>// f
           United States
           Environmental Protection
           Agency
Office of Water
(4303)
                                                        l0057
EPA-821-R-95-036
April 1995
•8-EPA    Report of the Method 1664
           Validation Studies

-------
vvEPA  ReP°rt °f the Method 1664 Validation
       Studies

-------
                                                    Report of the Method 1664 Validation Studies
Acknowledgments

This report was prepared under the direction of William A. Telliard of the Engineering and
Analysis Division within the EPA Office of Water. This document was prepared under EPA
Contract No. 68-C3-0337 by DynCorp Environmental. Other contributors to this study
included Interface, Inc., Commonwealth Technology, Global Environmental Services, Inc.,
ETS Analytical Services, and members of the Twin Cities Round Robin (TCRR) Group,
including Bay West, Inc., Ecolab, Inc., Interpoll Laboratories, Inc., Koch Refining Co., Land
O'Lakes, Inc., Legend Technical Services, Metropolitan Waste Control Commission, Minnesota
Department of Health, Minnesota Valley Testing Laboratories, Montgomery Watson, Serco
Laboratories, and Spectrum Laboratories, Inc.
Disclaimer

This report has been reviewed and approved for publication by the Engineering and Analysis
Division, U.S. Environmental Protection Agency.  Mention of company names, trade names
or commercial products does not constitute endorsement or recommendation for use.
Questions or comments regarding this report should be addressed to:

W. A. Telliard
USEPA Office of Water
Analytical Methods Staff
Mail Code 4303
401 M Street, SW
Washington, D.C. 20460
202/260-7120
Requests for additional copies should be directed to:

US EPA NCEPI
11029 Kenwood Road
Cincinnati, OH 45242
513-489-8190
NOTE:  This printing of the Report of the Method 1664 Validation Studies contains an
Addendum to Section 3, dated January 1996.  This version of the report supersedes any
previous publication of the Report of the Method 1664 Validation Studies.
                                                                        April 1995

-------
Report of the Method 1664 Validation Studies
                                                                TABLE OF CONTENTS
Section 1 Background/Introduction  	       1

Section 2 Twin Cities Round Robin Study  	       2
    2.1   Study Design	       2
    2.2   Data Validation  	       3
    2.3   Results	       3
    2.4   Discussion and Conclusions	       6

Section 3 Development of Quality Control Acceptance Criteria	       7
    3.1   Data Validation and Statistical Analysis	       7
    3.2   Results	       9
    3.3   Discussion and Conclusions	      10

Section 4 Method Detection Limit Studies	      17
    4.1   Study Design	      17
    4.2   Implementation  	      17
    4.3   Results	      19
    4.4   Discussion	      21
Appendix A  Phase II Draft Study Plan

Appendix B TCRR Study Analytical Requirements

Appendix C  Study Plans for the Method 1664 MDL Studies

Appendix D  Analytical Requirements for the Method 1664 MDL Studies

Appendix E 40 CFR  136, Appendix B

Appendix F References

Addendum to Section 3 Development of Quality Control Acceptance Criteria,
January 1996	   AA-1
    3.1   Data Validation and Statistical Analysis	   AA-2
    3.2   Results	   AA-3
    3.3   Discussion  and Conclusions	   AA-5
April 1995                               iv

-------
                                                  Report of the Method 1664 Validation Studies
                                                                   List of Tables
Table 1


Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10

Table 11

Table 12

Table 13


Table 14


Table 15


Addendum Table 2

Addendum Table 3

Addendum Table 4

Addendum Table 5
Twin Cities Round Robin Group Intel-laboratory Study of Method 1664
HEM (mg/L) in Field Samples

95 Percent Confidence Limits for Method 1664 QC Criteria

HEM IPR Data

HEM IPR Data - Upper and Lower Limits for Recovery

HEM IPR Data - Upper Limit for Precision

HEM OPR Data

HEM OPR Data - Upper and Lower Limits for Recovery

SGT-HEM IPR Data

SGT-HEM IPR Data - Upper and Lower Limits for Recovery

SGT-HEM IPR Data - Upper Limit for Precision

SGT-HEM OPR Data

SGT-HEM OPR Data - Upper  and Lower Limits for Recovery

Twin Cities Round Robin Group Interlaboratory Study of Method 1664
Mean RPDs of the HEM Field Sample Results

Results and Calculated MDLs and MLs for Seven Replicate Analyses for
HEM

Results and Calculated MDLs and MLs for Seven Replicate Analyses for
SGT-HEM

95 Percent Confidence Limits for Method 1664 QC Criteria

HEM IPR Data

HEM IPR Data - Upper and Lower Limits for Recovery

HEM IPR Data - Upper Limit for Precision
                                                                    April 1995

-------
Report of the Method 1664 Validation Studies
Addendum Table 6   HEM OPR Data




Addendum Table 7   HEM OPR Data - Upper and Lower Limits for Recovery




Addendum Table 8   SGT-HEM IPR Data




Addendum Table 9   SGT-HEM IPR Data - Upper and Lower Limits for Recovery




Addendum Table 10  SGT-HEM IPR Data - Upper Limit for Precision




Addendum Table 11  SGT-HEM OPR Data




Addendum Table 12  SGT-HEM OPR Data - Upper and Lower Limits for Recovery




Addendum Table 13B      Comparison of Method 1664 Criteria
April 1995                              vi

-------
                                                     Report of the Method 1664 Validation Studies
                                                                            SECTION 1
                                                         BACKGROUND/INTRODUCTION
   EPA Method 1664 is a gravimetric procedure applicable to aqueous matrices for the
determination of n-hexane extractable material and silica gel treated n-hexane extractable
material (oil and grease and total petroleum hydrocarbons, respectively).  Method 1664
resulted from a multiphase study conducted by the Environmental Protection Agency (EPA)
to determine a suitable replacement solvent for Freon-113, a Class I chlorofluorocarbon (CFC)
used in several EPA wastewater and solid waste methods for the determination of oil and
grease and petroleum hydrocarbons. Method 1664 is being proposed to replace two
currently approved 40 CFR part 136 methods for the determination of oil and grease.  These
methods are EPA Method 413.1, published in Methods for Chemical Analysis of Water and
Wastes, and Method 5520B, published in Standard Methods for the Examination of Water and
Wastewater, 18th edition.

   Development of Method 1664 was the  final objective of Phase II of the Freon Replacement
Study, the results and details of which are provided in the Report of EPA Efforts  to Replace
Freon for the Determination of Oil and Grease and Total Petroleum Hydrocarbons: Phase II. Method
1664 reflects not only  the information gathered under Phase II of the Freon Replacement
Study, but  also consideration of numerous peer review comments, survey results, data from
industry studies, results from an interlaboratory method validation study, and results from
several EPA single-laboratory method detection limit (MDL) studies.

   The quality control (QC) requirements incorporated into Method 1664 exceed and
improve upon that of the currently approved 40  CFR part 136 oil and grease methods, and
are consistent with 40 CFR 136, Appendix A methods for determination of organic analytes.
The QC acceptance criteria incorporated in the initial draft of Method 1664 (March 29,  1994)
reflected the data quality objectives  (DQOs) that were established at the beginning of Phase II
of the Freon Replacement Study. These  objectives are outlined in the Draft Study Plan for
Phase II of the Freon Replacement Study, September 29, 1993 (Appendix A).  A final goal of the
Phase II study was the validation and/or revision of the estimated DQOs and subsequent
development of final QC acceptance criteria (specifications)  for the updated version of
Method 1664.  These final specifications  were derived from the QC analyses conducted as
part of Phase II of the Freon Replacement Study and from the results of an interlaboratory
study conducted by 11 laboratories belonging to the Twin Cities Round Robin Group.  In
addition, five studies were performed to determine the method detection limit (MDL) and the
minimum level of quantitation (ML).

   This report gives details of the studies and associated results that were used to derive the
QC acceptance  criteria in the latest version of Method 1664 (April, 1995).  This report is
divided into three main sections: Section 2 presents the field sample results of the Twin Cities
Round Robin Group interlaboratory study of Method 1664; Section 3 details the procedures
used to develop the QC specifications in Method 1664; Section 4 provides the results and
discussion  of the five MDL studies.
                                                                         April 1995

-------
Report of the Method 1664 Validation Studies
                                                                            SECTION 2
                                                    TWIN CITIES ROUND ROBIN STUDY
    This section of this report provides the details of an interlaboratory study conducted by
the Twin Cities Round Robin (TCRR) Group for the determination of HEM by Method 1664.
The TCRR Group consists of a number of commercial, state, and municipal laboratories in the
Minneapolis/St. Paul region that volunteer their services to evaluate new procedures.  The
objectives of this study were to determine if the method instructions were clear and
comprehensive, and to assess whether or not the estimated QC acceptance criteria in the
March 29, 1994 draft version of Method 1664 could be achieved in a variety of laboratories.

    In addition, the results of the round-robin study were used to statistically derive QC
specifications to replace the estimated QC specifications in the draft version of Method 1664.
Field sample and QC analysis results from a total of eleven laboratories were submitted for
consideration under this study. Presentation and  evaluation of the QC results are included in
Section 3 of this report.  Results and assessment of the field sample analyses are provided
below.
2.1     Study Design

2.1.1   Analytical Study Design

    The TCRR study consisted of two parts: 1) performance of an initial precision and
recovery (IPR) test requiring the analysis of four spiked reagent water samples to
demonstrate the laboratory's ability to generate acceptable precision and accuracy, and 2)
analysis of two sets of field samples, one from a petroleum source and the other from a non-
petroleum source, in triplicate, for  HEM. Laboratories were required to submit the four
replicate IPR results for evaluation prior to analysis of the field samples.  Laboratories were
encouraged to perform analysis for silica gel treated-hexane extractable material (SGT-HEM)
as well, although this analysis was optional.

    In addition, analysis of a reagent water method blank with both the IPR set and the field
samples was required. An ongoing precision and recovery (OPR) analysis, the equivalent of
a single IPR analysis, was performed with the field sample analyses. Hexadecane and stearic
acid standards from a single source were provided to the laboratories for convenience.
Details of the analytical protocol and reporting requirements are provided in Appendix B.

2.1.2   Sample Source Selection

    In order to avoid the problems associated with the comparison and evaluation of non-
detect results, candidate field samples were screened by an independent laboratory to ensure
that the HEM concentrations were  within a range of 40 - 300 mg/L.  The screening was
performed  with Method 1664.  The two samples used in this study included an oily
wastewater from a shore reception facility (petroleum source) and a primary effluent from an
olive packaging plant (non-petroleum source).
April 1995

-------
                                                      Report of the Method 1664 Validation Studies
    At each site, a large-volume sample was collected in a clean polyethylene barrel either by
peristaltic pump or by hand with a clean polyethylene beaker.  Transfer of material was
minimized in order to prevent loss of extractable material during the collection process.  To
assure homogeneity, wastewater in the barrel was mixed by stirring with a polyethylene
paddle while sample was siphoned from the center of the barrel directly into individual
sample containers.  Sample containers were unused, pre-cleaned, 1-liter, wide-mouth, clear
glass bottles with fluoropolymer-lined caps.

    Each individual 1-L sample aliquot was preserved on-site with HC1 (1:1)  to a pH < 2.
Sample bottles were cooled with wet ice prior to shipment to the laboratories.  Prior to
collection, each sample bottle was labeled with a unique EPA sample number and identifying
information, including source  location, collection date, and preservatives used.  Pre-assigned
sample numbers were the primary means of identifying and tracking samples.  Traffic
Reports, which provide information on the samples collected and serve as a sample tracking
sheet, were completed at each site, and accompanied each shipment to the receiving
laboratory. Copies of these reports were used by field and Sample Control Center1 personnel
for tracking purposes.
2.2    Data Validation

    Data generated from this study were submitted to the Sample Control Center for review
and validation. All results received by August 20, 1994 were considered and are presented in
this report.  Final results, including all calculations, were verified from the raw data and
bench sheets provided.  All quality control requirements in the study plan and draft version
of Method 1664 were met, and all data reviewed were determined to be of acceptable quality.

2.3    Results

    Eleven laboratories submitted results for HEM determination.  Only one laboratory
performed the optional SGT-HEM determination; therefore, SGT-HEM results are not
included in this report.  For each laboratory, the HEM triplicate measurements were
averaged.  The standard deviation of the triplicate measurements and the relative standard
deviation (RSD) were then calculated, providing an indication of the intralaboratory
variability and precision.

    For each sample, the average HEM concentration across all laboratories was calculated,
along with the standard deviation of the HEM concentrations.  In addition, the average RSD
and the standard deviation of the RSDs across laboratories were calculated to estimate the
interlaboratory variability and precision. These calculations, along with the analytical results,
are presented by sample in Table 1.
       SCC, operated by DynCorp Environmental under contract to EPA's Engineering and Analysis Division.

                                         3                                 April 1995

-------
Report of the Method 1664 Validation Studies
Twin Cities
Table 1
Round Robin Group Inter laboratory Study of Method 1664
HEM (mg/L) in Field Samples
Sample=25101 Source=Petroleum
Lab
1
2
3
4
5
6
7
8
9
10
11


Rep 1
70.2
57.8
63.0
47.0
56.7
61.8
50.6
63.4
63.7
49.0
50.4


Rep 2
69.0
52.9
61.0
47.0
41.8
46.5
49.9
62.8
98.3
41.0
57.2
Mean HEM (m
Std. Deviation

Rep 3
71.5
53.8
69.0
30.0
47.8
53.4
47.5
63.6
96.3
65.0
28.4
g/L)
of HEM
Mean RSD
Triplicate
Mean
70.2
54.8
64.3
41.3
48.8
53.9
49.3
63.3
86.1
51.7
45.3
57.2
12.9
Across Labs
Triplicate
Std.
Deviation
1.3
2.6
4.2
9.8
7.5
7.7
1.6
0.4
19.4
12.2
15.1


Std. Deviation of RSDs
RSD (%)
1.8
4.8
6.5
23.7
15.4
14.2
3.3
0.7
22.6
23.7
33.2


13.6
11.0
April 1995

-------
                                                      Report of the Method 1664 Validation Studies
Twin Cities
Table 1 (Cont.)
Round Robin Group Inter laboratory Study of Method 1664
HEM (mg/L) in Field Samples
Sample=25104 Source=Non-petroleum
Lab
1
2
3
4
5
6
7
8
9
10
11
Rep 1
182.0
135.2
182.0
150.0
166.3
144.0
173.0
194.1
158.5
157.0
160.0
Rep 2
171.0
134.8
163.0
225.0
180.9
167.0
*
187.4
174.1
167.0
162.0
Rep 3
194.0
136.9
181.0
184.0
178.2
143.0
173.0
186.0
181.9
167.0
162.0
Mean HEM (mg/L)
Std. Deviation of HEM
Triplicate
Mean
182.3
135.6
175.3
186.3
175.1
151.3
173.0
189.2
171.5
163.7
161.3
169.5
15.8
Triplicate
Std.
Deviation
11.5
1.1
10.7
37.6
7.8
13.6
0.0
4.3
11.9
5.8
1.2

Mean RSD Across Labs



Mean RSD Across

Std.
Deviation of
Std. Deviation of
Labs for Samples
RSDs for Samples
RSDs
25101 and 25104
25101 and 25104

Combined
Combined
RSD (%)
6.3
0.8
6.1
20.2
4.4
9.0
0.0
2.3
6.9
3.5
0.7


5.5
5.7
9.5
9.5
Due to a laboratory accident, this aliquot of the sample was not analyzed.
                                                                           April 1995

-------
Report of the Method 1664 Validation Studies
2.4    Discussion and Conclusions

    Initially, a small number of laboratories encountered some difficulties with the analysis of
IPR samples but, after adjusting and improving their technique, were able to achieve
acceptable recoveries of hexadecane and stearic acid in both the IPR and OPR analyses.
Statistical evaluation of the results produced few outliers, indicating that Method 1664 is a
reproducible procedure sufficiently reliable to be used by a variety of laboratories. In
addition, the mean relative standard deviation across all laboratories for the petroleum
sample was 13.6 percent, and for the non-petroleum sample was 5.5 percent, resulting in a
combined mean RSD of 9.5 percent.  These results demonstrate that Method 1664 is capable
of producing precise results on real world samples.

    Since the TCRR Group was the first to perform  analyses according to Draft Method  1664,
input from the laboratories regarding inconsistencies, unclear instructions, or suggestions for
improvement in the procedure were encouraged.  An initial comment related to the
ambiguity surrounding the grade of n-hexane that should be used.  This comment was
addressed, and more specific criteria were included in the updated versions of Method 1664.

    Most other comments focused  on difficulties related to extracts containing excessive
amounts of water and the longer time required for the  evaporation of n-hexane from the
extracted material.  The former was addressed in  revised versions of Method  1664 by
recommending more careful separation of the aqueous  and solvent phases to avoid carryover
of the water into the extract and that more sodium sulfate be used in the filtering process.
The latter was addressed in revised versions by allowing the use of either a water bath or
steam bath set at a temperature that results in evaporation of the solvent within 30 minutes.
April 1995

-------
                                                     Report of the Method 1664 Validation Studies
                                                                            SECTION 3
                        DEVELOPMENT OF QUALITY CONTROL ACCEPTANCE CRITERIA
    This section of this report presents details of the statistical analyses used to develop the
QC acceptance criteria in Method 1664. As previously mentioned, EPA used data gathered
from two studies for the development of these criteria.  These two studies were the Twin
Cities Round Robin (TCRR) interlaboratory method validation study and Phase II of EPA's
Freon Replacement Study.

    Study data that were used to develop the QC acceptance criteria that are  in the  April
1995 version of the method consisted of results from initial precision and recovery (IPR) and
ongoing precision and recovery (OPR) tests. In the IPR test, four 1-L reagent water samples
are spiked with and analyzed for the analytes of interest (in this case hexadecane and stearic
acid), and the precision and recovery are measured; in the OPR test, a single  reagent water
aliquot is spiked, and the recovery is measured.

    An advantage of using data from both studies was that the specifications ultimately
developed for Method 1664 were based on results from both a breadth and depth of
laboratory analyses.  Use of data from the TCRR study provided EPA with a single set of IPR
and OPR results from each of 11 laboratories; use of data from the Phase II study provided
EPA with results from 30 OPR analyses performed by a single laboratory, along with IPR
results from that laboratory. The basic design of the Phase II study and the TCRR study are
described in detail in the respective study plan or  analytical requirements summary
(Appendices A and B). Evaluation and discussion of the Phase II field sample results is
provided in the Report of EPA Efforts to Replace Freon for the Determination of Oil and  Grease and
Total Petroleum Hydrocarbons: Phase II.  Evaluation of the TCRR study field sample results are
given in Section  2 of this report.

3.1    Data Validation and Statistical Analysis

3.1.1   Screening for  Outlier Laboratories

    Results from each of the twelve laboratories (11 from the TCRR study and one from the
Phase II study) were compared to results from all  laboratories to determine if any
laboratories were systematically biased.  First, the  median percent recovery of the results
from all laboratories was calculated.  Then the percent recovery values of each of the four
IPR samples were ranked by laboratory according to their deviation from the median percent
recovery.  The ranks  for each laboratory were summed across the four IPR samples, and the
summed ranks were  analyzed by an extreme rank sum test.  If all laboratories were
equivalent, then  theoretically the ranks of each laboratory relative to the other laboratories
would be random, and the summed ranks would be equal.  The extreme rank sum  test
determines if the summed rank of any laboratory is significantly different from those of the
other laboratories.  This test is based on the work of Youden in Ranking Laboratories by Round
Robin Tests (1963).
                                                                         April 1995

-------
Report of the Method 1664 Validation Studies
3.1.2   QC Acceptance Criteria for Hexane Extractable Material

    Criteria were developed for IPR, OPR, and matrix spike/matrix spike duplicate
(MS/MSD) analyses for both HEM and SGT-HEM. For HEM, the IPR and OPR acceptance
criteria were constructed using an analysis of variance (ANOVA) with laboratories as a
random variance component. For each QC criterion,  the mean result across laboratories was
determined, and the variability was calculated from the interlaboratory and intralaboratory
variance components using a formula appropriate to  the particular test.  The 95 percent
cutoff value from the Student's t distribution was determined based on the appropriate
degrees of freedom. The degrees of freedom  are dependent on the number of laboratories,
the number of sample analyses, and the variance components.  The QC limit was derived by
multiplying the t value by the variability, and subtracting this value from the mean result for
the lower  limit, or adding this value to the mean result for the upper limit.  Details of the
equations  used to derive these limits are presented in the document titled Interlaboratory
Validation  of U.S. Environmental Protection Agency Method 1625A, July 1984.

3.1.3   QC Acceptance Criteria for Silica Gel Treated  Hexane Extractable Material

    For SGT-HEM, EPA received results from only two laboratories, one from the TCRR
study and one from the Phase II study.  EPA  used these data to construct preliminary IPR
and OPR acceptance criteria for SGT-HEM and widened these preliminary criteria to those of
HEM in those instances in which the calculated SGT-HEM criteria were more stringent than
those for HEM.  The acceptance criteria were  widened based on the knowledge that the
determination of SGT-HEM follows the determination of HEM in Method 1664, and therefore
the result  for SGT-HEM is likely to be at least as variable as the result for HEM.

3.1.4   QC Acceptance Criteria for Matrix Spike and Matrix Spike Duplicate

    For HEM and SGT-HEM, the  criteria for recovery of a matrix spike (MS) or matrix spike
duplicate  (MSD) and for the relative percent difference between an MS and an MSD were
derived from the OPR criteria, since neither the TCRR study nor the Phase II study required
the spiking of field samples. EPA believes that this application of OPR criteria to MS/MSD
samples is acceptable because the determinative technique in Method 1664 is gravimetry,
which is not susceptible to interferences, and  because nearly all of the treated effluents to
which Method 1664 is to be applied in monitoring  will be similar to the reagent water used
in the OPR tests.  This transfer of data for development of specifications for acceptance
criteria is similar to that which EPA used in the organic  methods that are promulgated at 40
CFR 136, Appendix A.

    Determination of the relative percent difference (RPD) criteria consisted of setting the
limit at approximately one-half of the range between  the lower recovery limit and upper
recovery limit.  This estimation establishes the RPD limit at 18 percent for HEM and 24
percent for SGT-HEM.  These limits are  considered to be a reasonable first approximation of
method performance in the absence of MS/MSD data. In other methods, EPA  has
established the default limit at 20 percent.  If  MS/MSD data become available,  EPA may
revise this limit.
April 1995

-------
                                                      Report of the Method 1664 Validation Studies
3.2    Results

    Few laboratories in the interlaboratory study encountered difficulties with the analysis of
IPR and OPR samples, and most achieved acceptable recoveries of hexadecane and stearic
acid.  Statistical evaluation of the results from all twelve laboratories produced few outliers,
and the extreme rank sum test showed all laboratories to be equivalent. This indicates that
Method 1664 is a reproducible  procedure sufficiently reliable to be used by a variety of
laboratories.

Results are summarized in Table 2 below. Individual laboratory results and the statistical
analyses of these data  are presented in Tables 3 through 12.
                                         Table 2
                95 Percent Confidence Limits for Method 1664 QC Criteria*
Criterion
IPR
HEM Precision
HEM Recovery
SGT-HEM Precision
SGT-HEM Recovery
MS/MSD
HEM Recovery
HEM RPD
SGT-HEM Recovery
SGT-HEM RPD
OPR
HEM Recovery
SGT-HEM Recovery
95% Lower Limit (%)


83.0

83.2

79.0

65.8


79.0
65.8
95% Upper Limit (%)

10.9
100.7
13.3
116.0

113.9
17.5
105.7**
24.0

113.9
105.7**
*The values for the acceptance criteria in Method 1664 were derived by rounding these
values to the nearest whole number.

**These calculated values were widened to the limits generated from HEM, as explained in
Section 3.1.3. This change  also applies to the MS/MSD RPD calculation.
                                                                          April 1995

-------
Report of the Method 1664 Validation Studies
3.3    Discussion and Conclusions

    The QC acceptance criteria were evaluated in several ways.  First, the precision of the
HEM results from the triplicate analyses of field samples in the TCRR study were compared
to the QC acceptance criteria for RPD.  Since there were three replicates, and RPD is
determined using duplicates, RPDs were calculated for each of the three pair-wise groupings
of the three replicates.  These were then averaged to give the mean RPD for each laboratory.
In addition, the mean RPDs of all laboratories were averaged across each sample and across
both samples.  The mean RPD found was 12.3 percent, which is well within the HEM RPD
acceptance criterion of 18 percent.  These  RPD values, along with the mean and standard
deviation of the three RPDs for each laboratory, are shown  in Table 13.

    Second, the RSD of results for the two field  samples in the TCRR study was computed.
As is explained in Section 2 of this report, the TCRR group  analyzed two sets of field
samples, one from a petroleum source and the other from a non-petroleum source, in
triplicate, for HEM.  The mean RSD of the results across all laboratories and all samples was
9.5 percent (see Table 1), further demonstrating that Method 1664 and the laboratories used
in the study are capable of producing precise results on real world  samples.

    Third, IPR and OPR results from the Phase II study and the TCRR study were compared
with the respective QC acceptance criteria.  For the IPR test of HEM, all of the mean percent
recovery values were within the IPR recovery criterion and  only one of the 12 IPR sets failed
the IPR precision criterion. For the OPR test, none failed the OPR recovery criterion.  For the
IPR test of SGT-HEM, both laboratories met the recovery criterion, but failed the precision
criterion.  These higher standard deviations can be attributed to one of the four IPR results
for each of the labs. For the OPR test, only two of the numerous data points did not meet
the recovery criterion.

    From the results presented above, it can be concluded that nearly all laboratories are
capable of performing Method 1664 successfully.  This  demonstrates that the QC acceptance
criteria are realistic and reflect the performance of Method 1664 on  both reagent water and
on real world sample  matrices.
April 1995                                10

-------
                                                          Report of the Method 1664 Validation Studies

Lab
1
2
3
4
5
6
7
8
9
10
11
12


% Rec.
IPR 1
98.8
78.7
87.5
90.3
93.9
86.8
82.3
91.0
85.0
87.5
89.5
107.5


% Rec.
IPR 2
103.8
103.5
92.5
87.5
97.9
92.0
83.8
103.5
93.3
77.5
89.3
110.0
Std. Deviation of
Table 3
HEM IPR Data
% Rec.
IPR 3
96.3
105.7
85.0
89.0
82.2
92.5
81.5
94.5
89.3
87.5
92.3
95.0
Mean % Recovery
Mean % Recoveries

% Rec.
IPR 4
98.3
83.2
87.5
89.3
90.1
93.8
86.5
95.3
102.0
82.5
89.0
90.0
Across Labs
Across Labs
Mean Std. Deviation
Std. Deviation of the Std. Deviations

Mean
% Rec.
99.3
92.8
88.1
89.0
91.0
91.3
83.5
96.1
92.4
83.8
90.0
100.6
91.5
5.3
Across Labs
Across Labs

Standard
Deviation
3.2
13.8*
3.1
1.2
6.7
3.1
2.2
5.3
7.2
4.8
1.5
9.7


5.2
3.7
                                             Table 4
                       HEM IPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
83.0
95% Upper
Limit
100.7
No. of Mean % Rec.
Values Below 95% Lower
Limit
0
No. of Mean % Rec.
Values Above 95% Upper
Limit
0
                                             Table 5
                             HEM IPR Data - Upper Limit for Precision
95% Upper Limit
10.9
No. of Standard Deviation
Values Above 95% Upper
Limit
1
Result that failed the precision specification.
                                             11
April 1995

-------
Report of the Method 1664 Validation Studies
Table 6
HEM OPR
Lab %
1
2
3
4
5
6
7
8
9
10
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
11
Mean % Recovery Across Labs
Standard Deviation of % Recoveries Across
Data
Recovery
83.0
80.8
82.5
87.8
91.0
98.8
89.8
89.8
82.5
91.0
99.5
91.5
91.5
91.5
91.5
91.5
91.0
91.0
91.0
91.0
91.0
91.5
91.0
112.0
112.0
93.8
93.8
93.3
93.3
101.8
101.8
101.8
112.0
112.0
112.0
89.8
89.8
111.8
111.8
94.0
95.2
9.1
                                      Labs
April 1995                                     12

-------
                                                          Report of the Method 1664 Validation Studies
                                             Table 7
                        HEM OPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
79.0
95% Upper
Limit
113.9
No. of % Rec. Values
Below 95% Lower
Limit
0
No. of % Rec. Values
Above 95% Upper
Limit
0
                                             Table 8
                                        SGT-HEM IPR Data
       Lab

          1

          2
 % Rec.
 IPR 1
% Rec.
 IPR 2
% Rec.
 IPR 3
% Rec.
 IPR 4
  110.0        115.0          80.0        115.0

  105.0        100.0          95.0         65.0

                  Mean % Recovery Across Labs

Std. Deviation of Mean % Recoveries Across Labs
Mean
% Rec.

 105.0

  91.3
                                                                      98.2

                                                                       9.7
                                                 Mean Std. Deviation Across Labs

                                  Std. Deviation of the Std. Deviations Across Labs
Standard
Deviation

  16.8*

  18.0*
                                                                   17.4

                                                                    0.8
                                             Table 9
                    SGT-HEM IPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
83.2
95% Upper
Limit
116.0
No. of Mean % Rec.
Values Below 95% Lower
Limit
0
No. of Mean % Rec.
Values Above 95% Upper
Limit
0
                                             Table 10
                          SGT-HEM IPR Data - Upper Limit for Precision
95% Upper Limit
13.3
No. of Standard Deviation
Values Above 95% Upper
Limit
2
Results that failed the precision specification.
                                             13
                                                             April 1995

-------
       Report of the Method 1664 Validation Studies
                                                   Table 11
                                            SGT-HEM OPR Data

                                            Lab        % Recovery
                                               1               88.0
                                               1               82.0
                                               1               82.0
                                               1               82.0
                                               1               82.0
                                               1               82.0
                                               1               76.0
                                               1               76.0
                                               1               76.0
                                               1               76.0
                                               1               76.0
                                               1               82.0
                                               1               76.0
                                               1               75.5
                                               1               75.5
                                               1              100.0
                                               1              100.0
                                               1              107.5*
                                               1              107.5*
                                               1               80.0
                                               1               80.0
                                               1               80.0
                                               1               75.5
                                               1               75.5
                                               1               75.5
                                               1               79.0
                                               1               79.0
                                               1               96.0
                                               1               96.0
                                               1              100.0
                   Mean % Recovery Across Labs               84.0
        Standard Deviation  of % Recoveries Across               10.2
                                            Labs
                                                   Table 12
                          SGT-HEM OPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
65.8
95% Upper
Limit
105.7
No. of % Rec. Values
Below 95% Lower
Limit
0
No. of % Rec. Values
Above 95% Upper
Limit
2
* Results that failed the percent recovery specifications.
       April 1995
14

-------
                                                               Report of the Method 1664 Validation Studies
Table 13
Twin Cities Round Robin Group Interlaboratory Study of Method 1664
Mean RPDs of the HEM Field Sample Results
Sample=25101 Source=Petroleum

Lab Rep 1
1 70.2
2 57.8
3 63.0
4 47.0
5 56.7
6 61.8
7 50.6
8 63.4
9 63.7
10 49.0
11 50.4


Rep 2
69.0
52.9
61.0
47.0
41.8
46.5
49.9
62.8
98.3
41.0
57.2


Rep 3
71.5
53.8
69.0
30.0
47.8
53.4
47.5
63.6
96.3
65.0
28.4


RPD 12
1.7
8.9
3.2
0.0
30.3
28.3
1.4
1.0
42.7
17.8
12.6
Standard Deviation

RPD 23
3.6
1.7
12.3
44.2
13.4
13.8
4.9
1.3
2.1
45.3
67.3
Mean RPD
of Mean RPDs

RPD 1-3
1.8
7.2
9.1
44.2
17.0
14.6
6.3
0.3
40.7
28.1
55.8
Across Labs
Across Labs

Mean RPD
2.4
5.9
8.2
29.4
20.2
18.9
4.2
0.9
28.5
30.4
45.2
17.6
14.5
RPD Std.
Deviation
1.0
3.7
4.6
25.5
8.9
8.1
2.5
0.5
22.9
13.9
28.8

15                                     April 1995

-------
Report of the Method 1664 Validation Studies
Table 13 (Cont.)
Twin Cities Round Robin Group Interlaboratory Study of Method 1664
Mean RPDs of the HEM Field Sample Results
Sample=25 104 Source=Non-petroleum

Lab Rep 1
1 182.0
2 135.2
3 182.0
4 150.0
5 166.3
6 144.0
7 173.0
8 194.1
9 158.5
10 157.0
11 160.0

Rep 2
171.0
134.8
163.0
225.0
180.9
167.0
*
187.4
174.1
167.0
162.0

Rep 3
194.0
136.9
181.0
184.0
178.2
143.0
173.0
186.0
181.9
167.0
162.0

RPD 12
6.2
0.3
11.0
40.0
8.4
14.8
*
3.5
9.4
6.2
1.2

RPD 23
12.6
1.5
10.5
20.0
1.5
15.5
*
0.7
4.4
0.0
0.0

RPD 1-3
6.4
1.2
0.6
20.4
6.9
0.7
0.0
4.3
13.7
6.2
1.2
Mean RPD across Labs
Standard Deviation




Mean RPD Across Labs
Std. Deviation of RPDs
for Samples
for Samples
of Mean RPDs Across Labs
25101 and 25104
25101 and 25104
Combined
Combined

Mean RPD
8.4
1.0
7.4
26.8
5.6
10.3
0.0
2.8
9.2
4.1
0.8
7.0
7.5
12.3
12.5
RPD Std.
Deviation
3.6
0.7
5.9
11.4
3.6
8.3
*
1.8
4.7
3.6
0.7



         Due to a laboratory accident, this aliquot of the sample was not analyzed, and the associated statistical analyses could not be performed.
       April 1995                                    16

-------
                                                     Report of the Method 1664 Validation Studies
                                                                            SECTION 4
                                                   METHOD DETECTION LIMIT STUDIES
   This section presents the results of five MDL studies for Method 1664. These studies
were performed by two laboratories that used reagent water as the reference matrix.  Global
Environmental, Inc. performed MDL Studies 1,4, and 5, and ETS Analytical Services
performed MDL Studies 2 and 3.  The purposes of these studies were to 1) determine and
confirm the MDLs for HEM and SGT-HEM from the analysis of seven reagent water samples
spiked with hexadecane and stearic acid using Method  1664 and 2) use the MDL results to
establish minimum levels (MLs) for both analytes.

   The MDL is defined as the minimum concentration  of a substance that can be measured
and reported with 99 percent confidence that the analyte concentration is greater than zero.
To determine the  MDL, the laboratories were required to follow the procedure in 40  CFR 136,
Appendix B.  This procedure consists of analyzing seven aliquots of reagent water that are
spiked with the analyte (s) of interest. The MDL is calculated by multiplying the standard
deviation of the seven replicate analyses by the Student's t value for (n - 1) degrees of
freedom, where n equals the number of replicates. The  Student's t value for seven replicates
is 3.143.

       The ML is defined as the level at which the entire analytical system produces a
recognizable signal and an acceptable calibration point,  and is determined by multiplying the
MDL by 3.18 and rounding the resulting value to the number  nearest to (1, 2, or 5) x 10n,
where n is an integer.  The  value "3.18" represents the ratio of the  10 times multiplier used in
the American Chemical Society (ACS) Limit of Quantitation (LOQ) calculations to the
Student's t multiplier of 3.143 that is used to determine  the MDL (i.e., 10 ^ 3.143 = 3.18).  For
example, if the calculated MDL is 5.8, the ML will be equal to 5.8 x 3.18 = 18.4. Using (1, 2,
or 5) x 10n, the number nearest to 18.4 establishes the ML at 20.

4.1     Study Design

       The design of the MDL studies is described in detail in two HEM and SGT-HEM
study plans (Appendix C).  The study plan dated June 23,  1994 applies to MDL Studies 1 and
2; the study plan dated November 3,  1994 applies to MDL  Studies 3, 4, and 5.  The analytical
requirements for each study were modified slightly to reflect revisions of the estimated MDL
and corresponding adjustments to the suggested spike levels and blank contamination limits.
Examples of these analytical requirements are provided in  Appendix D. All studies were
conducted in accordance with 40 CFR 136, Appendix B,  "Definition and Procedure for the
Determination of the Method Detection Limit", Revision 1.11 (Appendix E).

4.2    Implementation

4.2.1   Sample Matrix

       Because EPA regulates more than 600 sub-categories of wastewater discharge, it was
not possible to determine the MDL for each discharge type. EPA considers treated effluents
from well-designed, well-operated best available technology (BAT) treatment systems to be
                                       17                               April 1995

-------
Report of the Method 1664 Validation Studies
similar in nature to reagent water and, therefore, reagent water was judged to be a suitable
sample matrix for the MDL studies.  Reagent water is a homogeneous reference matrix
available to all laboratories and is the primary matrix for determining the MDL using the
procedure in 40 CFR 136, Appendix B. Further, because the determination of HEM and SGT-
HEM are total bulk measurements, matrix interferences are unlikely.

4.2.2  Sample Preparation and Spiking

      For each MDL study, the laboratories were required to prepare and analyze seven
separate aliquots of spiked reagent water. The preparation of a single stock sample and
subsequent distribution into seven separate containers was not allowed due to the propensity
for the analyte to adhere to container walls,  which would prevent equal distribution and
precise allocation of the spiked analytes into each aliquot.  Separate containers for each
sample ensured that an exact, known amount of HEM and SGT-HEM was present in each
aliquot.

      MDL Studies  1, 2, and 3 were performed to determine MDL/ML values for both
HEM and SGT-HEM, and therefore entailed two separate sets of analyses.  In these three
studies, the SGT-HEM samples had a final concentration of the combined amounts of hexa-
decane and stearic acid that were at least double the final concentration in the HEM samples.
This was necessary in order to compensate for the adsorption of stearic acid by silica gel.
MDL Studies 4 and 5 were performed to determine the MDL/ML values for HEM only. The
spike concentrations in the sample aliquots for the HEM and SGT-HEM analyses in each
study are provided below.

MDL Studies 1 and 2

   Samples were prepared by spiking each  one-liter sample with the hexadecane/stearic acid
spiking solution to produce a final concentration of 10 mg/L (5 mg hexadecane and  5 mg
stearic acid) for the HEM MDL study and a  final concentration of 20 mg/L (10 mg
hexadecane and 10 mg stearic acid) for the SGT-HEM MDL study. As recommended in the
MDL procedure in 40 CFR 136, the spiking levels chosen represent a target analyte
concentration that is one to five times the then estimated MDL of 3 mg/L for both HEM and
SGT-HEM.

MDL Study 3

   The final concentration of the hexadecane/stearic acid  solution in the sample aliquots
analyzed in Study 1 was more than five times the MDL values determined for HEM and
SGT-HEM.  To investigate the possible bias that may have been introduced by
disproportionately high spike levels, samples in Study 3 were prepared by spiking each one-
liter  sample with the hexadecane/stearic acid solution to produce a final concentration of 4
mg/L (2  mg hexadecane and 2 mg stearic acid) for the HEM MDL study and a final
concentration of 10 mg/L (5 mg hexadecane and 5 mg stearic acid) for the SGT-HEM MDL
study.

MDL Studies 4 and 5

   Because SGT-HEM results produced in Study 1 were verified by another laboratory in
Study 3, Studies 4 and 5 were conducted for HEM only. Samples for Study 4 were prepared
by spiking each one-liter sample with the hexadecane/stearic acid spiking solution to

April  1995                                18

-------
                                                     Report of the Method 1664 Validation Studies
produce a final concentration of 5 mg/L (2.5 mg of hexadecane and 2.5 mg of stearic acid).
For Study 5, the laboratory chose to spike each one-liter sample to produce a final concentra-
tion of 2.5 mg/L (1.25 mg of hexadecane and 1.25 mg of stearic acid).

       With each MDL study, the laboratories were also required to analyze a reagent water
blank.  The reagent water blanks were run through the entire extraction and analysis
procedure by which the MDL  samples were run.  The HEM or SGT-HEM concentration in
each blank was required to be less than the estimated MDL.

4.2.3   Method Detection Limit Calculations

       Using the data obtained from the analysis of the seven replicate aliquots, the MDLs
for  HEM and SGT-HEM were determined in accordance with the calculations found in 40
CFR 136, Appendix B.  The standard deviations of the replicate results were multiplied by
the  Student's t value, which is 3.143 at the 99% confidence level for six  (n -  1) degrees of
freedom, to obtain the MDLs.

4.2.4   Minimum Level Calculations

       Minimum levels were calculated upon receipt of the data. Exact interim MLs were
determined by multiplying the MDLs by 3.18,  and final MLs were determined by rounding
the  interim ML to the number nearest to  (1, 2 or 5) x 10n, where n is an integer. The 3.18
factor is the quotient of the 10 times multiplier used in the American Chemical Society's
Limit of Quantitation (LOQ) and the Student's t value used in the MDL study calculations
(3.143 for seven replicates).  The exact interim  MLs are equivalent to the 10 standard
deviations used to establish the LOQ, and the  rounded final MLs are close to this value.

4.2.5   Data Verification and Validation

       All data generated from this study were submitted to SCC for review and validation.
Hardcopy data, including all calculations, were verified from the raw data and bench sheets
provided. All QC specifications were met, and all data were determined to be of acceptable
quality.

4.3     Results

4.3.1   HEM Analysis

       Results of the five HEM MDL studies are provided below in Table 14. This table
presents the values for the individual analyses of the seven replicate sample aliquots, the
mean concentration of each set of replicates, the standard deviation of the seven results, and
the  calculated MDLs and MLs for each study.  Results are in mg/L.
                                        19                                April 1995

-------
Report of the Method 1664 Validation Studies
                                        Table 14
      Results and Calculated MDLs and MLs for Seven Replicate Analyses for HEM
                                     (Units = mg/L*)

Spiked Concentration
Replicate Analysis 1
Replicate Analysis 2
Replicate Analysis 3
Replicate Analysis 4
Replicate Analysis 5
Replicate Analysis 6
Replicate Analysis 7
Mean (X)
Standard Deviation (s)
Student's t value
MDL
ML
Study 1
10
10.1
9.8
10.1
10.2
10.1
10.2
9.4
9.99
0.29
3.143
0.91
2
Study 2
10
9.1
9.3
6.0
9.9
6.1
8.0
10.1
8.36
1.7
3.143
5.4
20
Study 3
4
5.2
4.0
3.8
4.1
5.4
5.7
4.3
4.64
0.77
3.143
2.4
10
Study 4
5
4.6
4.4
5.1
4.9
5.0
5.1
5.1
4.89
0.28
3.143
0.88
2
Study 5
2.5
2.7
2.8
2.1
1.6
1.8
1.9
2.0
2.13
0.45
3.143
1.4
5
 *Standard Deviation (s) and Student's t value excluded.

4.3.2   SGT-HEM Analysis

       Results of the three SGT-HEM MDL studies are provided below in Table 15. This
table presents the values for the individual analyses of the seven replicate sample aliquots,
the mean concentration of each set of replicates, the standard deviation of the seven results,
and the calculated MDLs and MLs for each study. Results are in mg/L.
April 1995
20

-------
                                                   Report of the Method 1664 Validation Studies
                                      Table 15
   Results and Calculated MDLs and MLs for Seven Replicate Analyses for SGT-HEM
                                    (Units = mg/L*)

Spiked Concentration
Replicate Analysis 1
Replicate Analysis 2
Replicate Analysis 3
Replicate Analysis 4
Replicate Analysis 5
Replicate Analysis 6
Replicate Analysis 7
Mean (X)
Standard Deviation (s)
Student's t value
MDL
ML
Study 1
20
9.3
10.1
9.8
9.6
10.3
9.6
8.8
9.64
0.50
3.143
1.6
5
Study 2
20
6.7
4.7
6.7
5.7
5.6
5.4
6.9
5.96
0.82
3.143
2.6
10
Study 3
10
5.5
5.7
4.7
4.2
5.5
5.1
4.7
5.06
0.55
3.143
1.7
5
             *Standard Deviation (s) and Student's t value excluded.

4.4    Discussion

       The first MDL study yielded an MDL of 0.91 mg/L and a resultant ML of 2 mg/L for
HEM and an MDL of 1.6 mg/L and a resultant ML of 5 mg/L for SGT-HEM. These
MDL/ML results were well below those expected when using the lower limit of the ranges in
EPA Method 413.1 as a guideline.  The MDLs produced in this study were also less than one-
fifth the spike levels used to conduct the study. Finally, though the study was conducted in
a commercial laboratory, it was performed by a Ph.D. level chemist with more than 20 years'
experience in the determination of oil and grease and TPH. Based on these factors, a second
MDL study was conducted in another commercial laboratory to verify the values obtained in
the first study.

       The second MDL study was also performed by a commercial laboratory experienced
in the determination of oil and grease and TPH, though the analysts performing the study
were not at the  Ph.D. level.  In order to expedite  the proposal of Method  1664, the laboratory
was required to perform the second MDL study within 24 hours.  An MDL of 5.4 mg/L and
an ML of 20 mg/L for HEM, and an MDL of 2.6  mg/L and an ML of 10 mg/L for SGT-HEM
were determined in the second MDL study.
                                      21
April 1995

-------
Report of the Method 1664 Validation Studies
       The laboratory that performed MDL Study 2 noted that the results produced were the
best that could be obtained under the imposed 24 hour turnaround time constraint, and that
they believed they could achieve lower MDLs given more time.  As a result, this laboratory
performed another MDL study (MDL Study 3) using the same turnaround constraints
applied in Study 1, with the analytical objective of confirming the MDLs/MLs that had been
obtained in the first MDL study. An MDL of 2.4 mg/L and an associated ML of 10 mg/L for
HEM, and an MDL of 1.7 mg/L and an associated ML of 5 mg/L for SGT-HEM were
obtained from this third MDL study.  Although closer to the MDL and ML  for HEM obtained
in the first MDL study, the Study 3 ML  of 10 mg/L for HEM was still above the lower limit
of the concentration range in EPA Method 413.1, and  the result for SGT-HEM, the more
complex procedure,  was still less than the result for HEM.

       From these results, EPA concluded that the MDLs/MLs for HEM and SGT-HEM pro-
duced in MDL Study 1  were self-consistent, whereas the results produced in MDL Studies 2
and 3 were not. Because the MDL Study 2 results were determined to have been compro-
mised by the short timeframe required, results from MDL Study 3 were compared to those
produced in MDL Study 1.  The SGT-HEM results produced in MDL Study 3 supported the
MDL values produced for SGT-HEM in  Study  1. Results for HEM, however,  were less
consistent.  EPA concluded that the difference in the HEM results might be explained by the
elevated spike levels used in Study 1. (While the spike levels used in Study 3 were within 1
- 5 times the calculated MDLs, spike levels used in Study 1 exceeded this objective.)

       Consequently, the laboratory that performed Study 1 was asked to perform a new
MDL study (MDL Study 4), for HEM only, using lower spike concentrations.  As with Study
1, the same Ph.D.  level  chemist with extensive  analytical experience performed the analyses.
Despite the lower spike levels used, the  results obtained in Study 4 were  consistent with
those produced by the same chemist in Study 1 (an MDL of 0.88 mg/L, with  a resulting ML
of 2 mg/L).

       In response to comments received from laboratories and other interested parties
regarding the difficulties encountered when attempting to achieve the HEM MDL of 0.91
mg/L specified in the October 1994 and January 1995 versions of Method 1664, and because
most technicians performing HEM analysis for commercial laboratories will not have the
experience or qualifications of the Ph.D. level chemist who performed MDL Studies 1 and 4,
an analyst with a bachelor's degree and  one month's laboratory experience performed
another HEM MDL study at this laboratory.  The results of MDL Study 5 were an HEM
MDL of 1.4 mg/L and a resulting ML of 5 mg/L.

       EPA has concluded that the MDL appropriate  for Method 1664 should be
representative of a better performing laboratory. However, to realistically address the
qualifications of the laboratory personnel most likely to perform this procedure, the MDL
should reflect the  results obtained when using  qualified,  but not Ph.D. level, personnel.
Therefore, the HEM MDL specified in the April 1995 version of Method 1664  (the version
being proposed) is 1.4 mg/L and the HEM ML is 5 mg/L.  This ML is consistent with the
low end of the range given in EPA Method 413.1. Unchanged from the January 1995 version
of Method 1664, the SGT-HEM MDL is 1.6 mg/L and the SGT-HEM ML is  5 mg/L.
April 1995                              22

-------
             APPENDIX A
PHASE II DRAFT STUDY PLAN

-------
Report of the Method 1664 Validation Studies
                              USEPA OFFICE OF WATER

         STUDY PLAN FOR PHASE II OF THE FREON REPLACEMENT STUDY


1.  INTRODUCTION

The discharge of chlorofluorocarbons (CFCs) has been shown to be a primary contributor to
the depletion of the earth's stratospheric ozone layer.  The United States, as a party to the
Montreal Protocol on Substances that Deplete the Ozone Layer and as required by law under
the Clean Air Act Amendments of 1990 (CAAA), is committed to controlling and eventually
phasing out CFCs. Under both the Montreal Protocol and the CAAA, Class I CFCs will be
phased out by January 1, 1996.

Freon 113 is a Class I CFC that is required for use in several U.S. Environmental Protection
Agency (EPA) wastewater and solid waste methods for the determination of oil and grease
and petroleum hydrocarbons. These analytes are included in various regulatory compliance
monitoring programs and are therefore measured on  a continuous basis. As part of the  effort
to  eliminate the use of CFCs, EPA is studying the use of alternate solvents that would
produce results nearly identical to results produced with Freon 113 for these analytes.

Initial efforts to find a solvent alternative  to Freon 113 were conducted by the Office of
Research and Development's Environmental Monitoring Systems Laboratory in Cincinnati,
Ohio (EMSL-Ci).  EMSL-Ci focused its study on Method 413.1 (promulgated at 40 CFR Part
136), which is  used in Clean Water Act (CWA) programs to gravimetrically determine the  oil
and grease  content of surface and saline waters and domestic and industrial wastes.
Aqueous samples, most of which were synthetically prepared by spiking reagent water with
various oils and greases, were analyzed using several different extraction solvents in place of
Freon 113.  Results of the study, presented in the document titled A Study to Select a Suitable
Replacement Solvent for Freon 113 in the Gravimetric Determination of Oil and Grease, by F.K.
Kawahara, October 2,  1991, suggested the use of an 80/20 mixture of n-hexane and methyl
tertiary butyl ether (MTBE) in place of Freon 113 for oil and grease determination. Following
this study, an  Office of Air and Radiation (OAR) proposal (56 FR 30519) suggested
replacement of Freon 113 by the n-hexane:MTBE mix in CWA and RCRA analytical  methods
for determination of oil and grease.

Based on comments submitted concerning the EMSL-Ci study results, and the  need to further
investigate alternative solvents,  the Office of Water and the Office of Solid Waste initiated  a
multi-phase Freon Replacement Study. The objective of Phase I was to evaluate alternative
solvents and extraction systems for equivalency across a range of real world effluent and
solid waste samples from a variety of  industrial categories. This phase of the study focused
on 1) the use of five alternative solvents for gravimetric determination of oil and grease  in
aqueous samples  by MCAWW Method 413.1 (with modifications) and in solid samples by
SW-846 Method 9071A (with modifications) and 2)  the use of alternative techniques  for oil
September 29, 1993                        A-2

-------
                                                     Report of the Method 1664 Validation Studies
and grease analysis including sonication extraction, solid phase extraction (SPE) using
cartridges and disks, and a solvent/non-dispersive infrared technique.

The results of Phase I yielded the following conclusions:  n-hexane should be retained as a
possible extraction solvent for further study using gravimetric techniques; perchloroethylene
should be retained for consideration in the use of infra-red techniques; and cyclohexane
should be introduced for consideration with gravimetric techniques based on its similarity to
n-hexane and because of its lower neurotoxicity when compared to n-hexane.  Results of the
alternative techniques indicated that only sonication extraction produced results equivalent to
existing techniques  that use Freon 113. Specifics of the study design, results, and conclusions
can be found in the Preliminary Report of EPA Efforts to Replace Freon for the Determination of
Oil and Grease, September 1993.

In accordance with  the conclusions of Phase I of the Freon Replacement Study, Phase II will
focus on the evaluation  of n-hexane and cyclohexane as extraction solvents in the gravimetric
determination of oil and grease in aqueous samples by Method 413.1 (with modifications).  In
addition, determination of petroleum hydrocarbons shall be performed on the extracted oil
and grease samples. Of secondary importance is the continuing evaluation of alternative
techniques, which will be performed by manufacturers on splits of samples collected as part
of Phase II.  This  study  plan outlines  the continuing efforts to determine an appropriate
replacement solvent for Freon 113, in order to develop a revised method for oil and grease
and petroleum  hydrocarbons.
2.  OBJECTIVES

The objectives of this study are to further assess the use of n-hexane as a replacement solvent
and, as a result of concerns over the possible neurotoxic effects of hexane, to formally
evaluate the use of cyclohexane as a replacement solvent for oil and grease determination in
aqueous samples by Method 413.1.

In addition, the gravimetric determination of petroleum hydrocarbons by Standard Methods
5520F coupled with Method 413.1 will be evaluated using Freon 113, n-hexane, and
cyclohexane as extraction solvents in order to determine an appropriate replacement solvent
for this analysis. The gravimetric determination of petroleum hydrocarbons is being
incorporated  into the Freon Replacement Study for several reasons. These include 1) the
increasing occurrence of this analyte for regulation in NPDES permits, which has resulted in
the need to address the same issues that are affecting oil and grease analysis; 2) the claim
that petroleum hydrocarbons measurements should replace oil and grease measurements for
certain industrial categories (such as Industrial Laundries) because detergents are extracted
with oil and grease, thereby misrepresenting the oil and grease measurement; and 3) the
apparent need to focus on the gravimetric procedure due to concerns about the susceptibility
of infra-red measurement techniques to  matrix interferences.  These interferences  are found in
matrices associated with certain industrial categories (such as Industrial Laundries) and may
bias the results.

                                       A-3                        September 29, 1993

-------
Report of the Method 1664 Validation Studies
EPA will also supply additional volume of each sample collected under this study to a
number of vendors who are interested in testing alternative oil and grease extraction
techniques similar to those addressed in Phase I of the Freon Replacement Study. These
include sonication extraction, solid phase extraction using cartridges and disks, and a non-
dispersive infrared technique.

As discussed in the Phase I report, analysis of oil and grease by infra-red techniques
(different than the non-dispersive technique cited in the previous paragraph) is another
analytical protocol that warrants further study for replacement solvents. This investigation
will not be incorporated into Phase II of the Freon Replacement Study, but will be addressed
in the future as funding and time allow.

The final objective of Phase II is to utilize the results of Phase II to identify a replacement
solvent for Freon 113 for the gravimetric analysis of both oil and grease and petroleum
hydrocarbons. Once this is determined, a draft method will be written, and an
interlaboratory study may be conducted using the replacement solvent and draft method to
generate method specifications.
3.  SOURCE/TYPE OF SAMPLE

Approximately 38 samples from both in-process and effluent waste streams are scheduled for
collection at over 30 facilities encompassing 15-20 different industrial categories. Samples
containing between 40-300 mg/L oil and grease, some from petroleum and some from non-
petroleum sources, will be targeted for collection. The study will focus on this  concentration
range to avoid the usability problems associated with the comparison and evaluation of non-
detect results. In order to increase the  types of matrices considered by the Agency, many of
the industrial categories sampled as part of Phase II will be different from those collected
during Phase I of the study.

4.  ANALYTICAL STUDY DESIGN

       4.1     Management:  The study will be managed by the Office of Water's
              Engineering and Analysis Division  through the Analytical Methods Staff
              (AMS).  Day-to-day management and coordination of study activities will be
              provided by the contractor-operated Sample Control Center (SCC) under AMS
              guidance.  SCC will contract a laboratory experienced with the determination
              of oil and grease and petroleum hydrocarbons through the competitive
              solicitation process.  SCC will then  coordinate laboratory analyses, receive  and
              validate  all analytical data, and perform statistical analyses. AMS will draw
              conclusions from the results, and produce a report providing the results of the
              study. Upon request, AMS will share data and results with all interested
              parties.
September 29, 1993                        A-4

-------
                                               Report of the Method 1664 Validation Studies
4.2     Analytical Plan: The contracted laboratory will be required to analyze
       approximately 38 wastewater samples for oil and grease by MCAWW Method
       413.1 and for petroleum hydrocarbons by Standard Method 5520F using Freon
       113, n-hexane, and cyclohexane as extraction solvents.  Analysis of each sample
       will be performed in triplicate for each of the three extraction solvents.  The
       Method 413.1 modifications that are necessary when using n-hexane and
       cyclohexane as extraction solvents are included as attachments to this study
       plan.

       Prior to the analysis of field samples, the contracted laboratory will be required
       to make an initial demonstration of its ability to generate acceptable accuracy
       and precision of each of the above required procedures. To  do so, a series of
       Initial Precision and Recovery (IPR) analyses for oil and grease and for
       petroleum hydrocarbons analysis using Freon, n-hexane, and cyclohexane as
       extraction solvents will be performed. Performance of IPRs consists of the
       extraction, concentration, and analysis of a set of four 1-L aliquots of spiked
       reagent water using each of the analytical procedures.  All IPR analyses will
       incorporate the sodium sulfate and filtering steps required for emulsion
       problems (as specified in Method 413.1) in order to test any effects these
       method procedures may introduce.

       A reagent water method blank will be analyzed with each of the IPR sets and
       with each of the sample sets associated with each of the three different solvent
       procedures.  These reagent water blanks will be run through the entire
       extraction and analysis procedure by which the samples are  run.  It will be
       required that method blanks contain a concentration of oil and grease less than
       5 mg/L.  If contamination is detected in any reagent water blank, the
       laboratory will need to isolate the source of contamination, and associated
       samples will be reanalyzed.  Results for the sample analyses will be blank
       subtracted using the value from the reagent  water blank.

       Multiple aliquots will be collected for each sample in order to accommodate
       the numerous analyses required.  For each sample, the multiple aliquots will
       be taken from a homogenized sample and are  expected to have similar oil and
       grease  concentrations. Within each of the three different solvent procedures
       and two  methods (413.1 and 5520F) it is expected that the standard deviation
       of the triplicate measurements will be less than or equal to 10 percent. The
       contracted laboratory will be required to notify appropriate personnel if the
       triplicate results  exceed a relative standard deviation of 10 percent.

4.3     Quality Assurance/Quality Control (QA/QC) Plan:  The contracted laboratory
       will have a comprehensive QA program in place and operating throughout the
       duration of the contract to ensure that data produced are of the highest
       quality.  The laboratory will also follow all QC procedures defined in the
                                 A-5                        September 29, 1993

-------
Report of the Method 1664 Validation Studies
              methodology (except matrix spikes and duplicates) in addition to the following
              QC tests:
       4.4
             QC Sample

             Initial Precision
             and Recovery
             (IPR)

             Ongoing
             Precision and
             Recovery (OPR)

             Method Blanks
                                                           Data Quality
                                                            Objectives

                                                            < 16 % RSD,
                                                              80 - 120%
                                                              recovery

                                                              75 - 125%
                                                              recovery


                                                              < 5 mg/L
Frequency                 Spike Level

Four (4) reagent water      20 mg/L of
aliquots for each of the six  Diesel Fuel
(6) techniques.                  #6

One (1) with each sample   20 mg/L of
batch of the six (6)          Diesel Fuel
analytical techniques.            #6

One (1) with each sample      N/A
batch for each of the six
(6) analytical techniques
and for each of the  six (6)
IPR sets.
Reporting Requirements: The laboratory will be required to submit summary
data in hardcopy and electronic format. The laboratory will also be required
to submit raw data, including copies of worksheets and laboratory notebooks
showing tare and sample weights, sample volumes, solvent volumes, and other
data that will allow the final results to be traced to the analytical steps
performed.  A detailed narrative describing any modifications of the analytical
techniques, problems, and implemented corrective action procedures will also
be provided as part of the data package.
5.  SAMPLE COLLECTION ACTIVITIES

Samples will be collected between November 1993 and April 1994 from facilities in Maine,
Maryland, Massachusetts, New Jersey, New York, and Virginia.  It is estimated that collection
activities will involve 30 facilities encompassing 15-20 industrial categories. Based on
historical information from the facilities, samplers will attempt to collect samples with oil and
grease concentrations between 40-300 mg/L.  At each sample point, approximately 60 1-L
aliquots will be collected.  Samples will be distributed to the contracted laboratory as well as
to vendors of alternative extraction techniques who have expressed interest in participating in
the study.  In addition, back-up aliquots will be stored at Gascoyne Laboratories, Inc., which
provides chemical repository services for AMS activities.

6.  DATA EVALUATION

Data will be evaluated for conformance with data quality objectives for the study, including
completeness and validity. QC test data will be reviewed for compliance with specified
September 29, 1993
                          A-6

-------
                                                      Report of the Method 1664 Validation Studies
objectives, and appropriate corrective action procedures will be implemented where
necessary.  Results presented in the hardcopy summary forms will be verified by
recalculating from the raw data.  Any discrepancies found will be clarified with the analytical
laboratory.

7.  STATISTICAL ANALYSIS OF SAMPLE RESULTS

Standard deviations and percent recoveries will be calculated for the appropriate QC
samples, and for each of the triplicate sample analyses. It is expected that within each of the
six analytical techniques, the relative  standard deviation of the triplicate measurements will
be less than 10 percent.

It is expected that the data will be evaluated by techniques similar to those used in the Phase
I study. This would include using the Root Mean Square Deviation as the main criterion of
equivalence when comparing results obtained with alternative solvents to results obtained
with Freon. Specifics of these statistical techniques are explained in Section 4 of the
Preliminary Report of EPA Efforts to Replace Freon for the Determination of Oil and Grease,
September 1993.
                                        A-7                        September 29, 1993

-------
                                                                                         p. 3
                     ATTACHMENT I - Freon Replacement Study Phase II

Modifications to Method 413.1 when using n-hexane or cyclohexane as the extraction solvent:

       Replace Section 1.1 to read:

       This  method includes the measurement of n-hexane (or cyelphe^jjane) extractable matter from
       surface and saline  water, industrial and domestic wastes.  It is- applicable to determination of
       relatively non-volatile hydrocarbons, vegetable oils, animal fats, waxes, soaps, greases and related
       matter.

       Replace Section 1.2 to read:

       The method is not applicable to measurement of light hydrocarbons that volatilize at temperatures
       below 8§|C. Petroleum fuels from gasoline through #2 fuel oils are completely or partially lost
       in the solvent removal operation.

       Replace Section 1.3 to read:

       Some crude oils and heavy fuel oils contain a significant percentage of residue-type materials that
       are not soluble in n|Mexaae (or cyclohexane).  Accordingly,  recoveries of these materials will be
       low.

       Replace Section 2.1 to read:

       The sample is acidified to low pH (<2) and serially extracted with nrb^^ane^or cyclohexane) in
       a separatory funnel. The solvent is evaporated from the extract and the residue  weighed.

       Replace Section 6.2 to read:

       n-sHe&ane, b^;6g?©(0K:C^clohexane, b,p. 81°C).

       Add Section 6.4 to  read:
       Replace Section 7.3 to read:

       Tare a boiling flask (pre-dried in an oven at 103°C and stored in a desiccator) containing 4 small
       Replace Section 7.4 to read:

       Add 30 ml p^h^^^|6:p^^iR^cl^he^an©:) to the sample bottle and rotate the bottle to rinse the
       sides.  Transfer the  solvent into the  separatory funnel.  Extract by shaking vigorously for 2
       minutes.  Allow the layers to  separate^^^^l^ftS-sfy^^^^ll^^l^^^^ffl^^Ssa^Bi6
       <|i||l||ae|i^
       NOTE:  An emulsion that fails to dissipate can be broken by pouring about 1 g sodium sulfate

-------
                                                                                p. 4
 (6.3) into the filter paper cone and slowly draining the emulsion through the salt. Additional 1
 g portions can be added to the cone as required.

 Replace Section 7f7 to read:

 Connect the boiling flask to the distilling head and evaporate the solvent by immersing the low
 half of the flask in  water at 85IG.   Collect the solvent for reuse.  A solvent blank  should
 accompany each set of samples.

 Replace Section 7.8 to read:

When the temperature in the distilling head reaches 70°C mstem^^^i'^^^ble ^^
uwg«i^neias€i&splvent. Whenuling cyclohexane, the temperature wftise&to%e higir)
or the flask appears dry remove the distilling head.  Sweep out the flask for 15 seconds with air
to remove solvent vapor by inserting  a glass tube connected to a vacuum source.  Immediately
remove  the flask from the heat source and wipe the outside to remove excess moisture and
fingerprints.

-------
                       APPENDIX B
TCRR STUDY ANALYTICAL REQUIREMENTS

-------
Report of the Method 1664 Validation Studies
                      EPA EAD ANALYTICAL METHODS STAFF
                         ANALYTICAL REQUIREMENTS FOR
                 THE FREON REPLACEMENT STUDY ROUND ROBIN
                                        3/21/94

The laboratory's strict adherence to the required analytical methods and quality control
procedures is essential to assure data validity for EPA use. The laboratory shall adhere to
the Quality Assurance/Quality Control (QA/QC) measures prescribed below, and shall
otherwise employ accepted good laboratory practices in all aspects of analytical performance.

Summary of analytical requirements:

1)      Preparation and analysis of two industrial wastewater samples, each in triplicate, for
       n-hexane extractable material (HEM) by Draft Method 1664.

2)      The determination of silica gel treated n-hexane extractable material (SGT-HEM), as
       specified in Draft  Method 1664, is optional and may be  performed at the laboratory's
       discretion.

NOTE:       The laboratory must make an initial demonstration of the ability to generate
             acceptable  accuracy and precision of the above required  procedures. To do so,
             a series of  Initial Precision and Recovery (IPR) analyses must be performed.
             Performance of IPRs consists of the extraction, concentration, and analysis of a
             set  of four  1-L aliquots of spiked reagent water, as specified in Draft Method
             1664. These IPR analyses  must be completed prior to the analysis of field
             samples. A method blank must be run with each IPR set.  If SGT-HEM
             determination is performed, these analytical steps must be incorporated into
             the IPR analyses.

NOTE:       A reagent water method blank must be analyzed with the  IPR set and with the
             sample sets.  These reagent water blanks shall be run through the entire
             extraction and analysis procedure by which the samples are run.  Method
             blanks must have a value less than 5 mg/L.  If contamination is detected in
             the reagent water blanks, the laboratory will need to isolate the source of
             contamination, and associated  samples will need to be reanalyzed.

Special Technical Instructions:

     Multiple aliquots are being collected  for each of the samples.  For each sample, the
     multiple aliquots are from a homogenized sample and are expected to have identical
     concentrations.  It is expected that  the relative standard deviation of the triplicate
     measurements will  be 1)  less than or equal to 10% for those samples with values that
March 21, 1994                           B-2

-------
                                                     Report of the Method 1664 Validation Studies
     are greater than or equal to the 20 mg/L minimum level and 2) less than or equal to
     20% for those samples with values that are less than 20 mg/L.

     All IPR analyses shall incorporate the sodium sulfate and filtering steps required for
     emulsion problems in order to test any effects these steps may introduce.

     If a filter other than the method-specified Whatman 40 filter is necessary in order to
     prevent passage of fine particulates, the laboratory should incorporate such changes
     into the spiked reagent water and method blank analyses.  These changes  must be
     documented in the laboratory narrative.

     The laboratory has the option of using whichever filter it judges to be suitable for this
     procedure, as long as contamination of the extract does not occur. It is advised, but not
     required, that 0.45 u membrane filters be used.  If contamination occurs (i.e. method
     blanks produce values > 5 mg/L) samples associated with these blanks will need to be
     reanalyzed.  The laboratory should report contamination problems to SCC as they
     occur.

     These above mentioned considerations for particulate contamination also apply to the
     silica gel procedure.

Analytical protocol required:

Draft Method 1664:  Determination of n-hexane extractable material and silica gel treated n-
hexane extractable material (attached).

Analytical results required:

     a)    A narrative that details any problems with or deviations from the referenced
           methods and reports problems associated with the analysis of specific samples.
           The narrative should also provide comments on the method performance on
           various analytes and matrices.

     b)    A list of samples analyzed,  and  a run chronology showing (in sequence)
           extraction dates and times and weighing dates and times.

     c)    Summary reports of all sample,  IPR, OPR, and blank  analyses.  Data reporting
           forms are included in this information package.

Note:  The calculations  performed by the laboratory in generating sample data must be able
       to be reproduced by a 3rd party from the data package.
                                       B-3                           March 21, 1994

-------
Report of the Method 1664 Validation Studies
QC Requirements:

The following QC parameters are required:

Initial Precision and Recovery Analyses (IPR) - IPR analysis prior to the analysis of field
samples. Data quality objectives for spiked reagent water sets are as follows:  An average
percent recovery between 80-120% and a relative standard deviation less than or equal to
16%.

Ongoing Precision and Recovery Analysis  (OPR) - One OPR analysis with the two sample
sets. Data quality objectives are as follows:  A percent recovery between 75-125%.

One Method Blank is  required for analysis with the set of IPR samples and with the samples.
Method blank values  must be less than 5 mg/L.

NOTE:        If contamination is detected in the method blank, the source of contamination
              must be identified and corrected. The blank and all samples associated with
              that contaminated blank must be reprepared and reanalyzed.

Instrument Calibration shall be  confirmed through the use of Initial and Continuing
Calibration Verification Standards per method instructions.

If any of these instructions are not clear please call Carrie Buswell at (703) 519-1140.
March 21, 1994                           B-4

-------
Laboratory
Parameter
Extraction
Sample
Method
Blank
IPR 1
IPR 2
IPR 3
IPR 4
Reporting Format for the Twin City Interlaboratory Study
Initial Precision and Recovery (IPR) Samples and Associated Method Blank
f Name:
	 HEM
/Analysis Date
Sample
Volume (L)





	 SGT-HEM
Date I

Initial
Weight of
Flask (g)





Final
Weight of
Flask (g)





ixtractable Material Weighed:
Sample
Weight (g)





Sample
Cone.
(mg/L)







Blank
Cone.
(mg/L)






Sample - Blank
(mg/L)
N/A




Spike
Cone.
(mg/L)
N/A





Percent
Recovery
N/A




IPR Mean Recovery = %
IPR Relative Standard Deviation = 	 %
Data Quality Objectives for the average IPR recovery = 80 - 120%
Data Quality Objectives for the relative standard deviation = less than or equal to 16%
Method Blank values must be less than 5 mg/L
•p
 en

-------
Laboratory
Parameter:
Extraction//
Lab
Sample #








Reporting Format for the Twin City Intel-laboratory Study
Field Samples, Associated Method Blank, and Associated Ongoing Precision and Recovery Sample
Vame:
	 HEM
Analysis Date
EPA
Sample #
Method
Blank






OPR
_ SGT-HEM
D

Rep#

1
2
3
1
2
3

Sample
Volume
(L)








ate Extractable Material Weij
Initial
Weight of
Flask (g)








Final
Weight of
Flask (g)








died:

Sample
Weight (g)








Sample
Cone.
(mg/L)









Blank
Cone.
(mg/L)








Sample - Blank
(mg/L)
N/A







Spike
Cone.
(mg/L)
N/A
N/A
N/A
N/A
N/A
N/A
N/A

Percent
Recovery
N/A
N/A
N/A
N/A
N/A
N/A
N/A

Data Quality Objectives for the OPR recovery = 75 - 125%
Method Blank values must be less than 5 mg/L
o

-------
                              APPENDIX C
STUDY PLANS FOR THE METHOD 1664 MDL STUDIES

-------
                                                     Report of the Method 1664 Validation Studies
                                    STUDY PLAN
  Determination of MDL for EPA Draft Method 1664 - "Oil and Grease" and "Petroleum
 Hydrocarbons"  [N-Hexane Extractable Material (HEM) and Silica Gel Treated N-Hexane
             Extractable Material (SGT-HEM) by Extraction and Gravimetry]

1.      Introduction

       As a party to the Montreal Protocol on Substances that Deplete the Ozone Layer and as
       required by law under the Clean Air Act Amendments of 1990 (CAAA), the United States
       is committed to controlling and eventually phasing out the use of chlorofluorocarbons
       (CFCs).   In support of these  efforts, the U.S. Environmental Protection Agency is
       searching for suitable solvents to replace Freon-113, which is required for use in several
       U.S. EPA wastewater and solid waste methods for the determination of oil and grease
       and petroleum hydrocarbons. The Freon Replacement Study, which to date has consisted
       of two phases,  is part of these efforts.

       In Phase  I, five extraction solvents  for liquid/liquid  extraction and gravimetric
       determination of oil and grease in aqueous samples were evaluated, as was the use of
       alternative techniques  for oil and grease analysis including sonication extraction, solid
       phase extraction, and a non-dispersive infra-red technique.

       The results from Phase  I were used to narrow the list of alternative solvents  under
       consideration in Phase II to n-hexane and cyclohexane.  These Phase II solvents were
       used for liquid/liquid extraction and gravimetric determination of both oil and grease
       and total petroleum hydrocarbons (TPH).  Based on  the results  of Phase II and the
       analytical issues and observations consolidated from the study,  Draft Method  1664 was
       written, with n-hexane as the extraction solvent.

       The most significant changes in Draft Method 1664 compared to previous oil and grease
       and petroleum hydrocarbons methods  were 1) the use of n-hexane as the extraction
       solvent,  2)  the use of hexadecane and stearic acid as the spiking materials for  QC
       analyses, and 3) the introduction of 1600 series method QC, including initial  precision
       and recovery analysis (IPR), ongoing precision and recovery analysis (OPR), reagent
       water method  blanks,  and  matrix  spike/matrix   spike  duplicates  (MS/MSDs).
       Interlaboratory studies  are currently underway to develop method specifications for IPRs
       and OPRs.

       As part of QC  development, a method detection limit (MDL) must be determined for
       Draft Method 1664.  A previous study determined the MDL for a modified version of
       MCAWW Method 413.1  using Wesson oil and No. 6 Fuel oil  as analytes and  n-hexane
       as the extraction solvent. Though the modified Method 413.1  procedure was similar to
       Draft Method 1664, the spiking materials differed from the hexadecane and stearic acid


                                        C-1                         Novembers, 1994

-------
Report of the Method 1664 Validation Studies
       spiking materials specified in  Draft Method  1664.  This study plan describes EPA's
       approach to determining the MDL for n-hexane extractable material (HEM) and silica gel
       treated n-hexane  extractable material (SGT-HEM) ("oil  and grease" and  "petroleum
       hydrocarbons", respectively), using Draft Method 1664.

2.      Objective

       The objective of this study is to determine the MDLs for HEM and SGT-HEM by Draft
       Method 1664 from the analysis  of seven reagent water samples spiked with hexadecane
       and stearic acid. As is specified in Section 9.2.1 of Draft Method 1664, MDL analyses will
       be performed according to the  procedure in 40 CFR 136, Appendix B.

       Data quality objectives  (DQOs) include the following:

             To achieve an MDL that is equal to or within a factor of five lower than the level
             spiked in  order to ensure  that the MDLs determined in this  study  are not
             understated or overstated.

             To achieve an MDL of 3 mg/L for both HEM and SGT-HEM.

       In order to meet these DQOs, the laboratory will be required to have a comprehensive
       QA program in place and operating throughout the duration of this study. This will
       ensure that the data produced are of the highest possible quality.  The laboratory will be
       required  to follow all  QC procedures  defined in this study plan  and in EPA Draft
       Method 1664 with the following exceptions:

             Demonstration of ongoing precision and recovery will not be required.

             Performance of matrix  spike and  matrix spike duplicate analyses will not be
             required.

             Instrument calibration must be performed at  a range that will encompass the
             estimated detection limit being studied.

             Analysis of a Quality Control Sample, as defined in Section 9.7 of Draft Method
             1664, will not be required.

       If the procedures described in this study plan conflict with those  described in Draft
       Method 1664 and 40 CFR 136, Appendix B, the study  plan will take precedence.

3.      Study Management/Limitations

       The study will be managed by  the Office of Water's Engineering and Analysis Division
       through the Analytical Methods Staff (AMS). Day-to-day management and coordination
       of study activities will  be provided by the contractor-operated Sample Control Center
       (SCC) under AMS guidance. SCC will contract a laboratory experienced with gravimetric
       determination of oil and grease and petroleum hydrocarbons.  SCC will then coordinate

November 3, 1994                           C-2

-------
                                                     Report of the Method 1664 Validation Studies
       laboratory analysis, receive  and validate  all analytical data, and perform  statistical
       analyses. AMS will draw conclusions from the results, and produce a report providing
       the  results  of the  study.  Upon request,  AMS will share data and results with all
       interested parties.

       Analyses will be performed and results obtained by July 15, 1994, in order to incorporate
       these results into Draft Method 1664 prior to the proposal of this method in the Federal
       Register.

       If they have  not already done so, the laboratories  considered  for this study will be
       required to  perform initial precision and recovery analyses (IPR) for Draft Method 1664
       and to meet the IPR data quality objectives specified in this method.

4.      Technical Approach/Limitations/Procedures

       Two separate MDL studies will be  performed, one for HEM analysis and one for SGT-
       HEM analysis.  MDLs will  be determined according to the protocol in  40  CFR  136,
       Appendix B using the apparatus, reagents, and standards  that are specified in Draft
       Method  1664.  Reagent water  will  be  used  as  the  sample matrix because  it  is
       homogeneous, readily available, and included as a matrix option in 40 CFR 136.

       Targeting other matrices was not considered a viable option  for a number of reasons.  If
       samples from a particular industrial category were used, for example, results  would be
       biased for a particular industry. It would also be nearly impossible to test every sample
       matrix  to which this method may be applied, since  EPA regulates more  than  600
       categories of wastewater discharge.  Consequently, laboratory evaluation of Draft Method
       1664 on each  discharge would be both time-consuming and expensive.

       Some exceptions and clarifications to the procedure in 40 CFR 136, Appendix B will need
       to be incorporated  into this study.  They are as follows:

             The  laboratory will be required to prepare and analyze  seven  separate aliquots
             of spiked reagent water, each prepared in its own sample container, for each of
             the two MDL studies. Separate containers are required for each sample to ensure
             that  an exact, known  amount of HEM and SGT-HEM is present in each aliquot.
             If aliquots were to be  taken from a single container, components of the HEM  and
             SGT-HEM might  adhere to the sample container,  thereby preventing  equal
             distribution and precise allocation of the spiked components into each aliquot.

             The  MDL study  protocol requires  spiking seven replicates with  the targeted
             analytes at  a concentration of one to five  times the  estimated MDL.   The
             estimated MDL for both HEM and SGT-HEM is 3 mg/L.  For the HEM MDL
             study, the laboratory will be required to spike reagent water samples to produce
             a  combined hexadecane/stearic acid  concentration of 10  mg/L  (5 mg of
             hexadecane and  5 mg of stearic acid). For the SGT-HEM  MDL study,  the


                                        C-3                         Novembers, 1994

-------
Report of the Method 1664 Validation Studies
             laboratory will need to compensate for the silica gel adsorption of the stearic acid.
              Consequently, the laboratory will be required to spike reagent water samples to
             produce a combined hexadecane/stearic acid concentration of 20 mg/L (10  mg
             hexadecane and 10 mg of stearic acid) for SGT-HEM analysis.

             Samples are to be prepared according to the instructions in Section 7.9 of Draft
             Method 1664,  with the exception that the amount of the hexadecane/stearic acid
             spiking solution added to the one liter reagent water sample will be decreased in
             order to produce a final concentration of 10 mg/L (5 mg of hexadecane and 5  mg
             of stearic acid) in each one liter sample for the HEM MDL study and 20 mg/L (10
             mg of hexadecane and 10 mg of stearic acid) in each one liter sample for the SGT-
             HEM MDL  study.

       To ensure that the  laboratory is spiking at the appropriate level, the laboratory will be
       required to analyze two aliquots to evaluate the spike levels  as described in Section 4b
       of the 40 CFR 136, Appendix B procedure. The laboratory will report the results of these
       two analyses to SCC  before continuing with any other analyses.  If these measurements
       indicate that the sample is in the desirable range for determination of the MDL, SCC will
       instruct the laboratory to proceed with the analysis of the remaining five aliquots.  All
       seven measurements will be used  for calculation of the  MDL.   If  the first  two
       measurements indicate that the samples are not in the  desirable range, the laboratory
       shall repeat Section 4b (analysis of two aliquots) until the desired level is achieved.

       A reagent water blank will be analyzed with each set of seven analyses.  These reagent
       water blanks will be run through the entire extraction and analysis procedure by which
       the MDL samples are run. It will be required that method blanks contain a concentration
       of 3 mg/L or less.   If contamination is detected in  any  reagent water  blank,  the
       laboratory will need to isolate the source of contamination, and the associated MDL
       studies will need to be repeated.

       If the MDL resulting  from the study is not equal to or within  a factor of five lower than
       the level spiked, the spiking, measurement, and calculation process will be iterated until
       the measured MDL is within these specifications.

5.      Reporting Requirements/Statistical Analysis of Laboratory Results

       The laboratory will be required to submit summary data in hardcopy and electronic
       format.  The summary data will consist  of a table that will group data under  the
       following categories:  concentration of HEM or SGT-HEM in the seven aliquots, the mean
       of the seven concentrations, standard deviations, Student's value used, and calculated
       MDL for HEM and SGT-HEM.

       The laboratory will also be required to submit raw data, including copies of worksheets
       and laboratory notebooks showing tare and sample weights, sample volumes, solvent
       volumes,  and other data that will allow the final results to be traced to the analytical
       steps performed.  A  detailed narrative describing any  modifications of the analytical

November 3, 1994                           C-4

-------
                                               Report of the Method 1664 Validation Studies
techniques,  problems, and implemented  corrective action procedures  will  also  be
provided as part of the data package.

All data from this study will  be  submitted to SCC for review and validation.  The
computer-readable results will  be verified against the hardcopy data, then loaded into
the study database. Hardcopy  data, including all calculations, will be verified from the
bench sheets provided.

The MDLs will be calculated using the procedures in 40 CFR 136, Appendix B.
                                  C-5                         Novembers, 1994

-------
Report of the Method 1664 Validation Studies
                                    STUDY PLAN
 Determination of MDL for EPA Method 1664 - N-Hexane Extractable Material (HEM) and
     Silica Gel Treated N-Hexane Extractable Material (SGT-HEM) by Extraction and
             Gravimetry (Oil and Grease and Total Petroleum Hydrocarbons)

1.      Introduction

       As a party to the Montreal Protocol on Substances that Deplete the Ozone Layer and as
       required by law under the Clean Air Act Amendments of 1990 (CAAA), the United States
       is committed to controlling and eventually phasing out the use of chlorofluorocarbons
       (CFCs).   In support of these  efforts, the U.S. Environmental Protection Agency is
       searching for suitable solvents to replace Freon-113, which is required for use in several
       U.S. EPA wastewater and solid waste methods for the determination of oil and grease
       and petroleum hydrocarbons. The Freon Replacement Study, which to date has consisted
       of two phases,  is part of these efforts.

       In Phase  I, five extraction solvents for liquid/liquid  extraction and gravimetric
       determination of oil and grease in aqueous samples were evaluated, as was the use of
       alternative techniques for oil and grease analysis including sonication extraction, solid
       phase extraction, and a non-dispersive infra-red technique.

       The results from Phase I were used to narrow the list of alternative  solvents under
       consideration in Phase  II to n-hexane and cyclohexane.  These Phase II solvents were
       used for liquid/liquid extraction and gravimetric determination of both oil and grease
       and total petroleum hydrocarbons (TPH).  Based on the results of Phase II and the
       analytical issues and observations consolidated from the study,  Draft Method  1664 was
       written, with n-hexane  as the extraction solvent.

       The most significant changes in Draft Method 1664 compared to previous oil and grease
       and petroleum hydrocarbons methods were 1) the use of n-hexane as the extraction
       solvent,  2)  the use of  hexadecane and stearic acid as the spiking materials for  QC
       analyses,  and 3) the introduction of 1600 series method QC, including initial  precision
       and recovery analysis  (IPR), ongoing precision and recovery analysis  (OPR), reagent
       water method blanks, and matrix spike/matrix spike duplicates (MS/MSDs).

       An interlaboratory study was conducted and the results obtained were used to develop
       method specifications for IPRs,  OPRs, and MS/MSDs, which were incorporated into the
       October  1994 revision of Method 1664.

       As part of this QC development,  a method detection limit (MDL) also needed to be
       determined for Method 1664. An MDL study was conducted using reagent water as the
       sample matrix and the method  specified hexadecane and stearic acid spiking materials.
       The MDL values produced were 0.91 mg/L for HEM and 1.6 mg/L for SGT-HEM.


June 23, 1994                             C-6

-------
                                                     Report of the Method 1664 Validation Studies
       In  order  to confirm  the values from the original study,  a second MDL study was
       initiated.  Results from this study (5.4 mg/L for HEM and 2.6 mg/L for SGT-HEM) were
       higher than those produced in the original study but, after evaluating the conditions
       under which the laboratory performed the second study, which included  considerable
       time constraints (i.e., a 48 hour turnaround)  and what the laboratory  admitted  as
       performance not reflective  of their best abilities, it was decided that the MDL values to
       be  included in the October version of Method 1664 should be those representing the best
       laboratory performance. Therefore, the MDL and associated Minimum Level values from
       the original Method 1664 MDL study were incorporated into the October 1994  revision
       of Method 1664.

       Though a decision was made regarding the MDL  and ML specifications for the October
       1994 version of Method 1664, several factors suggest the need to perform additional MDL
       studies. First, though a second MDL study was performed, the laboratory stated that the
       work performed did  not reflect their  ability.  Therefore, the results are suspect and do
       not represent a reliable comparison to the original study.  Second, the HEM and SGT-
       HEM values calculated from the first MDL study  are at levels below five times  the then
       estimated detection limit of 3 mg/L,  indicating that even lower spike levels could have
       been used.  Given these circumstances,  a third MDL study will  be  conducted using
       reagent water samples spiked at lower levels than  previously used to reflect the updated
       (lower) MDL levels in the October version of Method 1664.

       This study plan  describes EPA's approach to  determining  the  MDL for n-hexane
       extractable material (HEM) and silica gel treated n-hexane extractable material (SGT-
       HEM) using Method  1664, October 1994.

2.      Objective

       The objective of this study is to determine the MDLs for HEM and SGT-HEM by Method
       1664 from the  analysis of seven reagent  water samples spiked with hexadecane and
       stearic acid.  As is specified in Section  9.2.1 of  Method 1664, MDL analyses will  be
       performed according to the procedure in 40 CFR  136, Appendix B.

       Data quality objectives (DQOs) include the following:

             To achieve an MDL that is equal to or within a factor of five lower than the level
             spiked  in  order to  ensure that the MDLs determined in  this study  are not
             understated or overstated.

             To achieve an MDL of 0.91 mg/L for HEM and 1.6 mg/L for SGT-HEM.

       In  order to meet these DQOs, the laboratory will  be required to have  a comprehensive
       QA program in place and operating throughout  the duration of this  study. This will
       ensure that the data produced are of the highest possible quality.  The laboratory will be
       required to follow all QC procedures defined in this study plan and in the October 1994
       version of Method 1664 with the following exceptions:


                                       C-7                            June 23, 1994

-------
Report of the Method 1664 Validation Studies
             Demonstration of ongoing precision and recovery will not be required.

             Performance of matrix spike  and matrix spike duplicate analyses will not be
             required.

             Instrument calibration must be performed at a range that will encompass the
             estimated detection limit being studied.

             Analysis of a Quality Control  Sample, as defined in Section 9.7 of Method 1664,
             will not be required.

       If the procedures described in this  study plan conflict with those described in Method
       1664 and 40 CFR 136, Appendix B,  the study plan will take precedence.

3.      Study Management/Limitations

       The study will be  managed by the Office of Water's Engineering and Analysis Division
       through the Analytical Methods Staff  (AMS).  Day-to-day management and coordination
       of study activities will be provided by the contractor-operated Sample Control Center
       (SCC) under AMS guidance.  SCC will contract a laboratory experienced with gravimetric
       determination of oil and grease and petroleum hydrocarbons. SCC will then coordinate
       laboratory analysis, receive and validate  all analytical data,  and perform statistical
       analyses. AMS will draw conclusions from the results, and produce a report providing
       the  results of  the  study.   Upon request,  AMS will share data and results with all
       interested parties.

       Analyses will be performed and results obtained by November 21,  1994, in order to
       incorporate these results prior to a revision of Method 1664 for proposal in the Federal
       Register.

       If they have not already done so, the laboratories  considered for this study will be
       required to perform initial precision and recovery analyses (IPR)  for Method 1664 and
       to meet  the IPR data quality objectives specified in this method.

4.      Technical Approach/Limitations/Procedures

       Two separate MDL studies will be  performed, one for HEM analysis and one for SGT-
       HEM analysis.  MDLs will be determined according to the protocol in  40 CFR 136,
       Appendix B using the apparatus, reagents, and standards that are specified in Method
       1664.  Reagent water  will be used as the sample matrix because  it  is homogeneous,
       readily available, and included as a matrix option in 40 CFR 136.

       Targeting other matrices was not considered a viable option for  a number of reasons.  If
       samples from a particular industrial category were used, for example, results would be
       biased for a particular  industry. It would also be nearly impossible to  test every sample
       matrix  to  which this  method may be  applied, since  EPA regulates more than 600
June 23, 1994                            C-8

-------
                                              Report of the Method 1664 Validation Studies
categories of wastewater discharge. Consequently, laboratory evaluation of Method 1664
on each discharge would be both time-consuming and expensive.

Some exceptions and clarifications to the procedure in 40 CFR 136, Appendix B will need
to be incorporated into this study. They are as follows:

       The laboratory will be required to prepare and analyze seven separate aliquots
       of spiked reagent water, each prepared in its own sample container, for each of
       the two MDL studies. Separate containers are required for each sample to ensure
       that an exact, known amount of HEM and SGT-HEM is present in each aliquot.
       If aliquots were to be taken from a single container, components of the HEM and
       SGT-HEM might adhere  to  the sample container, thereby preventing  equal
       distribution and precise allocation of the spiked components into each aliquot.

       The MDL study protocol requires spiking seven replicates with the  targeted
       analytes  at a  concentration of one  to  five times the estimated  MDL.  The
       estimated MDL for HEM is 0.91 mg/L  and for SGT-HEM is 1.6 mg/L.  For the
       HEM MDL study, the laboratory will be required to spike reagent water samples
       to produce a combined hexadecane/stearic acid concentration between 0.91 and
       4.55 mg/L (note that equal amounts of hexadecane and stearic acid will comprise
       the total concentration). For the SGT-HEM MDL study,  the laboratory will need
       to compensate for the silica gel adsorption of the stearic  acid. Consequently, the
       laboratory will be required to spike reagent water samples to produce a combined
       hexadecane/stearic acid concentration of 3.2 to 16 mg/L (note that equal amounts
       of hexadecane and stearic acid will comprise the total  concentration) for SGT-
       HEM analysis.

       Samples  are to be prepared according to the  instructions in Section  7.11 of
       Method 1664, with the exception that the amount of the  hexadecane/stearic acid
       spiking solution added to the one liter reagent water sample  will be  decreased in
       order to produce a final concentration within the ranges specified above.

To ensure that the laboratory is spiking at the  appropriate level, the laboratory will be
required to analyze two aliquots to evaluate the spike levels as described in Section 4b
of the 40 CFR 136, Appendix B procedure. The  laboratory will report the results of these
two analyses to SCC before continuing with any other analyses.  If these  measurements
indicate that the sample is in the desirable range for determination of the MDL, SCC will
instruct the laboratory to proceed with the analysis of the remaining five aliquots.  All
seven  measurements will  be used for  calculation of the MDL.  If  the  first two
measurements indicate that  the samples are not in the desirable range, the laboratory
shall repeat Section 4b (analysis of two aliquots) until the desired level is achieved.

A reagent water blank will be analyzed with each set of seven analyses.  These  reagent
water blanks will be run through the entire extraction and analysis procedure by which
the MDL samples are run.  It will be required that method blanks contain a concentration
of 0.91 mg/L or less for HEM and 1.6 mg/L or less for SGT-HEM.  If contamination is

                                C-9                            June 23, 1994

-------
Report of the Method 1664 Validation Studies
       detected in any reagent water blank, the laboratory will need to isolate the source of
       contamination, and the associated MDL studies will need to be repeated.

       If the MDL resulting from the study is not equal to or within a factor of five lower than
       the level spiked, the spiking, measurement, and calculation process will be iterated until
       the measured MDL is within these specifications.

5.      Reporting Requirements/Statistical Analysis of Laboratory Results

       The laboratory will be required to submit summary data in  hardcopy and electronic
       format.  The summary data will  consist of a table that will group data under  the
       following categories: concentration of HEM or SGT-HEM in the seven aliquots, the mean
       of the seven concentrations, standard deviations, Student's t value used, and calculated
       MDL for HEM and SGT-HEM.

       The laboratory will also be required to submit raw data, including copies of worksheets
       and laboratory notebooks showing tare and  sample weights,  sample volumes, solvent
       volumes, and other data that will allow the  final  results to be traced to the analytical
       steps performed.  A  detailed narrative describing any modifications of the analytical
       techniques,  problems,  and  implemented corrective action procedures  will  also  be
       provided as part of the data package.

       All data from this study will be submitted to SCC for review and validation.  The
       computer-readable results will be verified against the hardcopy data, then loaded into
       the study database. Hardcopy data, including all calculations, will be verified from the
       bench sheets provided.

       The MDLs will be calculated using the procedures in 40 CFR  136, Appendix B.
June 23, 1994                            C-10

-------
                             Report of the Method 1664 Validation Studies
                                               APPENDIX D
ANALYTICAL REQUIREMENTS FOR THE METHOD 1664 MDL STUDIES

-------
Report of the Method 1664 Validation Studies
    ANALYTICAL REQUIREMENTS FOR THE DRAFT METHOD 1664 MDL STUDY

                                        6/24/94
The  laboratories strict adherence to the required analytical methods  and quality control
procedures is essential to assure data validity for EPA use. The laboratory shall adhere to the
Quality Assurance/Quality Control (QA/QC) measures prescribed below, and shall otherwise
employ accepted good laboratory practices in all aspects of analytical performance.

Summary of analytical requirements:

       Performance of a method detection limit (MDL) study for HEM analysis by Draft Method
       1664.

       Under this purchase order, a minimum of seven HEM analyses and a maximum of nine
       HEM analyses will be performed for the MDL determination.  Note: This number  does
       not include the four IPR analyses that may be necessary and the method blank analysis.

       Performance of a method detection limit (MDL) study for SGT-HEM analysis by Draft
       Method 1664.

       Under this purchase order, a minimum of seven SGT-HEM analyses and a maximum of
       nine SGT-HEM  analyses will be performed for the MDL determination.  Note:   This
       number does not include the four IPR analyses that may be necessary and the method
       blank analysis.

NOTE 1:      Two separate MDL  studies are required, one for HEM and the other for SGT-
             HEM, because the concentrations at which the  samples  will be spiked with
             hexadecane and stearic acid  for  the  HEM MDL study  will be too low for
             satisfactory determination and quantitation of SGT-HEM.  This is due to the fact
             that  the silica  gel procedure will  adsorb the stearic acid, thereby halving the
             concentration available for SGT-HEM measurement.

NOTE 2:      The maximum number of nine analyses for each MDL study accounts for the
             possibility that, after performing the procedure  in Section 4b of 40 CFR 136,
             Appendix B, the laboratory may find that it is necessary to adjust spiking levels
             and repeat the analysis of those two aliquots.

NOTE 3:      Laboratories who have not previously done so must satisfactorily analyze  four
             initial precision and recovery  (IPR) samples  according to the instructions  in
             Section  9.2.2 of Draft Method  1664 prior to  the analysis of the MDL study
             samples. The data for these IPR samples must be faxed to SCC upon completion
             at  (703) 684-0610.  These data must also be submitted with the data package.  For
             the purposes of the  IPR analyses only, the same extract used for HEM analysis
             may be used for  SGT-HEM analysis,  meaning that the HEM residue may be

June 24, 1994                           D-2

-------
                                                     Report of the Method 1664 Validation Studies
             redissolved  in  n-hexane  and subjected  to  the  silica  gel  procedure  for
             determination of SGT-HEM.

NOTE 4:     A reagent water method blank, as is specified in Section 9.4 of Draft Method 1664,
             must be analyzed with the set of four IPR samples and with each of the two MDL
             studies (three method blanks total). Method blanks must have a value of less
             than  3 mg/L.   If contamination is detected in the reagent  water blanks,  the
             laboratory will need to isolate  the source of contamination, and the associated
             MDL studies will need to be repeated.

Data Turnaround Requirements: Data must be submitted to SCC by July 15,  1994.

Spiking Instructions:

       IPR samples are to be spiked as described  in Section 9.2.2 of Draft Method 1664.

       For the HEM MDL study, reagent water samples are to be prepared according to  the
       instructions in Section 7.9 of Draft Method 1664, with the exception that the amount of
       hexadecane/stearic acid spiking solution will be decreased in order  to  produce a final
       concentration of 10 mg/L (5 mg of hexadecane and 5 mg of stearic acid).

       For the SGT-HEM MDL study, reagent water samples are to be prepared according to
       the instructions in Section 7.9 of Draft Method 1664, with the exception that the amount
       of hexadecane/stearic acid spiking solution will be decreased in order to produce a final
       concentration of 20 mg/L (10 mg of hexadecane and 10 mg of stearic acid).

Analytical Protocol Required:

       EPA Draft Method 1664.

       40  CFR Part  136, Appendix B - Definition and Procedure for the Determination of the Method
       Detection Limit - Revision 1.11  (attached).

Special Technical Instructions:

       The laboratory must follow the technical instructions outlined in the Study Plan,  along
       with those specified in the required analytical protocol, with the exception of the bullet
       listed below.

       A minimum of seven analyses and a maximum of nine analyses will be required for each
       of  the two  MDL  studies (IPRs are not included). The Study  Plan states  that if  the
       resulting MDL(s)  does  not meet the study plan criteria of being equal to or within a
       factor of five lower than the level spiked, the laboratory is to repeat the study. For  the
       purposes of this project, however, the laboratory is NOT to repeat an MDL study without
       formal modification of this purchase order by DynCorp Viar.
                                       D-3                            June 24, 1994

-------
Report of the Method 1664 Validation Studies
Analytical Results Required:

       A narrative that details any problems with or deviations from the referenced methods
       and reports problems associated with the analysis of specific samples.  The narrative
       should also provide comments on the method performance on each analyte.

       A list of samples analyzed, and a run chronology.

       Summary reports of all sample results. An explanation of the summary format table is
       provided in the Study Plan.

       Raw data, including bench sheets, etc.

       The laboratory must report the MDL values to three significant figures.

QC Requirements:  As specified in the Study Plan - QC Minimums provided below.

       Initial Precision and Recovery Analyses - Data quality objectives are specified in Draft
       Method 1664.

       Analysis of a method blank is required with the IPR set and with each of the two MDL
       studies. Method blank values must be less than 3 mg/L.

       NOTE:       If  contamination  is detected  in  the  method  blank,  the source  of
                    contamination must be identified and corrected.  The blank and any IPR
                    sets or MDL  studies associated with that contaminated blank must be
                    reprepared and reanalyzed.

       Instrument  calibration shall be confirmed  per Draft Method 1664 and  Study  Plan
       specifications.
June 24, 1994                            D-4

-------
                                                    Report of the Method 1664 Validation Studies
          ANALYTICAL REQUIREMENTS FOR METHOD 1664 MDL STUDY

                                        11/3/94

The  laboratory's strict adherence to the required analytical methods and  quality control
procedures is essential to assure data validity for EPA use.  The laboratory shall adhere to the
Quality Assurance/Quality Control (QA/QC) measures prescribed below, and shall otherwise
employ accepted good laboratory practices in all aspects of analytical performance.

Summary of analytical requirements:

       Performance of a method detection limit (MDL) study for HEM analysis by Method 1664.

       Under this purchase order, a minimum of seven HEM analyses and a maximum of nine
       HEM analyses will be performed for the MDL determination. Note:  This number does
       not include  the four IPR analyses that may be necessary and the method blank analysis.

       Performance of a method detection limit (MDL) study for SGT-HEM analysis by Method
       1664.

       Under this purchase order, a minimum of seven SGT-HEM analyses and a maximum of
       nine SGT-HEM analyses will be performed for the MDL  determination.  Note:  This
       number does not include the four IPR analyses that may be necessary and the method
       blank analysis.

NOTE 1:      Two separate MDL  studies are required, one for HEM and the other for SGT-
             HEM, because  the concentrations  at which the samples will be spiked with
             hexadecane and stearic acid for the  HEM  MDL  study will be too  low for
             satisfactory determination and quantitation of SGT-HEM.  This is due to the  fact
             that  the silica gel procedure will adsorb the stearic acid, thereby halving the
             concentration available for SGT-HEM measurement.

NOTE 2:      The  maximum  number of nine analyses for each MDL study accounts for the
             possibility  that, after performing the procedure in  Section  4b  of 40 CFR  136,
             Appendix B, the laboratory may find that it is necessary to adjust spiking levels
             and  repeat the analysis of those two aliquots.

NOTE 3:      Laboratories who have not previously done so must satisfactorily analyze  four
             initial precision and recovery  (IPR) samples according  to the instructions in
             Section 9.2.2 of Method 1664 prior to the analysis of the MDL study samples.   The
             data for these IPR samples must be faxed to SCC upon completion at (703)  684-
             0610.  These data must also be submitted  with the  data  package.   For  the
             purposes of the IPR analyses only, the same extract used for HEM analysis  may

                                      D-5                         Novembers,  1994

-------
Report of the Method 1664 Validation Studies
             be  used  for  SGT-HEM analysis,  meaning  that  the HEM residue may  be
             redissolved  in  n-hexane  and  subjected to  the  silica  gel  procedure  for
             determination of SGT-HEM.

NOTE 4:     A reagent water method blank, as is specified in Section 9.4 of Method 1664, must
             be analyzed  with the set of four IPR samples and with  each of the two MDL
             studies (three method blanks total). Method blanks must have a value of less
             than 0.91 mg/L for HEM and  1.6 mg/L for SGT-HEM.  If contamination is
             detected in the reagent water blanks, the laboratory will need to isolate the source
             of contamination, and the associated MDL studies will need to be repeated.

Data  Turnaround Requirements:  A  formal  data package must be  submitted  to SCC  by
November 21, 1994.

Spiking Instructions:

      IPR samples are to be spiked as described  in Section 9.2.2  of Method 1664.

      For the HEM MDL  study, reagent water samples are to be prepared according  to the
      instructions in Section  7.11  of Method 1664, with the exception  that the amount of
      hexadecane/stearic acid spiking solution will be decreased in order to produce a final
      concentration  that is one to five times the 0.91 mg/L MDL value  that is specified in the
      October 1994 version of Method 1664, i.e. 0.91  - 4.55 mg/L. Note that the concentration
      chosen will consist of equal portions of hexadecane and stearic acid. For example, if a
      final concentration of 3 mg/L is chosen, it  will consist of 1.5 mg  of hexadecane and  1.5
      mg of stearic acid.

      For the SGT-HEM MDL study, reagent  water samples are to be  prepared according to
      the instructions in Section 7.11  Method 1664, with the exception  that the amount of
      hexadecane/stearic acid spiking solution will be decreased in order to produce a final
      concentration that is two to ten  times the 1.6 mg/L MDL  value  that is specified  in the
      October 1994  version of Method 1664, i.e.  3.2 - 16 mg/L.  Note  that the  concentration
      chosen will consist of equal portions of hexadecane and stearic acid. For example, if a
      final concentration of 10 mg/L is chosen, it will consist of 5 mg of hexadecane and 5 mg
      of  stearic acid.  Also note that the two to  ten times multiplier applied to the MDL is
      double that in the HEM study in order to account for the adsorption of the stearic acid
      by the silica gel procedure.

Analytical Protocol Required:

      EPA Method  1664, October 1994.

      40  CFR Part 136, Appendix B - Definition and Procedure for the Determination of the Method
      Detection Limit - Revision 1.11 (attached).
Novembers,  1994                         D-6

-------
                                                     Report of the Method 1664 Validation Studies
Special Technical Instructions:

       The laboratory must follow the technical instructions outlined in the Study Plan, along
       with those specified in the required analytical protocol, with the exception of the bullet
       listed below.

       A minimum of seven analyses and a maximum of nine analyses will be required for each
       of the two  MDL studies (IPRs  are not included).   The Study Plan states that if the
       resulting MDL(s) does  not meet the study plan criteria of being equal to or within a
       factor of five lower than the level spiked, the laboratory is to repeat the study.  For the
       purposes of this project, however, the laboratory is NOT to repeat an MDL study without
       formal modification of this purchase order by DynCorp Viar.

Analytical Results Required:

       A narrative that details any problems with or deviations from the referenced methods
       and reports problems associated with the analysis  of specific samples. The narrative
       should also provide comments on the method performance on each analyte.

       A list of samples analyzed, and  a run chronology.

       Summary reports of all sample results. An explanation of the summary format table is
       provided in the Study Plan.

       Raw data, including bench sheets, etc.

       The laboratory must report the MDL values to  three significant figures.

QC Requirements:  As specified in the Study Plan - QC Minimums provided below.

       Initial Precision and Recovery Analyses - Data quality objectives are specified in Method
       1664.

       Analysis of a method blank is required with the IPR set and with each of the two MDL
       studies.  Method blank  values must be less than 0.91 mg/L for HEM and 1.6 mg/L for
       SGT-HEM.

       NOTE:  If contamination is detected in the method blank, the source of contamination
       must be  identified and corrected.  The blank and any IPR sets or MDL studies associated
       with that contaminated blank must be reprepared and reanalyzed.

       Instrument calibration shall be confirmed per Method 1664 and Study Plan specifications.
                                       D-7                         Novembers, 1994

-------
           APPENDIX E
40 CFR136, APPENDIX B

-------
Report of the Method 1664 Validation Studies
APPENDIX B TO PART 136 - DEFINITION AND PROCEDURE FOR THE DETERMINATION OF THE METHOD
DETECTION LIMIT - REVISION 1.11

                                        Definition

 The method detection limit (MDL) is defined as the minimum concentration of a substance that
can be measured and reported with 99% confidence that the analyte concentration is greater than
zero and is determined from analysis of a sample in a given matrix containing  the analyte.

                                   Scope and Application

 This procedure is designed for applicability to a wide variety of sample types ranging from
reagent (blank) water  containing analyte to wastewater containing analyte. The MDL for an
analytical procedure may vary as a function of sample type. The procedure requires a complete,
specific, and well defined analytical method. It is essential that all sample processing steps of the
analytical method be included in the determination of the method detection limit.

 The MDL obtained by this procedure is used to judge the significance of a single measurement
of a future  sample.

 The MDL procedure was designed for applicability to a broad variety of physical and chemical
methods. To accomplish this, the procedure was made device- or instrument-independent.

                                        Procedure

 1. Make an estimate of the detection limit using one of the following:

 (a) The concentration value that corresponds to an instrument signal/noise in the range of 2.5
to  5.

 (b) The concentration  equivalent of three times the standard deviation of replicate instrumental
measurements of the analyte in reagent water.

 (c) That region of the standard curve where there is a significant change in sensitivity, i.e., a
break in the slope of the standard curve.

 (d) Instrumental limitations.

 It is recognized that the  experience of the analyst is important to this process. However, the
analyst  must include the above considerations in the initial estimate of the detection limit.

 2. Prepare reagent (blank) water that  is  as free of analyte as possible. Reagent or interference
free water is defined as a water sample in which analyte and interferent concentrations are not
detected at the method detection limit of each analyte  of interest. Interferences are defined as
systematic  errors in the measured analytical signal of an established procedure caused by the
presence of interfering species (interferent). The interferent concentration is presupposed to be
normally distributed in representative samples  of a given matrix.

April 1995                               E-2

-------
                                                     Report of the Method 1664 Validation Studies
 3. (a) If the MDL is to be determined in reagent (blank) water, prepare a laboratory standard
(analyte  in reagent  water)  at  a concentration  which is at least equal to  or  in  the  same
concentration range as the estimated method detection limit. (Recommend between 1 and 5 times
the estimated method detection limit.) Proceed to Step 4.

    (b) If the  MDL is to be determined in another sample matrix, analyze the sample.  If the
measured level of the analyte is in the recommended range of one to five times the estimated
detection limit, proceed to Step 4.

 If the measured level of analyte is less than the estimated detection limit, add a known amount
of analyte to bring the level of analyte between one and five times the estimated detection limit.

 If the measured level of analyte is greater than  five times the estimated detection limit, there
are two options.

 (1) Obtain another sample with a lower level of analyte in the same matrix if possible.

 (2) The sample may be used as is for determining the method detection limit if the analyte level
does not exceed 10 times the MDL of the analyte in reagent water. The variance of the analytical
method changes  as the  analyte concentration  increases  from the MDL,  hence  the  MDL
determined under these circumstances may not truly reflect method variance at lower analyte
concentrations.

 4. (a) Take a  minimum of seven aliquots of the sample to be used to calculate the method
detection limit  and process each through the entire analytical method. Make all computations
according to the defined method with final results in the method reporting  units.  If a  blank
measurement is required to calculate the measured  level of analyte, obtain a separate  blank
measurement for each sample aliquot analyzed. The average blank measurement is subtracted
from the respective sample measurements.

    (b) It may be economically and technically desirable to evaluate  the estimated method
detection limit before proceeding with 4a. This will: (1) Prevent repeating this entire  procedure
when the costs of analyses are high and (2) insure that the procedure is  being  conducted  at the
correct concentration. It is quite possible that an inflated MDL will be calculated  from data
obtained at many times the real  MDL even though the level of analyte is less than five times the
calculated method detection limit. To insure that the estimate of the method detection limit is
a good estimate, it is necessary to determine that a lower concentration of analyte will not  result
in  a significantly lower method detection limit. Take two aliquots of the sample  to be used to
calculate the method  detection limit and process each through the entire method, including blank
measurements as described above in 4a. Evaluate these data:
  (1) If these measurements indicate the sample is in desirable range for determination of the
MDL, take five additional aliquots and proceed. Use all seven measurements for calculation of
the MDL.


                                        E-3                             April 1995

-------
Report of the Method 1664 Validation Studies
  (2) If these measurements indicate the sample is not in correct range, reestimate the MDL,
obtain new sample as in 3 and repeat either 4a or 4b.

  5. Calculate the variance (S2) and standard deviation (S) of the  replicate measurements, as
follows:

1
1
n - 1

n
EY2
A,
i=lS
n
(X, X )
i=l
n
                       S2 =  ——  £  X,2 - -^	  S= (S2)1/2
Where:

Xi; 1=1 to n, are the analytical results in the final method reporting units obtained from the n
sample aliquots and S refers to the sum of the X values from 1=1 to n.

 6. (a) Compute the MDL as follows:

                                  MDL = t(n_u_0.99) (S)

where:

 MDL = the method detection limit

 t(nua = .99) = the students t value appropriate for a 99% confidence level and a standard

 deviation estimate with n-1 degrees of freedom. See Table.

 S = standard deviation of the replicate analyses.

   (b) The 95% confidence interval estimates for the MDL derived in 6a are computed according
to  the following equations derived from percentiles of the chi square over degrees of freedom
distribution (xVdf).

 LCL = 0.64 MDL

 UCL =  2.20 MDL

 where:  LCL and UCL are the lower and upper 95% confidence limits respectively based on
seven aliquots.
  7. Optional iterative procedure to verify the reasonableness of the estimate of the MDL and
subsequent MDL determinations.

April 1995                               E-4

-------
                                                      Report of the Method 1664 Validation Studies
  (a) If this is the initial attempt to compute MDL based on the estimate of MDL formulated in
Step 1, take  the MDL as calculated in Step 6,  spike the matrix at this calculated MDL  and
proceed through the procedure starting with Step 4.

  (b) If this is the second  or later iteration of the  MDL calculation, use S2 from the current MDL
calculation and  S2 from  the previous MDL calculation to compute the F-ratio. The F-ratio is
calculated by substituting the larger S2 into the numerator S2 A and the other into the denominator
S2B The computed F-ratio is then compared with the F-ratio found in the table which is 3.05 as
follows: if S2A/S2B>3.05, then compute the pooled standard deviation by the following equation:
                                               -    6S
                                    pooled
                                                      *i
                                                12
 if S2A/S2B>3.05, respike at the most recent calculated MDL and process the samples through the
procedure starting with Step 4. If the most recent calculated MDL does not permit qualitative
identification when samples are spiked at that level, report the MDL as a concentration between
the current and previous MDL which permits qualitative identification.

  (c)  Use the Spooled as calculated in 7b to compute the final MDL according to the following
equation:

                                   MDL=2.681  (Spooled)

where 2.681 is equal to t(12 j_a = 99)

  (d) The 95% confidence limits for MDL derived in 7c are computed according to the following
equations derived from percentiles of the chi squared over degrees of freedom distribution.

  LCL=0.72 MDL

  UCL=1.65 MDL

where LCL and  UCL are the lower and upper 95% confidence limits respectively based on 14
aliquots.
                                        E-5                              April 1995

-------
Report of the Method 1664 Validation Studies
TABLES OF STUDENTS' T VALUES AT THE 99 PERCENT CONFIDENCE LEVEL
Number of
replicates
7
8
9
10
11
16
21
26
31
61
00
Degrees of freedom
(n-1)
6
7
8
9
10
15
20
25
30
60
00
t cn-l' .99)
3.143
2.998
2.896
2.821
2.764
2.602
2.528
2.485
2.457
2.390
2.326
                                       Reporting

 The analytical method used must be specifically identified by number or title and the MDL for
each analyte expressed  in the appropriate  method reporting units. If the analytical method
permits options which affect the method detection limit, these conditions must be specified with
the MDL value. The sample matrix used to determine the MDL must also be identified with
MDL value. Report the mean analyte level with the MDL and indicate if the MDL procedure was
iterated. If a laboratory standard or a sample that contained a known amount analyte was used
for this determination, also report the mean recovery.

       If the level of analyte in the sample was below the determined MDL or exceeds 10 times
the MDL of the analyte  in reagent water, do not report a value for the MDL.

[49 F.R. 43430, Oct. 26, 1984; 50 F.R. 694, 696, Jan. 4, 1985, as amended at 51 F.R. 23703, June 30,
1986]
April 1995
E-6

-------
 APPENDIX F
REFERENCES

-------
Report of the Method 1664 Validation Studies
                                                                            APPENDIX F

                                                                           REFERENCES
              Interlaboratory Validation of U.S. Environmental Protection Agency Method 1625A, July
              1984.  Prepared by SRI  International, 333 Ravenswood Avenue,  Menlo  Park,
              California 94025, (415) 326-6200.  Available  from the Sample Control Center
              (operated by DynCorp Environmental), 300 N. Lee Street, Alexandria, VA 22314,
              (703) 519-1140.

              Method 1664:  N-Hexane Extractable Material (HEM) and Silica Gel Treated N-Hexane
              Extractable Material (SGT-HEM) by Extraction and  Gravimetry (Oil and Grease and
              Total Petroleum Hydrocarbons), April 1995, Document No. EPA-821-B-94-004b, EPA
              Water Resource Center, Mail Code RC-4100, 401 M Street, S.W.,  Washington, B.C.
              20460.

              Methods for Chemical Analysis of Water and Wastes, 3rd Edition;  Environmental
              Protection Agency,  Environmental Monitoring Systems Laboratory-Cincinnati
              (EMSL-Ci): Cincinnati, Ohio 45268, EPA-600/4-79-020, 1983; Method 413.1.

              Report of EPA Efforts to Replace Freon for the Determination of Oil and  Grease:  Phase
              II, April 1995, Document No. EPA-820-R-95-003, EPA Water Resource Center, Mail
              Code RC-4100, 401 M Street, S.W., Washington, B.C. 20460.

              Standard Methods for the Examination of Water and Wastewater,  18th Edition;
              American Public Health Association: 1015 Fifteenth Street, NW,  Washington, B.C.
              20005, 1992, Method  5520B.

              Youden, W.J., "Ranking Laboratories by Round-Robin Tests",  Materials Research
              and Standards 3, 13-17, 1963.
April 1995                                 F2

-------
                           ADDENDUM TO SECTION 3
DEVELOPMENT OF QUALITY CONTROL ACCEPTANCE CRITERIA
                                   JANUARY 1996

-------
                      Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996



                                                            ADDENDUM TO SECTION 3

                        DEVELOPMENT OF QUALITY CONTROL ACCEPTANCE CRITERIA

                                                                      JANUARY 1996
   Section 3 of the Method 1664 validation study report details the development of the
quality control (QC) acceptance  criteria that were published in the April 1995 version of
Method 1664, and that were included in the version of Method 1664 incorporated by
reference into the proposal scheduled for publication in the Federal Register in January of
1996.  After the preamble and regulatory language had been submitted for publication, and
after it was too late to recall the proposal of Method 1664, it was noted that some of the data
sets used to derive the QC acceptance criteria required modification.  These discrepancies
consisted of several occurrences of replicate values of the same results and two missing
results.  Also, data that were received too late for inclusion in the original QC derivations
were included in the  revised data set, and an alternate statistical approach was used for
generation of the IPR and OPR acceptance criteria. This addendum gives details of changes
resulting from the modifications to the data sets used for statistical analyses and provides re-
derived QC acceptance  criteria based on the revised data sets.

Replicate and Missing  Data

   A close inspection of the data sets used to establish the ongoing precision and recovery
(OPR) QC acceptance criteria for both HEM and SGT-HEM showed that they contained
replicate values for what should have been a single result, and were each missing one OPR
result.  The database  was corrected by removing the replicate results and adding the missing
result.

Late Data

   A set of HEM data from the TCRR study were received from a laboratory that had not
submitted results in time for inclusion in the database that was used for development of the
original QC acceptance  criteria.  In addition, an SGT-HEM OPR result from the TCRR
laboratory that performed this analysis was added to the SGT-HEM OPR data set.
Originally, this datum had not been  included.  These additions served to further broaden the
data set.

Alternate Statistical Analysis

   The IPR and OPR QC acceptance criteria in the April 1995  version of Method 1664, and
the version incorporated by reference into the proposal scheduled for publication in January
of 1996, were each constructed using the respective IPR or OPR data that had been provided
by the Twin Cities Round Robin (TCRR) study and Phase II of EPA's Freon Replacement
Study. Because an OPR sample is the equivalent of each of the four samples that make up
an IPR set, an alternate, and equally acceptable approach would be to  pool the  IPR and OPR
data.  The advantage of this approach is that it would provide a broader data set and greater

                                     AA-1                            January 1996

-------
Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996


statistical power for calculation of the variance components. These variance components are
factors in the equations used to determine the IPR and OPR QC acceptance criteria limits for
both recovery and precision.  For derivation of the revised QC acceptance criteria given in
this addendum, the variance components were recalculated using the pooled data set.
Effect of Changes to the Database

   With the exception of the SGT-HEM IPR precision specification, the SGT-HEM OPR
upper recovery limit, and the OPR-based matrix spike (MS) and matrix spike duplicate
(MSB) relative percent difference (RPD) limit, the recalculated values are not substantially
different from those published in the April 1995 version of Method 1664. These recalculated
values are not in the version of Method 1664 incorporated by reference into the proposal
scheduled for publication in January 1996.  At the close of the comment period on the
proposal, EPA will consider comments as well as other data and, depending upon the
conclusions reached, may find it appropriate to revise one or more of the specifications in
Method 1664.

   The remainder of this addendum is a revision of those subsections within Section 3 of the
validation study report that required modification to reflect the recalculations. Those
subsections that did not require revision have not been reprinted here.

3.1 Data Validation and Statistical Analysis

3.1.2   QC Acceptance Criteria for Hexane Extractable Material

   Criteria were developed for the IPR, OPR,  and matrix spike/matrix spike duplicate
(MS/MSD) tests for both HEM and SGT-HEM. For HEM, the IPR and OPR acceptance
criteria were constructed using an analysis of variance (ANOVA) with laboratories as a
random variance component. For each QC criterion, the  mean result across laboratories was
determined. The IPR and OPR data were pooled, and the interlaboratory and intralaboratory
variance components were  calculated from the consolidated data. The individual IPR and
OPR variability values were then calculated from these interlaboratory and intralaboratory
variance components using a formula appropriate to the  particular quality control test. The
95 percent cutoff value from the Student's t distribution was determined based on the
appropriate degrees of freedom.  The degrees of freedom are dependent on the number of
laboratories, the number of sample analyses, and the variance components. The QC limit
was derived by multiplying the t value by the variability, and subtracting this value from the
mean result for the lower limit, or adding this value to the mean result for the upper limit.
Details of the equations used to derive these limits are presented in the document titled
Interlaboratory Validation of U.S. Environmental Protection Agency Method 1625A, July 1984.

3.1.3   QC Acceptance Criteria for Silica Gel Treated Hexane  Extractable Material

   For SGT-HEM, EPA received  results from only two laboratories, one from the TCRR
study and one from the Phase II study. EPA used these  data to construct the revised IPR
and OPR QC acceptance criteria.

3.1.4   QC Acceptance Criteria for Matrix Spike and Matrix Spike Duplicate
January 1996                             AA-2

-------
                       Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996


    For HEM and SGT-HEM, the criteria for recovery of a matrix spike (MS) or matrix spike
duplicate (MSB) and for the relative percent difference between an MS and an MSB were
derived from the OPR criteria, since neither the TCRR study nor the Phase II study required
the spiking of field samples.  EPA believes that this application of OPR criteria to MS/MSB
samples is acceptable because the determinative technique in Method 1664 is gravimetry,
which is not susceptible to interferences, and because nearly all of the treated effluents to
which Method 1664 is to be applied in monitoring will be similar to the reagent water used
in the OPR tests. This transfer of data for development of specifications for  acceptance
criteria is similar to that which EPA used in the organic methods that are  promulgated at 40
CFR 136, Appendix A and in other  methods.

    Betermination  of the relative percent difference (RPB) criteria consisted of setting the
limit at approximately one-half of the range between the  lower recovery limit and the upper
recovery limit. This estimation establishes the RPB limit at 14.4 percent for HEM and 33.4
percent for SGT-HEM. These limits are considered to be a reasonable first approximation of
method performance in  the absence of MS/MSB data. If MS/MSB data become available, it
may be appropriate to revise these limits.

3.2 Results

    Few laboratories in the interlaboratory study encountered difficulties with the analysis of
IPR and OPR samples, and most achieved acceptable recoveries of hexadecane and  stearic
acid.  Statistical  evaluation of the results from all thirteen laboratories produced few outliers,
and the extreme rank sum test showed all laboratories to be equivalent. This indicates that
Method 1664 is a reproducible procedure sufficiently reliable to be used by a variety of
laboratories.

Results are summarized in Addendum Table 2 below.  Individual laboratory results and the
statistical analyses of these data are presented in Addendum Tables 3 through 12.
                                       AA-3                             January 1996

-------
Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996
                                    Addendum Table 2




                95 Percent Confidence Limits for Method 1664 QC Criteria
Criterion
IPR
HEM Precision
HEM Recovery
SGT-HEM Precision
SGT-HEM Recovery
MS/MSD
HEM Recovery
HEM RPD
SGT-HEM Recovery
SGT-HEM RPD
OPR
HEM Recovery
SGT-HEM Recovery
95% Lower Limit (%)


85.4

86.3

78.0

64.9


78.0
64.9
95% Upper Limit (%)

10.7
97.8
27.9
101.8

106.7
14.4
131.7
33.4

106.7
131.7
January 1996
AA-4

-------
                      Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996
3.3 Discussion and Conclusions

    As a check on the validity of the revised QC acceptance criteria, IPR and OPR results
from the Phase II study and the TCRR study were compared with the respective criteria.  For
the IPR test of HEM, four of the mean percent recovery values did not meet the IPR recovery
criterion, and only one of the 13 IPR sets failed the IPR precision criterion.  In all cases, the
IPR recoveries that did not meet the lower or upper limits were within 3 percent of that limit.
For the OPR test, only three of the 22  results failed the OPR recovery criterion, one of which
was within 0.09 percent of the limit. For the IPR test of SGT-HEM, one of the two
laboratories did not meet the recovery criterion (3 percent outside the limit), and both
laboratories met the precision criterion.

    Upon comparing these recalculated QC acceptance criteria to the criteria given in the
version of Method 1664 being proposed, it is evident that the majority of the QC limits are
similar (Addendum Table 13B). The most obvious exceptions are the SGT-HEM IPR
precision criteria, the SGT-HEM OPR upper recovery limit, and the OPR-based SGT-HEM
MS/MSD RPD limit.  When determining those criteria presented in Section  3 (and published
in the proposed version of Method  1664), the IPR and OPR data were considered separately.
This approach is acceptable because an OPR analysis,  though considered to  be the equivalent
of one of four IPR analyses,  is performed at a frequency of one per analytical batch. IPR
analyses are performed  at the same time and should theoretically be less susceptible to the
variability that would occur were the samples prepared days apart,  as they  are with the OPR.

    Because each individual IPR analysis is performed in exactly the same way as each OPR
analysis, it is also acceptable to consider these analyses to be equivalent, and it is therefore
acceptable to pool data from both analyses to derive the variance components.  The
advantage to this latter approach is that this will provide a broader data set, which is
especially important given the limited number of laboratories that performed SGT-HEM
analysis. The effect of this pooling  is a widening of the specifications for SGT-HEM.  Even
when pooled, however,  the SGT-HEM data set is limited and may not provide an adequate
representation of interlaboratory effects. If more data become available, it may be necessary
to re-examine these specifications.

    Regardless of the approach to data pooling, nearly all results produced by the
contributing laboratories are within the QC acceptance criteria contained in  the April 1995
version of Method 1664 that is to be proposed, and demonstrate that the QC acceptance
criteria are realistic and reflect the performance of Method 1664 on both reagent water and
on real world sample matrices.  For this reason, and because the April 1995  version of
Method 1664 has been in circulation for more than nine months with an almost unanimously
favorable response, EPA intends to  retain the QC acceptance criteria currently given in the
April 1995 version of Method 1664,  which is to  be incorporated by reference into 40 CFR part
136 by the proposal scheduled to be published in the Federal Register in January of 1996.
                                      AA-5                            January 1996

-------
      Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996


Lab
1
2
3
4
5
6
7
8
9
10
11
12
13



% Rec.
IPR 1
98.8
78.7
87.5
90.3
93.9
86.8
82.3
91.0
85.0
87.5
89.5
107.5
95.1



% Rec.
IPR 2
103.8
103.5
92.5
87.5
97.9
92.0
83.8
103.5
93.3
77.5
89.3
110.0
94.9
Std. Deviation of
Addendum Table
HEM IPR Data
% Rec.
IPR 3
96.3
105.7
85.0
89.0
82.2
92.5
81.5
94.5
89.3
87.5
92.3
95.0
97.1
Mean % Recovery
Mean % Recoveries
3

% Rec.
IPR 4
98.3
83.2
87.5
89.3
90.1
93.8
86.5
95.3
102.0
82.5
89.0
90.0
97.8
Across Labs
Across Labs
Mean Std. Deviation
Std. Deviation of the Std. Deviations


Mean
% Rec.
99.3*
92.8
88.1
89.0
91.0
91.3
83.5*
96.1
92.4
83.8*
90.0
100.6*
96.2
91.9
5.3
Across Labs
Across Labs


Standard
Deviation
3.2
13.8*
3.1
1.2
6.7
3.1
2.2
5.3
7.2
4.8
1.5
9.7
1.4


4.9
3.7
Results that failed the recovery or precision specification.
      January 1996                                AA-6

-------
     Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996
                 Addendum Table 4




HEM IPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
85.4
95% Upper
Limit
97.8
No. of Mean % Rec.
Values Below 95% Lower
Limit
2
No. of Mean % Rec.
Values Above 95% Upper
Limit
2
                 Addendum Table 5




      HEM IPR Data - Upper Limit for Precision
95% Upper Limit
10.7
No. of Standard Deviation
Values Above 95% Upper
Limit
1
                     AA-7
January 1996

-------
Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996


                                    Addendum Table 6

                                      HEM OPR Data
                                          Lab        % Recovery
                                             1              83.0
                                             2              80.8
                                             3              82.5
                                             4              87.8
                                             5              91.0
                                             6              98.8
                                             7              89.8
                                             8              89.8
                                             9              82.5
                                           10              91.0
                                           11              99.5
                                           11              91.5
                                           11              91.0
                                           11             112.0*
                                           11              93.8
                                           11              93.3
                                           11             101.8
                                           11              89.8
                                           11             111.8*
                                           11              94.0
                                           11             106.8*
                                           12              92.2
                Mean % Recovery Across Labs              93.4
    Standard Deviation of % Recoveries Across               8.7
                                         Labs
January 1996                             AA-8

-------
                 Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996


                             Addendum Table 7
            HEM OPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
78.0
95%
Upper
Limit
106.7
No. of % Rec. Values
Below 95% Lower
Limit
0
No. of % Rec. Values
Above 95% Upper
Limit
3
Results that failed the recovery specification.
                             Addendum Table 8
                             SGT-HEM IPR Data
Lab
1
2
% Rec.
IPR 1
110.0
105.0
% Rec.
IPR 2
115.0
100.0
% Rec.
IPR 3
80.0
95.0
% Rec.
IPR 4
115.0
65.0
Mean
% Rec.
105.0*
91.3
Standard
Deviation
16.8
18.0
                        Mean % Recovery Across Labs       98.1
       Std. Deviation of Mean % Recoveries Across Labs        9.7
                                      Mean Std. Deviation Across Labs
                       Std. Deviation of the Std. Deviations Across Labs
         17.4
         0.8
                                 AA-9
January 1996

-------
Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996











                                  Addendum Table 9




               SGT-HEM IPR Data - Upper and Lower Limits for Recovery
95% Lower
Limitl
86.3
95% Upper
Limit
101.8
No. of Mean % Rec.
Values Below 95% Lower
Limit
0
No. of Mean % Rec.
Values Above 95% Upper
Limit
1
 Result that failed the recovery specification.
                                  Addendum Table 10




                     SGT-HEM IPR Data - Upper Limit for Precision
95% Upper Limit
27.9
No. of Standard Deviation
Values Above 95% Upper
Limit
0
 Result that failed the recovery specification.
January 1996
AA-10

-------
                   Report of the Method 1664 Validation Studies - Addendum to Section 3, January 1996




                              Addendum Table 11

                              SGT-HEM OPR Data
                                    Lab       % Recovery
                                       1             88.0
                                       1             82.0
                                       1             76.0
                                       1             75.5
                                       1            100.0
                                       1            107.5
                                       1             80.0
                                       1             79.0
                                       1             96.0
                                       1            100.0
                                       1            111.0
                                       2             75.0
           Mean % Recovery Across Labs
Standard Deviation  of % Recoveries Across
                                    Labs
89.2
13.1
                              Addendum Table 12

           SGT-HEM OPR Data - Upper and Lower Limits for Recovery
95% Lower
Limit
64.9
95% Upper
Limit
131.7
No. of % Rec. Values
Below 95% Lower
Limit
0
No. of % Rec. Values
Above 95% Upper
Limit
0
                                  AA-11
              January 1996

-------
Report of the Method 1664 Validation Studies
                                                          Addendum Table 13B

                                                   Comparison of Method 1664 Criteria



Published
Recalculated
HEM IPR
95% Lower
% Rec.
83.0
85.4
95% Upper
% Rec.
100.7
97.8

Precision
10.9
10.7
HEM OPR
95% Lower
% Rec.
79.0
78.0
95% Upper
% Rec.
113.9
106.7
HEM MS/MSD
Lower
% Rec.
79.0
78.0
Upper
% Rec.
113.9
106.7
RPD

17.5
14.4



Published
Recalculated
SGT-HEM IPR
95% Lower
% Rec.
83.2
86.3
95% Upper
% Rec.
116.0
101.8

Precision
13.3
27.9
SGT-HEM OPR
95% Lower
% Rec.
65.8
64.9
95% Upper
% Rec.
105.7*
131.7
SGT-HEM MS/MSD
Lower
% Rec.
65.8
64.9
Upper
% Rec.
105.7*
131.7
RPD

24.0
33.4
*These calculated values were widened to the limits generated from HEM, as explained in Section 3.1.3 of the validation study report.  This change
also applies to the MS/MSD  RPD calculation.
       January 1996
AA-12

-------