Solvent Minimization i&A €tf£Atfdm£ritftSus Liquid/Liquid Extraction
           of Aqueous Samples for Semivolatile Organics

                  Joseph Slayton, Susan Warner,
        Philip Shreiner, Carole Tulip, and Edward Messer
               U.S.  Environmental Protection Agency
             Central Regional Laboratory, Region III
             839 Bestgate Road, Annapolis, Md. 21401
                           410-266-9180

                           Introduction

Continuous extraction (CE)  of aqueous samples is quickly replacing
separatory  funnel  extraction for  semivgiatile  organics.    The
advantages of continuous liquid/liquid extfdi&ion over separatory
funnel extractions include the following:

1) improved extraction efficiencies and accuracy due to the
   increased number of theoretical plates associated with the re-
   distilled solvent being continously exposed to the sample;
2) savings in manpower due to the reduction of both time and
   physical labor;
3) the effectiveness of the CE technique in highly contaminated
   matrices containing suspended  solids (a problem with Solid Phase
   Extractions);
4) the effective elimination of emulsions common with separatory
   funnel extractions of environmental samples; and the
5) improved precision using CE.

One  disadvantage   of  the  traditional  CE   procedure   is  the
considerable volumes  of  solvent  that is required  to perform the
analysis versus  the separatory  funnel  method.  A  commonly used
"macro-sized" extractor  is  illustrated in Figure 1.  The continuous
extraction  technique frequently requires  600  to 1000  mLs  of
methylene chloride solvent  to perform a single extraction.  Compare
this to the 180 to 360 mLs to perform a routine separatory funnel
extraction. Given the overall expense of using methylene chloride,
both the initial purchase  cost and  the  extremely costly disposal
fee  ($200+  per  55  gallon  drum),   it  would be  desirable  to
miniaturize the procedure  in order  to minimize  the  volume  of
solvent.  Miniaturization was considered more desirable than Solid
Phase  Extraction  (SPE)  or  other  technologies,  which  involve
different chemistries than  liquid/liquid extraction, since the use
of these  techniques  for EPA's programs would require  obtaining
analytical  "variances".   Such variances  may take many years  to
obtain.
This  work  was  funded  by  the  USEPA  Office  of  Research  and
Development.   The authors  recognize  the contribution  of  Angela
Cogswell, NNEMS  Fellowship  student,  on the  preliminary  studies
associated with this work.

-------
A design  for a miniaturized  continuous  extractor (Figure 2) was
developed so as to maintain the  sensitivity of the procedure, yet
minimize the  solvent necessary  to perform the  analysis.   A full
liter of sample was extracted, as per the current  Agency protocols
(SW-846, EPA NPDES Methods 625 and 608, SDWA Method 508, Superfund
CLP Statement of Work) to  assure sensitivity and  to help assure a
sample aliquot of sufficient size to  be accurately representative.
This sample volume also avoids the necessity for concentration of
the extract obtained to a  smaller final volume to maintain
sensitivity,  e.g.,  less than 0.5 mL.   It was decided  to avoid
attempts to  reduce the final extract volume to  less  than 1 mL,
since the extract could easily  go to dryness.   Going to dryness
would result in the loss of the  more  volatile compounds.
                       '(
A series of  extraction recovery experiments were performed using
the prototype extractor design to determine the:

    *    Necessity  for design  modifications  and/or  extraction
protocols necessary to maximize target compound recoveries with the
goal being to obtain the performance specifications (% recovery and
standard deviation),  required by current Agency protocols  (EPA
Methods 625 and 608).
    *  Effect upon the analytical results (accuracy and precision) .
Recovery of semivolatile  organics, pesticides and PCBs listed as
target compounds under the Superfund  Contract Laboratory Program ,
EPA methods  608  and  625  (NPDES) and 508 (SDWA)  were determined.
This  work  followed the   "initial demonstration  of  capability"
procedures specified per the 600  series methods.   These procedures
test  the  performance  of   the  method  (all  steps  of the  method)
against specified accuracy and  precision criteria  specified for
each  target  compound.    In  addition,  the performance  of  the
miniature extractors (employing 200 mLs of solvent—Figure 2)  were
compared to  "macro" size  extractors  (employing  700-1000  mLs  of
solvent—Figure 1).  These larger "macro" CEs have been routinely
used by our  laboratory for the  analysis  of  semivolatile  organic
compounds since 1986.
    *  Ease and practicality of  use.
    *    Consistency   with  the   Agency's  mandatory  analytical
procedures.    As  part of  this  work,  it was determined  whether
special variances by the  Agency are necessary  for use of these
protocols in the NPDES, SDWA,  Superfund and RCRA programs.
    *  Effectiveness of the extractor in recovering compounds from
wastewater samples.

                           .Disclaimer

Although the research described in this document has been supported
by the U.S.  Environmental  Protection  Agency and is awaiting Agency
wide review,  it  does  not necessarily reflect  the views  of  the
Agency, and no  official   endorsement should  be  inferred.    The
mention of trade names  or commercial products in this report is for
illustrational  purposes  and  does not constitute endorsement  or
recommendation by the U.S. Environmental Protection Agency.

-------
                         I.  Experimental

A. Reagents  & Equipment

Note:   Brand  names   and  catalog  numbers   are  included   for
illustrational purposes  only.

1   Methylene chloride,  B&J high purity  solvent,product #300,
    contains cyclohexene preservative to inhibit  HC1 formation.
2.  Sulfuric acid, Baker,  Instra-Analyzed, #9673-03,
    6N H2S04 prepared  by slowly adding 167 mLs of
    concentrated H2SO4 to 833 mLs of reagent water.
3.  Multi-range pH paper strips, EM-Reagents ColorpHast,
    pH indicator strips,  pH 0-14.
4.  Boiling  Stones, Hengar Co., carborundum #12 granules,
    #133-B.  Conditioned by muffling at  450°C for 3-4 hrs.
5.  Sodium sulfate, anhydrous, granular, Mallinckrodt, product
    #8024 .  Muffled for 3-4 hours at 450°C.   Stored in glass.
6.  Glass wool, Pyrex  brand, fiber glass, sliver  8 micron,
    Corning  Glass Works.  Muffled for 3-4 hours at 450°C.
7.  Muffle furnace, Blue M Power-0-Matic 80.
8.  Muffle furnace, Blue M Touch Master, Model #CFD-20F-6.
9.  Heating  mantle, Glas-Col Apparatus Co.,Terre  Haulte,  In 47802
    Cat. No. TM98, (80 Watt, 115 V).
10.  Variable transformer, Staco Energy Products Co., type  3PN1010.
11.  125 mL  boiling flask.
12.  1000 mL graduated cylinder.
13.  Allihn condenser,  45/50 joint,  4 ball, with special drip ring
     to catch condensation from room humidity.
14.  3-ball  Snyder columns, (macro- and  micro-).
15.  500 mL  Kuderna-Danish evaporative flask.
16.  10 mL graduated Kuderna-Danish concentrator  tube.
17.  Continuous extractor, one piece, glass, obtained from LAB
     Glass,  Inc., Vineland, Pa.  (Figure #2—"Micro-" and Figure #1
     —"Macro-").  The "macro-" size extractors are routinely
     employed in environmental laboratories.  The miniature
     ("micro-") extractors cost about $100/each.
18.  Drummond pipet, 100  uL dispensing pipettor,  Model #375, used
     for pipetting spikes.
19.  Volumetric pipet, 1 mL.
20.  Volumetric flasks,  1 mL, 2 mL and 10 mL.
21.  Milli-R015 Millipore (10 megaohm-cm, deionized water) System.
22.  Carbon  filter system  (made internally, 5 Ibs. activated
     charcoal)  for final polishing of lab pure water.
23.  Screw cap vials with teflon-faced silicone septa, 1.8 mL,
     Cat. Nos.  3-3286  (vials) and 3-3210 (caps and septa),
     Supelco, Beliefonte, Pa.
24.  Pyrex funnels (for sample addition), 60°, 145 mm stem length.
25.  Pyrex stirring rods  (for sample pH adjustment), 370 mm length,
     and 15  mm diameter.
26.  Finnigan MAT 4500 GC/MS.  The system equipped with:  a

-------
 N21344
                                                 Figure  I
                           IOOMM _
                                O.D.
                 350MM
                                                            I35MM
                                                            22MM
                                                               O.D,
                                                                                             524/40
                                                                                          'INNER Ji
   THESE DRAWINGS AND SPECIFICATIONS
   ARE THE PROPERTY OF LABGLASS. INC AND
   SHALL NOT BE REPRODUCED OR COPIED
   OR USED AS THE BASIS FOR THE MANU
   FACTURE  OR SALE OF APPARATUS OR
   DEVICES WITHOUT WRITTEN PERMISSION
   OF LABGLASS. INC. VINELAND. N 1.
CUSTOMER
                                 CATALOGUE
                                                      DESCRIPTION
                  RAW MATERIAL COMPONENTS
QTY.
    PART NO
                             QTY I PART NO
                                                                                           DATE   -
                                                                                           SCALE
                                                          APR.
                                                          PRINTS
 LABGLASS INC.
NORTHWEST BLVD & OAK
   VINELAND. N.J.

-------
                                               Figure 2
                                                              S 45/50
                             80MM
                                 O.D,

                      irw     /
                      G"0/A
                                                                  -80MM—H
                                                                                       22MM
                                                                                         "O.D.
                                                                                      4MM TEFLON
                                                                                      STOPCOCK

                                                                                    \   S 24/40
                                                                                      INNER JT.
       DRAWINGS AND SPECIFICATIONS
    (U THE PROPERTY OF LA8CLASS. INC AND
    HALL NOT BE REPRODUCED OR COPIED
  OR USED AS THE BASIS FOR THE MANU
  FACTURE OR SALE OF APPARATUS OR
  DEVICES WITHOUT WRITTEN PERMISSION
  OF LABGLASS. INC VtNELANO. N J.
CUSTOMER
                               CATALOGUE
                                                   DESCRIPTION
                 RAW. MATERIAL COMPONENTS
QTY.|PART NO
QTY I PART NO
                                                       DATE
                                                                                                -- 9?
                                                                                     SCALE
                                                       APR.
                                                       PRINTS
                            LABGLASS INC.
                           NORTHWEST BLVO & OAK
                              VINELAND. N J.

-------
     quadrupole analyzer  and El  source;  an HP 7673 automatic
     sampler and an Incos data system.  The Fused Silica  Capillary
     Column  (FSCC) was  a  DB-5, J&W Scientific, 30M x 0.32mm
     ID with a film thickness of 1 urn.  The GC temperature program
     was: 30°C for 2 minutes, ramped to 300°C at  10°C/minute.
27.  HP 5890  Series II Gas Chromatograph/BCD/FID system, with an HP
     7673 automatic sampler  and  HP 3365  Chemstation data system.
     The ECD was equipped with a Supelco SPB-608, FSCC, 30M  x 0.53
     mm ID, with a film thickness  of  0.50 urn.  The GC program for
     the pesticides,  (ECD) was from 150°C to 280°C at  10°C/minute,
     with a final  hold  of 10 minutes.
     The FID was equipped with a Supelco #2-4050, SPB-5 FSCC,  60M,
     0.32 mm ID, with a 0.25 um film thickness.   For FID  analyses,
     the GC was programmed from 50°C to a 280°C at 5°C/minute,  with
     a final hold  of 10-25 minutes (compound dependent).
28.  S-EVAP, solvent recovery system  (during K-D process),
     Organomation, Inc., South Berlin, MA.
29.  Re-circulating water bath (condenser cooling), FTS  Inc., model
     RC-25.
30.  Methanol, B&J, purity suitable for Purge &  Trap analysis.

B. Calibration standards and Spiking  Solutions/Procedures

The calibration and spiking  solutions used were  all from EPA's QA
Materials Bank  in RTP, NC   or from certified  CRADA vendors.   A
detailed listing of the sources  and preparation procedures for the
following solutions is included in the Appendix:  spiking solutions
(general);   calibration  standards   and  multiple  point  curve
preparation;  internal  standards;  Superfund  CLP "Matrix Spikes"
(MS); Superfund CLP  "Surrogate Compound" spikes; BNA  spiking QC
solutions  (CRADA); benzidine/s  and aniline/s spiking  solutions;
Pesticide   spikes   (single  component   analytes,    toxaphene,
chlordane,and PCBs).


C. General Procedures;

GC Screening;

The initial phases of this work involved numerous  re-designs of the
dimensions of the miniaturized  CE.   As a consequence,  a simple
spiking  mixture  was  used (matrix spike and/or surrogate  spike
delineated above).  This reduced the expense of  using  more costly
spiking cocktails  and provided relatively simple mixtures  which
could be analyzed via GC/FID.  Once the design was  optimized, the
more complex mixtures  were  tested using a GC/MS  system.   This
tiered approach saved much expense  (reference materials and costly
GC/MS use), and helped  speed the progress.

Loading the Continuous  Extractors;

   Miniature or "Micro-" CE  Extractors;

In a  fume hood the  stopcock on the  CE solvent return  line  was

-------
closed  (see figure  2),  a 125  mL flat  bottom  boiling flask was
attached  (containing several boiling stones).  200 mLs of methylene
chloride  were placed  into the  continuous  extractor.  A 500 mL
volume  of the 1  liter sample was added using a glass  funnel with
145 mm  glass  stem.   This procedure helped assure  that the aqueous
sample would not displace the solvent (avoid water break-through of
the solvent layer).  The remaining volume of sample was then poured
into the  extractor.  The dense solvent was thus layered below one
liter of  aqueous sample and therefore exposure of the analyst to
solvent vapor was minimized.   The extractors were  secured in a
ringstand with the solvent flask placed in a heating  mantle.  Each
one liter aliquot was adjusted to a pH <2 using 6N  H2S04, (5mL auto-
dispenser) , except for spikes  of anilines  and benzidines,  which
were  extracted  at  pH  >11  (6N  NaOH pH  adjustment).  Also the
pesticides/PCB extractions were  performed  at pH  5.5-6.5 (no pH
adjustment  necessary).   The samples were stirred using a 370 mm
glass rod and a drop  of  the  solution  was tested  with  pH test
strips.  The samples were then spiked with 100 uL-1 mL aliquots of
the appropriate  stock solutions.

An  Allihn  condenser was  attached.   An FTS  refrigerated  water
recirculator was used to cool the condensers.  The temperature of
each condenser was 5°C.  The  stopcock was opened and  approximately
50 mL of methylene chloride siphoned over into the boiling flask.

The heating mantles were turned on after the condensers were cold
to  the  touch  and extraction  continued  for  24+  2  hours.    All
extractions were generally carried out in the dark (no sunlight and
a minimum of  exposure  to  fluorescent lighting)  to  avoid photo-
decomposition of light-sensitive  compounds.

The extracts  obtained  were  concentrated via  the Kuderna-Danish
procedure  specified  by EPA NPDES method 625   (macro followed by
micro  K-D/Snyder  columns).    However,  a  condenser  device  was
employed  during  the macro  K-D  step  to  assure prevention  of
emissions  (S-EVAP from  Organomation Assoc.,Inc., South  Berlin,
MA) .    Also,  less Na2S04 was employed for  extract drying (20-30
grams was used).

   J'Macro-" CE Extractors  (figure /I);

The operation of the these  larger  extractors was very similar.
However,  600-1000 mL  of  solvent was  used and  loading  of  the
methylene  chloride was  without  the benefit of a  stopcock in the
solvent return-line.   The flow rate through the extractors  was
approximately 6  mLs/minute (rheostat set at 60% full scale).

Instrumental Analysis:

A tiered  approach  of analysis was employed  when working with the
design  configuration of the  CE.  To help resolve and correct CE
design  problems, simple spiking solutions were  used  to verify
performance and  analyses were performed on the GC/FID.
Once the design was finalized all analyses were performed by GC/MS,

-------
except for toxaphene and PCBs, which were via GC/ECD.

GC/MS analysis was  performed (GC/MS via 70 eV electron impact as
per  EPA method  625  for the  reported recoveries  of  the target
compounds, except for Toxaphene  and PCBs, for  which  GC/ECD was
employed.

Calculations; (The details of calibration solution preparation are
              included in the Appendix).

A reference solution (same volume  of material that was added as the
spike), was  prepared in a  volumetric  flask (same  volume as the
final volume for the K-D process).

The reference solution was analyzed and concentration was verified
versus a 100, 50, 20 and  10  ng calibration standard curve prepared
from AccuStandard stock solutions.  The guantitation was based on
internal standard (response  factor) calculations as per method 625.

The percent recovery of spiked material was determined as follows:

        % Recovery = ng measured  in extract  X  100
                        ng measured in reference

This technigue took advantage of the improved precision associated
with the internal standard technigue.

The exception to this reference solution approach was the analyses
of  the   "BNA  Spiking  Solution,"  (48   priority   pollutants).
Quantitation  was  performed  as  based  on  the   AccuStandard
"Calibration Standards (GC/MS)".   This  guantitation procedure for
the BNAs  was the same for  the "macro-" and "micro-"  continuous
extractors.   As  resultant   extracts were  ideally  100  ng/uL,  %
recovery was the  same  as the ng/uL measured  from the calibration
curve.

D.General Quality Control:

     a.  All  glassware including CEs were solvent rinsed, soap and
         tap water cleaned,  deionized water rinsed and heated in a
         high temperature oven (400-450°C)  for  6-8 hours prior to
         use.

     b.  All surrogate, matrix spike and priority pollutant
         standards, and spiking materials were certified materials
         (CRADA)  or were obtained directly  from the  USEPA Quality
         Assurance Materials Bank in RTF, NC.

     c.  The  GC/MS mass assignments were calibrated with FC43 prior
         to analyses.

     d.  The  GC/MS relative mass abundances were tuned by obtaining
         the spectrum of DFTPP.  This was stressed for the
         identification of non-target compounds in the analyses of

-------
         wastewater.

     e.  Immediately before GC/MS analysis,  each sample was spiked
         with an internal  standard mixture.  GC/FID/ECD used the
         external standard quantitation technique.  Each batch of
         samples analyzed  by GC/MS and GC/FID/ECD included
         calibration check standards, analyzed throughout each
         analytical run.

     f.  The sensitivity of the GC/MS instrument to 40 ng of
         dlO-phenanthrene  was at least 50,000 area counts.  The
         sensitivity of the GC/FID/ECD was confirmed by GC/MS
         analyses.

     g.  All surrogate and matrix spike recovery limits
         referenced in this study were from the Super£und
         Contract Laboratory Program  (CLP) protocols
         The recovery limits for the BNAs and pesticides were
         as per EPA Methods  625 and 608.

     h.  Compound spike recoveries were computed against a
         the response to a  reference standard prepared the same day
         the samples were  extracted. Reference standards and
         samples were analyzed on the same day and on the same
         GC/MS or GC/ECD.  The exception to this procedure was for
         the analyses of BNA spikes (48 compounds),  in which a
         freshly prepared  multiple point calibration curve was
         employed to determine the concentration of the analytes
         and % recovery was calculated vs. the certified values
         for the spiked QC materials.

     i.  Data quantitation was performed by automated
         procedures using  Incos software (GC/MS Finnigan MAT, San
         Jose, California) and HP Chemstation software (GC/FID,
         GC/ECD Hewlett Packard, Palo Alto,  California).

     j.  All compound identifications via GC/MS were made by
         comparing known reference spectra to those of the
         unknowns.  GC/FID identifications were based on retention
         time matches to reference material, with GC/MS
         confirmation.  GC/ECD identifications were based on
         retention time matches to reference materials.

     k.  All spikes into aqueous matrices were prepared in a
         hydrophilic solvent.

     1.  Precision and accuracy were routinely based on four (or
         more) replicate spikes carried through the entire
         analytical process.

              II.  Results and Discussion

A  major initial  challenge  of  this  work  was  to eliminate  the
carry-over of water into the solvent reservoir (boiling flask).  A

-------
combination  of design  changes  and procedural  changes  (loading
technique) has eliminated this  problem and the associated poor
recoveries of hydrophilic compounds  (erratic and low recoveries).

Miniature  Continuous Extractor  Design  (Height  of the  S-shaped
Solvent Return Line):

Initial trials were performed on continuous extractors  (.prototype
that used approximately 100 mLs of methylene chloride) by analyzing
laboratory pure  water  fortified with  the  CLP matrix  spike and
surrogate  compounds.   This  afforded  a relatively  inexpensive
mixture which resulted  in simple chromatographic runs which could
employ  GC/FID analyses.   A  number of difficulties with these
preliminary designs were encountered.  One critical parameter was
the height of the "S"  shaped solvent  return-line.   The distance
from  the  base of the  extractor to the  top of  the  return-line
directly  determined  the depth of  the solvent below  the aqueous
sample.  At the initial height of 175 mm,  water routinely "broke
through" the solvent reservoir during the extraction. In addition,
this,  design  included a return glass tubing line with  an inside
diameter of 4 mm and  a 2 mm Teflon stopcock.   It was found that any
water  droplets in the  solvent  return-line  stopped  the  flow  of
solvent (surface tension).  The height of the "S"  tube was adjusted
to 185 mm and the  inside diameter of the tubing was adjusted to 10
mm.  A 4 mm Teflon stopcock was bored out to interface with the 10
mm tubing.   These adjustments  largely avoided  water  carryover.
Matrix spike  and  surrogate results  were much improved,  but water
(aqueous  sample)  would periodically  break  through the  solvent
reservoir  (base of the continuous  extractor).   As indicated  in
Table #1,  the recovery of surrogates were acceptable but were more
erratic and  lower when water was  observed  in the extract.   The
greatest reduction in recovery was  associated with the phenolic
compounds (2-fluorophenol and d5-phenol) , which have great affinity
for water  and could be easily lost during the  drying  step  with
sodium sulfate (significant quantities of water  causes  the Na2S04
to  form  lumps that  could   entrap the  associated  hydrophilic
compounds).   Matrix  spike recoveries  were  similarly  acceptable.
Only trace amounts (a few drops)  of water were observable in the
extracts associated with Table 2. With a height  of 185 mm for the
"S"  line.  150  mL  of  solvent  was  necessary  for  continuous
extraction.

In previous  work  with  the "macro" CEs  (used  routinely by  our
laboratory since 1986),  little concern  had been given to losses of
methylene chloride  during  the extraction  period.   However,  re-
circulating chillers were routinely  employed (Neslab-Coolflow 75) .
During  the  work   with miniature  or   "micro"  CEs,   slight  but
repetitive losses of solvent were  visually apparent.   For  this
reason, volumes of methylene chloride were measured before  and
after the extraction  period (24 hours + 2 hours) .  An FTS model RC-
25 recirculation chiller was  employed and set at  5°C.   The chiller
was  charged   with laboratory pure  water  mixed with  1:1  (v/v)
ethylene glycol.   Two hundred  mLs of methylene chloride and l  L of
water  were loaded  and the  routine  extraction  procedures  were
                              10

-------
followed.
measured:
After the extractions  the distribution  of solvent was
       LOSSES OF METHYLENE  CHLORIDE DURING CE EXTRACTION
      ("MICRO" CE, WITH CHILLER AT 5°Cf 24 HOUR EXTRACTION)

      CEs were initially charged with 200 mL of CH2CL2

 DISTRIBUTION OP METHYLENE  CHLORIDE (mL) FOLLOWING EXTRACTION
EXTRACTOR
**
1
2
3
4
AVE.
CE
RESERVOIR
135
105
135.
140
128
SIDE
FLASK
43
75
48
41
52
TOTAL
178
180
183
181
181
LOSS
VS 200 mL
21
20
17
19
19
% LOSS
10.5
10
8.5
9.5
9.6
  ** Four separate extractor units were tested

Essentially  10%  of the  solvent was lost  during  the extraction.
Since  methylene  chloride  is  about 2%  soluble  in water  and the
extractors were charged with 1000 mL of water, 20 mL of methylene
chloride could be dissolved in the aqueous sample and could account
for the loss in solvent.

It was determined that a number of variables associated with the
miniature  extractors  could affect  water  breakthrough.    These
included:  how vertical the extractors were  placed  in the ring-
stand;  the height of the "S" return line;   and the possible losses
of solvent during the  extraction (10%  on  average).   To allow for
such variables and  to reliably  avoid  water  breakthrough,  it was
decided  to extend  the height  of the  "S" line to  195rom.   The
miniature  CEs  with the 195 mm return line  required 200 mL of
methylene chloride.

Continuous Extractor Design (Overall Length) ;

An additional variable that proved important was the length of the
CE extractor  above the  aqueous sample.   During the extraction,
solvent  collected at  the drip  tip of the Allihn condenser and
dropped to the aqueous sample  surface.   In  the "Macro" extractors,
this "drop distance" was sufficient to have the solvent droplets
easily break the water surface tension  and  pass through the sample
as small droplets.   Such solvent droplets with large surface to
volume ratios were thought to provide the best extracting exposure
as they  "fall" through  the sample.   However,  the  miniature or.
"micro" extractor droplets were accumulating on the sample surface
until  large pooled  droplets of  solvent would "fall"  through the
sample.  The micro  extractor  design was altered by extending the
length of the CE  extractor 30 mm to 400 mm (figure 2) .
                               11

-------
Final Miniature  ("micro") CE design:

The  final  dimensions of the one piece, all glass extractor  (with
Teflon stopcock in the solvent return line) were as follows [Figure
2]:  overall height  400 mm; height of  solvent return  line 195 mm;
overall diameter 80  mm; inside diameter of the return line tubing
10 mm; and 4 mm Teflon stopcock bored  to  interface with the  10 mm
tubing.  The simple  one piece units were  easy to charge with both
solvent and  sample and to mount  in the ring stands.   since they
were all glass (except for the stopcock), they were easy  to  clean
and the entire extractor could be placed in a high temperature oven
(425-450°C) as the final step in the cleaning procedure.

Loading the Extractor:

A combination of closing the stopcock, followed by adding all 200
mLs  of the  extracting solvent  to the  CE  unit,  plus the  slow
addition of the aqueous sample (a funnel  used for about the  first
half of the sample volume) has minimized water carryover  into the
boiling  flask   (described   in   detail  in  Section  c   (General
Procedures) .  This loading technique also minimized the exposure of
the analyst to the organic solvent.

Determination of solvent flow rate;

The flow rate of solvent was set at 7.5 mLs per minute.   This was
adjusted by varying the rheostat settings  (80% of full  scale).  The
flow rate was measured by marking the  level of the solvent in the
reservoir  at  equilibrium  (at a given  rheostat  setting)  and  then
closing the stopcock for a measured time  interval and marking the
level of the solvent.  The volume  of solvent distilled over during
the  measured  interval was  determined  by  filling  the  emptied
extractor with solvent (between the two marked levels).  Without a
stopcock it would be difficult to  determine the solvent flow rate.

Sodium Sulfate and possible alternatives;

The  final design of  the  miniature   continuous  extractors  had
minimized solvent use (about a five fold reduction).  However,  once
the extraction was completed, the extract was  dried  (water removed)
by passing it through sodium sulfate.  The  drying columns specified
by EPA Method 625 are  19 mm ID and long enough to allow 100 mm of
sodium sulfate.  The method  specifies  that  20-30  mL  of methylene
chloride be used for rinsing the flask  and the drying column after
passing the  extract  through the  column.   Large volumes of rinse
solvent  (greater  than 20-30  mLs)  will  negate  the  effort  at
minimizing  the  solvent  used  during the   extraction.     Very
preliminary results  are  listed  in the Appendix for one  possible
alternative to the use of sodium sulfate dryirig  columns, namely
hydrophobia filters. A sample of prototype filters was provided by
Varian Corporation  (Sample Preparation Division).  100 ng of the
BNA target compounds were placed  in 60 mLs  of methylene  chloride
(about  the volume of  the  extract  resulting  for miniature  CE
extractions).  The filters were rinsed with  5 mLs  of  solvent and
                               12

-------
the extracts were KD concentrated to 1 mL.  The recoveries for the
target compounds are listed in the Appendix.  The use of these type
of filters for drying extracts with small amounts of water (2-4 %)
would avoid  the necessity for sodium  sulfate  columns and should
require far  less  solvent volume  for rinsing.  Similarly, syringe
"micro- sodium sulfate columns" are now available, which may prove
effective.

"Demonstration of Capability"  (miniature CEs):

Since the  final  CE  design  (figure  #2),  an extensive series of
spiking experiments were performed into laboratory pure water.  The
experimental design (four replicates, at specified concentrations,
and performing all aspects of the analytical method)  is that listed
in EPA's organic protocols, e.g.,  625,  608, 508.  This is referred
to in these methods as the "initial demonstration of capability".
The corresponding  analytical methods list  specifications for the
accuracy (% recovery) and precision (standard deviation) , which are
to be obtained.

The spiking solutions and procedures for CE  as well as quantitation
have been described previously.

   CLP Surrogate Compounds

The results for the replicate spikes of the CLP surrogates (n=5),
were excellent (Table #3) .  All six  compounds were well within the
specified recovery criteria.

   Priority Pollutant ("BNA  Spiking Solution")

Forty-eight BNA  compounds (included in the  Supelco CRADA mixes)
were spiked into laboratory  pure water in four separate miniature
and macro  CEs.   Figure  #3  is a chromatogram  (Reconstructed  Ion
Current Profile)  resulting from the GC/MS analysis of a "BNA" spike
extract. The resulting recoveries for both  the miniature and macro
extractors were all within the acceptance  limits as specified by
EPA method 625  (Table  4  and 5) .   All but ten compounds  were
recovered  by  the miniature  extractor  in excess of 90%  (average
recovery).   The troublesome  compounds included:

                               Average Recovery (Std.  Dev.)
                                            with n=4
Compound
Phenol
1 , 3-dichlorobenzene
1 , 2-dichlorobenzene
1 , 4-dichlorobenzene
N-Nitroso-di-n-propylamine
hexachloroethane
1,2, 4-trichlorobenzene
1,1,2,3,4, 4-hexachloro-l , 3-
butadiene
"Miniature"
CE
89.
79.
82.
83.
85.
75.
85.

74.
4
3
9
8
3
7
7

2
(1.
(1.
(2.
(2.
(0.
(1.
(1.

(1.
3)
6)
2)
6)
5)
6)
6)

8)
"Macro"
CE
90.
71.
75.
71.
86.
67.
78.

67.
2
4
7
6
9
5
7

2
(4
(6
(6
(3
(6
(2
(3

(2
•9)
.5)
.4)
•1)
.1)
•7)
•3)

.6)
                                13

-------
                               Average  Recovery  (Std. Dev.)
                                         with n=4
Compound
2 , 4-dinitrophenol
2-methyl-4 , 6-dinitrophenol
"Miniature"
CE
73.6 (2.0)
80.3 (2.3)
"Macro"
CE
66.1 (4.5)
80.8 (3.2)
The recoveries obtained for these difficult compounds
were  in close agreement between the  macro and micro extractors,
indicating that the recoveries of these compounds were not related
to the volume of solvent used for extraction.

Figure #4 depicts the BNA spike recoveries (Tables  #4 and #5) .  As
indicated,  the agreement  in  recoveries  was  good with  the most
notable  exceptions  occurring when  the  "macro-"  CE  recoveries
exceeded  100%.   In these cases,  the  miniature  extractors gave
recoveries closer to 100%.  The five  fold reduction in solvent had
no adverse affect on the recoveries of  these target compounds.
NOTE:  These extractions were performed at pH<2.  It was found that
extraction under  acidic conditions greatly  improved recovery of
short-chain phthalate esters  .   Also,  floe and emulsion formation
were minimized, even for the continuous extractor method, which is
far  less  prone  to  these  difficulties  than separatory  funnel
methods.  All compounds,  except the most basic ones such as aniline
or benzidine, were effectively extracted at a pH <2.
Spikes of Benzidines and Anilines

Basic semivolatile compounds (substituted benzidines and anilines)
extracted effectively under basic conditions (pH >11), giving spike
recoveries near 100% using  the miniature extractors (Table  6) .  One
problem encountered involved the recovery of aniline.  Though this
compound is not a Priority  Pollutant (not an NPDES  analyte) and has
been dropped  from the Superfund CLP  target  list, it was in the
spiking mixture and the difficulties encountered were of interest.
The recoveries  obtained for aniline were  routinely in  excess of
150%  and were  thought  related to  a  adverse "solvent  effect"
(presence of  methanol in  the reference  material  suppressing the
response of this polar compound.  This topic is discussed in more
detail in a later  section  on recoveries of phenolic compounds in
the.CLP "Matrix Spikes".

      Single Component  (Priority Pollutant) Pesticides

Single component pesticides (EPA Method 625,  608 and  508)  gave
average spike  recoveries  over 90%,  with the exception  of endrin
(80%).  As indicated in Table 7, all of the accuracy and precision
requirements specified for EPA Method  608 were met (these criteria
were used since they are more demanding) , though GC/MS was used as
the detector for these analyses).
                                14

-------
      Chlordane

The analyses of  50 ug/L spikes (50 ng/uL in the extracts) proved
challenging, but not beyond the quantitation range of the GC/MS.
The chromatographic peaks selected for analysis are presented in
Figure  5 and  the  mass spectra  associated with  the heptachlor
component of technical chlordane is presented  in  Figure 6.   The
recoveries averaged 108%  with a standard deviation of 2.7 (Table
8) .  The range  of recoveries specified by EPA Method 608 were 55.2-
109% (limits more restrictive than those in 625).

      Toxaphene

The analyses  of  50 ug/L spikes  (50  ng/uL in  the extracts)  was
performed by GC/ECD.   The recovery data ranged from 100-112% for.
the replicate  spikes  (n=5)  and are presented  in  Table  9.   The
accuracy and precision requirements specified for EPA 608 were met.

      PCB-1242 and 1260

The average recoveries for spikes of Aroclor 1242 and 1260 (Tables
10 and 11) were 100 and 97.6 % respectively.  The accuracy  (average
% recovery) specified for these PCBs  in EPA 608 are (24.8-69.6%)
and (18.7-54.9%),  respectively.

Figure  #7 is the  chromatogram associated with  one of the spiked
extracts (PCB-1260).

WasteWater Sample

Four  liters of  waste from  a  secondary  effluent at   a  local
wastewater treatment plant was extracted  in replicate employing
both the "mini-"  and  "macro" extractors.  These analyses included
Superfund Surrogate Spikes (Table 12) .   All the  recoveries for the
surrogates  were  within  CLP  specified  limits.   The  compounds
recovered from this sample are presented in Table 13.  The sample
was selected because  it  contained significant  suspended solids.
The resultant CE extracts ("mini-11 and  "macro") were orange/brown.
The results indicate that the environmental matrix was successfully
extracted using  the miniature extractors,  as no apparent bias is
indicated.

S-EVAP

This work was conducted without solvent emissions.   A device (the
S-EVAP) was used  during the K-D concentration procedure to assure
that methylene chloride  was  not emitted.   The S-EVAP  employed
special  Hopkins  condensers  during  the   KD   (macro)   solvent
concentration step prior to GC/MS or GC/FID analyses.  The use of
this condensing  instrumentation  for  the  recovery  of  methylene
chloride  and  hexane  had been  reported as  effective  and of  no
significant effect upon the recovery of semi-volatile compounds .
This  current  work,  "validation" of  the  miniature  continuous
extractors,  has provided additional data supporting the use of this
                                15

-------
pollution prevention device.

Methanol Solvent Effect;

The  "solvent  effect"  of  methanol  upon  acidic  semi-volatile
compounds became an inadvertent area of study when spikes of acidic
compounds (prepared in methanol) into deionized water were compared
to  spikes  made  directly  into methylene  chloride.    The  spiking
solution contained methanol.  The  adverse affects  were  compound
specific with the greatest effect associated with compounds with an
affinity for  methanol.   The methanol apparently suppressed the
response  of  H-bonding  compounds  (acidic  compounds  such  as
pentachlorophenol)   in the  analysis of  the  reference standard.
Since methanol does not extract from water using the  CE procedures,
the test CE extracts were free of  methanol and the associated
suppression.    This  effect  was   related to   several  different
capillary columns  and chromatographic  conditions (Appendix). The
higher the initial  GC temperature,  the  less pronounced the solvent
effect.   Similarly the  thinner the  stationary phase, the  less
"effect" was measured.  A tight narrow band of methanol condensed
after  injection would  be  associated with  the   lower  initial
temperatures  (30C)   and the  thicker column phase.   As the initial
temperature increased, the methanol would be expected to occupy a
wider band, and the "effect" would be reduced.


                        III.  Conclusions
The miniature continuous extractors:

*   Effectively extracted  the target semivolatile  organics  (EPA
Method 625, Superfund Contract Laboratory SOW) with recoveries
(accuracy)  and precision  within the performance  specifications
specified in EPA Methods 625 and 608  (in laboratory pure water and
the tested wastewater).

*   Required less  methylene chloride per  sample  extracted,  which
should result in significant savings in solvent costs.  A 3- to 5-
fold  reduction of  methylene  chloride required for  continuous
liquid/liquid  extraction  was obtained.    The use of this  device
represents a laboratory pollution prevention measure.

*    Required  significantly less  time to  K-D  concentrate  the
resulting extracts,  since less volume- of extract was involved (50-
75  mLs  for the miniature  extractors  versus 200-300 raLs for  the
macro CE extractors).

*    Significantly  reduced  the  volume  of  waste solvent  to  be
recycled.  This  will  save  on  the  expense of  recycling  this
halogenated  solvent  (externally)  and/or  reduce the  number  of
distillation runs to  recover  the solvent  for direct reuse  by  the
laboratory.
                              16

-------
*  The extractors were relatively inexpensive ($100 from Lab Glass,
Vineland, New  Jersey).   The manufacturer has indicated that they
are less expensive to produce than the "macro" extractors because
of the smaller diameter of the CE extractor  (smaller stock tubing
is needed for  construction).

*  Were applicable to most of the Agency's methods for the analyses
of semivolatile  compounds.   The use of "miniaturized techniques"
(reduction in  scale, which preserve the basic chemistries) do not
require the  application  for a "variance" for use under  SDWA and
NPDES.  This is  an  extensive confirmation  that  the  analytical
results  are  consistent  with the  mandated  EPA  methods.    The
semivolatile techniques  for the RCRA and Superfund programs have
few specifications for the extractors.  Therefore the use of this
device should be applicable to most environmental  laboratories for
use in the determination  of semivolatile  organics  for the Agency's
programs.   One  notable exception  is  EPA's Method  608.   This
pesticide method for wastewater  does not  include the use  of a
continuous  extractor.   However,  EPA Method 625 (which uses  a
different detector, namely MS) includes the extraction of some of
the method 608 pesticides  using  CE.   The reality is that the 608
analytes extract well by  CE and  this extraction technique has been
used for years by the Superfund and  RCRA programs.  However, the
analysis via 608 must be followed for compliance under the NPDES
program (an issue of program compliance as opposed to an analytical
problem).

With  this exception,  employing  this apparatus  for the  NPDES,
Superfund and  RCRA programs would require only a demonstration by
the laboratory of analytical capability  ("initial demonstration of
capability" procedure as specified in EPA Methods SDWA, NPDES and
RCRA).   Though  this  "demonstration" is  not mandated under the
Superfund  program,  such  a  procedure   should  be part  of  the
laboratory's routine QC procedure.

In addition, as  many  of  these programs  are  now operated  by State
Authorities, prior  to use of this  device,  the  State  Authority
should be  consulted . (since under delegated  programs,  States are
able to be more restrictive than the Agency).

*  Were easy to setup (ring stands, etc.)  and load with sample and
solvent.

*  Retained the necessary analytical sensitivity,  since the initial
1 L sample volume was retained as specified by the Agency's organic
protocols and  the final extract' volumes were as  mandated.   The
miniature extractors resulted in no loss  of analytical sensitivity.
If the  sample  volume had been reduced,  the  final extract volume
would  have to be correspondingly  reduced.    This is  generally
undesirable  for  environmental  samples,  since  these  extracts
generally have significant  suspended/dissolved solids and/or foam.

*  Were safer  to load, because  they  use less solvent and because
the stopcock allowed the  solvent to be loaded and the sample placed
                                17

-------
on top.  The sample served as a barrier to escaping solvent fumes.

*  Unlike extracting devices which employ hydrophobia membranes to
help reduce  the  volume of solvent for extraction  (as  low as 100
mLs), the miniature extractors do not suffer reduced recoveries for
hydrophilic compounds, e.g., phenolics.

Additional Observations;

Several of the BNAs (1,4-,  1,3-,  1-2-dichlorobenzenes and 1,2,4-
trichlorobenzene), which are poorly recovered by CE actually should
be  dropped  from  the  Agency's  semivolatile  protocols.    These
compounds are too volatile and too hydrophobia to be sampled in 1
L amber  containers  (too  easily lost  to the headspace  of  these
containers) .  These compounds are already redundantly listed in the
Agency's volatile organics methods and their measurement should be
restricted to these protocols (required zero head-space septa vial
sample containers are mandatory for these methods).

Though the  time for  extraction in this study was routinely  24
hours, because of  the reduction in the solvent reservoir volume
with this design, the time required to complete extraction should
be reduced (less  time for the extracted compounds to be washed from
the solvent reservoir below the sample).

"Drip Lips" placed on the condensers used during CE extraction have
helped avoid contamination at the condenser/CE joint (45/50)  when
room  humidity condenses.    Also  this  has helped  avoid  water
(condensation) near electrical  equipment  (mantles,  rheostats),
which would be an obvious hazard for the analyst.

The use of condensers  during the  K-D  concentration  step (S-EVAP)
allowed the recovery of solvent, which would otherwise be vented up
the fume hood.
                      IV.  References

1.  U.S. EPA  Contract  Laboratory  Program,  "Statement of Work for
Organic  Analysis,  Multi-media,  Multi-Concentration",   OLM01.0,
12/1990.

2.  "Guidelines  Establishing Test  Procedures for the Analysis  of
Pollutants Under the Clean Water  Act;  40  CFR Part 136,  Federal
Register,  October 8, 1991.

3.  Slayton, J.,  Molnar, J. and Alvero, M.,  "Recovery  of Solvents
Utilized  in  EPA  Methods for  Extractable   Organics",   Pittsburgh
Conference, March 1992.

4.  Slayton, J.  and Trovato, R., "Acid-Neutral  Continuous Liquid-
Liquid  Extraction  of  EPA  Priority  Pollutants and   Hazardous
Substances List  Compounds",  28th Rocky  Mountain Conf.,  Aug.,  1986.

5.   Slayton, J.,  "EPA Case Study",  International Conference and
Exhibition on Pollution Prevention in the Laboratory,  June 1993.
                                18

-------
Data Tables and Additional
         Figures:
            19

-------
                                 TABLE  1

                      MINIATURE CONTINOUS  EXTRACTORS
                      185mm RETURN LINE

                      SURROGATE AQC X RECOVERY MICRO
SAMPLE    • 2-FLUORO-  05-        D5-NITRO-
           PHENOL     PHENOL     BENZENE
2-FLUORO-
l.l'-BI-
PHENYL
2,4.6-TRI- D14-TER-
BROMO-     PHENYL
PHENOL
                           CLP TARGET LIMITS

           (21-100)   (10-94)    (35-114)   (43-116)   (10-123)   (33-141)

NO WATER IN EXTRACT:
T2062403
T2062404
T2062405
T2062406
T629-01
T629-03
T629-04
T629-05
T629-06
AVERAGE
STO.DEV.
67.80
70.60
79.60
83.75
66.45
75.55
65.00
70.17
74.75
72.63
6.28
71.06
72.80
80.26
84.70
71.19
79.65
69.64
74.30
79.50
75.90
5.25
77.13
79.86
85.34
90.36
78.36
85.50
77.36
79.79
85.80
82.17
4.68
84.36
88.70
85.50
81.05
72.72
78.70
74.00
74.39
81.01
80.05
5.58
87.54
89.27
96.07
99.00
86.40
91.00
82.15
. 84.28
89.46
89.46
5.37
91.80
90.30
90.70
89.54
93.07
99.86
92.68
91.79
97.70
93.05
3.48
*************************************************************************
**********************************************************************
WATER OBSERVED IN EXTRACT:
T2062401
T2062402
T629-02
AVERAGE
STD.DEV.
56.46
41.80
49.70
49.32
7.34
57.46
40.90
53.15
50.50
8.59
67.30
48.47
57.77
57.85
9.42
85.40
51.06
50.50
62.32
19.99
71.18
57.37
62.30
-63.62
7.00
90.40
89.50
69.32
83.07
11.92
                                                  20

-------
                          TABLE 2

                          MINIATURE CONTINOUS EXTRACTORS
                          185nro RETURN LINE
                          GCXFID
EXPERIMENT:  -MATRIX SPIKE RECOVERY
GC RUN DATE:  19 Aug 92
REFERENCE: T2081701
REF. FILE #:  A081992\001F0101.D
COMPOUND NAME
Phenol
2-Chlorophenol
1 , 4-Di chl orobenzene
N-Ni troso-Di -n-Propyl ami ne
1,2, 4-Tr 1 chl orobenzene
4-Chl oro-3-Methyl phenol
Acenaphthene
4-Nitrophenol
2.4-Dinitrotoluene
Pentachl orophenol
Oi-N-Butylphthalate
Pyrene
SJ
T2081702
83.954
85.746
80.058
78.046
79.523
87.769
84.490
92.677
88.453
94.340
86.253
88.283
\MPLE NUMBER
T2081703
85.569
85.447
61.947
72.118
62.032
87.201
76.035
92.966
81.652
93.709
77.747
82.337
	
T2081704
82.658
82.834
72.310
80.156
72.652
88.104
83.893
95.578
91.285
93.473
87.073
90.322
T2081705
	
74.619
71.044
62.896
70.276
63.317
78.788
73.806
85.881
78.054
87.736
76.754
78.493
	
AVE
X REC
	
81.700
81.268
69.303
75.149
69.381
85.466
79.556
91.776
84.861
92.315
81.957
84.859
STO.OEV
4.024
5.464
4.338
3.738
4.352
3.648
3.814
3.550
4.865
2.410
4.073
4.295
                                                  21

-------
180.0-1
.00

 0)
 p

 00
 •1-1
 RIC
                                             DATA:  Al
                                             CALI:  CALi
            RIC
            11/12/92  14:14:00
            SAMPLE: FC43
            CONDS.: UPGRADE 30C 2 MIN TO 300 AT 10C/MIN
            RANGE: G    1/3675  LABEL: N  0, 4.0  QUAN: A
 #3
            SCANS  600 TO 3400
                                                       0,  1.0 J  0  BASE:  U 20,   3

                                                                                                                       225536.
                   1000
                   13:20
                                           1500
                                           20:00
2000
26:40
2500
33:20
  i
3000
40:00
SCAN
TIME

-------
                      TABLE 3
Surrogate Recovery (Miniature Continuous Liquid/Liquid Extraction)           CESUR.WK1
195mm RETURN LINE

50 ug/L Spike into Lab Pure Water (100 uL of 500 ug/mL of USEPA RTF Repository  Standard).

                                        % Recovery
                                                                    Average     Std.   CLP
Compound                Run 1    Run 2    Run 3     Run 4    Run 5  Recovery    Dev.   Limits
                                                                               (n-1)
2-Fluorophenol              87     84.7        91     86.3     88.5     87.5      2.39   21-100
D5-Phenol                 92.7     90.6      94.3     90.7     93.9     92.4      1.74   10-94
05-Nitrobenzene           95.6       93      94.4     93.6     96.4     94.6      1.40   35-114
2-Fluoro-l.r-biphenyl    95.8     93.7      99.7       97     99.5     97.1      2.54   43-116
2,4.6-Tribromophenol      101.6     96.6      96.7     92.2       97     96.8      3.33   10-123
014-p-Terphenyl             96       95      93.2     89.6     93.6     93.5    .  2.44   33-141
                                                   23

-------
                                TABLE 4
    Validation  of Miniature Continuous  Extractors BNA TARGETS(195 im S-tube)
                   Calibration Standard (ACCU Standard 100  ng  Z014A.B.O.E.GR.H) With Ultra Sc. Int.
                   Supelco CRADA QC SAMPLES

NO.  COMPOUND

  2  PHENOL *CCC*
  3  ETHANE,  1.1'-OXYBIS\2-CHLORO-
  4  2-CHLOROPHENOL
  5  1.3-DICHLOROBENZENE  (COELUTES)
  7  1.4-DICHLOROBENZENE  (COELUTES)  *CCC*
  8  1,2-DICHLOROBENZENE  (COELUTES)
 13  1-PROPANAMINE.  N-NITROSO-N-PROPYL-
 14  ETHANE.  HEXACHLORO-
 15  BENZENE,  NITRO-
 16  2-CYCLOHEXEN-l-ONE,  3,5,5-TRIMETHYL- *CCC*
 17  2-NITRO-PHENOL
 18  PHENOL,  2,4-DIMETHYL-
 20  ETHANE,  1,1'-[METHYLENEBIS(OXY)]BIS[2-CHLOR
 21  2,4-DICHLORO-PHENOL
 22  BENZENE,  1,2.4-TRICHLORO-
 24  NAPHTHALENE
 25  1,3-BUTAOIENE,  l.l^.S^^-HEXACHLORO-  *CCC
 26  PHENOL,  4-CHLORO-3-METHYL-
 29   2,4,6-TRlCHLOROPHENOL HSL
 31  2-CHLORONAPHTHALENE
 32  1.2-BENZENEDICARBOXYLIC ACID, OIMETHYLESTER
 33  2.6-DINITRO-TOLUENE
 36  ACENAPHTHYLENE,  1,2-DIHYDRO-  *CCC*
 37  2,4-DINITROPHENOL *SPCC*
 40  4-NITROPHENOL  (SEC ION)
 41  2,4-DINITRO-TOLUENE
 42  1,2-BENZENEOICARBOXYLIC ACID, DIETHYLESTER
 43  l-CHLORO-4-PHENOXY-BENZENE
 44  9H-FLUORENE
 45  2-METHYL-4.6-DINITROPHENOL
 47  BROMOPHENOXYBENZENE
 48  BENZENE,  HEXACHLORO- *CCC*
 49  PENTACHLOROPHENOL
 51  PHENANTHRENE
 52  ANTHRACENE
 53  1.2-BENZENEDICARBOXYLIC ACID, DIBUTYLESTER
 54  FLUORANTHENE CCC*
 55  PYRENE
 56  N-BUTYL  BENZYL  PHTHALATE
 57  BIS(2-ETHYLHEXYL)PHTHALATE
 59  BENZO/A/ANTHRACENE
 60  CHRYSENE
 61  DIOCTYLPHTHALATE *CCC*
 62  BENZO\B\FLUORANTHENE
 63  BENZO\K\FLUORANTHENE
 64  BENZO/A/PYRENE
 67  DIBENZO(A,H) ANTHRACENE
 68  BENZO\GHI\PERYLENE


Al
90.5
91.5
95.7
78.3
84.1
84.3
85.1
76.9
91.6
96.0
95.2
111.8
112.7
95.8
85.3
96.2
71.0
98.3
96.8
95.1
93.8
95.8
95.4
76.1
91.2
97.8
95.8
114.1
99.9
83.8
92.9
98.6
95.1
101.4
96.1
94.7
92.8
93.0
96.0
94.3
95.4
96.1
95.2
106.5
101.3
90.2
94.4
103.7


A2
89.7
91.0
94.1
79.6
79.8
79.5
85.7
73.4
93.7
96.1
98.0
99.7
113.5
96.7
84.5
92.3
74.8
101.7
94.9
90.6
90.0
95.3
93.3
76.0
103.4
97.4
92.0
104.8
92.9
76.9
89.0
91.6
101.7
94.1
88.6
88.6
93.3
91.7
96.5
87.5
95.5
92.5
93.6 .
92.8
112.6
83.0
85.0
92.0


A3
90.4
91.3
94.9
81.1
85.4
86.2
84.4
77.8
91.5
95.6
94.2
104.9
112.9
96.1
87.7
96.0
76.2
98.0
94.2
93.2
91.0
93.0
95.9
72.3
91.2
94.8
92.8
108.0
96.3
81.6
90.7
93.2
90.3
101.1
95.9
96.9
94.1
92.6
99.5
96.0
95.3
95.5
97.9
92.7
102.3
86.6
94.4
105.6


A4
89.6
90.2
94.6
80.8
84.4
86.6
86.0
76.0
90.9
94.6
94.0
113.0
111.8
94.2
87.5
94.9
75.5
95.3
93.2
91.3
90.7
93.0
96.1
72.1
85.8
94.3
93.0
111.7
97.3
79.6
92.8
94.8
88.9
100.3
95.9
95.5
100.8
97.0
103.0
101.5
94.0
98.4
97.0
98.1
95.9
84.0
93.8
10.7.7
REQUIRED 625
AVE STD. RANGE FOR
A5 % REC DEV. AVERAGE
86.9
88.5
91.5
76.8
81.0
82.6
85.3
74.5
88.0
92.7
91.4
111.3
108.5
92.5
83.7
92.8
73.7
94.7
92.1
89.5
89.2
92.6
92.1
71.4
85.2
86.8
92.6
109.6
95.6
79.8
91.4
96.2
91.1
99.5
94.9
94.2
98.6
96.9
98.1
98.0
93.7
95.8
92.2
95.6
96.1
83.3
97.0
100.1
89.4
90.5
94.2
79.3
82.9
83.8
85.3
75.7
91.1
95.0
94.6
108.1
111.9
95.1
85.7
94.4
74.2
97.6
94.2
91.9
90.9
93.9
94.6
73.6
91.4
94.2
93.2
109.6
96.4
80.3
91.4
94.9
93.4
99.3
94.3
94.0
95.9
94.2
98.6
95.5
94.8
95.7
95.2
97.1
101.6
85.4
92.9
101.8
1.3
1.1
1.4
1.6
2.2
2.6
0.5
1.6
1.8
1.3
2.1
5.1
1.8
1.5
1.6
1.6
1.8
2.5
1.6
2.0
1.6
1.3
1.6
2.0
6.5
4.0
1.3
3.2
2.3
2.3
1.4
2.4
4.6
2.7
2.9
2.8
3.2
2.3
2.5
4.6
0.8
1.9
2.1
5.1
6.1
2.7
4.1
5.5
16.6—100
42.9--126
36.2--120.4
16.7—153.9
37.3—105.7
48.6—112.0
13.6—197.9
55.2—100.0
54.3—157.6
46.6—180.2
45.0—166.7
41.8-109.0
49.2-164.7
52.5-121.7
57.3-129.2
35.6-119.6
37.8-102.2
40.8—127.9

64.5-113.5
0—100^
68.1— 13|fl
60.1-132H
D— 172.9
13.0—106.5
47.5—126.9
D-100
38.4—144.7
71.6-108.4
53.0-100.0
64.9—114.4
7.8—141.5
38.1-151.8
65.2—108.7
43.4-118.0
8.4-111.0
42.9-121.3
69.6-100.0
D--139.9
28.9—136.8
41.8-133
44.1—139.9
18.6-131.8
42.0—140.4
25.2-145.7
31.7-148.0
D-199.7
D— 195.0
                                                           24

-------
                                    TABLE 5
   GC/MS Method Validation  Study   (October 30, 1992)
   Performed by Ed Messer.  EPA Central Regional (III) Laboratory
   TARGET COMPOUNDS
 2 Phenol
 3 Bis(2-Chloroethyl)Ether
 4 2-Chlorophenol
 5 1,3-Oichlorobenzene
 7 l.4-D1chlorobenzene
 8 1,2-Dichlorobenzene
13 Nitrosodi-n-propylamine
14 Hexachloroethane
15 Nitrobenzene
16 Isophorone
17 2-Nitrophenol
18 2.4-Dimethyl Phenol
20 Bis(2-Chloroethoxy)Methane
21 2-4-Oichlorophenol
22 1.2,4-Tri-chlorobenzene
24 Naphthalene
25 Hexachlorobutadiene
26 4-Chloro-3-Methyl Phenol
29 2,4,6-Trichlorophenol
31 2-Chloronaphthalene
32 Dimethyl Phthalate
33 2,6-Dinitrotoluene
36 Acenaphthene
37 2,4-Dinitrophenol
40 4-Nitrophenol
41 2,4-Dinitrotoluene
42 Diethyl' Phthalate
43 4-Chlorophenyl Phenyl Ether
44 Fluorene
45 2-Methyl-4.6-Dinitrophenol
47 4-Bromophenyl Phenyl Ether
48 Hexachlorobenzene
49 Pentachlorophenol
51 Phenanthrene
52 Anthracene
53 Di-n-Butyl Phthalate
54 Fluoranthene
55 Pyrene
56 Benzyl Butyl Phthalate
57 Bis(2-Ethylhexyl)Phthalate
59 Benzo(a)Anthracene
60 Chrysene
61. Di-n-Octyl Phthalate
62 Benzo(b)Fluoranthene
63 Benzo(k)Fluoranthene
64 Benzo(a)Pyrene
67 Dibenzo(a,h)Anthracene
68 Benzo(g,h,i)Perylene
TRUE
MV1
MV2
MV3
MV4
AVERAGE
VALUE
100
100
100
100
100
100
100
100
100
100
100
100
100
100
.100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
. 100
100
100
100
100
100
100
100
100
100
100
82.9
84.1
85.6
61.1
66.7
65.8
76.6
63.0
98.9
97.9
96.1
84.7
112.0
92.1
73.8
85.7
62.9
105.0
95.1
87.9
98.1
87.0
94.2
68.6
109.0
94.7
90.9
103.0
91.1
80.5
87.8
87.0
94.9
86.9
85.7
85.6
90.9
84.5
102.0
83.2
91.5
86.9
85.6
97.5
87.6
78.0
71.6
66.4
88.3
94.5
92.1
71.3
72.0
75.0
88.2
68.1
91.6
102.0
87.3
104.0
119.0
92.9
77.5
88.7
67.2
108.0
97.8
91.6
97.6
92.3
92.4
61.0
114.0
100.0
96.2
105.0
92.2
85.9
91.1
95.1
102.0
94.1
89.3
87.0
' 95.3
94.3
104.0
86.7
99.2
87.8
94.9
96.0
93.2
79.9
71.9
65.5
95.2
102.0
99.9
78.6
75.5
83.3
92.2
70.1
93.6
104.0
92.3
102.0
124.0
94.8
82.1
88.6
69.6
108.0
98.2
95.3
99.3
96.1
94.3
62.5
111.0
104.0
101.0
107.0
93.8
77.1
90.4
95.3
95.5
97.9
89.5
91.0
103.0
104.0
108.0
94.9
103.0
92.1
98.4
83.0
98.3
73.2
70.4
80.6
94.2
98.6
96.0
74.6
72.0
78.
90.
68.
96.
105.
97.
102.
125.
95.
81.
89.
69.
109.
95.
92.
94.
94.
91.
72.
110.
102.
97.
103.
90.
79.
87.
95.
101.
94.
86.
87.
93.
93.
100.
86.
98.
86.
92.
91.
94.
75.
70.
70.
,7
.5
,9
,2
0
2
0
0
4
2
6
0
0
8
8
6
8
2
1
0
0
6
0
7
7
8
8
0
6
2
0
3
5
0
2
4
5
7
3
1
2
7
6
90.2
94.8
93.4
71.4
71.6
75.7
86.9
67.5
95.1
102.2
93.2
98.2
120.0
93.8
78.7
88.2
67.2
107.5
96.7
91.9
97.4
92.6
. 93.0
66.1
111.0
100.2
96.4
104.5
92.0
80.8
89.3
93.3
98.4
93.4
87.7
87.7
95.6
94.1
103.5
87.8
98.0
88.3
92.9
92.0
93.3
76.6
71.2
70.8

RANGE
REQUIRED
STD
FOR AVG
16
42
36
16
37
48
13
55
54
46
45
41
49
52
57
35
37
40
52
64

68
60

13
47

38
71
53
64
7
.6-100
.9-126
.2-120
.7-153
.3-105
.6-112
.6-197
.2-100
.3-157
.6-180
.0-166
.8-109
.2-164
.5-121
.3-129
.6-119
.8-102
.8-127
.4-129
.5-113
D-100
.1-136
.2-132
D-172
.0-160
.5-126
D-100
.4-144
.6-108
.0-100
.9-114
.8-141
38.1-151,
65.2-108,
43.4-118,
.0
.0
.4
.7
.7
.0
.9
.0
.6
.2
.7
.0
.7
.7
.2
.6
.2
.9
.2
.5
.0
.7
.3
.9
.5
.9
.0
.7
.4
.0
.4
.5
.8
.7
.0
8.4-111.0
42.9-121,
69.6-100,

D-139,
28.9-136.
41
.8-133.
44.1-139.
.3
.0
.9
.8
,0
,9
18.6-131.8
42.0-140.4
25.2-145.7
31.
.7-148.0
D-199.7
D-195.0
DEV
4.9
6.7
5.3
6.5
3.1
6.4
6.1
2.7
2.7
2.7
3.9
7.8
5.1
1.3
3.3
1.5
2.6
1.5
1.3
2.7
1.7
3.5
1.3
4.5
1.9
3.5
3.6
1.7
1.2
3.2
1.5
3.6
3.2
4.0
1.7
2.0
4.5
6.9
3.0
4.3
4.1
2.2
4.7
5.7
3.8
2.6
0.6
6.0
625
LIMIT
STD DEV
22.6
55.0
28.7
41.7
32.1
30.9
55.4
24.5
39.3
63.3
. 35.2
26.1
34.5
26.4
28.1
30.1
26.3
37.2
31.7
13.0
23.2
29.6
27.6
49.8
47.2
21.8
26.5
33.4
20.7
93.2
23.0
24.9
48.9
20.6
32.0
16.7
32.8
25.2
23.4
41.1
27.6
48.3
31.4
38.8
32.3
39.0
70.0
58.9
                                                  25

-------
                            Figure
                  MINI vs  MACRO  EXTRACTION
K.
U
>
O
O
U
a:

i-

u
O
a:
u
a.

u
     130
     120 -
     110 -
100 -
                                                               vO
                                                           64
                           COMPOUND NUMBERS

                       D   MACRO        -h  MINI

-------
                           TABLE 6

   Recovery of Anilines and Benzidines  ("Basic Compounds"):

   Min. Cont. Liq/Liq. Extractors (  195 mm  )


Spikes:         100 ug/L (1.0 mL of  100 ug/mL Working Stock)

Stocks:         5000 mg/L from EPA RTP Repository (MEOH);

Working Stock:  diluted 200 uL of each RTP  Stock to 10 ML MEOH.

Extraction:     24 hrs; pH >11 (6N NaOH); Chiller 3C;
                         200 mL Fisher "Optima" CH2C12; 1L Lab Pure Water.

Recovery:       Determined Relative  to Direct Analysis of Spike.


Compound ***


2-Nitroaniline

3-Nitroaniline

   izidine

3,3'-Di chlorobenzi di ne


*** Aniline was tested as well, but  the MEOHMeoH in the-reference solution
    resulted in multiple peaks (6-8  peaks)  in the reference resulting
    in exaggerated recoveries (»150% in the extracts (MeOH does  not CE extract).
    This chromatographic problem was not observed for the substituted anilines or benzidines.
Run 1
96
101
92.

.2
,9
.7
100
Run 2
101
107
98
101

.3
.8
.7
.5
ng or %
Run 3
103.9
114
113.5
116.7
Recovery
Run 4
100.8
107.6
101.4
103.8
Run 5
99,
108,
109,
110.

.4
.3
.9
,5
Average
Recovery
100.3
107.9
103.2
106.5
EPA
625
RECOVERY
-
-
-
8.2-212
Std.
Dev.
(N-l)
2.8
4.3
8.4
7.0
EPA
625
Std.Oev
-
-
-
71.4
                                                           27

-------
                            TABLE 7
 Pesticide Validation  (Miniature Continuous Liquid/Liquid Extraction)   [ FILE :  PESTCE.VK1 ]

 100 ug/L Spike into Lab Pure Water (1.0 mL of 500 uL of AccuStandard Z-014C diluted to 10.0 ML with
 MEOH).
 6C/MS ANALYSIS

 Compound

 Alpha-BHC
 Oelta-BHC
 Gamma-BHC
 Beta-BHC
 Heptachlor
 Aldrln (HHDN)
 Heptachlor Epoxide
 Endosulfan I
 4,4'-DDE
 Dieldrin
'Endrln
 4.4'-DDD
 Endrin Aldehyde
 4.4'-DDT
 Endosulfan Cyclic  Sulfate
 Endosulfan II
Run 1

96.1
98.8
95.9
99.1
93.5
90.9
89.8
99.2
96
90.9
86.6
95.6
94.5
95.9
91.2
98.9
Run

103
% Recovery
2 Run 3 Run

.6
104
103
106
103
101
100
. 107
99
.5
.3
.8
.2
.4
.5
.5
101
91
100
98
102
98
103
.2
.7
.3
.4
.2
.1

92.2
91.8
98.8
105.3
109
101.9
94.2
108.3
96.6
95.6
99
94.2
92.9
86
87.7
102.7

101


91
• 101
98
96
99
4

.1
92
91
.7
,9
.2
.7
.9
95.3
95
81
94
93
95
.1
.3
.2
.3
.6
97
98
.4
Run

98
101
98
103
100
98
97
103
99
98
80
101
108
105
107
Average
5 Recovery

.4
.2
.9
.3
.3
.4
.7
.7
.3
.9
.3
.5
.4
.9
.2
108

98.
97.
97.
101.
101.
98.
95.
103.
97.
96.
87.
97.
97.
97.
96.
102.

,3
.6
6
1
7
1
8
7
3
3
7
2
5
2
3
2
40 CFR
Requi red
Recovery

37-134
19-140
32-127
17-147
34-111
42-122
37-132
45-153
30-145
36-146
30-147
31-141
-
25-160
726-144
D-202
Std.
Dev.
(n-1)
4.42
5.49
4.59
5.96
5.63
4.36
4.00
4.19
1.94
3.87
7.70
3.58
6.47
7.62
7.46
3.87
                                                           28

-------
                                                    Figure 5
100.0-1
 100
482.1-
                                            DATA: CHLOR02 #1
                                            CALI: CAL0715 #3
RIC+MASS CHRGMATOGRAM     '          DATA: CHLOR02 #1          SCANS  1700 TO I860
82/17/93 14:17:00
SAMPLE: FC43
CONDS.: UPGRADE 38C 2 MIN TO 300 AT 10C/MIN
RANGE: G   1,3675  LABEL: N  1, 4.0  QUAN: A -1.. 4.0  J   8   BASE:  U  20,   3
                                                              1799

                                        1136.                  3253!
                                        3001.
                        1729
                         15.
                         29.
                         T
                                                                      1799
                                                                      5769.
                                                                      13547.
                  1198.
                100.030

               ±  0.509
1841

 12.
 RIC
.176
392
946


6
9.
1.
J
1


L k
• ' ' ' ' ^L ' ' 1 ' 1 ' ^ I
m 1720 1740 1760 •l0 1800 1820 1843 •360
T40 22:56 23:12 23:28 T3:44 24:00 24:16 24:32 ~4:48

SCAN
TIME

-------
                                                     Figure 6
SAMPLE
                LIBRARY SEARCH
                82/17/93 14:17:68 + 23:59
                SAMPLE: FC43
                CQNDS.! UPGRADE 30C 2 HIM'TO 380 AT  10C/MIN
                ENHANCED 3,4-METHENO-2H-CYCLOBUTACCD3PEHTALEN-2-OHE, 1,1A,3,3ft,4,5,5,5A,5B-N!
 M/Z
                           100
150
280
256
308
358
480

-------
                    TABLE 8

Miniature Continuous L/L Extraction:  Chlordane

    50 ug/L spike (Methanol.  EPA-RTP).  GCMS (100 m/z)
                    % RECOVERY:                                    STO.
                                                                  OEV.
         Replicate  Replicate  Replicate  Replicate    AVE.        ug/L
             #1          #2          #3         #4         %         (n-1)

             100       112.1        108       112        108        2.7
                                      608 REQ.      (55.2  -  109)    10.0
                                                   31

-------
                              TABLE 9

         PERCENT RECOVERY OF TOXAPHENE via MINIATURE CONTINUOUS EXTRACTION
=====================:====:===:================================================
TARGET|            PERCENT RECOVERY
CPD    I
t      j    REP-1      REP-2     REP-3     REP-4     REP-5    AVERAGE std (n-1)
====== I =================s======:============:========s:=============:==========
     1  |     100        112       108       112       108       108      4.9
                             TABLE 10

        RECOVERY  OF  PCB-1260 VIA MINIATURE CONTINUOUS EXTRACTION
      ============
TARGET)            PERCENT RECOVERY
CPO    |                                                             LIMIT
#      I PCB1242-2  PCB1242-3  PCB1242-4 PCB1242-5    AVE    std (n-l)STO DEV
==== = = I ============
    1  |     94        100        101        105       100       4.6       12
                             TABLE  11

        RECOVERY OF PCB-1242 VIA MINIATURE CONTINUOUS  EXTRACTION
TARGET|                 '     PERCENT  RECOVERY     .
CPD   j
«     j  1260CE1   1260CE2    1260CE3   1260CE4   1260CE5    AVERAGE std  (n-1)
====== I == ================================ ========= === ======================
    1 I     91       100        101        95       101        97.6    4.6
                                             32

-------
                                              Figure 7    PCB  1260
                                                            0
1 .

                                                                                                                             0
                                                                                                                             E
                                                                                                                             H
                                                            D
                                                            >
                                                            H
                                                            >

                                                            (D
                                                            U
                                                            0
                                                            CO
                                                            M
                                                            M
                                                            td
                                                            /
                                                            0
                                                            0
                                                            p
                                                            fl
                                                            0
                                                            p
                                                            0
                                                            M

                                                            b
                                     i o
so
C3O

-------
                           TABLE 12







Surrogate Recovery (Miniature & Macro-  Continuous Liquid/Liquid Extraction)









50 ug/L Spike into Secondary Effluent (100 uL  of  500 ug/mL of USEPA RTP Repository Standard).




                                            % Recovery
Compound
2-Fluorophenol
05-Phenol
05-Nitrobenzene
2-Fluoro-1,1'-biphenyl
2,4,6-Tribromophenol
D14-p-Terphenyl
Min. CE 1
Run 1
85.3
89.5
91.8
88.6
96.8
92.6

-------
                         TABLE 13




Continuous Extraction of Secondary Effluent




      Miniature (HIN.) vs.  Macro- CEs




Quantitat ion Based on Assumed Response Factor = 1.




Qualitative Identifications Based Upon EPA-NIH Mass Spectral Library Match.
Scan
413
680
794
815
948
'1012
1103
1106
1125
1393
1404
1583
1743
1755
1911
Compound
Name
dimethyl disulfide
sulfonylbismethane
1-(2-methoxy-1 -methyl ethoxy)-2-propanol
1-(2-methoxypropoxy)-2-propanol
2-(methylthio)pyridine
4,4,5-trimethyl-2-hexene
2-(2-hydroxypropoxy)-1-propanol
3-Ethyl-4-Methyl-1H-pyrrole-2,5-dione
2-methyl-2-(,1-methylethoxy)propane
N,N-diethyl-1,2-ethanediamine
2,6-bis(1,1-dimethylethyl)-4-methylphenol
4-(dimethylamino)-3-methyl-2-butanone
Caffeine
4-(dimethylamino)-3-methyl-2-butanone
methoxycylobutane
MIN. CE
4.5
4
3.3
4.6
0.7
0.6
8.4
0.7
0.7
2.6
1.9
7.6
2.9
2.9
0.7
Estimated
Concentration
(ug/L)
Max. CE
3.7
4.2
2.9
4.2
0.6
1.3
7.5
0.6
0.7
2.5
1.7
8
2.5
4.2
1.5
                                                 35

-------
                            Appendix:
**  Calibration standards and Spiking Solutions/Procedures.
**  Hydrophobic Membranes—possible mode of extract drying with
       minimum use of rinsing solvents.
**  Possible "Solvent Effect" associated with methanol.
                               36

-------
                    Calibration Standards and
                  Spiking Solutions/Procedures:
Spiking Solutions  (General);

The  spiking cocktails  were methanol  or acetone  (miscible with
water) .  Additions were made using volumetric pipets, or calibrated
uL pipets.   A "reference  standard"  was prepared using  the same
volume as was the spike into a volumetric flask of the same volume
as the final K-D volume  (final extract volume) .

Calibration Standards fGC/MS);

Calibration standards (10,  20, 50, 100 ng/uL in MeCL2)  for all of
the tested semivolatile compounds (excluding pesticides and PCBs)
were  prepared  volumetrically  from AccuStandard  (New Haven,  CT)
stock solutions (ampules at 2 mg/mL in MeCl2).  The procedure was
as follows:
AccuStd .
(Stock ID)
Z-014A
Z-014B
Z-014D
Z-014E
Z-014G-R
Z-014P
Z-014A
Content
(in MeCl,)
Base Neutral Mix 1
Base Neutral Mix 2
Tox. Sub. Mix 1
Tox. Sub. Mix 2
PNA Mix
Phenols Mix
Int. Std. Mix
Cone.
(ng/uL
2000
2000
2000
2000
2000
2000
4000
Vol. (uL)
AccuStd.
Stock a
Vol. (uL)
AccuStd .
Int. Std.
Final
Vol. (mL) c
Final Cone.
(ng/uL)
10.0
20.0
2.0
10.0

20.0
20.0
2.0
20.0

50.0
20.0
2.0
50.0

50.0
10.0
1.0
100.0

  50 uL syringe.    20 uL syringe.   Volumetric flask  (Class "A")
                              37

-------
The  compound  names  are  delineated  in  Table  #4.    The  final
concentrations of the  internal  standards were 40 ng/uL.

Internal Standards;

AccuStandard  Internal  Standard Mix  (Z-014J)  was employed,  which
consisted  of  4000 ng/uL  (in methylene chloride) of  each of the
following: dlO-acenaphthene; d!2-chrysene; d4-l,4-dichlorobenzene;
d8-naphthalene;  d!2-perylene;  and dlO-phenanthrene.   Because of
concern for the stability  of dlO-perylene in solution,  it was not
employed as an internal standard  (quantitation).

All extracts were spiked with the internal standards mix just  prior
to GC/MS analysis.

Superfund CLP "Matrix  Spikes" (MS):

Stock  solutions  (ampules)  at  5000  ng/uL  were in  methanol  were
obtained from the EPA Quality  Assurance Materials Bank  in  RTF,
North Carolina.  These were diluted  10 fold with methanol to give
500 ng/uL spiking solutions.  One hundred uL spikes were performed
directly into 1 L of the aqueous samples (deionized water) prior to
continuous extraction  (50 ug/L spike).  The  compounds included:
1,2-dichlorobenzene; N-nitroso-n-propylamine; 2,4-dinitrotoluene;
di-n-butylphthalate; acenaphthene; 1,2,4-trichlorobenzene; pyrene;
4-nitrophenol; pentachlorophenol; 4-chloro-3-methylphenol; phenol;
and 2-chlorophenol.

This spiking cocktail  was  used during the design/re-design phases
of this work,  in which a  relatively simple mixture,  with a  wide
range of chemical qualities could be accurately measured via GC/FlD
(a means of quickly screening different CE designs, without costly
GC/MS analyses).

Superfund CLP "Surrogate Compound" Spikes:

Stock solutions (ampules at 5000 ng/uL in methanol)  were obtained
from  the  EPA Quality  Assurance Materials  Bank  in   RTF,  North
Carolina.  These were  diluted 5 fold  to give  1000  ng/uL  spiking
solutions.  One hundred uL  spikes were placed directly into 1  L of
the  aqueous   samples  prior  to  continuous  extraction  (100  ug/L
spike).  The  compounds included: 2-fluorophenol; d5-phenol; d5-
nitrobenzene;   2-fluoro-l,l-'biphenyl; 2,4,6-tribromophenol;  and
d!4-p-terphenyl.

BNA QC Spiking Solution:

Quality  Control  solutions  (CRADA  ampules)  were  obtained  from
Supelco Inc.,  Bellefonte,  PA.   Base/Neutral #1, Base/Neutral  #2
ampules were in acetone and contained 37 target semivolatile target
priority pollutant (base/neutral)  compounds.    Acid   #1  ampules
contained 11 priority pollutant  (acid)  compounds in methanol.   All
compounds were at a concentration of 100 ng/uL.   These were added
as 1.0  mL  (class  A  volumetric pipets) into  1000  mL  of  aqueous
                               38

-------
sample  (deionized  water)  to result in  a 100 ug/L spike for each
compound.  The compound names are delineated in Table # 4.

Benzidine/s and Aniline/s Spikes;

Stock  solutions  (ampules)  were  obtained  from  the  EPA  Quality
Assurance Materials  Bank,  RTF/  North Carolina. C-075 and 62-53-3
(aniline) at 5000 ng/uL in benzene were diluted 50 fold  (200 UL to
10 mil in methanol)  to result in a working stock of 100 ng/uL.  The
excessive  dilution  was  performed  to  maximize  the  quantity  of
methanol  (hydrophilic solvent).   This mixed  stock  solution was
added  as 1.0 mL   (class  "A"  volumetric  pipet)  into 1000  mL  of
aqueous sample (deionized water) to result in a 100 ug/L spike for
each compound.  The  compounds are delineated in Table # 6.

Pesticide Spikes;

   Single Component  Analvtes;

Stock solutions (ampules) were obtained from Accustandard Z-014C at
2000 ng/uL in 1:1 toluene/hexane.  These were diluted  500 uL to 10
mL in methanol (to maximize the hydrophilic  solvent).  This stock
solution was added as 1.0 mL (class "A" volumetric  pipet) into 1000
mL of  aqueous sample  (deionized water)  to  result in a 100 ug/L
spike  for  each compound.   The  compound  names are delineated  in
Table # 7.

   Toxaphene;

Stock  solutions  (ampules)  were  obtained  from  the   EPA  Quality
Assurance Materials Bank, RTF, North Carolina. These were as 1000
ng/uL solutions in methanol.   Spikes were prepared by the addition
of 50  uL of  the  stock  solutions  to  1000  mL of  aqueous  sample
(deionized water) to result in a 50 ug/L  spike for this compound.
The resulting CE extracts were exchanged  to  hexane (as per method
608/508).

   Chlordane;

Stock  solutions  (ampules)  were  obtained  from  the   EPA  Quality
Assurance Materials Bank, RTP, North Carolina. These were as 1000
ng/uL solutions in methanol.   Spikes were prepared  by the addition
of 50  uL of  the  stock  solutions  to  1000  mL of  aqueous  sample
(deionized water) to result in a 50 ug/L  spike for this compound.

   PCBs;

Stock solutions (ampules) were obtained from Supelco, Inc.  (CRADA
QC material) for Aroclor  1260 and  1242.  These solutions  were  in
acetone at 50 ng/uL.   Spikes were prepared by the addition of 1 mL
(class  "A"  volumetric  pipet)   to  1000 mL of   aqueous  sample
(deionized water)  to  result in a 50 ug/L spike for these compounds.
The resulting  extracts were exchanged  to hexane  (as per  method
608/508).
                                39

-------
                                            Hydrophobic Membranes
                                                                      [filters2.wkl]
EXPERIMENT: Validation of the hydrophobic filters.  60 mL of methylene
           chloride spiked with the list below and gravity filtered.
"Extracts were concentrated via K-D (EPA 625)  and
GC/MS analysis was performed vs. a "reference  spike".
COMPOUND
	 	 ;_
— •-- 	
1 METHANAMINE. N-METHYL-N-NITROSO-
2 PHENOL *CCC*
3 ETHANE. l.l'-OXYBIS\2-CHLORO-
4 2-CHLOROPHENOL
5 1.3-DICHLOROBENZENE (COELUTES)
6 *** D4-1.4-DICHLOROBENZENE ***INTERNAL STD.***
7 1.4-DJCHLOROBENZENE (COELUTES) *CCC*
8 1.2-DICHLOROBENZENE (COELUTES)
9 BENZENEMETHANOL HSL
10 2-METHYLPHENOL HSL
11 BIS(2-CHLOROISOPROPYL)ETHER
12 4-METHYLPHENOL HSL
13 1-PROPANAMINE. N-NITROSO-N-PROPYL-
14 ETHANE, HEXACHLORO- •
15 BENZENE. NITRO-
16 2-CYCLOHEXEN-l-ONE. 3.5.5-TRIMETHYL- *CCC*
17 2-NITROPHENOL
18 PHENOL. 2,4-DIMETHYL-
19 BENZOIC ACID HSL
20 ETHANE. l.l'-[KETHYLENEBIS(OXY)]BIS[2-CHLORO-
21 2.4-DICHLOROPHENOL
22 BENZENE. 1,2.4-TRICHLORO-
23 *** 08-NAPHTHALENE *** INTERNAL STD.***
24 NAPHTHALENE
25 1,3-BUTADIENE. 1,1,2.3.4.4-HEXACHLORO- *CCC*
26 PHENOL. 4-CHLORO-3-METHYL-
27 NAPHTHALENE, 2-METHYL- HSL
28 1.2.3.4.5,5-HEXACHLORO-1,3-CYCLOPENTADIENE *SPCC*
29 2.4,6-TRICHLOROPHENOL *CCC*
30 2.4.5-TRICHLOROPHENOL HSL
31 2-CHLORONAPHTHALENE
32 1,2-BEHZENEDICARBOXYLIC ACID. DIMETHYLESTER
33 2.6-DINITROTOLUENE
34 ACENAPHTHYLENE
35 *** D10-PHENANTRENE ***INTERNAL STD.***
36 ACENAPHTHYLENE. 1.2-DI HYDRO- *CCC*
37 2.4-DINITROPHENOL *SPCC* . .
38 PHENOL. 4-NITRO- *SPCC*
39 DIBENZOFURAN HSL " '
40 4-NITROPHENOL (SEC ION)
41 2.4-DIN1TROTOLUENE
42 1.2-BENZENEDICARBOXYLIC ACID. DIMETHYLESTER
43 l-CHLORO-4-PHENOXYBENZENE
44 9H-FLOURENE
45 2-METHYL-4.6-DINITROPHENOL
Percent R
T073101

85.500
93.354
92.498
90.780

91.007
86.060
85.535
86.503
82.977
69.216
84.193
83.023
97.464
98.508
96.241
104.924
100.037
95.042
90.548
93.348

91.623
93.790
92.721
90.745
64.775
84.823
80.859
83.988
85.173
76.472
81.324

82.432
69.712
67.536
84.284
81.055
77.978
81.574
87.701
87.246
80.430
scoveri es
SAMPLE NU
T073102

90.268
92.885
96.790
95.814

92.881
87.442
89.520
88.717
82.477
97.299
86.212
86.868
91.929
92.175
99.470
107.665
71.771
96.104
97.734
96.470

90.906
98.902
95.399
93.457
66.911
93.313
92.302
86.651
92.398
87.642
88.287

87.826
83.007
75.770
87.618
95.171
84.672
86.406
96.009
92.556
89.049
IB
AVERAGE

87.884
93.120
94.644
93.297

91.944
86.751
87.528
87.610
82.727
93.258
85.203
84.946
94.697
95.342
97.856
106.295
85.904
95.573
94.141
94.909

91.265
96.346
94.060
92.101
65.843
89.068
86.581
85.320
88.786
82.057
84.806

85.129
76.360
71.653
85.951
88.113
81.325
83.990
91.855
89.901
84.740
STD. DEV.

2.4
0.2
2.1
2.5

0.9
0.7
2.0
1.1
0.3
4.0
1.0
1.9
2.8
3.2
1.6
1.4
14.1
0.5
3.6
1.6

0.4
2.6
1.3
1.4
1.1
4.2
5.7
1.3
3.6
5.6
3.5

2.7
6.6
4.1
1.7
7.1
3.3
2.4
4.2
2.7
4.3
                                                        40

-------
                                             Hydrophobic  Membranes  (cont'd)
46 ^»HO STD.  AVAILABLE
47  BROMOPHENOXYBENZENE
48  BENZENE,  HEXACHLORO- *CCC*
49  PENTACHLOROPHENOL
SO  *** D10-PHENANTHRENE ***INTERNAL STD.***
51  FHENANTHRENE
52  ANTHRACENE
53  1,2-BENZENEDICARBOXYLIC ACID. DIBUTYLESTER
54  FLUORANTHENE CCC*
55  PYRENE
56  N-BUTYL BENZYL PHTHALATE
57  B1S(2-ETHYLHEXYL)PHTHALATE
58  *** D12-CHRYSENE***INTERNAL STD.***
59  BENZ/A/ANTHRACENE
60  CHRYSENE
61  DIOCTYLPHTHALATE *CCC*
62  BENZO\B\FLUORANTHENE
63  BENZO\K\FLUORANTHENE
64  BENZO/A/PYRENE
65  *** D12-PERYLENE ***INTERNAL STD.***
66  INDENO(1.2,3-CD)PYRENE
67  DIBENZO(A.H) ANTHRACENE
68  BENZO\GHI\PERYLENE
N-PHENYLBENZENEAMINE DECOMP. OF NNDPA
78.500
79.194
78.996
79.313
82.499
79.466
86.907
91.250
89.885
97.158
103.051
87.408
94.722
74.639
74.949
82.893
83.144
89.225
92.429
88.329
82.387
87.924
89.361
92.561
89.901
87.524
93.386
107.378
103.851
106.327
108.143
91.997
97.892
93.802
86.738
92.354
90.382
90.171
99.621
96.353.
80.444
83.559
84.179
85.937
86.200
83.495
90.147
99.314
96.868
101.743
105.597
89.703
96.307
84.221
80.844
87.624
86.763
89.698
96.025
92.341
1.9
4.4
5.2
6.6
3.7
4.0
3.2
8.1
7.0
4.6
2.5
2.3
1.6
9.6
5.9
4.7
3.6
0.5
3.6
4.0

-------
                                        Possible  "Solvent Effect"
                                                (Methanol)
                            MATRIX SPIKE X RECOVERY MICRO—REFERENCE CONTAINING 200 UL OF METHANOL
(1) PHENOL
(2) 2-CHLOROPHENOL
(3) 1,4-DICHLOROBENZENE
(4) N-NITROSO-N-PROPYL-1-PROPANAMINE
(5) 1,2,4-TRICHLOROBENZENE
(6) PHENOL-4-CHLORO-3-METHYL-
(7) ACENAPHTHENE
(8) 4-NITROPHENOL
(9) 2,4-DINITROTOLUENE
(10) PENTACHLOROPHENOL
(11) 1,2-BENZENEDICARBOXYLICACID,DIBUTYLESTER
(12) PYRENE
SAMPLE

1

(12-89)
30C( initial temp.
921101-01
921101-02
921101-03
921101-04
REF.ffl
REF. #2
REF. #3
AVE. SAMPLES
STD.DEV. SAMPLE
AVE. REF.
) via
83.7
78.3
77.2
84.5
91.9
100.0
97.8
80.9
3-7
96.6
STD.DEV.REF. 4.2
2

(27-123)
GC/MS: 30M
87.7
84.2
82.8
91.0
96.4
100.0
99.0
86.4
3.7
98.5
1.9

50C(initial temp.)GC/MS 30M SPB-5
921101-01
921101-02**
921101-03
921101-04
REF.#1
REF. #2 **
REF. #3
AVE. SAMPLES
STD.DEV. SAMPLE
AVE. REF.
STD.DEV.REF.
79.6

78.7
81.5
91.3

100.0
79.9
1.4
95.7
6.2
81.5

82.8
87.4
92.8

100.0
83.9
3.1
96.4
5.1
3 4
TARGET X
(36-97) (41-116)
SPB-5, 1uM FILM
74.4 82.7
72.7 78.5
69.2 77.1
77.1 84.3
97.3 94.4
100.0 100.0
101.0 99.3
73.4 80.7
3.3 3.4
99.4 97.9
1.9 3.1
5
WATER
(39-98)
, 0.32 mm
77.4
75.2
75.9
80.8
97.9
100.0
102.3
77.3
2.5
100.1
2.2
6

(23-97)
ID
86.6
83.1
82.1
88.1
95.1
100.0
93.7
85.0
2.8
96.3
3.3
7

(46-118)

92.5
91.3
90.1
96.8
98.4
100.0
.101.5
92.7
2.9
100.0
1.6
8

(10-80)

146.0
142.6
142.6
151.7
77.0
100.0
106.1
145.7
4.3
94.4
15.3
9

(24-96)

95.6
93.2
95.3
97.1
97.0
100.0
97.7
95.3
1.6
98.2
1.6
10

(9-103)

212.1
229.9
221.6
216.5
100.1
100.0
90.8
220.0
7.6
97.0
5.3
11

(11-117)

95.3
94.2
94.0
93.6
97.3
100.0
96.8
94.3
0.7
98.0
1.7
12

(26-127)

99.0
95.9
97.7
93.8
101.2
100.0
98.9
96.6
2.3
100.0
1.2

, 1uM FILM, 0.32 mm ID
68.8 80.9

73.5 83.9
77.8 87.0
88.5 95.5

100.0 100.0
73.4 83.9
4.5 3.1
94.3 97.8
8.1 3.2
74.0

76.6
78.1
97.9

100.0
76.2
2.1
99.0
1.5
81.5

84.5
85.6
97,2

100.0
83.9
2.1
98.6
2.0
91.1

93.1
92.3
97.2

100.0
92.2
1.0
98.6
2.0
91.7

92.6
94.3
93.7

100.0
92.9
1.3
96.9
. 4.5
93.4

98.7
97.9
89.4

100.0
96.7
'2.9
94.7
7.5
207.3

241.5
251.4
92.0

100.0
233.4
23.1
96.0
5.7
90.3

92.9
96.0
90.5

100.0
93.1
2.9
95.3
6.7
84.2

91.1
91.1
92.1

100.0
88.8
4.0
96.1
5.6
»****»*******»**«,:***»*********»**********************************************************
                                                  42

-------
                                        Possible  "Solvent  Effect"
(Methanol)
Cont'd
fMb
^Rinitial temp.) GC/FID: 60M SPB-5, 0.25uM FILM,
921101-01
921101-02
921101-03
921101-04
REF.tfl
REF.#2
REF.tt
AVE. SAMPLES
STO.OEV. SAMPLE
AVE. REF.
STD.OEV.REF.
85.5
87.0
81.6
82.6
100.0
94.4
92.8
84.2
2.5
95.7
3.8
84.1
86.1
81.3
82.2
100.0
94.7
93.2
83.4
2.1
96.0
3.6
69.9
75.9
73.5
74.2
100.0
95.8
93.9
73.4
2.5
96.6
3.1
77.1
77.9
73.7
74.7
100.0
94.8
92.3
75.9
2.0
95.7
3.9
0.32imi ID
74.1
78.1
77.4
78.0
100.0
95.0
93.4
76.9
1.9
96.1
3.4

87.4
87.8
81.9
83.3
100.0
95.2
93.3
85.1
2.9
96.2
3.5

88.0
86.1
85.3
86.2
100.0
95.5
93.8
86.4
1.1
96.4
3.2

109.7
111.2
98.4
101.4
100.0
96.4
91.1
105.2
6.2
95.8
4.5

90.4
87.8
86.2
87.4
100.0
95.8
94.6
88.0
1.8
96.8
2.8

130.1
129.7
120.4
127.4
100.0
102.3
91.6
126.9
4.5
98.0
5.6

94.9
90.9
88.7
'••90.. 8
100.0
95.7
93.4
91.3
2.6
96.4
3.4

94.7
92.2
89.8
90.7
ioo.o
96.2
94.0
91.9
2.1
96.7
3.0
A*****************************************************************************************
                                                                 Solvent minimization in the continuous liquid/liquid
                                                                 extraction of aqueous samples for semivolatile
                                                                 organics
                                                                 OC:35112543
                                                                                       *.,

-------