Groundwater moving through the geosphere appears to be a simple enough process, yet a groundwater realm of cryptic underground rivers and channels has remained in our culture since early historic times, sustained by water diviners and rural myths. Surface waters seem easier to understand, their intricacies more apparent because we can see them flow and follow them to their source. But to quench your thirst, they must first be flocculated, sedimented, filtered, limed, chlorinated and often chilled, provided industrial effluents have not already damaged the supply. For groundwaters, the geosphere provides these treatments naturally. Groundwater represents more than 50 times the freshwater resource that surface waters do, yet in North America groundwater is used for less than half of freshwater needs; in Central Europe, groundwater is the dominant source for drinking water.
The dawn of hydrogeology as a science began with Darcys early experimenting with the plumbing for the fountains of Dijon. Today, the over-exploitation and contamination of this resource has moved groundwater research to the forefront of the geosciences. Nonetheless, like the diviners of historic times, hydrogeologists still wrestle with the questions of groundwater provenance, its renewability and the subsurface processes affecting its quality. These questions become increasingly relevant as we continue to test the limits of groundwater resource sustainability.
Environmental isotopes now routinely contribute to such investigations,
complementing geochemistry and physical hydrogeology. Meteoric processes,
for instance, modify the stable isotopic composition of water, and so the
recharge waters in a particular environment will have a characteristic
isotopic signature. This signature then serves as a natural tracer for
the provenance of groundwater. On the other hand, radioisotopes decay,
providing us with a measure of circulation time, and thus groundwater renewability.
Environmental isotopes provide, however, much more than indications of
groundwater provenance and age. Looking at isotopes in water, solutes and
solids tells us about groundwater quality, geochemical evolution, recharge
processes, rock-water interaction, the origin of salinity and contaminant
processes. Lets start with the basics.
The nuclear structure of a nuclide (an isotope-specific atom) is classically defined by its number of protons (Z) which defines the element, and the number of neutrons (N) which defines the isotope of that element. For a given nuclide, the sum of protons and neutrons gives the atomic weight (A), expressed by the notation . For example, most oxygen has 8 protons and 8 neutrons, giving a nuclide with 16 atomic mass units () while about 0.2% of oxygen has 10 neutrons (). In reality, the mass of a nuclide is slightly less than the combined mass of its neutrons and protons. The "missing" mass is expressed as the nuclear binding energy (according to Einsteins mass-energy relationship E = mc2), which represents the amount of energy required to break the nucleus into its constituent nucleons. Conventional notation for a nuclide uses only the elemental symbol and atomic weight (e.g. 18O or 34S).
Whereas the number of neutrons in the nucleus can vary, the range is
limited by the degree of instability created by having too many or too
few neutrons. Unstable isotopes or radioactive nuclides have a certain
probability of decay. Stable isotopes, on the other hand, do not spontaneously
disintegrate by any known mode of decay. To date, some 270 stable nuclides
and over 1700 radionuclides have been identified. For the light elements
(Z up to 20) the greatest stability occurs with a Z:N ratio close to 1,
and increases towards 1.5 for heavy elements. In a chart arranged according
to Z and N (Fig. 1-1), the stable isotopes of the elements form a stable
valley from hydrogen to uranium. Departures from this stable valley produce
radionuclides of decreasing stability (shorter half-lifes). Oxygen, for
example, has eleven isotopes (12O to 22O) although
only the median isotopes, 16O, 17O and 18O,
are stable (Fig 1-2). The others are radioactive with half-lives varying
from 122 seconds to less than a femtosecond (10-15 s).
Fig. 1-1 Plot of Z vs. N for nuclides up to tin (Z=50) showing the "stable" valley of the nuclides. The Z : N ratio is 1 for the light nuclides and increases towards 1.5 for the heavier nuclides. Increases or decreases in N for given element produces increasingly unstable isotopes (decreasing T½).
The distribution of stable isotopes reflects the structure of the nucleus. Like electron orbits, the most stable nuclei have filled neutron and/or proton shells. Those nuclides with a "magic number" (2, 8, 20, 28, 50, 82 and 126) of neutrons and/or protons are the most common (e.g. = 99.99986% of all helium; = 99.76% of all oxygen; or = 96.9% of all calcium) whereas others have lower abundances (e.g. = 19.9% of all boron). As the nuclear binding energy occurs between nucleon pairs (protons or neutrons), stable nuclides with even numbers for N and Z, dominate. Thus, 161 of the known stable nuclides have an even N and Z while only 4 have an odd values for N and Z. There are 105 with either an odd N or Z.
Although all elements present in hydrogeological systems have a number of isotopes, only a few are of practical importance to us. The environmental isotopes are the naturally occurring isotopes of elements found in abundance in our environment: H, C, N, O and S. These are principal elements of hydrological, geological and biological systems. The stable isotopes of these element serve as tracers of water, carbon, nutrient and solute cycling. They are also light elements. As a consequence, the relative mass differences between their isotopes are large, imparting measurable fractionations during physical and chemical reactions. For example, 2H has 100% more mass than its sister isotope 1H, whereas the two stable isotopes of bromine (81Br and 79Br) have a mass difference of only 2.5%. Radioactive environmental isotopes are also important in hydrogeology. From their decay we have a measure of time and so environmental radionuclides such as 14C and 3H can be used to estimate the age or circulation of groundwater.
The family of environmental isotopes is growing as new methods allow the routine analysis of additional isotopes. Accelerator mass spectrometry (AMS) analysis has brought 36Cl into mainstream isotope hydrogeology. Refinements in solid source mass spectrometry and inductively coupled plasma mass spectrometry (ICP-MS) allows high precision measurement of the isotopes of trace elements such as U, Th, Li and B. The major stable environmental isotopes used in hydrogeology are presented in Table 1-1.
Environmental isotopes are now used to trace not only groundwater provenance, but also recharge processes, subsurface processes, geochemical reactions and reaction rates. Their importance in studies of biogeochemical cycles and soil-water-atmosphere processes is increasingly being recognized, and new applications in contaminant hydrogeology are being made.
Table 1-1 The stable environmental isotopes
|Isotope||Ratio||% natural abundance||Reference (abundance ratio)||Commonly measured phases|
|2H||2H/1H||0.015||VSMOW (1.5575 · 104)||H2O, CH2O, CH4, H2, OH minerals|
|3He||3He/4He||0.000138||Atmospheric He (1.3 · 106)||He in water or gas, crustal fluids. basalt|
|6Li||6Li/7Li||7.5||L-SVEC (8.32 · 102)||Saline waters, rocks|
|11B||11B/10B||80.1||NBS 951 (4.04362)||Saline waters, clays, borate, rocks|
|13C||13C/12C||1.11||VPDB (1.1237 · 102)||CO2, carbonate, DIC, CH4, organics|
|15N||15N/14N||0.366||AIR N2 (3.677·103)||N2, NH4+, NO3, N-organics|
|18O||18O/16O||0.204||VSMOW (2.0052 · 103) VPDB (2.0672 · 103)||H2O, CH2O, CO2, sulphates, NO3, carbonates, silicates, OH minerals|
|34S||34S/32S||4.21||CDT (4.5005 · 102)||Sulphates, sulphides, H2S, S-organics|
|37Cl||37Cl/35Cl||24.23||SMOC (0.324)||Saline waters, rocks, evaporites, solvents|
|81Br||81Br/79Br||49.31||SMOB||Developmental for saline waters|
|87Sr||87Sr/86Sr||87Sr = 7.0 86Sr = 9.86||Absolute ratio measured||Water, carbonates, sulphates, feldspar|
The variations in numbers of neutrons in an element provides for the
different masses (atomic weights) of the element and the molecules of which
they may be a part. For example, heavy water, 2H216O,
has a mass of 20 compared to normal water, 1H216O,
which has a mass of 18. Molecules with differences in mass have different
reaction rates. This leads to the isotope partitioning or fractionation
described by Urey (1947).
Stable environmental isotopes are measured as the ratio of the two most
abundant isotopes of a given element. For oxygen it is the ratio of 18O,
with a terrestrial abundance of 0.204%, to common 16O which
represents 99.796 of terrestrial oxygen. Thus the 18O/16O
ratio is about 0.00204. Fractionation processes will of course modify this
ratio slightly for any given compound containing oxygen, but these variations
are seen only at the fifth or sixth decimal place.
Measuring an absolute isotope ratio or abundance is not easily done
and requires some rather sophisticated mass spectrometric equipment. Further,
measuring this ratio on a routine basis would lead to tremendous problems
in comparing data sets from different laboratories. However, we are mainly
interested in comparing the variations in stable isotope concentrations
rather than actual abundance, and so a simpler approach is used. Rather
than measuring a true ratio, an apparent ratio can easily be measured by
gas source mass spectrometry. The apparent ratio differs from the true
ratio due to operational variations (machine error, or m) and will not
be constant between machines or laboratories or even different days for
the same machine. However, by measuring a known reference on the same machine
at the same time, we can compare our sample to the reference. Isotopic
concentrations are then expressed as the difference between the measured
ratios of the sample and reference over the measured ratio of the reference.
Mathematically, the error (m) between the apparent and true ratios is cancelled.
This is expressed using the delta (d) notation:
The range of oxidation states of carbon makes it a fundamental element of the biosphere and hydrosphere. Carbon-13 traces carbon sources and reactions for a multitude of inter-reacting organic and inorganic species. The paleotemperature scale developed in the early 1950s using the 18O/16O ratio in marine carbonates adopted PDB as the international reference material (Urey et al., 1951). PDB was the internal calcite structure (rostrum) from a fossil Belemnitella americana from the Cretaceous Pee Dee Formation in South Carolina. In 1957, Craig formally introduced PDB as the standard for both 13C and 18O in carbonate minerals. It has subsequently been adopted as the 13C standard for all carbon compounds, including CO2, dissolved inorganic carbon species (DIC), dissolved organic carbon (DOC), cellulose and other fixed-C solids (CH2O), organic liquids, methane and other hydrocarbons. However, VSMOW is the standard for measurements of 2H or 18O in organic molecules (e.g. CH4, CH2O etc.).
Before the limited PDB supply was exhausted, Friedman et al. (1982)
used it to calibrate a crushed slab of white marble of unknown origin,
designated as NBS-19:
d13CNBS-19 = +1.95
The measurement of isotopes in carbonate minerals is done on CO2
gas that is normally produced by acidification, a method developed by McCrea
(1950). Carbon dioxide is produced from carbonate minerals by reaction
with 100% phosphoric acid (H3PO4) at 25°C (Urey
et al., 1951). The conversion of calcite to CO2 follows the
|CaCO3 - Calcite||25||10.20|
|CaCO3 - Aragonite||25||10.29|
|CaMg(CO3)2 - Dolomite||25||11.03|
|CaMg(CO3)2 - Dolomite||25||11.712|
|SrCO3 - Strontianite||25||10.43|
|BaCO3 - Witherite||25||10.91|
|FeCO3 - Siderite||25||10.11753|
|FeCO3 - Siderite||50||10.10753|
|MgCO3 - Magnesite||50||11.53|
2 Rosenbaum and Sheppard, 1986 .
3 Carothers et al., 1988.
Both VPDB and VSMOW are recognized international standards for 18O.
While waters are exclusively referenced to VSMOW, carbonates can refer
to either. VPDB was originally introduced for paleoclimatic studies, where
the 18O content of carbonate was used as a paleotemperature
scale. However, the use of carbonate isotopes has gone far beyond this
field, and in water-carbonate studies it is common to express d18O
data for carbonate against the VSMOW scale. Conversion is also necessary
when deriving information about the d18O
content of the water in which a carbonate has formed. The conversion chart
in Fig. 1-3 or the following equations can be used (Coplen et al., 1983):
d18OVPDB = 0.97002
· d18OVSMOW 29.98
Fig. 1-3 Conversion chart for 18O between VSMOW and VPDB, with fractionation factors for 25°C. The bold line equates values on the VPDB scale to values on the VSMOW scale according to the two reciprocal equations: d 18OVSMOW=1.03091 · d 18OVPDB + 30.91 , and d 18OVPDB=0.97002 · d 18OVSMOW 29.98 .
In 1947, Alfred Nier developed the first dual-inlet, double-collector gas-source mass spectrometer. The double collector allowed the simultaneous measurement of two isotopes and the dual inlet allowed ratio measurement on both a sample and a standard by alternating between inlets. Gas source mass spectrometry has since become the measurement technique of preference for isotope ratios of most of the light elements (e.g. H, C, N, O and S) because of its relative simplicity and because the use of international standards allows comparison of data bases from different laboratories. A host of preparation methods have been developed and improved to convert different sample compounds to an appropriate gas including CO2, SO2, H2 and N2.
A heated tungsten-coated iridium (thoria) filament inside the source block cavity ionizes a laminar stream of gas entering the ultra-high vacuum source (Fig. 1-4). The gas molecules are stripped of one electron, producing positive ions (e.g. CO2+) which are then accelerated through a voltage gradient and focused into the flight tube upon exiting the source. The ionization efficiency varies between 0.01 and 0.1% for different instruments. The ion beam bends as it passes through the field of a magnet installed over the flight tube. Here, the beam separates into a spectrum of masses according to the isotopes present. Each mass beam continues to the ion detectors where preset faraday cup collectors measure each ion current. By collecting two or three ion beams simultaneously, the ion currents can be expressed as mass ratios. For example, CO2 would contribute three principal peaks at mass 44 (12C16O2), mass 45 (13C16O2 or 12C17O16O) and mass 46 (12C16O18O). A dual-inlet system allows the mass spectrometer to alternately measure ratios in the sample and a working or laboratory standard. Thus, the extreme fractionation imparted during ionization in the source is resolved. The early mass spectrometers suffered from drifting electronics, which precluded accurate abundance measurements. These instabilities have since been overcome with solid-state and fibre optic signal transfer systems.
Fig. 1-4 Schematic of a gas source isotope ratio mass spectrometer (IRMS), showing both continuous flow and dual inlets. The continuous flow inlet here is shown with a sample combustion and gas chromatograph configuration. Capillary tubes ensure laminar, non-fractionating gas flow. Example shows mass range of CO2 gas, and includes the short radius flight tube for H2 found on many designs. Other mass ranges (for SO2 and N2) are attained by either additional fixed-position faraday collectors or by adjusting the beam. For manufacturers details, see <http://beluga.uvm.edu/geowww/isogeochem.htm>.
Quadrupole mass spectrometers employ a method of isotope separation and measurement that offers economies in size and cost, but lack the precision of a Nier-type instrument. However, the portability of quadrupole mass spectrometry allowed the Viking mission to perform isotope analyses of Martian soil and atmosphere.
The direction of gas source mass spectrometry is now towards continuous flow systems where the sample and standard gases are carried into the mass spectrometer source in a stream of helium gas (Fig. 1-4). In this way, pulse injections of sample gas can be analysed, reducing volume constraints and sample size. Measurement in the nano-mole (109 moles of gas) size range is now possible, opening a host of research possibilities. These systems also greatly increase sample through-put when configured with automated sample preparation systems. In particular, an elemental analyzer on the front end of a continuous flow mass spectrometer allows the analysis of solid, liquid, and mixed matrix samples which can be combusted and the gases separated on a chromatographic column. Helium carries the sample through the column and into the mass spectrometer. Continuous flow mass spectrometers configured with a laser ablation sampling system provide researchers with a tool for micro-analysis of sulphides and carbonates, with a spatial resolution in the order of 10 to 50 mm.
The radioisotopes routinely employed in hydrogeology include tritium and carbon-14. Measurement of chlorine-36 is becoming increasingly available to non-specialists, while the use of other radioisotopes like argon-39 and the krypton (85Kr and 81Kr) is restricted to a limited number of research laboratories due to complications in sampling, analysis and interpretation. For the noble gases, huge volumes of water must be vacuum extracted in the field for analysis in low-level counters. Details on these radionuclides are given in Table 1-4. Sampling methods are covered in detail in Chapter 10.
Table 1-4 The environmental radioisotopes
|Principal Sources||Commonly measured phases|
|3H||12.43||b||Cosmogenic, weapons testing||H2O, CH2O|
|14C||5730||b||Cosmogenic, weapons testing, nuclear reactors||DIC, DOC, CO2 CaCO3, CH2O|
|36Cl||301,000||b||Cosmogenic and subsurface||Cl, surface Cl-salts|
|39Ar||269||b||Cosmogenic and subsurface||Ar|
|85Kr||10.72||b||Nuclear fuel processing||Kr|
|81Kr||210,000||ec||Cosmogenic and subsurface||Kr|
|129I||1.6 · 107 yr||b||Cosmogenic, subsurface, nuclear reactors||I and I in organics|
|222Rn||3.8 days||a||Daughter of 226Ra in 238U decay series||Rn gas|
|226Ra||1600||a||Daughter of 230Th in 238U decay series||Ra2+, carbonate, clays|
|230Th||75,400||a||Daughter of 234U in 238U decay series||Carbonate, organics|
|234U||246,000||a||Daughter of 234Pa in 238U decay series||UO22+, carbonate, organics|
|238U||4.47·109||a||Primordial||UO22+, carbonate, organics|
a - alpha emission.
ec - electron capture.
Tritium, 3H, is a short-lived isotope of hydrogen with a half-life of 12.43 years. It attracted considerable interest during the era of thermonuclear bomb testing. Dr. R. Brown with Atomic Energy of Canada Limited (AECL) was the first to begin monitoring 3H fallout from atmospheric weapons tests. His data for Ottawa precipitation begins in 1952 and documents the dramatic increases in atmospheric 3H produced during the ensuing two decades of hydrogen bomb testing.
Small but measurable amounts of tritium are also produced naturally in the stratosphere by cosmic radiation on 14N. Both natural and anthropogenic tritium enter the hydrological cycle via precipitation. Its presence in groundwater provides evidence for active recharge. As it is part of the water molecule, it is the only direct water dating method available.
Tritium has also gained importance in the medical field to tag compounds in biological reactions, although the concentrations used here exceed environmental concentrations by several orders of magnitude. Environmental concerns limit the use of artificial 3H as a tracer in hydrological studies.
Tritium concentrations are expressed as absolute concentrations, using tritium units (TU) and so no reference standard is required. One TU corresponds to one 3H atom per 1018 atoms of hydrogen. For 1 litre of water, its radioactivity is equivalent to 0.12 Bq (1 Becquerel = 1 disintegration/second), or 3.2 pCi/l (1 pCi is 1012 Curies and a Curie is the radioactivity of 1 gram of 226Ra; 1 Curie = 3.7·1010 Bq). Groundwaters today seldom have more than 50 TU and are typically in the <1 to 10 TU range.
Tritium is measured by counting b decay events in a liquid scintillation counter (LSC). A 10 mL sample aliquot is mixed with the scintillation compound that releases a photon when struck by a b particle. Photomultiplier tubes in the counter convert the photons to electrical pulses that are counted over a several-hour period. Results are calculated by comparing the count to those of calibrated standards and blanks. Increased precision is gained through concentration by electrolytic enrichment of 3H in the water before counting, or by conversion to propane (C3H8) for gas proportional counting. Direct liquid-scintillation counting carries a precision of ±7 TU, whereas with enrichment and LSC this is improved to better than ±0.8 TU. With propane synthesis, a precision of ±0.1 TU can be obtained.
Although it was shown early this century that stable isotopes of oxygen and hydrogen existed, it was not until much later that their abundance and variation in the hydrosphere were investigated. Friedman, in 1953, first noted that in precipitation a change in concentration of H218O was accompanied by a change in 2HHO. In 1961, Craig published his landmark finding that these two isotopes are partitioned by meteorological processes in a rather predictable fashion. Subsequent work has shown how isotopes are partitioned through other systems, such as 13C in the carbon cycle.
How are environmental isotopes partitioned? Thermodynamic fractionation is a fundamental process. Here we will explore the details of isotope fractionation, using 18O and 2H in the hydrological cycle as an example. Fractionation of 13C, 34S, 15N and other isotopes will be discussed in other chapters.
Isotope fractionation occurs in any thermodynamic reaction due to differences
in the rates of reaction for different molecular species. The result is
a disproportionate concentration of one isotope over the other on one side
of the reaction. It is expressed by the fractionation factor a
which is the ratio of the isotope ratios for the reactant and product:
1. What is the relative enrichment or depletion of VSMOW, in d
notation, relative to the average terrestrial abundance of 18O
given in Table 1-1? What about VPDB?
= 17.1 vs. terrestrial abundance
Calculating RVPDB: determine VPDB on the VSMOW scale
d18O = 1.03091 · d18OVPDB + 30.91 = 30.91 VSMOW (for 0 VPDB)
= 13.3 enriched over the terrestrial abundance.
Accordingly, 103lna18Ocalcite-water = 30.4
From the table at the front of the book, 103lna18Ocalcitewater = 30.4 at a temperature of about 15.5°C. Considering that the bottom temperatures in Cretaceous seas are not well known, this seems pretty close to isotopic equilibrium.
SLAP = 55.5 VSMOW, therefore:
10Be ® 10B
H2Ol « H2Ov 103lna2Hl-v = 75.6 @ 25°C; 22.3 @ 100°C (evaporation)
18O: H2O - CO2(g)
CaCO3 + H2O + CO2 [CO32, HCO3, H2CO3, CO2] Ca2+ + HCO3
103lna18OCaCO3H2O = 28.4 @ 25°C; 17.1 @ 100°C (rapid precipitation of calcite)
13C: CO2(g) CaCO3
CO2(g) + H2O + CaCO3 « [Ca2+, CO32, HCO3, H2CO3, CO2(aq), CO2(g)] « Ca2+ + 2HCO3
103lna18OCaCO3CO2 = 10.4 @ 25°C; 3.4 @ 100°C (rapid degasing of CO2)
34S: reduction of sulphate to sulphide
CH4 + SO42 ® [sulphite + other intermediary S species] ® HCO3 + HS + H2O
103lna18OSO4HS = 72.7 @ 25°C; 49.9 @ 100°C (rapid degasing of CO2)
Biologically mediated reaction, and equilibrium fractionation is never attained at low T. Biological fractionation generally varies between 20 and 30.
18O: H2Oice - H2Ovapour
H2Oi « H2Ov 103lna18Oi-v = 14.7 @ 0°C (sublimation a surface reaction with little to no fractionation)
a2HH2 = 1.2055 - greatest fractionation
a3He/4He = 1.1371
a13CCO2 = 1.0044
|2H||18O or 13C||2H||18O or 13C|
|Isotope separation dX - dY||45||18||228||122|
|Deviation dX - dref. (dref. = 0)||32||12||45||18|
8. Isotope fractionation effects are expressed as a-values
and isotope abundances are given as permil differences from a reference
(d-value). Establish the relationship between
a and d in the form
of a simple equation.
d2HH2 = 754 VSMOW