Document 8VQQDMDDkMOBONJZGXjkZzxvB
A P, A a6_ :0 g6
3M ENVIRONMENTAL LABORATORY REPO R T NO . E06-0199
: S u breport
Analysis of PFOA, PFOS, PFHS, and PFBS in Aqueous Samples : 3M Decatur Offsite Sampling Event (BPP Sample Sites )
Laboratory Request Number. E06-0199
Method Requirement : 3M Method ETS-8-154 .1 (modified)
Testing Laboratory
3M EHS Operation s 3M Environmental Laboratory
Building 2-3E-0 9 935 Bush Avenue St. Paul, MN 55106
Requeste r
Gary Hohenstein 3M Building 42-02-E-27
PO Box 333 1 Saint Paul, MN 55133-3331
Phone : (651) 778-5150 Fax: (651) 778-720 3
The testing reported herein most the requirements of ESOAEC 17025-1999 "General
aorla Requirements for th e Competence of Testing and Calibration I.Aboratorbs^, in accordance with
ACCREDI ED
OW A21LA Certificate 02052-0 1. Testing that complies with this International Standard also
operate In accordance with ISO 900111180 9002 (19$4 ~
Cert ificate #2052 -0 1
PAGE 1 OF 21
p. 2
3M ENVIRONMENTAL LABORATOR Y REPORT NO . E06-01 99
3tsfi Environmental Laboratory 3M Environmental Laboratory Managc:r : V4liizm K. Reagen, Ph .D . 3M Technical Lead : Michelle Nlalinsky, Ph .[) . Report Author: Michelle Malinsky, Ph .D .
Analytical Subreport B96-0999
Surface Water Samples BPPO1, BPP02, and BPPO3; Ground Water Sample BPWeII
Decatur Offsite Sampling Event Report Date* May 1, 200 6
9 tntr,aductidNSummarX- _
.,. .
The 3M Environmental Laboratory extracted and analyzed ground water and surface water samples coflected by Weston Solutions personnel on April 12 and 13, 203G near the 3M Decatur facility . Samples were returned to the 3M Envir-)nment~i Laboratory for analysis of perfluorooctanoate (PFOA), perfluorooctane sulfonate (PFOS), perfluorohexane sulfonate (!'FH`') and perfluorobutane sulfonate
(PFBS) under laboratory project number E06-0199 using modified 3M Environmental Laboratory Method ETS 8-154 .1 "Determination of Perfluorinated Acids, Alcohols, Amides, and Sulfonates in Water by Solid Phase Extraction and High Performance Liquid Chromatography/Mass Spectrometry' .
Per the client's request, an initial subrepert is being issued for the `roliowing four sample locations only :
DAL SWS BPPO 1 DAL SWS BPPO2 DAL SJVS BPPO 3 DAL. GWS BPWELL
This report will contain information relevant to these four sample collection points only . These results will be reissued in a comprehensive final project report that includes results for all samples collected under E06-0199 .
The 3M Environmental Laboratory prepared sets of sample contairterc . for thirty-tNe separate sampling
locations. At a minimum, each sample sat consisted of a field sample, field sample duplicate, low field spike and high field spike . For some locations, abird mid-levz; field spike, was prepared . Each empty container was marked with a"fill to here" line which corresponded to a final volume of 450 mL . Containers reserved for field matrix spikes were fortified with an appropriate matrix spike solution containing the four target anatytes prior to being sail `.a the field for sample collection . The spike amounts prepared for each, of the four locations presented here vvili be provided later in this report .
jEflw ACCREDITEO
Certificate #2052-01
The testing reported herein meet the requirements of ISO/lEC 17025-1999 "General Requirements for the Competence of Testing and Calibration Laboratories", in accxsre!ance with the P21-4 Certificate #2052-01 . Testing that complies with this Iniemational Standard also operate in accordance with ISO
PAGE 2 OF 21
p. 3
3M ENVIRONMENTAL LABORATOR Y REPORT NO. EO 6-0199
Samples were initially extracted and analyzed on April 19, 2006 . Samples were diluted,, reextracted, and analyzed on April 21, 2006 . Laboratory matrix spikes were .prepared, diluted, extracted, and analyzed on April 25, 2006 in the analytical batch initiated on April 24, 2Q06 : ,
Table 1 below summarizes the sample results . All results for quality control samples prepared and analyzed with the samples will be reported and discussed elsewhere in this report . All samples were diluted 1 :50 in ASTM Type I water prior to extraction .
Table 1 . Sample Results Summary . -
3M L W ID
Lamle Desc "
roPfOA APFOS 'VPFHS
19PFBS
Coneenlra0ior! ConcenbaOion ConcentrrBon Concenba tion
tion ml mL
E06-0199-055 DAL SWS BPPOI 0 060413 293 14>423 53 .3 . 6.98
E06-0199-056 DAL SWS BPPO1 DB 060413 307
'4>380 54 .3 7 .1 7
Avqrage 300
(4>402 , 53'8 708
%.RPD $am e/5am d Dup r ' 4.7 ~ NA
1.8 27
E06-0199-060 DAL SWS BPPO2 0 060413, 206 t-0>295 362 430
E06-0199-061 DAL SINS BPP02 DB 060413 219 1>314 39 .6 4 .7 2
,AyeiaQe
212 M>304
39.0
4. 68
SSRPf7 Sam e Dup,
6.1 NA 4.1
2. 6
E06-0199-065 DAL SW5 BPP03 0 060413 219 '4>772 39 .1 4.69
E06-0 199- 066 DAL SWS BPPO3 08060413
214
r4>256
40.1
5.0 5
Average 216 >264 39.6 ~ 4.87
%RPD Sam elSam 2.3 NA 2.5 7. 4
E06-0199-070 DAL GWS BPWELL 0 060413 1=1>774 s>1460 17k>145 17.2
E06-0199-071 DAL GWS BPWELL DB 060413 ">779
'4>1170 ta>145 17.6
Averape
M >778 '*>1320
0 >145
17.4
~___ ____ YeRPD Sanp/d~aritple'Dup NA
NA
NA
2.3
(1) The analytical method uncertainty for PFOA, PFHS, and PFBS was 100t11,6,100t5 .8%, and 100t6 .8% , respedively based on method accuracy and precision. Sel''SeQion 3.7 for additional ezplanation. The limit of quantka tion (LOO) for these samples only was 250 nghn . . (PFOA) and 1 .25 nglmt. (PFHS and PFBS). Results and average values are rounded to three significant figures according to EPA rounding rules. Pe rcent relative difference (%RPO) values are rounded to two significant fpurds. Because of ro urding, .vakies may vary slightly from tttose listed'in the raw d hta:
(2)
All PFOS sample results preserited herd should be considered estimated minimum values as quality
contrat samples (field matrix,s,pikes and lab mabfir spikgs) associated witl Moss four samples did not
demonstrate adequate method perfonmvice. The accuracy of the PFOS results could be Within an order
of magnitude. 9LRPD values not calculated. BPWeN results for PFOA and PFHS should afso be
consideeed estimated minimums as field matnbr spike recoveries did notmeet method acceptance criteria
of 100.t30%. The analytical accuracy of the PFOA and PFHS estimated values for this sample only could
be within an order of magnidude. B:RPD values. ,not calculated.
., . , , ,. I
.
,..
2 Mefhoals ~ Analyflca and l~paratorjr'` r - 4r`
r;F . , ~' . . :i ft, : .zv . . ... . .
2 .1 Sample Collectio n
Samples were collected in NalgeneT"' (low-density polyethylene) bottles prepared at the 3M Environmental Laboratory . Prior to sample collection, bottles designated for field matrix spikes were spiked in the laboratory with a known volume of an appropriate matrix spiking solution containing th e
PAGE 3 OF 21
p. 4
3M ENVIRONMENTAL LABORATOR Y REPORT NO . E06-0199
f61r anaiytea of interest (PFOA ; PFOS, P~:HS and PFBS) . Collected sample bottles were received at the Laboratory at ambient conditions nr a,pri! 19, 2W6 . Table 2 tx4,YN details the samples collected and
spikes added to each hottle
Table 2. Sample Ca1Jer.tion and Spike inVarmation .
38t VMS 1D!1
Weston Sample Desc don
31W BoWe
Number
Bo se Deswwdon
Nominat Spike GPO Locedon Level (no(mL) Comment
E06-0199-055 DAL S'dVS BPP010 060413 109 Sample NA FSIA Drainage
E06-0199-056 DAL SWS BPPO1 DB D60a13 110 Sample Dup NA FSIA Drainage
E05-0199-^57 ; DAL SINS BPF^t LS DnC4113 11 FMS Low 1,11 FSIA Drainage
E06-0199-058 !;AL SVU8 BPPOt MS ^60413 '11.2 FMS Mid 10 FSiA Drainage
E06-0199-059 DAL SWS BPP O 1 HS 060413 113 FMS High
100 FSIA Drainage
E06-0199-060 DAL SWS BPPO2 0 060413 149 Sample NA FSIA Drainage
E06-0199-061 DAL SWS BPPO2 DE ; 060413 120 Sample Dup NA FSIA Drainage E06-0199-062 DAL SWS BPPO2 LS 0"13 121 FMS Low 1 .0 FSIA Drainage
E06-0199-063 DAL SWS BPPO2 MS 060413 122 FMS Mid 10 FSlA Drainage
E06-0199-064 DAL SWS BPPO2 HE 060413 123 FMS H;.Aah 100 FSiA Drainage
E06-0199-065 DAL SAS BPPO3 0 060413 114 Sample NA FSIA Drainage
E06-0199-066 DAL SWS BPPO3 DB 060413 115 Sample [AV NA FSIA Drainage
E06-0199-067 DAL SWS BPPO3 LS 060413 ~ 116 FMS Low
1 .0 FSIA Drainage
E06-0199()68 DAL SWS BPPO3 MS 060413 117 FNS Mid 10 FSIA Drainage
E06-0199-069 DAL SWS BPPO 3 HS 060413 ' 13 FMS High
100 FS1A Drain
E06-0199-070 DAL GWS BPWELL 0 060412 72 Sample NA Br- Property
E06-0199-071 DAL GWS BPWELL DB 060413 73 Sample Dup NA BP Property
E06-0199-072 DAL GWS BPWELL LS 060413 74 FMS Low 10 BP Property
E06-0199-073 DAL GWS E?PWELL MS 060417 75 FMS M41 100 BP Prc+Perty
E06-0199-074 DAL GWS BPV4ELL KS 060413 73 FMS High
DAL = Decatur, A L SWS = Surface Water Sample
GWS = Ground Water Sam* 0 = Sample DB = Duplicate LS = Low Spike
MS = Mid Spike HS = High Spike
1000 SP Property
2.2 Extraction
All samples, calibration standards, and associated quality control samples were extracted using a modified procedure of ETS-8-1 54.1 "Determination of Perfluorinated Acids, Alcohols, Amides, and
Sulfonates in Water by Solid Phase Extraction and High Performance Liquid ChromatographyMlass Spectrometry". Briefly, 40 mL of sample were loaded onto a pre-conditioned Waters C18 solid phase
extraction (SPE) cartridge (Sep-Pak, 1 .0 g, 6 cc) using a vacuum manifold . The loaded SPE cartridges were then eluted with 5 mL of methanol . This extraction procedure concentrates the samples by a factor of eight. (initial volume = 40 mL, final volume = 5 mL) . Lab control spikes extracted in the same
manner cross-validate all the method modificationsJdeviations from ETS-8-154 .1 . See Section 3 .6 for additional information.
PAGE 4OF21
p. 5
3M ENVIRONMENTAL LABORATOR Y REPORT NO . E06-0199
The samples and sample duplicates for BPPO1, BPPO2, and BPPO3 were initially extracted undiluted on April 19, 2006. Initial analysis of theundNute.d samOles exceeded the upper limit vf quantiation (25 ng/mL nominal) for PFOA and PFOS . Likewise, the sample and sample duplicate for BPWELL were initially extracted on April 19, 2006 . The final extracts were diluted 1 :10 prior to analysis . Analysis of these diluted extracts also exceeded the upper limitofquantitation . :Ali the samples, sample duplicates, and field matrix spikes were reextracted on April 21, 2006 . For BPPO1, BPPO2, and BPPO3, the samples and associated QC were diluted 1 :50 using ASTM Type I water prior to extraction. For
BspPikWeEwLaLs, dthileutseadm1p:l5e0,0s.ample duplicate,,4ow spike and mid spike were diluted 1 :50, while the hig h
2 .3 Analysi s All sample and quality control extracts were analyzed for PFOA, PFOS, PFHS, and PFBS using high performance liquid chromatography/ tandem mass spectrometry (HPLC1MSMtS). 'Pertinent insftment parameters, the liquid chromatography gradient program, and the specific mass transitions analyzed are described in the tables below . Table 3 . Instrument Parameters .
Instrument Name ETSStan
_!,quid Ch rs ilent 1100
Guard column
Betasil C18 2.1 mm .X100 mm), 5 m
Ana cal column Betasil C18 2.1 mm X 100 mm , 5
In Volume 5 L
Mass
-ion Source
-Electrode
Appfied Biosystems API 4000 Turbo Spray
Polarity five
Software
Ana 1 .4 .1
Table 4. Liquid Chromatography Gradient Program.
Step Number
Toto! Time Flow Rate Pement A
(min)
&Ah+u+)
mM smmoneum acehte
Percent a
0 0
1 1 .0 2 14 .5
3 15 .5
4 5
18.5 20 .0
300 80.0
300 300 300 300 300
80.0 10 .0 10 .0
80.0 80.0
20 .0 20 .0 90.0 90.0 20 .0 20 .0
PAGE 5 OF 21
p. 6 3M ENVIRONMENTAL LABORATOR Y
REPORT NO . E06-0199
Table S . Mass Transitions .
Finsgte
Mats Transition OweU Time (rtr,sec) Q11Q3
t''PFBS 299/80 125
299KJ9 125
"'PFHS 399180
125
3p3/99 125
""PFOS 49/13 0 1 25 499/99 125
499 I80 12 5
"}i'FGA
4 1Sl369_ 413P119 413/1ti9
125 125 ~ 1 2a
(1) The nxfivalual GaU~jc,ns were surtared to produoe a"total ior, dxomatogram" (T I:,) . The TICs were usea for quantitation .
. ..
,, . , ,,,.; . .
3 ; LiataAnalysls :
.1 Calibration
~ 3
Calibration standards were prepared by spiking known amounts of stock solutions containing the target
anaiytes into 40 rtL of ASTM type I water, Each spiked water standard was then extracted in the same
manner as the, =l;ected samples . A total of twelve spiked slandards r3nging from 0 .025 ng/mL to 25
ng1mL .(r.ominat) were prepared . A quadratic, 11x weig h. led, calibration curve was used to fit the data
for each analyte . The data were not forced trough zero during the fitting process, Calculating the
standard concentration using the peak area counts, and the resultant calibration curve confirmed
accuracy of each curve point. Each extracted calibration standard used to generate the final calibration
curve met the method calibration accuracy requirement of 10025% . The correlation coefficients (r)
were greater than 0.999 for all analytes .
3.2 Limit of QtJantitation . (LOQ)
The LOQ for this analysis, as defined :n ETS-8-154 .1, is the lowest nurr-zero calibration standard in the curve in which the area counts are at ieast twice those of the method and solvent blark(s) . The 1-00s
for PFOA, PFOS, PFHS, and PF3S were 0 .O5OC nglrnl ., 0 .0999, ng/mL, 0.0250 ng/mL, and 0 .0250
ng(mL, respectively . Note: these LnQ values have not been corrected for sample specific dilution factors . Area count comparison for the method blanks and the lowest calibration standard will be
provided in Section 3 .5 .1 .
3.3 System Suitability
..
Replicate injections of the 1 ng/mL extractec'-calibration standard were analyzed at the beginning and
end of the analytical sequence to demonstrate overell system suitability . All analytes met method
acceptance criteria of less than 5% relativc- standard deviation (RSD) for peak area and less than 2%
RSD for retention time for both opening and closing system suitability injections .
3.4 Continuing Calibratio n
During the course of the anayiical sequence, several continuing calibration verification samples (CCVs) were analyzed to confirm that the instrument response and the initial calibration curve were still
in control . One PFOS CCV (139l0) and one PFHS CCV (153%) exceeded method acceptance criteria of 10025% for accuracy . Both of these non-compliant CCVs followed a contaminated solvent blank and the high recovery is most likely carryover from the previous injection . A method deviation has bee n
PAGE 60F21
p. 7
3M ENVIRONMENTAL LABORATOR Y REPORT NO . E06-0199
issued. The overall impact to the data quality objectives for the sample results is considered minima l
as all sample injections and sample related QC injections were bracketed by compliant CCVs .
3.5 Blank s Three types of blanks were prepared and analyzed with the samples : method banks, solvent blanks, and field/trip blanks . Each blank type is described below .
35.1 Method Blanks
Several method blanks were prepared by loading 40 mL of ASTM Type i water onto a C18 SPE cartridge and eluting with 5 mL of methanol using the same extraction procedure as the samples . Method blanks were prepared to evaluate the levels of background contamination in the overall extraction process (reagent water, glassware, SPE cartridges, etc .) The .calibration curve was prepared on a separate day as the samples . (Calibration standards: April 10, 2006 ; diluted samples : April 21, 2006) . The method blanks prepared with the calibration standards as well as with the samples were both analyzed during the April 21, 2006 analysis . Additional lab matrix spikes were prepared and analyzed in a separate analytical run initiated on Apri l 24, 2006 . Again, the method blanks prepared with the calibration curve and the method blanks prepared with the spikes were analyzed . Method blank area count comparisons to the LOQ standard are provided in the Attachments section at the end of this report .
3.5.2 Solvent Blanks
Several methanol solvent blanks were analyzed to assess system contamination andlor instrument canyover. For the April 21, 2006 anaysis, three separate vials of methanol solvent blanks were ~ contaminated and produced area counts of the target analytes greater than,one-hatf those of the LOQ standard . A method deviation has been issued . The solvent blank area count comparison for the April 21, 2006 is provided in the Attachments section at the end of this report . For the lab matrix spike
ril 24, 2006 , only the initial solvent blanks injected before the opening systerr, sanlyzedoAp suitabilities ar,a the solvent blanks analyzed after Ire highest calibration standard exhibited area counts greater than one-half the LOQ standard .
3.5 .3 Fieki/Trip Blanks
Prior to sample collection, seven sets of trip/field blanks were prepared : A trip blank consists of a sample container filled with 450 mL of ASTM Type I water, sealed, and shipped to the sample collection site along with the empty containers . The trip blanic serves as an additional method blank that accounts for any storage conditions and/or holding time issues that the samples may experience . Results of the trip blank analyses will te provided in the final comprehensive report .
3.6 Lab Control Spikes (LCSs )
Lab control spikes at nominal concentrations of 0 .050 ng/mL, 5 .0 ng/mL, 100 ng/mL, and 1000 ng/mL spikes were prepared and analyzed each day samples were extracted and analyzed . The 0 .050 ng/mL and 5 .0 ng/rnL LCSs were prepared by spiking Ynovm zmounts of the analytes into 40 mL of ASTM Type I water to produce the desired concentration . The spiked water samples were then extracted and analyzed in the same manner as the samples, . For the 100 ngJmL and 1000 ng/mL LCSs, target analytes were added to 450 mL of ASTM Type I water to produce the final desired concentration . An aliquot of the LCS was then diluted 1 :50 or 1 :500 with ASTM Type I water prior to extraction, in the same manner as the diluted samples. This technique was done to demonstrate that the dilution procedure used for the samples was able to produce accurate results . Analysis of replicate LCSs at the various levels cross-validates the analytical method, as used here, for any modifications/deviations from ETS &-154 .1 . Additionally, LCS results will be used to determine overall method uncertainty in Section 3 .7. Table 6 summarizes the LCS recrvery results .
PAGE 7 OF 21
p. 8 3M ENVIRONMENTAL LABORA TOR Y
REPORT NO. E06-0199 LCS Percent RCaelcuclaotevd Ceornycen=trati1on0,0%
Spike Conce ntrat io n standard deviation LCS replicates 100% LCS% RSD =
average LCS recovery
PAGE 8 OF 21
p. 9
~O Op ~0
c`~ Of h Ol t"7 uY c+7 P7 t~ tD O V;
ch N C 1 C7 M Of tD
rn
S to 'coo ~ap0tOf Q ~} 1 CAO Ocffp O N1 OO n$ f
o
r~
n
ui ~p T Q)
~p r~ Q/ T O/
m ~J
s~
p
Q~
~p
cp
v UM, O
O'. P~~ V O Y
Q` C O O O O O O
Qf r, O~ a+ O~i O K V . oO
CS Qf Q1
WLU
0
31
O1 Oyf OS O~i Q~ O Of Of Qf
~~ S O ~~t ~t Y Y V
V
O
C
O
O
plj OQif pp~~ pp~~pp~~ Of Oa~ [71 ~ p~ W C
~1' 01 Of O~ O~ C~ O~ Q Od+ d'
O
O
OO
Opli pOy/ Of Qf
pp~~ pQ~~ Qf
w o so ~o co o $ a m ~n o o .- 4 v o ao o ao w c~ rn m oo <a CO C, no gn' a rn rn rn rn O0rn rn rn
cp pp
m
al C O 01 OJ O O Y V Qf Q~ Ol C G < tA O~ Q! O
~~ssssgs~~~~ssss~~ W U O O O N 1n [j O C U/., 0 1A " - .- - e- in tif r' r' Q!
O
O
O
G
O
O
O
O
~
n~p O) v cp~~s ln c0 pO O~~ Orn Q~prn yspw 07 O
.~ Z Z Z n O ,~ Z_Z np r
~
- C E ^
a0 000 cr-O 47 000 Ql O~i F nL
Q~i Qs aD MW 'CIO,
Q.'
O
u~ o
989 v
WO , ~8 ~99~<O .:~~ n~O ~~j ~N ~o
r;
~ Z O 1~ O Z O O N V O h CO n t~f ~ .~j O tY M C~ tn t~ N
CD
rn rn rn CD o,
O ~ 4`
tf7 aD 1'1 ~ ~ iA O a7 O t\ Q~ ~0 n c'! O
n t7 tD ~.. ~y
Q ~ VU c s 8 o d v ~ri e~J a~i v< rn g'i `8i NCh f o o c v rn CD C D
ooo~~~ o ao ~- 6-6 - - - m
C6
rrr
N 7
N N ~ N N N~ N~ N ~ ~ N N N d
;Q ~ J J J J J .J J J J J~ JJJ J J J~ U J J ...1 J J J J J
(a J J c~
2- 1- 12J
-01
N cJ c) J c~.~ `Jt.J
2
c.~)
J
JJ
a
a~ $ a a$$$ S n a~ n c~i~ S~~ ~~ $
$
N a o 0 0 0 0 0`~a', g o 0 o 8 g g$$$`b' `d c g g~$ }- ~ o co o I n on un 0 o 0 U7 LO o ri ,- .- .- .- .- o 0 ui
P . 10
CZ,
~ SAQ
0m o
1a
0 ~~~ LU Uj 32
ul
E
ZZ g ~
a ~ ~ Lp Cp O
LL
Q
:P :13
O
~Qi
01
w
N
C)
~
l9
O 18 C W
0~8 ~ oN o o ~.
72
El c O ~s &
ti ~ a nn
0 toon: 4
P~ '.!. g
P . 11
3M ENVIRONMENTAL LABORATOR Y REPORT NO . E08-0199
3.7 Analytical Method Uncertainty
Both the accuracy (percent recovery) and precision (%RSD) of the lab control spikes were used to estimate the overall method's analytical uncertainty for a given analyte . For example, the overall accuracy and precision for PFOA based on LCS results was 101 0/9 .7to . The measured precision (%RSD) is then used to determine the range of the accuracy .
Example : 101`(0.097) = 9 .84 101+9 .84=111 .1 ; 101 -9.84=91 .45
Thus, LCS accuracy results range from 91 .45% to 111 .1 h . The absolute difference of the low and high ends of this range, when compared to 100Jo, are then calculated .
111 .1%-100= 11 .1% 100%-91 .45 = 8 .54% .
The most conservative (largest) absolute difference is then used as the analytical uncertainty for the given analyte. Therefore, the analytical uncertainty for PFOA is given as 100t11% (two significant f1ig0u0ret5s).8f%or,tahnedse10re0su8lt.s8%. F,orresPpFeOctSiv,ePlyF.HISt s, haonudldPbFeBSem, tphheaasnizaelydtitchaaltmtheethaondaulynticcearltaminettyhoisd10016 % , uncertainty described above only applies to the method and the procedures perforrrted within . Additional quality control samples specific to the sample matrix, in the form of field matrix spikes and lab matrix spikes, demonstrate that the individual sample results are accurate to within 100300/o, the targeted data quality objectives .
3.8 Field Matrix Spikes (FMS )
At a minimum, low and high field matrix spikes were collected at each sampling point to verify that the analytical method is applicable to the collected matrix. For some locations, a mid-level spike was also collected. Field matrix spike recoveries within method acceptance criteria of 10030% confirm that .unknown" components in the sample matrix do not interfere with the extraction and analysis of the analytes of interest. Field matrix spikes will be presented in the next section with the sample data . The field matrix spike should be at least 50% of the endogenous sample concentration to be considered an appropriate spike level for the given sample matrix .
FMS Recovery = (Sample Concentration of FMS- Average Concentration : Field Sample & Field Sample Dup .) . Spike Concentraton 100 %
3 .9 Lab Matrix Spikes (LMS )
If the sample and sample duplicate produced concentrations where none of the associated field matrix spikes were at least 50% of the endogenous concentration, then laboratory matrix spikes (LMS) were prepared. LMSs were prepared by spiking a known amount of target analytes into a measured aliquot of the sample matrix. The spiked matrix was then diluted, if necessary, and extracted in the same manner as the samples . LMS recoveries within 10030% demonstrated that the analytical method
was appropriate for the given. s, am. ple matrix only in the absence of an appropriate field matrix spike . 4 : t7afa Summary~ e
The data tables below summarize the sample results, field matrix spike recoveries, and laboratory matrix spike recoveries for the four locations (BPP O 1, BPPO2, BPPO 3, and BPWeII) . Each table provides the average concentration and the relative percent difference (RPD) of the sample and sample duplicate .
PAGE 11 OF 21
p . 12
3M ENVIRONMENTAL LABORATOR Y REPORT NO E06-0199
BPPO 1 The endogenous concentration for BPPO1 was at least twice the spike concentration for the low spike (1 ng/mL), mid spike (10 ng/mL) and high spike (100 ng/mL) for both PFOA and PFOS . A laboratory matrix spike at 500 ng/mL was prepared and analyzed. The LMS recovery met method acceptance criteria of 10030% for PFOA (79 .6%), but not for PFOS (60.6%) . Therefore, the sample results for PFOS should be considened estimated minimums with a sample analytical accuracy within an order of magnitude as the analytical method has not been demons trated to be suitable for the given sample matrix . FMS andlor LMS recoveries within 10030% for PFOA, PFHS, and PFBS demonstrate that the anaiytical method is appropriate for the given sample matrix for these analytes .
BPP O 2 The endogenous concentration for BPPO2 was at least twice the spike concentration for the low spike (1 ng/mL) and mid spike (10 ng/mL) for both PFOA and PFOS as well as the high spike (100 ng/mL) for PFOS only . A laboratory matrix spike at 250 ng/snL was prepared and analyzed . The LMS recovery met method acceptance criteria of 10030% for PFOA (75 .0%), but not PFOS (59.8%) . Therefore, the sample results for PFOS should be considered estimated minimums with a sample analytical accuracy within an order of magnitude as the analytical method has not been demonstrated to be suitable for the given sample matri x. FMS and/or LMS recoveries within 10030% for PFOA, PFHS, and PFBS demonstrate that the analytical method is appropriate for the given sample matrix for these analytes .
BPP O 3 The endogenous concentration for BPPO3 was at !east twice tne spike concentration for the !ow spike (1 ng/mL) and mid spike (10 ng/mL) for both PFOA and PFOS as well as the high spike (100 ng/mL) for PFOS only . A laboratory matrix spike at 250 ng/mL was prepared and analyzed . The LMS recovery met method acceptance criteria of 10030% for PFOA (73 .8%), but not PFOS (60 .8%). Therefore, the sample results for PFOS should be considered estimated minimums with a sample analytical accuracy within an order of magnitude as the analytical method has not been demonstrated to be suitable for th e given sample matrix. The high level PFBS FMS exceeded method acceptance criteria of 10030% (134%) . However, the PFBS results for this sample are still considered accurate within 10030% because the criteria exceedence was minimal (within 4%) and mid level FMS and LMS recoveries were excellent . 101 % and 90 .2%, respectively . High level FMS and LMS recoveries within 10030% for PFOA and PFHS also demonstrate that the anaiytical method is appropriate for the given sample matrix for these two analytes as well .
BPWeIi The endogenous concentration for BPWeII was at least twice the spike concentration for the low spike (10 ng/mL) for PFOA, PFOS, PHFS, and PFBS as well as the mid spike for PFOA, PFOS, and PFHS . Although the high level spike (1000 ngJmL)was an appropriate level for all four analytes, the recoveries were consistently less than 30% for all analytes, with PFOS showing no measuring recovery . The high FMS was reextracted to rule out a potential preparatory error during the extraction process . The second extraction produced similar recoveries as the first . A laboratory matrix spike at 1000 ng/mL was also prepared . The LMS recovery met method acceptance criteria of 10030% for PFOA (85 .4%), PFHS(93 .5%), and PFBS(97 .2l0) but not for PFOS (13 .5%). Therefore, the sample results for PFOS should be considered estimated minimums with a sample analytical accuracy within an order of magnitude as the analytical method has not been demonstrated to be suitable for the given sample mat rix . Although the LMS produces acceptable recoveries for PFOA and PFHS, the sample results should be considered esti mated minimums as no app ropriate supplementary FMS recovery was available. Low and mid level FMS and LMS recoveries within 10030% for PFBS demonstrate that the analytical method is appropriate for the given sample matrix for this analyte only. Further method development for the BP Well sample matrix will be performed to see if improved PFOA, PFOS, and PFHS recoveries can be achieved .
PAGE 12 OF 21
p . 13
gw m
.~o
a a 4 h ia
~ V C tD f~ f~ W~ o t~2~j5
g g 4.' 2' N N m
.,
~ ri m v, UD
w~
~p
u~i
V41~1
~
s'
d
la
O_3
e~-
W
!`'~ E r
o = ~
4 C N [") N .O !h t0
u-~0
. M M, M M p~ ~~~~
w o O S a
IL a.
a v~ cn u~ u~ v~ ~n
5 e ~~
.
m
~p
to
-- Q .~CI ~t'f ~~f1 ' N Z h ~~ J ~
~ ~~~~~ v` ~ ~
LU uj
p . 14
>m
o M a $ m
~b
.
00
~
N
"~ ~ ~~``j tp N Of f N 7 Q' !6
a CC Y .ra' ~fi
N
O
,q Q
N pp tD M O ~p a ~Q Y ~ C C f") en V O f`/ Nn
e[
0
\
O ~ E Cv `~ ~iTry
Z
cT
~
C', ~~~~
~
w
a
M'c L
z C4 < N 0
CL
E
a
el
C ~Op
0~y a pp
~ Vd N N~ N M'7
.~
l~! IL
El ~SY?
~ Q }m3 ~ ~
a 1
_9
M
o J~ S a~ ~~LL o~O NNNNN N
1=A m m[0 m m os a0 E
Ra
2~
C.
op =o
. co
ro ,~,~-~z ~, j a
~ WWW
oo ~ $$? ~
co Z
4
~
~
o
u
o ~
. kn W, C4
~
a Ix
w a $~~ o cc
N O
N
A gS 9 ~N
~7Yw
mmm CO z a ez a*: ~s~ .2
N N N ~~ N CO ~
a~
en
~ 9 1Ep .4t
W o~
C, LD
N N N
-p
G ~9t
E O
~ ~~~$~
~2 I I I I
jfl h
0 0 J ~ S o ~ L
.
~{
a
a
a ~sa ~~a ~
a~
a p
E~
mmmmmm
co V)
a~~ ~
o IL
z11 cn0 z~ a
F~ W W W W W J~ Z '
p. 15
U. 0
p . 16
In W tq
t o ~~~ ~ E
F-- O W W ~ ti
g~
.V rD T y Tr i~D~ N 14 4 M )
u~
~o
o
>
(
W I Z Z 7Q aN { ~ o C'%
ro
C.)
^~LLL_
LO LO
C
C-4 ~_
C
oc V
r m
~p
3
X
? Z Z Z
co
L
f
V
Ix
in
LL
m
0
w
d`
U"i o p 0 p o p
"
~A
~
df O "- r p
r
~i ~
en C.
S
E L~L
r
$ y0
? z co
J
y} y}~'~
N tp
o
0
M _
Y }a{
M TM
q~j
M 'D
CD (p ~" V) (J7 O
4~ ~p
J J J J J J J t~
I
E t0 (/) ,
q gc m m co in co co in E
ooo o ~
n^
oH
Q _ w it
a 2lp cnIt ~ cn ~ O Y LL Z LL t1. LL Z J d tp
a yr : " n n
;,::,
~ W W W W 9~ _ v~j a Z
p . 17
3M ENVIRONMENTAL LABORATOR Y REPORT NO E06-0199
5 Conclusio n
Sample matrix QC samples for the four samples reported herein did not meet method acceptance criteria of 100t30% for PFOS . Therefore, th e values-fisted in Table 1. and Tables 7-10 for PFOS should be considered estimated minimums. The sample analytical accuracy is arbitrarily estimated to be within an order of magnitude. Further:method development is needed for this matrix as the QC results demonstrated that unknown matrix components preclude accurate analysis with the method as performed here. Field matrix spikes yielding recoveries within 10030% demonstrated that the analytical method was suitable for the given sample matrices for PFOA, PFHS, and PFBS for BPPO1, BPPO2, and BPPO3 . The analytical method uncertainty for these analytes was presented in Tablet PFOA and PFHS results for BPWeN should also be considered estimated minimums with a sample analytical accuracy within an order of magnitude as no acceptable field matrix spikes recoveries were produced.
8 _. . : , . . .
~ DataI Sample Retention
All remaining sample and associated project data (hardcapy and electronic) will be archived according to 3M Environmental Laboratory standard operating procedures .
PAGE 17OF21
p . 18
3M ENVIRONMENTAL LABORATOR Y REPORT NO . E060 1 99
7 Signatures
~..
Mich Ile D . Malinsky, Plf.`D ., M Technica9ead Date
Kent R . Lin dstrom, 3M Technical Reviewer
Date
e~ ~~`~~-%?~-'
C! S Ci,~~
William K. Reagen, Ph .D ., Environmental Latrxzto;y M anagem ent Date
The 3M Environmental Laborato ry's Quali ty Assurance Unit has audited the data and report for this project.
r n
Quality As ance Representative
-flb
Dat e
3M CONFIDENTIAL.
PAGE 1 8 OF 21
P . 19
3M ENVIRONMENTAL LABOR4 TOR Y REPORT NO. E06-0199
8 ' Attachments ,- "
.~
8.1 Method Blank/LOQ Area Count Comparison .
Table 11 . Method BianklLOQ Area Count Comparison : April 21, 2006 .
An*ftal Run: Apr# 21 2006
DataBfe
Sample
S le Nan,
PFOA
Area Counts
PFOS
PFHS
PFBS
Method Blanks Pn pared wfth the Extuded Curve s060421a025 Method Blank-1 MB-0604101 55982 3413 1591 2760 s060421 a026 Method Blank-2 MB-060410-2 18376 1082 668 2001
s060421a027 Method Blank-3 MB-O604103 16127 1718 1970 2336 s060421a028 Method Blank-4 MB-060410-4 22714 1116 639 1305
s060421 a029 Method Blank-5 MB-060410-5 21587 792 1686 2109 s060421 a030 Method Blank-6 MB-060410-6 25928 4338 2077 5711
s060421a031 Method Blank-7 MB-060410-7 21060 2143 2436 3561
s060421 a032 Method Blank-8 MB-06041 M
16383 1776 1660 4235
Method BJwks
with the Exbacted Saawl"
s060421a045 Method Blank-1 MB-060421-1 12933 1437 "1121128 277
s060421a046 Method Blank-2 MB-060421-2 12809 1446 " )129379 715
s060421447 Method Blank-3 MB-060421-3 13875 2228 " 66834 595
s060421a048 Method Blank-4 MB-060421-4 22242 2053
1~42902 1052
s060421a049 Method Blank-5 MB-060421-5 19675 1762 " 1 33923 826
5060421450 Method Blank-6 MB-060421-6 17774 1568
1113105
2071
s060421a051 Method Blank-7 MB-060421-7 13843 1084 6059 J19419 s060421a052 Method Blank-8 MB -060421-8 14073 1492 5570 450
Area Counts of the LOQ standard 163373 14600 13176 25766
Area Counts of the LOQ standard'O .ti 81686.5 7300 6588 12883
Concentration of LOQ standard n L 0 .0500 0 .0999 0 .025 0.025 (1) Method blanks injected immediately after a highly contaminated solvent blank . Carryover present . Values excluded when establishing the LOQ for PFHS .
(2) Method blank determined to be a statistical outlier at the 99% confidence level using Dixon's Q-test . Value excluded when establishing the LOO for PFBS .
PAGE 19 OF 21
p . 20
3M ENVIRONMENTAL LABORATOR Y REPORT NO. E06-0199
Table 12 . Method BlanklLOG1 Area Count Comparison :! . April 24, 2006 .
Datafile Run:
24, 2006 S Deec /e Heme
Area Counts
PFOA
PFOS PFHS PFBS
Method Blanks
with the Ex tracted Curve
s060424a025 Method Blank-1 MB-06000-1 "~36625 3514 2029 3078
s060424426 Method Blank-2 MB-060410-2 11)42239 2274 2890 3938
s060424027 Method Blank-3 M8-060410-3 25302 2950 1147 2248
060424428 Method Blank-4 MB-0604104 22151 804 1197 1638
s060424a029 Method Blank-5 MB-060410-5 30401 2565 1955 2593
s060424a03O Method Blank-6 MB-060410-6 17458 827 1275 5242
s060424aU31 Method dianc-7 MB-060410-7 2's074 1195 1974 3516 s060424a032 Method Blank-8 MB-06N1 9-8 22049 1192 1512 4294 Method Blanks w i t h the E x tr a c t e d Laboratory 8felrix S O&" s060424090 Method Eiank-1 MB-395: 425-1 30448 2301 1778 1906 s060424a091 Method Blank-2 MB-060425-2 15006 1266 1311 2990
s060424a092 Method Blank-3 MB-060425-3 11262 991 1039 1703
s060424aU93 Method Blank-4 MB-060425-4 11867 836 707 6142
s060424a094 Method BtanK-5 MB-060425-5 16236 1024 717 2723
s060424a095 Method Blank-6 M13-060425-6 15924 1107 589 5446
s0604~.4a0'3i Method B1a1k-7 MB-050425-7 22753 1530 1038 435E
s060424a097 Method Blank -8 Area Counts of the LOQ standard Area Counts of the LOQ standard'0 .5 Concentration of LOQ standard n
MB-06L 425-8 1928,i 1071 1135 1677
36G60 6724 12166 27032
L
18030 3362 6083 13516 0 .0-1 5 0 .04 0 .025 0,025
(1) Value excluded for the LOO determination for the given set . Area counts are very close to one-half the LOC value . Prior wanealllyasbiosvoef tthheisLeOxtOra.ct demonstrated area courts below one-half the 25 ppt standard . LMSs analyzed on this day are
PAGE 20 OF 21
p . 21
3M ENVIRONMENTAL LABORATOR Y REPORT NO. E06-0 199
8.2 Solvent Blank/1-00 Comparison .
Table 13 . Area Counts of Solvent. Blank .
Area Counts
i FJle
ViN PosNbn FFOA
PFOS
PFH
PFBS
s060421a001 91 70233 3067 1853 192
s060421a002 91 16249 2196 1851 563
s060421a003 91 18222 16~2 567 471 .
s060421a009 91 48287 4524 . 1505 447
s060421a010 92 (184762 5006 2424 889
060421423 92 48308 . . 4192 4524 6609
s060421a024 92 n'84666 5317 . 4738 7007
s060421 a035 93 s060421437 93
01162166 . (11283547 . n)52950 23806 . n 8904 2706
( 'i1582 2357 .,
s060421a042 93 15939 . 2135 1693 1823
s060421044 94 49169, 42~8, . +'~274~54 . 1118,_
s060421a055 94 1268 P
~ 1141
" 328841 547
s060421a057 94
47071
( 3382
"'257946 946
s060421a068 95 53698 2525 5237 1137
s060421a070 95 74125 5907 6588 975
s060421a081 s060421a083
95 95
62139 68329
2124 3139
3945 3951
1225 1149
s060421a094 96 76799 4192 3512 1054'
s060421a096 96 77269 .
.3358 3905 1173
s060421a107 97 14824 1166 852 335
s060421a109 97 19120, 1763 1499 1289
s060421a120 97 9918 1028
.1024 698
s060421a122 98 24599 1589 2321 3275
s060421a127 98 37573 1305 6127 2190
s060421a129 99 18673 110d 1909 937
s060421a135 99 31743 1944 1472 63 2
LOQ Standard 163373 14600 13176 25766
'/~'LOQ Standard 81687 7300 6588 12883
(1)
Area counts of the solvent blank were greater than %z'ihe LOQ standard area counts .
PAGE 21 0F 21