54 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

Geochemical Analysis (Including X-ray)

R H Worden, University of Liverpool, Liverpool, UK signal over a range of concentrations (Figure 1). The
ß 2005, Elsevier Ltd. All Rights Reserved.
detector outputs from the standards must be converted
into some sort of equation that permits the eventual
back-conversion of output signals into concentra-
Introduction tions for unknown samples. In the example in Figure 1,
Geochemistry is the study of the occurrence and dis- the output must be determined as a function of concen-
tribution of elements, isotopes, minerals, and com- tration (equation 1) and then inverted to allow con-
pounds in the natural environment. Geochemical centration to be determined from the output signal for
analysis, the measurement of the quantities of elem- unknown samples (using equation 2).
ents, isotopes, minerals, and compounds in a rock, The quality of the analytical output from a device is
natural liquid, or naturally occurring gas, is a colossal fundamentally a function of the quality of the calibra-
topic. tion curve. No calibration is perfect, and the degree of
The applications of geochemistry stretch from the imperfection can be described in terms of accuracy
core to the outer atmosphere and beyond. The objects and precision. These discrete characteristics are best
of study include minerals, metals, and salts, organic described for a dataset composed of repeated analyses
biomolecules, coal, bitumen, kerogen, and petrol- of the same standard sample (Figure 2). When the
eum, atmospheric-, river-, ground-, and formation- results are widely dispersed about the known answer,
water, and gases in the crust, sediments, and the output is said to be imprecise. When the results are
atmosphere. Analysis is also a huge topic, with sub- tightly clustered about the known answer, the output
jects ranging from rocks, minerals, compounds, and is said to be precise. If the results cluster around the
species to isotopes, elements, and atoms. correct figure, they are said to be accurate, whereas if
they cluster about a figure other than the correct
The Role of Analysis: Hypotheses, Questions, output, they are said to be inaccurate (Figure 2). Ac-
Problems, and Theories curacy and precision are the combined result of the
quality of calibration, the sensitivity of the device,
In the context of Earth sciences, geochemical analysis
commonly implies quantitative work with a numer-
ical output. Geochemical analysis is thus the quantita-
tive examination of the composition of natural Earth
materials. Most geochemical work is undertaken to
address a hypothesis, answer a question, or solve a
problem. Even routine environmental geochemical
monitoring is done in order to address the question
of the continued safety of natural materials over time.
It is typically important to formulate a hypothesis,
question, or problem carefully before embarking on
an analytical programme. Such an approach is best
since the outcome should be cost-effective while being
statistically rigorous and ultimately defensible.

Producing Geochemical Data

When any analytical geochemical device is used to
quantify the composition of a material, the detector
output is usually in the form of some sort of electrical
signal. The electrical signal must be converted into a Figure 1 Example of a calibration curve using standards of
meaningful geochemical parameter, such as a unit of known concentration and their output signal strengths. The
concentration. This conversion is usually achieved via small black dots represent the calibration. Equation 1 describes
a calibration curve. For any technique, a series of how concentration could be converted to signal strength. The
more useful equation 2 describes how signal strength from an
previously characterized standards must be analysed,
unknown sample can be converted into concentration. Unknown
in exactly the same way as an unknown sample, and sample A yields a valid result since it falls within the calibration
the output signals measured. In essence, the calibration range. The analysis of sample B is not valid since the output
is a plot of known concentration versus the output signal falls outside the calibration range.
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 55

Figure 2 Representation of analytical precision and accuracy. Precision is a measure of how tightly the results cluster about the mean
result. Accuracy measures the proximity of the mean result to the expected result. Accuracy can be improved by recalibrating the device.

the attainment of constant operating conditions, and technique. Finally, it is worth running standards
the skill of the device operator. throughout an analytical run at regular intervals.
Developing and checking the calibration lines are
Geochemical Analytical Protocol (Blanks, the most crucial steps in geochemical analysis, and
Standards, Repeats)
being aware of the upper and lower limits of the
One of the most important aspects of any geochem- calibration is important. Many calibration lines do
ical analytical protocol is to ascertain the credibility not vary linearly with concentration. Thus, if any
of the data. Every geochemical analysis has some geochemical analytical result falls outside the cali-
degree of associated uncertainty (see Figure 2, where brated range of the detector and its output, then the
uncertainty is expressed in terms of precision). Every result is theoretically invalid since it is not certain
technique also has a minimum detection limit, below how to convert the instrumental signal (e.g. amps,
which the concentration cannot be determined. The volts etc) into geochemically significant units.
theoretical lower detection limit is an intrinsic func-
The Range of Geochemical Analytical Techniques
tion of the technique, although the quality of the
calibration and the cleanliness of the equipment As stated earlier, geochemical analysis is a large topic,
used in sample and standard preparation also limit and many analytical techniques have been employed
detection. It is seldom correct to report a figure of in geochemical research and studies over the last
zero concentration. It is more correct to report the four or five decades. Techniques may be grouped
detection limit and state that the sample had a con- according to the attribute being measured (Table 1).
centration below that value (‘below detection’ or Many important techniques use X-rays in a variety
‘none detected’). In any analytical run it is worth- of ways for analytical purposes. Other techniques
while running blanks to check the lower detection use the optical effects of samples, typically taking
limit. It is also worth discretely running repeat advantage of electromagnetic radiation emitted or
samples to check the reproducibility of the analytical absorbed by the material at elevated temperatures.
 56 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)
Table 1 Summary of some of the main techniques of geochemical analysis and their sample types

Solid (rock, mineral, Artificially Water Fluids

sediment, soil, dissolved (natural Fluid Solid trapped in
Technique Output types Abbreviation filtrate) solid solution) organic organic Gas solids

X-ray techniques
X-ray analysis (electron Point mineral composition MPA Yes Yes Yes
microprobe)
X-ray fluorescence Rock composition XRF Yes Yes Yes
X-ray diffraction Mineralogy XRD Yes Yes
Spectroscopic techniques
Atomic absorption Solution composition AAS Yes Yes Yes
spectroscopy
Inductively coupled plasma Solution composition ICP-OES Yes Yes Yes
optical emission
spectroscopy
UV spectroscopy Composition UVS Yes Yes Yes

Chromatography
Gas chromatography Gas composition GC Yes Yes Yes Yes
Ion-exchange Aqueous-liquid composition IC Yes Yes Yes
chromatography
Mass spectroscopy
Mass spectroscopy Composition, isotope content MS Yes Yes Yes Yes Yes Yes Yes
Isotope dilution mass Isotope ratio IRMS Yes Yes Yes Yes Yes Yes
spectroscopy
Inductively coupled plasma Composition ICP-MS Yes Yes Yes Yes Yes
mass spectroscopy
Gas chromatography mass Composition GC-MS Yes Yes Yes Yes
spectrocopy
Thermal analysis techniques
Pyrolysis Elemental analysis Yes Yes
(C,H,O,S,N), organic type
Evolved water analysis, Clay content, clay mineralogy EWA-TG Yes
thermogravimetry
Fluid inclusion Temperature of trapping, FI Yes Yes
microthermometry composition
Related techniques
Wet chemistry techniques Composition Yes Yes Yes
Microscopy (light and electron Mineralogy, fabric, SEM,TEM Yes Yes Yes
optics) crystallography
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 57

Other techniques use chromatography, the time taken by the atom during the slowing of the electrons
for one substance to move through another or through is excessive to the atom and will be radiated as
a capillary under a given gradient. A wide family of X-radiation of equal energy to that absorbed. Brems-
geochemical analytical techniques use mass spec- strahlung X-rays tend to have a broad range of ener-
trometry to split propelled material, converted into gies since the degree of slowing can be variable and
charged particles, using electromagnets. Other tech- materials composed of mixtures have atoms with
niques involve examining the products of heating different properties (Figure 6). Bremsstrahlung tend
Earth materials under controlled conditions and not to be used for geochemical analysis; that is the
studying either the evolved fluids or the changes in preserve of electron-shell emission.
the properties of the residual solids.
Analysis of X-rays: Electron-Shell Emission
A common geochemical application of X-ray analysis
X-ray Techniques is to direct a focussed electron beam at a polished rock
or mineral surface and then collect and quantify the
Origin of X-rays
resulting secondary characteristic X-rays (Figure 7).
X-ray technologies have proved to be useful in geo- The secondary X-rays help to reveal the elements
chemical analysis (Table 2). X-rays are part of the present in that part of the sample that is directly
electromagnetic spectrum (Figure 3) and have wave- under the electron beam. This technique is known as
lengths ranging between 0.01 nm and 10 nm (0.1– electron-beam microanalysis, or microprobe analysis,
100 Å). They are waveforms that are part of a family and gives spatially resolved major- and trace-element
that includes light, infrared, and radio waves. Since geochemical data from solid samples, including rocks,
X-rays have no mass and no electrical charge, they are minerals, sediments, soils, and glass. Many ordinary
not influenced by electrical or magnetic fields and electron microscopes are fitted with a secondary
travel in straight lines. X-rays, like all parts of the X-ray detector, making them suitable for geochemical
electromagnetic spectrum, possess a dual character, analysis. All of these devices rely on electron optics,
being both particles and waves. The name that has using electromagnetic lenses to focus and direct a
been given to the small packets of energy with these stream of electrons, generated by an electron gun,
characteristics is photon. onto a polished mineral or rock surface (Figure 7).
The simple model of the atom, proposed by Niels The focused electron beam has a variable radius, but
Bohr in 1915, is not completely correct, but it has can typically be maintained at slightly greater than
many features that are approximately correct. The about 1 mm. The spatial resolution of a microprobe
modern theory of the atom is called quantum mech- is actually somewhat greater than 1 mm. The impin-
anics; the Bohr model is an approximation to quan- ging electron stream interacts with the polished sur-
tum mechanics that has the virtue of being much face and produces a wide range of signals, including
simpler than the full theory. In the Bohr model neu- secondary and backscattered electron and cathodolu-
trons and protons occupy a dense central region (the minescence (light) as well as the secondary X-rays of
nucleus), and electrons orbit the nucleus. The basic concern here. There is an activation volume from
feature of quantum mechanics that is incorporated in which X-rays are generated, below the polished sur-
the Bohr model is that the energies of the electrons face, which is several times larger than the primary
in the Bohr atom are restricted to certain discrete beam. Samples must be highly polished (flat) to avoid
values (the energy is quantized) – only certain electron scattering.
orbits with certain radii are allowed. When a sample is bombarded by an electron beam,
X-rays are generated when free electrons from some electrons are knocked out of their quantum shells
an electron gun give up some of their energy when in a process called inner-shell ionization (Figure 5).
they interact with the orbital electrons or the nucleus Outer-shell electrons fall in to fill a vacancy in a process
of an atom (Figure 4). The energy given up by the of self-neutralization. The shells are termed K, L, M,
electron during this interaction reappears as emitted and N starting from the innermost most strongly bound
electromagnetic energy, known as X-radiation. Two shell.
different atomic processes can produce X-ray Electrons moving from one shell to another pro-
photons. One is called bremsstrahlung and the other duce characteristic X-rays. K-shell ionizations are
is called electron-shell emission (Figure 5). Brems- commonly filled by electrons from the L shell (Ka
strahlung means ‘braking rays’. When an electron radiation) or M shell (Kb radiation). There are two
approaches an atom, it is affected by the negative Ka peaks (Ka1 and Kb2) corresponding to two dis-
force from the electrons of the atom, and it may be crete states of the in-falling electron. When outer-shell
slowed or completely stopped. The energy absorbed electrons drop into inner shells, they emit a quantized
 58 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)
Table 2 X-ray techniques commonly used in geochemical analysis

Technique Output-1 Output-2 Output-3 Sample type Advantages Disadvantages

X-ray analysis Quantitative Major elements Trace elements Polished rock High spatial resolution; gives Sample preparation can be slow;
elemental detected using detected using samples, or grains quantitative data with errors can be large for some
composition of energy dispersive wavelength set in resin and estimate of uncertainty; elements; problems of
small volume spectrometer dispersive polished well-established technology; analysis statistics for
(several cubic (SEM or spectrometer wide range of elements heterogeneous minerals
mm) microprobe) (microprobe)
X-ray fluorescence Quantitative major Crushed and Gives major and trace elements Sample preparation can be slow;
spectroscopy and trace homogenized simultaneously; produces relies on sample type-specific
elemental sample then data for Al and Si in rocks as calibration curve; matrix
composition of a compressed or well as a range of metals; effects can be large;
crushed melted and well-established technology instrumental set-up can be
homogenized quenched into arduous
sample small disks
X-ray diffraction Qualitative Relative proportion Mineral Fairly quick; well-established Semi-quantitative in most cases;
presence or of minerals in a composition technology; large database of difficult for minerals with solid
absence of rock or sediment minerals for comparison solutions; difficult for poorly
minerals in mixture – for crystalline materials; difficult
crushed sample suitably prepared with very complex mixtures
samples
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 59

Figure 5 Incident electrons cause electrons in metal atoms to

become excited and jump to outer orbitals. Ka and Kb X-rays (etc.)
are emitted as electrons fall back to their original orbitals. The
Figure 3 Representation of the electromagnetic spectrum with
X-ray energy is characteristic of the element and of the starting
the various wavelengths employed in geochemical studies.
and final orbitals. Background ‘white’ radiation (bremsstrahlung)
Visible light is not discussed here but forms the basis of optical
is due to collisions between incident and orbital electrons.
microscopy. Gamma rays are used to differentiate radioactive
fission reactions and thus identify and quantify radioactive
elements.

Figure 4 General set of processes involved in the generation of Figure 6 Example of a secondary X-ray trace from a sample of
X-rays by electron bombardment of atoms. The small grey filled calcite with a minor amount of iron substituting for the calcium
circle represents the nucleus. The outer rings represent the (known as ferroan calcite). The X-ray energy values are charac-
quantized electron orbitals of the Bohr atomic model. The thicker teristic of the elements. Note that the energy of the K lines in-
black circle represents the location of a given electron (e
). With creases with increasing atomic number. Note also the increasing
electron bombardment, the highlighted electron jumps to a separation of the Ka and Kb lines with increasing atomic number.
higher orbital. The energized electron quickly falls back to its The bremsstrahlung can cause problems with quantification,
original state, releasing an X-ray. especially at low secondary X-ray energies.

photon characteristic of the element. The resulting electrons, which arrive about every 10
12 s, can
characteristic spectrum is superimposed on the brems- repeatedly ionize an atom.
strahlung (Figure 6). An atom remains ionized for a It is common practice to measure X-rays in units of
very short time (about 10
14 s) and thus the incident thousand electron volts (KeV). One electron volt is
60 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

Figure 7 Basic equipment used for secondary X-ray analysis following electron-beam bombardment of Earth materials. The various
types of signals generated in the samples and the various detectors are highlighted in the enlarged diagram. Characteristic X-ray
emission occurs several micrometres below the sample surface in a volume with a radius of a few times the electron-beam radius
(spot size). Bremsstrahlung are generated in the outlying volume where adsorption of X-rays also occurs.

the same as 1.6021  10
19 J. An electron volt is the Secondary X-rays can be detected and quantified
amount of energy gained by one electron when it in two ways. One, which is best for major elements
is accelerated by one volt. The output characteristic (>0.1 wt%), involves measuring the energy of the
X-rays have an energy and wavelength controlled by emitted X-rays using a scintillation counter (known
the element that is present and the specific nature of as energy dispersive analysis). The other, best for
the electron orbital transition. Elements with low trace elements (>1 ppm at best), measures the wave-
atomic numbers produce low-energy X-rays, and length (and intensity) of the emitted X-rays (known as
many X-ray detectors have difficulty quantifying the wavelength dispersive analysis). Energy dispersive an-
output from elements with atomic numbers of less alysis is typically faster than wavelength dispersive
than 11 (sodium). The energy and wavelength of the analysis but has much lower sensitivity.
characteristic X-rays are related by the equation Most modern devices are computerized and have
E ¼ hc/l, where E is the energy (in KeV), h is the inbuilt quantification and correction systems. The
Plank constant (6.626  10
34 J s), c is the speed of geochemical output from electron microprobes is in
light (2.99782  108 m s
1), and l is the element and the form of oxide or element weight percentages.
electron orbital-specific wavelength of the X-rays.
X-ray Fluorescence
Conversion of X-ray energy into wavelength is thus
achieved using E (eV) ¼ 123985/l (in nm). X-ray Although X-ray beams cannot be focussed or bent,
spectra are typically plotted in terms of energy. The they can be directed by a series of diffraction gratings,
outputs of characteristic X-rays have intensities collimators, and slits. Such X-ray beams have been
that are a complex function of the quantity of the used for geochemical analysis by a technique known
element in the analysed volume. In general, the in- as X-ray fluorescence spectroscopy (XRF). The output
tensity is approximately proportional to the relative from this technique gives the concentrations of most
concentration. Several matrix (mineral) dependent major elements (with atomic numbers of 11 (sodium)
processes alter the primary intensity of the secondary and above) and many useful trace elements.
X-rays. The atomic mass of the element influences It was stated earlier that some secondary X-rays
the efficiency of X-ray generation (Z-correction), can be absorbed by material and that some of this
some secondary X-rays are absorbed by the material absorption can result in localized X-ray fluorescence
(A-correction), and some of the absorbed X-rays (Figure 8). Just as with primary electron beams, when
result in localized X-ray fluorescence (F correction). a primary X-ray excitation source (e.g. from an X-ray
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 61

Figure 8 General set of processes involved in the generation of fluorescent X-rays by X-ray bombardment of atoms. The small grey
filled circle represents the nucleus; the outer rings represent electron orbitals. The thicker black circle represents the location of a
given electron (e
). With X-ray bombardment, the highlighted electron jumps to a higher orbital. The energized electron quickly falls
back to its original state, releasing a secondary X-ray. The range of elements in a sample leads to a range of characteristic fluorescent
X-rays with peak heights that are functions of the amounts of the elements in the sample.

tube or even a radioactive source) strikes a sample, X-ray Diffraction

the X-ray can be either absorbed by the atom or X-rays have a similar wavelength to the lattice spacing
scattered through the material. Some incident
of common rock-forming minerals and have been used
X-rays are absorbed by the atom, and electrons are
to characterize the crystal structure and mineralogy
ejected from the inner shells to outer shells, creating
of Earth materials by using X-ray diffraction (XRD)
vacancies. As the atom returns to its stable condition,
analysis. This is most commonly used to define the
electrons from the outer shells are transferred to the
presence of minerals, mineral proportions, mineral
inner shells and in the process give off a characteristic
composition (in favourable circumstances), and other
X-ray with an energy equal to the difference between subtle mineralogical features of rocks, sediments, and
the two binding energies of the corresponding shells. soils.
XRF is widely used to measure the elemental com-
X-rays, even from a pure elemental source bom-
position of natural materials, since it is fast and non-
barded with electrons, have a collection of peaks –
destructive to the sample. It can be used to measure
X-rays characteristic of the quantized electron energy
many elements simultaneously. XRF can be used dir-
levels – and bremsstrahlung. X-ray beams of a tightly
ectly on rock surfaces, although there is a danger of
defined energy (and thus wavelength) have been used
natural heterogeneity resulting in variable results.
to investigate and characterize the minerals present
Rock, soil, and sediment samples are typically crushed in rocks, sediments, and soils by removing all but one
and made into pellets by compressing them or by X-ray peak from the spectrum of wavelengths gener-
melting the whole sample and then quenching to
ated by a source element. X-rays are useful in investi-
make a glass disk. XRF is useful for the geochemical
gating mineral structure since they can be selected to
analysis of a wide range of metals and refractory and
have a wavelength that is only just smaller than the
amphoteric compounds (such as SiO2 and Al2O3) and
interlattice spacing (d-spacing) of common rock-
even some non-metals (chloride and bromide). XRF
forming minerals. A number of X-ray sources have
is also routinely used to measure the natural metal
historically been selected, but copper is the most
content of liquid petroleum samples. The quality of commonly employed, and the copper Ka peak is the
XRF data is a function of the selection of appropriate one that is directed onto samples. This has a charac-
standards. It is considered to be best practice to
teristic wavelength of 1.5418 Å (0.15418 nm). This is
use standards that are similar to the samples in ques-
ideal for many minerals since they have high-order
tion to minimize matrix effects. XRF can measure
(dominant, most obvious) lattice spacings of this size
down to parts-per-million concentrations and lower,
or up to 10–15 times greater than this wavelength.
depending on the element and the material.
62 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

The melee of X-radiation from copper can be reduced for constructive interference (‘diffraction’) to occur:
to the Ka peak and then directed by a series of 2dsiny ¼ nl, where d is the lattice spacing, l is the
diffraction gratings, collimators, and slits. wavelength of the incident X-ray source, and n is an
The main features of an XRD analyser include an integer (typically one in many cases). The value of y,
X-ray source with collimators, slits, etc, a sample, defined in Figure 9, is a function of the variable
and an X-ray detector (Figure 9). The source and geometry of the XRD equipment.
detector are both rotated about the sample, an arbi- X-ray diffraction is most commonly used on
trarily fixed point, and define the same angle (y) crushed (powered) rock samples to ensure homo-
relative to the sample. The angle between the source geneity of the sample and randomness of the orien-
and the detector is thus 2y relative to the sample. tations of all the crystal lattices represented by
Diffraction occurs when X-rays, light, or any other different minerals. This is known as X-ray powder
type of radiation passes into, but is then bounced diffraction.
back out of, a material with a regular series of layers. XRD works by rotating the X-ray source and the
Diffraction occurs within the body of the material detector about the sample from small angles (e.g. 4 )
rather than from the surface (and so is quite different through to angles of up to 70 . The low angles can
from reflection). Regular layers are a characteristic detect large interlattice spacings (large values of d)
of all crystalline materials (minerals, metals, etc). while the high angles detect smaller interlattice
Each rock-forming mineral has a well-defined set spacings. For CuKa radiation these angles equate to
of these layers, which constitute the crystal lattice. d-spacings from about 30 Å down to about 1.5 Å,
No two minerals have exactly the same crystal struc- covering the dominant d-spacings of practically all
ture, so fingerprinting a mineral by its characteristic rock-forming minerals.
set of lattice spacings helps to identify minerals. A For a pure mineral sample, the diffraction peaks
radiation beam from a pure source has a defined from different lattice planes with discrete d-spacings
wavelength, and the rays from such a pure source have different relative intensities. This is a function of
will be ‘in phase’. Constructive interference occurs the details of the crystal structure of a particular
only when all the outgoing (diffracted) X-rays are mineral, but the maximum-intensity trace (peak) for
also in phase. Destructive interference, the norm, many minerals has a low Miller Index value (a simple
occurs when the diffracted X-rays are no longer in notation for describing the orientation of a crystal).
phase. Constructive interference occurs when the For example, many clay minerals dominated by sheet-
extra distance that X-rays travel within the body of like crystal structures have (001) as the maximum-
the material is an integer (whole number) multiple of intensity peak. All other XRD traces have intensities
the characteristic wavelength of the incident X-ray that are fixed fractions of the intensity of the max-
(Figure 10). The geometry of the XRD equipment, imum-intensity trace. The result for each pure min-
the wavelength of the incident radiation, and the eral is a fingerprint of XRD peaks on a chart
lattice spacing are all important in defining whether of intensity on the y-axis and 2y on the x-axis
constructive interference occurs. The key equation (Figure 11). This can be compared with collections
is known as the Bragg Law, which must be satisfied of standards to identify the mineral.

Figure 9 Basic elements of an X-ray diffraction device. An

X-ray source is directed at a sample at a controlled and variable
angle (y). The X-ray detector is at the equivalent angle on the Figure 10 Diffraction from a crystal. The incident X-rays are in
opposite side of the pivot point. The source and detector are at an phase as they hit the mineral surface. The grey line shows the
angle of 180
2y to one another. The source and detector are path-length difference between the two X-rays. Constructive
thus simultaneously rotated about the pivot point. When diffrac- interference occurs when the extra path length (2d siny) is an
tion occurs the X-ray detector records a signal above the back- integral multiple (typically one) of the wavelength of the X-rays.
ground. The sample is usually a powder and preferably randomly Constructive interference leads to an X-ray diffraction peak set
orientated. against a low level of background noise.
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 63

Figure 11 The X-ray diffraction output from a rock composed of the clay minerals muscovite (pale grey peaks) and kaolinite (black
peaks). The two minerals have discrete d spacings, which are reflected in their discrete peaks on the chart. The peaks at the low 2y end
have high d-spacings, and vice versa. The dominant basal spacings of these two sheet silicates (the (001) planes) are labelled. The
maximum-intensity diffraction peaks from these minerals are produced by the (001) planes. From this diagram it would appear that
there is much more kaolinite than muscovite in the mixture (approximately three times as much) since its maximum-intensity peak is
much more intense. The mixture therefore has about 25% muscovite and 75% kaolinite.

The intensity of the collection of diffraction peaks

from a given mineral in a mixture (e.g. rock, soil, or
sediment) is broadly a function of the proportion of the
mineral in the mixture. This is a subject of ongoing
research since the issue is complicated by different
minerals having different efficiencies at diffracting
X-rays. In simple terms, the intensities of the max-
imum-intensity peaks from a mixture of minerals
give a guide to the relative proportions of the different
minerals (Figure 11).
There is a range of problems for XRD when dealing
with natural Earth materials. The most obvious is the
simple identification of minerals when each one has its
Figure 12 X-ray diffraction for the same peak from the same
own collection of diffraction peaks. These must be mineral may occur at a different 2y value if there is solid solution
carefully deconvoluted by a process of elimination. and the substituted element has a different ionic radius. For
Computer-based programs can be of great help in example, in calcite the dominant (104) diffraction peak moves
this task. Many rock-forming minerals do not conform systematically as magnesium replaces calcium. Magnesium
ions are smaller than calcium ions so the structure collapses
to a perfectly defined composition. The set of interlat-
slightly. In favourable circumstances, this approach can be
tice d-spacings in minerals is a function of the way used to help determine mineral composition.
atoms are packed together in the lattice, so that com-
positional variation affects the precise 2y position of a
given crystal orientation. In some cases, this variability the transformation of a poorly defined crystal struc-
can be put to good use since it can be used to identify ture into a well-defined structure (typically as a funct-
the composition of a given mineral (Figure 12). ion of time and temperature) during diagenesis or
Another issue with XRD analysis, especially of metamorphism (Figure 13).
sediments, soils, and sedimentary rocks is that many
minerals have poorly defined crystal structure. This
Optical Techniques
problem is typical of clay minerals, pedogenic oxides,
etc. Poorly defined crystal structure translates into Spectroscopy is the use of the absorption, emission, or
wider XRD peaks. This presents problems for quan- scattering of electromagnetic radiation by atoms or
tification and results in a need to measure the area molecules (or atomic or molecular ions) to study the
under an entire peak rather than the height of the atoms or molecules qualitatively or quantitatively.
maximum-intensity peak. However, this too can be Isolated atoms can absorb and emit packets of
put to good use by enabling the quantification of electromagnetic radiation with discrete energies that
64 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

atoms. Since samples are usually liquids or solids, the

sample atoms or ions must be vaporized in a flame
or graphite furnace (Figure 14). The atoms absorb
ultraviolet or visible light. Concentrations are usually
determined from a working curve after calibrating the
instrument with standards of known concentration
(Figure 15). The Lambert–Beer Law is the relationship
between the change in light intensity for a given
wavelength and the relative incident light energy:
log Io/I ¼ aLc, where I is the light intensity after the
metal is added, Io is the initial light intensity, a is a
machine-dependent constant, L is the path length of
light through the torch, and c is the concentration.
The light source is usually a hollow-cathode lamp
of the element that is being measured. A major disad-
vantage of these narrow-band light sources is that
only one element can be measured at a time. AA
spectroscopy requires that the target atoms be in the
gas phase. Ions or atoms in a sample must be desol-
vated and vaporized in a high-temperature source
Figure 13 X-ray diffraction outputs from a mineral (illite) with such as a flame or graphite furnace. Flame AA can
variable degrees of crystallinity. The top image represents a analyse only solutions. Sample solutions are usually
well-crystalline sample, e.g. a slate. The lower image represents aspirated with the gas flow into a nebulizing and
a poorly crystalline material, e.g. from a juvenile sedimentary
rock, soil, or weathering horizon. The degree of crystallinity can
mixing chamber to form small droplets before
be quantified by measuring the peak width at half the peak height. entering the flame. AA spectrometers use monochro-
mators and detectors for UV and visible light. The
main purpose of the monochromator is to isolate the
absorption line from background light due to inter-
are dictated by the detailed electronic structure of ferences. Photomultiplier tubes are the most common
the atoms. When the resulting altered light is passed detectors for AA spectroscopy.
through a prism or spectrograph it is separated This technique can be used to analyse aqueous
according to wavelength. Emission spectra are pro- samples with negligible sample preparation. It can
duced by high-temperature gases. The emission lines also be used for rock and mineral samples if they are
correspond to photons of discrete energies, which are quantitatively dissolved (typically using acids of vari-
emitted when excited atomic states in the gas make ous strengths). Flame AA spectroscopy can typically
transitions back to lower-lying levels. Generally, detect concentrations as low as mg l
1 (ppm), al-
solids, liquids, and dense gases emit light at all wave- though graphite-furnace AA spectroscopy has been
lengths when heated. An absorption spectrum occurs shown to be able to detect concentrations that are
when light passes through a cold dilute gas and atoms orders of magnitude below this.
in the gas absorb light at characteristic frequencies.
Since the re-emitted light is unlikely to be emitted Inductively Coupled Plasma Optical Emission
Spectroscopy
in the same direction as the absorbed photon was
travelling, this gives rise to dark lines (absence of ICP-OES is short for optical (or atomic) emission
light) in the spectrum. spectrometry with inductively coupled plasma.
Emission and absorption spectroscopy have been Plasma is a luminous volume of atoms and gas at
used widely in geochemical analysis for many years extremely high temperature in an ionized state. The
(Table 3). They are commonly used to analyse waters plasma is formed by argon flowing through a radio
but can also be employed to analyse rock and other frequency field, where it is kept in a state of partial
solid samples that have been quantitatively dissolved ionization, i.e. the gas consists partly of electrically
(e.g. in acid solutions). charged particles. This allows it to reach very high
temperatures of up to approximately 10 000 C. At
Atomic Absorption and Atomic Emission
these high temperatures, most elements emit light of
Spectroscopy
characteristic wavelengths, which can be measured and
Atomic-absorption (AA) spectroscopy uses the absorp- used to determine the concentration of the elements
tion of light to measure the concentration of gas-phase in the solution.
Table 3 Optical spectroscopic techniques commonly used in geochemical analysis

Technique Output-1 Output-2 Output-3 Sample type Advantages Disadvantages

Atomic absorption Element concentrations Element Water sample from the Well-established Relatively high detection
spectroscopy (AAS) in water concentration in natural or altered technique; relatively low limit; relatively slow –
quantitatively environment (or costs one element at a time;
dissolved rock solid rocks and limited range of
samples minerals dissolved elements
in acid)
Inductively coupled plasma Element concentrations Element Water sample from the Many elements analysed Expensive equipment;
optical emission in water concentration in natural or altered simultaneously; technique not for the
spectroscopy (ICP-OES) quantitatively environment (or relatively fast; linear novice analyst; carrier
dissolved rock solid rocks and calibration over wide gases cause

ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 65

samples minerals dissolved concentration range; interference with some
in acid) wide concentration elements; minerals and
range; one calibration rocks must be dissolved
suitable for most prior to analysis;
material types; good for relatively new technique
water samples; excellent still undergoing
(ppb) resolution for development
some trace elements
Infrared spectroscopy Mineral proportions Water content in Clay and other Ultra-small samples; Complex to interpret
quartz minerals quantitative; relatively absorption spectra;
low cost; simple sample adsorbed water can
preparation interfere with diagnostic
peaks
Ultraviolet spectroscopy Presence of organic Presence of liquid Approximate Organic-bearing rock, Simple to make qualitative Some minerals have
inclusions in organics in indication of sediment, or soil observations; easily masking fluorescence;
minerals and rocks porous the density repeated; can be used difficult to quantity
materials (e.g. and on bulk and microscopic
pollution or oil) maturity of samples
petroleum
66 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

Figure 14 Atomic absorption spectroscopy equipment. The aqueous sample for analysis is injected into the burner and light
is shone through the flame. The mirrors (labelled M) are rotated to measure original and the absorbed light characteristics. The
path that is deflected away from the flame (labelled as unaltered energy path) measures the unaffected light intensity. The path that
goes through the flame (labelled as absorbed energy path) measures the intensity of the light after it has passed through a burner
containing the co-injected dissolved sample.

Ultraviolet Spectroscopy
Many organic-derived materials fluoresce under
ultraviolet (UV) light – that is, they absorb light from
the incident ultraviolet beam and release light, often
in the visible range, that has a different wavelength
from the primary UV light. UV spectroscopy can be
used qualitatively to determine whether complex or-
ganic molecules are present in sediment or a rock.
UV spectroscopy is commonly used to determine the
presence of diffuse oil-shows in petroleum-reservoir
cores.
There is a loose correspondence between the pre-
cise wavelength of the fluorescent light and the nature
Figure 15 Light absorption due to atomic excitation with base-
line and post-excitation light intensities indicated. The frequency of the organic material. This relationship has been
of the absorbed light depends on the element. Concentration is developed into an analytical technique to determine
proportional to Io/I, all other things being equal. the thermal maturity of petroleum in rock samples.
This technique has reached its apotheosis in its appli-
cation to petroleum trapped in inclusions in mineral
cements and healed fractures. These fluid inclusions
The sample being analysed is introduced into the have been used to track the petroleum generation of
plasma as an aerosol of fine droplets (Figure 16). Light source rocks and the petroleum migration history into
from the different elements is separated into different reservoirs (see Fluid Inclusions).
wavelengths by means of a grating and is captured
Infrared Spectroscopy
simultaneously by light-sensitive detectors. This per-
mits the simultaneous analysis of up to 40 elem- The electrical bonds between molecules continually
ents, and ICP-OES is consequently a multi-element vibrate as a function of interaction between neigh-
technique. In terms of sensitivity, ICP-OES is gener- bouring molecules. These bonds can be excited by
ally comparable with graphite furnace AA, i.e. detec- infrared radiation, resulting in higher amplitudes of
tion limits are typically at the mg l
1 level in aqueous vibration. Only discrete (quantile) increases in vibra-
solutions. tion energy are possible, and this results in an
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 67

Figure 16 Essential components of an inductively coupled plasma optical emission spectrometer. The induction coils produce a ring
of plasma at temperatures of ca. 10 000 C. The water sample, drawn in by the flowing carrier gas, is dispersed and drawn into the
plasma. Elements emit light due to thermal excitation. A wide range of light frequencies can be analysed simultaneously using a
diffraction grating to disperse the light.

absorption spectrum. The molecular environment is over them. Other components of the sample that are
different in all minerals, even in minerals of similar less strongly adsorbed on the stationary phase and
structure (e.g. the clay family), so that infrared spec- moved along more quickly by the moving phase.
troscopy can be used to identify minerals in rock and Thus as the mobile phase flows through the column,
sediment samples. The strength of the absorption components in the sample move down the column at
spectrum depends on concentration, so the technique different rates and therefore separate from one an-
can be used for quantitative analysis of minerals other. At the end of the column, molecules or ions of
under favourable circumstances. Infrared spectros- the fastest-moving substance (least tightly bound to
copy has been used to identify clay minerals in sand- the stationary phase) emerge first, usually with each
stones but is most useful for identifying organic compound emerging over a well-defined time inter-
inclusions and non-hydrocarbon gases trapped in val. A suitable detector analyses the output at the end
fluid inclusions. of the column. Each time molecules or ions of the
sample emerge from the chromatography column
the detector generates a measurable signal, which is
Chromatography recorded as a peak on the chromatogram. The chro-
Chromatography has proved invaluable in geochem- matogram is thus a record of detector output as a
ical analysis (Table 4). It is an analytical technique function of time and consists of a series of peaks
used to quantify and separate mixtures of fluid chem- corresponding to the different times at which com-
ical compounds. There are many different kinds of ponents of the sample mixture emerge from the
chromatography, among them gas chromatography, column.
organic liquid chromatography, and ion-exchange By running standards and mixtures of known con-
chromatography. All chromatographic methods share centration, it is possible to relate the arrival time to
the same basic principles and mode of operation. In species type and the size of the peak to concentration,
every case, a sample of the mixture to be analysed is making chromatography a valuable quantitative
applied to some stationary fixed material (the adsorb- technique. There are three main types of chromatog-
ent or stationary phase) and then a second material raphy employed in geochemical laboratories: liquid
(the eluent or mobile phase) is passed through or over chromatography, gas chromatography, and ion chro-
the stationary phase. The compounds contained in the matography. Gas chromatography is used to separate
sample are then partitioned between the stationary mixtures of gases or vaporized liquids. Ion chroma-
phase and the mobile phase. The success of the ap- tography is used to separate and analyse ions, typic-
proach depends on the fact that different materials ally but not exclusively anions, in aqueous solutions.
adhere to the adsorbent with different forces. Those Liquid chromatography is included in Table 4 but is
that adhere more strongly are moved through the mainly used as a sample-preparation procedure prior
adsorbent more slowly as the mobile phase flows to gas chromatography to split petroleum, e.g. into
 68 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)
Table 4 Chromatographic techniques commonly used in geochemical analysis

Technique Output-1 Output-2 Output-3 Sample type Advantages Disadvantages

Liquid chromatography Physical separation of Whole petroleum Excellent pre-separation Limited separation
different components of or extracted technique for GC and capability
complex liquid mixture bitumen samples GC-MS analyses; Gives
(petroleum) quantities of groups of
petroleum compounds
Gas chromatography Physical separation of Quantitative Either whole Splits gas and liquid range Co-elution of different
different components of measure of petroleum or compounds; easily compounds; requires
complex gas-phase proportions of separate parts quantified; well- sample preparation;
mixture (petroleum gas different (achieved established technology; unknown GC-peaks can
or heated volatilized compounds in using liquid good for samples with give ambiguous
liquid) gas and liquid chromatography) dominant alkanes interpretation
petroleum
Ion chromatography Physical separation of Quantitative Quantitative Water sample or Splits a range of anions in Unsuitable for bicarbonate
different charged measure of measure of solid sample water; high resolution; analysis
(aqueous) components proportions of proportions of quantitatively relatively fast and
in complex natural different anions different cations dissolved in simultaneously analyses
solutions in water in water (less water all anions; no real
(common common alternative
application) application)
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 69

saturated, aromatic, and resin groups, and will not be into one of a number of detectors such as a flame
covered further here. ionization detector, which has high sensitivity, a large
linear response range, and low noise. An example
Gas Chromatography of a flame-ionization-detector signal from a whole-
Gas chromatography (GC) – specifically gas–liquid petroleum sample injected onto a GC column is given
chromatography – involves a sample being vaporized in Figure 18.
and injected onto the head of the chromatographic One of the problems of GC analysis of geochemical
column (Table 4). The sample is transported through samples is that different compounds can have similar
the column by the flow of an inert gaseous mobile elution times, rendering identification and quantifica-
phase. The column itself contains a liquid stationary tion difficult. However, the output stream from a gas
phase that is adsorbed onto the surface of an inert chromatograph can be passed into other types of
solid (Figure 17). analytical instrument (e.g. a mass spectrometer) for
The carrier gas is chemically inert (e.g. helium). further analysis of the separated compounds over and
A sample of gas or petroleum is injected into the above simple quantification of compounds with a
column quickly as a slug to prevent peak broadening common elution time. GC is useful for analysing
and loss of resolution. The temperature of the sample organic compounds and can be used to quantify
port is somewhat higher than the boiling point of the mixtures if suitable standards have been employed.
least-volatile component of the sample. Sample sizes
Ion Chromatography
typically range from tenths of a microlitre to 20 ml.
The carrier gas enters a mixing chamber, where the Ion chromatography can be used for both cations and
sample vaporizes to form a mixture of carrier gas, anions. However, it is in the analysis of non-metal
vaporized solvent, and vaporized solutes. A propor- ions that the technique has proved most useful mainly
tion of this mixture passes onto the chromatography because there are no real alternatives for the simul-
column. Chromatography columns have an internal taneous quantitative analysis of these important
diameter of a few tenths of a millimetre and walls species in waters or synthetic solutions.
coated with a liquid but stationary phase. The opti- Ion chromatography is used to analyse aqueous
mum column temperature depends on the boiling samples containing ppm quantities of common anions
point of the sample; typically a temperature slightly (such as fluoride, chloride, nitrite, nitrate, and sul-
above the average boiling point of the sample results phate). Ion chromatography is a form of liquid chro-
in an elution time of 2–30 min. As the carrier gas matography that uses ion-exchange resins to separate
containing the chromatographically separated sample atomic or molecular ions based on their interaction
passes out of the end of the column, it is passed with the resin (Figure 19). Its greatest utility is for the

Figure 17 Essential components of a gas chromatograph. The sample injection port is heated to volatilize liquid-phase organics.
The GC column is held in an oven at a temperature above the boiling point of the compounds of interest. The inert carrier gas drives the
sample through the capillary with its stationary phase. The stationary phase retards larger molecules more efficiently than small
molecules. Smaller molecules thus pass out of the column more rapidly than large molecules.
70 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

Figure 18 Typical GC output from a black oil sample. Most of the peaks correspond to normal alkanes. The longest-chain alkane
detected is C35H72. The biomarkers, molecules of clear biological origin, form an area of low-level noise that is difficult to discern with
this technique. On the figure nc-3, nc-22, nc-35 refer to normal alkanes with 3, 22 and 35 carbon atoms. The three labelled peaks are
thus from C3H8, C22H46 and C35H72.

Figure 19 Essential components of an ion-exchange chromatograph, and a typical output trace for a low-concentration standard. The
suppressor column removes the bicarbonate and carbonate that are released from the ion-exchange chromatography column to avoid
the real sample signal being swamped. The trace has 0.02 ppm fluoride and 0.1 ppm of all the other anions. Note that the transit time
through the column increases with atomic number for the halogens and is greatest for the multivalent anions phosphate and sulphate.

analysis of anions for which there are no other rapid on the polymer surface and may replace (exchange
analytical methods. For anion chromatography the with) the bicarbonate and carbonate ions stuck to the
mobile phase is a dilute aqueous solution of sodium surface. Usually, the greater the charge on the anion
bicarbonate and sodium carbonate prepared with the more strongly it is attracted to the surface of the
pure water. polymer bead. Also, larger anions generally move
The ion-exchange column is tightly packed with more slowly through the column than smaller anions.
the stationary adsorbent. This adsorbent is usually The result is that the sample separates into bands of
composed of tiny polymer beads that have positively different kinds of ion as it travels through the column.
charged centres. These become coated with bicarbon- The detector, usually a conductivity cell, measures
ate and carbonate anions if no sample is passing the conductance of the solution passing through it.
through the column. As anions in the sample enter The conductance is proportional to the concentration
the column, they are attracted to the positive centres of ions dissolved in the solution. It is essential to
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 71

pass the sample–mobile-phase mixture through a

suppresser column – another ion-exchange column –
to remove the bicarbonate and carbonate ions and
avoid the sample signal being masked before entering
the detector. Anions can be qualitatively identified
by analysing standards and standard mixtures. The
concentrations of anions are quantified by the usual
geochemical technique of calibration (Figure 1).

Mass Spectroscopy
Principles of Mass Spectroscopy
If a moving charged molecule or atom is subjected to
a sideways electromagnetic force, it will move in a
curve. The amount of deflection in a given electro- Figure 20 Essential components of a mass spectrometer.
magnetic field depends on the velocity and mass of A source (from thermal ionization, inductively coupled plasma,
the ion and the number of charges on it. The latter a GC column, etc) feeds vaporized sample into an ionization
two factors are combined in the mass–charge ratio chamber. The vapour is ionized to produce positively charged
ions. These are drawn into the mass spectrometer via an accel-
(given the symbol m/z or sometimes m/e). Most of the
erating electrode. The ions are passed into an electromagnet,
ions passing through the mass spectrometer will have where they are deflected as a function of the mass–charge (m/z)
a charge of þ1, so the mass–charge ratio will be the ratio, their velocity, and the strength of the magnetic field. The
same as the mass of the ion. Atoms and molecules are electromagnets have their field strength varied to cause variable
ionized by removing one or more electron in an ion- deflection. The signal strength is recorded as a function of mag-
netic-field strength, and the signal output is given in terms of the
ization chamber to leave a positive ion. The beam of
relative intensity of the peak versus m/z.
ions passing through the machine is detected electric-
ally. It is important that the ions produced in the
ionization chamber have a free run through the device device can be calibrated to record current (which is a
without hitting air molecules so that equipment is measure of the number of ions) directly against m/z.
operated under a high vacuum. The vaporized sample The output from the chart recorder is usually simpli-
passes into the ionization chamber. The electrically fied into a ‘stick diagram’. This shows the relative
heated metal coil gives off electrons that are attracted current produced by ions of varying mass–charge
to the electron trap, which is a positively charged ratio (Figure 20). There are a wide range of mass-
plate. The particles in the sample (atoms or mol- spectroscopy techniques; three have been summarized
ecules) are therefore bombarded with a stream of in Table 5.
electrons, and some of the collisions are energetic
Thermal-Ionization Isotope-Ratio Mass
enough to knock one or more electrons out of the
Spectroscopy
sample particles to make positive ions. Most of
the positive ions formed will carry a charge of þ1 The technique of isotope dilution is being used
because it is much more difficult to remove further increasingly to improve precision and accuracy by
electrons from an already positive ion. These positive reducing the problems of calibration and sample-
ions are persuaded out into the rest of the machine by preparation effects. Some materials for analysis and
the ion repeller, which is another metal plate carrying some analytical methods result in a large degree of
a slight positive charge. uncertainty. Variability caused by such problems is
When an ion hits the detector, its charge is neutral- usually partly compensated for or monitored by
ized by an electron jumping from the metal to the ion. using internal standards and surrogate samples. An
That leaves a space amongst the electrons in the isotope-dilution standard is the ‘perfect’ internal
metal, and the electrons in the wire shuffle along to standard or surrogate.
fill it. A flow of electrons in the wire is detected as an An internal standard or surrogate is a compound
electric current, which can be amplified and recorded. similar to the sample of interest. An isotope-dilution
If the magnetic field is varied, the ion stream can be standard is an isotope of an element or a molecular
deflected on to the detector to produce a current that compound labelled with an isotope. A good example
is proportional to the number of ions arriving. The of this is 204Pb, a minor isotope of lead. The natural
mass of the ion being detected is related to the size of stable lead isotopes are 204 (1.4%), 206 (24.1%),
the magnetic field used to bring it to the detector. The 207 (22.1%), and 208 (52.4%). By adding a known
72 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

Table 5 Mass spectrometry techniques commonly used in geochemical analysis

Technique Output-1 Sample type Advantages Disadvantages

Thermal ionization Concentrations of Solid salt of the Relatively simple; gives isotope Only for limited range of
mass spectroscopy individual element in ratios; can be quantitative if elements; only for solid
elements and question sample diluted with known samples
isotopes, deposited on concentration of isotope; high
typically heavier a filament temperature of filament
elements causes ionization directly
Inductively coupled Concentrations of Water and High analytical resolution; Expensive equipment;
plasma mass trace and minor quantitatively small sample sizes; wide difficult technique
spectroscopy elements dissolved range of elements; rapid requiring expert
solids analysis; no matrix effects operator; solid samples
due to sample dissolution require quantitative
dissolution
Gas chromatography Concentrations of Volatile or High analytical resolution; easy Problems with ab initio
mass spectroscopy trace organic dissolved to quantify; large range of determination of
compounds organic compounds can be unknown compounds;
compounds determined from one sample expensive and tricky
technique

amount of 204Pb to a sample before testing for total source, so a group of related molecules can be traced
lead and by testing for each of the lead isotopes, it using the same m/z fraction for the different time-
is possible to determine accurately total lead and controlled GC fractions. A time plot of the same
individual lead-isotope ratios. This approach leads ionized molecular fragment can then be reconstructed
to much smaller errors than simply calibrating signal from the individual intensities of the mass spectra.
strength for individual m/z values. The data can be used to determine the identity and
quantity of an unknown chromatographic compon-
Gas Chromatography Mass Spectroscopy ent with an assuredness that is simply unavailable
One of the main problems with analysis of GC column by other techniques. This approach can be quantified
output is the co-elution (same rate of passage) of if the equipment is calibrated or if the sample is
groups of compounds. This problem has been tackled mixed with a known quantity of a standard. GC-MS
by using a mass spectrometer as the detection system analysis allows the quantification of trace organic
(rather than, for example, a flame ionization de- components including biomarkers, thermally con-
tector). Placed at the end of a chromatographic trolled optical isomers (e.g. steranes), and geological
column in a similar manner to other GC detectors, a age-dependent molecules (e.g. olearane, derived from
mass spectrometer detector is more complicated than post-Cretaceous flowering plants).
other GC detector systems (Figure 21). A capillary
Inductively Coupled Plasma Mass Spectroscopy
column must be used in the chromatograph because
the entire mass spectroscopy process must be carried Inductively coupled plasma mass spectroscopy (ICP-
out at very low pressures and in order to meet this MS) uses plasma of the same type as in ICP-OES
requirement a vacuum is maintained via constant (Figure 16), but here it is used to convert elements
pumping using a vacuum pump. to ions that are then separated by mass in a mass
The major components over and above the GC spectrometer. This allows the different elements in a
column are an ionization source, a mass separator, sample (and their natural isotopes) to be separated
and an ion detector. There are two common mass and their concentrations determined. The core of the
analysers or separators commercially available for ICP-MS system is the interface through which ions
GC-MS: the quadrapole and the ion trap. from the inductively coupled plasma source enter the
The power of this technique lies in its ability to high-vacuum chamber of the mass spectrometer.
produce mass spectra from each time-controlled GC ICP-MS combines the advantages of inductively
peak (Figure 18). Thus, for each GC peak, coeluted coupled plasma (simple and rapid sample handling)
molecules are ionized and separated into m/z frac- and mass spectrometry (high sensitivity, isotope
tions in the mass spectrometer. Complex organic mol- measurement) in a multielement technique. Detection
ecules tend to fragment in predictable ways in the ion limits can be much lower than in ICP-AES: certain
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 73

Figure 21 Essential components of a gas chromatography mass spectrometer and its output. The sample injection port is heated to
volatilize liquid-phase organics. The capillary GC column is held in an oven at a temperature above the boiling point of the compounds
of interest. The GC column separates the sample into groups of molecules according to their elution rates. The GC-separated sample
is fed into the ionization chamber, where the organic molecules are fragmented, drawn into the mass spectrometer, and separated into
different m/z fragments. (B) Each peak on a GC trace (Figure 18) has its own mass spectrum, permitting high-resolution analysis of
trace organic molecules. (C) The individual m/z fragments (e.g. 217) can be reconstructed as a plot of fragment concentration versus
time. The high resolution of this technique permits the resolution of the biomarkers shown on the GC trace in Figure 18.

elements can be detected at the ng l
1 (parts per accurate assessment of volatile loss from clay min-
trillion) level in aqueous solutions. erals. These techniques can be used to determine the
quantity of clay minerals in a rock under ideal condi-
Pyrolysis and Other Heating tions the exact types and quantities of clays can be
Techniques) determined since different clay minerals dehydrate at
different temperatures.
It is possible to characterize materials by heating
them and either by studying the resulting change in Pyrolysis
optical and physical properties or by analysing the Pyrolysis has been used to study organic material for
fluid evolved. This approach has been applied to a number of purposes. Heating organic matter is used
organic materials, minerals, rocks, and fluid trapped to study the resulting total quantity of evolved CO2,
as inclusions within mineral grains (Table 6). SO2 etc. (when done in an oxygen atmosphere), which
can be used to help determine the elemental compos-
Thermogravimetry and Evolved Water Analysis
ition of the organic matter. The elements in organic
If a small sample of rock is subjected to a controlled matter of all sorts, determined in this way, include
heating cycle and simultaneously weighed, the tem- carbon, hydrogen, sulphur, and nitrogen. The
perature at which volatiles are driven off from the resulting evolved gas phases can be split using a GC
sample can be accurately monitored. If the volatiles column (see above) or analysed using various optical
are carried to a detector using an inert gas (e.g. nitro- (e.g. infrared) techniques specific to the expected gas
gen), then it is also possible to analyse the evolved products. The output from this approach can be the
gases using an infrared water-vapour analyser to total organic carbon content (e.g. of a rock or sedi-
determine independently the exact quantity of water ment) or the elemental analysis of pre-separated
driven off at each stage in the heating cycle. The samples (Figure 22).
first technique is known as thermogravimetry; the Another approach is to heat solid samples (e.g. of
second technique is known as evolved water analy- organic-rich sediment, coal, separated kerogen, or
sis. Carbonate minerals also undergo volatile loss asphaltene exsolved from petroleum) in an oxygen-
during heating, so it is important to pair the weight free environment and analyse the resulting fluid-
loss with the identification of the mineral under- phase products during a programmed heating cycle
going volatile loss. The combined approach allows (known as rock eval pyrolysis). During heating the
 74 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)
Table 6 Pyrolysis and other thermal techniques commonly used in geochemical analysis

Technique Output-1 Output-2 Output-3 Sample type Advantages Disadvantages

Thermogravimetry– Estimate of total Small samples of Sub percentage level Difficult to resolve individual
evolved water clay mineral rocks, sediments, resolution; useful ally clay minerals in clastic rocks;
analysis content of or minerals to XRD analysis; rapid difficult to repeat; destroys
rocks sample
Pyrolysis Total organic Quantities of existing General character of Organic-bearing Rapid; approach allied Destroys sample; geologically
carbon (on petroleum, petroleum that would sediment or rock, to existing unrealistic rates of heating
decarbonated petroleum- be generated by solid asphaltene technology; (for rock-eval); results
samples) generation potential, organic matter with exolved from quantitative; can sensitive to geological history
CO2-generating further heating petroleum, reveal kinetic data of sample
potential (rock-eval) (pyrolysis-GC) bitumen about source rocks
Fluid-inclusion Salinity of water Growth temperature of Polished wafers of High level of precision; Difficult sample preparation;
microthermometry trapped in mineral rocks and easily repeated; high requires assumption that
minerals minerals spatial resolution; salinity is due to NaCl;
reveals geological requires assumption that
evolution of samples vapour-saturated fluid was
trapped
 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray) 75

Figure 22 Essential components of pyrolysis equipment, with either the volatiles sequentially emitted during a programmed heating
cycle or GC analysis of the pyrolysate. P1 represents evaporated pre-existing petroleum. P2 represents the petroleum generated from
kerogen. P3 represents the generation of oxygen-bearing species (CO2). Tmax is the peak temperature of the P2 trace. The pyrolysis-
GC trace is typically used to determine the gas (short alkanes, small elution time) versus oil (longer alkanes, longer elution time)
generation capability of immature kerogen.

free hydrocarbons contained in the sample are driven The products of pyrolysis in the absence of oxygen
off first (the P1 peak) followed by experimentally can also be passed into a GC column to study their
generated petroleum (the P2 peak) and then oxygen- composition. This is known as pyrolysis-GC and can
containing compounds (CO2; the P3 peak). Detection usefully simulate the type of petroleum that would be
in simple instruments is performed using a flame expected from the organic-rich source rock during
ionization detector (for free and laboratory-generated thermal evolution.
hydrocarbons) and a thermal conductivity detector
Fluid Inclusion Microthermometry
for carbon dioxide. Three peaks are thus usually pro-
duced and quantified during heating and analysis. Small samples of the fluid from which a mineral grew
These peaks are expressed in terms of mg g
1. P1 are commonly trapped in inclusions. When minerals
indicates how much petroleum exists in an organic- are fractured, they sometimes re-heal, trapping the
rich rock now. P2 indicates the petroleum-generating ambient fluid. Petroleum, which is immiscible with
capability of the rock and is used to calculate the the aqueous mineral growth medium, can also get
hydrogen index of the source rock. P3 is an indication trapped in these inclusions. These fluid inclusions
of the amount of oxygen in the kerogen and is used to have proved to be very valuable to geochemists since
calculate the oxygen index. The temperature at which they provide a snapshot of fluid evolution and rock
the maximum release of hydrocarbons from cracking geohistory. The fluid itself can be analysed if it is
of kerogen occurs during pyrolysis (P2 peak) is released by crushing. This approach is especially
termed Tmax and is an indication of the stage of useful for petroleum inclusions. Analysis is by GC,
maturation of the organic matter. The hydrogen GC-MS, or simply mass spectroscopy. UV spectros-
index (HI ¼ (100  P2)/total organic carbon) and the copy can also be used to analyse petroleum trapped in
oxygen index (OI ¼ (100  P3)/total organic carbon) inclusions to help reveal the broad characteristics of
correlate with the ratios of hydrogen to carbon and the petroleum.
oxygen to carbon, respectively. These parameters Aqueous inclusions can be analysed either by freez-
have been usefully employed to study the type and ing the sample and using electron micoprobe (second-
thermal evolution of sedimentary organic matter. ary X-ray) analysis, or by crushing the sample and
76 ANALYTICAL METHODS/Geochemical Analysis (Including X-ray)

collecting the fluid for conventional water analysis Titration, electrochemical techniques, and colorim-
(inductively coupled plasma techniques, ion chroma- etry are essential techniques that are used routinely
tography). Fluid inclusions are listed here since they in many geochemical studies.
are most commonly analysed by using a high powered
microscope and heating–cooling stage. Most fluid
inclusions are composed of discrete liquid and vapour
See Also
phases even though they would have been trapped as Clay Minerals. Fluid Inclusions. Minerals: Definition
a single-phase liquid, which is typically assumed to and Classification; Native Elements. Petroleum Geol-
have been saturated with vapour at the time of trap- ogy: Chemical and Physical Properties; The Petroleum
ping. During a heating cycle, the two phases hom- System. Rocks and Their Classification.
ogenize; the precise temperature of homogenization
reveals the temperature at which the mineral grew. Further Reading
The salinity of the water can be assessed by monitor-
ing the temperature at which it starts freezing, since Emery D and Robinson AC (1993) Inorganic Geochemistry:
this temperature can be related to salt content (assum- Applications to Petroleum Geology. Oxford: Blackwells.
ing that the water is dominated by dissolved NaCl). Farmer VC (1974) The Infrared Spectra of Minerals.
London: Mineralogical Society.
A combination of thorough petrography and thermo-
Faure G (1986) Principle of Isotope Geology. New York:
metric studies of aqueous inclusions can help to reveal John Wiley and Sons.
details of the thermal and mineral-growth history as Goldstein RH and Reynolds TJ (1994) Systems of Fluid
well as the fluid evolution history. Inclusions. Tulsa: Society of Sedimentary Geology.
Hagemann HW and Hollerbach A (1986) The fluorescence
behaviour of crude oils with respect to their thermal
Related Geochemical Techniques maturation and degradation. Organic Geochemistry 10:
The techniques listed and briefly discussed here are 473–480.
only some of the vast panoply of techniques that have Jarvis I and Jarvis KE (1992) Plasma spectrometry in the
been employed during geochemical studies over the last Earth sciences: techniques, applications and future trends.
50 years or so. Some have now fallen out of favour. Chemical Geology 95: 1–33.
Jenkins R (1999) X-ray Fluorescence Spectrometry. New
For example, a technique called neutron activation
York: Wiley Interscience.
analysis was used for a long time to measure trace Lico MS, Kharaka YK, Carothers WM, and Wright VA
elements in solids. It has fallen out of favour mainly (1982) Methods for the Collection and Analysis of Geo-
owing to developments in ICP-OES and ICP-MS. pressured Geothermal and Oil Field Waters. Water-
The vast range of light and electron optical tech- Supply Paper 2194. United States Geological Survey.
niques are routinely used in conjunction with a wide Moore DM and Reynolds RC (1997) X-ray Diffraction and
range of solid-state and even organic geochemical the Identification and Analysis of Clay Minerals. Oxford:
studies. Scanning electron microscopy has recently Oxford University Press.
been extensively developed and can now give fabric, Rollinson HR (1993) Using Geochemical Data: Evalu-
mineralogy, mineral chemistry, and high-resolution ation, Presentation, Interpretation. New York: Longman
crystallographic information. Transmission electron Scientific and Technical.
Tissot B and Welte D (1984) Petroleum Formation and
microscopy can provide ultra-high spatial resolu-
Occurrence. Berlin: Springer Verlag.
tion (of the order of tens of nm) geochemical data Tucker ME (1988) Techniques in Sedimentology. Oxford:
(using secondary X-rays) as well as fabric and Blackwell Scientific.
crystallographic data. Weiss J (2000) Ion Chromatography. New York: John
A wide range of wet geochemical techniques have Wiley and Sons.
been employed routinely in studies of natural waters Zussman J (1967) Physical Methods in Determinative
from all near-surface and surface environments. Mineralogy. London: Academic Press.