Protonation Thermochemistry of Ethyl Halides

4 downloads 0 Views 164KB Size Report
Organic halides are extensively used in numerous fields of condensed- and ... determination of the gas-phase basicity of ethyl halides was as large as 50 kJmolА1 and, ...... they are adsorbed and flattened (as the chain sketched in. [a] Prof.
[20] P. J. MacDougall, M. B. Hall, Trans. Am. Cryst. Assoc. 1990, 26, 105. [21] Z. Shi, R. J. Boyd, J. Chem. Phys. 1988, 88, 4375. [22] R. F. W. Bader, P. J. MacDougall, C. D. Lau, J. Am. Chem. Soc. 1984, 106, 1594. [23] R. F. W. Bader, P. J. MacDougall, J. Am. Chem. Soc. 1985, 107, 6788. [24] R. F. W. Bader, P. L. A. Popelie, C. J. Chang, J. Mol. Struct. (Theochem) 1992, 255, 145. [25] R. F. W. Bader, Atoms in MoleculesÐa Quantum Theory, Claredon, Oxford, UK, 1990. [26] Y. Aray, J. Rodríguez, Surf. Sci. Lett. 1998, 405, L532. [27] Y. Aray, J. Rodríguez, J. Rivero, D. Vega, Surf. Sci. 1999, 441, 344. [28] P. Blaha, K. Schwarz, J. Luitz, WIEN 97, Vienna University of Technology, 1997. (Improved and updated Unix version of the original copyrighted WIEN code, which was published by P. Blaha, K. Schwarz, P. Sorantin, S. B. Trickey, Comput. Phys. Commun. 1990, 59, 399). [29] J. P. Perdew, S. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865. [30] AIMPAC, R. F. W. Bader, P. Krugg, Department of Chemistry, Mc Master University, Hamilton, Ontario, Canada, 1990 (http://www.chemistry.mcmaster.ca/aimpac/). [31] Y. Aray, J. Rodríguez, R. Lopez-Boada, J. Phys. Chem. 1997, 101, 2178. [32] RuS2 : H. D. Lutz, B. Mueller, T. Schmidt, T. Sting, Acta Crystallogr. Sect. C 1990, 46, 2003. [33] PdS: N. E. Brese, P. J. Squatrio, J. A. Ibers, Acta Crystallogr. Sect. C 1985, 41, 1829. [34] MoS2 : K. D. Bronsema, J. L. de Boer, F. Z. Jellinek, Z. Anorg. Allg. Chem. 1986, 540, 15. [35] NbS2 : F. Jellinek, G. Brauer, H. Muller, Nature 1960, 185, 376. [36] Rh2S3 : E. Parthe, D. Hohnke, F. Hulliger, Acta Crystallogr. 1967, 23, 832 [37] TcS2 : H. J. Lamfers, A. Wieger, J. L. de Boer, J.A.L.C.E. 1996, 241, 34. [38] T. Hahn, International Tables for Crystallography, Kluwer Academic, Boston, 1996. [39] The optimized geometries and the electronic densities of the N2 and KF molecules were calculated with the ADF program (ADF 99, Scientific Computing Modelling, Chemistry Department, Vrije Universiteit, Netherlands, 1999), by using a full-electron Slater Basis set of triple-z quality contained in the ADF package. The Laplacian topology was determined by using a local version of BUBBLE adapted to the ADF program. [40] R. F. W. Bader, W. Henneker, J. Am. Chem. Soc. 1965, 87, 3063. [41] F. Frechard, P. Sautet, Surf. Sci. 1995, 336, 149. [42] Y. Aray, J. Rodríguez, D. Vega, J. Phys. Chem. B 2000, 104, 5225. [43] P. Raybaud, J. Hafner, G. Kresse, H. Toulhoat, Surf. Sci. 1998, 407, 237. [44] FASTRUCTURE, Release 1.0, MSI. Inc., San Diego, CA, 1999. [45] V. Smelyansky, J. Hafner, G. Kresse, Phys. Rev. B 1998, 48, R1782. Received: February 26, 2001 [Z 196]

Protonation Thermochemistry of Ethyl Halides Guy Bouchoux,*[a] François Caunan,[a] Danielle Leblanc,[a] Minh Tho Nguyen,[b] and Jean-Yves Salpin[c] KEYWORDS: ab initio calculations ´ ethyl halides ´ gas-phase basicities ´ proton affinities ´ thermokinetic methods Organic halides are extensively used in numerous fields of condensed- and gas-phase chemistry. The knowledge of their thermochemical properties and chemical reactivity is consequently of general importance. In the gas phase, protonated halides and dialkylhalonium ions are readily produced by ion ± molecule reactions involving the RX (X ˆ F, Cl, Br, I) compounds. Unimolecular[1] and bimolecular[2±7] reactions involving protonated methyl halides in the gas phase have been reported in the literature, and various aspects of the reactivity of dimethylhalide cations were documented.[8] Concerning the protonation thermochemistry of organic halides, only halomethanes and haloethanes have been examined.[2±4] However, the most recent study devoted to halomethanes[2] reveals that the previous estimates of the proton affinities of these molecules[3] were seriously overestimated. For the four CH3X molecules, the estimates of the gas-phase basicities, GB(CH3X), given by Beauchamp et al.[3] are situated from 8 to 31 kJ molÿ1 higher than the values obtained by McMahon et al.[2] Moreover, most of the time, the accurate GB value[2] is situated in the lower part of, or below, the bracketing range used by Beauchamp et al.[3] These observations raise some doubt on the validity of the presently accepted proton affinity values of haloethanes,[9] which were inferred from indications given in ref. [3]. Furthermore, we note also that the bracketing window used in ref. [3] for the determination of the gas-phase basicity of ethyl halides was as large as 50 kJ molÿ1 and, consequently, large uncertainties are associated with these measurements. The present study was undertaken with the intention to provide new and precise experimental and theoretical proton affinities and gas-phase basicities for the series of haloethanes, C2H5X. Experiments were done on a Fourier-transform ion cyclotron resonance (FT-ICR) [a] Prof. Dr. G. Bouchoux, F. Caunan, Dr. D. Leblanc DeÂpartement de Chimie Laboratoire des MeÂcanismes ReÂactionnels Ecole Polytechnique, 91128 Palaiseau cedex (France) Fax: (‡ 33) 1-69-33-30-41 E-mail: [email protected] [b] Prof. Dr. M. T. Nguyen Department of Chemistry University of Leuven Celestijnenlaan 200F, 3001 Leuven (Belgium) [c] Dr. J.-Y. Salpin UMR 8587 ªAnalyse et environnementº Institut des Sciences Universite d'Evry Val d'Essonne Boulevard Francois Mitterrand, 91025 Evry Cedex (France)

604

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

CHEMPHYSCHEM 2001, No. 10

mass spectrometer; the theoretical data were obtained by means of ab initio molecular orbital calculations at the G2 level. Both results agree nicely with each other and, as expected, show substantial differences from previous literature values which appear to be overestimated by 10 ± 30 kJ molÿ1. Computational Results One of the major feature of the protonation process of haloalkanes is the spectacular CÿX (X ˆ halogen) bond elongation when passing from the neutral to the protonated form. This phenomenon, which was demonstrated more than ten years ago,[16] is shown in Figure 1 for the haloethyl series. As already noted, the relative CÿX bond elongation increases with the electronegativity of the basic center. In the case of fluorine, the bond extension attains  25 % of its original value and it may be considered that protonated ethyl fluoride consists in a complex between a classic ethyl cation and a molecule of hydrogen fluoride. At the opposite end, the CÿI bond is only 3 % larger in protonated ethyl iodide than in the neutral molecule. A consequence of the existence of a large CÿX bond in protonated ethyl halides is that the corresponding rotational barriers are small. Therefore, the contribution to entropy and to internal energy of these motions should be considered in detail. In fact, at the MP2/6-31G* level, the barriers for rotation around the CÿX bond in protonated ethyl halides are in the range of 2 ± 4 kJ molÿ1; consequently the corresponding torsional mode is  close to a free rotation. This is confirmed by the fact that the Sÿ ÿ  term calculated using Pitzer's approximation (S p ) is within  1 J Kÿ1 molÿ1 of the value given by the free-rotation model (Sÿfr ; see the Computational Section). A summary of the entropy calculations on neutral and protonated ethyl halides is given in Table 1. From a general point of view, we found that the difference DSÿp (C2H5X) ˆ Sÿ[C2H5XH]‡ ÿ Sÿ(C2H5X) is close to a common value of 30 J Kÿ1 molÿ1, which is essentially due to the difference in vibrational contributions. The appearance of a free rotation when passing from C2H5X to [C2H5XH]‡ is responsible for a large part (17 ± 23 J Kÿ1 molÿ1) of DSÿp (C2H5X). The remaining (18 ± 7 J Kÿ1 molÿ1) is due to the two low frequencies associated with the deformation of the HXCC frame. Concerning the absolute Sÿ values, one may first observe that the method of entropy calculation adopted here provides numbers in close agreement with experiment. When a compar-

Figure 1. Optimized MP2/6-31G* bond lengths [Š] and angles [8] of neutral (left) and protonated (right) ethyl halides: a) ethyl fluoride, b) ethyl chloride, c) ethyl bromide, d) ethyl iodide.

ison is possible, namely for the neutral ethyl halides, the deviation is less than 1.2 J Kÿ1 molÿ1 (Table 1). This agreement leads to the expectation that Sÿ[C2H5XH]‡ values are also accurately estimated by our treatment based on the Pitzer model. The rotational barriers of the methyl groups have been found to be similar in both the neutral and protonated molecules, thus no significant effect of this torsional mode appears in the DSÿp (C2H5X) differences. A second important observation is that the entropy of the protonated ethyl halides Sÿ[C2H5XH]‡ is not comparable to the entropy of the corre-

Table 1. Calculated entropy of neutral and protonated ethyl halides [J Kÿ1 molÿ1]. Sÿ

C2H5F

C2H5FH‡

C2H5Cl

C2H5ClH‡

C2H5Br

C2H5BrH‡

C2H5I

C2H5IH‡

translational rotational vibrational CH3 torsion XH torsion

157.0 93.2 6.5 9.0 ±

157.3 96.5 20.9 9.3 17.1

160.6 98.2 9.2 8.6 ±

160.8 99.8 16.3 8.0 20.2

167.1 101.4 11.1 8.6 ±

167.3 102.7 17.7 8.2 21.6

171.7 103.5 12.8 8.9 ±

171.8 104.6 18.9 8.3 22.8

Sÿ total Sÿ exp.[a]

265.7 264.8

301.1 [282.4][c]

276.6 (277.6)[b] 276.6

305.1 (305.7)[b] [297][d]

288.2 (288.5)[b] 287.0

317.5 (317.8)[b]

296.9 295.8

326.4

[a] From ref. [19]. [b] Values calculated for the heaviest isotope are given in parentheses. [c] For comparison, Sÿ(C2H5OH); see ref. [19]. [d] For comparison, Sÿ(C2H5SH); see ref. [19].

CHEMPHYSCHEM 2001, No. 10

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

605

sponding isoelectronic species. For example, the value of Sÿ[C2H5FH]‡ is 301.1 J Kÿ1 molÿ1 (Table 1) and larger than (experimental value 282.4;[17] calculated Sÿ(C2H5OH) ÿ1 ÿ1 279.1 J K mol ) by no less than 18 J Kÿ1 molÿ1; a smaller difference holds for [C2H5ClH]‡ and C2H5SH (8 J Kÿ1 molÿ1, Table 1). This fact is essentially related to the longer CÿX bond in protonated halides, which is thus more spectacular in the fluorine case. By comparison, the rotational barrier in protonated ethyl fluoride is equal to only 1.9 kJ molÿ1 whereas it attains 7.7 kJ molÿ1 for ethanol (MP2/6-31G* calculations), and the corresponding participations to entropy are 17.1 and 13.5 J Kÿ1 molÿ1, respectively. Most importantly, the Sÿ terms due to the contribution of the vibrational modes are equal to 20.9 and 6.9 J Kÿ1 molÿ1 for [C2H5FH]‡ and C2H5OH, respectively, in keeping with the very low frequency numbers associated with the CCFH deformations. The 298 K proton affinities of ethyl halides were calculated using the usual relationship expressed in Equation (1), where E represents the total energy at 0 K, D298Hÿ the difference and Hÿ298 (C2H5X) ÿ Hÿ0 (C2H5X) ÿ (Hÿ298 [C2H5XH]‡ ÿ Hÿ0 [C2H5XH]‡ ) 6.2 kJ molÿ1 represents the translational enthalpy of the proton (52RT). PA(C2H5X) ˆ E(C2H5X) ÿ E[C2H5XH]‡ ‡ D298Hÿ ‡ 6.2 kJ molÿ1

(1)

The G2 method was utilized in order to obtain accurate total energies E[C2H5XH]‡ and E(C2H5X). The D298Hÿ terms were estimated using the standard statistical-thermodynamic equations and the scaled HF/6-31G* vibrational frequencies; the CÿXH torsions of the protonated species were treated as free internal rotations. These terms are not completely negligible, since the free rotation and the low frequencies associated with the CÿXH moieties contribute strongly to the vibrational part of Hÿ298 [C2H5XH]‡ . A summary of the calculated thermochemical data concerning the protonation of ethyl halides is presented in Table 2. The comparison with the experimental proton affinity values will be discussed in the following Section.

Experimental Results The gas-phase ion ± molecule reactions of pure ethyl halides yield a number of ions, among which are C2H5XH‡ and (C2H5)2X‡. Thus, both protonation and alkylation processes participate to the ion ± molecule chemistry of these species. Moreover, the

Table 2. Summary of calculated and experimental 298 K thermochemical data relevant to the protonation of ethyl halides. X

DSÿp [a] [J Kÿ1 molÿ1] calcd

D298Hÿ[b] [kJ molÿ1] calcd

PA(C2H5X) [kJ molÿ1] calcd exp

F Cl Br I

35.4 28.5 29.2 29.5

ÿ 3.7 ÿ 2.4 ÿ 2.3 ÿ 2.2

653.4 678.4 686.6 708.6

± 679.5  1.4[c] 684.3  1.6[c] 709.3  2.5[c]

(683.4)[d] (693.4)[d] (696.2)[d] (724.8)[d]

[a] DSÿp (C2H5X) ˆ Sÿ[C2H5XH]‡ ÿ Sÿ(C2H5X). [b] DHÿ298 ˆ (Hÿ298 [C2H5XH]‡ ÿ Hÿ0 [C2H5XH]‡ ) ÿ (Hÿ298 (C2H5X) ÿ Hÿ0 (C2H5X)). [c] This work, see text. [d] From the compilation by Hunter and Lias.[9]

606

latter process is generally dominant and this behaviour renders difficult a direct determination of the basicity of C2H5X by measurement of the equilibrium constant of a proton-transfer reaction such as in Equation (2). C2H5XH‡ ‡ B ÿ! C2H5X ‡ BH‡

(2)

For this reason, the data reported in ref. [3] were obtained by a bracketing technique in an ICR mass spectrometer. However, even with that technique, the low intensity of the C2H5XH‡ signals and the occurence of competitive reactions may affect the observation of Equation (2), particularly if the experiments are done in a mixture of alkyl halide and reference base B as recalled in the Introduction. To avoid such problems, the present study makes use of the advantage of a FT-ICR mass spectrometer equipped with an external ion source to separate the region where the reactant ion is produced from that where it reacts with a selected base. The ions of interest were produced in the chemical-ionization mode using methane as the protonating reagent. After introduction of a pulse of the ion mixture formed in the external source into the ICR cell, the reactant ions were selected and allowed to react by the usual procedures (see Experimental Section). The determination of the basicity of ethyl halide C2H5X has been done using the ªthermokineticº method,[17, 18] which constitutes a major improvement of the bracketting technique. Briefly, it consists in measuring the rate constant of Equation (2) for a set of bases B and then fitting the reaction efficiency as a function of the gas-phase basicities GB(B). The reaction efficiency RE is given by Equation (3), where kexp and kcoll are the experimental and collision rate coefficients, respectively, DGÿ1 is the standard free-energy change of Equation (2) and, consequently, DGÿ1 ˆ GB(C2H5X) ÿ GB(B), and DGÿa is an apparent energy barrier for Equation (2). RE ˆ kexp/kcoll ˆ 1/[1 ‡ exp((DGÿ1 ‡ DGÿa †/RT)]

(3)

The value of GB(C2H5X) can be deduced by plotting the RE of Equation (2) for a series of bases B of known basicities as a function of GB(B) and by fitting the data with a parametric function accounting for the theoretical relationship given in Equation (4). RE ˆ a/[1 ‡ exp(b(c ÿ GB(B))]

(4)

It has been established empirically[17] and frequently used with success[18] that DGÿa  RT and, consequently, that GB(C2H5X) ˆ c ÿ 1/b. The sets of reactions studied for the three commercially available ethyl halides, C2H5I, C2H5Br, and C2H5Cl, are summarized in Tables 3 ± 5. Their experimental GB values were determined from the curves presented in Figures 2 ± 4. The determination of the proton affinities can thus be done either by a curve fitting of the reaction efficiencies, RE versus PA(B), or more easily by using the relationship GB(C2H5X) ˆ PA(C2H5X) ÿ TDSÿ1 . In both cases the entropy difference DSÿp (C2H5X) ˆ Sÿ(C2H5XH‡) ÿ Sÿ(C2H5X) is needed. In the present treatment, the proton affinities are

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

CHEMPHYSCHEM 2001, No. 10

directly deduced from the corresponding gas-phase basicity obtained from Equation (4) by adding the corresponding TDSÿ1 term. The latter has been estimated using the third-law entropies calculated as described previously. Ethyl iodide: The basicity of ethyl iodide has been estimated between that of propene and H2S in ref. [4]. This suggestion has been retained in the compilation by Hunter and Lias,[9] who used GB(C3H6) ˆ 722.7 and GB(H2S) ˆ 673.8 kJ molÿ1, and obtained GB(C2H5I) ˆ 698.3  24.5 kJ molÿ1. A proton affinity value of PA(C2H5I) ˆ 724.8 kJ molÿ1 was then deduced assuming that the protonation entropy DSÿp (C2H5I) is equal to 20 J Kÿ1 molÿ1 (namely DSÿp (C2H5I)  DSÿp (C2H5Cl)).[9] Obviously the tabulated GB and PA values of ethyl iodide are associated with a large uncertainty due to the large gap between the basicities of propene and hydrogen sulfide. In our experiments, a set of nine bases B (Table 3) were considered and submitted to proton transfer with protonated ethyl iodide, as depicted by Equation (2). Our measurement of the gas-phase basicity GB(C2H5I) by the thermokinetic method (Figure 2) gives a value of 685.7  2.5 kJ molÿ1. The calculated entropy term, DSÿp (C2H5I) ˆ 29.5 J Kÿ1 molÿ1 (Table 2), is slightly higher than the approximate value used by Hunter and Lias.[9] This allows a proton affinity PA(C2H5I) of 709.2  2.5 kJ molÿ1 to be estimated. The agreement between the new experimental result and the calculated value of 708.6 kJ molÿ1 (G2 level, Table 2), is really good and confirms that the tabulated[9] GB(C2H5I) and PA(C2H5I) values should be revised downward by about 13 kJ molÿ1. Ethyl bromide: This molecule was claimed to have a basicity comparable to that of H2S and CH3I.[3] The GB values of the two latter molecules are 673.8 and 665.5 kJ molÿ1, respectively, and it has been concluded that GB(C2H5Br) may be approximated by the mean value of 669.7 kJ molÿ1.[9] The proton affinity value quoted in ref. [9] (PA(C2H5Br) ˆ 696.2 kJ molÿ1) has been calculated using DSÿp (C2H5Br)  DSÿp (C2H5Cl) ˆ 20 J Kÿ1 molÿ1. We investigated eight proton-transfer reactions between protonated ethyl bromide and a base B (Table 4). Our determination of GB(C2H5Br) by the thermokinetic method (Figure 3) leads to a value of 660.5  1.6 kJ molÿ1, namely a basicity close to that of water. Considering the theoretical protonation entropy DSÿp (C2H5Br) ˆ 29.2 J Kÿ1 molÿ1, we deduce that PA(C2H5Br) ˆ

Figure 2. Normalized reaction efficiency as a function of GB(B) for [C2H5IH]‡ ‡ B ! C2H5I ‡ [BH]‡ . Fitting parameters [kJ molÿ1]: b ˆ 0.31  0.07, c ˆ 689.0  1.0.

684.3  1.6 kJ molÿ1, a value in excellent agreement with the G2level estimate of 686.6 kJ molÿ1 (Table 2). It appears again that the PA(C2H5Br) value tabulated in ref. [9] is overestimated by  10 kJ molÿ1. Ethyl chloride: The basicity of ethyl chloride has been suggested to lie between those of water and of the groups H2S, CH3I, and C2H5Br.[3] Using GB(H2O) ˆ 660.0 and GB(H2S) ˆ 673.8 kJ molÿ1, Hunter and Lias[9] concluded that GB(C2H5Cl) ˆ 666.9 kJ molÿ1. Using DSÿp (C2H5Cl) ˆ Sÿ(C2H5SH) ÿ Sÿ(C2H5Cl) ˆ 20 J Kÿ1 molÿ1, a proton affinity value of PA(C2H5Cl) ˆ 693.4 kJ molÿ1 was proposed.

Table 3. Parameters relevant to proton-transfer reactions involving [C2H5IH]‡ ions produced by chemical ionization of ethyl iodide and several bases B. Base B

PA(B)[a] [kJ molÿ1]

GB(B)[a] [kJ molÿ1]

m[b] [D]

a[c] [Š3]

kexp( 1010) [cm3 molÿ1 sÿ1]

kcoll( 1010)[d] [cm3 molÿ1 sÿ1]

RE[e]

CH3I H2O CF3CH2OH 1,2,3,4-C6H2F4 CF3CO2H CH2O CF3COCH3 1,4-C6H4F2 CF3COOCH3

683.1 691.0 700.2 700.4 711.7 712.9 723.9 721.8 740.5

665.5 660.0 669.9 672.7 680.7 683.3 692.0 692.8 709.6

1.64 1.82 2.03 2.8 2.28 2.33 3.0 0 (1.8)

7.6 1.45 5.48 9.97 5.50 2.8 6.37 10.2 7.14

0.14 0 0.085 0.08 0.20 3.52 8.31 6.91 9.84

10.5 17.3 12.0 14.7 12.4 18.2 15.3 9.2 11.2

0.01 0.00 0.01 0.01 0.02 0.22 0.61 0.85 1.00

(12.0) (22.1) (14.5) (17.7) (15.1) (22.9) (19.0) (13.0)

(0.02) (0.01) (0.01) (0.02) (0.20) (0.58) (0.99) (1.00)

[a] Proton affinities PA, and gas-phase basicities GB from the compilation by Hunter and Lias.[9] [b] Dipole moment; experimental values from ref. [20] (values in parentheses estimated from comparison with homologous compounds). [c] Polarizabilities calculated using the method of Miller.[21] [d] Collision rate constant calculated using the ADO model[22] and, in parentheses, the VTST formalism.[23] [e] Normalized reaction efficiency, RE ˆ (kexp/kcoll)/(kexp/kcoll)max .

CHEMPHYSCHEM 2001, No. 10

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

607

Table 4. Parameters relevant to proton-transfer reactions involving [C2H5BrH]‡ ions produced by chemical ionization of ethyl bromide and several bases B. Base B

PA(B)[a] [kJ molÿ1]

GB(B)[a] [kJ molÿ1]

m[b] [D]

a[c] [Š3]

kexp( 1010) [cm3 molÿ1 sÿ1]

kcoll( 1010)[d] [cm3 molÿ1 sÿ1]

RE[e]

C2H4 (CF3)2CHOH H2O C6HF5 CF3CH2OH 1,2,3,4-C6H2F4 CF3CO2H CF3COCH3

680.5 686.6 691.0 690.4 700.2 700.4 711.7 723.9

651.5 656.2 660.0 662.7 669.9 672.7 680.7 692.0

0 (2.5) 1.82 1.6 2.03 2.8 2.28 3.0

4.25 7.37 1.45 10.1 5.48 9.97 5.5 6.4

0 0 3.31 3.87 8.93 11.1 9.25 11.4

10.1 13.9 17.9 12.0 12.9 16.1 13.5 16.5

0 0 0.28 0.46 1.00 1.00 1.00 1.00

(16.7) (22.7) (13.3) (15.4) (19.4) (16.4) (20.6)

(0.26) (0.50) (1.00) (0.98) (0.97) (0.97)

[a] Proton affinities PA and gas-phase basicities GB from the compilation by Hunter and Lias.[9] [b] Dipole moment; experimental values from ref. [20] (value in parentheses estimated from comparison with homologous compounds). [c] Polarizabilities calculated using the method of Miller.[21] [d] Collision rate constant calculated using the ADO model[22] and, in parentheses, the VTST formalism.[23] [e] Normalized reaction efficiency, RE ˆ (kexp/kcoll)/(kexp/kcoll)max .

Figure 3. Normalized reaction efficiency as a function of GB(B) for the reaction [C2H5BrH]‡ ‡ B ! C2H5Br ‡ [BH]‡ . Fitting parameters [kJ molÿ1]: b ˆ 0.47  0.07, c ˆ 662.6  0.3.

Figure 4. Normalized reaction efficiency as a function of GB(B) for the reaction C2H5Cl ‡ [BH]‡ ! [C2H5ClH]‡ ‡ B. Fitting parameters [kJ molÿ1]: b ˆ 1.8  0.5, c ˆ 656.2  0.2.

In the present study, seven proton-transfer reactions have been monitored with ethyl chloride acting as the neutral base and [BH]‡ ions as protonating agents. Our choice to investigate the reverse of Equation (2) for this species was guided by technical reasons. In fact, the basicity of C2H5Cl is in a GB region where only a few compounds may serve as reference base B. Moreover, among these species only one (CS2) has a known dipole moment. Since a precise value of the dipole moment of the neutral species is needed in the calculation of the collision rate constant, it was of interest to consider C2H5Cl as the neutral reactant. Under these conditions the thermokinetic graph is obviously reversed (Figure 4). The treatment of our data provides

a gas-phase basicity of GB(C2H5Cl) ˆ 655.6  1.4 kJ molÿ1; thus ethyl chloride appears to be less basic than water. Using the calculated entropy difference DSÿp (C2H5Cl) ˆ 28.5 J Kÿ1 molÿ1, we obtain a proton affinity value of PA(C2H5Cl) ˆ 679.5  1.4 kJ molÿ1. Satisfactorily enough, the G2-level calculation leads to a comparable value of PA(C2H5Cl) ˆ 678.4 kJ molÿ1 (Table 2). The discrepancy with the earlier tabulation[9] lies at about 14 kJ molÿ1. Ethyl fluoride: The basicity of ethyl fluoride was suggested to be situated between those of water and ethene.[3] Using GB(H2O) ˆ 660.0 and GB(C2H4) ˆ 651.4 kJ molÿ1, a mean value of GB(C2H5F) ˆ 655.8 kJ molÿ1 has been proposed by Hunter and Lias.[9] Using DSÿp (C2H5F) ˆ Sÿ(C2H5OH) ÿ Sÿ(C2H5F) ˆ

608

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

CHEMPHYSCHEM 2001, No. 10

16 J Kÿ1 molÿ1, a proton affinity value of 683.4 kJ molÿ1 can thus be derived (Table 5). Unfortunately, no commercial sample of ethyl fluoride was available for he present experimental investigation. However, G2-level calculations provide a proton affinity value of

Table 5. Parameters relevant to proton-transfer reactions involving [BH]‡ ions produced by chemical ionization of several bases B and neutral ethyl chloride. Base B

PA(B)[a] GB(B)[a] kexp( 1010) kcoll( 1010)[d] RE[e] [kJ molÿ1] [kJ molÿ1] [cm3 molÿ1sÿ1] [cm3 molÿ1sÿ1] [%]

CH3Br (CF3)3COH (CF3)2CHOH CS2 C2H5Br (CF3)2CCH3OH C6HF5

662.3 678.6[c] 685.9[c] 681.9 684.3[b] 691.2 690.4

638.0 648.5[c] 655.5[c] 657.7 660.5[b] 660.9 662.7

12.7 13.1 9.98 0.99 0 0 0

15.8 13.8 14.4 16.6 15.4 14.2 14.4

(18.8) (16.4) (17.1) (19.7) (18.3) (16.9) (17.1)

0.84 (0.84) 1 (1) 0.73 (0.73) 0.06 (0.05) 0 0 0

[a] Proton affinities PA and gas-phase basicities GB from the compilation by Hunter and Lias[9] except where otherwise specified. [b] This work. [c] Using the most meaningful PA and GB values from ref. [9]. [d] Collision rate constant calculated using the ADO model[22] and, in parentheses, the VTST formalism.[23] Calculations use m(C2H5Cl) ˆ 2.05 D[20] and a(C2H5Cl) ˆ 6.4 Š3.[21] [e] Normalized reaction efficiency, RE ˆ (kexp/kcoll)/(kexp/kcoll)max .

653.4 kJ molÿ1 (Table 2) and, using the calculated entropy difference DSÿp (C2H5F) ˆ 35.4 J Kÿ1 molÿ1, a gas-phase basicity GB(C2H5F) ˆ 631.5 kJ molÿ1 arises. It is difficult to assign an error bar to the PA and GB calculated by the G2 method in the specific case of ethyl fluoride; one suggestion is to use the standard deviation of about 6 kJ molÿ1 observed for the G2 ion test set.[12b] Thanks to these possible deviations, our results indicate that the true values of the proton affinity and gas-phase basicity of ethyl fluoride are below the tabulated figures[9] by no less than  30 kJ molÿ1. Concluding Remarks We have revisited the gas-phase basicities and proton affinities of the halogenated ethanes for which both proton-transfer reaction-rate measurement and high-level ab initio calculations gave consistent values. Experimental gas-phase basicities have been derived from the measurement of reaction rates using the thermokinetic method (GB(C2H5X) ˆ 655.6  1.4, 660.5  1.6, and 685.7  2.5 kJ molÿ1 for X ˆ Cl, Br and I, respectively). Proton affinities were deduced after consideration of the entropy change associated with the protonation process. The consequence of the CÿX bond elongation following protonation has explicitly been taken into account in the estimate of entropy and enthalpy changes. Experimental and theoretical proton affinities are in excellent agreement with each other (averaged values: PA(C2H5X) ˆ 653, 679, 685 and 709 kJ molÿ1 for X ˆ F, Cl, Br and I, respectively, with a deviation of less than 2 kJ molÿ1). These novel results confirm that, similar to the case of halomethanes, the currently tabulated values of GB(C2H5X) and PA(C2H5X) should be revised downward by 10 to 30 kJ molÿ1. CHEMPHYSCHEM 2001, No. 10

Experimental Section FT-ICR experiments were performed on a mass spectrometer (Bruker Spectrospin CMS 47X) equipped with an external ion source.[15] C2H5I and C2H5Br were ionized in the external ion source in the chemicalionization mode using methane as the protonating gas. Typical source conditions were filament current 180 mA, electron energy 100 eV, and ionizing pulse duration 100 ms. All ions were transferred to the reaction cell located inside the 4.7 T superconducting magnet. Selection of the [C2H5XH]‡ ion of interest was done by ejection of unwanted ions by a combination of chirp and soft rf pulses. The reactants were relaxed to thermal energy (T ˆ 300 K) by introducing argon inside the ICR cell at a pressure approximately one order of magnitude greater than the pressure of the neutral reactant. A relaxation delay of 2 ± 6 s after selection of the reacting ions was typically used. Subsequently, the selected ions were treated for a variable time with neutral base. Experiments were conducted at a constant pressure of neutral reactant in the range 10ÿ8 ± 10ÿ7 mbar. For C2H5Cl the reverse procedure has been used: The bases were protonated in the external ion source and treated with neutral C2H5Cl in the ICR cell. All the samples used in theses experiments were commercially available (Aldrich Chemical).

Computational Methods Ab initio calculations have been carried out using the Gaussian 98 series of programs.[10] Standard G2 theory[11] employs a geometry optimized at the MP2(full)/6-31G(d) level and a scaled (by a factor 0.893) HF/6-31G(d) zero-point energy (ZPE). A base energy calculated at the MP4/6-311G(d,p) level is corrected by several additivity approximations to approach the QCISD(T) energy with the 6-311 ‡ G(3df,2p) basis set. In an attempt to account for residual basis-set deficiencies, G2 theory introduces higher-level corrections (HLC) that depend on the number of paired and unpaired electrons. The G2 formalism yields, in general, reliable heats of formation, ionization energies, and proton affinities.[11, 12] An average deviation of 6.6 kJ molÿ1 has been obtained on the heats of formation of a set of 148 molecules using the G2 method[12a] and a similar result is obtained for ionization energies and electron affinities (a standard deviation of 5.8 kJ molÿ1 is observed for a series of 146 examples[12b] ). The calculation of the third-law entropies uses standard statistical thermodynamic formulae through a procedure comparable to the E2 method described by Radom and co-workers.[13] Each vibrational contribution to entropy was computed according to the standard Equation (5), where q ˆ hn/kB and the scaled harmonic vibrational frequencies n calculated at the HF/6-31G(d) level are used. Sÿ ˆ R[(q/T)/(eq/T ÿ 1) ÿ ln (1 ÿ eÿq/T)]

(5)

Entropies for internal rotations were computed by using the hindered-rotor model developed by Pitzer.[14] In this approach, the energy levels of a rotor associated with a potential energy barrier of the form V0/2(1 ÿ cos nf), where f is the dihedral angle, are found with the help of the one-dimensional Schrödinger equation. The results are presented as a function of two dimensionless variables V0/RTand 1/Qfr (namely the reciprocal of the partition function for the free rotation). In practice, the entropy of a given rotor is obtained by addition of a corrective term to the entropy calculated under the free-rotor approximation Sÿfr , Equation (6), where Ired is the reduced moment of inertia of the two rotating groups around the axis containing the twisting bond. Sÿfr ˆ 12R ln (8p3 eIredkBT/n2h2)

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

(6)

609

It is generally observed that the free-rotor approximation leads to correct entropy estimate if the rotational barrier V0 is less than about 1.5 RT, and that Pitzer's model correctly works for V0 values situated above this limit.[13, 14] In the present study, the required rotational potential-energy barriers V0 were obtained at the MP2/6-31G(d) level using a relaxed-rotation approach (that is, all geometrical parameters were optimized except for the dihedral angle considered). The detailed geometries and vibrational frequencies used in the present work are available upon request from the authors.

[1] A. J. R. Heck, L. J. de Koning, N. M. M. Nibbering, Int. J. Mass Spectrom Ion Processes 1991, 109, 209. [2] M. N. Glukhovtsev, J. E. Szulejko, T. B. McMahon, J. W. Gauld, A. P. Scott, B. J. Smith, A. Pross, L. Radom, J. Phys. Chem. 1994, 98, 13 099. [3] J. L. Beauchamp, D. Holtz, S. D. Woodgate, S. L. Patt, J. Am. Chem. Soc. 1972, 94, 2798. [4] T. B. McMahon, P. Kebarle, Can. J. Chem. 1985, 63, 3160. [5] R. Houriet, E. Rolli, A. Maquestiau, R. Flammang, G. Bouchoux, Org. Mass Spectrom. 1987, 22, 770. [6] G. Vaidyanathan, M. Y. M. Lyktey, J. J. Stry, R. L. DeLeon, J. F. Garvey, J. Phys. Chem. 1994, 98, 7475. [7] L. S. Nichols, M. L. McKee, A. J. Illies. J. Am. Chem. Soc. 1998, 120, 1538. [8] See, for example: H. W. Zappey, T. Drewello, S. Ingemann, N. M. M. Nibbering, Int. J. Mass Spectrom. Ion Processes 1992, 115, 193. [9] E. P. L. Hunter, S. G. Lias, J. Phys. Chem. Ref. Data 1998, 27, 413. [10] Gaussian 98 (Revision A.6), M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1998. [11] L. A. Curtiss, K. Raghavachari, G. W. Trucks, J. A. Pople, J. Chem. Phys. 1991, 94, 7221. [12] a) L. A. Curtiss, K. Raghavachari, P. C. Redfern, J. A. Pople, J. Chem. Phys. 1997, 106, 1063; b) L. A. Curtiss, P. C. Redfern, K. Raghavachari, J. A. Pople, J. Chem. Phys. 1998, 109, 42. [13] a) A. L. L. East, L. Radom, J. Chem. Phys. 1997, 106, 6655; b) A. L. L. East, B. J. Smith, L. Radom, J. Am. Chem. Soc. 1997, 119, 9014. [14] K. S. Pitzer, W. D. Gwinn, J. Chem. Phys. 1942, 10, 428. [15] P. Kofel, M. Alleman, H. P. Kelerhals, K. P. Wanczek, Int. J. Mass Spectrom. Ion Processes 1985, 65, 97. [16] M. Alcami, O. MoÂ, M. YaÂnÄez, J.-L. Abboud, J. Elguero, Chem. Phys. Lett. 1990, 172, 471. [17] G. Bouchoux, J.-Y. Salpin, D. Leblanc, Int. J. Mass Spectrom. Ion Processes 1996, 153, 37. [18] a) G. Bouchoux, J.-Y. Salpin, J. Phys. Chem. 1996, 110, 16 555; b) G. Bouchoux, J.-Y. Salpin, J. Am. Chem. Soc. 1996, 118, 6516; c) M. Witt, H.-F. Grützmacher, Int. J. Mass Spectrom. Ion Processes 1997, 164, 93; d) A. Ricci, M. Rosi, J. Phys. Chem. A 1998, 110, 16 555; e) B. K. Decker, N. G. Adams, L. M. Babcock, Int. J. Mass Spectrom. Ion Processes 1999, 185, 727; f) G. Bouchoux, J.-Y. Salpin, Rapid Commun. Mass Spectrom. 1999, 13, 932; g) M. Mormann, J.-Y. Salpin, D. Kuck, Eur. Mass Spectrom. 1999, 5, 441; h) G. Bouchoux, D. Leblanc, Eur. J. Mass Spectrom. 2000, 6, 109; i) G. Bouchoux, B. Gaudin, D. Leblanc, M. YaÂnÄez, O. MoÂ, Int. J. Mass Spectrom. 2000, 199, 59; j) F. Bernardi, F. Cacace, G. Occhioucci, A. Ricci, I. Rossi, J. Phys. Chem. A 2000, 104, 5545; k) M. Mormann, S. Bashir, P. J. Derrick, D. Kuck, J. Am. Soc. Mass Spectrom. 2000, 11, 544. [19] S. W. Benson, Thermochemical Kinetics, 2nd ed., Wiley, New York, NY, 1976. [20] A. C. McClellan, W. H. Freeman, Table of Experimental Dipole Moments, Vol. 1, Freeman, San Fransisco, CA, 1963; Table of Experimental Dipole Moments, Vol. 2, Rahara Enterprises, El Cerrito, CA, 1973. [21] K. J. Miller, J. Am. Chem. Soc. 1990, 112, 8533.

610

[22] T. Su, M. T. Bowers, Gas-Phase Ion Chemistry, Vol. 1 (Ed.: M. T. Bowers), Academic Press, New York, NY, 1979, chap. 3. [23] T. Su, W. Chesnavich, J. Chem. Phys. 1982, 76, 5183. Received: March 23, 2001 [Z 206]

The Desorption Process of Macromolecules Adsorbed on Interfaces: The Force Spectroscopy Approach Matteo Conti,[c] Yasser Bustanji,[a] Giuseppe Falini,[b] Paolo Ferruti,[d] Sergio Stefoni,[c] and Bruno Samorì*[a] KEYWORDS: interfaces ´ polymer adhesion ´ scanning force microscopy ´ single molecules The adsorption and desorption of macromolecules on interfaces are multifaceted phenomena, which play essential roles in many technologies, such as those based on lubrication, on material adhesion and coating, and also in medicine for the development of artificial organs. The conformations assumed by the polymer molecules at the interfaces (commonly described in terms of loops, trains and tails, as sketched in Figure 1),[1, 2] determine the configurational entropy and enthalpy of the adsorption processes and hence the structural organisation, the thickness and the physical properties of the resulting depositions. In this Communication, we demonstrate that the scanning force microscope (SFM) in the force spectroscopy mode makes it possible to monitor, at the level of single molecules, how the extent of their cross-linking affects the adhesion energy and their conformations upon adsorption at interfaces. The conformations that single DNA molecules assume when they are adsorbed and flattened (as the chain sketched in [a] Prof. B. Samorì, Dr. Y. Bustanji Dipartimento di Biochimica UniversitaÁ di Bologna Via Irnerio 48, 40126 Bologna (Italy) Fax: (‡ 39) 051-354387 E-mail: [email protected] [b] Dr. G. Falini Dipartimento di Chimica G. Ciamician UniversitaÁ di Bologna (Italy) [c] Dr. M. Conti, Prof. S. Stefoni Dipartimento di Scienze Nefrologiche Ospedale S. Orsola, Bologna (Italy) [d] Prof. P. Ferruti Dipartimento di Chimica Organica e Industriale UniversitaÁ di Milano (Italy)

 WILEY-VCH-Verlag GmbH, D-69451 Weinheim, 2001 1439-4235/01/02/10 $ 17.50+.50/0

CHEMPHYSCHEM 2001, No. 10