United States
                     Environmental Protection
                     Agency
Environmental Monitoring and
Support Laboratory
Cincinnati OH 45268
                     Research and Development
EPA-600/S4-84-006  Mar. 1984
&EPA          Project  Summary

                     Analysis  of the  NBS
                     Sediment  by  the  MRI
                     Sludge  Protocol
                     Viorica Lopez-Avila, Raymond V. Northcutt, Jon Onstot, and Margie
                     Wickham
                      An extraction technique, involving
                     homogenization of a sediment sample
                     with  dichloromethane at dual pH and
                     phase separation by centrifugation,
                     was used in the determination of 51
                     organic priority pollutants as identified
                     in  a  standard  reference sediment
                     sample. These compounds were spiked
                     into the sediment and equilibrated for a
                     defined period of time. The extraction
                     was  performed first at  pH > 11  to
                     isolate the base/neutral compounds;
                     acidic compounds were extracted at <
                     2. The extract containing the base/neu-
                     tral compounds was fractionated by gel
                     permeation  chromatography (GPC).
                     Ssparation and identification of individ-
                     ual compounds in the silica gel fractions
                     was accomplished by gas chromatogra-
                     phy/mass spectrometry (GC/MS).
                     Method  precision and  accuracy are
                     discussed. Tentative identifications  of
                     other organic compounds found in the
                     sediment are given.
                      This Project Summary was developed
                     by  EPA's Environmental Monitoring
                     and Support Laboratory. Cincinnati.
                     Ohio  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).
                      This project investigated the application
                     of the EPA  Interim Protocol  for the
                     Analysis of Extractable Organic Priority
                     Pollutants in Industrial and Municipal
                     Wastewater Treatment Sludge for isola-
                     tion of selected test compounds spiked
                     into a sediment sample. The extraction
                     procedure involved homogenization of a
                     sediment/water slurry with dichlorome-
                     thane at dual pH followed by centrifuga-
tion. A silica gel fractionation of the
base/neutral extract and a GPC fraction-
ation of the acidic extract were performed.
The compounds selected for this study
were 51 priority pollutants representative
of  nonpolar and polar, basic and acidic,
stable and reactive compounds. A stand-
ard reference sediment material available
from the National Bureau of Standards
(NBS) was used  as matrix.

Experimental
  The standard reference sediment
material  was purchased from  the NBS
and  is identified as  NBS Standard
Reference Material No. 1645. According
to NBS,  the material was  dredged
from the bottom of the Indiana Harbor
Canal near Gary, Indiana. The material
was screened to remove foreign objects,
freeze dried, and sieved (particle size <
180 fj). The material was sterilized by
radiation to minimize biological activity.
  The individual stock solutions for each
of the 51  compounds (concentration
approximately 4 mg/mL) were made in
methanol, except for hexachlorobenzene,
chrysene and dibenzo(a,h)anthracene
stock  solutions,  which were  made in
tetrahydrofuran, and anthracene and
benzo(a)pyrene, which were  made in
acetone.  Internal standards naphthalene-
da,  anthracene-dio,  and  chrysene-di2
were purchased from Merck, Sharp and
Dohme, Stable Isotope Division (Quebec,
Canada).
  A 20-g aliquot of the NBS  Standard
Reference  Material  was  loaded into a
225-mL centrifuge bottle and diluted with
60 ml_ purified  water.  The  sample was
homogenized with Tekmar  Tissuemizer
blending probe for 30 sec to enhance the

-------
wetting of the sediment. The spike was
added  immediately following homogen-
ization and allowed to equilibrate with the
sediment through continuous mechanical
tumbling for 24  hours at 4°C.  Three
sequential extractions with 80-mL aliquots
of dichloromethane were performed first
at pH > 11 and then at pH < 2. Following
each  addition of  dichloromethane, the
sample was homogenized for 30 sec with
a Tekmar Tissuemizer blending  probe,
and centrifuged for 30 min at 2,500 rpm
to promote phase  separation.  Three
distinct layers consisting of an aqueous
layer on top,  a dark solvent layer in the
middle, and a firm solid  layer on the
bottom were observed. The base/neutral
extracts were combined and  dried by
passage through a short column of
anhydrous sodium sulfate  prior to con-
centration to 5 ml_ in a Kuderna-Danish
evaporator. Following extraction at pH <
11, the sediment/water  slurry was
adjusted to pH < 2 with 6 N hydrochloric
acid, and the extraction, extract drying,
and concentration steps were repeated to
isolate the acidic compounds.
  Cleanup of the  base/neutral extracts
was performed by silica gel chromatogra-
phy  using 3% water deactivated silica
(Silica  Woelm, 70/150 mesh, ICN). The
elution patterns and the recoveries of the
42  base/neutral  compounds from the
silica gel  column were determined  in
duplicate at two levels (8 fjg, 80 fjg, per
column), corresponding to the spiking
levels in sediment samples. The compounds
were loaded onto the silica gel column by
mixing appropriate aliquots of the stock
solution with 2.5g of 3% water-deactivated
silica gel in a 10-mL beaker. The solvent
was removed under a gentle stream of dry
nitrogen, and the adsorbent containing
the spikes was transferred to the silica gel
column. Four fractions were collected:
Fraction I eluted with 100 mL of hexane;
Fraction II  eluted with 50 mL of  10%
dichloromethane in  hexane; Fraction  III
eluted  with 50 mL of 50% dichlorometh-
ane inhexane; and Fraction  IV eluted with
150 mL of 5% acetone in dichloromethane.
The sediment extracts were processed
identically.
  Extracts containing the acidic compounds
were fractionated by GPC using Biobeads
SX-3 gels (BioRad, Inc.). The GPC column
was calibrated using solutions of corn oil,
n-butylbenzylphthalate, and phenols (e.g.
phenol, 2,4-dichlorophenol, and  penta-
chlorophenol) as  described in the test
procedure.  The  eluting  solvent  was
dichloromethane.  GPC recoveries  of
phenols were determined at 8^g and 80
fjg  per column, corresponding  to the
spiking levels of the sediment samples.
  A Varian  3700  gas chromatograph
equipped with an effluent splitter and a
fused silica capillary column (30 m x 0.25
mm ID) coated with SE-54 (film thickness
0.25 /j)  was  used for  screening the
various fractions prior to GC/MS analysis.
The oven was held at 50°C for 4 min
during  injection  and then temperature
programmed  at 4°C/min to  300°C (10
min hold) for the base/neutral fractions.
The oven was held at 60 °C for 5 min and
temperature programmed at 8°C/min to
325°C  for the acid fractions.  A flame
ionization detector (FID) and an electron
capture detector (ECD) were used simul-
taneously (effluent split ratio  1:1).
  GC/MS analyses were performed on
a Finnigan 4021  quadrupole mass spec-
trometer  interfaced to a  Finnigan 9600
gas chromatograph and  operated under
the control of Incos 2300 data system. All
separations were performed by splitless
injection  using  a  SE-54 fused  silica
capillary  column (30 m  x  0.25  mm ID)
connected directly to the ion source. The
mass spectrometer was  operated in the
electron impact mode. Calibration for the
mass spectrometer was checked daily
using decafluorotriphenylphosphine (DFTPP):
resolution; focusing was adjusted to meet
the DFTPP criteria. Emission current was
in the 0.4-0.6  mA range;  ionization
energy was eV, and the electron multiplier
was operated at 1.7 kV with the preampli-
fier set at 10"8A/V.

Results and Discussion
  The  sample  preparation  technique
described here was originally developed
for POTW sludges. In order to be applied
to the NBS sediment, an aliquot of the dry
sediment is first slurried with water and
then extracted with dichloromethane
using a high speed homogenizer.  If the
sediment particles are  not within  a
relatively narrow particle size range (e.g.
20 to 200 /j), difficulties in isolating the
organic layer from the aqueous and the
solid layers will be experienced.
  Although the evaluation experiments
were conducted using a set of compounds
representative of halogenated hydrocar-
bons, polycyclic aromatic hydrocarbons,
phenols,  etc., the  results indicate that
this extraction technique can be used to
isolate other nonpolar and weakly polar
compounds.
  The cleanup of the  base/neutral and
the acidic fractions was performed by
silica gel and GPC, respectively. Most
of the halogenated hydrocarbons (e.g.,
dichlorobenzenes, trichlorobenzene,
hexachlorobutadiene,  hexachlorocyclo-
pentadiene,  hexachlorobenzene) eluted
in Fraction I, whereas the  polycyclic
aromatic hydrocarbons (e.g. fluoranthene,
pyrene,  benzo(a)pyrene, dibenzo(a,h)an-
thracene) eluted in Fraction III. The more
polar  compounds such as bis(2-chloro-
ethyl) ether, N-nitrosodi-jvpropylamine,
bis(2-chloroethoxy)methane,  dinitrotolu-
enes, phthalates) were  identified in
Fraction  IV. Several  compounds (e.g.,
naphthalene, chloronaphthalene, y-BHC,
p,p'-DDE) were found to spread over two or
three silica  gel fractions;  however, for
each  compound,  at  least 75%  of  the
amount spiked eluted  in only one of the
fractions. The data indicated  that with a
few  exceptions, the  recoveries  of  the
base/neutral compounds from the clean-
up procedures were quantitative and
reproducible.
  All phenols were recovered from spiked
blanks at both spike levels. The chromato-
graphic  behavior  of dinitrophenols may
have  precluded their identification at
levels below 8 ng/yuL in the GPC
fractions. Due  to  these problems, these
compounds  were not  reported for  the
spiked sediment.
  The results of the precision and
accuracy determinations for the base/
neutral compounds are given in Table 1.
Of all recovery determinations in Table 1,
51% fell  within the range 60  to 120%.
An  additional  16%  were below  60%
recovery, and 8% were above the 120%
recovery. Due to the high concentrations
of several compounds in  the  unspiked
sediment,  their recoveries could  not be
determined at the lower spike  level. For
example, fluoranthene, pyrene, chrysene,
and  benzo(a)pyrene  had  levels above
10,000 ng/g dry sediment. Furthermore,
because of the complexity of the sediment
sample, coeluting compounds may have
also contributed  to  the  high  standard
deviations of the  recoveries  of many of
the spikes and the very poor recoveries of
several  compounds.  Polar and reactive
compounds such as hexachlorocyclopen-
tadiene, benzidine, 3,3'-dichlorobenzidine,
heptachlor epoxide (only at 400 ng/g dry
sediment),  endrin, and a-endosulfan
were not recovered.
  The results of the precision and
accuracy determinations for the  acidic
compounds  are given in Table 2. Of all
recovery  determinations,  64% were
greater  than 70%,  29%  were between
40%  and  70%,  and 7% were zero.
Quantitative recoveries (>70%) were
found at both spike levels for phenol, 2-
chlorophenol, 2,4,-dichlorophenol, 2,4-
dimethylphenol, and pentachlorophenol.
4-Nitrophenol  was not detected at  400
ng/g; however, its recovery was satisfac-
tory  at the higher  spike  level.  The
chromatographic behavior of this compound

-------
Table 1.   Results of the Precision and Accuracy Evaluations for the Sludge Protocol Applied to the Analysis of Base/Neutral/Pesticide
          Compounds in the Spiked NBS Sediment
                                                                                                 % Recovery"
Compound
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Compound
Bis/2-ch/oroethyl) ether
1,3-Dichlorobenzene
1,2-Dichlorobenzene
Hexachloroethane
N-Nitroso-di-ji-propylamine
Nitrobenzene
Bis(2-chloroethoxy)methane
1 ,2,4-Trichlorobenzene
Naphthalene
Hexachlorobutadiene
Hexachlorocyclopentadiene
2-Chloronaphthalene
A cenaphthylene
2, 6-Dinitrotoluene
A cenaphthene
2,4-Dinitrotoluene
Fluorene
4-Chtorophenyl phenyl ether
Diethylphthalate
4-Bromophenyl phenyl ether
Hexachlorobenzene
Phenanthrene
Anthracene
Di-n -butylphthalate
Fluoranthene
Pyrene
Benzidine0
Endosulfan sulfate
n-Butylbenzylphthalate
Chrysene
3, 3 '-Dichlorobenzidine
Bis{2 -ethylhexyDphthalate
Benzofa)pyrene
Dibenzo(a,h)anthracene
Y-BHC
Aldrin
Heptachlorepoxide
p,p'-DDE
~Endrin
p.p'-DDD
p,p'-DDT
a-Endosulfan
Unspiked sediment*
concentration
(ng/g dry sediment)
ND
ND
ND
ND
ND
ND
ND
ND
880 + 850
ND
ND
ND
1.820 ± 350
ND
850 ± 120
ND
1,580 + 130
ND
ND
ND
ND
3,770+ 300
2,090 + 72
580 ± 65
14.760+ 2,810
20.020 ± 4,630
ND
1,160, 455
920
27.000 ± 6,500
ND
7,560. 10,790
22,850 + 6,200
5,960 + 420
3,050 + 1,270
ND
ND
340 ± WO
ND
120
840 + 20
ND
Spike level
400 ng/g
dry sediment
96+ 23
63+9
66 ± 14
41 + 15
95 + 38
43+3
94+ 39
67+4
64C
63+2
0
72+ 10
d
110+ 51
d
49C
d
86+ 42
33+ 14
91 + 50
46 + 10
d
d
156C
d
d
0
97, 137
d
d
0
d
d
d
d
44'
0
41C
0
121, 186
d
0
Spike level
4.000 ng/g
dry sediment
112+ 19
68+ 16
75 + 22
62+5
89 + 11
102 + 26
81 + 11
58+ 13
70 + 35
71 + 17
0
94 + 20
86+ 24
102 + 36
104 ± 24
64 + 50
82+7
74+7
109 + 13
39+2
53+4
83 + 26
25+ 11
100, 137
174+ 78
148, 252
0
0
W3C
d
0
d
128 ± 23
113+ 42
50+ 19
22 ± 13
54+9
76 ± 32
0
104 + 18
155 + 43
0
ND = Not detected.
a Mean + standard deviation for three determinations.
b Benzidine not detected by GC/MS in any of the standards at concentrations s 80 ng//j/.
c Single determination reported; levels in the other replicates were below the unspiked sediment level.
d Concentrations determined in the spiked sediment were below the unspiked sediment level.
 may have precluded its identification at
 low spike levels.
  The GC/MS analyses of the silica gel
 fractions and the GPC fraction generated
 from the NBS sediment extract resulted
 in the identification of over 100 base/
 neutral compounds and 11 acidic compounds,
 without considering the various isomers
 of many of these compounds.
  Because authentic reference standards
 were not analyzed for each compound,
 the identifications  are tentative (a few
 compound identifications were confirmed
 with standards). Most of the compounds
give mass spectra that matched the mass
spectra  in the NBS library. Identification
of specific positional  isomers (e.g.
dimethylnapthalenes, trimethylnapha-
lenes, etc.) could not be obtained from the
mass spectrometric information.
  Whenever an identification was made,
either by  library  search  routines or by
manual  interpretation, mass chromato-
grams for the most intense ions  in the
mass spectrum were  obtained.  In many
instances, this allowed identification of
other isomers  and assignment of coelut-
ing peaks. Although in many cases the
library search did indicate the positional
isomers  that best match a particular
unknown mass spectrum, this  was not
sufficient for confirming the compound
identity.  Consequently, no identification
has been made about the position of the
substituent or the particular alkyl radical.
  Fractionation  of  the base/neutral
extract by silica  gel  chromatography
reduced  sample complexity and, conse-
quently, made possible the identification
of minor components  in the  sample.
Furthermore, the chromatographic beha-
vior of several classes of compounds (e.g..

-------
Table 2.
          Results of the Precision and Accuracy Evaluations for the Sludge Protocol Applied to the Analysis of Acidic Compounds in the Spiked NBS
          Sediment
Compound
No.
1
2
3
4
5
6
7
8
9
Compound
Phenol
2-Chlorophenol
2,4-Dimethylphenol
2,4-Dichlorophenol
2.4,6- Trichlorophenol
2, 4-Dinitrophenol
4-Nitrophenol
4, 6-Dinitrocresol
Pentach/orophenol
Unspiked sediment
concentration
(ng/g dry sediment]
1,810, 2.005
ND
ND
ND
ND
ND
ND
ND
ND
% Recovery
Spike level
400 ng/g dry sediment 4,
112, 128
106+ 11
99 + 12
99+ 16
67+8
0
0
0
95+2

Spike level
000 ng/g dry sediment
78 + 21
78+ 18
55 + 3
76 ±5
63+ 10
0
46+11
0
71+3
ND = Not detected.
a - % Recovery determined from triplicate measurements;
Recovery , Spikedsed. cone - unspikedsed. cone
                   Spike added
sulfur containing heterocyclics, aldehydes
keytones) on  silica  gel  helped in  the
assignment of the identifications.
  For example, polar compounds such as
benzofluorenone and methylbenzofluo-
renone (m/z 230) was identified by reverse
library search in Fraction IV; however, no
mass spectral match was found for the
mass spectrum with a parent ion at m/z
244. Assigning a mass spectrum with a
parent ion at m/z 244 methylbenzofluo-
renone was based on two considerations:
similar mass spectral fragmentation  pat-
tern with benzofluorenone and similar
chromatographic behavior on silica gel of
benzofluorenone and methylbenzofluo-
renone.
  The GC/MS chromatograms showed
that most of the compounds of environ-
mental significance elute in Fractions II
and III.  Fraction I contains mostly
saturated hydrocarbons (alkanes) that
were not resolved under the conditions
used. Similarly, Fraction IV contains polar
compounds (naturally occurring materials)
that would interfere in the analysis,  had
the sample fractionation not been performed.

Summary
  Given the complexity of the matrix, the
precision and accuracy results presented
here  demonstrate that the extraction
method, involving homogenization of the
wet sediment with dichloromethane
followed by centrifugation, can be applied
with minor modifications to the analysis
of sediments. Although these experiments
were conducted using selected compounds,
the results indicate that other nonpolar or
weakly  polar  compounds can  also  be
isolated. In fact, the qualitative information
presented here does indicate that a large
number of polycyclic aromatic hydrocar-
bons and sulfur-containing heterocyclics
were isolated from the sediment matrix.
Polar compounds or reactive compounds
containing nitro or epoxy groups were not
amenable to the  analysis procedures
chosen for evaluation.
  Fractionation of  the sediment  extract
containing the base/neutral compounds
by silica gel chromatography and of the
acidic extract by GPC helped reduce the
sample complexity, thus  improving the
quantitation  of the  priority pollutants
without any significant  effect  on the
analyte recovery for many of the compounds.
  Analysis of the NBS sediment extracts
containing the base/neutral compounds
and the acidic compounds was success-
fully  performed  using a  fused silica
capillary column coated with SE-54. The
inertness and excellent resolution of the
capillary column allowed the detection of
nanogram levels of phenols  in  the
presence  of  high  levels of  interfering
coextractants.
  Tentative identifications were provided
for the organic  compounds  extracted for
sediment by the extraction  and analysis
procedures described in this paper. Most
of the compounds identified in the silica
gel  and GPC fractions  were  polycyclic
aromatic hydrocarbons, alkyl substituted
polyaromatics, sulfur-containing  hetero-
cyclics, alkyl substituted  sulfur-contain-
ing  heterocyclics, and several polycyclic
ketones.
   Viorica Lopez-Avila, Raymond V. Northcutt. Jon Onstot, and Margie Wickham are
     with Midwest Research Institute, Kansas City, MO 64110.
   Stephen Billets was the EPA Project Officer (see below for present contact).
   The complete report, entitled "Analysis of the NBS Sediment by the MRI Sludge
     Protocol," (Order No. PB 84-133 750; Cost: $16.0O. subject to change) will be
     available only from:
           National Technical Information Service
           5285 Port Royal Road
           Springfield, VA 22161
           Telephone: 703-487-4650
   For information James E. Longbottom can be contacted at:
           Environmental Monitoring and Support Laboratory
           U.S. Environmental Protection Agency
           Cincinnati, OH 45268

-------

-------

-------

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
      BULK RATE
POSTAGE & FEES PAID
         EPA
   PERMIT No  G-35
Official Business
Penalty for Private Use $300
                                                                                           U.S. GOVERNMENT PRINTING OFFICE: 1984-759-102/874

-------