-------
TABLE 2 (cowbjnued)
1. Acrolein
0. A:r,c; ;i:rije
3. a.nzen..
4. Benz ;t:nt
ierovrr. (
ó. &en:v cr.0,r;dr
Carbon tctra—
C l 0 Ci
9. Ohry.. ,n.
Crone 0
peroxidi
9. li— ;;-et . ..,2
) 0t alati Ilyb)
IC. 0,4—0loano
to...n. d c0cradr
.1. t .n. ox.d.
13. H .xao9ierc—1 .3
b i it& d i an e
04. P . nt . br t no1
15. SCyrece
It. t.tr.:r .0*ro, .tnyl.no
Thtra.en’I I..d
17. c l onn* — .4—
4 ;
16. ‘:inyb acetat.
IA. £COyOen.
d Iron idc
1, 3— 3tcr. loro—
pro .r.c
4 ,ALt0 0AAO
— $o n.
• M0d.nit.
3 High
0 — tnkno on
cao 0aZar0 .ntormation 0
9..ltr, tazktd ratings
C>c ISO C r syst
Orb Al lOng dAb
3 1 U U 0
Highly tool:
(1 0 3 3 3
U U 3 3 3
0 U U U U
1 , 3 3 3
Aatty n.zard 11100 1 1
bla.h ?o .to ig
point. t.nlpor.tore,
crr 0r
0—19 (<0)
—3 (373 482 (900)
-11 (12)
24,000
35,000
100 3,300
6,000
6ci ita *5 1 St - 6C00 . .yscenu.c
9.11 — A10.r g.n
Aoto iq - A iit o i9 a i it i on
Th : bc - CUOfO )C bc .)
Ott •y at — Chronic systemic
Dang — Dangerous
bipi - txp!oi ion
Exp I (S . .) - toplosian by r.actioit
Tog - Ongestion
Zn ), - rni ,.ixt x . n
Itt - Trr0t..flt
I4 od - Md 6r ntI
00 04144 - 9.acta II1tA
04124..
3410 — Skin kOICOPtIOn
53.0 — Slight
SV* — 5.vsr.
- tsp rott irn
0’ 0.09 — V.c’y Uanqsroua
61 Sax, 9. 3. Oanq.rous PYor ..rtx,a o Induatrt..X Ch ica1I. Van I so Struld Pa Inhobd company. 0000lnnItl., ohio. 1975.
71 Patty. F. A.. ii. Industrial Nygiun. md toxicoloqy, 004 Ed., Vol. 11. InterIciane. Publiahars. New York. )i 5 1 York, 1962.
Oral
710’,
pp.n o.g/k 9
66
30—90
General 0x 10nt3tiOn
Irritant, highly toxic: flanoncole.
Irritant, highly toxic; !lannrnacle.
toxic. flaa.iacl,.
Eotr.mely toxic .114 carcninogsnkc.
toxic md a inoqeriic.
Irritant. ret. with Oxiditora. toxic.
O U U U (3
3 0 2 2 1
U U 3 3 2
215 425)
1* (65)
3.
0
3 3 2 19 (65) 413 (775)
(1 -20 (—4)
)IOIOE non. 10 2.920 toxo c .
toxic and carcinogenic.
Irritant, toxic. Oxidizer rot, violentOy
tow toxicity-plaltict 0•11.
toxic, irritant; flantaclx .
100 680 Very toxic irritant; Olaininaila,
00 1.00—100 Very toxic, irritant. hign.y 00 .110 .1011;
coca with oxidizers; •xpi.o. ion naza rd.
toxic and careinoqaruc.
0.046 20-200 0 ;ghly toxic, Irritant.
100 Irrit*nt, cool:; d. fl.anonabl..
100 Anesth,tjc 1 toxic.
000rsm.3.y toxic.
U U 2
3 . 1
L o t ran.
3 U 3.
1 U
2 2
U U U
tex c
U
U
2
(I
U
30 (*6)
none
—22 (8)
non.
toxic.
2,000 1’lkomabja, toxic.
25 60-500 Kigoly toxic; Irritant.
Irrit.nt ; flrnnnabl.; tool:.
13
-------
TABLE 3.
Atrcae r St
MSSS SPECTRAL INFORMATION FOR SELECTED COMPOUNDS [ 8]
c ,.. a:
seio ’-t
ç ‘e,:
t1.per4C1
(or,-- , ) .
taabt aa,er MS peak. I.! . ratU (relative
zntennty)’
3
4
I
C,h.
2 ’
2’
2
21
7’)
(100)
(200)
(130)
(100)
(200)
56
56
Sf
56
26
(83)
(65)
‘70)
(14)
(57)
75
2€
35
25
64.
( ‘70)
(58)
(60)
(Sli
(55)
50
79
26
26
26
(94)
(53.
(5 )1
(S4)
(45)
25
70
57
50
29
(51)
(44
(19)
(53)
(44)
25
29
25
2’
55
(49)
(39)
47
(37j
(39)
29
2)
2)
2)
31
(43.
9)
( 6
8
5 )
15
14
0)
35.
3)
(32
1.
1
53
34 j9
25
76
13
53
1€
52
53
(100)
(103)
(100)
(300)
(100)
(100)
(120)
53
63
52
24
53
26
26
(99
(92)
(73)
(93)
(96)
(‘73)
(97)
57
52
26
52
62
53
52
(75)
(75)
(‘70)
(78)
(73)
(69)
(‘791
52
52
51
32
51
50
28
(32)
33)
(28)
(33)
(32)
(27)
(41)
27
27
28
77
7’
2’)
52
(25)
(23)
(10)
(22:
(221
(301
(36’
20
20
27
53
25
5)
27
(20)
(11)
( 9:
‘7)
(13)
1 7)
33)
SC
36
50
20
2)
25
20
9
(101
( 7)
Il
221
( 4)
(1)
35
0 )
2).
54
35
54
10)12’
‘
S
4
0.
I
2
6 r i. , :
‘76
0.5.
76
‘79
75
75
‘75
‘76
75
76
78
‘76
‘76
(100)
(100)
(103,
(103)
1)3,
(100)
(300)
(103)
(100)
(200)
(300)
71
53
52
92
52
52
‘77
‘07
77
77
52
(19)
(39)
(15’
(2:)
(18 ?
(19)
(19)
(24)
(19)
(39)
(25)
52
52
77
0
51
‘77
52
52
52
52
17
(39)
(25)
(17)
(21 ’:
(111
(18)
(15)
(18)
(25)
(16)
(23)
57
50
01
92
77
51
53
52
01
51
51
(17)
(29)
(15)
(35’
(1 )
(26’
(3 ,3)
(151
(33)
(15)
(2 ,6)
50
‘77
53
35
SC
39
50
39
39
53
50
(24)
(34)
25
(1’’
(li
(13)
(10)
112)
(10 ,
(13)
(19)
35
3”
39
7”
91
39
50
5 )?
39
‘19
122’
12)
110)
(1(:
22’
213)
9)
132)
9)
(11)
(10)
‘79
79
19
•7.
‘
79
79
‘79
. “
‘75
‘71
9,
I 6)
0
“
1 7)
I
( 7)
“
I 6,
1 1)
75
7)
7
3,
“
‘ 7
7)
‘ 7’
‘75
‘7)
1
6 ’
0.
I 4
I 0
)
1 0)
4)
5 n C
164
C,3 9 . . M a
184
164
003)
(100)
9;
2.59
(351
(15)
153
183
12
(11)
i8
52
(32)
I 5)
055
39
I 6
( ‘7)
16’
2,96
) 5)
1 4)
252
182
1 4
I 4)
(65
157
4
1 4
Ser OOa)pyr ene
212
C,.9,
262
(200)
126
(23)
253
(21)
250
(. 6 )
220
(15)
3,13
( 9)
132
( ‘7)
124
I
6 ,,z,’2 c52o: d.
125
C,H,0
92
91
92
52
(1001
((00)
(103)
(100)
2.2*
126
126
175
(25)
(23)
(27)
(38
128
2
52
65
) 5)
(15)
(16)
( 8)
45
19
43
52
1 9)
(13
(12)
( 8)
92
65
36
63
1 5)
(12)
( 6)
( 6)
63
63
126
59
I 6)
(10)
I 7)
I 5)
35
9;
76
39
( 0)
I 9
( 5)
I 5)
9’
126
45
126
‘ 7
“
I 0
4,
Carbo: t•trac’r,2 rid.
112
001,
117
137
11’
117
(200)
(000)
(105)
(1Q01
129
119
119
119
(99)
(95)
(97)
(96)
171
121
121
41
3 )4)
(32)
(32)
(45)
17
53
*2
35
(32)
(24)
(19)
(44)
48
47
47
82
(201
(73)
(13)
(30)
84
54
84
121
(15)
(35)
(13)
(30)
35
35
20
84
(14)
(31)
( 4.)
(29)
34.
45
45
49
(321
I ‘7)
I 4)
(10
Cr Iry.er
CW’ irt.
nroperDel4v
226
152
0,s8
C&l .aC ’a
226
226
226
43
43
(100)
(500)
(1001
(100)
(300)
326
228
221
2.19
105
(20)
(25)
(22)
(91)
(77)
229
229
229
105
2.21
(29)
(20)
(19)
(76)
(43)
3,34
3,13
134
721
‘77
(151
(15)
(16)
(74)
(42)
2,33
114
113
‘77
9
123)
(17)
(14)
(57)
(25)
44
131
101
93
3 .20
(20)
(10)
1 9)
(43)
(19)
501
227
121
120
78
) 7)
I 9)
7
(73)
(13)
321
102 ,
101,
52
35
I
I 8 :
) 4.,
(3:)
(211
(corttnia.4
14
-------
TABLE 3 (continued)
(a) Eoqht P .46 Ondix ae a.. Spectra, 2nd .d..
Vol. 0. s... Sp4ctr ty S .t. C.ntx.. k1dstw ast n. 8ssd nq, )2.X.
1974.
M o Icuar
•sq ’ t.
V.’ol
•r caL
formula -
Eight
(45
n/n r4tio (relatove
i a3cr
2 3
4
teneLty (
8 - S
297
Ca..141 1 0,
—(l—ny ( .6. ay )
t & at.
149
149
149
149
149
149
57
149
(100)
(100)
(200)
(100)
(100)
(100)
(100)
(100)
57
279
219
47
57
57
44
57
(72)
(5 1.)
(64)
(73)
(37)
(54)
(99)
(48)
43
167
167
43
167
168
55
43
(48)
(39)
(46)
(00)
(27)
(33)
(79)
(28)
41
57
57
41
43
71
41
11
(42)
(32)
(16)
(50)
(27)
(27)
(65)
(24)
54
70
213
105
70
70
70
43.
(39
(26)
(14)
(44
(23)
(27)
(60)
(20)
167
71
114
167
71
42
43
50
(38)
(21)
(14)
(42)
(23)
(20)
(46)
(19)
70
55
156
74
41
55
2
279
(30)
(19)
(14)
(37)
(22)
(19)
(45)
(17)
55
3
260
56
55
113
56
167
(28)
17)
(14)
(17)
(17)
(24)
28)
(18)
1.4— 2 .o ..n.
$9
C .J C
28
29
28
28
29
29
(130)
(100)
(100)
(100)
(100)
(100)
29
88
98
99
59
39
(37)
(53)
(36)
(54)
(68)
(32)
89
58
58
59
29
5$
(31)
(40)
(32)
(44)
(54)
(34)
54
29
29
39
31
89
(24)
(28)
U7)
(32)
(32)
(22)
15
33.
31
43
43
31
(17)
(16)
(14)
(17)
(27)
(16)
31
30
30
27
27
27
(17)
(15)
(14)
(11)
(24)
(15)
27
43
15
57
20
20
(15)
(12)
(11)
(11)
(24)
(11)
10
15
43
10
57
41
(23)
(1.(
10)
(11)
(19)
(11)
Etnyl.ni d2 .or.lor lds
99
CaiLCIa
62
62
62
(100)
4100)
(100)
27
27
27
(70)
(86)
(91)
49
49
49
(50)
(37)
(40)
64
64
64
(33)
(32)
(32)
63
26
26
(27)
(29)
(31)
90
83
63
(23)
17)
(19)
41
51
98
(17)
(21)
(14)
61
61
51
16)
(11)
13)
tv . y1.n. ox d.
44
C h.0
39
44
29
44
(100)
1300)
(100)
(200)
44
29
15
29
(79)
(997
(65)
(83)
15
15
44
15
(55)
(55)
(65)
(53)
43
43
14
43
(23)
(23)
(26)
(22)
14
42
43
42
(21)
(15)
(16)
(13)
42
14
26
14
(17)
(13)
(14)
(12)
28
16
42
16
( 9)
(6)
(12)
(2)
16
29
13
45
( 4)
(4)
(8)
( 3)
8e a. ar51 ro—2 .3—
259
C.C1.
225
225
1100)
(100)
327
227
165)
(64)
223
223
(63)
(83)
290
290
(43)
(43)
360
190
(36)
(39)
188
262
(32)
(34)
110
188
(30)
30)
141
141
(29)
(26)
P.r.ta c lloroph.no l
264
C. 140C1.
266
266
1100)
(100)
268
26$
(70)
(65)
264
264
(68)
(63)
165
165
(54)
(29)
(47
36
(53)
(29)
95
167
(44)
(28)
130
270
(30)
(23)
60
230
(28)
(18)
Styren.
104
C&4.
104
104
104
104
(100)
1100)
(100)
(100)
103
102
103
103
(38)
(691
(45)
(38)
79
St
79
78
(31)
(45)
(32)
(27)
51
103
51
51
(35)
(39
(21)
(25)
77
78
77
77
(20)
(24)
(17)
(17)
50
39
100
50
(2.5)
(23)
( 9)
( 9)
29
50
50
52
(11)
(1?
( ‘)
( 9)
52
52
52
115
(11)
(16)
( 7)
( 9)
T.tra:h loro—
•thy l.n.
164
Cad..
166
166
166
(100)
(100)
(100)
164
164
164
(79)
(79)
(79)
129
129
129
(64)
(99)
(61)
13).
131
133
(62)
(66)
(58)
168
168
168
(48)
(48)
(46)
91
47
54
(21)
(42)
(18)
133
94
133
(20)
(40)
(19)
96
35
96
(241
(35)
(14)
?Cluen.—2 , 4—iLan sn.
122
C,&4,aNa
121
(200)
2.22
(15)
94
(10)
105
( 8)
104
) 6)
77
1 5)
61
) 5)
123
( 4)
Vonyl aC etatS
96
C.3IsOa
43
43
43
43
(100)
(100)
(2.00)
(100)
15
15
45
35
(18)
(9)
126)
(13)
29
96
29
96
(10)
19)
(25)
(6)
27
27
60
27
) 8)
(6)
(14)
1 6)
42
42
29
42
1 9)
(5)
(14)
C 5)
$6
44
42
44
1 5)
(4)
(13)
1 5)
44
29
44
28
) 5)
(4)
(13)
( 3)
14
29
27
14
( 4)
I 2)
( 6)
1 3)
Ethylene dtbecsld.
196
Ca14 . .9r 9
101
27
(100)
(100)
109
2.07
(95 )
(77)
27
109
(54)
(72)
29
26
(11)
(24)
36
28
( 9)
(10)
93
79
C 5)
1 5)
109
25
( 5)
1 5)
95
93
C 4)
) 4)
1.3—Osthloroprep.n.
13.0
C )4..C1a
75
16
(100)
(100)
39
77
(55)
(33)
77
39
(32)
(33)
49
120
(20)
(25)
110
49
(20)
(17)
29
122
(14)
(16)
112
39
(13)
( 6)
36
52.
(10)
1 6)
15
-------
TABLE 4. DIRECTORY OF INFORMATION CONThINED IN LITERATURE REFERENCES
— this s,tipI. !Si 0* this . . ...J otis. .ii .&I.r k.s. stfls.1.
b t *1 i i .St.
SC - hrth .t 14 d. .s. .i I - tao lana- Podlapi Is -
9 .1 - Plush 0 DIuiut aaIe curbs!. I*aoItI&
- Pn,. .i.t U I - fapi..qun rhp
I C - Silts. Opt P - P uts re na - .tI.pl .Ut.tp
90 - 9 5 5 5 5 SC - .ni.n.*, t - th-s.Sl t Olti - 54 1 19I55 JId .IoriOp
55 5 - 5554055 It 5Th . liwl .kh.,
ks -
* 4 * ! — .—t I - ni
pet rib — Psililsan SetS,
SC - (el uh,StStOi lisp4IV
- C*p111557 aa rh .an
SC - Cs. rh ,n55io4 l*3iy
lwtç - 1111’ p .rro..5. 5r5
fl - This Isp.’r rhrr.
hi - lts.Ir 4 4 55Pp4i0 . .
SC - tI . t,O. ra pist.
P t - PIe,r*,rt..r,
P IG - Ph . . - l IpStI05 4 5.
- I..r ,. .. - 490i .
IC - 544 ,I’lc.
Ut - llItr.-vIaI ii * 1-fl.
SI C - iheihir ‘rrr.
55. ,.1. - e.. .., *1* 0*Ihen.iaq p 5 4 5 5.
p. I. r .
, . nsr- i i , , i i i ‘ -r’ -
P
- I - U
•
I
S
p
S P P
I
I
pp pp
Pt P
a -
I
•1
la(.. 0 5t 1s! . _____ — i r ir ivi1 i ii i ,
M ,y l s . hl *lI* P P
5 * I a a P
P I Pt
la usIalpy ra i p I a
*4 yl rhI*.IIs ‘ P
1-Opto .. iiitSkhP iLk ‘ P
P - a
C *. I.penep.oak*
U4-U— a l I.7 1 155y 1 1 p 1 .tMIai* a
Oh. . ... I P
55I.pI.IS 4Puh 1551 15 P P P P P
Ittyi 1 15 soils I a a -
Irnc$ .i a n*-I.1- t st alt*S p p
P. I t. chIo t4*4 . 1 5 0i P P n
P I P a I p
t at pSdd 0 5 e Ith 9 l lS P • I a p
TPttISth 7 t 100*
SI_pt o.Itatp P P P
45* 5155 5 Itb*55 1 1 5 P P P P P
P P P P p
- I
TICs W I
S I a a a
I S S U
Otter Sir a a
Solar ta.ly) a
t._I S alt*1,lb I at a I • I S S
90 Ta hr IS SC SC
P 0 055 P
pnatlcul*eMIec . a I
90 5450*5 .
ptj $M’ S SO? P 9 S I
P 0 * 5 4 t h 1 is, Op.
Tan, 11(1
______ SC san CSC IC SC IC
ta .thsb ii II .an P0
Stasi s.
Qs thtatI ss a a s * a a p
9 5 0l itak$.. I I
na1 EaIan05tI*S . I a
1 5 1. 1 7 I
P .5 . 1 . 55 Ut.
S 1i.., U . S.
Slat. Itatash
I
I . . I . P Ii C I 4 . SI 41 4! 44 4•,
a
• *
a a S P
* P . S I
P p a P P
P
a p
I P u P 4
I
• p
• P P a
a
o p
o P P
• p P
I
I
a
S
S S S S I a S -
• a- - --I I I U I I I U
a I - I
• • v ii • • • •
IS Pal 901 555 San 55 50 PC SC SC
- SC
P V P P 0 P p p p p
* 5. S S U I S = S
‘ . .P .$t ‘Ju..
is 1 t e 4. flk. W 14’s.P IS, (I Is
.41. alt
SC SC SC 5 5 W SC SC SC ala SC SC PC SC flCSC
P91 P10 P0 9 1155 SC hr 501 0*1 P10 TI SC PIG Fl FL
U S I I a S - U I I 0 I
I - I
p p
P P
• a a U
• I S S
• a a S
SC Usa OI P 55 55
F
• a
S
a
• S I I I I? P
Is V
OCI tC$ IOU
cc 15555 - SIC
‘•“ rio 9 ii
• S I S I I
• C S
S
a
a
• S T •
St
is, Itt
9 10
SC SC 55
— Pl o .1*
• a I
a I S a
I
U S
• 1
a S
• U
-------
TABLE 4. REFERENCES
(9 ) Analytical Methode for l isa in Occuptational hygiene: Detersilnation of Benzo(a)pyrene and Benzo(k) f luoranthene in Airborne P rticulates
• (dircnnto.Jraphy and ( tica1 Fluorescence). Pure and AppI ie.i Chenistry. 40(3): 36-1 — 36-7. 1974.
1101 laker. B. B. Neaaurlng Trace Impurities in Air by Infrared Spoctroscupy at 20 M .ACrS Path and 10 Atnospheree Pressure. Anerican Indus-
trial Hygiene Association Journal. 35( 111:735-840. 1974.
U I ) Ballon. B. V. Seoud I IIOSH Solid Sorbents Reundtable. HEW Publication No. N IoSII-76-193. U.S. Departuent of Health Education and
Wslfae, Wmaliington. D .C.. 1976.
(12 ) Baretta. 5. 0., C. A. Hawk. P. S. Bodeaw. J. 5. Anderson. C. Dickerson. and N. P. Adriauio. Cheaica l Material Handling Guides. Anerican
1ndu .tri l hhygte* Association Journal. 39(11):898-903. 1978.
£131 Burtadi. W., and B. Anderson. Trace Analysis of Organic Volatile. in Water by Gas Chiro.atography —Nass Spectro.etry with Glass Capillary
Co l i is. Journal a4 Cbroantogxapliy. 112(l) :70l-718. 1975.
(14) Bertsch. w.. a. C. chang. and A. Slattis. The Deter.inat lon of Organic Volatile. in Air Pollution Studies; Characterization of Profiles.
• Journal of CImenstograic Science. 12(41:175-IS?. 1974.
US) Burnett. H. 0. Evaluition of Charcoal Sampling Tubes. Anericafl lndgStrial Hygiene Association Journal • 37(1): 31—45, 1916.
(16j Ceatree la. V. • and k. Van Cauwenberghe. Experboenta on the Distribution of Organic Pollutants Between Airborne Particulate Matter and
Corresponding Gas Pkase. Atanspheric Envlronnent. 12(5) :1133-1141. 1978.
1171 Chroetak. J. Health Hazard Evaluation/Toxicity Deternination Report. NISOII-TRHIIS 74-23-216. P 8-249396. National Institute for
Ocaupatic inal Safety and Health. Cincinnati. Ohio. 1975. 7 p. -
-i tl ) Criteria for a Neca nded Standard: Oceupat ional Ezposure to Dioxana. HEW Publication No. NI0SlI-77-226. U.S. Departaent of Health.
Education, and Velfare. Washington, D.C.. 1977. 193 pp.
U I) D’iqostine. H. B. • and 3. C. Gillespie. Coanents on the llA Accuracy of Measurenent .llequire.ent for Nonitorin9 Bep loyee Exposure to
Beazene. Anerican Industrial Hygiene Asaociaticn Journal. 39(6) :5l0- 5L3. 1918.
120 1 Dietrich . H. W. and I .. N. Chap.nn. Deterhination of Benzyl Chloride in Delaware River Plant Air. Special Study; S-75-SS-13. Monsanto
Cc any. St. mis. MiSsouri. 1975.
(ill Hts,aherg, V. C. Fractionation of Organic Material Estracted fran Suspended Air Particulate Matter Using High Pressure Liquid Chronatog-
raphy. Journal of aironatographic Scienee. 16(4)145-151. 1978.
£22 1 linbbe i n . L. Potential Hazards of lu.igant Residues. Rnviron.enta l Health Perspectives. 14:39-45. 1976.
£23) Giui, C. S.. I I. S. Cha n, and G. S. Heft. Rapid and Inexpensive Method for Detection of Po lychlorinated Biphenyls and Phthalates in Air.
A a1ytical Chenistry , 47(l3):2319—2320. 1975.
(24) gLen, c. S.. H. S. Chan. G. S. Neff. and 8. L. Atlas. Phttia late Ester Plasticizers: A New Class of Marine Pollutant. Science.
199(4327) :419—421. l9lL
(25) a, a. J. Distribution of Airborne Polycyci ic Aro.atic Hydrocarbons Throughout Los Angeles. Environaefltal Science and Technology.
10(4) 370-373. 1976.
126) Scot,, A. A., V. S. El.. and R. E. Kupel. Establishing a Protocol f t c. Laboratory Studies to be used in Field Sampling Operations.
A.erican Industrial Hygiene Association Journal. 39(4) :880-884, 1978.
(37) Mactines. B. M. The Analysie of Tetraalky l i.ead Compounds sad their Significance as Urban AIr Pollutants. Atuospheric Environaent.
2(9) :847—852. 1917.
(281 Hill. H. H. • Y. V. Gagnos. and A. V. Teass. Evaluation and Control of Contaaination in the Preparation of Analytical Standard Solutions
of Hazardous Chesitcals. Anerican IndustriaL Association Journal, 39(2); 157-l6O. 1978.
(continued)
-------
TABLE 4 REFERENCES (continued)
(29) nu 1ey, C. F., and N. I I. Fetcham. A Solid Sorhent Pereonal Saaq.llnq Method for th” 0.’tsrmlnatinn i A roI”in in Air. An’ricnn 1n .Iu ’-
trial flyg )Pnn A . .soclatlon Journnl, .Glc-6lo 19777.
1301 .Jacobs , ft. W., and P. 8. Syrjaht. 11ic Usc ’ of infinroci Annlyzc’rs for Nonitorin.. Arrylotiltrilo. Amorir ,n i ti, l lIy .’n’ A c . ,i 7ir,n
Journal, 39(21:161—165, 19177.
(31) Li, P. ‘1., .2. F.. Coing, and 3. I.. Splqarc’llI. Sampling nd AnaI is of Sclr’ctr ’d To, .i .5 u nc-r ’ T . ’ .k 177 -
kPA-5f ,0/6-76-olS. Fnvironwuental Ptotc’vtion A .Jc’ncy. Wa Ii1nqt n , 0. C. • 1976. I 2
(32) l,loyd, 3. W., general chairman. Prceedlnge of 81051 1 Styrene — IUitadlenc’ Rrir .Iing. 77 .W Publication Nc ,. NlO II-17-l29. U.S. Department
of health, Education, and Welfare, Washington, 0. C., 1971. 169 pp.
(331 licCuirk. N., and 8. 3. Malnwaring. Pevc,rmed o-fl1mc.nm(on l Techniqu.. for Multiple Separations of Penrolalpyrone from Atmo ’q’heric Aerosol
Saep1e . Journal of Chromatography, 13511) :211-244 , 1971.
(34) Mulik, 3. 0.. H. Cooke, N. F. (2uyer. C. N. Semeniuk, and . Sawlcki. A Cam tlq.iir1 Chromatoqra ,hIc orrscc’nt Prooeduro for the Analy is
of $enzo(a)pyrene in 24 Hour Atmosç .heric Particulate Samples. Analytical 1,ettnr , Di ): ,ll-52d, 1915.
(35 ) Nelson, C. 0., and C. A. Harder. Ne pirator Cartridge Ffficlency Studies. V I Fffect of Concentration. American Industrial hygiene
Association Journal, 27(1) .205-2l8, 1976.
1361 Mutt, A. Measurement of Scums Potentially Harardous Materials in the At phere of Rubber Factories. RnvIronsw ntal health Perspectives,
17.117—123, 1976.
I3 1 Pierce. P. C., and N. Xatz. Chrosatographic Isolation and Spectral Analysis of Polycyclie Quinones. Application of Air PolluUon
Analysis. En ironmenta1 Science and Technology, 1O(1P :4S-5l, 1916.
1361 Reckner. I .. P., and 3. Sachdev. Collaborative te 5tinq of Activated Charcoal Sampling Tubes for Seven organic Solvents. IIFW Publication
No. NIOSH-15-184. 71.8. Department of Health, Education, and Welfare, Washington, D.C. • 1975. 221 pp.
(39) Robinson. 3. V. Further Comeentg on the Analysis of tetraalkyl Lead Coumpounds and Their Significane an Urban Air Po11utant . Atmospheric
Envitonment l2(5) 1247—1248, 1918.
1401 Ronaell, 3. N., and 1.. A. Shadoff. The Sampling and Determination of hI locarbon in Azthient Air Using Concentration on Porous Polymer.
Journal of Chroumatography, 134(21 .37 5-384, 1971.
141) SeVets. 1 .. N., 5. G. Mp1 her, and N. 1. ocsi . Dynamic U-Tube System for Solid Sorbent Sampling Method D,vc’lopment. American Indu tria1
Hygiene Association Journal, 39(4),321-326, 1978.
1421 Spangler. C., and N. de ihevera. Benzo(a)pyrene and Trace Metala in Charleston, South Carolina. Et’A-450/2-1S-004. U.S. Snvironpsental
Protection Agency, Research Triangle Park, North CarolIna, 1975. 58 pp.
(43) Yasuda, S. A., and S. Dan Loughran. Air Sampling Methods for a-Tetrachlroethane and Other Related Chlorinated Hydrocarbon S. Journal of
Chromatography. 137(2) .283-292. 1917.
)44) Trjanheikki, 8. A Method for Personnel Sampling and Analyzing of Phenol. American Industrial Hygiene Association Journal, 39(4 ) 327—
330, 1978.
(45J Zeidea, S. C., end P.. 71. Norton, Trapping and Deteriulnation of Labile Compounds in the Ca Phase of Ciqarette Smoke. Analytical Chemistry,
50(61 :179—78?, 1978.
-------
Be nzene 1.3—15 PPba
1.3-15 ppb
Benz.c (a) pyrene
(with B(e)P perylene)
8 — 40 ng/m
1 — 3 ng/m 3
200 — 300 ng/rn 3
0.1 — 2.0 ng/m 3
0.122 — 1.76 ng/rn 3
MO - 57 ng/m 3
0.003 — 1.98 ng/m 3
16 Ambient
21 Ambient
21 Do mwind froi t co)ce oven
25 X os Angeles, CA
34 Urban areas
36 Rubber factory (outside)
42 Charleston, Sc
30 — 130 ppt
twith B(s)A3
S—3 Onq/m 3
0.2 — 2.0 ng/m 3
0.002 — 0.09 ng/n 3
0.08 — 2.5 ng/m 3
(phthalate ester)
‘10 uig/m 3
126 - 200 ngfm 3
30 - 70 ng/m 3
0.4 — 3,370 ng/m 3
An biant
los Angeles, CA
Over open ocean
24 Over open ocean
3.1 Car interior
16 Ambient
23 Office area
27 Ambient (various)
P.ntach lorophefl o l
1 — 3 ng/m 3
16 Ambient
Pereh loto.thylene
(tetrachloroethylelle)
30 — 130 ppt
40 Ambient
polychio r in ated
biphanyls (PCB)
35 — 90 nq/m 3
0.1 — 1.0 ng/in 3
23 Office area
24 Over open ocean
£Ifoz. tion in both references provided by the sas e authors.
TABLE 5. CONCENTRATIONS OF VARIOUS COMPOUNDS IN AIR
SAMPLES REPORTED IN LITERATURE REFERENCES
Concound Concentration Reference Type of a r sa 2e
13. Ambient
14 Ambient
Carbon tetrachloride
C ’rysene
Dibutyl phthalats
Di- (2-.thylhexyl)
phtha late
40 Ambient
16
25
24
19
-------
The analytical technique determined to be most suitable for the
determination of the twenty compounds was gas chromatography
combined with mass spectroscopy (GC/MS). Gas chromatography
has been used in various applications for all twenty of the
compounds of interest. Mass spectroscopy is also widely appli-
cable and aids in compound identification/confirmation by pro-
ducing mass spectra for compounds elutirig from a chromatograPh.
To further aid in compound identification by retention time data,
maximum gas chrornatographic resolution, through the use of capil-
lary columns, was desired. As an alternative to mass spectro-
metric detection, simultaneous detection by three to five
different, ceneral and selective, gas chromatographiC detectors
(e.g., flame ionization, electron capture, photoionizatiOfli etc.)
was also to be quickly evaluated.
The technique of compound desorption, thermal or solvent, from
the sorbent materials prior to GC/MS analysis was to be deter-
mined experimentally. The greater sensitivity attainable by
thermal desorption was to be evaluated in comparison to the
greater compound stability and (frequently, though not necessary)
more reproducible desorption efficiencies attainable by solvent
desorpt ion.
The literature survey of the twenty, selected, potential carcirlo-
gens indicated significant difficulties associated with the
sampling of two compounds in particular, tetraethyl lead and
curnene hydroperoxide. Tetraethyl lead was found to be an
extremely toxic compound which to obtain required an on—site
inspection of laboratory facilities by its supplier. In addi-
tion, it was indicated to be a relatively unstable compound that
readily thermally degrades and that may not persist in the
atmosphere 119, 311 as a pollutant. Cumene hydropero xide was
found to be a very strong oxidizing agent which also would not
likely persist in the atmosphere due to its extreme reactivity.
Such a reactive compound was thought’to present a particUlar
problem with potential in-siti reactions when concentrated in
a sorbent cartridge. Therefore, these two compoundS - tetra—
ethyl lead and cuinene hydroperoxide, were replaced as compounds
of interest by two alternate compounds, ethylene dibrom .de and
1,3—dichioropropene (cis and trans isomers).
A few general problems were found to be associated with the
twenty selected compounds. For example, highly volatile com-
pounds could not be quantitatively retained on solid sorbents if
the sampling volume was very high; while non-volatile compounds,
such as benzo(a)pyrene, which exist in the atmosphere at con-
centrations lower than volatile compounds generally could not
be detected analytically without very high sampling volt2xfles.
Another problem was that reactive compounds, such as styrene
and ethylene oxide, which tend to polymerize on active surfaces,
were more effectively retained on sorbents which had relatively
20
-------
more active surfaces. These paradoxes indicated areas where corn-
promises would be required to obtain a sinqie sampling system.
21
-------
SECTION 6
SELECTION OF SORBENT MATERIALS
OS JEC TI yE
Based on the review of sampling and analytical techniques, the
sampling technique determined to be the most applicable to the
collection of the twenty selected compounds was sampling with a
system containing an appropriate combination of solid sorbents.
Sorbent materials having complimentary adsorption characteris-
tics were to be selected so that all twenty compounds could be
collected for subsequent analysis with a system using these
sorberits.
EVALUATION
The selection of sorbent materials to be incorporated into a
sampling system for this contract was largely based on work
previously performed for EPA Contract 68—02—2774. Under this
contract a combination of sorbent materials was to be selected
for .use in a portable collection system. Ideally, these sorbents
also were to have complimentary adsorption characteristics so that
this portable sampling system could collect compounds varying
widely in volatility, polarity, and functionality.
Before beginning studies to evaluate and select sorbent materials
for EPA Contract 68-02-2774, a comprehensive review of available
literature on the use of solid sorbent materials for sampling
organic vapors was conducted. Information was collected on
specific sampling applications for over 110 compoundS using more
than 30 different sorbent materials [ 46). The pertinent know-
ledge obtained from this review included:
• Capacity/efficiency information for various compounds
on various sorbents,
• A variety of deso ption techniques, thermal and. solvent,
[ 46) Brooks, J.J., and D. S. West. Solid Sorbent Digest.
Monsanto Research Corporation, Environmental Analytical
Sciences Center, Dayton, Ohio, 1978. 126 pp.
22
-------
• General knowledge of sorbent properties, capabilities,
and problems, and
• Possible experimental techniques for sorbent evaluation
and sampling validation.
Six sorbent materials were selected for further laboratory
characterizations. These sorbents were Tenax—GC, Porapak R,
Porapak N, Chromosorb 104, Ainbersorb XE-340, and SKC Inc.
activated charcoal (which is used in NIOSH tubes). Later,
Chrornosorb 104 was eliminated from consideration and laboratory
evaluations.
A matrix of test compounds representing a wide variety of
polarities, volatilities, and functionaj.ities, was selected for
laboratory characterizations of the sorbent materials. This
matrix of aliphatic and aromatic compounds is presented in
Table 6.
TABLE 6. ALIPHATIC AND AROMATIC
TEST COMPOUND MATRIX
.
Volatility
Polarity
.
Hydrocarbons
Halogenated compounds,
aldehydes,, ethers,
ketones and esters
Nitro compounds, nitriles,
amines, and strong acids,
alcohols and phosphates
Low
Medium
Hich
Low
n-Hexadecane
Phenanthrene
Hexachloro—l,3—butadjene
4-Bromodiphenyle ther
Succjnorijtrjle
o-Nitroaniline
d
Me iuin
iso-Octane
Naphthalene
bis—(2-Chloroethyl) ether
1,2,4—Trichlorobenzene
Ethylene glycol.
Nitroanisole
High
n-Butane
Benzene
Propylene oxide
Benzyl chloride
Acrylonitrile
Phenol
The sorbent materials (Tenax-GC, Porapak R, Porapak N, Ambersorb
XE—340, and SKC Inc. activated charcoal) were then evaluated in
the laboratory for their abilities to collect the compounds in
the test matrix and low molecular weight aliphatics. A gas
chromatographic technique involving the generation of Arrhenius
plots was used to estimate sorbent capacities. With chromatog—
graphic columns packed with the sorbents of interest, the test
matrix compounds were analyzed at Various temperatures. By
plotting the elution volume per gram of sorberit versus the
reciprocal of the absolute temperature (Arrhenius plot) elution
volumes could beextrapolated to room temperature (20°C). This
provides a estimate of the volume of air which could be drawn
through a .sorbent tube containing. a, given amount of sorbent be-
fore a teSt compound would begin to elute or “break through”.
The data obtained from these experiments are compiled in Table 7.
23
-------
a
N,tentl.tly 1.rg
b
fta loq,natpd.
TABLE 7. COMPILATION OF CAPACITY AND SOLtIBILITY PARAMETER DATA
“.3
.
,e
6 1 ,,14J.
lit
talc
P9
rrr
VIr, 0 . .
L,’g
r,’ ;
V
nx
i /
r
vg
_Por,r*k
rrr
L/ .j
N
iq
r 1
P- rnbfl- 4r)
n i-
Vq 20 ,,
I . q
rr 9
S ( fh.urto.%
r n
Vq 30 , ,
L/q
Ioq
Yrr 9
l th.n.
Ptkan.
Plopane
n*utans
a—Ppntw
Puopylen, ,,ald
90
190
292
395
505
534
-
—
2*2
401
519
—
ILOI
10.01
44.09
39.12
72.15
60.00
-
—
—
0.36
—
3.14
—
-
—
—0.900
—
0.417
-
-
-
1.93
—
12.4
-
-
—
0.693
—
1.097
-
-
—
3.9
—
17$
-
—
—
0.311
—
1.14$
0.009
o. r
3.11
10$
7 , Sr)t,
-
-2.046
-o. 4o
0.727
7.023
3.973
-
0.047
2.40
163
12 N 10’
2 * 10’
-
-1.329
0.190
7.710
4.079
5.3(11
-
Acr 1 lonttri Ip
53?
511
51.06
9.33
0.971
16.9
I.22S
$9
1.019
R-H w
,,,‘r.ne
$79
719
619
719
96. 17
79.11
-
82
-
1.914
-
175
-
2.244
231
-
2.179
4 N 10’
—
5.602
—
-
—
-
—
1n,-Oct.n.
79*
914
134.22
0.512
-0.274
1350
1.550
2,900
1.462
-
-
-
-
Ethyl.ne qlycol
906
$2.01
120
2.070
-
-
-
-
-
-
-
.
Succinonitrile
—
1,073
90.09
1.04 N 10’
4.015
-
—
—
.
Phenol
9 enayl thIn,14, 1 ’
1 .17$
—
—
1,253
94.11
126.59
5.46.0
1.02 1 10’
3.737
4.007
—
-
—
-
I N I0
6.6 a 10
I2.477
5.670
—
-
—
-
—
-
—
-
N. ,hthalene
1 ,26.9
1,410
129.1$
3,260
3.313
—
—
—
—
—
-
-
Pt. U-chlorosthyl)
ether 1 ’
1,402
1.402
141.02
1.0$ a 10’
4.025
—
-
2.9 N 10
6.44)
—
-
-
-
0— Pltroan I It99
—
1,6.37
139.12
2.9 a ID ’
4.460
—
—
—
•
—
-
-
m—N itrnaid.o l e
—
1 .694
151.13
1.14 a t o
4.OS*
—
—
—
—
-
-
-
-
PIwn.nthr,n,
1 .141
—
179.22
1.9 * l0
6.292
-
-
—
—
-
-
-
-
n—We *ad,c.n,
1.911
1.764
226.43
1.3 a 10’
5.113
—
—
I • 2,4-it ichloro—
benten, 1
—
2.011
191.46.
1.57 N $0’
4.195
—
—
—
—
—
-
4-Srenodlpheny I
etheib
—
2.616
249.11
2.4 a 10’
6.177
—
•
—
—
—
-
-
—
lleNachloTo-I • 3—
but d I en, °
—
3 .791
267.6
1.15 N ID’
4.059
eapeilmentil error.
-------
A means of correlating sorbent capacities with compound properties
was sought such that a sorbent’s ability to collect any given corn-
pound would be estimated (based on the properties of that compound).
A molecular weight-modified solubility parameter ( 6 m)’ a value
combining compound properties of intermolecular forces of attrac-
tion and molecular weight was found to give reasonable correlation
with the sorbent capacity data experimentally obtained. This is
demonstrated by Table 8 and Figure 2.
TABLE 8. CORRELATION COEFFICIENTS FOR 6
VERSUS LOG FETvg m
Correlation
Sorbent material
coefficient
SKC charcoal
0.996
Axnbersorb XE—340
0.999
Porapak N
0.993
Porapak R
0.944
Tenax-GC
0.806
Tenax-GC (excluding
halogenated compounds)
0.916
A number of experiments were also conducted to determine how
amenable the selected sorbents were to thermal desorption of
the test matrix compounds. These desorption properties were
very quickly evaluated by desorbing standard solutions of the
test matrix compound through small sorbent-filled tubes into a
gas chromatograph. Compounds having small elution volumes at
high chrornatograph temperatures during sorbent capacity deter-
Inination were found to be efficiently desorbed. The ability
of sorbents to efficiently desorb collected compounds also was
found to favorably correlate with a compound’s solubility
parameter. A qualitative comparison of 6 m with sorbent capacity
and desorption efficiency is given in Table 9. The enclosed
regions indicate compounds which are probably or possibly quanti-
tatively collected and are probably or possibly quantitatively
desorbed from the sorberit materials indicated (assuming a sample
volume of 480L and lg of sorbent). These regions correspond
to ranges of values listed in Table 10.
Other areas of laboratory evaluation of the sorbent materials
included an investigation of conditioning and handling procedures,
level and composition of sorbent background, and pressure drops
across various sorbent tube designs packed with the sorbent mate-
rials. Table 11 summarizes the major results of the sorbent
chazacterization (conducted under EPA Contract 68-02-2774).
25
-------
Methane
Ethane
Propane
n-butane
n-Pentane
Propylene oxide
Acrylonitri le
n-Hexane
Benzene
. so—Octane
EthyD ene glycol
Succinoriitrile
18 Pheno’
17 Bentyl chloride
13 Naphthalane
5 Eis —(2—ChlOrOethYl) ether
12 o —Njtroaniljne
15 m—NitroAflisole
10 Ph.nanthr*fle
1 n-Mexadecafle
14 1 ,2,4_Tr ch1OrObeflZene
1]. 4 roieodipheflyl ether
2 MexachlorO—1 • 3-butadiene
Correlation of
test compounds
FET
and log Vg 20 o
on sorbent materials.
for
>
15.0
14.0
13.0
12.0
11.0
10.0
9.0
8. 0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
-LU
-2.0
• . 0
0
Ti
UXX 2000
Name
s 1 u
3000 4000 5(
Name
A
B
C
(7)
6
9
F
16
4
6
3
Figure 2.
26
-------
TABLE 9. QUALITATIVE EVALUATION OF TWO SORBENT PROPERTIES COM-
Name
6_(4).
lit
6 m’ 31 ’ Tcnax—GC
caic Trap Desorb
Porapak ft
Trap Desorb
Porapak N
Trap flesorb
Pi hersorb
SI C Charcoal
Trap Desorb
Trap
Desorb
Methane
89
— 00
I
00
I
00
I
00
I
00
I
Ethane
100
— 00
I
00
I
00
I
00
I
00
I
Propane
n—Butane
282
395
202 00
401 00
I
I
00
00
1
I
00
00
I
I
00
00
1
I
00
/
I
I?
n-Pentane
Propylene oxide
Acrylonitrile
505
514
551
519 00
— 00
573 00
I
I
I
00
00
00
I
I
I
00
00
00
I
I
I
/
S
1
I?
I?
1?
/ I ?
/ I?
1’
n-Ilexane
629
630 00
I
00
I
00
1
1
I?
/
x
Beuzene
719
119 00
I
00
I
00
I
I
I?
I
X
iso—Octane
Ethylene glycol
Succinonitrile
Phenol
Benzyl chloride
Naphthalene
Dis(2-ch loroethy l)
ether
o—Ilitroani line
*—N itroaniao le
Phénanthrene
n— Ilexadecane
798
906
—
1,176
—
1.269
1,402
—
—
1,747
1,811
8)4 00
— 00
1,073 1
— /
1,253 /
1.410 1
1,402 /
1.657 /
1,684 /
— I
1,766 /
I
I
I
I
I
I
I
I
I
I
I
/
I S ’
IS’
IS’
I
I
I
If
S
I
.4’
I
I J
I
I
I ?
I?
I?
I?
X
X
x
X
I
I
I
I
/
I
I
I
I
I
.4’
I
I
I
I?
I?
I?
I?
X
I
x
X
I
I
I
I
I
I
I
I
1
I
I
x
X
X
X
x
X
X
X
x
I
X
I
I
I
I
I
1
I
I
I
I
I
x
X
X
x
X
X
X
X
x
x
X
1,2, 4—Trichloro—
benzene
—
2.014
1
I
/
X
I
X
I
x
I
x
4-Bromodiphenyl
ether
—
2,616
1
I
I
X
I
X
I
X
I
x
Ilexachloro—1 • 3—
butadiene
—
2,783
1
I
I
X
I
X
I
X
I
X
in
PARED TO 5 TO DETERMINE RANGES OF SORBENT UTILITY
—I
xEY
Probably
1
Posaibly
1?
hot
00
tikely
(Vq <400 f/g)
I
I?
X
Trap Quantitatively
Desorb Quantitatively
-------
TABLE 10. RkNGES OF UTILIT’ OF SOLD
SORBENTS BASED ON
Sorbent
(Trap)
to
6 m (Desorb)
SKC charcoal
“ 350
to
‘ . 600
An-tbersorb
‘ 450
to
‘ 750
Porapak N
750
to
.1,500
Porapak R
‘ \ 750
to
l,500
Tenax—GC
950
to
>2,800
A more detailed discussion of these evaluations is available in
the literature [ 47].
RE S ULTS
Of the five sorbent materials evaluated for EPA contract 68—02-
2774, a combination of three was selected for use in a portable
miniature collection system. These three sorberit materials have
complimentary adsorption characteristics as indicated by Table 9
and their 6 m ranges of utility. Therefore, this combination of
sorbent materials was selected for incorporation into a multi-
residue sampling system for collection of the twenty selected
compounds. These three sorbents are listed below along with the
major reasons for their selection.
Tenax-GC - The only high temperature (350°C) adsorbent available
which allows the quantitative thermal desorptiori of
low-volatility organic compounds.
Porapak R - One of the highest capacity polymeric adsorbents with
a reasonable background level (better than Porapak N)
and with an overlap in range of utility (ô ) with
Tenax—GC.
Ambersorb XE-340 - Less difficulty anticipated with the desorp-
tion of compounds of intermediate volatility,
fewer detrimental effects by water and reac-
tivity with collected samples than with char-
coal. Also, its range of utility Córn) leaves
the smallest gap between polyiner .c and carbo-
naceous adsorbents in the types of compounds
collected.
j47) Brooks, J. 3., D. S. West, D. 3. David, and 3. D. Mulik. A
Combination Sorbent System for Broad Range Organic Sampling
in Air. In: Proceedings of the Symposium on the Develop-
ment and Usage of Personal Monitors for Exposure and Health
Effect Studies. EPA—600/9—79—032, U.S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina,
June 1979. 525 pp.
28
-------
TABLE 11. ADSORBENT PROPERTIES CHART
Absorbent
?. mp
limit.
C
Cond 0
Leap.
C
0 esb
temp.
C
Chemical
composit ion
Major thermal
decomposition
products
Background
level
.
ItP
e . ig
Capacity
besorJ on
Rari .(e of
utility, in
terms
Tenax—GC
350
320
300
2.6—Diphenyl-
Alkyl benzene
(5530 (nOne do—
tected above
“.1.6
Should efficiently
(trap intermediately
Very amenable to
thermal .jesorptaon
‘°i i0->2 .8OQ
(35/60 mesh)
p-pheuyleibe
oxide
Styrene
Benzone
Alkyl phenols
system back—
ground)
( a n . l less) volatile
compounds with
slightly less ettin-
ity for polar
compounds.
f, ,r i.,terrnedjately
(arid all higher)
volatile compounds.
porapak 6
250
235
150
N—Vinyl
Vinyl Pryroli—
Air conditioning
at 235C. back-
, 111 e
Should efficiently
trap intermediately
Very amenable to
thermal desorption
“150- .l,500
(50/80 mesh)
to
pyrrolidone
desorbing
(and less) volatile
for intermedrately
.
220
pyrrolidone
Pyrrilidiene
upon
is: P00)1 8 220C
(well abovesys—
tern background)
FAIR lSOC
(slightly above
system background)
compounds with
slightly greater
affinity for polar
(and all higher)
volatile compounds.
Porapak 8
190
175
150
H-VInyl
Viny) pyrroli-
POOR (Well above
background)
1.1
Should efficiently
trap Intermediately
Very amenable to
thermal desorption
‘ .150-’.l.SOO
(50/80 mesh)
pyrrolidone
done
syste.
(and less) volatile
for intermediately
Pyrrolidone
compounds with
(and all higher)
.
Pyrrilidiene
slightly greater
affinity for polar
compounds.
volatile compounds.
A.bereorb
XE ’340
>400
320
300
Carbonized
styre ne—
None observed
(after con—
0000 (None detected
tected above
0.6
Should of liciently
trap highly (and all
less) volatile con-
Questionable amena-
bility to thermal
desorption for all
‘t450- ” .750
divinyl
benzene
ditioning
350C + ob—
served on
GCJi )
system
pounds with greater
affinity for polar
compounds.
but highly volatile
hydrocarbons.
SSC Acti-
vated
charcoal
>400
320
300
Carbonized
organics
.
None observed
(5300 (None de-
tected above
system background)
543.60
Should eftictently
trap highly (and all
less) voljtile com-
pounds with much
greater eftinity for
polar compounds.
Questionable anena-
bility to thermal
desorpt ion for all
but highly volatile
hydtocarbons.
.3SO-”-6OO
0
Conditioning.
b iorpt ion.
Cpeuge drop across tapered tubes containing 1.0 q of adsorbent at a 3-liter/mAn flow rate.
psig 6.9 x l0 Pa.
Estimated.
-------
SECTION 7
DEVELOPMENT OF THE Ai ALYTICAL METHOD
OEJ EC TI VE
The analytical technique determined to be most suitable for the
determination of the twenty selected compounds was capillary gas
chromnatogra hy ±(GC) 2 1 combined with mass spectroscopy (MS). In
order to be cost and time effective, it was decided that the cap-
illary gas chrornatographic technique should be developed prior
to interfacing with a mass spectrometer. Ideally, this (CC) 2
technicue was to allow the separation of the twenty compounds of
interest, to be adaptable to mass spectrometric detection, and
to permit the introduction of sorbent tube samples obtained by
either solvent or thermal desorption. This analytical technique
was to be used fc’r the laboratory development and evaluation of
the multi-residue sampling system, as well as for the analyses
of field samples (after MS interfacing).
INSTRUMENTATION
A Hewlett-Packard Model 5840 gas chromnatograph (CC) was dedicated
to this contract during the development of the analytical method.
This instrument had dual flame ionization detectors (FID) and a
printer/plotter terminal with microprocessing and integration
capabilities. In addition, the instrument could be used with
capillary GC columns, in either a split or splitless injection
mode, and with packed GC columns. A liquid nitrogen cryogenic
option was installed in the instrument during the course of the
analytical method development work. This instrument was selected
as being particularly suitable to this contract because a hewlett-
Packard Model 5985 GC/MS instrument was to be used for later
analyses. The chromatograph on the CC/MS instrument is also a
Hewlett-packard Model 5840 CC, almost identical to the instrument
selected for developing the analytical method.
EVALUAT I ON
Columns
A variety of GC capillary columns were evaluated for their suit-
ability for the analysis of the twenty selected compounds. Five
columns were evaluated during the analytical method development
work performed on the Hewlett-Packard Model 5840 CC; a few other
30
-------
columns were tried later during GC/MS interfacing and investigation
of multi-detector GC analyses. The five columns investigated with
the Hewlett-Packard Model 5840 GC included:
• An “ .48 in long, 0.25 nun ID, Carbowax 20 M wall-coated, open-
tubular (WCOT) glass column.
• An ‘ 50 in long, ‘ O.50 mm ID, Carbowax 20 M wall-coated, open-
tubular, glass column.
• An 50 in long, ‘.0.20 nun ID, Carbowax 20 M wall-coated, open-
tubular, fused—silica column.
• An ‘ 50 m long, ‘ 0.20 mm ID, OV—10l wall-coated, open—tubular,
fused—silica column.
• An “.45 in long, . 0.50 nun ID, SP2100 support-coated, open-
tubular (SCOT), glass column.
These columns were coated with three types of liquid phases,
Carbowax 20 N, 011-101, and SP2100. Carbowax 20 M is a polar phase,
and OV-l01 and $P2l00 are non—polar methyl-silicone phases that
are approximately equivalent. Different types of capillary col-
umns also are represented by these five columns including a wide-
bore SCOT glass column, a wide-bore WCOT glass column, a regular-
bore WCOT glass column and two narrow-bore fused silica columns.
These columns were assessed for their ability to chromatograph
the twenty compounds of interest by installing them within the
GC using direct injection port and detector connections. The
headspace samples or standard analytical mixtures of these com-
pounds were injected into the GC. The results of these experi-
ments are compiled in Table 12. Although chromatographic condI-
tions were varied with the evaluation of each column, Table 12
still demonstrates the comparative abilities of the columns to
chromatograph and separate the compounds of interest.
The retention times indicate the columns’ abilities to separate
these compounds. Compounds with retention time differences of
less than ‘ . 0.15 mm usually co—eluted during the analysis of a
mixture. For example, tetrachioroethylene (RT 5.25 mm) and
acrylonitrile (RT = 5.27 mm) were found to co—elute when ana-
lyzed together on an ‘ O.25 nun ID, Carbowax 20 M WCOT column
(column A in Table 12). All columns were found to have at least
two co-eluting compounds, although which compounds co-eluted
varied from column to column. To help separate compounds as best
as they could be separated, all columns but one were evaluated
using close to optimal conditions. These conditions included
very slow temperature programming (2 or 3°C/mm) from a low
temperature (0° or 35°C) to an upper temperature ‘ . 30°C below
the 0 1 nng’temperaturelimitE The exception was the wide-bore
31
-------
prop*n.8 v. uneJ to dpter iLnp eoLuui dead volu.ue.
9 lyvol is a polyi.erlzation product of ethylene oxide.
NE - Not evalnetsd
- Not detected
— UnknoWn
PT — P,tentioa tlue
Peak ShapeT
I - Excellent
2 - Good
3 - f air
4 — Poor
S — Very poor
- Carbowax 20 N wall-coated. op n-tuhuiat. glaxa colunn. ‘0.25 p In. ‘48 e long
P - Carbowax 20 N wail-coated. open-tubular. glans coluevi. ‘-050 sa. TI). 30 a long
C — Cathowax 20 P4 wall—coated. open-tubular. Iuned—nilfra cp ’Iuoo. 90.20 ixa ID. ‘.50 a long
0 — OV-I0t wall—coated. open-tubular. i,sed—nl Ilca . oIupn. ‘ 1 1.2 iii. II). ‘- 0 a long
- 5P2100 su 1 g ort-coated. o 1 .en- (ul,ular. glass cnl,pan. (1.60 ni, ID. 945 a long
Cniu .ound
TABLE 12. CIIROMATOCRAPHIC PROPERTIES OF TIlE TWENTY SELECTED COMPOUNDS
ON VARIOUS CAPIIPIJARY COLUMNS
Co I nixi A Loluewi 8 ..I inwi e ( . ii’,i, I ) . . .p ,
_!_ L_! _h L0 R .1. 1 1 , 10 I’v .ik li.T. n I H I f . T.,ni ,i I. .j.I . T.. . .n .
Propane
2.84
I
NE
NE
N i-:
NT:
tit
I II:
3’
71
I
6
Ethylene oxide
NF.
N I-
NE
Ni ’
NI
NI-:
PJI:
c .
P.crolein
341 ,
3
NE
Ni’
31 1
2
I I I:
9.27
3
Carbon tetrachioride
3.63
2
NE
NE
3.0 13
I n:
T IE
676
4
Vinyl acetate
3.76
2
NE
NE
4. 110
2
2
P e ezene
4.06
I
NE
NE
4.19
2
9.70
7
Tetrachloroethy iene
5.25
1
NE
NE
5.64
1
18.91
2
930
5
I (crylonitrtle
5.27
3
NE
N E.
5.12
2
5.54
2
7. 3;
Ethylene dichloride
5.99
2
Nt
NE
6.72
i
4.22
2
1 ,4—Dioxane
6.25
2
NE
NE
6.41
2
11.51
17.71
1
Ethylene dibroisida
14.77
2
NE
NE
NE
NE
ji .qi
‘. 17
3
ci a-I,3-Dichloropr o pefla
15.24
2
Nt
NE
NI:
NE
15.70
NE
N C
tran s-l.3—Dlch1oroproPefl
9.26
2
NE
NE
(if
NE
13.70
2
27.74
3
Styrene
17.23
3
NE
NE
tIN
2
25.82
36.4(3
2
27.87
2
Penzyl chloride
34.76
3
2.78
2
(‘N
2
Ethy I.ns
3944
4
372
2
UN
3
Rr,.46
2
2
20. 10
31.79
5
3
l lexa ch loto-i ,3-b uta diefl e
40.22
2
2.69
2
(IN
2
34.71
5
lb luene-2,4-d iaafne
84.83
3
NE
NE
NE
Nt
68.67
2
9.26
5
Di—(2 —ethylhexyL)phthallte
133.01
2
‘95
3
#i
2
151.13
3
ND
38.56
Pent.ch1oroph not
ND
Nt)
ND
NI)
NI)
NI)
2
48.33
S
Pensidine
H I )
NI)
NE
NE
NI)
Nt )
3
64.1,6
3
Citrysena
Pen ro(.)py?ene
Ni)
ND
NI)
ND
N E
lIE
NE
NE
NT)
ND
Nt)
NI)
131.15
213.03
4
9.107
4
-------
( ‘ 0.50 mm ID) Carbowax 20 M WCOT column (column B in Table 12).
Co1urr B was evaluated isothermally near its upper temperature
limit to determine whether high—boiling compounds such as di-
(2-ethylhexyl) phthalate and pentachiorophenol would elute sig-
nificantly faster (in less than ‘ .60 mm) than from a similar
narrow-bore column (column A). Unfortunately, these high—boiling
compounds did not quickly elute from the wide—bore column (column
B) either. Indeed, all compounds in Table 12 indicated as
“not detected” did not appear to elute during the analysis time
allotted (100 to 220 mm). High boiling compounds did not appear
to elute from any of the Carbowax 20 M coated capillary columns
under the analytical conditions evaluated.
How a column interacts with a compound eluting through it is
demonstrated by the peak shape of the resulting chromatogram.
A very sharp, well-shaped peak indicates “good” chromatographic
interaction, whereas a tailing, broad, or skewed peak indicates
“poor” chromatographic interaction. The chromatographic inter-
actions of the twenty selected compounds with the various cap-
illary columns were evaluated by rating the peak shapes of their
chromatograms as indicated in Table 12. Generally, tailing was
the major problem observed and ranged from slight to very ex-
treme. A peak shape rating of at least a 3 (Fair) , preferably a
2 (Good), was considered adequate for permitting compound quan-
titation. The SCOT column (Column E) demonstrated the poorest
chromatographic ability with badly tailing peaks observed for
polar compounds such as ethylene oxide, acrylonitrile, and
ethylene glycol. All of the other columns exhibited reasonable
chromatograrns for most of the compounds with the fused-silica
column giving especiallY good results.
As a result of these analyses, ethylene oxide was eliminated
from the list of compounds to be irvestigated because it so
readily reacted to form ethylene glycol and other products. In-
stead, both of the isomers of 1,3—dichioropropene were to be
evaluated. Ethylene g].yco]. was included in most of the subsequent
analyses as a compound which might be of interest later. Although
ethylene glycol doesn’t exhibit carcinogenic potential [ 1], its
presence in a sample could indicate the possible presence of
ethylene oxide.
Column test mixtures, as suggested in the literature (48], were
prepared to further evaluate the performance of the capillary col-
umns. These mixtures are given in Tables 13 and 14 and typical
chroxnatograzns are shown in Figures 3 and 4. They were useful in
evaluating a number of chromatographic phenomena arising from
column and system characteristics. Specifically, the normal
(48) Cram, S. P., F. J. Yang, A. C. Brown III. Characterization
of High Performance Glass Capillary Gas Chromatography.
10(8), 1977.
33
-------
T ELE 13. COMPOSITION OF COLUMN TEXT MIXTURE lib
Co n poun
A mouri n Conc.,
Density, 10 mLMeOH , ppm
mc (wt/wt)
Amount n
1 mL ir .,
Amount to
(GC) Det.
C,
170.33 0.751 28.9 3,910
3.10
15.5
C .
184.36 0.757 13.3 3,645
2.89
14.5
121.18 0.974 18.7 2,359
1.87
9.4
DMF
122.16 1.036 15.7 1,980
1.57
7.9
2—rthv he ano
130.23 0.834 316 3,986
3.16
15.8
t1er
128.16 1.145 31.0 3,910
3.10
15.5
2—Qctar o e
128.21 0.818 16.0 2,270
1.80
9.00
p. e 08 b
32.04 0.793 — —
aAssu ,nc a 200 to
bs 1 t
I split ratio.
TABLE
14. COMPOSITION OF COLUMN TEST
MIXTURE
lila
Coir pound
Concentration, Amount in 1. iL
.‘g/mL injection, rig
Amount to
detector,a
ng
n—C, 0 312 312 3.12
ri—C , 3 346 346 3.46
n—C, 6 356 356 3.56
158 158 1.58
n—C 304 304 3.04
Dimethyipheriol (L MP) 200 200 2.00
Dimnethylanuline (DMA) 570 570 5.70
2—Octanone 314 314 3.14
2—Ethyl hexanol. 300 300 3.00
Acetoneb
a 100:1 split ratio.
bSlt
34
-------
:c’ ri—i
:0 i .?
I I I S S
.i r
— -4 -4-4
MeOH (SOLVENT)
2-ETHYL
Figure 3.
NAP HTHALENE
1
Chromatogram of column test mixture lib.
1•
C 13
b
35
-------
Figure 4. Chromatograms of column test mixture lila.
c .
C20
C
z
C -,
36
-------
alkanes were used to characterize instrumental band—broadening
effects and to calculate column parameters such as theoretical
plates and separation number. Adsorption effects were indicated
by naphthalene , 2—octanone, and 2-ethyl hexariol. Naphthalene
tails in the presence of any metal adsorptive sites, 2—octonorie
tails in the presence of surfaces or interfaces which act as
Lewis acids, and the alcohol (2-ethyl hexanol) indicates active
siloxyl groups on glass capillary surfaces. The acid—base pair,
2,6—dimethylaflilirle (DMA) and 2,6-dimethyiphenOl (DMP), were used
to determine the acidity/basicitY of the capillary columns.
The chromatographiC performances of the various capillary columns
determined by these columns test mixtures is compiled in Table 15.
Generally, all of the columns demonstrated little or no tailing
for the alcohol and ketone peaks unless the column was deterio-
rated due to age. Naphthalefle was removed from the second test
mixture because the capillary inlet system (to be discussed
shortly) showed no major problems with adsorptive metal sites and
because directly connecting the capillary columns to the GC in-
jection port and detectiOn eliminated metal unions. Chromatog-
graphic efficiency (as demonstrated by HETP in Table 15) was
particularly good on the Carbowax 20 H WCOT column (column A) and
the OV—10i fused-Silica column (column D).
The ease of installing the chromatographic columns was also an im-
portant consideration. ‘Duriflgthe evaluation of the five columns
all were directly connected to theGC detector and injection port .
This is the “preferred” method of installing capillary columns
because it eliminates active sites and dead volumes in various
types of unions, such as unions made with heat-shrink Teflon
tubing or low-dead volume stainless steel fittings. To install
a glass column in a GC in this manner, the ends of the glass cap-
illaries needed to be carefully straightened, cleaned, and de-
activated. During the installation of the glass columns,
extreme care had tO be taken not to break, their ends, or the
procedure of 5 traighteniflg had to be repeated. The wide—bore
glass columns were especially brittle. By comparison, the j ’ :
stallation of the fused-Silica columns was extremely simple àn ’
rapid. The outer polymeric coating and thin wailsof the fused:.
silica columns gave them great enough flexibility and strength
to be directly connected to the GC without prior preparations and
without much risk of breakage.
hromatographiC ConditiOn
Plow rates through a capillary column are very small, ranging from
0.7 xnL/min to 5 mL/min depending on the internal diameter of the
column, the type of carrier gas, and column temperature. To get
maximum 0 omatographiC efficiency from any particular column art
optimum range of flow rate exists which is described by a
37
-------
T1 1ThE 15. CIIROMATOGRAPIIIC PERFORMANCE OF VARIOUS CAPILLARY COLUMNS
Column
Co ound
Theoretical plates,
N, plates
IIETP,
mm/plate
paration
Compounds
nu L
No.opr-ks
tii 7 i
ratio
A
n-Trldecane
%160,000
03
N E
NE
1.18/1
B
n-Eicosane
“ 37,OO0
1.l
NE
NE
NE
C
n-Eicosane
%G6,000
O.7
n—Octadecane
&n—Eicosane
48 -
1.31/1
D
n—Eicosane
%160,000
%O.3
n—Octadecane
&n—Eicosane
39
NE
E
n—Trldecane
“74,000
0.6
n—flodecane
26
O. 4/I
&n—Tridecane
Calculations:
N, plates = I(RT, min)/(W1/2li , mm ) ) 2 5.54
HETP, mm/plate = (I. ., nm )/(N , plates)
KEY TO TABLE AND EQUATIONS :
NE — Not evaluated.
HETP - Height equivalent of a theoretical plate
D MA — 2,6—Dimethylaniline
DMP - 2,6-Dimethyiphenol
N - Theoretical plates
L - Column length
NT - Retention time of compound
Wl/211 — Width at half height of compound
Columns:
A - Carbowax 20 N wall—coated, open-tubular, glass column, “0.25 sun 10, n 48 in long
B - Carbowax 20 N wall-coated, open—tubular glass column, n - 0.50 nun ID, 50 in long
C — Carbowax 20 P4 wall—coated, open-tubular, fused—silica column, ‘0.20 mm ID. i ,50 in long
o — ov-l0l wall—coated, open—tubular, fused-silica column, %0.20 sun ID, ‘50 in long
E — SP2IOO support-coated, open-tubular, glass column, “M.SO mm ID, 45 in long
-------
Van Deemter plot. Such a plot, shown in Figure 5, was determined
for the sP2].00 SCOT column. Injections of propane were used to
determine dead time at various column flow rates. By dividing
the column length by the retention times of propane, the linear
velocities were calculated. Analyses of column test mixture ha
were performed at each of these linear velocities. The height
equivalents of a theoretical plate (HETP) were calculated for
n—tridecane and plotted versus the linear velocities. An optimum
in chromatographic efficiency occurs at a linear velocity of
.2O cm/sec. The relationship of linear velocity, column flow
rate, split flow rate, split ratio, and GC oven temperature was
investigated. These values were determined at a number of GC
oven temperatures and plotted in Figure 6. This demonstrates
that for a temperature programmed analysis, the linear velocity
decreases as oven temperature increases, affecting the column
resolution during the course of the analysis. A linear velocity
of “. l7 cm/sec at the upper CC temperature usually provided op-
timal flow rate conditions during the course of a temperature—
programmed analysis. This usually corresponded to a flow rate
of “0.8 to ‘ .l.2 mL/min at this upper temperature. At the lower
temperature (0°C to 35°C) flow rates ranged from “1.2 to ‘ ..2.0
mL/mia for narrow and regular bore columns, and “.i3 to “6 mL/min
for wide—bore columns. (Capillary column flow rate is controlled
by column head pressure on an HP Model 5840 GC, which results in
changing flow rates during temperature programming. Other in-
struments, however, have flow controllers which can maintain
a constant, optimal flow rate throughout a temperature-programmed
analysis.)
Isothermal analyses of a mixture of low boiling compounds
[ Standard Mixture (GC) 2 Va] were performed at various tentper-
atures to determine whether maintaining sub—ambient temperatures
could improve chromatograpluC resolution. A typical chromatograin
of this mixture is depicted in Figure 7. Two isothermal analyses
of this mixture, at 40°C and at 5°C, are shown in Figure 8. As
indicated by the resolution of 1,4-dioxane and ethylene dichloride,
the analysis at 40°C provides better compound separation. The
reduced chromatograpiliC efficiency at the lower temperature was
thought to be due to a combination of faster linear velocity and
decreased effectiveness of the column liquid phase (in this case
Carbowax 20 14), which has a lower temperature limit of 60°C.
Therefore, aintaifliflg a sub—ambient temperature for a long time
(>10 mm) was found not to improve chromatographic resolution.
Slow temperature programming (2 to 3°C/mm) from sub-ambient to
higher temperatures generally produced the best chromatographic
results.
Capillary Inlet System
The analytical method developed for this contract was to allow
the introduction of sorbent tube samples obtained by either sol-
vent or thermal désorption. Of course, a solvent desorbed sample
39
-------
L HEWLETT PACKARD MODEL 58 OA CC
GAS CHROMATOGRAPH HP 5 ’1DA CC)
3.2 k=1. 11 AT 120°C C 13
3.1) 0.50 mm I. 9. x 45 m ( [ ST.) SP?100 SCOT COLUMN
2.8 EUUM CARRIER
NJ. PORT 250°C
2.6 DETECTOR 300°C ( [ ID)
24 OVEN 120°C ISOTHERMAL
2.2 -
2.0-
E
a. 1.8 -
———
U i —
16
14
o .
1. 2
1.0
08
• 0
0.6
0.4
0.2
I I I 1
0 10 20 30 40 50 60 70 80 90
[ INEAR V [ LOCIr ’ (cm!sec)
Figure 5. Van Decmter p’ot for n uly ic ii column.
-------
- 110
-160
- 150
140
• 130
• 120
• 110
•100
90
80
70
-60
• 50
40
30
-20
- 10
r t o f I ______________________________________
70 - 1.0 - 100
425 lIP 5840 AGC
400 2V OPEN 1.5 ps q
IVOP [ N —3 ,5p9p
60 -3. 15 6.0 - 600 INJ. PORT 250 C
DETECTOR 300°C (FID)
3. 50
0.5 mm 1.0. x 45 m ( [ ST.) 5P2100 SCOT COLUMN
3.25
50 5.0 500
300 REUNTION TIME OF C
2.15
40 -2.50 4.0 400
2.25 ._._ I
-------
SIARI
AC ROIF IN
CARBON TETRACIIWRII3 [
TETRACHWRO [ THfl(NE
Fiqure 7.
STOP
Typical chromLltOeram of ndard Mixture (CC) 2 V ..
*9.17
ETIIVEiNE DICHIORIDE
VINYt ACETATE
(1 EN/F NE
ACF YLONt1Iflt-E
OXANE
-------
A. 40°C isothermal
analysis
Figure 8.
B. 5°C isothermal
analysis
Chrornatograms of Standard Mixture
(GC) 2 Va.
43
-------
can he introduced into a chromatograph by syringe injection. How-
ever, a thermally desorbed sorbent sample must be reconcentrated
as it is slowly desorbing from a sorbent tube so that the entire
sample can be introduced into the GC as a discrete slug. Such
a discrete slug retains the chromatographic resolution capabilities
of the analytical system.
An inlet svster was designed and constructed to permit the thermal
desorption of sorbent tubes, reconcentration of the desorbed
of the desorbed samples in a trap, and the introduction of these
samples into the capillary GC for analysis. The system was corn—
prised of three major parts: a six—port, two-position high temper-
ature valve; a sorbent-tube desorption chamber; and, a nickel
capillary trap. A schematic of this system is shown in Figure 9.
In the ‘trap position a sorbent tube was desorbed by the tube
furnace and its sample swept by the auxiliary “desorptiOfl” carrier
through the valve into the nickel capillary trap, which was 5O%
submerged in liquid nitrogen. The flow out of the trap was passed
through the “A” injection port to the NA” FID for monitoring trap
break-through. When the valve was switched to the “analyze”
position, simultaneously the tube furnace was cooled, the liquid
nitrogen was removed from the trap, and the trap was heated. The
auxiliary “purge trap’ flow then swept the collected sample out
of the trap and into the capillary injection port. When the capil-
lary GC was operated in the “split” mode, most of the sample enter-
ing the injection port was vented and only a small portion entered
the capillary column for chromatographic separation and analysis.
In this injection mode, vented portions of samples could be re-
collected onto other sorbent tubes for subsequent analyses.
The auxiliary “desorptiorl” carrier was obtained simply by split-
ting flow off the instrument carrier supply line. However, the
auxiliary “purge trap’ carrier, which was responsible for sweeping
a sample from the nickel capillary trap into the capillary injec-
tion port, was controlled by the capillary injection port
pneumatics. This was accomplished, as sketched in Figure 10, by
installing a tee after toggle valve 1. In this configuration part
of the injection port carrier would purge the capillary inlet trap
and introduce samples into the injection port through a needle.
The rest of the injection port carrier would purge the GC injec-
tion port in a normal manner, sweeping the introduced (through
the needle) sample into the GC capi1lary co1utnn.
Initially, thermal desportiOn of sorbent tubes was accomplished by
a copper tube furnace wrapped with nichrome heating wire. A
thermocouple was incorporated with this furnace arid the unit was
wrapped with asbestos and/or glass tape insulation. The temper-
ature of this tube furnace was controlled by a Variac, and was
capable of ballistic heating and cooling. A digital voltmeter
was used to measure the furnace’s temperature. Although the
ballistic heating and cooling capabilities of this type of furnace
were found to be excellent, after constant use and manipulation,
44
-------
AUXILIARY
TRAP MODE
SAMPLE DESORBED FROM
SORBENT TUBE INTO TRAP
COLUMN CARRIER
ii mUmin)
ANALYZE MODE
SAMPLE DESOR BED FROM TRAP
INTO CAPILLARY COLUMN
SPLIT ‘ LLARY “A”
- 2 mUmin)
INJECTION PORT INJECTION PORT
CAPILLARY COLUMN. EMPTY NICKEL
AND “3”FID COLUMN, AND
“A” FID
Figure 9.
Flow schematic of capillary inlet system.
SOR BENT
TUBE
AUXILIARY “PURGE
TRAP” CAPILlARY
CARRIER I—2U) mLimIfl)
UQUID
CAPILLARY
INJECTION PORT,
CAPILLARY COLUMN,
AND “8” FID
“A”
INJECTION PORT,
EMPTY NICKEL
COLUMN, AND
“A” FID
SOR BENT
TUBE
AUXILIARY “ PURGE
TRAP” CAPILLARY
CARRIER I—2 mUminI
45
-------
SI’I I MOOF
TO CAPIllARY
INLU SYSI [ M
AUXILIARY
“PURG( TRAP”
CARRIER
Schematic of flow pallern at the
capillary GC injccLion port.
INJECTION PORT
!IOW MODULE
Ficjure 10.
-------
the insulation of the furnace would deteriorate and the thermo-
couple would short to the heating wire. Consequently a different,
more durable, desorption chamber was designed and constructed.
This desorption chamber is pictured in Figure 11. It consisted of
an electrically heated aluminum block hinged in two halves with
each half provided with its own thermocouple and heating element.
Sorbent tubes were desorbed by being placed in the center of this
device with its hinged “doors” enclosed around them. The advan-
tages of this desorption chamber included:
• Easy mounting/dismounting of the entire unit by attachment/
detachment of its mounting bars to a ring stand;
• Ballistic heating and cooling of sorbent tubes by closing
and opening the chamber’s hinged doors, while maintaining
the aluminum block at desorption temperatures;
• Durability due to the sturdiness of design (fragile parts
embedded in the aluminum block) and simplicity of frequent
manipulation (just open “doors” to change sorbent tubes);
and
• Non—interference with sample analyses since the chamber
was not within the desorption flow system.
Another improvement made to the capillary inlet system was the
design and construction of a thermal control unit, sketched in
Figure 12. This unit could accommodate all of the heating
requirements of the capillary inlet system, including power
supply and thermocouple monitoring. Five voltage regulators
were used to supply power to the various parts of the capillary
inlet system, replacing the Variacs that were previously used to
perform this function. A transformer used to heat the nickel
capillary trap by electrical resistance was incorporated into
the thermal control unit. The voltage regulators are controlled
by on/off switches, of which two can be activated/deactivated
by timing devices. DesorptiOn of the nickel trap by resistive
heating was usually controlled by one of these timers to prevent
the transformer from overheating. Pilot lights were also included
to indicate the status of the on/off switches and intensity in-
dicator lamps visually displayed the voltage settings of the
regulators. A pyrometer attached to a multi-positiOn selector
switch permitted the alternate monitoring of up to twelve
thermocouples. A sketch of the “improved” capillary inlet
system is given in Figure 13.
Another improvement to the capillary inlet system, made after
and as a result of analytical method development experiments and
evaluation of sorbent tubes, was a slight change in the design
of the nickel capillary trap. The two trap designs are shown
in Figure 14. Note that the major difference in the two de-
signs is that the “newer” trap has zones both above and below
47
-------
Figure 11. Front and rear view photograph of new, durable
capillary inlet system desorption chamber.
‘U
MOUNTING BARI
I.
aEcTRIcAL
CONNECTIONS
I
-------
PYROMETER
THERMOCOUPt.E
S ELECT
INTENSITY
INDICATOR
LAMPS
VOLTAGE REGULATORS
(i.e. VARIACS)
Figure 12. Sketch of capillary inlet system thermal control unit.
-------
t: F
C
HE 11TT-PA KARD MODEL 5S4OA CC
WITh CLASS cAPILLARY OPTION
TUBE DESORPTION CHAMBER (OVEN)
(53 HEATED VALVE OVEN-CONTAINING
6-POR 2-POSITION. LOW DEAD
VOLUME VALVE
13 1-METER, N CAPILLARY TRANSAXIAL
TRAP
® DEWAR CONTAINING LIQUID N 2 FOR COOLING
CAPILLARY TRAP
(j) THERMAL CONTROLLER FOR INSTANT HEATING
OF CAPILLARY TRAP, HEATING DESORPTION CHAMBER
HEAlING VALVL ETC.
(1) PYROMETER FOR TEMPERATURE READINGS
WITH MULTI-POSITION SELECTOR
( ) FLOW CONTROU.ERS AND SPLIT CONTROLLER
(13 RECOLLECTION VENT WITH TUBE
® DETECTOR ADAPTABLE TO FID,P4-P FID,
GLASS JET RD
Figure 13. Capillary GC with capillary inlet system.
50
-------
ELECTR I CAL —
CON NECT I ON
L VALVE
4 FROST
0 0 C
— p—’
— - —
A. “OLD Ni CAPILLARY TRAP DESIGN
(—lmeterLONGx —0.l6mm I.D.)
—10°C to
20 (
FROST,— 0°C
(
k-’- -20°C
-20°C -
NITROGEN
TO VALVE
—0°C
ELECTR I CAL
CONNECT I ON
— —
B. “NEW” Ni CAPILLARY TRAP DESIGN
1 —1.5 meter LONG x—0.51 mm 1.0.)
Figure 14. Comparison of trap designs and their temperature
zones during cryogenic sample reconcentration.
01
— — —
-------
0 ’C durinc cryocenic sample reconcentration. This was found to
reduce the probiem of cryogenic trap breakthrough due to the
presence of significant levels of water in samples.
The effect of this capillary inlet system on the chromatographic
performance of the analytical system was evaluated by analyzing
samples of the column test mixtures previously described. Im-
mediately prior to installing the capillary inlet system or the
GC, a test mixture would be analyzed to provide a basis for com-
parinc results. Once the system was installed and demonstrated
to be free of leaks, samples of the column test mixture were in-
jected into the inlet system. These injections were either made
into a stainless steel injection oven (installed in place of the
sorber.t tube desorptiori chamber) or through a tee into an empty
glass tube enclosed by the desorption chamber (Figure 15). These
samples were then cryogenically reconcentrated in the nickel
trap before being introduced into the CC. The results of these
evaluations are given in Table 16. The reproducibility of re-
sults for several injections of Standard Mixture Va (Figure 7)
was also evaluated. Injections were made through a tee into a
heated empty glass tube (Figure 15) . Tables 17 and 18 summarize
the results obtained. Overall, the capillary inlet system ap-
peared to perform very well. Reproducibility for standards was
reasonably good, the effect on chromatographic efficiency was
slight, and no noticable difference in peak shapes were noted
for any of the twenty selected compounds or the compounds in the
column test mixtures.
Injection Mode
In order to detect between 1 and 100 parts per trillion concen-
trations of various compounds in ambient air, one must be able
to collect at least 10 ng of a compound and then detect this
amount (10 rig) or less during sample analysis. To obtain max-
imum sensitivity for an analysis, the entire sample contained on
a sorbent tube should be introduced into the capillary column
(after cryogenic reconcentration) by using the “splitless” in-
jectior’. mode. However, such a “one—shot” analysis does not per-
mit a sorbent tube sample to be split and part recollected on
another sorbent tube for additional analyses. Therefore,
analytical sensitivity must be weighed against the requirement
for multiple sample analyses for any application.
The flame ionization detectors of the Hewlett—Packard r . odel
5840 GC were estimated to have a detection limit of 0.01 rig to
1 ng for most compounds. Therefore most laboratory evaluations
were conducted with the GC in the “split” injection mode, usually
with a split ratio of 100 to 1. The “splitless” mode of sample
introduction, however, was briefly investigated to evaluate the
performance of the analytical system under these conditions.
Two problems were noted. One was the analytical interference
of the solvent with the compounds of interest in standard
52
-------
TABLE 16. EFFECT OF CAPILLARY INLET SYSTEM ON THE
CHROMATOGRAPHIC PERFORMANCE OF VARIOUS
CAPILLARY COLUMNS
Inlet
system
Trap
design
Column
Compound
Theoretical plates,
N, plates
mm/plate
No
Yes
“Old”
A
A
n—Tridecafle
n—Tridecafle
%160,000
“ .84,000
‘ 0.3
.0.6
No
Yes
“New”
D
D
n-Octadecafle
ri-Octadecane
“ . .].50,0O0
‘ 53,000
“ ‘ 0.9
No
Yes
“Old”
E
E
n—Tride cafle
n—TrideCarie
‘ 74,OO0
‘ . .68,OO0
“ 0.7
Calculations:
N, plate [ (RT, mjn)/(Wl/2H, mm) ) 2 5.54
HETP, mm/plate = (L, xnm)/(N, plates)
KEY TO TABLE AND CALCULATIONS :
METP - Height equivalent of a theoretical plate
N - Theoretical plates
RT - Retention time of compound
Wl/2H - Width at half height of compound
L - Coli.iifln length
Columns:
A - Carbowax 20 ti wall-coated, open-tubular, glass column,
“ .0.25 nun ID, “.48 in long
D - OV—lOl wall-coated, open tubular, fused—silica coluitifl,
‘ ‘0.20 mm ID, 50 in long
E — SP2100 support-coateds open—tubular glass column,
“.0.50 nun ID, ‘ .45 in long
53
-------
SEPTUM
FUSED 114” SWAGELOK CAP
WITH CENTRAL OPENING
DESOR P11 ON CARRIER
TO CAPILLARY INLET SYSTEM
AND GAS CHROMATOGRAPH
DESORPTION CHAMBER
SORBENT TUBE OR
EMPTY GLASS TUBE
118” SWAGELOK TEE
U ’
Figure 15.
Injection system with modified septum tee.
-------
I -n
U i
TABLE 17. REPRODUCIBILITY OF RETENTION TIMES
Retention timf?S, min
Compound Run #468 Run #46’) Run #470 Run #471
Run #472 Pun #473
Stdtistics
Mc an psob,
Acetone (solvent) 3.60 3.60 3.60 3.59
Acrolein 3.74 3.74 3.75 3.73
Carbon tetrachloride 3.92 3.92 3.92 3.90
Vinyl acetate 4.05 4.05 4.05 4.03
Benzene 4.46 4.46 4.45 4.43
Acrylonitrile 5.23 5.23 5.21 5.17
Tetrachioroethylene 5.66 5.67 5.65 5.60
] 1 4-Dioxane 6.58 6.58 6.59 6.50
Ethylene dichioride 6.92 6.92 6.88 6.82
3.58 3.58
3.72 3.71
3.89 3.88
4.01 4.00
4.41 4.39
5.14 5.12
5.57 5.54
6.47 6.41
6.77 6.72
3.59 0.3
3.73 0.4
3.91 0.5
4.03 0.6
4.43 0.6
5.18 0.9
5.62 0.9
6.52 1.1
6.84 1.2
aTh 10 mm reconcentration time is subtracted.
b 1 standard deviation.
TABLE 18. REPRODUCIBILITY OF AREA
INTEGRATIONS
Statistics
Corrected for Corrected for
size of solvent
140 correction
Area Integrated. Computer counts
0472 0473
injection
pso .
Mean
Mean S Mean I
Co. ound Pun 1468 Run 0469 Pun 0479 Run Run
Acetone (solvent) 1,409.000 1.390.000 1,403.000 1,312,000 1.314,000 1.339,000
2,378
1.316.16) 2.9 — —
2,231 4.8 2.380 3.3
1,381,167 1.8
2.347 1.9
Acro lein 2,340 2,352 2,217 2.404
556 5.7 591 5.5
583 3.9
Carbon tetrachloride 569 552 584 605 576
2.201 2.9 2.341 0.8
2,310 2.4
VInyl acetate 2,360 2,354 2,345 2,272 2,308 2.221
6.868 2.8 7,305 0.3
7,207 1.8
Benzene 7,328 7,236 7.340 7,164 7,168
2.122 14.7 2,895 14.3
2,851 12.9
Acrytonitrile 2,535 2,469 2,554 3,199 3,220 3,110
1.055
993 11.3 1,055 9.9
1,040 8.8
Tetrach loroethylene 992 941 966 1.118 1,109
5.328
5.228 2.7 5,561 0.)
5,481 1.8
1,4-Dtosane 5,577 5,519 5,602 5,444 5,449
1.288 11.0 1,170 9.9
1.350 9.0
Ethylene dibrumide 1,250 1.216 1.307 1,496 1,500
a 11 . standard deviation.
-------
solutions. This is demonstrated in Figure 16 for Standard Mixture
(GC) 2 Va (Figure 7). The other problem was the tremendous increase
( lOO fold) in the level of background interferences. Therefore,
if instrumental sensitivity is sufficient to permit the detection
of compounds in a sample at an acceptable level (0.1 to 10 ng),
the “split” inode of injection provides better analytical perform-
ance (i.e. , fewer interferences) and potential multiple analyses.
RESULTS
The development of the analytical method resulted in an analytical
technique which provides major analytical conditions and hardware
requirements. This technique allows the separation of most of
the twenty selected compounds and permits the introduction of
sorbent tube samples obtained by solvent or thermal desorption.
An inlet system (Figure 13), shown to efficiently and reproducibly
introduce samples into a GC, provides for the introduction of
samples thermally desorbed from sorbent tubes. The capillary col-
umn found to chromatograph most of the twenty selected compounds
adequately (Table 12), to provide a high chrornatographic efficiency
(Table 15) , and to allow rapid, easy installation, was on n 50 m long,
‘ O.2 mn ID, fused-silica, OV—101 WCOT column. A glass WCOT column
of the same (OV—lOl) or equivalent (5? 2100) liquid phase also
would probably provide similar chroinatographic results, although
installation would be more difficult. Fused—silica or glass
WCOT columns coated with Carbowax 20 Mprovided complimentary resol-
ution of compounds (i.e., compounds changed elution order), but
did not chromatograph high—boiling compounds. Such polar c cl-
umris could be used secondarily for multiple sample analyses if
further confirmation of compound identities is desired. For any
analytical column, a slow temperature—programmed analysis (2 to
3°C/mm), from a low temperature (0°C to 35°C) to a temperature
‘ 30°C below the column temperature limit, provides the best
analytical conditions for compound resolution. However, analysis
for high-boiling compounds with these analytical conditions is
imnpracticalJ.y large (213 mm for benzo(a)pyrene). Therefore a
faster temperature-programming rate (5 to 10°C/mm) will probably
be required to analyze “field” samples within reasonable time
periods (45 to 90 mm). At the upper analytical temperature, a
column flow rate adjusted to obtain a linear velocity of 17 to
20 cm/sec (with helium carrier) provides optimal conditions for
chromatographic efficiency. Injection of a sorbent taken sample
with the GC in the “split” injection mode permits recollection
of the split portion for multiple analyses and keeps background
interferences to a minimum. However, this injection mode reduces
analytical sensitivity and its (“split” injection) advantages must
be weighed against sensitivity requirements.
56
-------
B. Splitless injection of Standard
Mixture (GC) 2 Va.
C. Splitless injection of Standard
Mixture (GC) 2 Va 100—fold dilution.
Ficiure 16.
Chromatograms of Standard Mixture (GC) 2 Va comparing
split and splitless modes of injection.
l -
q •
_4 __. —
— .
0
‘I;
U’
I ’
•1
a
a
A. 200:1 Split of Standard
Mixture (GC) 2 Va.
-------
In the actual analyses of field samples certain changes were made
in the analytical technique owing to either improvement or con-
venience of operation with the GC/MS system, or conditions/proper-
ties ci the particular samples (e.g., high water content). These
changes are described in detail in Section 10. Most notable among
these was the use of splitless injection and capillary direct in-
jection technicues arid the modification of the GC temperature
program for samples with high water content. The following suit-
marizes the preferred technique used for the analyses:
Instrument: Hewlett—Packard Model 59855 GC/MS
Column: Methyl Silicone (OVIO1, SE-30,
SP2100 or equivalent) capillary
(fused silica or glass) , 50 m,
0.2—0.25 imn ID.
Temperature Program: Subambient (-30°C) during tube
desorption. Rapid rise ( 30°C/
mm) to 0°C. More gradual
temperature rise ( 8°C/min) to
30° below upper temperature
limit of column.
Inlet System: utech Model 320 thermal desorption
system, capillary direct coupling.
Mass Spectrometer: Scan Mode
58
-------
SECTION 8
DEVELOPMENT OF THE SAMPLING SYSTEM
OBJECTIVE
A sampling system was to be developed so that the twenty
selected potential carcinogens could be collected effectively
for subsequent laboratory analyses. This system was to
incorporate a combination of solid sorbents of complimentary
adsorption characteristics. As discussed in Section 6, the
sorbent materials selected were Tenax—GC, Porapak R, and
Arnbersorb XE-340. Cartridges containing these sorbents were
to be included in a sampling system fabricated for the easy,
efficient collection of ambient air samples.
INSTRUMENTATION
The analytical system, described in Section 7, was developed
concurrently with the development and evaluation of the sampling
technique. This analytical system was used in a variety of
configurations (particularly with various capillary columns)
for the analyses of sorbent cartridges. The analyses of the
sorbent cartridge were measured against analytical standards
to allow the comparison of results obtained with the various
analytical configurations. Vaporous samples of the more vola-
tile compounds of interest were produced with the Sample
Generation System described in Appendix B. These samples
were “known” artificial atmospheric environments which could
be sampled with sorbent cartridges. This allowed the evaluation
of the sorbent cartridges under somewhat realistic sampling
conditions.
EVALUATION
Design Criteria
The first step in the development of the ambient air collection
system was to evaluate the basic design options and criteria of
an air sampling system. The most basic sorbent sampling system
(Figure 17) is comprised of only four major elements: a particu-
late filter, usually placed at the inlet of a system; a cartridge
to contain the sorbent material; a pumP to draw the air sample
and, tubing to connect the filter, cartridge, and pump. HOW
ever, specific parameters such as flow rate, cartridge design,
59 -
-------
PART ICU LATE
FILTER
SAMPLING TUBE
AIR ; __________
Figure 17. Basic sampling system.
and materials of construction need to be adapted to the specific
sampling application. Therefore, these four basic elements and
various sampling parameters were assessed with respect to this
project.
Since some of the compounds of interest, such as acrylonitrile
and benzvl chloride, tended to be reactive, inert materials were
chosen for the fabrication of the sampling system. Silanized
pyrex glass was the material selected for the sorbent cartridges.
Similarly, silanized pyrex glass wool was chosen for containing
the sorbents within cartridges and for use as a particulate
filter. Interconnecting tubing was preferably inert, flexible,
and non-contaminating, so Teflon was selected as the most suitable
material. Stainless steel was chosen for connecting fittings.
Another area of consideration was the size, weight, cost, and
durability of the sampling system versus the sampling capabilities
of the pump. Since a portable system was being developed under
EP?. Contract 68-02-2774, a larger system was desired for this
contract. The desk-top type of sampling system, shown in Fig-
ure 18, was selected for investigation. This system would be an
area-monitoring (versus personal-monitoring ) sampling system,
portable enough to be hand-carried to field locations without
heavy equipment. In comparison to the portable system developed
for the other contract, this system would be more durable and
be capable of sampling at higher flow rates. The size and weight
would be greater but the cost would be approximately the same
as the portable system. Although a large vacuum pump could be
placed in such a system, a more moderately sized pump was thought
to be potentially more suitable, especially because of weight
considerations. The sampling capabilities of two pumps consid-
ered are shown in Figure 19.
Combining the three selected sorberit materials within a single
sampling system also had to be contemplated. In a previous
application for the collection of jet exhaust fumes, different
PUMP
60
-------
Sampling Tube (3)
Holder
WE IGET
COST
SIZE
OPERATION TIME
METHOD OF SETTING
FLOW
DURABILITY PROBLEMS
Battery Pump
(May have 2)
Samp’ing Ports
(Hoses Opt.)
— ‘ 32 lbs (‘ 6O if 2 batteries)
- $385 (for case, 1 battery, charger
and pump only)
— 13” (long) x 8” (wide) x 5” (deep)
— 4—6 hrs (>8 hrs with 2 batteries
or wall adaptor)a
- Needle valves and flow meters or
critical orificesb
- Needle valves (if used) fragile
aBattery may be put into a separate pack and/or a 120V line
with transformer may be used to adapt this sampler to a
wall socket.
bDue to pressure drop required at orifices the capabilities
would be reduced.
Figure 18. Sketch and description of large desk-top sampler.
61
-------
3c
FLOW RATE, 1/mm
Figure 19.
Plots of requirements for various sampling
arrangements and the manufacturer’s speci-
fications for two lar’ge (desk-top) pumps.
250
2
150
‘I . ,
50
10
2 4 6 8 10
12
62
-------
sorbents were combined within a single cartridge [ 49). However,
the three sorbents selected for this application did not have
compatible temperature limits. The temperature limit of
orapak R (250°C) is significantly less than Tenax—GC (350°C) and
Arnbersorb XE-340 (>400°C). Therefore, if thermal desorption was
used to analyze these tubes, a low temperature would be required
for sorbents combined in a single cartridge, or optimal desorp-
tion temperatures would be possible for sorbents in separate
cartridges. The latter arrangement was chosen for investigation.
The separation of sorbents into different sorbent cartridges
allows the additional option of parallel or series sampling.
Table 19 qualitatively describes the performance of the three
sorbent materials anticipated for these two sampling modes. It
assumes the m range (for the sorbent cartridges given in Table 9
(in Section 6) which was based on l g per sorbent and a sample
volume of 480 L. An evaluation of the two sampling modes was to
be performed during laboratory development of the sampling system.
The relationship of atmospheric concentration, analytical detec-
tion limit, sample volumes, and sorbent capacities was investi-
gated with respect to sampling limitations. Concentration and
detection of organic compounds at atmospheric concentrations of
“1 to “slO parts per trillion (ppt)a was set as an objective.
Assuming an analytical detection limit of 10 ng, Figure 20
shows the relationship between atmospheric concentration and
sample volume necessary to collect a detectable sample of benzene.
The larger the volume of sample collected, the lower the atmos-
pheric concentration which is detectable. However, there are
limitations to the amount of sample that can be collected on
sorbent materials. One limitation is the capacity of a sorbent
or sorbent system for any particular compound. The capacities
per gram of sorbent of Tenax-GC, porapak R, and Ainbersorb XE-340
for benzene are also shown in Figure 20. As indicated, if a
sample is collected at a rate of 3 L/min for 8 hours, the
capacities of 1 gram of Tenax and Porapak R for benzene would
be exceeded. Benzerie would “break through” sorbent cartridges
containing 1 g of these sorbents before being concentrated to
an analytically detectable level. However, the capacity of 1 g
of Aznbersorb XE-340 for benzene is much greater than the sample
volume indicated. By collecting that sample volume, atmos-
pheric concentrations of benzene could be detected down to
2 ppt with an Ambersorb cartridge. Similar relationships
a 1 ppt X 24450 = 1 ng/L (in air at STP)
(49] Brooks, J. J., D. S. West, J. E. Strobel, and L. Stamper.
Jet Exhaust Analysis by Subtractive Chromatography.
SAM-TR-78-37, USAF School of Aerospace Medicine, Brooks
Air Force Base, Texas, Dec. 1978. 76 pp.
63
-------
TAflIJE 19. EVALUATION OF SORF5ENT ACK)\GE FOR TWO MODES
_____ Spries Parallel
?ena s-GC
• %9 O to 7, OO)
a
QuantItatively trap all
low and some inter—
mediately vol.tIIe
compounds •nd par-
tially ( ‘100%) trap
other compounds.
Quantitatively dc n,b
(ajisnet) all c-omç ,o ,in4c
trapped. including
those only partially
trapped.
QuantItatively trap all
l Ow and some intet
mediately volatile
COISpOIIHdS, aiR) par-
tially (‘100% ) trap
alt other compounds.
çiriantiratively de .urh
(airsn tl alL compounds
tia; ’pcol, *nrlurl&riq
tlwcp only f.QlrtialIy
tI al.p ed.
O ’i
Porapak R
•tlSO to “1,500)
Quantitatively trap all
*ntera,edtate)y vola—
tile compounds pas-
slug through the
lenas, and partially
(‘100%) trap all
other compounds pas-
aing through Tenav.
çuantitatively desorb all
compounds trapped from
Tenax breakthrough, in-
ciuding those only par-
tially (cloos) trapped.
Quantitatively trap all
low and intermediately
volatile compounds,
and partially ( <100 ’ )
trap alt other
compounds.
Quantttat ively decorb
,v ct of the lnt rnedt-
atety volatile and
partially trapped com—
i ’ r’ s, and partially
(sIOO%) desoib so me
interisediate y and low
volatility coepounds.
e:,orb E-)4O
(6_. t450 to “150)
Quantitatively trap
moat of the highly
volatile cc.çounds
breaking through both
the 1ena* and the
Porapak with other
more volatile cow—
1 ounds being partial-
ly t(L00%) tiapped.
Quantitatively deqoib most
of the compounds trapped,
except possibly those
highly volatile cow-
pouniL that border on
iiiter.ediate volatility
which may only partial-
ly ( 100%) desorb.
QuantitatfveLy trap all
of the low and inter—
mediately volatile
compounds and st of
the highly volatile
compounds aol partially
1’100%) trap even more
highly volatile
compounds.
Quantitatively decorb
most of the highly
volatile quantitatively
•nd partially trapped
tnmpound , partially
desorb ( < 100%) some
Intermediately land
border) volatile com-
pounds, and not desorb
I O%) most other
compounds.
Sor bent
Itesor b
Ira i’
l)e ’Oi h
-------
io2
a
z
b
1
10•1
io.2
io
Figure 20. Relationship between concentration, volume sampled,
and approximate detection limit (10 ng) for benzene
with sorbent capacities superimposed (assuming
1 grain of sorbent).
10
1
10
VOLUME SAMPLED. L
io6
65
-------
between sorbent capacity, atmospheric concentration, sample
volume, and analytical detection limit exist for all atmospheric
pollutants. Thus, for any given sample volume, different groups
of atmospheric pollutants will be concentrated to analytically
detectable levels on different sorberit materials. The lowest
detectable atmospheric concentration of any particular pollutant
is determined by the sample volume, as long as that volume
doesn’t exceed the collecting sorbent’s capacity for that
pollutant.
Theoretically, one could decrease the detectable atmospheric
concentration of a pollutant to an infinitely small level by
greatly increasinc the sample volume, as long as the amount of
sorbent is increased to quantitatively collect the pollutant
with that laroer sample volume. However, there are practical,
physical limitations to this approach. The major limitation is
the sampling capabilities of pumps. For any pump there are
limits to the flow rate capable of being drawn over a given pres-
sure drop, as indicated in Figure 19. For a given period of time,
the sample volume capable of being drawn by a pump is directly
related to the pressure drop across the system in which it is
incorporated. For sorbent cartridges, the pressure drop across
a single cartridge is determined by the amount, type, and mesh
size of the sorbent material, the diameter of the tube, the ty e
and amount of material (e.g., glass wool) used to contain the
sorbent, and the flow rate through the tubes. For a multiple
cartridge s\’stern, series or parallel arrangement of cartridges,
interconnecting tubing, and any flow controlling valves also
add to the pressure drop across a system. Therefore, minimum
detectable atmospheric concentrations of pollutants are limited
by both sorbent capacity and the physical design of a sampling
system.
The physical design of the sampling system for this project was
developed so that sorbent capacities were maximized and pressure
drops minimized (thereby allowing any pump to draw a maximum
sample volume). The first step was to select a cartridge design
for the three chosen sorbent materials which would potentially
allow the efficient collection of the twenty selected potential
carcinogens. Under EPA Contract 68-02-2774, pressure drop
studies were conducted with a variety of sorbent tube designs
[ 47). The tube design depicted in Figure 21 was selected as
offering a reasonable compromise between the requirement to have
the smallest possible diameter for efficient thermal desorption
(theoretically more sensitive than solvent desorption) and an
acceptable pressure drop. With this design, glass cartridges could
contain l g of Tenax, “ l.5 g of Porapak R and “ .3 g of Aznbersorb.
The abilities of these sorbent cartridges to collect the twenty
compounds of interest was estimated by determining the molecular
weight modified solubility parameters (óm) of the twenty selected
compounds. These compounds were arranged with the test matrix
compounds (previously discussed in Section 6) according to
66
-------
Figure 21. Specifications of selected
sampling tube design.
increasing value. The abilities of the sorbent cartridges to
collect and desorb these compounds were qualitatively evaluated
for a sample volume of “ l,500 L. The results of these evalua-
tions are suxtunarized in Table 20. The enclosed regions indicate
compounds anticipated to be quantitatively collected and desorbed
with these cartridges using this sample volume. As indicated,
all of the twenty compounds of interest theoretically are effec-
tively sampled with sorbent cartridges of this design. Therefore,
this sorbent cartridge design (Figure 21) was tentatively selected
for use in the Ambient Air Collection System pending laboratory
evaluation.
Sorbent Cartridges
Conditioning- -
Generally, all commercially available sorbent materials, includ-
ing the three selected for this project, contain significant
levels of background contamination upon receipt as the result
of production, packaging, and/or transportation. Cleanup pro-
cedures for sorbents range from short ( l hr) thermal condition-
ing to exhaustive conditioning techniques involving extraction
with multiple solvents and rigorous (>8 hr) thermal conditioning
[ 46). Since the sorbent cartridges for this project were to
collect ambient air samples containing trace levels of organic
compounds, an exhaustive cleanup procedure was selected. This
procedure is described in Appendix C. “Blank” sorbent cartridges
prepared according to these procedures and analyzed with the
analytical system described in Section 7 were found to have
acceptable backgrounds, as demonstrated by the chromatogramS in
Figure 22.
Capacity - -
The ability of the sorbent cartridge to retain some of the com-
pounds of interest was briefly investigated. As discussed in
Section 6, .a chromatographic technique was used to estimate
sorberit capacities (Table 7 in Section 6) which were then related
SORBENT MATERIAL
67
-------
TABLE 20. SUMM RY OF ANTICIPATED SAMPLING
CHARCTERISTICS AND CAPABILITIES
General
infornation
Sorbent
ranoes
of utility
Tenax
B?, C Trap besorb
Por. P A b.
Trap Desorb Trap
34G
Desort
Nan e of compour
M
Methane 89 16.04 -l64 0 / 0 / 0 /
£thine 180 30.07 —88.63 0 / 0 / 0 /
Propane 282 44.09 —42.07 0 / 0 / 0 /
n-B jtane - 395 58.12 —0.5 0 / 0 / 0
Ethylene oxid 489 44.05 10.7 0 / 0 / 7
n-Pentane 505 72.15 36.07 0 / 0 / V
Propylene ox de 534 58.08 34.3 0 / 0 / ,j
Acry lo nitrile 557 b 53.06 ‘78.9 0 / 0 / V
Acr o 1e n 560 56.06 52.5 0 / C / V
n—Hexane 629 86.17 69.0 0 / 0 / V
Benzene 719 b 78.1.1 80.1 0 / C / V x’
V .r vl aretete 769 86.09 72.2 0 / /? / /
.so-Ootane 788 114.22 99.3 0 / V / /
Ethylene dichloride 806 98.96 83.47 0 / / / /
1,4-Dioxane 881 88.10 101.5 0 / VI / /
Ethylene glycol 906 62.07 197.2 0 / / X? /
Ethylene dibrornide 946 187.87 131.36 VI? / / X? /
Styrene 969 104.14 145 / / V X? V X?j
Suc rinonitrj le 1,073b 80.09 267 V / / V x
Phenol 1,176 94.11 182 V / V x’ / x
Benzyl chloride 1,253 126.58 179 VI / / X? /
Naphtha lene 1,269 128.16 217.9 V / I X? / X
Carbon tetrachloride 1,323 153.84 76.8 / / I X? V x
Bis—(2—OhloEoethyl) ether 1,402 143.02 178 VI / / X? / X
Tetrachioroethylene 1,542 165.85 121.02 ,/ / V V X
Toluen.-2,4—a nj b 122.17 280 V / I X ? / X
o—Nitr oaj i1ine 1 ’ 657 b 138.12 284.11 / / VI x I x
—Mitroaniso1e 1,684 153.13 258 V / / x r x
Phenanthrene 1,747 178.22 340.2 V / / X / X
n—Nexadecarie 1,811 226.43 287.5 V / I X V X
1,2,4-Trith loro benzene 21014 b 181.46 21.3 V / / X V X
Benzidine 7 184.2 401. 7(s) if / / x V x
4—Bromod .pheny1 ether 21616 b 249.11 310.14 VI / / x V x
Penteoblorophenol 7 266.35 310(d) VI / V x VI x
I exathloro-1,3-butadjene 2 , 783 b 267.6 215 VI / / X / X
0i(2—ethylhexyl) phthalate ? 1 385 VI X l / VI x
iryse ne 7 228.3 498 f Xl / X / X
Benzo(a)pyrene 7 252.3 7 VI Xl / X / X
words are the 20 e1ected possible and proba± 1e carcinogens.
b fron calculated 6.
0 — Probahie trap breakthrough.
V I? - Possible trap breakthrough.
V - Quantitative trapping (with appropriate a mt of sorbent).
/ - Quantitative desorption.
— Possibly not desorbed.
X - Probably not desirbed.
S - S uk limed.
d — Derou poses.
68
-------
: L
A. Analysis of a blank Tenax-GC sorbent tube.
_ ___I
B. Analysis of a blank Porapak P. sorbent tube.
C. Analysis of a blank Ainbersorb XE—340 sorbent tube.
Figure 22. Backgrounds produced by desorbing blank
sorbent tubes.
69
-------
to the r r oeter . This chromatoqraphic technique is known as
the ‘elution’ method of sorbent capacity determination. Another
techniaue of sorbent capacity determination which mo -e realistic-
ally approximates sorbent sampling conditions is ‘frontal analy--
sis”. By this technique a sorbent cartridge is exposed to a given
(challence) concentration of a single compound and its effluent
monitored for breakthrough. Figure 23 depicts a comparison of the
two technicues. Using the Sample Generation System (Ap endix B)
“frontal analyses” were performed on the three types of sorberit
cartridaes with benzene, carbon tetrachloride, and acrylonitrile.
These results are compiled in Table 21 and compared to their est-
mated values. Generally the “frontal analyses” showed the sorbent
cartridges’ capacities to be less than those estimated. though
usually in the same order of magnitude (except for carbon tetra-
chloride) . “?reakthrough” of the challenge compounds from the
Ambersorb cartridges was very difficult to detect because slow
breakthro’jch’ was indistinguishable from baseline drift. Based
on these studies, the anticipated retention capabilities of the
sorbent cartridges shown in Table 20 was judeed to be plausibly
realistic. The analyses of backup tubes during the collection
of actual field samples was thought to be the best means of
demonstrating the collection capabilities of the selected sorbent
cartridges, since matrix effects could greatly alter sorbent
capacities.
Desorption Properties— —
After a sample has been collected and concentrated on a sorbent
tube, tw major techniques, thermal and solvent desor tion exist
for recovering the sample from the tube for analysis. In thermal
desor tion, a sorbent tube containing a sample is heated to a
temperature high enough to volatilize the sample, such that a
stream of carrier gas flowing through the sorbent will cause the
sample to quickly migrate from the tube into an analytical system.
The capillary inlet system, discussed previously, was designed
for the thermal desorption of sorbent cartridges. With solvent
desorption, the sample is recovered from the sorbent tube by
extracting the sorbent with an appropriate solvent. The result-
ing solution is then introduced into an analytical system by
syringe injection.
Thermal and solvent desorption have differing advantages. ‘or
example, solvent desorption permits numerous replicate analyses,
but the solvent dilutes the sample of interest (often >1000 fold).
This defeats at least some of the sample concentration accom-
plished by sorbent sampling and reduces overall method sensitiv-
ity. On the other hand, thermal desorption can achieve a maximum
sensitivity by introducing all of a collected sample into an
analytical system; however, this does not permit replicate
analyses. By splitting a thermally desorbed sample, replicate
analyses can be performed but sensitivity is decreased by the
amount of the split.
70
-------
“CHAILENG E”
CONCENTRATION
w
In
0
-------
T/\RIJE 21. RESULTS OF FRONTJ\L P.NALYSES OF OR13ENT TUflES
_________ tntaIn. y’ . ’ : 1r im . ______
“! Itit ton” flrth o iqIi
ii c S flur t i’fl
W( ’jqllt , P . ’ tP, Tt ’mp. • rr ’ iirc ’, tilnMQ Tim. ’, Amo ,nt, r .’.p city. Tn’’, 7rn ’ . I Ti . ”, Ar’,’ ir
Conpoun Sorb nt j E/mln’C aim _____ 2O’C____ mIn Iq _____ L/ min J9ninL
Carbon Tpnax—GC 0.9 “ .1.025 “.25 “1 >1.02 a 10” <0.5 <3i “057 “ .1k “ 71 “ . 3
t .tra- FOr .ISak P 1.5 M).PP .75 “.1 “66 a tO “10 “ .1R4 “ 1R -, — % 1 “ 3 13
chiori Asthetsorb 3.2 “.0.975 “.25 “.1 >4.0 x IO3 — — - - %4’O “ .7,810
*i - 340
?er ’ .aa-CC 0.9 “.0.979 “ .25 ‘ .1 02.’ “.8 “.30 h - “t.y i
Porapak P 1.5 - - 175” - - - - 0 0
Ajuther,orb 3.2 - — - >4.0 x 10’ - - - - - 0 0
XE- 340
& rylo— tpn aa-CC 0.9 “.1 ‘ .25 5.1 0.4: “24
nitrile Porapak P 1.5 “.1 5.25 5.1 25 ’ 7 h 5.11 5.44 ‘ .9 fl.S 5.94 5.38
Aaiherqorb 3.2 “.1 5.25 5.1 >24 ,000 - - >206 - - 5.660 2,25
X X — 340
‘Pace.) on the fact that 6 for carbon tetrachioride qre ,i . ’r than 6 for ben yl chiorki.’.
hEx;,olUre distonttnued before caturetion obtaIned.
Paaed on the fact that 6 for cetbon tetrachiorid.’ in greater than 6 for n-be, .nne.
Pecu1t were inconelucive.
“Previoucly deteriu,(ned.
foot
9 pa pd on the fact that 6 for benzene I c greater then 6 for n-hean,.p.
on the fact that 6 for •cry1on1trI1 La greater than 6 for n-pentane.
1 Expo cure diecontinued before breakthrough obtained.
-------
A brief study was conducted to compare thermal and solvent de-
sorption. Two solvent desorption techniques, shown in Figures
24 and 25, were compared to thermal desorption of benzene from
Ambersorb XE-340 cartridges. The benzerie samples were generated
with the Sample Generation System (Appendix B) and compared to
analytical standards. A sample of benzene was also thermally
desorbed from a Porapak R cartridge for comparison. The re-
coveries obtained, based on the theoretical amount of benzene
generated, are given in Table 22. Assuming the actual amount of
benzene generated was somewhat higher than the theoretical amount,
more efficient recovery was indicated by thermal desorption. In
another experiment, some laboratory air samples were collected
with the sorbent cartridges and then analyzed. One Ambersorb
XE-340 tube was solvent desorbed with little detected during its
analysis. A similar Ainbersorb XE-340 sample that was thermally
desorbed yielded a much more complex chromatogram, demonstrating
the superior sensitivity attainable by thermal desorption.
Therefore, thermal desorption was selected as the most suitable
sample recovery technique.
TABLE 22. EVALUATION OF THE DESORPTION EFFICIENCY
OF BENZENE FROM AMBERSORB XE-340
Sample
Theoretical
amount,
vg
Amount
recovered,
g
Percent
recovery
First wash of
benzene exposed
0.90
0.76
84
Ambersorb tube
Second wash of
benzene exposed
Azttbersorb tube
0
0.0].
5
(1.2)
Anthersorb tube
emptied in
0.90
0.59
66
acetone
Benzene exposed
Porapak R tube
thermally
desorbed
7.9
138
Benzene exposed
Am ersorb tube
thermally
desorbed
0.86
1.38
160
within the
percentage of the first wash of that tube.
Csampling rate and time were inaccurate, so that the
actual amount of the sample may be higher than that
indicated.
73
-------
44
B. ADD SOLVENT
“Pas —throU9h” 5 olvent
desorptiOfl lechniquP.
L’iqnre 25. S 0 IvC’flt “I’ th”
clesorptiOt1 techniqUe.
-J
A. EMPTY TUBE CONTENTS
MARk
Fiq JrC 24.
-------
Two methods were used to determine the efficiency of recovering.
some of the twenty selected compounds from the sorbent cartridges
by thermal desorption. One technique was to inject analytical
standards through a septum tee (Figure 15, Section 7) into a sor-
bent cartridge enclosed by the capillary inlet system desorption
chamber. The sample was then desorbed through the entire sorbent
cartridge into the analytical system. The recoveries summarized
in Table 23 were obtained by comparing results of standards de-
sorbed through sorbent cartridges to results of standards desorbed
through an empty glass cartridge. The other method was to collect
a vaporous compound mixture made with the Sample Generation System
on sorbent cartridges and thermally desorb these cartridges. The
results were quantitated by comparison to analytical standards
and these values compared to the theoretical amounts collected
from the Sample Generation System. The recoveries obtained are
compiled in Table 24. Assuming the actual concentrations of the
generated compounds were somewhat different than their theoretical
concentrations, certain recovered amounts were selected as repre-
senting true values. The recoveries obtained from various sorbent
tubes could then be compared to each other as given in Table 25.
These recovery experiments demonstrated that most of the compounds
evaluated were efficiently desorbed from the Tenax and Porapak R
sorbent cartridges; however, recoveries of compounds from
Ambersorb XE-340 cartridges were biased in favor of the less polar
compounds. If these sorbent cartridge are combined in series with
the Arnbersorb tube last, most high—molecular weight and polar com-
pounds should be adsorbed on the Tenax and Porapak R tubes.
Therefore, most compounds not efficiently desorbed from Ambersorb
should be retained on the sorbent tubes prior to it.
TABLE 23. RECOVERY OF STANDARDS FROM SORBENT TUBES
Compound
Sorbent materiala
Tenax
Porapak
R Ambersorb
Carbon tetrachioride
Vinyl acetate
79
105
101
107
10 b
—
Benzene
109
102
81
Tetrachioroethylene
110
91
51
Ethylene dichloride
1,4—Dioxane
Ethylene dibromide
Styrene
Hexachloro-l,3-butadiene
Benzyl chloride
108
105
129
106
108
59
104
100
9O
—
15 b
b
avalues under sorbents are percentages found versus
theoretical amounts.
bNo peak. of similar retention time observed.
CNOt evaluated.
75 -
-------
TM3LE 24. PERCENT RECOVERIES OF TiTFOflETICAt AMOUNTS OF GENEPATSI) SAMPLES
Tube
I
Tube
Type
Compounda
Avpra r
of
thr’nrc’ ic l
irnount__________
R.S.t. of
thr tir 1
porr r 1 ,
ero1ein
Carbon
tetra—
chloride
Vinyl
acetate
81
57
flrn—
zcne
126
71
cryio—
nitrile
47
17
ch1’ ro-
ethylnne__Dioxane__chlorid”
41 42
21
EUyI
di
180
105
18
61
Porapak ft
Porapak Rb
20
50
84
78
78
54
61
57
54
Ambersorb
11
17
18
143
52
23 20
124
51
103
119
Tena*-(C
41
212
72
66
38
6q 39
78
77
74
avalueg under compounds are the perceffl:ages found vernus the theoretical amounts.
bAld one week after tube *18.
TABLE 25. PERCENT RECOVERIES OF SELECTED AMOUNTS OF GENERATED SAMPLES
Tube
I
Tube
Type
Compounda
? verage
% of
selected
amount
R.S.D. of
percentages,
I
Acrolein
Carbon
tetra—
chloride
Vinyl
acetate
Ben-
zene
Acrylo—
nitrile
Tetra—
chioro—
et v ene
Dioxane
Ethylene
di-
chloride
—
18
67
Porapak R
Porapak pC
39
100 b
100 b
i
100 b
71
100 b
61
100 b
37
100 b
58
100 b
60
100 b
58
92
67
23
30
54
Ajabersorb
21
20
22
113
112
48
48
69
57
68
119
Tenax—CC
83
253
89
52
82
148
93
43
105
64
aValuea under c apounds are the percentages found versus the eciected amounts.
b
Selected amount -100% by definition.
one week after tube *18.
-------
As mentioned previously, the cartridge design given in Figure 21
was selected as potentially offering both acceptable thermal re—
cover ies and an acceptable pressure drop, while containing suffi-
cient sorbent to collect the compounds of interest (Table 20).
Increasing the diameter of the cartridge or the amount of sorbent
in the cartridge would further decrease the ability to thermally
desorb compounds from the sorbent cartridges. Furthermore,
solvent desorption results in an unacceptable decrease in method
sensitivity. Therefore, the cartridge design shown in Figure 21
was selected for use in the Ambient Air Collection System.
Sampling System
The ability of a pump to draw air through the selected sorbent
cartridges in parallel or series modes was compared under EPA
Contract 68-02-2774. For this work, a small personal pump was
attached to “breadboarded” configurations of sampling systems in
the two sampling modes. These configurations are shown in the
photograph in Figure 26 and the schematic in Figure 27. A com-
parison of these two sampling configurations demonstrated that
with the small personal pump a flow rate of l L/min per tube
could be attained with the series arrangement, while only 0.5 L/
mm per tube could be attained with the parallel arrangement.
The reduced flow rate attained in the parallel configuration
was attributed largely to the metering valves used to adjust the
flows through the different sorbent cartridges. The valves,
rotometers and other additional parts required for the parallel
sampling arrangement would result in a sampling system with
greater cost, weight, and complexity, and less accuracy, durabil-
ity, and flow capability. Furthermore, the series configuration
would have the additional advantage of fractionating compounds
in air samples into zones of different volatilities on the dif-
ferent sorbent cartridges, thereby aiding the subsequent chro-
matographic separation of the complex samples. Therefore, the
series arrangement was selected as the more suitable for the
Ambient Air Collection System.
For convenience, a tray was designed to contain the sorbent
cartridges during sampling. The specifications of this sorbent
tube tray and the procedure of “loading” it are given in the
Operation Manual for the Ambient Air Collection System (Appendix
C). Figure 28 shows a “loaded” sorbent tube tray ready for
sample collection.
A commercially available, gas sampling device was obtained for
use in the Ambient Air Collection System. This device, a
Nutech Model 221 Gas Sampler, contained a pump of the type
thought to be most applicable for this project (see Figure 19).
It also included a valve and rotameter for monitoring and
adjusting the flow rate, a dry gas meter to measure the pressure
drop across which the pump was drawing, and power adaptability
to either an AC (outlet) or DC (battery) supply. (A power supply
77
-------
1 - DUPONT P4 000 PUMP
2 - PUMP PRE-FILTER
3 - TYGON TUBING
4 - GLASS WOOL PLUG
(PARTICULATE FILTER)
5 - TENAX-GC CARTRIDGE
6 - PORAPAK R CARTRIDGE
7 - AMBERSORB XE-340
CARTRIDGE
Figure 26.
“Breadboarded” version of the portable
miniature sampling system with sorbent
tubes in series.
78
-------
1 - BROOKS 500 cc/mm ROTAMETERS
2- RUBBER TUBING
3 - TENAX-GC SOR BENT TUBE
4- PORAPAK R SOR BENT TUBE
5 AMBERSORB XE-340 SORBENT TUBE
6- ALTERNATE POSITIONS FOR TUBING
1 - NEEDLE VALVES
8 - SWAGELOK TEES
9 - PUMP PRE-FILTER
10 - DUPONT P4000 PUMP
11 - BUBBLE METER
12 - EXHAUST BARB
Figure 27. Schematic of the “breadboarded” version
of the sampling system with the sorbent
tubes in parallel.
II ,
79
-------
I,
Figure 28. The “tube tray” portion of the
Ambient Air Collection System.
80
;;
I
I
I
— . .
____ 0
-------
was riot included with this sampler). This system was to be
capable of attaining flow rates up to 10—15 L/mifl across a
substantial pressure drop (maximum of 6.8 x 10” - 8.1 x 10” Pa
(20-24 in. Hg)]. Laboratory evaluation of the Nutech Gas Sampler
showed that flow rates of less than 3 L/min were attained with
this, sampler attached to a sorbent tube tray containing three
sorbent tubes (Tenax-GC, Porapak R, and Ambersorb XE-340) in
series. The Nutech system was then modified. and connected to
a large vacuum pump to evaluate the additional flow capabilities
of the larger pump. Figure 29 gives schematics of the two
systems evaluated and Figure 30 shows a plot of the results
obtained. As indicated in Figure 30, the large vacuum pump
increased the flow rate obtainable only slightly because pressure
drop across the tube tray increased dramatically as flow rates
increased. The specifications of the Nutech system superimposed
on this plot demonstrate that it was performing properly. There-
fore, the Nutech Gas Sampler, with its supplied pump, was adapted
for use in the Ambient Air Collection System. More details on
this sampler and its use in the Ambient Air Collection System
are supplied in Appendix C of this report.
RESULTS
Based on laboratory evaluation, a multi—residue sampling system
was developed for the collection of ambient air samples. This
system, designed to collect a broad variety of organic compounds
from the atmosphere, should effectively collect and subsequently
release the compounds of interest for this project. An Operation
Manual, given in Appendix C, was written in draft form prior to
field evaluation of this sampling system and finalized for this
report. This manual describes the preparation, fabrication,
operation, and maintenance of the Ambient Air Collection System.
DISCUSSION
The Ambient Air Collection System combines, in series, three com-
mercially available solid sorbent materials, each within a dis-
crete glass cartridge. These sorbents, Tenax—GC, Porapak R, and
Ambersorb XE-340, have complimentary adsorption characteristics.
When combined in series in an air sampling system, each sorberit
retains a certain range of compounds for subsequent analysis.
Figure 31 displays the theoretical performance of a multi
residue sampling system. An ambient air sample contains a wide
variety of compounds including those of high, intermediate, and
low volatility (represented by squares, triangles, and circles,
respectively, in Figure 31). All of the compounds in the air
sample are concentrated to varying degrees on the first sorbent
tube (Tenax-GC). Low volatility compounds (squares) are con-
centrated quantitatively at the head of the first tube. Such
compounds experienced very little chromatographiC displacement
down the sorbent tube because they partition more into the
stationary (sorbent) than mobIle (air) phase of the system. Very
81
-------
ThGON filiNG
ROIM WR
R. SCHEMATIC O MODIrIED NuTECH STSYEM WITH CAST PUMP
VACUUM GAUGt
rtow CONIROL
MUIHIR
“ .3
Alp
AIR
EXflA I i St
A. SCHEMATIC OF NUTECII GAS SAMPtJNG SYSTEM
ROTAA TtR
VACUUM
STAIMi Sitfi CAL E
UflINGS QUItl
ORY GAS ME1ER
POTAf4ItR
FLOW CONTROL
VALVE
Figure 29. 5chema ics of two sampling systems.
-------
35
KEY
SYMBOL PLANATI ON
0 PERFORMANCE OF NuTECH SYSTEM WITH GAST PUMP
A PERFORMANCE OF NuTECH GAS SAMPLING SYSTEM
0 MANUFACTURERS SPECIFICATIONS FOR NuTECH GAS SAMPLER
25
a
A
MAXIMUM
20 OBTAINABLE
15
1• DELIVERABLE
cc . . . .-
If REQUIRED
BY SYSTEM
£ \
5]
5 10 15 20 25
FLOW RATE, LJmin
Figure 30. Plot of flow rate versus pressure drop
performance of two sampling systems
attached to a sorbent tube tray and
manufacturer s specifications for the
Nutech Gas Sampler.
83
-------
U
• ,
S
a
Figure 31. Theoretical performance of
multi-residue sampling system.
volatile compounds (circles), however, partition mostly into the
mobile (air) phase and rapidly pass through the first tube. They
frequently are not concentrated to analytically detectable levels.
Compounds of intermediate volatility (triangles) may be concen-
trated to detectable levels, but also eventually pass through the
first sorbent tube. Intermediate and high volatility compounds
which elute from the first tube pass into a second sorbent tube
(Porapak R) which has greater affinity for volatile organics.
Again, all compounds are concentrated on the second tube to
varying degrees. Compounds partitioning more into the stationary
(sorbent) than mobile (air) phase are retained while very vola-
tile compounds are concentrated -but eventually pass through this
second tube as well. The very volatile compounds which pass into
the third sorbent tube (Asnbersorb XE-340) are retained by this
tube. The ability of the Ambient Air Collection System to per-
form as theoretically described was evaluated with field samples.
The results of these field evaluations are given in SectionS 10
and 11.
In addition to the ability to collect atmospheric organics, the
sorbent cartridges should also be able to release organicS for
analysis. As indicated previously, not all organics may be
quantitatively released from the cartridges of the Ambient Air
Collection System. In particular, very volatile, polar organics
may be difficult to recover from the Ambersorb tubes. Conse-
quently, field evaluations included a study of spiking techniques
(see Sections 9 and 10).
84
-------
SECTION 9
COLLECTION OF FIELD SAMPLES
OBJECTIVE
The purpose of the field sampling phase of this program was to
test the newly developed sampling and analytical methodology
under actual field conditions. This included evaluation of the
difficulty of transporting the sampling system and accessories,
of operating this system in the field, and, later, of analyzing
complex environmental samples. The analyses of these samples
are discussed in Sections 10 and 11.
INSTRUMENTAT ION
The Ambient Air Collection System which was developed for this
project (see Section 8), was used to collect field samples.
The Operation Manual for this system, given in Appendix C,
describes the preparatory and sampling procedures for using
this system, as well as the details of its fabrication. Although
the operation manual was drafted prior to field sampling, experi-
ence gained during this work was used to modify the procedures in
the original draft. These modifications are included in the
final version of the manual presented in this report. Most of
the modifications were relatively minor, such as wrapping sorbent
tubes with glass wool to minimize breakage of sorbent and culture
tubes. One exception was the mechanical leak check procedure
(included in Appendix C), which was developed as the result of
field sampling experience. This procedure was first instituted
during the collection of samples in Houston, Texas.
FIELD SAMPLING
Dayton
Samples were collected at Carillon Park in Dayton, Ohio, under
EPA Contract 68-02-2774. They were obtained with the use of
the Portable Miniature Collection System developed for that
contract. The results of these analyses are not given in this
report; however, the results did cause a small modification to
be made in the analytical system ((GC) 2 /MS]. This modification
is described in Section 10.
85
-------
Los Anoeles
The sampling plan developed for the collection of samples in
Los Angeles is given in Table 26. In addition to those samples
shown, samples were also collected with two Portable Miniature
Collection Systems during the same time periods. Although total
..
- _--
The sampling site was on the roof top of the EduCatiofl Resource
Center Building located on the grounds of the California State
University, DomingueZ Hills campus. A map of the Los Angeles
area, provided in Figure 32, shows the Campus slightly north
of area A. Areas A, B, and C correspond to locations estimated
to be high in concentrations of the compounds of interest by
computer ClimatoloiC l Dispersion Modeling. Data for this
model were obtained from the MRC Source Assessment Data Base
which incorporates stack emission data from numerous sources.
Table 32 gives the computer estimated composition of air pol-
lution in these areas for some of the compounds of interest and
a few others.
The samples were collected Ofl 14 and 15 April 1980. On 14 April
1980 the temperature in Los Angeles ranged from approximately
18°C to 27°C (65°F to 80°F), the relative humiditY from about
25% to 49%, and the winds from the northwest to west_southwest.
On the following day, 15 April 1980, weather conditions were
similar: the temperature range was approximatelY the same; the
relative humidity slightly less and more stable (around 32%); and
the winds more gusty, mainly from the northwest. Several exhaust
stacks could be seen from the sampling location to the north,
northwest, and southwest of the DomingUeZ Hills campus.
86
-------
TABLE 26. SAMPLING PLAN - LOS ANGELES
Sa m ple
numijer
Sorbent
Puro s .
Ikay
position
Date Times Volume
Pussy ir .format on
— Ottn r information
1mw
rat.
I
Tenax
Sample
3
4—14-80
‘i0800
‘.1,00(1 I.
‘t .2-3
LJmin
Ambient Air
Syst.m.
Porapak R
Sample
2
-1600
AC power,
maximum
M bersorb
Sample
3
hr
flow
Amtsersorb
Backup
4
C
Tenax
Calibration
1
4-14-80
Calibration of sampling
Porapak R
CalIbration
2
and
,
system flow rates.
Am&)ersorb
Calibration
3
4-1 -80
2
Tenax
Sample
I
4-I --80
‘s0800
.I,000 L
2-3
L/min
Ambient Air
System,
Field spike — see Method
P.
Porapak H
Sample
2
-1600
•
AC power,
maximum
Arbersoib
Sample
3
hr
flow
Asnbersorb
Backup
4
B
Tenax
Blank
blanks
Porapak H
Blank
Assbersorb
blank
F’S
l ’enax
Spike
Transport spskes — see
Porapak K
Spike
Method 13
Aribersorb
Spike
LS
renax
Spike
Lab spikes — see Method B
Porapak K
Spike
Arbersorb
Spike
Spiking Method A — “.5 mm after starting sample collection, Inject 3 pL quantities each of spiking solutions A, C, and E (see Tables
27, 29. and 30) into the glass wool plug of the system inlet.
Spiking Method B — Inject I. IL. quantities of each spiking solution “3 vms (1/B in.) into the sorbent beds of the tubes below at their
unnumbered ends.
Tena* - C, E, F (See Tables 27 to 30)
Porapak K - A, B, C
Ambersorb — A
0 3
-J
-------
TAELE 27. FIELD SPIKING SOLUTION A
Concentration,
compound
Acrolein 93
Carbon tetrachioride 120
vinyl acetate 132
Ber.zene 76
Acrylonitrile 81
Ethylene dichioride 56
1,4-Dioxane 133
Tetrachloroethylene 59
Ethylene dibrornide 202
Acetone
aSolvent
TABLE 28. FIELD SPIKING SOLUTION B
Concentration,
Compound
Tetrachioroethylene 55
cis—1, 3—Dich1oropropar e 77
trans—i, 3—Dichioropropene 76
Hexach]jro-1 3—butadiene 1.63
Acetone
aSolvent
TABLE 29. FIELD SPIKING SOLUTION C
Concentration,
Compound pg/iriL
Styrene 92
iexach1oro-1, 3-butadiene 191
Benzyl ghioride 98
Acetone
88
-------
TABLE 30. FIELD SPIKING SOLUTION E
Compound
Concentration,
‘ig/inL
Ethylene glycol
151
Di-(2-ethylhexyl)
phthalate
42
PentachAoropheno l
Acetone
68
asolvent
TABLE 31. FIELD SPIKING SOLUTION F
Compound
Concentration,
ug/mL
Benzidine
73
Chrysene
24
Benzo(a pyrene
Toi.uene
47
asolvent
TABLE 32. LOS ANGELES - POTENTIAL CARCINOGENIC
COMPOSITION AT POINTS OF MAXIMUM CON-
CENTRATION (micrograms per cubic meter)
Compound
Location A
86, 46
Location B
80, 46.
Location C
82, 46
Acrylonitrile
0.0007
0.0030
0.0016
Benzene
0.01
0.01
0.01
Carbon tetrachioride
44
5
9
Chloroform
22
2
4
DDT
0.38
24.7
1.4
Ethyl hloride
6
5
6
Ethylene dichloride
191
22
42
Vinyl chloride
117
12
23
Vinylidene chloride
0.16
0.05
0.02
89
-------
:
- 4fff1 J
—. — I ... ,, III) A
SANTA I4ONIC . - .. .‘ j:- 1 ”: .;i- l• 1 ; ( I’1 . . 1.J:;. L ‘‘ “ ç r”
-‘V -‘ I I P L . - , w,.,v?Iv I I
\‘r < i - I ‘I I t V i,/ )‘ ./ ‘ / j
kh ’ 1 I. ‘‘ I 1 . .AcIrNnf ..,..
I ..Lr ‘IcQ fT l11 : I . irfn. ’
.‘ .. I -i \ iy i.aw&pi(( :
.1 r .i. . .7 1
‘ ‘t ‘r rr’ “ “ II 1 I ¶ Nn’1W4LI 1 • . -1
“.‘ iL _Lt L1 1 ltLI :t 1?fl 14 1 4”F 1 i t
-. .1 111 1 7. . ULI4 O
MAN NATTAM tAC II . . 4 .
. L •• . ._ -_-.
- - oI .H?1n - .
I AN1 1 U I . .. J !\ . ..C.4 ..e._yuII I .N - 7 i
Md.... ’77 T ‘:•.1’ L j . .... ... ..‘ ...
-. . -r. . ..— .• .. . , ____ • ‘ _ _ A A
.. 4 - - •1 - ___
. 1 • r • rn V
i•.. . . l . L. • •
.3 :
— •‘ .. A C H - . _ _
:-. — \ - L_. L .. . . -. - , ...i.’9 . / £ç
- - Li
— - - S.-- .. ._ P -
IACJ
/
I wPO T S ACH
Fiçpire 32. LocatiOns of hiqh qrourud level concentr t1i0flS -
Los Anqeles, Ca1ifOrfli .
-------
Table 33 describes the sampling conditions. No problems were
experienced with the Ambient Air Collection System.
Niagara Falls
The ability of the Ambient Air Collection System to sample un-
attended was tested during an indoor sampling trip in Niagara
Falls. The sampling plan is given in Table 34. As indicated,
three, consecutive, twelve—hour samples were to be taken with
the system unattended during the majority of the sampling
periods. Samples were collected in a private residence with
temperatures ranging from 24°C to 27°C (75° to 81°F). As
indicated in Table 35, this system did not function well under
these strenuous circumstances with discontinuous operation
during the second and third sampling periods. Maitenance
performed on the system’s pump, after returning to the labora—
tory, corrected the malfunction.
Houston
The sampling plan developed for the collection of Houston air
samples is given in Table 36. In addition to those samples
shown, samples were also collected with three Portable Miniature
Collection Systems and a critical orifice system developed for
another project. Although total sample volumes collected by
these three different systems were not identical, the samples
were essentially replicates. Extra, blank sorbent tubes were
taken to replace broken tubes and tubes spiked with the Sample
Generation System were take,n to provide transport and field
spikes.
As with Los Angeles, personnel and equipment were flown to
Houston. This time, airline officials were especially concerned
that the materials and/or equipment might be hazardous. They
were assured there was absolutely no hazard. Although the
sorbent tubes were wrapped with glass wool inside the culture
tubes, at least a dozen culture tubes broke during shipment.
This damage was attributed to the soft—sided case in which
they were packed. A sturdy case lined with cushioning mate-
rials would have been preferable.
The sampling was performed on the second floor balcony of a
motel located 0.4 Km (1/4 mile) north of Route 225 on South
Richie Road. This site was within area number 3 on the map
shown in Figure 33. As with Figure 32, the locations marked on
this map are areas estimated to have high levels of the pollut-
ants of interest. This information was obtained by computer
climatological Dispersion Modeling with data from the MRC
Source Assessment Data Base (1). The estimated composition of
some of the selected compounds and a few others in these areas
are given in Table 37. The terrain surrounding the sampling
area selected was relatively flat and oil refineries could be
seen approximately 1.6 Km (1 mile) away in all directions.
91
-------
TAI3LE 33. SAMPLING CONDITIONS - LOS ANGEL1 S
ipIe
number Date
Times
Pry g s reeler
volume, L.
Flow rato,
1./rein
Volume cimaled
from flow rate, L
Comments_____
1
4—14—80
1007—1807
hr
1111.3
t.3.7
“ .1,700
Rackup
ci ii ipec1
No pump
)\mbrr orb tube
upon removal
problems.
2
4—15—80
0810—1610
hr
914.5
“ .2.0
%9 10
No pump
Sptked
problems.
as di reeled.
0
p )
TA8LE 34. SAMPLING PLAN - NIAGARA FALLS
Sample
number
Sorbent
Purpose
‘tray
position
Sam il n9
Pump intotnuatien
Other itiori
Oate
Times
Volume
Flow
rate
1
Tenax
Porapak P
Ainbermorb
Ambersorb
Sample
Sample
Sample
BackUp
I
2
3
4
f.-18-BO
.0R0O-20O0
hr
“4,400 1.
,2
h /mm
Ambient Air
System,
AC uower,
maximum flow.
.
2
.
Tenax
Porapak
Aathereorb
Aa ersorb
Sample
Sample
Sample
Backup
1
2
3
4
6-10-80
to
6-19-00
“.2100-0900
hr
“ t,4o(l L “‘2
1./mm
Ambient Air
System,
AC power,
maximum flow.
3
‘t na*
Porapak P
Pimbermorb
ambermerb
Sample
Sample
Sample
Backup
1
2
I
4
6-19-00
“10r)0—2200
hr
“ . 1.400 L “‘2
b/tale
Ambient Air,
System,
AC rower,
maximum flow
B
Tenax
Porapak P
Ambersorb
Blank
Blank
Blank
Slankq.
C
Tenax
Porepak 0
Auubereorb
Ambersotb
Calibration
CalibratIon
Calibration
Calibration
6-18-00
a d
6-1 0-00
‘
Cal tbr t1on of
samt’1i09 ‘ .ystenu
how r trs.
-------
0
TABLE 35. SMIPLING CONDITIONS - NIAGARA FALLS
Sample
number
Date
Times
Dry gas
volume,
meter
L
Flow rate,
L/min
Volume estimated
from flow rate, L
Comments
1
6—18—80
0809—2010
hr
1430
%2
“-1,400
No problems.
2
6—18—80
to
6—19—80
2112—?
hr
680
“-2
?
Pump stopped during
sampling period.
3
6—19—80
1012—2202
hr
720
“-2
?
Pump stopped during
sampling, started
again by tapping.
-------
T/\BLE 36. S1 MPL1NG PL? N — 11OtJ TON
it. _______
n imt r’r SnrhrnI Prjo ” p v; it i rn P. t • Ti mr VnI nrn - — Ft r.w -.1 n r.’p jfl !r)rm.1’ fl h ’ r ri t,rr’
Irflax 1 •7 — 2’’— fifi Ofinfi— I 4,Ofl “-I ,(100 r, .7 ?/r’. in lflr). .‘n l I r
t’r ipak P 7 t r -i”.
1tnthPt! nFh 3 A 1 ’r ,w ’I
4 Ft .w ? I/em.
s n.ij -1e L 7—10- -flO nrtnn-Ir,n( ’ 14 OOh1 t, -7 7,/em 7Innt pit
pnrapak P SmaI ’lf’ 2 hr
pji cr nrb Semple 3 Pr p wr ’r
Pmbernnrh P rkIIp 4 ¶I ” ? t./T ’Ifl
Tenax plank RI Icc
T efl I X Plank
Lorapak P Plank
I’orapak P Plank
Itntberaorb Plank
AMbersorb Plpnk
S ‘Tana* Spike rik’ a - aptked with
Teflax Spike tiq rarh ni
I’orapak It Spike hnnzr ’ne. hen7.yi h3o-
Porapak P Spike ridn. cathr’n te ra-
Pa* erqorb Spike chtr,tid ”. tetra—
chlornnthylnne.
. .thylnnn ciihrn ld.’.
d c -1, 1—dichinrn- ’
prr,p ’nn. and hnxa-
ch lorc’-) .
r ,n Sarple c,nnnrat inn
SystnT (Pptnnr3ix P7
C Tenax cilibratton 7—79-Pt) Calihr tinfl ci
Porapak P CalIbration 2 anti ccw plln ’ cystem
lUabercorb CalibratIon 3 7-3fl-PO how ratec.
An bereorb calIbration 4
-------
Figure 33. Locations of maximum concentration
of pollutants in Houston.
95
-------
AVERAGE CONCENTRATIONS AT POINTS OF
MJ XiMUM POL .UT1ON IN HOUSTON
[ Micrograms per cubic meterj
Compound
Location 1
82,88
Location 2
90,92
Location 3
95,90
Acrylonitrile
1.0
0.1
0.1
Berizene
1,366.0
6.0
5.0
Carbon tetrachioride
6.0
190.0
251.0
Chloroform
2.0
92.0
123.0
Dich lorobutene
1.0
0.02
0.02
Ethylene dichioride
26.0
977.0
757.0
Tetrachioroethane
0.009
0.022
0.018
Epichiorohydrin
0.15
1.0
1.0
Trichloroethy lene
5.0
12.0
10.0
Vinyl chloride
7.0
15.0
2.0
Vinylidene chloride
0.004
0.165
0.220
Ethylene dibrornide
0.14
0.01
0.10
The samples were collected on 29 and 30 3uly 1980. Weather
conditions were hot and humid both days, with temperatures exceed-
ing 32°C (90°F) and relative humidity at approximately 90%. The
sampling conditions are provided in Table 38. As mentioned
previously, the leak check procedure specified in the Operation
Manual (Appendix C) was instituted during this sampling program.
TABLE 38. SAMPLING CONDITIONS - HOUSTON
£& npIe
t uiber
Date
Timea
Dry ças
volume,
meter
1.
Y oi rite,
L/e in
Volume •.tim*t.d
froi f w rate, L
Comesents
I
114 —i948
hr
461
.l
i.4$O
No
problems.
2
7—30—BQ OTh6—1556
hr
626
%1.S
%720
Mo
problems.
96
-------
SECTION 10
ANALYSIS OF FIELD SAMPLES BY CAPILLARY GAS CHROMATOGRAPIiY/
MASS SPECTROMETRY
OBJECTIVE
Samples of ambient air which were collected in the field with the
Ambient Air Collection System were to be analyzed by capillary
gas chromatography/mass spectrometry ((GC) 2 /MS). The analytical
method, described in Section 7, for (CC) 2 analyses was adapted
to a (GC) 2 /MS instrument for the evaluation of the field samples.
The presence of the twenty selected potential carcinogens was
investigated by reviewing mass spectral data. Compounds of
interest identified in the field samples were then quantitated.
INSTRUMENTATION
A Hewlett—Packard Model 5985 GC/MS system was used for the
analysis of field samples. This instrument had a Model 5840 GC
with a capillary inlet system which was very similar to the CC
used during the analytical method development phase of this
program (Section 7). A data system, that was also part of this
instrument, was used to display, interpret, and reduce mass
spectral data. Final interpretation of mass spectral data was
performed by a qualified mass spectroscopist.
To introduce samples from sorbent tubes into this analytical
system, the capillary inlet system (Section 7) was interfaced
to the Model 5985 CC/MS. Some modifications to the inlet system,
GC/MS, and analytical method were necessary to obtain an inter-
face which produced acceptable results. For the analysis of
field samples from Houston, Texas, the capillary inlet system
was removed from the GC/MS and installed on the multi-detector
(GC) 2 instrument described in Section 11. A Nutech Model 320
Thermal Desorption System, a commercially available sample intro-
duction system similar to the capillary inlet system, was then
interfaced with the GC/MS. Houston field samples were analyzed
with both the CC/MS and the multi-detector (GC) 2 instruments.
EVALUATION
The capillary inlet system devised during analytical method de-
velopment was initially installed on the CC of the Model 5985
97
-------
CC/MS with the inlet system/GC interface maintained in its exact
physical and functional configuration. Preliminary evaluations
demonstrated that this system was operating properly with this
configuration. These evaluations were performed by analyzing
column test mixtures which were previously used to specify
analytical performance during analytical method development
(discussed in Section 7). Analyses of these mixtures were per-
formed in the split injection mode, the mode used during method
development, with injections made through a septum tee into an
empty glass tube installed in the inlet system (see Figure 15,
Section 7) . Although the results demonstrated proper performance
of the inlet system, the mass spectrometer was not as sensitive
as the flame ionization detector (FID) used during method devel-
opiment. All of these experiments were performed with the mass
spectrometer operated under electron impact ionization in a con-
tinuously scanning mode.
To improve analytical sensitivity, the inlet system was evaluated
with very-low split and splitless injection operation. During
splitless operation some compounds in the test mixtures were
found to be catalytically altered. These in situ reactions were
detectable with mass spectral information, but would not be
identifiable with an FID (as previously used) . Splitless opera-
tion also decreased the flow of carrier gas removing sample from
the capillary inlet system trap. This extended the period of
time to effectively transfer the sample from this trap to the GC
and resulted in broadened chroinatographic peaks, especially for
low-boiling constituents. The problems of peak—broadening and
catalytic alteration were found to decrease as the injection
mode was changed from splitless to low—split to moderate—split
operation. By periodically changing the nickel capillary trap
of the inlet system, by maintaining uniform heating of the trap
during desorption, arid by adjusting instrument operating param-
eters, the problems of peak—broadening and catalytic alteration
with splitless injection were found to be significantly reduced.
As part of the preliminary evaluations of the (GC) 3 /MS analytical
system, field samples were collected near the laboratory to pro-
vide authentic samples for analysis. As indicated in Section 7,
these samples were collected at Carillon Park in Dayton, Ohio,
under EPA Contract 68—02-2774. They were obtained with the use
of the Portable Miniature Collection System developed for that
contract and, therefore, the results of these analyses are not
given in this report. However, the results impacted on this
project by demonstrating the necessity of splitless inject .on
operation in order to obtain suitable sensitivity for field
samples.
Frequent adjustment, trap changing, and cleaning of the capillary
inlet system were found to be necessary to maintain acceptable
operation of the analytical system using this system and split-
less operation. A commercial sample introduction system obtained
98
-------
previously was therefore interfaced to the Model 5985 GC/MS
for the analysis of Houston field samples. This system was a
Nute ch Model 320 Thermal Desorption System. Concurrently, the
capillary inlet system was interfaced to a multi—detector (CC) 2
system (Section 11). Both systems were comprised of the same
major elements: thermal desorption chamber, heated valve,
capillary trap, and thermal control unit. Sample reconcentra-
tion and introduction into the analytical systems were performed
in the same manner; however, some of the physical details of
these two systems varied. For example, the trap of the Nutech
system was contained in such a manner that the thermal control
unit controlled both the cooling and heating of this trap. This
was an advantageous convenience. Both systems were found to
perform satisfactorily.
Another change made prior to the analysis of the Houston field
samples was to change the mode of sample introduction from
splitless to direct injection. To accomplish this modification,
the carrier flow directed to the sample introduction system
(Nutech) was modified in the Model 5985 GC/MS system. Instead of
being obtained from a split off of the capillary inlet pneumatics
(see Figure 10, Section 7), the carrier gas was supplied from a
separate auxiliary system with an independent pressure controller.
Furthermore, instead of introducing a sample from the trap into
the GC through a needle inserted in the GC injection port, a
direct coupling was made between the Nutech system and the GC
column. In the GC/MS this was done with glass-lined stainless
steel tubing (passing through the GC injection port for heat)
with low-dead-volume stainless steel connections to the Nutech
system and capillary column. To accomodate potentially greater
sample quantities, the capillary column within the GC/MS was
changed from an ‘ 50 M long, 0.2 mm ID, OV-lOl fused-silica WCOT
column to an “. 60 M long, 0.25 mm ID, SE-30 glass WCOT column.
(Both columns had non-polar, methyl-silicone liquid phases making
them somewhat equivalent in chrotnatographic characteristics.)
Carrier gas flow through the Nutech system directly passed into
the capillary column. The flow rate of this carrier was con-
trolled by the pressure controller of the auxiliary system. The
multi—detector (GC) 2 system also was interfaced to the capillary
inlet system to permit direct sample injection as described in
Section 11.
RESULTS
As noted previously, analyses were conducted using a Hewlett-
Packard 5985A GC/MS/Data System having a 50—meter fused silica
column coated with methyl silicone liquid phase. This was
later changed to a 60-meter glass capillary column having an
equivalent phase. Chromatographic conditions were varied to
meet the individual demands of specific samples. When possible,
all sorbents from a single sampling run were analyzed using the
same temperature program so that retention values could be
99
-------
compared. Samples collected in Los angeles and Niagara Falls
were analyzed usir the original capillary trap and inlet system
designed and constructed as a part of this program (Section 7)
Samples collected in Houston were analyzed using a Nutech Corpo-
ration Model 320 Thermal Desorption System. The latter unit was
connected directly to the capillary analytical column of the
GC/MS system. The two units offer essentially the same features;
however, the Nutech system is more convenient to operate. This
research effort was conducted with a companion program for the
development of a portable miniature collection system. Thus, in
some instances, data from parallel sampling conducted under that
program . iil be used to illustrate results pertinent to this
proj ect.
A basic premise of the multiple sorbent sampling approach is that,
for ambient air sampling in the complex and often hostile environ-
ments frequently encountered, compounds of interest are often not
completely retained by a single sorbent. This would be particu-
larly true when the volume of air sampled is large. The results
presented for the three locations sampled adequately demonstrate
this fact.
Calibration for Compounds on the enty Potential Carcinogen List
Compounds on the list of twenty potential carcinogens being
studied in this program, if found to be present, were quanti-
tatively measured. Calibration was achieved by preparing sorbent
tubes with known weights of each material using a standard
gaseous Sample Generation System constructed at MRC. This system
is based on controlled syringe injection, vaporization and sub-
sequent serial dilution of the standard compounds (Appendix B).
Quantification by CC/MS was achieved by integration of the peak
corresponding to a major interference-free ion of the compound
which appeared at the correct predetermined retention value.
Qualitative identification of other components was achieved by
initially using a standard compilation of mass spectral data [ 81.
Several collections of complete mass spectra are available and
were used for more detailed spectral analyses. In some cases,
compounds have been identified by class and carbon number such as
C 10 alkane or C —alkylbenzene. In the latter case, the total
carbon number of the alkyl moieties is given. For the data
obtained on the Houston samples, an attempt was made to identify
the specific compound where possible. Due to the similarity in
frag-rnentation pattern of hydrocarbons under electron-impact, and
because of the lack of strong molecular ions for highly branched
hydrocarbons, boiling point data were used to select the one
compound from several which might fit a mass spectrum.
Los Anceles Air Sampling Analysis
The atmospheric conditions and the sampling location in inetropol-
itan Los Angeles are given in Section 9 of this report.
100
-------
Figure 34 presents the chromatograms (total ion current plots) of
materials retained on each of the sorbents from a single sampling
run. The chromatogram at the bottom of the figure shows compounds
retained in the Tenax-filled collector, while chromatograms for
materials from the Porapak R and Antbersorb XE-340 collectors are
shown in the middle and upper parts of the figure. At the top of
the figure, a chromatogram of the contents of the backup An bersorb
XE-340 collector is shown.
These data are for samples collected with the portable miniature
system; however, replicate samples showing the same features were
collected using the Ambient Air Collection System.
By comparing the compounds retained by each of the sorbents, it
is apparent that many compounds of interest escape collection on
the Teriax and are retained in the Porapak and Arnbersorb collec-
tors. Benzene, for example, appears largely in the latter two
collectors. TetrachiorOethYlefle, while appearing in the Tenax
collector, is heavily concentrated in the Porapak R collector.
Both of these compounds are included among the twenty potential
carcinogens selected for evaluation in this program. Another
compound of some interest is trichioroethane. This compound is
found to reside essentially in the Aithersorb and Porapak R
collectors.
Sampling system break-through is found to be minimal as shown by
analysis of the Axnbersorb backup collector which is the fourth
tube in the collection series. The single, large peak observed
from this collector is due to carbon dioxide.
The chromatograms in Figure 34 are normalized on the intensity of
the strongest peak in each chromatogram. For quantitative com-
parison of one chromatogram to any of the others shown in that
figure, absolute intensities must, of course, be compared. Ac-
cordingly, absolute peak area counts were used as the basis for
the foregoing observations. Table 39 presents the amount of
benzene and tetrachioroethylefle found in each sorbent tube and
the concentration of these compounds in the sampled air.
TABLE 39. SUBJECT COMPOUNDS PRESENT IN SAMPLES
COLLECTED IN METROPOLITAN LOS ANGELES
Total a
__________ weight, Concentration,
ng nc/rn 3 in air
Weight per collector, ng
Tenax-CC Porapak R Arnbersorb
Compound
.
Benzene
87
111
8
206
120
94
TetrachiOrOethYlefle
99
62
-
161
a 17113 liters of air sampled.
101
-------
AM (RSOR BAC1cUP WRE - PoSITION 4
Reconstructed total ion chron atogram of materials
collected in metropolitan Los Angeles by the
multisorbent air sampling technique.
‘-a
Figure 34.
-------
Niagara Falls Air Sampling Analysis
Sampling conducted in Niagara Falls, New York was of particular
interest to this program because it provided a rather sheltered
sampling environment. Sampling was conducted indoors where
atmospheric oxidants (ozone, NOR) were presumably at a low
level, temperature and humidity were moderate, and the contri-
bution to the sample matrix by various common mobile and station-
ary emission sources was minimal. The unattended performance of
the sampling systems was tested and a discussion of this aspect
of the operation is presented in Section 9.
Though the purpose of the sampling was to quantitate three speci-
fic compounds, benzene, tetrachioroethylene and trichloroethylene,
all twenty potential carcinogens were included in the data evalu-
ation. A total ion chromatogram of the contents of a Tenax col-
lector is presented in Figure 35. A list of detected compounds
is keyed by peak number to this chromatogram and is given in
Table 40. As is shown by the chromatogram, the sample size sig-
nificantly exceeded the capacity of the column, resulting in
broadened chrornatographic peaks. An important finding, however,
was that even with the high sample loading and the large volume
of air sampled, no sample breakthrough to the other sorbents in
the series occurred. Total ion chromatogram from analyses of the
Porapak R and Arnbersorb tubes are shown in Figure 36. The data
system software, as mentioned previously, normalizes the most
intense peak to full—scale. In these cases, trace components or
base-line perturbation due to water were normalized to full—scale.
Mass spectra, however, show that the only components present are
several silicone compounds, which are probably artifacts of the
analytical system.
Quantitation of four compounds among the twenty subject carcino-
gens was conducted and is presented in Table 41.
Houston Air Sample Analysis
The results from the Niagara Falls, New York sampling indicated
that sample breakthrough, and the resultant need for the multi-
sorberit sampling system, may depend upon sample matrix and/or
sampling conditions. To further evaluate this possibility, a
field sampling trip was scheduled to take place in Houston,
Texas during a sustained period of record high temperatures,
accompanied by high humidity. During the time of actual sampling
the temperature ranged in the nineties and relative humidity was
‘ 90%.
Several modifications in the analytical system were made prior
to analysis of the collected samples. Since a heavy loading of
compounds was expected in these samples, the narrow—bore, fused-
silica GC capillary column was replaced with a similar larger—
bore glass column of about the same length. A methyl sil .COfle
liquid phase was retained. The capillary trapping and inlet
103
-------
Qualita€ive scan of a Tenax collector sample from
Niagara Falls sampling with the Ambient Air System.
‘ —I
0
Figure 35.
-------
IA )
Figure 36. Reconstructed total ion chromatograms from GC/MS analysis
of second and third collectors from Niagara Fal.1.s sampling,
A) Porapak R; B) Ambersorb XE-340.
a
U i
(8)
-------
COMPOUNDS DETECTED IN ANBIENT AIR SANPLES
COLLECTED INDOORS AT NIAGARA FALLS
Co p
Chro atogr
retention
r n
aphic
tiTne,
r ientb
Air
Sampler
ientb
Air
Sa np1er
Portable
Sampler
1.
l,1,a-7: c ioroethane
10.6
X
X
X
2.
Ser.:e c
10.9
X
X
X
3.
Cvclohexan
10.9
X
X
4.
Carbc r tetrar lcr de
10.9
X
5.
C--A1’ ar €
11.6
X
X
X
6.
Tcjuere
12 .8
X
X
X
.
rcc -othylere
13.6
X
X
6.
C -Akane
13.6
X
X
X
9.
20.
Et yiber.zene
lene ‘
15.0
15.3
X
X
>
X
X
31.
Xyier e
15.9
X
X
X
X
3 :.
C 5 - A1kar e
16.2
X
1 .
c-rotoluerie
17.6
X
X
24.
15.
Acetophen ne
SiliconeC
16.1
18.5
X
V
V
V
V
3€.
chlorobenzene, m.i’p
19.0
X
X
V
37.
16.
C — lkv1ber zene
o 0ach1oi-cbenZene
194
19
V
X
V
V
19.
20.
C..—Alkylbenzene
Silicone
20.0
20.3
X
V
V
X
22.
22.
C-Alkylbenzene
S 1 .cone
20.8
21.7
X
V
V
23.
D th1orotO1uene
22.0
x
V
24.
Silicone
22.6
x
25.
Silicone
23.9
V
V
2€.
Trhlrrotoluene
26.4
X
X
27.
Silicone
27.1
x
2.
TetrachlorobeflZene
28.2
x
x
V
29.
Alkane
29.9
30.
2,3—di—t— tYl -4
methyipheriol
30.7
V
X
31.
S li one
40.9
number is used to key chromatogram shown in Figure 35.
bThese samples also show ch1orobenaer e and trichioroethylefle.
components ar se from system silanization and septum bleed.
AEnt ie amount retained in the
contained Tenax GC.
TAELE 41. SUBJECT COMPOUNDS DETECTED IN SAMPLES
COLLECTED INDOORS AT NIAGARA FALLS
Compound
555 ,a
ng
Volume
air sampled,
liters
ConoentratiOfl,
r g/ m’
Benzene
1,970
1136.9
1,730
Tetrachloroethylene
Benzyl chloride
Carbon tetrachioride
8,770
4,160
2,470
1136.9
1136.9
1136.9
7,710
3,660
2,170
106
-------
system, of MRC design, was replaced with a Nutech Model 320
Thermal Desorptiori System. Since the collection tubes used for
the ambient air sampler differed from those designed for the
Nutech Model 320, an external desorption furnace was connected
to the Model 320.
The extremes in humidity presented a problem in the analysis of
the Porapak and Arnbersorb collector contents. Best results were
obtained in previous analyses by maintaining the gas chromatograph
column at —30°C during warming of the cryogenic trap. This pro-
cedure insured that all trap contents would be on the head of the
column before any component began to travel through the column.
The large amount of water present in the latter collectors, once
condensed, tended to block the column or pass as droplets of
water through the column. No data could be obtained under these
conditions. By modifying the analytical conditions so that the
column was maintained at +30°C during the trap warming cycle,
water did not condense and an adequate analysis was possible.
Total ion chromatograms for the Tenax, Porapak and Axnbersorb
collectors are shown in Figures 37 to 39. Retention values for
the various analyses vary because of the change in chromatographic
conditions. Components present in the Tenax and Porapak collec-
tors are tabulated and presented in Tables 42 and 43. Note that
carbon dioxide appears as two separate peaks in the Porapak
sample, and several components such as toluene appear twice.
This phenomenon may represent a limitation of the Nutech sample
introduction system, with some reconcentrated compounds in the
Nutech’s trap not introduced into the CC/MS system in a discrete
slug.
No component breakthrough into the Ainbersorb tubes occurred.
The several small peaks which appear in those chromatograms
are due to silicone compounds, and are artifacts of the analyti-
cal system.
Only benzene and tetrachioroethylene, of the list of twenty
potential carcinogens, were detected in the Houston ambient air
samples. The distribution of these components between the vari-
ous sorbents is presented in Table 44.
DISCUSSION
The results presented indicate that components can breakthrough
a single Tenax collector under certain conditions. Though all
of the factors involved have not been fully defined, it is clear
that sampling volume is involved but is not the only factor.
Climatic conditions such as temperature and humidity, as well
as the composition of the sample, are doubtlessly factors as
well. For very low volumes of sample, breakthrough would not
be expected. The continually increasing need for lower detec-
tion limits, however, requires that a large volume of air be
107
-------
Total ion chromatogram ot materials on the Tenax collector
after ambient air sampling in Houston, Texas.
Figure 38. Total ion chromatograrn of materials on the Porapak R collector
after three—sorbent air sampling in Houston, Texas.
Figure 37.
I — ,
0
a,
-------
Figure 39. Total ion chromalograms of materials on the Ambersorb collectors
after three—sorbent a bient aIr saniplinq in houston, Texas.
I - .
0
-------
T.P 42. COMPOUNDS COLLECTED ON TENAX GO DURING THREE-
SOREENT AIR SAMPLING AT HOUSTON, TEXAS
Chrornato raph c’
peak nurr.ber C r pound
1 Acetone
2 Benzene
3 Trirnethv1but ne
4 Dirnethylpentane
5 Ethylpentane
6 2-! ethy1-2—hexene
7 Toluene
8 2, 3 , 4—Trin ethy1pentane
9 2, 3 , 3—Trirnethylpentane
10 3-Ethyihexane
11 Tetrachioroethylene & Ca—Alkane
12 Si l .iconeb
13 Ethy lbenzene
14 p-Xy lene
15 3, 4-Dirnethyiheptane
16 n —Xy1ene
17 3—Methy l—3—ethy lhexane
18 Curnene
19 n-Propy lcyc lohexane
20 Acetic acid
21 n-Propylbenzene
22 Methy lethy lbenzene
23 1—Nethy1—4—ethy1benze e & 3,4—Diethylhexane
24 3—Ethy l-3-methy lheptane
25 1 , 3, 5-Trirnethylbenzene
26 3, 3—Diethyihexane
27 3-Methylnonane
28 1,2, 4-Trimethylbenzene
29 3, 3, 4 , 4—Tetrarnethyihexane
30 n—tecane
31 n-Buty lcyc lohexane
32 1, 3—Diethylbenzene
33 1-Methy l—3--propy lbenzene
34 1 , 3—Dirnethyl—5—ethylbenzene
35 1-Methy l-2—propy lbenzerie
36 1 , 3-Dirnethyl—4—ethylbenzene
37 2-Methy l-1-cyc lohexy lbut .ane
38 1 ,2—Dimethyl-3—ethylbenzene
39 Methylbutylcyclohexane
40 C 11 —Alkane
41 1,2, 3,5—Tetrarnethylbenzene
42 n—Pen tylcyclohexene
43 C 12 —A lkane
44 C 12 -Alkane & C 5 —Alkylbenzene
(continued)
13.0
-------
TABLE 42 (continued)
Chromatographica
peak number Compound
45 C 1 2 -Mono—olefin
46 C 12 -Alkane & C 5 -Alkylbenzene
47 1, 2, 3, 4—Tetramethylbenzene
48 C 12 —Alkane
49 C 5 —A lky lbenzene
50 C 12 —A lkane
51 Alkane
52 Naphthalene
53 1—Nethy l-2—penty lcyc lohexane
54 C 12 —A lkane
55 Alkane
56 C —A1ky1benzene
57 Benzoic acid
58 Cycloalkane
59 C 6 -A lky lbenzene
60 C -A1ky1benzene
61 C 5 -A lky lbenzene
62 Alkylbenzene (s)
63 Alkane(s)
64 2—Methylnaphtha lene
65 C 13 —A lkane
66 1—Methy1 aphtha1ene
67 Si1icone
68 C 1 —A1kane
69 Biphenyl
70 C 15 —A lkane
72 EthyJ.naphtha lene
73 C 15 —A lkane
74 Dimethylnaphtha lene
75 Dimethylnaphthalene
76 Alkane
77 Alkane
78 Alkane
79 Cycloalkane
80 siliconeb
81 2-Phenylindole
82 Diethy lphtha late
84 Benzophenone
85 A].kane
86 Cycloalkane
87 n-Heptadecane
88 Alkane
89 2-Methoxy-di-t--buty lphenol
(continued)
11].
-------
TABLE 42 (continued)
Chromat
peak
ccr ica
number
90
91
Compound
Anthracene
Dibutyl phthalate
aRefe s to reconstructed total ion chrornatogram shown in
Figure 37.
components apparently due to system silanization
and/or column bleed.
112
-------
TABLE 43. COMPOUNDS COLLECTED ON PORAPAK R DURING THREE-
SORBENT AIR SAMPLING AT HOUSTON, TEXAS
Chromatographica
peak number Compound
1 Air
2 CO 2
3 Acetaldehyde
4 Benzene & Trimethylbutane
5 Dirrtethy lpentane
6 Ethylpentane
7 2, 2, 4-Trimethylpentaneb
8 2, 2, 3—Trimethylpentane
9 Tolueneb
10 2, 3, 4-Trimethylpentane
11 2,2, 3_Trimethylpentaneb
12 3-Ethy1 iexane
13 To lueneD
14 Trimethylpentane
15 Tetrachloroethylene
16 C —a1kan
17 Silicone
18 2, 3—Di ethylheptane
19 Xy1ene’
20 Ethy lbenzene
21 Xy leneb
22 2, 3-Dimethyiheptane
23 3, 4—Dimethyiheptane
24 Xylene
25 3-Methy l-3—ethylhexane
26 5-Methy l—2-ethylhexane
27 1—Methy l —2—ethy lcyc lohexane
28 n—Propylcyc lohexane
29 2,2, 3-Trimethyiheptane
30 C —Cyc1oa1kane
31 Ce—Cycloalkanes
32 C 10 —A lkane
33 C 3 -A lky lbenzene
34 3, 4-Diethyihexane
35 C 10 -A lkane
36 3-Ethy l-3—methy lheptane
37 C 10 -Cyc loparaffin
38 1,3, 5-Triinethylbenzene
39 3, 3—Diethyihexane
40 3-Methylnonane
41 1,2,4-Trixnethylbenzene & C 11 -Cycloalkafle
42 C 11 —A lkane
43 4—Methyldecane
.44 2-Methyldecane
(continued)
113
-------
TABLE 43 (continued)
Chromatocraph ca
peak number Compound
45 C 11 -Alkane
4€ MethylbutylcyCloheXane
C, 1 -A lkane
4 E 2, 2—DirnethylacetylcyCloheXafle
49 trans—Decalin
50 Alkane
51 C -Alkylbenzene
52 Alkane
53 o-Ethylstyrefle
54 p-Ethylstyrefle
55 C 11 -Alkane
5€ Diethyiphenol
57 2-Methyl-trans—deCalirl
•58 Alkane
59 2—Methyl—cis—decalin
60 Vinyiphenylcarbinol
61 Dimethyliridan
62 Ethylbenzaldehyde
63 Alkane
64 Ethylacetophenone
65 Alkane
aRefers to reconstructed total ion chromatograrri shown
in Figure 38.
bApparent doubling of peaks due to uneven trap warming
phenomenon.
CSilicone components apparently due to system silanizatiori
and/or column bleed.
TABLE 44. SUBJECT COMPOUNDS PRESENT IN SAMPLES
COLLECTED IN HOUSTON, TEXAS
Compound
Weight
per collector, rig
Total
weight,
rig
Cor icentration,a
ng/ 3 in air
Tenax-GC
Porapa
R
Be r izer le
151
55
—
—
206
35
450
76
Tetrach l or oethY le r ie
34
1
a 461 1iter of air sampled.
114
-------
sampled to assure sufficient material for analyses. The data ob-
tained during the Niagara Falls sampling shows that, even with a
large volume, Tenax may retain all of the organic constituents
under certain conditions. The sample volume taken at that loca-
tion was greater than that taken during Houston sampling and
nearly as great as that taken at Los Angeles. Moreover, the
loading of the specific compounds of interest was greater than
that found at the other locations.
Compound retention on Tenax sorbent may be affected by other
factors, the evaluation of which is beyond the scope of this
program. These may include selective displacement effects by
other matrix compounds; interaction with water whereby an
immiscible compound pair, having a combined vapor pressure higher
than that of either compound (as in steam distillation) is
formed; change in the surface characteristics of the Tenax due
to atmospheric constituents such as ozone or NOR.
Except under certain very controlled or specific conditions the
inultisorbent sampling approach is required to assure complete
retention of all compounds.
The methodology developed during the course of this program has
been applied to a wide variety of potential carcinogens to demon-
strate its effectiveness as a screening technique. As noted
previously, certain artifact compounds have been found in most
o the samples. Some of these, such as the silicones, are
recognized as originating from the analytical system. Great
importance has been placed on the recognition of artifact peaks
only if the compound is environmentally significant.
115
-------
SECTION 11
ANALYSIS OF FIELD SAMPLES BY MULTI-DETECTOR
CAPILLARY GAS CHROMATOGRAPHY
OBJECTIVE
The non-specificity of conventional gas chroinatographic (GC)
analyses has traditionally been a major drawback of the technique.
This has led to the heavy dependence on gas chromatography/mass
spectrometry (GC/MS) as the technique for accomplishing unequiv-
ocal identification of compounds. However, certain CC detectors
(e.g., electron capture, photoionization, nitrogen—phosphorus
specific flame ionization) offer degrees of selectivity for
certain compounds that have potential for decreasing the depend-
ence on GC/M5. if operated simultaneously during an analysis,
this permits the ratioing of different detector responses for
compounds “seen” by more than one detector. When used in con-
junction with GC retention times, detector response ratios pro-
vide an additional parameter which greatly improve the confidence
of compound specificity of a CC analysis. The objective of this
work was to conduct a preliminary investigation of multiple
selective detection and detector response ratioing as an alter-
native technique to mass spectrornetry for the detection of the
compounds of interest. This investigation was to include ana-
lyses of “field” samples and a comparison of multi—detector
capillary GC results with mass spectrometric results obtained
for comparable “field” samples.
INSTRUMENTATION
A multi—detector capillary gas chromatographic (MD(GC) 3 ) system
developed through independent research at Monsanto Research
Corporation was used to evaluate the applicability of this
analytical technique for this program. This system is shown in
Figure 40. It consisted of a Hewlett—Packard Model 5880 gas
chromatograph with dual printer/ plotters, three installed
detectors [ electron capture (ECD), flame ionization (FID), and
NP-flame ionization (NPDfl, dual computer interface boards, one
analog output board (installed during the course of these evalu-
ations), dual capillary injection ports, and a heated collection
vent. Each of the three installed detectors had independent
supplies of makeup carrier gases. Mounted to the side of the GC,
at the heated collection vent, was a HNU Inc. Model P1—52 photo-
ionization detector (PID). The PID was controlled by a power
supply/electrometer unit. Signals from the PID electrometer and
116
-------
NP-FID
4 - FID
6 - PID AND HEATED COLLECTION VENT
7 - PID POWER SUPPLY/ELECTROMETER
8 - HALL ELECTROLYTIC DETECTOR
9 - HALL POWER SUPPLY/ELECTROMETER
10 - HP A/D CONVERTERS
11 - HP STRIP CHART RECORDER
12 - CAPILLARY INLET SYSTEM THERMAL CONTROLLER
13 - CAPILLARY INLET SYSTEM VALVE OVEN
14 - CAPILLARY INLET SYSTEM DESORPTION CHAMBER
15 - CAPILLARY INLET SYSTEM TRAP
Figure 40. Multi-detector, capillary gas chromatograPh.
1
2
3
HP MODEL 5880 GC
PRINTER/PLOTTERS
5 - ECD
117
-------
the GC analog output board were plotted on a Hewlett-Packard
Model 7132A dual pen strip chart recorder and monitored by
computer through Hewlett—Packard Model 18652A analog to digital
(A/D) converters. A Tracor Model 700 Hall electrolytic conduct-
ivity detector was mounted above the PID so that the PID effluent
would be passed to the Hall detector. However, attempts to
detect compounds with the Hall detector had been unsuccessful
with the MD(GC) 2 system in this configuration. Therefore, the
Hall detector and its power supply/electrometer unit were dis-
connected and not used during these determinations. The capil-
lary inlet system, with desorption chamber, valve oven, nickel
capillary trap, and thermal control unit were mounted on top of
the GC.
A Hewlett—Packard, OV—l0]. WCOT, fused—silica column (‘ ..25 m long,
‘ 0.3 mm ID) was used as the analytical column during these
evaluations. The inlet of this column was passed up through the
GC injection port, through the injection port septum, and directly
connected to the capillary inlet system valve. Small exposed
portions of the column were ensheatheci in 1.6 iran (1/16 in.) OD
nickel tubing and wrapped with glass wool to ensure good heat
transfer. Figure 41 depicts a schematic of gas flows through the
capillary inlet system. The GC capillary carrier shown in this
schematic was obtained by inserting a 1.6 iran (1/16 in.) stainless
steel tee in the injection port carrier gas supply line immedi-
ately prior to the GC injection port. One arm of the tee was
capped and the other connected to the capillary inlet system.
(Other gas lines to the GC injection port were also capped to
prevent contamination of the GC.) The other end of the analyti-
cal capillary column was connected to a 1.6 mm (1/16 in..) stain-
less steel union with a nut and graphite ferrule. Four ‘ 30 cm
lengths of fused-silica columns were attached to the other end of
the union with a vent and specially formed graphite—ribbon fer-
rule. Two of these lengths were 0.3 iran ID, OV-lOl WCOT column
pieces and were connected to the PID and NP-FID. The other two
lengths were 0.2 mm ID, OV—lOl WCOT column pieces and were con-
nected to the ECD and FID. A photograph of the GC oven is given
in Figure 42.
The MD(GC) 2 system was interfaced to a Hewlett—Packard Model
3354 laboratory automation system. Prior to receiving the GC
analog output board, the system’s detectors were monitored as
described below:
• The PID signal was plotted on the strip chart recorder
and monitored by the computer with an A/D converter.
• The NP-FID signal was plotted on a printer/plotter but
was not monitored by the computer (i.e., plot only).
• The FID signal was plotted on the other printer/plotter
and monitored by the computer through one of the GC’s
computer interface boards.
118
-------
TRAP MODE
SAMPLE DESORBED FROM
SORBENT TUBE
INTO TRAP
GC CAPILLARY CARRIER
(—7 mLJmin)
ANALYZE MODE
• SAMPLE DESORBED
FROM TRAP INTO
CAPILLARY COLUMN
AUXIU ARY “DESORPTION”
CARRIER (—30 mLIm n)
CAPILLARY CARRIER
(—1 mLJmin)
Figure 41.
Flow schematic of capillary inlet system.
AUXILIARY “DESORPTION t ’
CARRIER
CAPILLARY
COLUMN
COLUMN
119
-------
Figure 42. Gas chromatograph oven with fused-silica
capillary column split to four detectors.
120
-------
• The ECD signal was not plotted but was monitored by the
computer through the other GC computer interface board
(i.e., computer only).
Signals monitored by the computer were integrated and the results
obtained from a DecWriter III Model LS—120 high speed terminal.
Figure 43 gives an example of a typical computer report. The
signals were also recorded and stored on the computer’s disc.
Chromatograms could be reconstructed from this stored information
on a Tektronix Model 4010—1 cathode ray tube (CRT) terminal and
printed by a Tektronix Model 4610 hard copy unit. After obtain-
ing the GC analog output board the following changes were made in
how the detectors were monitored.
• The NP-FID signal was outputted through the analog output
board, plotted on the strip chart recorder, and monitored
by the computer with an A/D converter.
• The ECD signal was plotted on a printer/plotter and
monitored by the computer through a GC computer interface
board.
• The PID and FID signals were monitored as described
previously.
RESULTS
A number of analyses were performed using the MD(GC) 2 system
with the chromatographic conditions given in Table 45. The only
variations in these conditions were the changes in the computer
interfacing, previously described, and a change in the level 2
program rate (from 8°C/mm to 3°C/mm) for the analysis of Tenax
tube #271 and one standard mixture.
One set of field sample tubes were analyzed, excluding the
Arnbersorb backup tube, on the MD(GC) 2 system. This sample was
collected in Houston, Texas, on 30 July 1980 with the ambient
air collection system. The chromatograms obtained for this set
of tubes are depicted in Figures 44, 45, and 46 for the Tenax,
Porapak R, and Ambersorb tubes, respectively. These chromato-
grams, except for the Tenax and Axnbersorb NP—FID tracers, are
computer reconstructed. They are normalized with the largest
peaks at full—scale and with the same time scale. Much chro
matographic resolution is lost by the reconstruction process
but this doesn’t affect computer integration. The two NP-FID
traces were obtained from a printer/plotter. Note that the
Tenax traces indicate most of the higher molecular weight
compounds collected on this sorbent tube, while only the most
volatile compounds are indicated as being collected by the
Ambersorb tube. For the Porapak R and Amnbersorb tubes the
first peaks eluting after temperature programming was initiated
121
-------
F EFDF T:
C
- CHANNEL: 23
MULT1-I ETECTO SYSTEM
SAMFLE: ECE INJECTEI A’T 140049 ON SE i , i eo
ZE D METHOr: CAFEC3
4CTUAL UtJ TIME 44.650 MiNUTES
F UN AF2RTEEI
AREA AREA NAME
7.04 88512 SF 1.301
814 c,s 15 11.94
8.17 45541 SF 6.69 ’S
9.47 8O323 S1 11.902
9.71 830431 11 12.202
279334 SF 4.090
10.44 20429 55 .300
10.64 21355 55 .334
10.79 179 3 11
11.16 5894473 FE 42.529
13.30 8106 51 .119’
16.45 7696 81 .113
17.36 6092 11’ .090
19.07 22663 51 .318
19.33 6544 51 .092
39.49 5205 51 .076
20.91 11022 bE .262
13552 85 .199
24.17 15179 5? .179
24.75 44532 55
25.58 9867 51 .145
27.49 39483 51 .272
28.32 5992 58 .007
29.90 6595 58 .097
31.09 5242 51 .077
31.78 12127 55 .163
33.5.4 5286 51 .078
33.49 6802 58 .100
34.53 13946 PS .205
35.07 40814 FE .600
35.42 5417 55 .080
36.74 5502 55 .081
37.23 280049 55 4.115
39.08 30310 55 .445
TCTA. AREA — 6805931 TOTAL AREA X 100.000
RDCESSEE’ 5A?A FILE: F’2OEP kA IIATA FILE: P22OER
Figure 43. Computer printout of computer integrated chrornatog-
raphic data from the analysis of Porapak R tube
22O with electron capture detection.
122
-------
TAI3LE 45. CIIROM! TOGRAPIITC CONDITIONS
I :ramel :r Vain. I’ararn.’ter Vain. — —
PID power supply/
electrometerr
mode
Plotter
Plotter
Plotter
Plotter
Signal
Channel
5i gnal
Channel
Signal
Signal
Channel
Signal
Channel
Chart speed
Pen Ii signal
Pen II and 12 range
Pen 12 signal
Tb *0 c
2 .l) C
2 .(IC
1 mm
—J0.C
S mm
30C mm
D.C
0 mm
8C/min
250C
20 sun
Splitless
Flu
0.5 cut/mm
ECO
0.5 cm/mm
FID
12
LCD
13
NP-FIn
NP-FlU
11
Pin
10
6.4 am/mm
(0.25 in./min)
[ ‘ID
10 utV
NP - rio
.230C
I0 Ajsp. ’j
Positive
5.1 marks
Vslv. t.nj 1 ’•rat nrc
rra( . ‘“‘i’ I •m ..ra [ nrc
“Analyz.” I .-.‘.p. I p., • ut ir
irs)J , I I r n ..
flesorit iOn t.mI.r .ralur .
nesorpI ion I 155
Type
Tank pressur”
Column back pressure
Column flow rates
Type
Tank pressure
CC prrssurr .
Plowrate
Type
Tank pressure
CC pressure
Flow rate
Type
Tank pressure
CC pressure
l’low rate
Type
Tank pressure
CC pressure
rrn [ tow rate
NP-FlU flow rate
Type
Tank pr .ssure
GC prr ’ssure
FlU (low rate
NP-Fit) flow rate
11.1 2
¶ ,nin
2 O . C,a Y)O.cb
V to 40 rntn
Helium
60 psi
21, i i
r .7 suL/min
Helium
60 psi
12 psi
30 eL/pain
Ilel ium
60 psi
10 psi
% 17 niL/sin
none
C
Arg/meth (5%)
50 psi
20 psi
“ .30 paL/mm
Air
40 psi
30 psi
%400 raL/min
55 mI,/mtn
Hydrogen
40 psi
30 psi
30 tnt/irAn
%4 inL/min
a 0 desorption of Porapak i i tubes.
bpor desorption of Tenax and Axiborsorb tubes.
methane in argon.
Detector zone 1 (NP—Flu)
Detector zone 2 (liD & Ecu)
lnjc ’Ct ion port un( . I
Auxiliary zone 1 (to Pl O)
Initial equilibration time
Initial value
Initial time
Level 1 program rate
Level I final value
Level 1 final time
Level 2 program rate
Level 2 final value
Level 2 final time
CC zone t.r .eraturs:
CC oven temperature
program:
SC capillary Injection
CC terminal 1:
C C terminal 2:
SC A/D loop board 1*
SC A/D loop board 2:
SC 0/A board:
A/ I) Converter:
A/P Converter:
Strip chart recorder:
p ..)
signal
chart speed
signal
chart speed
(a,i II .iry i ii I ci sy .L. pa:
Carrier r as:
Pl O makeup gas:
NP-Plo makeup gas:
PID makeup gas:
LCD makeup gas:
Detector qas:
Detector gas:
Detector temperature
Input attenuation
Recorder output polarity
Lamp intensity
-------
A.
B. P 10
D. ECD
Figure 44.
Analysis of Tenax tube U36,
a Houston air sample.
C. NPFID
124
-------
A. FID
B. Pio
Figure 45. Analysis of Porapak R tube #220,
a Houston air sample.
C. NP.FID
D. ec
125
-------
A. n
B. PID
Figure 46.
Analysis of Ainbersorb XE-340 tube #272,
a Houston air sample.
C. NP-FID
D. ECD
126
-------
were later than those for the Tenax tube. This was possible due
to a “solvent effect” created by large amounts of water in the
Porapak R and Ambersorb tubes.
Qualitative standard mixtures were prepared by putting lO mg
(i.e., ‘ l drop) of each mixture component into ‘ lO mL of solvent.
One mixture contained benzene, carbon tetrachloride, acrolein,
acrylonitrile, vinyl acetate, styrene, ethylene dichioride,
ethylene dibroinide, tetrachioroethylene, cis—l, 3—dichioroproperie,
trans-l,3-dichloropropene, and 1,4-dioxane in n-tridecane. A
typical analysis of this mixture on the MD(GC) 2 system is shown
in the chromatograrns in Figure 47. A similar mixture was prepared
containing toluene-2,4—diamine, benzyl chloride, hexachloro-l,3-
butadiene, and di—(2-ethylhexyl) phthalate in methanol. These
qualitative standards were used to establish retention time
ranges, detector responses, and detector response ratios for
compound identification. The results for a number of these
standards are compiled in Table 46.
Integration data from computer reports for sorbent tube analyses
were evaluated in comparison to the results of the standard
mixtures. An example of how these data were evaluated is given
in Table 47. Detector responses within the retention time ranges
of the compounds of interest were noted for each detector. The
retention times and areas of those responses were recorded beside
the compounds of interest in a table. Responses of different
detectors with close retention times were placed beside each
other and considered simultaneous responses to the same peak.
Determination of simultaneous responses was a subjective evalu-
ation. Allowances were made for differences in retention times
due to offset in computer integration and due to differences in
detector response intensities. The pattern of detector responses
was then noted for all simultaneous responses, and response ratios
were calculated as required, Simultaneous responses were then
assigned points according to the criteria given at the end of
Table 47. One mandatory criterion had to be met: the detector
responses had to be the correct pattern, or no points were
assigned. Responses potentially positively identifying the
compound of interest were then flagged with an asterisk. If
more than one group of simultaneous responses occurred within the
retention time ranges of some compounds, the group most likely
to indicate that compound was subjectively determined. The
number of points assigned, closeness to the retention time
optimunt, and potential identifications of other compounds were
among the factors considered. Tables of this type were pre-
pared for each analysis evaluated.
A compilation of results is given in Table 48. All compounds
flagged as being potentially positively identified are indicated
by the presence of FID retention times under the appropriate
analysis. The number of assigned points are then compared to
the total number possible and the percentage calculated. This
127
-------
A. rj
B. PID
I
Figure 47.
Multi-detector (GC) 2 analysis of a standard
mixture of acrolein, acrylonitrile, vinyl
acetate, ethylene dichioride, benzene, carbon
tetrachioride, 1, 4—dioxane, cis—l, 3-dichioro-
propene, trans-l, 3—dichloropropene, ethylene
dibromide, tetrachioroethylene, and styrene
in n—tridecane.
C
a
o 0
0
o a
C. NP-FID
D. ECD
1
0
9
128
-------
ot detected.
EEY FOR RESPONSES
TABLE 46. COMPILATION OF RESULTS FOR ANALYTICAL STANDARDS
+ Positive response
- Negative response
o No detectable response
I . . ,
‘.0
Compound
16 • tention
t imv.min
N . ‘ponses
U t j iii’. e ratios
Rarsjc
Optimum -
P11)
ECU
P 11)
N1’I)
ECU/FIt)
- P11)/FID
ECI)/PII)
1111 )/F it)
EC [ )JtIPIi
NPO/lID
Acrolein
5.6-6.6
6.09-6.11
4
0
4
0
1/1.23
Acrylonitrllo
5.9-6.9
6.38—6.39
+
0
0
31.7/1
Vinyl acetate
6.5-7.5
6.96-1.02
+
0
4
0
3.76/1
Ethylene dichloride
1.2-8.2
1.69—7.71
4
4
0
—
10.2/1
Renzone
7.4-8.5
7.93-8.02
4
0
4
0
8.17/1
Carbon tetz-achlortde
7.4-8.5
7.93-8.02
+
0
+
0
141011 a>
1121/1
( ) ) 8 / 1 a)
( 115 / 1 a)
1.4-Dioxane
8.2-9.3
8.73-8.80
4
0
4
0
3.40/1
cis-L.3-Dtch loroprene
8.8-9.9
9.34—9.41
4
4
+
131/1
3.24/1
46.6/1
trans—1,3—D ich loropropene
9.3-10.4
9.85—9.93
4
4
4
127/1
5.37/1
27.7/1
Ethylene dtbroøide
10.1-11.1
10.60-10.63
4
4
0
-
798/1
Tetrach loroethylene
10.4-11.5
10.93-11.01
4
+
4
1938/1
20.2/1
117/1
Styrene b
?o luene-2.4-diaine
12.2-13.3
14.1-15.4
12.70-12.79
14.60-14.90
4
4
0
+
+
4
0
4
90.4/1
9.99/1
8.20/1
15.9/1
NE
NE
NE
Ilexach lorc,—1 ,3-butadjene
19.0-20.2
19.47-19.69
-1
4
4
0
1528/1
23.5/1
64.9
Eenzyl chloride
21.0-22.3
21.65—21.74
4
+
+
0
5.94/1
3.02/1
1/5.94
D1—(2—ethylbexyl) phthalate
Ethylene g1ycolC
Pentach1o opheno1
Benzidin
Chrysene d
Benzo(a)pyrene
24.8-26.2
NEC
NE
NE
NE
NE
25.33-25.71
NE
NE
NE
NE
NE
4
ND
4
+
4
4
4
ND
4
4
-
+
NIt
0
0
0
0
0
NE
NE
NE
NE
NE
10.2/1
1098/1
2/6.7
63.5/1
1/6.50
apor co-eltuing peaks.
evaluated with computer integration of NP-FIn.
CDi.er of ethylene oxidej not detected at all, probably decomposed.
evaluated with appropriate chromatographic conditions.
eN 6 evaluated.
-------
I .-’
L)
0
Acrylonitrile 5 - 6 I 6 67 3.476
(6.34-6.39)
6.95 3,333
Vinyl .c.tate 6.5 - 7.5 6.67 3,476
(6 96—7.02)
6.05 3,333
7.04 2,353 7.04 *4,512
7.11 5.520
7.20 112.57*
6*4.433
7.27 6.301
7 30 6,14.2
7.32 5.462
7,38 26.151 7,41 814.205
7 46 10,5*0
ItIiyl.n. dich1.ri 7 2 - 4.2 7.20 212.519
(7.69-7.72) 7.23 6*4.433
7.27 6,3*1
7.30 6.142
7.33 5.462
‘7.3* 26.181 7.41 814.204
7.46 10.5*0
46nzsni id carbOn 7.4 - S.S 7.3* 26.1*1 7.41 *14,205
tetr.chlorid. (7*3—5,02)
7.46 10,540
4,37 7.546
4.45 5.379 8.47 403,23*
• • 0 0 37 4
/1
• 0 • 0
7.35 2,77* * • 0 • 31.1
/1
7.47 55.033 * 0 • 0
7.25 22,992 • 0 • 0
7.35 2.775 $ • 0 9 31.1
/1
7.47 55.033 • 0 * 0
1/ 2 3
9 42 /2
5.20 3
/1
3 60
/1
520
‘1
1/ 2*3
9.42 /1
1/ 2’93
942 /1
3.20 3
/1
S
S½
TABLE 47. EVALUATION OF RAW DATA OF PORAPAK R TUBE #220 ANALYSIS
Fl NLtroq. nI
Ieteu9tion •On t On unction r40tUre Phetoio..i&.tlon Phoup ) ,i $ os —
tint •T. UT. Area, 41’. Area. R’? Ar ... F ”D! PTD/ ECDI IIPL F ’DI
c ormd r.n , nit . n coe . 6 nm • b tin co*n7 . JrID*ECD PI0 NP0 flO Fit P1D — ‘
Ucroleil’ S 6 - 6,6 6 67 3,476
(6
6,85 1.545
6.55 1,545
7.53 402
3.60 3
/1
‘7.53 402
7.62 443
‘7.84 21,672 7.80
‘1.95 14.597 7.07
1.13 15.111 8.17 455.417 8.14
15,172
18.40*
48,610
7.47 55,033
7.53 402
7.62 643
7.84 21,672 7.80 18,172
7.05 14,5*7 7.97 15.40*
5.131 14,12* 11.171 455.417 $.14 45.610
5.34 25.118
7.35 2,714 6
5,24 25.118
• 0 • 0 1/
1.19
* 0 * 0 1.26
/1
* . • 0 25.1 2.65 9.37
/1 /1 j
• 0 * 31.1
/1
0 • 0
• 0 * 0 1/
1.19
* 0 * 0 1.26
/1
* . • 0 251 268 9.37
/1 /1 /1
• * 0 0 14*
/1
4
-------
TABLE 47 (continued)
lttr . I
50L* $ti 1005044 40 LJ.ctros o.plut. P 007I.t1.. 15p 64 10 ________
ha. T. Ire.. 0T , 6.0$. IT , Aros. 12. ICDI P401 LCD, NPD/ LCD!
C a’.J raa at. at. COe41 6 ala roa.t. 4 at. a.4a 4 •J , LCD 510 ISP P lO PlO SIb 5 17 WIll 5434444
l,4 -Dl .aa.. 1.2 1.3 1.14 15.11$
($73-ISO) S 37 1.144
5.43 5.374 5.47 101.131 • • ft 0 141
I l
5.65 17.747 5.6.1 4V 1 4 0 4 0 2.69 4
I L
1.73 $34 .433 0.72 10.400 0 4 • 0 0 I I
‘I
1.36 1.101
I 1$ $511
5.10 17.550
0.01 14.421
4.15 11.414
1.24 14.131 4,22 270 .334 1.20 45.405 4 4 • 0 16.1 3.76 600
/1 /1 /1
c-I,3-Ok6.la..- 1.1 - 3.4 4.74 5.101
P 00P. a . (1.34-0.41) 1.1$ 1,457
0.55 17,150
1.01 41.421
0.47 37,410
4.34 14.435 3.22 210,331 0.20 44,403 + * • 0 36.5 2 76 6.00 3 ’ .
/1 /1 Il
4.4* 11.156
4.63 23.431
460 1.071
0.31 1.474 0.31 14,404 • 0 * 0 7.34
( 1 /1
I ..a $-1,3-Uicl l.ro- 0.) - 10.4 0.44 11.11$
p( $p e O . (4.05-1.31) 3.63 23.411
4.40 7.473
0.51 1,074 3.31 14.404 • 0 4 0 7.34
‘I
10.02 4.331
10.23 143,4 0
30.34 40,201
30.44 20,420
Ehlyla.. (0 I . 13.1 10.24 343,440
dibrlde (10.60-10.43) 10.34 40,300
1044 20,421
15.66 30,4*3 10.64k 21.355 • $ 0 0 I/ 31
1_IS
10.71 25,030 10.70 17,033 • • 0 P 2/ 2
1.41
11.00 12,100 11.03 163 + 0 0 *
11.01 10,776
T.t ,.rLla ’r.- 30 4 - 11 S 30 44 20,435
•thyleoe (I0. 3-1l.O1) 10,44 25,444 10.04 21.355
10.70 25.031 10.70 17.113
11.00 U.140 11.03 163 0 0 4 1/
14 7
13.07 10,776
13.36 2,434,473
11 33 35.121
33.40 3,703
-------
TABLE 47 (continued)
potent In.. ._. L_ !A!! !.L! p1.it,5_
tie. 511, Ores, Pt, 0 . .
_r . -o . c i ’ sin v ..mts sin c mtu ’
12.2 - 10.1
(12 70-52.74) Il 25 2(40*
52.43 hOt?
52.72 12,145
12.45 24.444
11.01 14,111
teim.i ,eZ.4 24.2 -25.4
4 lm.lno (54 60-54.44) 14 10
24 32
1450 22.3*8
4 03 40 .410
15.04 60.844
15.24 0.Ofl
55.35 10.054
Im.d . l .ero ’I.l- 10 0 - 24.2 55.2’? 1.014
beI.dIon . fl9.41-20.t )
Ot...yl hlerI 22.0 — 22.3 20.44 15.201
(2 1. 65-21,74)
25 54 13.025
22.20 7,535 22.70
24.75 44,772
250*
74.59 5.4€?
25 64
25 15
25 44
Will iWJ fl
! !!_ A __!! !P’ ?!.9!.._
It T . 0 Are.. , *T. lr . ., _______!!, ._!!L___. frill rini furl erDl trill
ml.. coon.t , pin co r *ntl PIP PC I) LII I P I Po LID fID LID LIP liPS Poloto
52.56 7,450
22.17 24,250
12.14 12,744
13.34 1,457
1412 201
24.30 852 • 0 0 ft I!
56 0
14,47 11
________________________________ Ct (tot (a ______________________________
Correot tr ’lnatle ,q . of lpno ..r. . 110 10 PT . ..nq.
PT i.lthln rj ’timno PT r .flqe.
PT ..lthu , . 0 2 mm of ot’tln.n. PT ran
For peat; with one Tripod. rAtio:
000ponoe ri.t 4 ; wIthin rio other of .e *to4. I i. 0.5 to too range.
.04 wittitri too other, .5 magnitod. in 1 1.000 :00g.’,
0..peo.e r.tio not , .ltI.io ahOvo tritOn ..
Pot post; with more that, one ,r00.e ratio’
P ’ifo’noo ,.tio ..ithio on or ,t . . ’ of • . oito,io Is Si to
Ill S ranqr, .nd wIthin too order. of .. .g..itrrOe in 11,000 tong..
P.epoi:.e toUr. riot within above rltnI..
001poir.. f.clni, rorrect relouior to curb other.
00 .0 n .e. Ia 5l.0°0 to,..:. for LID or 150,000 trollS. for LID.
Fur. o.rj 00’,
Peat .r.o S. I . .. than th.t lodloated .hoo.
‘I por ratio
‘S 5,01 iaI lo
i’ . trot poaS
‘½ p .r p .0k
1121 344 0 0 0 • 5/
74.,
21.727
25,244
I-J
( A l
‘h - I
35 33
24 44
20,21 1.252
53.30 0.504
54.55 (.552
6.244
5.205
21.47 155,113
13.552 22.24k 315,7(3 4 0 0 0 6.8 63.5 1/
/1 /1 10 I I
Di.(7-rthylhney l) 24.4 - 20.2
pI.theI.te (25.31-25.71)
*1,7 (5
4.052
50,270
81. 154
RT - Petnett .o ti.o.
b 01 ,_, 5 , - Me.a ..,nI In .fIOe* .44.5.5. v i i .. ,arien aith tO.. hettrteo.
- Vim.. lonir.tin.. der.ot.r,
!CD - PlOrtOrin oaf.traro dotooter.
— Pi..rtolont..tto,. d t .ctor.
- P l troq rm/ pho.pI .oro. . - F 1 ismln.tIm. dotert..’.
0 Foint n - Fo(ritn are a , ;Iç io il and .44.4 recording to the nrh.. q(ven
.5 tO.. ri 3 ht.
etontion the. rang. optic...
• potential pooitbov Idnotifi,.t%o.. of rr o .ooi of interoat
.456*, Oh. s..ple.
thn .dntory onion. — if toil. totterS. 1. .itt ont, to... 11* prOntO are
coo;I.$e,pd to he mere and .rc ant ohor, In the above r.hi.
POlots
l2
02
I I
*0
-------
TAI3LE 48. COMPILATION OF ANALYTICAL RESULTS
Tenax lab spike Tcnax (hid blank Tenax tu1e 2M. A .ersorh tut,e
P ointS I’Otiitià I OU itS — POiflt
110 OT. b Pcr- 1 lID liT, Ptr FID liT, P ’r- lii) liT, Pe r-
______ C pound .in Act. ios. cent .in Act. Pox. cent .10 ACt. Pox. Cent am Act. Poe. cent
Acrotein 6.05 3 5 60 5.92 2 1/2 5 50 0 5 0 0 5 0
Acrylon itrtle 6.42 2 4 50 0 4 0 6.43 3 4 15 0 4 0
Vinyl acetate 0 5 O 1.41 3 S eo 6.96 5 5 100 0 5 0
Ethylene dichtortde 8.08 2 5 0 0 5 O 1.45 2 5 40 0 5 0
Hensene j7.51 4 1/2 8 1/2 5) 1.41 3 S 0 0.05 4 5 80 0 5 0
Carbon tetracbtorlde J 0 5 0 7.91 S 5 100 0 S 0
1.4-Dioxane 0 5 0 (1 5 0 8.58 3 5 60 0 5 0
cii—I.I -Oich loro-
UI pcoi.eume 9.56 1 1/2 8 1/2 88 0 8 1/2 0 0 8 1/2 0 0 8 1/2 (1
UI trans-h 3-Olebloro-
propehe 9.00 7 1/2 B 1/2 118 0 8 1/2 0 0 8 1/2 0 0 8 1/2 0
Ethylene dlbro.ide 10.55 4 5 80 0 5 0 10.11 3 S 60 10.59 3 5 60
TetgachtorOethylene 10.99 6 1/2 8 1/2 16 10.86 1/2 8 1/2 6 0 8 1/2 0 0 8 1/2 0
Styrene 0 5 0 0 5 0 0 5 0 0 5 0
To luene-2.4—dl aalne 0 8 1/2 0 14.84 2 8 1/2 24 0 81/2 0 0 8 1/2 0
Ilexachloro- I • 3-
butadjene 19.82 6 1/2 U 1/2 76 0 8 1/2 0 0 8 1/2 0 0 8 1/2 0
Benzyl chloride 21.23 4 1/2 0 1/2 5) 21.19 3 1/2 8 1/2 41 21.06 —1/2 8 1/2 —6 0 8 1/2 0
Di- (2-ethyihexyil
pbtha late 0 8 1/2 0 25.14 7 1/2 8 1/2 011 0 8 1/2 0 0 B 1/2 0
(continuedl
-------
TABLE 48 (continued)
Ac, -ole ln
Acry lr ’ntttlIA
Vinyl acetate
Ethy le ,w dtchtot ide
Benzrfle
Carbon tetr l ,Ioride
I l-Diox ane
c lq—l 1-flich1oro-
y ’topene
trAh -1, 3-DichiOro-
propelle
tthyI .ene dihrcaaide
T et t achlot oAthy lefle
Styrene
? ,,ltwne- 2,4—dlamine
Ik’xachloro-l. 3-
b uta ,itefl e
Pen yl chloride
Di- (2-ethylhexyl)
phthalatn
9.82 7 1/2 P 1/2 08
10.51 4 !‘ 770
10.90 5 5/2 8 1/2 65
12.74 5 5 100
o 81/2 0
o 81/ ? 0
o 8 1/7 0
O 0 7/2 0
o 8 1/2
10.06 3 5
o n 1/?
o S
o 81/2
2 1/2 0 1/? 29
0 77 1/7
(1 5
U 775/2
I I S
7761 Nt
Ion ‘,l)
777; I I , )
5 ’ 3 ’)
4’) 4’-) 2’)
no na 54
. .4 7,)
( .0 60 5))
tla.e recorded by flame Ionizat i on rletr ’ctor reupooxe.
1 ’Act. — Actual pointfl a!uaIqnerl .
- Potntx pomsibte.
4 Act. points
Percent — a 10(7.
Poc. points
rubes represent a single sample collected by the ambient Air collection sy tem.
1 Repiicat. sample collected with a portable miniature collectIon sy t e , analyzed wIth a slower temperature 1 .roqram.
9 ptean value of the percent poInts of the tenex field ampIes (tubes 216 and 2111.
value of the percent points of 1l field s.wap1e (tob, 230. 272, 270, and 271: or tubes 216. 272. and 720).
t tltmtnated because bentene/c rbon tetrachlorlde rea bud iarqc’r percent potuuts a,*t rioted prior to thus p..ak.
Co—eIutinq peaks.
tEll ted becauge peak I closer to thu. retention (lee of vinyl a.-etat ’.
1 NE - not evaluated.
1t. ,nd,uI -d fx Pora 1 ’.ak P Lu) ’- ‘‘u T . ,,.e I ,, ) ’ , ,‘
— -—--— —— — ---- - l u-Id t ’l-
Points I S,t,,(’ loh uIt’; —
P l O I1T. Per— TIP PT, — - ,— r io ci, - - Pr- 1 ‘
.n ,J ln Act. Ens, runt ml , , A ,t.Pos. c- nt gin Act. c S 5 .- c ) - nil
6.07
4
5
1;;
0
5
0
5.16
1,35
3
4
75
0
5
0
5.41
8.99
5
5
103)
7,27
1
5
rn
f ,(.6
7.93
9
1/2 01/2
:r::
:
0 1/2
ç
n .in
9.32
7
1/2 8 1/2
08
9.24
3
1/2 P 7/2
41
1 I. 79
4 5
4 5
2 5
7 1/7 77 1/7
0
00
(3
0
0
0
()
(3
Ut
IS 177
0 0
30 45
(7 ( 7
0 0
73 1 ; 0
O 0 1/? 0 771; NT. 74E NT 771; (3
22.20 3 1/2 0 1/2 41 741; NT U I NT NT 12
0 P 1/2 0 116 NT NI. NT N E 0
-------
percentage value does not represent probable certainty, but is
merely a score. The lab spike that was analyzed contained
3000 ng each of benzene, benzyl chloride, carbon tetrachioride,
tetrachioroethylene, ethylene dibromide, hexachloro—l, 3—butadiene,
and cis-l,3- -dichloropropene (some trans—l,3--dichloropropene also
was probably introduced as in impurity). It was prepared roTn a
vaporous sample from the Sample Generation System (Appendix 3).
Note that all of the spike components were potentially identified
during the analysis, but acrolein and acrylonitrile were also
indicated. The Tenax field blank indicated potential contazni.na-
tion by acrolein, vinyl acetate, tetrachioroethylene, benzyl
chloride, and di—(2—ethylhexyl) phthalate; however, some of these
compounds [ e.g., di—(2-ethylhexyl) phthalatel do not appear in
other analyses. Therefore, the blank may have been contaminated
in some manner to which the other sorbent tubes were not exposed.
The standard mixture (Figure 47) showed potential positive identi-
fications for all of its components when evaluated as previously
described (Table 47). Average values were calculated for the
percentage scores obtained for the Tenax and all the field samples.
Tenax tube #271 was analyzed using a slower temperature—program—
mirig rate (3°C/mm). It was the first tube of a three-tube series
collected in Houston on 30 July 1980 by a portable miniature
sampling system (developed under EPA Contract 68-02-2774). Ex-
cept for the sample volume, it would essentially be a duplicate
of the sample collected on Tenax tube #236.
The percentage scores determined in Table 48 were assigned relative
qualitative value (i.e., the higher the score the more likely the
potential positive identification was correct). These qualitative
values are compiled in Table 49 for the Houston field samples.
enzene, carbon tetrachioride, vinyl acetate, dioxane, ethylene
dibromide, and acrylonitrile were potentially identified in more
than one sample tube. Their identifications range in qualitative
value from possibly to probably correct. The chrornatograph in
Figure 48 presents an expanded view of the ECD trace obtained for
Porapak R tube #220. Detector responses (peaks) determined to
potentially represent some of the compounds of interest are
indicated.
DISCUSSION
As mentioned previously, during the analyses of the Porapak R and
Ambersorb tubes the first peaks eluting, after temperature program-
ming was initiated, eluted later than those for the Tenax tube.
This was attributed to a possible “solvent effect” created by
the large amounts of water in the Porapak R and Axnbersorb tubes.
Therefore, the water in these tube samples may have displaced
early eluting peaks causing retention times to deviate from the
retention time ranges of these compounds in the analytical stand—
ards. The shifting of compound retention times due to matrix
effects represents a major limitation of (GC) 2 analysis.
Such matrix effects can be due to any component of a sample
135
-------
F±gure 48.
Expanded presentation of Porapak R
tube *220, ECD trace.
ECD
C
9 0
3
0
I .-
L J -
=
c .)
C
0
C
136
-------
TABLE 49. RESULTS OF HOUSTON FIELD SAMPLES
Compound
U
4,
s -I
£4
.0
0
I
a.
Tenax tube
1236ã
44
4,
.-4
.0
.5.4
41 414
41 41
0 s-I
0. .0
..I
41
44 41
41 0
0.
Tenax tube
1211 b
Porapak RC
AmbersorI
Overall lloust(,n air
0
41
s-I
.0
4,
.0
0
44
0.
44
•
5_ •
. 0
s_I
41
41
0
0.
-.
4. .
44
414
44
s-I
.0
In
41
0
0.
0
4 )
s-I
.0
.4
£4
0
44
0.
41
.4
54
.0
. 1
41
41
0
0.
),
I - ,
4)
414
44
s-I
.0
.5.4
14
.4
0
0.
U
41
.-I
.0
41
.0
0
44
0.
41
44
.
.0
.5. 4
41 41.
41 41
0 s-s
0. .0
s -I
41
1. 4)
41 C)
0.
41
.0
-.4
u’ in
II 41
s-I 0
Q 0.
.4
.0 >.
0 4. .
4..
4... >
.4 -4-.
4’ ) V
.-4 .-4 s -I
.0 )J.O
--4 .c• - .
in 0 41
41 -.15
0 -1
0. U)
Carbon tetrachloride
I
X
x
X
Benrene
X
X
x
Vinyl acetate
I
X
x
X
Dioxane
X
X
X
X
Ethylene dibromide
X
X
I
I
Acrylonitrile
I
I
I
I
?.crolein
X
I
Ethylene dichloride
X
I
I
Benzyl chloride
I
I
Dichloropropene
I
I
aTu es represent a single sample collected by the ambient air collection system.
bReplicate sample collected with a portable miniature collection system; analyzed with a slower temperature program.
Cwater in sample may have displaced early eluting peaks causing retention times to deviate from the retention time
ranges of these compounds in an analytical standard.
dpercent points from Table 85.
epercent points from Table
percent points from table
9 Average percent points of all
hi veragc percent points of all
Average percent points of all
Average percent points of all
I- .
t .J
-4
58 to 84.
31 to 58.
field samples from Table
field samples from Table
field samples from Table
fIeld samples from Table
60.
= 49 to 59.
= 30 to 40.
— 1 to 37.
-------
(e.g., water) or the combination of components in that sample.
Consequently, compounds can elute within the retention time
ranges of other, similar compounds and be mistakenly identified
as the other compound. Alternately, compounds actually present
in a sample might not be identified at all. All (GC) 2 analyses
which rely heavily on retention times for compound identification
have this limitation. Analyses performed with spectral detection,
such as (GC) 2 /MS or (GC) 2 /FTIR, use spectra for compound identi-
fication and do not rely as heavily on retention times.
In addition to matrix effects, which are a limitation of all (GC)
analyses, MD(GC) 2 has other limitations. The determination of
which peaks are simultaneous observed by different detectors is
somewhat subjective. Compensation must be made for any offset
in the beginning of different detector integrations and allow-
ances made for differences in integration due to different in-
tensities in detector responses. Manual determination of
simultaneous peaks allows flexibility of interpretation but is
subject to the inconsistency of the interpreter. On the other
hand, computer determination allows consistency but not flexibil-
ity. Once simultaneous peaks are determined, detector responses
and response ratios can be obtained. These results also must be
subjectively evaluated to determine whether a compound is poten-
tially identified and the reliability of a potential identifica-
tiori. Compound identification is complicated by co—eluting com-
pounds (e.g., benzene and carbon tetrachioride); by variable
detector—response ratios, which occur if one or more detectors
are responding in a non-linear response range; and by multiple
compounds of similar type having similar detector responses and
response ratios eluting within a given retention time range
(e.g., cis and trans-l,3—dichloropropene).
The results of the MID(GC) analyses of the Houston field samples
are summarized in Table 49 and the results of the (GC) 2 /NS anal-
yses of replicate Houston field samples are summarized in
Table 44 (see Section 10). The presence of benzene in the Tenax
and Porapak R tubes is indicated in both sets of results. Tetra-
chioroethylene, which was determined by (GC) 3 /MS to also be
present in the Tenax and Porapak R tubes, was not indicated by
the MD(GC) 2 analyses. Conversely, carbon tetrachlorjde, acrylo-
nitrile, vinyl acetate, 1,4—dioxane, and ethylene dibrornide were
tentatively identified in the field samples by MD(GC) 2 .
The detectors of the MD(GC) 2 system are much more sensitive than
is a mass spectrometer, and some of these tentative identifica-
tions might be correct but below the detection level of the
mass spectrometer. Alternately, some of the discrepancies in
compound identification might be due to a combination of the
limitations discussed above for MD(GC) 2 analyses.
138
-------
These results indicate that although MD(GC) 2 is much more
selective and specific than (GC) 2 or GC, it cannot replace GC/MS
for the unequivocal identification of compounds. MD(GC) 2 can be
used to indicate the possible presence of selected compounds
although matrix effects and other limitations can imply the
presence of compounds actually not present and the absence of
compounds actually present. Perhaps a better use for MD(GC) 2
would be to use this technique to identify various types of
compounds in samples and their overall “total” amounts. This
would provide much more information than a total chromatograPh
able orgariics analysis (TCO) about the composition of an air
sample, and could perhaps be an indicator of overall air quality
in terms of organic pollutants.
139
-------
REFERENCE S
1. MeNillin, C. R., L. B. Mote, and D. G. DeAngelis. Potential
Atmospheric Carcinogens, Phase 1: Identification and Classi-
fication. EPA—600/2—80-015, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, January 1980.
253 pp.
2. Weast, R. C., ed. CRC Handbook of Chemistry and Physics,
55th Ed. CRC Press, Cleveland, Ohio, 1974. 2300 pp.
3. Small, P. A. Some Factors Affecting the Solubility of
Polymers. Journal of Applied Chemistry, 3(2):71—90, 1953.
4. Burrell, IL, and P. Immergut. Solubility Parameter Values.
In: Polymer Handbook, J. Brandup and E. H. Irrunergut, eds.
Interscience Publishers, New York, New York, 1966.
pp. IV—341—IV—368.
5. Chemical Abstracts, Ninth Collective Index, Formulas.
ISSN:0097—6474. Library of Congress Cat. No. 9—4698.
American Chemical Society, Chemical Abstracts Services,
Columbus, Ohio, 1978.
6. Sax, N. I. Dangerous Properties of Industrial Chemicals.
Van Nostrand Reinhold Company, Cincinnati, Ohio, 1975.
7. Patty, F. A., ed. Industrial Hygiene and Toxicology,
2nd Ed., Vol. II. Interscience publishers, New York,
New York, 1962.
8. Eight Peak Index of Mass Spectra, 2nd Ed., Vol. I. Mass
Spectrometry Data Centre, Altermaston, Reading, U.K., 1974.
9. Analytical Methods for Use in Occupational Hygiene: Deter-
mination of Benzo(a)pyrene and Benzo(k)flUOranthefle in
Airborne Particulates (Chromatography and Optical Fluores-
cence). Pure and Applied Chemistry, 40(3):36—l—36—7, 1974.
10. Bake:, B. Measuring Trace Impurities in Air by Infrared
Spectroscopy at 20 Meters Path and 10 Atmospheres Pressure.
American Industrial Hygiene Association Journal, 35(11):
735—840, 1974.
140
-------
11. Ballou, E. V. Second NIOSH Solid Sorbents Roundtable. HEW
Publication No. NIOSH-76-193. u.s. Department of Health
Education and Welfare, Washington, D.C., 1976.
12. Baretta, E. D., C. A. Hawk, P. S. Bodeau, J. E. Anderson,
C. Dickerson, and N. P. Adriano. Chemical Material Handling
Guides. American Industrial Hygiene Association Journal,
39(ll):898—903, 1978.
13. Bertsch, W., and E. Anderson. Trace Analysis of Organic
Volatiles in Water by Gas Chromatography-MaSS SpectrometrY
with Glass Capillary Columns. Journal of Chromatography,
1l2(l):701—718, 1975.
14. Bertsch, W., R. C. Chang, and A. Zlatkis. The Determination
of Organic Volatiles in Air Pollution Studies: Character-
ization of Profiles. Journal of ChromatographiC Science,
12(4):175—182, 1974.
15. Burnett, R. D. Evaluation of Charcoal Sampling Tubes.
American Industrial Hygiene Association Journal, 37(1):
37—45, 1976.
16. Cautreels, W., and K. Van Cauwenberghe. Experiments on the
Distribution of Organic Pollutants Between Airborne Parti-
culate Matter and Corresponding Gas Phase. Atmospheric
Environment, 12(5) :1133—1141, 1978.
17. Chrostek, J. Health Hazard Evaluation/Toxicity Determina
tion Report. NISOH—TR-HHE 74-23-215. PB—249396. National
Institute for Occupational Safety and Health, Cincinnati,
Ohio, 1975. 7 pp.
18. Criteria for a Recommended Standard: Occupational Exposure
to Dioxane. HEW Publication No. NIOSH-77-226. u.s. Depart-
ment of Health, Education, and Welfare, Washington, D.C.,
1977. 193 pp.
19. D’Agostifle, R. B., and J. C. Gillespie. Comments on the
OSHA Accuracy of Measurement Requirement for Monitoring
Employee Exposure to Benzene. American Industrial Hygiene
Association Journal, 39(6):510—513, 1978.
20. Dietrich, M. W., and L. M. Chapman. Determination of Benzyl
Chloride in Delaware River Plant Air. Special Study: S—75—
SS-13. Monsanto Company, St. Louis, Missouri, 1975.
21. Eisenberg, w. c. Fractionation of Organic Material Extracted
from Suspended Air Particulate Matter Using High Pressure
Liquid ChromatograPhY. Journal of ChromatographiC Science,
16(4) :145—151, 1978.
141
-------
22. Fishbeii-i, L. Potential Hazards of Fumigant Residues. En-
vironmental Health Perspectives, 14:39—45, 1976.
23. Giam, C. S., Ii. S. Chan, and G. S. Neff. Rapid and Inex-
pensive Method for Detection of Polychiorinated Biphenyls
and Phthalates in Air. Analytical Chemistry, 47(13):
2319—2320, 1975.
24. Giam, C. S., H. S. Chan, G. S. Neff, and E. L. Atlas.
Phthalate Ester Plasticizers: A New Class of Marine
Pollutant. Science, 199(4327):419—42l, 1978.
25. Gordon, R. J. Distribution of Airborne Polycyc].ic Aromatic
Hydrocarbons Throughout Los Angeles. Environmental Science
and Technology, 10(4):370—373, 1976.
26. Grote, A. A., W. S. Kiln, and R. E. Kupel. Establishing a
Protocol from Laboratory Studies to be used in Field
Sampling Operations. American Industrial Hygiene Associ-
ation Journal, 39(4):880—884, 1978.
27. Harrison, R. M. The Analysis of Tetraalkyl Lead Compounds
and their Significance as Urban Air Pollutants. Atmospheric
Environment, 2(9):847—852, 1977.
28. Hill, R. H., Y. T. Gagnon, and A. W. Teass. Evaluation and
Control of Contamination in the Preparation of Analytical
Standard Solutions of Hazardous Chemicals. American
Industrial Association Journal, 39(2; :157—160, 1978.
29. Hurley, G. F.,, and N. H. Ketcham. A Solid Sorbent Personal
Sampling Method for the Determination of Acrolein in Air.
Ainrican Industrial Hygiene Association Journal, 39(8):
615—619, 1978.
30. Jacobs, H. W., and R. H. Syrjala. The Use of Infrared
Analyzers for Monitoring Acrylonitrile. American Industrial
Hygiene Association Journal, 39(2):l6l—165, 1978.
31. Li, R. T., L E. Going, and J. L. Spigarelli. Sampling and
Analysis of Selected Toxic Substances: Task lB — Hexachioro-
butadiene. EPA/560/6—76—015. Environmental Protection
Agency, Washington, D. C., 1976. 152 pp.
32. Lloyd, J. W., general chairman. Proceedings of NIOSH
Styrene - Butadiene Briefing. HEW Publication No. NIOSH-
77—129. U.S. Department of Health, Education, and Welfare,
Washington, D. C., 1977. 169 pp.
142
-------
33. McGuirk, M., and S. J. Mainwaring. Reversed Two-Dimensional
Technique for Multiple Separations of Benzo(a)pyrefle from
Atmospheric Aerosol Samples. Journal of Chromatography,
l35(l):24l—244, 1977.
34. Mulik, J. D., M. Cooke, M. F. Guyer, C. M. Semeniuk, and
E. Sawicki. A Gas Liquid Chromatographic Fluorescent
Procedure for the Analysis of Benzo(a)pyrene in 24 Hour
Atmospheric Particulate Samples. Analytical Letters,
8(8) :511—524, 1975.
35. Nelson, G. 0., and C. A. Harder. Respirator Cartridge
Efficiency Studies: VI Effect of Concentration. American
Industrial Hygiene Association 3ourna1, 37(4):205—216, 1976.
36. Nutt, A. Measurement of Some Potentially Hazardous Mate-
rials in the Atmosphere of Rubber Factories. Environmental
Health Perspectives, 17:117—123, 1976.
37. Pierce, R. C., and M. Katz. Chromatographic Isolation and
Spectral Analysis of Polycyclic Quinones: Application of
Air Pollution Analysis. Environmental Science and Technol-
ogy, 10(l):45—51, 1976.
38. Reckner, L. R., and J. Sachdev. Collaborative Testing of
Activated Charcoal Sampling Tubes for Seven Organic Solvents.
HEW Publication No. NIOSH—75-184. U.S. Department of Health,
Education, and Welfare, Washington, D.C., 1975. 227 pp.
39. Robinson, J. W. Further Comments on the Analysis of Tetra-
alkyl Lead Compounds and Their Significane as Urban Air
Pollutants. Atmospheric Environment 12(5):1247—1248, 1978.
40. Russell, J. W., and L. A. Shadoff. The Sampling and Deter-
mination of Halocarbons in Ambient Air Using Concentration
on Porous Polymer. Journal of Chromatography, 134(2):
375—384, 1977.
41. Severs, L. W., R. G. Meicher, and M. T. Kocsis. Dynamic
U-Tube System for Solid Sorbent Sampling Method Development.
American Industrial Hygiene Association Journal, 39(4):
321—326, 1978.
42. Spangler, C., and N. de Nevers. Benzo(a)pyrene and Trace
Metals in Charleston, South Carolina. EPA-450/2-75-004.
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, 1975. 58 pp.
43. Yasuda, S. K., and E. Dan Loughran. Air Sampling Methods
for s-Tetrachlroethane and Other Related Chlorinated Hydro—
carbons. Journal of Chromatography, l37(2):283-292, 1977.
-------
44. Yrjanheikki, E. A Method for Personnel Sampling and
Analyzing of Phenol. American Industrial Hygiene Association
Journal, 39(4):327—330, 1978.
45. Zeldes, S. G., and A. D. Horton. Trapping and Determination
of Labile Compounds in the Gas Phase of Cigarette Smoke.
Analytical Chemistry, 50(6):779—782, 1978.
46. Brooks, J. J., and D. S. West. Solid Sorbent Digest.
Monsanto Research Corporation, Environmental Analytical
Sciences Center, Dayton, Ohio, 1978. 126 pp.
47. Brooks, J. J., D. S. West, D. J. David, and 3. D. Mulik. A
Combination Sorbent System for Broad Range Organic Sampling
in Air. In: Proceedings of the Symposium on the Develop-
ment and Usage of Personal Monitors for Exposure and Health
Effect Studies. EPA—600/9—79—03 2 , U. S. Environmental Pro-
tection Agency, Research Triangle Park, North Carolina,
June 1979. 525 pp.
48. Cram, S. P., F. J. Yang, A. C. Brown III. Characterization
of High Performance Glass Capillary Gas Chromatography.
Chrornatographia, 10(8), 1977.
49. Brooks, J. J., D. S. West, J. E. Strobel, and L. Stamper.
Jet Exhaust Analysis by Subtractive Chromatography.
SAN-TR-78-37, USAF School of Aerospace Medicine, Brooks
Air Force Base, Texas, Dec. 1978. 76 pp.
50. Sawicki, E. Analysis of Atmospheric Carcinogens and Their
Cof actors. In: Environmental Pollution and Carcinogenic
Risks, INSERN Symposia Series, Vol. 52, IARC Scientific
Publications No. 13, World Health Organization, Internation-
al Agency for Research on Cancer, 1976. pp 297-354.
51. Barger, W. R., and W. D. Garrett. Surface-Active Organic
Material in Air over the Ligurian Sea and over the Eastern
Equatorial Pacific Ocean. NRL—7897, Naval Research Labora-
tory, Washington, D.C., 30 June 1975. 20 pp.
52. Rondia, D., F. De Wiest, and H. Della Fiorentina. OrgafliCs
in Atmospheric Aerosols in Belgium. State UniversitY of
Liege, Belgium. International Conference on Environmental
Sensing and Assessment, Las Vegas, Nevada, 14-19 September
1975. Published by IEEE, New York, Vol. 2, Paper 24—5,
1976. 4 pp.
53. Method 5 - Determination of Particulate Emissions from
Stationary Sources. Federal Register, 42(160):4l77641782,
August 1977.
144
-------
54. Method 17 — Determination of Particulate Emissions from
Stationary Sources (In—Stack Filtration Method). Federal
Register, 43(37):7584—7596, February 1978.
55. Sharkey, A. G., J. L. Schultz, C. White, and R. Lett.
Analysis of Polycyclic Organic Material in Coal, Coal Ash,
Fly Ash, and Other Fuel and Emission Samples. EPA—600/2
76—075, u.s. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1976. 31 pp.
56. Okita, T. Filter Method for Determination of Trace Quanti-
ties of Amines, Mercaptaris, and Organic Sulfides in the
Atmosphere. Atmospheric Environment, 4(1): 93—102, 1970.
57. Leskovar, B., C. Lo, and B. Turko. Multiparameter Fluores-
cence Method for Monitoring In Situ Particulates. U.S.
Energy Research and Development Administration, Washington,
D.C., 1977.
58. Davis, W. D. Continuous Mass SpectrometriC Analysis of
Particulates by Use of Surface Ionization and Continuous
Mass SpectrometriC Determination of Concentration of Parti-
culate Impurities in Air by Use of Surface Ionization.
Environmental Science & Technology, 11(6):587, 1977.
59. CantreelS, W., and K. Van cauwenberghe. Determination of
Organic Compounds in Airborne Particulate Matter by Gas
Chromatography/Mass Spectrometry. Atmospheric Environment,
lO(6):447—457, 1976.
60. Eisenberg, w. c. Fractionation of Organic Material Ex-
tracted from Suspended Air Particulate Matter Using High
Pressure Liquid ChromatograPh. 1 L ChromatograPhiC Science,
16(4):145—l51, 1978.
61. Stanley, T. W., M. J. Morgan, and J. E. Meeker. Rapid
Estimation of 7,8_Benz(delaflthracen- 7 —one and Phenalefl-l
one in Organic Extracts of Airborne ParticulateS from
3—Hour sequential Air Samples. Environmental Science &
TechnolOgy, 3(11) :1198—1200, 1969.
62. Kurasek, F. W., D. W. Denney, K. W. Chan, and R. E. Clement.
Analysis of Complex Organic Mixtures and Airborne Particu-
late Matter. Analytical Chemistry, 50(1):82, 1978.
63. Dong, M. W., C. Locke David, D. Hoffman, and D. Naylor.
Characterization of Aza-Arenes in Basic Organic Portion of
Suspended Particulate Matter. Environmental Science &
Technology, ll(6):612, 1977.
145
-------
64. Bove, J. L., and V. P. Kukreja. A New Mechanical Disruption
•Technique for Organic Enrichment of Hi—Vol Samples. Envir-
oniriental Letters, l0(2):89, 1975.
65. Bove, J. L., and V. P. Kukreja. An Enrichment Method for
Polycyclic Aromatic Hydrocarbons Collected on Glass Fiber
Filters Using Hydrofluoric Gas. J. Environmental Science &
Health — Environmental Science & Engineering., ll(8—9):5l7,
1976.
66. Dietrick, M. ., L. M. Chapman, and J. P. Misure. Sampling
for Organic Chemicals in Workplace Atmospheres with Porous
Polymer Beads. American Industrial Hygiene Association
Journal, 39(5):385—392, 1978.
67. Adams, J., K. Menzies, and P. Levins. Selection and Evalu—
atior of Sorbent Resins for the Collection of Organic Com-
pounds. EPA—600/7-77—044, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, April 1977,
67 pp.
68. Gallant, R. F., J. W. King, P. L. Levins, J. F. PieCeWiCZ.
Characterization of Sorbent Resins for Use in Environmental
Sampling. EPA—600/7-78—054, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, March 1978.
163 pp.
69. Snyder, A. D., F. N. Hodgson, M. A. Keminer, and J. R.
McKendree. Utility of Solid Sorbents for Sampling Organic
Emissions from Stationary Sources. EPA—600/2—76-201, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, July 1976. 77 pp.
70. Rhoades, J. w., and D. E. Johnson. Evaluation of Collection
Media for Low Levels of Airborne Pesticides. EPA—600/l-77
050, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, October 1977. 140 pp.
71. Strup, P. E., P. W. Jones, R. D. Giarnmar, and T. B. Stanford.
Efficient Collection arid Analysis of Hazardous Organic Com-
pounds from Combustion Effluents. Battelle—Coluinbus Labora-
tory, Ohio. International Conference on Environmental Sens-
ing and Assessment, Las Vegas, Nevada, 14-19 September 1975.
Published by IEEE, New york, Vol. 2, Paper 22-3 1976.
4 pp.
72. Winterlin, W., D. G. Crosby, and W. W. Kilgore. Analytical
Research on Economic and Environmental Toxicants. Grant
67347, U.S. Dept. of Agriculture, Washington, D.C., 1978.
146
-------
73. Bunn, W. W., E. R. Deane, D. W. Klein, and A. D. Klecpfer.
Sampling and Characterization of Air for Organic Compounds.
Water, Air and Soil Pollution, 4(3 —4):367380 , 1975.
74. Barrett, W. J. Identification of Organic CompoundS in the
Atmosphere. Grant PHS-AP- O045 4 , U.S. Public Health Service,
1968.
75. Smith, M. S., A. 3. Francis, and 3. M. Duxbury. Collection
and Analysis of Organic Gases from Natural EcosyStCXflS
Application to Poultry Manure. Environmental Science &
Technology ll(l):51—55, 1977.
76. Pellizzari, E. D., 3. E. Bunch, R. E. Berkley, and 3. McRae.
Determination of Trace Hazardous Organic Vapor Pollutants
in Ambient Atmospheres by Gas romatographY/MaSs SpeCtrO
rnetry/COtflPUter. Analytical Chemistry, 48(6) :803, 1976.
77. EricksOn, M. D., R. A. Zweidinger, L. c. Michael, and E. D.
Pellizzari. Environmental Monitoring Near Industrial Sites:
PolychlorortaPhthale EPA-560/6—7 7 — 019 v U.S. Environ-
mental Protection Agency, Washington, D.C., June 1977.
267 pp.
78. Lewis, R. G., A. R. Brown, and M. D. Jackson. Evaluation
of polyurethane Foam for sampling of pesticides, polychlorifl
ated BiphenylS and Polychioriflated Naphthalefles in Ambient
Air. Analytical Chemistry, 49(12):16681672 , 1977.
79. MargesOn, 3. H. Methodology for Measurement of PolychlOrifl
ated BiphenylS in Ambient Air and Stationary Sources - A
Review. EPA—600/4- 77 O 2 l ’ u.s. Environmental ProtectiOn
Agency, Research Triangle Park, North Carolina, April 1977.
39 pp.
80. Bidleman, T. F., and C. E. Olney. High—Volume Collection
of Atmospheric polychloriflated Biphenyls. Bulletin of
Environmental Contamination and ToxicOlOgY, 1l(5):442449 ,
1974.
81. Haile, C. L., and E. Baladi. Methods for etermifliflg the
Polychioriflated Biphenyl Emissions from Incineration and
Capacitor and Transformer Filling Plants. EPA-600/4 77048 ’
u.s. Environmental Protection Agency, Research Triangle
Park, North Carolina, November 1977. 94 pp.
82. Mann, 3. B., H. F. Enos, 3. Gonzalez, and 3. F. Thompson.
Development of sampling and Analytical Procedure for Deter-
mining HexachlOrObeflzene and HexachlOrO_1,3_but iene in
Air. Environmental Science & Technology, 8(6):584, 1974.
147
-------
83. Reckner, L. R. and J. Sachdev. Collaborative Testing of
Activated Charcoal Sampling Tubes for Seven Organic Sol-
vents. NIOSH-75/184, National Institute for Occupational
Safety and Health, Cincinnati, Ohio, June 1975, 229 pp.
84. Flisak, F., and J. Kubarewicz. Sampling Methods and Instru-
mental Analyses o Airborne Toxic Chemicals in Workroom
Environments. In: Proceedings of a Special Conference on
Industrial Pollution Control Measurement and Instruinenta-
tion, Technamic Publishing Co., Westport, Connecticut,
1976. pp 162—165.
85. Carter, R. V. and B. Linsky. Gaseous Emissions from
Whiskey Fermentation Units. Atmospheric Environment,
8(l):57—62, 1974.
86. Zweidinger, R. A. Organic Emissions from Automobile
Interiors. EPA-600/7—77-149, U.S. Environmental Protection
Acency, Research Triangle Park, North Carolina, December
1977. 90 pp.
87. Distenfeld, C. H., and J. R. Klernish. High Efficiency
Mixed Species Radioiodine Air Sampling, Readout, and Dose
Assessment System. Presented at IAEA/OECD Handling of
Radiation Accidents Symposium, Vienna, 28 February —
4 March 1977, p 463.
88. Sprouse, J. F. New Sampling Technique for Identifying
Organic Film Accumulation on Army Material in the Tropics.
Presented at 6th Annual Symposium on Environmental Research,
Edgewood Arsenal, Aberdeen Proving Ground, Maryland,
29 April - 1 May 1975. Published by Dept. of Army,
Edgewood Arsenal, Aberdeen Proving Ground, Maryland,
EO—SP—76001, 1976.
89. Feairheller, W. R., P. J. Marn, D. H. Harris, and D. L.
Harris. Technical Manual for Process Sampling Strategies
for Organic Materials. EPA—600/2—76—122, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, April 1976. 172 pp.
90. Hamersma, J. W., S. L. Reynolds, and R. F. Maddalone. IERL-
Procedure Manual: Level I Environmental Assessment. EPA-
600/2-76-160a, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, June 1976. 147 pp.
91. Kutz, F. W., and H. S. C. Yang. A Note on Polychiorinated
Biphenyls in Air. U.S. Environmental Protection Agency,
Washington, D.C. (PB—275 978).
148
-------
92. Conkle, 3. P., and R. L. Miller. Cryogenic Sampling of
Ambient Atmospheres on a Means of centratiflg Trace
Organic Compounds for Chemical Analyses. presented at the
6th Annual Symposium on Environmental Research, EdgewoOd
Arsenal, Aberdeen Proving Ground, Maryland, 29 April —
1 May 1975. Published by Dept. of Army, EdgeWOOd Arsenal,
Aberdeen Proving Ground, Maryland, EO—SP—7 6001 , 1976.
93. Salo, A. E., W. L. Oaks, and R. D. MacPhee. Measuring the
Organic Carbon Content of Source Emissions for Air Pollution
Control. 3. Air Pollution Control Association, 25(4):390
393, 1975.
94. Appel, B. R., P. Colodny, and 3. 3. wesolowski. Analysis
of Carbonaceous Materials in Southern California Atmosphere
Aerosols. Environmental Science & Technology, l0(4):359-
363, 1976.
95. Guerin, 0. G., and W. T. Rainey. Gas romatogra c
Determination of Nonvolatile Fatty Acids in Cigarette Smoke.
Analytical ChemistrY, 46(6) :761—763, 1974.
96. Akin, F. J., M. E. Snook, R. B. Severson, W. 3. Chamberlain,
and D. B. Walters. Identification of Polynuclear Aromatic
Hydrocarbons in Cigarette Smoke and Their Importance as
Tumorigens. J. National Cancer Institute, 57(1):191—195 ,
1976.
97. Sherina, 3., and T. M. Shafik. A MulticlaSS, Multiresidue
Analytical Method for Determining pesticide Residues in Air.
Archives of Environmental Contamination and Toxicology,
3(l):55—7l, 1975.
98. WittgenSteifl , B., E. Sawicki, and R. L. Antonelli. Freeze
Drying in the Recovery of Organic Material from Extracts of
Air particulate Matter. Atmospheric Environment, 5(9):80l
810, 1971.
99. Russel, . w. Analysis of Air Pollutants Using Sampling
Tubes and Gas romatography. Environmental Science &
TechnolOgy, 9(13) :1175—1178, 1975.
100. Frankel, L. S., and R. F. Black. Automatic Gas ChromatO
graphic Monitor for the Determination of Parts_PerBilli0
Levels of Bis(ChlOrOmethYl) Ether. Analytical Chemistry,
48(4):732—737 , 1976.
101. Feairheller, W. R., M. D. Schumacher, and W. D. Ross.
Research on the Thermolytic Dissociation of Molecules.
APML—TR-74 83 , Air Force Materials Laboratory, Wright
Patterson AFB, Ohio, May19 74 . 67 pp.
149
-------
102. }Zunen, S. M., K. J. Voorhees, C. A. Hill, F. D. Hilernan,
and D. N. Osborne. Chemical Analysis of the Insoluble
Carbonaceous Components of Atmospheric Particulates with
pvrolvsis/GaS Chromatograph/Mass Spectrometry Techniques.
In: p oceedingsof the 70th Annual Meeting, Air Pollution
Control i ssociat1On, Toronto, Ontario, 20—24 June 1977.
Published by Air Pollution Control Association, Pittsburgh,
pennsylvania, 1977. Papers 77-36.
103. Jewett, K. L., and C. E. Parris. Environmental Chemistry.
Contract 3130117. National Bureau of Standards, Washing-
ton, D.C. 1977.
104. Williams, F. W., M. E. Urnstead, and J. E. Johnson. Deter-
mination of Trace Quantities of Halogenated Hydrocarbons
in Gas Samples. NRL6964, Naval Research Laboratory,
Washington, D.C., 6 November 1969. 18 Pp.
105. Su, Shih-Wu. Low Molecular Weight Halocarbons. NSF/IDOl-
77-36, National Science Foundation, Washington, D.C., 1976.
15 pp.
106. Environmental Monitoring Near Industrial Sites: Methyl-
chloroform. EPA-560/677-025, U.S. Environmental Protec-
tion Agency Washington, D.C., August 1977. 82 Pp.
107. Environmental Monitoring Near Industrial Sites: Trichicro-
ethylene. EPA-SGO/ 677 O 24 , U.S. Environmental Protection
Agency, Washington, D.C., August 1977. 75 pp.
108. Vick, R. D., and N. J. Avery. Extraction and Identification
of Organic Materials Present in Soot from a Natural Gas Flue.
Contract W-7405—ENG82, Dept. of Energy, January 1978.
19 pp.
109. Crittenden, A. L. Analysis of Atmospheric Organic Aerosols
by Mass SpectrOSCOPY. EPA—600/3—76—093, u.s. Environmental
protectiori Agency, Research Triangle Park, North Carolina,
August 1976. 296 Pp.
110. Sirnoneit, B. R., D. H. Smith, arid G. Eglinton. Application
of Real-Time Ma Ss SpectroinetriC Techniques to Environmental
Organic Geochemistry: I. Computerized High Resolution
Mass Spectrornetry and Gas Chromatography - Low Resolution
Mass Spectrometry. Archives of Environmental Contamination
and toxicology, 3(4):385, 1975—1976.
111. MartineZ, R. I., R. E. Huie, and 7. T. Berron. Mass Spec-
trometriC Detection of Dioxirane, H 3 CO, and Its Decomposi-
tion products, H 2 and CO, from the Reaction Ozone with
Ethylene. Chemical Physics Letters, 57(3) :457—459,
November 1977.
150
-------
112. Mendenhall, G. D., P. W. Jones, P. E. Strup, and W. L.
Margard. Organic Characterization of Aerosols and Vapor
Phase Compounds in Urban Atmospheres. EPA-600/3—7 8 -O 3 l,
u.s. Environmental Protection Agency, Research Triangle
Park, North Carolina, March 1978. 81 pp.
113. Niki, H., P. Maker, C. Savage, and L. BreitenbaCh. Par-
Participating in Photochemical Smog Formation. Ford Motor
Company, Dearborn, Michigan. International Conference on
Environmental Sensing and Assessment, Las Vegas, Nevada,
14—19 September 1975. Published by IEEE, New York, Vol. 2,
Paper 24—4, 1976. 4 pp.
114. Della Fiorentina, H., J. Degraeve, and F. Dewiest. Deter-
mination of Non—Volatile Organic Matter Associated with
Airborne Particulates by Infrared SpectrometrY — 1. Oper-
ating Conditions. Atmospheric Environment, 9(5):5l3, May
1975.
115. Kuhiman, C. M. Monitoring Toxic Air Contaminants. Plant
Engineering, 31(15):149150, 21 July 1977.
116. Lukens, H. R., C. B. Williams, S. A. LevinsOn, and W. B.
Dardliker. Sensitive, Specific Fluorescence IrnrnunoaSSay
Methods for Detecting Pesticides and Other Organic Envir-
onmental Contaminants Arising from Biological or Chemical
Sources. NSF/RA—770472, National Science Foundation,
Washington, D.C., 21 November 1977. 95 Pp.
117. Smith, E. M., and P. L. Levins. Sensitized Fluorescence for
the Detection of Polycyclic Aromatic Hydrocarbons. EPA—600/
7—78—182, u.s. Environmental Protection Agency, Research
Triangle Park, North Carolina, September 1978. 31 pp.
118. Foo, M. A., and S. W. Staley. Determination of Polycyclic
Aromatic Hydrocarbons in Atmospheric particulate Matter by
High Pressure Liquid Chromatography Coupled with Fluores-
cence Techniques. Analytical Chemistry, 48(7) :992-998,
June 1976.
119. El, w. Laser-Excited Matrix—Insulation FluorometriC
Analysis of Po].ycyclic Aromatic Hydrocarbons. Grant CHE77
12542, National Science Foundation, Washington, D.C., 1978.
120. Hodgeson, J. A., and W. A. McClenny. Analytical Technique
for Gaseous Pollutants Based on PhotolysiS and Detection
of Primary Fragments. presented at American Chemical
Society, Division of Water, Air and Waste ChemistrY, 28
August — 1 September 1972, Preprint 12—2.
151
-------
121. Fontijn, A. Chemiluminescent Method and Apparatus for
Determining the Photochemical Reactivity of Organic Pol-
lutants in a Gaseous Mixture. Patent Application 548,471
(U.S. Environmental Protection Agency) , filed 10 February
1975. 21 pp.
122. Riggle, C. J., D. L. Sgontz, and A. P. Graffec. The Analy-
sis of Organotins in the Environment. In: 4th Joint Con-
ference on Sensing of Environmental Pollutants, New Orleans,
Louisiana, 6—11 November 1977. p. 761.
123. I-iermann, T. S. Development of Sampling Procedures for
Polycyclic Organic Matter and Polychlorinated Biphenyls.
EPA-650/2-75—007, U.S. Environmental Protection Agency,
Washington, D.C., August 1974.
124. Nwankwo, J. N. , and A. Turk. Electrochemical Analysis of
Sulfjdjc and Amine Odorants. EPA—600/2—76—02J., U.S. Envir-
onmental Protection Agency, Research Triangle Park, North
Carolina, June 1976. 46 pp.
125. Shaw, M. Prototype Hydrazines Electrochemica.l Analyzer.
Contract F336l5-78—C—0607, Air Force School of Aerospace
Medicine, Brooks Air Force Base, San Antonio, Texas, 1978.
126. Fine, D. H., D. P. Rounbehier, E. Sawicki, K. Krost, A.
Rounbehier, A. Silvergleid, and G. A. Demarrais. Determin-
ation of Dimethylnitrosoarnine in Air, Water, and Soil by
Thermal Energy Analysis: Validation of Analytical Pro-
cedures. Environmental Science & Technology, ll(6):577,
June 1977.
127. Scheide, E. P., and G. C. Guilbault. Piezoelectric
Detectors for Organophosphorus Compounds and Pesticides.
Analytical Chemistry, 44(1l):1764—1768, September 1972.
128. Driscoll, J. N., and F. P. Spaziani. Trace Gas Analysis
by Photoionization. In: 21st Annual ISA Analytical
Instrumentation Symposium, Philadelphia, Pennsylvania,
6—8 May 1975. p 111—114.
129. Trace Analysis arid Detection in the Environment. Presented
at 6th Annual Symposium on Environmental Research, Edgewood
Arsenal, Aberdeen Proving Ground, Maryland. 29 April -
1 May 1975. Published by Dept. of Army, Edgewood Arsenal,
Aberdeen Proving Ground, Maryland, EO-SP-76001, 1976.
130. Leithe, W. The Analysis of Air Pollutants. Ann Arbor
Science Publishers, Ann Arbor, Michigan, 1971.
152
-------
131. Operating Manual: Model 221—lA, AC/DC Sampler for Remote
sample Collection Using Wet Chemical Collectors and Gas
Chromatography Cartridges. The Nutech Corporation, Research
Triangle Park, N. C.
132. Quality Assurance Handbook for Air Pollution Measurement
Systems, Vol. II, Ambient Air Specific Methods. EPA—600/4—
77—027a, U.S. Environmental Protection Agency, Research
Triangle Park, N. C., May 1977, 339 pp.
133. Haynes, W. M., D. M. Haile, D. H. Toy, and W. N. Mees.
Guide to Good Laboratory Practices. Monsanto Research
Corporation, Environmental Analytical Sciences Center,
Dayton, Ohio, March 1979.
153
-------
APPENDIX A
METHODS FOR THE SAMPLING AND ANALYSIS OF ORGANIC TERIALS
SAMPLING
Various sampling methods have been employed to collect organic
pollutants for subsequent laboratory analysis. Other methods are
available to analyze these pollutants in the field by passing the
polluted air or stack gas through analytical equipment as it is
collected. This section discusses those methods which can be
employed for field use.
Organic pollutants that are present in stack gases, ambient air,
and workplace environments exist as particulates, gases, or
vapors and therefore require collection techniques specific to
those media. Particulates are commonly collected on filters and
in cyclones; gases and vapors are either collected in cryogenic
traps, polymeric bags, or evacuated cylinders, or trapped in or
on various media. Nedia employed for trapping gases and vapors
have included porous polymers, charcoal, silica gel, alumina
glass filaments or beads, and organic solvents. The EPA source
assessment sampling system (SASS train) has been recently de-
veloped to collect stack gas samples which contain particulates
and condensed organic vapors adsorbed onto a porous polymer
resin. Recommended sample volumes greatly exceed those of con-
ventional stack sampling systems, thus minimizing the problem
of sample quantities below the detection limit of analytical
equipment. Specific methods of sample collection are discussed
in more detail below.
Particulate Collection
Filtering has probably been the most thoroughly investigated of
all methods for the collection of atmospheric arid stack gas
samples. These methods trap particulates and condensed vapors
on a filter media. Glass-fiber filters have been most frequently
used however, other filters such as quartz, cellulose, Teflon,
154
-------
and cellulose esters have also been used [ 501. Material collected
on filters can subsequently be extracted and analyzed for organic
components. For ambient air sampling with filters, high—volume
sampling systems are needed to collect a sample mass large
enough to be above the detection limits of the analytical tech-
niques. High-volume particulate samples have been used in a
variety of situations where it was desired to measure organic
material in ambient air.
Airborne particulate material has been collected on glass-fiber
filters with high—volume air samples mounted aboard ships during
cruises of the Mediterranean Sea and Pacific Ocean. These fil-
ters were extracted with chloroform, and the lipid components
of the extracts were converted to methyl esters for analysis by
gas chromatography (GC) [ 511. In a survey conducted in Belgium,
particu].ates were collected at 14 sites employing high—volume
samplers. The samples were analyzed for total organics and
benzo(a)pyrene (BaP). The results showed that high-volume sam-
pling volatilizes a large portion of the organics adsorbed on
the particles (52]. Organic materials are also recovered from
particulates collected from stack gases. Stack sampling devices
collect much lower volumes of gas than do ambient air samplers
because of the higher concentration of particles usually occur-
ring in stack gases.
EPA Method 5, illustrated in Figure A-i, is the most widely used
stack sampling method for particulate collection (53]. Stack
gas is drawn into the system by a vacuum pump through a probe
[ 50) Sawicki, E. Analysis of Atmospheric Carcinogens and Their
Cofactors. In: Environmental Pollution and Carcinogenic
Risks, INSERN Symposia Series, Vol. 52, IARC Scientific
Publications No. 13, World Health Organization, Internation-
al Agency for Research on Cancer, 1976. pp 297-354.
(Si) Barger, W. R., and W. D. Garrett. Surface—Active Organics
Material in Air Over the Ligurian Sea and Over the Eastern
Equatorial Pacific Ocean. NRL—7897, Naval Research Labora-
tory, Washington, D.C., 30 June 1975. 20 pp.
(52] Rondia, D., F. De Wiest, and H. Della Fiorentina. Organics
in Atmospheric Aerosols in Belgium. State University of
Liege, Belgium. International Conference on Environmental
Sensing and Assessment, Las Vegas, Nevada, 14-19 September
1975. Published by IEEE, New York, Vol. 2, Paper 24-5,
1976. 4 pp.
(53] Method 5 — Derinination of Particulate Emissions from
Stationary Sources. Federal. Register, 42(160): 41776—41782,
August 1977.
155
-------
and heated filter and into a set of water impingers. Sampling
is conducted isokineticallY while traversing the stack to obtain
a representative particle size distribution. Another procedure
for stack gas particulate collection is EPA Method 17, illus-
trated in Figure A-2, which is similar to Method 5, except that
an in—stack filter is employed [ 54].
Schematic of EPA Method 5 sampling
train for particulates [ 55).
In—stack filters have been employed to collect particulate
samples from coking, smelting, and similar industrial operations.
The extracts of these particulateS were subsequently analyzed for
14 highly carcinogenic polynuclear aromatic (PNA) compounds [ 55].
A variety of samplers is available to collect particles of
various size ranges. The best known samplers are the Anderson
size—fractionating cascade impactors. Particles collected on
stages 2 through 7 of an Anderson sampler, representing aero-
dynamic sizes from 0.43 to 7.0 pm, are purported to be deposited
in the pulmonary, tracheobronchial and nasopharyrigeal compart-
ments of the respiratory system [ 50).
[ 54] Method 17 — Determination of Particulate Emissions from
Stationary Sources (In—Stack Filteration Method). Federal
Register, 43(37):7584—7596, February 1978.
[ 55] Sharkey, A. G., J. L. Schultz, C. White, and R. Lett.
Analysis of Polycyclic Organic Material in Coal, Coal Ash,
Fly Ash, and Other Fuel and Emission Samples. EPA—600/2-
76-075, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, March 1976. 31 pp.
n o
i i D Il&lU MUNALI
.DFI
•MV
Figure A-i.
156
-------
b øI* III lt*, s.aII . $
S., L ,.lII kl• / I
/ typt $
• lI s.dl sJ 5 1 151 1US$
— AI lutist
PUMP
,Acu
I 101
MANIuMI 11*
$UtttIV$S lIUtt$I INtNCt tilt) IPACS$G$
Figure A—2. Particulate-sampling train, equipped with in-stack filter [ 54].
rffiGli TRAIN RPISONAINAY P( RIPISCID
it amtsusvaItu( c000Usf S
rim
S INS I NS
S n
-J
1 0 1 1* 1 1
ISLIS S
•INI Pit
iewNu lIrt
PIlOT lUll
t$ I RM IM IIIN
£ 01 1 1
V*101
MAIN VAt Vt
SRY LAS Ml i ii
-------
A coniparison of the performance of a membrane—filter personal
air sampler with that of a high-volume sampler showed good cor-
relation when suitable care was taken in weighing the membrane
filters and in compensating for absorbed water (50).
A similar filtering method has also been employed for collection
of organic materials. Nercuric salts and sulfuric acid impreg-
nated filters have been employed in an effort to determine low—
molecular—weight rnercaptans, dirnethyl sulfide, and methylamine
in urban and industrial areas. The gases were recovered from
the filter, concentrated, and analyzed by GC. This procedure
allowed short—term, high-volume sampling to determine the concen—
ration of gases at less than 1 ppb [ 56]. Glass—fiber filters
have also been treated with glycerin tricaprylate and reported
to trap tetracyclic arenes with 100% collection efficiency. The
method requires the additional step of hydrolyzing the ester
before extraction with cyclohexane [ 50].
A multiparameter fluorescence method for in-situ monitoring of
stack gas particulates has been reported. The fluorescent emis-
sions from the particulates could be used to characterize the
chemical composition of particulates as well as to classify their
size. Early measurements of aromatic compounds showed that each
sample had a characteristic emission spectra and lifetime.
Measurements and discrimination were accomplished using photon
counting and fluorescence spectra correlation [ 57).
An instrument for real—time analysis of airborne particulates
has been described in the literature. T e average concentration
of particulates in the air was determined by impinging the par-
ticles on a heated rhenium ribbon and analyzing the ion current
with a small, magnetic sector, mass spectrometer. Heavy metals
and organic compounds in the particles could be analyzed by this
technique [ 58].
[ 56] Okita, T. Filter Method for Determination of Trace Quanti-
ties of Amines, Nercaptans, and Organic Sulfides in the
Atmosphere. Atmospheric Environment, 4(l):93-102, 1970.
[ 57) Leskovar, B., C. Lo, and B. Turko. MuJ.tiparameter Fluores-
cence Method for Monitoring In-Situ Particulates. U.S.
Energy Research and Development Administration, Washington,
D.C., 1977.
[ 58) Davis, W. D. Continuous Mass Spectrometric Analysis of
Particulates by Use of Surface Ionization and Continuous
Mass Spectrometric Determination of Concentration of Parti-
culate Impurities in Air by Use of Surface Ionization.
Environmental Science & Technology, ll(6):587, 1977.
158
-------
Extraction of Organic Material from Particulate Samples
Because reliable techniques for the direct analysis of particles
for organic compounds are not readily available, the organics
associated with collected particulates are often extracted. In
many cases organic solvents are employed for extraction, and the
resulting solutions are normally concentrated and separated into
fractions prior to analysis.
Extraction of polyaromatic hydrocarbons (PAH) from particulates
is usually accomplished by a 5— to 12-hour soxhlet extraction
with benzene or cyclohexane at the boiling point. The disadvan-
tages of this method are 1) benzene is a human leukernogen,
2) during the extraction and evaporation steps, some of the polar
material decomposes and some polymerizes, 3) the extraction time
is too long, 4) benzene is a more efficient extractant than
cyclohexane but contains much more extraneous polar material in
the extract than does cyclohexarie, and 5) the reproducibility
of the method is not as good as that obtained with some of the
more recently developed methods. An independent investigation
of the benzene-soluble extraction method pointed out many of its
faults and emphasized its poor reproducibility (50].
Airborne particulate matter collected on glass-fiber filters has
been extracted with benzerie and separated into neutral, acidic
and basic fractions. In this study, the acidic fraction was con-
verted to the methylated derivatives before analysis. More than
100 compounds were identified in the acidic fraction consisting
mainly of fatty acids and aromatic carboxylic acids. In the
neutral fraction, saturated aliphatic hydrocarbons, PAH, and
polar oxygenated substances were identified. The basic fraction
consisted of nitrogen—containing analogs of the important PAH
present in the neutral fraction (59].
In another study particu].ates collected on glass—fiber filters
were soxhiet extracted with cyclohexane, and the organic matter
was concentrated and fractionated into chemical classes by ad-
sorption chromatography. The PAM fraction was collected and
further fractionated by bond—reverse phase-partition chromatog-
raphy (60]. Development work in the separation of organic
(59] Cantreels, W., and K. Van Cauwenberghe. Determination of
Organic Compounds in Airborne Particulate Matter by Gas
Chromatography/Mass Spectrometry. Atmospheric Environment,
10(6):447—457, 1976.
(601 Eisenberg, W. C. Fractionation of Organic Material Ex-
tracted from Suspended Air Particulate Matter Using High
Pressure Liquid Chromatograph. 3. Chromatographic Science,
16(4):i.45—151, 1978.
159
-------
compounds by instant thin layer chromatography (TLC) has also
been reported [ 61). Soxhiet extraction of filtered airborne
particulates with methanol has been employed prior to the analy-
sis 0± complex organic mixtures [ 62). More than 20 aza—arenes
and other nitroaen—bases have been identified by employing an
isolating technique applied to ambient particulates. An extrac-
tion, solvent partitioning, and a sequence of chromatographic
separations were employed to identify aza—arenes, quinolines,
isoquinolines, and their alkyl derivatives (63).
Methods other than solvent extraction have also been employed to
extract organic material from collected particulate matter. A
mechanical disrupting technique capable of extracting the organ-
ic content from filtered ambient particulate has been tested
and found to compare favorably with the soxhiet extraction method
[ 64). Hydrofluoric acid has been employed to enrich the PAH
content from filtered airborne particulate. The hydrofluoric
acid dissolves the glass—fiber filter and collected particles,
leaving the organic residue free for solvent extraction. This
method is useful for rapid preparation of samples for GC/MS
analysis (65).
Soxhlet and ultrasonic extraction procedures were compared for
seven different solvents having boiling points ranging from 34°C
to 81.4°C. For benzo(a)pyrene and aromatic compounds, the ultra-
sonic method using methanol as solvent proved to be best. Both
[ 61) Stanley, T.W., N. J. Morgan, and C. E. Meeker. Rapid
Estimation of 7H-Benz [ de)anthracen-7—one and Phenalen—l-
one in Organic Extracts of Airborne Particulates from
3—} our Sequential Air Samples. Environmental Science &
Technology, 3(11) :1198—1200, 1969.
[ 62) Kurasek, F. W,, D. W. Denney, K. W. Chan, and R. E. Clement.
Analysis of Complex Organic Mixtures and Airborne Particu-
late Matter. Analytical Chemistry, 50(l):82, 1978.
[ 63) Dong, N. W., C. Locke David, D. Hoffman, and D. Naylor.
Characterization of Aza—Arenes in Basic Organic Portion of
Suspended Particulate Matter. Environmental Science &
Technology, ll(6):612, 1977.
[ 64) 3ove, C. L., and V. P. Rukreja. A New Mechanical Disruption
Technique for Organic Enrichment of Hi-Vol Samples. Envir-
onmental Letters, 1O(2):89, 1975.
[ 65] Bove, C. L., and V. P. Kukreja. An Enrichment Method for
Polycyclic Aromatic Hydrocarbons Collected on Glass Fiber
Filters Using Hydrofluoric Gas. C. Environmental Science &
Health — Environmental Science & Engineering, 11(8—9):517,
1976.
160
-------
rnetho•ds were applicable for ether, dichioromethane, chloroform,
and benzene. The ultrasonic method has the advantages of being
rapid, operating at relatively low temperatures (0°C-lO°C), and
causing no changes in the spectra [ 50).
Decomposition problems can occur with PAH and polar compounds [ 50]
when solvent desorbed. However, sublimation of PAH from airborne
particulate at about 250°C—300°C does not require the use of
solvents, can prevent decomposition of compounds when performed
under vacuum, can be automated, and can discriminate for low
boiling hydrocarbons.
Temperature programming essentially consists of extraction and
separation of gaseous pollutants from polymer resins by heating
the resin gradually from 0°C to 150°C. This desorbs the mate—
rial collected, passing it directly to a mass spectrometer for
analysis. The disadvantage is artifact formation, the problem
of “synthesizing” compounds which were not present originally
in the air. Temperature programming cannot separate Ba? from
benzo(e)pyrene or perylene. TLC cannot separate BaP completely
from other PAH. However, with two—dimensional alumina-cellulose
acetate TLC, BaP is easily separated from other PAH and the
other fluorescent pollutants in organic extracts of airborne
particulates [ 50).
Polytron extraction utilizes a Brinkmann homogenizer which ccm—
bines the tearing and ripping strength of mechanical energy with
the cavitating energy of sonic and ultrasonic sound. The ex-
traction of a large filter (500 cm 2 ) can be accomplished below
room temperature within 5—10 mm. Cyclohexane or water work well
as the extractant. The disadvantages of this technique are:
1) a pure solvent must be used, and 2) ultrasonic treatment
should be kept to a minimum so as to minimize decomposition [ 50].
A mechanical disruption technique has been employed to extract
the organic material from a fiber—glass filter sample. The
method compared favorably with soxhiet extraction [ 64).
Solid Sorbents for Collecting and Concentrating
Organic Compounds
Solid sorbent methods for collecting and concentrating organic
substances from point source or ambient air employ the adsorpt .ofl
or partitioning properties of the sorbent materials. These mate-
rials retain organic substances selectively while passing the
major diluent gases like nitrogen and oxygen of air.
The retentive characteristics, varied polarity, and high thermal
stability of charcoal, molecular sieves, and porous polymers
suggest that these materials might be the best media for effi-
ciently collecting and enriching organic substances although the
varied nature of sources requires their evaluation for specific
applications.
16 ].
-------
in selecting a sorbent for high collection efficiency, the re-
coverv of the compound must also be considered. Often a inutal
compromise Tray be necessary to obtain an acceptable system.
Table A-l lists some of the sorbents most used and the types of
compounds which could be collected [ 66).
TABLE A-i. GENERAL SORPTION-DESORPTION
SYSTEMS FOR ORGANIC COMPOUNDS
Sc che
Activatec carbon
DesorptiOfl solvent
Carbon disulfide, inethylene
chloride, ethyl ether
(1% methanol or 5% iso-
propanol solnet Lines added).
Type of compound
Miscellaneous volatile
organics: methyl
chloride, vinyl chloride
and other chlorinated
aliphatics, aliphatic
and aromatic solvents,
acetates, ketories, al-
cohols, etc.
Silica gel
Methanol, ethanol, diethyl
ether, water.
Polar compounds: alcohols,
phenols, chlorophenc.s,
chlorober.Zefles, ali-
phatic and aromatic
a.mines, etc.
Activated alumina
Water, diethyl ether,
methanol.
Polar compounds: alcohols,
glycols, ketones, alde-
hydes, etc.
Porous rolvmers
Ether, hexane, carbon
disulfide, alcohols.
Wide range of compounds:
Phenols, acidic and
basic orgariics, multi-
functional organics, etc.
Chemically bonded
and other CC
packings
Ether, hexane, methanol.
Specialized: high boiling
compounds, pesticides,
herbicides, polynuclear
aroxnatics, etc.
[ 66) Dietrick, M. W., L. N. Chapman, and 3. 3. Mieure. Sampling
for Organic Chemicals in workplace Atmospheres with Porous
Polymer Beads. American Industrial Hygiene Association
3ournal, 39(5):385-392, 1978.
162
-------
Polymer Sorbents- —
Five groups of porous polymers have been found to be most useful
as solid sorbents for collecting the concentrating organic com-
pounds from polluted atmospheres and stack gases; they are: XAD
resins, Tenax-GC, the Porapak series, the Chromosorb Century
series, and polyimides. These sorbents are finding increased
application in stack and ambient air sampling. Several studies
have evaluated the effectiveness of sampling with polymer
sorbents [ 67-75]. Polymer sorbent samplers have been shown to
[ 67] Adams, J., K. Menzies, and P. Levins. Selection and Evalu-
ation of Sorbent Resins for the Collection of Organic Com-
pounds. EPA-600/7—77—04 4 , u.s. Environmental Protection
Agency, Research Triangle Park, North Carolina, April 1977,
67 pp.
[ 68) Gallant, R. F., J. W. King, P. L. Levins, J. F. Piecewicz.
Characterization of Sorbent Resins for Use in Environmental
Sampling. EPA—600/7—78-054, u.s. Environmental Protection
Agency, Research Triangle Park, North Carolina, March 1978.
163 pp.
(69] Snyder, A. D., F. N. Hodgson, M. A. Keminer, and J. R.
McKendree. Utility of Solid Sorbents for Sampling Organic
Emissions from stationary Sources. EPA—600/2— 7 O— 2 O 1 , U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, July 1976, 77 pp.
(70] Rhoades, 3. W., and D. E. Johnson. Evaluation of Collection
Media for Low Levels of Airborne Pesticides. E?A-600/l-77
050, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, October 1977. 140 pp.
(71] Strup, P. E., P. W. Jones, R. D. Giarnxnar, and T. B. Stanford.
Efficient Collection and Analysis of Hazardous Organic Com-
pounds from Combustion Effluents. Battelle-C01UIthUS Labora-
tory, Ohio. International Conference on Environmental Sens-
ing and Assessment, Las Vegas, Nevada, 14—18 September 1975.
Published by IEEE, New York, Vol. 2, Paper 22-3, 1976.
4 pp.
[ 72] Winterlin, W., D. G. Crosby, and W. W. Kilgore. Analytical
Research on Economic and Environmental Toxicants. Grant
67347, U.S. Dept. of Agriculture, Washington, D.C., 1978.
[ 73] Bunri, W. W., E. R. Deane, D. W. Klein, and A. D. Kleopfer.
Sampling and Characterization of Air for Organic Compounds.
Water, Air and Soil Pollution, 4(3—4):367—380, 1975.
[ 74] Barrett, W. J. Identification of Organic Compounds in the
Atmosphere. Grant PHS—AP—00454, U.S. Public Health Service,
1968.
[ 75] Smith, M.S., A. 3. Francis., and J. M. Duxbury. Collection
and Analysis of Organic Gases from Natural Ecosystems:
Application to Poultry Manure. Environmental Science &
Technology 11(l):51—55, . 1977.
163
-------
be more efficient than cornnton impinger methods with nearly
quantitative recovery for many organic species including poly—
cyclic organic compounds (POM) [ 71] . Some of the specific
applications of polymer sorbents are presented in more detail
below.
A study to develop methods for pesticide collection and analysis
employed tubes packed with cross—linked polystyrene resin [ 72].
In a program to sample and characterize ambient air for organic
compounds, several polymer resins were employed. Sorption on
XAD-2 resin followed by solvent elution, and sorption on
Tenax-GC followed by direct desorptiori onto a GC column identi-
fied 62 compounds including large quantities of phenol and phenol-
formaldehyde resin intermediates [ 73). In another study a
0.15—rn long, 3.2—mn diameter copper tube was filled with an
ordinary GC column packing for concentrating traces of oxygenated
organic compounds in air. Samples of approximately one liter
were drawn through the tube using a rubber aspirator bulb [ 74].
Porapak QS-Carbosieve B traps were used to identify volatile
compounds generated from chicken manure. Various alcohols,
ketones, esters, and carboxylic acids together with dimethyl
sulfide were detected [ 75).
Tenax-GC cartridges have been used to collect organic vapors
from the ambient air of three major cities. Vapors were thermal-
l.y desorbed and analyzed by a capillary GC/MS technique.
Twenty—one halogenated hydrocarbons were detected, including
vinyl chloride and trichlorOethylefle [ 76). Epoxides such as
ethylene oxide, propylene oxide, and styrene oxide have been
tentatively identified in air samples collected on Tenax—GC as
well as dimethylnitrosamifle [ 50) . Tenax—GC has been suggested
as the best available sorbent for general use. It has been
thoroughly studied in the laboratory and in the field and has
been applied to the sampling of carcinogenic vapors in the atmos-
phere. It has a high operating temperature in that even up to
350CC little background bleeding is observed. The effects of
transportation and storage of bis(chloron’ethYl) ether on
Tenax-GC have been investigated. Immediate analysis gave 100%
recovery; following transportation of a cartridge from Research
Triangle Park, N.C., to San Francisco, CA and back (total trans-
portation time of 6 days), the recovery dropped to 58 ± 5% after
3. week and 41 ± 4% after 2 weeks. Without transportation the
recovery was 65 ± 5% in 1 week [ 50].
[ 76) Pellizzari, B. D., 3. E. Bunch, R. E. Berkley, and 3. NcRae.
Determination of Trace Hazardous Organic Vapor Pollutants
in Ambient Atmospheres by Gas C ro atographY/MaSs Spectro-
rnetry/Coinputer. Analytical ChemistrY, 48(6) :803, 1976.
164
-------
A method to sample for environmental contamination of polychlo-
rinated naphthalenes (PCN) was developed and consisted of a
glass—fiber filter and two precleaned polyurethane foam plugs
in tandem. Recovery of PCN from the foam and filter was ac-
complished by triple extraction with toluene (77]. Polyurethane
foam, known for its capacity to remove PCB’S from sea water, is
finding increased potential for removing PCB’s from air samples.
Several studies report polyurethane foam being used to sample
ambient air in high—volume samplers for collection of PCB’s and
chlorinated pesticides. Removal of the collected organics from
the foam is normally accomplished by organic solvent extraction
[ 78—80].
PCB emissions have been measured from stacks of municipal waste,
industrial waste, and sewage sludge incinerators and from capaci-
tor and transformer filling plants. Collection was accomplished
by water impingement and adsorption on Florisil or direct adsorp-
tion on Florisil. Samples were extracted with hexane and concen-
trated through evaporation of the solvent (811. Chromosorb 101
has been used to efficiently trap vapors of hexachlorobenzene
[ 77] Erickson, M.D., R. A. Zweidinger, L. C. Michael, and E. 0.
Pellizari. Environmental Monitoring Near Industrial Sites:
Polychloronaphthalenes. EPA-560/6-77—019, U.S. Environ-
mental Protection Agency, Washington, D.C., June 1977.
267 pp.
(78) Lewis, R. G., A. R. Brown, and M. D. Jackson. Evaluation
of Polyurethane Foam for Sampling of Pesticides, Polychlorin—
ated Biphenyls and Polychiorinated Naphthalenes in Ambient
Air. A.nalytical Chemistry, 49(12):1668—1672, 1977.
(79] Margeson, J. H. Methodology for Measurement of Polychiorin-
ated Biphenyls in Ambient Air and Stationary Sources - A
Review. EPA—600/4—77-021, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, April 1977.
39. pp.
(80] Bidleman, T. F., and C. E. Olney. High—Volume Collection
of Atmospheric Polychiorinated Biphenyls. Bulletin of
Environmental Contamination and Toxicology, 11(5) :442-449,
1974.
(81] Haile, C. L., and E. Baladi. Methods for Determining the
Polychiorinated Biphenyl Emissions from Incineration and
Capacitor and Transformer Filling Plants. EPA-600/4-77-048,
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, November 1977. 94 pp.
165
-------
(liCE) and hexachioro—l,3-butadiene (HCBD) from air at a sampling
rate of 3 L,/min decreased the efficiency by 20% [ 82].
Other Sorbent Collection Materials— —
Activated charcoal has found widespread usage for sampling
orcariics especially in the collection of solvent vapors. A
charcoal—tube sampling procedure was tested on vapors of benzene,
carbon terrachloride, chloroform, dioxane, ethylene dichioride,
trichloroethylene, meta—xylene, and two solvent mixtures for
NIOSH methods development [ 83]. A later study reports the
development by NIOSH of a personnel atmospheric sampler consist-
ing of a glass tube containing layers of activated charcoal,
urethane foam, and glass wool for organic pollutant sampling.
Organics trapped on the charcoal are removed with carbon di—
sulfide [ 84). Activated carbon was also used to sample emissions
from grain fermentation units in a whiskey distillery. Ethyl
acetate, ethyl alcohol, n—propyl alcohol, isobuty]. alcohol, iso-
arnyl alcohol, and isoamyl acetate were adsorbed on the carbon
and identified by GC [ 85]. Vinyl chloride monomer (VCM) was
quantified using charcoal cartridges to sample automobile
interior atmospheres. Qualitative sampling with Tenax—GC
identified 147 organic compounds in the automobiles not present
in the surrounding air [ 86]. Dimethyl nitrosamine was also
reported to be collected by activated charcoal [ 50].
[ 82] Mann, J. B., H. F. Enos, 3. Gonzalez, and 3. F. Thompson.
Development of Sampling and Analytical Procedure for Deter-
mining Hexachlorobenzene and Hexachloro-l,3-butadiene in
Air. Environmental Science & Technology, 8(6):584, 1974.
[ 83] Reckner, L. R. and 3. Sachdev. Collaborative Testing of
Activated Charcoal Sampling Tubes for Seven Organic Sol-
vents. NIOSH-75/184, National Institute for Occupational
Safety and Health, Cincinnati, Ohio, June 1975, 229 pp.
[ 84] Flisak, F., and 3. Kubarewicz. Sampling Methods and Instru-
mental Analyses of Airborne Toxic Chemicals in Workroom
Environments. In: Proceedings of a Special Conference on
Industrial Pollution Control Measurement and Instrumenta-
tion, Technamic Publishing Co., Westport, Connecticut,
1976, pp 162—165.
[ 85) Carter, R. V. and B. Linsky. Gaseous Emissions from
Whiskey Fermentation Units. Atmospheric Environment,
8(1):57—62, 1974.
[ 86) Zweidinger, R. A. Organic Emissions from Automobile
Interiors. EPA-600/7-77—149, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, December
1977. 90 pp.
166
-------
Other sorbent materials reported include silica gel, alumina,
and glass plates. Sulfuric acid—coated silica gel was found t
be effective in the collection of low—boiling aliphatic amines,
low-boiling hydrazines, pyridines, and formaldehyde. Solvent
elution was used to remove the sample from the silica gel.
Alumina was also studied for sampling formaldehyde vapors [ 87).
The glass plate sampling technique has been shown to be a simple
method for sampling organic films that accumulate on items ex-
posed to a tropical environment. The glass plate can be placed
adjacent to a test item stored in the tropics and the effects of
environmental chemicals studied without destroying the test
item [ 88).
Desorption of Solid Sorbents- —
After sample collection, tubes containing the solid sorbent and
adsorbed organic chemicals are returned to the laboratory and
the organic chemical is recovered, most frequently by solvent
desorption. Many solvent desorption techniques are identical
to those described for extraction of organic material from
particulate samples.
Porous-polymer-bead sorbents offer the option of thermal desorp-
tion for sample recovery. The trapped materials are thermally
desorbed from the collection column and directly transferred to
the GC analytical column within the gas chromatograph oven. This
technique is used frequently in air pollution studies and is
equally applicable for industrial hygiene applications. The
thermal desorption process has several advantages over solvent
desorption procedures. High sensitivity is possible with thermal
desorption because the entire collected sample can be analyzed
at one time. Also, no solvents are present which not only
reduce sensitivity but obscure chromatographic regions and may
present potential health hazards.
(87] Distenfeld, C. H., and J. R. Kleinish. High Efficiency
Mixed Species Radioiodine Air Sampling Readout, and Dose
Assessment System. Presented at IAEA/OECD Handling of
Radiation Accidents, Symposium, Vienna, 28 February —
4 March 1977, p 463.
(88] Sprouse, 3. F. New Sampling Technique for Identifying
Organic Film Accukulation on Army Material in the Tropics.
Presented at 6th Annual Symposium on Environmental Research,
Edgewood Arsenal, Aberdeen Proving Ground, Maryland,
29 April — 1 May 1975. Published by Dept. of Army,
Edgewood Arsenal, Aberdeen Proving Ground, Maryland,
EO—SP—76001, 1976.
167
-------
Coramercial instrumentation for thermal desorption is now avail-
able from several sources. These units are designed primarily
for the analysis of volatile materials such as process solvents.
They are less applicable for high boiling chemicals and those
sensitive to metal surfaces. A simple method has been developed
to modify a laboratory gas chromatograPh for thermal desorption
studies, including those of high boiling and sensitive materials.
A three-way valve is utilized which allows ready return of the
gas chromatograph to the conventional operating mode. A rnodi-
fication is also described which will permit splitting of the
collected sample so that a portion may be preserved for a second
analysis [ 89].
Solid Sorbent Sampling Systems— —
very simple sampling system using the porous polymer approach
is shown in Figure A—3. This system, suitable for higher molecular
weight hydrocarbons, consists of a 6.3 mm or 9.5 mm stainless-
steel tube, 0.15 m long, packed with porous polymer bead mate-
rial. This sampling tube is attached to a pump and a gas
meter [ 89]. The pump is required because there is an appreciable
pressure crop due to the packing in the tube.
SOURCE
Porous polymer vapor sampling method [ 40].
[ 89) Feairheller, W. R., P. J. Marn, ID. H. Harris, and ID. L.
Harris. Technical Manual for Process Sampling Strategies
for Organic Materials. EPA—600/2— 76 l 22 U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, April 1976.
STEEL PROBE
ROTOMETER
HO SE
GAS METER
Figure A-3.
168
-------
In a similar train, shown in Figure A-4, the pump is replaced with
an evacuated cylinder to eliminate pump fluctuations in the
sample lines [ 89]. The cylinder is equipped with a thermocouple
or other temperature readout and a vacuum guage. A glass sampling
bulb is used to back up the polymer—packed tube to permit at
least qualitative identification of low-molecular-weight gaseous
species. Typical flow rates of 100—200 mL/min for 20 minutes
will provide two to four liters of sampled gas. The temperature
and pressure differential in the cylinder before and after col-
lection provides the necessary volume of gas sampled. Some of
the organic materials detected by this procedure from a point
system and polymer curing ovens are listed in Table A—2 [ 89].
A more complex system for sampling organic pollutants in stack
gases is the EPA source assessment sampling system (SASS train)
[ 90]. This system was developed to efficiently collect at a
detectable level all nonvolatile organic compounds present in a
flue gas stream including those that exist as a solid, liquid,
or vapor. In addition, this system is useful for collecting
volatile materials, thus yielding a complete evaluation of the
emission source. This provision adds a great deal of complexity
to the sampling system. In order to provide the entire range of
capability, this system combines a triple cyclone approach for
size fractionation of particulate, an organic adsorber employing
a polymer sorbent, and an inorganic trace element impinger system.
A schematic diagram of the sampling system is shown in Figure A-5
[ 90). A major feature of the SASS train is that it samples in-
organic and organic emissions simultaneously. Inorganic species
are primarily collected by the cyclones and filters. Volatile
inorganics are collected in the solid sorbent trap and impinger
solutions. Organic species are primarily collected in the solid
sorbent trap, although other portions of the train (particulate
samples, impinger solutions, rinse solutions) are solvent ex-
tracted to recover any other organic material.
Other Means of Entrapment of Organic Pollutants
Probably the most basic technique for collecting air samples for
analysis of organic pollutants is the use of evacuated bags or
cylinders. Samples are drawn into these containers and trans-
ported to a laboratory for analysis of the collected gas either
directly or after a concentration step. Commonly used bags are
made of Scotchpac, Tedlar and Teflon. Solvent traps have also
been used to collect organic compounds. In one study ethylene
(90) Hamersma, J. W., S. L. Reynolds, and R. F. Maddalone. IERL-
Procedure Manual: Level I Environmental Assessment. EPA-
600/2—76-160a, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, June 1976. 147 pp.
169
-------
STAI . S S1EEL PR EE
__ POLYMER PACKED TUBE
‘¼
Lj—tr ___ —j
250 ML
DOUBLE ENDED
I FLASK
Figure A-4. Alternate porous polymer system
for organic vapors [ 89].
TABLE A-2.
ORGANIC SUBSTANCES COLLECTED FROM PAINT
AND POLYMER CURING OVENS BY POROUS POLYMER
ADSORPTION AND ANALYZED USING GC/MS [ 89).
Methanol
Ethanol
I sopropanol
2 -Ethoxyethanol
Isobutarioj.
n—Butanol
C 5 Alcohols
n—Propanol
2-Me thybutanol
Ethylene glycol/rnonoethyj. ether
2- (2-Ethoxyethoxy) ethanol
Formaldehyde
Acetaldehyde
Ac role in
Acetone
Methyl ethyl ketone
Diethy]. ether
Butyl acetate
Saturated hydrocarbons
2—Ethoxyethylacetate
Chloroform
Methylene chloride
Cyclohexane
Dimethylcyc lohexarie
Benzene
Toluene
Cylenes
Styrene
!4ethylstyrerie
Dimethyistyrene
C 3 Alkylbenzenes
CM Alkylbenzenes
CM Substituted styrene
Trichioroethane
Dichioroethylene
Carbon disulfide
Isopropylbenzene
Phenol
Benzaldehyde
TEMPERATURE GAUGE
PRESS JRE GAUGE
EVACUATED CYLINDER VACUL’M PUM
170
-------
STACK
THERMOCOVI
CONVECTION
z_ _ _ -
FILTER
3iim
I ’m
GAS COOlER
I _
t—.
-3
XAD-2
CARTRIDGE
TRACE ELEMENT
COLLECTOR
CONDENSATE
COLLECTOR
CONTROL MODULE
THE RMOCOLJ PIE
IO-CFM VACUUM PUMP (2)
Figure A—5. Schematic of source assessment sampling system 1901.
-------
glycol was the trapping solvent for pesticides in the environment.
The trapping solvent was extracted with methylene chloride fol-
lowed by fractionation and cleanup by elution through a silica
gel column [ 91].
Cryogenic sampling has also been employed as a means of concen-
trating trace organic compounds for chemical analysis. A three-
stage cryogenic sampler commonly employs a series of condensers
at 0°C (ice water) , —78°C (dry ice), and —175°C (liquid nitrogen)
Liquid oxygen formation in the third stage is prevented by using
a gaseous nitrogen flush [ 92]. Total organic carbon has been
determined on samples from a system employing a dry—ice trap
followed by an evacuated 8-liter tank [ 93]. A dry—ice cold trap
has been reportedly used to collect dirnethyl—nitrosamine followed
by methylene chloride extraction [ 50).
An alternative method of sampling (wherein chloromethyl methyl
ether and bis(chlorornethyl) ether can be collected) consists of
collection in glass impingers containing a methanolic solution
of the sodium salt of 2,4,6—trichioropheflol. Reaction of the
chlorornethyl ethers with phenol forms stable derivat ves with
significantly increased detector sensitivity; the chiorornethyl
methyl ether and the bis(chloromethyl) ether are detected at the
3.3 g/m 3 and 4.7 g/m 3 levels, respectively. Two impinge:s must
be used in series to trap the compounds in approximately 90%
yield [ 50).
SEPARATION AND ANALYSIS
Samples collected from ambient air and stack gases for organic
material analysis will be in the form of particulate, condensed
vapors, materials trapped in solution or adsorbed onto solids,
or gases. Gases and some liquid samples may be analyzed directly,
but particulates, solid sorbents, and other liquid samples need
to be extracted and put into solution. Methods for extraction
[ 91) Kutz, F. W., and H. S. C. Yang. A Note on Polychiorinated
Biphenyls in Air. U.S. Environmental Protection Agency,
Washington, D.C. (PB-275 978).
192) Conkle, . P., and R. L. Miller. Cryogenic Sampling of
Ambient Atmospheres as a Means of Concentrating Trace
Organic Compounds for Chemical Analyses. Presented at the
6th Annual Symposium on Environmental Research, EdgewoOd
Arsenal, Aberdeen Proving Ground, Maryland, 29 April —
1 May 1975. Published by Dept. of Army, Edgewood Arsenal,
Aberdeen Proving Ground, Maryland, EO—SP—76001, 1976.
[ 93] Salo, A. E., W. L. Oaks, and R. D. MacPhee. Measuring the
Organic Carbon Content of Source Emissions for Air Pollution
Control. 1. Air Pollution Control Association, 25(4):390-
393, 1976.
172
-------
have been discussed in the previous section. Because of the com-
plexity of most samples, a separation step must be employed prior
to analyses. This step separates a sample into fractions thus
reducing potential ambiguities in the determination of organic
components.
Sample analysis can be carried out by a number of techniques
which can be grouped into the general categories of 1) chromat-
ographic methods, 2) mass spectrometry, 3) visible and near-
visible spectrum measurement, 4) electrochemical techniques, and
5) other methods. These analytical techniques along with methods
of separation are discussed in this section.
Sample Separation
Techniques for separating a liquid sample into various fractions
for analysis usually rely on the difference in physical proper-
ties between organic components. A technique has been used
which separates the organic extract of a particulate sample into
fractions by means of successive solvent extractions. Primary
organics were estimated from the carbon solubilized in cyclo—
hexane, and secondary organics were estimated from the carbon
solubilized in successive extraction with benzene and methanol-
chloroform. An upper limit estimate of the elemental carbon was
obtained from the carbon that remained insoluble (94].
In another study, airborne particulates were collected and
extracted with benzene. •The extract was then separated into
neutral, acidic, and basic substances. The acid fraction was
converted to methylated derivatives and found to consist of a
homologue series of fatty acids and aromatic carboxylic acids.
The neutral fraction consisted of saturated aliphatic hydrocar-
bons, PAH, and polar oxygenated substances. The basic fraction
consisted of nitrogen-containing analogs of the important poly-
aromatic hydrocarbons present in the neutral fraction (59].
In a study of the nonvolatile fatty acids in cigarette smoke,
particulate matter underwent a serial extraction, methyl ester
formation, Florisil column cleanup, and analysis by GC. Fatty
acids in the 16-18°C range were determined in tobacco smoke by
this technique [ 95].
[ 94] Appel, B. R., P. Colodny, and J. J. Wesolowski. Analysis
of Carbonaceous Materials in Southern California Atmosphere
Aerosols. Environmental Science & Technology, lO(4):359—
363, 1976.
[ 95] Guerin, 0. G., and w. T. Rainey. Gas Chromatographic
Determination of Nonvolatile Fatty Acids in Cigarette Smoke.
Analytical Chemistry, 46(6) :761—763, 1974.
173
-------
The most widely used methods of separation probably employ some
type of chromatography. ChromatographiC separation has been per-
formed on various media. Silica gel columns have been used in
the chrcTatographiC fracticnatiCr and cleanup of sample extracts
containing chlorinated pesticideS polychlcronaphth lE.ne . x 6
PAH compounds [ 77, 96, 97].
A method has been described utilizing paper chromatography and
consisting of column chromatograPhiC cleanup and separation of
B(a)P from the other PAIl on acelylated paper. After evaporation,
the B(a)P spot is dissolved in sulfuric acid and analyzed by a
Iluorirnetric technique [ 50] . A similar separation technique,
thin layer chromatography (TLC), was coupled with a spectrophoto-
fluorornetric procedure and used for the rapid determination of
7,8 -benz [ de)anthracen-7—Ofle and phenalen—l—one in organic ex-
tracts of airborne particulates [ 61]. Because of the complex
nature of combustion emissions, liquid chromatography has been
employed to initially separate the collected species into
chemical classes after extraction from polymer sorbent collec-
tion media [ 71].
An unambiguous identification of more than 20 aza—arenes and
other nitrogen—bases previously unidentified in ambient particu—
lates was accomplished through isolation of the compounds of
interest. After extraction from the particulate matter, the
organic sample was separated using solvent partitioning and a
sequence of chromatographic separations [ 63].
Manual column chromatography has been extensively used for the
separation of B(a)P and other PAIl obtained from air samples.
B(a)P is usually separated out with benzo(e)pyrene, perylene,
benzo(b)fluoranthene and benzo(k)flUoraflthefle. With poor
separation this mixture can even contain benzo(ghi)perYlefle
and anthracene. With a long enough column B(a)P can be separ-
ated from the benzofluorantheneS. Unfortunately this type of
separation can take half a day. The fairly good resolution of
a longer column is tempered with the longer times necessary for
separation and for a spectral examination [ 50 ).
[ 96] Akin, F. J., M. E. Snook, R. E. Severson, W. T. Chamberlain,
and D. B. Walters. Identification of Polynuc].ear Aromatic
}1ydrocarbons in Cigarette Smoke and Their Importance as
Turnorigens. . National Cancer Institute, 57(1):191195,
1976.
[ 97] Sherina, J., and T. M. Shafik. A MulticlasS, Multiresidue
Analytical Method for Determining Pesticide Residues in Air.
Archives of Environmental Contamination and Toxicology,
3(1):55—71, 1975.
174
-------
High performance liquid chromatography (HPLC) has the potential
to separate B(a)P from other PAH within minutes. Using a 40% cel-
lulose acetate column, B(a)P was readily separated from other PAH
by HPLC. With columns of Durapak OPN and cellulose acetate, a
much larger number of PAH were resolved. The excitation and
emission spectra of PAH separated by HPLC could be recorded dur-
ing the separation by stopping the elution at peak maxima. The
main problems of cellulose acetate separation in HPLC is that
1) cellulose acetate packs down during separation and stops the
flow, and 2) cellulose acetate washes off the column [ 501.
Very good resolution for the separation of B(a)P and B(e)P using
sodium chloride—Chrornosorb columns has been reported. However,
other workers attempting to reproduce these results have had
difficulties. Excellent separation of B(a)P from B(e)P has been
obtained with a nematic liquid crystal column, but the column
bleeds so badly at the recommended elevated temperature that the
resolution disappears after a few runs (50].
An HPLC method has been developed for class fractionation of the
organic material extracted from atmospheric particulate samples.
The PAH fraction was collected and further fractionated using
bonded reverse phase partition chromatography (60]. Chromato-
graphic separation techniques are often coupled directly to a
detector or other means of measurement and identification;
applications of these methods are discussed later in this section.
Depending on the analytical technique employed, the separate
sample fractions may need to undergo a volume reduction or be
dried and redissolved. Drying at elevated temperatures is not
possible because of potential evaporation or degradation of
organic components. The Kuderna—Danish apparatus has been used
for reducing the volume of organic sample fractions after
chrornatographic separation [ 77]. A small, easily assembled,
freeze—drying unit was used to dry a number of samples of various
sizes simultaneously. Evaporation from the frozen or cooled
sample surface reduced the danger of damaging sensitive organic
components [ 981.
Analytical Techniques
Various techniques exist for determining the identity and
quantity of organic pollutants. Many methods are useful for
screening or serniquantitatiVe and qualitative determinations.
[ 98] Wittgenstein, E., E. Sawicki, and R. L. Antonelli. Freeze—
Drying in the Recovery of Organic Material from Extracts of
Air Particulate Matter. Atmospheric Environment, 5(9);8Ol-
810, 1971.
175
-------
These methods usually rely on inference techniques for identifi-
cation of organic components rather than on direct identification.
Mass spectrornetry and electron capture detectors are used for
quantitative identification of organic components. The use of
several techniques in tandem (multi-instrument approach) can
result in reduced ambiguity and improved sensitivity.
Gas Chromatographic Techniques— -
Gas chromatography, coupled with one of several detectors, has
been frequently applied in the measurement of organic pollutants.
A gas chrornatograPh with a flame ionization detector (GC/FID)
was used in the analysis of PCB emissions from various indus-
trial sources. Sample preparation included extraction with
hexane and perchlorinatiOfl. PCB content was measured as the
de-chiorinated isomer [ 81]. In another study, atmospheric
organic pollutants were concentrated on GC packing material in
the field. Analysis consisted of thermally desorbing the organic
material directly into a GC/FID or a GC with a flame photometric
detector (GC/FPD). Quantitative recovery and part-per—billion
(ppb) sensitivity have been reported using this technique [ 991.
Many studies report the use of gas chromatography without speci-
fying the detection technique. One such study reports the use
of GC to determine ppb concentrations of sulfur- and nitrogen-
bearing organic gases [ 56). Pyridine and 2—aminopyridifle were
identified by using a GC technique developed to analyze vapors
collected on a solid sorbent [ 87). Gas chromatography was em-
polyed to obtain quantitative and qualitative information on the
methanol extract of airborne particulate samples [ 62]. Samples
of air pollutants from grain fermentation units were analyzed
using GC. Identified in the sample were ethyl acetate, ethyl
alcohol, n—propyl alcohol, isobutyl alcohol, isoamyl alcohol,
and isoarnyl acetate [ 85). In other studies, GC has been used
for the analysis of organochiorine and organophosphate pesti-
cides, PCB’s, polychiorinated naphthalenes [ 78), benzene,
carbon tetrachloride, chloroform, dioxane, ethylene dichioride,
trichioroethylene, meta—xylene [ 83), fatty components of air-
borne particulate matter containing 12 to 18 carbon atoms in
length [ 511, and benzo(a)pyrene [ 52]. An automatic contiflouS
GC has been used to monitor bjs(chloromethyl) ether in air
samples at less than ppb levels. The monitor utilizes solid
sorbents for gas-phase enrichment, thermal elutiori, and two GC
columns [ 100).
[ 99) Russel, J. W. Analysis of Air Pollutants Using Sampling
Tubes and Gas Chromatography. Environmental Science &
Technology, 9(13):1175—1178, 1975.
[ 100] Frankel, L. S. and R. F. Black. Automatic Gas ChromatO
graphic Monitor for the Determination of Parts -PerBilliOn
Levels of Bis(ChlorOmethYl) Ether. Analytical Chemistry,
48(4):732—737, 1976.
176
-------
A system employing GC and catalytic combustion has been reported
which yields the total organic carbon content of a sample col-
lected from a stack or vent in a freeze—out trap 1931. A
pyrolysis-GC system provided with a data handling unit has been
employed in the separation and identification of organic mate-
rials. The system separates a sample mixture into components
and then pyrolyzes these components one at a time. The pyrolysis
products are separated by two chromatographic systems and, from
their identity, the original material can be identified (101).
This same system, when coupled with a mass spectrometer, has been
used to analyze more than 175 organic molecules comprising the
insoluble carbonaceous components of atmospheric particulate
matter extract 11021.
A CC/atomic absorption apparatus has been constructed for the
measurement of toxic volatile organornetallic species released
from aquatic sediments (1031. A gas chromatographic on—column
concentrating technique has been developed for the analysis of
halocarbons in atmospheric gas mixtures, thus eliminating the
use of cold traps. A coulometric detector is employed which is
specific for the quantitative analysis of chlorinated, bromi—
nated, or iodated hydrocarbons (1041.
Electron Capture Gas Chromatography— -
Electron capture gas chromatographic (EC/GC) analysis schemes
have been reported for effective quantification of low—molecular
weight halocarbons (1051. Ambient air levels of methyichioroform
(1011 Feairheller, W. R., M. D. Schumacher, and W. D. Ross.
Research on the Therinolytic Dissociation of Molecules.
AFNL-TR—74-83, Air Force Materials Laboratory, Wright-
Patterson AFB, Ohio, May 1974. 67 pp.
[ 102] Kunen, S. N., K. J. Voorhees, C. A. Hill, F. D. Hileman,
and D. N. Osborne. Chemical Analysis of the Insoluble
Carbonaceous Components of Atmospheric Particulates with
Pyrolysis/Gas Chromatograph/MaSS Spectroinetry Techniques.
In: Proceedings of the 70th Annual Meeting, Air Pollution
Control Association, Toronto, Ontario, 20-24 June 1977.
Published by Air Pollution Control Association, Pittsburgh,
Pennsylvania, 1977. Paper 77—36.
[ 1031 Jewett, K. L., and G. E. Parris. Environmental Chemistry.
Contract 3130117. National Bureau of Standards, Washing-
ton, D.C. 1977.
[ 104] Williams, F. W., N. E. Urnstead, and J. E. Johnson. Deter-
mination of Trace Quantities of Halogenated Hydrocarbons
in Gas Samples. NRL-6964, Naval Research Laboratory,
Washington, D.C., 6 November 1969. 18 pp.
[ 105] Su, Shth—Wu. Low Molecular Weight Halocarbons. NSF/IDOE-
77—36, National Science Foundation, Washington, D.C., 1976.
15 pp.
177
-------
arid trichloroethylene have been determined on site by direct
irijection of ambient air into a GC followed by detection and
quantification with an electron capture detector [ 106, 107). The
most pronising method of analyzing source and ambient air samples
for PCE’s is reported to be analysis by electron capture gas
chromatography. Quantitation is achieved by perchlorination of
PCB’s to decachlorobiphenyJ. [ 79). Bis(chloroinethy) ether (BCME)
and chiorornethyl methyl ether (CMME) have been quantified using
electron capture/GC. The method involved the collection of air
samples in a methanolic solution of the sodium salt of 2,4,6-
trichloropheriol thus forming stable derivatives with BCME and
CMJ 1E [ 50]. Electron capture-GC has also been used to determine
chlorinated pesticide residue in air. Prior to analysis, samples
were collected in ethylene glycoJ. traps, extracted with rnethylene
chloride, and fractionated in a silica gel column 197).
Nass Spectrometry— —
r ass spectrornetry is another technique that has been used lately.
It has been interfaced with temperature programming, thin-layer
chromatography, and gas chrornatogrphy for the analysis of organic
pollutants. The problem with temperature programming as a sepa-
ration step is that isomers and other compounds with closely
similar sublimation or boiling points would not be separated; for
example, compounds like B(a)P and B(e)P would not be separated.
In addition, there would be possibilities of decomposition and
artifact formation [ 50 ).
Gas chromatography—mass spectrometry (GC—NS) with computerized
data acquisition has been applied to the analysis of the extract
from airborne particulates. Acidic, basic, and neutral fractions
of the extract were analyzed. The generation of mass chromato-
grams for specific ion-fragment masses permitted the location
of mass spectra. Comparison with reference data led to identif i-
cation of more than 100 compounds including aliphatic hydrocar-
bons, PAH, polar-oxygenated substances, fatty acids, carboxylic
acids, and nitrogen—containing analogs of the important PAH
compounds present [ 59]. Analysis by GC—MS of the ether extract
of automobile exhaust fine particulates showed the presence of
hundreds of compounds including about 50% saturated aliphatics,
[ 106) Environmental Monitoring Near Industrial Sites: Methyl-
chloroform. EPA-560/6—77—0 25 , u.s. Environmental Protec-
tion Agency, Washington, D.C., August 1977. 82 pp.
[ 107) Environmental Monitoring Near Industrial Sites: TrichlOrO-
ethylene. EPA-560/6-77—024, u.s. Environmental Protection
Agency, Washington, D.C., August 1977. 75 pp.
178
-------
5% PAH, and 30% oxygenated hydrocarbons [ 50]. Aliphatic hydro-
carbons and PAH have also been identified in natural gas flame
soot using CC-MS preceded by capillary column GC (108].
Ambient air sample extracts were analyzed for the presence of
polychloronaphthalenes (PCN’s) using a GC-quadrupole mass spectro-
meter with computerized data acquisition. The instrument was
operated in the multiple—ion detection mode. The presence of
PCN’s was confirmed from full—scan mass spectra or by monitoring
the chlorine isotope ratio [ 77]. Volatile compounds generated
from chicken manure were collected and analyzed by GC-MS. Vari-
ous alcohols, ketones, esters, carboxylic acid, and dimethyl
disulfide were found [ 75]. GC-MS has also been used for the
analysis of hazardous organic compounds from combustion efflu-
ents [ 71] and the analysis of trace organics in ambient atmos-
pheres [ 92]. Trace, hazardous, organic vapor pollutants in
ambient atmospheres have been determined by capillary gas—liquid
chromatography mass spectrornetry. Twenty—one halogenated hydro-
carbons were detected including the carcinogens, vinyl chloride
and trichioroethylene and numerous oxygen, sulfur, nitrogen, and
silicon compounds [ 76]. A similar procedure was employed to
identify 147 organic compounds emitted from automobile interiors
(86].
High resolution mass spectrornetry (HRNS) has been frequently
cited as being useful in identifying atmospheric pollutants.
Analysis of the volatilized material from temperature-programmed
automobile exhaust particulates by HRMS showed that aliphatics
and aliphatic-substituted benzenes constituted 99% by volume of
the total volatile organic matter associated with the particu—
lates [ 50]. HRMS has also been a useful means of characterizing
the organic fraction of urban aerosols. Compounds detected
included phenols, aromatic carboxylic acids, and aliphatic di-
basic acids. Quantitative accuracy, however, was limited. Some
ambiguities were found in the assignment of origins of ions
formed in the mass spectrometer [ 109].
HRMS has been used to identify bis(chloromethyl) ether in the
atmosphere. In the high resolution mass spectrometric procedure,
the collected organic material is desorbed into the reservoir of
the mass spectrometer heated inlet system and then examined at
high resolution. One part—per—billion of BCME can be determined.
(108] Vick, R. D., and M. J. Avery. Extraction and Identification
of Organic Materials Present in Soot from a Natural Gas Flue.
Contract W-7405-ENG-82, Dept. of Energy, January 1978.
19 pp.
[ 1093 Crittenden, A. L. Analysis of Atmospheric Organic Aerosols
by Mass Spectroscopy. EPA-600/3-76—093, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina
August 1976. 296 pp.
179
-------
The difficulties with the method are 1) chiorornethyl methyl ether
gives the same ion, 2) control of adsorption and
effects, and 3) possible decomposition of BCME in the reservoir
and its associated values [ 50).
HRMS was used to identify the major PAH compounds in samples
obtained from fuel conversion process emissions. The analysis
considered 14 highly carcinogenic PAH having nine Ufli u formulas.
Additional quantitative data were obtained by low ionizing voltage
techniques [ 55]. HRNS combined with GC/MS and auxiliary techniques
such as stable isotopic labeling, has been employed to illustrate
the application of automated, real-time MS in the Study of atmos-
pheric organic pollutants [ 110). Additionally, dioxirane has been
detected in the gas phase by means of photoionization mass
spectrornetry [ 111).
Visible and Near Visible Spectrum Measurement Tech — —
Ultraviolet (UV) infrared (IR) spectra have been used to identify
atmospheric organics [ 52, 112). Ultraviolet visible absorption
spectrophotometry has been used in the analysis of B(a)P and
other PAH following column chromatography. However, a more
selective and faster method of analysis for B(a)P is the GC—UV
absorptiornetric method which has been used to analyze the PAH
of auto exhaust and coke oven effluents (50).
Fourier—transform IR spectrornetry has been employed to study the
organic species participating in photochernical smog formation.
[ 110) Sirnoneit, B. R., D. H. Smith, and G. Eglinton. Application
of Real-Time Mass Spectrornetric Techniques to Environmental
Organic Geochemistry: I. Computerized High Resolution
Mass Spectrornetry and Gas Chroinatograph - Low Resolution
Mass Spectrornetry. Archives of Environmental Contamination
and Toxicology, 3(4):385, 1975—1976.
[ ill) Martinez, R. I., R. E. Huie, and J. T. Herron. Mass Spec—
trometric Detection of Dioxirane, H 2 COO, and Its Decomposi-
tion Products, H 2 and CO , from the Reaction Ozone with
Ethylene. Chemical Physics Letters 57(3):457—459,
November.l977. -
[ 112) Mendenhall, G. D., P. W. Jones, P. E. Strup, and W. L.
Margard. Organic Characterization of Aerosols and Vapor
Phase Compounds in Urban Atmospheres. EPA—600/3-78-031,
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, March 1978. 81 pp.
180
-------
Product analyses of ozone—olefin reactions at ppm levels have
been made by means of computer—aided subtraction techniques [ 113].
IR spectroscopy has also been used for the preliminary screening
of chemical classes in air samples from combustion effluents (71].
An IR technique has been used for determining the nonvolatile
organic matter associated with airborne particulates (114].
Recently developed, portable, IR gas analyzers have been used on
solvent—laden atmospheres. Because all organic compounds have
characteristic IR spectral lines, the analyzer can be used to
detect and measure essentially all of the organic materials
listed in the OSHA standards (115).
Fluorescence spectroscopy has been used frequently but has also
exhibited problems in instrumental reproducibility, quenching
phenomena, excimer formation (as evidenced by a change in the
intensity and broadness of the usual emission spectral band),
sensitized fluorescence, and photooxidation. One of the most
specific and most sensitive methods for B(a)P involves its
fluorescence in concentrated sulfuric acid. Preliminary separa-
tion can be by PC, TLC, GC or liquid—liquid extraction, dependent
on the type of sample being analyzed. It has been recommended
that in the TLC method alumina-coated aluminum foil sheets be
used and the sulfuric acid solvent be purged with nitrogen before
fluorimetric measurement. The advantages of fluorescence analy-
sis in the analysis of B(a)P and other PAN are the greater sens-
tivity the method and sometimes the increase in selectivity as
compared to absorption spectral methods of assay. Low tempera-
ture fluorescence methods are also available for the analysis
of B(a)P and other PAN. Quasi—linear spectra are obtained at
liquid nitrogen temperatures. Since much fine structure is
obtained, the method is capable of unequivocal characterization
of B(a)P. By combining the high resolution of the spectrofluori-
meter with that of the gas chrornatograph, highly selective
methods for B(a)P are possible. A GC-gas phase fluorescence
instrument has been developed for such analyses and utilized in
the analysis of a variety of mixtures (50].
(113] Niki, H., P. Maker, C. Savage, and L. Breitenbach.
Participating in Photochexnical Smog Formation. Ford Motor
Company, Dearborn, Michigan. International Conference on
Environmental Sensing and Assessment, Las Vegas, Nevada,
14—19 September 1975. Published by IEEE, New York, Vol. 2,
Paper 24—4, 1976. 4 pp.
(1141 Della Fiorentina, H., J. Degraeve, and F. Dewiest. Deter-
mination of Non—Volatile Organic Matter Associated with
Airborne Particulates by Infrared Spectrometery - 1. Oper-
ating Conditions. Atmospheric Environment, 9(5):513, May
1975.
[ 115] Kuhiman, C. M. Monitoring Toxic Air Contaminants. Plant
Engineering, 31(15):149—150, 21 July 1977.
181
-------
An imrnurnoassay method with instrumentation has been developed
for the detection of organic contaminants of environmental
concern. The basic technique utilizes an antibody to the con-
tamirAant of concern with subsequent detection by means of
fluorescence polarization or intensity measurements. Successful
assays have been developed for diquat, hexachlorophene, and
diethylstilbestrol [ 116). A fluorescence spot test has been
devised for PAI based on the sensitization of the inherent
fluorescence of such compounds. The basic procedure involves
spotting a filter paper with sample solution, adding naphthalene
to the spot, and visually observing the fluorescence under UV
illumination. In the case of B(a)P, one picogram has been
detected. The method is specific for PAH with minimum iriterfer-
ences and can be used to estimate the general level of PAH
within a factor of 10 [ 11?].
PAH present in atmospheric particulate matter has also been
analyzed using high pressure liquid chromatography coupled
with on—line fluorescence detection [ 118). Laser—excited matrix
isolation fluorescence spectrometry is being developed as an
analytical technique for characterizing polycyclic organic
matter (POM) with special emphasis on the characterization of
chemical carcinogens present in coal liquids and shale oil [ l19 .
Both fluorescence and chemiluminescence techniques have been
proposed to detect toxic air pollutants such as nitrogen dioxide,
ozone, formaldehyde, halogens, hydrogen halides, hydrogen per-
oxide, and organic peroxides. Each of these pollutants undergoes
[ 116) Lukens, H. R., C. B. Williams, S. A. Levinson, and W. B.
Dardliker. Sensitive, Specific Fluorescence Iir nunoassay
Methods for Detecting Pesticides and Other Organic Envir-
onmental Contaminants Arising from Biological or Chemical
Sources. NSF/R —770472, National Science Foundation,
Washington, D.C., 21 November 1977. 95 pp.
[ 117) Smith, E. N., and P. L. Levins. Sensitized Fluorescence
for the Detection of Polycyclic Aromatic Hydrocarbons.
EPA-600/7-78-182, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, September 1978.
31 pp.
[ 118] Foo, N. A., and S. W. Staley. Determination of Polycyclic
Aromatic Hydrocarbons in Atmospheric Particulate Matter by
High Pressure Liquid Chromatography Coupled with Fluores-
cence Techniques. Analytical Chemistry, 48(7): 992—998,
dune 1976.
[ 119) El, W. Laser—Excited Matrix—Insulation F].uorometric
Analysis of Polycyclic Aromatic Hydrocarbons. Grant CH77-
12542, National Science Foundation, Washington, D.C., 1978.
182
-------
photolysis in the solar region or the near-ultraviolet (1201.
A chemiluminescent technique has been developed to determine
the photochernical reactivity of organic pollutants in gaseous
mixtures such as air, engine exhausts, and organic solvent
vapors. In this method, the pollutants react with oxygen atoms
to produce chemiluminescence which results in radiation emitted
at two separate wavelengths and the difference in intensity is
measured (121).
Atomic absorption (AA) has been employed in the analysis of
organic air pollutants. Organotin compounds have been detected
in air at 0.1 mg/rn 3 levels employing AA after concentration on
silica gel and separation by chromatography [ 122]. A GC-A.A
apparatus has been mentioned earlier which measures the release
of organometallic species, such as trimethylarsenic, from aquatic
sediments [ 103). The results of several studies in spectral
methods are presented in Table A-3 [ 123).
Electrochernical Techniques- —
Oxidation of odorous vapors at the anode of an electrochernical
cell was studied as an approach to the analysis of odors. The
technique of linear, potential sweep cyclic voltairinietry was used
to investigate the oxidizability of several amines, sulfides, and
their mixtures. The results indicate that a mixture of sulfides
and arnines can be characterized (124]. A portable hydrazine
(120] Hodgeson, J. A., and W. A. McClenny. Analytical Technique
for Gaseous Pollutants Based on Photolysis and Detection
of Primary Fragements. Presented in American Chemical
Society, Division of Water, Air and Waste Chemistry, 28
August — 1 September 1972, Preprint 12—2.
[ 121] Foritijn, A. Cheiniluminescent Method and Apparatus for
Determining the Photochemical Reactivity of Organic Pol-
lutants in a Gaseous Mixture. Patent Application 548,471
(U.S. Environmental Protection Agency), filed 10 February
1975. 21 pp.
[ 122] Riggle, C. J., D. L. Sgontz, and A. P. Graffeo. The Analy-
sis of Organotins in the Environment. In: 4th Joint Con-
ference on Sensing of Environmental Pollutants, New Orleans,
Louisiana, 6—11 November 1977. p. 761.
[ 123] Herxnann, T. S. Development of Sampling Procedures for
Polycyclic Organic Matter and Polychlorinated Biphenyls.
EPA—650/2—75—007, U.S. Environmental Protection Agency,
Washington, D.C., August 1974.
(124] Nwankwo, J. N., and A. Turk. Electrochemica]. Analysis of
Sulfidic and A nine Odorants. EPA-600/2-76—021, U.S. Envir-
onmental Protection Agency, Research Triangle Park, North
Carolina, June 1976. 46 pp.
183
-------
TABLE A-3. SUMMAR ’ OF SOME INVESTIGATIONS
IN SPECTRAL METHODS [ 74]
Normal
Motbo
Procedure
Compounds
sen itiv t y—
ranoe
U travic ct
Benzene
extractiOr
Benzo(cx)pyrene
0 to 45 pg/100C
air
r 3
itrav:olet
Hexane
Benzo(ci)pyrene
10 lJg/rr.L
extraction
L 1traviolet
Columr
chromatography
Benzo(Cx)pyrene
1 to 6 ,Jg/rnL
tltraviolet
Hexane
Dibenzanthrerie
0.5 to 5 Jg/mL
extraction
Fluorescence
Pentane
extraction
Perylene
Anthracene
senzo(o)pyrene
0.3 pg/mL
0.3 pg/rnL
0.4 pg/rnL
electrochemical analyzer is being designed and built to meas’ :e
trace quantities of hydrazine in air [ 125).
Other Techniques- -
Thermal energy analysis has been employed to determine dimethyl-
nitrosoamine in air. The technique is sensitive at the 1 g/m 3
level [ 126).
A quartz piezolelectric crystal coated with a substrate has been
used for the detection of small mass changes caused by selective
adsorption of organophosphorOus compounds and pesticides. In-
corporation of the crystal into a variable oscillator circuit and
measurement of the change in frequency of the crystal due to the
increase in mass allowed a sensitive quantification of the com-
pounds present down to the ppm level [ 127].
[ 125) Shaw, N. Prototype Hydrazines Electrochemical Analyzer.
Contract F33615—78—C—OGO 7 , Air Force School of Aerospace
Medicine, Brooks Air Force Base, San Antonio, Texas, 1978.
[ 126) Fine, D. H., D. P. Roundbehler, E. Sawicki, K. Frost, A.
Rounbehler, A. silvergleid, and G. A. Dernarrais. Deterrnin—
ation of DirnethylnitrOSOaIr ifle in Air, Water, and Soil by
Thermal Energy Analysis: Validation of Analytical Pro-
cedures. Environmental Science & Technology, 11(6):577,
June 1977.
[ 127] Scheide, B. P., and G. C. Guilbault. Piezoelectric
Detectors for organophosphorus Compounds and Pesticides.
Analytical Chemistry, 44(1l):17641768, September 1972.
184
-------
A portable battery—operated instrument has been developed for
monitoring hydrocarbons and other trace gases in industrial
atmospheres by photoionization (128].
Other techniques include laser intracavity Raman cell, enzymatic
systems for environmental monitoring, gas detection sensors,
liquid crystals, microwave-induced emission spectroscopy, and
direct-current polarography (129]. In one study, the application
of the piperonal test to the benzene—soluble fraction of air-
borne particulates was investigated. Piperonal reacts with
aromatic compounds to give a colored product, which obeys Beer’s
Law. The correlation coefficients between benzo(a)pyrene con-
centrations and the piperonal test was 0.95 and 0.89 for 174
urban and 25 nonurban samples, respectively (123]. In two other
studies, thermochromic tests were used for determining the
amounts of polynuclear compounds containing the fluorenic
methylene group and polycyclic p-quinories. In both cases,
borohydridies were used at reflux temperatures for color develop-
ment. Identification limits ranged from 3 to 40 g. A some-
what similar study was reported where 4—azobenzene—diazOfliurfl
fluoroborate was used to develop color in the determination of
aniline, naphthylamine, and anthramine derivatives. The identi-
fication limits were in the range of 2 to 20 .ig (123).
Gas detector tubes are also available for the semiquantitative
determination of a specific pollutant. These tubes can be used
for monitoring workplace environments and waste gases from auto-
mobiles, industrial processes, etc. The instrument consists of
a hand-operated pump that sucks a measured volume of air per
stroke thrugh the detector tube which is inserted into the pump.
The detector tube is calibrated, and the extent of color change
in the tube indicates the quantity of pollutant present. Wolfgang
Leithe gives a more detailed description of this analytical
technique [ 130].
[ 128] Driscoll, J. N., and F. F. Spaziani. Trace Gas Analysis
by Photoionization. In: 21st Annual ISA Analytical
Instrumentation Symposium, Philadelphia, Pennsylvania,
6—8 May 1975. p 111—114.
(129] Trace Analysis and Detection in the Environment, presented
at 6th Annual Symposium on Environmental Research, Edgewood
Arsenal, Aberdeen Proving Ground, Maryland. 29 April -
1 May 1975. Published by Dept. of Army, Edgewood Arsenal,
Aberdeen Proving Ground, Maryland, E0—SP—76001, 1976.
(130] Leithe, W. The Analysis of Air Pollutants. Ann Arbor
Science Publishers, Ann Arbor, Michigan, 1971.
18 5
-------
APPENDIX B
STANDARD SAMPLE GENE TION SYSTEM
OBJECTIVE
In order to evaluate the sorbent sampling systems used in this
project, it was necessary to have a method for generating known
concentrations of organic compounds in dynamic gas streams. MRC
has developed a dynamic standard Sample Generation System based
on the controlled syringe injection of organic liquids into a
flowing stream of gas (N 2 ), vaporization, and subsequent dilution.
INSTRUMENT
The Sample Generation System is shown schematically in Figure E-l.
The main frame of the system is a modified gas chrornatograph
(F&M Model 700) with its protective metal covering and detector
removed. The oven [ 30 cm (12 in.) x 30 cm (12 in.)] is in a re-
located position on the left side of the main frame. Of the four
heated zones available on the original GC, one controls the oven
temperature, two are used to separately control the temperature
of the two three-port injection blocks, arid the final zone is
available for heating transfer lines used for direct interfacing
with a detector in frontal analysis capacity studies. A pyro-
meter (Simpson 0-500°C), a selection switch, thermocouples, and
necessary circuit modifications allow the monitoring of tempera-
tures in the four zones.
An aluminum plate [ 45.7 cm (18 in.) x 46.4 cm (18 1/4 in.) x 0.6
cm (1/4 in.)) is attached to the main frame to the right of the
oven to accommodate the syringe drives. Necessary bulk—head
fittings, tees, toggle valves, needle valves, tubing, pressure
gauges, and rota neters are used in the configuration indicated
in the figure. Materials of construction for the flow system
are either stainless steel or nickel, and the entire system
is contained in a large exhaust hood [ 2.4 in (8 ft) x 1.2 in
(4 ft) x 1.5 in (5 ft)) with front and rear access through
double sliding doors. Flow control/sensing instrumentation,
pressure gauges, and rotameters are mounted on a panel and rack
assembly in the hood above the generation system.
186
-------
flAM IONIZAIION
OflICIOR
Standard Sample Generation System (SGS)
with effluent monitor modification.
Itow SINSORSICONIROLURs
H
co
-J
PORI
SWUCIIING
VALVE
(ItAfto L
ENCI OSED I
• 106611 VALVE
Wra
(D & • NEWEl VALVE
o DUAL CONTROL N.Y.
o PRESSURE GAUGE
S0RBLNT TUNE
RO IAARI LR
HLAJ1D TRANSFER
(“IS
10-1,10) NtimlnI
Figure B—i.
-------
The G ner tion System functions in the following manner. A
source cf carrier cas (N 2 ) is introduced into the system and
split into primary, secondary and tertiary flows. The primary
flow passes throuch a mass-flow sensor/controller (Brooks Model
5841) which maintains the flow rate at a preset value (0—1,000
mL/min) . This flow passes throuch individually heated and con-
trolled injection blocks which are each designed to accept
three syringes mounted on variable—control syringe drives
(Sage Model 355) . This permits the simultaneous introduction
of six pure licuid components, or potentially many more, if
mixtures of compounds are used in the syringes. The rate of
injection can be varied from submicroliter/hour to milliliter/
minute by the appropriate choice of syringe size and drive rate.
Liquid samples are vaporized in the injection blocks and swept
with the carrier gas into a constant temperature oven.
Most of the primary flow passes through the oven and through a
needle valve used to regulate the pressure in the primary flow
path. It is then expelled to the vent (Vi) . A small portion
(usually l0 mL/min) of the primary flow is removed through a
dual-control needle valve and passes into a switching valve
(SV1) which is actually one—half of an eight—port switching
valve. The second half of the switching valve is used for an
identical function in the second dilution stage. The switchin
valve makes it possible to insert a mass—flow sensor (Brooks
Model 5810, 0-100 rnL/min) in the flow path (indicated by the
dashed line in Figure B-i) to measure the exact split of the
primary flow. Once the split has been established, the flow
sensor is switched out of the system so that the sensing
elements are not subjected to sample -cor.t ining organic vapors.
The split is maintained by carefully controlling the pressures
in each dilution stage.
Pressures are monitored with precision pressure gauges [ Heise
Model CM, 0—6.9 x l0 Pa (0—100 psi)). By maintaining a con-
stant pressure in each dilution stage, a constant pressure
differential is established between the stages assuring a con-
sistent split ratio across the needle valves.
The split—out portion of the primary flow passes from the
switching valve and is combined with the secondary carrier flow
which enters the constant temperature oven through a mass-flow
sensor/controller (Brooks Model 4841, 0-2,000 mL/min) . As in
the case of the primary flow, most of the secondary flow passes
through the oven and is expelled to the vent (V2) through a
needle valve. A small portion (usually 10 mL/min) of the
secondary flow is removed through a dual—control needle valve
and passes into a switching valve (SV 2 ) . This valve functions
in an identical manner to SV1 and places a second mass—flow
sensor (Brooks, Model 5810, 0—100 mL/min) in the flow path for
measuring the second split.
188
-------
The split—out portion of the secondary flow passes from the
switching valve and is combined with the final (tertiary) di-
lution stream which enters the constant temperature oven through
a mass-flow sensor/controller (Brooks Model 5841, 0-50 L/min).
This final concentration is available for sampling through a
five-port manifold system that will accommodate sorbent sampling
tubes. sampling rates through the sorbent tubes are maintained
by rotameters (Brooks Model 1110) at the tube exits. The unused
portion of the final dilution stage is expelled to the vent (V3).
The sample generation system is connected through an eight-port
(Valco) selection valve and stainless—steel tubing [ 1.6 mm
(1/16 in.)] to a flame ionization detector. This arrangement
permits the selective monitoring of concentration levels in each
of the three dilution stages and the additional monitoring of
effluent from five sampling locations.
Evaluation studies have demonstrated linear concentration re-
sponses over the operational ranges of the syringe drive rate,
the two split monitors, and the three dilution stages. By the
appropriate selection of syringe size, drive rate, split ratios,
and dilution rates, it is possible to achieve continuously vari-
able concentrations from percent to low part—per-trillion levels
using this system.
189
-------
APPENDIX C
OPERATION MANUAL
PURPOSE
The purpose of this manual is to provide the information neces-
sary to construct, prepare, calibrate, and operate the ambient
air collection system which was designed, under EPA contract
68-02-2773, to collect and concentrate a certain number (15 to
20) of selected compounds of potential carcinogenic nature from
ambient air for later analytical assessment.
PREPARATION, CLEANING, AND CONDITIONING OF REPLACEABLE SORBENT
CARTRIDGES
The ambient air collection system is comprised of two major
parts: the sorbent tube tray, which contains replaceable sc:bent
cartridges that collect and concentrate organic vapors from an
air stream passing through them; and the Nutech Model 221-lA Air
Sampler, which draws the air stream through the sorbent tube
tray. This section will provide the procedures for preparing the
replaceable sorbent cartridges for use in field sampling, while
other aspects of preparing, calibrating, and operating the ambient
air collection system will be discussed later in this manual.
Cleaning of Sorbent Materials
The replaceable sorberit cartridges used within the ambient air
collection system’s tube tray consist of three types, each type
containing a different sorbent material (Tenax-GC, Porapak R, and
Amnbersorb XE-340). These sorbents are listed in Table C-l along
with other pertinent information. They may be obtained from their
manufacturers, or Tenax-GC and Porapak R may be obtained from a
chromatography supply distributor (e.g., Supelco, Ailtech, Applied
Science, etc.).
Generally, all commercial sorbent materials as received from
suppliers contain significant levels of background contamination
as the result of production, packaging, and/or transportation. To
prepare the sorbent materials for cartridge sampling, their back-
grounds must be reduced to minimal levels and then maintained at
these levels. To start, one can remove most of the initial
contamination from the sorbent materials by an extensive solvent
cleanup. The procedure for this solvent cleanup is given below:
190
-------
TABLE C-i. PERTINENT SORBENT INFORMATION
Mesh size
Manufacturer’ s tempera-
ture limit
Conditioning temperature
Desorption temperature
Desorption time
Physical description
Amount packed in a
cartridge
Chemical composition
Major thermal decomposi-
tion products
Solvent incompatability
Manufacturer address
35/60 mesh
350°C
325°C
300°C
30 mm
off-white powder
0.9 to Li g
2 ,6-diphenyl-p-phenylene
oxide
alkyl benzenes, styrene,
benzene, alkyl phenols
chlorinated solvents
Enka, Inc.
Netherlands
Information Tenax-GC
Ambersorb XE-340
Sorbent
P pakR
50/80 mesh
250°C
230°C
<200°C
60 mm
off-white sphere
1.6 to 1.8 g
N-vinyl pyrrolidone
vinyl pyrrolidone,
pyrrolidone
Waters Associates
Milford, MA
>350°C
325°C
300°C
60 mm
black, shiny sphere
3.0 to 3.4 g
styrene-divinyl benzene
styrene, benzene, alkyl
benzenes
7
Rohm & Haas
Philadelphia, PA
-------
1) Soxhiet extract sorbent for 16 to 24 hr with distilled-in-
glass (high—purity) methanol.
2) Vacuum dry sorberit at 120°C and approximately 1.02 x iO Pa
(30 in Hg) vacuum for 16 to 24 hr.
3) Soxhlet extract sorbent for 16 to 24 hr with distilled-in-
glass (high-purity) pentane.
4) Vacuum dry sorbent at 120°C and approximately 1.02 x lOs
(30 in Hg) vacuum for 16 to 24 hr.
5) Store sorbent in a clean glass jar with a screw-on cap
within a dessicator.
At this point the sorbent is clean enough to be loaded into glass
cartridge tubes (which will be discussed later), and may be stored
for a few months ( 3 to 6 months) without re-extraction as long as
it is protected from contaminants. Sorbent may also be stored
for several weeks under solvent after one of the extraction
procedures and prior to vacuum drying, if desired.
The purpose of this sorbent cleanup procedure is to remove most
contaminants from the sorbent material; however, contaminants may
also be added during this cleanup procedure, such as from trace
impurities present in the solvent, if one is not careful to pre-
vent such contamination. Therefore, one should always use high-
purity solvents and clean glassware during soxhlet extractions,
and the glass joints should be sealed with Teflon tape or Teflon
sleeves (not with silicone grease). Further nore, a small charcoal
trap should be used in the vacuum line of the vacuum oven to
prevent the contamination of the drying sorbent material by the
back-diffusion of vacuum-pump lubricants.
Preparation of Glass Cartridge Tubes
In addition to cleaning the sorbent materials which are contained
within the replaceable sorbent cartridges, the glass cartridge
tubes must also be appropriately prepared. The first step of
this preparation is to make these tubes (using pyrex glass)
according to the specifications given in Figure C-i. This tapered
cartridge design has been found to allow an appropriate compromise
of sorbent cartridge properties, such that pressure drop is kept
to a minimum, thermal desorption and sorbent capacity properties
are kept to a maximum, and the sorbent material is adequately
contained. (The reasoning behind the selection of this sorbent
cartridge design is more thoroughly discussed in the final report
of this contract.) During the manufacture of these glass cartridge
tubes, each tube should be numbered (at one end, as indicated in
Figure C-i) to permit easy identification of sorbent cartridges
for record-keeping and sampling purposes. These numbers should
192
-------
-14cm
0.6 cm
7.ócm
I 12 mm 0.D . 10 mm 1.0 . 1
NUMBER4rnrnI.D.
Figure C-i. Specifications of selected sampling tube design.
be able to withstand high temperatures (400°C) and should not be
washed off by common laboratory solvents (acetone, methanol, toluene,
methylene chloride, or pentane). Ceramic numbers annealed to the
glass at +400°C have been found to be quite suitable. Preferably,
the numbers should be a distinct bright color, such as red, although
other colors are permissible as long as they can be distinguished.
Once the glass cartridge tubes have been constructed, they may
be stored indefinitely in their empty condition. However, before
sorbent is packed within them, the glass tubes should undergo the
cleaning and deactivating procedures listed below.
1) Clean glass tubes in a heated ultrasonic bath containing
glassware detergent (e.g., Microclean, LiquiNoX , etc.) for
4 to 8 hr, or meticulously “hand-clean” with glassware
detergent and a soft test tube brush.
2) Rinse interior and exterior with methanol (reagent grade).
3) Bake tubes dry at 300°C to 400°C for 8 to 16 hr.
4) Prepare a 10% solution of dimethyldichiOrOSilafle in high-
purity, distilled-in-glass toluene (deactivating solution).
5) Soak glass tubes in deactivating solution for at least
15 mm.
6) Thoroughly rinse tubes with high-purity, distilled-in-glass
methanol.
7) . Bake tubes dry at .300°C for 8 to 16 hr.
8) Store glass tubes in a clean, dry environment.
193
-------
While following the procedures just described, care should be
taken to avoid contaminating the tubes (with fingerprints, etc.).
Therefore, the tubes should always be handled while wearing
protective gloves and/or with clean utensils (e.g., tweezers,
tongs, etc.). Furthermore, after cleaning and deactivating, the
tubes should be packed with the appropriate sorbent material(s)
within a reasonable period of time ( one month).
Cleaning of Other Hardware Materials
The other materials (other than the sorbent materials and glass
tubes) which are used in the preparation and storage of the replace-
able sorbent cartridges include glass wool, which is used to con-
tain sorbent within the glass cartridges, and stainless steel
fittings, viton 0-rings, and culture tubes, which are used to seal
and store the sorbent cartridges.
Generally, glass wool as received from a chromatography supply
distributor (e.g., Supelco, Ailtech, Applied Science, etc.)
should be clean enough for packing the sorbent cartridges since
little of this material is used and the sorbent cartridges are
conditioned after packing. The type of glass wool ordered should
be a soft, fine pyrex wool which has been silanized (preferably
with dimethyldichiorosilane). If the glass wool which is avail-
able is probably contaminated (through use), then either new glass
wool should be ordered or the contaminated glass wool should
undergo the following procedures.
1) Soxhiet extract glass wool for 16 to 24 hr with high-purity
distilled—in—glass methanol.
2) Vacuum dry glass wool at 120°C and approximately 1.02 x l0 Pa
(30 in. Hg) vacuum for 16 to 24 hr.
3) Soxhiet extract glass wool for 16 to 24 hr with high-purity
distilled-in-glass pentane.
4) Vacuum dry glass wool at 120°C and approximately a 30 in. Hg
vacuum for 16 to 24 hr.
5) Prepare a 10% solution (deactivation solution) of diinethyl-
dichlorosilane in high-purity, distilled-in-glass toluene.
6) Soak the glass wool in the above deactivating solution for
at least 30 mm.
7) Soxhlet extract glass wool for 8 to 16 hr with high-purity,
distilled-in-glass methanol.
5
8) Vacuum dry glass wool at 120°C and approximately 1.02 x 10 Pa
(30 in. Hg) vacuum for 24 to 48 hr.
194
-------
As previously discussed during the cleaning of the sorbent
materials, the extraction glassware should be clean; Teflon tape
or Teflon sleeves should be used at the glass joints instead of
silicone grease; and a charcoal trap should be used in the vacuum
line of the vacuum oven to prevent contamination of the glass
wool by the back-diffusion of vacuum-pump lubricants. This pro-
cedure, however, has been found to create a problem with silicon
contamination due to excess silanizing reagent retained on the
glas wool. Blank analyses of glass wool, therefore, should be
performed to determine whether it is suitable for packing sorbent
tubes. If unacceptable, the extraction and drying procedures
should be repeated.
After the glass cartridges are packed with the sorbent materials
(which will be discussed later), the sorbent cartridges are
sealed with stainless steel fittings and viton o-rings, and
placed within pyrex culture tubes (200 mm long by 25 mm diameter)
with Teflon-lined, screw—on caps for storage (as depicted in
Figure C-2). Since these materials are to be used to protect the
sorbent cartridges from atmospheric contaminants, they should be
prepared in such a manner that they introduce minimal contaminants
to the sorbent cartridges. To do so, the viton 0—rings, stainless
steel fittings, pyrex culture tubes, and the Teflon—lined, screw-
on culture tube caps should be cleaned according to the procedures
listed below.
1) Clean indicated items in a heated ultrasonic bath containing
glassware detergent (e.g., Microclean, LiquiNox , etc.) for
4 to 8 hr.
2) Thoroughly rinse these materials with high-purity, distilled-
in-glass methanol.
3) Bake items dryat .120°C for 8 to 16 hr.
4) Store items in a clean, dry environment until used.
SCR(W ONP .AST1C CAP REVERSED STAINLESS
g. PYft CUL1URE TUBE ( 2 mm Ionq bp 3mm hm. v) / STW. FRONT FERRULE
VITOP4 0-RING
STAINLESS SIEEI.
TTTINGS
Figure C-2. Materials used to seal and store a sorbent
cartridge.
195
-------
Preparation of Sorbent Cartridges
With the sorbent materials solvent cleaned, and dried, the glass
ca ridae tubes cleaned and deactivated; and the other hardware
materials (glass wool, stainless steel fittings, etc.) cleaned;
all the required materials are ready for the sorbent cartridges to
be prepared for sampling. This preparation consists of two major
steps: (1) packing the sorbent tubes and (2) thermally conditioning
them to remove most remaining contaminants, thereby reducing their
backgrounds to minimal levels.
To pack a sorbent cartridge the procedures given below should be
followed, with all weights carefully recorded (this will be
discussed more thoroughly later).
1) Weigh an empty glass cartridge tube.
2) Weigh two small pieces of glass wool (‘ l0 to 20 mg each).
3) Place one piece of glass wool into the glass cartridge tube
at one of the tapered ends just before the 6 mm O.D. portion
of the tube (see Figure C-3A).
4) Draw a vacuum at this end of the sorbent tube (see Figure
C—3B).
5) Place several grams of one of the three sorbent ma eriais in
a clean, shallow, glass container (e.g., watch glass, petri
dish, etc.).
6) “Vacuum” the sorbent material into the glass cartridge tube
until just a small space is left at the unplugged tapered
end just before the 6 mm O.D. portion of the tube (see
Figure C-3C and C-3D).
7) Remove the vacuum and place the remaining piece of glass
wool at the other end of the tube (see Figure C-3E).
8) Weigh the sorbent cartridge and determine the amount of
sorbent within the cartridge.
9) Modify the amount of sorbent placed within the cartridges
to try to attain the sorbent quantities specified in Table C-i
(0.9 to 1.1 g of Tenax—GC, 1.6 to 1.8 g of Porapak R, and
3.0 to 3.4 g of Ambersorb XE-340).
While packing the sorbent cartridges as just described, care
should be taken to avoid extraneous contamination of the sorbent
materials, glass wool, or glass cartridge tubes. Therefore, as
previously indicated, all materials should be handled while wear-
ing protective gloves and/or with clean utensils (e.g., tweezers,
tongs, etc.). The vacuum line should include a charcoal trap to
196
-------
VACUUM
.WATCH GLASS
C. VACUUM” SORBENT INTO TUBE.
B. ATTACH TUBE TO A VACUUM.
C U71:
0. LEAVE ROOM FOR OTIER GLASS WOOL PLUG.
NUMBER
GlASS WOOL SORBENT MATERIAL GLASS WOOL
E. PLACE REMAINING GLASS WOOL Ar UNPUJGGEI) 1APERED END.
Figure C-3. Procedure for packing a sorbent cartridge.
A. PLACE GLASS WOOL AT ONE TAPERED EM).
NUMBER
I - ’
NUMBER
GLASS WOOL
SORSENI
-------
prevent contamination by the back-diffusion of pump lubricants.
Generally, these procedures may be followed in the open labora-
tory. Although, the sorbent will quickly adsorb vapors present
in the lab air, the thermal conditioning procedure should remove
these contaminants. However, if a clean-box or clean-room is
available, then these facilities would be preferred while packing
the sorbents since a clean, inert atmosphere would further reduce
potential background and contamination.
After the sorbent cartridges have been packed, they need to be
thermally conditioned to remove most remaining contaminants there-
by reducing their backgrounds to minimal levels and making them
suitably clean for sampling. The current technique which is used
to condition sorbent cartridges has been found to be adequate for
most sampling applications.
The sorbent cartridge thermal conditioning technique is straight-
forward and simply involves passing a purified, inert stream of gas
(helium or nitrogen) at 30 mL/mifl through a sorbent cartridge for
16 to 24 hr with the sorbent cartridge within an oven heated to
an appropriate temperature (230°C for Porapak R, and 325°C for
Tenax-GC and Arnbersorb XE-340). The gas stream should be purified
by placing a drierite trap, a molecular sieve trap, and a charcoal
trap in the gas stream (in that order) before passing the gas stream
through the sorbent cartridges. This gas stream is then attached
to a manifold comprised of stainless steel fittings and contained
within a laboratory oven (with a maximum temperature of >325°C).
A small section of such a manifold is sketched in Figure C-4,
shown with a few sorbent cartridges attached. A manifold of this
type can be extended to condition as many tubes as there is room
for w±th n the laboratory oven, as long as a flow rate of
30 mL/min can be attained through the tubes. The tubes are con-
nected to the manifold using graphite ferrules.
Figure C-4. Section of manifold used for thermally
conditioning sorbent cartridges.
CAITIIDGE
J
198
-------
The critical portion of this sorbent cartridge thermal condition-
ing technique occurs after the cartridges have been purged with
purified gas in a heated environment and are ready to be removed
from the conditioning manifold to be stored for later sampling.
As mentioned previously, sorbent will quickly adsorb vapors
present in the lab air. This will occur as soon as thermal
conditioning is stopped (i.e., heat and purge gas are turned
off). Therefore, it is imperative that the sorbent cartridges be
sealed from contaminating laboratory atmosphere as soon as
possible after thermal conditioning is discontinued. When the
sorbent cartridges are barely cool enough to be handled (while
wearing protective gloves and/or using clean utensils), the nuts
(with graphite ferrules) which are used to connect the cartridges
to the manifold should be carefully (the cartridges easily break
while hot) loosened. As rapidly as possible the sorbent cart-
ridges should then be removed by slipping them out of the manifold
nuts and then sealed with metal fittings and 0-rings (as shown
previously in Figure C-2). After all the sorbent cartridges are
sealed in this manner, they should be placed inside individual
culture tubes and the culture tubes sealed with their Teflon-
lined, screw-on plastic caps (see Figure 2). These sorbent tubes
are now properly prepared for collecting samples.
To assure the sorbent backgrounds are at their minimal levels,
sampling should be done as soon as possible after the tubes are
conditioned (preferably < one week). Tubes stored for more than
“one week should again be thermally conditioned prior to
sampling. Any labels which are used to designate the samples
the sorbent cartridges are to be used to collect (or blanks, etc.)
should be affixed to the outside of the culture tube to prevent
the ink or adhesive from contaminating the sorbent. Finally, if
rough handling is anticipated, especially during transportation,
clean glass wool should be placed within the culture tube
surrounding the sorbent cartridge to offer the cartridge addi-
tional cushioning and protection from breakage.
Storage, Background, and Contamination
As mentioned several times previously, precautions should be
taken to prevent contamination of a sorbent cartridge during its
preparation. These precautions include handling all materials
while wearing protective gloves and/or using clean utensils,
using high-purity, distilled-in-glass solvents, using Teflon-tape
or Teflon-sleeves instead of silicone grease at soxhiet extractor
glass joints, placing a charcoal trap in the vacuum line of a
vacuum oven system to prevent the contamination of drying materials
by the back-diffusion of pump lubricants, and many others (which
have been previously stated). The purpose of cleaning the sorbent
and other materials, taking precautions to prevent contamination
during sorbent cartridge preparation, and conditioning the sorbent
cartridges within one week of sample collection is to assure
that the sórbent cartridges have minimal levels of background and
contamination.
199
-------
It should be noted here, however, that although contamination can
contribute to the background (materials present during a sorbent
cartridge analysis when no materials have been added to the
cartridge [ i.e., a blank]) present on a sorbent cartridge,
background can also be contributed by the sorbent material through
slow degradation (or “aging”). For this reason a sorbent cartridge
should be reconditioned just prior to sampling if it has been
stored for more than one week. However, the thermal or chemical
decomposition of a sorbent caused by sampling/analytical condi-
tions also can significantly contribute to sorbent cartridge
backgrounds. Thermal decomposition of a sorbent results, of
course, from overheating the sorbent cartridge during condition-
ing, sampling, or desorption. Therefore, not exposing the sorbent
cartridges to temperatures above their temperature limits (given
in Table C-i) should minimize thermal decomposition problems.
Chemical decomposition caused by exposing the sorbent cartridges
to inappropriate solvents (also given in Table C-i) should also be
avoided. Chemical decomposition may also be caused by reactive
species (such as NO , SO 2 , and HC1) in the sample matrix (which
if anticipated can e potentially removed by selective filters
placed prior to the sorbent cartridge), but they should not
present a major problem with ambient air samples.
Usually, the analyses of “blank” sorbent cartridges is enough tc
indicate the type of compounds and their levels which comprise
the sorbent background. Quite often, background (of a particular
sorbent) from many “blank” tubes vary somewhat in their amounts,
as indicated by peak sizes (or areas), but retain a certain
pattern (i.e., peaks from various analyses have similar retention
times and similar chemical type). This paTtern is probably due
to compounds produced by sorbent decomposition which is typical
for that particular sorbent because of the chemical composition
of the sorbent. For example, Table C-i lists some of the major
thermal decomposition products that might be expected for the
sorbent materials used for this project based on their chemical
composition (also given in Table C-i). Knowing the types of
decomposition products to expect from a particular sorbent, one
can determine whether thermal or chemical decomposition of the
sorbent likely occurred during sample collection, such as from
matrix effects. Therefore, one can evaluate whether the compounds
detected during an analysis of a sorbent cartridge sample are the
result of the collected sample or some type of sorbent
decomposition.
Records
It was briefly noted when discussing the packing of the sorbent
cartridges that records should be kept of all glass cartridge
tube and sorbent weights. Indeed, the reason that numbers are to
be permanently annealed to the glass cartridge tubes is for record-
keeping purposes. It is essential that accurate histories be
kept for each sorbent tube in order to assure the quality control
200
-------
of all sorbent cartridge samples. Therefore, the sorbent cart-
ridges must be marked in some manner (the annealed numbers) in
order to assure individual identification. FurthermOre, all
manipulations of the sorbent cartridge or the sorbent material,
including weights, cleanup and conditioning procedureS and
storage times, need to be recorded and associated with each
sorbent cartridge. By keeping such accurate histories of each
sorbent cartridge, one avoids such poor quality control situa-
tions as trying to compare the results of two duplicate samples,
one collected on a freshly packed sorbent cartridge and the other
collected on a sorbent cartridge in use for at least five years.
Obviously, a great disparity of results could be causedbY the
difference in “age” and treatment of the two sorbent cartridges.
In addition, such recordkeepiflg helps to prevent the possibilitY
of getting different types (i.e., packed with different sorbent
materials) of sorbent cartridges mixed-up and overheating some
during conditioning because the wrong thermal conditioning tempera-
ture was used. For example, if a few Porapak R sorbent cartridges
are conditioned with a number of Tenax-GC sorbent cartridges at
the conditioning temperature of Tenax-GC (325°C), the temperature
limit of the Porapak R sorbent cartridges (250°C) would be
exceeded and the Porapak R would thermally decompose. Therefore,
the history of each sorbent tube should be kept meticulously,
beginning with the “initiation” of the sorbent cartridge, which
should include sorbent cleanup procedures 1 sorbent cartridge
weights, and conditioning information; continuing through the
sorbent cartridge’s “sampling lifetime,” which should include
storage, packaging, and transportation information, sampling
conditions and dates, and analytical log-in numbers, procedures 1
and results; and ending with the “termination” of the sorbent
cartridge, which should include the procedure and date the
• cartridge was either reconditioned or unpacked. All of this
information should be kept on each sorbent cartridge according
to its annealed number and retained within a sorbent history log
book. Figure C-5 depicts the information which should be kept
within the log book associated with the “initiation” of each
sorbent cartridge, information which should be entered while
following the procedure in this section of this Operation Manual.
The records associated with the aspects of a sorbent cartridge’S
“sampling lifetime” and “termination” will be further discussed
later in this manual.
FLOW CHART
For convenience, a flow chart which summarizes the procedures for
the preparations cleaning, and conditioning of the replaceable
sorbent cartridges is given in Figure C-6.
201
-------
Name:
Tube /: Sorbent Type: Date Packed:
Weight of: Glass Cartridge: Glass Wool: Sorbent:
Sorbent N terja1 Information
Supplier: Mesh Size: Lot /:
Date Opened: Date Arrived: Generation: 1 2 3
Cleanup Procedure:
Cleanup Dates:
Sieve Size: Sieve Date:
Storage Procedures and Conditions:
Storage Dates:
Storage Procedures and Conditions:
Storage Dates:
Sorbent Cartridge Conditioning
Date Packed: Storage Procedure & Dates:
Conditioning Procedure:
Conditioning Date:
Conditioning Procedure:
Conditioning Date:
Conditioning Procedure:
Conditioning Date:
Conditioning Procedure:
Conditioning Date:
Figure C-5. Information which should be recorded in the
sorbent cartridge logbook.
202
-------
p p..’
11
RCIOCLT t0TU T 1624
343 OI l IWOR
2) VOC DRY S 120C l ID
10 P. (30 00
VOCUOJI FOR 16 .25 341
3) 000*1ST IOTUCT 14-24
MD. 1* pLNflJl0
4) VOC DRY S 120C WIlT
1.03 . 00’ P. (30 IS
1(q) VOCUY$I FYI 14 .2*
5) STOOl ON 4 00.1511 00*31
u 1 110 5051*-CAP
IN DL1010000 1 (
Fig’ure C-6. Flow chart summarizing sorbent cartridge
preparation procedures.
1)
TORSO
D OD CULTOBI
.11IIINC 5000USD PIIPII00 10* or 10.033 CAlMING
OF
TOIL CAPS
CCC
*01 50 141.5 CA4T112GI TOILS
-
1)
ClAN I. MAY10 1) c .uo $000MOOOb 1) sorT p7 5 55 GLASS C I I I IN A MACID 11.T*ADC$IIC
Ul.1 5010 1 11C lATH WITH Ia t. P7 5*0 GLANS WOOL OP GUI5110SDILI *001(4 0010 047141510 44 14 11.
3410*00*1 4- 5 ISA. M C I. flOII IUPP0101 WC.I I1 $
2) 45 .0* zosoos US OUT 2) STORE ZN I C.LMI 3) 505411 ZITIICY 2) 01341 0007R0000.Y 0071(4 51DR
SIN N*A IY 0*034 051 OIY0SAIUMT 1424 345. 11 11.0* P 5 1 200 41.104
*00*] 1
3) 1031 DI I S 300405C 3) 030. 05? 5 120C IWO 3) 5*41 III S I.24 .C FOR 4-10 45.5
505 4—14 153 1.02* 10’ Ps (30 10
$551 510005 FOR 14 10
24.3.
4) P1.01(421 10% 50007005 4 ) 105021 511UCO 4) 3100*0 0 * 4 01.030. DI I DV1004
OP DI34T1(ft1.0100 .US 1424 453. II P 1 5 0 MM 531
110010 15 41004-PY*20Y
?oL0L*I
5) 500* TUNIS II 010*1 1) P 0 0. DRY S 120010
400.21005 FOR 0.03 0 10’ (30 04 1(q)
15 WIN. 00S 16-24 55.
4) HORSE 715505311.? OtTO 4) 55 1 0 4 * 1 L0, S000TISI
5 .104 (Ko l5-PUIITT ) D1lUT l 4Y I.0 00 4 -C l0-
SOUND IN 41104 POLITY
7) W A SIT I 500 ’C 7) 30*4 10*35 L
FOR 5-10 $ 52. LW 5JOVI SOLOTIOR
I 501 30503
I) 110*1 011 COAL. DRY I) 105111? 510500?
DV 1 0Gu R IT P CI N—CA 455.
1) 61051 SOIT 10.031 CIITP.1401 (MOORS! I
2) 6115 1 TM 10-SIRS 510051 05 00.13* WOO l.
(450500?)
1 .
3) FRY 51*05 INTO 50 05 OP 155*
4) 007004 450055 DO OIlS 105
5) 5 5 .000 *000*52. 0*1151 0? 101151? I N
051.0.34 10030 TAOI
5) “0500IN 1 401151* ONTO TOM LUO’IOIO
03005 501 53 *1 l .
7) US 0100* ? 10450 WOOL
55 6110* COITSISM (*00551?)
203
-------
)urowt:’TTc trc 50JEW CAfiTIIDOES
s c 3Th A T
7
CA) Ufl- A ITit. AN T-
0R2 3SS / TM taouTtr; C VV ANT AflAO
GA ( AT M
F .ATi AT T I TTSRI / 2 ATiACN SC AC3T cApmTrTcTs ic
T i flCM-LTKLc P1.ISiI I MIM!FO G (T4IWC .APlT T IMU $
AlGAl IA jAI DAT ‘TAT3t Tl 3 TV flIA UTI ‘ OT T AOUflr.
A JT .’Tl it 3T L.J T
r 4 I A T’Y r
IJANt POM A8TQ.IN fl 3tLATVE (32 ATN
____________________ hA ll AAIALA-ADAA Z IA 1GM
DACOMIT DA-TT .DITI’ \ ?OMIAA A
PALPAA! £ L .TAG PAOGM.AT T tLAltTTI M D I IA MM 2* lO G
)) IAAA . GITT. DE O CTt1T*X ‘ I rTG c1l2c3: G A*
\
P ____
5c Ill F 1IM OMAJTT
Figure C-6 (continued)
204
-------
PREPARATION, CALIBRATION, ANI) OPERATION OF THE AMBIENT MR
COLLECTION SYSTEM
This section of this manual describes all aspects of the prepara-
tion, operation, and calibration of the ambient air collection
system. Included are instructions on how to assemble the replace-
able sorbent cartridges within a sorbent tube tray, how to operate
the Nutech Air Sampler, how to calibrate the Nutech Air Sampler
without and with a sorbent tube tray attached, and how to set up
arid inonitor the ambient air sampling system. In addition quality
control, storage, recordkeeping, and other aspects of a field
sampling program are discussed. This section should provide
enough information for the collection of ambient air samples using
replaceable sorbent cartridges, and the handling of these samples
until analysis.
Q eration arid Calibration of the Nutech Model 221-lA AC/DC
Air Sampler
The Nutech Model 221-lA AC/DC Air Sampler is an air pumping system
which is used in the ambient air sampling system to draw an air
stream through a sorbent tube tray. A major portion of the manu-
facturer’s “Operating Manual: Model 221-lA, AC/DC Sampler for
Remote Sample Collection Using Wet Chemical Collectors and Gas
Chromatography cartridges,” is presented below with the permission
of the Nutech Corporation. Sections, tables, and figures drawn
from this manual are referenced. Metric values have been added
and some minor corrections made.
INTRODUCTION [ 1313
The Model 221 gas sampler is optimally designed for remote sampling
of hydrocarbon vapors using CC cartridges. The pumping system is
powerful enough to provide 10 Lpm and a large pressure drop to
accommodate a wide variety of cartridge collectors. However, the
unit is small enough to be hand carried, and power requirements
are within the range of automobile storage batteries when AC power
is not available.
Analysis of cartridges is done using a laboratory CC. A special
interface unit is normally required to obtain reproducible results.
A CC Thermal Desorption Unit (Nutech Model 320) or a custom
designed apparatus of individual choosing serves this purpose.
[ 131] Operating Manual: Model 221-lA, AC/DC Sampler for Remote
Sample Collecton Using Wet Chemical Collectors and Gas
Chromatography Cartridges.. The Nutech Corporation,
Research Triangle Park, North Carolina.
205
-------
Spec .f .cat ons [ 131]
DIMENSIONS:
WEIGHT:
POWER:
GAS FLOW:
SA ?LE INLET:
39.4 cm (15.15 in.) wide, 36.8 cm (14.5 in.)
high, 27.9 cm (11 in.) deep.
15.9 Kg (35 ib).
110 V, 60 Hz and/or 12 V DC.
DC current varies with flow.
221-1A20 draws 2.0 - 2.5 amps,
22l-1A14 draws 2.0 — 3.5 amps.
Built-in battery charger.
Recharging time approximately
equals running time.
Max. Flow—-lO — 15 Lpm.
Max. Vacuum--6.8 x i0 - 8.1 x l0 Pa
(20—24 in. Hg).
Flowmeter--O.5 - 10 Lpm.
Dry Gas Neter--280 rn 3 (10,000 cu ft) total
readable to 2.8 x iO- in 3 (0.001 cu ft).
6.4 nun (1/4 in.) Quick Connect in upper side
for easy mounting of GC cartridge hcläer
or other sampling connections.
Vacuum gauge monitors pressure
drop across cartridge.
Operating procedure [ 131]
• The Model 221 gas sampler comes equipped with a 120 volt
power cord and external battery supply leads for a 12 volt
automobile storage battery which the user must supply.
• The Model 221 is ready to be put into operation.
Figure C-7 shows the switching arrangement for AC and DC
operation and the charging mode.
• For AC operation, turn on AC switch and then DC switch.
• For DC operation, connect battery terminals to battery
and plug into side receptacle. Turn on DC switch to start
pump. (AC switch is inoperative in this mode.)
AC SW1104
a
PUMP OPUATN&
ttCflflW AC
AL5ostwct1r .
O4UGC
P SATJaT.
C !
sAr r CAl cKALL
SnJaO
Figure C-7. Switching arrangement [ 131].
206
-------
XHAIJST
Figure C—8. Flow diagram for Nutech Model 221 AC/DC Gas Sampler [ 131].
SAMPEE
IMfl
0
-J
VACUUM
GAUCt
QUick
CONNFCT
t
TUBE
MUFRER
fl.OWMEIER DRY GAS
METER
PUMP
-------
1aintenanoe [ 131]
P urn r - -
purnp ‘see F±gure C-9) is a permanently lubricated, permanent
magnet low battery drain, DC operated diaphragm pump.
Reciacinc a Diaohraam-—Remove the four head screws to remove the
head. Then remove the two diaphragm hold—down plate screws and
replace diaphragm. To reassemble reverse the above procedure.
Reniacinc the Intake Valve--Remove four head Screws to remove the
head. Remove the valve hold-down screw and replace valve. Reverse
procedure to reassemble.
cing the Exhaust Valve--Remove the four head screws and remove
the head. Turn the head upside-down and remove the flat Phillips
head screws in the valve plate.
After the screws have been removed, turn the head over on a flat
table. Introduce a slight amount of air from an air line or blow
gun into the exhaust part of the head. Be careful that not too
high a pressure is used [ approximately 1.4 X 10 Pa (20 psi)].
This will blow the valve plate down on the table if the exhaust
valve itself is intact. If the valve plate gasket becomes torn
or mutilated through this disassembly, it should be replaced.
The valve on the valve plate can now be replaced by removing the
valve hold-down screw and replacing the valve. In order to
reassemble reverse this procedure.
Replacing the Eccentric & Bearing Assembl --Remove the two front
cover screws and remove cover. With a 4.0 irun (5/32 in.) Allen wrench
remove the two pipe plugs on right side of housing. Loosen the set
screw in the eccentric (Key No. 22) with a 3.2 mm (1/8 in.) Allen
wrench by going through the port hole in the side of the housing.
Next loosen the cap screw on the lower end of the connecting rod
with a 4.0 mm (5/32 in.) Allen wrench. After this has been done,
the eccentric and bearing assembly (Key No. 21) can be removed.
To reassemble, reverse procedure. Do not over-tighten connecting
rod screw, 1.7 J(15 in.-lb) is sufficient. The eccentric screw
cannot be over-tightened.
Renlacing Brushes--The complete brush holder and lead wire
assembly (Key No. 11) must be replaced. Remove the two and cap
screws and remove end cap. Then remove two screws holding brush
cord. Pull complete brush holder and rubber grommet out of housing.
To replace, reverse procedure. Be sure to keep brushes inside of
brush guide before placing over commutator.
Valve - -
A specially designed stem point in the metering valve permits
positive shut-off on a separate seat. The fine metering needle
208
-------
PARTS LIST FOR PUMP
; -
L
ilS_
1i
627072 (KE•Y # 11 )
KEY PART No .
1 614377
2 623260
3 602492
4 625 8
5 614372
6 625266
7 607139
8 625114
9 63364
11 627072
12 625006
13 617045
DESCR PTION
MOTOR END CAP ASSEMBLY
SCREW - M.E.C.
MOTOR HSG & MAGNET ASSEMBLY
SCREW - SLEEVE TO HSG
FRONT COVER
SCREW - FRONT COVER
CONN. ROD
SCREW - CONN. ROD
FLUSH SEAL PIPE PLUG
BRUSH & LEAD WIRE ASSEMBLY
SCREW - BRUSH ASSEMBLY
VALVE KEEPER
KEY PART No.
14
15
16
17
18
19
20
22
23
24
25
26
27
621102
654129
625 141
660131
625109
654649
608148
625244
608092
660437
625160
615403
633439
DESCRI PTION
VALVE FLAPPER
VALVE PLATE
SCREW -VLV. & H.D.PL
HEAD
SCREW-HEAD
DIAPH. H.D. PLATE
DIAPHRAGM
SET SCREW - ECC
ARMATURE FAN & BRNG. ASSEMBLY
HOUSING
SCREW - FLAPPER
SPACER
GASKET - VLV. PLATE
Figure C .-9. Parts list for pump [ 131].
209
-------
is used only for flow control and can be shut off just like a
blunt nose valve without the metering qualities.
There are 10 full turns from full close to full open, 8 of which
are for fine metering.
Materials are the following:
Body: Brass forging
Stem: 316 stainless steel
Packing: Teflon
Maximum operating pressure: 2.1 x l0 Pa (3000 psi) at
21°C (70°F).
Temperature range: -51°C (—60°F) to 230°C (450°F).
Flow characteristics of this valve are shown in Figures C-lO and
C—li for two different pumping capacities.
Flowmeter - -
Range; 0 - 10 LPM
Accuracy: 5% of full scale
Calibration: The flow meter has been factory calibrated. If
you wish to re-check calibration do so only with devices of
certified accuracy.
Temperature Limit: 66°C (150°F)
Pressure Limit: 6.9 x l0 Pa (100 psi)
Atmosphere: Do not expose to strong chlorine atmospheres or
solvents such as benzene, acetone, carbon tetrachloride, etc.
Disassembly: The flowrneter can be disassembled for cleaning by
disconnecting the piping, dismounting the unit from the panel
and removing the top-plug-ball stop. Remove ball by inverting
the body and letting the bell fall into your hand.
Cleaning: The flow tube and body can be cleaned with pure soap
and water and a soft bottle brush. Do not use benzene, acetone,
carbon tetrachloride, alkaline detergents, caustic soda or
liquid soaps.
Reassembly: Reinstall float, remount, connect piping and put
the unit back in service.
Dry Gas Meter- -
To replace front gasket, remove four mounting screws from dial
cover. Install new gasket and reassemble. To replace large
top mounting gasket, remove dial face cover, remove dial face
assembly by removing two dial face mounting screws. Next remove
8 screws holding meter top to meter bottom; remove old gasket,
clean surfaces, install new gasket and reassemble.
210
-------
—
=
E
C,
C)
*
L J
0
U-
I-
10
8
6
4
2
0
0
VACUUM, nchesHg
Typical flow rate versus inlet pressure
drop for Model 221—1A14 (131].
Figure C-1O.
10
8
6
4
2
0
VACUUM, inches Hg
Figure C - il. Typical flow rate versus inlet pressure
drop for Model 221—lA2O (131).
211
4 8 12 16 20 24
0 4 8 12 16 20 24
-------
Since meter body is made of aluminum, use force sparingly to
tighten pipe nipple.
Vacuum Cau e Oto 1.02 x 10 Pa (0 to 30 in. Hg) ——
The Bourdon tube type gauge is made of brass. A pin hole permits
the sensor element to come to equilibrium with the gas in the
sample line. If this pin hole should become plugged making the
gauge inoperative, the gauge may be removed by removing two screws
in the rear of the panel. A small wire can be used to clear the
hole. Care should be taken not to tamper with the sensor mechanism
and alter the calibration.
Drying tube- -
Make sure end caps are all the way on. To make ends more
secure wrap each end once with pipe dope tape before replacing
end caps.
When installing drying agent, place glass wool in one end, fill
with drying agent, finish with glass wool on the other end and
secure end caps.
Power Supply- -
Requires 120 V, 60 Hz input; 12 V DC output.
Input and output connections are shown in Figure C-12 with the
numbers corresponding to the barrier strip on the power supply
with terminal one toward the front of the sampler console. The
DC output supplies the pump motor when the AC and DC switches are
in the ON positions; the DC output charges an external battery with
the AC switch in the ON position and the DC switch in the OFF
position. The sampler can be operated from an external 12 V DC
source with the D switch in the ON position.
Be careful not to short the battery terminal connector leads
when AC power is in ON position. DC power connection should
always be made or broken by unplugging from the cabinet.
Battery Connections- -
Battery connections are made by plugging the plug on the battery
leads into the battery lead receptacle on the side of the sampler
and connecting the clips to the + and - poles of the battery.
Be careful to observe polarity--red is positive; black is
negative.
Parts List [ 131)——
221—001 Base
221—002 Cabinet
221—003 Panel
221—004 Cabinet Cover
221-005 Battery Charger
212
-------
ts.) 12OVAC
( )
BATTERY
CONNECTIONS
Figure C-12. Wiring block diagram for Nutech Model 221 AC/DC
Gas Sampler 11311.
1.5
DC AMMETER
+
AC POWER
__ I
+
-------
221-006 6 Terminal Barrier Strip
221—007 AC Plug
221-008 DC Socket
221-009 DC Plug & Battery Leads
221—010 Fuse (1 1/2 amp)
221—012 AC Switch (SPST or SPDT)
221-013 DC Switch (DPDT)
221-014 Ammeter
221-OlE ’ 6.40 nun (1/4 in.) Quick Connect
221-016 6.4 mm (1/4 in.) SS Elbow
221-017 Mtg. Bracket for SS Elbow
221—018 t PTU
221—019 3.2 mm (1/8 in. NPT to 4.8 mm (3/16 in.) Hose
Connector
221-020 Vacuum Gauge [ 0-1.02 x iO Pa (0-30 in.
Hg.)
221—021 Drying Tube
221-022 Metering Valve
221-023 3.2 nun (1/8 in.) NPT Street Elbow
221-024 Muffler
221-025 3.2 mm (1/8 in.) to 6.4 mm (1/4 in.)
Male Union
221-026 6.4 nun (1/4 in. ) NPT to 4.8 mm (3/16 in) Hose
Connector
221-027 Flow Meter (0 — 10 LPM)
221-028 Dry Gas Meter
221—029 12.7 nun (1/2 in.) to 6.4 nun (1/4 in.) N?T
Reducing Bushing
221-030 3-Conductor Power Cord
221—031 4.8 m m (3/16 in.) ID Tubing
221-032 Pump (Specify Size 35.6 m (14) or
50.8 cm (20 in.) Brackets)
To order replacement parts or to obtain further information about
the Nutech Air Sampler write or telephone;
Nutech Corporation
P.O. Box 12425
Research Triangle Park, N.C. 27709
(919) 682—0402
Calibration of the Nutech Air Sampler for the Ambient Air
Collection System
Prior to using the Nutech Air Sampler as part of the ambient
air collection system for obtaining air samples, certain parts
of the Nutech Air Sampler should be properly calibrated. The
most important variable to determine during sample collection is
the total volume of the sample collected. This is true because
this volume is used in combination with the amounts of various
compounds analyzed in the collected samples to determine the
concentrations of the compounds in the original air samples.
Therefore, the dry gas meter of the Nutech Air Sampler should
214
-------
be properly calibrated prior to each sampling program to allow
an easy and accurate determination of the total sample volume for
each sample. Calibration of the dry gas meter can be performed
using a calibrated wet test meter by standard procedures. For
additional certainty, the flow meter (rotameter) of the Nutech
Air Sampler should also be calibrated prior to each sampling
program (and rechecked immediately before and after each sample
is collected). This can be done by using a calibrated mass flow
meter or bubble meter [ 132]. This rotameter can be used during
sampling to monitor the flow rate through the ambient air collec-
tion system. The average or integrated flow rate can then be
multiplied by the sampling time to yield a total sample volume
which can be compared to the sample volume determined by the
dry gas meter.
Design and Preparation of the Sorbent Tube Tray
This part of this manual describes the preparation of the sorbent
tube tray which is used to contain the replaceable sorbent car-
tridges that collect and concentrate organic vapors from an air
stream passing through them. The preparation, cleaning, and
conditioning of these replaceable sorbent cartridges was de-
scribed in a previous section of this manual. At the end of that
section, the sorbent tubes, consisting of three types (Tenax-GC,
Porapak R, and Axnbersorb XE-340), were ready to be used for
sample collection. Through cleaning and conditioning their
background levels had been reduced to a minimum and this minimal
background level maintained by storing the tubes capped with
metal fittings and 0-rings inside a culture tube with a
Teflon-lined screw-top cap. The hardware needed fcr the storage
of each sorbent tube is described in Table C-2 and the use of
these materials for tube storage is demonstrated by Figure C-l3.
For sample collection, four sorbent tubes of specific types will
be connected in series within a special tray, such that the
collected air sample first passes through a Tenax sorbent tube,
then through a Porapak R sorbent tube, then through an Amber-
sorb XE-340 sorbent tube, and finally through a second, backup,
Ainbersorb XE-340 sorbent tube. With these sorbent tubes con-
nected in series in this order, the highest molecular weight
compounds will be collected on the Tenax tube, which has the
greatest ability of the three sorbents to desorb such compounds.
Any intermediate or low molecular weight compounds which “break
through” the Tenax tithe are then collected by the Porapak R
sorbent tithe. Very volatile, low molecular weight compounds will
“break through” the Porapak R tube also and pass into the first
Aznbersorb tube which, of the three sorbents, has the greatest
ability to retain such compounds. Hopefully, this combination of
[ 1321 “Quality Assurance Handbook for Air Pollution Measurement
Systems, Vol. II”. U.S. Environmental Protection Agency, May
1977.
215
-------
TABLE C-2.
HARDWARE MATERIALS REQUIRED FOR STORAGE OF
EACH SORBENT TUBE
Nut
Stainless steel
6.4 mm (1/4 in.)-Swagelok
Two
Front ferrule
Stainless steel
6.4 mm (1/4 iri.)-Swagelok
Two
Cap
Stainless steel
6.4 mm (1/4 in.)-Swagelok
Two
0-ring
Vi.ton
“6.4 mm (‘ l/4 in.) 0.D.
Two
Culture tube
Glass
20 cm long x 2.5 cm
diameter-Pyrex,
Kimax or equiva-
lent
One
Culture tube
Plastic with
for 2.5 cm diameter
One
screw-top cap
teflon liner
culture tube
TEFLONtINED,
SCREW-ON PLASTIC CAP
REVERSED STAINLESS
GLASS W
STAINLESS STEEL
FITflNGS
Figure C-13. Storage of a sorbent tube.
PYREX CULTURE lUBE (200 mm long by 25 mm die meter
216
-------
three tubes will quantitatively collect all compounds of interest
to this project, however, some of the very volatile, low molecular
weight compounds may “break through” the first Ainbersorb tube and
pass into the second, backup Ainbersorb tube. This backup tube
serves to check the ability of the first three tubes to retain
the compounds of interest to this project. The quantities of
compounds on one or more of the first three sorbent tubes, and
not found on the backup tube, should be summed to obtain the
total amounts collected. These amounts can then be divided by the
sample volume collected to determine the approximate original
concentrations of the compounds found in ambient air. For
compounds found on the backup sorbent tube as well as one or more
of the first three sorbent tubes, their sununed quantities should
be considered to represent the minimum amounts of these compounds
present in the sampled air. When dividing these quantities by
the sample volume collected, the concentrations of these com-
pounds in ambient air should be reported as greater than or equal
to the values obtained.
In some instances, such as if a number of samples are collected
simultaneously, the analysis of one or two Ambersorb backup tubes
may suffice to ascertain which, if any, compounds have “broken
through”. It may then not be necessary or desirable to analyze
the rest of the Ambersorb backup tubes collected. Where such a
situation is anticipated, using backup tubes with each sample
collected may not be desired. Therefore, the backup tubes can be
replaced in some of the sorbent tube trays by a piece of stain-•
less steel tubing (6 mm O.D., ‘l4 nun long). If any question exists
about whether or not backup tubes should be used on certain
samples, it would probably be advisable to collect the backup
samples and later decide whether they should be analyzed.
Figure C-l4 gives the dimensions of the aluminum box to be used as
a tray for containing the sorbent tubes. It also shows the
locations and sizes of holes to be drilled for attaching the tube
retaining clips, the tray cover, and the 6.4 nun (1/4 in) stainless
steel bulkhead unions. The rest of the hardware necessary for use
within the sorbent tube tray are listed in Table C-3. For cleaning,
the aluminum box and retaining chips should be rinsed with tap
water and dried. The rest of the hardware materials for the
sorbent tube tray should be prepared according to the procedures
given previously for the cleaning of hardware materials for sorbent
tube storage. After cleaning the sorbent tube tray and fittings,
three stainless steel/teflon connectors, as depicted in Figure
C-15, should be assembled in the laboratory using appropriate care
such as by wearing protective gloves to prevent contamination
from handling. The stainless steel/teflon connectors should then
be placed within the tube tray and the cover placed on the tray
to assure that all necessary parts are readily available for the
“loading” of this tray with sorbent tubes in the field.
217
-------
TABL E C-3. HARDWARE REQUIREMENTS FOR EACH TUBE TRAY
er
Tray U rove:
Bu1 head un on
ur um
Stainless
steel
20 cm (6 in.) x 11 cm x (4
6. mit (1/4 in. )—Swageiok
1/2
in ) x 4 cm (1 1/2 in. ) On
T o
educin; union
Nut
Stainless
Stainless
steel
steel
6.4 mit (1/4 in.) to 3.2 mm
3.2 nut (1/8 in.)—Swagelok
(1/B
in. )—Swagelok
Six
$ i
Front ferrule
Stainless
steel
3.2 mm (1/8 in. (—Swagelok
Six
Back ferrule
Stainless
steel
3.2 mm (1/8 in. ) —S a0elOk
Six
Pluc
T .th ing
Sta3nless
Teflo n
steel
6.4 mm (1/4 in. ) —Swagelok
57 mix (2 1/4 in.) to 64 rex
5.2 mm (2/S in.) 0.D.,
1.6 mit (1/16 in.) i.t.
(2
1/2
in.)
long,
Ten
Three
Clips
(Metal)
- for 12 mm 0.D.
tubes
Eight
Screw s
(Metal)
- for tray cover
- for reta lfllfl9
clips
Four
Eight
m3 0,e) 12.1mm
S 1i2ej 6m,r
01 5
v 1so 76 mm (ILES coo 6 mm . 1i2 ix.
17. (3 J ix i a in. S .S.SwA iE LO
(L1J in. RiaXKLA0 U½1O m
______________________ -I-
27/3L is.
D mm US OR f
0.4mm (1 /4 ix. ) ‘ (S )32 In.
Ii Yam Ix. TRAY VER
17. 5mm
27.0 liii 27.0 mn 2 7.0mm 1k/lb n
*—.-U k / l O in. /. t (-1/IA ir., )u- (1l0 In. —
! F
HOUSRW 7 53
UlE RlT6It lNG
Ollsfl.) (Iii.
•7\ ,
3.2mm
(k/Re.)
01 5.
2.6mm
1 (a in.
(Mar
141)2 in.)
Figure C—14. Dimensions of tube tray, and locations and
sizes of drilled holes.
218
-------
(PREPARE THREE FOR
EACH TUBE TRAY)
SS1TEF’ CN CONNECTOR
STAINLESS STEEL
FITTINGS
TEFIC
TUBING
3.2mm 18 in.)
S. S. SWAGELOK NUT WITH
FRONT & BACK FERRULES -
6.4 mm (114 in.) TO 3.2 mm
(1)8 in.) REDUCING UNION
6.4mm (114 in.) S.S.
SWAGELOK PLUG
GLASS WOO). PLUG
(INSERTED PRIOR
TO EACH SAMPLE
COLLECTED
Figure C-15. Tube tray with all hardware assembled.
219
-------
Step by step instructions for the ‘loading” of the tube tray with
four sorbent tubes are given in Figures C-16 through C-27. it
should be remembered that sorbent tubes quickly adsorb vapors from
the atmosphere when they are opened, which introduces an error to
the quantitation of compounds collected. Therefore, the tubes
should be assembled quickly and, as shown by the figures, caps and
plugs should remain in position until a connection is to be made
at that position. Furthermore, once a connection is made, it
should immediately be tightened with wrenches to assure a good seal.
To prevent the introduction of contamination from handling the
sorbent tubes, they should be assembled while wearing protective
latex gloves. In addition, the ends of the sorbent tubes (i.e.,
at the metal fittings) should not be touched or otherwise dis-
turbed, because these ends will later be exposed to the analytical
system during sorbent tube desorption. After the sorbent tube
tray is “loaded 1 , it is ready to be connected to the Nutech Air
Sampler for sample collection. Since the sorbent tubes are not
as well protected from contamination within the tube tray as they
are when stored, a sample should be collected with the tube tray
within a few hours after being loaded.
Calibration and Operation of the Ambient Air Collection System
Before the ambient air collection system is used for sample
collections, several tasks should be performed. One task is t
prepare a pump/tray connector such as depicted in Figure C-2
This hardware should be cleaned by the procedures described pre-
viously fcr cleaning hardware materials for sample tube storage.
This connector may be placed in the tube tray of the ambient air
collection system, along with the stainless steel/teflon sorbent
tube connectors, for transportation into the field. Also, a setond,
unattached tube tray should be “loaded” with four sorbent tubes (not
necessarily freshly conditioned) in the manner described in the pre-
vious section. This tray will be used for measuring the flow rate
of the Nutech Air Sampler prior to sample collection. The “cali-
bration” tube tray may be transported into the field while loaded,
since some contamination of the tubes is unimportant as they will
not be analyzed. In addition, the dry gas meter and rotarneter of
the Nutech air samples should be calibrated prior to a field sam-
pling trip as previously described. Finally, any solutions to be
used for spiking the samples collected should be prepared in the
laboratory prior to field sampling.
When going into the field to collect ambient air samples, at
least the items listed below should be taken.
1) Ambient Air Collection System with Nutech Air Sampler,
tube tray, and all necessary connectors.
2) Power cord for the Nutech Air Sampler and a long extension
cord.
220
-------
STAIM.(Ss STEEL
FITTLNGS
STAD&ESS SilO.
Fm1 4Gs
Figure C-16. Remove cap from numbered end of a Tenax tube,
and plug from one end of a stainless steel!
teflon connector.
Figure C-17. Attach the Tenax tube to the opened end of the
stainless steel/teflon connector. Tighten with
wrenches.
TEMx lUlL
TDLOi 1UIIrIG
TUgS
221
-------
Figure C-18.
Remove plug from other end of stainless steel/teflon connector and
the cap from the non-numbered end of a Porapak R tube.
t\)
t\)
-------
Figure C-19. Attach the Porapak R tube to the remaining end of the first stainless
connector. Tighten with wrenches. Then remove cap from the other end
of the Porapak R tube and a plug from one end of a second stainless
steel/teflon connector.
ts )
t )
t )
POIblp 4 ; N
FU t
1 U8b4G
-------
scm ft ON
CONNI dON
Figure C-20. Attach the Porapak R tube to the opened end of the second stainless
steel/teflon connector. Tighten with wrenches. Then remove the plug
from the other end of the stainless steel/teflon connector and the cap
from the non-numbered end of an Amhersorb tube.
AMBER ORB IUL3E
I . ’)
fiNAl ThBt
PONAPAK R ! I I
/
ssrnF ON
COWMC1 O N
-------
SSfIFflON
ONN(CTOR
stilt
FIllINGS
Figure C-21. Attach the Ambersorb tube to the remaining end of the second stainless
steel/teflon connector. Tighten with wrenches. Then remove the cap
from the other end of the Ambersorb tube and a plug from one end of a
third stainless steel/teflon connector.
ILILON UJOING
t%)
I -n
TENAX-GC fliRt
PORAPM( R lUR(
AMBIRSt*B UJBI
SS/IFFLON
GONNECIOR
-------
fl5ACKUP)
Figure C-22. Attach the Ambersorb tube to the opened end of the third stainless
steel/teflon connector. Tighten with wrenches. Then remove the plug
from the other end of the third stainless steel/teflon connector and
the cap from the nonnwrtbered end of a second Ambersoi.b tube (backup).
ssIiiIt.ON
CONNI C1OR
SS(TEftON
CONNI CTOR
t )
a’
ns*x-4C lUBE
PORM AKRThBE
AM8ERSORB TUBE
flE11 0N
COPINEC1OR
-------
SSIEII(ON
CONNECTOR
Figure C-23. Attach the backup Aznbersorb tube to the remaining end of the third
stainless steel/teflon connector. Tighten with wrenches.
SS(WItON
CONNECTOR
—3
TENAX TUBE
PORAPAI( R TUBE
AMBERSORU TUBE
AMBERSORB TUBE
4BACKUP)
SSII(FLON
CONNECTOR
-------
1/h” S.S. SWAGELOI( BULKHEAD UNIONS
AMBERSORB TUBE
(POSITION #3)
AMBERSORB TUEE
(POSITION 4)
ACKU P
Figure C-24.
ALUMINUM TRAY
TUBE RETAINiNG
CLIPS
TE AX TUBE
(POSITION #1)
PORAPAK R TUBE
(POSITION 2
Place series of sorbent tubes near the aluminum tube
tray as showii above. Remove the plug from the inside
portion of the right and bulkhead union and remove the
cap from the free end of the Tenax tube.
SCREW THREADS FOR
ATTACHING TRAY COVER
SS/TEFLON
CONNECTORS
228
-------
AMBERSORB TUBE
(POSITION #3)
AMBERSOR B
(POSITION #4)
BACKUP
Figure C-25.
ALUMINUM TRAY
RETAINING
CLIPS
TENAX TUBE
(POSITION #1)
PORAPAK R TUBE
(POSITION #2)
Careful Y slip the Tenax tube into the tube retaining
iips in position #1. Slide the tube up to and attach
to the opened bulkhead union. Tighten with wrench.
114”S.S. SWAGELOK
BULKHEAD UNIONS
SCREW THREADS FOR
AIrACHING TRAY COVER
CONNECTORS
229
-------
AMBERSORB TUBE
(POSITION *3)
AMBERSORB TUBE
(POSITION #4)
BACKUP
Figure C-26.
ALUMINUM TRAY
TUBE RETAINING
CLIPS
TENAX TUBE
(POSITION ! 1)
PORAPAK RTUBE
(POSITION • )
Carefully slip the Porapak R tube and the first Anther-
sorb tube into the tube retaining clips in positions
#2 and #3. Remove the plug from the inside portion of
the left-hand bulkhead union and remove the cap from the
free end of the second (backup) Arnbersorb tube.
1/4” S.S. SWAGELOK
‘BULKHEAD UNIONS
SCREW THREADS FOR
ATtACHiNG TRAY COVER
230
-------
AMBERSORS TUBE
(POSITION #4)
BACKUP
AMBERSORB TUBE
(POSITION #3)
SCREW THREADS FOR
ATIACHING TRAY COVER
ALUMINUM TRAY
TUBE RETAINING
CLIPS
TENAX TUBE
(POSITION #1)
PORAPAK R TUBE
(POSITION #2)
Figure C-27. Carefully slip the second Ambersorb tube (backup)
iñ o the tube retaining clips in position #4. Slide
the tube up to and. attach to the opened bulkhead
union. Tighten with wrenches.
1/4” S.S. SWAGELOK
BULKHEAD UNIONS
CONNECTORS
231
-------
I.
(114 in.) S. S. SWAGELOK
REDUCER
Figure C-28. Pump/tray connector.
TEFLON TUB1NG
3.2mm (1/8 in.)O.D.,
1.6mm (]Jl6in.) ID.,
460 mm (1-1/2 ft. ) TO
915mm (3ftiLONG
3.2mm (1/8 in.)
S. 5. SWAGELOK NUT,
FRONT FERRULE, AND
BACK FERRRULE
3.2 mm (118 in.) TO 6.4 mm
6.4 mm (1/4 in.) S. S. SWAGELOK CAP
232
-------
3) A sufficient number of stored, freshly conditioned
sorbent tubes (as determined by the sampling plan, which
will, be discussed later).
4) A bubble meter, stop watch, and calculator for flow rate
calibrations.
5) A “calibration” sorbent tube tray.
6) A stopwatch or wrist watch for measuring the sampling
times.
7) Necessary tools, including wrenches, screw drivers, and
hex key.
8) Clean, silanized glass wool for particulate plugs for each
sample.
9) Vials for containing the glass wool plugs after sample
collection.
10) Spiking solutions and clean syringes ,
11) Thermometer and hydrometer (and any other meteorological
equipment des red) or, evaluating the sampling conditions.
12) “sampling Fact Sheets” and a pen.
13) This manual!!
14) Necessary boxes, trunks, tape, and packing materials
for transporting the above items.
15) Any required paperwork (e.g., shipping receipt, etc.).
Once in the field, before sample collection is initiated the
Ambient Air Collection System should have its sampling flow rate
measured with a calibrated bubble meter. The bubble meter may
be attached to the exhaust of the Nutech Air Sampler, and the
exhaust flow rate monitored whenever necessary. This exhaust flow
rate can be compared to the inlet flow rate before and after sample
collection. Measurement of the inlet flow rate must be taken with
a “calibration” tube tray attached to the Nutech Sampler (see
Figure C-29) in order to approximate the pressure drop incurred
with a sorbent. tube tray in this position during actual sampling
and to prevent the contamination of the Nutech Sampler with sur-
factants. This tube tray should be quickly leak-checked as de-
scribed below. With the flow needle valve fully opened and the
Nutech Sampler plugged into an AC power source, the Nutech Air
Sampler should be turned “on” by first switching the AC switch
and then the DC switch to their “up” positions. The flow needle
233
-------
valve is adjusted to obtain the flow rate specified by the sam—
pling plan. The flow at the inlet of the “calibration” tube tray
should be measured with the bubble meter at least three times.
This should be compared with exhaust and the rotameter flow rates,
and all flow rate values recorded. This procedure re-checks the
calibration of the rotameter which can be used during actual
sampling, in conjunction with exhaust flow rate determinations,
to determine the approximate sampling flow rate and to detect any
major changes in this flow rate during sample collection. This
value can be used to determine a sample volume that can be compared
to the volume measured by the dry gas meter. Other conditions of
the Nutech Air Sampler, such as temperature and measured pressure
drop, should be recorded during its “calibration” for comparison
with these conditions measured during sample collection.
With a tube tray attached to the Nutech Air Sampler, a leak check
must be performed to assure that air is being drawn only through
the tray inlet. Since air is being drawn, not pushed, through the
sorbent tubes and various connectors during sampling or calibration,
a vacuum is created at any leak within the system. Therefore, this
system cannot be effectively leak—checked with a leak-detection
liquid because that liquid would be drawn into the sampling system
at the point of any leaks and contaminate that system. However,
the vacuum phenomenon can be successfully used to perform a
“mechanical” leak check. This is done by first attaching the
pump/tray connector to the Nutech Air Sampler with the end to be
attached to the tube tray still capped. With a bubble meter
attached to the exhaust of the Nutech Air Sampler, the system
should be turned “on”. The connector and pump should be quickly
evacuated and no flow observed if no leaks are present within
the system. if flow is observed, tighten all fittings until none
is evident. Check internal connections within the Nutech Air
ampler, if necessary. The Nutech Air Sampler should be quickly
turned “off” after a leak free system is obtained. The pump/tray
connector is then attached to the tube tray (“calibration” or
sample) outlet, with its inlet still capped (see Figure C-30).
The above procedure is then repeated. If all connections were
made securely when the tube tray was assembled, a relatively
leak-free system should result. if a flow rate of less than
5 mL/min is observed, then the leak-check should be considered
“acceptable”. A flow of 5 mL/min through small leaks is
insignificant compared to the sampling flow rate of 1 or 2 L/min.
The results of the leak check should be recorded and the pump
turned “off” prior to opening the tube tray inlet.
After calibrating the flow rate of the Ambient Air Collection System 1
the final preparation should be made for sample collection with
the Nutech Air Sampler turned “off”, the “calibration” tube tray
should be removed from the pump/tray connector and replaced with
a sorbent tube tray, “loaded” with four sorbent tubes according to
the procedures described previously. Care should be taken to
234
-------
CURRENT METER
F LJ
Ac DC
ROTAMETER
PRESSURE GAUGE
I DRY GAS
I 101 515j41
[ METER
TEMPERATURE. °C
/ NUTECH CORPORATiON
P4U1tCH AIR SAMPLER
Figure C-29.
“Calibration” tube tray attached to
Nutech Air Sampler az2d bubble meter.
“CALl BRATION”
TUBE TRAY
235
-------
Figure C-30.
AjTi ierit air collection system ready for sample
collection.
CURRENT_METER
n ii
0
PRESSURE GAUGE
AC DC
ROTAMETER
DR ’ GAS
IoTs1 14
[ METER
0
ThMPERATURL CC
/ NuTECHCDRPORATI ON /
NuTECH Al R SAMPLER
236
-------
assure that all connections are secure by leak—checking the
sampling system by the above procedure.
Now the Ambient Air Collection System is ready for sample collec-
tion. First, the sampler should be placed in an appropriate
location, such that the sorbent tube tray is not exposed to the
exhaust of the Nutech Air Sampler. Next, the inlet plug of the
tube tray should be removed and a small amount of glass wool
inserted in this inlet to serve as a particulate filter After
recording the initial volume of the dry gas meter, the ambient
air sampler should be turned “on” by switching first the AC
switch, then the DC switch to their “up” positions. The time of
sample initiation and all conditions of the Nutech Air Sampler
should be recorded. Approximately every other hour after sampling
is initiated, the conditions of the Nutech Air Sampler should again
be recorded. Exhaust flow rates should be determined at least
every other hour. This will allow the approximate flow rate to be
averaged over the sampling time to obtain a sample volume which
can be compared to the sample volume measured by the dry gas meter.
The only other disturbance to the sampling system during sample
collection, should be sample spiking, if required for the sample
by the sampling plan. Both the techniques of sample spiking and
the development of a sampling plan will be discussed later in this
manual.
After sampling iscomplete (i.e., after 8 hrs.), the Nutech Air
Sampler should be switched “off” and the final volume of the dry
gas meter recorded. Then the glass wool plug should be removed
from the sorbent tube tray inlet and stored in a glass vial The
tray inlet should be plugged, the System quickly leak-checked, and
the results recorded. The pump/tray connector should be removed
from the sorbent tube tray, and the outlet of the tube tray plugged.
The sorbent tube tray should be disassembled in the opposite manner
of assembly with the sorbent tubes removed, capped, and placed in
their appropriate culture tubes. Again, this should be done as
quickly as possible to prevent additional sorbent contamination and
to prevent loss of sample due to diffusion.
The flow rate through the Nutech Air Sampler should again be
checked after sampling and after securing the sorbent tubes. This
should be done by reconnecting the “calibration” tube tray to the
Nutech Air sampler, quickly leak-checking and determining the flow
rate at the “calibration” tray inlet as previously described. This
should be determined wi.th at least three trials, and all flow rates
(inlet, exhaust, and rotaineter) and conditions recorded. This will
allow a direct comparison to the conditions originally determined
prior to sample collection.
Quality Control Considerations
In order to assure that data obtained from sorbent samples are
reasonable and valid, various aspects of quaaity control/quality
237
-------
assurance (QC/QA) must be maintained. One aspect of QC/QA which
should be addressed is the demonstration that the laboratory,
which will be performing the sampling and analyses of sorbent sam-
ples, is competent and follows accepted good laboratory practices.
(A more extensive discussion of quality control procedures and
considerations is presented in a special MRC-EASC publication [ 133].)
To meet the demands for QC/QA of samples for numerous environmental
and industrial hygiene programs, such a laboratory should develop
and continuously improve a good laboratory practice program.
Another aspect of QC/QA involves testing the sorbent saznpling/
analytical method presented by various quality control measures.
This should be done to practically demonstrate the validity of
the method and of the samples. These control measures include
evaluating standards, blanks, replicate, splits, backups, and spikes.
Standard solutions of mixtures of compounds of interest (to this
project) or other compounds for instrument evaluation/calibration
should be analyzed to allow quantitation of compounds present in
samples to generate calibration curves, to check detector linearity,
and to assure that the analytical instrument and method are func-
tioning properly. “Blanks” are sorbent tubes that are treated in
a manner identical to that used for the sorbent tubes on which
samples are collected except that no sample is drawn. Such “blanks”
can indicate problems with contamination (e.g., from transport)
and/or sorbent degradation. By collecting multiple samples on a
number of tubes under identical conditions, replicate samples are
collected which can provide data for determining the precision of
the sampling/analytical method. When thermally desorbing the sorbent
tubes.for analysis, the tube effluent.can e split and recollected
for subsequent analyses using the same or different analytical method.
Such split samples can be used to verify analytical precision or
to validate analytical results by analyzing them under different
conditions or using columns with different liquid phases. To
evaluate the overall sampling/analytical method, backup sorbent
tubes, and laboratory and field. spikes are analyzed. A backup
sorbent tube (Ambersorb XE-340) is placed in the fourth position
during sampling to collect compounds “breaking through” the first
three sorbent tubes. The compounds analyzed on this backup tube
are, therefore, the compounds which were not quantitatively retained
by the three sorbent system. A spike is a sorbent tube with a known
quantity of the compound(s) of interest loaded on it. If the loading
is done in a laboratory after sample collection or on a “blank” sor-
bent tube, that tube is called a “laboratory” spike. If the loading
is done before or during sample collection, that tube is termed a
“field” spike. Laboratory spikes can indicate sample stability and
lifetime to some extent, but they are mainly used to determine
desorption-efficiencies. Whether or not the compound of interests
[ 133) W.M. Haynes, D.M. Haile, D. H. Toy, and W.M. Mess, “Guide
to Good Laboratory Practices”, the Environmental Analytical
Sciences Center, NRC, tayton, Ohio, March 197g.
238
-------
is altered during sampling or lost for another reason should be
indicated by the recoveries obtained for field spikes.
There, are numerous procedures by which laboratory and field spikes
can be produced. One technique is to simply “inject” a small
quantity of an analytical standard solution into the “inlet” end
of a sorbent tube or the inlet of a tube tray. Although this
technique is very simple and cost-effective, problems can arise
due to the solvent interfering with the analyses or changing the
sorbent retention characteristics. Another technique is to
generate a vaporous standard mixture and “load” a small amount
of this mixture into the “inlet” end of a sorbent tube. Usually
this spiking technique requires special equipment (see Appendix B)
that is more easily operated in a laboratory. Under these con-
ditions al]. spikes must be performed prior to field sampling.
Whatever technique is used should be quickly evaluated in the
laboratOrY prior to preparing a fieldsampling trip.
For this project each air sample is collected in a sorberit tube
tray containing three or four sorbent tubes. The analysis times
for these sorbent tubes ranges from 1 to 3 hours. depending upon
the sorbent type; the amount of material adsorbed, and the analyt-
•jcal technique. Therefore, there is a great deal of time and cost
involved in the analysis of each air sample. .n extensive quality
control program, using a large number of sample replicates, spikes,
splits, etc . , would be.. extrern ly costly. and,..,therefore,, impractical.
However, at least some quality control samples need to be collected
and analyzed in order to demonstrate the validity of sample data,
even if not enough are collected to specify the accuracy and pre-
cision of the sampling/analytical method. The best approach is
‘to devise a sampling plan specifically designed for a particular
sampling program, such that there are. at least one set of blanks,
a few splits, spikes. backups, replicates, and calibration/
quantitatiofl standards.
The sampling plafl which is devised for a field sampling program
should describe the following information for each sorbent tube:
• Tube number and sorbent type.
• purpose of the sorbent tube (blank, sample, replicate
sample, lab transport spike, sample spike, field
transport spikes, backup, etc.).
• The position of the tubes in a tube tray (if any).
• The sampling conditions, (if any), including date, time,
sample volume and flow rate.
The pump and pump conditions, if sampling is done.
239
-------
• Any additional, pertinent information, such as the identity
arid amount of spikes, as well as method and time of spiking,
or additional tube purpose.
An example of such asainpling plan is given in Table C -4. The
information given in this table should also be marked on the out-
side of each of the sorbents’ culture tubes to prevent switching
of sorbent tubes. Grouping tubes together according to purpose
during packaging will also help to keep the sorbent tubes from
getting mixed-up.
Storage, Transportation, and Contamination
This part of this manual briefly reviews and summarizes informa-
tion on sorberit tube storage, transportation, and protection from
extraneous contamination given previously in a number of different
sections. Generally, there are five phases in using sorbent tubes
to collect ambient air samples. The first phase is the preparation
of the sorbent tubes for sample collection, including preparation
of some sorbent tubes as quality control samples; second is the
protective storage arid packaging of the sorberit tubes, grouped
according to the sampling scheme, for safe and rapid transport to
the field sampling location; third is the collection of air samples
at the field location, including preparation of additional quality
control samples; fourth is the protective storage and packaging of
the sorbent tubes for safe and rapid transport back to the labora-
tory for analysis; and fifth is the analysis of the field and
quality control samples. At each stage of preparation, storage,
and transport, the sorbent tubes need to be protected from
contamination and other adverse phenomena (e.g., breakage,
sunlight, excessive beat, etc.) which could affect the materials
adsorbed on the tubes and, hence, the validity of their data.
In review, to prevent the contamination of the sorbent tubes with
extraneous materials during their preparation, the following guide-
lines should be observed.
1) All glassware and other hardware materials used during
sorbent tube preparation should be meticulously cleaned
prior to preparing the sorbent tubes. -
2) Hardware and other materials which will be used to
prepare the sorbent tubes should never be handled with
bare bands. Protective gloves an /or clean utensils
should be used to handle these materials.
3) tlltra-high purity solvents should be used during the
soxhiet extraction of sorberit materials and teflon tape
on sleeves (not silicone grease) should be used to seal
the glass joints of the extraction glassware.
240
-------
TABLE C-4. FIELD SAMPLING PLAN
Tube I Sorbent typp
40 Tenar
SO Por pak B
63 Aabetsorb
60 Aabersorb
43 Tenax
46 Porapak B
56 abersorb
38 Tenak
55 Porapak B
61 AMersoib
58 AMersotb
36 Tenax
52 Porapak B
66 Asbersoib
41 Tenax
49 Porapak R
62 sbersorb
57 Asberaorb
39 Tenax
47 Pórapek B
53 Asbersorb
37 Ten*x
45 Tenax
51 Porapek I
46 Porapak I
64 Asbersorb
64 Mbersoxb
Sample II
Sample I I
Sample 61
Backup-Sample Il
1ank
Blank
Blank
Calibration
Cal ibrat ion
CalibratIon
Calibration
Transport spike
Transport spike
Transport sp&k*
Sample 44 — spike
Sample 14 — spike
Sasple 44 — spike
Backup-aesple *4 spike
Transport spike
Transport spike
Transport spike
Era — ‘Blank”
Xtra - ‘Blank’
Xtra - ‘Blank’
Era — Blank
Xtr C -
Xtra — Blarak”
‘.1100 1. ‘.2.3 1./sin Field spike. method B
spike *2 , spike 15 • n
after start1i q LamphOg
- Field spike. method A
- — spike 82. spike
— — ‘.3:00 P I t 4/14/80
To replace any other
tubes broken (except
lab transport spikes)
Tray
Purpose position Date
__ ._ ._ 9_. _ Flow rate Other infoimation
I
2 4/14 ‘ . —‘.4Pfl ‘.1100 1. ‘.2.3 1./sin.
3
4
k)
I
2
3 ,
4
I
2
3
4
— ‘.2.3 1./am
4/14 & 4/15
4/15
lab spike, method A
spike *2. spiked
11:30 Alt, 4/11/80
-------
4) A charcoal trap should be used in the vacuum line of a
vacuum oven during the vacuum drying of the sorbent
materials
5) The environment to which the sorbent materials, or tubes
are exposed should be as clean and dry as possible.
6) During the thermal conditioning of the sorbent tubes,
the gas stream should be pre-purified by passing it
through a drying trap and a charcoal trap.
7) After thermal conditioning, the sorbent tubes should be
quickly capped and stored.
8) If a sorbent tube is to be spiked, the spiking solution
or vaporous mixture should be carefully and “freshly”
prepared, the spiking materials meticulously cleaned, and
the spiking performed as quickly as possible.
After preparation, the sorbent tubes should be stored tightly capped
with metal fixings and viton 0-rings, and contained within culture
tubes. They should be used to collect air samples within one week
after thermal conditioning. This insures sorbent background is
at a minimal level during sample collection. In addition, all
the sorbent tubes, stored the same way as described above, should
then be analyzed at least within one month, preferably within two
weeks, after sample collection. While being stored or transported
the capped and contained sorbent tubes should be kept in an environ-
rnent as clean and dry as is practical. Furthermore, they should not
be exposed to excessively hot temperatures, which would induce
sorbent degradation and/or in-situ sample reactions, or to exces-
sively cold temperatures, which could cause a change in sorbent
properties and/or condense water out of the collected samples. A
cool environment, at approximately ambient temperatures (15°C to
25DC), would be preferable. The sorbent tubes should also be
protected from light, especially sunlight, during storage and
transport to prevent light-induced reactions or degradations. To
help protect the sorbent tubes from breakage during transport,
clean glass wool can be wrapped around the sorbent tubes within
the culture tubes. These culture tubes can be labeled and grouped
according to the field sampling plan, and packaged with protective
packing materials (e.g., styrofoam, bubble-sheets, etc.) inside a
box or small suitcase. Finally, all paperwork associated with the
transport of the sorbent tubes, sampling systems, and associated
supplied should be prepared in advance of a field sampling trip.
During sample collection, certain quidelines, which are similar
to the ones described for sorbent tube preparation, should be
followed to prevent the contamination of the sorbent tubes with
extraneous materials. These guidlines are listed below.
242
-------
1) AU. hardware materials used during sample collection
should be meticulously cleaned prior to field sampling.
2) ProtectiVe gloves and/or clean utensils should be used
when handling the sorbent tubes during field sampling.
Furthermore, extra care should be taken not to touch or
otherwise contaminate the sorbent tube ends.
3) The sorbent tube tray should be assembled quickly,
with the sorbent tubes exposed to the atmosphere as
little as possible.
4) AU. fittings should be securely tightened with wrenches.
They should be leak-checked according to the procedure
previously described, not with a leak-detection fluid
since this can contaminate the sorbent tubes.
5) 11 a sorbent tube is to be spiked, with a spiking solu-
tion, the solution should be carefully and “freshly”
prepared, the spiking materials meticulously cleaned,
and the spiking performed as quickly as possible.
6) The tube tray lid should be on the tube tray during
sampling to protect the sorbent tubes from sunlight.
7) If possible, the..Sorbent tue tray should be kept. out,
of thot” locations during sampling. For example,
sampling should be done in a shaded location rather than
on asphalt or concrete in hot, direct sunlight
8) sampling should be. performed in a location away from
sources of atmospheric contamination (e.g., cigarette
smoke, fumes from automobile and truck exhausts, flowers,
etc.), if they are not of interest.
Records
As mentioned numerous times in this manual, it is necessary to
keep accurate, detailed records of all aspects of sorbent sampling.
These records should include information on the history of each
sorbent tube, the sampling schemes for each sampling program, all
sampling conditions and flow rates, the sorbent tube storage and
transportation conditions, the log in and transfer of samples, and
the analytical methods, conditions, and results. All of these
records help to’ insure the accuracy of the final results obtained
and/or to identify the cause of any problems or discrepancies
observed during sampling or in the analytical results.
The purpose o.f keeping histories for each sorbent tube is to allow
the quick review of how and when the tube was prepared, and how,
when, and what types of samples to which the sorbent tube was
exposed after . preparation. TMs allows variations in sampling
243
-------
results to be traced to variations in the histories of the sorbent
tubes used in a particular sampling program. Furthermore, accurate
histories help to prevent sorbent tubes from getting “mixed-up”,
from being stored for unreasonably long periods of time (e.g.,
5 years), or from being used an unreasonable number of times (e.g.,
1000 times) for collecting samples. The information that should
be recorded in a sorbent log-book while preparing sorbent tubes
was given previously Figure C-5, in this manual. Such records
should be kept for each sorbent tube. n addition, information
such as shown in Figure C-31 should also be recorded in the sorbent
log book for each sorbent tube each time it is used to collect a
sample. Although not all the information shown may be available
for all tubes and all samples, the sorbent log book should be kept
as complete as possible.
For field sampling, record-keeping begins with the development of
a sampling plan, as discussed previously. This sampling plan
is used as a master scheme to coordinate sampling and sampling
information. Information on all sampling conditions, including
weather, flow rate, and sorbent cartridges, should be recorded
on a sampling fact sheet for each sample specified in the sampling
plan. An example of such a sampling fact sheet which should be com-
pleted when collecting a sample with the ambient air collection
system is depicted in Figure C—32. A fact sheet should also be
completed for each set of quality control samples. Keeping such
detailed information during a field sampling program can help to
define problems with the sampling technique and/or method. Informa-
tion is also recorded in the fact sheets regarding sorbent tube
storage, transportation, and transfer for analysis.
Once field sampling is completed and the sorbent tubes are received
at a laboratory for analysis, these samples should be logged in
through the laboratory’s sample log-in system for proper transfer
to analytical personnel. As part of that laboratory’s quality
control/quality assurance program, this sample log-in system should
allow the easy review of the chain—of-custody (record of individ-
ual(s) charged with responsibility for these samples) of a set of
samples and ready coordination of sorbent tube preparation, field
sampling, and analytical information and reports.
Reconditioning Sorbent Cartri4 es
After collecting and analy2ing a set of ambient air samples, -the
sorbent tubes can be reconditioned for subsequent sampling programs.
This reconditioning can be done in one of two ways. The first
method is simply to thermally condition the sorbent tubes as des-
cribed in previously for the original preparation of these car-
tridges for sampling. This should be done soon after the sorbent
air samples are analyzed in order to remove, as quickly as possible,
any compounds from the sorbent tubes not removed during these
analyses. This helps to insure that any compounds of the air sample
still remaining on the sorbent tubes wil]. not irreversibly adsorb
on or cause the degradation of the sorbent materials. ! owever,
244
-------
“ Sampling Lifetime ” Name:
Tube #: Sorbent Type: Date Packed:
Last Conditioning Procedure:
Date:
Storage procedure:
Storage Dates:
Transportation Procedure:
Packaging Procedure:
Lab to Field Dates: Field to Lab Dates:
Sampling Conditions
sampling Date: Sampling Fact Sheet #:
Type of Sample: Lab Blank Field Blank Lab Spike
Backup Transport Spike Field Sample Spike
Field Sample
Log In Date: Log In *:
Analysis Request #: Analysis Request Date:
Analysis Procedure’:
Analysis Dates Name.
Analytical Results (briefly)
Analytical Results Reference (NB. Pg. or Report #):
Date Sample Log Closed:
Sorbent cartridge Fat
Storage Procedure: Dates:
Reconditioning Procedure:
Reconditioning Date:
Log Book Reference Pg:
Date Sorbent Tube Unpacked: Comment:
Recycled? Sorbent: Glass Cartridge:
Log Book Pg:
Figure C- .31. Information which should be recorded in a sorbent
logbook.
245
-------
AMBIENT AIR COLLECTION SYSTEM:
SAMPLING FACT SHEET
Date of Sarnpl rig: Name(s):
Location (City, State, Street, Etc.):
Atmospheric Conditions:
TYPE OF SAMPLE Lab Blank Field Blank Lab Spike
Transport Spike Field Sample Spike Field Sample Sample Backup
Spike Backup Calibration
SORBENT CARTR IDGES
Condition of Tray
Tube * Sorbent Type Cartridge Position Comments
IF SPIKE : Date Spiked: Notebook Page:
Procedure:
Amounts & Names of Compounds:
PUMP CONDITIONS Power: AC DC No. of Valve Turns:
Date of Last Dry Gas Meter Calibration:
Initial Time: Final Time:
Initial Dry Gas Volume: Final Dry Gas Volume:
Figure C-32. Sampling fact s1 eet.
246
-------
Flow Rate Observed for System Calibration and Sample Monitoring
Time for wL of flow, mm Ave. Flow Hotameter Pressure Power Temperature. Dry Gas
Setup Time Trial 1 Trial 2 Trial 3 Rate, mi./min Reading. ( .1mm Gauge, in 1120 Gauge. Amps Neter, ai 3
Figure C-32. (continued)
-------
EVALUATION OF FLOW MEASUREMENTS
Estimated
Sample Volu
me,
Location of
Flow Measurement
Flow
M .
Rates
Ave.
mLjmin
Lo
Sampling
Time, mm
liters
Hi Ave.
Lo
TRANSPORTATION AND STORAGE
Storage Procedures,/Conditiofls:
Transportation Procedures/Conditions:
Packaging Procedures/Conditions:
Lab to Field Dates: Field to Lab Dates:
SANPLE TRANSFER
Log-In nate: Log-In *(s):
Analysis Request : Analysis Request t)ate:
Reguestor: Sent to:
Figure C-32 (continued)
248
-------
after the sorbent tubes are thermally reconditioned, .a set of blanks
should be analyzed to determine if the background of the sorbent
tubes is “acceptable”. To be “acceptable’ t , these backgrounds
should be the same types and levels as determined for the first
air sample quality control blanks. If the reconditioned sorbent
tubes are found to be “acceptable”, they should be used for col-
lecting another set of air samples within one week of being
thermally conditioned, or be thermally conditioned again immediately
prior to the next sampling program.
If the sorbent tube backgrounds are found to be unacceptable, or
if these tubes have been used to collect a number of air samples
(%5), then a second, more rigorous, reconditioning method should
be used to rejuvenate the sorbent materials. By this method,
after the sorbent tube air samples have been analyzed, the sorbent
tubes should be “unpacked”. The glass wool plugs retaining the
sorbent materials within the sorbent tubes should be removed and
discarded. The sorbent materials should be “poured” into clean
glass jars, keeping the different sorbent materials separate.
The sorbent materials should be sieved (60 mesh for Tenax and
Antbersorb, 80 mesh for Porapak R) to remove fine particles. Then
the sorbents should undergo the extraction/drying procedures
given previously for cleaning “new” sorbent materials. After
cleaning and re-deactivating the glass cartridges (now empty)
by the method given previously (discarding and replacing car-
tridges with badly broken ends), the sorbent materials are then
“repacked” in the glass cartridges. These “re-packed” sorbent
tubes should receive a new page in the sorbent log-book, with
reference back to the original sorbent tubes, and should be
indicated as “second generation” sorbent tubes. Within one week
of the next sampling program these sorbent tubes should be
thermally conditioned. The hardware materials for the storage of
these sorbent tubes should also be re-cleaned according to the
procedures given previously. (Hardware used within the sorbent
tube tray or for connecting the pump and tube tray should be re-
cleaned between each sampling program.)
Since each sorbent tube has a definite “lifetime” after which
its capacity for retaining compounds diminishes or its background
level becomes excessively high, limits must be placed on the
number of times a sorbent can be used. For this project, a sorbent
tube (with “acceptable” background) may be used to collect up to
five air samples or be thermally conditioned up to fifteen times
(whichever comes first), before being re—extracted or discarded.
The sorbent materials may be re-extracted twice (i.e., three
sorbent tube “generations”) for rejuvenating these materials.
This allows a total up to fifteen air samples, forty-five thermal
conditionings, and three extractions for a lot of sorbent material.
249
-------
Flow Chart II
The 1ow chart given in Figure C-33 summarizes the procedure for
preparing the Ambient Air System, for collecting air samples,
and for reconditioning sorbent tubes.
250
-------
hO. PRLPAAATIIN. CALI1ISATIE , 5101 OPLOATISU Of 100 AIIIIIT 0 (0 COLLECTiON SVSTON
PILl ININEJY PUE?WTSA. CON FIE LD 5*001.100
j _ f 0001 Iu . COON? I I I) CLEAN. 50Cr. I ) CEEWA IN * ONATED 23
a—.t 51(001 Wi, Pflfl ULT50 50IIIC SA m
51*15 51AIL FOIlS VITO LOI.tJ lI
I ) CALINOATE ANT GA S 54 (.00 5 LIWTSON I ) If .C NO 001014.100 SUIPL lEa 4-S 5*5
001111 500 SOTANGTEO 15*1 50* 5 (SPOCTW 00*ING I I
4001.7 ‘3 Of W’TYCU I ‘T0* ISZPGST. PU? I 2) 01031 SITU l*P 2)
A AI 5 1 Ll 0 I CLLAN 5 1 .1 .53 5004. 4 5 * 5 0 0
I AOIN*O.I 1 5lSf S 3) P1 -Aol 505(4.
I I C51.11*t UlS15. PIOCES IN 000U .. 3 ) lINT * - 1Z5C 21
I 4 PLASTIC G*PPHA. PA. t
3) 5*1111.3 IlUCI I O U 2) ClOOl LID 000 3) fACE V TONOPOUT 00*11 VIALS ION 4
P*0I ’LN 3 11151100 15101 NI* DS USE AZ PA ST ICuI .*1T 4) lAG. Vol IUAIISPOU T
________ I Pu01S )
SI ( JOEL 3, s* oW 3) (lUlL 05* lOGO 4 4)
PAID ViA SUANSPOUT 11* TUSOS000T 3) 1*004. VI 51% *00
PIGS FOP TOAIISPONT
04
4)
fl
4 4
*5511014.1 111*/TUT 3) OSThIN ON FuEl Alit *0
*00 50 1 1EV? 11*1 51.0011501 1500 01
STAINLESS STtEL/1tfI AhI0UI’I iIAIt 0113551005
CC 1 0 IEC T IONS 4
4 2) A.IU. 004.15 Of COllECT
?I51ITLA £CCI10ELV Sill IN APPNOEAIAT,.
WIN WRIIICIO.S LOCATIONS
CL&III ’CONIIIC?OUN *1111 3) IS5TA4 I . TI RflAINiI51
511151 FI l lINGS IN A CLIPS
0051511 IILT%ASONIC 0*70 II
SITU DITTOGONT S-N 00$. 4) NIPO AU. PANTI CAZAII
111031’ WIN A
TWOIIISILT WIN NO.. *4.511 1 TO 5355 10
Ul I Si-PI01IT Y 00(01 aNEW ooxsi 00 LAO
lOON i S V AT .I2(I.PI [ _ ’Ti ’j IUL ON FITTINGS,
S-lA 713.
TO TOWN lIST 4 (01 NOT
CAP OS P 500 *1L
ITIIIINAT5II)
•.I15 ) PI lISbON NW
hAd C( 501C7A.3 II Cf OIIIXIILI.D FITTINGS.
T5*C 1UAV41I 005 _________
TA Il S POP 1
7) pIn LID A. TIlT
14111 0101 P TOO
FITTINGS NO) TIIUV NAY NW
00T513 50I .AGICAI. 5003 51)11
NOTCCII LIP IWIAO SA. Nf .IIT p.*is .*ss . T S TEFLON 15*100 111* 101010 CLIPS
000300100 0L03 1ITW07
II CLEAN 0053 CALI500TE
405 P1. 5010113 AL l.
U lTt i i SC,ACIC*L It 3..
T 115P*3NITES3 0003
£1111 005500)15 : (e.g
O.INAI.UETYN) (Q OI I I IS35t
2) PACE FOP TOMISPOWT
‘5)
(‘I
1-
70*0510117 INTO 11 ( 1 .0
I) FOCI ILL TOOLS *10) SUPPLIES 4(14 1? FOUI3ET GLOVES)
INTO 1101 OS A ITS 001905 SUITCASES.
2) P011*00 PACKING LIST 0l Ii C1 TO IN SIMS ALL
111115 011 IECIEII. —
33 ISOEN NAT AL!. IDCESSAIT 10*314. AMAISIMOIITS *00
hA ITI NONE 1 1*51 UCLA CO I15LETLD.
4) ?001ISPOUT 1011110 LOIIItING LOCATIA.
SI 10511515*7 TO FIELD 501051110 SITE
Figure C-33. Flow Chart II.
-------
rxtw r — Afl CA. 1 kA l04
2 ) *2 tOoLs NC r npLxts
2 *TT* P10’tLOY C O CTO to 40fl ?t3 NC LZJ.5 TV
2 NC 5 VJU r1t 5N60 255 7153 04L21 153” 0010
4) A0T& CA L1 £At153 5 101 T1A TV C03 ) PC DO 71C
P’V T!AT C53*I2 TVI, LIA CT .. NC W4 VL ThAt I5I2.V
31 .3 2
V53O53 3202101 5555 5551531.20 70 55C P 20 W*2t
5) 003.55 2LC5 555101.20 535 (SC, DC), 1 ( 2Z VA2YD 0P71 FULL
5 75 noFULLO 7tJ52 4 ___
o .005501 0551.1 I0AT 501552.2 10112 10 •CALIWt1 0010
03.M 5551.27 5510 10550052 ?‘L At 1.15507 3 02101 INC 522052)
I) 10V01Z 11.2. 2055355 00110 527222 5 3.1* 5551531.2*, 1SC1. l255lI2
t cu5k2*7. ?52*5U551 0507 5552 22 *55120
5) TU*1 3070101 1551531.05
30) SD52YI C55L5 51552225” YUlE 052” 15535 37153/1120 C !T!0P
3.52 31.32 50115 S OF 110 11520
rzMA1. 1352.2 ? A55552153
I 1.010 210 IC5* 1010 ThAT 5522 ? ‘7*110 ’ 505 )101 15*21
*10 ?5 rD TI s SWIM 05 UIt)
3) *12*25 110 3550 ,1553.! C7 TV $010101 7010 115*7
55101.2553101023. 4
3) 120552 1551113.2. 5 5* 0 5 1 023 )0 3)2 (0105
4) POT 110 1*1531.212 *151535 III 55$ 3*5*115570 0355015*
0) 0310 110 *5*1107 1052 1120 2?Lrl
e ) *5332c’t 05.101 W20 1 . 35. 552 7*5* 10350311411 08552. 5552
1555057 20 2153 50100500 YUlE 1*557 255227
L 1*501.2
1) 7012 5532101 5*1535.05 5512 10Cm 11*0
2) *105*2 02 55.1*1 0100 iA?35*
3) 5512 3225*0 *2.2. C53 !T55*1 0 ? 110 55772201 5515 155101.20
4) 10555232 fl ,5* 5552! 3.0 5212? *01*353205 55505 *200*0 3312 551111
*55101.110 SS 105555052. 0OVsy *00 5 551705 1*3535.1552 15 1?5Zt52 512
350 005110 1 2.1l53. 122052 AUtt UT! 10550
p ‘ Y1* ’— .4plV1 5511* 110 101.2121531 *31
3*00205*2 131251820 ) 710
1*5531.110 P25555 40 110 001*220
1*1535.1552 2 10. 1? UQV1* .
*1
1) LIMO 110 SWL.1152 *00110 *LATIVZ5.2 ’ 5501*1525320. 55175 2*552 LID
4) k? 5350 03 1 .55105.112, 13*5 210 5001.01 P% ‘oyp• sjc, *12052
Figure C-33. (continued)
252
-------
[ .sMo.t.nIc AC!IVI?)U
1) l. VO O lJAl OL PU TI 1014110 TOll 01*1 IEIT All PtAC!
ip A1PIOPAi*l1 O1.AZ* ViAL .
a) PLil )l0t 11 $014 17 UU OIlY A .413
3) WUVO Pt3OtllAT CIO)ZCTQP TO 1044110 lUll 11*1. 1101
Till aft lZT All CA? Ill C 1CT
4) Ti-lOAD” $013117 lull 0*40 II I $T & 1 1.411 a
ppiOpliAfl CUUU*Z TOILS
4) PUll )11D 01*0 WUILAOI All 1$? C PLCTOSS
I) IJC OID ALl. P1 *10 £AllI.2I0 0)00041 AT 10 04
1) ATTACO r$0YT1AY CDIWCC7OI TO A1U40004 TOIL 71*1 , l I lA
*111 IDIOt ! AY 1*1 . 40 PL$0
U) 1)
4) UCO ALL. C041101041 Of Tll $00411 All $*IGLD. IPCLJllI$0
AC 1*4010, ,USSU*1 0401, All ?101 1*041
10) 470401 CAL OIIATW *1hI0D00Z1 TO •CALXI&I? ) OW. 110 )1 11*1
WIlY All IO.AOU*E 0 fL OW $401 AT LU4T 01111 71101 gAll
4
L I) 1)11 $00111 11L11 ‘OfT.
U) 11011 7110 fOllY 0011*10101 PlOW •CALOU*?104 ’ 1UI1 014 !
All 010 71*4. CAP O0C5OW 101 All Pull 110 0 1* 1
r
13) ._. ALL UlWTUUS 11 010 101410 1111 01*1 *10 100 04 LI I.
14) AU All llflLiU.
LI) $0 TAXW ALL 1*101110 UC I AU 0L.l01I
LI) TIA4U T 1)11.0 lAllIOll 4)71.
Figure C-33. (continued)
TO 1*10 101041
1) TOAJISPOU? IL ). 1041$ All $0171210 $411 00 110
1 *0 — I l
1 )1 114111 IN 01* Pl 0LT TO UAL701CAL P1 012.
A) 0401114 I 1.llIi 1*00100111, 71)70 11 UCTIWY
10171 1* 101110 7)27 111010
253
-------