United States
 Environmental Protection
 Agency
 Environmental Monitoring and
 Support Laboratory
 Cincinnati OH 45268
 Research and Development
 EPA-600/S4-85-001   Feb. 1985
Project  Summary
 Evaluation of  EPA  Method  603
 (Modified)

 S. V. Lucas. T. F. Cole, A. Riggin, and W. M. Cooke
  EPA Method 603 was modified and
evaluated with newly established chro-
matographic conditions for the deter-
mination of acrolein, acrylonitrile, and
acetonrtrile. The recommended chroma-
tographic column and conditions speci-
fied in the original method were found
to be unsuitable when acetonitrile was
included as a method parameter. Pora-
pak QS wa» found to perform well as a
gas chromatography (GC) packing for
the three method parameters, and
method validation using reagent water.
Publicly  Owned Treatment  Works
(POTW)  outfall water,  and  industrial
wastewater was successfully perform-
ed.  Recoveries and precisions were
determined at two spiking levels in each
of the three matrices. Seven replicates
were analyzed for each of the six spike-
level/matrix  sets.  For acetonitrile,
recoveries for the six sets ranged from
86 to 111 percent (average. 99 percent),
with precisions ranging from 3 to 17
percent relative standard  deviation
(RSD), (average, 11 percent RSD). The
corresponding values for acrylonitrile
were recoveries ranging from 84 to 104
percent (average, 97 percent) and pre-
cisions ranging from 3  to 8 percent
RSD (average. 4 percent RSD). Acrolein
was unstable in  the industrial waste-
water, even though the water was
buffered at pH 4.5 before spiking. For
the other two matrices, the acrolein
recovery ranged from 80 to 104 percent
(average, 94 percent) and the precisions
ranged from 1  to 4  percent RSD
(average, 3 percent RSD). The loss of
acrolein  due to apparent chemical or
biological degradation in the industrial
wastewater underscores the  current
lack of understanding of the chemical
factors that control  the stability of
acrolein in aqueous systems. All of the
data sets were quantified using both an
internal standard method and an ex-
ternal  standard  method, and  both
methods of quantification provided
essentially identical accuracy and pre-
cision results.
  This  report was submitted in partial
fulfillment of Contract Number 68-03-
1760, Work Assignment Number 5, by
Battelle's Columbus Laboratories. The
research reported here was conducted
under  the  sponsorship of the  U.S.
Environmental Protection Agency.
  This Project Summary was developed
by EPA's Environmental Monitoring and
Support Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).


Introduction
  The Environmental Protection Agency's
(EPA's)  Environmental  Monitoring and
Support Laboratory (EMSL), Cincinnati,
Ohio, has primary responsibility for devel-
oping and evaluating monitoring methods
for organic pollutants in industrial waste-
waters.  Method 603, developed by EPA,
for wastewater monitoring involves the
use of a heated purge-and-trap technique
and gas chromatographic (GC) analysis
for the  determination of acrolein and
acrylonitrile.
  This  method validation study  was
undertaken in response to the experience
of various laboratories which suggested
that Method 603 was  not a reliable
analysis method. For example, a previous
single laboratory method validation study
performed for EPA by another contracting

-------
laboratory produced no  useful data. In
addition, several investigators have quer-
ied EPA regarding difficulties with the
reproducibilityof thegaschromatography
and problems encountered from acetone
interference. Thus, development and use
of improved chromatographic conditions
was a critical first step in this work.
  In an effort to increase the usefulness
and  scope  of  Method 603,  Battelle
Columbus Laboratories was contracted to
conduct research to:  (1)  recommend
modifications to Method 603 for improv-
ing the GC analysis;  (2)  evaluate the
feasibility of including acetonitrile as a
method parameter; and (3)  validate the
method, as modified, using three matrices
(reagent water, POTW wastewater and a
relevant industrial wastewater) and two
spiking  levels ([a "low" level near the
method detection limit [MDL] and a "high"
level at least 10-fold higher than the
"low" level]).


Analytical Method Evaluation

GC Column Evaluations
  Since acetonitrile was to be included in
the parameter list of Method 603, the
identification of a suitable GC column for
the separation of all three  compounds,
acetonitrile, acrolein,  and acrylonitrile,
was a necessary first step. The separation
of methylene chloride and acetone from
the method parameters was also a con-
sideration in selecting the column to be
used for validation.
  Both of the columns specified  in the
1982 revision of Method 603 were de-
ter mined to be unacceptable. Chromosorb
101 resulted in exact coelution of acrolein
and acetonitrile and partial resolution of
acetone from acrylonitrile but otherwise
it performs adequately for acrolein and
acrylonitrile in the absence of acetonitrile.
Durapack CW400/Porasil C resulted in
coelution of acetone and  acrolein  and
such severe peak tailing that low level
injections (100 ng) did not provide usable
data. Other column packings were tried,
and the results obtained were as follows:
(1) 1  percent SP-1000/Carbopack  B
(moderate tailing and acrolein-acetone
coelution); (2) 0.1 percent CW400/Car-
bopack C  (inadequate retention  and
severe peak tailing); and (3) Chromosorb
104 (high temperature of elution, unac-
ceptable isothermal  elution time  for
acetonitrile and  moderate  tailing).  The
column packing used in this work (Porapak
QS) was the only one tested that provided
good GC peak shape at low levels and
baseline separation between the method
analytes as well as two potential inter-
ferences  of interest  to EPA (acetone,
which elutes between acrolein and acryl-
onitrile, and methylene chloride, which
elutes  between methanol  and acetoni-
trile).

Purge-and-Trap Conditions
  The purge-and-trap conditions for this
validation effort included an 85°C purge
with helium at 20 mL/minute flow rate
for 15 minutes, a 1.5  minute desorption
at 180°C, and a 210°C trap bake for 10
minutes. These conditions  were recom-
mended by EMSL staff, and were compared
to the original Method 603 conditions. In
summary, the conditions gave improved
peak heights for acetonitrile and acrolein
and better reproducibility, due, apparently,
to the  more rapid delivery  of trapped
analytes to the GC column and the
resulting production of narrower  and
taller GC peaks. Actual recoveries, how-
ever, were  about the same for the two
conditions. The trap bake temperature of
210°C was employed, since it is generally
desirable that the trap bake temperature
be higher than the desorption tempera-
ture. The low (100°C)  desorption  tem-
perature specified in the 1982 version of
Method 603 was required  by the initial
GC column  temperature of 45°C and the
sensitivity of Durapak CW400/Porasil C
to the  presence of liquid water on the
column, which would be caused by rapid
trap heating to 180°C.  For the Porapak
QS column  packing, problems caused by
liquid water are of no concern, since the
initial  column  temperature is 110°C.
Moreover,  macroreticular  phases are
generally not susceptible to water dam-
age.

Method Validation
  Validation consisted of the analysis of
seven replicates at two spiking levels for
each of three  matrices: reagent water,
Columbus POTW secondary effluent, and
industrial wastewater from the outfall of
a manufacturing plant producing two of
the analytes. The water samples were
buffered before spiking at pH 4.5 ±0.2 by
the addition of sodium citrate to approxi-
mately 0.1 m/W and pH adjustment using
a glass electrode pH  meter.  Spiked
samples  were allowed to equilibrate
overnight  at  ice temperature  before
analysis on the following day.
  A summary of the results for the method
validation using reagent water, Columbus
POTW effluent, and the industrial waste-
water are presented in Tables 1,2, and 3,
respectively.  All  quantifications were
performed using both external standard
calibration  and  internal standard (IS)
calibration with 1 -chloropropane, but only
the external standard results  are  pre-
sented in the  tables.  In general, no
significant differences were observed in
the results for the two methods of quanti-
fication. Acceptable recoveries and pre-
cisions were obtained for all  method
parameters at both spiking levels in the
reagent water  and  Columbus POTW
effluents. For the industrial wastewater,
acrolein apparently decomposed or was
biodegraded despite pH adjustment to a
value of 4.5 prior to  spiking,  and the
recoveries for this matrix were less than
12 percent for all replicates analyzed.
  With the  results for  reagent water
spiked at the lower level, MDLs of 2.7,
0.7, and 0.5 were calculated for aceto-
nitrile, acrolein,  and acrylonitrile, respec-
tively.

Summary and Conclusions
  The validation data collected in this
research support the  usefulness  of
Method 603 for the analysis of acetoni-
trile, acrolein, and acrylonitrile using the
newly established chromatographic con-
ditions. The following conclusions can be
drawn:

• Both of the previously recommended
   GC columns  are unsuitable  for the
   separations needed if acetonitrile is to
   be included as a method parameter. A
   Porapak  QS column provides the re-
   quired separation of all three analytes.
• The method appears to be precise and
   accurate for acetonitrile determination
   even though  the purging efficiency of
   acetonitrile is only about 50 percent.
• No apparent advantage is gained  by
   the use of an internal standard calibra-
   tion procedure. However, the quality
   assurance feature  of an  internal
   standard  may have  merit in some
   cases.
• Factors that affect acrolein stability in
   aqueous  samples are still not  well
   understood.

-------
Table 1.
Method
Parameter
Acetonitrile
Acrolein
Acrylonitrile
Method Validation Result* for Reagent Water Using Modified Method 603
Low Level Spike Results
Background
Level, fjg/L
0.9
0.0
1.1
Level
Spiked,
M9/L
10.0
5.0
5.0
Average
Recovery,
W/L
9.5
5.2
5.3
SD1
0.8
0.2
0.1
RSD,2
%
9
4
3
Average
Percent
Recovery
86
104
84
Background
Level, fjg/L
1.3
0.2
1.0
High Level Spike Results
Level
Spiked,
pg/L
100.0
50.0
50.0
A verage
Recovery,
fjg/L
112
51.6
524
RSD,
SD %
12 11
0.7 1 4
1.5 3
A verage
Percent
Recovery
111
103
103
^Standard Deviation.
'Percent Relative Standard Deviation.
Table 2. Method Validation Results for Columbus POTW Secondary Effluent Using Modified Method 603
Low Level Spike Results High Level Spike Results
Method
Parameter
Acetonitrile
Acrolein
Acrylonitrile
Background
Level, fig/L
0.3
0.2
0.9
Level
Spiked,
ng/L
10.0
5.0
20.0
Average
Recovery,
fg/L
10.4
4.2
21.0
SD1
1.5
0.2
0.8
RSD,2
%
14
4
4
Average
Percent
Recovery
101
80
100
Background
Level, fig/L
0.3
0.2
0.9
^Standard Deviation.
'Percent Relative Standard Deviation.
Table 3. Method Validation Results for Industrial Wastewater Using Modified Method 603
Low Level Spike Results
Method
Parameter
Acetonitrile
Acrolein
Acrylonitrile
Background
Level, /jig/L
1.2
0.3
2.1
Level
Spiked,
ug/L
10.0
5.0
10.0
Average
Recovery,
V9/L
11.0
0.4
11.2
SD1
0.3
0.1
0.8
RSD.2
%
3
25
8
Average
Percent
Recovery
98
2
91
Background
Level, fjg/L
0.3
0.6
0.8
Level
Spiked,
fjg/L
100
50
100
Average
Recovery,
ug/L
98
61
107
RSD.
SD %
9 9
1.1 1.8
1.6 1.5
A verage
Percent
Recovery
98
121
106
High Level Spike Results
Level
Spiked,
W/L
100.0
100.0
100.0
Average
Recovery,
ng/L
99
9.9
105
RSD.
SD %
17 17
1.1 11
3.2 3
A verage
Percent
Recovery
99
9
104
^Standard Deviation.
2Percent Relative Standard Deviation.

-------
     S. V. Lucas, T. F. Cole, A. Riggin, and W. M. Cooke are with Battelle-Columbus
       Laboratories, Columbus. OH 43201-2693.
     James E. Longbottom is the EPA Project Officer (see below).
     The complete report, entitled "Evaluation of EPA Method 603 (Modified)," (Order
       No. PB85-146 629/AS; Cost: $11.50, subject to change) will be available only
       from:
            National Technical Information Service
            5285 Port Royal Road
            Springfield. VA 22161
            Telephone: 703-487-4650
     The EPA Project Officer can be contacted at:
            Environmental Monitoring and Support Laboratory
            U.S. Environmental Protection Agency
            Cincinnati, OH 45268
  * U.S. GOVERNMENT PRINTING OFFICE; 1985 — 569-016/7907
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
        OCOC329   FS

-------