vvEPA
                United States
                Environmental Protection
                Agency
d   Water Issue
   Ground Water Sample Preservation at In-Situ Chemical Oxidation
                            Sites - Recommended Guidelines
                                Saebom Ko1, Scott G. Huling2'*, Bruce Pivetz3
             Table of Contents
  1. Introduction                           1
   1.1  Reasons to Sample and Analyze Binary Mixtures 2
   1.2  Binary Mixtures of Oxidant and Organic
      Contaminants in Ground Water Samples       2
   1.3  Impact of Binary Mixtures -Previous Studies    2
  2. Ground Water Sample Collection, Oxidant
    Measurement, and Oxidant Neutralization/Sample
    Preservation                           4
   2.1  Permanganate (MnO4)                  5
   2.1.1 Analysis by Visual Observation            5
   2.1.2 Spectrophotometric Analysis             5
   2.1.3 Results                            6
   2.1.4 Oxidant Neutralization and Sample
       Preservation                        6
   2.2  Persulfate (S2O82 )                      7
   2.2.1 Analysis by Field Test Kit Colorimetry       7
   2.2.2 Analysis by Spectrophotometric Analysis
       (Ferrous Ammonium Sulfate Method)       7
   2.2.3 Results                            7
   2.2.4 Oxidant Neutralization and Sample
       Preservation                        8
  3. Additional Information                    9
  4. References                            9
  5. Acknowledgements                      10
  Appendix - Recommended Operating Procedures -
    Preservation of Ground Water Samples at ISCO
    Sites Using Ascorbic Acid                   11
                   Figures
  Figure 1. Conceptual model of hydrogeologic, and oxidant
       and contaminant fate and transport conditions
       that contribute to binary mixture ground water
       samples.                          3
  Figure 2. MnO4 Calibration Curve.                 6
  Figure 3. Calibration curve for S2O82 (Ferrous Ammonium
       Sulfate Method).                      8
                   Tables
  Table 1. Permanganate concentration, Spectrophotometric
       absorbance, ascorbic acid.                5
  Table 2. Persulfate concentration, Spectrophotometric
       absorbance, ascorbic acid.                8
  Table A1. Corresponding concentration of sodium
       permanganate and potassium permanganate to
       permanganate.                       13
  Table A2. Corresponding concentration of sodium
       persulfate to persulfate (S2O82).             14

1 National Research Council, Robert S. Kerr Environmental
   Research Center, P.O. Box 1198, Ada, OK, 74820; Phone:
   (580) 436-8742; ko.saebomepa.aov
2" (Corresponding Author) U.S. Environmental Protection
   Agency, National Risk Management Research Laboratory,
   Robert S. Kerr Environmental Research Center, P.O. Box
   1198, Ada OK, 74820; Phone: (580) 436-8610; huling.scotm
   epa.gov
3 Shaw Environmental & Infrastructure, Inc., Robert S. Kerr
   Environmental Research Center, P.O. Box 1198, Ada, OK,
   74820; Phone: (580) 436-8998; E-mail: pivetz.bruce@epa.gov
                                           1.  INTRODUCTION
                                           In-situ chemical oxidation (ISCO) involves  the  introduction  of a
                                           chemical oxidant into the subsurface for the purpose of transforming
                                           ground water and/or soil contaminants into less harmful chemical by-
                                           products (Huling and Pivetz, 2006; Rivas, 2006; Ferrarese etal., 2008;
                                           Kao etal., 2008). Often, ground water samples collected specifically to
                                           analyze organic contaminants may contain the oxidant and the organic
                                           contaminants in a "binary mixture" (Huling etal.,  201 la; Johnson et
                                           al., 2012). When organic contaminants and oxidants are commingled
                                           in the ground water sample, there is significant potential for oxidative
                                           transformation of contaminants to occur after the sample is collected
                                           and the results of the sample analysis to become non-representative of
                                           in-situ conditions at the time of sampling. Consequently, the quality
                                           of the ground water sample may be compromised and a false negative
                                           result may occur.

                                           An integral component of ISCO is the collection and analysis  of
                                           ground water samples  to assess ISCO treatment performance.  A
                                           technical issue faced by Remedial Project Managers is the collection
                                           and analysis of representative, high quality ground water samples that
                                           can be used to support a site assessment and remedial performance
                                           monitoring at sites where ISCO is being deployed. The purpose of this
                                           Issue Paper is to provide background information and general guidelines
                                           involving methods and procedures that can be used to detect whether
                                           an oxidant (i.e., permanganate or persulfate) is present in ground water,
                                           to approximate the oxidant concentration, and to estimate and deliver
                                           the volume or mass of preservative, specifically ascorbic acid, required
                                           to preserve the binary mixture ground water sample. The focus of this
                                           Issue Paper is on permanganate and persulfate, two oxidants that can
                                           persist for long periods of time in the subsurface and therefore represent
                                           the greatest potential for  binary mixture ground water samples. An
                                           Appendix to this Issue Paper (Recommended Operating Procedures -
                                           Preservation of Ground Water Samples at ISCO Sites Using Ascorbic
                                           Acid) provides specific details regarding the preservation procedures for
                                           use by EPA Regional personnel, contractors, and other environmental
                                           professionals engaged in ground water sample collection and analysis.

                                           The guidelines are also applicable to bench-scale studies where oxi-
                                           dants are  used to investigate the feasibility  of ISCO treatment. For

-------
example, aqueous samples collected from bench-scale
soil reactors are analyzed for organic contaminants, but
may also contain the oxidant amended to the reactor to
destroy the contaminant. Consequently, the guidelines
described below also extend to bench-scale studies where
the potential for binary mixture aqueous  samples may
occur, and are analyzed for organic contaminants.

7.7. Reasons to Sample and Analyze Binary
     Mixtures
It is often desirable for oxidants in ground water to fully
react prior to  collecting and analyzing ground water
samples for organic contaminants. However, there are
circumstances where the collection and analysis of binary
mixture ground  water samples  may not be avoided.
These reasons vary widely and some examples  include
the need to:
(1)  conduct  an immediate preliminary assessment of
    ISCO to validate in-progress treatment performance,
(2)  establish design parameters from  interim ISCO
    pilot-scale studies needed to design full-scale ISCO
    deployment,
(3)  assess the potential redistribution of the ground water
    contaminant plume as affected by ISCO activities,
    and
(4)  evaluate reaction kinetics during oxidative treatment.
Rapid turnaround of field data and information may be
needed to meet specified milestones and deadlines for
full-scale remedy selection, design, construction, and
implementation. In addition, regulatory-driven goals
and associated  timelines may require rapid completion
of pilot-scale testing and full-scale deployment of ISCO.
Therefore, a significant emphasis may be placed on the
collection of ground water samples at ISCO sites  prior to
complete reaction of the oxidant (Huling et al, 201 la).

7.2. Binary Mixtures of Oxidant and Organic
     Contaminants in Ground Water Samples
Heterogeneous distribution of oxidant and contami-
nants, and hydraulic conductivity variations in  hetero-
geneous aquifers are two main causes of binary mixtures
(Figure 1) (Huling etctl, 201 la). For example, oxidants
and contaminants can enter a monitoring well screen
from different  lithologic zones. These solutes may  be
captured as separate  solutes from different lithologic
zones, or as separate  or commingled solutes from the
     same lithologic zone. Insufficient contact time (i.e., reac-
     tion time) between the oxidant and contaminants prior
     to, or after, entering the well leads to binary mixtures in
     the ground water sample.

     Commingling of organic contaminants and oxidants
     in the ground water sample impacts the quality of the
     ground water sample, but may also impact the analyti-
     cal instruments used to measure the concentration of
     analyte(s) in the ground water sample (Johnson et al.,
     2012). Although rarely reported and  documented,
     the impact of oxidants on analytical instruments is
     exclusively reported for permanganate and predomi-
     nantly involves instrument malfunction resulting from
     MnO2(s)-clogged lines and ports. No information was
     found that documented the impact of hydrogen peroxide
     or persulfate on analytical instruments despite numerous
     studies where binary mixtures were analyzed.

     7.3.  Impact of Binary Mixtures  - Previous
          Studies
     A detailed  study involving  the  impact  of residual
     persulfate on the quality of ground water samples was
     performed (Huling et al., 201 la). A significant  decline
     (49 to 100 percent (%)) in volatile organic compound
     (VOC) concentrations was measured in unpreserved
     binary mixture samples using gas chromatography
     (GC) purge and trap, and GC mass spectroscopy (MS)
     headspace analytical methods. In that study, preservation
     of the binary mixture samples was achieved through the
     addition of ascorbic acid and resulted in 99 to 100%
     VOC average recovery relative to  oxidant-free  control
     samples. Adding high concentrations of ascorbic acid (42
     to 420 millimolar (mM)) to the samples did not interfere
     in the measurement of the VOCs and did not negatively
     impact the analytical instruments. These results indicated
     that if persulfate is present in the sample, and the binary
     sample is not appropriately preserved, the quality of
     the sample will be compromised. A companion study
     involving the impact of permanganate on the quality of
     ground water samples and analytical instruments, and
     the use of ascorbic acid yielded similar results (Johnson
     et al., 2012). The results of these studies (Huling et al.,
     2011 a; Johnson et al., 2012)  serve as the basis  for the
     guidelines provided in this Issue Paper.
        Ground Water Issue
Ground  Water Sample Preservation at ISCO Sites

-------
                         Oxidant Injection Well,
                      Well Point, or Other Oxidant
                           Injection Method
                       Monitoring
                          Well
             HYDRAULIC
             CONDUCTIVITY
                LOW
               VERY
                LOW
                HIGH
              MEDIUM
                LOW
               VERY
               HIGH
               VERY
               LOW
                                                 SANDY CLAY
 CLAY
                                            SAND
  SANDY CLAY
SAND/GRAVEL
            CLAY
 Figure 1. Conceptual model of hydrogeologic, and oxidant and contaminant fate and transport conditions that contribute
         to binary mixture ground water samples. The oxidant illustrated in purple, conceptually represents any oxidant
         (permanganate, persulfate) used for in-situ chemical oxidation (Ruling etal., 201 la).
 The analytical methods used in these studies are com-
 monly used in commercial analytical laboratories. The
 analytes, including benzene, toluene,  xylene (BTX),
 perchloroethylene (PCE), and trichloroethylene (TCE),
 are representative of contaminants commonly found at
 hazardous waste sites. Similarly, empirical results were
 obtained in the analysis of binary mixtures comprised
 of persulfate and pentachlorophenol (PCP) by high
 performance liquid chromatography (HPLC)  where
 significant loss of PCP was measured in unpreserved
 samples  relative to persulfate-free control samples and
 ascorbic acid-preserved samples (data  not  included).
 Currently, we do not have a firm explanation for a viable
 mechanism responsible for persulfate activation and PCP
 oxidation in these samples.

 Overall, results are applicable to a broad set of analytical
 methods, analytes, and site conditions.  It is unclear to
 what extent these results extend to analytical methods
 and contaminants that were not tested in these studies,
 however. Additional specific studies are needed in cases
                  where different analytical  methods and ground water
                  contaminants are involved.

                  Specifically, analysis involved the measurement of
                  (1) BTX, PCE, and TCE using the GC/MS headspace
                  method, and (2) BTX using the  GC purge and trap
                  method (Huling et al, 201 la). The GC/MS headspace
                  method is involved in EPA Method Nos. 8260C and
                  5021 A. The automated headspace GC/MS  method is
                  used to confirm the identity and quantity of purgeable
                  VOCs in water samples  in 40 mL volatile organic
                  analysis (VOA) vials. This  method is used to quantify
                  over sixty VOCs in drinking water, including aromat-
                  ics, haloalkenes, haloalkanes, haloaromatics, and fuel
                  oxygenates. This automated method involves the transfer
                  of an aqueous sub-sample (10 mL) to a sealed headspace
                  vial which is heated from room  temperature to 80
                  degrees Celsius (°C) in 30 minutes. A sample of the
                  headspace gas is then transferred to the capillary column
                  in the GC. After separation on the GC column and
                  introduction into the MS, the VOCs are identified and
Ground Water Sample Preservation at /SCO Sites

-------
quantified using the MS. We propose that contaminant
loss occurs during the heating step of the sub-sample
where residual persulfate is thermally activated resulting
in VOC oxidation.

The automated purge and trap GC (Agilent, Model 6890,
Wilmington, DE) method was used to quantify BTX
in water samples (40 mL VOA vials). This method is
most similar to EPA Methods 602 and 8020, but shares
similarities with several other EPA methods that involve
purge and trap, including: EPA 501, 502.2, 503.1, 524.2,
601, 602,  624, 8010, 8020, 8021, 8240, and 8260. In
this method, a sub-sample (10 mL) is transferred to a
sparge chamber and purged with helium (6 minutes). The
VOCs are transferred to a K VOCARB 3000 Encon trap
and dry purged with helium to remove water vapor. The
VOCs are thermally desorbed and transferred to the GC
column for separation and measurement. Sample transfer
is through a heated  1.9 mmxl.O m Silcosteel  (Restek,
Bellefonte, PA) transfer line coupled directly to  the
analytical column. Following separation on the column,
the presence of VOCs is determined and quantified with
photoionization and flame ionization detectors. It was
proposed that the contaminant loss was due to the helium
sparging step where aerosols  are formed containing
persulfate and are transfered to the VOC granular acti-
vated carbon trap (Huling etal., 201 la).  Subsequently,
during the VOC thermal desorption step where the trap
is heated from room temperature to 260 °C (25 min),
the persulfate residing in the trap is thermally activated
resulting in the oxidation of the VOCs immobilized and
concentrated on the trap. Similarly, highly efficient oxida-
tion of organics immobilized in solid media (i.e., granular
activated carbon) by thermally activated  persulfate has
been demonstrated (Huling etal., 201 Ib).

The impact of residual permanganate was evaluated
in water samples prepared in the  lab using a multi-
component standard, and in ground water  samples
collected at ISCO sites (Johnson et al., 2012). Binary
mixture aqueous samples were prepared that contained
a 52-component standard of organic compounds and
permanganate. Ascorbic acid was added to the binary
mixture which reacted  rapidly with the MnO4 , pre-
served the sample,  and limited the reaction between
MnO4 and the organic compounds. Consequently, the
concentrations of the  majority of the compounds in
     the multi-component standard were within the control
     limits established for quality assurance. However, despite
     timely efforts to preserve the laboratory-prepared binary
     mixture samples, the quality of the sample was impacted;
     concentrations were  generally lower than oxidant-free
     controls, and the concentration of several compounds
     (r»-l,3-dichloropropene, styrene, ?rara-l,2-dichloro-
     ethene, ?rara-l,3-dichloropropene, vinyl  chloride) fell
     below the applicable  lower control limit.

     Concentrations of VOCs measured  in field-preserved
     binary mixture ground water samples were greater than
     in replicate samples refrigerated in the field and preserved
     with ascorbic acid upon arrival at the lab (Johnson etal.,
     2012). These results indicate that the VOCs reacted
     in transit despite refrigeration.  Excess ascorbic acid
     did not negatively impact the quality of the simulated
     ground water samples containing a 52-component stock
     standard, or actual ground water samples collected from
     two field sites, and did not negatively impact the GC/MS
     instruments used in the analysis.

     2.  GROUND WATER SAMPLE COLLECTION,
        OXIDANT MEASUREMENT, AND OXIDANT
        NEUTRALIZATION/SAMPLE PRESERVATION
     Specific details regarding the procedures used in amend-
     ing ground water samples with ascorbic acid are provided
     in the Appendix  entitled,  "Recommended Operating
     Procedures - Preservation of Ground Water Samples  at
     ISCO Sites Using Ascorbic Acid".

     It is recommended that a representative ground water
     sample be collected at the well head in a test vial for the
     specific purpose of measuring the oxidant concentration.
     Ground water sample collection for this purpose should
     follow the normal ground water sampling protocol
     established at the site. This initial screening ground water
     sample is not collected for the purpose  of measuring
     organic contaminant concentrations. If contaminant
     analysis of the ground water sample is desired, additional
     samples must be subsequently collected and preserved,
     if necessary. Normal sampling procedures appropriate
     for site conditions and regulatory acceptance are recom-
     mended. Sample preservation and handling requirements
     are based on the type of analyses being performed and
     should be specified in project-specific documents such
     as the quality assurance project plan, field sampling
        Ground Water Issue
Ground  Water Sample Preservation at ISCO Sites

-------
plan, or in general EPA documents such as the Resource
Conservation and Recovery Act (RCRA) guidance docu-
ment (U.S. EPA, 1992) or EPA SW-846  (U.S. EPA,
1982). Additional direction on ground water sampling
techniques can be found in Yeskis and Zavala (2002).

2.1. Permanganate (MnO4)
Data and information presented below are reported in
terms of the permanganate anion (MnO4; 118.9 grams
per mole (g/mol)).  Permanganate is purchased either
as sodium permanganate (NaMnO4; 141.9 g/mol) or
potassium permanganate (KMnO4; 158.0 g/mol) and as
a result conversion to the permanganate anion concentra-
tion is needed to determine sample preservation needs as
per the Issue Paper.  Specifically, the ratios 118.9/141.9
(g-mole/g-mole) and 118.9/158.0 (g-mole/g-mole) are
used to convert NaMnO4 and KMnO4, respectively to
MnO4.
2.1.1.  Analysis by  Visual Observation
The characteristic pink or purple color of MnO4 in a
40 mL VOA vial can be used as a general guideline to
estimate the concentration by using the MnO4 colori-
metric scale (Table 1). This method should be used with
caution because ground water turbidity and colloidal
manganese dioxide solids (MnO2(s)) can affect sample
color and result in deviations from the tabulated color
scale. Field filtration can help minimize these interfer-
ences, but may not fully remove all color if sub-micron
colloidal and/or dissolved constituents are present.
2.1.2.  Spectrophotometric Analysis
The  permanganate concentration can be determined
using commercially available field test kits (SenSafe M,
2011; CHEMetrics, 2011). Additionally, an accurate
measurement of the permanganate concentrations can be
determined using a field spectrophotometer (maximum
absorbance wavelength (A,) =  525 nanometers (nm)
(A525)) and a calibration curve involving a linear correla-
tion  between MnO4 concentration and A525 (Figure 2,
Table 1). Filtered samples (0.2-0.45 micron)  may be
required to eliminate background colloidal or suspended
solid materials that can absorb light at 525 nm and inter-
fere with permanganate measurement. Volatilization of
 Table 1.  Permanganate concentration, Spectrophotometric absorbance at 525 nm, and required amount of ascorbic acid
         required to neutralize the oxidant in a 40 mL vial. The color scale represents actual photos of MnO4" vials and is
         included for conceptual guidance. Actual colors vary based on background lighting, and color printers. Additionally,
         photographs of low concentrations (i.e., clear solutions) do not accurately capture transparency.
[MnO4~] (mg/L) (millimolar in parentheses)
0
(0)
0.75
(0.01)
3.8
(0.03)
7.5
(0.06)
11.3
(0.09)
18.8
(0.16)
30.1
(0.25)
37.6
(0.32)
56.4
(0.47)
75.3
(0.63)
113
(0.95)
151
(1.27)
188
(1.58)
376
(3.16)
Absorbance(1), wavelength (X) = 525 nm
0
0.011
0.059
0.134
0.197
0.329
0.516
0.627
NL
NL
NL
NL
NL
NL
Ascorbic Acid Stock Solution (M)(2)
-
0.015
0.015
0.15
0.15
0.15
0.15
0.15
1.5
1.5
1.5
1.5
1.5
1.5
Volume of Ascorbic Acid solution (\iL)
0
30
150
30
46
76
121
152
23
30
46
61
76
152
Mass of Ascorbic Acid (mg)
0
0.08
0.4
0.79
1.21
2.1
3.32
4.17
6.1
7.9
12.2
16.1
20.1
40.2
(1) [MnO4~] (mg/L) = 58.8 x A525', A525 is the absorbance at 525 nm; non-linear above 38 mg/L MnO4".
(2) To minimize sample dilution, the ascorbic acid stock solution used was 0.015, 0.15, and 1.5 M.
Ground Water Sample Preservation at /SCO Sites

-------
contaminants is not a concern since the initial screening
ground water sample is used specifically to determine the
concentration of permanganate.
    0.7 -i
                       20      30
                      [Mn04-] (mg/L)
Figure 2. Calibration curve of MnO4~ concentration versus
         absorbance at wavelength (A,) of 525 nm.

2.7.3.  Results
If MnO4 is not detected in the ground water sample,
it is recommended that normal ground water sampling
and analysis procedures be used. If MnO4 is detected,
there  are two general options to  consider.  The first
option is to delay the collection  and analysis  of the
ground water sample for a sufficient time allowing the
MnO4 concentration to fully diminish in the subsurface,
if desired. In some cases, MnO4 persistence  is lengthy
and this option is not possible (as discussed above in
Section 1.1). Due to the site-specific time-dependency
of contaminant mass transfer and transport, the time
required to approach chemical equilibrium in ground
water will likely require additional time after the oxidant
is fully consumed.  Subsequently, ground water sampling
would follow routine guidelines and requirements. The
second option is to collect and preserve the ground water
sample (i.e., neutralize the oxidant) prior to analysis to
minimize the impact of the commingled oxidant. The
second option may be desirable for a number of reasons
described in Section 1.1.
     2.1.4.  Oxidant Neutralization and Sample
            Preservation
     Given the MnO4 concentration, the volume of ascorbic
     acid stock solution (0.015,0.15, or 1.5 mol/L), orweight
     of crystalline ascorbic acid (176.12 g/mol)  required
     to preserve the binary mixture  is determined (Table
     1). Sample preservation involves the addition of the
     appropriate amount of ascorbic acid to preserve a binary
     mixture in a 40 mL VOA vial. In a lab study (Johnson et
     ctl, 2012), the mass of ascorbic acid required to neutralize
     MnO4 ranging in concentration from 1-750 milligrams
     per liter (mg/L) was determined empirically. The aver-
     age molar ratio (n=l4) was 1.64 mol ascorbic acid/mol
     MnO4 and values ranged from 1.45 to 1.75 mol/mol.
     Therefore, the weight of ascorbic acid that corresponded
     with the  MnO4  colorimetric scale was conservatively
     based on a stoichiometric ratio  of 1.8 mol  ascorbic
     acid/mol MnO4, since,  as noted below, no  negative
     side-effects were noted with over-dosing.  Detailed
     recommended operating procedures are provided in the
     Appendix to estimate the volume of crystalline ascorbic
     acid or ascorbic acid stock solution required to neutralize
     the MnO4. Once the oxidant is neutralized, it is recom-
     mended that normal ground water sample handling and
     procedures be followed.

     The recommended volume and  mass of ascorbic acid
     included in Table  1 is  a guideline.  The addition of
     ascorbic acid will rapidly reduce  the MnO4 concentra-
     tion and eliminate the pink/purple color. The formation
     of colloidal or particulate  MnO2(s)  (i.e., Mn+ ) may
     occur causing a brown tinge appearance of the solution.
     Incremental amendment of ascorbic acid is  required
     to further reduce the Mn+  to Mn+2, and eliminate the
     brownish tinge color. Mn+2 is highly soluble and the most
     desirable  form of Mn to minimize the impact of col-
     loidal or particulate matter on the laboratory analytical
     instruments. Overall, Table 1 is used as a guideline but
     the actual amount of ascorbic acid to be added should
     be based on the amount required to fully eliminate the
     MnO4 and MnO2(s), and to achieve a clear solution.

     Excess ascorbic acid did not have a negative impact on
     the quality of the ground water sample involving GC and
     GC/MS analysis  of a broad range of organic chemicals
     (Johnson et al.,  2012). The volume of ascorbic acid
     solution added to the sample vial should be recorded so
         Ground Water Issue
Ground Water Sample Preservation at /SCO Sites

-------
 appropriate dilution calculations can be performed to
 obtain an accurate estimate of the contaminant concen-
 trations. Pre-amending sample vials with ascorbic acid is
 also an option and is discussed further in Section 7.F of
 the Appendix. Other sample preservation requirements
 are based on the analyses being performed and are speci-
 fied in the quality assurance project plan, field sampling
 plan, RCRA guidance document (U.S. EPA, 1992) or
 EPA SW-846  (U.S. EPA, 1982). Additional direction
 on ground water sampling techniques  can be found in
 Yeskis and Zavala (2002)

 2.2.  Persulfate (SzOf)
 The data and information below are presented in terms
 of the persulfate  anion (S2O82~; 192.0 g/mol). However,
 persulfate is predominantly purchased as sodium per-
 sulfate (Na2S2O8; 238.1  g/mol). As a result, conversion
 of sodium persulfate to persulfate anion concentrations
 is necessary to determine sample preservation needs as
 per the Issue Paper. Specifically, the ratio of 192.0/238.1
 (g-mol/g-mol) is used to convert Na2S2O8 to S2O82".
 Persulfate is colorless and requires field measurement at
 the well head to determine its presence and concentration
 in the ground water sample.
 2.2.1.  Analysis  by Field Test Kit Colorimetry
 Field test kits are commercially available to measure per-
 sulfate concentration in aqueous samples (CHEMetrics,
 2011; FMC, 2012). CHEMetrics persulfate test kits
 are available for two sodium persulfate concentration
 ranges (0-7,7-70 mg/L). Given the high concentrations
 of persulfate injected into the subsurface at ISCO  sites,
 significant dilution may be required in the use of these
 test kits. FMC commercial test kits are dependent on
 whether the persulfate activator is base or thermal (test kit
 "K"), or whether persulfate is activated by iron chelates
 or H2O2 (test kit  "C") (FMC, 2012). The lower detection
 limit of persulfate using the current FMC  test kits is
 500 mg/L, a sufficient quantity of oxidant to significantly
 impact the concentrations of VOCs and the quality of
 the sample. Based on the current detection limit using
 the FMC test kit, it is recommended that the minimum
 amount of ascorbic acid added to  the sample vessel
 should conservatively account for 500  mg/L persulfate.
2.2.2.  Analysis by Spectrophotometric Analysis
       (Ferrous Ammonium Sulfate (FAS) Method)
A Spectrophotometric method can be used to analyze the
persulfate concentration in aqueous samples. The ground
water sample should be filtered (0.2-0.45 micron)  to
eliminate background material (i.e., turbidity) that may
interfere with S2O8" analysis. A small volume of de-ion-
ized (DI) water (0.9 mL) and sulfuric acid (H2SO4) (10
mL, 2.5 normal (N)) (or, add 10.9 mL of 2.3 N H2SO4)
is placed in a 20 mL glass or plastic test vessel. These can
be prepared prior to  transport to the field. A blank is
prepared by mixing 1 mL DI water with H2SO4 (10 mL,
2.5 N). The filtered sample (0.1 mL) is placed in the test
vessel,  followed by the addition of ferrous ammonium
sulfate (FAS) (Fe(SO4)2(NH4)2.6H2O)  (0.1 mL, 0.4
N)  (prepared immediately before use). Adding a couple
drops of H2SO4 (cone.) to the FAS reagent increases the
stability of the ferrous iron for several more hours (5  to
10 hours). The mixture is swirled/mixed and allowed to
react for 30 to 40 minutes. Subsequently, the mixture is
amended with ammonium thiocyanate (NH4SCN) (0.2
mL, 0.6 N) and the absorbance of the solution is analyzed
immediately with a spectrophotometer at a wavelength
of A, = 450 nm (A450) (Huang etal., 2002; Huling etctl,
201 la; b). The general colorimetric scale provided below
can be used to estimate the persulfate concentration  in
a ground water sample  (Table 2)  analyzed by the FAS
method. Alternatively, a calibration curve involving a
linear correlation between S2O8" concentration and A450
can be  used to determine a more precise estimate of the
persulfate concentration (Figure 3).
2.2.3.  Results
If S2O8" is not detected in the ground water sample, it
is recommended to proceed using normal ground water
sampling and analysis procedures. If S2O82" is detected,
there are two general options to consider. The first is  to
delay collection and analysis of the ground water sample
for sufficient time which allows the persulfate concentra-
tion to fully diminish in the subsurface,  if desired. Due
to the site-specific time-dependency of contaminant mass
transfer and transport,  the time required to  approach
chemical equilibrium in ground water will likely require
additional time after the oxidant is fully consumed.
Subsequently, ground water sampling would follow
routine guidelines. The  second option is to collect and
Ground Water Sample Preservation at ISCO Sites
               Ground Water Issue

-------
 Table 2.  Persulfate concentrations  resulting from the ferrous ammonium sulfate analytical method involving the
          spectrophotometric measurement (A, = 450 nm) of the solution, and the required amount of ascorbic acid required
          to neutralize the oxidant in a 40 mL vial. The color scale represents actual photos of S2O82" vials and is included for
          conceptual guidance. Actual colors vary based on background lighting, and color printers. Additionally, photographs
          of low concentrations (i.e., clear solutions) do not accurately capture transparency.
                                [S2O8  ] (mg/L) (millimolar in parentheses)
            80
           (0.42
        200
        (1.1)
400
(2.1)
610
(3.2)
810
(4.2)
1210
(6.3)
1610
(8.4)
2020
(10.5)
2420
(12.6)
2820
(14.7)
3230
(16.8)
3630
(18.9)
4030
(21.0)
                                  Absorbance(1), wavelength (1) = 450 nm
          0.011  0.019   0.04    0.062   0.076   0.121   0.164   0.204
                                                                     0.245
                                                                             0.275   0.313
                                                                                            0.349   0.397
                                  Volume of Ascorbic Acid solution (mL)
           0.04    0.11    0.22   0.34   0.45    0.67   0.89    1.12    1.34    1.57    1.79    2.02    2.24
                                 Mass of Ascorbic Acid (176.12 g/mol) (g)
   0
0.01  I 0.03  | 0.06 |  0.09  |  0.12 |  0.18  |  0.24    0.3   |  0.35  |  0.41  |  0.47  |  0.53 |  0.59
   (1) Solubility of ascorbic acid in water = 330 g/L (1.87 mol/L); 80% solubility (1.5 mol/L) used as stock
   solution; [S2O82~] (mg/L) =  10,000 x A45o; where A450 is the absorbance at 450 nm.
 preserve the ground water sample prior to analysis to
 minimize the impact of persulfate on the ground water
 sample. The second option may be desirable for a number
 of reasons described in Section 1.1.
      0.45

       0.4 -

      0.35

       0.3 -

      0.25

       0.2 -

      0.15

       0.1 -

      0.05
         0
    Y = 1.0e-4x; R2 = 0.999
              [S20821 (mg/L) = 10,000 xA450
           0      1000    2000    3000    4000    5000

                       [S2082 ] (mg/L)
 Figure 3.  Calibration curve for S2O8" concentration versus
          absorbance at wavelength 450 nm using the ferrous
          ammonium sulfate method.
                              2.2.4.  Oxidant Neutralization and Sample
                                     Preservation
                              Guidelines for the volume of ascorbic acid stock solution
                              (1.5 mol/L) or the weight of crystalline ascorbic acid
                              (176.1 g/mol) required to preserve the binary mixture in
                              a 40 mL sample vial are provided (Table 2). The mass of
                              ascorbic acid that corresponds with the persulfate colo-
                              rimetric scale is based on a stoichiometric ratio of 4 mol
                              ascorbic acid/mol persulfate and was determined empiri-
                              cally in a laboratory study (Hiding etal.,  2011a). Detailed
                              recommended operating procedures are provided in the
                              Appendix to estimate the volume of crystalline ascorbic
                              acid or ascorbic acid stock solution required to neutralize
                              the S2O82". This stoichiometric ratio is in excess of the
                              ideal stoichiometry for mineralization of persulfate by
                              ascorbic acid. Excess ascorbic acid (4 - 40 mol ascorbic
                              acid/mol persulfate)  did  not have a negative impact on
                              the quality of the  ground water sample involving GC
                              and GC/MS analysis of BTX, TCE, and PCE (Huling
                              et al., 201 la).  The basis for this quantity of ascorbic
                              acid is to achieve  favorable reaction kinetics  between
                              •SO4 and ascorbic  acid, relative to the reaction between
                              the sulfate radical (-SO4~) and the VOCs. Following
                              oxidant neutralization, it is recommended that other
                              approved sample preservation and handling  methods
,
                                                  Ground Water Sample Preservation at /SCO Sites

-------
in ground water sample handling be performed. For
example, acidification of the sample is normally carried
out to minimize biochemical and reduction reactions.
Other sample preservation requirements are based on the
analyses being performed and are specified in the qual-
ity assurance project plan, field sampling plan, RCRA
guidance document (U.S. EPA, 1992) or EPA SW-846
(U.S. EPA, 1982). Additional direction on ground water
sampling techniques can be found in Yeskis and Zavala
(2002).

3.  ADDITIONAL INFORMATION
It is recommended that the analytical laboratory be noti-
fied that the aqueous samples contain residual persulfate
or permanganate and were preserved with ascorbic acid.
The volume of ascorbic acid solution added to the sample
should be recorded so the appropriate calculations can
be used to correct for dilutions. If MnO2(s) has settled
on the bottom of the VOA vial, it is important that the
sample not be disturbed prior to analysis. This precau-
tion in sample handling prevents the  suspension  of
the MnO2(s) particles and the potential for accidental
injection into the analytical instruments.

Other preservatives have been used to  successfully
neutralize  these oxidants, but may negatively impact
the quality of the sample (Huling etal., 201 la). Despite
efforts used to  neutralize the oxidant and to preserve
the quality of the ground water sample, the presence of
oxidant in ground water samples introduces uncertainty
in the precise measurement of contaminant concentra-
tions in the subsurface. This is attributed to the potential
impact of the oxidant on contaminant concentrations
in the ground water sample prior to neutralization, the
transient nature of contaminant fate and transport in the
subsurface where ISCO  activities were deployed, and the
site-specific oxidant injection and hydrogeologic condi-
tions contributing to binary mixtures.  Consequently,
additional ground water sample collection and analysis
will likely be required to achieve an accurate evaluation
of post-ISCO performance, and regulatory adherence
with  US  EPA ground  water compliance monitoring
requirements.

Numerous examples exist where elevated permanganate
and VOC concentrations have been measured in ground
water samples collected  over extended periods of time at
hazardous waste sites. It can be concluded from a simple
kinetic analysis that long term VOC persistence can
primarily be explained by spatial separation between the
ground water containing the oxidant and contaminant
(Figure 1) (Johnson etal., 2012). Ground water samples
derived from wells screened over spatially separate vertical
intervals  indicate an in-well mixture of ground water
containing either oxidants  or contaminants. Limited
contact between the oxidant and contaminant within the
same lithologic unit can be due to specific mass transfer
or mass transport conditions including the dissolution
of non-aqueous phase liquids (NAPLs) or slow diffusion
of contaminants from low permeability materials. These
fate and transport conditions indicate the oxidant has not
been uniformly delivered to the contaminated zone(s).
A critical analysis of screened intervals, injection inter-
vals, contaminated intervals, oxidant and contaminant
transport characteristics, and ground water sample results
from analyzing preserved binary mixtures, could provide
valuable insight for the development of a more accurate
site conceptual model that could be used to  design and
deploy a more effective oxidant delivery system.

4.  REFERENCES
CHEMetrics. 2011. http://www. chemetrics. com/Persulfate.
Ferrarese, E., Andreottola, G., and Oprea,  LA. 2008.
  Remediation of PAH-contaminated  sediments by
  chemical oxidation./. Haz. Mat. 152(1), 128-139.
FMC Environmental Solutions. 2012. Field Measurement
  http://environmental.fmc.com/solutions/soil-g'round-
  remediation/field-measurement.
Huang, K.C., Couttenye, R.A., and Hoag, G.E. 2002.
  Kinetics of heat-assisted persulfate oxidation of methyl
  tert-butylether (MTBE). Chemosphere 49(4),413-420.
Huling, S.G., and Pivetz, B. 2006. In-Situ Chemical
  Oxidation - Engineering Issue. US Environmental
  Protection Agency,  National  Risk  Management
  Research Laboratory,  R.S. Kerr  Environmental
  Research Center, Ada, OK. EPA/600/R-06/072. http://
  www. epa.pov/nrmrl/nuerd/publicatiom. html#oxidation.
Huling, S.G.,  Ko,  S.,  and  Pivetz, B. 201 la. Ground
  water sampling at ISCO  sites - Binary mixtures of
  volatile organic compounds and persulfate. Ground
  Water Monit. Remed. 31(2), Spring 72-79.
Huling, S.G.,  Ko, S., Park, S., and Kan,  E.  201 Ib.
  Persulfate-driven oxidation of contaminant-spent
  granular activated carbon. /.  Haz.  Mat. 192(3),
  1484-1490.
Ground  Water Sample Preservation at ISCO Sites
               Ground Water Issue

-------
 Johnson, K.T., Wickham-St. Germain, M., Ko, S. and   International) for graphics preparation and Ms. Martha
   Hiding, S.G.  2012. Binary Mixtures of Permanganate   Williams (SRA International) for desktop publishing.
   and Chlorinated Volatile Organic Compounds in   A   rtable document format (PDF)  version of this
   Groundwater samples: oample  Preservation and   ,         .     ., , ,  r             ,    ,   ,.   r
   Analysis. GroundWaterMonit. Remed., 32(3), Summer   document is available for viewing or downloading from
   g4_92.                                          http://www.epct.govlnrmrllgwerdlpubliccttiom.html (please
 Kao, C.M.,  Huang, K.D., Wang,  J.Y.,  Chen, T.Y.,   ffer l° "In Situ Chemical Oxidation; "Issue Paper"; or
   and Chien, H.Y. 2008. Application of potassium   "2012").
   permanganate as an oxidant  for  in-situ oxidation
   of tricmoroethylene-contaminated groundwater: A
   laboratory and kinetics study. /.  Haz.  Mat. 153(3),
   919-927.
 Rivas, F.J. 2006. Polycyclic aromatic hydrocarbons sorbed
   on soils: A short review of chemical oxidation based
   treatments./. Haz. Mat. 138(2), 234-251.
 SenSafe,  2011.  http://www.sensafe. com/product.
   php?recordID=481 138.
 U.S. EPA. 1982. Test Methods for Evaluating Solid Waste,
   Physical/Chemical Methods, EPA SW-846. Office of
   Solid Waste and Emergency Response, Washington,
   D.C. http://www.epa.mv/epawaste/hazard/testmethods/
   sw846/online/index. htm.
 U.S. EPA. 1992. RCRA Ground-Water  Monitoring:
   Draft Technical Guidance.  Office of Solid  Waste,
   Washington,  DC EPA/530/R-93/001, NTIS PB 93-
   139350.
 Yeskis, D., andZavala, B. 2002. Ground-Water Sampling
   Guidelines for Superfund and RCRA Project Managers
   - Ground Water Forum Issue Paper. Office of Solid
   Waste and Emergency Response, Washington, D.C.
   EPA 542-S-02-001 May 2002.  http://www.epa.yov/
   tio/tsp/download/?w sampling guide.pdf.

 5. ACKNOWLEDGEMENTS
 This Ground Water Issue Paper was prepared for the
 U.S. Environmental  Protection Agency, Office of
 Research and Development, National Risk Management
 Research Laboratory. The authors were Dr. Saebom Ko
 (National Research Council), Dr. Scott G. Huling (U.S.
 EPA) and Dr. Bruce E. Pivetz (Shaw Environmental &
 Infrastructure Inc.). The authors acknowledge the U.S.
 EPA Ground Water Forum members for their valuable
 input and peer review comments. The authors also wish
 to acknowledge the valuable  input and  peer  review
 comments provided by Dr. Phil Block (FMC Corp.),
 Dr. Daniel Cassidy (Western Michigan  University),
 Dr. Wilson Clayton (TriHydro Corporation), Mr. Tom
 Palaia (CH2M Hill), and Mr. Mike Wireman (U.S. EPA
 Region 8). The authors thank Ms. KathyTynsky (SRA
10       Ground Water Issue
Ground Water Sample Preservation at /SCO Sites

-------
                 Appendix
     Recommended Operating
               Procedures -
   Preservation of Ground Water
    Samples at  ISCO Sites Using
              Ascorbic Acid
                 Table of Contents
  1. Purpose (Scope and Application)                   11
  2. Method Summary                            11
  3. Reagents                                  11
  4. Equipment/Apparatus                          11
  5. Health and Safety Precautions                     11
  6. Interferences                               12
  7. Procedures                                 12
    A. Ascorbic Acid                            12
    B. Sample Filtration                           12
    C. Concentration Measurement                   12
      1) Commercially available test kits                12
        a. Permanganate                         12
        b. Persulfate                            12
      2) UV-VIS absorbance                        12
        a. Permanganate                         12
        b. Persulfate                            12
      3) Colorimetric scales                        12
    D. Quality Assurance and Quality Control (QA/QC)        12
    E. Calculations                              12
      1) Concentration conversion                   12
        a. Permanganate                         12
        b. Persulfate                            13
      2) Required volume and mass of ascorbic acid to
         neutralize oxidants                      13
        a. Permanganate                         13
        b. Persulfate                            14
    F. Pre-amending Sample Vials With Preservative         14
  8. References                                 15
  9. Disclaimer                                 15
 1.  PURPOSE (SCOPE AND APPLICATION)
 The commingling of organic contaminants and oxidants
 in ground water or aqueous samples represents a condi-
 tion in which there is significant potential for oxidative
 transformation of the contaminants after the sample
 is collected. Consequently, the quality of the ground
 water or aqueous sample may be compromised and a
 false negative result may occur. These recommended
 operating procedures describe the steps used to preserve
 ground water samples containing the oxidants  per-
 manganate (MnO4), or persulfate (S2O8~) and organic
 contaminants of concern (COCs) prior to analysis. It is
 applicable for ground water samples containing volatile
 and non-volatile organic contaminants to be analyzed by
gas chromatography (GC), or gas chromatography-mass
spectroscopy (GC-MS), using either the purge and trap
or headspace sample introduction methods, and high
performance liquid chromatography (HPLC).

These procedures are also applicable to bench-scale stud-
ies where oxidants are used to investigate the feasibility of
ISCO treatment. For example, aqueous samples collected
from bench-scale soil reactors are analyzed for organic
contaminants, but may also contain the oxidant amended
to the reactor to destroy the contaminant. Consequently,
the guidelines and general procedures described below
also extend to bench-scale studies where  the potential
for binary mixture aqueous samples may occur, and are
analyzed for organic contaminants.

2. METHOD SUMMARY
Based on the measured or estimated oxidant concentra-
tion in a ground water or aqueous sample, a specific
quantity of the preservative, ascorbic acid, is added to the
ground water or aqueous sample to either neutralize or
to limit the impact of the residual oxidant on the quality
of the sample. Tables 1 and 2 in the Issue Paperare used
as guidelines to estimate the amount of ascorbic acid to
add to  a 40 mL VOA vial to preserve binary mixture
ground water and/or aqueous samples.

3. REAGENTS
Ascorbic Acid (C6H8O6; 176.1 g mol"1)
De-ionized (DI) water
Ferrous  amonium sulfate  (FAS)  reagents - sulfuric
   acid (H2SO4), ferrous ammonium sulfate
   (Fe(SO4)2(NH4)2-6H2O), ammonium thiocyanate
   (NH4SCN).

4. EQUIPMENT/APPARATUS
Pipette,  volumetric flasks, spectrophotometer (or field
   test kits)
SenSafe   or CHEMetrics field test kits for permanganate
   measurement (if used), or direct measurement.
CHEMetrics or FMC field test kits for persulfate measure-
   ment (if used), or measurement using  FAS method.

5. HEALTH AND SAFETY PRECAUTIONS
The Materials Safety  Data Sheet for ascorbic acid indi-
cates potentially acute health effects: slightly hazardous
in case of skin contact (irritant), of eye contact (irritant),
of ingestion, of inhalation. In case of skin  contact: wash
Ground  Water Sample Preservation at ISCO Sites
              Ground Water Issue
11

-------
 with soap and water. Cover the irritated skin with an
 emollient. Get medical attention if irritation develops.
 Cold water may be used. Other guidelines are available
 based on exposure (http://www.sciencelab.coml'msds.
 php?msdsld=9922972). It is recommended to  wear
 gloves and safety glasses during all of the procedures
 described herein due to the potential for exposure to
 oxidants, impacted ground water sample,  and other
 chemicals involved in these procedures. Always consult
 site-specific health and safety plans prior to sampling.

 6. INTERFERENCES
 Colloidal and/or suspended solids in ground water sam-
 ples may adsorb light and interfere with the measurement
 of oxidant concentration. For this reason, the ground
 water sample may require filtration (0.2-0.45 |J.m) to
 eliminate background material (i.e., turbidity).

 7.  PROCEDURES
 A.   Ascorbic Add
 Prepare ascorbic acid stock solution either in the lab
 prior to  ground water  sampling, or in the field. The
 appropriate use of these stock solutions is dependent on
 concentrations of the oxidant measured in the ground
 water samples. The stock solution  should be stored in
 a refrigerator or cooler  until used, and discarded after
 150 days.

 High Concentration Stock Solution: 1.5 M ascorbic acid
 (e.g., add 264 g of ascorbic acid (MW= 176.1 g/mol) to
 1L  volumetric flask and fill with DI water). This stock
 solution can be diluted in the preparation of 0.015 and
 0.15 M ascorbic acid stock solutions.

 Medium Concentration Stock Solution: 0.15 M ascorbic
 acid: Dilute 1.5  M ascorbic acid stock solution  1:10
 (e.g., dilute 100 mL of 1.5 M stock solution to 1L with
 DI  water).

 Low Concentration Stock Solution: 0.015 M ascorbic
 acid: Dilute 1.5 M ascorbic acid stock solution 1:100
 (e.g., dilute 10 mL of 1.5 M stock solution to 1L with
 DI  water).

 B.   Sample Filtration
 Filter the ground water  or aqueous sample using
 0.2-0.45 um filter (as needed in accordance with the
                                           site QAPP or Sampling and Analysis Plan) to eliminate
                                           background material (i.e., turbidity) that may interfere
                                           with oxidant analysis.

                                           C.   Concentration Measurement
                                           Determine the oxidant concentrations (permanganate or
                                           persulfate) through one of three methods below.

                                           1)  Commercially available test kits
                                           a.  Permanganate: SenSafe™ or CHEMetrics
                                           b.  Persulfate: CHEMetrics or FMC
                                                                          measurement):
     2)  UV-VIS absorbance
     a.  Permanganate   (direct
        wavelength = 525 nm
     b.  Persulfate (Ferrous Ammonium Sulfate method):
        wavelength = 450 nm (Huang et al, 2002; Huling
        etal.,20U)

     3)  Colorimetric scales presented in Tables 1 and 2.
     Based on the oxidant concentration determined, ascorbic
     acid stock solution is added to an empty sample vial
     according to Tables 1 and 2.

     D.   Quality Assurance and Quality Control
          (QA/QC)
     Quality control includes regularly scheduled analysis of
     method blanks and sample replicates, and the verification
     of stock solutions of known concentration via the analysis
     for concentrations of secondary solutions prepared from
     the stocks. Results of the  analyses of method blanks,
     replicate analyses, and the verification of stock solution
     concentrations are logged and maintained in  record
     books specific to the research being conducted. The
     frequency, control limits, and corrective actions  should
     be appropriately developed for specific applications.

     E.    Calculations
     1) Concentration conversion
     a.  Permanganate.
     The concentrations of permanganate (MnO4~) have
     been presented  in terms of the permanganate anion
     (118.9  g/mol)  (Table  1).  However, permanganate is
     purchased either as sodium permanganate (NaMnO4;
     141.9 g/mol) or potassium  permanganate (KMnO4;
     158.0 g/mol) and as a  result conversion to permanga-
     nate anion concentrations may be desired to determine
12
Ground Water Issue
Ground  Water Sample Preservation at /SCO Sites

-------
adequate sample preservation needs. Specifically, the
ratios 118.9/141.9 (0.84) and 118.9/158.0 (0.75) are
used to convert NaMnO4 and KMnO4 respectively, to
MnO4 (Table Al).

Because 1 mmole of either sodium or potassium perman-
ganate produces 1 mmole of permanganate (Eqs 1 and
2), the molar concentrations of sodium and potassium
permanganate are the same as permanganate (Table 3).
    NaMnO4 -» Na+ + MnO4

    KMnO, -» K+ + MnO;
(1)

(2)
Converting sodium and potassium permanganate con-
centrations from mg/L to millimolar, and calculating
their permanganate equivalence,

    X mg/L NaMnO4 =
    (Xmg/L) x (1 mmol/141.9 mg) =
    X/l4l.9mMNaMnO4 =
    X/141.9 mM MnO4 X/141.9 mM MnO4 =
    ((X/141.9) mmol/L) x (118.9 mg/mmol) =
    0.84X mg/L MnO4
    NOTE: 1 mmol = 0.001 mol; mM= mmol/L

    Y mg/L KMnO4 =
    (Ymg/L) x (1 mmol/158.0 mg) =
    Y/158.0mMKMnO4 =
    Y/158.0 mM MnO4

    Y/158.0mMMnO4 =
    ((Y/158.0) mmol/L) x (118.9 mg/mmol) =
    0.75Y mg/L MnO4
    NOTE:  1 mmol =  0.001 mol;  mM= mmol/L
b.  Persulfate.
The concentration of persulfate is presented in terms of
the persulfate anion (S2O82~; 192.0 g/mol) (Table A2).
However, persulfate is purchased as sodium persulfate
(Na2S2O8; 238.1 g/mol)  and as a result a conversion
may  be desired to correct for the anionic form of the
oxidant and to determine adequate sample preservation
needs. Specifically, the ratio of 192.0/238.1 (0.81) is
used to convert Na2S2O8 to S2O8 " .  Persulfate is color-
less and requires field measurement  at the well head to
determine its presence and concentration in the ground
water sample.

Converting sodium persulfate concentrations from mg/L
to millimolar, and calculating the persulfate equivalence,
   Na2S2O8 -> 2Na+ +S2O82~

   Z mg/L Na2S2O8 =
   (Z mg/L) x (1 mmole/238.1 mg) =
   Z/238.1 mMNa2S2O8 =
   Z/238.1 mMS2O82-
                                                  (3)
          Z/238.1
          (Z/238.1) mmole/L) x (192 mg/mmole) =
          0.8 1Z mg/L S2O82~
      2)  Required volume and mass of ascorbic acid to neutral-
          ize oxidants.
      a.  Permanganate.
      1.8 mole ascorbic acid per mole of permanganate was
      empirically determined to effectively neutralize perman-
      ganate in an aqueous sample containing VOCs (Johnson
      etal., 2012). Therefore, the mass balance equation (Eq 4)
      can be set up as follows,
                  V-
                    MnO4- -
                               H2A
                                            (4)
      Where,
      CMn04. = permanganate concentration determined in
          step 7.C,
Table Al. Corresponding concentration of sodium permanganate and potassium permanganate to permanganate.
NaMn04
KMn04
Mn04
mg/L
mM
mg/L
mM
mg/L
mM
0.90
0.006
1.00
0.006
0.75
0.006
4.5
0.032
5.0
0.032
3.8
0.032
9.0
0.063
10.0
0.063
7.5
0.063
13.5
0.095
15.0
0.095
11.3
0.095
22.4
0.16
25.0
0.16
18.8
0.16
35.9
0.25
40.0
0.25
30.1
0.25
44.9
0.32
50.0
0.32
37.9
0.32
67.3
0.47
74.9
0.47
56.4
0.47
89.9
0.63
100
0.63
75.3
0.63
135
0.95
150
0.95
113
0.95
180
1.27
201
1.27
151
1.27
224
1.58
250
1.58
188
1.58
449
3.16
500
3.16
376
3.16
Ground  Water Sample Preservation at /SCO Sites
                    Ground Water Issue
                                               13

-------
 Table A2. Corresponding concentration of sodium persulfate to persulfate (S2O8").
Na2S208
SA2-
mg/L
mM
mg/L
mM
99
0.42
80
0.42
248
1.0
200
1.0
496
2.1
400
2.1
756
3.2
610
3.2
1004
4.2
810
4.2
1500
6.3
1210
6.3
1996
8.4
1610
8.4
2504
10.5
2020
10.5
3000
12.6
2420
12.6
3496
14.7
2820
14.7
4004
16.8
3230
16.8
4500
18.9
3630
18.9
4996
21.0
4030
21.0
 VMno4  = volume of permanganate solution in the VOA
     vial (0.04 L),
 CH2A = ascorbic acid concentration (0.015, 0.15 or 1.5
     M), and
 VH2A = volume of ascorbic acid required to neutralize
     permanganate.
 VH2A can be calculated (Eq 5) through rearranging Eq. (4)
     VH2A= (1.8
                                     (5)
 For example, a 40 mL permanganate concentration of
 1.27mM (151 mg/L) is neutralized using 1.5 M ascorbic
 acid. The volume of stock solution and mass of ascorbic
 acid can be calculated as follows.

     VH2A= (1.8 x 1.27 mmol/L x 0.04L/1.5 mol/L) x
      (1 mol/1000 mmol) x (106 ixL/lL) = 61 iiL

     MH2A = 1.5 mol/L x 61 ixL x (1L/106 iiL) x
     (176.12 g/mol) x (1000 mg/g) = 16.1 mg

 Where,
     MH2A = mass of ascorbic acid

 The formation of colloidal or particulate MnO2(s) (i.e.,
 Mn+4) may occur causing a brown tinge appearance of
 the solution. Incremental amendment of ascorbic acid
 may be required to  further reduce the Mn+ to Mn+2,
 and eliminate the brownish tinge color. Mn+  is highly
 soluble and the most desirable form of Mn to minimize
 the impact of colloidal or particulate matter on the
 laboratory  analytical instruments. Overall, Table 1 is
 used as a guideline  but the actual amount should be
 based on the amount required to fully eliminate the
 MnO4 and MnO2(s), and to achieve a clear solution. The
 volume of ascorbic acid solution added to the sample vial
 should be recorded so appropriate dilution calculations
 can be performed to obtain an accurate estimate of the
 contaminant concentrations.
 b.  Persulfate.
 4 mole  of ascorbic acid per mole  of persulfate was
                                           empirically determined to effectively limit the impact
                                           of the oxidant on VOCs in aqueous samples (Huling et
                                           ctl, 2011). Therefore, the mass balance equation (Eq 6)
                                           can be set up as follows,
                                                      V'
                                                       S2O82- ~
                                                                   H2A
                                                 (6)
   Where,
   CS2082- = persulfate concentration determined in step 7.C,
   VS2082 = volume of persulfate solution in the VOA vial
       0.04 L,
   CH2A = ascorbic acid concentration (1.5 M),
   VH2A = volume of ascorbic acid required to neutralize
       persulfate
   VH2A can be calculated (Eq 7) through rearranging Eq. (6)
                                              VH2A=(4xC
                                                           'S2O82-
                                                                  ' S2O82-
                                                                          'H2A
                                                 (7)
                                           For example, persulfate  concentration  is 10.5 mM
                                           (2020 mg/L) and neutralized using 1.5 M ascorbic acid.
                                           The volume of stock solution and mass of ascorbic acid
                                           can be calculated as follows.

                                              VH2A= (4 x 10.5 mmol/L x 0.04L / 1.5 mol/L) x
                                               (1 mol/1000 mmol) x (1000 mL/lL)= 1.12 mL

                                              MH2A= 1.5 mol/L x 1.12 mL x
                                               (1 L/1000 mL) x (176.12 g/mol) = 0.3 g

                                           Where,
                                              MH2A = mass of ascorbic acid

                                           The volume of ascorbic acid solution added to the sample
                                           vial should be recorded so appropriate dilution calcula-
                                           tions can be performed to obtain an accurate estimate
                                           of the contaminant concentrations.

                                           F. Pre-amending Sample Vials With Preservative
                                           Pre-amending  the 40 mL sample vials prior to per-
                                           forming ground water sample collection in the field is
                                           one step that may help  simplify sample preservation
                                           procedures. The advantage  is that all sample vials are
14
Ground Water Issue
Ground  Water Sample Preservation at /SCO Sites

-------
amended with the preservative in a uniform manner,
and this reduces the number of steps and time required
during ground water sampling activities in the field.
Specifically, this  would involve amending the sample
vial with an appropriate quantity of ascorbic acid using
the procedures recommended above. Successful sample
preservation would be immediately obvious in the case
with permanganate binary mixtures as the pink/purple
color would disappear and the sample would become
clear. A persistent pink/purple or brown tinge color
would indicate the need for additional  preservative.
The immediate visual feedback would not occur in the
preservation of persulfate binary mixtures due to the
absence of oxidant coloration. Success of the preservation
method will most likely require prior knowledge of oxi-
dant concentrations in ground water samples to support
the selection of an appropriate quantity of preservative.
A quality assurance step could include the collection of
duplicate samples, and subsequent analysis for persulfate,
when time permits, to confirm that a sufficient quantity
of preservative was amended. Other appropriate quality
assurance steps could be developed.

8.  REFERENCES
Huang, K.C., Couttenye, R.A., and Hoag, G.E. 2002.
  Kinetics of heat-assisted persulfate oxidation of methyl
  tert-butyl ether, Chemosphere49(4), 413-420.
Huling, S.G., Ko, S., and Pivetz, B. 2011. Ground water
  sampling at ISCO sites - binary mixtures of volatile
  organic compounds and persulfate. Ground Water
  Monit. Remed.  31(2), Spring 72-79.
Johnson, K.T., Wickham-St. Germain, M., Ko, S. and
  Huling, S.G. 2012. Binary Mixtures of Permanganate
  and Chlorinated Volatile Organic Compounds  in
  Groundwater  Samples: Sample Preservation and
  Analysis. GroundWaterMonit. Remed., 32(3), Summer
  84-92.

9.  DISCLAIMER
This recommended operating procedure has been pre-
pared for general use. This is not an official approved U.S.
Environmental Protection Agency method and has not
undergone  the Agency's peer review process.
Ground  Water Sample Preservation at ISCO Sites
Ground Water Issue
15

-------
United States
Environmental Protection
Agency
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand corner.

If you do not wish to receive these reports CHECK HEREd;
detach, or copy this cover, and return to the address in the
upper left-hand corner.
PRESORTED STANDARD
 POSTAGE & FEES PAID
          EPA
    PERMIT No. G-35
National Risk Management
   Research Laboratory
Cincinnati, OH 45268

Official Business
Penalty for Private Use
$300

EPA/600/R-12/049
August 2012
                                                                            Printed on 100% recycled/recyclable paper
                                                                            with a minimum 50% post-consumer
                                                                            fiber using vegetable-based ink.

-------