GEOS 4430/5310 Lecture Notes:Groundwater Chemistry
Dr. T. Brikowski
Fall 2012
Vers. 1.33, Printed: December 3, 2012
Introduction
1
What use is Groundwater Geochemistry?
Example applications of low-temperature aqueous
geochemistry:
Simpson-Arbuckle Aquifer a potential “new” source of water
for Oklahoma city. See USGS Factsheet1
• how good is this water for drinking
• how much can be taken out safely (i.e. where does water
come from)
• significance of nearby saline and sulfurous wells? (can this
water be avoided?)
Southern Nevada Region Groundwater System a contaminant1http://www.owrb.ok.gov/studies/groundwater/arbuckle_simpson/pdf/
ArbuckleSimpsonFactSheet2008.pdf2
risk (Nevada Test Site)? A source of water for thirsty Las
Vegas2??
• can shallow contaminated water reach the currently-clean
deep aquifers?
CO2flooding A common secondary recovery method in
petroleum reservoirs. Why CO2, will it accidentally reduce
porosity? Examples include DOE-20093, Shiraki and Dunn
[2000]
2http://www.latimes.com/news/nationworld/nation/la-na-radiation-nevada13-2009nov13,0,3038881.story?track=rss
3http://www.netl.doe.gov/technologies/oil-gas/publications/EP/small_CO2_eor_primer.pdf
3
Basic Principles: Measurement
• concentration generally specified as milligrams mgL , this is
essentially equivalent to ppm (except in saline waters, where
TDS>∼ 7000 mg
L )
• thermodynamic measures [Sec. 9.2, Fetter, 2001]:
– molality: one mole of solute per 1000 gm of solvent is a 1
molal solution
– molarity: one mole of solute per liter of solvent is a 1
molar solution
• comparative chemical behavior: millequivalents:
– concentration of ionic species in mgL , multiplied by ionic
charge, and divided by molecular (formula) weight4
– a quantitative measure for comparing the chemical
behavior of dissolved species
5
Reactions
• Reversible reactions reach equilibrium easily. Typically
equilibrium is assumed in hydrochemistry, therefore
reversibility is implied
• dissociation: most common type, separation of molecules
into individual ions (e.g. NaCl)
• solvent (water) can directly participate in reaction (e.g.
carbonation reactions)
• oxidation-reduction: exchange of electrons between ions, e.g.
electrons appear in the reaction equation, and one or more
cations change atomic charge6
Equilibrium Constant (Law of Mass Action)
• an equilibrium constant can be expressed for an arbitrary
reaction as
cC + dD ⇀↽ xX + yY
K =[X]x[Y]y
[C]c[D]d(1)
where [X] is the molal concentration of X for an ideal
solution, and is the chemical activity a of X for non-ideal
solutions
• a = γm, where m is the molal concentration
• activity coefficient γ can be calculated using the Debye-
Huckel equation [Eqn. 10-17, Fetter, 2001] if the ionic7
strength I of the solution is known [Eqn. 10-16, for TDS<∼
5000 mgL Fetter, 2001]
• K values are tabulated at standard T & P (STP), can be
determined under other conditions using the expression
K = exp(−∆Go
r
RT
)where ∆Go
r is the Gibbs Free-Energy change of the reaction
at STP
• solubility product:
– equivalent to the equilibrium constant for dissolution of
a mildly-soluble salt, e.g. NaCl ⇀↽ Na+ + Cl− (at8
equilibrium):
Keq =[Na][Cl][NaCl]
= [Na][Cl] ≡ Ksp
– because [NaCl]≡ 1, in reality the solubility product is
equivalent to the equilibrium constant at equilibrium
• ion activity product
– Kiap is the numerator of the equilibrium constant (1)
– when Kiap > Keq, the salt is supersaturated and will
precipitate (ignoring kinetic effects)
• common-ion effect9
– is simply that when multiple sources for an ion exist, less
of at least one of those sources will dissolve than if it was
the only source
– for this reason, quantitative calculations involving
groundwater chemistry must consider all dissolved and
solid species
10
pH
• pH is the negative logarithm of [H+] ion concentration (really
H3O)
• a neutral solution has equivalent [H+] and [OH−], i.e. the
concentrations are controlled by water dissociation (Fig. 1)
• since equilibrium constants vary considerably with T, so does
the neutral pH (at 0◦C neutral pH is 7.5, at 25◦C it is 7.0,
Fig. 2)
• should be measured when water sample is collected, since it
can change after a few minutes of atmospheric contact
• acids are proton donors to the aqueous solution. Strong11
acids tend to completely dissociate, weak acids generally
dissociate partially. E.g. dissociation of carbonic acid [eqn.
9.21A-B, Fetter, 2001]
H2CO3 ⇀↽ H+ + HCO−3 , K = 10−6.4
HCO−3 ⇀↽ H+ + CO2−3 , K = 10−10.3
• pH is important because it controls/indicates the distribution
of many species. E.g. the three-component system C-H-O
contains seven species H+, OH−, H2CO3, HCO−3 , CO2−3 ,
CO2, H2O. Given the concentration of the components and
the pH, the concentrations of the species can be determined.
12
Water Autoionization
Figure 1: Autoionization reaction of water.
13
pH vs. T
Figure 2: Variation of pH vs. temperature.
14
pH of Common Solutions
Figure 3: pH of common solutions (after Purves et al., Life:
The Science of Biology, 4th Edition).15
Alkalinity
• ability of solution to neutralize acid (i.e. to react with H+)
• generally reported as meq of bicarbonate (HCO−3 ) +
carbonate (CO2−3 )
• CO2−3 stable only in highly basic water, so alkalinity is
essentially [HCO−3 ] [Table 9.5, Fetter, 2001]
• in common usage also a measure of the tendency to form
carbonate scale
16
Oxidation potential (Eh)
• in effect Eh is the negative log10 of [e−]
• field-measured ORP is essentially (Eh − 200mV ) (e.g. see
YSI manual4)
• oxidation refers to the removal of electrons from an
atom, increasing its oxidation number (the net charge of
a compound), reduction is the addition of an electron
• redox pairs generally control (buffer) conditions, e.g. Fe+2–
Fe+3 in near-surface oxygenated waters (Fig. 5)4http://www.ysi.com/media/pdfs/T608-Measuring-ORP-on-YSI-6-Series-Sondes-Tips-Cautions-and-Limitations.
pdf17
• in nature such redox reactions are invariably mediated by
bacteria, which use the reactions as an energy source
• difficult to obtain accurate value in the field, since any
contact with O2 dramatically changes Eh
• often analyzed using Eh-Ph diagram, for determining stability
of redox pairs (e.g. iron)
O2 + 4Fe+2︸ ︷︷ ︸Reduced
+ 4H+ ⇀↽ 4Fe+3︸ ︷︷ ︸Oxidized
+ 2H2O (2)
O02 + 4FeII + 4H+ ⇀↽ 4FeIII + 2H2O−II
• every redox reaction can be expressed as a pair of half-18
reactions, e.g. (2) can be expressed as:
O2 + 4H+ + 4e−⇀↽ 2H2O
4Fe+2 ⇀↽ 4Fe+3 + 4e−
• important in many remediation schemes (e.g. air sparging
can clog local aquifer with iron oxide precipitates)
• presence of organic compounds leads to reducing conditions
by consumption of oxygen e.g. carbohydrate breakdown
[section 12.5, Domenico and Schwartz, 1998]
O2(g) + CH2O ⇀↽ CO2(g) + H2O (3)
19
Redox in Batteries
Figure 4: Redox in a crude battery. Shown is an early battery called the “Daniell Cell” (for a true battery the two
beakers are connected by a salt bridge, allowing ions to flow between them, and avoiding explosive H2 gas formation). Zinc is
oxidized at the anode, donating an electron e− as it dissolves. Cu2+aq accepts electrons (is reduced), plating onto the cathode.
The total reaction is Zn(s) + Cu2+(aq)
→ Zn2+(aq)
+ Cu(s).
20
Redox Ladder
Figure 5: Redox pairs vs. Eh, most oxidizing environment on
the top, electron donor on the right, acceptor on the left. After
NIEH Meeting Report5.21
Eh-pH Diagram
Figure 6: Eh-pH diagram, showing effect on the solubility of
metal ions (Fe), after [Fig. 9.4, Fetter, 2001].22
A Civil Action: Woburn PCE
• a famous lethal groundwater chemical plume court case,
memorialized in the movie, jury blamed the most innocent
party
• see Woburn Spreadsheet Model Homework6
• see also animation of best model results7
6http://www.utdallas.edu/~brikowi/Teaching/Hydrogeology/Homework/woburn_spreadsheet_model.pdf
7http://serc.carleton.edu/files/woburn/resources/tce_animation.avi23
Dissolved oxygen (DO)
• critical for redox reactions and biological processes
• in last decade became standard to measure DO, wasn’t done
much before that
• easily measured with a specific probe, useful indicator of Eh
• can remain in groundwater far from recharge sites
24
Major Ion Chemistry
25
Water Analyses
• collection methods important, see USGS manual [USGS,
1998]
• generally reported in concentrations of actual ions, some (like
SiO2, nitrate NO3) are lumped together, and/or reported as
oxides
• Analytic methods standardized for EPA and environmental
applications in general [USGS, 1979, WEF, 1998]
• check error in analysis by performing charge balance (sum
of cations and anions expressed as milli-equivalents). This
sometimes fails, e.g. for strongly colored fluids which may
contain organic complexes)26
Graphical Analysis
• Stiff Diagram Fetter [Fig. 9.10, 2001] or Parkhurst et al.
[Fig. 11, 1996], Fig. 7
– plot most common milli-equivalent cations on one side of
a line, most common anions on the other
– intended to give distinctive geometric pattern, allowing
mapping of groundwater bodies/facies
– Typically plot Cations on left: Na+K, Ca, Mg, Fe; right
side anions: Cl, HCO−3 , SO4, CO2−3
– try free software from TWDB8
• Piper diagram8http://www.twdb.state.tx.us/publications/reports/GroundWaterReports/
Open-File/Open-File_01-001.htm27
– plot natural groupings on two trilinear diagrams (one for
cations, one for anions), the combination of these two
plots is made by projecting these onto the quadrilateral
diagram above (Fig. 8)
– classification of the water chemistry is based on the sum
of cation and anion classifications Fetter [Fig. 9.9, 2001]
– evolution of waters along flow path is often revealed by
these diagrams [e.g. Floridan aquifer, p 377-9, Fetter,
2001], or water sources (e.g. Floridan aquifer9)
– try free GW-Chart software10 from USGS
• activity-activity plots
– usually used in modeling studies, or determination of9http://sofia.usgs.gov/publications/wri/02-4050/distsources.html
10http://water.usgs.gov/nrp/gwsoftware/GW_Chart/GW_Chart.html28
mineral reaction effects
– plot chemical activity (often ratios) of two ions in the fluid
and superimpose mineral stability fields [Fig. 2 Parkhurst,
1995]
29
Stiff Diagram Example
Figure 7: Stiff diagram example, after [Fig. 9.10, Fetter, 2001].
30
Piper Diagram Example
Figure 8: Piper diagram example, after [Fig. 9.8, Fetter, 2001].
31
Carbonate Equilibrium
• most common reaction series in groundwater (any water in
contact with atmosphere equilibrates with CO2: rainwater
reaching equilibrium with atmospheric CO2 acquires acidic
pH ∼ 5.7; sufficient to dissolve limestone/dolomite)
• Important reactions
– solution of CO2
H2O + CO2 ⇀↽ H2CO3 (4)
KCO2 =aH2CO3
PCO2
where PCO2 is the partial pressure of CO2 (equivalent32
to its activity, essentially mole fraction times total gas
pressure)
– dissolution (hydrolysis) of calcite. The following reactions
are equivalent, for modeling the first equation would be
used
CaCO3 ⇀↽ Ca2+ + CO2−3
CaCO3 + H2O ⇀↽ Ca2+ + HCO−3 + OH−
CaCO3 + CO2 + H2O ⇀↽ Ca2+(aq) + 2 ·HCO−3 (aq)
– dissociation of bicarbonate
HCO−3 ⇀↽ H+ + CO2−3 (5)
33
– dissociation of carbonic acid
H2CO3 ⇀↽ H+ + HCO−3 (6)
• all of (4)– (6) are strongly influenced by pH (e.g. Fig. 9)
(4), (5), (6) control fluid pH for fluids in contact with the
atmosphere
• since atmospheric reservoir of CO2 is enormous, fluid is
buffered with respect to PCO2
34
Carbonate Species vs. pH
0
0.2
0.4
0.6
0.8
1
2 4 6 8 10 12 14
Fra
ctio
n of
Spe
cies
Pre
sent
pH
Dissolved Inorganic Carbon Species at 20oC
H2CO3 HCO3- CO3
2-
Figure 9: Carbonate species vs pH.
35
Ion exchange
• important process in groundwater chemistry, especially where
clays prominent (e.g. many soils)
• typically Ca exchanges for Na in the clay, naturally
“softening” the groundwater [e.g. Central Oklahoma aquifer,
Parkhurst et al., 1996] Figs. 10–11
• can be a problem in irrigation, eventually defloculating the
clays (destroying soil texture)
36
Cross-Section, Central Oklahoma Aquifer
Figure 10: E-W cross-section through Central Oklahoma
Aquifer, after [Fig. 5, Parkhurst et al., 1996]. Henessy Group
contains abundant clays, providing ion exchange medium.37
Stiff Diagram, Central Oklahoma Aquifer
Figure 11: Stiff diagram, Central Oklahoma Aquifer, after [Fig.
11, Parkhurst et al., 1996]. Significant Ca-Na exchange is
visible along flow path (down vertical axis of diagram).38
Applications of Major IonChemistry
39
Flow Path Determination
• Nevada Test Site/Yucca Mountain, NV [Winograd and
Thordarson, 1975], [Case Study: Great Basin (p. 250-254),
Fetter, 2001]. Also Death Valley Springs Study11
• Central Oklahoma Aquifer [Parkhurst et al., 1996]
– covers central OK, from Oklahoma City eastward about 50
miles [Fig. 2, Parkhurst et al., 1996]
– mostly composed of Permian red-bed aquifers (sandstone),
some Qal & Qt [Fig. 3, Parkhurst et al., 1996]
– unconfined to East, confined to West by Hennesy Group
(clays)11http://hydrodynamics-group.net/yucca.html
40
– groundwater flow from W to E, most streams gain from
aquifer; 100-1000 ft. of freshwater above saline (thinnest
at aquifer edges)
– Ca-Mg-HCO−3 waters in unconfined, Na-HCO−3 in confined
(ion exchange of originally atmospheric-influenced Ca-Mg-
HCO−3 recharge waters)
– high DO, nitrates, 3H, indicating relatively rapid travel
time
– Chemical reactions:
∗ after recharge, soil CO2dissolves, allowing dolomite
dissolution, generating Ca-Mg-HCO−3 water. Usually
dolomite ± calcite saturated after short travel distance
[Parkhurst, 1995]
∗ then Ca+Mg exchange for Na (natural softening) [Table
11, Parkhurst, 1995].41
– flow path calculations based on this conceptual model
indicate 50 mile flow paths requiring 103–105 years
– core analysis supports the model, indicating much of the
soluble carbonate is gone from parts of the aquifer
– mass-balance calculations also clarify the evolution of the
groundwater (e.g. probable early ET⇒ increased N, TDS)
[Fig. 119, Parkhurst et al., 1996], (Fig. 12)
42
Rock Reaction Model, Central Oklahoma Aquifer
Figure 12: Rock reaction model, Central Oklahoma Aquifer, after [Fig. 11, Parkhurst et al., 1996]. As Ca-water is
added, exchange capacity of rock is eventually exceeded at about 100 pore-volumes of water added. Note reduction in mobile
arsenic (As) by surface complexation.43
Remediation: Active Barriers
• One option is to bury a passive, permeable medium that can
remove pollutants
• Example, Elizabeth City, NC chromate plume EPA project
[Puls et al., 1996].
• Remediated by permeable iron wall (grid) reducing the
chromate and precipitating it (Fig. 13)
• Wall is 150’ long, 24 ft high, positioned to intercept the Cr
plume, made up of Fe cylinders (trench filled with iron chips
became impermeable, this design is the second try, Fig. 14.44
• Barrier is effective at removing Cr, and most of the TCE (a
common poly-chlorinated solvent, Figs. 15
45
Elizabeth City Location Map
Figure 13: Location of plume and permeable Fe wall, Elizabeth City, NJ. Hydraulic gradient is directly toward river,
contaminant is Cr-bearing TCE (solvent). After Puls et al. [Fig. 1, 1996].
46
Design of Elizabeth City Fe Wall
Figure 14: Design of permeable Fe wall, Elizabeth City, NJ,
after Puls et al. [Fig. 2, 1996].47
Chemical Changes With Time, Elizabeth City
Figure 15: Chemical changes with time, Elizabeth City, NJ,
after Puls et al. [Fig. 6, 1996]. Note immediate drop
in Cr to near zero, order-of-magnitude drop in TCE, and
seasonally-varying changes in Fe.
48
Bioremediation
• typically need to maintain nutrients (N, P), substrate (the
contaminant) and electron donor (O2)
• provides natural attenuation of many petroleum-based
contaminants [Semprini et al., 1995, Weaver et al., Sept.
11-13, 1996]
• e.g USGS study of degrading TCE plume12
• EPA natural attenuation13 overview12http://pubs.usgs.gov/sir/2006/5030/images/fig03.png13http://www.clu-in.org/techfocus/default.focus/sec/Natural_
Attenuation/cat/Overview/49
• natural attenuation in Macondo Well leak, Gulf of Mexico,
2010 (previously unknown bacteria save the Gulf, Fig. 16).
This interpretation controversial14.
14http://dx.doi.org/10.1126/science.119969750
Natural Attenuation, Macondo Well, 2010
Figure 16: Natural attenuation, Macondo Well blowout, Gulf of Mexico, 2010. FTIR shows development of petroleum
breakdown products (left), these are spatially-correlated with bacteria (right). From Hazen et al. [2010].
51
Isotope Hydrology
52
Introduction to Isotope Hydrology
Very important for determining recharge setting of and
impacts of surficial processes on any water; can be used to
estimate travel time for many waters
53
Light-Stable Isotopes
Typically these are18O16O
, DH
• usually expressed as a normalized ratio (relative to a
standard) in per-mil ( ooo). Increasing values mean enrichment
in the heavy isotope
• lighter isotope partitions into gas phase. Evaporation leads
to increasing δ18O and δD at a slope of about 5:1 on a
δ18O –δD diagram [Fig. 12-12, Domenico and Schwartz,
1998], see Fig. 17
• meteoric water line shows T-dependence of precipitation (low
T/high elev ⇒ lower δ18O –δD )54
• various processes are indicated by scatter/mixing in samples
(Fig. 18)
• local meteoric water lines can develop resulting from local
climate effects (usually rain shadows). These are invariably
more depleted than the world meteoric water line
• examples: Claassen [1985], δ18O & 14C indicate spring
waters recharged during last glacial maximum. See also
Fetter [Fig. 9.7, 2001]
55
Process of Isotopic Fractionation
Figure 17: Isotopic fractionation processes in the water cycle.
56
Isotopic Fractionation Trends
Figure 18: Isotopic fractionation trends, on δ18O -δD diagram,
after [Fig. 12.12, Domenico and Schwartz, 1998].57
Radiogenic isotopes
• 14C
– generated by cosmic radiation in the atmosphere, decays
with approximately 5000 yr half-life
– can date groundwater using14C12C
, use13C12C
ratio to correct
for isotopically “dead” carbon 12C input during subsurface
flow
• 3H (Tritium)
– produced as fallout from atmospheric nuclear testing, 13
year half-life
– 3H in precipitation peaked around 196715, useful as15http://gwadi.org/sites/gwadi.org/files/diagram1.gif
58
“natural” tracer of groundwater [Fig. 14.28, Domenico
and Schwartz, 1998]
– most often used qualitatively to discern water recharged in
the last 40 years, and to make crude velocity estimate
• 36Cl - also from fallout. Useful for waters up to 2 million years
old. Requires particle accelerator for analysis (expensive,
usually done at National Laboratories). Examples: Yucca
Mtn notes16, Scanlon [1991]
16http://www.utdallas.edu/~brikowi/Teaching/Applied_Modeling/GroundWater/LectureNotes/YuccaMtn/yuccaMtn.pdf
59
Bibliography
60
H. C. Claassen. Sources and mechanisms of recharge for ground-water in the west-centralamargosa desert, nevada – a geochemical interpretation. Prof. paper 712-f, U.S. Geol.Survey, Washington, D.C., 1985.
P. A. Domenico and F. W. Schwartz. Physical and Chemical Hydrogeology. John Wiley &Sons, New York, 2nd edition, 1998. ISBN 0-471-59762-7.
C. W. Fetter. Applied Hydrogeology. Prentice Hall, Upper Saddle River, NJ, 4th edition, 2001.ISBN 0-13-088239-9.
Terry C. Hazen, Eric A. Dubinsky, Todd Z. DeSantis, Gary L. Andersen, Yvette M. Piceno,Navjeet Singh, Janet K. Jansson, Alexander Probst, Sharon E. Borglin, Julian L. Fortney,William T. Stringfellow, Markus Bill, Mark E. Conrad, Lauren M. Tom, Krystle L. Chavarria,Thana R. Alusi, Regina Lamendella, Dominique C. Joyner, Chelsea Spier, Jacob Baelum,Manfred Auer, Marcin L. Zemla, Romy Chakraborty, Eric L. Sonnenthal, Patrik D’haeseleer,Hoi-Ying N. Holman, Shariff Osman, Zhenmei Lu, Joy D. Van Nostrand, Ye Deng, JizhongZhou, and Olivia U. Mason. Deep-sea oil plume enriches indigenous oil-degrading bacteria.Science, 330(6001):204–208, OCT 8 2010. ISSN 0036-8075. doi: 10.1126/science.1195979.
D. L. Parkhurst. User’s guide to PHREEQC-A computer program for speciation, reaction-path, advective-transport, and inverse geochemical calculations. Water-Resources Investig.Rept., 95-4227, 1995. URL http://wwwbrr.cr.usgs.gov/projects/GWC_coupled/phreeqc/.
D. L. Parkhurst, S. Christenson, and G. N. Breit. Ground-water quality assessment of the CentralOklahoma Aquifer, Oklahoma–geochemical and geohydrologic investigations. Water-SupplyPaper, 2357:101, 1996. URL http://pubs.er.usgs.gov/usgspubs/wsp/wsp2357C.Chp. C.
R. W. Puls, C. J. Paul, and R. M. Powell. Remediation of chromate-contaminated ground waterusing zero-valent iron: Field test at uscg support center, elizabeth city, north carolina. InHRSC/WERC Joint Conference on the Environment: Proceedings, Manhattan, KS, May1996. Great Plains/Rocky Mountain Hazardous Substance Research Center Kansas StateUniversity, HSRC. URL http://www.engg.ksu.edu/HSRC/96Proceed/.
B. R. Scanlon. Evaluation of moisture flux from chloride data in desert soils. J. Hydrol., (inpress), 1991.
61
L. Semprini, P. K. Kitanidis, D. H. Kampbell, and J. T. Wilson. Anaerobic transformation ofchlorinated aliphatic hydrocarbons based on spatial chemical distributions. Water Resour.Res., 31(3):1051–1062, 1995.
R Shiraki and TL Dunn. Experimental study on water-rock interactions during CO2 floodingin the Tensleep Formation, Wyoming, USA. Applied Geochemistry, 15(3):265–279, MAR2000. ISSN 0883-2927. doi: 10.1016/S0883-2927(99)00048-7.
USGS. Methods for determination of inorganic substances in water and fluvial sediments.Techniques of Water Resources Investigations of the U. S. Geological Survey, Book 5,Chap. A1:626, 1979.
USGS. National field manual for the collection of water-quality data. Techniques of WaterResources Investigations of the U. S. Geological Survey, Book 9, Sec. A:626, 1998. URLhttp://water.usgs.gov/owq/FieldManual/index.html.
J. W. Weaver, J. T. Wilson, and D. H. Kampbell. Case study of natural attenuation oftricholoroethene at st. joseph, michigan. page 9, Dallas, TX, Sept. 11-13, 1996. Symposiumon Natural Attenuation of Chlorinated Organics in Ground Water.
WEF. Standard methods for the examination of water and wastewater.Water Environment Federation, Washington, D.C., 1998. URL https://www.e-wef.org/timssnet/products/tnt_products.cfm?primary_id=S82010&Action=LONG. 20thedition.
I. J. Winograd and W. Thordarson. Hydrogeologic and hydrochemical framework, south-centralGreat Basin, Nevada-California, with special reference to the Nevada Test Site. Prof. paper712-c, U. S. Geol. Survey, 1975.
62