6 pages

Stabilities of halonium ions from a study of gas-phase equilibria R+ + XR' = (RXR

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Stabilities of halonium ions from a study of gas-phase equilibria R+ + XR' = (RXR
  J. Am. Chem. SOC. 985, 107, 3151-3162 3151 vicinity of the positive charge and this solvent exclusion reduces the nonspecific solvation. The same should be true for Nb' (relative to t-Bu+ or c-Pe+). The above observations have some bearing on the Nb' con- troversy (see preceding section). The AH2 s01) changes based on Arnett's data39 or reactions 15-20 were given in the preceding section. Clearly the solution data appear to present a less dramatic case for the special stability of Nb'. Thus Nb' is less stable than t-Bu' in solution reaction 15, while the opposite is true in the gas phase. For the change i-Pr', c-Pe', Nb', reactions 16 and 17, one observes also a reduced trend in solution. Thus the change c-Pe' to Nb+ is only -4 kcal/mol more exothermic than the change i-Pr' to c-Pe+ while in the gas phase the value is -6 kcal/mol. Also the changes in the methyl substituent series reactions 18-20 are less pronounced in solution. When one considers the above AH changes in solution one must remind oneself that they are used as models to the stability of Nb' in connection with the exo-endo solvolysis rate difference^ ^^ The experimental solvolysis rates establish a difference of about 6 kcal/mol between the exo and endo transition states (R'-X-)* in solution.30 In the nonclassical view only the exo transition state is significantly stabilized by bridging. Since the exo state is electronically more stabilized one expects that the nucleophilic solvent stabilization will be somewhat less for that state than for the endo state. As was pointed out above some differential nu- cleophilic stabilization is probably affecting the Arnett solution data. Thus, in this respect these data may be suitable models for the exo-endo norbornyl transition states. However, it was seen that Arnett's data are much more strongly affected by differential nonspecific solvation, Le., solvent exclusion by bulky hydrocarbon structures that lead to poorer solvation. This change of general solvation present in the Arnett results makes the energy changes in eq 16-20 in solution not suitable for modeling of the exeendo transition-states energies. No gross changes of the size of ion occur for the exo relative to the endo transition state, and thus no significant changes of nonspecific solvation for these transition states can be expected. Thus Arnett's solution data, if they are to be applied for modeling energy changes of exo-endo transition states, should be corrected for the presence of nonspecific solvation. The corrections, which cannot be quantitatively assessed, will be in the direction of greater agreement with the gas-phase results and increased support of reactions 16-20 energy change models for the unusual stability of Nb+ and by implication also of the exo-Nb+-X- transition state in solution. Registry No. Cl-, 16887-00-6; H-, 12184-88-2; PhCH,', 671 1-19-9; t-Bu+, 14804-25-2; -FPhCH,', 291 80-23-2; o-MePhCH,', 63246-55-9; p-MePhCH,', 57669-14-4; PhCHCI+, 56683-65-9; o-MePhCH,CI, 552-45-4; p-MePhCH,Cl, 104-82-5; PhCCl,, 98-07-7; p-MeOPhCH,CI, 824-94-2; PhCHCH,', 25414-93-1; (CH,)$H+, 19252-53-0; m- FPhCH,+, 65108-06-7; o-FPhCH,+, 65108-14-7; p-CIPhCH,', 29180- 24-3; m-MePhCH,', 60154-94-1; PhCC12+, 4154-22-1; AdC1, 935-56-8; 2-NbC1, 29342-53-8; Ad', 19740-18-2; Nb+, 24321-81-1; I-C1-c-Pe, 930-28-9; 2-Me-NbC1, 96246-73-0; 1-Me-c-Pe+, 17106-22-8; c-Pe+, 25076-72-6. Stabilities of Halonium Ions from a Study of Gas-Phase Equilibria R+ XR' = (RXR')+ Dilip K. Sen Sharma, Sarah Meza de Hojer,+ and Paul Kebarle* Contribution from the Chemistry Department, University of Alberta, Edmonton, Alberta, Canada T6G 2G2. Received December 11, 1984 Abstract: The gas-phase ion equilibria R+ B = RB', where R' = Et', i-Pr+, c-Pe', t-Bu+, 2-Me-2-Bu', and 2-Nb' and B = CH3C1, CH2C12, CHC12, CHCI,, S02F2, CF3H, and CF4 were determined in a pulsed electron beam high pressure mass spectrometer. van't Hoff plots provide AGO,,, AHo, and ASo. For the chloronium ions the following trends were observed. The bond energy D(R'-CIR,), where R' changes and R, is constant, decreases with increasing electronic stabilization of R+, Le., in the order Me', Et', i-Pr', c-Pe', t-Bu', Nb'. The same order was observed earlier in this laboratory for D(R+-CI-), i.e., the chloride affinity of R'. However, the changes of D(R+-CIR,) for R+ = 2-Me-2-Bu+, Nb', and t-Bu' are very small. This means that little differential, specific nucleophilic solvation of these ions in solution is to be expected when solvents of low nucleophilicity like CH2CI2 and S0,CIF are used. The bond energies D(Me'-CIR) increase in the order R = Me, Et, i-Pr, t-Bu. The bond energies D(t-Bu'-B) decrease in the order B = C2H5C1, CH2C1, = CH,CI, CC13H, S02F2, CF3H, CF4. The significance of these trends is discussed. Measurements of ion equilibria in the gas phase' include ac- ceptor-donor (Lewis acid-base) equilibria of the type in eq 1, R' + B = RB+ R' XR' = (RXR')' where R' is a carbocation and B a u, r r n donor base. De- termination of the equilibrium constant K, with a pulsed electron beam high pressure mass spectrometer leads via van't Hoff plots to the corresponding AGO AHo,, and ASo,. The present work describes results for systems where R = ethyl (Et), isopropyl (i-Pr), tert-butyl (t-Bu), 2-methyl-2-butyl (t-Pe), cyclopentyl (c-Pe), and 2-norbornyl (Nb), while B = XR' = CH3C1, CH2C12, CHCI,, C2H5C1, CHF,, CF,, and S02F2 The Chemistry Department, University of Mexico, (1) Permanent address: Mexico City, Mexico. 0002-7863/85/1501-3751 01.50/0 work is an extension of measurements described in an earlier publicationZ in which reactions involving Et+, i-Pr+, and MeCl were studied. The present results allow one to observe the change of bonding in R+-B with increasing stabilization of the carbocation R+. They also give the bonding changes for a given R+ with changing donor character (nucleophilicty) of B. Chloronium ions are important alkylating agents in ~olution.~ heir usefulness as alkylating agents in the gas phase was pointed out re~ently.~ Thus, the dimethylchloronium ion can be used4 for the clean gas-phase preparation of tertiary oxonium and quaternary am- monium ions as shown in reaction 2. The product ions can then (1) Kebarle, P. Annu. Rev. Phys. Chem. 1977, 28, 455. (2) Sen Sharma, D. K.; Kebarle, P. J. Am. Chem. Soc 1978, 100,5826. (3) Olah, G. . Halonium Ions ; Wiley: New York, 1975. 4) Sen Sharma, D. K.; Kebarle, P. J. Am. Chem. Sor 1982, 104, 19. 1985 American Chemical Society  3758 J. Am Chem SOC., Vol. 107, No. 13, 1985 CH,ClCH,+ + (CH3)ZO = CH3C1 (CH3)30+ (2) Sen Sharma et al. be used in other measurements. Thus Meot-Ner5 has studied the gas-phase hydration of Me4N+ prepared in the above manner. More recently many Me+ transfer equilibria between two bases B were determined and a Me+ affinity ladder established in this laboratory.6 The reagents most often used for the preparation of Me+B were the dimethylchloronium and the dimethyl- fluoronium ions. Alkylation reactions like (2) also can be useful in analytical chemical ionization work, but this potential has not been realized yet. Alkyl cation transfer reactions involving halonium ions and bases B are SN2 eactions. The gas-phase kinetics of these reactions have been studied4 and have proven of considerable interest to the development of the theory of gas-phase SN2 on-molecule reaction^.^ Knowledge of the energetics provided by equilibria 1 is of significant utility in all the above cases. Arnett and co-worker~~*~ ave measured enthalpy changes AH3 for the ionization processes 3 in S02C1F and CH2CI2 olutions. RCl + SbCls = R+ SbCl,+ (3) The AH3 were then used to evaluate enthalpy changes for the chloride-transfer reactions 4 and relative heats of formation 4) 5) AHr(R+) of the ions in s~lution.~~~ n the basis of comparisons with available corresponding data in the gas phase, Arnett et al. concluded that there is little differential solvation of the carbenium ions R+ in solution. More recently, relative chloride affinities in the gas phase were determined in this laboratory.I0 These pro- vided a somewhat better basis for comparison with the solution data8s9 and indicated that significant differential solvation can be present in some cases. To be able to separate the (differential) solvent effect into the components of specific nucleophilic solvation of R+ by one solvent molecule and the nonspecific solvation by the solvent dielectric, information on the energetics of ion solvent molecule complexes like R+C1CH2CI s required. The energetics provided by measurement of equilibria 1 also prove useful for this purpose. Experimental Section The measurements of the equilibrium constant K, or equilibria 1-R+ B = RB+ (where R+ were carbocations like c-Pe+, t-Bu+, etc., and B various halo compounds like CH,CI, CH2C12, CHF,, etc.)-were exec- uted with a pulsed electron beam high ion source pressure mass spec- trometer which has been described previously.'. Briefly, the principle of the method is as follows. A short pulse of 2000 V electrons produces the primary ionization in the temperature-controlled ion-source-reaction chamber maintained at some 2-8 torr of total pressure. The ions grad- ually diffuse to the walls where they become discharged. During their motion through the gas they also engage in ion-molecule reactions. By proper choice of neutral reactant concentrations, the reactions may be speeded up such that the ions reach ion-molecule reaction equilibria. The relative ion concentrations are monitored by allowing some of the gas to bleed out through a very narrow slit (1 X 0.01 mm) into an evacuated chamber, where the ions are captured by electric fields and subjected to conventional mass spectrometric analysis and detection. Results and Discussion (a) Description of Reaction Systems in Which Equilibria R+ + B = RB+ Were Observed. The required ions R+ were generated ROf + RC1 = RoC1 R+ AH4 = AH3(R+) AH3(R0+) (5) Meot-Ner (Mautner), M; Deakyne, C. A. J. Am. Chem. SOC., n (6) McMahon, T. B.; Nicol, G.; Heinis, T.; Kebarle, P., in preparation. (7) Farneth, W. E.; Brauman, J. I. J. Am. Chem. SOC. 976, 98 5546. Olmstead; Brauman, J. I. Ibid 1977, 99, 219; 1979, 101, 3715. Caldwell, G.; Magnera, T. F.; Kebarle, P. J. Am. Chem. SOC. 984, 106, 959. (8) Arnett, E. M.; Petro, S. C. J. Am. Chem. SOC. 978, 100, 2563, 5402, 5408. Arnett, E. M.; Petro, S. C.; Schleyer, P. v. R. Ibid 1979, 101, 329. (9) Arnett, E. M.; Pienta, N. J. J. Am. Chem. SOC. 980, 102, 3399. Arnett, E. M.; Hofelich, T. C. Ibid 1982, 104, 522. (10) Sharma, R. B.; Sen Sharma, D. K.; Hiraoka, K.; Kebarle, P. J. Am. Chem. SOC., receding paper in this issue. (11) Cunningham, A. J.; Payzant, J. D.; Kebarle, P. J. Am. Chem. SOC. 1971, 93, 7627. preess. 05 (m ec) IO Figure 1 Ion intensities in percent of total ion current of major ions, observed after an ionizing electron pulse in 4 torr of CH, containing 72 mtorr of C3H8 and 1.16 mtorr of CH2CI2. Temperature -4 OC. i-C3H7+ is obtained by reactions of the ultimate ions in methane, CH5+, and C2H5+ with propane. The iPr+ engages in the following reaction: C3H7+ + CH2CI2 = C3H7CICH2C1+, hich reaches equilibrium. A side reaction occurs and also reaches equilibrium: C3H7+ C3H8 = C3H7-C3H8+. Both adduct ions are slowly drained by exchange reactions with the strong base H20 present as a minor impurity. in -4 torr of methane. The ultimate ions in pure methane are12 CH5+ and C2H5+. The ions R+ were generated by reactions of CH5+ and C2H5+ with suitable reagent gases added to the methane. As an example we will discuss the setup involved in the measurement of the following equilibrium: i-C3H7+ + CHzC12 = i-C3H7C1CH2CI+ see Figure 1). To create the isopropyl cation some 70 mtorr of propane were added to 4 torr of methane. This results in a very fast conversion of the CH5+ and C2HS+ o C3H7+ by reactionsi2 6 and 7. In the presence of small amounts of B = CH2C12, (6) and (7) are followed by (8), Le., the formation of CH5+ + C3H8 = i-C3H7+ + H2 + CH, C2H5' + C3H8 = C2H6 + i-C3H7+ i-C3H7+ CH2C12 = i-C3H7ClCH2CI+ i-C3H7+ C3Hs = i-C3H7.C3H8+ 6) (7) (8) (9) the desired chloronium ion. Reaction 8 is third body dependent. Reactions 2 and 3 are complete in microseconds and therefore CHs+ and C2H5+ do not appear in Figure 1. The C3H7+ on, initially - 00% of the total ion current, disappears rapidly by reaction 8 forming the chloronium adduct. The adduct formation of C3H,+.C& by (9) is also observed. At times longer than -0.4 ms the ratio C3H7CICH2Cl+/C3H7+ ecomes constant. This must mean that reaction 8 reaches equilibrium. After a similar time reaction 9 also reaches equilibrium. The concentrations of iso- propyl, the chloronium adduct, and the propane adduct are seen to gradually decrease while hydrates are formed. Since water is much stronger base than either methylene chloride or propane, the displacement of these bases by water, present as a minor impurity in the apparatus, leads to reactions 10-12. The C3H&H2Cl,+ + OH2 = C3H7*0H2CH2C12+ (10) (1 1) (12) C3H7*OHyCH2C12+ + OH2 = C3H7(0H2)2+ + CH2C12 C3H7-C3H8+ OH2 = C3H7(0H2)+ + C3H8 etc. equilibrium constant for the chloronium equilibrium 8 was ob- tained by using the constant ratio of the ions C3H7C1CH2C1+ nd (12) (a) Munson, M. S. B.; Field, F. H. J. Am. Chem. SOC. 965.87, 3294. (b) n-C3H7+ s also produced by reactions 6 and 7. This either isomerizes to i-C3H7+ r is converted by hydride abstraction from propane to i-C3H,+.  Stabilities of Halonium Ions C3H: 50 r 1000 - qr loo J. Am. Chem. SOC. Vol 107, No. 13, 1985 3759 - 67 u 52 -40 1 -32' . 05 10 m sec) Figure 2 Same reaction mixture and reaction mechanism as in Figure 1, but at the lower temperature (-14 C). Both equilibria now contain larger equilibrium concentrations of the adduct ions C3H7CICH2CI+ nd C3H7C3H,+ elative to C3H7+. C3H7+ bserved at longer reaction times. Since reaction 8 is much faster than the hydration reaction 10 (see Figure l), the equi- librium concentrations of the above ions should be little disturbed by the slow removal via (10). The connentration changes shown in Figure 2 were obtained with the same gas mixture as used in Figure 1 but at a lower (-14 C) temperature. Correspondingly, the concentration ratio C3H7+ClCH2C1+/C3H,+ s much higher, as expected, since re- action 8 is exothermic, Le., the equilibrium constant Klo ncreases with decreasing temperature. Changes of the methylene chloride pressures by a factor of 2 or more, in separate runs, had no effect on the equilibrium constant. The equilibrium 8 was observed at several different temperatures in between 29 and 20 C. The determination of the equilibria 8 described above is fairly typical of the reactions encountered in all determinations of equilibria 1. The t-Bu+ ion was produced by the addition of smail amounts of isobutane to the methane. The c-Pe+ was produced via hydride abstraction by i-Pr+ from cyclopentane. The desired reaction sequence was obtained by adding propane in roughly tenfold excess over cyclopentane in the major methane gas. With this mixture i-Pr+ is produced by reactions 2 and 3 and then i-Pr+ abstracts H- from cyclopentane to give c-Pe+. The hydride ab- straction by i-Pr+ s considerably less exothermic than the direct reactions of CH5+ nd C2H5+ with cyclopentane. Since these more exothermic reactions, in principle, could produce acyclic C5H,+ isomers, the milder two-step route was chosen. The 2-methyl-2-butyl and the 2-norbornyl cations were also produced by milder, two-step processes. Thus Nb+ was prepared in a mixture of 5 torr of CH4, 20 mtorr of propane, and 4 mtorr of norbornane. The i-Pr+, produced by the first two gases, hydride abstracts from norbornane, and this reaction leads to Nb+. Hydride abstractions involving alkyl cations and various RH have been described by Solomon, Meot-Ner, and Field.I3 A more complex but still tractable reaction system occurred in the measurement of the equilibria involving Et+ and CH2C12. The ions observed are shown in Figure 3. The Et+ resulting from methane engages in the desired equilibrium reaction 13. However, Et+ also engages in reaction 14. This normally very slow reaction C2H5+ CH2CI2 = (C2H,CH2C12)+ C2H5+ + CH4 = C3H7' H2 (13) (14) CHS' + CHzCl2 = CH2Cl+ + HC1 CH, CH2Cl' + CH2CI2 = CH3C1 CHC12' (15) (16) observed earlier in this laboratoryi5 has a positive temperature 13) Solomon, J. J.; Field, F. H. J. Am. Chem. Sor. 1965,87, 1567; 1975, 97, 625. 05 O (msec) Figure 3. Reactions occurring in the experiments for determination of equilibrium C2HS+ CH2C12 = (C2HSCH2Cl2)+. on source pressures: 4 torr of CHI, 13 mtorr of CH2C12. Temperature = 252 OC. The C2Hs+ involved in the equilibrium can also react with CHI at this temperature: C2Hf CH4 = C3H7+ H2. However, this reaction is slower than the equilrhrium rates. Pressure C2H5CI [m torr) Figure 4. Equilibrium constants for the reaction f-Bu+ + EtCl = t-Bu- CIEt'. Dependence on pressure of EtCl. dependence and at the elevated temperature of the experiments, 252 C, becomes sufficiently fast to remove a significant fraction of Et+ (see Figure 3). An analogue computer simulation of reaction systems, very similar to reactions 13 and 14 performed in the earlier work,2 shows that reaction 14 would cause the (C2HSCH2Cl2)+/C2H,+ atio to increase relative to the equilibrium 13 ratio. However, the increase is very small and can be neglected. A side reaction that can be observed in Figure 3 is the pro- duction of CHC12+ by the hydride abstraction 16. From the half-life of (16) in Figure 3 one can estimate a k16 = 8 X lo-'' molecules-' cm3 s-I. This reaction was not observed in an ICR study of Lias and AusloosI6 presumably because it was too slow for detection at the low CH2C12 pressures used. The lack of dependence of the equilibrium constants Kl on the pressure of B is illustrated in Figure 4, for the equilibrium t-Bu+ EtCl = BuEtCP. Similar results were obtained for the other systems. (b) Binding Energies of Chloronium Ions R'CIR'. The results from the measurements of the equilibrium constants Kl are summarized in the van't Hoff plots shown in Figures 5-7. The results are divided into three groups. Figure 5 shows a constant B = CH2C12 and a cnanging R+, Figure 6 gives a constant R+ = t-Bu+ and changing B, and Figure 7 gives results for two different B (CH3C1 and CH2C12) and three different ions R+ (Et+, i-Pr', t-Bu+). 14) Meot-Ner, M.: Solomon, J. J.; Field, F. H. J. Am. Chem. SOC. 976, 15) Hiraoka, K.; Kebarle, P. J. Chem. Phys. 1975, 63 394. 16) Lias, S. G ; Ausloos, P. nt. J. Mass. Spectrom. Ion Phys. 1977, 23, 98 1025. 213.  3760 J. Am. Chem. SOC., Vol. 107, No. 13, 1985 Sen Sharma et al. Table I. Thermochemical Data for Halonium Ions AH,^ -AH,' AH,' -a , -AGI9 (298), reaction MIND0/3 (exptl) (alt) cal/deg kcal/mol 1. CH,CI.Me' = CH3CI + Me' 2. CH,CIEt' = CH3CI + Et' 31.4 30.7e 30.7 29.4' 22Se 4. CH~CI-C-BU' = CHjCl ?-Bu' 13.8 8.4 9.3 19.3 2.6 5 CH2C12Et' = CH2CI2 + Et' 36.0 33.2 45.1 22.6 6. CH2C12+Pr' = CH2C12 i-Pr' 15.6 14.9 31.0 6.4 7. CH2CI2.c-Pe' = CH2CI2 + c-Pe' 9.8 10.6 20.2 3.8 8. CH2C12*t-But = CH2C12 + ?-But 9.5 10.0 22.3 2.8 9. CH,CI,.f-Pe' = CH2CI2 t-Pet 9.5 9.8 23.3 2.5 10. CH2CI.Nb' = CH2C12 + Nb' 10.6 9.8 31.1 1.3 11. CHpCI*?-Bu' = CHjCI t-Bu' 13.8 8.4 9.9 19.3 2.6 13. CH2Cl2.t-But = CH2C12 + ?-But 12.6 9.5 10.0 22.3 2.8 15. S02Fz.t-Bu' = SO2F2 Z-BU' 10.4 11.5 10.4 1.6 16. CHFx-t-Bu' = CHF, + C-BU' (44.0)d 6.8 7.7 19.8 0.9 64.2 3. CH,CI.i-PR = CH3Cl i-Pr' 21.3 22.9' 22.4 43.1e 10.1e 12. C2H5C1*t-But = C2HSC1 + t-Bu' 17.4 9.2 10.6 16.3 4.3 14. CHCI,*t-Bu' = CHC13 t-Bu 10.2 9.1 9.5 23.6 2.1 17. CF4.iBu' = CF4 ;-Bu' (40.1 d 3.4 5.7 10.4 0.3 All data without suoerscriuts are from uresent work all energies are in kcal/mol. From MIND0/3 calculations of Affkreactants): AHdMe') = 260.3; AHdEt') = 2b5.7 (dridged structire); AHdsec-Pr') = 789.7; AHf(t-BUt) = 170.8; AHdMeCi) = -15.4; AH,(EtCl) = -26.0; AHf(CH2C12) = -22.3; AHf(CHCI3) = -26.0. 'AH(expt1) was obtained from the slope of the van? Hoff plots. AH(alt) is an alternate value, considered more reliable, calculated from AG0298(exptl) nd a AS(alt) = (26 + ASo(exptl))/2, Le., the AS'(a1t) is assumed to be an average of the experimental AS' from the van't Hoff plot and a constant AS = 26 eu. This is an averaging procedure based on the observation that the entropy change for this type of association reactions is often close to -26 eu. dMIND0/3 calculations, present work. The MIND0/3 method is obviously unreliable for fluoronium ions. e Previous work from this laboratory.2 F t 1° 40 50 00 1 '~'''~1'''11''1'1~11'~~~~'~ CH2Cl2+R+: RCICH2CI+ 11~x10~ Figure 5. van't Hoff plots of equilibrium constants for the reactions R' CH2C12 = (RCICH,CI)'. Stability of chloronium ions decreases as stabilization of R increases. These results reproduce stability order for R+: i-Pr, c-Pet, ?-But, Nb' observed from measurements of chloride affinities of R (Sharma'O). 1 t-C4 H;+RX -C4 HpXR' IT x 103 Figure 6. van't Hoff plots of equilibrium constants for the reactions t-Bu' + B = (r-BuB)+. Stability of adducts decreases in the order B = EtC1, CHZCIZ, CHClp CHpCI, SOzFz, CHF,, CF,. It is interesting to note the wide range of temperatures which had to be covered in order to observe the different equilibria 1. Thus a temperature as high as 390 OC was required for the most strongly bonded adduct C,H,+-ClCH,, while temperatures as low as -160 OC were used in measurements of weakly bonded adducts 10000 bl Figure 7. van't Hoff plots of equilibria R' R'CI = (RCIR')'. involving stabilized R' like t-Bu nd a very weakly nucleophilic B like CF, (see Figures 5-7). The AH1', AG10(298), and ASlo values obtained from the van't Hoff plots in Figures 5-7 are given in Table I. Results from two equilibria measured earlier2 are also included. MIND0/3 predicted AHlo for a number of reactions were obtained by calculating the MIND0/3 (Dewar18) enthalpies of formation of the reactants Rf, R'X, and (RXR')'. The results for these calculations are also given in Table I. McManus19 has also calculated MIND0/3 energies involving various chloronium ions. Since the geometry optimization procedure is part of the MIND0/3 program,'* essentially identical results to those of McManus19 were obtained whenever calculations on the same systems were performed. Comparing the MINDO results with experiment for the re- action series 1-4, Table I, involving Rf = Et', i-Pr+, -But, and CH,Cl, one finds good agreement particularly for Etf and i-Pr'. This suggests that the MINDO result for D(Me+-ClMe) = 64.2 kcal/mol is also reliable to a few kcal/mol. This energy change is too high to be determined experimentally via van't Hoff plots (17) Complete geometry optimization was performed. The resulting (18) Bingham, R. C.; Dewar, M. J.; Lo, D. H. J. Am. Chem. SOC. 975, (19) McManus, S. . J. Org. Chem. 1982, 47, 070 structures are available on request from one of the authors (P.K.). 97, 1307.  Stabilities of Halonium Ions J. Am. Chem. SOC., ol. 107, No. 13, 1985 3761 Table 11 Methyl Cation Affinities of CIR = D(Me+-CIR)' and D(R+-ClR) D Me+-ClR) D(R+-CIR) R AHf( R+) AHf(RCI)' AHf(MeCIRC)d exptld (MINDO/3)C (MIND0/3)/ Me 261.0 -20.6 64 64 Et 215.6 -26.1 164.4 71 72 38 i-Pr 191.0 -33.6 150.1 77 79 30 t-Bu 166.5 -43.1 136.6 81 85 23 All values in kcal/mol. bRosenstock.22 Cox and Pilcher.21 dCalculated from AHl, reactions 2-4, Table I, and literature data.21,22 McManus.I9 /Present work. 5 I I 150 16G 170 180 I90 D(R+ CI-) (kcal./mole) Figure 8. Plot of heterolytic bond dissociation energies D(R+-ClCH,CI) vs. D(R+-Cl-). Data show that D(R+C1CH2Cl) decreases as D(R+-CI-) decreases, Le., as stability of R+ increases. However, for relatively stable ions R+, (R+-CICH,CI) is almost constant (R+ = i-Pe', Nb+, t-Bu', and even c-Pe'). This shows that there is little differential nucleophilic solvation of these R+ by the weakly nucleophilic solvent molecule CH2C12. of the association equilibria 1. An experimental result based on Me cation transfer equilibria6 which is in agreement with the MIND0/3 value will be reported in the near future. The results for the series R+-C1CH3 show that the bond energy decreases rapidly with increasing stability of R+, Le., in the order Me', Et', i-Pr+, t-Bu+. The overall change being from -64 kcal/mol for Me+ to -9 kcal/mol for t-Bu+. The same type change is also observed in the series R+-CICH2C1 given by reactions 5-10 (Table I). The results for this series are based on the van? Hoff plots of Figure 5 which very graphically display the decrease of stability of Rf-C1CH2C1 with increasing stability of R+. It is interesting to note that these results give the stability order Et', i-Pr+, c-Pe+, t-Bu+, Nb', Le., place the nominally secondary norbornyl cation at a higher stability than the tertiary species, t-Bu+. This result is in line with measurements of the gas-phase hydridel0%l3 nd chloridei0 affinities of the above R+ ions which have shown that Nb' has a lower hydride and chloride affinity than t-Bu+, Le., is more stable than t-Bu+. A plot of D(R+-CICH2CI) vs. D(Rf-C1-), i.e., the chloride affinity of R+, is shown in Figure 8. The chloride affinities are from ref 10 except that for t-Pe+.20 The decrease of D(R+- CICH,CI) with decrease of D(R+-Cl-) slows down as the stability of R+ increases. Thus t-Pe+, Nb+, t-Bu+, and even c-Pef have almost the same D(R+-C1CH2C1) values. The differences for the corresponding free energy changes (-AGIO, see Table I) are somewhat larger but still relatively small, Le., within about 2 kcal/mol. The formation of the chloronium ions R+C1CH2C1 can be considered as representing the specific nucleophilic solvation of R+ by the solvent CH2CI2. Thus, the small changes for the more stable Rf ions observed above mean that the specific sol- vation of R+ by weakly nucleophilic solvents like CH2C1, is almost equally strong as long as R+ is a fairly stable cation. Notice that the above generalization would not at all hold for R+ = Et+ or Me+ (see Figure 8). The above findings are in agreement with deductions made by Ar~~ett~,~ nd othersz3 that there is little (20) n he basis of the hydride affinity of Field, AH,(t-Bu+) = 166.5 kcal/molZ2 and AHf(r-PeC1) = -48.4 kcal/mol. differential (nucleophilic) solvation of carbocations R+ in weakly nucleophilic solvents like CH2C12 and S02C1F. A recent com- parisonlo of the stabilities of carbocations in the gas phase with Arnett'sss9 solution data indicates that there can be significant differential nonspecific solvation. Bulky cations like 1-adamantyl' were found to be significantly less stable when compared to t-Bu+ in CH2C12 and SOzCIF solution than in the gas phase. This observation was attributed to poor general solvation of ions which, due to their bulky structures, displace significant amounts of the solvent dielectric from the vicinity of the ionic charge center.I0 Examined above were bonding changes in the series R+-ClR, where the nature of Rf was changed while R, was kept constant. Bonding changes for the situation R,+-ClR where the leaving cation is kept constant while the nature of RCl is changed are given in Table I1 for Mef-CIR where R = Me, Et, i-Pr, t-Bu. These values were obtained from the experimental AHi for re- actions 2-4 (Table I) and literature data.21.22 Also given in Table I1 are the MIND0/3 results for the same bonds obtained by McMan~s.'~ he two sets of results are seen to be very close. Examining the data one finds that D(Me+-ClR) increases as R changes in the order R = Me, Et, i-Pr, t-Bu. Electron donation by R to C1 increases in the above order and should be the factor responsible for the observed Me+-ClR stability increases. Another series Rof-CIR is present in Table I reactions 11-17, where R,+ = t-Bu+ and RC1 = MeC1, EtCI, CH2C12, CHC13. Included for the same R, are three solvent molecules, which contain no chlorine: SO2F,, CF,H, and CF4. The observed increase of D(t-Bu+-ClR) for R = Me to Et is in the same di- rection as that observed for the series Me+-ClR (Table 11). The changes for t-Bu+ and the three chloromethanes, CH3C1, CH2C12, and CHC1, (Table I), are small and somewhat erratic, so a significant regular trend cannot be deduced for this series. The corresponding MIND0/3 results predict a small decrease of the t-Bu+-ClR bond energy with increased chlorine substitution. This trend, if true, would be in line with the expectation that substituents on R with an electron-withdrawing field-inductive effect, like C1, will decrease the Ro+-CIR bond energy. In fact, this expected trend is observed in the experimental results for the pair Et+-ClCH3, Etf-CICH2C1 and the pair i-Prf-C1CH3, i- Pr-ClCH,Cl (see Figure 7). While the exact AH values are probably not sufficiently accurate to reveal the trend, the position of the van't Hoff lines shows a weaker bond (free) energy for dichloromethane. Thus, the reversal observed for t-Bu+ (Figure 7) is unexpected and its cause unclear. The fluoromethanes CF3H and CF4 are seen to bond much more weakly to t-Buf than the chloromethanes (Table I). This result is expected since the formation of a halonium ion on com- bination with R+ causes the participating halide atom to acquire some positive charge. The greater electronegativity of fluorine relative to chlorine results in weaker bonding R+-XR' for the fluoronium ion. This electronegativity effect overpowers the greater ability of fluorine to form covalent bonds. Me+ affinity studies6 reveal the same trend. Thus D(MeC-CIMe) was deter- mined to be some 10 kcal/mol higher than D(Me+-FMe). The t-Bu+-CF4 bond energy is found to be lower than that of t- Bu+-CF3H (Table I, Figure 6) which is expected since the (21) Cox, J. D.; Pilcher, G Thermochemistry of Organic and Organo- metallic Compounds ; Academic Press: New York, 1970. (22) Rosenstock, H. M.; Buff, R.; Ferreira, A. A,; Lias, S. G.; Parr, A. A,; Stockhauer, R. L.; Holmes, J. L. J. Am. Chem. SOC. 982, 104 2337. (23) Taft, R. W. Prog. Phys. Org. Chem. 1983, 4, 247.
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks