JOURNAL OF RESEARCH of the National Bureau of Standards - A. Physics and Chemistry netic Mass Spectrometric Investigation of the Reactions of t-C. Hot lons With Some Simple Polar Molecules at Thermal Energies* Lucette Hellner** and L. Wayne Sieck Institute for Materials Research, National Bureau of Standards, Washington, D.C. 20234 (April 21, 1971) The interactions of t-C4H ions from neopentane with some simple polar molecules have been investigated in a high pressure photoionization mass spectrometer at thermal kinetic energies. Proton transfer was found to occur from t-CH to acetone, ammonia, and the various methylamines, but not to molecules having estimated proton affinities <195 kcal/mol (815 kJ/mol). Macroscopic thermal rate coefficients are reported for the various proton transfer reactions, all of which are on the order of 10-9 cm3/molecule second. The reaction t-CH+2CH3OH → (CH3OH)1⁄2H++C4H8 was found to occur, but not the analogous reaction with H2O. On the basis of supplementary experi- Key words: Ion molecule reactions; mass spectrometry; neopentane; photoionization; proton affinity; The ion-molecule reactions involving t-C4H ions and certain organic molecules have also been investigated by the techniques of high pressure electron impact mass spectrometry, using both conventional single stage [2] and modified chemical ionization instruments [3, 4]. However, absolute rate constants have been determined and/or estimated for only a relatively few (1) systems. ich was monitored by determining the isotopic mposition and yield of the stable chemical endoduct, isobutane. Although the relative reactivities. some 22 C5-C8 alkanes were determined with a high ecision by this method, absolute rate coefficients for ocess (1) could only be estimated by assuming the aximum theoretical value (4.0 × 10-9 cm3/mole 'Supported in part by the U.S. Atomic Energy Commission. "Guest Worker, 1970-1971. Permanent Address, Equipe de Recherche du C.N.R.S., As part of a continuing program involving ionchemistry in the vapor phase we have investigated the interactions of t-C4H with simple polar molecules at thermal kinetic energies in the NBS high pressure photoionization mass spectrometer. Neopentane was chosen as the source of t-C4H since the photoionization of this molecule at 106.7-104.8 nm (argon resonance lamp) yields only t-CH, and i-CH as fragment ions [5]. It is also known that neopentane is very unreactive towards t-C1H [1, 5–7]; consequently these ions are available for reaction with any additive compounds chosen for investigation. In pursuing this study we hoped to convert the relative rate constants which were previously measured in this laboratory to an absolute basis, as well as provide information concerning the overall reactivity of t-CH; and, perhaps, the proton affinities of some simple polar molecules in the vapor phase. 2. Experimental and Results All experiments were carried out at 298 K with the NBS high pressure photoionization mass spectrometer described in detail elsewhere [8, 9]. API neopentane was used without further purification. The additive compounds, which were of research grade quality obtained from various sources, were also used without further purification except for outgassing and fractional distillation in vacuo. 2.1. Photoionization of Neopentane at 116.5 and 106.7-104.8 nm Results obtained from the photoionization of pure neopentane (ionization threshold = 10.37 eV [10], 128.2 nm) at 116.5 and 106.7-104.8 nm are displayed in figure 1. In agreement with previous electron impact and photoionization results parent ions from neopentane were not detected at any wavelength. At 116.5 nm (Kr resonance lamp) the primary mass spectrum consists of 42.5 percent CH and 57.5 percent t-C4H. At 106.7-104.8 nm (argon resonance lamp), where the majority of the experiments were carried out, photoionization of neopentane yields 89 percent t-C4H and 11 percent C4H. No additional ions were detected at any wavelength at pressures up to approximately 1 torr, and the CH/CH ratio remained constant over this range. This behavior indicates an upper limit for the bimolecular rate constant for reaction of CH and C4H with neopentane of 10-14 cm3/molecule second at 298 K, which is consistent with the conclusions of earlier mass spectrometric [7, 11] and radiolysis studies [1, 6]. 2.2. Photoionization of Neopentane in the Presence of Polar Additives Neopentane was photolyzed in the presence of vary ing amounts of H2O, CH3OH, C2H5OH, (CH3)2CHOH. CH3CHO, CH3COCH3, CH3OCH3, NH3, CH3NH (CH3)2NH, and (CH3)3N at 106.7-104.8 nm. A syste matic search was made in each neopentane-additive combination for a bimolecular chemical reaction involving t-CH, and the polar component (X) in the mixture, particularly the unimolecular decomposition process involving the collision complex which is equivalent to proton transfer: t-C4H+XXH++C4H8 (or dissociative proton transfer products). (3 The compositions of the mixtures were generally varied from 3 to 90 percent neopentane. No reactions involv ing neutral neopentane and the ionized polar component in the mixtures were detected. The results of the various experiments are summarized in table 1. which is arranged in the order of increasing protor affinity (P.A.) of the polar additive. Absolute values for the P.A.'s of the additives are also included, insofa: as they are known, except in those cases where only the relative basicities have been established (as in the amines). The following comments may be made concerning the specific behavior found for the various reaction pairs: (a) Water-Due to the higher ionization potential of H2O (I.P. = 12.6eV [17] the photolysis of neopentane H2O mixtures at 106.7-104.8 nm resulted in a primary mass spectrum containing only CH and CH Process 3 was not detected in any experiment. The ter molecular reaction t-CH+2H2O→H(H2O)1⁄2 +C‚H. involving two water molecules was also not found to occur in mixtures containing as much as 97 percent H2O. The only process detected was the formation of the stabilized t-CH-H2O adduct at pressures in excess of 20 to 30 millitorr, presumably via a termolecular association reaction. The rate constant for this extremely inefficient process was not estimated. (b) Methanol-The photoionization of neopentaneCH3OH mixtures was carried out at 116.5 nm (10.6 eV) since primary ionization of CH3OH (I.P. CHOH = 10.85 eV) does not occur at this wavelength. No bimolecular reactions involving t-CH and CH3OH were detected. However, the extremely efficient termolecular process TABLE 1. Reactions of t-C4H with some polar molecules at 298 K t-C4H+X→ products a M t-C1H + H2O→C4H9H2O++M t-C1H$+2CH3OH → H*(CH3OH)2+C4H8 M t-C1H$+CH3OCH3 →C,H,CH3OCH++M M t-C1H$+i-C3H7OH → C4H9C3H7OH++M t-C4H+NH3 NH*+CH k=0.90±0.08; (ka=0.96, kμ = 4.2) k=1.2±0.1; (ka= 1.25, kμ = 1.2) Where rate coefficients are given, the units are cm3/molecule second × 10−9. "First entry, kcal/mol; second entry, kJ/mol. Ref. [12]. Ref. [13]. Ref. [14]. 'Ref. [15]. *See discussion. Ref. [12b). Order of basicities given in Ref. [16]. was observed. The rate coefficient for this Process was found to be 4.2±0.5 × 10-11 cm3/molecule second. (d) Ethanol - No bimolecular reactions were found involving t-C4H and C2H5OH. (e) Dimethyl ether-No bimolecular reactions were observed, although the formation of the [C4H9CH3OCH3* adduct was detected at pressures in excess of 15 to 20 millitorr. (f) Isopropanol-Again no bimolecular reactions were found, although the stabilized adduct [C4H9(CH3)2CHOH]+ was detected at pressures in excess of 20 millitorr. was found to occur efficiently in neopentane-acetone mixtures. Typical data obtained for the rates of forma tion and disappearance of the major ions resulting from the photolysis of a 1:1 mixture at 106.7-104.8 nm as a function of chamber pressure are shown in figure 2. At these wavelengths the primary mass spectrum of acetone contains 84 percent CH COCH and 16 percent CH3CO+, which accounts for the appearance of CH,CO+ in figure 2. Separate experiments involving acetone along verified that CH CO reacts with acetone via proton transfer, while CH COCH; reacts both by proton transfer and via formation of (CH3COCH3)CH CO+. The detailed results pertinent to pure acetone and other ketones will be discussed at a later date. The macroscopic rate coefficient for Process 6 was derived by the method described [9] previously through consideration of the slope of the decay curve in that FIGURE 2. Photoionization of a 1:1 mixture of neopentane and acetone at 106.7-104.8 nm. Composite mass spectrum as a function of chamber pressure for major ions. region of low chamber pressures where the semilogarithmic decay plot is linear. The downward curvature of the semilog decay for m/e 58, 57, etc., at higher pressures is due to an increase in the ionic residence time due to nonreactive scattering. Experiments with mixtures of various compositions yielded an average thermal bimolecular rate coefficient of 1.1±0.1 × 10-9 cm3/molecule second for the proton transfer reaction involving t-CH; and acetone. Proton transfer from i-CH was found to be slightly more efficient. A thermal rate coefficient of 1.3±0.1 x 10-9 cm3/molecule. second was derived for this process. (h) Ammonia and methylamines - Process 3 was also found to occur efficiently in neopentane-NH3 mixtures, as well as mixtures with methyl-, dimethyl- and trimethylamine. The thermal rate coefficients derived for these systems are given in table 1. + Additional experiments were carried out with neopentane-d12, which yields 85.7 percent t-C4D and 14.3 percent CD at the argon resonance lines. In mixtures of neopentane-d12 with trace amounts of NH: it was determined that the reaction t-C4D+NH; → products yields 3.6 percent NH2D and 96.4 percent NH3D as products, indicating that hydrogen exchange within the collision complex is a relatively inefficient process. No exchange was detected in any other reaction pairs except t-C4D-CH3NH2, which yielded approximately 2 percent (CND2H1)*. Trace amounts of the additives were used in order to minimize the probability for successive exchange reactions such as NH2D+NH3 →NH3D++NH2D which appear to remove the deuterium content of the primary reaction product. Supplementary experiments involving acetone-NH3 mixtures were also performed. These will be discussed in the appropriate portion of the text. 3. Discussion 3.1. Magnitude of the Rate Constant for Proton Transfer The total rate constant for collision involving t-C4H ions and polar molecules may be approximated by assuming that the collision is described by a cross section for both ion-induced dipole [18] and ion-dipole interactions [19]. Calculated values for the ion dipole (k) and ion-induced dipole (ka) rate coefficients are included in table 1 for those reaction pairs in which proton transfer was found to occur. Within the limits of experimental error, the calculated values for ka are in excellent agreement with the thermal macroscopic rate coefficients found in the present study. Although this agreement may be fortuitous, this equivalence suggests that ion-dipole interactions have little, if any, effect on the total cross section for collision if proton transfer occurs with unit efficiency (at every collision). The fact that very little hydrogen exchange occurs in the t-CH-NH and t-CH-CH3NH2 reaction pairs, and none in the other systems, is a further indication 8 may be calculated provided that AH (MH+) is known. Recently Lossing and Semeluk [20], using a refined energy resolved electron beam mass spectrometer, have determined the AH, for the various isomeric butyl ions. Their value for AH(t-C,Hg) of 167 kcal/mol (698 kJ/mol) yields a P.A. for i-C4H8 of 195 kcal/mol (815 kJ/mol) using Reaction 7. In order for proton transfer to occur from t-C4H to any of the molecules (X) investigated in this study, the P.A. (X) must be > P.A. (i-C4H8). Experimentally, t-C4H is found to transfer a proton to acetone but not to isopropanol. Since the value of Semeluk and Lossing for AHƒ(t-CH‡) appears to have been determined very accurately, the P.A. of isopropanol must be 195 kcal/mol unless the proton transfer reaction involving t-C.H is exothermic and exhibits an extremely high activation energy, which we feel is unlikely. On this basis, the value for the P.A. of isopropanol of 193 kcal/mol (807 kJ/mol) quoted by Beauchamp [14] seems reasonable. Under results we pointed out that the termolecular proton transfer reaction was not observed within our sensitivity limits, although reactions of this type have been observed previously for a variety of alkane and cycloalkane molecular ions [12c]. Based on the value of 36 kcal/mol (150 kJ/mol) [21] for the overall exothermicity of the solvation reaction process H3O+H2O ⇒ H†(H2O), and an average value for the AH (H3O) of 145±5 kcal/mol (606±21 kJ/mol) derived from various sources, Reaction 8 would be approximately 4 kcal/mol (16 kJ/mol) Lossing and Semeluk. However, the apparent exoexothermic taking the value for AHƒ(t-C4H§) given by thermicity is equivalent in magnitude to the error limits associated with AH (H3O+); consequently the negative experimental result and the apparent exothermicity are not necessarily contradictory, particularly if the overall reaction is essentially thermoneutral. We cannot comment on Reaction 4, since the overall enthalpy change associated with the solvation process |