English

• In the atmosphere, aromatic hydrocarbons are an important fraction of volatile organic compounds (VOCs) [1], among which benzene (C₆H₆) is the simplest but ubiquitous pollutant. Mineral dust particles including various metal oxides provide adsorption sites of VOCs, and reactive surfaces for heterogeneous reactions [2, 3]. The oxidation reactions of VOCs such as benzene can result in the formation of oxygenated products with lower volatilities and these highly-oxidized compounds are quite important to the generation of secondary organic aerosols (SOAs) [4, 5]. SOAs are one of the major components throughout the haze-fog episodes in China [6], and particle-phase reactions are one of important processes of the SOA formation [7, 8]. Vanadium, manganese, nickel, chromium, and copper are abundant heavy metals found in the atmosphere in the north of China [9, 10], and there are various reactive oxide species on the mineral oxide surfaces, which are key species of atmospheric chemistry [11]. Apart from the above mentioned transition metal oxides, Fe₂O₃ as well as TiO₂ are also typical mineral aerosols. It is quite desirable to obtain the detailed reaction kinetic data and reaction mechanisms to understand the intrinsic properties of reaction centers of mineral dust aerosols at a molecular level and offer basic parameters for atmospheric models. However, the enormous complexities of environmental factors and relative reaction products render the reactions as well as the nature of active centers obscure.

  Gas-phase experiments conducted under properly defined conditions provide an ideal arena for probing the energtics and kinetics of a reaction at a strictly molecular level [12-22]. Investigations on benzene oxidation by gas-phase ions and clusters gives fundamental insights for how benzene molecules can be activated and the nature of active sites. There are some reports about the reactions of metal ions [23-38], metal clusters [32, 36, 39], and several transition metal oxide ions [40-43] with benzene. Large amount of experimental efforts have been devoted to the study of 3d transition metal atoms (M = Sc-Ni) interacting with C₆H₆, and sandwich and rice-ball structures for M_n(C₆H₆)_m were observed [25]. In addition, the adsorption, C-H bond activation [44], dehydrogenation and dissociation processes [24, 45, 46] of benzene molecules mediated by gas-phase ions and clusters were also reported. Notably, there is a lack of clear picture about the reactivity of small 3d transition metal oxide cluster (TMOC) cations toward C₆H₆ in the literature, which is useful in elucidating compositions of reactive clusters, the reaction trends and the possible oxidative ability of mineral aerosols.

  In this study, we systematically investigate 103 reactions of first-row transition metal oxide cations M_xO_y⁺ (M = Sc-Cu, x = 1–6) with C₆H₆ in the linear ion trap at near room temperature. Thirtynine clusters show high reactivity toward benzene and display up to six primary reaction channels, as shown in Scheme 1a: 1) adsorption channel, 2) dehydrogenation of C₆H₆, 3) charge exchange, 4) hydrogen atom transfer (HAT), 5) oxygen atom transfer (OAT) and 6) the formation of C₆H₅O radical. In addition to M_xO_yC₆H₆⁺, five types of oxidation products are generated during these reactions, that is, C₆H₄ unit, C₆H₅^· (phenyl) radical, C₆H₆O_{0, 1}⁺ cations, C₆H₆O^· and C₆H₅O^· radical. The adsorption channel exists in most of the investigated cluster cations under our experimental conditions, except for ScO₃⁺, MnO₄⁺ and other 25 inert clusters (Table S1 in Supporting information), and this pathway will not be included in the following discussion. Generally speaking, only series of scandium oxide clusters and vanadium oxide clusters exhibit similar reactivity toward benzene in the 39 reactive 3d TMOC cations; however, more than two types of reactions coexist in other 3d TMOCs. Co_xO_y⁺ and Cr_xO_y⁺ cations exhibit variety of the products in the reactions with C₆H₆. The number of reactive Mn_xO_y⁺ and Ni_xO_y⁺ clusters is the least. The detailed reaction channels will be discussed in the following.

  Scheme 1

  

  Scheme 1.  Simplified reaction channels for the reactions of M_xO_y⁺ (M = Sc-Cu cluster cations (a) and atmospheric OH^· radicals (b) with benzene.

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. 1 and Fig. S1 (Supporting information), dehydrogenation of C₆H₆ is prevalent in the C₆H₆ oxidation reactions mediated by the reactive metal oxide cations, except for Sc_xO_y⁺ clusters. The pseudo-first order rate coefficients (k₁) [47], which were estimated on the basis of a least-square fitting procedure (Fig. S2 in Supporting information), and the reaction efficiencies (Φ= k₁/k_calc; k_calc is the theoretical collision rate [48, 49]) for these reactions are given in Table 1 and Table S2 (Supporting information). Among the investigated clusters, Ti₂O₄⁺, (V₂O₅)_1–3⁺, (V₂O₅)_0-1VO₃⁺ [47] Cr_xO_y⁺ (x, y = 1, 2; 2, 5; 2, 6; 3, 2; 3, 8), Mn₂O₃⁺, FeO₄⁺, Fe₂O₃⁺, Co_xO_y⁺ (x, y = 2, 2; 2, 5; 3, 4; 4, 5; 5, 7), Ni₅O₅⁺, as well as Cu₅O₃⁺ and Cu₆O₄⁺ can induce dehydrogenation of C₆H₆, with the formation of M_xO_y-1C₆H₄⁺ and H₂O. The estimated rate constants varied from 10^-9 cm³ molecule^-1 s^-1 to 10^-10 cm³ molecule^-1 s^-1, and branching ratios (BRs) of this channel dramatically changes from 1.3% for Cr₂O₆/C₆H₆ to 78% for V₃O₈⁺/C₆H₆ [50]. More information of reactivity and reaction channels is shown in Fig. 2. Based on the reported B3LYP-calculated results for the reactions of V₂O₅⁺ and VO₃⁺ with C₆H₆, the dehydrogenation pathway proceeds by two steps of C-O bond making and two steps of hydrogen atom abstraction alternately, with the formation of OH moiety and H₂O unit successively [50]. The formed C₆H₄ unit is bonded with two oxygen atoms of oxide clusters by forming a ···-M-O-C-C-O-··· ring in the product [50]. Similar reaction mechanisms may be followed for other reactive TMOCs. Normally, a small change in cluster composition can completely change the reactive properties of the clusters [51, 52]. Herein, the dehydrogenation of C₆H₆ generally exists in various 3d TMOC cations, and the reactive clusters contain both of oxygen-rich and oxygen-poor ones. Like Cr₃O₂⁺, which is very oxygen deficient, can even offer one oxygen atom to form H₂O unit in the C₆H₆ oxidation reactions. Unlike the reactions of stoichiometric early TMOC cations with CH₄ [53], no uniform structural compositions can be summed up for these reactive 3d TMOCs. In addition, the dehydrogenation of C₆H₆ is one of the major channels (BR > 25%) in most of the investigated reactions. In literature, there are some oxide ions which also can bring about the dehydrogenation of C₆H₆, such as FeO⁺[54], MnO⁺ [40], V₂O₅ and (V₂O₅)_0–2VO₃ [42], and the branching ratios of this channel for MnO⁺ and FeO⁺ are only 10% and 5%, respectively.

  Figure 1

  

  Figure 1.  Time-of-flight mass spectra for the reactions of the mass-selected Sc₄O₆⁺ (a), Ti₃O₇⁺ (b), Cr₃O₂⁺(c), Co₃O₄⁺(d) and Cu₃O₂⁺(e) with C₆H₆, respectively. The reaction time and the C₆H₆ pressures are given. The peaks marked with stars are present in the background spectra.

  DownLoad: Full-Size Img PowerPoint

  Table 1

  

  Table 1.  Products, pseudo-first-order rate constants, and reaction efficiencies for the investigated reactions.

  DownLoad: CSV

  Figure 2

  

  Figure 2.  Map showing of the reactivity and reaction channels for 3d transition metal oxide cations with C₆H₆. The red number shown in each block represents the oxygen atoms y in M_xO_y⁺ cations. The blank block means these clusters have not been investigated.

  DownLoad: Full-Size Img PowerPoint

  In addition to dehydrogenation of C₆H₆, charge exchange channel is also observed in many investigated 3d TMOC cations, and two kinds of products C₆H₆⁺ and C₆H₆O⁺ are present, as shown in Fig. 1 and Fig. S1. For V₃O₈⁺, V₅O₁₃⁺, Cr₂O₆⁺, Cr₄O₁₁⁺ and Ni₄O₅⁺, the charge exchange channel is concomitant with the OAT channel, resulting to the formation of C₆H₆O⁺ ions as the products of the reactions with C₆H₆. For the reactions of (V₂O₅)_1–3⁺, Cr_xO_y⁺ (x, y = 1, 2; 2, 5; 2, 6; 3, 2; 3, 8; 4, 10; 4, 11), MnO₂⁺ and Co_xO_y⁺ (x, y = 1, 4; 2, 3; 2, 5) with C₆H₆, C₆H₆⁺ cations are the reaction products. The charge transfer happens when cluster cations collide with benzene molecules [50], and this channel is closely related to the ionization energies (IEs) of benzene, phenol and relative neutral TMOCs. If C₆H₆⁺ is present in the mass spectra, the IEs of neutral TMOCs can be assigned as larger than that of C₆H₆. Mass discrimination effect due to large mass difference between C₆H₆O_{0, 1}⁺ and oxide clusters makes it difficult to exhibit the precise intensity of C₆H₆⁺ in some reactions, and this phenomenon is more serious for larger clusters. Thus, it is hard to accurately compare the productions of C₆H₆O_{0, 1}⁺ for the reactions of the small clusters with those of the larger clusters. This charge exchange channel also takes place in the reactions of benzene with Co⁺, Cu⁺, Nb⁺ [55] and Au⁺ [37] ions, forming C₆H₆⁺ cations. However, the pathway of generating C₆H₆O⁺ has not been reported in the studies of C₆H₆ oxidation reaction mediated by gas-phase clusters. In the atmosphere, charge exchange reactions, such as H + O⁺ →H⁺ + O and O₂⁺ + NO →O₂ + NO⁺, have been studied extensively [56]. From our gas-phase reactions, one may expect that this reaction type may also exist between mineral oxides and some VOC molecules, such as C₆H₆. Then the formed C₆H₆⁺ radical cations can further react fast with other atmospheric radicals through radical reactions.

  Different from the generally existing pathways 1–3 in Scheme 1a, the HAT process prevails predominantly only in (Sc₂O₃)_1–5⁺/C₆H₆ systems, with the formation of (Sc₂O₃)_1–5H⁺ and C₆H₅ (Fig. 1a and Fig. S1). The Sc₈O₁₂⁺ is the most reactive one and the BRs for this HAT channel varied from 38% (for Sc₁₀O₁₅⁺) to 90% (for Sc₆O₉⁺). As reported in literature, Δ≡2y – nx + q (q: the charge number, n: the highest oxidation state of element M') can be used to classify oxygen-richness or -poorness of early transition metal oxide clusters M'_xO_y^q [15, 57], and the Δ = 1 clusters (Sc₂O₃)_n⁺, which possess oxygen-centered radicals (O^-·), can activate the C H bond of C₄H₁₀ [58]. Although the C H bond energy (BE) of C₆H₆ (4.89 eV) is higher than that of CH₄ (4.55 eV) [59], there is no doubt that O plays a key role for the HAT process in (Sc₂O₃)_n⁺/ benzene systems. Larger scandium oxide cluster cations with stoichiometry of (Sc₂O₃)_n may also be able to activate C₆H₆ through HAT, which is not further investigated herein. Among other studied early TMOC cations with Δ = 1, such as (V₂O₅)_1-4⁺ and (CrO₃)_{1, 2}⁺, only (V₂O₅)_1-4⁺ clusters can abstract one hydrogen atom from C₆H₆, and this channel is not the dominant one [50]. Schwarz et al. pointed out that an increase or decrease in electronic multiplicity of the bare MO⁺ (M = Sc-Ni) coincides with a similar trend in the richness and variety of the products formed in the reactions with C₆H₆ [40]. However, this phenomenon is not suitable for the benzene oxidation on polynuclear metal oxide cation clusters studied herein.

  It is interesting that in benzene oxidation reactions with FeO₂⁺, Co₂O₃⁺, Co₃O₄⁺, Cu₂O₂⁺ and Cu₄O₃⁺ clusters (M_xO_y⁺), oxygen atom transfer accompanying HAT happens, leading to the formation of C₆H₅O radicals and M_xO_y-1H⁺ cations [60]. Note that the bond dissociation energies of FeO⁺, CoO⁺, and CuO⁺ are relatively weak in the first-row transition metal monoxide ions MO⁺ [61]. According to the thermodynamic requirement that must be met by these M_xO_y⁺, the reactive M-O bond energies for M_xO_y⁺ should be within this range: BE(M–O) < BE(C₆H₅–O)-BE(C₆H₅–H) = (8.126-4.896) eV = 3.23 eV (The bonds energy for C₆H₅–O is calculated at the B3LYP/6-311++G^** level of theory, and that for C₆H₅–H is from Ref. [60]). Besides, we systematically studied the reactions of nanosized vanadium oxide cluster cations with benzene, and phenol is the product; surprisingly, the size is not the major factors influencing the clusters' reactivities [50].

  In the atmosphere, OH^· radicals are one of the most important oxidant, and there are lots of products observed from the OH^· radical-initiated reactions of benzene [62], which leads to the removal of benzene and other VOC molecules from the atmosphere or transformation in the atmosphere. For instance, in Scheme 1b, the abstraction pathway can form H₂O and a phenyl radical in OH^·/benzene system, accounting for 5% of the overall reaction at 298 K [63, 64]. The C₆H₅^· radical can further react with O₂ to produce C₆H₅O^· (C₆H₅^· + O₂ →C₆H₅O^· + O, Progress (2)). The following reaction C₆H₅O^· +M →C₅H₅^· + CO + M (Progres (3)) will obtain cyclopentadienyl (C₅H₅^·)radical, which is a key intermediate species for aromatic ring growth [65]. Phenol is also generated via OH^· addition reaction to benzene (Progress (4)). Three (intermediate) products for OH^· +C₆H₆ were also observed in our gas-phase model reactions, such as C₆H₅^· radical, C₆H₅O^· radical, and C₆H₅OH. In contrast, H₂O is formed through breaking of two C-H bonds in benzene and generating of C-O bonds in cluster reactions, and this dehydrogenation channel is generally present with large BRs. Therefore, the mechanisms and BRs of dehydrogenation in benzene oxidation reactions mediated by OH^· radical and 3d TMOC cations are different. Physical adsorption by mineral dusts is the one of major sinks for VOC molecules [65], and the adsorption channels are also generally present in most of the investigated cluster cations. Based on the cluster reactions, other two types of channels existing in gas-phase cluster/C₆H₆ systems, that is, charge exchange and dehydrogenation, may contribute to the correspond-ing VOC oxidation reactions in the atmosphere. From these results, we may speculate that similar reaction types observed in OH^·+C₆H₆ in the atmosphere may also exist in the heterogeneous reactions of benzene molecules over mineral dusts, but the mechanisms and importance of each channels may be different; there are also some specific reactions on mineral oxide surfaces, which are not present in the gas-phase reaction of OH^· with C₆H₆. These ring-retaining radical products obtained in the benzene oxidation reactions can further react with other radicals or species in the atmosphere.

  In summary, we systematically investigate 103 reactions of small 3d transition metal oxide cations with C₆H₆ by using mass spectrometry, and a clear picture of the reactivity of TMOCs toward benzene is built. As summarized in Fig. 2, a total of 39 cations can oxidize C₆H₆ efficiently, with the rate constants being in the order of 10^-9–10^-10 cm³ molecule ^-1 s ^-1, regardless of adsorption channel. The shade of color indicates the approximate range of reaction rates. And the darker the color, the faster the reaction. The blank block means these clusters have not been investigated. As shown in Fig. 2, In the obtained six primary reaction channels, dehydrogenation of benzene and charge transfer between cations and C₆H₆ generally exist in the investigated reactions, and the formation of C₆H₆O⁺ has not been reported previously. In comparison with hydroxyl radical, the most important gas-phase oxidant in the atmosphere, three (intermediate) products are observed in the reactions of TMOCs with benzene, that is, C₆H₅^· radical, C₆H₅O^· radical, and phenol. The reaction information is particularly valuable for gauging of future nano- and multiscale modeling of benzene oxidation over mineral oxide aerosols in the atmosphere, and one may expect that the heterogeneous conver-sion of VOCs such as C₆H₆ on mineral dusts is indispensable.

  Acknowledgments

  This work was supported by National Key R & D Program of China (No. 2016YFC0203000), the National Natural Science Foundation of China (No. 21503015), and the Fundamental Research Funds for the Central Universities (Nos. 22050205, 2017CX01008).

  Appendix A. Supplementary data

  Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2019.05.015.

  
  
• 1. [1]

     J.G. Calvert, R. Atkinson, K.H. Becker, et al., The Mechanisms of Atmospheric Oxidation of Aromatic Hydrocarbons, Oxford University Press, New York, 2002.

  2. [2]

     C.R. Usher, A.E. Michel, V.H. Grassian, Chem. Rev. 103(2003) 4883-4939. doi: 10.1021/cr020657y

  3. [3]

     T. Kameda, E. Azumi, A. Fukushima, et al., Sci. Rep.-UK 6(2016) 1-10. doi: 10.1038/s41598-016-0001-8

  4. [4]

     J.L. Jimenez, M.R. Canagaratna, N.M. Donahue, et al., Science 326(2009) 1525-1529. doi: 10.1126/science.1180353

  5. [5]

     E. Borrás, L.A. Tortajada-Genaro, Atmos. Environ. 47(2012) 154-163. doi: 10.1016/j.atmosenv.2011.11.020

  6. [6]

     G. Wang, R. Zhang, M.E. Gomez, et al., Proc. Natl. Acad. Sci. U. S. A. 48(2016) 13630-13635.

  7. [7]

     M. Kalberer, D. Paulsen, M. Sax, et al., Science 303(2004) 1659-1662. doi: 10.1126/science.1092185

  8. [8]

     B. Zheng, Q. Zhang, Y. Zhang, et al., Atmos. Chem. Phys. 14(2015) 2031-2049.

  9. [9]

     J. Duan, J. Tan, Atmos. Environ. 74(2013) 93-101. doi: 10.1016/j.atmosenv.2013.03.031

  10. [10]

      H.Z. Tian, C.Y. Zhu, J.J. Gao, et al., Atmos. Chem. Phys. 15(2015) 12107-12166. doi: 10.5194/acpd-15-12107-2015

  11. [11]

      H.J. Tong, P.S.J. Lakey, A.M. Arangio, et al., Faraday Discuss. 200(2017) 251-270. doi: 10.1039/C7FD00023E

  12. [12]

      H. Schwarz, Catal. Sci. Technol. 7(2017) 4302-4314. doi: 10.1039/C6CY02658C

  13. [13]

      H. Schwarz, S. Shaik, J.L. Li, J. Am. Chem. Soc. 139(2017) 17201-17212. doi: 10.1021/jacs.7b10139

  14. [14]

      S.M. Lang, T.M. Bernhardt, Phys. Chem. Chem. Phys. 14(2012) 9255-9269. doi: 10.1039/c2cp40660h

  15. [15]

      X.L. Ding, X.N. Wu, Y.X. Zhao, et al., Acc. Chem. Res. 45(2012) 382-390. doi: 10.1021/ar2001364

  16. [16]

      R.A.J. O'Hair, G.N. Khairallah, J. Cluster Sci. 15(2004) 331-363. doi: 10.1023/B:JOCL.0000041199.40945.e3

  17. [17]

      J. Roithová, D. Schröder, Chem. Rev. 110(2010) 1170-1211. doi: 10.1021/cr900183p

  18. [18]

      A.W. Castleman Jr., Catal. Lett. 141(2011) 1243-1253. doi: 10.1007/s10562-011-0670-7

  19. [19]

      S. Yin, E.R. Bernstein, Int. J. Mass Spectrom. 321-322(2012) 49-65.

  20. [20]

      N. Dietl, M. Schlangen, H. Schwarz, Angew. Chem. Int. Ed. 51(2012) 5544-5555. doi: 10.1002/anie.201108363

  21. [21]

      D. Caraiman, G.K. Koyanagi, D.K. Bohme, J. Phys. Chem. A 108(2004) 978-986. doi: 10.1021/jp0307194

  22. [22]

      P. Cheng, A. Shayesteh, D.K. Bohme, Cheminform 40(2010) 241-246.

  23. [23]

      G.K. Koyanagi, D.K. Bohme, Int. J. Mass Spectrom. 227(2003) 563-575. doi: 10.1016/S1387-3806(03)00091-5

  24. [24]

      X.P. Xing, Z.X. Tian, H.T. Liu, et al., Rapid Commun. Mass Spectrom. 17(2003) 1743-1748. doi: 10.1002/rcm.1115

  25. [25]

      T. Hanmura, M. Ichihashi, T. Kondow, J. Phys. Chem. A 106(2002) 11465-11469. doi: 10.1021/jp021275z

  26. [26]

      T. Kurikawa, H. Takeda, M. Hirano, et al., Organometallics 18(2012) 1430-1438.

  27. [27]

      D. Caraiman, D.K. Bohme, J. Phys. Chem. A 106(2002) 9705-9717. doi: 10.1021/jp0208900

  28. [28]

      D.J. Trevor, R.L. Whetten, D.M. Cox, et al., J. Am. Chem. Soc. 107(1985) 518-519. doi: 10.1021/ja00288a049

  29. [29]

      M.R. Zakin, D.M. Cox, R.O. Brickman, et al., J. Phys. Chem. 93(1989) 6823-6827. doi: 10.1021/j100355a048

  30. [30]

      Y. Huang, B.S. Freiser, J. Am. Chem. Soc. 112(1990) 1682-1685. doi: 10.1021/ja00161a004

  31. [31]

      X.P. Xing, H.T. Liu, Z.C. Tang, PhysChemComm 6(2003) 32-35. doi: 10.1039/b302761a

  32. [32]

      C. Berg, M. Beyer, U. Achatz, et al., J. Chem. Phys. 108(1998) 5398-5403. doi: 10.1063/1.475972

  33. [33]

      S. Roszak, D. Majumdara, K. Balasubramaniana, J. Phys. Chem. A 103(1999) 5801-5806. doi: 10.1021/jp9907950

  34. [34]

      X.P. Xing, Z. Tian, H.T. Liu, et al., J. Phys. Chem. A 107(2003) 8484-8491.

  35. [35]

      H.T. Liu, S.T. Sun, X.P. Xing, et al., Rapid Commun. Mass Spectrom. 20(2006) 1899-1904. doi: 10.1002/rcm.2524

  36. [36]

      M. Tombers, L. Barzen, G. Niednerschatteburg, J. Phys. Chem. A 117(2013) 1197.

  37. [37]

      Y.P. Ho, R.C. Dunbar, Int. J. Mass Spectrom. 182-183(1999) 175-184.

  38. [38]

      R.C. Dunbar, G.T. Uechi, B. Asamoto, J. Am. Chem. Soc. 116(1994) 2466-2470. doi: 10.1021/ja00085a029

  39. [39]

      L. Barzen, M. Tombers, C. Merkert, et al., Int. J. Mass Spectrom. 330-332(2012) 271-276.

  40. [40]

      M.F. Ryan, D. Stoeckigt, H. Schwarz, J. Am. Chem. Soc. 116(1994) 9565-9570. doi: 10.1021/ja00100a021

  41. [41]

      C. Heinemann, H.H. Cornehl, D. Schroder, et al., Inorg. Chem. 35(1996) 2463-2475. doi: 10.1021/ic951322k

  42. [42]

      F. Dong, S. Heinbuch, Y. Xie, et al., J. Am. Chem. Soc. 131(2009) 1057-1066. doi: 10.1021/ja8065946

  43. [43]

      K.A. Zemski, R.C. Bell, A.W. Castleman, Int. J. Mass Spectrom. 184(1999) 119-128. doi: 10.1016/S1387-3806(98)14276-8

  44. [44]

      B. Butschke, H. Schwarz, Organometallics 30(2011) 1588-1598. doi: 10.1021/om101138d

  45. [45]

      K. Judai, M. Hirano, H. Kawamata, et al., Chem. Phys. Lett. 270(1997) 23-30. doi: 10.1016/S0009-2614(97)00336-9

  46. [46]

      G.S. Jackson, F.M. White, C.L. Hammill, et al., J. Am. Chem. Soc. 119(1997) 7567-7572. doi: 10.1021/ja970218u

  47. [47]

      Z.Y. Li, Z. Yuan, X.N. Li, et al., J. Am. Chem. Soc. 136(2014) 14307-14313. doi: 10.1021/ja508547z

  48. [48]

      G. Kummerlöwe, M.K. Beyer, Int. J. Mass Spectrom. 244(2005) 84-90. doi: 10.1016/j.ijms.2005.03.012

  49. [49]

      T. Su, M.T. Bowers, J. Chem. Phys. 58(1973) 3027-3037. doi: 10.1063/1.1679615

  50. [50]

      J.T. Cui, Y. Zhao, J.C. Hu, et al., J. Chem. Phys. 149(2018) 074308. doi: 10.1063/1.5038175

  51. [51]

      S.M. Lang, D.M. Popolan, T.M. Bernhardt, Chem. Phys. Solid Surf. 12(2007) 53-90. doi: 10.1016/S1571-0785(07)12002-2

  52. [52]

      M. Arenz, S. Gilb, U. Heiz, Chem. Phys. Solid Surf. 12(2007) 1-51. doi: 10.1016/S1571-0785(07)12001-0

  53. [53]

      Y.X. Zhao, X.N. Wu, Z.C. Wang, et al., Chem. Commun. 46(2010) 1736-1738. doi: 10.1039/b924603g

  54. [54]

      S. Detlef, S. Helmut, Helv. Chim. Acta 75(1992) 1281-1287. doi: 10.1002/hlca.19920750429

  55. [55]

      H. Higashide, T. Oka, K. Kasatani, et al., Chem. Phys. Lett. 163(1989) 485-489. doi: 10.1016/0009-2614(89)85173-5

  56. [56]

      N.S. Shuman, D.E. Hunton, A.A. Viggiano, Chem. Rev. 115(2015) 4542. doi: 10.1021/cr5003479

  57. [57]

      Y.X. Zhao, X.N. Wu, J.B. Ma, et al., Phys. Chem. Chem. Phys. 13(2011) 1925-1938. doi: 10.1039/c0cp01171a

  58. [58]

      X.N. Wu, B. Xu, J.H. Meng, et al., Int. J. Mass Spectrom. 310(2012) 57-64. doi: 10.1016/j.ijms.2011.11.011

  59. [59]

      Y.R. Luo, Comprehensive Handbook of Chemical Bond Energies, CRC Press, Boca Raton, 2007.

  60. [60]

      J.T. Cui, C.X. Sun, Y. Zhao, et al., Phys. Chem. Chem. Phys. 21(2019) 1117-1122. doi: 10.1039/C8CP06807K

  61. [61]

      D. Schröder, H. Schwarz, Angew. Chem. Int. Ed. 34(1995) 1973-1995. doi: 10.1002/anie.199519731

  62. [62]

      R. Atkinson, J. Arey, Chem. Rev. 103(2003) 4605-4638. doi: 10.1021/cr0206420

  63. [63]

      R. Atkinson, J. Phys. Chem. Ref. Data 26(1997) 215-290. doi: 10.1063/1.556012

  64. [64]

      T. Seta, M. Nakajima, A. Miyoshi, J. Phys. Chem. A 110(2006) 5081-5090. doi: 10.1021/jp0575456

  65. [65]

      X. Shen, Y. Zhao, Z. Chen, et al., Atmos. Environ. 68(2013) 297-314. doi: 10.1016/j.atmosenv.2012.11.027

• 

  Scheme 1  Simplified reaction channels for the reactions of M_xO_y⁺ (M = Sc-Cu cluster cations (a) and atmospheric OH^· radicals (b) with benzene.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Time-of-flight mass spectra for the reactions of the mass-selected Sc₄O₆⁺ (a), Ti₃O₇⁺ (b), Cr₃O₂⁺(c), Co₃O₄⁺(d) and Cu₃O₂⁺(e) with C₆H₆, respectively. The reaction time and the C₆H₆ pressures are given. The peaks marked with stars are present in the background spectra.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Map showing of the reactivity and reaction channels for 3d transition metal oxide cations with C₆H₆. The red number shown in each block represents the oxygen atoms y in M_xO_y⁺ cations. The blank block means these clusters have not been investigated.

  下载: 全尺寸图片 幻灯片

  Table 1.  Products, pseudo-first-order rate constants, and reaction efficiencies for the investigated reactions.

  下载: 导出CSV
•