Anal. Chem. 1990, 62, 1265-1271
1265
Membrane Mass Spectrometry for the Direct Trace Analysis of Volatile Organic Compounds in Air and Water M a r k A. LaPack,*JV* J a m e s C. Tou,’ a n d Christie G . Enke2
Analytical Sciences Laboratory, T h e Dow Chemical Company, Midland, Michigan 48667, and Department of Chemistry, Michigan State University, East Lansing, Michigan 48824
Very slmple membrane inlets can be constructed that are selecllvely permeable to organlc molecules while discriminating agalnst the air or water sample matrix. The membrane thickness dkectty affects the flow through the membrana such that thlnner membranes provide hlgher throughputs and shorter response times. Higher temperatures result in shortened response times and reduced selectlvlty of the permeation process. Comparisons of two hdlow fiber membrane Inlet geometries show that permeallon rates are improved In the configuration where the sample flows through the lnslde of the hollow flber and the permeating gases are analyzed on the outslde, as opposed to the reversed conflg uration.
INTRODUCTION In recent years, the need for real-time trace analysis of air and water streams has increased. Methods are being sought for the in situ analysis of process streams for process optimization and control. The challenge of environmental monitoring of air and wastewater emissions grows with increasingly stringent regulations. In many cases, the extraction, concentration, desorption, and analysis of organic compounds from air or water can be accomplished in a single step by using a membrane sampling device. Membranes have been used as molecular separators in gas chromatography-mass spectrometry (GC-MS) to reduce the amount of helium carrier gas flowing into the mass spectrometer. The organic analytes are enriched in the analyzer. Llewellen and Littlejohn (I)have used a flat silicone membrane that is selectively permeable to organic molecules over helium. The GC effluent is flowed across one surface of the membrane, through which the organic molecules permeate into the analyzer while the helium is largely rejected. Lipsky et al. (2) have used a heated Teflon tube to act as a molecular separator interface for GC-MS. In this case, the helium preferentially permeates the porous Teflon membrane, resulting in an enrichment in the organic analytes flowing into the analyzer as the reject stream. Westover et al. (3) evaluated various hollow fiber membrane materials for MS analysis of organic compounds in air and water without GC separations. The membrane was the direct interface between the sample and the analyzer, with the inside of the hollow fiber exposed to the MS vacuum and the outer surface exposed to the sample. This particular configuration will hererafter be termed the “flow-over” hollow fiber inlet. Cooks and co-workers (4) reversed the configuration such that the sample flows through the inner volume of the hollow fiber while the outer surface is exposed to the MS vacuum. Hence, “flow-through” will be the terminology describing this configuration. There are advantages to analyzing the sample directly without chromatographic separation. In cases where interferences between analytes are minimal or can be acIThe Dow Chemical Co. *MichiganState University. 0003-2700/90/0382-1265$02.50/0
counted for, the analysis or screening of batch samples can be shortened (5). In addition, transient processes can be much better studied by a direct and continuous analysis (6). Much of the literature on the subject of sampling with membranes is devoted to useful applications. In this study, generalized forms of the permeation rate equation for analytical applications to both air and water samples are developed. The results and a discussion of the effects of various experimental parameters on permeation rates are presented to serve as a guide for membrane sampling. In general, membrane processes are composed of the feed stream (sample), the reject stream (waste or vent), and the permeate stream (sample extract). The permeate stream is enriched in organic molecules due to the selective permeation properties of the membrane. The permeation of a substance through a membrane is defined by three processes: (1) selective partitioning of the substance into the membrane polymer matrix, (2) selective diffusion of the substance through the membrane, and (3) desorption of the substance from the membrane into the MS vacuum. Diffusion through the membrane is assumed to be the rate-determining process, while partitioning a t the highpressure surface and desorption from the low-pressure surface are considered to be instantaneous. The sensitivity of a membrane separation technique is determined by the steady-state permeation response, while the non-steady-state permeation characteristics of the analyte in the membrane determine the response time. EXPERIMENTAL SECTION Chemicals. Aqueous solutions were prepared in deionized water with reagents from Scientific Services and with reagentgrade solvents from Fisher Scientific. Gas samples were prepared in a 100-L Saran bag filled with prepurified nitrogen from Scott Specialty Gases. Apparatus. The mass spectrometer used in this study is a Balzers QMG 511 quadrupole instrument with unit resolution and a 1-1023 mass range. A gas-tight ion source was used with analyzer pressures maintained below 3 X lO-’ Torr while sampling. The MS base pressure was 2 X lo4 Torr. Data acquisition was controlled by a DEC PDP 11/73 based computer system. The hollow fiber membranes are Silastic medical-grade tubing from Dow Corning Corp. The membrane material is a poly(dimethylsiioxane) elastomer. The sampling devices were constructed from hollow fiber tubes of two different radial dimensions. One inlet contains a 2.5-cm-long,0.0305-cm-id., 0.0635-cm-0.d. hollow fiber. The second inlet contains a 2.5-cm-long, 0.147-cm-i.d., 0.196-cm-0.d. hollow fiber. As discussed above, two basic hollow fiber sampling geometries have been reported in the literature, flow-over and flow-through. For performance comparisons, each inlet was constructed from 2.5cm lengths of 0.0305cm-i.d., 0.0635-cm-0.d. hollow fibers. The simple construction of the two inlets is shown in Figure 1. The flow-over inlet is made by using Dow Corning RTV 734 Silastic sealant to close one end of the hollow fiber and to attach the other end to a ‘/*-in.-o.d. (0.32-cm-0.d.)stainless steel tube connected to the MS ion source. The flow-through inlet is made by sealing the ends of the hollow fiber into two legs of a tubing-fitting tee. The third leg of the tee is connected to the MS ion source via a 1/8-in.-o.d.(0.32-cm-0.d.) stainless steel tube. Samples are drawn 0 1990 American Chemical Society
1266
ANALYTICAL CHEMISTRY, VOL. 62, NO. 13, JULY 1, 1990 1oooO.03 Hollow Rber Membrane
@
4
1oO0.00-
.ri
d
::
100.00-
Id c, .rl
-2 n
Feed Stream
8 3 3 2
low-
1.00-
0.10-
1
Sealant
Feed Stream
Flgure 1. Two hollow fiber inlet geometries. Fbwover geometry (top): the feed or sample stream flows across the outer surface of the membrane, and the inner surface is exposed to the MS vacuum. Flow-through geometry (bottom): the sample flows across the inner surface of the membrane while the outer surface is exposed to the MS vacuum.
rather than pushed through the membrane inlets so that the membrane is not unduly pressurized. Except for the comparison study, the flow-through inlet was used for all experiments in this study.
RESULTS AND DISCUSSION
I. Steady-State Permeation Response. Steady-state permeation is described by Fick’s first law, which in general terms is given by permeation rate = (dimension factor)(permeability) (concentration) where the dimension factor is usually some function of the surface area to membrane thickness ratio. For a hollow fiber membrane, Fick’s first law gives F = 2aLD(C1 - C2)/ln (ro/ri) (1) where F is the flow rate (permeation rate) of a substance in the extract stream; D is the diffusivity of the substance in the membrane polymer; L is the length of the hollow fiber; C, and C2 are the concentrations of the substance in the high- and low-pressure surfaces of the membrane, respectively; and ro and ri are the outer and inner radii of the hollow fiber, respectively. If the low-pressure side of the membrane is exposed to the mass spectrometer vacuum or swept with a canier gas, C2 becomes very small relative to C1 and can be ignored. The concentration C, is established by the partitioning process and, for gas samples, is directly proportional to the partial pressure, p, of the substance in the sample. The proportionality constant is S, the Henry’s law solubility coefficient, such that C1 = Sp. However, since the desired measurement for most analytical methods is concentration and not partial pressure of a substance in the sample, eq 1 can be rewritten as follows:
F = 2*LDSptCp/pJ/In
(ro/ri) (2) The partial pressure/total sample pressure ratio, p/pt, gives
10
loo
lam
1 m
lomm
l&
ANALYTICAL CHEMISTRY, VOL. 62, NO. 13, JULY 1, 1990
1267
Table I. Effect of the Radial Dimensions of the Hollow Fiber Membrane on the Analytical Response, I , for Organic Compounds in Nitrogena
compd dichloromethane 1,l-dichloroethene chlorobenzene acetone analyzer pressure, Torr
2.8 0.6 1.2
0.8 6.2
X
1,111 exptl theor
Zzb
Ilb
5.5 1.3 2.4 2.0 lo4 1.4 X lo-'
2.0
2.5 2.5 2.5 2.5 2.5
2.2
2.0 2.5 2.3
BEOYOFOFU IN WATER
The membranes are approximately 2.5-cm-long silicone hollow fibers. Temperature = 23 O C . *I, and 1, are given in arbitrary units. For 11,ri = 0.0153 cm and ro = 0.0318 cm; for I,, ri = 0.0735 cm and r, = 0.0980 cm.
,.A/ /
T E U P E u m (C)
Figure 4. Temperature effects on MS response.
78
o
1
235
a
3
390 4
I
703
646
8
7
8
e
io
FLOW RATE (cc/min.) Figure 3. Response for toluene vs sample flow rate through a
0.0305-cm-Ld., 0.0635-cm4.d. hollow fiber.
greater than through water for some compounds. The transition from laminar flow to turbulent flow occurs at Reynolds number values of 2000-3000 (9), NRe
= dup/@
(4)
where d is the inner diameter of the hollow fiber membrane, u is the sample linear velocity, p is the density of the fluid, and p is the fluid viscosity. To achieve these values, however, volume flows of over 25 cm3/min for water (25 "C)are required in a 0.0305-cm-i.d. hollow fiber. The effect of the linear velocity on the response for toluene in water is shown in Figure 3. Flow rates approaching 10 cm3/min appear to reduce the boundary layer to dimensions where the change in permeation rate with change in sample flow rate becomes small for most volatile organic analytes studied. When drawing the sample through the 0.0305-cm-i.d. hollow fiber by pumping at the outlet, flow rates higher than 10 cm3/min often result in the formation of bubbles in the sample stream. Increased turbulence (and thus permeation rates) for a given flow rate has been achieved by segmenting the flow stream (10)and packing the interior of the membrane with inert beads (11). These techniques have not been employed in this work. As described below, changes in response times also become small as flow rates approach 10 cm3/min. C. Inlet Configuration Considerations. Comparisons of the absolute signal showed a significant advantage for the flow-through configuration (Figure 1, bottom) when compared with the flow-over configuration (Figure 1, top). Approximately 50-fold greater response was observed for the flowthrough inlet than for the flow-over inlet. Collapsing of the soft membrane in the flow-over configuration resulted in a significant reduction in the effective membrane area. On the other hand, the inverted pressure drop in the flow-through configuration takes maximum advantage of the membrane
inner radius/thickness ratio. For identical volume flows, higher Reynolds numbers are more easily achieved with the flow-through configuration. There may, however, be cases where the flow-over configuration is preferred, as it can be made into a probe convenient for in situ monitoring of hazardous or reactive chemicals (12, 13). D. Temperature Effects. Permeation is a temperature-dependent phenomenon obeying the Arrhenius relation: P = Po exp[-E,(l/RT - l / R T O ) ] (5) where the initial permeability, Po,is given at some initial temperature, To; the activation energy for permeation, E,, is the sum of the activation energy for diffusion, E d ; and the difference in the heats of solution between the membrane and the sample matrix, AH, = H,(membrane) - H,(matrix). Substituting DK = P into eq 5 gives DK = D&o exp[-(Ed + AH,)(l/RT - 1/RT0)] (6) The activation energy Edis greater than zero, while for most volatile organic compounds AHs is less than zero. The direction for the change in the permeability with changing temperature is dependent upon whether the change in the diffusivity or the distribution ratio dominates, as determined by the relative magnitudes of Ed and AH,. The relative trends for the responses of air, water, and organic compounds with increasing temperature are shown in Figure 4. At higher temperatures, the permeabilities for air and water increase because increasing diffusivities for air and water dominate their permeability-temperature relationship. Organic permeabilities from water samples also increase at least in part due to increasing diffusivities. In contrast to aqueous samples, air samples showed a decrease in permeabilities for organic compounds due to their reduced partitioning into the membrane at higher temperatures. Since organic permeabilities decrease for air samples with increasing temperatures, their detection limits increase, as shown in Table 11. Because of the greater permeabilities for organic compounds from water at higher temperatures, their detection limits improve. Further observations on these temperature-related phenomena are presented below in the discussion on organic enrichment. Increases in organic permeation rates from water may also be due to the fact that the Reynolds number increases with
1268
ANALYTICAL CHEMISTRY, VOL. 62, NO. 13, JULY 1, 1990
Table 11. Temperature Effects on the Detection Limits (SignaVNoise = 2) for Organic Compounds in Air and Watera
compd
m/z
chloromethane dichloromethane chloroform carbon tetrachloride chloroethene 1,l-dichloroethene trichloroethene tetrachloroethane dibromomethane bromoform benzene toluene ethylbenzene chlorobenzene 1,3-dichlorobenzene 1,2,4-trichlorobenzene
50 84 83 117 62 61
detection limit, ppb in water 26 O C 45 O C
in air 26 "C 45 O C
325 669 182 130
368
269
301
101
125 39
128
30
166 172
29
171 78 91
742
6
5
233 163
7 18
5
28 14 42
106 112
31 45 12
146 180
3 3
34 36 21
54 38 57 14 3 5
13
6
5
29 32 106 62 3 2 6 8 23 51
25
25 93
0.
34 2 2 5 7 16 29
Figure 5. Plot of solutions to non-steady-state permeation (eq 8) calculated from t(50)values and permeation response data for chloroform at 26 and 85 O C . The thin lines are solutions using only the first term (n = 1) of the polynomial. The heavy lines are solutions using the first five terms ( n = 1-5).
temperature according to the relation N b = du(0.00538T2 + 2.38T + 48.7), so that boundary layers are reduced a t higher temperatures. This relation was determined from curve fitting a plot of water density/viscosity ratios vs temperature (14). 11. Non-Steady-State Response. Non-steady-state permeation is governed by Fick's second law:
(7)
The mathematical solution for diffusion through a membrane of thickness 1 following a step change in sample concentration is (15)
Ft = F,(1
+ [2C(-1)"e x p ( - ( n ~ / l ) ~ D t ) ] )
(8)
The permeation process exhibits an asymptotic approach to steady state; thus, the time required to achieve steady state, or even 95% steady state, t(,), may be difficult to determine. It is convenient, therefore, to relate the response time to the time required to achieve 50% steady-state permeation, t(50), which is more easily determined. The first-order approximation a t t(,,), neglecting all but the first exponential term (n = l),can be used to determine the diffusion coefficient (16), where
D = 0.14(12/t(50))
(9)
Further evaluation yields a t(95)/t(50) theoretical ratio of 2.7. Good agreement is obtained for times greater than t(50)when the first approximation is applied to most experimental response curves as shown in Figure 5. Thus, the response time approximation t(%)= 2.7t(,) is valid. For times less than t(50), additional terms in the polynomial are required for a good fit. Response times are reported in this work in terms of t(M) measurements. A. Effects of Hollow Fiber Dimensions. Since diffusivity is a constant for a given substance in a given polymer, the response time-thickness relationship can be expressed as follows: t(50)2/t(50)1
=
(l2/l1l2
1.a
0.6
TIME (MIN.)
a The membrane is a 2.5-cm-long, 0.0305-cm-i.d.,0.0635-cm-0.d. silicone hollow fiber.
d C / d t = -D(d2C/dx2)
0.0
(10)
The increase in the response time observed for a hollow fiber membrane with a 0.0216-cm wall thickness compared with that for a 0.0165-cm wall thickness is shown in Table 111. The experimental measurements of t(50)are dependent upon errors in defining the time when the sample makes initial contact
Table 111. Effect of the Membrane Thickness, I , on the Response Time, t(,o)(Equation IO), for Organic Compounds in Nitrogen' t(m)z/t(m)l
compd
t(m),b
t(m)2*
dichloromethane 1,l-dichloroethene chlorobenzene acetone
0.38 0.28
0.60
0.72 1.88
1.60 3.02
0.48
exptl
theor
1.6 1.7 2.2 1.6
2.2 2.2 2.2 2.2
a The membranes are approximately 2.5-cm-long silicone hollow fibers. Temperature = 23 "C. bt(m)land t(w)zare given in minutes. For trm,,,1, = 0.0165 cm; for trW,.,.1, = 0.0245 cm.
with the membrane, t(o),and the steady-state permeation response. For membrane separations, thinner is nearly always better, but the choice of membrane thickness is limited to commercial availability or the ability to cast thin, supported membranes that will withstand the pressure differential imposed by the sample and the mass spectrometer. B. Flow Rate Effects. In addition to a reduced steadystate response, poor mixing also results in longer response times due to the additional boundary layer through which the analyte must diffuse. Increasing the water flow rate through the hollow fiber from 1.3 to 3.5 cm3/min resulted in an approximately 50% decrease in t(w) measurements for both toluene and dichloromethane. This effect on the response time is dependent upon the diffusivity of the substance in the sample matrix and upon the thickness of the boundary layer. The elimination of this boundary layer reduces the response time to the limit imposed by the membrane. For toluene and dichloromethane, response times appear to become constant at flow rates greater than about 8 cm3/min. C. Inlet Configuration Considerations. Neither the flow-through inlets nor the flow-over inlets displayed an advantage in response times when similar sample flow conditions were established. Again, the flow-through configuration is generally preferred because of the ability to more easily achieve higher linear velocities and thus minimize boundary layers. D. Temperature Effects. Higher temperatures result in increased diffusivities and therefore faster response times. The permeation rate response curves for a step change in the sample concentration at various temperatures are shown for 1,2,4-trichlorobenzene in Figure 6. The t(50)values for several organic compounds a t different temperatures are given in Table IV. These data are given for gas samples. As expected,
ANALYTICAL CHEMISTRY, VOL. 62, NO. 13, JULY 1, 1990
1289
m 6.00
U
c5.6< P
5.0m
4.5-e,
4.0.P P
3.5-
m
c < 3.0L \
u 2.5-z 2.0.U
P 0
1.5-
TIME (min.) Flgure 8. Normallzed permeation response curves for 1,2,4-trichlorobenzene as a function of temperature. 1
Table IV. Temperature Effects on the Response time, for Organic Compounds in Nitrogeno
t(loI,
compd
26 O C
45°C
65 "C
85 "C
chloromethane dichlormethane chloroform carbon tetrachloride chloroethene 1,l-dichloroethene trichloroethane tetrachloroethene dibromomethane bromoform benzene ethylbenzene chlorobenzene 1,3-dichlorobenzene 1,2,4-trichlorobenzene
0.23
0.19 0.30 0.36 0.38 0.20
0.18 0.24 0.30 0.32
0.17
0.25
0.36 0.41 0.78 0.57
1.30
0.27
0.16
0.30
0.35 0.55
0.28 0.41
0.21 0.33
0.40 0.88
0.33
0.25
0.62 0.26
0.50 0.24 0.30 0.37
0.32
0.86 0.93
0.60
4.40 8.70
0.20 0.24
0.18 0.25
0.43
0.63 1.70 4.60
0.42 0.45
1.00 3.00
0.23
0.95
2.00
The membrane is a 2.5-cm-long, 0.0305-cm-id.,0.0635-cm-0.d. silicone hollow fiber. aqueous sample response times mimic gas samples when boundary layers are minimized because the diffusivity is dependent only upon the interaction of the analyte molecule with the bulk membrane polymer matrix. 111. Organic Enrichment. The relative permeability through the membrane for one compound over another determines the membrane selectivity. The enrichment factor, E, of one component, i, over another, j (e.g., an organic compound over the sample matrix), is defined by the ratio of their permeabilities:
E = DjKj/D,Kj = (Fi/Cj*)/(Fj/Cj*)
1
2
3
I
4
6
i 6
FLOW-THBU OBGANIC/WATEB BESPONSE RATIOS
trmb. min
0.34 0.42 0.52
0
(11)
A. Effects of Hollow Fiber Dimensions. The enrichment factor is independent of membrane dimensions, as seen in eq 11. Increasing the length or the number of hollow fibers will increase the permeation rates of all components proportionally, with no effect on the observed organic enrichment, if Ci* does not change over the length of the hollow fiber. Increasing the membrane surface area may be more efficiently accomplished by increasing the number, rather than the length, of hollow fibers exposed to the sample. B. Flow Rate Effects. When an analyte-depleted boundary layer is established, the observed enrichment factor is reduced due to the depressed analyte/matrix ratio at the sample-membrane interface. A plot of the analyte/matrix
Flgure 7. Relative organiclwater response ratios for the flow-over vs flow-through inlets obtained at comparable sample linear velocities. (a) Dichloromethane, (b) chloroform, (c) bromodichloromethane, (d) dibromochloromethane, (e) bromoform, (f) 1, ldichloroethene, (9) trichloroethene, (h) tetrachloroethene, (i)1, ldichloroethane, (j) l,l, 1trichloroethane, (k) toluene, (I)ethylbenzene, (m) chlorobenzene, (n) 1,2-dichlorobenzene.
response ratio vs flow rate follows the same trend as that for the analyte response shown in Figure 3. The matrix is in such large excess that changes in its response are relatively small. When the permeation rates become independent of the sample flow rate, so do the organic enrichments. C. Inlet Configuration Considerations. Under similar sample linear velocity conditions, the enrichment factor is independent of configuration. A plot of organic/water signal ratios for the flow-over inlet vs that for the flow-through inlet with comparable feed linear flows is shown in Figure 7. As expected, no enrichment advantage was observed for either configuration since the permeation rate for water is also affected by the effective membrane area. D. Temperature Effects. With increasing temperature, the enrichment factors for organic components are reduced because the permeabilities for air and water increase more than do the permeabilities for most organic compounds studied, as shown in Figure 4. For trace-level organic samples, the analyzer pressure follows the permeabilities of the air or water matrix; therefore, lower temperatures are desired for lower analyzer pressures and to minimize oxygen and water in the mass spectrometer. On the basis of the steady-state response data used to determine the detection limits in Table 11,the permeabilities from aqueous samples increased an average 35% (range 14-83%) when the temperature was increased from 26 to 45 " C for the compounds analyzed in water. The diffusivities through the membrane for these same organic compounds increased an average 46% (range 13-159%) based upon the response time data in Table IV, where [0(45') - 0 ( 2 6 O ) ] / 0 (26") = [ 1/t(W)(45")- 1/t(W)(26")] / [ 1/t(W)(26")]. Although the organic permeabilities increased, the water permeability increased 130% for the same temperature change. With a given membrane, the desire for increased sensitivity and reduced response time must be balanced with the need for efficient separations. Separation efficiencies are preserved when the sensitivities are increased with greater membrane
1270
ANALYTICAL CHEMISTRY, VOL. 62, NO. 13, JULY 1, 1990
Table V. Temperature Effects on the Enrichment over the Sample Matrix for Organic Compounds in Nitrogen and Watera response
enrichment factor in nitrogen in water 26 "C 45 "C 26 "C 45 "C
compd
factor
chlorom2thane dichloromethane chloroform carbon tetrachloride chloroethene 1,l-dichloroethene trichloroethene tetrachloroethene dibromomethane bromoform benzene toluene ethylbenzene chlorobenzene 1,3-dichlorobenzene 1,2,4-trichlorobenzene
0.82
40
27
1.41 1.91
48 91
33 58
3.69 1.10 1.87 2.51
110
74
37
26 35
3.08 4.37
5.87 0.65 1.00
0.87 1.11
1.30 1.45
57 140 310 84 320 170
81
180 50 170 110
380 450 550 520
230 280
780
400
330
350
1900 2100 1800 3300 970 600
1100 1100
1100 1800 530 360
190 570 2600 2300
330 1400
2200
1200
1300
730
470
300 390
450
100 1300
"The response factors are the weight/signal of the listed compounds relative to the weight/signal ( m / z = 91) for toluene. The response factor for nitrogen = 0.39; the response factor for water = 1.04. The membrane is a 2.5-cm-long, 0.0305-cm-i.d., 0.635-cm-0.d. silicone hollow fiber. surface area and response times are shortened with thinner membranes. For air samples, organic responses decreased an average 20% (range 10-3370) when the temperature was increased from 26 to 45 O C , even though diffusivities increased an average 43% (range 13-159%). Therefore, the decrease in the membrane/air distribution ratio dominates the permeabilities for these compounds. While the organic permeabilities decreased, the nitrogen permeability increased 29%. Here, the only advantage to heating the membrane is to shorten response times, as both enrichment and sensitivities decay upon heating. Again, thinner membranes should be used, or the needs of the analysis must be balanced. The enrichment factors that were calculated from the following equation using standard mass spectral library response factors, k (relative to toluene) (In, are given in Table V: where ki and kj are the mass spectral response factors, by weight, for the organic species and the matrix, respectively; Iiand ljare their analytical molar responses; and MI and M 2 are their molecular weights, correcting for discrimination due to the reference orifice inlet. In general, the more volatile compounds exhibit higher enrichment factors from water samples but lower enrichment factors from gas samples.
PRACTICAL IMPLICATIONS For chemical process analysis, temperatures, flow rates, and inlet geometries are often dictated by the process itself. Since the goal of in situ analysis is to characterize a process without disturbing it, it is desirable to minimize the amount of sample withdrawn from the process stream. Small-diameter hollow fibers allow for adequate mixing at relatively low sample flow rates. In cases where samples cannot be withdrawn from a stream, flow-over hollow fiber sampling may be required. Due to the decay in separation efficiency with increasing temperatures, it is generally desirable to perform the membrane sampling at lower temperatures. Processes have been studied a t temperatures exceeding 100 "C (13), and performance evaluations have been carried out at temperatures greater than 200 O C (18). The membrane material is rated for a workable
temperature range of -54 to 249 "C (19). However, the lifetime of the membrane is expected to be reduced a t elevated temperatures. At room temperatures, the Silastic material has a reported shelf-life (for medical uses) of 5 years. The operating lifetime will depend upon the conditions of the application. The flow-through configuration sampling may provide a fast and simple screening technique for analyzing priority pollutants in water samples. By pumping a water sample from a volatile organic analysis (VOA) vial through the membrane and back into the vial, small samples have been analyzed without appreciable consumption of the sample and without formation of headspace in the vial. The amount of analyte removed by the permeation process is actually quite small. A water sample containing 10 ppm of dichloromethane was analyzed from a 48-cm3 VOA vial for over 30 min with no observed loss in the analyte signal. A water sample spiked with toluene and dichloromethane was pumped a t various rates through three flow-through hollow fiber membrane inlets in series, and their permeate streams were analyzed. No significant drop in the permeation rate between the first and the third membrane in the series was observed.
CONCLUSIONS Membrane mass spectrometry has been shown to provide parts per billion level analyses for organic compounds in air and water. Increasing the inner radius/thickness ratio of a hollow fiber results in increased permeation rates. Faster response times are achieved with thinner membranes. The most efficient membrane separations are achieved a t lower temperatures and at high sample linear velocities. Higher temperatures may be warranted for kinetic studies requiring fast response. In general, the flow-through configuration is preferred, since higher linear velocities are more easily obtained and the effective membrane surface area is maximized. On the basis of the response time data, the diffusivities for most volatile, nonpolar organic compounds, nitrogen, oxygen, and water are generally within an order of magnitude. Thus, the enrichment observed is due predominantly to the distribution ratio which favors (by 10-103) the organic components over water and air. ACKNOWLEDGMENT We gratefully acknowledge the assistance of Jim Shao for the development of computer software used in this study. Registry No. Chloromethane, 74-87-3; dichloromethane, 7509-2; chloroform, 67-66-3;carbon tetrachloride, 56-23-5; chloroethene, 75-01-4; 1,l-dichloroethene, 75-35-4; trichloroethene, 79-01-6;tetrachloroethene, 127-18-4; dibromomethane, 74-95-3; bromoform, 75-25-2; benzene, 71-43-2; toluene, 108-88-3; ethylbenzene, 100-41-4;chlorobenzene, 108-90-7; l,3-dichlorobenzene, 541-73-1; 1,2,4-trichlorobenzene,120-82-1;acetone, 67-64-1; water, 7732-18-5. LITERATURE CITED Liewellen, P. M.; Lktlejohn. 0. P. U.S. Patent 3,429,105, Feb 1969. Upsky, S. R.; Horvath, C. G.; McMurray, W. J. Anal. Chem. 1908, 38, 1565.
Westover. L. B.; TOU,J. C.; Mark, J. H. Anal. Chem. 1974, 46. 568. Bier, M. E.; Cooks, R. G.; Tou, J. C.; Westover, L. B. U.S. Patent 4,791,292, 1969. Bier, M. E.; Cooks, R. G.-Anal. Chem. 1987, 59, 597. Langvardt. P. W.; Brzak, K. A.; Kasti. P. E. Proceedings of the 34th ASMS Annual Conference on Mass Spectrometry and Allled Tropics, Cincinnati, OH, 1988; p 195. Brodbelt. J. S.; Cooks, R. 0.;Tou. J. C.; Kallos. G. J.; Dryzga. M. D. Anal. Chem. 1987, 59, 454. Lee, C. H. J . Appl. pdym. Sd. 1975, 79, 83. Harland, B. J.; Nicholson, P. J. D.; Gllllngs, E. Water Res. 1987, 27, 107. Perry, R. H.; Green, D. Perry's Chemical Engineer's Handbook, 8th ed.;McGraw-Hill Book Co.: New York, 1984; Chapter 5. Peters. T. L.; Stevens, T. S. US. Patent 4,664,470. 1987. Stevens, T. S.; Jewett, 0. L.; Bredeweg, R. A. Anal. Chem. 1982, 54. 1206. Kallos, G. J.; Tou, J. C. Envlron. Sci. Technol. 1977, T I , 1101.
Anal. Chem. 1990, 62, 1271-1274 (13) Savickas, P. J.; LaPack, M. A.; Tou, J. C. Anal. & e m . 1080. 67, 2332. (14) CRC Hendbodc of Chemlsby and phvsics, 64th ed.; CRC Press, Inc.: Boca Raton. FL, 1983; pp F-5 and F-38. (15) Pasternak, R. A.; Schimscheimer, J. F.; Heller, J. J . Porn. Sci. 1070, 8, 467. (16) Ziegei, K. D.; Frensdorff. H. K.: Blair, D. E. J . Pobm. Sci. 1089, 7 , 809. (17) Analytical Sciences Mass Spectral Data Base. The Dow Chemical
1271
Co., MMiand, MI, internal communication. (18) Bbr, M. E.; Kohiato, T.; Cooks, R. G. Anal. Chim. Acta, in press. (19) Medical Materials Buslness Product Form 51-772589; Dow Corning Corp.: MMland, MI.
RECEIVED for review October 4, 1989. Revised manuscript received February 26, 1990. Accepted March 2, 1990.
Mass Spectrometry of Technetium at the Subpicogram Level Donald J. Rokop,* Norman C. Schroeder, and Kurt Wolfsberg Isotope and Nuclear Chemistry Division, Los Alamos National Laboratory, Los Alamos, New Mexico 87545
We have developed a method to determine 210' atoms of @% or @'Tcand 5 X IO' atoms of @@Tcby negatlve thermal lonlzatlon mass spectrometry. Interferences from Isobaric lmpurlties or hydrocarbons are equlvalent to 5 X IO6 atoms of technetlum or less. Lanthanum oxlde Ion enhancers In conJunctlonwith Ca( NO3), are added to slngle, zone-refined rhenlum fllaments to achleve lonlzatlon efflclencles that are >2% for the formatlon of Tc0,-.
INTRODUCTION As an essential part of our program to measure the fluence of high-energy solar neutrinos over the past several million years (I, 2), we have developed a high-efficiency mass spectrometric technique to determine technetium isotopic compositions using negative ions. In the molybdenum-technetium solar neutrino experiment, we anticipate levels of IO6 atoms cof' @ lg and g7Tc and 10l2 to 1013 atoms of gsTc in purified samples isolated from lo7 kg of molybdenite ore. Previously developed methods (3) for the mass spectrometry of technetium with positive ions have detection l i m b of 1 pg (6 X log atoms). This limit is mainly due to the high ionization potential of 7.3 V for technetium. In addition, isobaric impurities produced by positive ion methods are so large that these techniques cannot be used a t the low levels required for this experiment. Kastenmayer (4),Heumann et al. (9,and Delmore (6) have previously shown negative thermal ionization as a possible means of performing high ionization efficiency mass spectrometry on selected elements, including technetium. Delmore proved that pertechnetate ions could be formed, while Kastenmayer and Heumann developed methods for microgram samples, sizes that are still lo3 to lo6 times larger than our experiment allows. The attraction of negative thermal ionization is that technetium is formed and measured as the Tc04- ion while molybdenum, the isobaric impurity which is most common and difficult to remove, is preferentially formed as the Moo3- ion. A disadvantage of this method is that the only available spikes, 98Tc and 9 7 T ~ form , ions with *'O and l80a t the gsTc04- position. This limitation establishes a measurement limit of a few million atoms of V c . However, this is not a serious constraint for the solar neutrino experiment. EXPERIMENTAL SECTION Instrumentation. We use a tandem magnetic mass spectrometer with pulse counting and movable Faraday cage detectors
(7,8)both at the intermediate (between magnets) position and in front of the multiplier. The ion source is a modified Nier thin-lens source (9)designed for thermal ionization with "2"axis focusing. The tandem instrument is necessary to discriminate against large scatter tails of Reo4- ions that are produced by the ionization of the rhenium filaments and by residual hydrocarbons. The Reo; signals are between lo4 and 10" A. Full-peak resolution with a
