
l
wbw
-41r
0.2-b
t
w
1. j: Z)
1
0 I
WAVELENGTH - MICRONS
I
I2
c- Ijw
2)251
!5
50
SS1 / SS2
I.~~1.
I-s
CD
f-
ro
L
-
A5
VE25NG)
2
WAVELENGTH
MRN
-MC
O
1125
2109
2j25
2.)50
II
i~. ~
I
1.25
I
1.50
I
1.25
Figure 6
I
1.00
RATIO
0.75
0.50
9.25
218
to
z
r-
0
Lf9 L..J
Rit
8S5 / ss8
(ll
s.eb
ka. lb
1 je
-L
g
WAVELENGTH -
I
MICRONS
4- tA
p
h
Pips
2s
so
se
SS8 / SS6
-J
LFI
u
Ln
;co
-4.
-J
-
t 1
WAVELENGTH -
MICRONS
75
.'Es
)5
fU
S
SS11 /
..
SS10
01
U1
00
\D
OS
SU,
S
U1
---
-j
- - -
0)500S25
0100
t 25
WAVELENGTH -
t se
MICRONS
I.
Z
i /s
ww
G~ Q
t
c
A
5 t0f
SS11 / SS12
-".
Ul
U1
L,
.F S
IS
- o)25
Oj
i)
tj5
WAVELENGTH -
1iJs
IS0
1 )25
MICRONS
w
e-.
2/25
2
5b0.,Z
ful
;Q
IS3
SS13 /
SS9
S-
Ul
Ln
Q.sbibL2G
100
U 25s
t1 b50
WAVELENGTH - MICRONS
I J 1 /5
2.00
2.125
2.150
SS14 /
SS12
U1
ru
-
61 -
ru
(A
gjg
Us
-soS
f
1GTes
00)
WAVELENGTH - MICRONS
ins
e
ej
)eis
2es
8S19 /
ss16
S
~
225
0)50
0)5
I.,
e
i5 so
WAVELENGTH - MICRONS
U /b
2j75.1
SS19 /
SS20
-J
U,
U,
U,
H'
2'
CD
4:::-
-J
U,
S
-
'I,
S
,-I--
SO
-00 IC5o
V25
1.25
WAVELENGTH - MICRONS
I,
-
/
Cw
)
2)5
C,2
SS18 / SS12
m4~
lo11
11,"All
11,111
116
1,11111110
, 4l,lilill
3>1
MIVELENG TH
-
MICRONS
SS11 / SS2
(D
0\
l
G-
cn
U1'
II
0s5
(b2
U00
V-7Stjes
1iiCo
1I7
WAVELENGTH - MICRONS
2y0
2 25
-2'50
SS16 / SS2
-I
(ft
Li~
(ft
-4
W6VELENGTH
-
MICRONS
I P7S
2joo
2j~s
~2.is
SS20 /
SS2
Ul
Ln
I
Ln
4v
"All''P,
11F'll
101111,1%111
Ln
4-
ro
WAVELENGTH -
MICRONS
ins
eles
25
e.es
se
Ss3 / Ss6
-J
ru
L
S
Ul
-.-
U1
Ul
ru
J2s e~uso
0.75
1)00
1.M3s
WAVELENGTH -
iise
MICRONS
i j-s
easo
2.12s
e.so
SS4 / SS6
r-
<n
L1q
CD
-450
ED
a
U,
-J
L9
isojso
0)7S
100J
tjs
t-so NS0 iMCRO
WAVELENGTH - MICRONS
2
e.so
SS7 / SS6
01
Cr'
Ul,
L2
Ul
ru
.25
s
I25
02~55
t) 5 NS
M25
9!;;lWAVELENGTH - MICRONS
s
SS7 /
SS8
U
-i
'A
i
f-f,
ru
Uo
p-
I1
A . 1 )CI
iz
L
0
AI
lj/b
WAVELENGTH
I. /b
-
MICRONS
2.100
2j~s
-2.1s
Vn
SS7 / SS10
D-
U'
!i,
S
-
da
Su
0
WAVELENGTH
-
1)so ----MICRIONS
7 S
2)oo
2j2S
.S
SS15
/ SS16
-j
U,
U,
S
-4..
(D
-Jq
UX=
IQ
1
i
WAVELENGTH -
MICRONS
)?5
Ejs
~gg
E25
S
SS15 / SS17
Ul
m
(Jl
-
ru
n
pls qi A
1
1
. 0
1
1 )5
WAVELENGTH -
15
MICRONS
-
--- --------
SS18 / SS17
S U1
U'
1-1PL
015
02S
1 go
i.2s
WAVELENGTH -
I
MICRONS
1Utb
i:'Jiow
2125
~250
2150
SS2 /
SS9
(f
09
(D
fr>
Lo
ru
-
s
s
ibeIn
-- i !es
MICRONS
WAVELENGTH - MICRONS
I.sns
eJOT
2s 2
eEs
SS2 / SS15
-L
c0
00
(f?
SI
.!e
0j Sib
a !-7c:
4i
ICA~
WAVELENGTH
"A 41c
-
MICRONS
n