Volume 33 Issue 5
Nov.  2020
Turn off MathJax
Article Contents

Bin Wu, Xu-dong Wang, Xiao-fei Gao, Hao Li, Shan Xi Tian. Dissociative Electron Attachment to Carbon Dioxide†[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 521-531. doi: 10.1063/1674-0068/cjcp2008152
Citation: Bin Wu, Xu-dong Wang, Xiao-fei Gao, Hao Li, Shan Xi Tian. Dissociative Electron Attachment to Carbon Dioxide[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 521-531. doi: 10.1063/1674-0068/cjcp2008152

Dissociative Electron Attachment to Carbon Dioxide

doi: 10.1063/1674-0068/cjcp2008152
More Information
  • Corresponding author: Shan Xi Tian, E-mail: sxtian@ustc.edu.cn
  • Part of the special issue for "the Chinese Chemical Society's 16th National Chemical Dynamics Symposium".
  • Received Date: 2020-08-28
  • Accepted Date: 2020-09-13
  • Publish Date: 2020-10-27
  • Part of the special issue for "the Chinese Chemical Society's 16th National Chemical Dynamics Symposium".
  • 加载中
  • [1] F. C. Fehsenfeld, A. L. Schmeltekopf, H. I. Schiff, and E. E. Ferguson, Planet. Space Sci. 15, 373 (1967). doi:  10.1016/0032-0633(67)90201-2
    [2] T. J. Millar, C. Walsh, and T. A. Field, Chem. Rev. 117, 1765 (2017). doi:  10.1021/acs.chemrev.6b00480
    [3] T. J. Millar, C. Walsh, M. A. Cordiner, R. N. Chuimn, and E. Herbst, Astrophys. J. Lett. 662, L87 (2007). doi:  10.1086/519376
    [4] V. Vuitton, P. Lavvas, R. V. Yelle, M. Galand, A. Wellbrock, G. Lewis, A. J. Coates, and J. E. Wahlund, Planet. Space Sci. 57, 1558 (2009). doi:  10.1016/j.pss.2009.04.004
    [5] I. Čadež, S. Markelj, Z. Rupnik, and P. Pelicon, J. Phys.: Conf. Ser. 133, 012029 (2008). doi:  10.1088/1742-6596/133/1/012029
    [6] B. BoudaÏffa, P. Cloutier, D. Hunting, M. A. Huels, and L. Sanche, Science 287, 1658 (2000). doi:  10.1126/science.287.5458.1658
    [7] Y. Zheng and L. Sanche, Rev. Nanosci. Nanotech. 2, 1 (2013). doi:  10.1166/rnn.2013.1022
    [8] S. W. Hla, L. Bartels, G. Meyer, and K. H. Rieder, Phys. Rev. Lett. 85, 2777 (2000). doi:  10.1103/PhysRevLett.85.2777
    [9] P. C. Deshmukh, E. Krishnakumar, S. Fritzsche, M. Krishnamurthy, and S. Majumder, Quantum Collisions and Confinement of Atomic and Molecular Species, and Photons, The 7th Topical Conference of ISAMP (2018).
    [10] T. F. O'Malley and H. S. Taylor, Phys. Rev. 176, 207 (1968). doi:  10.1103/PhysRev.176.207
    [11] G. H. Dunn, Phys. Rev. Lett. 8, 62 (1962). doi:  10.1103/PhysRevLett.8.62
    [12] M. Tronc, C. Schermann, R. I. Hall, and F. Fiquet-Fayard, J. Phys. B: At. Mol. Phys. 10, L459 (1977). doi:  10.1088/0022-3700/10/12/005
    [13] L. D. A. Siebbeles, M. Glass-Maujean, O. S. Vasyutinskii, J. A. Beswick, and O. Roncero, J. Chem. Phys. 100, 3610 (1994). doi:  10.1063/1.466402
    [14] R. Azria, Y. L. Coat, G. Lefevre, and D. Simon, J. Phys. B 12, 679 (1979). doi:  10.1088/0022-3700/12/4/016
    [15] R. Azria, Y. L. Coat, D. Simon, and M. Tronc, J. Phys. B 13, 1909 (1980). doi:  10.1088/0022-3700/13/9/020
    [16] A. Moradmand, D. S. Slaughter, D. J. Haxton, T. N. Rescigno, C. W. McCurdy, Th. Weber, S. Matsika, A. L. Landers, A. Belkacem, and M. Fogle, Phys. Rev. A 88, 032703 (2013). doi:  10.1103/PhysRevA.88.032703
    [17] R. Buchdahl, J. Chem. Phys. 9, 146 (1941). doi:  10.1063/1.1750866
    [18] E. Illenberger and J. Momigny, Gaseous Molecular Ions an Introduction to Elementary Processes Induced by Ionization, Darmstadt, New York, Berlin, Steinkopff Verlag: Springer, (1992).
    [19] O. Inglfsson, F. Weik, and E. Illenberger, Int. J. Mass Spectrom. Ion Process. 155, 1 (1996). doi:  10.1016/S0168-1176(96)04392-3
    [20] H. Hotop, M. W. Ruf, M. Allan, and I. I. Fabrikant, Resonance and Threshold Phenomena in Low-energy Electron Collisions with Molecules and Clusters, Advances In Atomic, Molecular, and Optical Physics (vol 49), Amsterdam: Elsevier, (2003).
    [21] M. Orzol, C. König-Lehmann, E. Illenberger, and J. Kopyra, J. Chem. Phys. 133, 194503 (2010). doi:  10.1063/1.3505550
    [22] A. Ryzhkova and P. Swiderek, Surf. Sci. 605, 963 (2011). doi:  10.1016/j.susc.2011.02.015
    [23] O. Hoegh-Guldberg, D. Jacob, M. Taylor, T. Guillén Bolaños, M. Bindi, S. Brown, I. A. Camilloni, A. Diedhiou, R. Djalante, K. Ebi, F. Engelbrecht, J. Guiot, Y. Hijioka, S. Mehrotra, C. W. Hope, A. J. Payne, H. O. Pörtner, S. I. Seneviratne, A. Thomas, R. Warren, and G. Zhou, Science 365, 6459 (2019).
    [24] W. H. Cheng, M. H. Richter, I. Sullivan, D. M. Larson, C. X. Xiang, B. S. Brunschwig, and H. A. Atwater, ACS Energy Lett. 5, 470 (2020). doi:  10.1021/acsenergylett.9b02576
    [25] S. Gao, Y. Lin, X. C. Jiao, Y. F. Sun, Q. Q. Luo, W. H. Zhang, D. Q. Li, J. L. Yang, and Y. Xie, Nature 529, 68 (2016). doi:  10.1038/nature16455
    [26] C. Hepburn, E. Adlen, J. Beddington, E. A. Carter, S. Fuss, N. M. Dowell, J. C. Minx, P. Smith, and C. K. Williams, Nature 575, 7781 (2019).
    [27] X. D. Wang, X. F. Gao, C. J. Xuan, and S. X. Tian, Nat. Chem. 8, 258 (2016). doi:  10.1038/nchem.2427
    [28] B. Wu, L. Xia, H. K. Li, X. J. Zeng, and S. X. Tian, Phys. Rev. A 85, 052709 (2012). doi:  10.1103/PhysRevA.85.052709
    [29] H. Li, X. F. Gao, X. Meng, and S. X. Tian, Phys. Rev. A 99, 032703 (2019). doi:  10.1103/PhysRevA.99.032703
    [30] D. Rapp and D. D. Briglia, J. Chem. Phys. 43, 1480 (1965). doi:  10.1063/1.1696958
    [31] C. R. Claydon, G. A. Segal and H. S. Taylor, J. Chem. Phys. 52, 3387 (1970). doi:  10.1063/1.1673501
    [32] P. J. Chantry, J. Chem. Phys. 57, 3180 (1972). doi:  10.1063/1.1678736
    [33] A. Stamatovic and G. J. Schulz, Phys. Rev. A 7, 589 (1973). doi:  10.1103/PhysRevA.7.589
    [34] D. Spence and G. J. Schulz, J. Chem. Phys. 60, 216 (1974).
    [35] M. J. W. Boness and G. J. Schulz, Phys. Rev. A 9, 1969 (1974). doi:  10.1103/PhysRevA.9.1969
    [36] R. Abouaf, R. Paineau, and F. Fiquet-Fayard, J. Phys. B 9, 303 (1976). doi:  10.1088/0022-3700/9/2/017
    [37] W. B. England, B. J Rosenberg, P. J. Fortune, and W. C Wahl, J. Chem. Phys. 65, 684 (1976). doi:  10.1063/1.433081
    [38] S. K. Srivastava and O. J. Orient, Phys. Rev. A 27, 1209 (1983). doi:  10.1103/PhysRevA.27.1209
    [39] O. J. Orient and S. K. Srivastava, Chem. Phys. Lett. 96, 681 (1983). doi:  10.1016/0009-2614(83)80073-6
    [40] R. Dressler and M. Allan, Chem. Phys. 92, 449 (1985). doi:  10.1016/0301-0104(85)85038-2
    [41] M. A. Huels, L. Parenteau, P. Cloutier, and L. Sanche, J. Chem. Phys. 103, 6775 (1995). doi:  10.1063/1.470356
    [42] P. Cicman, G. Senn, G. Denifl, D. Muigg, J. D. Skalny, P. Lukac, A. Stamatovic, and T. D. Märk, Czech J. Phys. 48, 1135 (1998). doi:  10.1023/A:1022849832444
    [43] D. S. Slaughter, T. N. Rescigno, D. J. Haxton, A. E. Orel, C. W. McCurdy, and A. Belkacem, J. Phys. B 44, 205203 (2011). doi:  10.1088/0953-4075/44/20/205203
    [44] A. Moradmand, D. S. Slaughter, A. L. Landers, and M. Fogle, Phys. Rev. A 88, 022711 (2013). doi:  10.1103/PhysRevA.88.022711
    [45] P. Nag and D. Nandi, Phys. Rev. A 91, 052705 (2015). doi:  10.1103/PhysRevA.91.052705
    [46] E. Illenberger, Chem. Phys. Lett. 80, 153 (1981). doi:  10.1016/0009-2614(81)80079-6
    [47] C. W. Walter, K. A. Smith, and F. B. Dunning, J. Chem. Phys. 90, 1652 (1989). doi:  10.1063/1.456058
    [48] R. J. Van Brunt and L. J. Kieffer, Phys. Rev. A 2, 1899 (1970). doi:  10.1103/PhysRevA.2.1899
    [49] R. J. Van Brunt and L. J. Kieffer, Phys. Rev. A 10, 1633 (1974). doi:  10.1103/PhysRevA.10.1633
    [50] S. Trajmar and R. I. Hall, J. Phys. B 7, L458 (1974). doi:  10.1088/0022-3700/7/16/009
    [51] D. W. Chandler and P. L. Houston, J. Chem. Phys. 87, 1445 (1987). doi:  10.1063/1.453276
    [52] A. Eppink and D. H. Parker, Rev. Sci. Instrum. 68, 3477 (1997). doi:  10.1063/1.1148310
    [53] C. R. Gebhardt, T. P. Rakitzis, P. C. Samartzis, V. Ladopoulos, and T. N. Kitsopoulos, Rev. Sci. Instrum. 72, 3848 (2001). doi:  10.1063/1.1403010
    [54] D. Townsend, M. P. Minitti, and A. G. Suits, Rev. Sci. Instrum. 74, 2530 (2003). doi:  10.1063/1.1544053
    [55] B. Whitaker, Imaging in Molecular Dynamics: Technology and Applications, UK: Cambridge University Press, (2003).
    [56] D. Nandi, V. S. Prabhudesai, E. Krishnakumar, and A. Chatterjee, Rev. Sci. Instrum. 76, 053107 (2005). doi:  10.1063/1.1899404
    [57] B. Wu, L. Xia, H. K. Li, X. J. Zeng, and S. X. Tian, Rev. Sci. Instrum. 83, 013108 (2012). doi:  10.1063/1.3678328
    [58] X. D. Wang, X. F. Gao, H. Li, B. Wu, and S. X. Tian, J. Phys.: Conf. Ser. 1412, 052010 (2020). doi:  10.1088/1742-6596/1412/5/052010
    [59] A. Moradmand, J. B. Williams, A. L. Landers, and M. Fogle, Rev. Sci. Instrum. 84, 033104 (2013). doi:  10.1063/1.4794093
    [60] E. Krishnakumar, S. Denifl, I. Čadež, S. Markelj, and N. J. Mason, Phys. Rev. Lett. 106, 243201 (2011). doi:  10.1103/PhysRevLett.106.243201
    [61] P. Nag, M. Polášek, and J. Fedor, Phys. Rev. A 99, 052705 (2019). doi:  10.1103/PhysRevA.99.052705
    [62] L. Xia, X. J. Zeng, H. K. Li, B. Wu, and S. X. Tian, Angew. Chem. Int. Ed. 52, 1013 (2013). doi:  10.1002/anie.201206948
    [63] X. J. Zeng, L. Xia, H. K. Li, K. C. Lau, and S. X. Tian, Phys. Rev. A 87, 012711 (2013). doi:  10.1103/PhysRevA.87.012711
    [64] H. K. Li, L. Xia, X. J. Zeng, and S. X. Tian, J. Phys. Chem. A 117, 3176 (2013). doi:  10.1021/jp400026c
    [65] S. X. Tian, B. Wu, L. Xia, H. K. Li, X. J. Zeng, Y. Luo, and J. L. Yang, Phys. Rev. A 88, 012708 (2013). doi:  10.1103/PhysRevA.88.012708
    [66] L. Xia, X. D. Wang, C. J. Xuan, X. J. Zeng, S. X. Tian, H. K. Li, Y. Pan, and K. C. Lau, J. Chem. Phys. 140, 041106 (2014). doi:  10.1063/1.4862684
    [67] S. X. Tian and Y. Luo, Phys. Rev. A 91, 056702 (2015). doi:  10.1103/PhysRevA.91.056702
    [68] X. D. Wang, C. J. Xuan, W. L. Feng, and S. X. Tian, J. Chem. Phys. 142, 064316 (2015). doi:  10.1063/1.4907940
    [69] X. D. Wang, C. J. Xuan, Y. Luo, and S. X. Tian, J. Chem. Phys. 143, 066101 (2015). doi:  10.1063/1.4928639
    [70] M. Y. Li, X. F. Gao, X. D. Wang, H. Li, and S. X. Tian, Inter. J. Mass Spectrom. 404, 20 (2016). doi:  10.1016/j.ijms.2016.03.010
    [71] H. Li, X. F. Gao, X. Meng, and S. X. Tian, J. Phys. Chem. A 123, 9089 (2019). doi:  10.1021/acs.jpca.9b07399
    [72] X. F. Gao, H. Li, X. Meng, J. C. Xie, and S. X. Tian, J. Chem. Phys. 152, 084305 (2020). doi:  10.1063/1.5135609
    [73] R. E. Imhof, A. Adams, and G. C. King, J. Phys. E 9, 138 (1976). doi:  10.1088/0022-3735/9/2/024
    [74] A. Galejs and C. E. Kuyatt, J. Vac. Sci. Tech. 15, 865 (1978). doi:  10.1116/1.569615
    [75] A. Stamatovic and G. J. Schulz, Rev. Sci. Instrum. 41, 423 (1970). doi:  10.1063/1.1684534
    [76] S. Matejcik, A. Kiendler, A. Stamatovic, and T. D. Märk, Int. J. Mass Spectrom. 149, 311 (1995).
    [77] V. Grill, H. Drexel, W. Sailer, M. Lezius, and T. D. Märk, Int. J. Mass Spectrom. 205, 209 (2001). doi:  10.1016/S1387-3806(00)00378-X
    [78] P. Cicman, J. D. Skalny, J. Fedor, N. J. Mason, P. Scheier, E. Illenberger, and T. D. Märk, Int. J. Mass Spectrom. 260, 85 (2007). doi:  10.1016/j.ijms.2006.07.011
    [79] D. C. Cartwright and S. Trajmar, J. Phys. B 29, 1549 (1996). doi:  10.1088/0953-4075/29/8/018
    [80] M. Allan, J. Electron Spectrosc. Relat. Phenom. 48, 219 (1989). doi:  10.1016/0368-2048(89)80018-0
    [81] M. Allan, Phys. Rev. Lett. 87, 0332201 (2001).
    [82] K. Mitsuke, S. Suzuki, T. Imamura, and I. Koyano, J. Chem. Phys. 93, 1710 (1990). doi:  10.1063/1.459098
    [83] Z. Lu, Y. C. Chang, Q. Z. Yin, C. Y. Ng, and W. M. Jackson, Science 346, 61 (2014). doi:  10.1126/science.1257156
    [84] J. S. Lee, J. P. Doering, T. A. Potemral, and H. Brace, Planet. Space Sci. 28, 947 (1980). doi:  10.1016/0032-0633(80)90058-6
    [85] R. A. Frahm, J. D. Winningham, J. R. Sharber, J. R. Scherrer, S. J. Jeffers, A. J. Coates, D. R. Linder, D. O. Kataria, R. Lundin, S. Barabash, M. Holmström, H. Andersson, M. Yamauchi, A. Grigoriev, E. Kallio, T. Säles, P. Riihelä, W. Schmidt, H. Koskinen, J. U. Kozyra, J. G. Luhmann, E. C. Roelof, D. J. Williams, S. Livi, C. C. Curtis, K. C. Hsieh, B. R. Sandel, M. Grande, M. Carter, J. A. Sauvaud, A. Fedorov, J. J. Thocaven, S. McKenna-Lawler, S. Orsini, R. Cerulli-Irelli, M. Maggi, P. Wurz, P. Bochsler, N. Krupp, J. Woch, M. Fränz, K. Asamura, and C. Dierker, Icarus 182, 371 (2006). doi:  10.1016/j.icarus.2006.01.014
    [86] F. E. Meyen, M. H. Hecht, J. A. Hoffman, and the MOXIE Team, Acta Astron. 129, 82 (2016). doi:  10.1016/j.actaastro.2016.06.005
    [87] E. Hinterman and J. A. Hoffman, Acta Astron. 170, 678 (2020). doi:  10.1016/j.actaastro.2020.02.043
    [88] Y. X. Yao, P. Shushkov, T. F. Miller, and K. P. Giapis, Nat. Comm. 10, 2294 (2019). doi:  10.1038/s41467-019-10342-6
    [89] V. S. Prabhudesai, A. H. Kelkar, D. Nandi, and E. Krishnakumar, Phys. Rev. Lett. 95, 143202 (2005). doi:  10.1103/PhysRevLett.95.143202
    [90] D. Davis, S. Kundu, V. S. Prabhudesai, Y. Sajeev, and E. Krishnakumar, J. Chem. Phys. 149, 064308 (2018). doi:  10.1063/1.5032172
    [91] E. Bohler, J. Warneke, and P. Swiderek, Chem. Soc. Rev. 42, 9219 (2013). doi:  10.1039/c3cs60180c
    [92] D. Davis, V. P. Vysotskiy, Y. Sajeev, and L. S. Cederbaum, Angew. Chem. Int. Ed. 50, 4119 (2011). doi:  10.1002/anie.201005129
    [93] D. Davis, V. P. Vysotskiy, Y. Sajeev, and L. S. Cederbaum, Angew. Chem. Int. Ed. 51, 8003 (2012). doi:  10.1002/anie.201204162
    [94] A. Studer and D. P. Curran, Nat. Chem. 6, 765 (2014). doi:  10.1038/nchem.2031
    [95] L. L. Chen, Z. W. Chen, Z. Y. Li, J. Hu, and S. X. Tian, Rev. Sci. Instrum. 89, 103102 (2018). doi:  10.1063/1.5022394
    [96] Z. W. Chen, C. F. Fu, Z. Y. Li, J. Hu, H. Li, J. L. Yang, and S. X. Tian, J. Phys. Chem. Lett. 11, 7510 (2020). doi:  10.1021/acs.jpclett.0c02097
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(12)

Article Metrics

Article views(29) PDF downloads(5) Cited by()

Proportional views
Related

Dissociative Electron Attachment to Carbon Dioxide

doi: 10.1063/1674-0068/cjcp2008152
Part of the special issue for "the Chinese Chemical Society's 16th National Chemical Dynamics Symposium".
Bin Wu, Xu-dong Wang, Xiao-fei Gao, Hao Li, Shan Xi Tian. Dissociative Electron Attachment to Carbon Dioxide†[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 521-531. doi: 10.1063/1674-0068/cjcp2008152
Citation: Bin Wu, Xu-dong Wang, Xiao-fei Gao, Hao Li, Shan Xi Tian. Dissociative Electron Attachment to Carbon Dioxide[J]. Chinese Journal of Chemical Physics , 2020, 33(5): 521-531. doi: 10.1063/1674-0068/cjcp2008152
  • Dissociative electron attachment (DEA) is one of electron-molecule collisional reactions, in which a low-energy (typically 0-20 eV) electron is captured to form a temporary negative ion (TNI), and then the molecular TNI dissociates into a negative ion and neutral fragment(s). DEA plays important roles in various processes such as atmospheric chemistry, astrochemistry, plasma technique, radiation damage and therapy, and nano-fabrication [1-8]. As described in Eqs.(1, 2) and FIG. 1, electron autodetachment (AD) of TNI is a competitive process of DEA,

    Figure 1.  (A) Two typical pathways after electron attachment to XYZ and Morse potential curves of XYZ and XYZ- (left) and (B) differential cross sections of $\sigma_{\rm{EA}}$ and $\sigma_{\rm{DEA}}$ (right).

    DEA process:

    AD process:

    The resonant AD is a typical inelastic collision process, and as shown in FIG. 1(B) the target XYZ can be pumped to high vibrational ($ v $) states.

    If the potential energy curve of (XYZ-)$ ^* $ is of a repulsive Morse type, as described in FIG. 1(B), the electron attachment has a neutral-to-TNI transition probability which is proportional to the Franck-Condon factors (i.e., the square of an integral of the vibrational wave functions of the neutral ground-state target and the resonant-state TNI). $ R_0 $ is the equilibrium internuclear distance of the neutral molecule XY-Z, and $ R_ \rm{c} $ is the crossing point of the neutral and anionic potential energy curves along XY-Z bond. The AD only occurs as $ R $$ \leq $$ R_ \rm{c} $; while beyond $ R_ \rm{c} $, DEA is an available channel with high possibility. The electron attachment cross section ($ \sigma_{\rm{EA}} $) comprises the components of DEA and AD, and

    where $ \tau_ \rm{a} $ is the AD time and $ \tau_ \rm{d} $ is the dissociation time. The exponential term, exp($ -\tau_ \rm{d}/\tau_ \rm{a} $), represents the survival probability of the TNI against AD, and it is also given as exp$ \left(-\int_{R_\varepsilon}^{R_c}\frac{\Gamma_ \rm{a}(R)}{\hbar v(R)} \rm{d}R\right) $ where $ \Gamma_ \rm{a}(R) $ is the AD energy width, $ v(R) $ is the radial velocity between XY and Z-, and $ R_\varepsilon $ is the internuclear distance of the electron attachment [9]. The thermodynamics threshold of the DEA process leading to fragments Z- and XY is,

    where $ D $(XY-Z) is the bond energy of XY-Z, and EA(Z) is the electron affinity of Z. If the electron attachment energy $ E_ \rm{e} $ is higher than $ E_{\rm{th}} $, then the excess energy $ E^* $ will deposit into multiple degrees of freedom, transforming to translational or kinetic energy $ E_k $ of fragments and possibly to internal energies $ E_{\rm{int}} $ (electronic, vibrational or rotational energy) of XY. Therefore, in practice, the appearance energy (AE) of Z- is frequently used in experiment,

    At a given $ E_ \rm{e} $, the $ E_k $(Z-) is determined as,

    where $ m_{\rm{XY}} $ and $ M_{\rm{XYZ}} $ are the masses of the XY fragment and the parent molecule XYZ, respectively. Eqs.(4-6) describe the two-body dissociation. For three-body or many-body dissociations, Eq.(4) should be extended as,

    where the right first two terms denote the energy difference of the bond cleavages and formations in the dissociation.

    Besides the above thermodynamics analyses, in the sixties of the last century the DEA process was investigated with quantum scattering theory. O'Malley and Taylor [10] derived the angular differential cross section for the DEA to diatomic molecule on the basis of the symmetry arguments by Dunn [11]. Tronc et al. [12] simplified the expression as,

    where $ k $ is the incident electron momentum, $ a_{l, |\mu|}(k) $ is the energy-dependent expansion coefficient, and $ Y_{l, \mu}(\theta, \varphi) $ are the spherical harmonics. Here, $ \mu $ is the difference in the projection of the angular momentum along the internuclear axis for the neutral molecular state and the TNI's resonance state, given as $ |\mu| $ = $ |\Lambda_f-\Lambda_i| $; $ l $ is the angular momentum of the incoming electron with values given as $ l $$ \geq $$ |\mu| $. This equation requires axial-recoil approximation [13] and assumes that the TNI states do not interfere or couple with each other. If more than one uncoupled TNI states are involved in the DEA process, $ \sigma_{\rm{DEA}} $ equals a summation over different $ |\mu| $ valves. In principle, the target's electronic state, usually the electronically ground state, is known, the resonant state symmetry of the TNI can be obtained by fitting the experimental angular distribution of the negative fragment. This also provides us an experimental method to evaluate the quantum scattering predictions about TNI states.

    Although Eq.(8) is derived for diatomic molecules, it is still applicable for polyatomic molecules if the axial-recoil approximation is valid or in an impulsive dissociation. The majority of the excess energy is transformed into the fragments' translational motions along the dissociating axis [14, 15]. If the experimental angular distribution of the negative fragment may have a remarkable difference from the axial-recoil prediction, the axial-recoil approximation is invalid, which implies that TNI undergoes bending and conical intersections involved in dissociation dynamics [16]. If the lifetime of a resonance is long enough to allow the intramolecular nuclear motions in TNI, then understanding of angular distributions should resort to sophisticated dynamic simulations of the nuclear motions that need to be performed on the multi-dimensional potential energy surface of the TNI state.

  • The DEA process was firstly discussed in 1920s and an experimental study about the DEA to I2 (e-+I2$ \rightarrow $I+I-) was reported by Buchdahl [17]. Research interest on this topic continues and is extended for gas-phase molecules [18-20], molecular clusters [19, 20], and molecular films [21, 22]. As an important atmospheric molecule, CO2 emission by burning fossil fuels leads to global warming and diverse environmental crisis [23]. Its reduction into value-added chemicals or fuels has been a long-standing issue in both heterogeneous and homogeneous catalysis [24-26]. In the catalytic reduction of CO2, DEA to CO2 can be considered as a fundamental step to produce CO which is one constituent of Fischer-Tropsch syngas. Moreover, the DEA to CO2 profoundly influences the C and O cycles in the interstellar atmosphere [27]. Thus, it is vital to understand the mechanism of DEA to CO2, especially the kinetic energy, state, and angular distributions of the fragments.

    DEA to CO2 has been extensively studied [16, 27-45]. These studies mainly focused on the production efficiencies of the anionic fragments of O-, C-, and O2- [32, 34]. FIG. 2(A) indicates a typical efficiency curve of O- which primarily consists of three bands, i.e., two strong bands around 4.4 and 8.2 eV and a much lower one around 13.0 eV [32]. As shown in FIG. 2(B), only one band around 18.7 eV together with a broad shoulder from 15.0 eV to 22.0 eV was observed in the C- efficiency curve [34]. Two peaks around 11.3 and 12.9 eV were observed with the large uncertainties in the O2- efficiency curve and the total production cross-section was in the order of 10$ ^{-24} $ cm$ ^2 $ which was four or five orders lower than that of O- [34].

    Figure 2.  (A) O- and (B) C- production efficiency curves of the DEA to CO2 (reproduced from [32, 34]).

    It is well known that the O- production at 4.4 eV is attributed to the DEA process around a shape resonance $ ^2\Pi_{\rm{u}} $ of CO2-, but the arguments about some fine structures of this O- efficiency band continues for a half century [16, 33, 36, 40, 42]. Stamatovic and Schulz assigned these structures with the vibrational states of CO ($ v $ = 0-3) fragment produced in CO2-($ ^2\Pi_{\rm{u}} $)$ \rightarrow $CO($ ^1\Sigma^+ $)+O-($ ^2 $P) [33]. However, using the same technique, Abouaf et al. [36] observed the different structures which are attributed to the nuclear motions of intermediate CO2-. Some ensuing studies tried to bridge this divergence, for example, the fine structures with an energy spacing of 0.26 eV were observed by filtering all O- yields with non-zero kinetic energy and assigned with the CO vibrational states, while additional much narrower structures (with a spacing of 0.10 eV) appeared when the kinetic energy discrimination was removed; the latter was attributed to the nuclear motion of intermediate CO2- [40, 42]. However, more dynamics information is still encouraged due to the limitation of anion efficiency curve measurement. A recent experimental study by using the anion velocity map imaging (VMI) technique brought new information but the results were still puzzling [16]. Since this DEA process undergoes the $ ^2\Pi_{\rm{u}} $ shape resonant state of CO2-, the O- angular distribution should exhibit a typical $ \Pi $-symmetric feature. On the contrary, the forward-backward scattered distribution of the O- yields, i.e., $ \Sigma $-symmetric feature, was observed [16].

    The nature of the CO2- resonance(s) around 8.2 eV was debated for a long time. Firstly it was regarded as a $ ^2\Sigma_{\rm{g}}^+ $ shape resonance [31, 37], but afterwards was recognized as a $ ^2\Pi_{\rm{g}} $ Feshbach resonance [38, 40, 41]. Recently a conical intersection between $ ^2\Pi_{\rm{g}} $ state at 8.2 eV and $ ^2\Pi_{\rm{u}} $ state at 4.4 eV was proposed, and the axial-recoil approximation should be invalid in the DEA process around 8.2 eV [43, 44]. Accompanying the O- production at 13.0 eV and the higher energy, the co-yield CO was proposed to be in electronically excited states and this DEA process might undergo the high-lying resonance $ ^2\Phi_{\rm{g}} $ of CO2- [32].

    At the higher attachment energy, the C- ion was proposed as the yield of three-body dissociation CO2-$ \rightarrow $C-+O+O, and two C-O bonds were simultaneously broken around $ ^2\Sigma_{\rm{u}}^+ $ and $ ^2\Pi_{\rm{g}} $ states of CO2- locating between 15 and 20 eV [34]. O2- was also reported as the yield of a two-body dissociation, but the competitive pathway leading to C-+O2 was denied [34]. Considering that the much faster C- ions could be produced in the above pathway, we conjecture that most of these fast ions might fly away and could not enter the quadrupole mass filter (QMF) spectrometer used in their measurements [34]. Therefore, the collection efficiency of the C- ions should be enhanced and three- and two-body dissociations are hopefully disentangled by the $ E_k $(C-)-resolved measurements.

  • The $ E_k $-resolved experiment was partially realized by modifications of the QMF or time-of-flight (TOF) apparatus [46, 47], but the energy resolution was unsatisfied. On the other hand, the angular distributions of anionic fragment were measured using a mechanically rotating analyzer [48-50]. The detectable angle range was usually restricted, and the forward (near 0$ ^\circ $) and backward (near 180$ ^\circ $) directions were inaccessible due to the geometrical limitations of the conventional turn-table arrangement. The $ E_k $ distribution in full scattering angle range can be obtained with ion imaging technique innovated by Chandler and Houston [51], and this technique was further developed into VMI by Eppink and Parker [52], and time-sliced velocity imaging by Townsend and Gebhardt et al. [53, 54]. In the VMI and time-sliced imaging techniques, the product ions at different positions (e.g., in a volume of 2 mm$ \times $2 mm$ \times $2 mm) but with the same velocity can be effectively focused in an inhomogeneous extraction field and expanded to be a Newton sphere in the downstream homogeneous field. A two-dimensional (2D) velocity image is recorded when the ions within a Newton sphere are projected onto a 2D position-sensitive detector, while time-sliced image at the center of the Newton sphere is recorded when the detector is working in a pulsed mode. Complete kinetics information, i.e., the $ E_k $ and full angular distribution of the fragment ions, can be obtained with the Abel transformation from 2D velocity image [52, 55], while it is directly viewed in the central time-sliced image. The VMI and time-sliced imaging techniques have been extensively used in photodissociation, photoionization, and crossed-beam reaction. Recently, these state-of-the-art techniques were introduced into the DEA studies by different groups [56-61].

    The pulsed electron beam in the anionic velocity imaging apparatuses (including ours [57]) is produced with a thermal filament, and the energy spread of these thermally emitted electrons is about 500-600 meV. Our first VMI apparatus was developed in 2012 [57] and schematized in FIG. 3(A). Briefly, the low-energy pulsed electrons are collimated and guided into the reaction region by a pair of Helmholtz coils, then collide with an effusive molecular beam which is perpendicular to the electron incident direction. The anionic fragments are periodically pushed out and then fly through the TOF-VMI tube, afterwards are detected by the detector which consists of a set of microchannel plates (MCPs), phosphor screen, and a CCD camera. The time-sliced imaging is realized by applying a high voltage pulse (typically with a time width corresponding to 1/5-1/10 diameter of a Newton sphere) on the rear MCP, meanwhile, this voltage pulse serves as a mass gate to selectively detect the anionic yields. Since there are millions of electrons in each bunch of the electron beam, the combinational usage of phosphor screen and CCD camera is more suitable in the case of the multiple anionic yields in each pulse, indicating the detection efficiency higher than that of position- and time-sensitive detectors (e.g., the wedge-and-strip anode [56] or the delay-line anode [59]). Lots of the DEA studies were accomplished with this apparatus and some interesting results have been reported [27, 28, 62-70].

    Figure 3.  Schematics of our (A) low- and (B) high- resolution VMI apparatuses.

    Currently, the velocity- or momentum-resolution of the anionic image of DEA process suffers from the energy spread of the thermal electrons. The effects of ro-vibrational motions of the target XYZ or the molecular fragment XY are blurred or invisible in the Z- images. Several types of electron monochromator, such as hemispherical electron monochromator [73], Wien filter [74], and trochoidal electron monochromator (TEM) [75], are applicable in the DEA experiments. TEM, as a typical one in combination with a QMF or TOF mass spectrometer, is frequently used to measure the cross sections or production efficiency curves of the DEA anionic fragments [32, 33, 34, 41, 42, 46, 47, 50, 76-78]. Using the TEM and a retarding field filtering lens, we established a high-resolution apparatus [58] (while the previous one [57] is named as low-resolution apparatus). As schematized in FIG. 3(B), the monochromatized electron beam is perpendicular to a continuous supersonic molecular beam. A retarding field analyzer and a Faraday cup installed at the downstream of the TEM are used to determine the electron beam energy resolution and dump the scattered electrons, respectively. The current 1.5 nA of the continuous monochromatized electrons is almost invariable at the energy above 3 eV, and the derivatives in terms of the retarding field strength of the analyzer show the FWHM (full width at half maximum) values in a range of 60-110 meV below 10 eV while a little larger value at the higher energy. Sometimes, to improve the anionic counting rate, we enhanced the electron beam current by increasing the electron energy spread up to 250 meV [71, 72]. The ion detection system is similar to the previous one [57], but the anionic VMI optics is updated. The kinetic energy resolution ($ \Delta E/E $) of the image is about 2% for the O- anion with the kinetic energy less than 3.5 eV but reaches 5.5% at a very low kinetic energy (0.1 eV) [58], which is much better than our previous design ($ \Delta E/E $ = 5% when $ E $$ \leq $1.0 eV) [57]. As we proposed previously [57], the image can be zoomed in or out by adjusting the voltages of the VMI lenses with a certain proportion. This flexible VMI system has two benefits for the MCP detector with a fixed size (here is 75 mm diameter, while the pervious one is 40 mm): the large Newton sphere of the fast anionic products can be contracted without any deformations by increasing the voltages of the lenses; on the contrary, more structures in the small Newton sphere of the slow anionic products can be observed in the magnified image by decreasing the voltages [57, 58].

  • Considering the half-century controversy on the DEA process (e-+CO2$ \rightarrow $CO2-($ ^2\Pi_{\rm{u}} $)$ \rightarrow $CO($ ^1\Sigma^+ $)+O-($ ^2 $P), $ E_{\rm{th}} $ = 3.988 eV) around 4.4 eV, we reinvestigate it using our high-resolution apparatus (FIG. 3(B)). The time-sliced images of O- are recorded at $ E_ \rm{e} $ = 4.15, 4.55 and 4.95 eV and shown in FIG. 4(A-C). The images at 4.55 eV (FIG. 4(B)) and 4.95 eV (FIG. 4(C)) distinctly show an asymmetric feature about the relative intensities in the forward-backward directions, which is in line with the previous observations [16]. However, the image at 4.15 eV (FIG. 4(A)) exhibits the relatively stronger intensity in the forward direction, which is a little different from the previous result at 4.1 eV [16]. Besides the double-petal distributions in the forward (down) and backward (up) directions, some fine structures stand out in the present images, owing to the high resolution of our apparatus. Moreover, one can find the dramatic changes of the images with the $ E_ \rm{e} $ increase, which is definitely related to the $ E_ \rm{e} $-dependence of the intramolecular nuclear motions of the intermediate CO2-($ ^2\Pi_{\rm{u}} $).

    Figure 4.  High-resolution O- velocity images of the DEAs to CO2 at the electron energies of (A) 4.15, (B) 4.55, and (C) 4.95 eV. The electron incident direction (along y axis) is from top (backward) to bottom (forward) and through the image center.

    According to Eq.(5), we have

    where the excess energy $ E^* $ = $ E_ \rm{e}-E_{\rm{th}} $ and $ E_k $(O-) is determined experimentally in images, thus the internal energy $ E_{\rm{int}} $(CO), corresponding to different vibrational ($ v $) and rotational ($ j $) states of the electronically ground state $ X^1\Sigma^+ $ of CO, is derived. The images of 4.55 and 4.95 eV can be assigned with the vibrationally excited states $ v $ = 0, 1 and $ v $ = 1, 2 of CO respectively, as denoted in FIG. 4 (B) and (C). Only the vibrationally ground state ($ v $ = 0) CO is observed in FIG. 4(A). Furthermore, the rotational states $ j $ = 0-25 of CO ($ v $ = 0) in the forward (from -15$ ^\circ $ to 15$ ^\circ $) and backward (from -165$ ^\circ $ to 165$ ^\circ $) directions are assigned in FIG. 5. Although the present resolution is not high enough to disentangle the isolated rotational states of CO, the undulating $ E_k $(O-) profiles indicate the non-Boltzmann distributions of these rotational states.

    Figure 5.  Assignments with the rotational states $j$=0-25 of CO ($X^1\Sigma^+$, $v$=0) of the O- kinetic energy distributions in the forward (A, $\theta$ from -15° to 15°) and backward (B, $\theta$ from -165° to 165°) scattering directions. The solid circles are the relative intensities of O- ions observed in FIG. 4(A), the vertical lines show the intensities of different rotational states, and the red curves show the fitting profiles of all rotational states considered here. In the data fittings, a gaussian function with an energy width of 0.03 eV is used for each rotational state.

    The O- angular distribution of 4.15 eV is plotted in FIG. 6 for the ion intensities within an annular area of the image which corresponds to certain rotational state(s) of CO. The fastest O- ions near the outside edge of the image in FIG. 4(A) are related to $ j $ = 0, 1, 2, 3 states of CO ($ X^1\Sigma^+ $, $ v $ = 0), while the slowest ones corresponding to $ j $ = 25 state locate in the most inside region. FIG. 6 clearly exhibits that the O- angular distributions heavily rely on the CO rational states. The O- ions corresponding to the highest rotational state $ j $ = 25 of CO indicate a nearly isotropic distribution, while an anisotropy appears for the rotational states $ j $ = 23-25. Only the forward and backward distributions are observed for the lower rotational states $ j $ = 0-3. This forward-backward peaking profile (roughly as a function of 1/$ \sin\theta $) is a typical feature of a reaction experienced with an intermediate complex with lifetime long compared to its rotational period, thus the axial-recoil approximation is completely invalid. The period of the slowest CO rotation is about 10$ ^{-11} $ s, thus we can conjecture that the lifetime of CO2- at $ ^2\Pi_{\rm{u}} $ state could be in this time scale. The nuclear motions in the predissociative CO2- lead to the intramolecular energy partitions and finally determine the asymptotic energy distributions. Around the resonant state $ ^2\Pi_{\rm{u}} $ of CO2-, the vibrational excitations of the symmetric stretching $ \nu_1 $($ \sigma_{\rm{g}}^+ $), and the odd-quanta bending $ \nu_2 $($ \pi_{\rm{u}} $), rather than the asymmetric-stretching $ \nu_3 $($ \sigma_{\rm{u}}^+ $), of CO2 are permitted [79]. Namely, the vibrationally excited CO can be produced by the bond stretching via $ \nu_1 $ mode of CO2-, while the rotating CO is formed by receiving a torque in the bond bending $ \nu_2 $ of CO2-. The combination and Fermi coupling between $ \nu_1 $ and $ \nu_2 $ lead to the ro-vibrationally excited CO [80, 81]. Our results support the previous conjecture about the quantum states of the CO product [33] and clarify the long-time controversies [16, 33, 36, 40, 42] on the dissociation processes of CO2- at the $ ^2\Pi_{\rm{u}} $ shape resonant state.

    Figure 6.  Angular distributions of O- ions produced in the DEA to CO2 at 4.15 eV. The co-product CO ($X^1\Sigma^+$, $v$=0) is populated at different rotational states ($j$).

  • As mentioned above, the vibrational resonant excitation through AD is a competitive pathway of DEA. In a previous experimental study, two CO2 vibrational modes $ \nu_1 $ and $ \nu_2 $ were found to be activated at 3.5 eV, namely ($ n $, 0, 0) and ($ n $, 1, 0) [80]. Due to the symmetry selection rules [79, 81], the bond bending motion $ \nu_2 $ = 1 is forbidden. However, the excitation to ($ n $, 1, 0) is attributed to the fact that the CO2 structure is slightly bent at room temperature ($ \nu_2 $ = 1 is only about 0.065 eV above its ground state). With the energy losses of 0.5-1.0 eV in the electron inelastic collisions, the cross sections of the excitations to $ n $ = 3, 4, 5 and 6 of $ \nu_1 $ were quite larger [80]. These studies provide us for opportunities to investigate the DEA process of vibrationally excited CO2, although there are no satisfactory results from the previous efforts [38].

    The DEA to vibrationally excited CO2 experiments were performed with high-resolution apparatus, in which a long 400 ns (while a typical value is 250 ns) width of the pulsed electron beam was used [29]. In comparison with the DEA to cold (vibrational ground state) CO2 [40, 42], the O- production efficiency curve for the present hot CO2 target shows the strong intensities below the $ E_{\rm{th}} $ of 3.988 eV. More interestingly, we observed three weak peaks at 2.95, 3.05, and 3.16 eV and three strong ones at 3.34, 3.59, and 3.83 eV. The former are attributed to the vibrational nonresonant excitations [80] of CO2, while the latter are closely related to the vibrational resonant excitations around $ ^2\Pi_{\rm{u}} $ state. Besides the O- production efficiency curve for the present hot CO2, we also measured the O- momentum images and carried out the analyses of the O- angular distributions [29].

    Here some points deserve attention: First, the cross section of the DEA to vibrationally excited target is usually larger than that of the cold target. Second, the hot molecules at high vibrational states can be efficiently prepared in the inelastic resonant collisions (i.e., AD process, see FIG. 1), moreover, specific vibrational modes of a polyatomic molecule can be selectively excited. Third, as discussed in our previous paper [29], the long electron pulse is essential, because the front part is responsible for the CO2 vibrational excitation and the residual part is responsible for subsequent DEA process. This inspires a pump-probe method in our future studies of the excited-state target.

  • The DEA process, e-+CO2$ \rightarrow $CO2-($ ^2\Pi_{\rm{g}} $)$ \rightarrow $CO($ ^1\Sigma^+ $) +O-($ ^2 $P), happens around $ E_e $ = 8.2 eV. We performed the VMI experiments with the low-resolution apparatus (FIG. 3(A) and Ref.[57]) in 2012. In this work, we stressed on the important role of dynamic Renner-Teller effect around $ ^2\Pi_{\rm{g}} $ resonance of CO2- (FIG. 7(A)) on the basis of the analyses of three O- images recorded at 7.7, 8.2, and 8.7 eV (FIG. 7(B)-(D)) [28]. In the $ ^2\Pi_{\rm{g}} $-state splitting due to the Renner-Teller effect, neither $ \Lambda $ nor $ l $ is a good quantum number, but the vibronic angular momentum $ K $ about the axis is a good quantum number, $ K $ = $ | $$ \pm $$ \Lambda $+$ l $$ | $. Therefore, at the $ ^2\Pi_{\rm{g}} $ resonant state, when one quantum of the bending vibration mode $ v_2 $ is excited, and parities correspond to $ K $ = 0 and $ K $ = 2 ($ \Pi_{\rm{g}} $$ \times $$ \Pi_{\rm{u}} $ = $ \Sigma_{\rm{u}}^+ $+$ \Sigma_{\rm{u}}^ˉ $+$ \Delta_{\rm{u}} $), respectively. Since the excitation of the vibrational bending mode $ v_2 $ participates in the DEA, namely the nuclear motions are involved in the dissociation processes, the axial-recoil approximation will break down. However, if the dissociation happens at the moment of the slight bond bending (as shown in FIG. 7(A)), the axial-recoil approximation may be still valid. Thereby, the $ \Sigma $ and $ \Delta $ parities, rather than $ \Pi $, control the O- momentum distributions at 7.7 eV [28]. On the contrary, at 8.2 and 8.7 eV, the O- angular distributions were fitted well with the $ \Pi $ parity [28], implying the impulsive dissociations around $ ^2\Pi_{\rm{g}} $ resonance of the linear CO2-.

    Figure 7.  Renner-Teller split states coupling with the dissociation pathway in Franck-Condon region of electron attachment (A). Time-sliced O- velocity images of the DEA to CO2 are recorded at electron energies of 7.7 eV (B), 8.2 eV (C), and \mbox{8.7 eV} (D) \cite{28}. The electron incident direction (along $x$ aix) is from left (backward) to right (forward) and through the image center.

    Slaughter et al. [43] and Moradmand et al. [44] reported their O- momentum images at 8.1 eV and 8.2 eV, which were similar to our image at 7.7 eV. They further inferred that the nonaxial recoil dissociations occurred at some specific bending angles [43, 44]. Nag and Nandi [45] presented an incredible result and they still believed that some $ \Sigma $-symmetry resonance was responsible for the O- angular destitutions. The dominating state of two anionic resonant states involved was a $ \Sigma $ symmetry, instead of a $ \Pi $ symmetry with minor contributions coming from an additional $ \Delta $ state at lower energy and a $ \Pi $ state at higher energy [45]. Their differences [45, 56] from ours [28] should be due to their problematic apparatus and operational mistakes that have been pointed out in our work published in Ref.[67].

  • The O- ions produced in the electron energy range 10-20 eV were observed by Rapp and Briglia (two peaks at 12 and 17 eV) [30], Chantry (one peak at 13 eV) [32] and Orient and Srivastava (three peaks at 13, 16.9 and 19.4 eV) [39]. An electronically excited CO ($ d^3\Delta $ or $ a^3\Pi $) may be produced in the DEA process through $ ^2\Phi_{\rm{g}} $ resonance of CO2- [32]. The O- kinetic energy distribution measured at 13 eV showed a profile similar to that at 4.4 eV by Chantry [32]. Note that the O- cross sections at 13 eV are two orders of magnitude less than those at 8.2 eV. We performed the VMI experiments at this energy with the high-resolution apparatus, and the results are shown in FIG. 8. The present O- kinetic energy distribution shows an agreement with that of Chantry [32], and the tetrad-petal like pattern is visible in the momentum image. At this electron energy, the ion pair dissociation e-+CO2$ \rightarrow $CO$ ^+ $+O- ($ E_{\rm{th}} $ = 17.99 eV [82]) is inaccessible, the DEA process is the only pathway to produce O- ion.

    Figure 8.  High-resolution O- image (A) and the O- kinetic energy distribution (B) for the DEA to CO2 at the electron energy of 13 eV. In (A), the electron incident direction (along the $y$ axis) is from top (backward) to bottom (forward) and through the image center.

  • Using the QMF method, the C- production efficiency curve (FIG. 2(B)) was assigned with the three-body dissociation to C-+O+O [34]. In the electron energy range 15-20 eV, the fast C- ions could be produced with its co-product O2 in the two-body dissociation (e-+CO2$ \rightarrow $C-+O2, $ E_{\rm{th}} $ = 10.18 eV). As pointed out in the above description, a small aperture is usually set in the front of the QMF, which largely reduces the collection efficiency of the fast C- ions. In the VMI apparatus of FIG. 3(A), the large aperture (with a diameter of 20 mm) of the second lens is beneficial to the fast C- collection. This is a key to identify the existence of the two-body dissociation in the electron energy range 15-20 eV. In the VMI measurements, the C- ions with the kinetic energy up to 2.0 eV can be detected [27], and we confirmed that the two-body dissociation coexists with the three-body dissociation in the same energy range [27]. As shown in FIG. 9, the C- momentum image at 15.9 eV indicates that the center spot corresponds to the C- ions produced in the three-body dissociation while the outside anisotropic distribution is the yield of the two-body dissociation. It the latter, the vibrationally excited and electronic ground-state O2 together with C- is produced in the symmetric bending motion of CO2- [27].

    Figure 9.  Three-dimensional image of C- of the DEA to CO2 at the electron energy of 15.9 eV. The red circle is the demarcation of three-body dissociation and two-body dissociation.

    A profound importance of this finding arises from the mystery of molecular oxygen origin in the prebiotic Earth. The three-body combination reaction, O+O+M$ \rightarrow $M+O2, is a widely acceptable mechanism to produce O2 in atmosphere, where atomic oxygen is produced in photolysis of CO2 [83]. CO2 is the predominant component (more than 95%) in the atmospheres of the prebiotic Earth and the current Mars, and as shown in FIG. 10 there are lots of the low-energy free electrons in the ionospheres of the present Earth and Mars [84, 85]. These low energy electrons are produced in the ionizations of molecules or dust exposed to extreme vacuum ultraviolet light or collided with high energy particles [27, 84]. Considering the extensive presences of both the low-energy electrons and CO2 in atmosphere, we conclude that the two- and three-body dissociations of the DEA process that were ignored previously should be incorporated into the general oxygen-circulation model of atmospheric chemistry.

    Figure 10.  (A) Daytime photoelectrons spectrum in the Earth ionosphere (reproduced from [84]), in which each line represents a different height. (B) Energy spectra of oversampled photoelectrons in the Martian ionosphere, where the electron flows toward and away from Mars are marked in red and green respectively (reproduced from [85]). The vertical bars shaded in blue correspond to the electron energy range investigated in the DEA study [27].

  • We know that the Martian atmosphere is full of CO2 molecules while the O2 density is extremely low (less than 2%). If the future explorers could produce O2 from the Martian atmosphere for propellant and for breathing, immigration to Mars becomes highly possible. Recently, a NASA's new project, MOXIE (Mars Oxygen In situ Resource Utilization Experiment [86, 87]), was launched in July, 2020. In virtue of the above issue and the CO2 catalytic transformations [24-26], we need to reexamine the dynamics bases of the catalytic reactions.

    In thermodynamics, CO2 is quite stable and its reduction to CO+O is a highly endothermic process (with an energy cost of 5.43 eV) and that to C+O+O requires an energy of 16.46 eV [88]. As shown in FIG. 11(A), the reduction through DEA, such as, e-+CO2$ \rightarrow $CO+O-, reduces about 1.46 eV of the energy cost, owing to the high EA of atomic O. This is the real reason that charge transfer mechanism plays a central role in the catalysis, which is applicable in catalyzing the other noble molecules. As schematized in FIG. 11(B), the negative charge transfer to the CO2 absorbed on the substrate is a key step of the transformation catalysis. Besides improvement of the CO2 absorption efficiency on a suitable substrate, an efficient charge transfer or electron capture is essential, sometime, with help of the injection of excess electrons (under radiation of electrons) or by using the electrochemical cathode. This idea brings two terminological items, electron-induced or electron-driven chemical reaction [89-91] and bond-breaking-by-a-catalytic-electron (BBCE) [92-94]. Owing to the bond-breaking specificity by tuning electron attachment energy, the DEA process is a promising technique to control the electron-induced reactions.

    Figure 11.  (A) Energetics comparison between the neutral dissociation and DEA processes of CO2. (B) Surface catalytic processes of CO2.

  • The combined application of VMI and TEM techniques enables us to have more insights into the complicated dynamics of the DEA processes. The experimental studies of the DEA to CO2, as a typical example of this field, are reviewed in this article, indicating that the dissociative pathways and yields are primarily determined by the intramolecular nuclear motions of CO2- at different states. Our recent progresses encourage us to introduce more techniques into the experiments, such as pump-probe technique to investigate the DEA dynamics of vibrational or electronic excited molecules, cavity-enhanced laser-induced fluorescence and resonance-enhanced multiphoton ionization techniques to detect the neutral radical products. The latter helps us to realize a complete detection of all DEA products, providing for a full and state-resolved picture of the DEA dynamics. We also notice the potential applications of the DEA process, and a project about the electron-induced reactions in liquid and liquid-vapor surface is being carried out in our group [95, 96].

  • This work was supported by the National Natural Science Foundation of China (No.21727804, No.21625301, No.21273213). The authors appreciate the experimental contributions from the previous students (Lei Xia, Hong-kai Li, Xian-jin Zeng, Chuan-jin Xuan, Xin Meng) in our group.

Reference (96)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return