This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 128.194.161.133

This content was downloaded on 17/06/2017 at 22:12

Please note that terms and conditions apply.

Dissociative ionization of acetonitrile in intense femtosecond laser fields

View the table of contents for this issue, or go to the journal homepage for more

2017 J. Phys. B: At. Mol. Opt. Phys. 50 135003

(http://iopscience.iop.org/0953-4075/50/13/135003)

Home Search Collections Journals About Contact us My IOPscience

You may also be interested in:

Angular distributions of fragment ions in dissociative ionization of CH2I2 molecules in intense

laser fields

Xia Zhang, Dongdong Zhang, Hang Liu et al.

Dissociative ionization at high laser intensities: importance of resonances and relaxation for

fragmentation

S A Trushin, W Fuß and W E Schmid

Sub-pulselength time resolution of bond softening and Coulomb explosion using polarization control

of laser-induced alignment

J H Posthumus, J Plumridge, P F Taday et al.

Angular distributions of fragment ions

P Graham, K W D Ledingham, R P Singhal et al.

Pathway of H+ in fragmentation of H2 in 810 nm intense femtosecond laser fields

Wei Guo, Xingqiang Lu, Dan Zhao et al.

Field ionization and Coulomb explosion of CO in an intense femtosecond laserfield

Haizhen Ren, Ri Ma, Jianxin Chen et al.

Field-ionization, Coulomb explosion of diatomic molecules in intense laser fields

J H Posthumus, A J Giles, M R Thompson et al.

Dissociative ionization of acetonitrile inintense femtosecond laser fields

Y Boran1, A A Kolomenskii1, M Sayrac1, N Kaya2,3, H A Schuessler1,2 andJ Strohaber4

1 Texas A&M University, Department of Physics, College Station, Texas 77843-4242, United States ofAmerica2 Texas A&M University at Qatar, Science Program, Doha 23874, Qatar3Department of Physics, Faculty of Arts and Science, Giresun University, Giresun, Turkey4 Florida A&M University, Department of Physics, Tallahassee, Florida 32307, United States of America

E-mail: yakupboran@tamu.edu

Received 12 October 2016, revised 13 April 2017Accepted for publication 25 April 2017Published 5 June 2017

AbstractWe investigate the formation of positively charged fragments of acetonitrile (CH3CN) in intense800 nm, 50 fs pulses of radiation using a reflectron time-of-flight (TOF) ion mass spectrometer.Singly-charged ions of CHnCN

+ = -( )n 0 3 , HCN+, CN+, +CH ,3+CH ,2 CH+, C+ and H+; and

the multiply charged ions of C2+, C3+, +CH ,22 and +CH2

3 were observed in the mass spectra.Quantum chemical calculations with GAMESS (General Atomic and Molecular ElectronicStructure System) of appearance energies for the parent molecule and daughter fragments havebeen carried out. Intensity dependent ion yields were measured for intensities between

´ -4.4 10 W cm13 2 and ´ -3.3 10 W cm .14 2 Angular distributions of most fragment ions werefound to peak when the laser radiation was polarized parallel to the TOF axis, while the carbonions, C+ and C2+, were found to have maxima for both polarizations parallel and perpendicularto this axis. Kinetic energies of H+ fragments were experimentally measured and three differentphoto dissociation mechanisms were identified.

Keywords: acetonitrile, femtosecond, photoionization

(Some figures may appear in colour only in the online journal)

1. Introduction

The interaction of intense laser fields with atoms and mole-cules leads to strong field phenomena, such as multiphotonionization (MPI), tunneling ionization (TI), and Coulombexplosion (CE) [1–7]. By focusing amplified femtosecondlaser radiation, peak intensities of -10 W cm15 2 are routinelyachieved. Strong laser fields affect the properties and structureof molecular systems that are different than those found foratomic systems. In previous experiments, the ionization anddissociation of small molecules such as CH4, C2H6, CO2,NO2, H2, and N2 have been studied [8–13], and variousprocesses, such as asymmetric fragmentation, were observed.Photodissociation can occur after one or several electrons areremoved from a molecule by the laser field. At low peakintensities, singly-charged ions may be observed and theyields are described by MPI, where the power law

dependence µY I n can be used to determine the number ofphotons required to ionize the molecules. Above an intensitywhere a significant portion of molecules are ionized, theionization mechanism tends to scale as µ /Y I3 2 due to thevolume effect [8]. Currently, the understanding of the dis-sociation of polyatomic molecules in an intense laser field isan ongoing challenge, and more experimental and theoreticalstudies are required to understand relevant dissociationmechanisms.

In this study, we report on experimental results for theionization and photofragmentation of acetonitrile using areflectron time-of-flight (TOF) ion mass spectrometer.Acetonitrile, also known as methyl cyanide, is of interest toresearchers, since it is the simplest organic nitrile compound[14, 15]. Acetonitrile and its fragments are essential in thestudy of the origins of life. It has astrophysical importance[16] and was found in comets, interstellar clouds and the

Journal of Physics B: Atomic, Molecular and Optical Physics

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 (10pp) https://doi.org/10.1088/1361-6455/aa6f52

0953-4075/17/135003+10$33.00 © 2017 IOP Publishing Ltd Printed in the UK1

Earth’s stratosphere [17]. For these reasons, it is important tounderstand the strong field fragmentation of acetonitrile.Ionization and dissociation of acetonitrile has previouslybeen studied experimentally and theoretically by severalgroups using electron impact and photoelectron techniques[18–21]. However, data on the dissociative ionization ofacetonitrile in femtosecond fields are scarce. From electronimpact ionization experiments the dissociation of acetoni-trile, and appearance energies have been determinedfrom energy distribution [19]. Photoionization and photo-dissociation of acetonitrile were reported in the intensityrange of ´6.3 1013 to ´ -1.2 10 W cm14 2 using 50 fs,800 nm pulses of radiation [22]. In this previous work,detection of the singly charged fragment ions CHnCN

+

(n=0–3) and the mechanism of photodissociation wasreported, but no information on other fragments was pro-vided. We performed a detailed study of all photo-fragmentsof acetonitrile and measured the angular and intensitydependences of the observed fragments. Our work is sup-plemented by quantum chemical calculations for the ioniz-ation and appearance energies using GAMESS (GeneralAtomic and Molecular Electronic Structure System), and theresults are compared with those of the available experiments.To gain additional insight into possible fragmentationmechanisms, we analyzed the kinetic energy of H+ ionsformed through different processes.

2. Experimental setup

Our experimental setup consisted of a Ti:sapphire laser systemdelivering 50 fs, 800 nm, 1 mJ pulses at a repetition rate of1 kHz. For intensity scans, a half-wave plate and a polarizationcube were placed into the path of the laser beam. By rotatingthe half-wave plate, the polarization was changed, and since thecube polarizer transmitted only the horizontal component of thepolarization, the intensity could be adjusted as needed. Whenrotating the polarization for angular scans, the polarizationcube was removed, and the wave-plate was positioned directly

in front of the entrance window. Figure 1 depicts the TOFapparatus consisting of reflected and direct ion paths. Theionization chamber and the flight tube of the spectrometer wereevacuated to a pressure of ∼9×10−9 mbar. Acetonitrile(99.9% purity) was vacuum distilled in a separate chamber andthen introduced into the ionization region through a leak valveto an operating pressure of 5×10−7 mbar. Laser radiation wasfocused on the ionization chamber between a slit and a repellerplate by a lens having a focal length of 22.7 cm. The size of theslit was 400×12 μm. This size was chosen to reduce theinfluence of the spatially varying intensity distribution, so thations were mostly collected from the center of the focal area(Rayleigh range is ∼500 μm and the radius of beam waist is∼12 μm). The repeller plate and the slit plate were separated bya distance of 3 mm. The repeller plate had a positive voltage of1.5 kV, while the slit plate was held at ground potential. Ionscreated in the focal region were accelerated toward the slit bythe electrostatic field created between the plates. Because thelaser beam has a spatial extent, ions having the same mass-to-charge ratio and originating at different positions enter theflight tube with different kinetic energies. For this reason, theseions will have different flight times. To solve this problem, theTOF spectrometer was constructed with a set of grids (ionmirrors) to reflect ions back onto a detector (MCP1). Ions withlarger kinetic energies penetrate deeper into the reflecting grids,while ions with smaller kinetic energies penetrate less deep, sothat with suitably adjusted voltages on the grids all ions arriveat the detector at the same time. Figure 2 shows calculated TOFdispersion curves of the parent and deprotonated ions,CH3CN

+ and CH2CN+, with respect to their initial ion posi-

tions in the ionization region. The flat part of the curves(around 1.7 mm–2.0 mm) demonstrates that ions having thesame mass-to-charge ratio and originating within this region(within the focal spot) arrive at the detector at nearly the sametime. The TOF separation of these ions having masses of41 amu and 40 amu is shown in figure 2 to be about 400 ns. Forlighter fragments, the TOF separations will be even larger e.g.,the time separation between C+ and +CH2 is around 1400 ns.To detect ions, microchannel plate (MCP) detectors were used,and a broadband preamplifier (Mini Circuits ZKL-2, 500MHz

Figure 1. Schematic of the reflectron time-of-flight mass spectro-meter showing the reflected and direct paths. The dotted linesrepresent grids (G) held at fixed potentials. The microchannel platefor the linear mode is MCP2 and that of the reflected mode is MCP1.

Figure 2. TOF for CH3CN+ (dashed blue) and CH2CN

+ (solid red)ions. Time separation between these ions of about 400 ns is shown.Focal position shows the distance from the slit plate.

2

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

frequency bandwidth) was employed to amplify the ion signalfrom the MCPs. The signals were then recorded using amultiscalar (MCS6, FAST ComTec) having a 100 ps timeresolution.

3. Results

3.1. Mass spectra

The TOF mass spectrum of acetonitrile irradiated by laserpulses with a peak intensity of ´ -3 10 W cm14 2 is shown infigure 3. The mass spectra were measured using linearpolarization (horizontal, along the TOF axis; and vertical,perpendicular to the TOF axis). The CH2CN

+ ion peak at m/q=40 was the dominant peak for the different polarizationorientations. The ions of CHnCN

+ (n=0–3) show similarshapes in the mass spectra for both vertical and horizontalpolarizations. However, the H+ and +CH2 peaks werestrongly suppressed when vertical polarization was used. Thepeaks with mass to charge ratios of m/q=18, 28 and 32 arefrom H2O

+, +N2 and +O ,2 and originate from residualatmospheric gases. The peak at m/q=14 is dominated by

+CH ,2 since the appearance energy for N+, which has the

same mass, is greater (33.34 eV) than the appearance energyfor +CH2 (14.50 eV) (calculated with GAMESS). Hishikawaet al [23] also observed that breaking the C–C bond is easierthan breaking the C–N bond by comparing CH3CN

+ anddeuterated CD3CN

+. Some contribution to the m/q=14peak can also come from the dissociation of residual N2, Thelowest required energy for this dissociation was found tobe 24.3 eV [24]. This value is high in comparison to theappearance energy of +CH .2 We also show the mass spectrumfor residual gases at the same intensity, and only H2O

+, +N2

and +O2 are present in the mass spectrum, so that there wasno noticeable contribution to any peaks of interest in the massspectrum from residual gases. These arguments support ourassumption that the peak at m/q=14 and m/q=1 are dueto +CH2 and H+. The HCN+ peak at m/q=27 occursthrough a hydrogen migration process [23].

3.2. Angular distributions

The most noticeable feature in the yields of figure 3 is thesuppression of ion peaks when the laser polarization is orien-tated vertically (i.e. perpendicular to the TOF axis). For thisreason, we measured angular distributions of the fragment ionsat an intensity of 3×1014W cm−2. A half-wave plate, with its

Figure 3. Upper and middle panels show TOF mass spectra for acetonitrile with horizontally and vertically polarized laser beams at theintensity of ´ -3.3 10 W cm .14 2 The bottom panel shows TOF mass spectra for residual gases with horizontally polarized light at the sameintensity with no acetonitrile.

3

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

rotation angle controlled by a LabVIEW program, was used torotate the polarization angle of the laser beam. We measuredthe ion yields from 0° to 180° that are parallel to the TOF axisand for better viewing symmetrically reflected data for theangles between 180° and 360°. Although we expect left–rightsymmetry of these dependences, we observed slight asymme-tries, which may be due to laser power variations. Figure 4shows the measured angular-dependent yield for CHnCN

+

(n=0–3) ions. All ion yields have been normalized to theparent ion CH3CN

+ yield to mitigate pressure effects. Whilethe parent ion, CH3CN

+, is nearly isotropic, the ions CH2CN+,

CHCN+ and CCN+ show an increasing degree of anisotropywith maxima at 0° and 180°, and minima at ∼90° and ∼270°.In figure 4, we also fitted each fragment ion yield with

ås s p b qW = +⎡⎣ ⎤⎦( ) ( )Pd d 4 1 cos ,n n n where s Wd d is

the differential cross section, σ is the integrated cross section,Ωis the solid angle, q( )P cosn is the Legendre polynomial oforder n. β is the anisotropy or asymmetry parameter which hasvalues between −1 and 2 [25, 26]. Table 1 shows the values ofσ and β parameters up to sixth order as well as the maximumnumber of terms used in the fitting. Error bars for our data aresmaller than the symbols used in the angular plots.

Angular-dependent ion yields for the lighter fragments+CH ,2 C+, C2+ and H+ are shown in figure 5. The fragments

H+ and +CH2 were found to exhibit strong anisotropicbehavior with yield maxima at 0° and 180°. The decreases inyields for these ions were observed at 90° and 270°, and theyare the strongest for H+. The angular distribution for C+andC2+ shows two peaks, one for horizontal and one for verticalpolarization. The vertical yield was found to increase as thecharge state increases. The reason for the two maxima is

Figure 4. Angular-dependent normalized ion yields of CH3CN+ (red), CH2CN

+(black), CHCN+(blue), CCN+ (green) at the intensity of3×1014 W cm−2. Solid lines are fitted curves to the differential cross section.

Table 1. Fitting parameters for each angular dependent fragment to ås s p b qW = +⎡⎣ ⎤⎦( ) ( )Pd d 4 1 cos .n n n

s b0 b1 b2 b3 b4 b5 b6 n

H+ 0.2437 1.0000 0.1176 1.7760 0.0005 0.6257 −0.0504 0.2454 14C2+ 0.5909 1.0000 0.0590 −0.1233 0.0021 0.7067 0.0161 −0.0154 10C+ 0.6891 1.0000 0.0273 0.1881 0.0369 0.2254 −0.0067 −0.0301 10

+CH2 0.2395 1.0000 0.0953 1.2936 0.0426 0.8150 −0.0329 0.4310 15CCN+ 0.7319 1.0000 0.0628 0.3478 −0.0253 −0.0065 −0.0351 0.0056 8CHCN+ 0.8493 1.0000 0.0400 0.1687 −0.0235 −0.0114 5CH2CN

+ 0.8751 1.0000 −0.0029 0.0766 3CH3CN

+ 1.0000 1.0000 1

4

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

attributed to the different locations of the carbon atoms inacetonitrile. The ions C+ and C2+ most probably originatefrom the CCN fragment, which is obtained by removing thehydrogens from the parent molecule CH3CN. According toour calculations with GAMESS, to form C+ from CH3CNrequires relatively high energy, i.e. the fragmentation channel

n+ + + ++ - ( )nh eCH CN C H HCN 13 2

requires 17.55 eV; however, much less energy is needed to getC+ from CCN molecules. This observation of perpendicularfragmentation is similar to that observed by Graham et al [27],in which Nq+ (q=1–4) ions are ejected perpendicular to theN2O molecule, which is structurally similar to CCN. Accordingto their work, the peak for horizontal polarization comes fromthe peripheral N atom, while the vertical peak comes from thecentral N atom. Analogously, we expect that the central C atomis responsible for the vertical peak, and the peripheral C atom isresponsible for the peak in the horizontal direction.

3.3. Intensity dependences

We have also studied the intensity dependence of photo-fragmentation of acetonitrile in the range of intensities from4.4×1013W cm−2 to 3.3×1014W cm−2. Measurements at50 different laser intensities were performed while all otherlaser parameters were kept constant. Figures 6 and 7 show theresults of the intensity scan for acetonitrile. The almoststraight segments of the dependences on a log–log scale for

relatively low intensity values between 5×1013W cm−2 and1×1014W cm−2 suggest a multiphoton mechanism underthe assumption of a multiphoton mechanism of photo-dissociation, µ( )Y I I ,n where Y(I) is the ion yield and I is thepeak intensity of the laser beam. The number of photons, n,

Figure 5. Angular-dependent normalized ion yields of +CH2 (green), C+ (red), C2+ (blue), H+ (black) at the intensity of 3×1014 W cm−2.Solid lines show fitting curves.

Figure 6. Intensity dependences of CH3CN+ (blue), CH2CN

+ (red),CHCN+ (black) and CCN+ (green) fragment ions at intensitiesbetween 4.4×1013 W cm−2 and 3.3×1014 W cm−2. All ionsreach saturation around 1.5×1014 W cm−2. Vertical lines show theintensity interval used for determination of the slopes.

5

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

required for the appearance of each ion can be calculated as= ( ( )) ( )n Y I Ilog log , which is just the slope of a straight

line fitted to the ion yield curve plotted on a log–log scale (seefigures 6, 7). The fitting of the dependences with a power-lawfor CH3CN

+, CH2CN+, CHCN+ and CCN+ before the onset

of saturation (5×1013W cm−2�I�1014W cm−2) gives8, 10, 9, and 11, respectively. These slopes are not exactlyinteger values, but we chose the closest integer values foreach slope. The error for the appearance energies will be lessthan one photon energy which is 1.55 eV for 800 nm wave-length. In order to ionize the parent molecule, 8 photons areneeded, so we can conclude that an upper limit of ionizationenergy is 12.40 eV, which is close to the literature values of12.20 eV and 12.38 eV [18, 20]. The appearance energies forother fragments: CH2CN

+, CHCN+ and CCN+ can beobtained in the same way and give the values 15.5 eV,13.95 eV and 17.05 eV, respectively. These values comparefavourably with the appearance energies found by the inter-action of acetonitrile with low energy electrons [22]. Thisindicates that multiphoton ionization is indeed the dominantmechanism in this intensity range. This is further supportedby the Keldysh parameter (between 1.536 and 0.56) for theionization potential of CH3CN in the same intensity range of4.4×1013W cm−2 and 3.3×1014W cm−2. In figure 6, theintensity dependences for H+, C+, +CH2 and CN+ are shown.Similar to CHnCN

+ (n=0–3), these fragments also saturatenear an intensity of about -10 W cm .14 2 The hydrogen ion,H+, yield shows an increase even at intensities higher than

-10 W cm .14 2 Appearance energies calculated from exper-imental results for H+, C+, +CH2 and CN+ are 12.40 eV,12.40 eV, 13.95 eV, and 12.40 eV, respectively.

The ionization and appearance energies of the fragmentswere also calculated using GAMESS with the restrictedHartree–Fock (RHF) approach and the 6-311G basis set.Different pathways were calculated with GAMESS for each

fragment, and the ones that are matching our experimentalresults were chosen. The only possible mechanism forappearance of CH3CN

+ is

n+ ++ - ( )nh eCH CN CH CN , 23 3

where n is the number of required photons and nh is thephoton energy. The calculated value of 12.16 eV is in goodagreement with the measured value of 12.40 eV for the pro-cess in equation (2). Other calculated and experimentalappearance energies are compared in table 2.

3.4. Kinetic energy release of H+

To gain further insight into the mechanisms associated withthe appearance of different ion species, the kinetic energy ofthe H+ ions was measured using the direct path mode of theTOF spectrometer. In this mode of operation, the voltages onthe ion mirror were set to zero, and a small potential wasapplied to the repeller plate, which allows ions with differentkinetic energies to arrive at different times to the MCP2located at the back side of the flight tube (figure 1). Angulardistributions of H+ ions were measured using the samemethod as discussed in the previous section. A two-dimen-sional plot of H+ ion yields as a function of the TOF andpolarization angle is shown in figure 8. In this data, weobserved three different distinct peaks for the H+ ions sug-gesting that not all of the H+ ions arise from the samemechanism. In figure 8, the peak, labelled quasi-equilibriumtheory (QET), located around ∼9.4 μs, is seen to be largelyindependent of the polarization angle. The peak labeled fieldassisted dissociation (FAD) appears around∼9.1 μs and showsa strong enhancement when the polarization is along the TOFaxis. The peak appearing around ∼8.0 μs, and labeled Cou-lomb explosion (CE), is also polarization dependent andbroader. Time intervals of 7.7–8.4 μs, 9.0–9.2 μs, 9.3–9.45μswere used to integrate the peaks for CE, FAD and QET foreach angle. The resulting angular distributions of each peak areshown in polar plots in figure 8(b). The polarization indepen-dence of the QET peak, on the top in figure 8, is explained bythe QET or the Rice–Ramsperger–Kassel–Marcus (RRKM)dissociation mechanism [28]. According to these theories, aftera molecule has been ionized, the new potential energy surfaceis different from the initial potential energy surface, and thesystem is left in an excited state. Excess energy from these‘hot’ molecules is redistributed among the various internaldegrees of freedom. This excess energy is not localized, but itis statistically distributed in the molecule and depending on thesystem there might be enough energy to break a specific bond,resulting in the molecule dissociation. A characteristic of thistype of dissociation is that it is described statistically and dis-sociation only slightly depends on the polarization. In experi-ments on dissociation of +H2 [30] the KER distribution for H+

also exhibited three peaks, and the one with the lowest KERvalues was attributed to bond softening. However, contrary toour results, the bond softening peak is strongly polarizationdependent, while the QET peak we observed shows almostisotropic behaviour, so that the first peak in our data does notcome from bond softening process. The peak labelled FAD

Figure 7. Intensity dependences of yields of H+ (green), C+ (black),+CH2 (red) and CN+ (blue) fragment ions at the intensities between

4.4×1013 W cm−2 and 3.3×1014 W cm−2. All ions excluding H+

reach saturation around 1.5×1014 W cm−2. Vertical lines show theintensity interval used for determination of the slopes.

6

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

shows maxima in the polarization at 0° and 180°, and minimaat 90° and 270°. These peaks are attributed to FAD. In theFAD mechanism, the laser field is considered large enough, sothat it can sufficiently distort the potential energy curve along abond axis thereby effectively ‘pulling off’ a chemical bond[29]. This mechanism strongly depends upon the laser polar-ization and results in a small kinetic energy release from bondsoftening [30]. The peak labeled CE has a broader TOF dis-tribution compared to the other peaks and shows an angulardependence similar to the FAD peak. The broad distributionand angular dependence of the peak indicates that these ionsoriginate from CE. CE produces broader peaks in time becauseof the backward and frontward directed ejection of H+ havinga larger difference in arrival time [31]. As discussed above, it isexpected that CE occurs at larger intensities, while dissociationfollowing QET occurs at lower intensities. In figure 9, theintensity dependent yields of H+ for each process are shown.This data depicts that the peak associated with QET has thelowest appearance intensity similar to the appearance inten-sities of the singly charged ion, where MPI is known to be thedominant ionization mechanism. This is followed by the FADpeak and the CE peak. The CE peak has a larger appearanceintensity and is near that of the doubly charged species. Theorder in which the yields appear on the intensity graph supportsour claim that the H+ ions arise from different processes. Asimilar observation was first observed by Strohaber et al in thedissociative ionization of methane [8].

We note that contributions to the total fragment yieldsfrom different multiphoton processes, ionization from severalmolecular orbitals [32–34] (and not only HOMO) as well asdifferent fragmentation channels can result in a broadening of

Table 2. Comparison of appearance energies of ionized acetonitrile and its fragments obtained by GAMESS calculations and fromexperimental data.

Fragment Reaction GAMESS Exp.

CH3CN+ CH3CN+nhv → CH3CN

++e− 12.16 eV 12.40 eVCH2CN

+ CH3CN+nhv → CH2CN++H+e− 15.77 eV 15.50 eV

CHCN+ CH2CN+nhv → CHCN++H+e− 14.33 eV 13.95 eVCCN+ CH2CN+nhv → CCN++H2+e− 15.10 eV 17.05 eV

+CH2 CH2CN+nhv → +CH2 +CN+e− 14.70 eV 13.95 eVC+ CCN++nhv → C++CN+e− 14.53 eV 12.40 eV

Figure 8. TOF of H+ as a function of the laser polarization using direct TOF path. On the right y axis KERs are shown corresponding toTOFs of H+ (a). Corresponding angular distributions for each mechanism are shown in polar plots (b).

Figure 9. Intensity dependences of H+ yields. Yields for three peaksof figure 8 are shown, which correspond to three different processes:Coulomb explosion (CE, black), field assisted dissociation (FAD,red), and quasi-equilibrium theory (QET, blue).

7

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

the KER distribution of the fragments and also to a mod-ification of the angular dependences of the yields [35]. Therole of the additional contributions can increase with theintensity, when saturation effects become significant and theyield discrimination depending on the minimal requiredenergy for a particular channel becomes less strict.

In order to determine the kinetic energy of the measuredH+ ions, the TOF as a function of focal position in theacceleration region was calculated for frontward and back-ward ejected ions using Newton’s second law,

=+ +

+

⎛⎝⎜⎜

⎞⎠⎟⎟ ( )t

M

Q

x

V

x L v a

x v a

v

a

2

2. 3R

Rlin

0 02

0 02

0

Here M and Q are the mass and charge of an ion, L is thelength of flight tube, xR is distance between the repeller plateand slit plate, x0 is the focal position, VR is the potential onrepeller plate and =v c KE mc20

2 is the initial velocity.The TOFs for =KE 0, =KE 3 eV and =KE 15 eV areplotted in figure 10 using equation (3). In the case ofzero initial kinetic energy =KE 0, the dependence ofequation (3) is shown by the solid red curve in figure 10.For the case where =KE 3 eV, equation (3) gives theblack circles and crosses shown in figure 10. The circlesrepresent the forward directed fragments and the crossesrepresent the backward directed fragments. Similar data isshown by the blue circles and crosses for a kinetic energyof =KE 15 eV.

In order to determine the initial kinetic energy of thehydrogen ions from their TOF in the measured spectrum, theposition of the focus in the acceleration region must bedetermined. This can be accomplished by solving equation (3)for position x ,0

= - - -⎛⎝⎜

⎞⎠⎟

⎛⎝⎜

⎞⎠⎟ ( )x

t V

x

Q

ML

t V

x

Q

ML L

1

2 2 2. 4lin R

R

lin R

R0

2 2 22

Since it is difficult to measure the position of the focalpoint between the repeller and slit plates during the experiment,two different potentials were applied to the repeller plate andthe yields were measured using the same peak intensities(figure 11), which allowed the position of the focus to bedouble checked. Here, we used perpendicular polarizationsince the main peak (highest one in figure 11) is well defined,and also it is easy to isolate from other peaks when thepolarization direction was perpendicular to TOF axis. When ahigher voltage is used, ions arrive at the detector earlier and thepeaks become narrower. The time difference between the twoQET peaks is around 1.3 μs. Inserting all relevant quantitiesinto equation (4), we find that the focal position x0 is equal to1.868 mm for 90V and 1. 864 mm for 117 V, giving anaverage focal position of 1.866±0.002 mm. Using this focalposition, a list of TOFs for H+ ions was calculated using a listof kinetic energies from 0–30 eV. Gaussians were fitted to eachpeak to find the TOF positions for different intensities (Gaus-sian fits are shown in figure 12). Kinetic energies corresp-onding to each centre point of the Gaussians were found using

Figure 10. TOF of H+with different kinetic energy releases. 0 eV(red), 3 eV (black), 15 eV (blue). Circles and crosses show TOF offrontward and backward ejected H+. An inset shows the TOF at thefocal position, around 2 mm, indicating a small time difference(vertical displacement) between arrival times of frontward andbackward ejected H+ fragments.

Figure 11. TOF of H+ ions obtained with linear part of TOFMSusing 90 V (red) and 117 V (black) on the repeller plate at theintensity of 3×1014 W cm−2. This data is used to determine thefocal position in the interaction region. For both measurementspolarization direction was perpendicular to the TOF axis.

Figure 12. Measured TOF at the intensity of 3×1014 W cm−2 withthe polarization parallel to the TOF axis and Gaussian fits for H+.Red line shows Gaussian fits while blue line shows exper-imental TOF.

8

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

the interpolation method in Matlab. Kinetic energy release as afunction of intensity is shown in figure 13 for each of the threepeaks. This figure indicates that the KER of H+ formed via CEincreases with increasing laser intensity from about 9 eV to15 eV. This trend is consistent with reported kinetic energyrelease of H+ formed from propane (C3H8) [36]. The expla-nation is that as the intensity increases, ionization occurs earlierand at a smaller internuclear separation distance resulting in alarger Coulomb force and therefore a larger CE. The middlepeak FAD shows a smaller increase in KER from 2 eV to 3 eVand the QET peak shows no change. In [37], the first peak isattributed to bond softening and the second peak is attributed toabove threshold dissociation. In that paper, KER of these twopeaks decreases with the increasing intensity, but in our resultsKER of these peaks shows a different trend. KER of FAD peakis increasing while the KER of QET peak is not changing.These results further support our claims that the first and sec-ond peaks do most likely not originate from bond softening andabove threshold ionization, but they are coming from QETand FAD.

4. Conclusions

We have performed a detailed experimental study of the ioniz-ation and dissociation of acetonitrile (CH3CN) including obser-vations of the fragment ions’ angular distributions with intense50 fs laser pulses and a reflectron TOF mass spectrometer.Lighter ions showed a strong angular dependence, while theyields of heavier fragments were less affected when the polar-ization was varied. Power dependences of ions in the intensityrange from 4.4×1013W cm−2 to 3.3×1014W cm−2 showedthat the parent CH3CN molecules are excited by a multiphotonprocess, and molecules in the excited states dissociated intosmaller fragments. The measured kinetic energy release of H+

provides evidence that in releasing of protons different dis-sociation mechanisms are involved, which we identified as CE,FAD and through direct ionization (QET).

Acknowledgments

This work was supported by the Robert A Welch FoundationGrant No. A1546 and the Qatar Foundation under the grantNPRP 6-465-1-091. YB acknowledges support from theMinistry of National Education of the Republic of Turkey.

References

[1] Bayfield J E and Koch P M 1974 Multiphoton ionization ofhighly excited hydrogen atoms Phys. Rev. Lett. 33 258–61

[2] Rajgara F A, Krishnamurthy M and Mathur D 2003 Electronrescattering and the dissociative ionization of alcohols inintense laser light J. Chem. Phys. 119 12224–30

[3] Cornaggia C, Schmidt M and Normand D 1994 Coulombexplosion of CO2 in an intense femtosecond laser fieldJ. Phys. B: At. Mol. Opt. Phys. 27 L123

[4] Ilkov F A, Walsh T D G, Turgeon S and Chin S L 1995Dissociative tunnel ionization of H2 in an intense mid-irlaser field Phys. Rev. A 51 R2695–8

[5] Giusti-Suzor A, Mies F H, DiMauro L F, Charron E andYang B 1995 Dynamics of H2

+ in intense laser fields J. Phys.B: At. Mol. Opt. Phys. 28 309

[6] Cornaggia C and Hering P 2000 Nonsequential doubleionization of small molecules induced by a femtosecondlaser field Phys. Rev. A 62 023403

[7] Baldit E, Saugout S and Cornaggia C 2005 Coulomb explosionof N2 using intense 10- and 40 fs laser pulses Phys. Rev. A71 021403

[8] Strohaber J, Zhu F, Kolomenskii A A and Schuessler H A 2014Observation of anisotropic fragmentation in methanesubjected to femtosecond radiation Phys. Rev. A 89 023430

[9] Tian C and Vidal C R 1998 Electron impact dissociativeionization of ethane: cross sections, appearance potentials,and dissociation pathways J. Chem. Phys. 109 1704–12

[10] Oppermann M, Weber S J, Morales F, Richter M,Patchkovskii S, Csehi A, Ivanov V Á , M, Smirnova O andMarangos J P 2014 Control and identification of strong fielddissociative channels in CO2 via molecular alignmentJ. Phys. B: At. Mol. Opt. Phys. 47 124025

[11] Singhal R P, Kilic H S, Ledingham K W D, Kosmidis C,McCanny T, Langley A J and Shaikh W 1996 Multiphotonionisation and dissociation of NO2 by 50 fs laser pulsesChem. Phys. Lett. 253 81–6

[12] McKenna J et al 2006 Ultrafast ionization study of N2 inintense linearly and circularly polarized laser fields Phys.Rev. A 73 043401

[13] Posthumus J H 2004 The dynamics of small molecules inintense laser fields Rep. Prog. Phys. 67 623

[14] Sorokin D Y, van Pelt S, Tourova T P, Takaichi S andMuyzer G 2007 Acetonitrile degradation under haloalkalineconditions by Natronocella acetinitrilica gen. nov., sp. novMicrobiol. 153 1157–64

[15] Mui C, Filler M A, Bent S F and Musgrave C B 2003Reactions of nitriles at semiconductor surfaces J. Chem.Phys. B 107 12256–67

[16] Wakelam V et al 2010 Reaction networks for interstellarchemical modelling: improvements and challenges SpaceSci. Rev. 156 13–72

Figure 13. Measured KER of H+ ions as a function of intensity fordifferent dissociation mechanisms: Coulomb explosion (CE, black),field assisted dissociation (FAD, red), and quasi-equilibrium theory(QET, blue).

9

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al

[17] Delsemme A H 1980 Pristine nature of comets as revealed bytheir UV spectrum Appl. Opt. 19 4007–14

[18] Harland P W and McIntosh B J 1985 Enthalpies of formationfor the isomeric ions HxCCN

+ and HxCNC+ (x = 0−3) by

‘monochromatic’ electron impact on C2N2, CH3CN andCH3NC Int. J. Mass Spectrom. Ion Process. 67 29–46

[19] Nesk̄ovic O, Veljković M, Goloboćanin D A, Miletić M andZmbov K F 1982 Electron impact study of the appearanceenergies of positive ions from CH3CN Int. J. MassSpectrom. Ion Phys. 44 61–80

[20] Lake R F and Thompson H 1970 The photoelectron spectra ofsome molecules containing the C–N group Proc. R. Soc.London Ser. A 317 187–98

[21] Staley R H, Kleckner J E and Beauchamp J L 1976Relationship between orbital ionization energies andmolecular properties. Proton affinities and photoelectronspectra of nitriles J. Am. Chem. Soc. 98 2081–5

[22] Wu C, Xiong Y, Gao Z, Kong F A, Lu H, Yang X and Xu Z2000 Ionization and dissociation of acetonitrile by intensefemtosecond laser pulse Chin. Sci. Bull. 45 1953–5

[23] Hishikawa A, Hasegawa H and Yamanouchi K 2004 Hydrogenmigration in acetonitrile in intense laser fields in competitionwith two-body Coulomb explosion J. Electron Spectrosc.Relat. Phenom. 141 195–200

[24] Locht R, Schopman J, Wankenne H and Momigny J 1975 Thedissociative ionization of nitrogen Chem. Phys. 7 393–404

[25] Sanov A 2014 Laboratory-frame photoelectron angulardistributions in anion photodetachment: insight intoelectronic structure and intermolecular interactions Annu.Rev. Phys. Chem. 65 341–63

[26] Nakajima K et al 2015 Photoelectron diffraction from laser-aligned molecules with x-ray free-electron laser pulses Sci.Rep. 5 14065

[27] Graham P, Ledingham K W D, Singhal R P, McCanny T,Hankin S M, Fang X, Tzallas P, Kosmidis C, Taday P F andLangley A J 2000 The angular distributions of fragment ions

from labelled and unlabelled N2O in intense laser fieldsJ. Phys. B: At. Mol. Opt. Phys. 33 3779

[28] Baer T and Mayer P M 1997 Statistical Rice–Ramsperger–Kassel–Marcus quasiequilibrium theory calculations in massspectrometry J. Am. Soc. Mass Spectrom. 8 103–15

[29] Wang S, Tang X, Gao L, Elshakre M E and Kong F 2003Dissociation of methane in intense laser fields J. Chem.Phys. A 107 6123–9

[30] Pavičić D, Hänsch T W and Figger H 2005 Vibrationallyresolved strong-field dissociation of +D2 ion beams Phys.Rev. A 72 053413

[31] Graham P et al 1999 An investigation of the angulardistributions of fragment ions arising from the linear CS2and CO2 molecules J. Phys. B: At. Mol. Opt. Phys. 325557

[32] Arthur Z, Péter S, Tamás R and Thomas W 2014 Removingelectrons from more than one orbital: direct and indirectpathways to excited states of molecular cations J. Phys. B:At. Mol. Opt. Phys. 47 204023

[33] Sayres S G, Ross M W and Castleman A W 2011 Influence ofclustering and molecular orbital shapes on the ionizationenhancement in ammonia Phys. Chem. Chem. Phys. 1312231–9

[34] Boguslavskiy A E, Mikosch J, Gijsbertsen A, Spanner M,Patchkovskii S, Gador N, Vrakking M J J and Stolow A2012 The multielectron ionization dynamics underlyingattosecond strong-field spectroscopies Science 335 1336

[35] Xie X et al 2014 Selective control over fragmentation reactionsin polyatomic molecules using impulsive laser alignmentPhys. Rev. Lett. 112 163003

[36] Jianxin C, Ri M, Xia L, Haizhen R, Hong Y and Qihuang G2004 Coulomb explosion of propane in intense femtosecondlaser fields J. Phys. B: At. Mol. Opt. Phys. 37 2501

[37] Wei G, Xingqiang L, Dan Z and Xinlin W 2014 Pathway ofH+ in fragmentation of H2 in 810 nm intense femtosecondlaser fields Phys. Scr. 89 025401

10

J. Phys. B: At. Mol. Opt. Phys. 50 (2017) 135003 Y Boran et al