(HCN)m(NH3)nH+ clusters formed by the reaction of carbon vapor with jet-cooled ammonia

(HCN)m(NH3)nH+ clusters formed by the reaction of carbon vapor with jet-cooled ammonia

Chemical Physics Letters 372 (2003) 121–127 www.elsevier.com/locate/cplett ðHCNÞm ðNH3ÞnHþ clusters formed by the reaction of carbon vapor with jet-c...

248KB Sizes 0 Downloads 8 Views

Chemical Physics Letters 372 (2003) 121–127 www.elsevier.com/locate/cplett

ðHCNÞm ðNH3ÞnHþ clusters formed by the reaction of carbon vapor with jet-cooled ammonia G. Raina, G.U. Kulkarni, C.N.R. Rao

*

Chemistry and Physics of Materials Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur P. O., Bangalore 560 064, India Received 7 January 2003; in final form 20 February 2003

Abstract New cluster species of the type ðHCNÞm ðNH3 Þn Hþ with m up to 4 and n up to 7 alongwith ðNH3 Þn Hþ species are observed using mass spectrometry, on the reaction of carbon vapor with jet-cooled NH3 in admixture with helium. The most preponderant species correspond to m ¼ 1 and 2, viz. ðHCNÞ1 ðNH3 Þ4 Hþ , ðHCNÞ1 ðNH3 Þ3 Hþ and ðHCNÞ2 ðNH3 Þ3 Hþ . These clusters involving the tetrahedral coordination in the first solvation shell of the NHþ 4 ion are the most stable species, as corroborated by molecular-orbital calculations. The incremental complexation energy due to successive addition of the NH3 molecule to ðHCNÞðNH3 Þn Hþ , decreases monotonically with increasing n, for n > 2. Ó 2003 Elsevier Science B.V. All rights reserved.

1. Introduction Mass spectrometry has been used extensively to study reactivity of carbon clusters including fullerenes [1]. Bare carbon clusters larger than C40 , in particular C60 , are essentially found to be unreactive with respect to NH3 , SO2 , NO2 , H2 , CO and O2 [2,3]. On the other hand, smaller carbon clusters react with NH3 and CH3 CN molecules to produce polar cyanopolyynes species [4]. It was proposed that CN radicals formed as intermediate species, react with free protons to give rise to cyanopolyynes with H and CN adding to the ends of linear carbon chains. Indeed, atomic carbon in the

*

Corresponding author. Fax: +91-80-846-2766. E-mail address: [email protected] (C.N.R. Rao).

vapor state reacts readily with NH3 and N2 to produce the CN radical, which has been studied by various spectroscopic techniques [5–9]. Shevlin and co-workers [10,11] have reported the formation of HCN by co-condensing arc-generated carbon vapor and NH3 : 

þ

C þ NH3 ! CANHþ 3 ! HACANH2 ! H2 C@NH ! HCN þ H2

ð1Þ

In this Letter, we present a time-of-flight mass spectrometric analysis of the reaction products formed by co-expanding through a pulsed supersonic valve, the carbon vapor produced by laser ablation of graphite and a NH3 –He mixture. We observe that in the presence of NH3 , the nucleation of carbon clusters is completely suppressed and instead, adducts of the type ðHCNÞm

0009-2614/03/$ - see front matter Ó 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0009-2614(03)00375-0

122

G. Raina et al. / Chemical Physics Letters 372 (2003) 121–127

ðNH3 Þn Hþ are formed (m up to 4; n up to 7), the relative intensity of the clusters depending on the NH3 =He ratio employed. These species are protonated in contrast to adducts of the type ðHCNÞn ðNH3 Þ obtained by co-expanding HCN and NH3 seeded in a mixture of argon and neon through a pulsed nozzle [12,13]. We also carried out molecular-orbital calculations to understand the stability of these clusters. The study has yielded new results considering that a previous study [4] of this reaction was brief, reporting a limited number of product species.

2. Experimental We have used an indigenously built apparatus [14], consisting of a cluster generation chamber, which is connected to a linear time-of-flight (TOF) mass spectrometer through a gate valve. Ammonia (purity, 99.9%) was introduced into a sample cell [15] along with He (99.999%) at a total backpressure of 9.5 atm through a pulsed supersonic valve and the ratio of ammonia and helium in terms of their pressures was monitored using a Quadrupole Mass Spectrometer (SRS RGA 300). A graphite target rod (diameter, 3 mm; length, 20 mm) was mounted on a translating and rotating arrangement driven externally by a stepper motor. The second harmonic of a pulsed Nd-YAG laser (60 mJ/pulse, 6–7 ns and 10 Hz) was used for the sample vaporization. It was fired 120 ls after the peak of the current pulse driving the spring of the pulsed valve with typical pulse-width of 60 ls. The carbon plume formed was mixed with the NH3 –He gas pulse in the sample cell. The cluster species formed were ionized using the 355 nm harmonic of a pulsed Nd-YAG laser (80 mJ/pulse, 5–6 ns and 10 Hz). The mass spectra were recorded both in off and on conditions of the vaporization laser, for various concentrations of NH3 in He (0–1.8) keeping all other parameters fixed. In the laser-off condition, the mass spectrum showed prominently protonated ðNH3 Þn clusters, formed due to multi-photon ionization of the neutral clusters via intracluster ion-molecule reactions [16,17]. Such cluster species contain a central NHþ 4 ion with four hydrogen bonding sites in the first

solvation shell [18,19]. Accordingly, the ðNH3 Þ5 Hþ species was prominent in the spectrum, due to its magic nuclearity.

3. Results and discussion In the laser-on condition, the carbon cluster spectrum obtained with pure helium as the carrier gas reveals Cþ n species with n up to 30. When ammonia is introduced along with He, the spectrum does not show carbon clusters (even with 1% of ammonia). Instead, we observe several new peaks as shown in Fig. 1. The peak at 27 amu is assigned to the HCNþ species, the series of peaks at 45, 62, 79, 96 amu and so on, representing successive addition of up to 7 NH3 molecules to HCN, corresponding to the series ðHCNÞ ðNH3 Þn Hþ . Ammonia attachment to HCN is indeed not surprising, since such adducts are observed from a jet-cooled admixture [12,13]. The peaks corresponding to n ¼ 2, 3 and 4 exhibit relatively high intensity. The other prominent mass series observed is 72, 89, 106, 123 amu and so on, corresponding to ðHCNÞ2 ðNH3 Þn Hþ where n ¼ 1 to 6. Here, the prominent peak is one with n ¼ 3. In addition to these two series, we also observe the ðHCNÞ3 ðNH3 Þn Hþ and ðHCNÞ4 ðNH3 Þn Hþ series, albeit with lower intensity. Smalley and co-workers [4] reacted carbon vapor with ammonia seeded in helium and reported the formation of HC8 H, HC7 N, HC10 H and HC9 N cyanopolyynes appearing as satellite features to the carbon peaks. This is not, however, the case under the experimental conditions employed in the present study. The low-intensity peaks corresponding to ðHCNÞ3 ðNH3 ÞHþ (mass, 99) and ðHCNÞ2 ðNH3 Þ4 Hþ (mass, 123) could, in principle, be assigned to HC7 N and HC9 N, respectively, but we do not see evidence for other related species in the spectrum. Similarly, it can be argued that the main products in Fig. 1 cannot result from waterattached species. The assignments are also substantiated by the systematics observed in the mass spectra. It appears that the formation of HCN involves the mechanism given by Eq. (1). The relative populations of the various species with m ¼ 1 and 2, and n up to 6 vary with the

G. Raina et al. / Chemical Physics Letters 372 (2003) 121–127

123

Fig. 1. Time-of-flight mass spectrum of Am Bn Hþ cluster species, A ¼ HCN, B ¼ NH3 ; 1 6 m 6 4 and 1 6 n 6 7, produced by laser ablation of graphite in presence of NH3 seeded in helium (NH3 =He  0:2). To guide the eye, the prominent peaks containing HCN monomer and dimer are connected by dashed and dotted lines, respectively. Ammonia related peaks common with Fig. 1 are denoted by asterisk.

NH3 =He ratio (see Fig. 2). The populations reach maximum values at a NH3 =He ratio of 0.2. It appears that cluster formation is suppressed at low concentrations of NH3 ðNH3 =He < 0:2Þ primarily due to the dilution effect, while there is a gradual decay due to decreased thermalization in the presence of excess NH3 . In the laser-off experiments, the intensities of ammonia clusters were also highest when the NH3 =He ratio was 0.2, thereby confirming optimal cooling at this ratio. We notice from Fig. 2a that for the ðHCNÞ1 ðNH3 Þn Hþ species, the population is particularly high for n ¼ 3 and 4, and lowest for n ¼ 1. Fig. 2b shows similar plots for the ðHCNÞ2 ðNH3 Þn Hþ species. Thus, species with m ¼ 1, n ¼ 3; 4 and with m ¼ 2, n ¼ 3 or in general those with n þ m close to 5, exhibit higher intensities suggesting a greater stability. We have performed RHF/6-31G(d) level calculations using the GA U S S I A N 98W [20] package for the medium sized systems, ðHCNÞm ðNH3 Þn Hþ (m ¼ 1; 2 and n up to 4) and ðNH3 Þn Hþ (n up to 5).

Our results on the ðNH3 Þn Hþ species corroborate well with those of Kassab and Evleth [21] who used the RHF method with 4-31G+3S basis set. Calculations on ðHCNÞm ðNH3 Þn Hþ species were performed starting with an input geometry where a NHþ 4 ion is centrally located and surrounded by HCN and NH3 molecules. Given the nature of the molecules present in the coordination shell of the NHþ 4 ion, both the proton affinity and dipolar forces are expected to come into play [17]. The proton affinity of NH3 is 854 kJ/mol while that of HCN is somewhat lower, 717 kJ/mol. On the other hand, the dipole moment of HCN is much higher (3.2 Debye) compared to that of NH3 (1.77 Debye). The optimized geometries and energies of some of the initial members of the ðHCNÞm ðNH3 Þn Hþ species are shown in Fig. 3. The geometry-optimized structure for ðHCNÞðNH3 ÞHþ gives an intermolecular H    N hydrogen bond distance of  and an angle of 179.9°, as shown in Fig. 3a. 1.88 A The complexation energy of the cluster species was

124

G. Raina et al. / Chemical Physics Letters 372 (2003) 121–127

(a)

(b)

Fig. 2. Variation in the intensity of ðHCNÞm ðNH3 Þn Hþ peaks in the mass spectrum with the NH3 =He ratio: (a) m ¼ 1 with 1 6 n 6 5, (b) m ¼ 2 with 2 6 n 6 6; n ¼ 1 (squares), 2 (circles), 3 (uptriangle), 4 (down triangle), 5 (diamond) and 6 (cross). For the sake of clarity, some of the curves are presented in the insets.

estimated to be 20.6 kcal/mol, by subtracting its total energy from the sum–total of energies of the NHþ 4 ion and the HCN molecule. Mayer [22] has reported a complexation energy value of 20.3 kcal/ mol based on the (MP2/6-31+G(d)) method. The corresponding energy for the ðNH3 Þ2 Hþ species is somewhat higher (26.3 kcal/mol). In ðHCNÞðNH3 Þ2 Hþ (Fig. 3b), the N–H    N hydrogen bond along HCN deviates considerably from linearity (177.5°) and the bond distance in as against 1.88 A  in ðHCNÞ creases to 1.97 A þ ðNH3 ÞH (Fig. 3a). The H    N distance along the

NH3 molecule has a more favorable value of , owing to its relatively higher proton af1.82 A finity. The incremental complexation energy of this species, obtained by subtracting its total energy from the total energy of the ðHCNÞðNH3 ÞHþ species and NH3 molecule, is 22.0 kcal/mol. This energy is comparable to the corresponding value in ðNH3 Þ3 Hþ (21 kcal/mol). In ðHCNÞðNH3 Þ3 Hþ (Fig. 3c), the NHþ 4 ion is solvated by HCN on one H-bonding site and by two NH3 molecules on the other two H-bonding sites resulting in further relaxation of the solvent shell. The incremental

G. Raina et al. / Chemical Physics Letters 372 (2003) 121–127

125

Fig. 3. Optimized geometric parameters for five of the ðHCNÞm ðNH3 Þn Hþ species obtained at RHF/6-31G(d) level of theory using GA U S S I A N 98W. Bond lengths are reported in angstroms, angles in degrees and total energies in atomic units.

complexation energy in this case is 18.1 kcal/mol which is slightly higher compared to that in ðNH3 Þ4 Hþ (17.2 kcal/mol). With the addition of the fourth NH3 , the first solvation shell of the NHþ 4 ion becomes complete (Fig. 3d). The H-bonds between NHþ 4 ion and the  and so NH3 molecules further increase to 1.96 A  does the distance involving HCN (2.11 A) and are quite linear. Its incremental complexation energy is found to be 15 kcal/mol, which may be compared with that of the ðNH3 Þ5 Hþ species (14.4 kcal/mol). The completion of the solvation shell around NHþ 4 explains the high intensity observed for this species in the mass spectrum (Fig. 1).

It is interesting to examine how the presence of the HCN molecule in the solvation shell of NHþ 4 influences the complexation energy of the system. In Fig. 4, we have plotted the incremental complexation energy due to added NH3 for both ðHCNÞðNH3 Þn Hþ and ðNH3 Þn Hþ species. For the sake of comparison, we have shown the values obtained by Kassab and Evleth [21] for ðNH3 Þn Hþ . The ðNH3 Þn Hþ system, in both cases, shows a monotonic decrease in the incremental complexation energy with the successive additions of NH3 molecules to NHþ 4 ðNH3 Þ. This stepwise decrease has been explained as due to increased charge dispersion and dielectric

126

G. Raina et al. / Chemical Physics Letters 372 (2003) 121–127

Fig. 4. Variation in the incremental complexation energy with successive additions of NH3 molecules, nc , in the solvent shell of þ NHþ species (n ¼ 2 to 5) and 4 . Circles represent ðNH3 Þn H squares, the ðHCNÞðNH3 Þn Hþ species (n ¼ 1 to 4). The crossed square stands for HCN in the first coordination site. The values obtained by Kassab and Evleth [21] for ðNH3 Þn Hþ are also shown (filled circles). Inset shows experimentally measured enthalpy values for ðNH3 Þn Hþ ! ðNH3 Þn1 Hþ þ NH3 : Castleman and co-workers [23], uptriangles; Kebarle and co-workers [24], downtriangles.

higher (1.7 kcal/mol). Thus, it is clear that the solvent preference of ammonium ion changes as the cluster size increases, just as in water-amine ion clusters [25]. In Fig. 3, we have shown the optimized structure of ðHCNÞ2 ðNH3 Þ3 Hþ as an example of the cluster species containing two HCN molecules (see Fig. 3e). Here, the NHþ 4 ion is solvated by two HCN molecules and two NH3 molecules, on the four available H-bonding sites. The solvent shell being complete, its mass peak is associated with a high intensity (Fig. 1). The proton affinity of NH3 molecules being greater than that of HCN, the NH3 molecules get closer to the central NHþ 4 ion with H-bond distances  (angle, 180°) than the HCN moleof 1.94 A ). The incremental comcules (2.09 and 2.10 A plexation energy that results following the addition of HCN to ðHCNÞ1 ðNH3 Þ3 Hþ is 9.4 kcal/mol. It is to be noted that the complexation energies of the two magic nuclearity clusters differ by about 5.6 kcal/mol indicating that ðHCNÞ2 ðNH3 Þ3 Hþ is slightly more stabilized than ðHCNÞ1 ðNH3 Þ4 Hþ .

4. Conclusions shielding in high nuclearity clusters [21]. We have also compared the values with the corresponding experimentally measured enthalpies, DH 0 [23,24] (see inset of Fig. 4), to demonstrate similar trends as the calculated values. The energy required to complex a HCN molecule with the NHþ 4 ion is much lower (20.6 kcal/mol) than that in the case of an NH3 molecule (26.3 kcal/ mol), primarily due to the lower proton affinity of the former. Interestingly, the complexation energy increases to 22.2 kcal/mol with the addition of an NH3 molecule to ðHCNÞðNH3 ÞHþ (at the second site of NHþ 4 Þ. Such an increase could be related to weaker dielectric shielding in the presence of the HCN molecule. With further additions of NH3 , the energy follows a similar trend as that of the pure ammonia system. The ðHCNÞn ðNH3 Þm Hþ species being more polarized, the stabilization energies are also somewhat

We have examined the reaction products formed by interaction of carbon vapor generated by laser ablation of graphite with NH3 seeded in He. The presence of even a small concentration of ammonia (1%), inhibits the formation of carbon clusters and leads to the formation of cluster species of the type ðHCNÞm ðNH3 Þn Hþ along with the ðNH3 Þn Hþ clusters. The populations of these products reach maximum values for a NH3 =He ratio of 0.2 implying that there is a balance between the adiabatic cooling provided by helium and the reactivity of ammonia. The cluster species with n þ m  5, corresponding to fully coordinated NHþ 4 ions, are the most preponderant species as one would expect intuitively. Ab initio molecular orbital calculations show that the complexation energy afforded by addition of each NH3 to the HCN–NHþ 4 cluster decreases after the first addition.

G. Raina et al. / Chemical Physics Letters 372 (2003) 121–127

Acknowledgements The authors thank the Department of Science and Technology, Government of India, for support. References [1] C. Lifshitz, Int. J. Mass Spectrom. 200 (2000) 423. [2] Q.L. Zhang, S.C. OÕBrien, Q. Zhang, Y. Liu, R.F. Curl, H.W. Kroto, R.E. Smalley, J. Phys. Chem. 90 (1986) 525. [3] J.A. Zimmerman, J.R. Eyler, S.B.H. Bach, S.W. McElvany, J. Chem. Phys. 94 (1991) 3556. [4] J.R. Heath, Q. Zhang, S.C. OÕBrien, R.F. Curl, H.W. Kroto, R.E. Smalley, J. Am. Chem. Soc. 109 (1987) 359. [5] D.B. Geohegan, Laser Ablation: Mechanisms and Applications, Springer, Heidelberg, 1991. [6] J. Hermann, A.L. Thomann, C. Boulmer-Leborgne, B. Dubreil, M.L. De Giorgi, A. Luches, I.N. Mihalescu, J. Appl. Phys. 77 (1995) 2928. [7] L.G. Coccia, T.G. Tyrrel, I.W. Boyd, Appl. Surf. Sci. 109/ 110 (1997) 413. [8] S. Acquaviva, A.P. Caricato, M.L. De Giorgi, G. Dinescu, A. Luches, A. Perrone, J. Phys. B 30 (1997) 4405. [9] C. Vivien, J. Hermann, A. Perrone, C. Boulmer-Leborgnem, J. Phys. D 32 (1999) 518. [10] D.W. McPherson, M.L. McKee, P.B. Shevlin, J. Am. Chem. Soc. 105 (1983) 6493.

127

[11] P.B. Shevlin, D.W. McPherson, P. Melius, J. Am. Chem. Soc. 105 (1983) 488. [12] G.T. Fraser, K.R. Leopold, D.D. Nelson Jr., A. Tung, W. Klemperer, J. Chem. Phys. 80 (1984) 3073. [13] R.S. Ruoff, T. Emilsson, C. Chuang, T.D. Klots, H.S. Gutowsky, J. Chem. Phys. 93 (1990) 6363. [14] G. Raina, G.U. Kulkarni, R.T. Yadav, V.S. Ramamurthy, C.N.R. Rao, Proc. Indian Acad. Sci. (Chem. Sci.) 112 (2000) 83. [15] S. Maruyama, L.R. Anderson, R.E. Smalley, Rev. Sci. Instrum. 61 (1990) 3686. [16] A.O. Echt, P.D. Dao, S. Morgan, A.W. Castleman Jr., J. Chem. Phys. 82 (1985) 4076. [17] S. Wei, A.W. Castleman Jr., Int. J. Mass Spectrom. Ion Processes 131 (1994) 233. [18] K. Fuke, R. Takasu, F. Misaizu, Chem. Phys. Lett. 229 (1994) 597. [19] Th. Freudenberg, W. Radlolf, H.-H. Ritze, V. Stert, F. Noack, I.V. Hertel, Z. Phys. D. 41 (1997) 267. [20] M.J. Frisch et al., GA U S S I A N 98 (Rev. A.1); Gaussian, Inc., Pittsburgh, PA, 1998. [21] E. Kassab, E.M. Evleth, J. Am. Chem. Soc. 109 (1987) 1653. [22] P.M. Mayer, J. Phys. Chem. A 103 (1999) 5905. [23] I.N. Tang, A.W. Castleman Jr., J. Chem. Phys. 62 (1975) 4576. [24] J.D. Payzant, A.J. Cunningham, P. Kebarle, Can. J. Chem. 51 (1973) 3242. [25] A.J. Stace, J. Amer. Chem. Soc. 106 (1984) 2306.