Document MGmVb7aQVK4zdQOJXrGLGzvK9
FOR DU POUT USE ONLY
AR226-2911
Copies co
Haskeli
E. I. duPont de tfemours and Co., Inc.
Laboratory for Toxicol'ogy and Industrial
Silicon Road, P. 0. Box 50, Neva-k, Delaware 19711
Medicine
HASKELL LABORATORY REPORT MO. 109-83
Material Tested*
Study Initiated/Completed ------11^82-2728/8^------
Le titters Seaford, DE
EVALUATION OF METHODS AND DETERMINATION OF AIRBORNE!
FOR
Summary:
The ther-n-al desorptic
determination of airbornef whether historical data cc laboratory experiments with cfh< historical data are acceptable, to overestimaceflBBir concentrations.
method for the re evaluated to assess
ire acceptable* Based on
it is concluded that
desorption method tends
A new method for sampling and analysis of alt^ornafffvas developed and_
evaluated. and solvent
The c hod desor^cion
makes luse of Tenax sample tubes for of the Ttenax yith isopropyi alcohol
collection prior to
oSaSfj (^^^
analysis. This method meets the NIOSH criterion for overal^accuracy and it
is recommended for future sampling and analysis of airbornefl^^H^
Introduction;
At the request of Textile fibers Department at the Seaford Plant, work was undertaken with the following goals:
1) Co evaluate an existlnBmethod
for sampling and
analysis of aJ.rborne|J|BI|^l_____
h employs impingers
containing isopropynicono^rbr sampJ
Kith subsequent gas
chromatograrhic (GC) analysis of the iaopl-opyi cohol solution.
Company Sanitized. Does not contain TSCA CS
r i 2) co evaluate a mechod for sampling and analysis of airborne
currently in use at the Seaford^lant, which employs a solld'
adsorbent tube for sampling ofUKollowed by thermal desorpeion
and GC analysis.
^ JL/
3) to developand evaluate a new method for sampling and analysis of
airbornejHRrhich employs a. solid adsorbent cube for sampling of t^^followeaby solvent desorptlon and GC analysis of the desorbing
soTution.
This report describes the:experimental procedures and results for the evaluation work and the method development described above. Appendix I
describes the procedure ior use of the solvent desorpcionmejhod developed at Hasteil Laboratory for sampling and analysis of airborneQBy^
Experimental Procedure:
Because of difficulty in maintaining a constant proportion of
coapoaenes, it was not feasible to use afl|^B|^HH|iH|MBUjto generate a
A
test atmosphere* Nor was it practical co use a mixture as an-^na^tical
standard for GC analyst Sj Therefore, a. sample of each of chef^----oalnj^^
components was obtained (^------^----Krom Jackson. Laboratory^^fhel^^Hj
lllBiB^8 selected as^^rB^^Sentacive component of the test materTa^and
wasuserexclusively for the generation of test atmospheres
tfehkAeG eVvaaAUlukabt^iWokn wWVofcrCk. It was assumed chat the adsorption on
1------------------^^ 41|HR Uo^ch--e_--rooli--igo--mHeUrsfnwMroeoturledusneodt
deviate significantly from chat for deaorpcion efficiency studies
cnnrrooHBBaTTB^gPrafifpl'n^rpcpeerxfonnance checks.
and for most of leoaj^of the
All
andTor
It
contaia
should be noted chat trace aaounts of the
the standard Broduct^iin^cerial was found
sasvesFaTtdissiSSSSSSllSSSS^ These
to
components interfered significantly in attempts co use electron capture
detection for the GC determination ofIBB ^
Generation of Test Atmospheres >- Test atmospheres oo----^^Huere generated
in an industrial hygiene calibration manifold (IHCM) depicted schematically in Figure 1 In the IHCM, a vapor of the test material is generated at a
constant temperature and at a diffusion limited rate. The vapor is mixed
with conc
a
en
d tr
iluen ation
t
ogfHau^t^^H^
oHnIJs
tant Seri
; a
l
'
.low
dilu
rate tion
to of
produce a known airborne the source concentration
produces several <rori3ngalr concentrations.
^BH|^a With the material placed in a diffusion tube (10 mm 1.0. x 130 nm long)
at a temperature of iqO''CC^^nndd^^^^..llLLaa nitrogen flow of IL/mlnuce a test atmosphere of 50 mg/n oq^^^^^H&as obtained. This value was determined
Company Sanded. Dees no^inTSCAOr
from ehe measured weight loss Of Che^^^^^Rfroni che diffusion tube with
cine. This source concentration offlBBB^B diluted with air in the
ratios of 1:5, 1:10, and 1:20 to obtainnoninal air concentrations of 10, 5, and 2.5 mg/m , respectively. These valuea correspond to twice the Du Font
allowable exposure limit (AEL), the AEL, and one-half the AEL, respectively .
A Miller-Nelson Model HCS-201 atmosphere generator was used to regulate the
relative humidity (RH) of the djilution air at either 502 RH or 80" BH (at
23C). Air flow rates were monitored with mass flow meters at five locations
in the IHCM.
|
Igpinger Method Evaluation - Recovery experiments Here carried out with three
25 mL iapingers connected in series. The first iopinger contained 20 mL of
isopropyi alcohol (IPA) andjhj^ecend and third implngfirs contained 10 rnL of
IPA each. A 5 nig saaple of^B^^^Rwas vaporized in a glass tube connected
co the inlet of the first lopiqger^while air was drawn through the iapinger
train at a rate of 1 L/minute. ' Vaporization of the sample took place within 5 minutes and the flow of air lias continued for a total period of 1 hour to
simulate a 60-L air sample. Each IPA solution waa broughc ^o^tsinicial
volume and an aliquot was takeri to determine the aaount offUJRby GC
analysis. These amounts were compared to a recovery standard to determine percent recoveries in the three! impingers.
In another set of experiments, recovery,of he_fflpinger method was
checked with an air concentration of 50 mg/m o^j^----f Prior to the start
of^iecomoarative sample tube jloadings in the IHcff^che source concentration offfljjj^BjBras sampled with tws iiapingers in series that contained 20 sL and lO^iLorTpA, respectively. TBe total flow from the diffusion chamber in the IHCM at 1 L/rainute was passed through the iapingers-for a 1-hour period.
These solutions were analyzed 4s describedabove and recovery was determined
by comparison of the total amount ofyUBBRfound to the amount calculated from the known generation race of themater^il.
Evaluation of Thermal Desorptidn and Solvent Pesorption Methods - A six-port manifold was present in the IHQM system for each of the three test
concentrations of airborneM|1H|jdescribed earlier. A vacuum pump was used
ie draw air
approxiaate
froa rate
the aa&ifSTaSBPeriors through of 50 aL/minufie for collection
othf&^^^f^iaHoRle^tubAe. scoanttroanlled
vacuum source was created for c|ach port by plaeeBenc^r^Tcritical orifice in
line that limited the flow rate. Each line was calibrated for exact flow
with a representative sample Cube of each type of media in place.
With the vacuum pump off, three each of the thermal desorption and the
solvent desorption Cubes were placed in position at the three manifolds. The
vacuum pump was started to initiate sampling and after a two-hour period
sampling was stopped and the sample cubes were removed and capped.
The tbermal desor wich che use of tiheUB saaple Cubes were anaTyz?
^bes were analyzed at the Seaford Plant
L
a
ab
sC ora
anda tory
r
d. fri
t
h
T
h u
e s
e
(------/: soalvent dessoorrppttion
o^l--------rs s
Company Sanitized. Does not contain TSCA C^
standard. Reported results wee normalized Co a sampling rate of 50 BL/ninute for both media and the dacia were subjected to statistical analysis to quantitatively evaluate the tWo methods
Additional sample cubes fjor the solvent desorptlon method were loaded to obtain four-hour and six-hour samples at a rate of 50 mL/minute. Solvent desorptlon sample Cubes were also loaded at flow races of 100 nL/ninute and 150 mL/minute over varying tln(e intervals to determine the practical limits of absolute capacity and sampling rates for this medium..
^fiBB}" Stock standards ^H^^^Hf-11 IPA were exchanged by Seaford and Haskell.
these standards were analyzzeeddlb'byy cthhe normal method used at each site to cross-check the preparation of' standards.
Choice of Sampling Medium for .Solvent Deaorption Method - Three SflJ-id ^
adsorbents were initially investigated for collection of airbornejHHIrThese
adsorbents were silica gel, Anibersorb XE-340, and Tenax*. Based on^"
preliminary tests, silica gel ;was rejected as a candidate because of poor
desorpeion characteristics wttih methanol. The other two adsorbents were
found to have good desorptlon characteristics, but Tenax was chosen for all
further work because of generally easier availability.
The desorption efficiency of Tenax with methanol and with IPA as desorbing solvents was evaluated by two different meCiiods^tn the phase
equilibrium method, 1.0 nL of |a standard solution ofl^lHBjin either
solvent was placed in contaet 'with 100 ug of Tenax rora^TSaec ale-half hour. The concentration of c^ls equilibrated solution_a8thy compared to
the starting concentration to check for adsorption oqJH^B^Jln solution.
The desorption efficiency is expressed as the ratio oj^theequilibrated solution concentration to the standard solution concentration, expressed as a
percentage.
In the second method, a 10-uL aliquot of a standard solution ofj___ is deposited in the 100-mg section of the Tenax0 cube and the solvent Is allowed to evaporate. The 10^-mgsection of Tenax is then desorbed with 1.0 aL of either solvent for at least one-half hour. The concentration of the desorbing solution is then coinpared_othe concentration of a recovery standard chat contains 10 )iL pijIHHBNstandard/1.0 mL of solvent. The desorptlon efficiency Is expr^sse^ascne ratio of the concentration of the desorbing solution to the concentration of the recovery standard, expressed
as a percentage.
Gas Chromatographic Analysis - A Hewlett-Packard Model 5880A Gas
oy|HMp Chromacograph equipped with a;fla*-i ionlzatlon detectpr(FID) and a Model
7672A Automatic Sampler was uied for the analysis
the desorbing
solutions. Details of the operating conditions can Derounain Appendix I of
this report.
;
Company Sanitized. Does nol contain TSCACIH
It should be noted chat under the conditions reported, "finish solvents" Seaford Plant were found not to interfere with any of the
components when the solvents were present at a concentration
Results:
Imningaer Methhod Evaluation - Recovery experiments in which a 5-mg. sample of
I ^^----K--*a^aavaa~p^orized into a aeries of three impingers indicate an overall ~ aSvveerraagg^f^reeccoovvery of 972 for t^o trials. ' Average individual recoveries for
the first through third impingers were 872, 92, and IX, respectively with
very good reproducibility betjireen two trials. It was concluded from these
results that two impingers itt aeries with 20 mL IPA in the first impinger and
10 mL IPA in the second werf 'sufficient for quantitative recovery at > 952.
i
Impinger samples collected from the diffusion chamber at 5^a/B^tefa>re
sample Cube loadings on five separate occasions gave values Sotf^ll^Wj
within +5Z of those predicted: from measured weight losses with Cine. It was
necessary to sum the reeults froa both impingers since carryover (72) into the second impinger always occurred.
The tmpinger method was foundtobe^ccurate and reproducible for sampling and analysis of airhorne|H|^^Bat relatively high concentrations (worst case) under the conditions deserved.
Evaluation of Thermal DesorpCion and Solvent Desorption Methods - The results ale comparative analyses between the thermal desorption and solventdeso^ion
methods for sample cubes loaded on three separate occasions witluflHHHare presented in Tables 1 through 4 and are plotted in Figures 2 chrougn/.^ Nominal loadings are calculated from air concentrations, sampling rates, and
drration of sampling.
50 mL/minute.
All data are normalized to a sampling rate of
of
Three the two
qualitative methods for
statements regarding oveza^^esulcs of
sampling and analysis of^^^^Hjere:
the
comparison
^^^^^^~f
1) the thermal desorption (TD) method, with one exception, consistently results in reported'values that are higher than nominal.
2) the solvent desorptlon (SD) method consistently results in reported values that are lower than nominal.
3) the precision of thie TD method is much poorer than chat of the SO method for replicate ;. '-mple loadings.
The results of a statistical analysis of all of the reported values from
the two methods are presented in Table 4. These results indicate performance ac the three nominal loadings over time. A summary of the statistical
analysis is as follows:
5 -
S,,^.Do.sn<,lcon,ainTSCACBr,
Company as?
1} chere is a statistically significant difference (952 confidence level) between che nominal values and the reported values from both
methods.
2) chere is no statistically significant difference (952 confidence level) between the reported results at 502 KH (first and third loadings) and the reported results at 802 RH (second loading) for either method.
3) the mean bias (average relative" deviation from nominal) is +272 for the ID method and la -152 for the SD method.
4) the mean coefficient of variation for replicates is 21Z for the TD method and 42 for che SD method.
5) the overall accuracy of the TD method was besc at the lowest loading (+62 bias) and was significantly worse at the other levels (+37-382 bias). The coefficient of variation was best at the intermediate level (82) and was significantly worse at the other two levels (212 and 412).
6) the relative accuracy (-112 bias) and the coefficient of variation (22) for the SD method were best at the high loading, but there was no significant diffeirence in these values for the other loadings.
A concise evaluation of the two methods Is best obtained by application of the NIOSH criterion for acceptability of a new air sampling method . This criterion states Chat in the jrange of 0.5 to 2.0 times the environmental standard the overall accuracy' of a sampling method should be within +252 for 952 of Che samples tested. Six or more samples are to be collected at a
minimum of three concentrations and are analyzed for the amount of material collected. This criterion cap be expressed by the following equation:
Overall
;
Accuracy - ;+ [Absolute Mean Bias + (2 x Mean Coefficient of Variation)]
The equation incorporates both the accuracy (bias) and precision (coefficient of variation) of a test method.
The application of this equation to the thermal desorption method yields an overall accuracy value of +692 while that for the solvent desorption method is +232. Therefore, tne thermal desorption method does not meet the
NIOSH criterion of +252 overall accuracy, but the solvent desorption method '
does meet the criterion.
Experiments with the solvent desorpclon tubes at sampling rates of 50, 100 and i50 nL/ninute at^chrafi.concentrations indicate chat unacceptable
breakthrough (^252) ofjj^^HlJinfco the back section of Tenax occurs at
different times for each^HoWrate. The results are summarized in Table 5.
TSWCB>
C.^san^0----.^--co1ntain
Those results indicate chac a sampling race of 100 mL/oinuce for six hours gives the best detection limit (0.3 ^g/component) foilHfkrhen present at a
level of 2.5 mg/ or lower, sith little or no breakcnrougfi.
t ^ ^ ^ Sample tubea with 30 og of Tenax backed by-l^Bgwere found not to be
suitable because of unacceptable breakthrough ofu^HHIjinto the backup
section during sampling at 50 ^al/ninute for 4 hcursor^onger at concentrations as low as 2.5 ite/m offl------^1
A 1.002 stock solution ofJ^^^^Hi'n IPA received from the Seaford Plant was analyzed at Haskell Laboracoryand-Che reported value was 1.15% (+15Z).
A 0.3042 stock solution sent from Haskell to the Seaford Plant ess analyzed there and the reported value was 0.3232 (+7.6X). These results are within the known accuracy of the two methods and indicate no significant error in standard preparation.
Choice of Sampling Medium for' Solvent Desorption Method - Sample tubes
containing 100 mg of Tenax^ with a backup section of 50 mg were chosen as the
medium for collection of airborne^----HDesorption efficiencies of these
tubes with 1.0 niL of either in^thaSoT^it IPA by the phase equilibrium method and the spike/dry/desorb methdd were found to be 100% within experimental
error far allj^^Hmf the inajior oligomers tested. Spiking was done at
levels of 20 pganT'200 fig wicjh each of the three components in a composite standard. IPA was chosen as fihe better desorbing solvent because of an
improved baseline during GC analysis.
Conclusions:
'
A major reason for the evialuation^fthe impinger and thermal desorption methods for determination of a'irbornmUywas to assess whether historical data collected in the field.bvthese.aietnods are acceptabi Based on the
laboratory experiments withMHHHescribed in this report the conclusion is chat the historical daCaAarteaccepcable.
The impinger method appears to have good precision and accuracy, but in general suffers from a higher detection limit because of the inherent dilution factor caused by the volumes of solvent used. However, under the conditions used in field sampling the detection limit was sufficiently below the AEL for practical use*
The.thermal desorption isethod in general appears Co overestimate the
airbomaH^xoncentrations, but this is a bias on the safe side in terms of assuring^har the AEL is not eke-faded.
The solvent desorption method chat was developed in the coucse^f the
work is recosaasr.ded for future sampling sad analysis or airbomeifffj The
method is convenient and easy fco use relative to the others and i^meets the NIOSH criterion for overall accuracy.
^^^ 7 ^ed.OO65"010 ^P^551^
References;
1. Unpublished Ou Pone Co. data: Final AEL List, January 26, 1983. 2. U.S. Department of Health, Education and Welfare (NIOSH), "Docuaentafcion
of the NIOSB Validation Tests," Publication No. 77-185 (April, 1977).
Report by! ^ ^Resea .rch ^Chemirst ^
/j^^ / ^ ^ ^ Approved by: ~["Bruce Bruce M. Monroe Section Supervisor Analytical Chemistry
Date:
DJK:jtd:WP;12.9 Date Issued: March 29. 1983
There are 24 pages in this report.
/I ^.2.^1^3
APPENDIX I
PROCEDURE FOR U^EOF THE SOLVENT DESORPTION METHOD
Equipment and Maceriala Tenax* sample cubes
Personal sampling pump Isopropyi alcohol (2-propanol)' GC column Gas chromacograph
candards
100/50 mg with glass sealed enus. Available from SKC Inc., EightyFour, PA (412) 941-9701 Catalog ^226-35-03
calibrated with an accuracy of ~+52 at the recommended flow race
ACS Certified grade. Check for impurities
before use.
5' x 1/8" stainless steel packed with 10Z FFAP on 80/100 Chromasorb VAW
equipped with flame ionization detector and temperature program capability
Air Sampling
Prior to use, break the ends off a Tenax sample cube and attach it in
the proper orientation (arrow bointing cowards the pump) to a calibrated sampling pump accurate co^^^hin +52. For personnel monitoring, where the
airborne concentration oa|^|jls 'expected to be ^2.5 mg/m , a sampling flow
race of 100 mL/minute is Tecofflniended over a period of at least 6 hours. Other combinations of flow race and sample time can be used but Table 5 should be used as a guideline io avoid sicuations where breakthrough is likely to occur. Record samples flow rates and elapsed time for each sample and seal the open cubes with ttie caps provided.
Deaorption of Samples
Score and break each tube la.t the. glass wool plug that separates the two Tenax* sections. Pour the front (100 mg) and back (50 mg) sections of
Tenax* into separate 2.0 mL crjmp-cap vials. Pipet 1.0 mL of isopropyi
alcohol into each vial and cap it. Provide gentle mechanical shaking or
periodicnanual shaking of the vials during a minimum of one-half hour to desorbj^Huoaponents from the Tenax*.
o.^a---------00""'"""081
Preparation of Standards
From theIH^H^----^ MB*l| prepare a compo
Accurately weigh and record an amount near 0.025 g (25 og) of each
component and quantitatively kransfer all components to a 25-mL volumetric flash. Add isopropyi alcohol.(IPA) Co dissolve the sample and dilute to the
mark with IPA. Calculate the exact concentration ^f each component in fig/mL and label this solution "stock TBA." This solution can be kept for six months with refrigeration.
Prepare dilutions of the stock TBA as follows:
1)
Pipet co the
2.0 mL of mark with
-w- stocUHwJ.nto a IPPAA.^-L'taabbee]l this
10-mL volumetric standard "A".
flash
and
dilute
2) Pipet 1.0 mL of stockj^Kinto a 10-mL volumetric flash and dilute to the mark with IPA^Label this standard "B".
3) Pipet 0.5 mL of stoc
nto a 10-ttL volumetric flash and dilute
to the mark with IPA. Label this standard "C".
4) Pipet 1.0 mL of standard A into a 10-nL -olumetric flask and dilute
to the mark with IPAi, Label this standard "D".
I
5) Pipec 0.8 mL of standard B into a 10-mL volumetric flask and dilute to the mark with IPAi Label this standard "E".
I
The approximate concentrations for each component in standards A through E are 200, 100, 50, 20, and 8 ug/nL, respectively. Calculate exact concentrations for each component in standards A through E by multiplying the concentration of each componeijit in the stock standard solution by the following factors:
Standard
Factor
0.1
0.05
0.02
;
0.008
Freshly prepare these dUr-ions from the stock standard each time samples are analyzed. Use standards B through E for gas chromatographic
analyses.
10
=oesno,c*TSCAC.,
Compaq san,,l ,,,.<,.
GC Analysis
Operating conditions for analysis ofjfl
ere as follows :
n isopropyi alcohol
Column
5' x 1/8" stainless steel packed with 10Z FFAP on 80/100 aesh Chromasorb WAW
Oven temperature
85C for 3 ninuces then 15C/mlnute. to
200 "C
Dececcor temperature
275-C
Injector temperature
275'C
Carrier gas
Nitrogen at 30 mL/mlnute
Injection volume
1 ML
Attenuation
16
Injecc ac least 1 fit of standards and samples In duplicate for analysis. If an autosaapler Is used, place two IPA blanks between the last standard and the first sample to alalaize the chance of carryover of standard. A
representative chromatogtam fol- a standard solution Is shown In Figure 8.
Measure and record peak mights for standards and samples. Peak height
measurements were found to be (sore reproducible than peak area measurements because of peak calling. Plot'average measured peak height vs. standard concentration ()tg/mL) for each component or perform a linear regression on
the same data for each component. A representative set of calibration curves for the three major components:Is shown In Figure 9. Determine the sample concentration of each component from the calibration curve for chat
component. The amount oiu^^fys) Is numerically equal to the concentration since the desorbing volunelsT.0 mL.
FoEeach sample, sum the amounts on^^^^^^^^^^^^^^^^HBj
----A to obtain the total amount o^^^Biececcea^^erRrB^nia^ BBBbn on both the front fad backTeccxms of Tenax* and report the combined results. If the amoupt found In the beck section Is >25X of that found In the front section for:any one component, report this also since this
n^of^)s is an indication of breakthrough.
If
detected, de^rmlne the detection limit: for each component
on the gas chromacograph and report this value as a minimum detection limit.
11
^conte"
Con^81"^d.^5
DJK:jtd:WP;12.9
^^1 ^py^ """by!
Research Chemist
Approved by::
/f^^ 1^^ ^
fBruce Bruce Section
M. Monroe
Supervisor
Analytical Chemistry
Date: /ftf^ ^ W?
- 12 -
Companv Saniiizsd. Doss not contain TSCA Cy
Table 1
Results qfj^^^^Loading of 1-12-83
Method of Desorpcion
Thermal Solvent
Thermal Solvent
Thermal Solvent
Nominal Loading(
0*g)
Rep< irced
Va: ,ue (M !)
60
68 + 32
60
55.6 ^ 1.0
30
38.5 + 2.0
30
27.0 ^ 0.8
15
17.2 + 4.8
15
12.9 ^; 1.1
Biaa
+r3X -72
+282 -10X +15Z -14Z
Coefficient of Variation
472 22
52 32
282 92
Number of
Samples
3 2
3 2
3 2
Table 2
Results
----www----
4O
i*
i
--
^
^
^
HqE^Loaa
d^in^z
of
1-18-83
Method of Desorptlon
Nominal
Loading
W
Hepo rted Va^ue <W )
Bias
Coefficient of Variation
Thermal
60
65 +i 24
+82
372
Solvent
60
53.4 f 0.4
-11Z
12
Thermal
30
39.6 + 2.0
+322
52
Solvent
30
24.5 ? 0.9
-182
42
Thermal
15
18.6 + 2.1
+242
112
Solvent
15
11.9 +: 0.1
-212
12
)
Number of
Samples
3 2
3 2
3 2
-
13 -
Company ^e^noUon^TSCACB
Hechod of Deaorpcion
Thermal Solvent
Thermal Solvent
Thermal Solvent
Resuica of
Table 3 lading of 2-4-83
Nominal Loading
</*g)
60 60
Rape>rced
Vailue <rt;)
i
115 +i 11
51.4 ? 1.5
30
45.0 + 3.0
30
25.5 ^; 1.1
15
12.0 +: 1.4
15
12.3 f 0.3
Bias
+?22
-142 +502 -15Z -202 -182
Coef ficient of V<ariacion
102 32
72 42
122 22
Niimber of
! iamplec
3 3
3 3
3 3
Method of Desorption
Table 4
^^^^r Summary of ResuJLts for|
adings
Nominal
Loading (.ps)
Average
Deviation
From Nominal
Bias
Replicate
Standard Deviation
Coefficient of
Variation
Number of
Samples
Thermal Solvent
Thermal Solvent
Thermal Solvent
Thermal Solvent
60 60
30 30
15 15
Overall Overall
+22.3
-6.a
+11.0
-4.4
+0.9>7
-2.7
-
;
+382 -112
+372 -152
+62 -182
+272 -152
24.1 1.15
2.40 1.00
3.15 0.57
292
9
22
7
62
9
42
7
202
9
52
7
212
27
42
21
-
14 -
Coppan)? Sanit'issd, Doss not contain TSCA CB)
Table 5
(----------0^ Linfputth 100/510 mg Ten.ax* Tubes
Floii Sate (oL/'oinuce)
Time
(hr s)
50
2
4
6
100
2
4
6
150
2
4
6
Volume
(liter?)
6 12 18
12
24 36
11 0A 36 54
AjLr Conceqtirations (Bg/B-3)
^ 1
5 S S
s s s
JL2
lo.
S
S
S
S
S
S
s
S
s
BT
BT
BT
S
3T
BT
BT
BT
BT
BT
Approximate MDL* (ag/m-)
.
1.3 1.7 0.5
0.7 0.4 0.3T
0.5
-
S BT *MDL
- indicates satisfactory perfo'nuance - indicates unacceptable break.through
- B< in8lf,fuluga/codmepcoencecniot aonlimthiet GpCer component in air, with an MDL of
< 2.5
PSpresent at
15 -
50^ CsW? 3t^
AnTSC ,oiconia1
C<5-QOSS n
?1e
^
--
^i>i"(-
?iI1"?* ^
^
i
Figure 2
(1-12-83) MEASURED VS NOMINAL
" --f--i--r-i 1 | i r i i | i i i i | i r i i ) i i i i j ) i i -i
140.OC.) i (
Figure 3
(1-18-83) MEASURED VS NOMINAL
i -r--r-1--i i i j T--I 1 |--i i i 1--]--i--i i i--J--T--n--
120.00 -
S 3
100.00
^
{3 80.00
0
Q 60.00
1
'W
1
1.
0
0
%
1
o
1.
0
5
^
^
3" 0
5
'-
TO
^ 40.00
^
0
^-''"^^^
-'"'^ '
,^'" ^^
20.00
^^-^|B
\ ^ ^ 0.00\^S^1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1
>0
1 1 1
1
1
O.C
10.00
20.00
30.00
1 1 1
1 111
40.00
60.00
60.
--- -
NMIfML.tUT
10 ClKLVGdfi.UT
NOMINAL LOADING (fJG)
0 0 HBmili.lUT
*-1
p
Figure 4
( 2 - 4 - 8 3 ) 140.00p-^--p.-,
MEASURED VS NOMINAL
--p-r--,--i--T~r j 11--i i j i -r~r > j i--r^
WOML.W
D OttW.lUT
0 A01HGBIUHlUi.IBUT
I 5
.
g W
\^
\ i\^ 1\s0. \'&. ^l. \\o0 ^ ^ 'I 0o \1. \\^5. \w \y \.Qa.
Figure 5
RESULT VS DAY (15 JLJG)
- I I I 56.00 1 1( 1 1 j 1I 1 1. 1. | 1 1. 1. 1 1( 1l 1i 1i 1i 1 i i i j i i i i ( r r-r-
S4.00
25.00
:
0
20.00
0
:
0
+25
18.00
16.00 . ...... .......... ...
i144..00Q0 ~ 12.00 10.00
B 0 0
JBL
*" -25
0
8.00
B.OO
--
'
4
.
0
0
LL1 00
-
J1-
.
1i..jL--11..
1i
1 i
.1i .
i 1
1 i1
i 1
i 1
i 1
j1 1i .
i
| 1
, 1
1 1
11 1
1
1 1
1
1
1
1 1
1
1
0.
0.50
1.00
1.50 '
!5.00
'
5.SO
3.00
.
------ MulnellB.UT
-~ lW.IUT
Dva ^Y
-
-
InnwweeBB..OUtT
0
0B 0Q1 MlvuwktlUl..lUHT "^
0 thtfltfi.lUT
^VOSiu^osfou^^Q.,.Ps^ueg^tf -TZ.
RESULT
'
I
' '
1 !
5 5
8
5
.5
g
K
g
g
g^TTT ;,
! ! ' ! j ! !!! i j | i | j ,
i j in
'
^
,
01
0
g 0 0
i | i | 1 1 11 1 I -r
^
B I
,, 'a
5
sh
BA ~ 0 -
-
--"
;
BB
:
0
-
- tfl
CO
a
E
-3
0 >
<
0
^
-
ro
b -
0
--
(U i^
<*
8:
W a ~ 0 .
-
-
BE
QBQ
10 w >
i
- ^ ?.
A
- C/)^
(BO
^: --
> ^
~
-- ^
0
^
^ Q--
"*
+
i4
;
u
tfp
"
----
-
:
;0 .1111 1 1 1 1 t ( 1 ill 1 III I
Figure 7
(60 JJG) RESULT VS DAY 130.00 i
r - r .
i
i
i
| i
i
i
i
j
i
i
i
( 1"
t-i--r-T- i i | i i .- ^ i i i i
150.00
110.00 -
0
:
0
^ 0
100.00
&
5
N
|
S
w
1
?'
p
o3
? 1
|
0
i-
90.00 '...,.. - .........,;.........
.. ...
-. .. ..
.
..
.... ..... ..-0......
........
..
80.00
:
70.00
....
+25%
80.00
-
ni
:
^-oo :r
B
o
B B
40.00
-25%
30.00 L-J--L-l- 1 1 1 1 1 1 1 I I 1 I 1 I I | | 1 | | | | I | | | l 1 1 1
1
o't)0
0.50
31.00
1.50
8.00
2.50
3.00
3.5
----- NONIMU10.1UT
--- (Utta.mr
DAY
- -. NMfia.lUT
0
0 (g LVB(T.(MT k<a
0 IHMHLK.nW
Figure 8. Rfepres a Mixed
Chromatogram of tandard
^
jj
<c
^
<M
c 0
CO
0
*> t --
(> Cf
<^*^
f^.
T
a
r
~
J:S^
^ <r
^irt
Iyf
f<i k
-<
*
?^
p
'0 ;
-<J:
c
t-
^ ^-
-
m
<ft
u. (I.
v
0
.j - x ?
-
i tfU a
x o
> -
x x -'.
tf u *j \r y k 'c . "
.C >i -
W- 0. 0' a:
-^<! ^ff(k
-VM19-
se ? (*
(I
ft
f^
^^
.^tk
0 ^
0 .' A*
'4 tift u>
r. C
> ^ w
< T 0
il:
<!<? -
a if*
--
.
coicio-.. -. ^ 0 0 .; >5 r <j.
..
v
..-
'"> '/ 0 C -> r tfi <> .
?.SKfflzad. Does.^ipc..tqnfain. TSCA.CgS.
-23-
80.00
Figure 9
70.00
60.00
M
M
^ 50.00
0
fe
I 40.00
IsJ
P o ,
0
I
S . ts
1 V> i 0)
a.
Fi"
TO
P
a
0
8
30.00L 20.00 10.00.
O.OOLA_l_i i 1 i i i . I | | i i I i | | | | i i i i | i i i i | i .
0.00 S5.00 50.00 75.00 100.00 1S5.00 150.00
1 (JLJG/ML) CONCENTRATION
SI
0 nUOM.UT
1BU2IH.IUT
A iutm.ur
------ TBAffiIPA.LIN
- WKWUW
- luionpA.iu
tinfe A l(A
i <"-
