J. Phys. Chem. 1986, 90, 6201-6210
6201
Optimization of Weak Neutral Dimers in Nozzle Beams J. Robb Grover* Chemistry Department, Brookhaven National Laboratory, Upton, New York 1 1 973
and Edward A. Walters* Chemistry Department, University of New Mexico, Albuquerque, New Mexico 871 31 (Received: February 19, 1986; In Final Form: July 16, 1986)
A method is described for the optimization of conditions to produce neutral dimers in molecular beams collimated from gas jet expansions, including expansions of binary gas mixtures. The method makes use of the dependence on nozzle pressure of the observed relative ion intensities produced by energetic (120 eV) photons or electron impact. Relative dimer beam densities are obtained, and the ratio of neutral trimers to neutral dimers around the nozzle pressure at which the neutral dimers are a maximum can be estimated. For the expansion of gas mixtures, the neutral dimer optimization is in principle for a composite of heterodimer and homodimers, but nevertheless conditions often permit the optimization of a single species, or nearly so. In the several cases studied to date the photoionization of clusters is always highly dissociative, with dimers forming mainly monomer ions, and trimers forming mainly dimer and monomer ions. Therefore the direct use of cluster ion intensities to estimate the relative beam intensities of the corresponding neutral clusters is invariably misleading.
Introduction The synthesis of weakly bound dimers and larger clusters in jet expansions is in very widespread use. This technique is quite general and simple to implement,' but normally results in an array of cluster sizes, while accurate analysis of the resulting cluster distributions is often a significant problem. If a high-energy electron impact or photoionization mass spectrometer is used for analysis, extensive fragmentation of the clusters into smaller cluster ions and the monomer ion essentially always occurs (for recent discussions see ref 2-5), so that quantitative analysis by this means alone has not yet proved practicable. However, if one wishes to carry out observations on one particular neutral cluster, e.g. a dimer, it may be important to maximize the beam density of the desired cluster among the products emerging from the nozzle and minimize the densities of the neighboring or otherwise interfering clusters. It is unfortunate that a number of workers have already published results that should be remeasured or reinterpreted due to the difficulty in establishing exactly which species is being studied in a jet expansion. It would be very convenient to be able to effect the necessary optimization with respect to a given cluster utilizing only a mass spectrometer because it is a familiar instrument that is usually available. Indeed van Deursen and R e d published a useful algorithm for estimation of the conditions for the preparation of dimer beams essentially free of interference from trimers and larger clusters, which they obtained from the study of expansions of unmixed simple gases, viz. H2, Ne, Nz, Oz,Ar, and COz. Their prescription, though convenient, is too limited for application to the problems in which we are most interested; it does not locate the nozzle pressure at which the dimer beam density is at its maximum, and it does not treat expansions of gas mixtures. Also, their method gives a generalized estimate which is often not accurate enough for the study of specific systems. It is the purpose of this paper to report a partial but useful solution to this vexing problem, specifically the optimization of jet expansions of mixed gases for the investigation of weak dimers with minimization of interference by trimers and larger clusters, that we have developed (1) For discussions and references see (a) Levy, D. H. Annu. Reu. Phys. Chem. 1980,31, 197-225. (b) Balle, T.J.; Flygare, W. H. Rev. Sci. Instrum. 1981,52, 33-45. ( 2 ) Buck, U.;Meyer, H. Phys. Reu. Lett. 1984,52, 109-1 12. (3) Jonkman, H. T.;Even, U.; Kommandeur, J. J . Phys. Chem. 1985,89, 4240-4243, (4) La,N.; Fenn, J. B. Reu. Sci. Instrum. 1978,49, 1269-1272. Fenn, J. B.; Lee, N. Rev. Sci. Instrum. 1982, 53, 1494-1495. (5) Gentry, W. R. Reu. Sci. Instrum. 1982, 53, 1492-1493. (6) van Deumn, A.; Reuss, J. Int. J . Mass Spectrom. Ion Phys. 1977.23, 109-122.
in the course of a series of photoionization studies of weakly bound dimers.' Explanation of the Method The method is first described for the expansion of a pure gas, then generalized to mixtures. It is assumed that the jet expansion of a polyatomic gas A produces the clusters Az, A3, A4, ..., according to the well-established behavior in which, as the pressure of the expanding gas (i.e. the nozzle pressure) increases from low values, the density of Az in the expansion products at first rises and becomes the most prominent cluster, goes through a maximum, and then falls while the density of A3 rises and becomes dominant; the A3 maximizes and then falls as the density of A4 rises in its turn, etc. It is also assumed that the mass spectrometric analysis of these species in a molecular beam collimated from the expansion and crossed by an electron or photon beam is carried out at a fied high electron or ionizing photon energy to produce the ions A+, Az+,A3+, ..., plus various ion fragments. The goal of the analysis is only to obtain the nozzle pressure dependence of the beam densities of the neutral dimer, and of the neutral trimer around the dimer's maximum. For this analysis the intensities of the ions A+, A2+, and AS+are to be measured, for a given nozzle and temperature, from low nozzle pressures up to the pressure at which the intensity of A3+maximizes, above which pressure neutral clusters larger than trimers are significant (for simplicity we ignore the ion fragments here, but discuss their role further on.) Let [A'], be the observed (uncorrected) intensity of monomer ions A+ produced only from the dissociative ionization of clusters (obtained from the total A+ intensity by subtraction of the A+ intensity due only to the ionization of monomer A), let [Az+]be the observed intensity of dimer ions A2+, and let [A3+] be the observed intensity of trimer ions A3+. The beam number densities of the neutral dimer, Pd, and trimer, pt, are related to the observed ion intensities via the following expressions, neglecting tetramers and larger clusters
(3)
where I is the intensity of the photon (or electron) crossing beam, (7) (a) Walters, E. A.; Grover, J. R.; Newman, J. K.; White, M. G. Chem. Phys. Let?. 1984, I l l , 19C-194. (b) Walters, E. A.; Grover, J. R.; White, M. G.; Hui, E. T.J . Phys. Chem. 1985, 89, 3814-3818. (c) Grover, J. R.; Walters, E. A.; Newman, J. K.; White, M. G. J . Am. Chem. SOC.1985, 107, 7329-7339.
0022-3654/86/2090-6201$01 SO10 0 1986 American Chemical Society
Grover and Walters
6202 The Journal of Physical Chemistry, Vol. 90, No. 23, 1986 assumed to be smaller in diameter than the molecular beam, 1 is the effective path length of the photons or electrons through the molecular beam, where the electrons or photons are assumed to be negligibly attenuated, a d and a, are the total ionization cross sections of the dimers and trimers, respectively, a is the probability that the ionization of the dimer produces the monomer ion, /3 is the probability that the ionization of the trimer produces the monomer ion, 6 and E are the probabilities that the dimer ion is produced by ionization of dimer and trimer, respectively, TJ is the probability that the ionization of the trimer produces the trimer ion, and fM, fD, and fT are the respective probabilities that the mass spectrometer detects the monomer, dimer, and trimer ions. These expressions may be simplified via the substitutions
comparison with the experimental value from eq 12. (The pressure pairs should be chosen as close together as possible, but not too close because of the limited experimental accuracy.) Once one or more pairs of pressures are found by trial-and-error for which &/afrom eq 15 agrees with 6/a from eq 12, the remaining three coefficients can be evaluated via the following expressions, where C E (M2 - MI) + ( 0 2 - 4 )+ (Tz - TI).
P
= (M2 - M J / C
(16)
( 4- D l ) / C
(17)
= (T2 - T J / C
(18)
e =
v
Equations 16-18 are easily derived from eq 7, 8, 9, and 1 1 for the condition d , = d2. With all five coefficients known, eq 7 and 8 become two linear equations with two unknowns at any given nozzle pressure, so d and t can be evaluated point-by-point. to give
M = ad
+ @t
(7)
Note that the coefficients are so defined that a+6=1
(10)
P+t+q=l
(11)
Equations 5 and 6 define new quantities d and t proportional to the beam densities Pd and ptrthe common instrumental factors being collected inside the parentheses, because we are interested only in the functional shape of the dependence of the dimer and trimer beam densities on nozzle pressure, i.e. we need only the relative beam densities. Equation 4 describes the correction of the observed ion intensities for the mass dependence of the detector efficiency, normalized to the efficiency for detecting monomer ions; we neglect corrections to the detector efficiency for ions born with high translational energies. As a first step in the solution of eq 7-1 1 we see that extrapolation of experimentally determined values of DIM to low nozzle pressures P such that t D(A.B), the more strongly bound dimer will be formed and become optimum at much lower nozzle pressures than the more weakly bound dimer. Similar considerations apply to the trimers, so that if D(A.A) D(C6H6.Ar). A crucial requirement of the method presented above is the availability of the quantity M , which must be obtained by subtraction of the monomer ion intensity due only to the ionization of neutral monomers from the total monomer ion intensity. The procedure is described below. An electron impact or photon energy much larger than the ionization energy of A must be used, so that fragment ions F+ of A are formed that require a substantial amount of energy beyond that needed just for the production of A+, for example the formation of C4H4+from C6H6. We have observed that such high-energy fragments F+ are not detectable from clusters A,,, an observation that seems quite general (see Results section). This is very convenient, because then the intensity of F+ can be used as a measure of A alone in the presence of clusters A,,. Thus, the intensity ratio [A+]/[p] can be measured under conditions where there are no clusters, e.g. at low nozzle pressure
and used to evaluate M at nozzle pressure P.
[A+], = M = [A+], - ([A+l/[F+l)o[F+l,
(22)
Here, the square brackets mean the observed intensity of the ion that they enclose. Of course this procedure can only be applied to polyatomic gases. However, there is a complication caused by the polyatomicity of A and B. In general, ions larger than A+ or B+ other than A,+ or A,B,+ are formed. Usually this does not seriously interfere with the limited objective of locating the nozzle pressure for which the beam density of neutral dimers is at a maximum, but complicates the determination of the ratio d / t . To obtain d / t the intensities of the fragment ions must be included in the sums for M,D, and T to preserve the validity of eq 10 and 11, so we must further generalize the definitions of M , D, and T for this purpose. Equations I and 8, our original definitions of which confine M only to monomer ion and D only to dimer ions, also apply to the broadened definitions of M and D because both of these quantities include products of the ionization of the dimers plus contributions from trimers. We are then free to apportion the intensities of the dimer ion, its fragment monomer ion, and all of its other fragment ions to M o r D in such a way as to facilitate the solution we seek. In general, inspection of the data will indicate that the intensities of some fragment ions rise more rapidly with increasing nozzle pressure than the intensities of others, because they are produced to a greater extent by the ionization of trimers (and larger clusters). Thus, in eq 4'' we may advantageously redefine M as the sum of the intensities of fragment ions A+ that rise relatively more slowly with increasing nozzle pressure, and D as the sum of the intensities of ions a)+ that rise relatively more rapidly, where each term in the summations is appropriately corrected for detector
efficiency. In practice, we have found that with this rule the fragment monomer ion intensity [A'], is always included in M , the dimer ion intensities always go to D, and that it is usually straightforward to decide which term is more appropriate for each fragment ion intensity. Similarly, let T be the sum of the intensities of all ions ?+that can come from trimers or larger clusters, excluding those that are required to come from clusters larger than trimers. This further generalization of M , D, and T, following the pattern of eq 4 and 4', can be written
where, as before,A, is the detector efficiency for ion i+. In most systems that we have studied, the cluster ion spectra are dominated by species of the form A,+ or AnBm+,where the fragments are unimportant, but a few molecules, such as ammonia, provide conspicuous exception^.^ One sometimes wants to know the relative proportions of the neutral dimers and trimers in a molecular beam. It is useful to understand how well this information can be recovered from the values of d and t if the latter are obtained from the generalized definitions of D and T given above, The total ionization cross sections are contained in d and t as defined in eq 5 and 6 . It is reasonable to expect that at the high electron impact or photon energies at which the measurements are made the total ionization cross section of A, is approximately proportional to n, and therefore t should be multiplied by 213 for its proportion to d to be a reasonable estimate of the number density of neutral trimer clusters relative to dimer clusters, i.e. from eq 5 and 6 However, for mixed clusters one must not only use the relative total ionization cross sections of A and B, but also know the relative proportions of the different dimer and trimer species in the beam. Since the latter information is not forthcoming from this method, estimates of the proportions of mixed dimers to trimers must remain rather uncertain. As a rough rule-of-thumb we compare ( 2 / 3 ) t with d for mixed clusters, but recognize that the true relationship can deviate considerably from this rule. In summary, it should be emphasized that the above-described method of locating the nozzle pressure that maximizes the beam density of neutral weak dimers is only approximate. However, we ask only that it serve as a guide for selection of conditions for the easiest possible observations of the dimers consistent with minimal interference from larger clusters. This purpose demands only modest accuracy. Probably the most serious source of inaccuracy for pure gases is the truncation of the calculation at the trimers, while for mixed gases the method suffers in addition from our neglect separately to consider all of the different molecular combinations involved for each size cluster. In particular, it is to be expected that often the dimer maximum will occur at a somewhat higher nozzle pressure than is indicated by this method, because the trimer is increasingly overestimated with increasing pressure causing the dimer to be underestimated; fortunately both inaccuracies are such that any error in choosing the optimum experimental conditions for dimers will cause an overestimate of the possible interference from trimers. The prescription has proven more than adequate for the limited purpose for which it is intender', and is sufficiently straightforward that it can be applied while the data are being taken. At its simplest it is only necessary to find the optimum nozzle conditions by matching 6 / a from eq 15 with S/a from eq 12, and do ancillary measurements around this pressure to see if a lower pressure must be used to avoid interference from trimers, i.e. the five coefficients and d and t themselves need never be explicitly evaluated. In fact, it is easily shown (9) Ceyer, S. T.; Tiedemann, P. W.; Mahan, B. H.; Lee, Y.T. J Chem. Phys. 1979, 70, 14-17.
6204
The Journal of Physical Chemistry, Vol. 90, No. 23, 1986
Grover and Walters
that the shape of the d function is independent of the mass dependence of the detector efficiency, so that this correction can be neglected if the position of the maximum is the only information sought.
Experimental Section The experiments reported here were carried out at the 750-MeV electron storage ring of the National Synchrotron Light Source at Brookhaven National Laboratory. The apparatus has already been described in ref 7c and 10, but certain details pertinent to the present study are amplified here. The molecular beams containing clusters were produced by jet expansion of gas through a nozzle 0.01 cm in diameter. The nozzle chamber was evacuated by an 8000 L s-I diffusion pump backed by a 240 m3 h-' Roots blower, backed in turn by a 30 m3 h-' mechanical pump. At lo00 Torr of nozzle pressure, the jet expansion would typically be into Torr. A 0.1-cm skimmer was used, the pressures of - 5 X nozzle-skimmer distance of 0.5 to 1 cm being adjusted for maximum beam intensity. The post-skimmer collimator was 0.3-cm in diameter and was evacuated by a 500 L s-l turbomolecular pump. Under the above conditions the pressure in this Torr. The molecular beam intersected region is typically 2 X the ionizing photons in an interaction chamber at a point 11 cm from the nozzle and in the optimal ion-extraction region of a lens system that focused the ions on the entrance aperture of a quadrupole mass spectrometer equipped with a channeltron detector operated in the ion-counting mode. The interaction chamber and beam dump were pumped by two turbomolecular pumps with a combined pumping speed of 700 L s-l plus a 15 K helium refrigerator trap into which the molecular beam was directed. Interaction chamber pressures were -2 X lo-' Torr under fhe above conditions. Ionizing photons of wavelength 584 8, (21.2 eV) were employed almost exclusively for the studies described here, to allow direct comparison with work done off-line using the H e I line produced by conventional laboratory lamps. This photon energy is high enough to ionize anything of interest for the present work and also provide the high threshold fragment ions F+ required in the analysis. The gas mixtures supplied to the nozzle were accurately mixed off-line and stored in large stainless steel tanks at pressures sufficiently low to avoid any condensation. For mixtures of gas plus the vapor of a liquid, the gas was bubbled through several centimeters of the liquid from an immersed frit, keeping a constant head pressure maintained by a hand-operated needle valve and monitored by a capacitance manometer. The nozzle pressure itself could be accurately controlled electronically to f l Torr via a capacitance manometer + servomotor drive valve system. For calibration of the mass spectrometer plus detector system for the mass dependence of the detection efficiency, a 1:l mixture of COz and SF6 was expanded at a nozzle pressure of only 100 Torr to avoid fractionation and clustering effects, and the relative intensities of the ions COz+,SF3+, SF,+, and SF5+were measured. These intensities were compared with the published cross sectionsll,'* to provide relative efficienciesup to mass 127. Efficiencies between masses 43 and 25 were estimated via plausible extensions of the efficiency curves. Approximate efficiencies at masses 156 and 234 were obtained from benzene cluster ions as described in the Appendix. It is necessary to repeat this calibration whenever the quadrupole mass spectrometer plus ion lens system is retuned. A typical calibration curve is shown in Figure 1. Results and Discussion First, evidence for the two important phenomena exploited in the analytical method described above is presented, viz. the tendency for dissociative ionization of clusters to make monomer parent ions with higher probability than monomer fragment ions, and the practicabilit. of measuring the ratio of coefficients S/a (IO) White, M.G.; Grover, J. R. J. Chem. Phys. 1983, 79, 4124-4131. (11) Masuoka, T.; Samson, J. A. R. J. Chim.Phys. 1980, 77, 623-630. (12) Hitchcock, A. P.; Van der Wiel, M. J. J. Phys. B 1979, 12, 21 53-2169.
I.5
I .c
>
0
z
w
0
\
LL LL
W W
2
0.5
c 4:
_I
W [L
\
0.2 60
I20 ION M A S S
180
24 0
Figure 1. A typical curve of the mass dependence of the detection efficiency of the ion lens-quadrupole mass spectrometer system as a function of ion mass. The curve is arbitrarily normalized to unity at mass 78.
;
.... 5 1 (I)
z
U
z W
c
z
I O N MASS
Figure 2. Ratio of ion intensities from the photoionization at 584 A of molecular beams collimated from the jet expansion of 1:9 trans-2-butene + SO2at 800 Torr nozzle pressure, and of neat trans-2-butene at 800 Torr. All organic ions whose fragment yield in the neat expansion is greater than 0.4% have been included. The ratios of masses 27 to 42 have been normalized to unity, shown by the dashed line. (Note that these are different data than those presented for the same mixed gases in Table 1.)
by extrapolation of DIM to low nozzle pressures. Finally, three examples of nozzle cluster beam analyses are presented. Monomer Fragments from Clusters. It is to be expected that the dissociative ionization of a molecule bound in a weak complex is different from that of the isolated molecule, because there are additional pathways available for the disposal of excitation energy. That such is indeed the case is illustrated for complexes formed in the jet expansions of several different gas mixtures, in Table I, and in more detail for one mixture, trans-2-butene SOz, in Figure 2. For each mixture, the intensities of a few prominent
+
The Journal of Physical Chemistry, Vol. 90, No. 23, 1986 6205
Weak Neutral Dimers in Nozzle Beams I
I
I
I
I
I
I
I
I
( a ) 1,3 BUTADIENE + S O 2 I: 4 I
0 0
0
0. I
1
:I 0.8
0
0
O
0
-
I
I
I
I
I
I
I
( b ) 1,3- BUTADIENE
I
I
‘
-
8
7;= 0.020
0
1
I
200 400 600 N O Z Z L E PRESSURE ( T o r r )
0
r
0.4
0
I
O
O
800
I
I
+ Ar
(
0.4
(d)
I
I
I
I
I
I
I
I
(
C2H4+HCI 1: 4
I: 4
0 0
I,i-
0.2
y D
D -
1.01
0
M
0
0.i 0
0 0
0.I 7
0
Y ----&-ee---- - - - B.o.039 a
01
I
0
I
I
I
400 N O Z Z L E PRESSURE ( T o r r )
200
I
6(
0 0
I
I
200
I
I
400
I
I
600
I
I
800
I
IO00
NOZZLE PRESSURE ( T o r r )
Figure 3. Extrapolation of the intensity ratio DIM to low pressures to determine the value of 6/a for jet expansions of several gas mixtures: (a) 1,3-butadiene + SO2, 1:4 mixture; (b) 1,3-butadiene + Ar, 1:4; (c) benzene vapor + Ar, 1:9; (d) C2H4+ HC1, 1:4. The compositions of D and M for (a), (c), and (d) are given in the text. For (b), M = [(C4H6+),]and D includes [(C4H,+),]and [C4H7+],where the subscript c means ions produced
only from clusters. ions associated with the organic component (Le. the component of lower ionization potential) are compared, at lower nozzle pressures (or for the neat gas of the organic component) where there is essentially no complex formation, to the intensities of the same ions at higher nozzle pressures where there is substantial complex formation. Nozzle pressures in Torr are indicated at the tops of the columns, and the nozzle temperatures are all 23 OC. In each case, the intensity of the mass peaks are normalized to one peak that represents a fragment ion whose production requires energies significantly larger than the ionization energy of the organic partner molecule. All the mixtures tried so far (including several not shown) display the same pattern. The intensity of the parent ions of the organic moiety increase by a conspicuously larger factor than any of the fragment ions as the nozzle pressure is increased, shown in the column labeled “ratio”, which gives the ratio of relative mass peak intensity a t the higher nozzle pressure to relative mass peak intensity at the lower nozzle pressure. The same effect is depicted graphically in Figure 2, where the parent ions of the organic moiety, C4H8+and 13C’2C3H8+,at masses 56 and 57, are greatly enhanced over the fragment ions. This means that much of the energy that would lead to fragment ions in the isolated molecule is disposed of in other channels available to the clusters, namely to break the ion-neutral bond that was formerly the van der Waals bond, to contribute to the kinetic energy of separation of the former partners, and to supply excitation energy to either or both of them. In some cases fragment ions that require only a little more energy to be formed than do the parent ions
of the organic moiety show a similar, though diminished, effect. These observations are consistent with a picture in which the dissociative photoionization of van der Waals clusters produces the parent ion of the more easily ionized partner molecule with high probability. The allyl bromide-hydrogen chloride mixture is an apparent exception, wherein the lowest energy ion fragment of the allyl bromide, C3H5+,dominates; however, even this conclusion depends on the energy of formation of the as yet unknown molecule BrHCl. Determination of b/a. It is expected that the neutral dimer is the first cluster to become important in a jet expansion as the pressure behind the jet increases, and that the trimer then grows in at higher pressures, i.e. that there is a low pressure regime in which t I- I
I
cn
1,3 BUTADlENE+SO2 1: 4
z
W
n
w
D
I/
U
W
m
/
>
I
/
6' H6+
1:s
W
m
>
-
m -
I-
z
cn z w
W I--
I
z-
I-
z -
0
i
200
600 NOZZLE PRESSURE (Torr)
400
0 z -
aoo
+ sulfur dioxide. Dashed lines: relative ion intensities corrected for mass dependence of detector efficiency and smoothed by eye. Composition of M,D, and Tare given in the text. Solid lines: relative densities of neutral dimers, d, comprised of comparable amounts of 1,3-butadiene-sulfur dioxide and (1,3-butadiene)*,and of neutral trimers, (2/3)?, in a molecular beam collimated from the expansion. TABLE II: Trial Values of 6/a Calculated from the Smoothed Curves of M , D , and Tin Figure 5 according to Eq 15, for the System 1,lButadiene Sulfur Moxide
+
Pp
A P = 150
600 650 675 700 725
0.0312
A P = 200 0.0132 0.02 19
AP= 250
A P = 300
0.0263 0.0369
0.0290
0.0450
0.0400 0.045 1
750 800
0.0739 682
690
702
715
'Pressures PB, AP, and PMare in Torr. 4b also shows the uncorrected intensities of all of the cluster ions and cluster ion fragments of significant intensity. After correction for the mass dependence of the detector efficiency, we set M = [(C&+)C], D = [C4H&02+1 + [(C4H6)2+] + [C&.SO+I, and T = [c&*(s02)2+] [(c&)280,+] [(C,H,),+]. These data, smoothed by eye, are plotted in Figure 5 as dashed lines. The value 6/a = 0.033 was obtained from the extrapolation of the (unsmoothed) ratio DIM to low pressures, as already seen in Figure 3a. Equation 15 was applied for various estimated pressure values, PE, of the neutral dimer maximum, and various pressure differences AP (such that PI = PE- 1/2(AP) and P2 = PE 1/2(AP)), using the smoothed curves of Figure 5, to obtain the trial values of S/a shown in Table 11. Linear interpolation gives the pressure PMfor which 6/a = 0.033, shown in the bottom row of Table 11. The interpolated position of the maximum in d is only weakly dependent on AP in this case, and we estimate that it approaches about 670 Torr as AP 0. Application of eq 13, 14, and 16-18 for the conditions PE= 675 Torr, AP = 150 Torr (Le. PI = 600, P2= 750) then reveals the coefficients of eq 7-8 to be a = 0.968, 6 = 0.537, 6 = 0.032, e = 0.158, and 11 = 0.305. With these parameters known, the relative beam densities of the neutral dimer and trimer can be evaluated by using eq 19 and 20, and are displayed in Figure 5 as the solid curves. Although no independent measurements of the neutral dimer intensity were carried out here and may indeed not be possible with the present apparatus, such a study is possible for the related system trans-2-butene SOz,and was used to verify the results of the analysis for that system,7c which analysis is analogous to that just presented. Also, we observed that the intensity (normalized to constant light) of the ion (C4H6),+just above threshold, a t 1300 A, is nearly the same a t 800 Torr as it is a t 500 Torr, which is consistent with the neutral dimer curve in Figure 5 and
+
+
+
-
+
/
4 -
Figure 5. Analysis of products of the jet expansion of 1:4 1,3-butadiene
FM
I
a
/
k
0
I
1
I
1' 0/
I
I
1
6208 The Journal of Physical Chemistry, Vol. 90, No. 23, 1986 I
l
l
I
l
I
I
l
I
i
>
k v) z
/
/
W
/
n 2
/
a
/
W
/
/
m >
/ /
c
/
v)
z w I-
z
z
El
Grover and Walters factor 1.753 accounts for contributions from ions containing two chlorine atoms, one or both of which are 37Cl. The ion C2H5' was included in M because its pressure dependence is remarkably similar to that of CzH4+, and this choice therefore enhances the difference between the pressure dependences of M and D. For practically the entire pressure range studied [C2H5+Ici= (1/3)M. It was necessary to correct [C4H9+] for the contribution of [13C12C3H8+] (= 0.0448[C4Hs+]), and [C2H635CI+]for the contributions of [CzH437C1+] (= 0.324[C2H435Cl+]) and [ 13C12CH535C1+] (= 0.0224[CzH535Cl+]). The hydrocarbon-only cluster ions (C2H4)2+ and (c2H4)3+ rise a t nozzle pressures significantly larger than those at which the corresponding mixed dimer and trimer ions become important, and on a semilog plot they rise much more steeply, which indicates that the homoclusters (C2H4)2 and (CZH,), are more difficult to form than the corresponding heteroclusters. This indication is confirmed by a study of the jet expansion of CzH4 Ar, for which the heat capacity ratio y = 1.491 is larger and hence the beam temperature significantly lower than for the C2H4 HCl expansions, for which y = 1.353. For C2H4 Ar the cluster ions (C2H4)2+ and (C2H4)f only become significant above 500 and 900 Torr, respectively, while the ratio [C2H4+]c/[CZH4+] only rises to about 15% of the corresponding value achieved in the C2H4 HCl system. We conclude that D((C,H4)2)
