Absolute cross-sections of fragment negative ions in electron collisions with difluoromethane
aa r X i v : . [ phy s i c s . a t m - c l u s ] J un Absolute cross-sections of fragment negative ions inelectron collisions with difluoromethane
Dipayan Chakraborty and Dhananjay Nandi Indian Institute of Science Education and Research Kolkata, Mohanpur 741246, India email: [email protected], [email protected] Abstract
Dissociative electron attachment (DEA) and ion-pair dissociation (IPD) processes of Difluoromethane(CH F ) have been studied in the incident electron energy range 0 to 45 eV. Three different fragmentanions (F − , CHF − and F − ) are detected in the DEA range and two anions (F − and CHF − ) are detectedin IPD range. Absolute cross-section of the F − fragment ion is measured for the first time. Three differentresonances for both F − and CHF − ions and one single resonance peak for the F − ions are observed.Constant increase in ion counts above 8 eV incident electron energy indicates the involvement of IPDprocess. From the experimental observation, it is speculated that near 11 eV incident electron energy bothDEA and IPD processes occur simultaneously. Total elastic and inelastic electron scattering cross-section studies of fluoromethanes is a topic of interestthese days [1]. Cross-section values of these moleculeshave a demand because of its application to plasmaprocessing in the semiconductor industry. Besidesthis, Chlorofluorocarbons (CFC) and Hydrofluoro-carbons (HFC) are the main reason behind the Ozonelayer depletion due to their photolytic decomposition.However decomposition can also occur due to the lowenergy electron attachment process [2]. So, it is abso-lutely necessary to have accurate electron attachmentcross-section values of these molecules. Several stud-ies are performed to address this problem [3, 4, 5] butfor Methylene fluoride (CH F ) it is very rare.Electronic structures and energies of the Fluo-romethanes have been studied long ago by Brundle et al. [6]. In 1997, Tanaka et al. [7] obtain theirelastic differential cross sections below 100 eV inci-dent electron energy, later the integral cross-sectionis also calculated [8]. Later Tamio Nishimura [9] the-oretically calculate the vibrationally elastic scatter-ing cross section of CH F with electron collision below 30 eV. All these studies are limited to onlyelastic electron scattering cross section of CH F butthe inelastic scattering processes like dissociative ion-ization (DI), DEA and IPD process of CH F withelectron collisions are not studied in detail. In 1998,Motlagh and Moore [10] studied the electron impactDI process of CH F molecule up to 500 eV incidentelectron energy range. Later Torres et. al [11] havestudied the same upto 100 eV energy range by us-ing time-of-flight mass spectrometry method. Theauthors have discussed the appearance energy, abso-lute total and dissociative ionization cross sectionsand the corresponding kinetic energy of the fragmentions in details. The other inelastic scattering pro-cesses like DEA and IPD are still ignored for CH F though it is equally important like other moleculesin fluoromethane group. As per authors concern theonly reported studies of DEA and electron impactIPD process of CH F is done by Scheuremann et al. [12]. In this article, the authors have observed the for-mation of F − and CHF − fragment ions up to 20 eVincident electron energy range. Excitation functionof the fragment anions beyond 6 eV incident electronenergy is reported, although the corresponding ab-1olute cross-section values are not measured. In thepresent article, DEA and IPD processes of CH F are studied from 0 to 45 eV incident electron energy,using an advance time of flight mass spectrometer(TOFMS) developed in our group [13]. Three differ-ent fragment anions are observed. Dissociation path-ways and corresponding appearance energies of thefragment ions are discussed based on the experimen-tal observations. One low energy temporary negativeion (TNI) state around 2 eV incident electron energyis observed for the first time followed by three higherenergy TNI states, in agreement with the previousreport [12]. Absolute DEA and IPD cross-sectionsof the negative ions are measured in the above men-tioned energy range. Details of the experimental setup and the measure-ment procedure is described elsewhere [13]. Here themeasurement technique is discussed briefly. Basictheme of the experiment is magnetically collimatedpulsed electron beam with 200 ns pulse width and 10kHz repetition rate is interacted perpendicularly withan effusive molecular beam produced through a nee-dle of diameter 1 mm. Tip of the needle was kept 4mm away from the interaction region. Negative ionsformed in the interaction region are guided througha spectrometer and collected by the detector. Theenergy of the emitted electrons is controlled by an ex-ternal power supply which is connected with the elec-tron gun filament. Filament current is provided by aconstant current supply and the electrons are emit-ted via thermionic emission process. The electronbeam current has been measured by using a Fara-day cup, placed opposite to the electron gun in theinteraction region. Time-averaged electron beam cur-rent during the measurement was around 3 nA. Twomagnetic coils in Helmholtz configuration are used tocollimate the electron beam. Typical strength of themagnetic field is 30 Gauss. Axis of the spectrome-ter is situated perpendicular to both electron beamand molecular beam. Spectrometer contains a pusherplate, a puller plate, three lens electrodes, a drift tubeand a mesh grid. The electron-molecule interaction occurs in between pusher and puller plates. Thesepusher and puller plates consist of wire mesh of 90%transmission efficiency to avoid field penetration intothe interaction region. After the puller plate, threelens electrodes are placed in Einzel lens configura-tion to focus the negative ions. Applied voltage tothe three electrodes are 90, 1030 and 90 volts respec-tively. In order to increase the mass resolution of thespectrometer, one field free drift tube is placed af-ter the lens electrodes. At the end of the drift tube,one cap electrode with wire mesh is placed to avoidfield penetration from the detector. Both the drifttube and the cap electrode is biased to 1590 V. Af-ter the drift tube, MCP based detector is placed tocollect the negative ions. The detector consists oftwo micro channel plates (MCP) in Chevron config-uration, along with one collector plate. The TOF ofthe detected ions are determined from the back MCPsignal. Experiments were performed under ultra highvacuum conditions with base pressure 10 − mbar andwith 99.9% pure commercially available CH F gas.The absolute cross-section of the F − fragment anionhas been measured by using relative flow technique(RFT) [14, 15, 16]. RFT is basically a calibrationprocedure where one just needs to compare the rela-tive intensities of the species of interest with a stan-dard species of known cross section, by keeping theother experimental conditions unchanged. For exam-ple, in the present case, the absolute cross section forthe formation of F − fragment ions from CH F canbe determined by using the dissociative electron at-tachment (DEA) cross section [17] of O − ion from O by using the equation as σ (F − / CH F ) = σ (O − / O ) N (F − ) N (O − ) I e (O ) I e (CH F ) × (cid:18) M O M CH F (cid:19) / F O F CH F K (O − ) K (F − ) . (1)Here N is the number of fragment ions collected fora fixed time, F is the flow rate of the correspondinggases, I e is the time average electron beam current, M is the molecular weight of the parent molecules, K is the detection efficiency and σ is the absolute crosssection. All these factors and their contributions in2igure 1: Mass spectra of the CH F at 11 eV in-cident electron energy. Three different masses F − ,CHF − and F − are shown respectively.the overall measurements are discussed in details inthe previous study [13]. When low energy electrons are collided with the neu-tral molecules, temporary negative ion (TNI) states(CH F ) ∗− are formed via dissociative electron at-tachment (DEA) process. This TNI further decaysvia three possible dissociation channelsCH F + e − → (CH F ) −∗ → F − + CH FCHF − + HFF − + CH (2)Here F − channel is a simple bond cleavage whereas,CHF − and F − channels are associated with the re-arrangements in the TNI. Fig. 1 represents the massspectra of CH F molecule obtained at 11 eV inci-dent electron energy. Here X-axis is calibrated in themass unit which reveals that, in this energy region three different fragment ions F − , CHF − and F − areformed. In the previous experimental study, first twochannels (F − and CHF − ) were observed [12] but thethird channel(F − ) is observed for the first time. Fig.2, 3 and 4 are the corresponding excitation functionof the three fragment ions. From those excitationfunctions one can observe that for F − ions, one lowenergy peak around 2 eV followed by two higher en-ergy peaks around 11 eV and 15.2 eV are present.Whereas for CHF − ions, along with 2 eV peak, onlyone broad peak around 11 eV is present. With closeinspection one small hump near 10 eV incident elec-tron energy for F − ion can also be observed. But forF − ions only one broad peak near 2 eV incident elec-tron energy is observed. From these observations onecan conclude that in the Franck-Condon (FC) tran-sition region of neutral CH F molecule, temporarynegative ion (TNI) states are present around thoseenergies. Presence of lower energy peak due to DEA ofchlorofluorocarbon is quite natural and has beenobserved in several studies [18]. In 1992 Modelli et. al. [19] studied the electron attachment to thehalomethanes via electron transmission spectroscopyand observed low energy electrons are attached withthe halomethanes though for fluoromethanes theywere unable to measure any low energy ( < et. al. measured thenegative ion formation due to low energy electroncollision to CF Cl . In this study, the authorsobserved low energy NIR state which dissociates viadifferent fragment negative ions. They conclude thatthe NIR state, which dissociates via Cl − negative ionformation, is formed due to Shape Resonance, wherethe incoming electron is occupying the molecularorbital with σ ∗ (C-Cl) character. However theyare unable to comment on the nature of the NIRstate which dissociates via F − dissociation channel.According to author’s concern for CH F molecule,below 6 eV incident electron energy region noexperimental study has been done till date.Within the FC transition window the low energyelectron attachment to the ground state CH F − ion. Two lowerenergy peaks at 1.9 eV and 11.4 eV are observed.Two small humps at 10.1 eV and 15.2 eV are shownby the arrow.creates the TNI states which further dissociatesto the negative fragment ions via correspondingdissociation channels. After formation of the TNI,there will be two competing channels. One is thedissociation of the TNI via one negative and otherneutral fragment, other is the auto detachment (AD)where the TNI ejects the electron and back to theparent neutral molecule (may be with vibrationallyexcited states). The life time of the TNI actuallydetermines the cross-section or the probability ofnegative ion formation. If the life time of the TNIis subsequently high so that it can cross the criticaldistance R c , then the DEA cross section will be highotherwise AD of the TNI occurs. In the presentcontext, measured DEA cross section is around10 − cm for the F − channel. For CHF − and F − channels the cross-section is too low to calculate anyreliable absolute value, thus only differential valuesare reported. Incident electron energy (eV) I on c oun t s ( a r b . un i t s ) Figure 3: Ion yield curve for CHF − ion. Two lowerenergy peaks at 1.9 eV and 11.4 eV are observed. In the measured excitation function of the three chan-nels (Fig. 2, 3 and 4) one resonance peak near 11 eVis observed. Two small humps near 10 eV and 15.2eV incident electron energy for F − ions are also ob-served. This higher energy resonance peaks can bedescribed from previous experimental as well as the-oretical studies. Using electron transmission spec-troscopy (ETS) and multiple scattering X α (MS-X α )bound state calculations, Modellii et al. [19] stud-ied the electron attachment to halomethanes. Inthis study, the experimentally obtained electron at-tachment energies and theoretically calculated val-ues match nicely. Using MS-X α values the authorspredict that in CH F molecule, two broad σ ∗ reso-nances with symmetries b and a with comparableintensities around 10 eV are responsible for the elec-tron attachment cross-section. In the present study,clear signature of the TNI states around those above-mentioned energy ranges are observed. Two peaks asmentioned in the theoretical study is not possible toobserve separately due to poor electron gun resolu-tion. Besides these two DEA resonances, one peaknear 15.2 eV of the F − ion yield curve has been ob-served, but due to unavailability of any theoretical4 I on C oun t s ( a r b . un i t s ) Figure 4: Ion yield curve for F − ion. One lower en-ergy peak around 1.9 eV electron energy is prominentand one small hump around 11.4 eV is shown.studies we are unable to make any comment on that.However, the presence of 15.2 eV peak in the excita-tion function of F − ion has been observed previouslyby Scheuremann et al. [12]. With close inspection of the excitation function, onecan observe the cross-sections of F − , CHF − ionsgradually increasing beyond 8 eV incident electronenergy. This clearly indicates that the ion-pair disso-ciation (IPD) process starts around this energy. Samebehaviour is observed in the previous measurement[12]. Unlike DEA process, in IPD process the elec-trons are not captured by the molecule. Here the in-cident electron transfers some energy to the moleculeand excites it to the ion-pair state that eventuallydissociates into an anion and a cation. This ion-pairdissociation is possible as long as the energy of the in-cident electron is equal to or more than the excitationenergy required because the extra energy is always carried out by the outgoing electron [20]. The IPDprocess in the present experiment can be expressedas CH F + e − → (cid:26) F − + CH F + + e − CHF − + HF + + e − (3)Due to low cross-section, we are unable to commentanything about the F − channel. For F − channel thecross section is quite large in the IPD range com-pared to DEA range whereas, for CHF − channel thecross-sections in the two regions are comparable. Bylooking at the experimental results and the presentunderstanding it is speculated that in the FC tran-sition region the TNI state lies above the ion-pairstate. As a result, around 11 eV incident electronenergy both the DEA and IPD processes occurred si-multaneously.Obtained absolute cross section value of F − ion islisted in Table 1. Extreme care has been taken toconfirm that these peaks are not coming from im-purities. The mass resolution of the spectrometer ishigh enough that one can separate the mass differ-ence of 1 amu within this range with kinetic energyof fragments up to 5 eV. As mentioned earlier theexperiments have been performed with 99.9 % pureCH F gas and the chamber is kept in ultra high vac-uum (10 − mbar) for more than one week before theexperiment. So we confirmed the error due to the im-purity is negligible here. As the cross-section is verylow, there is a probability that the peaks in the cross-section curve are present due to the collision with sec-ondary electrons [21]. This can be identified by nonlinear pressure dependence of the peak intensities. Toavoid this secondary effects we have used very lowbackground pressure (10 − mbar). Thus we confirmthe peaks appearing in both the cross-section curvesare completely due to the dissociation of CH F withelectron collision. It is to be mentioned here that theoverall uncertainty in our measurement is within 15%[13].5able 1: Absolute cross section for the formation of F − and CHF − ions due to electron collisions with groundstate CH F molecule. The peak positions are in units of eV and the cross sections ( σ ) are in units of 10 − cm . Ion Peak position (eV) Peak cross section ( × − cm )RFTF − /CH F The absolute cross-section of F − ions due to elec-tron collision with CH F has been measured for thefirst time from 0 to 45 eV electron energy range byusing relative flow technique (RFT). One low en-ergy peak near 2 eV followed by higher energy peaksnear 10, 11 and 15.2 eV has been observed. Thesepeaks are the signature of dissociative electron at-tachment process. The ground state symmetry ofCH F is C v [9]. After colliding with lower ener-getic electrons ( <
15 eV), it forms temporary nega-tive ion (TNI) state (CH F ) −∗ and dissociates viathree different dissociation channels producing F − ,CHF − and F − fragment anions. Depending uponthe energy and possible symmetries of the remainingneutral fragments, threshold of different dissociationchannels may vary. More than one different symme-tries of (CH F ) ∗− may involve in this dissociationprocess. In the present measurements, we are unableto comment on the possible symmetry of the TNIstates. It was theoretically predicted that two nega-tive ion resonant states of symmetries b and a arepresent around 10 eV incident electron energy. Inthe present measurements the presence of TNI statesaround 10 eV electron energy range is experimen-tally verified. Though due to poor electron energyresolution, we are unable to separate them out. Theconstant increase in the cross-section curve indicatesthat beyond 8 eV incident electron energy IPD pro-cess starts. More than one different symmetries ofion-pair states may involve in the process. From theexperimental observation it is speculated that near11 eV electron energy, both DEA and DD processes occur simultaneously. D. N. gratefully acknowledges the partial fi-nancial support from “Indian National Sci-ence Academy (INSA)” under project No.“SP/YSP/80/2013/734” and from “Science andEngineering Research Board (SERB)” under theproject No. “EMR/2014/000457”. DC is thankfulto IISER Kolkata for providing research fellowship.
References [1] L. G. Christophorou, J. K. Olthoff, and M. V.V. S. Rao, J. Phys. Chem. Ref. Data , 1341(1996).[2] E. Illenberger and J. Momigny, Gaseous Molec-ular Ions: An Introduction to Elementary Pro-cesses Induced by Ionization
Topics in PhysicalChemistry (Steinkopff, 2013).[3] M. Stano et al. , J. Phys. B: At. Mol. Opt. Phys. , 443 (2003).[4] P. Rawat, V. S. Prabhudesai, M. Rahman, N. B.Ram, and E. Krishnakumar, Int. J. Mass Spec-trom. , 96 (2008), Electron-induced atomicand molecular processes: A special issue honor-ing Eugen Illenberger on his 65th birthday.[5] J. H. Moore, P. Swiderek, S. Matejcik, andM. Allan, Nanofabrication Using Focused Ion nd Electron Beams, Chapter 4. (Oxford Uni-versity Press, 2012).[6] C. R. Brundle, M. B. Robin, and H. Basch, J.Chem. Phys. , 2196 (1970).[7] H. Tanaka, T. Masai, M. Kimura, T. Nishimura,and Y. Itikawa, Phys. Rev. A , R3338 (1997).[8] M. T. d. N. Varella, C. Winstead, V. McKoy,M. Kitajima, and H. Tanaka, Phys. Rev. A ,022702 (2002).[9] T. Nishimura, J. Phys. B: At. Mol. Opt. Phys. , 3471 (1998).[10] S. Motlagh and J. H. Moore, The Journal ofChemical Physics , 432 (1998).[11] I. Torres, R. Martnez, M. N. S. Rayo, andF. Castao, J. Phys. B: At. Mol. Opt. Phys. ,3615 (2000).[12] H.-U. Scheunemann, M. Heni, E. Illenberger,and H. Baumgrtel, Ber. Bunsenges. Phys. Chem. , 321 (1982).[13] D. Chakraborty, P. Nag, and D. Nandi, Rev.Sci. Instrum. , 025115 (2018).[14] E. Krishnakumar and S. K. Srivastava, J. Phys.B: At. Mol. Opt. Phys. , 1055 (1988).[15] O. J. Orient and S. K. Srivastava, J. Chem.Phys. , 2949 (1983).[16] S. K. Srivastava, A. Chutjian, and S. Trajmar,J. Chem. Phys. , 2659 (1975).[17] D. Rapp and D. D. Briglia, J. Chem. Phys. ,1480 (1965).[18] J. Langer et al. , J. Chem. Phys. , 11063(2000).[19] A. Modelli, F. Scagnolari, G. Distefano,D. Jones, and M. Guerra, J. Chem. Phys. ,2061 (1992).[20] D. Chakraborty, P. Nag, and D. Nandi, Phys.Chem. Chem. Phys. , 32973 (2016).[21] P. J. Chantry, J. Chem. Phys. , 1851 (1971). 7, 1851 (1971). 7