Dipolar dissociation dynamics in electron collisions with carbon monoxide
DDipolar dissociation dynamics in electron collisionswith carbon monoxide
Dipayan Chakraborty , Pamir Nag and Dhananjay Nandi Indian Institute of Science Education and Research Kolkata, Mohanpur 741246, India email: [email protected] [email protected], [email protected] Abstract
Dipolar dissociation processes in the electron collisions with carbon monoxide have been studied usingtime of flight (TOF) mass spectroscopy in combination with the highly differential velocity slice imaging(VSI) technique. Probing ion-pair states both positive and/or negative ions may be detected. The ionyield curve of negative ions provides the threshold energy for the ion-pair production. On the other hand,the kinetic energy distributions and angular distributions of the fragment anion provide detailed dynamicsof the dipolar dissociation process. Two ion-pair states have been identified based on angular distributionmeasurements using VSI technique.
Ion-pair states are generally superexcited states ofmolecules embedded in the ionization continuum.Such superexcited states can be accessed throughphoton or electron collisions with isolated moleculesthat in turn may dissociates into positive and nega-tive ions. [1] Photoion pair formation has been stud-ied quite extensively over the past few decades. [2]Main emphasis was to determine threshold energyfor the process more accurately employing thresholdion pair production spectroscopy (TIPPS).[3, 4] How-ever, detailed dynamics have been obtained in the re-cent times due to the availability of high resolutionion pair imaging spectroscopy (IPIS). [2, 5] On theother hand, in the electron collisions, the same ionpair states can be accessed and detailed dissociationdynamics can be obtained by probing fragment an-ion using high resolution velocity slice imaging (VSI)spectroscopy.VSI technique has been successfully applied tostudy negative ion formation due to dissociative elec-tron attachment (DEA) process.[6, 7, 8] The sametechnique has been extended here to study dissoci-ation of the long-range ion-pair states of CO. Un- like DEA process, the anion formation due to dipolardissociation does not proceed via resonant electroncapture.[9] In the latter case, the incident electrontransfers some energy to the molecule and excite it toion-pair states that eventually dissociates into cationand anion. The ion-pair formation is possible as longas the excitation energy is equal and more than theasymptotic ion-pair dissociation energy. The min-imum energy position of the ion-pair state usuallyfar away from the equilibrium position of the neu-tral molecule in the potential energy curve formalism.The ion-pair state may be accessed via direct excita-tion or indirectly through predissociation of an ini-tially excited Rydberg state of the neutral molecule.The indirect mechanism is more commonly applied inthe studies of photoion pair formation.[2] The detec-tion of ion pair provides information on the electronicstructure of a molecule and the dissociation dynam-ics of its exited states. For electron collision studies,both direct and indirect mechanism may be applica-ble as discussed in the present article. In electroncollision with CO the ion-pair states can give rise tomomentum matched anion and cation products, ei-1 a r X i v : . [ phy s i c s . a t m - c l u s ] A ug her C + and O − or C − and O + channels:CO + e − → CO ∗ + e − → (cid:26) C + + O − + e − C − + O + + e − (1)The ion pair formation from CO was reported byVaughan [10] and Lozier [11] in the electron collisionstudies quite long ago. In the dipolar dissociationrange, Lozier [11] observed equal intensity of C + andO − formation with threshold energy of 20.9 ± − and O − have been observed in the dipolar dissociation range.However, the count rate for C − is too low to per-form any meaningful VSI study and is not reportedin the present article. It is well accepted [11, 13] thatthe anion formation in the electron collision studieswith reported primary electron energy range can onlybe possible through ion-pair states. Here, it is as-sumed that the O − ions are always accompanied byC + ions but to verify the claim conclusively a coin-cidence measurement is absolutely necessary. In thisarticle, we first outline the method and provide de-tailed studies of the dipolar dissociation dynamics inthe electron collisions with carbon monoxide (CO)using VSI. The O − ions produced due to dipolar dissociation arestudied using highly differential time sliced velocitymap imaging technique. The current experimentalsetup is similar to the previous report of Nandi etal. [14] with minor modifications as described byNag and Nandi.[15] The same setup has been usedto study the dissociative electron attachment to Cl ,[6] CO [7] and CO [8] in the recent time. In brief,the experimental setup consists of an electron gun, aFaraday cup to measure electron current situated inthe same axis, a needle of 1 mm diameter to pro-duce effusive molecular beam and a time of flight(TOF) based velocity map imaging (VMI) spectrom-eter. The needle directed towards the detector isplaced in the spectrometer axis and perpendicular to the electron beam axis. The basic theme of the ex-periment is the effusive molecular beam interacts per-pendicularly with the magnetically collimated pulsedelectron beam. As a result, ions are formed in theinteraction zone that are pulsed extracted into thespectrometer and detected by a two dimensional po-sition sensitive detector. The electrons are producedby thermionic emission and the energy of the elec-trons is controlled by a programmable power sup-ply. Typical energy resolution of the electron beam isabout 0.8 eV. The pulse width of the electron beamis about 200 ns and the repetition rate is 10 kHz.After passing through the interaction region the elec-trons are collected using the Faraday cup that mea-sures the time averaged electron beam current. TheVMI spectrometer is a three field time of flight (TOF)type mass spectrometer capable to map all the ionswith a given velocity vector to a point on the de-tector irrespective of their place of birth. The de-tector consists of three micro channel plates (MCP)with Z-stack configuration and three layers delay linehexanode.[16] The TOF of the detected ions are de-termined from the back MCP signal [15] whereas thex and y positions of the ions are calculated from thethree anode layers of the hexanode placed behind theMCPs. The TOF (t) and (x, y) position of eachdetected ions are stored in list-mode format (LMF)using the CoboldPC software from RoentDek. Theexperiments are performed under ultra high vacuumcondition with base pressure as low as 10 − mbar and99.9% pure commercially available CO gas.To obtain the ion yield curve a different set of dataacquisition system has been used. Only the MCPsignal is used for this purpose. The MCP signal isfirst amplified by a fast amplifier and then fed to aconstant fraction discriminator (CFD). The outputof CFD is fed to STOP of a nuclear instrumentationmodule (NIM) standard time to amplitude converter(TAC) and the START pulse is generated by a mas-ter pulse which is synchronized with the electron gunpulse. The time difference between this START andSTOP is the TOF of the O − ion. The output of theTAC is connected to a multichannel analyzer (MCA,Ortec model ASPEC-927). Finally, it is communi-cated to a computer via USB 2.0 interface used fordata acquisition. Our own LabVIEW based data ac-2uisition system [15] is used to obtain the mass spec-tra and the ion yield curve.When the electrons are collied with the molecule,‘Newton Sphere’ of ion is formed. One can obtainthe angular distribution information from the projec-tion of the ‘Newton spheres’ onto a two-dimensionalposition sensitive detector. Ions with higher kineticenergy will fall onto the detector with bigger diam-eter. In the current experiment, a moderate pulsedextraction field is applied and negative ions are ex-tracted from the source region of the spectrometer.The extraction pulse duration is 2 µ s and applied 100ns after the electron beam pulse. This delayed ex-traction provides sufficient time to expand the ‘New-ton Sphere ’ so that we can obtain better time slicedimages and also prevent the electrons from reachingthe detector. The aim is to obtain the central sliceof the Newton sphere containing the kinetic energyand angular distribution information of the detectedions. To obtain the central slice a suitable time win-dow has been selected during offline analysis usingCoboldPC software. These sliced images contain theions ejected in the plane parallel to the detector andcontaining the electron beam axis. The typical fullwidth at half maximum (FWHM) in TOF of the O − is about 500 ns and a 50 ns time window has beenselected for slicing purpose. For low energy ions athiner slice (25 ns) may be less erroneous. The elec-tron energy calibration has been done using the res-onant peaks of O − /O at 6.5 eV and the O − /CO at9.9 eV.[17] The calibration for the kinetic energy dis-tribution measurements have been performed usingthe kinetic energy released by O − /O at 6.5 eV.[18]Further, this energy calibration has been checked bymeasuring the kinetic energy of O − ion produced bydissociative electron attachment to CO [19, 7] at 8.2eV. Fig. 1 shows the ion yield curve of the O − ions pro-duced from CO due to 0-45 eV energy electron colli-sions. The ion yield curve is in good agreement withprevious report.[17] A resonant peak at 9.9 eV due todissociative electron attachment (DEA) is observed. I on c oun t s ( a r b . un i t s )
25 eV 30 eV 35 eV40 eV45eV
Figure 1:
Ion yield curve of O − ion produced due to elec-tron collision with gas phase CO molecule. The arrowsindicate the energies at which the images are taken. The detailed DEA dynamics have been recently stud-ied and reported elsewhere.[8] Increasing the electronenergy revealed an access to the dipolar dissociation(DD) process that results the feature observed in theO − ion yield curve. The main focus of the presentstudy is to understand the detailed dynamics occur-ring at the DD region, i.e., the process beyond 18eV incident electron energy. In Fig. 1, the arrowsindicate the electron energies at which the VSIs aretaken. In the following the threshold behaviour of theDD process seen in the ion yield is discussed withinthe limited energy resolution. The kinetic energy andangular distribution data extracted from VSI of thenegative ions formed due to DD process are analysedthoroughly.Fig. 2 shows the ion yield curve around the thresh-old of the DD region. Due to the finite energy resolu-tion of the electron beam instead of being sharp, thecurve gets smooth near the threshold value. Fromthe experimental data the appearance energy of theanions is found to be near 19.8 eV. The appearanceenergy for the DD process can be calculated usingthe accepted values of the thermochemical parame-ters. [20]Using the conservation of energy, one can3rite the following expression for the DD process as, V e = ( E i + D − A + IP ) + E + E (2)where V e is the amount of energy transfer from in-cident electron to molecule, E i is the energy associ-ated with the possible excited states of the cation, D is the bond dissociation energy, IP is the ionizationpotential of carbon atom and the electron affinity ofoxygen atom is A . The E and E are the kineticenergy associated with the C + and O − ions, respec-tively, at threshold both E and E are zero. As-suming both C + and O − ions are formed in groundstate, [21, 11, 12] i.e. E i = 0 the threshold can becalculated using the expression E T h = ( D − A + IP ) (3)Using the values for thermochemical parameters [20]threshold energy for DD process of CO molecule canbe obtain as 20 .
19 20 21 22 23 24 25 26Incident electron energy (eV) I on c oun t s ( a r b . un i t s ) Figure 2:
Ion yield for the O − ions produced in the dipo-lar dissociation range. The small circles represent the ex-perimental data points . The threshold energy is shownby the arrow. each image and from left to right as indicated by anarrow. Close inspection of each image shows a maxi-mum intensity at the center and a ring pattern withlarger diameter signifying the production of ions hav-ing two kinetic energy bands. The diameter of cen-tral pattern as well as annular pattern remain almostunchanged with increasing incident electron energy.These observations indicate two different mechanismsfor the ion pair formation.The kinetic energy distributions of the O − ionshave been extracted from the above sliced images andare shown in Fig. 4. The distributions have been nor-malised near zero eV. One strong peak near zero eVfollowed by another broad band between 0.7 to 2.0eV are observed. In order to obtain a better per-spective of the second kinetic energy peak, the re-gion 0.5 to 2.5 eV, in Fig. 4, has been magnified by5 times. Low and high kinetic energy bands arisesin the kinetic energy distribution can be explainedby the formation of indirect and direct ion pair pro-cess respectively. For the low kinetic energy bandsthe molecule first excites into a Rydberg state whichcrosses the ion pair state near the ion-pair dissoci-4igure 3: (a)-(e): Time sliced images taken with 50 ns time window of O − ion created due to the ion pair productionat the indicated incident electron energies. (f) represents the same image as shown in (d) but without low energypart for better perspective. The arrows indicate the electron beam direction. − ions (eV) N o r m a li s ed i on c oun t s
25 eV30 eV35 eV40 eV45 eVX5
Figure 4:
Kinetic energy distribution of the O − ions cre-ated due to dipolar dissociation process for five differentincident electron energies. ation limit. This results the predissociation of theRydberg state via the ion pair state. In this case thedynamics of the ion pair dissociation is restrict bythe degree of coupling between the initially excitedRydberg state and the ion pair state. The presenceof low kinetic energy ions clearly indicate that thepredissociation process occurs throughout the entireenergy range. The higher kinetic energy band occursdue to the direct excitation to the ion pair state. Thedynamics of the direct process is determined by theFranck-Condon factor. The initial increase of the ki-netic energy with increasing electron energy is due tothe access at different repulsive part of the ion pairstate. One can calculate the appearance energy forthe direct excitation to the ion pair state by calculat-ing the total kinetic energy release by the molecule.From Fig. 4 it can be observed that the higher kineticenergy band centred at 1.5 eV. From conservationof energy and momentum the kinetic energy of C + ion accompanying a O − ion of 1.5 eV can be foundto be 2 eV. Using the values in expression (2) one5an obtain that nearly 24.4 eV energy is transferredfrom the incident electron to the molecule. So in theFranck-Condon transition region the separation be-tween the ground state of CO molecule and the ionpair state is around 25 eV. The truncated shape seenat 25 eV clearly indicates that such an ion pair reso-nant state enters into the Franck-Condon transitionregion around that energy. From Fig. 4 it can be ob-served that the kinetic energy of the ions remainsunchanged for both lower energy as well as for higherenergy with increasing incident electron energy. Thepossible explanation for this behaviour is as the in-cident electron energy increases, the available energyfor the system is increases. Which can turn on someother excitation and ionisation process but the ionpair production proceeds through the population ofthe very same excited states throughout the energyrange. So from these observations one can concludethat for incident electron energy near threshold theions are formed due to the indirect excitation processwhereas, the direct excitation starts near 25 eV ofincident electron energy.The angular distribution of fragment O − ions com-ing from the ion pair formation process have beenanalysed for both the kinetic energy bands. The ob-served angular distribution from VSIs taken at in-dicated electron energies and within the O − kineticenergy range of 0-0.4 eV and 0.7-2.3 eV are shownin Fig. 5 and Fig. 6, respectively. A 25 ns and 50ns thin time slices have been considered for the an-gular distribution analysis for low and high kineticenergy bands, respectively. In a flat slicing techniquefor the ions with higher kinetic energy only a frac-tion of the entire ‘Newton Sphere’ is considered. Butthe entire ‘Newton Sphere’ contributes in the slicedimages of low kinetic energy ions.[22, 23] In order tominimise this effect, thinner slices must be used forlow kinetic energy ions. We used VSIs taken with 25ns slice (not shown here) for the angular distributionof low energy ions. In the Figs. 5 and 6, symbols arethe experimentally obtained data points and the solidcurves are the fit-to-data using the model discussedbelow. All the data points have been normalised at90 ◦ . For low kinetic energy ions one dominant for-ward lobe and one backward lobe are seen whereas,for high kinetic energy ions one dominant forward (a)(c)(b) ° ) I ( θ ) /I ( ° )
25 eV30 eV35 eV40 eV45 eV0 30 60 90 120 150 18000.511.52 Angle ( ° ) I ( θ ) /I ( ° )
25 eV30 eV35 eV40 eV45 eV0 30 60 90 120 150 1800.811.21.41.61.8 Angle ( ° ) I ( θ ) /I ( ° )
25 eV30 eV35 eV40 eV45 eV
Figure 5:
Angular distribution of the low kinetic energyO − ions created due to the ion pair formation process.Angular distribution data for all incident electron energiesare fitted with (a) Σ to Σ transition, (b) Σ to Π transition,and (c) Σ to Σ+Π transition. Symbols represent the datapoints and lines are the fitted curve. et al. [25] in the study ofoxygen molecule in the dipolar dissociation region.The reduction of anisotropy in angular distributionwith increasing electron energy could be explainedwith the similar argument as given by Zare.[26] Inthe study of angular distribution from electron im-pact dissociation of H +2 ion, Zare concluded that thedecreasing anisotropic nature is due to the K (mo-mentum transfer vector) dependence on I ( θ ) whichis more near threshold.In order to obtain the symmetry of the ion-pairstate(s) involved in the process, the angular distribu-tion data have been fitted using the similar expres-sion described by Van Brunt. [27] According to VanBrunt [27], the experimental data can be fitted usingthe expression as, I ( θ ) = K − n (cid:12)(cid:12)(cid:12)(cid:12)(cid:12)(cid:12) ∞ (cid:88) l = | µ | i l (cid:115) (2 l + 1) ( l − µ )!( l + µ )! j l ( κ ) Y l,µ ( θ, φ ) (cid:12)(cid:12)(cid:12)(cid:12)(cid:12)(cid:12) (4) This is similar with the equation derived by Zare.[26]Where K is the momentum transfer vector betweenthe incident and scattered electron, κ denotes theproduct of the momentum transfer vector K and thedistance of closest approach between the impingingelectron and the center of mass of the molecule, j l ’sare the spherical Bessel function, Y l,µ ’s are the spher-ical harmonics and µ = | Λ f − Λ i | , where Λ i and Λ f are the projection of the electronic axial orbital angu-lar momentum along the molecular axis for the initialand final states, respectively. The l is the angularmomentum of the electron that is participating inthe process. For hetero-nuclear diatomic moleculeslike the present case, l ≥ | µ | , whereas, for homo nu-clear diatomic molecules, l values are restricted toonly even or odd depending upon whether the initialand final states are of same or opposite parity. Thesummation over l takes care of the involvement ofdifferent partial waves. For a transition between twoparticular states for a given incident electron energy,the values of K and n are fixed and can be treated (a)(c)(b) ° ) I ( θ ) /I ( ° )
25 eV30 eV35 eV40 eV45 eV0 30 60 90 120 150 18000.511.522.5 Angle ( ° ) I ( θ ) /I ( ° )
25 eV30 eV35 eV40 eV45 eV0 30 60 90 120 150 1800.511.522.5 Angle ( ° ) I ( θ ) /I ( ° )
25 eV30 eV35 eV40 eV45 eV
Figure 6:
Angular distribution of the high kinetic energyO − ions created due to the ion pair formation process.Angular distribution data for all incident electron energiesare fitted with (a) Σ to Σ transition, (b) Σ to Π transition,and (c) Σ to Σ+Π transition. Symbols represent the datapoints and lines are the fitted curve.
7s parameters. Thus the angular distribution datadue to the involvement of one or more than one finalstate(s) can be fitted using the expression as I ( θ ) = (cid:88) | µ | (cid:12)(cid:12)(cid:12)(cid:12)(cid:12)(cid:12) (cid:88) l = | µ | a l i l (cid:115) (2 l + 1) ( l − µ )!( l + µ )! j l ( κ ) Y l,µ ( θ, φ ) e iδ l (cid:12)(cid:12)(cid:12)(cid:12)(cid:12)(cid:12) (5) Where a l ’s are energy dependent weight factors forthe different partial waves, and δ l ’s denote the phasedifferences between each partial wave responsible forthe transition with respect to the lowest one. Thesummation over µ takes care of the involvement ofmore than one ion-pair states in the process. Inpresent case, we considered upto four lowest partialwaves for each of the state involved.The solid curves in Fig. 5 and 6 represent the fit-to-data using the expression (5). The ground stateof the neutral CO molecule is Σ + (Λ i = 0). The ex-perimental angular distribution along with fits havebeen displayed in Fig. 5 (a, b, c) by considering onlyΣ, only Π and Σ + Π final state(s) and lowest fourpartial waves for the low kinetic energy ions. Sim-ilar data are displayed in Fig. 6 (a, b, c) for highenergy ions. If we consider only Σ final state transi-tion, the angular distributions are well fitted in theforward and backward directions but underestimatein the perpendicular direction. Similarly, if we onlyconsider a Π final state transition, we underestimatein the forward and backward directions while angu-lar distribution in the perpendicular direction fittedquite well. Finally, in order to obtain best fitted data,we need to consider both the Σ and Π final statestransition. The above observations are applicable forboth the low and high energy ions. However, we areunable to comment on whether same Σ and Π stateare responsible for the two cases. The best fittedcurve is shown in Fig. 5(c) and Fig. 6(c) for low andhigh energy ions, respectively. The values of differ-ent parameters used in the fit function are enlisted inTable 1 and Table 2 with R values for low and highenergy ions, respectively. In both the Tables, a , a , a , a are the contribution of each partial wave forthe transition to Σ state, b ’s are the contribution fortransition to Π state and κ ’s are adjustable param-eters. The observed forward-backward asymmetry may be explained in the light of permanent dipolemoment as described by Hall et al. [28] in the DEAto CO. A Σ state contains a large negative dipolemoment that favours backward peak in the angulardistribution, whereas, a Π state contains a positivedipole moment favouring forward peaking. The rela-tive contribution of these two states may explain theobserved forward-backward asymmetry. We have studied dipolar dissociation dynamicsthrough ion-pair states of CO populated in electroncollisions using velocity slice imaging, a well estab-lished method for dissociative electron attachmentstudies. The anion yield has been measured forthreshold energy determination of ion-pair dissocia-tion process. The threshold energy is in good agree-ment with previous reports. Velocity slice imageshave been taken at five different incident electron en-ergies in the dipolar dissociation region. The kineticenergy and angular distributions have been extractedfrom the slice images. Low and high kinetic energybands have been discussed using the indirect and di-rect ion pair formation process. A fixed kinetic en-ergy release with increasing primary electron energyindicates that the molecule absorbs a fixed amountof energy from the incoming electron and that therest of the energy is carried by the out going elec-trons. The truncated nature in the kinetic energydistribution at lower impact energy allow us to lo-cate the position of the ion-pair states with respectto the ground state. Measured kinetic energy dis-tributions clearly indicate that both direct and in-direct ion-pair formation mechanism are responsiblefor the dipolar dissociation of CO. The angular distri-bution data strongly suggest the involvement of twoion-pair states in the studied electron energy range.The symmetry of these observed states are Σ and Πfor both direct and indirect ion pair formation. Wecannot conclude whether the same ion-pair states areinvolved or not in the direct and indirect ion pair8able 1: Fitting parameters for the angular distribution of the O − /CO ions arising from dipolar dissociationprocess with low kinetic energy and fitted with Σ → Σ + Π transition.
25 eV 30 eV 35 eV 40 eV 45 eVWeighting ratio ofdifferent partial waves a : a : a : a : 1.05:1.10:0.12:0.92: 1.08:1.17:0.33:0.53: 0.87:1.01:0.27:0.44: 1.21:1.14:0.33:0.92: 1.02:1.08:0.27:0.55: b : b : b : b δ s − p , δ s − d , δ s − f (rad) 1.188, 1.269, 1.083 0.858, 1.233, 0.941 1.306, 2.072, 1.616 0.858, 1.755, 1.301 0.719, 1.456, 1.008Phase difference (Π) δ p − d , δ p − f , δ p − g (rad) 0.615, 0.707, 0.675 0.941, 0.498, 0.577 0.606, 0.266, 0.186 0.489, 0.148, 0.085 0.693, 0.163, 0.205Parameter κ , κ R value 0.98 0.95 0.91 0.89 0.88 Table 2: Fitting parameters for the angular distribution of the O − /CO ions arising from dipolar dissociationprocess with high kinetic energy and fitted with Σ → Σ + Π transition.
25 eV 30 eV 35 eV 40 eV 45 eVWeighting ratio ofdifferent partial waves a : a : a : a : 0.26:1.08:0.33:0.56: 0:0.96:0.46:1.03: 0.90:0.91:0.27:0.44: 0.16:0.89:0.16:0.21: 1.14:1.14:0.18:0.72: b : b : b : b δ s − p , δ s − d , δ s − f (rad) 1.862, 1.442, 1.532 0.281, 0.276, 0.097 0.888, 1.344, 1.004 0.185, 0.376, 0.204 0.9364, 1.675, 1.275Phase difference (Π) δ p − d , δ p − f , δ p − g (rad) 1.082, 1.225, 0.074 0.825, 0.637, 0.567 1.064, 0.514, 2.072 2.287, 0.239, 0.271 1.853, 0.161, 0.296Parameter κ , κ R value 0.99 0.99 0.96 0.87 0.88 D. N. gratefully acknowledges the partial financialsupport from “Indian National Science Academy” forthe development of VSI spectrometer under INSAYoung Scientist project “SP/YSP/80/2013/734”.
References [1] K. Kuchitsu.
Dynamics of excited molecules .Studies in physical and theoretical chemistry. El-sevier, 1994.[2] Arthur G. Suits and Jhon W. Hepburn. Ion pairdissociation: Spectroscopy and dynamics.
Annu.Rev. Phys. Chem. , 57:431, 2006.[3] X. K. Hu R. C. Shiell, Q. J. Hu, and J. W. Hep-burn. A determination of the bond dissociationenergy (d0(hsh)): threshold ion-pair productionspectroscopy (tipps) of a triatomic molecule.
J.Phys. Chem. A , 104(19):4339–4342, 2000.[4] James DD Martin and JW Hepburn. Elec-tric field induced dissociation of molecules inrydberg-like highly vibrationally excited ion-pairstates.
Phys. Rev. Lett. , 79(17):3154, 1997.[5] Chang Zhou Yusong Hao and Yuxiang Mo. Ve-locity map imaging study of the o2 ion-pair pro-duction at 17.499 ev: Simultaneous parallel andperpendicular transitions.
J. Phys. Chem. A ,109:5832, 2005.[6] Pamir Nag and Dhananjay Nandi. Identificationof overlapping resonances in dissociative electronattachment to chlorine molecules.
Phys. Rev. A ,93:012701, 2016. [7] Pamir Nag and Dhananjay Nandi. Dissociationdynamics in the dissociative electron attachmentto carbon dioxide.
Phys. Rev. A , 91:052705,2015.[8] Pamir Nag and Dhananjay Nandi. Fragmen-tation dynamics in dissociative electron attach-ment to co probed by velocity slice imaging.
Phys. Chem. Chem. Phys. , 17:7130–7137, 2015.[9] E. Illenberger and J. Momigny.
Gaseous Molec-ular Ions: An Introduction to Elementary Pro-cesses Induced by Ionization . Topics in PhysicalChemistry. Steinkopff, 2013.[10] Alfred L. Vaughan.
Phys. Rev. , 38:1687–1695,Nov 1931.[11] W. Wallace Lozier. The heat of dissociation of coand the electron affinity of o.
Phys. Rev. , 46:268,August 1934.[12] Homer D. Hagstrum and John T. Tate. Ioniza-tion and dissociation of diatomic molecules byelectron impact.
Phys. Rev. , 59:354, February1941.[13] F. H. Dorman, J. D. Morrison, and A. J. C.Nicholson. Threshold law for the probability ofexcitation by electron impact.
J. Chem. Phys. ,32:378, February 1960.[14] Dhananjay Nandi, Vaibhav S. Prabhudesai,E. Krishnakumar, and A. Chatterjee. Veloc-ity slice imaging for dissociative electron attach-ment.
Rev. Sci. Instrum. , 76(5):053107, 2005.[15] Pamir Nag and Dhananjay Nandi. Completedata acquisition and analysis system for low en-ergy electron molecule collision studies.
Meas.Sci. Technol. , 26:095007, 2015.[16] O. Jagutzki, A. Cerezo, A. Czasch, R. D¨orner,M. Hattas, Min Huang, V. Mergel, U. Spillmann,K. Ullmann-Pfleger, T. Weber, H. Schmidt-B¨ocking, and G.D.W. Smith. Multiple hit read-out of a microchannel plate detector with athree-layer delay-line anode.
IEEE Trans. Nucl.Sci. , 49(5):2477–2483, Oct 2002.1017] Donald Rapp and Donald D. Briglia. Total crosssections for ionization and attachment in gasesby electron impact. ii. negative-ion formation.
J.Chem. Phys. , 43(5):1480–1489, 1965.[18] Dhananjay Nandi and E. Krishnakumar. Dis-sociative electron attachment to polyatomicmolecules: Ion kinetic energy measurements.
Int. J. Mass Spectrom. , 289(1):39 – 46, 2010.[19] D S Slaughter, H Adaniya, T N Rescigno, D JHaxton, A E Orel, C W McCurdy, and A Belka-cem. Dissociative electron attachment to car-bon dioxide via the 8.2 ev feshbach resonance.
J. Phys. B: At. Mol. Opt. Phys.
Int. J. Mass Spectrom. Ion Phys. ,7:121, August 1971.[22] A. Moradmand, D. S. Slaughter, A. L. Landers,and M. Fogle. Dissociative-electron-attachmentdynamics near the 8-ev feshbach resonance ofco . Phys. Rev. A , 88:022711, Aug 2013.[23] A. Moradmand, J. B. Williams, A. L. Landers,and M. Fogle. Momentum-imaging apparatus forthe study of dissociative electron attachment dy-namics.
Rev. Sci. Instrum. , 84(3):033104, 2013.[24] R. J. Van Brunt and L .J. Kieffer. Electron en-ergy dependence of the energy and angular dis-tributions of o from dissociative ion pair forma-tion in o2.
J. Chem. Phys. , 60(7):3057–3063,1974.[25] D. Nandi, V.S. Prabhudesai, and E. Krishnaku-mar. Velocity map imaging for low-energy elec-tronmolecule collisions.
Radiat. Phys. Chem. ,75(12):2151 – 2158, 2006.[26] Richard N. Zare. Dissociation of h2+ by elec-tron impact: Calculated angular distribution.
J.Chem. Phys. , 47(1):204–215, 1967. [27] Richard J. Van Brunt. Breakdown of the dipole-born approximation for predicting angular dis-tributions of dissociation fragments.
J. Chem.Phys. , 60(8):3064–3070, 1974.[28] R. I. Hall, I. ˇCadeˇz, C. Schermann, andM. Tronc. Differential cross sections for dissocia-tive attachment in co.