NC0004685_Staff Comments_20010709

Re: oiI/g><<3Qe meth. 1664
/(/C
Subject: Re: oil/grease meth. 1664
Date: Mon, 09 Jul 2001 13:31:04 -0400
From: Dana Satterwhite <[e-mail]>
Organization: NC DENR DWQ Chemistry Laboratory
To: Susan Wilson <[e-mail]>
Susan -
Traditionally, the non -polar material (such as petroleum hydrocarbons)
measured using the SGT-HEM methodology is analyzed using other organic
methods (i.e. TPH by modified 8015 or California LUFT method). The HEM
procedure measures both non -polar plus polar materials, which include
vegetable oils, animal fats, waxes, greases, etc. The HEM procedure was a
replacement procedure for the freon extractable oil and grease method.
Did that help any?
Dana
Susan Wilson wrote:
> ok, I'm back to oil/grease now. Dana, in reviewing some of the info.
> you gave me it appears that the silica gel treatment - hexane
> extractable material (SGT-HEM) is part of the approved methodology?? i
> really can't tell - can someone clear me up on that? i can't tell if
> they want a facility to do some side -by -side testing with SGT and
> without SGT. anybody know what the deal is with that?
> i thought you guys were saying they would need a variance from EPA in
> order to use the SGT method? Thanks for any help!
Dana Satterwhite <[e-mail]>
QA/QC Coordinator
NC DWQ CHEMISTRY LABORATORY
Department of Environment and Natural Resources
-7'/'l
V' T)Aaa S.
POP -
A
I of 1 7/9/01 1:45 PM

Re: oil/grqu+e method 1664
or
Subject: Re: oWgrease method 1664
Date: Fri, 06 Jul 2001 17:36:56 -0400
From: [e-mail]
To: Susan Wilson <[e-mail]>
Susan,
Certain treatment to wastewater can affect results. There is an EPA
Guidance document that can give you some help. It is titled Analytical
Method Guidance for EPA Method 1664A Implementaion and Use (40 CFR Part
136), EPA/821-B-=99-xxx, January 2000. If you want to contact Bill
Telliard office for inforamtion on an aternate in special circumstances, a
name to start with would be Maria Gomez -Taylor who works for Telliard. The
alternate would have to come through the Region but you could find out your
options since his office gives the opions on alternates. I will be out of
the office next week but will return on the July 16. I will be glad to
discuss with you then. Maria's number is 202-260-1639.
Wayne Turnbull
SptD Sri fIAD 5?oK�� t>�FPR
O,a1 MCrrfo�. Tti�4 FIG( WE
T '1 �fl -(a zv ��
MM oFPKl Tke 5GT ��tlo� �I
TNc- W.*AA) .m�, rNPc7 AIJ MPhs �tvwt i
PJO-k r 1u �7Ko� oleo E,nbu�rK , Ds F
T+ J rNk 7. )
1 Pr• flAafiSlt4� �� D4N4!•
I of 1 7/9/01 9:22 AM

i
LI&f ru�sT PEn.a �- �koy i3rsrn o� t^firav Z57t c'/
Mm Mho ,uamasr % sA,a"
EAl
%Q(NK�is �[LsO. r(l�s Su2�itGik�`li - �9s l✓e2G �� �
i
0-
A
ft&,-;.JFr &Q-•
VA
M

Y
511 e!f ' - � �
waNj Ar sPLc %, 6Cse 1460N
1� z yar 5 c,41v Po 5r7r- Br
/�&I AIW,4,V�
%lr- Ta 6 �
PVVA- c.4y5 NQ-VZ47 Saw6 RWAFU44iiS
L--7 A90
lF !m R �*o
€ l
ty anrC-r
U9-uf
0'2 " "C> f7�/tfi sc
y7AII> RIA4r 7yf AG Fu efr�,�.�T
5 0� GAS
#W
g,,,,�/(JYprulJlr GC�eLa• r3cs��c��7s
—Ague C)r6rrcteLT f}i✓�Urs� s �/eyA�v� u4!✓ 2 7ri

..
N��rk� - -

I9su� `i"%I `r g7
r n G LASS AT
MFa-T[pl F, 3 )
" l3nvoC� " , tir s v n rWv't t?uAt- i ii37 /�5
NoY'� X.1'�� Fzau, ru/4J9cc5 �Cnr� nlS L *Z(
D,41
(?dICjCNnC Of </ LI/ht T /��'/ �A✓�
p✓r'2 tF5 S� Jc :/ 76.
� r,CGy ,� cam; � �� � � �� . h� oi� � �,e� sc ,4.✓,4Ly s rs
(11wevc.9L'(, of y CAwc-. twat- Wit" "AtTs)
G>4�5 CAAN7 u5C !�N-ili�Csz�os
AX41- 6.
CotaC�C7'�i���`o �7t SP[t i S�7t 4E5 �Vy" q� 3.1 Low T 41V /66¢) - au,—
G7tlt,j pcvaoo A- Ca¢xAAT/ei✓ FAC-7a2.

PPG Oil apd Grease
11
Subject: PPG Oil and Grease
Date: Fri, 25 May 2001 11:22:57 -0400
From: Dana Satterwhite <[e-mail]>
Organization: NC DENR DWQ Chemistry Laboratory
To: Susan A Wilson <[e-mail]>
Susan -
I asked around and didn't get much feedback on the issue. Everyone fell
back on ...consult EPA.
I am curious to know what they said or what you found out. If you can
find an extra minute, would you mind filling me in?
Sorry I wasn't much help.
Dana
Dana Satterwhite <[e-mail]>
QA/QC Coordinator
NC DWQ CHEMISTRY LABORATORY
Department of Environment and Natural Resources
I of 1 5/25/01 11:30 AM

GV iZRer-kC_�_
PPG Industries, 940 Washburn Switch Road, Shelby, N.C. 28150
Lacy Ballard, R. E. M.
Staff Engineer, Environmental
Fiber Glass Products
Mr. Dave Goodrich
NCDENR, DWQ
Permitting Division
1617 Mail Service Center
Raleigh, NC 27699-1617
MAR 14 2001
L
WAifQ C'�:.11.ITY
ch S, Q k^. art r a f H
Re: Request for NPDES Permit Modification for Oil and Grease
PPG Industries - Shelby, NC - NPDES No. NC0004685
Dear Mr. Goodrich:
704-434-2261
Extension 544
[e-mail]
The purpose of this letter is to provide an update of the testing described in PPG
Industries' (PPG) June 5, 2000 letter sent to NCDENR, and to request a modification
to PPG's NPDES permit limits. The June 5' testing program was designed to
investigate the differences in methods for determining oil and. grease in the PPG
wastewater treatment plant final effluent.
This letter summarizes the recent inconsistency in oil and grease results in the PPG
final effluent and the inadequacy of the new hexane oil and grease method for
measuring oil and grease in the PPG final effluent. Furthermore, PPG requests a
permit modification for oil and grease analysis, which is more appropriate for this type
of wastewater.
Background
PPG is a fiberglass manufacturer located in Shelby, NC. Process wastewater from our
production units is treated in an on -site WWTP before being discharged to Brushy
Creek. The waste treatment system is a highly effective system that provides optimum
treatment and maintains high effluent quality. It includes primary treatment with the
addition of lime, an aluminum -based coagulant and a polymer. In the secondary
biological treatment system, sodium bentonite (clay) is added to the mixed liquor to
remove refractory organic compounds and ferric chloride is added prior to the
secondary clarifiers to improve settling. It is expected that these processes would
significantly reduce hydrocarbon oil and grease to very low or non -detectable levels.
This has been shown in our historical effluent oil and grease data. The NPDES
discharge permit limits effluent oil and grease to a daily maximum of 15 mg/L and a
maximum monthly average of 10 mg/L. Historically, the oil and grease effluent
concentrations have been well below PPG's permitted discharge limits.

Page 2
As of January 1, 2000, CFC-113 (Freon) is no longer being imported into or produced
in the United States. Additionally, analytical laboratories will no longer be allowed to
use any existing Freon after the year 2005. In April 1998, the analyzing laboratory
(Pace Analytical Labs, Huntersville, NC) switched from the EPA Method 413.1, which
uses freon as the extracting solvent, to the new EPA Method 1664, which uses n-
hexane (hexane) as the extracting solvent in anticipation of this action. Since the
switch, PPG effluent oil and grease concentrations have occasionally been higher than
normal. Historical effluent oil and grease data is presented in Table 1.
Both EPA Methods 413.1 and 1664 were developed for the determination of relatively
non-volatile hydrocarbons, vegetable oils, animal fats, waxes, soaps, greases, and
related materials, commonly referred to as total oil and grease. Under EPA Method
413.1, the freon extracted constituents were referred to as "oil and grease". Under
EPA Method 1664, the hexane -extracted constituents are referred to as "hexane
extractable material," implying that Method 1664 measures not only oil and grease, but
any material that can be extracted using hexane. This is problematic for two reasons.
First, hexane has a higher affinity for extracting petroleum products, compared to
Freon which has a higher affinity for extracting natural fats and oils. Second,
surfactants appear to be more readily extracted by hexane than Freon, which can cause
higher results in Method 1664 compared to Method 413.1. This was discussed in
PPG's letter to you dated June 5, 2000. PPG uses a significant amount of surfactants
in the production process. Hexane extracted material treated with silica gel may reduce
interference from materials such as surfactants but does not interfere with the
measurement of petroleum products. In order to reduce the interference from
surfactants, it was recommended in phone conversations with the EPA that hydrocarbon
oil and grease be analyzed instead of total oil and grease. Method 1664 measures total
oil and grease. In order to measure the hydrocarbon portion of oil and grease, the
hexane -extracted material in Method 1664 is treated with silica gel, which removes
polar organic material and is expected to remove surfactants. Since hydrocarbon oil
and grease is nonpolar, it should not be affected by the silica gel treatment.
Oil and grease is a method -defined parameter; therefore, any change in method
protocol, such as use of another solvent, has the potential to affect results. The EPA
Guidance Document, Analytical Method Guidance for EPA Method 1664A
Implementation and Use (40 CFR part 136), indicates that there may be differences in
results obtained using Method 1664 compared to Method 413.1. This Guidance
Document outlines recommendations for performing side -by -side testing between
Method 413.1 and Method 1664. Based on the EPA method of side -by -side testing, a
conversion factor is developed. This conversion factor can be applied to oil and grease
results obtained by Method 1664 (hexane extraction) in order to determine the expected
result that would be obtained by Method 413.1 (Freon extraction).

t Page 3
Initial Testing and Results
In order to determine if PPG's higher than normal effluent oil and grease
concentrations were a result of the change in the analytical method, split samples were
analyzed using EPA Methods 413.1 and 1664. This testing was performed on several
final effluent samples from PPG's WWTP. The results of this initial testing showed
that Method 413.1 consistently resulted in lower oil and grease concentrations than
Method 1664. However, these results suggested that there was not a consistent
correlation between the two methods and that a representative correlation factor could
not be developed.
PPG then initiated a program to perform additional testing. PPG began collecting split
samples for effluent oil and grease. The first effluent sample was analyzed using
Method 1664. If Method 1664 resulted in a concentration greater than 10 mg/L, the
second sample was analyzed using Method 413.1. Effluent oil and grease data
collected from December 1999 through April 2000 is presented in Table 2. Based on
this data, it can be seen that Method 1664 typically results in a higher oil and grease
concentration. Furthermore, there is not a consistent correlation between Methods
413.1 and 1664 when determining oil and grease in PPG's wastewater. This data
shows that Method 1664 can result in an oil and grease concentration from 1.5 to 27
times as high as Method 413.1. Since there is not a good correlation between these
results, a conversion factor obtained by side -by -side testing would need to account for
the variation and inconsistency between methods.
Additional oil and grease data was obtained from Pace Analytical Labs. These data are
summarized in Tables 3 and 4 and include the quality control (QC) analyses for the
PPG samples analyzed from December 1999 to April 2000. Table 3 is a summary of
the oil and grease matrix spike results obtained by Method 1664. Table 4 is a summary
of the oil and grease DI water control spike results obtained by both Method 1664 and
Method 413.1. This spike recovery results in Table 4 show that DI water spiked with
oil and grease provides fairly consistent results for both Method 1664 and Method
413.1. The spike recovery results in Table 3 show that oil and grease recovery from
typical samples is highly variable under Method 1664. This type of variability is
consistent with the variability in the PPG effluent oil and grease data obtained by
Method 1664.
Expanded Testing Program and Results
Based on the data presented in Tables I and 2, it appears that PPG will be subjected to
a greater risk of noncompliance for effluent oil and grease under Method 1664. In
order to address this issue while Freon was still available, PPG expanded their testing
program. This testing program was initiated the week of May 29, 2000 and is the
program described in PPG's June Sm letter. Each week, approximately six (6) split
effluent samples were collected for oil and grease analysis. Sample No. 1 was
analyzed for oil and grease by Method 1664 and Sample No. 2 served as the QC matrix
spike for Method 1664. If the oil and grease result by Method 1664 was greater than
10 mg/L, additional analyses were performed as follows: Sample No. 3 was to be
analyzed for oil and grease by Method 1664 with the addition of a silica gel treatment
and Sample No. 4 was to serve as the QC matrix spike.

Page 4
Sample No. 5 was to be analyzed for oil and grease by Method 413.1 and Sample No.
6 was to serve as the QC matrix spike.
The purpose of this testing was to help clarify some of the issues associated with the
new analytical Method 1664 and determine a more appropriate oil and grease limit for
PPG's effluent. This testing program was originally designed for a total of eight (8)
full sets of oil and grease data, which is the number of samples required for EPA's
method of side -by -side testing. However, after a five (5) month period, only one (1)
sample exceeded an oil and grease concentration of 10 mg/L by Method 1664. Further
analysis of this sample using the Freon extraction and the silica gel treated hexane
extraction showed that no detectable quantity of oil and grease was present, as shown in
Table 5.
Although there is only one data set, this information is very important. It shows that
although Method 1664 indicated that oil and grease was present at 23.9 mg/L, the
material that resulted in the high reading was neither a hydrocarbon nor a natural oil or
grease. Method 413.1 resulted in no detectable oil and grease, which indicates that
there was no natural oils or greases in the sample. Method 1664 with a silica gel
treatment resulted in no detectable oil and grease, which indicates that there was no
petroleum hydrocarbons in the sample. Therefore, the high oil and grease
measurements are likely caused by a surfactant or other constituent that is neither
hydrocarbon nor natural oil or fat.
The program has been suspended due to the time, expense and effort required for
sample collection. Because the treatment facility operates very effectively, there is no
outward indication of when a high oil and grease measurement may occur. The data
show that even with hexane, the frequency of high oil and grease measurements is low.
Although Method 1664 does not consistently result in high effluent oil and grease
levels, the method does put PPG at risk for non-compliance.
Guidance Document Summary
As outlined in USEPA's Interim Guidance for Performance -Based Reductions of
NPDES Permit Monitoring Frequencies, it is within the power of the NPDES
authorities to grant relief to regulated facilities that have established a history of both
compliance and an ability to consistently reduce the pollutants in their discharge below
the levels needed to meet their permit requirements. This relief can come in the form
of reduced reporting and monitoring requirements when a facility has demonstrated
long term compliance for a given parameter. The following steps are to be taken when
determining if a facility is eligible for reductions:
1) Evaluation of the facility's enforcement history. The history includes any
environmentally related criminal actions, and any NPDES/Clean Water Act
related civil judicial or administrative actions.

r
Page 5
2) Evaluation of the facility's parameter -by -parameter compliance history. In
order to be considered, a facility may not have had any Significant
Noncompliance violations in the last two (2) years, and no selected (permit -
specific) violations within the last year.
3) Application of the parameter -by -parameter performance history. This uses a
two year monthly average of effluent data to establish a long term discharge
rate. This rate can be used to correlate the monitoring frequency required.
4) The performance levels used to establish the monitoring reductions must be
maintained in order to remain eligible for these reductions.
Request for Permit Modification
PPG has demonstrated both a history of compliance and a genuine interest in
compliance with environmental regulations. Data shows that prior to the change in
methodology, PPG's effluent oil and grease levels were low to non -detectable. Based
on this information, PPG Shelby would like to request that the monitoring frequency of
oil and grease be reduced to one event every two months. Prior to the change in
laboratory methodology, PPG had a monthly average oil and grease level of
approximately 1.8 mg/L (Table 5). This level is only 18% of their NPDES maximum
monthly average limit of 10 mg/L. According to the guidelines established in the
interim report, a monthly average (for a weekly sample) that is less than 25 % of the
permit limit only needs to be monitored on a bimonthly basis.
Also, through parallel sample analysis with freon, PPG has established that the current
hexane extraction method can result in false discharge violations. The data generated
using the hexane method have shown no correlation to those produced by freon
extraction. Therefore, PPG would like to use Method 413.1 for bimonthly, reportable
oil and grease measurements. However, due to the pending depletion of Freon, PPG
would like to begin analyzing the oil and grease samples using Method 1664 with silica
gel. These samples would be monitor -only and used to establish a baseline value for
this method.
After the baseline has been established, PPG would like to review the data with
NCDENR to establish a new oil and grease limit using Method 1664 with silica gel.
We feel that the use of Method 1664 with silica gel will provide a more effective and
adequate means of monitoring the PPG effluent than the current hexane method. We
also feel that the use of silica gel in the oil and grease method will allow us to maintain
compliance without altering the intention of the test parameter, which is to determine
the quantity of non-volatile hydrocarbons, vegetable oils, animal fats, waxes, soaps,
greases, and related materials.

Page 6
Conclusions
PPG has demonstrated long-term compliance with environmental regulations. The
recent decline in the supply of freon has forced laboratories to use the hexane extraction
for the oil and grease analysis. Since the change in methodology, PPG has experienced
seemingly random spikes in oil and grease concentration. Further analysis of these
samples with freon has shown that the oil and grease levels in PPG's wastewater have
not changed. This is further supported by the consistency of both our production and
wastewater treatment operations. PPG would like to request that the oil and grease
monitoring be reduced to one (1) event every two (2) months. We would also like to
analyze oil and grease samples using both Method 413.1 and 1664 with silica gel
treatment to prevent surfactant interference. Method 413.1 would be used for reporting
purposes. Method 1664 with silica gel treatment would be used to establish a baseline
with which to establish future oil and grease limits.
We are available to discuss the proposed permit changes with you after you have had a
chance to review this information. Please call me at the above number or contact our
consultant Lili Gellner of Aware Environmental at 704-845-1697.
Sincerely,
4 rdYBalla
Attachment
cc: M. LeCroy
R. Daman
B. Blackwell
J. Buchanan
Rex Gleason, NCDENR-Mooresville
Bob Stein, AWARE Environmental Inc.
Lili Gellner, AWARE Environmental Inc.
259011003

TABLE 1
PPG HISTORICAL EFFLUENT OIL AND GREASE VALUES
1997 Effluent Data
1998 Effluent Data
1999 Effluent Data
2000 Effluent Data
Date
O&G
Monthly
Date
O&G
Monthly
Date
O&G
Monthly
Date
O&G
Monthly
(mg/L)
Avg.
(mg/L)
Avg.
(mg/L)
Avg.
(mg/L)
Avg.
(mg/L
L
m /L)
(m tL
22
9.3
01/OS/2000
T8
_.._.._..._.--
7.3
.01/07/1997__......_1
..............._3:0........
. ..
._.._01/05/1998.....
....-3__.
__2_0_.___.
_.._01105/1999
..............
......
_......_......__......_..........._._..__-._..
.....01/1a1.997..............3...............................
1___..,
,....01,/73i1.998
01%20/1.998...�
_,1
ii.....................__..__................................................................._............................_...._
Ot/19/1999
<5
01/19/2000
491.8
01/28/1997
.......,_..._.
01I28/1998
_2.
2
Ot/26/1999
<5
01/26/2000
<5 .......
.......... ......
..._......_...__._...._........
OZ!04/1997.,._.
...,..7_ ..._
........................_................................__.._..---.
OZ03/1.998.....
2
_...._.__.---
1.5
...........__.........---.........._..._.._..._......................._...._...........................__.._....
OZ01/1,999
<5
5.0
._........................--'--....._......_..._...
OZOZ2000
6.1
-.........
7.0
_._-.._.-
..
02/11/1997
.....
4
_..._3_0...._
_
OZ1011998
--
2
........_..._._..._............
02109/1999
<5
...----
02/09/2000
..-..y_.........-...
<5 _........._....
_ .......
---._..__........__..__..
..�.._.._..._............._
_
.................-'-
--.................................-_
....._<5................................_OZ1
:3.�......
....... _.....
...._OZ18/1,997..............
02/25/1997
..............................
....02/17/1.998......
02/2a1998
__��.-.
<i.
........................_OZ1611.999.....
OZ23Ii 999
<5
.... _
()00.......»..
OZ23I2000
........_........_....---_.....
10
..__............_
... .............
997
.........
_.................
.____........_........
`
..........._..............................................._.............-..............
`�.: 9...._....._0310112000......._._<..5............._5
0 .._
..............3_.
03/11/1997
_...._.._3:0._......_03/03/1.998
3
03/10/1998
--_..........:.........._03/OZ7.999.........._.5............_..
1
03/09/1.
7.6
03/0812000
<5
03/18/1997
3
03J17A998
<1
03/16/1999
<5
<5
03/25/1.997,.,.._
<1
03/23/1999
<5
.............03l15/2
03/22/2000
28(3.5� ...._...._........_...
Oa01/1997
.,-3
2..._..I...........4
.............
....-03/2a1.999
03J31/1998.._..._<'..__........__.._._.........03%30�1999.......__8.7
..........................
......03�2912000.....
<5..
.................................._.
Oa08/1997
.._.__-._.
1
................. .-
Oa07/1998
<5
6
.................. ___....._..
.----.-
_........_._...........
--........ _............._._......._...._......_........_
........
..........................
04/13/1999
<5
Oa12/2000
15
__._..........
.......................
.......-........_..............._
04/2?y1997
..---........................
<i
__................_..............
Oa21/199a
....... .............
<s
......................................_......................_.__................................................._..................
Oa2011999
<$
oat9/2000
<5
.. 04f29/1997...
_.. <1__..
............................04128/1998-....__10......
........... _..........-04i2711999
<5
Oa2612000
<5
05/O6/1997
1
1.5
05/05/1998
49
16.5
05/04/1999
<5
5.6
05/03/2000
<5
__._...__..
6.3
_......._......__.
...._.....___._.................._._............................................
OS/13/1997
2
................._............................................................_.....__._..._
OS/1Z7998
<5
OS/1111999
....._........_....................................._._.......................
<5
O5l10/2000
<5
05/20/1997.._
......_.2_................................OS/1911995
7
O511811999
c5
OS/17/2000
7.1
OS/2711997
........1_-OS/26/19
_ _
._...............................---..._......__....................................._.__.
_....._................._.........
_.._..._........._..............
_._...__...._._..........-
06/03/1997
0.095
1.3
.
06/03/1998
<5
5.0
06/01/1999
.
<5
5.9
_............................-......
06/07/2000
<5
7.5
.......... ....
__......._7
_.................
......06/09/1998
.�..__....
<5
......_..................
B/1999
<5
--------
_._..........._....I ..............
O6/14/2000
_............
10
.
O6/18A.
3
<5
O6/21/2000
8.1
.....-._
-06/24/1997
<1
.. .......I.......
.. _ ..
.....06/16/1998
O6/23/1998
_T.�
<5
._..._.............
_06/15/1.. ...
06/23/7999
......
<5
__....._...._........._.....-..__..._...._..............
06128/2000
7
�07i01/i 997
_
<i
__ _
...0
<5
O6/30/1999
07/05/200.0
9.4
7.0
07/08/1997
<1
�������������������
. ffibW1998
07I0711998
07lO6/1999
c5
9.9
07/1 Z2000
5
07115/1997
<7-
'_._.......__
07/1 a1998.............
07/13/1999
23
07l19/2000
8
07/2Z1997
<1
0712111998
c5
07/20/1999
6.4
07/26/2000
5.5
._.............._....._...........
07/29/1997
..__.._. ..
<T
..............................
...................__._.-
07128/1998
67
........-_....__..
.....................__..........._..............
07/27/1999
<5
_.........................................-..........._................................--
08/02/2000
<5
8.8
- 08105...................
08/05/1997
__.-..
1
......_..................................................
1.3
OB/Oa1998
. _.._
<5
...-18.5
t8.5
..........03119_...--
08/03/
-...........
c5
....5......................................_.........._....................................
5.9
08/09/2000
<5
�08/1Z1997
2
-OB/11%1998
32
.
OB/10/1999
<5
08/16/2000
<5
....-08/19/1997.....
.....-<1..._._.......'.
08/18%1898..
32
.....- ...............
....08/1711999...
_
._.8.7
08/23/2000
24
O6/26/1997
c1
08/25/1998
<5
08/2a1999
<5
OB/30/2000
...........
<5 -_
..... _6....... _....
...._.........................................................._...........111998
09/09/1997
<1
2.2
09/Ot/1998
-...__.........
<5
7..........................................._.
27.6
09/07/1999
__..__....__
c5
._.....5............
5.3
..........................................
09/07/2000
<5
6.3
_ -os/Deli ssi
<t
_ _
os/oe/l sea
<s
os/15/l sss
<s
................................__..............................5....................._....._.
osn az000
a
................................................._.........._....._...._..._..
09/i6/1997
6
09/15/1998
..__._..............................,................_.............__........._
2.4
09/21/1999
6.3
09/20/2000
5.4
09/23/1997
<7
09/22/1998
<5
09/28/1999
09/30!1997........
2
___.........
_-09i28i1998
'120- .......................
10J05/1999
<5
13.0
10/07/1997
3
1.8
10/06/1998
6
5.3
70/12/1999
.. <5
.........
10N1/2000
---.................................
<5
..._t0if 697..._.....__Z._......___....
-10i13%1998
_
__<5 _....----........
...1 W1917999....
25
10/18l2000
6.6
-_.__...
__
1O121/1997�
......._.._..........__..........._........----
.<7
10/20/1998
-.........._........_............_
5
...........................
t0/28/1999
._.._._
17
..................._...._..._........_......__.
10/25/2000
<5
����t0i2a/ts97
<i
10/28n998
<5
11IOZ1999
<5
15.2
11/O7/2000
<5
<5.0
.1/04/......._7
11/Oa1997
_..._...._._............1
q
..............._.........._.._..........
1.0
11/03/1998
..._.-.....-._
<5
5.0
..._...._....---�--......................._..---.._..._.
11I09/1999
56
_._._....._...�....__.
11/082000
__
<5
_.._.__.._............
.-..._.._.._
..................ii..........._...........i.................................................._.............._
11/11/1997
1
11/10/1998
........................................................----....._..
c5
11/18/1999
---..._................._.................._................................
<5
11/16/2000
<5
.............<t
i t/1811997
11/17/1998
.. <5
.................-.
11/23/1999
<5
11/22/2000
<5
_
5I1997.............
j_....._..........._.........
......11%24/i�998 -._...yam_..._.
......_........_.
11/3011999
<5
11/29/2000
c5
............_._...._.......-...._
i ZOZi997
..........
<t
....... i...............................7i..........
1.6
.....
1Z01/1996
.....................................
46
.....
22.2
.................._....................__._._....................._.................0.......................
iZ07J1999
<5
5.0
1.......... ... 00
__....._....................6
<5
<5.6
1
12/08/1998
<5
1 Z1 a1999
<5
1 Z732000
<5
--'7.4_...'_.__.__
_ _12/09/1997
1 Zt6/1997
z .....
........................._.1
Z75/199a..................................._...
._..1 Z21/1999....
11 2.8
5_.....?................................................................_.................................._.
.......................
......1 Z20/2000.....
--
1212a1997
......_..........................................._............................_........................................_....
2
1Z2Z1998
<5
......__
12/28/1999
._................_...-....................................._......_........._......_..
<5
12127/2000
._. <5
-1?l31/1997
2 ......................
_-'12i29%1998.....
......<5._...
ANNUAL "G.
1.8
1.8
11.2
10.9
7.8
1 7.6
6.6
6.6
Notes: All analyses performed four 1l7/97 through 3131/98 performed using freon extraction.
Laboratory switched to hexane extraction starting with 417198 data Values in parentheses indicate
freon results from split samples.
25901 s008.x1s 7

TABLE 2
SUMMARY OF PPG EFFLUENT OIL AND GREASE RESULTS
(12/1/99 - 4/12/00)
Date
Sample
Collected
PPG Effluent Oil and Grease (mg/L)
Ratio of Oil and Grease
(Hexane Extraction/Freon
Extraction)
Hexane Extraction
(EPA Method 1664)
Freon Extraction
(EPA Method 413.1)
12/01/99
< 5.0
12/07/99
<5.0
12/15/99
<5.0
--
12/21/99
11.0
2.6
4.2
12/28/00
<5.0
--
01/05/00
7.8
01/12/00
9.1
--
01/19/00
49.0
1.8
27.2
O1/26/00
<5.0
--
02/02/00
6.1
--
02/09/00
<5.0
--
02/16/00
11.0
7.3
1.5
02/23/00
10.0
--
03/01/00
< 5.0
03/08/00
< 5.0
03/15/00
< 5.0
--
03/22/00
28.0
3.5
8.0
03/29/00
<5.0
--
04/05/00
5.8
-
04/12/00
15.0
--
Notes: Split sample not analyzed by Method 413.1 unless result from Method 1664 is greater than
10 mg/L.

TABLE 3
SUMMARY OF LABORATORY MATRIX SPIKE RESULTS
(NON PPG SAMPLES)
Oil and Grease Matrix Spike
(EPA Method 1664 - Hexane Extraction)
Date
Background O&G
Concentration of
Sample (mg/L)
Spike O&G
Concentration (mg/L)
Result (mg/L)
% Spike Recovery
(mg/L)
12/01/99
11.0
42.1
67.5
134.0 %
12/07/99
1 0.0
40.0
41.3
103.0 %
12/15/99
6.8
42.1
28.7
52.2 %
12/21/99
14.0
42.1
28.0
33.3 %
12/28/00
11.2
42.1
46.2
83.1 %
O1105100
7.8
40.0
13.8
15.0 %
O1/12/00
13.3
40.0
86.2
182.0 %
O1/19/00
0.0
40.0
19.0
47.5 %
01/26/00
53.1
40.0
165.1
280.0 %
02/02/00
24.3
40.0
38.6
60.0 %
02/09/00
10.6
40.8
65.2
134.0 %
02/16/00
43.8
48.8
65.7
44.9 %
02/23/00
19.3
40.0
51.2
79.8 %
03/01/00
7.2
40.0
31.4
60.5 %
03/08/00
0.0
40.0
29.3
73.3 %
03/15/00
2.9
40.0
17.3
36.0 %
03/22/00
8.8
40.0
37.1
70.8 %
03/29/00
0.1
40.0
34.7
86.5 %
04/05/00
0.0
40.0
52.9
132.0 %
04/12/00
10.5
40.0
39.8
73.3 %
Minimum Spike Recovery
15.0 %
Maximum Spike Recovery
280.0 %

TABLE 4
SUMMARY OF LABORATORY DI WATER CONTROL SPIKE RESULTS
(NON PPG SAMPLES)
Oil and Grease DI Water Control Spike
Hexane Extraction
(EPA Method 1664)
Freon Extraction
(EPA Method 413.1)
Date
Spike O&G
Concentration
(m /L)
Result (mg/L)
Percent Spike
Recovery (mg/L)
Spike O&G
Concentration
(m /L)
Result (mg/L)
Percent Spike
Recovery (mg/L)
12/01/99
40.0
34.4
86.0 %
12/07/99
40.0
36.1
90.3 %
12/15/99
40.0
36.1
90.3 %
12/21/99
40.0
37.9
94.8 %
520.0
477.0
91.7 %
12/28/00
40.0
39.1
97.8 %
01/05/00
40.0
36.6
91.5 %
O1/12/00
40.0
38.9
97.3 %
01/19/00
40.0
40.1
100.0 %
501.0
513.0
102.0 %
01/26/00
40.0
44.8
112.0 %
02/02/00
40.0
38.6
96.5 %
02/09/00
40.0
36.2
90.5 %
02/16/00
40.0
37.0
92.5 %
267.0
241.0
90.3 %
02/23/00
40.0
38.2
95.5 %
03/01/00
40.0
36.2
90.5 %
03/08/00
40.0
41.5
104.0 %
03/15/00
40.0
42.4
106.0 %
03/22/00
40.0
47.9
120.0 %
200.0
182.0
91.0 %
03/29/00
40.0
48.3
121.0 %
04/05/00
40.0
53.8
135.0 %
04/12/00
40.0
44.4
111.0 %
Maximum Spike Recovery
135.0 %
102.0
Minimum Spike Recovery
86.0 %
90.3 %

TABLE 5
SUMMARY OF PPG EFFLUENT OIL AND GREASE RESULTS (5/24/00-9/20/00)
Date Sample
Collected
PPG Effluent OR and Grease (mg/L)
Silica Gel
Hexane Treated Freon
Extraction Hexane Extraction
Extraction
Ratio of Hexane
Extraction/Freon
Extraction
Total Petroleum
Hydrocarbons - Freon
Extraction (mg/L)
05/24/00
7.9
05/31/00
<5 (0.1)
06/07/00
<5 (2.0
06/14/00
10.0
06/21/00
8.1
06/28/00
7.0
07/05/00
9.4
07/12/00
8.6
07/19/00
8.0
07/26/00
5.5
08/02/00
<5 (0.0
08/09/00
<5 (4.2)
08/16/00
<5 (3.0)
O8/23/00
23.9
<5
< 1
> 23.9
0.161
08/30/00
<5 1.9
091( 7100
<5 (1.7)
09/13/00
8.0
09/20/00
5.4
11

United States Office of Water (4303) EPA-821-F-98-005
Environmental Protection Office of Solid Waste (5307W) February 1999
Aoency
=,EPA Fact Sheet
Approval of EPA Methods 1664, Revision A, and
9071 B for Determination of Oil and Grease and Non -
polar Material in EPA's Wastewater and Hazardous
Waste Programs
Summary
EPA announces publication of a final rule approving use of EPA Methods 1664, Revision A, and
9071 B for determination of oil and grease and non -polar material (NPM) in EPA's wastewater
program (40 CFR part 136) and hazardous waste program (40 CFR part 260). Approval of
these methods supports EPA's effort to protect Earth's ozone layer by reducing dependency on
use of chlorofluorocarbons (CFCs). Methods 1664 and 9071 B employ n-hexane as the
extraction solvent in place of 1, 1,2-trichloro- 1,2,2-trifluoroethane (CFC-113), a Class 1 CFC.
Background
The U.S. Environmental Protection Agency
(EPA) publishes analytical testing methods that
are used by industrial and municipal facilities to
analyze chemical and biological components of
wastewater, drinking water, sediment, and other
environmental samples (for the purpose of data
gathering and compliance monitoring under the
Clean Water Act and the Safe Drinking Water
Act.)
EPA is approving Method 1664, Revision A,
for use under the Clean Water Act and as an
additional analytical method for the
determination of oil and grease and non -polar
material in aqueous matrices in EPA's
wastewater and hazardous waste programs.
EPA Method 1664, Revision A: N-Hexane
Extractable Material (HEM; Oil and Grease) and
Silica Gel Treated N-Hexane Extractable
Material (SGT-HEM; Non -polar Material) by
Extraction and Gravimetry is being approved as
part of EPA's effort to reduce dependency on
use of chlorofluorocarbons (CFCs). EPA is also
approving Method 9071 B for use in the
hazardous waste program for solid and semi-
solid materials.
Method 1664, Revision A
Method 1664, Revision A, is a liquid/liquid
extraction (LLE), gravimetric procedure that
employs normal hexane (n-hexane) as the
extraction solvent, in place of 1,1,2-trichloro-
1,2,2-trifluoroethane (CFC-113; Freon-113), a
Class 1 CFC, for determination of the
conventional pollutant oil and grease. Because
the nature and amount of material determined is
defined by the solvent and by the details of the
method used for extraction, oil and grease and
NPM are "method -defined analytes"
Method 1664, Revision A is capable of
measuring HEM and NPM in the range of 5 to
1000 mg/L, and may be extended to higher
levels by analysis of a smaller sample volume
collected separately. The method detection
limit (MDL) for HEM in Method 1664, Revision A
is 1.4 mg/L and the minimum level of
quantitation (ML) is 5.0 mg/L.
Method 1664, Revision A may be modified to
reduce interferences and take advantage of
advances in technology or to lower the costs of
measurements, provided that all method
equivalency and performance criteria are met.
This performance -based approach is consistent
with the Agency's streamlining proposal and the
Agency's performance -based measurement
system policy.
Improvements and Changes Included in
Revision A of Method 1664
Revision A to Method 1664 is based on
comments and analytical data received during
both comment periods for the proposed rule (61
FR 1730; January 23, 1996 and 61 FIR 26149;
May 24, 1996) and the comment period for the

notice of data availability (62 FR 51621;
October 2, 1997). The significant improvements
include: (1) the term "total petroleum
hydrocarbons" (TPH) has been replaced by
"non -polar material" (NPM) to avoid confusion
with TPH measurements by other methods; (2)
the requirement for a matrix spike duplicate
(MSD) has been changed to a suggestion; (3)
the size of an analytical batch has been
increased to a maximum of 20 samples per
discharge or waste stream; and (4) use of solid -
phase extraction (SPE) has been allowed
without a demonstration of equivalency,
provided that the discharger/generator assumes
the risk associated with any disparity in results
from liquid -liquid extraction (LLE).
Timing of Required Use of Method 1664,
Revision A and Phaseout of Use of CFC-113
EPA proposed to withdraw the currently
approved methods six months after publication
of the final rule in the Federal Register in an
effort to provide for use and depletion of
existing laboratory stocks of CFC-113. EPA has
decided not to withdraw the CFC-113 based on
EPA's and commenters' concerns about
potentially differing results using the new
method that could bring a permittee into
noncompliance under certain circumstances.
However, EPA believes that direct replacement
of the new method is warranted in most cases.
Therefore, EPA strongly encourages
dischargers/permittees to use Method 1664
rather than the CFC-113 methods for existing
permits. EPA also recommends the use of the
new method for all new permits and reissued
permits.
Method 9071B
EPA is also approving the use of Method
9071 B for use in EPA's hazardous waste
program. This method is for solid and semi-
solid materials. Method 9071 B also uses n-
hexane instead of CFC-113 as the extraction
solvent. This method is not required by any
hazardous waste program regulation but it can
be included as part of a hazardous waste
delisting demonstration.
Additional Information and Copies
For further information concerning the final
rule approving use of Method 1664, Revision A,
please contact Dr. Maria Gomez -Taylor, in
EPA's Office of Water (4303), USEPA Office of
Science and Technology, 401 M Street, SW,
Washington, DC 20460, (Phone: 202-260-
1639); E-mail: [e-mail]).
For information regarding the use of Method
1664, Revision A in the Office of Solid Waste,
contact Gail Hansen, in EPA's Office of Solid
Waste (5307W), USEPA, 401 M Street, SW,
Washington, DC 20460, (703-308-8855; E-mail:
ha nsen. gai I @epa mail.epa.gov).
The final rule published in the Federal
Register contains instructions on how to obtain
additional information and how to review the
public record for this rulemaking.
The complete text of the Federal Register
notice containing this final rule may be
accessed through the Internet and the
Superintendent of Documents homepage at
http://www.access.gpo.gov/su_docs/. The full
text of EPA Method 1664, Revision A may be
viewed or downloaded from the Internet at
hftp://www.epa.gov/OST/Methods.
You may also obtain copies of Method 1664,
Revision A through the U.S. EPA National
Service Center for Environmental Publications
(NSCEP), 11029 Kenwood Road, Cincinnati,
OH 45242, telephone: [phone], Internet:
http://www.epa.gov/ncepi. Method 1664,
Revision A will also be made available at the
National Technical Information Service (NTIS),
5285 Port Royal Road, Springfield, VA 22161,
[phone] or [phone]. The NTIS
number is PB99-121949.

PPG Industries, 940 Washburn Switch Road, Shelby, N.C. 28150
Lacy Ballard, R. E. M.
Staff Engineer, Environmental
Fiber Glass Products
June 5, 2000
Mr. Dave Goodrich
NCDENR
Permitting Division
1617 Mail Service Center
Raleigh, NC 27699-1617
704-434-2261
Extension 544
[e-mail]
JUN i 2000
Re: Effluent Oil and Grease Method Study/Permit Modification Request
PPG Industries Fiber Glass Products, Inc. — Shelby, NC
NPDES No. NC0004685
Dear Mr. Goodrich:
PPG Industries Fiber Glass Products, Inc. (PPG), a manufacturer of fiber glass for the reinforced
plastics industry, is located in Shelby, NC. Wastewater from our production units is treated in an
on -site WWTP before being discharged to Brushy Creek. The NPDES discharge permit limits
effluent oil and grease to 10 mg/L maximum daily and 15 mg/l average. Historically, the oil and
grease effluent concentrations have been below PPG's 10 mg/L permitted discharge limit.
Approximately one and a half years ago, the analyzing laboratory (Pace Analytical Labs,
Huntersville, NC) switched from the old EPA Method 413.1, which uses CFC-113 (freon) as the
extracting solvent, to the new EPA Method 1664, which uses n-hexane (hexane) as the extracting
solvent. As a result of this switch, PPG effluent oil and grease concentrations have been
periodically higher than normal and have exceeded the 10 mg/L discharge criteria.
Freon is no longer being imported or produced in the U.S. as of January 1, 2000. Additionally,
analytical laboratories will no longer be allowed to use any existing freon after the year 2005.
Therefore, future oil and grease analysis will be performed using Method 1664 once freon is
permanently phased out.
Both EPA Methods 413.1 and 1664 were developed for the determination of relatively non-
volatile hydrocarbons, vegetable oils, animal fats, waxes, soaps, greases, and related materials.
Under EPA Method 413.1, the freon extracted constituents were referred to as `oil and grease"
Under EPA Method 1664, the hexane -extracted constituents are referred to as "hexane
extractable material". Method 1664 measures not only oil and grease, but any material that can
be extracted using hexane.

Oil and grease is a method -deft e0arameter; therefore, any change in method protocol, such as
use of another solvent; as the potential to affect results. The EPA Guidance Document,
Analytical Method Guidance for EPA Method 1664A Implementation and Use (40 CFR part
136), indicates that there may be differences in results obtained using Method 1664 compared to
Method 413.1. This Guidance Document outlines recommendations for performing side -by -side
testing between Method 413.1 and Method 1664. Based on the EPA method of side -by -side
testing, a conversion factor is developed. This conversion factor can be applied to oil and grease
results obtained by Method 1664 (hexane extraction) in order to determine the expected result
that would be obtained by Method 413.1 (freon extraction).
In order to investigate potential causes for the higher than normal effluent oil and grease
concentrations, Mr. Jim Meyer (NCDENR), Mr. Wayne Turnbull (EPA — Region 4), Mr. William
Telliard (EPA, Office of Water) and Ms. Marta Jordan (EPA, Office of Water) were contacted by
our wastewater consultant Lili Gellner of Aware Environmental, Inc.. Based on discussions with
these individuals, several key issues were identified. They are as follows:
Hexane has a higher affinity for extracting petroleum products, compared to freon
which has a higher affinity for extracting natural fats and oils.
Surfactants appear to be more readily extracted by hexane than freon, which can cause
higher results in Method 1664 compared to Method 413.1. Hexane extracted material
treated with silica gel may reduce interference from materials such as surfactants but
does not interfere with the measurement of petroleum products.
Hexane versus Freon
Hexane has a higher affinity for petroleum products than freon. Higher oil and grease
results under Method 1664 as compared to Method 413.1 would suggest that more of the
hydrocarbon products are being extracted with hexane than with freon. Based on the
operation and performance of the waste treatment system, it is not expected that higher
than normal oil and grease levels are associated with hydrocarbons. The PPG wastewater
trea en ess is a unique and highly a ectrve system t at uses a variety of treatment
chemicals, along with physical, chemical and biological processes in order to provide
optimum treatment and maintain high effluent quality.
The treatment system includes primary treatment with the addition of time, an aluminum -
based coagulant and a polymer. In the secondary biological treatment system, sodium
bentonite (clay) is added to the mixed liquor to remove refractory organic compounds and
ferric chloride is added prior to the secondary clarifiers to improve settling and turbidity.
It is expected that thes�ocesses would significantly reduce hydrocarbon oil and ereas�
to very low or non -detectable levels. This corresponds to our historical oil and grease
data.
Surfactants
Based on EPA testing of laundering facilities, surfactants can cause a positive result for
oil and grease. PPG uses a significant a
2

mount of surfactants in the production process. In order to reduce the interference
from surfactants, it was recommended that hydrocarbon oil and grease be analyzed
instead of total oil and grease. Method 1664 measures total oil and grease. In order to
measure the hydrocarbon portion of oil and grease, the_fte—x—ant-txtracWA-matenaLin
Method 166A i5 treated with silica gel, which removes polar organic material and
s ould remove most surfactants. Since hydrocarbon oil and grease is nonpolar, it
should not be affected by the silica gel treatment.
In order to determine if PPG's higher than normal effluent oil and grease concentrations were a
result of the change in the analytical method, split samples were analyzed using EPA Methods
413.1 and 1664. This testing was performed on several effluent samples from PPG's WWTP.
The results of this initial testing showed that(Method 413.1 consistently resulted in lower oil and
grease concentrations than Method 1664. However, these results suggested that there was not a
consistent correlation between the two thods.
PPG initiated a program to perform additional testing. PPG routinely collects split samples for
effluent oil and grease. The first effluent sample is analyzed using the new Method 1664. If
Method 1664 results in a concentration greater than 10 mg/L, the second sample is analyzed using
Method 413.1. Method 413.1 has consistently resulted in an effluent oil and grease concentration
lower than Method 1664. Effluent oil and grease data collected from December 1999 through
April 2000 is presented in Table 1. Based on this data, Method 1664 results in a higher oil and
grease concentration. Furthermore, there is not a consistent correlation between Methods 413.1
and 1664 w en determining oil and grease in PPG's wastewater. This data shows that Method
1664 can result in an oil and grease concentration from 1.5 to 27 times as high as Method 413.1.
Since there is not a good correlation between these results, a conversion factor obtained by side -
by -side testing would need to account for the high ratio results.
Additional oil and grease data was obtained from Pace Analytical Labs. These data are
summarized in Tables 2 and 3 and includes the quality control (QC) analyses for the PPG samples.
Table 2 includes a summary of the oil and grease matrix spike results obtained by Method 1664.
Table 3 includes of summary of the oil and grease DI water control spike results obtained by both
Method 1664 and Method 413.1. This spike recovery results in Table 3 show that DI water
spiked with oil and grease provides fairly consistent results for both Method 1664 and Method
413.1. The spike recovery results in Table 2 show that oil and grease recovery from typical
samples is highly variable under Method 1664. This type of variability is consistent with the
variability in the PPG effluent oil and grease data obtained by Method 1664.
Based on the data presented in Table 1, it appears that PPG will be subjected to a greater risk of
noncompliance for effluent oil and grease under Method 1664. In order to address this issue
while freon is still available, PPG is expanding their testing program. This testing program was
initiated the week of May 29, 2000. Each week, approximately six (6) split effluent samples will
be collected for oil and grease analysis. Sample No. 1 will be analyzed for oil and grease by
Method 1664 and Sample No. 2 will serve as the QC matrix spike for Method 1664. If the oil
and grease result by Method 1664 is greater than 10 mg/L, additional analyses will be performed
as follows. Sample No. 3 will be analyzed for oil and grease by Method 1664 with the addition of
3

a silica gel treatment and Sample No. 4 will serve as the QC matrix spike. Sample No. 5 will be
analyzed for oil and grease by Method 413.1 and Sample No. 6 will serve as the QC matrix spike.
It is expected that this testing will help to clarify some of the issues associated with the new
analytical Method 1664 and will help to determine a more appropriate oil and grease limit for
PPG's effluent. This testing program will continue until a total of eight (8) full sets of oil and
grease data are obtained, which is the number of samples required for EPA's method of side -by -
side testing. Because Method 1664 does not consistently result in an oil and grease result greater
than 10 mg/L, it is expected that testing will take longer than eight weeks to complete. //
We appreciate NCDENR's cooperation with this testing program. During this testing program,
we would like to request that PPG's NPDES permit be temporarily modified from an oil and
grease limit of 10 mg/L to a "monitor only" status. PPG will continue to maintain optimum
performance of the treatment system throughout this testing program.
Sincerely,
Lacy allard
Attachment
cc: Rex Gleason, NCDENR-Mooresville
Chet Whiting, NCDENR- Mooresville
Bob Stein, AWARE Environmental Inc.
Lili Gellner, AWARE Environmental Inc.
M. LeCroy
R. Daman
J. Buchanan
zsvouooi

TABLE 1
SUMMARY OF PPG EFFLUENT OIL AND GREASE RESULTS
(12/1/99 — 4/12/00)
Date
Sample
Collected
PPG Effluent Oil and Grease (mg/L)
Ratio of Oil and Grease
(Hexane Extraction/Freon
Extraction)
Hexane Extraction
(EPA Method 1664)
Freon Extraction
(EPA Method 413.1)
12/01/99
<5.0
12/07/99
<5.0
12/15/99
<5.0
12/21/99
11.0
2.6
4.2
12/28/00
<5.0
01/05/00
7.8
_
01/12/00
9.1
_
01/19/00
49.0
1.8
27.2
01/26/00
<5.0
02/02/00
6.1
_
02/09/00
<5.0
02/16/00
11.0
7.3
1.5
02/23/00
10.0
_
03/01/00
<5.0
03/08/00
<5.0
03/15/00
<5.0
03/22/00
28.0
3.5
8.0
03/29/00
<5.0
04/05/00
5.8
04/12/00
15.0
Notes: Split sample not analyzed by Method 413.1 unless result from Method 1664 is greater than
10 mg/L.
NA = Not Analyzed
5

TABLE 2
SUMMARY OF LABORATORY MATRIX SPIKE RESULTS
(NON PPG SAMPLES)
Oil and Grease Matrix Spike
EPA Method 1664 - Hexane Extraction
Date
Background O&G
Concentration of
Sample (mg/L)
Spike O&G
Concentration (mg/L)
Result (mg/L)
o �� Spike Recovery
(mg/L)
12/01/99
11.0
42.1
67.5
134.0 %
12/07/99
1 0.0
40.0
41.3
103.0 %
12/15/99
6.8
42.1
28.7
52.2 %
12/21/99
14.0
42.1
28.0
33.3 %
12/28/00
11.2
42.1
46.2
83.1 %
O1/05/00
7.8
40.0
13.8
15.0 %
O1/12/00
13.3
40.0
86.2
182.0 %
O1/19/00
0.0
40.0
19.0
47.5 %
O1/26/00
53.1
40.0
165.1
280.0 %
02/02/00
24.3
40.0
38.6
60.0 %
02/09/00
10.6
40.8
65.2
134.0 %
02/16/00
43.8
48.8
65.7
44.9 %
02/23/00
19.3
40.0
51.2
79.8 %
03/01/00
7.2
40.0
31.4
60.5 %
03/08/00
0.0
40.0
29.3
73.3 %
03/15/00
2.9
40.0
17.3
36.0 %
03/22/00
8.8
40.0
37.1
70.8 %
03/29/00
0.1
40.0
34.7
86.5 %
04/05/00
0.0
40.0
52.9
132.0 %
04/12/00
10.5
40.0
39.8
73.3 %
Maximum Spike Recovery
15.0 %
Minimum Spike Recovery
280.0 %
0

TABLE 3
SUMMARY OF LABORATORY DI WATER CONTROL SPIKE RESULTS
(NON PPG SAMPLES)
Oil and Grease DI Water Control Spike
Hexane Extraction
PA Method 1664
Freon Extraction
EPA Method 413.1
Date
Spike O&G
Concentration
Result
(mg2)
Percent Spike
Recovery
Spike O&G
ConcentrationConc'eentration
Result
(mg/L)
Percent Spike
Recovery (mg/L)
12/01/99
40.0
34.4
86.0 %
12/07/99
40.0
36.1
90.3 %
12/15/99
40.0
36.1
90.3 %
12/21/99
40.0
37.9
94.8 %
520.0
477.0
91.7 %
12/28/00
40.0
39.1
97.8 %
O1/05/00
40.0
36.6
91.5 %
01/12/00
40.0
38.9
97.3 %
O1/19/00
40.0
40.1
100.0 %
501.0
513.0
102.0 %
O1/26/00
40.0
44.8
112.0 %
02/02/00
40.0
38.6
96.5 %
02/09/00
40.0
36.2
90.5 %
02/16/00
40.0
37.0
92.5 %
267.0
241.0
90.3 %
02/23/00
40.0
38.2
95.5 %
03/01/00
40.0
36.2
90.5 %
03/08/00
40.0
41.5
104.0 %
03/15/00
40.0
42.4
106.0 %
03/22/00
40.0
47.9
120.0 %
200.0
182.0
91.0 %
03/29/00
40.0
48.3
121.0 %
04/05/00
40.0
53.8
135.0 %
04/12/00
40.0
44.4
111.0 %
Maximum Spike Recovery
135.0 %
102.0 %
Minimum Spike Recovery
86.0 %
90.3 %
7

EPA
United States Office of Water
Environmental Protection 4303
Agency
EPA/821-R-00-003
February 2000
Analytical Method Guidance for
EPA Method 1664A Implementation
and Use (40 CFR part 136)
V9
.- �

Table of Contents
Section 1 Introduction...........................................................
1-1
Background.............................................................
1-1
Subjects Addressed in the Guidance ...........................................
1-2
Other Subjects .................................... ...........
.......... 1-3
Section 2 Side -By -Side Comparison of CFC-113 and n-Hexane ........................
2-1
Are there differences in results produced using EPA Method 1664A
versus CFC-based methods? .................................................
2-1
When should I conduct a side -by -side comparison? ................................
2-2
What constitutes side -by -side comparisons? .....................................
2-3
How do I derive a conversion factor? ...........................................
2-3
What are the detailed components of a side -by -side comparison? ......................
2-4
Section 3 Solid -phase Extraction (SPE)............................................. 3-1
What is solid -phase extraction and under what conditions may I use it? ................. 3-1
What waste streams were tested to attempt to establish that liquid -liquid extraction and solid -phase
extraction are equivalent?................................................... 3-2
How can I demonstrate the applicability of SPE to compliance monitoring with my discharge
orwaste stream?......................................................... 3-3
How can I demonstrate equivalency of liquid -liquid and solid -phase extraction? ........... 3-4
Section 4 Flexibility in EPA Method 1664A......................................... 4-1
Is there flexibility in EPA Method 1664A?...................................... 4-1
What types of modifications may I make? ....................................... 4-1
How can I demonstrate equivalent or superior performance for a method modification? ..... 4-1
Section 5 Frequently Asked Questions (FAQs) Concerning
EPA Method 1664A..................................................... 5-1
Section 6 Sources of Information .................................................. 6-1
Regulatory Background.................................................... 6-1
Data Gathering for EPA Method 1664A........................................ 6-1
Interpretation of Requirements in Clean Air Act Amendments ........................ 6-2
Section 7 Where to Get Additional Help ............................................ 7-1
Documents on Compliance Monitoring and Methods ............................... 7-2
Websites................................................................ 7-2
Appendix A Example Calculations for Side -by -Side Comparisons ........................ A-1
Appendix B Annotated Initial Demonstration of Laboratory Capability for SPE Annotated ... B-1
Appendix C Annotated Equivalency Demonstration for Application of a Method Modification to
Compliance Monitoring for SPE.......................................... C-1

Tables
Table 1 Example 1: Example Results for Side -By -Side Study ........................ A-1
Table 2 Example 1: Log -Transformed Results ..................................... A-2
Table 3 Example 1: Triplicate Means ............................................ A-2
Table 4 Example 1: Triplicate Standard Deviations ................................ A-3
Table 5 Example 2: Example Results for Side -By -Side Study ....................... A-5
Table 6 Example 2: Log -Transformed Results ..................................... A-6
Table 7 Example 2: Triplicate Means ............................................ A-6
Table 8 Example 2: Triplicate Standard Deviations ................................ A-7
ii

Acknowledgements
Analytical Method Guidance for EPA Method 1664A Implementation and Use (40 CFR part 136) (Guidance) was
prepared by the U.S. Environmental Protection Agency's (EPA's) Office of Water (OW), Engineering and Analysis
Division (EAD). EPA gratefully acknowledges assistance from DynCorp Information and Enterprise Technology
and Interface, Inc. in preparing this document. Additionally, EPA gratefully acknowledges Horizon Technology,
Inc. and Mallinckrodt Baker, Inc. for their comments and review.
Disclaimer
This Analytical Method Guidance for EPA Method 1664.4 Implementation and Use (40 CFR part 136)
(Guidance) is provided to help implement national policy on use of EPA Method 1664A. This Guidance does not,
however, substitute for the Clean Water Act (CWA), the Resource Conservation and Recovery Act (RCRA), or
EPA's regulations, nor is it a regulation itself. Thus, it cannot impose legally binding requirements on EPA, States,
Tribes, or the regulated community and may not apply to a particular situation based upon case -specific
circumstances. EPA and State decision makers retain the discretion to adopt approaches on a case -by -case basis
that differ from this Guidance where appropriate. This Guidance may be changed in the future based on any future
information made available to EPA. The material presented is intended solely for guidance and does not alter any
statutory requirements.
iii

Executive Summary
0 n May 141999, the U.S. Environmental Protection Agency (EPA) promulgated Method 1664, Revision
A: N-HeLne Extractable Material (HEM; Oil and Grease) and Silica Gel Treated N-Hexane
Extractable Material (SGT-HEM, Non polar Material) by Extraction and Gravimetry (Method 1664A;
the Method) at tide 40, part 136 of the Code ojFederal Regulations (40 CFR part 136) for use in
EPA's CWA monitoring programs and at 40 CFR part 260 for use in EPA's RCRA monitoring programs. The
purpose of this Guidance is to assist dischargers, generators, industrial users, and laboratories in the application of
Method 1664A to wastewaters and wastes, to address potential differences between results obtained with Method
1664A and an approved method that employs CFC-113 (1,1,2-trichloro-1,2,2-trifluoroethane) as an extracting
solvent, and to address questions regarding use and application of Method 1664A.
iv

Introduction
0 n May 14, 1999, we promulgated Method 1664, Revision A: N-Hexane Extractable Material (IMM; Oil
and Grease) and Silica Gel Treated N-Hexane Extractable Material (SGT-HEM,• Non polar Material)
by Extraction and Gravimetry (Method 1664A; the Method) at 40 CFR part 136 for use in EPA's CWA
monitoring programs and at 40 CFR part 260 for use in EPA's RCRA monitoring programs. The
purpose of this Guidance is to assist you in the application of Method 1664A to wastewaters and wastes, and to
address potential differences between results that could be obtained using Method 1664A and approved methods that
employ CFC-113 as the extracting solvent. This Guidance also further explains the flexibility allowed within the
scope of Method 1664A, particularly the use of solid -phase extraction (SPE), and answers frequently asked
questions (FAQs) concerning Method 1664A.
This Guidance is specifically written to provide you with procedures for side -by -side testing of specific discharges
(when necessary), and with clarification on the use of solid -phase extraction. To help in this process, the following
topics are addressed:
# Section 2 discusses differences in results produced by EPA Method 1664A and CFC-based
methods and how to determine if results are equivalent or if determining a conversion factor is
necessary;
# Section 3 discusses use of solid -phase extraction (SPE) and how to determine if results produced
with SPE are equivalent to those produced by liquid -liquid extraction (LLE) using CFC-113
(LLE/CFC-113) or Method 1664A (LLE/n-hexane);
# Section 4 discusses flexibility in EPA Method 1664A;
# Section 5 presents responses to specific questions raised by the regulated community regarding the
use of Method 1664A.
Background:
The Clean Air Act Amendments of 1990 require the production phaseout and cessation of import of ozone -depleting
substances (DDSs), including chlorofluorocarbons (CFCs), in order to protect Earth's ozone layer. The major
reliance on CFC-113 in EPA's regulatory programs is in analytical methods for determination of the conventional
pollutant "oil and grease." To address the phaseout, as applied to methods for determination of oil and grease, we
began studying extraction solvent alternatives to CFC-113 in the early 1990s. We proposed Method 1664 on
1-1

January 23, 1996 (61 FR 1730). Method 1664 used normal hexane (n-hexane) in place of CFC-113. We
promulgated Method 1664A on May 14, 1999 (64 FR 26316). Method 1664A also uses n-hexane as the extraction
solvent.
Subjects Addressed in this Guidance:
Phase -out ofproduction and cessation of import of ozone -depleting substances (ODSs), including CFCs.
The Montreal Protocol allows limited production of CFCs for certain uses that are considered to be essential. Under
the Montreal Protocol these essential uses include laboratory and analytical use. In the preamble to the final rule
promulgating EPA Method 1664A, we stated that since the laboratory use exemption for CFC-113 under the
Montreal Protocol had been extended to 2005, EPA had decided to allow continued use of methods that use CFC-
113 as the extraction solvent until that time.
However, the Clean Air Act Amendments of 1990 (CAAA) require the cessation of production and import of CFCs,
including CFC-113 for laboratory uses, by January 1, 2000. Because we are governed by CAAA and the Montreal
Protocol, we must enforce the more stringent of the two. Therefore, CFC-113 may not be available for continued
use with analytical methods that use CFC-113 as the extraction solvent once current supplies are exhausted. You
may continue to use existing supplies of CFC-113 after January 1, 2000 until existing supplies are depleted.
However, we strongly recommend the switch to Method 1664A, and the discontinuation of use of CFC-113. (See
the memo from Drusilla Hufford dated August 30, 1999 cited at the end of this Guidance, the Federal Register
notice dated September 15, 1999 (64 FR 50083), and the Federal Register notice of the proposed mlemaking dated
November 2, 1999 (64 FR 59141)).
Differences in amount of material extracted by CFC-113 and n-hexane.
We included in the administrative record at proposal studies comparing results obtained using CFC-113 and several
other solvents, among them n-hexane. These studies showed that n-hexane produced results closest to results
produced by CFC-113 but that, when applied to a given discharge, n-hexane could extract more or less material than
the amount extracted by CFC-113. For details of the results of the studies, see the administrative record for Method
1664A in the Water Docket. (Chapter 7, Where to Get Additional Help, provides contact information for the Water
Docket.)
At proposal, and at the fatal rule, we addressed the differences in results produced by CFC-113 and n-hexane, and
the preamble to the final rule provided a procedure for gathering data to support that a statistically significant
difference in results could be produced. This Guidance details how the procedure in the preamble is to be applied to
a specific discharge.
1-2

Solid phase extraction
Method 1664A employs liquid -liquid extraction (LLE) with a separatory funnel as the means for extracting oil and
grease from water. As a part of the studies conducted in the early 1990s, we studied solid -phase extraction (SPE),
also known as liquid -solid extraction. SPE uses a cartridge or disk to adsorb the components of oil and grease from
the water, with subsequent desorption by an organic solvent. SPE has advantages over LLE, in that less
manipulation of the sample is required and the amount of solvent used is considerably less. In general, SPE and
LLE will produce nearly identical results when applied to final effluents.
In early versions of EPA Method 1664, we required a side -by -side comparison of SPE and LLE to demonstrate that
equivalent results would be produced. We solicited comment on the need for this requirement. Commenters stated
that the comparison procedure was cumbersome and that SPE should be allowed without a side -by -side comparison.
(See Section VI of the final rule and the administrative record in the Water Docket for comments and responses.) As
a result of the comments, we allowed use of SPE in Method 1664A without a side -by -side comparison, but cautioned
that you and your laboratory must take responsibility for the results produced. However, for some discharges,
particularly untreated effluents and in -process wastewaters, and for some final effluents, SPE can produce results
different from those produced by LLE. This Guidance expands the discussion on use of SPE in Method 1664A.
Other Subjects:
This Guidance also gives details concerning the flexibility inherent in Method 1664A and provides answers to
frequently asked questions (FAQs).
We trust that this Guidance will assist you in the use of Method 1664A for determinations of oil and grease (HEM)
and non -polar material (SGT-HEM) in wastewaters and wastes. While this Guidance attempts to address issues and
situations pertinent to Method 1664A, this Guidance also identifies and references other sources that you may wish
to consult concerning selection and use of analytical methods. Also included is a list of EPA and other authorities to
contact for more guidance.
1-3

2 Side -By -Side Comparison
of CFC-113 and n-Hexane
his section provides a discussion of a difference in results that could be produced by EPA Method 1664A
Tand CFC-based methods, how to determine if the difference is statistically significant, and how to calculate
a conversion factor if the difference is statistically significant. For more background information on
method comparisons, refer to Guidelines Establishing Test Procedures for the Analysis of Oil and Grease
and Non -Polar Material Under the Clean Water Act and Resource Conservation and Recovery Act; Final Rule at
40 CFR parts 136 and 260, published in the Federal Register (64 FR 26315; May 14, 1999).
Are there differences in results produced using EPA Method 1664A versus CFC-based methods?
Prior to the advent of EPA Method 1664, the most commonly used EPA methods for determination of oil and grease
were EPA Method 413.1 in our CWA program and EPA Method 9070 in our RCRA program These methods use
CFC-113 as the extraction solvent. CFC-113 is a Class I ozone -depleting substance under Section 604 of CAAA.
In comments on the proposal of EPA Method 1664A, several of you expressed concerns about the impact of
differences resulting from substitution of CFC-113 with n-hexane on determinations of compliance under the
NPDES program and pretreatment programs. We discussed this issue at proposal (61 FR 1730; January 23, 1996),
and discussed the issue in workshops, conferences, and seminars between proposal and development of the final rule.
After proposal, we received numerous requests from States and EPA Regions for guidance on implementation of
Method 1664. On July 9, 1996, we issued guidance to Pretreatment Coordinators and Regional NPDES Contacts.
A copy of the memorandum was included in the administrative record for the final Wile. In part, this memorandum
stated:
"EPA acknowledges that, due to the diverse nature of discharges, there may be
instances in which n-hexane will extract an amount of oil and grease greater or less
than the amount extracted by Freon-113. If these instances affect compliance, the
permitting authority may wish to consider establishing a conversion factor,
multiplier, or divisor to account for these differences in the permit. EPA
emphasizes that few, if any, instances will likely be found in which the differences
affect compliance and, therefore, urges direct substitution of the presently approved
methods with Method 1664 when the date of substitution is announced in the
Federal Register."
We still believe that the approach outlined in the memorandum appropriately accommodates any significant
discrepancies that could arise in determining compliance with limitations or standards for oil and grease using
2-1

Method 1664A. Based on results from the Freon Replacement Studies, we found that, on average, n-hexane
extracted approximately 96% of the material extracted using CFC-113. Therefore, while there may be some effluent
matrices from which n-hexane will extract more material than CFC-113, on the whole, most dischargers would have
little risk of a determination of non-compliance with existing limits.
The slightly smaller average amount of oil and grease extracted by n-hexane (96% versus 100% by CFC-113) is not
statistically significant because errors in oil and grease measurement are in the order of 10% relative standard
deviation. A coarse estimate of 95% confidence limits around the 961/o recovery by n-hexane is 96% plus or minus
20%, or the true difference lies somewhere between 76-1161/o. This range encompasses 100%, which indicates that
there is no significant difference. Given the lack of significance of the 4% difference, the measurement error that
would be encountered in the side -by -side comparison (estimated at 10% for each measurement), the potentially
significant cost of a side -by -side comparison with each discharge, and the low anticipated likelihood that a
significant difference would be found (based on our studies), we do not recommend a side -by -side comparison for
each discharge. Instead, we continue to recommend a direct replacement of the approved CFC-based methods with
Method 1664A.
When should I conduct a side -by -side comparison?
The Montreal Protocol on substances that deplete the ozone layer allows production and use of Class I substances
for certain critical uses, whereas Section 604 of CAAA requires production phase -out and cessation of import of
Class I substances, with certain exceptions, by January 1, 2000. Until 1999, we allowed production and import of
Class I CFCs under the exemptions in the Montreal Protocol. One of these exemptions is production and import of
CFCs for laboratory use. We had planned to extend the exemption of production and import for laboratory use until
2005. However, in 1999, our Office of General Counsel clarified requirements under CAAA and concluded that
CAAA does not allow any exemption to the phase -out of production and cessation of import of CFC-113 and other
Class I substances for which there is no exemption. Therefore, at the end of 1999, production and import of CFC-
113 will cease, and new supplies of CFC-113 will not be available. You may continue to recycle and use existing
supplies of CFC-113 until stocks are exhausted.
In the preamble to the proposed and final rules for Method 1664A, we recommended that laboratories switch to
Method 1664A rather than wait until expiration of the production exemption for Class I substances, and that
regulatory/control authorities authorize the switch. The re -interpretation of CAAA requirements will hasten the
switch to Method 1664A and could, possibly, result in compliance difficulties for you, although our studies have
indicated that this possibility is remote. If you wish to use Method 1664A and believe that the switch will cause a
non-compliance, we suggest that you perform a side -by -side comparison with an approved method that uses
LLE/CFC-113. We addressed this possibility in the final rule promulgating Method 1664A, and stated that although
it is unlikely that a non-compliance situation may occur, in a few cases it may be necessary to perform side -by -side
testing with Method 1664A and an approved method that employs CFC-113 to demonstrate a significant difference
in results between the solvents and methods. The production phase -out and cessation of import of CFC-I 13 may
preclude this side -by -side demonstration at some point in the future because CFC-113 may not be available.
Therefore, if you believe that you could be adversely affected by the switch from CFC-I 13 to n-hexane, you should
2-2

perform the side -by -side demonstration while supplies of CFC-113 are available, because this option may not be
possible after stocks of CFC-113 are exhausted, unless Congress further amends the Clean Air Act.
The side -by -side demonstration is detailed below and examples are provided in Appendix A to this Guidance. To
allow you to make the calculations detailed below and in Appendix A easily, we have produced an Excel spreadsheet
that does all calculations automatically after you have entered the necessary data. This spreadsheet is available via
an E-mail attachment from our Sample Control Center (SCC). Please E-mail your request to [e-mail],
or you may phone your request to 703461-2100. If you telephone, please ask for the "Method 1664A Comparison
Spreadsheet" and provide your E-mail address. You must have Microsoft's Excel 97 to nut the spreadsheet.
What constitutes a side -by -side comparison?
For the side -by -side comparison of Method 1664A to an approved method that uses LLE/CFC-113, we suggest, at a
minimum, analysis of three replicates of each sample by each method on any seven days over a minimum 30-day
period, for a total of 42 analyses (21 by the previously used method and 21 by Method 1664A). For this side -by -
side comparison, you should use the sepamtory funnel liquid/liquid extraction (LLE) procedure, not the optional
SPE procedure, in Method 1664A because of the possible confounding of results that could occur when two
variables (SPE and the solvent) are changed simultaneously. If all six results associated with any sample are less
than the minimum level (<ML), these results should not be used in the comparison because it is necessary to have
actual measured values to test equivalency. In the event that a test result less than the ML is obtained, samples
should be collected on an additional day (i.e., the number of tests should be increased to provide a minimum of seven
paired triplicate results for the comparison). If all or nearly all results are less than the ML, we suggest that no
further testing be done and that a direct substitution of Method 1664A for the CFC-based method be made because a
reliable conversion factor cannot be derived. Further details are provided below in response to the questions: "How
do I derive a conversion factor?" and, "What are the detailed components of a side -by -side comparison? ".
How do I derive a conversion factor?
Step I: Test for statistical significance using procedures for development of the root -mean -square deviation
(RMSD) detailed in our Freon Replacement Study reports, and shown by example in Appendix A to
this Guidance, using results obtained with the CFC-based method as reference. If the RMSD is
within the acceptance limit, the results obtained using the different methods are equivalent and a
conversion factor is not warranted. If the RMSD is not within the acceptance limit, the difference is
significant.
Step 2: If the difference is significant, a second step is necessary to test whether the bias between methods
varies significantly from sample to sample (a significant method/sample interaction). If there is no
significant interaction, a conversion factor can be developed for application to future results
obtained with Method 1664A to convert these results to what they would have been had an
alternate, approved, CFC-based method been used. If a significant interaction is present, a
conversion factor would not be appropriate. Appendix A to this Guidance gives detailed example
calculations for a side -by -side comparison and development of a conversion factor (if appropriate).
2-3

What are the detailed components of a side -by -side comparison?
Step 1: Collect eight samples on a total of seven days spaced over a minimum of a 1-month period
Observe the required preservation and holding time for oil and grease (per the methods and 40 CFR
136.3(e), Table H).
Step 1: For the eight samples collected on each day, analyze three by EPA Method 1664A and three by
EPA Method 413.1 or another, approved, CFC-based method (for a total of 6 samples). Spike and
analyze the remaining two samples and analyze one by the CFC-based method and the other by
Method 1664A as a matrix spike sample with each method. For analyses by Method 413.1 or
other, approved, method, the QC requirements in Method 1664A must be applied (including the
demonstration of initial precision and recovery, on -going precision and recovery, blank, and matrix
spike) to assure that the QC used with Method 413.1 or other, approved, method are equivalent to
those used with Method 1664A.
Step 3: Tabulate the results of the analyses.
Step 4: Calculate the logarithms of the results to assure that a single result does not unduly influence the
outcome.
Step S: Calculate the means of the triplicates of the log -transformed results for each method on each day.
Label each mean value M;j, for sample j, ranging from 1 to 7, and method i, ranging from 1 to 2 (for
example, let Method 1664A be method 1 and Method 413.1 be method 2).
Step 6: Calculate the standard deviations of the triplicates of the log -transformed results for each method on
each day:
Sj= 2Zei..... 3)(Ytjk-Mj)2
where Y;jk is the k" replicate for the f" sample, analyzed using method i.
Step 7• Calculate the mean -square error (MSE):
MSE= 14�,..,2)t,...,�)s2ij
Step 8: Calculate the method specific -means MM, and MM2, the sample -specific means Msi+ Ms2, Ms3,
Ms„ and the overall mean K.
2-4

MMt= I t r1Mt.., for method i (i= 1, 2)
J
Step 9: Calculate the root -mean -square deviation (RMSD) for the results using Method 1664A and the
alternate, approved method (e.g., EPA Method 413.1):
Msi= Z t„z)Mti, for sample j (i in 1,...,7)
Mo = 2 F(w1-"2)MM' = J �t,..,nMsi
Step 10. Calculate RMSDmAx as:
RMSD= I�;t,...,7)(Mti-M2,
i)2
7
RMSDM _ 2 * MSE * 2.36
where 2.36 is the 95" percentile F value with 7 degrees of freedom in the numerator, and 28 degrees
of freedom in the denominator.
Step 11: Compare results. If RMSDK� > RMSD, there is not a significant difference
between methods, and therefore no conversion factor is necessary.
If RMSD x < RMSD, the difference between results produced by the two methods is significant.
However, it is necessary to test whether this difference is constant for each sample. Therefore,
continue to step 12 only if RMSDmx < RMSD.
Step 12: Calculate the mean squares attributable to a method -by -sample interaction
(MSINT):
MSINT = Z�t,2)�t,...,z)(Mti-MMi-Msi+M)2
2-5

Step 13: Divide MSINT by the MSE from Step 7, and label this ratio F,.k. Compare this to 2.45 (a 95
percentile F value with 6 degrees of freedom in the numerator and 28 degrees of freedom in the
denominator). If F� ,, > 2.45, there is a significant interaction between method and sample (i.e., the
bias between methods varies significantly from sample to sample) and a conversion factor is not
appropriate. Stop here.
If RMSDmm < RMSD, and F.I. < 2.45, there is a consistent bias between methods, and a
conversion factor should be calculated. Proceed to Step 14 for calculation of a conversion factor.
Step 14: Calculate the conversion factor
In order to calculate the conversion factor, calculate the ratio of the averages of the method means
of the log -transformed results as shown below. The result for Method 1664A must be in the
denominator.
MM2
CFL= M
MI
This mean is the log of the conversion factor that could be applied to all future values to transform a
result for Method 1664A to the result that would have been produced had an alternate, approved
method been used.
For a given log -transformed result from Method 1664A, the converted value is:
XR = EXP[YN • CFL]
where XR is the converted value in the original scale, and Y. is the log -transformed value obtained
using 1664A.
See Appendix A for examples using the procedure above.
2-6

Solid -Phase Extraction
(SPE)
his section discusses use of solid -phase extraction (SPE) and how to determine if results produced with
TSPE are equivalent to those produced by liquid -liquid extraction (LLE).
What is solid -phase extraction and under what conditions may I use it?
SPE uses a cartridge or disk for removal of the oil and grease from the sample. A detailed description of the SPE
technique was provided when Method 1664 was proposed (61 FR 1730). Even prior to proposal of Method 1664,
proponents of SPE devices had requested that SPE be an allowed technique in Method 1664. Proposed Method
1664 allowed use of SPE, but required a demonstration that SPE produced results equivalent to results produced by
separatory funnel liquid -liquid extraction (LLE). Proponents objected to this requirement, claiming that SPE
provided sufficient advantages in solvent reduction, reduced analysis time, reduced emulsion formation, and other
advantages so that its use should be allowed without a prior demonstration of equivalency. We discussed the issue
extensively at proposal and in public workshops and meetings, and specifically solicited data demonstrating
equivalency of results produced by SPE and LLE.
Some data demonstrated that results produced by LLE and SPE are equivalent, whereas other data demonstrated
that results produced could be significantly different. We reopened the comment period (61 FR 26149) to allow
submission of further data, and provided a notice of availability (62 FR 51621) of those and other data so that we
could consider those data for the final rule.
Discussions of the detailed issues on SPE were summarized in Section VI of the preamble to the final rule
promulgating Method 1664A, and were given in the detailed comments and responses included in the Docket for the
final rule. Based on comments received and supporting data, we allowed use of SPE in Method 1664A without a
prior demonstration of equivalency. Therefore, if SPE in Method 1664A produces results equivalent to results by
LLE/CFC-113 in a approved method or to LLE in Method 1664A, SPE can be used as a directly equivalent
extraction technique. However, we added a note at the beginning of the extraction procedure (Section 11.3) in
Method 1664A to indicate that it is your responsibility to assure that the results produced are equivalent. The note
states:
"The procedure detailed below is for separatory funnel liquid -liquid extraction.
Solid -phase extraction (SPE) may be used at the discretion of the discharger/
generator and its laboratory. However, if SPE is used, it is the responsibility of
the discharger/generator and laboratory to assure that results produced are
3-1

equivalent to results produced by the procedure below." [The procedure following
the note is the LLE procedure.]
We emphasize in this Guidance that SPE may be used at the discretion of you and your laboratory without a
demonstration of equivalency, provided that you assure that the results produced by SPE are equivalent to results
produced by LLE and that you take responsibility for the results. Such an assurance, for example, could be based
on the fact that monitoring of a given discharge produces results that are always below the minimum level (ML) of
Method 1664A. For this instance, a side -by -side comparison cannot be performed because there is, essentially,
nothing to measure. In this case, you could assure that results produced by SPE are equivalent to results produced
by LLE. Another example would be that your discharge is comprised almost exclusively of a pure substance and
you know that SPE extracts this substance with the same efficiency as LLE.
If there is doubt about the equivalency of SPE and LLE, a test should be performed as described below in the
section titled: "How can I demonstrate equivalency of liquid -liquid and solid -phase extraction? ". If the SPE(n-
hexane results differ significantly from the LLE/CFC-113 results or results produced by Method 1664A, you may
wish to consider performing the 42-sample study described in Section 2 of this Guidance to determine if a
conversion factor is warranted. In the final rule, we also acknowledged that if a Region, State, or other permitting
or control authority has concerns about the difference in results produced by SPE and LLE, that authority may
specify the use of either technique in the permit.
What waste streams were tested to attempt to establish that liquid -liquid extraction (LLE) and
solid -phase extraction (SPE) are equivalent?
We evaluated LLE and SPE in the Phase I and Phase II Freon Replacement Studies to attempt to demonstrate that
the two extraction techniques are equivalent. In the initial portion of the Phase I study, we collected and analyzed
treated effluent samples from a variety of industrial facilities. These samples frequently did not contain detectable
levels of oil and grease. Therefore, it was not possible to compare solvents or extraction procedures with these
samples. As a result, we conducted further studies on untreated effluents, in -process wastes, or treated effluents
spiked with treatment system influents, so that we would have something to measure. For many of these samples,
results from LLE and SPE were significantly different; for others, results were not significantly different.
In addition to our data gathering, we shared samples with vendors of SPE devices. Some results provided by these
vendors showed significant differences and others did not. For our data gathering and data gathering by vendors of
SPE devices, multiple solvents were tested, including CFC-113 and n-hexane, with both disk and cartridge SPE
devices.
In addition to our data gathering and data gathering by vendors of SPE devices, some of you took the initiative to
perform additional comparison studies. As with our studies and vendor studies, some results showed significant
differences and others did not. This conclusion was valid regardless of solvent (n-hexane or CFC-113) or technique
(SPE or LLE).
3-2

Based on all of these results, it is likely that LLE and SPE would not produce results that are significantly different
when final effluents are analyzed. However, there may be significant differences when untreated effluents and in -
process wastes are analyzed. As a result, we stated in Method 1664A that you may use SPE without a
demonstration of equivalency provided that you assure that results produced are equivalent to the LLE procedure.
You may compare results produced by SPE in Method 1664A to results by LLE/CFC-113 in an approved method
or to LLE/n-hexane in Method 1664A.
How can I demonstrate the applicability of SPE to compliance monitoring with my discharge or
waste stream?
We allowed use of SPE in Method 1664A provided that you take responsibility to assure that results produced are
equivalent to results produced by the liquid/liquid extraction (LLE) procedure in Method 1664A. (See the note in
Section 11.3 of Method 1664A). Therefore, if you want to apply SPE as an integral part of Method 1664A to
compliance monitoring, you may apply it directly so long as you meet the requirements in the Method, in the same
way that you would apply the LLE procedure in Method 1664A. This demonstration may be done in three steps:
Step 1: If you have not performed the initial demonstration of laboratory capability required for use of
Method 1664A, you must perform this demonstration to assure that your laboratory is capable of
practicing Method 1664A. This demonstration includes the method detection limit (MDL) and
initial precision and recovery (IPR) tests given in Sections 9.2.1 and 9.2.2 of Method 1664A,
respectively. These tests are reproduced and annotated in Appendix B of this Guidance. For these
tests, SPE is included as an integral part of the Method.
If you have performed the initial demonstration of laboratory capability using liquid -liquid
extraction (LLE), you must repeat the 1PR test using SPE as an integral part of Method 1664A. If
the MDL will be affected by use of SPE, you must repeat the MDL test in Section 9.3.1. See
Section 9.1.2.1 of Method 1664A. Unless the sample volume or another significant change is
made to Method 1664A, it is unlikely that you would need to perform the MDL test as a
consequence of using SPE.
Step 2: Measure the concentration of HEM or NPM in your discharge or waste stream using Method
1664A with SPE as an integral part, in the same way that you would apply Method 1664A
directly. If the result is satisfactory; i.e., concentration of HEM or NPM meets your regulatory
compliance limit in the same way in which the compliance limit was met using an approved CFC-
based method, you may use Method 1664A with SPE as an integral part. If the result is
unsatisfactory, you may wish to establish a conversion factor using the procedure in Section 2 of
this Guidance.
Step 3: Perform the on -going QC tests (on -going precision and recovery, blank, and matrix spike) as a part
of the routine practice of Method 1664A with SPE as an integral part of the Method and meet the
QC acceptance criteria in Method 1664A.
3-3

How can I demonstrate the equivalency of liquid -liquid and solid -phase extraction?
One example means of assuring equivalence of LLE and SPE was given earlier in this guidance (that all results are
<ML). Assurance could also be provided by (1) making SPE a modification to EPA Method 1664A, (2)
performing the tests required for demonstrating equivalency that are given in Method 1664A, and (3) meeting the
QC acceptance criteria for equivalency specified in Method 1664A. This testing would be conducted in three steps,
as follows:
Step 1: If you have not performed the initial tests required for use of Method 1664A, you must first
perform these tests to demonstrate that your laboratory is capable of practicing Method 1664A as
written, including the liquid -liquid extraction procedure given in Section 11.3 of Method 1664A.
These tests include the initial precision and recovery (IPR) and method detection limit (MDL) tests.
See Section 9 of EPA Method 1664A.
Step 2: The second step is detailed in Section 9.2 of EPA Method 1664A. This test is a repeat of the
initial precision and recovery (IPR) test on reagent water with SPE as an integral part of Method
1664A. Section 9.2 of Method 1664A is reproduced in Appendix B of this Guidance and
annotated in italics and brackets [text] to provide further guidance. If the MDL will be affected by
the change to SPE, you must repeat the MDL test in Section 9.3.1. See Section 9.1.2.1 of Method
1664A. Unless the sample volume or another significant change is made to Method 1664A, it is
unlikely that you would need to perform the MDL test as a consequence of using SPE. Once the
equivalence of SPE is demonstrated, proceed to Step 3.
Step 3: The third step test is a demonstration of equivalency on each sample matrix type, as detailed in
Section 9.2.3 of EPA Method 1664A, with SPE as an integral part of Method 1664A and with the
matrix type being the actual discharge or waste stream to which SPE will be applied. Section 9.2.3
of Method 1664A is reproduced in Appendix C of this Guidance and annotated in italics and
brackets to provide further guidance. After you demonstrate the equivalence of SPE using the IPR
test (Step 1) and on the sample matrix type (Step 2), SPE may be used on that sample matrix type
(discharge or waste stream) thereafter. For application to another sample matrix type, the
equivalency demonstration would need to be repeated using Step 3; i.e., so long as the method is
not modified fiuther, the IPR test in Steps 1 and 2 do not need to be repeated. The on -going QC
tests (on -going precision and recovery, blank, and matrix spike) must be performed and the QC
acceptance criteria requirements in Method 1664A must be met when you apply the modification
routinely to a sample matrix type.
3-4

Flexibility in EPA
Method 1664A
This section discusses flexibility inherent in EPA Method 1664A and discusses the process of demonstrating
equivalent performance when the Method is modified This discussion is summarized from Guidance on
Evaluation, Resolution, and Demonstration of Analytical Problems Associated with Compliance Monitoring
(EPA 821-B-93-001).
Is there flexibility in EPA Method 1664A?
Flexibility exists for making changes to EPA Method 1664A. However you must demonstrate that your
modification achieves equivalent or superior performance to the performance of Method 1664A.
What types of modifications may I make?
The types of modifications that you may make to Method 1664A are the typical changes that would make it easier
for you to practice the Method without compromising performance. For example, change of a beaker to an
Erlenmeyer flask or change of a boiling flask to a Kudema-Danish is allowed, after a demonstration of equivalency,
because they would not be expected to adversely affect method performance. Certain changes, including change of
the extraction solvent, are not allowed as modifications, because the Freon Replacement Study data demonstrated
that significant changes in results occur when an alternate solvent is used. A change in the n-hexadecane and
stearic acid standards is not allowed because the performance data for Method 1664A were developed using these
standards. (Please refer to section 9.1.2 of EPA Method 1664A, excerpted in response to the question below).
How can I demonstrate equivalent or superior performance for a method modification?
You can demonstrate equivalent or superior performance by showing that results produced by your modification are
equal to or superior to results produced by the unmodified Method The performance of a modified method is
measured by precision and recovery (bias), and can be extended to include detection limit as well as other measures
of method performance. You must perform the initial precision and recovery (IPR) test prior to practicing Method
1664A. This test is described in detail in Section 9 of the Method. Results of the IPR test must meet the precision
and quality control (QC) acceptance criteria in Method 1664A. After the requirements are met for the unmodified
Method, you must repeat the IPR test with the modification as an integral part of the Method. Your modification is
permitted if the QC acceptance criteria for precision and recovery of Method 1664A are met.
As stated in Section 9.1.2 of EPA Method 1664A, the purpose of allowing changes to Method 1664A is to improve
Method performance or lower the cost of measurements. Section 9.1.2 of Method 1664A states:
4-1

9.1.2 In recognition of advances that are occurring in analytical technology, the laboratory is permitted
certain options to improve separations or lower the costs of measurements, provided that all
performance specifications are met. These options include alternate extraction and concentration
devices and procedures such as solid -phase extraction, continuous liquid -liquid extraction, and
Kudema-Danish concentration. Alternate determinative techniques, such as infrared spectroscopy
or immunoassay, and changes that degrade method performance, are not allowed. If an analytical
technique other than the techniques specified in this method is used, that technique must have a
specificity equal to or better than the specificity of the techniques in this method for HEM and/or
SGT-HEM in the sample of interest. Specificity is defined as producing results equivalent to the
results produced by this method for analytical standards (Section 9.2.2) and, where applicable,
environmental samples (Section 9.2.3), and that meet all of the QC criteria stated in this method.
4-2

Frequently Asked
Questions (FAQs)
Concerning EPA
Method 1664A
he questions below are those that we have been most frequently asked. In comments on proposal of
Method 1664, and in comments on notices that EPA published, commenters raised more than 100
questions/issues concerning Method 1664. EPA responded to these questionstissues and placed the
comments/responses in the Water Docket to support the final rule. We urge dischargers/generators/
industrial users, regulatory/control authorities, laboratories, and others that have questions concerning Method
1664A to review the comments/responses in the administrative record at the Water Docket. (Chapter 7, Where to
Get Additional Help, provides contact information for the Water Docket).
NOTE: In the questions below, the pronouns "P' and "we" refer to the questioner whereas the pronoun "you" refers
to EPA. In the answers, the pronouns "we," "us," and "our" refer to EPA whereas the pronouns "you" and "your"
refer to the questioner and to other users of Method 1664A.
We are having difficulty in achieving the recoveries of HEM and SGT-HEM specified in the QC
acceptance criteria in Method 1664A. Why?
This has been the most frequently asked question since proposal of Method 1664A. Low recoveries appear
attributable to two main sources: (1) failure to adequately rinse glassware and other surfaces that the sample
contacts and (2) use of a hot plate or Roto-vap for evaporation of the solvent.
Stearic acid sticks to nearly everything. Therefore, to adequately recover stearic acid, you must rinse every surface
that stearic acid touches with a small amount of n-hexane to assure that the stearic acid is removed and recovered.
This includes the inside surfaces of the sample bottle (and liner), the separatory funnel or SPE device used for
extraction, the sodium sulfate used for drying the extract, and the surfaces of the flask used for evaporation of the
solvent.
A technique that will aid in recovery of HEM and SGT-HEM is use of a small stream of n-hexane from a squeeze
bottle to rinse glassware surfaces. For example, after the third extraction using the separatory funnel, use a squeeze
bottle to allow a small amount of n-hexane to stream down the inside of the separatory funnel while rotating the
funnel to assure that the stream will rinse all of the inside surface. In Method 1664A, this technique is termed
"quantitative transfer." (See the definition in the Glossary at the end of Method 1664A).
5-1

Use of a hot plate or Roto-vap for evaporation of the n-hexane can result in loss of n-hexadecane if the solvent is
allowed to go to dryness, because n-hexadecane will evaporate rapidly as the temperature rises above the boiling
point of n-hexane.
Our laboratory is having difficulty achieving the MDL in Method 1664A. Are there any tips that
you can give us?
Obtaining the MDL in Method 1664A is dependent on performing the analysis precisely; i.e., performing it the
same way each time. To perform the analysis precisely requires analyst skill and technique. For example, and in
response to another FAQ, all surfaces that the sample contacts must be rinsed carefully and in the same way with
solvent to recover HEM and SGT-HEM. Certain devices that automate the extraction and/or solvent evaporation
processes can improve method precision, and may allow the MDL to be met more readily.
How can we overcome emulsions formed in the determination of oil and grease?
We have allowed great flexibility in overcoming emulsions, as listed in Section 11.3.5 of Method 1664A. The
optimum technique depends upon the sample, but may include stirring, filtration through glass wool, use of solvent
phase separation paper, centrifugation, use of an ultrasonic bath with ice, addition of NaCl, or other physical
methods. Alternatively, solid -phase extraction (SPE), continuous liquid -liquid extraction, or other extraction
techniques may be used to prevent emulsion formation, provided that the requirements in Section 9.1.2 are met.
The most effective ways that we have found to overcome emulsions are use of solid -phase extraction (SPE), as
described in EPA Method 1613 (40 CFR Part 136, Appendix A) and other methods, and continuous liquid -liquid
extraction (CLLE), as described in EPA Methods 625 and 1625 (40 CFRPart 136, Appendix A) and in other
methods. The advantage of SPE and CLLE is that the sample is not shaken with the extracting solvent, thereby
preventing the incorporation of air into the mixture. Saturation of the sample with salt (NaCl) prior to extraction
using SPE or CLLE may also aid in preventing emulsion formation.
Our discharge contains sulfur that is being converted to thiosulfate upon acidification. The
thiosulfate is being extracted inton-hexane and is inflating results for HEM. How can we solve
this problem?
For this particular instance, where it is known that acidification of the sample is causing a problem, we recommend
that the sample not be acidified. However, it may be difficult to recover stearic acid from the matrix spike if the
sample is not acidic. You should work with your regulatory/control authority to resolve the problem by gathering
data to demonstrate that the problem exists with the particular discharge and by gathering data to demonstrate that
the matrix spike can be recovered effectively if the sample is not acidified.
5-2

Is Non -Polar Material (NPM) another name for Total Petroleum Hydrocarbons (TPH)? Can the
test for NPM be used to analyze for TPH?
Non -Polar Material, or NPM, is another name for Silica -Gel -Treated Hexane -Extractable Material (SGT-HEM),
and is operationally defined (or Method -defined) by the SGT-HEM procedure specified in Method 1664A. In other
words, NPM are the compounds that are extracted from a sample with hexane after that sample has been treated
with silica gel to remove polar materials, as per Method 1664A.
Total petroleum hydrocarbons, or TPH, may or may not be equivalent to NPM. A number of different methods
exist for the analysis of TPH, and many use analytical procedures that differ significantly from Method 1664A.
For example, EPA Method 418.1 for TPH employs sample extraction without silica -gel -treatment, followed by
quantitation of TPH content using infrared analysis. The materials captured by this and other TPH methods may
include or exclude organic compounds that are part of NPM using Method 1664A, and vice versa. Therefore, we
cannot say conclusively that TPH is equivalent to NPM.
How can the non -polar material contribution to a total oil and grease measurement be
distinguished from the fatty acid contribution?
If an environmental sample is composed of non -polar material (such as petroleum hydrocarbons), plus polar
materials (such as glycerides, detergents, and fats), then the hexane -extractable material procedure will measure all
of these substances, whereas the silica -gel -treated -hexane extractable material (SGT-HEM) procedure will measure
the non -polar material after the polar material is removed. The difference between the two measurements will give
the amount of polar material present.
When should the silica -gel -treated, hexane -extractable material (SGT-HEM) procedure be used?
We provided the SGT-HEM procedure for those instances in which a regulatory/control authority believes that it is
appropriate to exclude certain polar materials from a compliance determination.
When CFC-113 becomes unavailable, what analytical method will replace CFC-113-based
procedure for the Coil Coating Point Source Category at 40 CFR 465.03(c)?
Commenters pointed out that we had not proposed revisions to the CFC-based procedure in the Coil Coating
standards at 40 CFR 465.03(c), the Metal Molding and Casting standards at 40 CFR part 464, the Aluminum
Forming standards at 40 CFR 467, and other sections of the CFR that specify use of Method 413.1 or "Freon or
equivalent"
We responded in the administrative record for the final Wile that we would not replace the procedure at 40 CFR
465.03(c) with Method 1664A because we were in the process of re-evaluating the Coil Coating standards. We
also pointed out that the NPM procedure in Method 1664A was virtually identical to the procedure in section
465.03(c), with the exception of the use of n-hexane in place of CFC-113. We further pointed out that the Davison
Grade 950 silica gel specified in section 465.03(c) was no longer commercially available, that Standard Methods
5-3

503A and 503E referenced in section 465.03(c) were outdated, and that Method 1664 contained QC and QC
acceptance criteria for NPM that were not present in the procedure in section 465.03(c).
Our response was based on an interpretation that CFC-113 would be available until 2005 under the exemption in
the Montreal Protocol. However, and as stated in Section 2 of this Guidance, a clarification of requirements in the
Clean Air Act Amendments of 1990 (CAAA) by our Office of General Counsel is that the production and import of
CFC-113 must cease on January 1 of 2000. Supplies of CFC-113 may continue to be recycled and used until the
laboratory use exemption expires in 2005. However, CFC-113 may not be available because of the production
phaseout and cessation of import required by CAAA.
When CFC-113 is no longer available, you will have no choice but to use the silica -gel -treated, hexane -extractable
material (SGT-HEM; non -polar material; NPM) procedure in Method 1664A. As stated throughout this Guidance,
we do not expect significant differences between results produced by CFC-113 and n-hexane. However, if you
believe that a difference could be significant and could affect compliance, you may use the procedure described in
Section 2 of this Guidance, as supported by the examples in Appendix A, to demonstrate to your regulatory/control
authority that a conversion factor is warranted.
In order to adsorb a larger amount of polar material for the SGT-HEM measurement, can I use a
larger amount of silica gel?
YL's.
Does Method 1664A allow use of composite versus grab sampling?
Section 8 of Method 1664A gives requirements for sample collection, preservation, and storage, as outlined below.
Section 8.3 of Method 1664A states that samples must be collected as grab samples because extractable material
may adhere to sampling equipment and result in measurements that are biased low. This condition precludes
collection of a composite sample in the field.
Section 8.3 also provides two approaches to obtaining a composite measurement. Option I consists of collecting
individual grab samples at prescribed time intervals, analyzing each grab separately, and averaging the determined
concentrations. Option 2 consists of collecting four individual 250-mL samples over the course of a day and
compositing these grab samples at the laboratory. At the laboratory, the four 250-mL grab samples are poured into
the separatory funnel, and each sample container and cap liner are rinsed with n-hexane to ensure that all materials
adhering to the walls of the sample containers and to the cap liners are transferred to the separatory funnel. For
more explicit details, please refer to Method 1664A.
5-4

In making its decision to switch from CFC-113, did EPA consider the additional hazards
associated with n-hexane in comparison to CFC-113?
Yes. We searched for a non-flammable solvent capable of producing results equivalent to results produced by
CFC-113. In Phase I of our Freon Replacement Study, we evaluated DuPont 123 (2,2-difluoro- 1, 1, 1 -
trifluoroethane) and tetrachloroethylene. However, neither of these solvents produced results as close to results
produced by CFC-113 as did n-hexane. As a result, we chose n-hexane. To more extensively address the safety
issue, we expanded the section on safety in Method 1664A promulgated in the final rule.
The quality control (QC) in Method 1664A is more extensive than in predecessor oil and grease
methods. Why does you believe that the additional QC is necessary?
We based the decision to add QC to Method 1664A based on concerns from the regulated industry and from a
history that some laboratories were practicing inadequate QC. We have added similar QC to other methods and
believe that this minimum QC is necessary to demonstrate that results are reliable.
Why didn't EPA make solid -phase extraction (SPE) the reference technique in EPA Method
1664A?
We desired to match, as closely as possible, results produced by EPA Method 413.1 and other methods approved
for determination of oil and grease. All of these previously approved methods use liquid -liquid extraction, and the
effluent guidelines were developed using these methods. Had we been starting from scratch, SPE would have been
the reference technique chosen.
If there are such problems with n-hexadecane and stearic acid, why were these compounds
selected as standards rather than Wesson oil or other substances?
In our Freon replacement studies, industries were segmented into two main categories: petroleum and non -
petroleum. The major component of petroleum -based samples is diesel oil, and the main component of diesel oil is
n-hexadecane. The major component of non -petroleum -based samples is animal fat, and the main component of
animal fat is stearic acid. We chose n-hexadecane and stearic acid as the compounds to be used for QC tests in
Method 1664A so that we can be sure that the major components of petroleum -based and non -petroleum based
samples are recovered.
We have preliminary data indicating that some of the hexadecane/stearic acid standard may be
lost by evaporation using the procedure in Section 11.4.4 for drying the boiling flask. Has EPA
seen other evidence of this problem?
Not until you indicated that the problem may exist. The procedure for drying the boiling flask at 70 °C for 30 - 45
minutes followed by dessication for 30 minutes was suggested by the American Petroleum Institute as a means for
assuring that residual water or n-hexane would not affect the measurement. If this problem is experienced, we suggest
a lower temperature or a shorter drying time.
5-5

An SPE vendor has stated that we should spike the hexadecane/stearic acid into the SPE reservoir,
and not into the sample container, so that we can recover the matrix spike more efficiently. Is this
allowed?
No. You must spike into the sample container, as stated in Section 9.3.2.2 of Method 1664A, because you must
demonstrate recovery of the matrix spike (MS) from the sample container.
You can improve HEM and NPM recovery when SPE is used by rinsing the sample bottle with small portions (e.g.,
3 x 10 mL) of n-hexane and using these portions to elute the HEM or NPM from the SPE disk. The procedure in
Section 11.3.3 of Method 1664A gives detail on how to rinse the sample container.
The weighing procedure in Section 11.4.4 requires drying and re -weighing until constant weight is
achieved. Should the final weight be used or should the final two weights be averaged?
The final weight should be used as the value for HEM or NPM.
What is the status of a method to replace EPA Method 418.1 ?(Method 418.1 uses CFC-113 and infra-
red spectroscopy (IR) to measure IR-sensitive substances in the sample).
We continue development work on a replacement for Method 418.1. The solvent being studied most intensely at the
time of writing of this Guidance is tetrachloroethene (tetrachloroethylene; perchloroethylene; perc). The number of
solvents available for the IR method is severely limited because any solvent that contains a carbon -hydrogen bond will
interfere in the determination. Most of the remaining solvents are hazardous (e.g., carbon disulfide), and perc is not
without hazard. The most significant problem encountered with perc is that a stabilizer is needed to prevent
degradation. To date, stabilizers have had a carbon -hydrogen bond, thus compromising IR measurements. We are
working with manufacturers of perc to attempt to overcome this problem.
Instead of dealing with the solvent issue, why didn't EPA simply abandon the oil and grease test
and rely on measurements of specific organic compounds for pollution control?
The Clean Water Act (CWA) requires regulation of "conventional pollutants," and oil and grease has been designated
as a conventional pollutant under CWA. In order to eliminate oil and grease as a conventional pollutant, either CWA
would need to be changed or designation of oil and grease as a conventional pollutant by the EPA Administrator would
need to be rescinded. Changing CWA would take an act of Congress, and such a change would likely be lengthy and
contentious. Similarly, the rescission of designation by the EPA Administrator would need to be proposed and would
likely be contentious. Rather than attempt to change the CWA or the designation, we decided to move forward with
a change in solvent.
(1) Section 18.2.2 of Method 1664 defines an analytical batch as a group of up to 20 samples
processed together in a 12-hour shift. It does notspecify that the samples be of the same matrix
type. It does specify that each analytical batch include a laboratory blank, anongoing precision
and recovery sample (OPR), and a matrix spike (MS); (2) Section 18.2.2 refers to Section 9.3
5-6

regarding a definition of the MS; (3) Section 9.3 requires "The laboratory must spike a minimum of
5 percent of all samples from agiven sampling site or, if for compliance monitoring, from a given
discharge/waste stream." The dilemma is that the requirements for MS in Sections 18.2.2 and 9.3
appearto be different with respect to both frequency and matrix type. Several interpretations seem
possible. Our questions are:
1. Is there a requirement for a separate analytical batch (blank, OPR, and MS) for each
discharge/waste stream?
Requirements for a batch and discharge/waste stream are different. (For a definition of discharge/waste stream,
see the Glossary at the end of Method 1664A.) The requirement for an analytical batch is that it must include a
blank, OPR, and MS. A discharge/waste stream must be spiked at a minimum frequency of 1 in 20 samples; i.e.,
the first sample must be spiked, the 21 st sample must be spiked, etc.
2. Or, is the requirement a blank and an OPR per 12 hour shift in the lab, with an MS per
sample?
The batch is comprised of the samples started through the extraction process during a given 12-hour shift, to a
maximum of 20 samples. If only 1 sample is started through the extraction process, the batch size is 1; if 20
samples are started through the extraction process, the batch size is 20.
3. Or, is there a requirement for a blank and an OPR per analytical batch (18.2.2), with a matrix
specific MS per 20 sampling events per effluent stream (9.3) overtime, i.e. across analytical
batches?
Both. The requirement for a batch is that it must include a blank, OPR, and MS. The requirement for the
frequency of the MS per discharge/waste stream is that the sample from the discharge/waste stream must be spiked
at a frequency of 5 percent (a minimum of I in 20 samples).
NOTE: The requirement for an MS per batch could result in a spike of the matrix type at a frequency
of greater than 20 percent. For example, if a discharge is monitored daily and is the only sample in
the batch, it must be spiked every day.
4. Or, is it acceptable to perform a blank, an OPR, and a randomly selected MS per analytical
batch?
The MS can be selected randomly, provided that each discharge/waste stream is spiked at a frequency no less than
5 percent. For example, if there are 20 samples in the batch, and all had been spiked during the previous ten times
that they were analyzed, none are required to be spiked. However, because of the requirement for an MS with the
batch, any one of the samples may be chosen for the MS.
So, for example, if 2 waste streams are in a 20 sample are run within a 12-hour shift, it is clear
that a blank and an OPR are required, but must 2 matrix spikes (one for each waste stream) be
5-7

run to be certain of achieving the 5% spike requirements for both the discharge/waste stream
and the analytical batch?
If the waste streams had not been spiked in the previous 20 samplings, an MS must be run on each. Thereafter,
the MS can be staggered between the waste streams so that 2 MSs are not required with each batch.
Explanations:
Batch and discharge/waste-stream-specific quality control (QC)
Batch -specific QC is required to demonstrate that the analytical process is in control during the 12-hour shift that
samples, blanks, and standards (OPRs) will be analyzed. Discharge/waste-stream-specific QC is required to assure
that the method continues to be applicable to that specific dischargelwaste stream-
Discharge/waste stream (matrix type)
Discharge/waste stream are synonymous with "matrix type." The term "matrix type" was created to address
wastewater streams in a given industrial subcategory. Industrial subcategories are defined in the wastewater
regulations at 40 CFR parts 400-699. "Matrix type" means a sample medium with common characteristics across
a given industrial subcategory. Examples include: C-stage effluents from chlorine bleach mills in the Pulp, Paper,
and Paperboard industrial category; effluents from the continuous casting subcategory of the Iron and Steel
industrial category; publicly owned treatment work (POTW) sludges; and effluents being discharged to POTWs
fromplants in the Atlantic and Gulf Coast Hand -shucked Oyster Processing subcategory. "Sample medium" means
the physical phase of a sample matrix. Air, water, soil, sediment, rock, and sludge are sample media.
Strictly speaking, the discharge/waste-stream-specific QC would be applicable on the basis of matrix type; i.e., if
a commercial laboratory was analyzing C-stage effluents from several chlorine bleach mills in the Pulp, Paper, and
Paperboard industrial category, it would be necessary to collect and analyze an MS from any one of the C-stage
effluents, only, to satisfy the discharge/waste-stream-specific QC. The reason that a single sample can be used to
represent a particular wastewater stream in a given industry is that the industry is subcategorized based on the
product produced and the process used to produce that product. Wastewater streams from the same product/
process will have the same characteristics.
In practice, it is unlikely that a commercial laboratory would know the specific stream sampled or the specific
industrial subcategory. Therefore, it would be prudent for the discharger/industrial user to provide an extra sample
for the matrix spike to meet the frequency requirement of 5 percent (1 in 20 samples).
5-8

6 Sources of Information
his section provides sources of information related to the final guidelines establishing test procedures for the
analysis of oil and grease. Specifically, this section provides a listing of documents pertaining to the
regulatory background and data gathering for oil and grease.
Regulatory Background
Acts
Clean Water Act (CWA) - Public Law 92-500, et. seq.; 33 U.S.C. 1251 et. seq.
Resource Conservation and Recovery Act (RCRA) - Public Law 94-580, et. seq.; 42 U.S.C. 6901 et. seq.
Clean Air Act Amendments (CAAA) - Public Law 101-549, et. seq.
Analytical methods under CWA Section 301, 304, and 601
History: see Federal Register, February 7, 1991 (56 FR 5090)
Support for effluent guidelines: see Federal Register, October 18, 1995 (60 FR 53988).
Proposal of Method 1664: See Federal Register, January 23, 1996 (61 FR 1730).
Analytical methods under RCRA
History: see Federal Register, June 13, 1997 (62 FR 32452)
Applicability to methods under CAAA: see Federal Register, July 3, 1991 (56 FR 30519)
Data Gathering for Method 1664A
Proposal
See the administrative record in the Water locket for the proposed rule (61 FR 1730) for the following study
reports
Phase I Freon Replacement Study (EPA-821-R-93-011)
Phase H Freon Replacement Study (EPA-820-R-95-003)
Method Validation Study (82 1 -R-95-036)
Notice of availability
See Federal Register, October 2, 1997 (62 FR 51621)
Water Pollution performance evaluation (PE) data
See Table 1 in the preamble to the final nde (64 FR 26316) for pertinent data
Data received from commenters
See the administrative record in the Water Docket for the final rule
6-1

Interpretation of Requirements in Clean Air Act Amendments
Letter from Drusilla Hufford, Director, Stratospheric Protection Division (6205J), Office of Atmospheric
Programs, Office of Air and Radiation, to Regional Quality Assurance Managers, August 30, 1999
"Request for Applications for Essential Use Exemptions to the Production and Import Phaseout ofOzone Depleting
Substances Under the Montreal Protocol" (64 FR 50089, September 15, 1999)

Where to Get Additional
Help
rovided in this section are additional sources of information and EPA contacts that may provide additional
information related to the final guidelines establishing test procedures for the analysis of oil and grease.
Specifically, this section presents a list of documents and websites relating to the fatal rule for the analysis of
oil and grease. These lists also include information on how to reach EPA program personnel and how to access
these information sources.
Questions specifically related to Method 1664A should be directed to:
Maria Gomez -Taylor
Engineering and Analysis Division (4303)
U.S. EPA
401 M Street, SW
Washington, DC 20460
Tel: [phone]
Fax: [phone]
E-Mail: [e-mail]
Questions specifically related to the use of Method 1664A in the Resource Conservation and Recovery Act program
should be directed to the regional delisting coordinator in each Regional Office:
U.S. EPA Reeion I
U.S. EPA Re£1011 lI
One Congress Street
290 Broadway
John F. Kennedy Federal Building
New York, NY 10007-1866
Suite 1100
Tel: [phone]
Boston, MA 02114-2023
Fax: [phone]
Tel: [phone]
Fax: [phone]
U.S. EPA Reeion III
U.S. EPA Reeion iV
1650 Arch Street
61 Forsyth Street, S.W.
Philadelphia, PA 19103-2029
Atlanta, GA 30303-3415
Tel: ( 215) 814-5000
Tel: [phone]
Fax: [phone]
Fax: [phone]

U.S. EPA Region V
U.S. EPA Region VI
77 West Jackson Boulevard
Fountain Place 12th Floor, Suite 1200
Chicago, IL 60604-3507
1445 Ross Avenue / 6PD
Tel: [phone]
Dallas, TX 75202-2733
Fax: [phone]
Tel: [phone]
Fax: [phone]
U.S. EPA Region VII
U.S. EPA Region VIII
726 Minnesota Avenue
999 18th Street, Suite 500 / 8P-HW
Kansas City, KS 66101
Denver, CO 80202-2466
Tel: [phone]
Tel: [phone]
Fax: [phone]
Fax: [phone]
U.S. EPA Region IX
U.S. EPA Region X
75 Hawthorne Street
1200 Sixth Avenue
San Francisco, CA 94105
Seattle, WA 98101
Tel: [phone]
Tel: [phone]
Fas:[phone]
Fax: [phone]
The administrative record (public comments, EPA responses, and all supporting documents for Method 1664A) are
available for review at the Water Docket. For access to docket materials, phone the Water Docket between 9:00 a.m.
and 3:30 p.m. for an appointment:
Water Docket
U.S. EPA
401 M Street, SW
Washington, DC 20460
Tel: [phone]
Documents on Compliance Monitoring and Methods
# Guidance on Evaluation,Resolution, and Documentation ofAnalytical Problems Associated with
Compliance Monitoring, EPA-821-B-93-00" June 1993.
Websites
# EPA's homepage on the World Wide Web:
http://www.epa.gov
# EPA's Monitoring Science in the RCRA Program (OSW methods Team Homepage):
http://www.epa.gov/SW-846
# EPA's Office of Science and Technology's analytical methods page on the World Wide Web:
http://www.epa.gov/OST/Methods/ (water methods)
http://www.epa.gov/ost/guide/ (water documents)
http://www.epa.gov/ttn/oarpg (air documents)
7-2

Appendix A
Example Calculations for Side -by -Side Comparisons
To allow you to make the calculations detailed in this Appendix easily, we have produced an Excel spreadsheet that
does all calculations automatically after you have entered the necessary results. This spreadsheet is available via an
E-mail attachment from our Sample Control Center (SCC). Please E-mail your request to [e-mail], or you
may phone your request to 703-461-2100. If you telephone, please ask for the "Method 1664A Comparison
Spreadsheet" and provide your E-mail address. You must have Microsoft's Excel 97 to run the spreadsheet.
Examples of a side -by -side comparison to determine if a conversion factor is warranted and to
calculate the conversion factor if it is.
Example 1:
Steps 1-3: Eight samples were collected on a total of seven days spaced over a minimum of a 1-month period. For the
samples collected on each day, three were analyzed by EPA Method 1664A and three by EPA Method 413.1(for a total
of 6 samples per day). All QC tests were performed and the results of the QC tests (including the matrix spike test)
met the QC acceptance criteria in Method 1664A. Results of the triplicate analyses by each method on each day are
listed in Table I.
Table 1. Example results for side -by -side study*
Results (mg/L)
Method 1664A
Method 413.1
Date
Sample
rep #1
rep #2
rep #3
rep #1
rep #2
rep #3
8/13
1
13
18
12
23
28
29
8/17
2
10
12
14
21
19
17
8/22
3
22
26
22
41
36
35
8/25
4
12
17
15
26
27
24
8/30
5
20
17
12
26
23
27
9/4
6
37
35
31
52
48
40
9/9
7
11
14
13
21
20
25
*Data in this data set are estimations of data that could be produced in a side -by -side study and are not "real -world"
data. For application of the comparison of results produced by alternative solvents to results produced by CFC-113
using "real -world" samples, see the Freon Replacement Study reports.
A-1

Step 4: The (natural log) log-hansformed results are listed in Table 2:
Table 2. Log -transformed results
Results (log mg1L)
Method 1664A
Method 413.1
Date
Sample
rep #1
rep #2
rep #3
rep #1
rep #2
rep #3
8/13
1
2.56
2.89
2.48
3.14
3.33
3.37
8/17
2
2.30
2.48
2.64
3.04
2.94
2.83
8/22
3
3.09
3.26
3.09
3.71
3.58
3.56
8/25
1 4
2.48
2.83
2.71
3.26
3.30
3.18
8/30
5
3.00
2.83
2.48
3.26
3.14
3.30
9/4
6
3.61
3.56
3.43
3.95
3.87
3.69
9/9
7
2.40
2.64
2.56
3.04
3.00
3.22
Step 5: The 14 triplicate means of the log -transformed results are listed in Table 3:
Table 3: Triplicate means (M,1)
Sample Q)
Method 1664A (i=1)
Method 413.1 (i=2)
1
2.65
3.28
2
2.48
2.94
3
3.15
3.62
4
2.68
3.24
5
2.77
3.23
6
3.53
3.84
7
2.53
3.09
A-2

Step 6: The 14 triplicate standard deviations of the log -transformed results are listed in Table 4:
Table 4: Triplicate standard deviations (s,t)
Sample 0)
Method 1664A (i=1)
Method 413.1 (i=2)
1
0.215
0.125
2
0.168
0.106
3
0.096
0.084
4
0.176
0.060
5
0.261
0.084
6
0.090
0.134
7
0.124
0.117
Step 7: The mean -square error is:
MSE= 14[(0.215)Z+(0.168)2+(0.096)2+...+(0.134)2+(0.117)']=0.020
Step 8: The method -specific means are:
MMi = � K,,..,7IMli = l[2.65+2.48+3.15+2.68+2.77+3.53+2.53]=2.83
Similarly, MMZ = 3.32.
MS,= 1 [2.65 + 3.28] = 2.97
Similarly, Msz = 2.71, Ms3 = 3.38, Ms, = 2.96, MS5 = 3.00, M,, = 3.69, and Ms7 = 2.81.
Ma= 1[2.83+3.32] = 1F/d'Ct..,7l[2.96+2.71+3.38+2.96+3.00+3.69+2.81] = 3.07
A-3

Step 9: The RMSD is:
RMSD=Y![(2.65-3.28)2+((2.48-2.94)2+...+(2.53-3.09)z=0.503
Step 10: The RMSD�m is:
RMSDMA = 2 * 0..020 * 2.36 = 0.177
Step H: The compared result is:
Because 0.503 > 0.177, the difference between the two methods is significant. Therefore, move on to step 12
Step 12: The mean square attributable to method/sample interaction is:
MSRgT =Z[(2.65-2.83-2.97+3.07)z+(2.48-2.83-2.71+3.07)Z+...+
(3.09- 3.32- 2.81 + 3.07)2] = 0.017
Step 13: Fw is:
F,,k = (0.017 / 0.020) = 0.85. Because 0.85 < 2.45, it cannot be concluded that there is a significant interaction
between method and sample.
Because the RMSD was significant, and there was not a significant interaction between method and sample, a
conversion factor should be calculated.
Step 14: The log -transformed factor is:
CFI = 3.32 / 2.83 = 1.17.
A-4

Concentration adjustment for a new result
Suppose a new value of 23 mg/L is obtained using method 1664A. The log -transformed value of this result is = 3.14.
The converted value is:
XR = EXP[3.14 * (1.17)] = 39 mg/L
where XR is the converted value in the original scale, and Y,, is the log -transformed value obtained using 1664A.
Therefore, the result that would have been obtained had Method 413.1 been used is 39 mg/L.
Example 2:
Eight samples were collected on a total of seven days spaced over a minimum of a I -month period. For the samples
collected on each day, three were analyzed by EPA Method 1664A and three by EPA Method 413.1. Results of the
analyses are listed in Table 5.
Table 5. Example results for side -by -side study`
Results (mg/L)
Method 1664A
Method 413.1
Date
Sample
rep #1
rep #2
rep #3
rep #1
rep #2
rep #3
8/13
1
20
17
21
25
28
32
8/17
2
21
20
23
29
26
26
8/22
3
29
27
31
23
25
26
8/25
14
34
36
34
25
29
28
8/30
5
30
27
32
22
24
24
9/4
6
16
18
14
19
18
20
9/9
7
22
18
21
21
19
23
*Data in this data set are estimations of data that could be produced in a side -by -side study and are not "real -world"
data. For application of the comparison of results produced by alternative solvents to results produce by CFC-113
using "real -world" samples, see the Freon Replacement Study reports.
A-5

Step 4: The log -transformed results are listed in Table 6:
Table 6: Log -transformed results
Results (log mg/L)
Method 1664A
Method 413.1
Date
j
rep #1
rep #2
rep #3
rep #1
rep #2
rep #3
8/13
1
3.00
2.83
3.04
3.22
3.33
3.47
8/17
3.04
3.00
3.14
3.37
3.26
3.26
8/22
t3::]
3.37
3.30
3.43
3.14
3.22
3.26
8/25
4
3.53
1 3.58
3.53
3.22
1 3.37
3.33
8/30
5
3.40
3.30
3.47
3.09
3.18
3.18
9/4
6
2.77
2.89
2.64
2.94
2.89
3.00
9/9
7
3.09
2.89
3.04
3.04
2.94
3.14
Step 5: The 14 triplicate means of the log -transformed results are listed in Table 7:
Table 7: Triplicate means (M;t)
Sample Q)
Method 1664A (i=1)
Method 413.1 (i=2)
1
2.96
3.34
2
3.06
3.29
3
3.37
3.20
4
3.55
3.31
5
3.39
3.15
6
2.77
2.94
7
3.01
3.04
A-6

Step 6: The 14 triplicate standard deviations of the log -transformed results are listed in Table 8.
Table 8: Triplicate deviations (s;;) of the log -transformed results
j
Method 1664A (i=1)
Method 413.1 (i=2)
1
0.111
0.124
2
0.071
0.063
3
0.069
0.063
4
0.033
0.078
5
0.086
0.050
6
0.126
0.053
7
0.105
0.096
Step 7: The mean -square error is:
MSE= 1) [(0.111)2+(0.071)z+(0.069)s+...+(0.053+(0.096)2]=0.0072
Step 8: The method -specific means are:
MMt= 7 Ft,..,7IMti = I[2.96+3.06+3.37+3.55+3.39+2.77+3.01]=3.16
Similarly, M65: = 3.18.
Mst= L[2.96+3.34]=3.15
Similarly, M11 = 3.18, MS3= 3.28, MS4 = 3.43, M55 = 3.27, M$6= 2.86, and MS7= 3.03.
Mo= 2[3.16+3.18] = �[3.15+3.18+3.28+3.43+3.27+2.86+3.03] = 3.17
A-7

Step 9: The RMSD is:
RMSD= 6[(2.96-3.34)Z+(3.06-3.29)2+...+(3.01-3.04)Z=0.231
Step 10. The RMSDkp�x is:
2 * 0.0072 * 2.36 = 0.106
RMSDMAX= 3
Step H: The compared result is:
Because 0.231 > 0.106, the difference between the two methods is significant. Therefore, move on to step 12.
Step 12: The mean square attributable to method/sample interaction is:
MSINT = 1[(2.96-3.18-3.15+3.19)2+(3.06-3.18-3.18+3.19)2+...+
(3.04- 3.18- 3.03+ 3.17)2] = 0.092
Step 13: F,, is:
F�,i (0.092 / 0.0072) = 12.77. Because 12.77 > 2.45, there is a significant interaction between method and sample.
Therefore, a conversion factor is not warranted.
A-8

Appendix B
Annotated Initial Demonstration of Laboratory Capability -for SPE
NOTE: The section numbers below refer to section numbers in EPA Method 1664A. Text in brackets and italics [text]
is additional guidance and is not part of Method 1664A.
9.2 Initial demonstration of laboratory capability.
9.2.1 Method Detection Limit (MDL)—To establish the ability to detect HEM and SGT-HEM, the
laboratory shall determine the MDL per the procedure in 40 CFR 136, Appendix B using the
apparatus, reagents, and standards that will be used in the practice of this method. An MDL less than
or equal to the MDL in Section 1.6 or less than 1/3 the regulatory compliance limit must be achieved
prior to the practice of this method.
[The MDL part of the demonstration with SPE does not need to be performed if the MDL was demonstrated
previously with unmodified Method 1664A, and if the MDL is not expected to be affected by the measurement (see
Section 9.1.2.1 ofMethod 1664A for this requirement). If the MDL is expected to be affected by use ofSPE (or other
modification), the MDL must be demonstrated per requirements in Sections 9.1.2.1 and 9.2.1 ofMethod 1664A.]
9.2.2 Initial precision and recovery (IPR�—To establish the ability to generate acceptable precision and
accuracy, the laboratory shall perform the following operations:
9.2.2.1 Determine the concentration of HEM and/or SGT-HEM in four samples ofthe PAR standard
(Section 7.11) according to the procedure beginning in Section 11.
[For this test, four I -L aliquots ofreagent water are spiked with hexadecane and stearic acid and processed through
the entire analytical Method, including sample preservation, using SPE (or other mod (cation) as an integral part
of the Method.]
9.2.2.2 Using the results of the set of four analyses, compute the average percent recovery (X) and
the standard deviation of the percent recovery (s) for HEM and for SGT-HEM (if determined).
When determining SGT-HEM, the true concentration (T) must be divided by 2 to reflect the
concentration of hexadecane that remains after removal of stearic acid. Use the following
equation for calculation of the standard deviation of the percent recovery:
Equation 1
&2 - (S)2
s =
n
n - 1
where:
n = Number of samples
x = % Recovery in each sample

9.2.2.3 Compare s and X with the corresponding limits for initial precision and recovery in Table 1.
If s and X meet the acceptance criteria, system performance is acceptable and analysis of
samples may begin. If, however, s exceeds the precision limit or X falls outside the range for
recovery, system performance is unacceptable. In this event, correct the problem and repeat
the test.
[Table I of Method 1664A gives the QC acceptance criteria for the IPR test. These criteria must be met with SPE
(or other modification) as an integral part of Method 1664A. If not, the modification must be improved and the test
repeated. Only after the QC acceptance criteria have been met with SPE (or other modification) as an integral part
of the test may the laboratory apply SPE (or other modification) to the sample matrix (see Appendix Q.]

Appendix C
Annotated Equivalency Demonstration for Application of a Method Modification to Compliance
Monitoring for SPE
NOTE: The section numbers below refer to section numbers in EPA Method 1664A. Text in brackets and italics [text]
is additional guidance and is not part of Method 1664A.
9.2.3 Equivalency demonstration for application of a method modification to compliance monitoring —To
establish the ability of a modification of this method to recover an amount of HEM and/or SGT-HEM
equivalent to the amount recovered by this method from a specific discharge/waste stream, proceed
as follows:
9.2.3.1 Collect, extract, concentrate, and weigh the HEM or SGT-HEM in two sets of four aliquots
of unspiked wastewater. One set of four wastewater aliquots is analyzed according to the
protocol in Section 11 of this method and the other set of four aliquots is analyzed using the
modified method
[For the equivalency demonstration with SPE, SPE constitutes the modification.]
9.2.3.2 Calculate the average concentration of HEM and SGT-HEM for the set of results from this
method and for the set of results from the modified method. The average concentration using
the modified method must be 78 to 114 percent of the average concentration produced by this
method for HEM and 64 to 132 percent of the average concentration produced by this method
for SGT-HEM. If not, the modified method may not be used.
[The QC acceptance criteria (78-114%for HEM and 64-1325Yofor SGT-HEM) are identical to thosefor the 1PR test
(Section 9.2.2) but are applied to results obtained with LLE as the reference. This assures that if LLE recovers a
lower amount of material from the sample matrix than recovered in the 1PR test, the test will not be more stringent
for SPE (or other modification).]
NOTE: If the average concentration of the four results produced using this method and the average
concentration of the four results produced using the modified method are below the minimum level
(Section 1.6), and if the equivalency test of the modified method is passed for spikes of reference
standards into reagent water (Section 9.2.2), the modified method is deemed to be equivalent to this
method for determining HEM and or SGT-HEM on that specific discharge/waste stream.
[This allowance for use of SPE (or other modification) when the concentration of HEM or SGT-HEM is below the
ML addresses the issue that a reliable comparison between the LLE and SPE (or other modification) cannot be made
at levels too low to be measured reliably.]
C-1