The occurrence and geochemistry of arsenic in groundwaters of …scp2/reprints/Peters_2007_AG_The_occurre… · The occurrence and geochemistry of arsenic in groundwaters of the Newark

Available online at www.sciencedirect.com

Applied Geochemistry 23 (2008) 85–98

www.elsevier.com/locate/apgeochem

AppliedGeochemistry

The occurrence and geochemistry of arsenic in groundwatersof the Newark basin of Pennsylvania

Stephen C. Peters *, Lori Burkert

Lehigh University, Department of Earth and Environmental Sciences, Bethlehem, PA, USA

Received 6 June 2007; accepted 1 October 2007Editorial handling by R. B. WantyAvailable online 10 October 2007

Abstract

Elevated As concentrations in groundwater in the eastern United States have been recognized predominantly in theaccretionary geologic terranes of northern New England. A retrospective examination of more than 18,000 existing ground-water samples from the Pennsylvania Department of Environmental Protection (PA DEP) Drinking Water and SamplingInformation System database indicates that elevated groundwater As concentrations occur throughout the northern half ofthe Piedmont Province of Pennsylvania. Chemical analyses of 53 samples collected in 2005 from drinking water wells in thisarea all had detectable As, and 23% of these samples contained elevated (>133 nmol/L or >10 lg/L) concentrations of As.Elevated concentrations of As in the groundwater samples were most common in the Mesozoic sedimentary strata com-posed of sandstone and red mudstone with interbedded gray shale, and gray to black siltstone and shale. Arsenic was typ-ically not elevated in groundwater of diabase intrusions of the Newark Basin or in crystalline and calcareous aquifers to thenorth of the Newark Basin. Geochemical parameters such as pH and oxidation–reduction potential can indicate mobilitymechanisms of As in some regions. In this area, measured groundwater conditions were predominantly oxidizing(Eh > +50 mV), and more than 85% of samples contained arsenate as the dominant As species. Variations in pH werestrongly correlated to the As concentration, with highest As concentrations observed at pH values greater than 6.4. Theoriginal source of As is most likely the black and gray shales that contain some arsenian pyrite with groundwater concen-trations likely to be controlled by adsorption/desorption reactions with Fe oxides in the red mudstone aquifer materials.� 2007 Elsevier Ltd. All rights reserved.

1. Introduction

1.1. Arsenic geochemistry

Arsenic is a ubiquitous trace element foundthroughout the environment. Arsenic concentra-

0883-2927/$ - see front matter � 2007 Elsevier Ltd. All rights reserveddoi:10.1016/j.apgeochem.2007.10.008

* Corresponding author. Tel.: +1 610 758 3957; fax: +1 610 7583677.

E-mail address: [email protected] (S.C. Peters).

tions in groundwater vary greatly due to the hetero-geneous distribution of source materials andsubsequent geochemical controls on aqueous Asmobility in aquifers (Cullen and Reimer, 1989).The causes of elevated As concentrations in ground-water, including the complex interactions betweenwater, geologic substrate and biological processes,are not yet completely understood. Recent workaround the globe, in such locations as West Bengal(Nickson et al., 2000), Bangladesh (Ahmed et al.,

.

mailto:[email protected]

86 S.C. Peters, L. Burkert / Applied Geochemistry 23 (2008) 85–98

2004), Vietnam (Berg et al., 2001), Spain (Garcia-Sanchez et al., 2005), Argentina, Mexico, Chile,China, Ghana, Hungary, Taiwan and Thailand(Smedley and Kinniburgh, 2002) underscore howlittle was known about As concentrations ingroundwater around the world until recently.

Elevated As in groundwater has been identifiedwithin the United States in the West and SW(Savage et al., 2000; Welch and Lico, 1998), upperMidwest (Schreiber et al., 2000), and New England(Lipfert et al., 2006; Peters and Blum, 2003; Peterset al., 1999). Dissolution and desorption of As fromnaturally occurring As-containing minerals, geo-thermal water, and mining activity appear to bethe key contributors to high-As groundwater prov-inces within the United States.

Arsenic is present in the environment in bothinorganic and methylated forms, though the inor-ganic species are considered to be the most preva-lent in groundwater (Cullen and Reimer, 1989;Welch et al., 2000). Inorganic forms are commonin As-containing minerals (O’Day, 2003; Thornburgand Sahai, 2004), sorbed on amorphous Fe(III) oxy-hydroxides (Raven et al., 1998; Wilkie and Hering,1996), sorbed on crystalline Fe oxide phases (Man-ning et al., 1998), as surface precipitates on sulfidesor pyrite (Bostick and Fendorf, 2003), and as dis-crete nanoparticulate phases (Utsunomiya et al.,2003). Widespread occurrences of elevated As con-centrations can be derived from oxidation of sulfideminerals, particularly trace-substituted pyrite, arse-nian pyrite, and arsenopyrite, and also from desorp-tion of As from mineral surfaces, or dissolution ofthe minerals with adsorbed As (Welch et al.,2000). In the case of As desorption from Fe oxideminerals, redox conditions, pH, solid-to-solutionratios, specific surface area of minerals, and compet-ing ions such as phosphate may affect As mobilityand thus As concentrations in surrounding waterse.g., (Dixit and Hering, 2003; Peters et al., 2006).

Typical geochemical parameters that may indi-cate desorption from mineral surfaces include ele-vated Eh (>100 mV), dissolved O2 (0.1–0.2 mmol/Land greater), pH (>8), and alkalinity (5–8 mmol/Land greater), and possibly high F (up to 400 lmol/L), U (>420 lmol/L), B (up to 6.7 mmol/L), Se(up to 12 lmol/L) and Mo (up to 156 lmol/L)(Smedley and Kinniburgh, 2002). Typical geochem-ical parameters that may indicate reductive dissolu-tion mechanisms include no dissolved O2, lowEh (<50 mV) and SO4 (<50 lmol/L), elevated Fe(>3 lmol/L), Mn (>9 lmol/L) and alkalinity

(5–8 mmol/L and greater), and possibly high dis-solved organic C (> 1 mmol C/L) (Smedley andKinniburgh, 2002).

The harmful health effects related to the con-sumption of elevated As in drinking water are welldocumented in groundwater provinces with As con-centrations in excess of 667 nmol/L (50 lg/L)(Chowdhury et al., 2000; Morales et al., 2000; Smithet al., 2000). Health effects that may occur at lowerconcentrations have prompted the evaluation oflarge regions where groundwater As concentrationsdo not appear to be as significant, though are exac-erbated when combined with other high risk activi-ties, such as smoking (Karagas, 2002; Karagas et al.,2004).

1.2. Local geologic setting

The state of Pennsylvania, USA, consists of fivephysiographic provinces, the Central Lowlands,Appalachian Plateaus, Ridge and Valley, the NewEngland Province, and the Piedmont (dashed lines,Fig. 1). The detailed field study area is located inthe Newark Basin within the Piedmont physio-graphic province. This physiographic province iscomposed primarily of Upper Triassic and LowerJurassic units that filled a half graben coincidentwith Mesozoic rifting. This portion of the failed riftzone is locally called the Newark Basin and therocks primarily consist of red sandstone, shale, silt-stone and conglomerate with igneous diabase intru-sions (Berg et al., 1986). Local lithologic unitsinclude playa-lacustrine deposits named the Passaicformation that laterally transition into lacustrinedeposits called the Lockatong formation. Althoughthey do not outcrop in the study area, fluvial depos-its of the Stockton Formation interfinger with andare overlain by Lockatong deposits to the south ofthe study area.

The Passaic Formation consists of quartzosesandstone grading into red mudstone, with some int-erbedded gray shale and argillite whereas Lockatongdeposits are dark siltstones, shales and argillites(Low et al., 2000). The diabase sheets are quartz nor-mative continental tholeiites of the York Haven type(Lyttle and Epstein, 1987; Smith et al., 1975). Unitsof gray mudstone, argillite, red siltstone, and redmudstone are metamorphosed to hornfels near thediabase intrusions (Froelich and Gottfried, 1999).Cambrian limestone conglomerate, quartzite con-glomerate and Precambrian gneiss occur to the northof the rift basin, but still within the field area.

Fig. 1. Map of arsenic concentration in dringking water. Circles are proportional to measured arsenic concentration. Dashed linesseparate physiographic provinces. n = 18,277.

S.C. Peters, L. Burkert / Applied Geochemistry 23 (2008) 85–98 87

The hydrologic properties of the sedimentaryrocks of the Newark Basin create thin aquifers sep-arated by thicker aquitards, while the crystallinebedrock aquifers are considerably more complex,anisotropic and heterogeneous (Low et al., 2000).Water-bearing zones in the crystalline rocks tendto be located in the weathered zones near theground surface and through fractures and joints atdepth (Sloto and Schreffler, 1994). Previous studiesof the aqueous geochemistry in this area indicatethat waters are neutral to slightly basic in the Lock-atong Formation whereas water is neutral to slightlyacidic in the Passaic Formation and diabase, andelevated concentrations of Fe, Mn and SO4 occurbut are not common throughout the aquifers (Lowet al., 2000).

A comprehensive analysis of the extent, levels,and causes of naturally occurring As enrichmenthas not been conducted in Pennsylvania, USA.The present research occurs at two scales, and withtwo intents. At the regional scale, a GIS analysiswas conducted using existing groundwater quality

data from state and federal databases (PA DEP,2004; USGS, 2001) to highlight areas within thestate that are more likely to have elevated As. Basedon these data, a field study was then conducted at alocal scale to focus on As concentration, speciationand potential mobilization mechanisms. The twoprimary intents of this paper are to highlight theimportance of using existing data to guide researchefforts and to evaluate field evidence for experimen-tally observed As geochemical phenomena. Thisstudy, which uses southeastern Pennsylvania as anatural laboratory, has specific applicability locally,but also potentially within other Mesozoic riftbasins along the Atlantic Ocean margin.

2. Methods

2.1. Statewide database and GIS analysis

A GIS database was prepared to analyze the rela-tionship between As occurrence, geography, andbedrock geology in Pennsylvania to determine an


appropriate field study location (Fig. 1). The maincriteria used to determine an appropriate locationwere evidence of both low (<133 nmol/L, 10 lg/L)and elevated (>133 nmol/L) As concentrations inthe groundwater within a single region. Test resultsfor As concentration in drinking water samples wereobtained from the Pennsylvania Department ofEnvironmental Protection (PA DEP) DrinkingWater and Sampling Information System for 1994through 2004 (PA DEP, 2004), and the USGSNational Water Information System (USGS,2001). Data entry errors in the PA DEP databaserequired extensive manual review and correctionbefore analysis of the geochemical data.

2.2. Field methods

Fifty-three water samples were collected (July–October, 2005) in southeastern Pennsylvania(Fig. 2) from 49 private homes, 2 schools and onemunicipal government building using groundwaterwells as their source of drinking water, and one hill-side spring used by local residents who fill their owncontainers. Groundwater wells were screened inaquifers in the following lithostratigraphic units:Passaic Formation, Lockatong Formation, diabase,Leithsville Formation, felsic to mafic gneiss, andhornblende gneiss. Well and spring locations werecollected with a Garmin GPSMAP 76S handheldGPS unit. Temperature, pH (Corning 3-in-1 Combi-nation IP67) and Eh (Hanna HI 98201) were deter-mined in the field using a continuous flow throughcell. The probes were calibrated daily with 4.0, 7.0and 10.0 pH standards for the pH probe and+468 and +220 mV standards for the Eh probe.Groundwater samples were collected from wellsusing outside taps after thoroughly flushing thetap for at least 15 min or until temperature measure-ments indicated that all onsite storage was purgedand water originated from the well bore or aquifer(�11–14 �C). All sampling containers were rinsedthree times with sample water before collecting thefinal sample using a syringe and a 0.45 lm polypro-pylene filter (Whatman PP Puradisc). Samples fortotal As, major anion and cation analysis were col-lected in acid washed LDPE bottles. Samples formajor cation analysis were fixed to pH 2 with dou-ble-distilled ultrapure HNO3. Samples for As speci-ation analysis were collected in 40 mL cleaned glassvials with Teflon-lined caps with no headspace. Allsamples were kept on ice and in the dark until anal-yses were complete.

2.3. Analytical methods

Within 24 h, groundwater samples were ana-lyzed for As speciation using continuous-flow cou-pled ion chromatography (IC, Dionex IONPACAG4A Guard and AS4A Analytical Columns),hydride generation (HG, in-house design) withultraviolet oxidation and inductively coupledplasma-mass spectrometry (ICP-MS, Thermo X-Series CCT) generally following the method ofKlaue and Blum (1999). Arsenic species were sepa-rated on a strong anion exchange column using thestandard eluent of 1.8 mM Na2CO3 and 1.9 mMNaHCO3 at a flow rate of 1 mL/min and �75 barpressure. The eluent containing separated As spe-cies was mixed online with a solution of 1 MHNO3 and allowed to react for �10 s before mix-ing with a solution of 1% NaBH4 (w/v) stabilizedin 0.1 M NaOH (Klaue and Blum, 1999). Crosscalibrations using single species and total As stan-dards were used to verify the column separationand the performance of the entire analytical appa-ratus. Samples were also analyzed for total Asusing hydride generation with ultraviolet oxidationcoupled to ICP-MS. The analytical detection limitwas approximately 0.1 nmol/L total As concentra-tion, which is lower than all sample concentrations.Arsenic concentrations are generally reported asnmol/L, and where applicable, breaks between cat-egories correspond to US regulatory concentra-tions (e.g. 133 nmol/L = 10 lg/L). Major cationsand other trace elements were determined usingICP-MS with pneumatic nebulization and anionswere determined using IC with suppressed conduc-tivity detection.

3. Results

3.1. Statewide survey

Groundwater samples (n = 18,277) submitted tostate testing agencies were compiled, entered into aGIS, and plotted in Fig. 1. Concentrations rangefrom below detection to 1.1 lmol/L with a medianvalue that was below detection. All the PA DEPdrinking water sample results from within a givengeologic formation were averaged and assigned tothat formation wherever it occurs (Fig. 2). Areaswith no shading have As concentrations less than66 nmol/L. Dark gray and black regions have aver-age As concentrations greater than 66 nmol/L, andin some cases greater than 133 nmol/L. This map

Fig. 2. Map of geologic units coded by arsenic concentration (upper). Field area is indicated by the box. Map of geologic units in fieldstudy area, including drinking water arsenic concentrations (lower).


Table 1Summary data table

# T (�C) pH Eh (mV) As* % As(V) B Na Mg K Ca Mn Fe Cl NO3 SO4 Map Unit

1 12.6 5.9 460 16.9 96 0.4 1800 0.0 1.0 0.0 0.0 0.0 78 55 260 hg Hornblende gneiss2 12.1 7.4 300 7.20 100 3.1 120 720 29 660 0.0 0.45 300 35 44 gn Felsic to mafic gneiss3 12.3 5.1 320 0.16 83 0.2 380 270 53 320 0.0 0.06 700 190 220 gn Felsic to mafic gneiss4 13.0 4.9 290 0.16 100 16 400 150 38 190 0.0 0.03 580 110 120 gn Felsic to mafic gneiss5 13.1 5.3 260 0.19 86 1.3 1500 590 85 700 0.4 0.10 2400 160 310 gn Felsic to mafic gneiss6 13.0 4.9 260 1.43 100 0.3 170 130 28 180 0.0 0.06 110 44 150 hg Hornblende gneiss7 14.7 5.3 230 0.88 100 3.6 620 300 46 340 0.0 0.00 460 780 130 hg Hornblende gneiss8 14.3 6.9 110 0.41 97 0.8 1500 13.9 40 1400 0.0 6.5 1800 200 580 Civ Leithsville Formation9 13.0 7.2 120 1.19 95 0.2 230 730 65 700 0.0 0.59 210 190 50 Civ Leithsville Formation

10 13.9 5.5 330 2.51 94 2.2 430 440 18 560 0.0 0.0 180 170 350 Jd Diabase11 15.4 5.6 180 0.92 65 5.5 280 520 20 800 0.0 0.10 650 210 340 Jd Diabase12 13.7 6.0 22 27.6 29 2.4 550 520 69 600 17 7.4 340 8.5 610 Trb Brunswick Formation13 14.1 6.8 130 135 99 0.9 500 840 26 1100 0.0 0.03 250 340 280 Trb Brunswick Formation14 14.1 5.6 150 1.76 85 4.1 760 720 44 930 0.0 0.44 2000 170 270 Jd Diabase15 13.1 5.3 170 0.32 83 1.7 490 740 16 730 0.0 0.00 1300 410 330 Trb Brunswick Formation16 15.8 6.9 100 524 72 88 1200 1000 29 5100 2.1 1.5 240 8.5 5400 Trb Brunswick Formation17 13.5 7.0 120 111 97 11.0 530 870 22 1100 0.0 0.16 530 230 270 Trb Brunswick Formation18 13.1 7.2 130 63.4 98 5.5 430 670 18 880 0.0 0.19 280 140 230 Trb Brunswick Formation19 14.1 7.1 140 20.9 95 2.2 490 890 25 1100 0.0 0.21 520 320 280 Trb Brunswick Formation20 16.6 7.2 140 71.6 87 11 960 1700 65 2100 0.0 0.33 2900 110 160 Trb Brunswick Formation21 38.3 48 3.4 460 960 23 15.0 .0 0.23 370 300 140 Trl Lockatong Formation22 247 85 12 510 730 19 990 0.0 0.15 110 58 300 Trb Brunswick Formation23 165 91 2.0 350 730 21 830 0.0 0.16 140 87 160 Trb Brunswick Formation24 114 88 38 430 590 28 1000 0.0 0.09 330 76 260 Trb Brunswick Formation25 61.5 88 3.0 380 730 12 890 0.0 0.09 320 100 160 Trb Brunswick Formation26 64.0 84 4.5 360 800 16 890 0.0 0.04 270 48 580 Trb Brunswick Formation27 14.2 6.9 360 306 79 9.1 4100 0.0 3.5 0.0 0.0 0.00 120 8.5 660 Trb Brunswick Formation28 13.5 6.7 170 14.2 83 2.4 730 1300 31 1700 0.0 0.25 1400 190 260 Trb Brunswick Formation29 14.6 7.0 67.9 73 2.2 600 1100 26 1400 0.0 0.36 1300 130 360 Trb Brunswick Formation30 14.0 7.0 68.1 84 2.9 540 1300 29 1400 0.0 0.25 1800 200 72 Trb Brunswick Formation31 14.3 7.2 �230 615 9 68 1200 390 24 1100 0.7 1.6 79 8.5 360 Trb Brunswick Formation32 14.3 7.2 110 265 83 29 620 850 22 1000 0.0 0.05 350 55 540 Trb Brunswick Formation33 13.9 7.3 160 51 51 3.0 420 760 21 920 0.0 0.09 330 170 110 Trb Brunswick Formation34 14.0 7.2 110 97.1 80 4.5 500 1200 30 1100 0.0 0.11 1300 250 120 Trb Brunswick Formation35 16.0 7.2 120 105 83 16 530 840 21 1000 0.0 0.00 340 150 420 Trb Brunswick Formation36 14.1 7.0 110 30.5 92 3.1 370 800 19 930 0.0 0.10 600 260 200 Trb Brunswick Formation37 13.4 5.8 52 6.60 52 12.7 3900 2600 100 1300 6.8 10.0 8900 6.8 690 Trb Brunswick Formation38 14.2 6.6 83 39.5 83 4.1 860 1100 24 1500 0.0 0.24 1300 240 250 Trb Brunswick Formation39 14.2 6.6 98 184 89 23 600 990 23 1300 0.0 0.04 200 97 1500 Trb Brunswick Formation40 14.0 6.7 100 85.3 87 19 680 840 39 1100 0.0 0.16 200 100 280 Trb Brunswick Formation41 13.9 6.4 87 2.25 45 2.0 810 1500 79 1500 0.0 0.28 1200 510 350 Civ Leithsville Formation42 18.6 5.2 160 0.67 50 0.7 160 160 55 830 0.1 0.12 270 74 250 hg Hornblende gneiss

90S

.C.

Peters,

L.

Bu

rkert

/A

pp

liedG

eoch

emistry

23

(2

00

8)

85

–9

8

4313

.56.

717

526

1722

990

480

914

000.

54.

213

08.

541

0T

rlL

ock

ato

ng

Fo

rmat

ion

4416

.16.

583

865

922.

251

036

050

1400

1.6

0.13

820

8.5

260

Trl

Lo

ckat

on

gF

orm

atio

n45

13.9

6.5

807.

6850

4913

0023

0089

3800

0.0

3.7

11,0

0011

055

0T

rlL

ock

ato

ng

Fo

rmat

ion

4613

.17.

476

29.1

7826

079

094

347

00.

00.

0016

015

150

JdD

iab

ase

4717

.05.

518

03.

6184

4.6

480

740

3867

00.

00.

0015

0020

025

0Jd

Dia

bas

e48

15.0

6.4

140

158

7925

340

1000

4795

00.

00.

0020

013

500

JdD

iab

ase

4912

.95.

416

00.

7374

4.2

260

700

951

00.

00.

0012

0031

020

0Jd

Dia

bas

e50

13.4

6.7

509.

3170

3.3

330

680

3196

00.

00.

4034

058

170

Qt

Tre

nto

nG

rave

l51

12.7

6.7

110

95.9

8310

480

800

3399

00.

00.

0024

012

026

0T

rlL

ock

ato

ng

Fo

rmat

ion

5212

.76.

811

071

.184

4.9

380

620

3086

00.

00.

0033

011

020

0T

rlL

ock

ato

ng

Fo

rmat

ion

5312

.65.

117

00.

8476

0.9

240

260

3437

00.

00.

0022

041

022

0h

gH

orn

ble

nd

egn

eiss

All

con

cen

trat

ion

sex

pre

ssed

asl

mo

l/L

exce

pt

As*

wh

ich

isre

po

rted

asn

mo

l/L

.


suggests the occurrence of elevated concentrationsof As in three geographic areas: sandstone aquifersin both the NE and east-central regions of Pennsyl-vania, and in mudstone and shale aquifers in the SEregion of the state. The main criteria used to deter-mine an appropriate location for a detailed fieldinvestigation were evidence of both low (<133nmol/L) and elevated (>133 nmol/L) As concentra-tions in the groundwater within a single region.From these three regions, the SE corner was chosenas the study area because it met the criteria for thepresence of elevated concentrations of As and wasalso located close to the Lehigh University geo-chemical laboratory where time-sensitive chemicalanalyses could be conducted.

3.2. Field survey

Arsenic was detected in all 53 samples from thestudy area, and concentrations ranged from0.16 nmol/L up to 864 nmol/L with a median valueof 38 nmol/L (Tables 1 and 2). Approximately 23%of all groundwater samples contained elevated(>133 nmol/L) concentrations of total As. Whenexamined in the context of aquifer host material,33% of groundwater samples from wells withintwo sedimentary formations, the Lockatong andthe Passaic, of the Newark Basin contained elevatedconcentrations of total As. The average and maxi-mum As concentration in water samples from wellswithin the Lockatong Formation is greater thansurrounding formations, followed by the PassaicFormation (Table 2). Samples from wells locatedin the Precambrian gneisses and Cambrian con-glomerate rocks in the northern part of the studyarea did not contain elevated As. Arsenic was alsonot elevated in the diabase aquifers except for onesample (ID# 48) located in the northeastern partof the study area near the contact between the dia-base and the Passaic Formation.

The statistical distribution of As concentrations inthe detailed study area (Table 2) compares favorablywith summary data from the same geologic units(Table 3) within the Pennsylvania Drinking WaterInformation System database drinking water dataset(PA DEP, 2004). In both datasets, samples from thePassaic and Lockatong formations containedthe highest concentrations of As. In both datasets,the percentage of samples with elevated As concen-trations are similar for all formations except for theLockatong Formation, which is due to a largenumber of nondetect results in the PA DEP dataset.

Table 3Summary statistics for samples in PADEP database

Formation # of As samples # of As samples >133 nmol/L Arsenic concentration (nmol/L)

Mean Median Max Min

Combined Formationsa 2310 630 120 26.7 1100 NDPassaic 1538 577 173 66.7 1100 NDLockatong 283 42 53 ND 853 NDDiabase 142 7 26.7 ND 560 NDPrecambrian/Cambrianb 347c 4 ND ND 320 ND

ND = Nondetect.a Represents data from PA DEP Drinking Water and Sampling Information System database (PA DEP, 2004) from the same for-

mations as this study. PA DEP data may represent parts of some formations that are outside of this study’s extent.b Precambrian and Cambrian lithologic units: combined results of felsic to mafic gneiss, Hornblende gneiss, and Leithsville Formation.c One sample result not included due to a data entry decimal point error.

Table 2Summary statistics for samples collected in Newark Basin

Formation # of As samples # of As samples >133 nmol/L Arsenic concentration (nmol/L)

Mean Median Max Min

Entire study 53 12 103 38.3 865 0.1Passaic 27 9 135 85.3 615 0.3Lockatong 6 2 267 83.5 865 7.7Diabase 7 1 28.0 2.5 158 0.8Precambrian/Cambriana 12 0 2.7 0.9 16.9 0.1Otherb 1 0 9.3 9.3 9.3 9.3

a Precambrian and Cambrian lithologic units: combined results of felsic to mafic gneiss, Hornblende gneiss, and Leithsville Formation.b Other = Trenton Gravel.


3.3. Geochemical indicators of arsenic release

In this study, elevated total As concentrationsoccur generally at circumneutral to high pH(Fig. 3a). All samples less than pH 6.4 contained lessthan 27 nmol/L total As, while samples above pH6.4 contained up to 894 nmol/L. This transitionalzone is noted on Fig. 3a by a dashed line. None ofthe samples collected as part of this study had ameasured pH greater than 7.5. Measured Eh valuesof the groundwater samples ranged from �232 mVto +456 mV with most of the samples having valuesin the +25 mV to +300 mV range. The behavior ofFe in this system may help explain some of the Asdata due to the close association between Fe andAs in mineral form, and the ability of Fe-oxides toadsorb As (Smedley and Kinniburgh, 2002). Ironconcentrations range from below detection(<0.02 lmol/L) up to 10.5 lmol/L (Fig. 3b). Thelowest concentrations of Fe occur throughout therange of measured pH values. The highest concen-trations of Fe occur at pH �5.8 and decrease withincreasing pH. The pH dependence of dissolved Feconcentrations and oxidation/precipitation reac-

tions has been well characterized (Stumm and Mor-gan, 1996). In this system, the stability of Feminerals can be represented by an activity diagramof dissolved Fe vs. pH at varying Eh values(Fig. 4). According to this diagram, samples fromthis area plot near the boundary of Fe oxyhydrox-ides, suggesting precipitation is likely to occur ifpH values increase or O2 becomes more available.The precipitation of Fe-oxyhydroxides may affectAs concentrations by providing substrate for Asadsorption (Pichler et al., 1999). Dissolved Fe con-centrations do not appear to be directly correlatedto As, and the highest concentrations of As occurat a wide range of Fe concentrations (Fig. 3c). Inter-estingly, the sample containing the highest concen-tration of Fe contains very low As.

Sulfate concentrations range from 43 to5400 lmol/L, with a median value of 260 lmol/L.Iron and SO4 in water samples do not appear to havea direct dissolution relationship (Fig. 5a), suggestingeither Fe sulfide dissolution followed by Fe oxideprecipitation or non-Fe sources of SO4, such as gyp-sum. Gypsum has not been reported in the field area,and is unlikely to explain the excess of the SO2�

4 .

0

250

500

750

1000

4.0 5.0 6.0 7.0 8.0pH

Ars

enic

(nm

ol/L

)

As(V)<20%As(V) 20-80%As(V)>80%

pH =

6.4

0

2

4

6

8

10

12

4.0 5.0 6.0 7.0 8.0pH

Iron

(um

ol/L

)

As <13 nmol/LAs 13-133 nmol/LAs 133-266 nmol/LAs >266 nmol/L

0

250

500

750

1000

0 2 4 6 8 10 12Iron (umol/L)

Ars

enic

(nm

ol/L

)

a

b

c

Fig. 3. Plots of total arsenic and iron concentrations vs pH(a and b), and arsenic vs iron (c) in groundwater samples.


Total As appears inversely correlated with bothNO3 and Cl, with high values of As occurring atlow values of these two ions (Fig. 5b and c). Rela-tionships did not exist between total As concentra-tion and F, SO4, Mg, Ca, Na or K (R2 range of0.001–0.1431, not shown). Nitrite was not detectedin enough samples to determine a relationship,though the few samples that did contain detectableNO�2 did not have different ranges of As concentra-tions than those that did not (p > 0.20). Phosphatewas not detected in any samples (detection limit of5 lmol/L). Manganese was only detected in 8 sam-ples; however, 4 of the 8 samples also containedthe 4 highest concentrations of As.

3.4. Arsenic speciation

Approximately 70% of groundwater samplesfrom the study region contained mostly (>80%)the oxidized As species arsenate. Slightly more than3% of the samples contained mostly arsenite, andthe remaining samples had an intermediate mix of20–80% of both species. Of the 12 samples with ele-vated concentrations of total As, 10 are mostly arse-nate while the remaining two samples were mostlyarsenite. Arsenate-dominated samples occurthroughout the pH range of the samples while thearsenite-dominated samples occur within a smallerpH range (pH 5.8–7.2). Not surprisingly, the pro-portion of arsenate is generally positively relatedto Eh (R2 = 0.383), with arsenate-dominated sam-ples generally occurring at high Eh. An Eh-pH dia-gram was constructed of the Fe–As–SO4–H2Osystem using the software Geochemist’s Workbenchto evaluate the aqueous and mineral thermody-namic stabilities of the system (Fig. 6). This study’sdata were plotted in the context of the modelaccording to measured Eh and pH. Samples domi-nated by arsenate plot within the arsenate stabilityfields indicating that the As species in these samplesare at or close to equilibrium. Samples with interme-diate compositions of each species (20–80%) plotwithin both arsenate and arsenite stability fields,but occur close to the boundary between these spe-cies. A few samples plot in stability fields oppositeto their measured speciation and indicate that eitherthey are not in equilibrium with their environment,or that the measurement technique for Eh did notaccurately record the redox conditions. The samplethat plots within the orpiment stability field is likelydue to the inaccurate measurement of Eh that is notsurprising when using a probe-type system.

4. Discussion

The prevalence of As in this groundwater systemis generally consistent with a natural source of Aswhose concentrations are moderated by exchangereactions between the aqueous and solid phases.In the following paragraphs, the relationshipbetween this study and other examinations of Asbehavior in both the laboratory and field environ-ments are discussed.

In regions where it is unclear whether As isderived from natural or anthropogenic sources,geochemical indicators may help identifypotential anthropogenic sources. For example, high

Fig. 4. Diagram of Fe activity under variable pH conditions. Stability fields are based on thermodynamic calculations from Bethke (2000)and the LLNL Thermodynamic dataset V8 R6. Stability fields were calculated and contoured for Eh = 0 mV, Eh = 100 mV andEh = 200 mV. SO4 concentration fixed at 312 lmol/L. Temperature fixed at 13 �C. This study’s data are plotted by measured Fe activityand pH according to different measured Eh ranges. Eh ranges are represented by different markers. Closed squares represent Eh < 0 mV.Closed circles represent Eh 0–100 mV. Open squares represent Eh 100–200 mV. Open circles represent Eh > 200 mV. Goethite, hematite,magnetite, and malanterite were suppressed during modeling. n = 32.


concentrations of Cl might indicate landfill leachateor road salt (Christensen et al., 1994; Fritz et al.,1994; Panno et al., 2006) and NO3 could be due toagricultural fertilizer infiltration (Hamilton and Hel-sel, 1995). It would expected that if water samplescontained high concentrations of these two com-pounds (or other indicator compounds), that theAs in the groundwater may be related to the pro-cesses that contribute these indicator elements.Arsenic concentrations in groundwater from this siteshow either inverse or no correlation with either ofthese parameters (Fig. 5b and c), suggesting thatAs is derived from natural sources. Theoretically, ifAs were due to mineral dissolution and possiblyadsorption/desorption reactions, then evidence ofthese processes should be present in the water chem-istry. Arsenic concentrations appear dependent on athreshold value of pH, above which elevated As ispresent in groundwater, and below which As con-centrations remain low. The moderately oxidizingconditions demonstrated by the dominance of arse-nate and measured Eh values greater than 100–

200 mV along with the observed pH dependencesuggest that the controlling mechanism is likelydesorption from mineral surfaces (e.g., Peters andBlum, 2003). At low concentrations of aqueousFe2+ in solution, the adsorption and desorption ofAs anions from Fe-oxyhydroxide minerals isstrongly controlled by pH-controlled changes in sur-face affinity of available mineral substrate asobserved by others (Smedley et al., 2005). In someareas where a few results indicate weakly oxidizingto strongly reducing conditions, reductive desorp-tion or primary sulfide dissolution may be possibleas localized mechanisms. Some support for reducingmobilization triggers exists in water samples thathave intermediate proportions of each As species,and those that plot in opposing stability fields sug-gest an ongoing redox transformation or possiblymixing with waters of different redox status (Petersand Blum, 2003).

Work conducted by Serfes (2005) in the samegeologic formations in the New Jersey portion ofthe Newark Basin is consistent with desorption pro-

10

100

1000

10000

0 2 4 6 8 10 12Iron (umol/L)

Sulfa

te (u

mol

/L)

0

250

500

750

1000

0 5000 10000 15000Chloride (umol/L)

Ars

enic

(nm

ol/L

)

0

250

500

750

1000

0 200 400 600 800 1000Nitrate (umol/L)

Ars

enic

(nm

ol/L

)

a

b

c

Fig. 5. Plots of sulfate vs iron concentration (a), and total arsenicconcentration vs nitrate and chloride concentration in ground-water samples (b and c).


cesses leading to the presence of aqueous As inwaters of the Passaic Formation; clay mineralscoated with early hematite were determined to likelybe the source of the As via desorption (Serfes, 2005;Serfes et al., 2005). Sulfide dissolution mechanismswere predicted in the black shales of the LockatongFormation (Serfes, 2005; Serfes et al., 2005).Although not common, reducing conditions werefound with elevated concentrations of As in somesamples from this study. Samples with As concen-trations dominated by arsenite were generally foundin both the Passaic and Lockatong formations.Although oxidizing to mildly reducing conditionspresent in all the Lockatong samples (Table 1; Ehrange 17–112 mV) may support oxidative dissolu-

tion of pyrite, other geochemical factors measuredin this study are less supportive. For example, SO4

is far more abundant in solution relative to Fe.Molar ratios close to 2:1 to 1:1 for SO4 and Fewould be expected from pyrite or arsenopyrite oxi-dation and subsequent congruent dissolution inthe mildly reducing conditions, especially at neutralto low pH. Actual ratios range from a minimumSO4:Fe of 6:1 to a maximum of more than 3600:1and median of 120:1. Sulfate and Ca in solutiondo not correlate and therefore gypsum dissolutionis unlikely to explain the elevated SO4. Excess SO4

likely remains in solution from the primary dissolu-tion of pyrite during the evolution of groundwaterwhereas the Fe is precipitated as amorphous andcrystalline forms of Fe-oxyhydroxides. A consider-able amount of Fe could easily have precipitatedafter initial sulfide dissolution, but there is no con-clusive evidence that pyrite dissolution deliveredthe initial As to these groundwaters, and it wouldbe difficult to conceive a study that would deliverdefinitive proof. Comparing the concentrations ofSO4 and As to the possible source materials can testthe feasibility of As being derived from the dis-solution of arsenian pyrite. For the sample contain-ing the highest As (865 nmol/L) and lowest SO4

(260 lmol/L) concentrations, arsenian pyrite ofapproximately 0.3% As (mol/mol) is needed. Thiscalculation assumes congruent dissolution and con-servative transport, but illustrates that very littlearsenian pyrite is needed to explain the observedconcentrations of As in groundwater.

The strength of arsenate adsorption does not dif-fer between more crystalline forms of Fe-oxyhy-droxide minerals such as goethite and magnetite toless crystalline forms such as hydrous ferric oxides(HFO); however, more crystalline forms exhibit anoverall decrease in specific surface area and there-fore site density (Dixit and Hering, 2003). There-fore, a conversion from less crystalline forms tomore crystalline forms could result in decreasedAs adsorption and increased As mobility. In thecase of precipitation of aqueous Fe to amorphousforms of Fe-oxyhydroxides, the newly formed Feminerals may adsorb more As from solution relativeto the surrounding crystalline Fe minerals. Overtime, as these amorphous phases mature into morecrystalline phases, As may be slowly released asadsorption site density decreases.

A tradeoff exists at higher pH values betweenan increase of the precipitation rate of Fe oxyhy-droxides and the decrease in the affinity of the

Fig. 6. Eh–pH diagram of the Fe–As–SO4–H2O system. Stability fields modeled using Geochemist’s Workbench (Bethke, 2000) and theLLNL thermodynamic dataset V8 R6. As concentration fixed at 133 nmol/L for modeling purposes. Fe concentration fixed at 180 nmol/L.SO4 concentration fixed at 312 lmol/L. Temperature fixed at 13 �C. Dashed line indicates lower bound of water stability. This study’s dataare plotted by measured Eh and pH according to different percentages of arsenate (As(V)) measured in solution Open circles contain lessthan 20% As(V). Open squares contain between 20% and 80% As(V). Solid triangles contain more than 80% As(V). n = 45.


precipitated substrate for As. At low pH Fe-oxyhy-droxides are stable in a solid state and have a posi-tively-charged surface that readily attracts Asanions. As the pH increases through the circumneu-tral range, the net Fe precipitation rate increases butthe surface charge of the mineral substrate becomesless positive thereby decreasing the total affinity ofthe Fe-oxyhydroxide surface for As anions. Thepresence of As in solution at both circumneutraland high pH values may be related to a combinationof the availability of precipitated Fe substrates atcircumneutral pH and possibly the adsorption prop-erties of Fe minerals for As at higher pH values.This study’s results were plotted in the context ofthermodynamic modeling of Fe activity under dif-ferent redox conditions using Geochemist’s Work-bench (Fig. 4). Iron concentrations are mostly inagreement with pH and redox conditions as con-strained by the model; however, samples plottingon or near solubility boundaries may either be

waters seeking equilibrium that are kineticallyinhibited from precipitating due to pH conditions,or have reached steady state.

The lack of a positive correlation between totalAs concentration and aqueous Fe concentration(Fig. 3c) suggests that As concentrations are gov-erned by a complex set of reactions, and not by sim-ple dissolution of an As bearing Fe phase. Atcircumneutral pH, kinetic controls on the precipita-tion of Fe (Peters and Blum, 2003), forms of Fepresent (Raven et al., 1998), surface structure attri-butes of Fe (Goldberg and Johnston, 2001; Shermanand Randall, 2003), and speciated forms of As(Dixit and Hering, 2003) simultaneously affect Asmobility in the groundwater.

5. Conclusions

Geographic analysis of an existing drinking waterquality database has proven to be a useful tool to


identify a region for further study with elevated con-centrations of As in the groundwater. While devel-oping a research program utilizing this approach,researchers should be aware of the sensitive natureof drinking water source locations, and the presenceof erroneous results due to data entry errors.

The presence of elevated As in groundwaters ofthe Newark Basin, Pennsylvania, USA, is primarilydue to desorption from Fe oxide minerals underoxidizing conditions at pH values greater than6.4. The original source of the As may be fromAs-bearing Fe sulfide minerals in the shales ofthe Lockatong Formation and interbedded shalesof the Passaic Formation (Serfes, 2005). The spe-cific geologic observations from this study may beuseful in directing efforts to understand As mobil-ity in other Mesozoic rift basins along the AtlanticCoast.

Considerable emphasis has been placed on study-ing reducing conditions in Quaternary river deltaenvironments (e.g. Bangladesh) because of the grav-ity and magnitude of human health impacts; how-ever, aquifers located in older and more crystallinegeologic environments can also have elevated Asconcentrations in groundwater, although the rangeof concentrations is considerably less than concen-trations reported from Bangladesh and geologicallysimilar areas. With the implementation of morestringent As standards in drinking water in the Uni-ted States, there is also a need to examine As mobilityin groundwater environments previously consideredto be low risk.

Acknowledgements

The authors would like to thank the US EPA forsupporting L.B. as a STAR Fellow (EPA-FP-916633) and to the Lehigh University FacultyResearch Grant program and to the Earth andEnvironmental Sciences Palmer Field Program forgrants to S.C.P. and L.B. that partially supportedthis project. L.B. would like to acknowledge MauraSullivan, Christopher Call, Michael Kutney andRobert Stanfield for assistance in the field and thelaboratory. This manuscript was improved by sug-gestions from three anonymous reviewers.

References

Ahmed, K.M., Bhattacharya, P., Hasan, M.A., Akhter, S.H.,

Alam, S.M.M., Bhuyian, M.A.H., Imam, M.B., Khan, A.A.,

Sracek, O., 2004. Arsenic enrichment in groundwater of the

alluvial aquifers in Bangladesh: an overview. Appl. Geochem.

19, 181–200.

Berg, M., Tran, H.C., Nguyen, T.C., Pham, H.V., Schertenleib,

R., Giger, W., 2001. Arsenic contamination of groundwater

and drinking water in Vietnam: a human health threat.

Environ. Sci. Technol. 35, 2621–2626.

Berg, T., McInerney, M., Way, J., MacLachlan, D., 1986.

Stratigraphic correlation chart of Pennsylvania. 2nd Printing,

Revised. Pennsylvania Geol. Surv., 4th Series, General

Geology Report #75.

Bethke, C.M., 2000. The Geochemist’s Workbench, v.3.1.

Bostick, B.C., Fendorf, S., 2003. Arsenite sorption on troilite

(FeS) and pyrite (FeS2). Geochim. Cosmochim. Acta 67, 909–

921.

Chowdhury, U.K., Biswas, B.K., Chowdhury, T.R., Samanta,

G., Mandal, B.K., Basu, G.C., Chanda, C.R., Lodh, D.,

Saha, K.C., Mukherjee, S.K., Roy, S., Kabir, S., Quamruzz-

aman, Q., Chakraborti, D., 2000. Groundwater arsenic

contamination in Bangladesh and West Bengal, India. Envi-

ron. Health Perspect. 108, 393–397.

Christensen, T.H., Kjeldsen, P., Albrechtsen, H.J., Heron, G.,

Nielsen, P.H., Holm, P.L., Bjerg, P.E., 1994. Attenuation of

landfill leachate pollutants in aquifers. Critical Rev. Environ.

Sci. Technol. 24, 119–202.

Cullen, W.R., Reimer, K.J., 1989. Arsenic speciation in the

environment. Chem. Rev. 89, 713–764.

Dixit, S., Hering, J.G., 2003. Comparison of arsenic(V) and

arsenic(III) sorption onto iron oxide minerals: implications

for arsenic mobility. Environ. Sci. Technol. 37, 4182–4189.

Fritz, S.J., Bryan, J.D., Harvey, F.E., Leap, D.I., 1994. A

geochemical and isotopic approach to delineate landfill

leachates in an RCRA study. Ground Water 32, 743–750.

Froelich, A., Gottfried, D., 1999. Early Mesozoic igneous and

contact metamorphic rocks. In: Shulte, C. (Ed.), The Geology

of Pennsylvania Special Publication #1. Pennsylvania Geo-

logical Survey and Pittsburgh Geological Soceity, Harris-

burg–Pittsburg, PA, pp. 202–209.

Garcia-Sanchez, A., Moyano, A., Mayorga, P., 2005. High

arsenic contents in groundwater of central Spain. Environ.

Geol. 47, 847–854.

Goldberg, S., Johnston, C.T., 2001. Mechanisms of arsenic

adsorption on amorphous oxides evaluated using macro-

scopic measurements, vibrational spectroscopy, and surface

complexation modeling. J. Colloid Interf. Sci. 234, 204–

216.

Hamilton, P.A., Helsel, D.R., 1995. Effects of agriculture on

groundwater quality in 5 regions of the United States.

Ground Water 33, 217–226.

Karagas, M., 2002. Assessment of cancer risk and environmental

levels of arsenic in New Hampshire. Int. J. Hygiene Environ.

Health 205, 85–94.

Karagas, M.R., Tosteson, T.D., Morris, J.S., Demidenko, E.,

Mott, L.A., Heaney, J., Schned, A., 2004. Incidence of

transitional cell carcinoma of the bladder and arsenic

exposure in New Hampshire. Cancer Causes Control 15,

465–472.

Klaue, B., Blum, J.D., 1999. Trace analysis of arsenic in drinking

water by inductively coupled plasma mass spectrometry: high

resolution versus hydride generation. Anal. Chem. 71, 1408–

1414.

Lipfert, G., Reeve, A.S., Sidle, W.C., Marvinney, R., 2006.

Geochemical patterns of arsenic-enriched ground water in


fractured, crystalline bedrock, Northport, Maine, USA. Appl.

Geochem. 21, 528–545.

Low, D., Hippe, D., Yannacci, D., 2000. Geohydrology of

southeastern Pennsylvania. US Geol. Surv. Water Resour.

Invest. Rep. 00-4166.

Lyttle, P., Epstein, J., 1987. Geological map of the Newark 1� · 2�quadrangle, New Jersey, Pennsylvania, and New York. US

Geol. Surv. Miscellaneous Invest. Series Map I-1715.

Manning, B.A., Fendorfi, S.E., Golderg, S., 1998. Surface

structures and stability of arsenic(III) on goethite: Spectro-

scopic evidence for inner-sphere complexes. Environ. Sci.

Technol. 32, 2383–2388.

Morales, K.H., Ryan, L., Kuo, T.L., Wu, M.M., Chen, C.J.,

2000. Risk of internal cancers from arsenic in drinking water.

Environ. Health Perspect. 108, 655–661.

Nickson, R., MacArthur, J., Ravenscroft, P., Burgess, W.,

Ahmed, K., 2000. Mechanism of arsenic release to ground-

water, Bangladesh and West Bengal. Appl. Geochem. 15,

403–413.

O’Day, P.A., 2003. A general biogeochemical model for arsenic

cycling in shallow aquifers. Abstracts of Papers of the Am.

Chem. Soc. 226, U586.

PA DEP, 2004. Drinking water and sampling information system

data (1994–2004). Pennsylvania Department of Environmen-

tal Protection, Harrisburg, PA.

Panno, S.V., Hackley, K.C., Hwang, H.H., Greenberg, S.E.,

Krapac, I.G., Landsberger, S., O’Kelly, D.J., 2006. Source

identification of sodium and chloride in natural waters:

preliminary results. Ground Water 44, 176–187.

Peters, S.C., Blum, J.D., 2003. The source and transport of

arsenic in a bedrock aquifer, New Hampshire, USA. Appl.

Geochem. 18, 1773–1787.

Peters, S.C., Blum, J.D., Karagas, M.R., Chamberlain, C.P.,

Sjostrom, D.J., 2006. Sources and exposure of the New

Hampshire population to arsenic in public and private

drinking water supplies. Chem. Geol. 228, 72–84.

Peters, S.C., Blum, J.D., Klaue, B., Karagas, M.R., 1999. Arsenic

occurrence in New Hampshire drinking water. Environ. Sci.

Technol. 33, 1328–1333.

Pichler, T., Veizer, J., Hall, G.E.M., 1999. Natural input of

arsenic into a coral reef ecosystem by hydrothermal fluids and

its removal by Fe(III) oxyhydroxides. Environ. Sci. Technol.

33, 1373–1378.

Raven, K.P., Jain, A., Loeppert, R.H., 1998. Arsenite and

arsenate adsorption on ferrihydrite: kinetics, equilibrium, and

adsorption envelopes. Environ. Sci. Technol. 32, 344–349.

Savage, K.S., Tingle, T.N., O’Day, P.A., Waychunas, G.A., Bird,

D.K., 2000. Arsenic speciation in pyrite and secondary

weathering phases, Mother Lode Gold District, Tuolomne

County, California. Appl. Geochem. 15, 1219–1244.

Schreiber, M.E., Simo, J.A., Freiberg, P.G., 2000. Stratigraphic

and geochemical controls on naturally occurring arsenic in

groundwater, eastern Wisconsin, USA. Hydrogeol. J. 8, 161–

176.

Serfes, M., 2005. Arsenic occurrence, sources, mobilization,

transport and prediction in the major bedrock aquifers of

the Newark Basin. Ph.D. Thesis, Rutgers Univ.

Serfes, M.E., Spayd, S.E., Herman, G.C., 2005. Arsenic occur-

rence, sources, mobilization, and transport in groundwater in

the Newark Basin of New Jersey, advances in arsenic

research. ACS Symp. Ser. 915, 175–190.

Sherman, D.M., Randall, S.R., 2003. Surface complexation of

arsenic(V) to iron(III) (hydr)oxides: structural mechanism

from ab initio molecular geometries and EXAFS spectros-

copy. Geochim. Cosmochim. Acta 67, 4223–4230.

Sloto, R., Schreffler, C., 1994. Hydrogeology and groundwater

quality of northern Bucks County, Pennsylvania. US Geol.

Surv. Water Resour. Invest. Rep., 94–4109.

Smedley, P.L., Kinniburgh, D.G., 2002. A review of the source,

behavior and distribution of arsenic in natural waters. Appl.

Geochem. 17, 517–568.

Smedley, P.L., Kinniburgh, D.G., Macdonald, D.M.J., Nicolli,

H.B., Barros, A.J., Tullio, J.O., Pearce, J.M., Alonso, M.S.,

2005. Arsenic associations in sediments from the loess aquifer

of La Pampa, Argentina. Appl. Geochem. 20, 989–1016.

Smith, A., Lopipero, P., Chung, J., Haque, R., Hernandez, A.,

Moore, L., Steinmaus, C., 2000. Arsenic in drinking water

and cancer risks estimated from epidemiological studies in

Argentina, Chile, Taiwan and Japan. Epidemiology 11, S93.

Smith, R., Rose, A., Lanning, R., 1975. Geology and geochem-

istry of Triassic diabase in Pennsylvania. Geol. Soc. Am. Bull.

86, 943–955.

Stumm, W., Morgan, J.J., 1996. Aquatic chemistry: chemical

equilibria and rates in natural waters, Environ. Sci. Technol..

Wiley-Interscience, New York.

Thornburg, K., Sahai, N., 2004. Arsenic occurrence, mobility,

and retardation in sandstone and dolomite formations of the

Fox River Valley, eastern Wisconsin. Environ. Sci. Technol.

38, 5087–5094.

USGS, 2001. National water information system: arsenic data

(1973–2001). US Geol. Surv., Reston, VA.

Utsunomiya, S., Peters, S.C., Blum, J.D., Ewing, R.C., 2003.

Nanoscale mineralogy of arsenic in a region of New Hamp-

shire with elevated As-concentrations in the groundwater.

Am. Mineral. 88, 1844–1852.

Welch, A.H., Lico, M.S., 1998. Factors Controlling As and U in

shallow ground water, southern Carson Desert, Nevada.

Appl. Geochem. 13, 529–539.

Welch, A.H., Westjohn, D.B., Helsel, D.R., Wanty, R.B., 2000.

Arsenic in ground water of the United States: occurrence and

geochemistry. Ground Water 38, 589–604.

Wilkie, J.A., Hering, J.G., 1996. Adsorption of arsenic onto

hydrous ferric oxide: effects of adsorbate/adsorbent ratios and

co-occurring solutes. Colloid Surface A 107, 97–110.

Documents

The occurrence and geochemistry of arsenic in groundwaters of …scp2/reprints/Peters_2007_AG_The_occurre… · The occurrence and geochemistry of arsenic in groundwaters of the Newark