dynamics of cl ( p) atom formation in the ... · and 4p 2p 1/2 r3p 2p 1/2 transitions,...

Published: February 15, 2011

r 2011 American Chemical Society 1538 dx.doi.org/10.1021/jp1100279 | J. Phys. Chem. A 2011, 115, 1538–1546

ARTICLE

pubs.acs.org/JPCA

Dynamics of Cl (2Pj) Atom Formation in the Photodissociation ofFumaryl Chloride (ClCO - CH = CH - COCl) at 235 nm: A ResonanceEnhanced Multiphoton Ionization (REMPI) Time-of-Flight (TOF) StudyMonali Kawade, Ankur Saha, Hari P. Upadhyaya,* Awadhesh Kumar, Prakash D. Naik, and P.N. Bajaj

Radiation & Photochemistry Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India

ABSTRACT: The photodissociation dynamics of fumarylchloride (ClCO—CHdCH—COCl) has been studiedin a supersonic molecular beam around 235 nm usingresonance enhanced multiphoton ionization (REMPI)time-of-flight (TOF) technique by detecting the nascentstate of the primary chlorine atom. A single laser has beenused for excitation of fumaryl chloride and the REMPIdetection of chlorine atoms in their spin-orbit states, Cl(2P3/2) and Cl* (2P1/2). We have determined the transla-tional energy distribution, the recoil anisotropy parameter, β, and the spin-orbit branching ratio for chlorine atom eliminationchannels. To obtain these, measured polarization-dependent and state-specific TOF profiles are converted into kinetic energydistributions, using a least-squares fitting method, taking into account the fragment recoil anisotropies, βi. The TOF profiles for bothCl and Cl* are found to be independent of laser polarization; i.e., β is well characterized by a value of 0.0, within the experimentaluncertainties. Two components, namely, the fast and the slow, are observed in the translational energy distribution, P(ET), of Cl andCl* atoms, and assigned to be formed from different potential energy surfaces. The average translational energies for the fastcomponents of the Cl and Cl* channels are 14.9( 1.6 and 16.8( 1.6 kcal/mol, respectively. Similarly, for the slow components, theaverage translational energies of the Cl and Cl* channels are 3.4( 0.8 and 3.1( 0.8 kcal/mol, respectively. The energy partitioninginto the translational modes is interpreted with the help of various models, such as impulsive and statistical models. Apart from thechlorine atom elimination channel, molecular hydrogen chloride (HCl) elimination is also observed in the photodissociationprocess. The HCl product has been detected, using a REMPI scheme in the region of 236-237 nm. The observation of themolecular HCl in the dissociation process highlights the importance of the relaxation process, in which the initially excited parentmolecule relaxes to the ground state from where the molecular (HCl) elimination takes place.

I. INTRODUCTION

Photodissociation dynamics of carbonyl compounds alwayshas been an area of extensive research that has been pursued withboth experimental and theoretical approaches. The existence ofclose-lying excited states and their interactions is mainly a greatattraction for the researchers. Again, the compounds with twocarbonyl groups (diketones), provide useful information for thestudy of intramolecular energy transfer, and interactions invol-ving remote carbonyl groups.1 It is very interesting to study andidentify themechanism of these types of interactions between thecarbonyl groups. Besides these diketones, there is another class ofcompounds, generally termed as R,β-enones, in which thecarbon-carbon double bond (-CdC-) is in conjugation withcarbonyl group (-CdO) giving rise to altogether a differentand interesting photochemistry. These R,β-enones undergo avariety of photochemical rearrangements,2,3 and their photo-chemistry, in general, has attracted much attention in recentyears. In our earlier studies, we investigated the dynamics offormation of the OH radical in photodissociation of few enones,such as acrylic acid [H2CdCHCOOH],3 propiolic acid,4 eno-lic-acetylacetone [CH3C(O)CHC(OH)CH3],

2 and pyruvic

acid [CH3COCOOH]5 using laser photolysis-laser inducedfluorescence (LP-LIF) technique.

Numerous studies on the photodissociation of carbonyl chlo-rides are available in the literature; very few of these are focusedon the photochemistry of fumaryl chloride. The UV photodisso-ciation of the most elementary acid chlorides, acetyl chloride(CH3COCl),

6,7 and propionyl chloride (CH3CH2COCl), havebeen previously studied in both gas and solid phases.8 Acetylchloride on photodissociation at 236 nmproduces acetyl and chlorineradicals while propionyl chloride gives ketene (H2CdCdO) andHCl as well. Photodissociation and isomerization processes ofacryloyl chloride (CH2dCHCOCl), an enone similar to fumarylchloride, have been studied both experimentally9,10 andtheoretically.11 Apart from the Cl atom formation channel, itphotodissociates through two additional channels, one producesketene and HCl (same as in propionyl chloride), while theother channel involves a 1,3-sigmatropic rearrangement, produc-ing 3-chloro-1,2-propenone (H2ClCCHdCdO). Extensive UV

Received: October 20, 2010Revised: January 12, 2011

1539 dx.doi.org/10.1021/jp1100279 |J. Phys. Chem. A 2011, 115, 1538–1546

The Journal of Physical Chemistry A ARTICLE

photodissociation studies of oxalyl chloride, (ClCO)2, a lowerhomologue of fumaryl chloride, have been carried out.12-14 Suitsand co-workers12,13 investigated photodissociation dynamics ofoxalyl chloride near 235 and at 193 nm, with a photofragmentimaging technique, using REMPI detection of Cl and COmoieties. They suggested that the photodissociation of(ClCO)2 proceeds via an impulsive three-body mechanismyielding translationally hot spin-orbit excited Cl* and groundstate Cl, together with translationally hot and rotationally excitedCO, and a translationally cold ClCO. The primary product COClundergoes subsequent dissociation to yield slow CO and Cl.They also measured the translational energy distribution andanisotropies of Cl, Cl*, and CO.

Fumaryl chloride, as shown in Figure 1, is a planar moleculeand has three stable isomers, out of which two isomers, namely,trans-trans and cis-cis, belong to the C2h point group and thethird trans-cis isomer belongs to the Cs point group.

15-18 Therelative abundances of these isomers are found to be 30%, 24%,and 46%, respectively, at 480 K.18 A negligible effect of tempera-ture on the relative abundance indicates only small energydifferences among the isomers. The complete IR and Ramanstudies of liquid and crystal forms have identified two conformersas trans-trans and cis-trans.19 It has a symmetric structure withboth carbonyl groups associated with an identical functionality.The thermal decomposition of fumaryl chloride at 900 �C givesrise to various products, such as C3O(CdCdCdO), HCl, andCO.20 Besides, fumaryl chloride has many synthetic applications,mainly since it contains unsaturated carbon-carbon doublebonds, which can be used for in situ cross-linking. Jabbari etal.21 have suggested that it can be copolymerized with a poly(ε-caprolactone) for constructing a biocompatible, bioresorbable,injectable, and self-cross-linkable polymer for bone tissue en-gineering. Poly(ethylene glycol) fumarate, synthesized fromfumaryl chloride and polyethylene glycol in the presence ofpropylene glycol, is used as a rigid coating for the superpara-magnetic iron oxide nanoparticles, which are increasingly beingevaluated for clinical applications such as hyperthermia, drugdelivery, magnetic resonance imaging, transfection, and cell/protein separation.22 Several ultrastructure cross-linked andfunctional polymer thin films, for potential nonlinear optical(NLO) application, or other applications requiring molecularorientation, are synthesized from fumaryl chloride and maleicanhydride.23

The present study is undertaken to understand the photo-dissociation dynamics of fumaryl chloride (ClCO—CHdCH—COCl), a molecule with two CdO groups and a CdC group inconjugation, in a supersonic molecular beam around 235 nm.The chlorine atom product in both the ground state (2P3/2) andthe spin-orbit excited state (2P1/2) has been detected using theREMPI-TOF technique. We have also measured the transla-tional energy distributions for both types of chlorine atoms and

relative quantum yield of Cl*. To gain further insights, theanisotropy parameters (βi) were also measured.

II. EXPERIMENTAL SECTION

The experiment was performed, using a molecular beam time-of-flight mass spectrometer system (MB-TOF-MS), as describedin our earlier studies on photodissociation of PCl3.

24 Briefly, itconsists of a supersonic beam source and an ionization (beam -laser interaction) region, which were differentially pumped, usingturbomolecular pumps (TMP), backed by rotary pumps. Anadditional 4 in. TMP was used to pump the TOF tube andmounted near the detector. Typical operating pressures in thesource and the ionization regions were 5 � 10-5 and 8 � 10-7

Torr, respectively. A pulsed supersonic free jet was generated,using a solenoid valve. The free jet was skimmed off by a 1.9 mmdiameter conical skimmer generating a supersonic molecularbeam (MB), which was directed to the interaction region. Thepulsed valve was located ∼2 cm from the skimmer, and 7.5 cmfrom the interaction region. The MB pulse profile and thenumber density were characterized, using a fast ion gauge(FIG), located beyond the ion optics. The fumaryl chloridesample (95% purity, Aldrich) was used, without further purifica-tion. Helium was bubbled through the sample maintained atroom temperature, and the mixture was expanded through thenozzle at a stagnation pressure of 1000 Torr of He. It was ensuredthat any interference to the measurements due to cluster photo-fragmentation was absent by operating at a low stagnation pressure,and using only the rising part of the molecular beam pulse.

The detector system consists of a two-stage Wiley-McLaren25 time-of-flight mass spectrometer (TOF-MS), withextraction and acceleration regions. The system is mountedvertically, perpendicular to the horizontal MB. The extractionregion consists of a repeller electrode and an extractor grid,mounted 10 mm above the repeller electrode. The accelerationregion is defined by the extractor electrode and a grid held at theground potential, separated from each other by 10 mm. Boththese grids (5 cm � 5 cm) are constructed from stainless steelmesh, with 90% transmission. To collect the total ions, theextraction region was held at ∼300 V/cm, and the accelerationregion was held at ∼3900 V/cm. After passing through theacceleration region, the ion packet passed through a 1035 mmlong field-free flight tube to the detector. Two deflector plates,placed perpendicular to the detector axis (z axis), allowed the ionpacket to be translated in the (x, y) plane, to center it on thedetector. The typical field strength for the deflector plates was2-6 V/cm. The ions were detected by a set of 18 mm dualmicrochannel plates (MCP). A single compact voltage generator,having multiple output voltage ports, was employed to power theTOF ion optics, the deflection plates and the MCP detector.

The chlorine atoms were probed, using 2 þ 1 resonanceenhanced multiphoton ionization (REMPI) transitions in theregion 234-236 nm. The laser pulses were generated by a dyelaser (TDL 90, Quantel), using rhodamine 101 dye. The dyelaser was pumped by 532 nm light from the second harmonic of aNd:YAG laser (YG-981-C, Quantel), operating at 20 Hz. Thefundamental dye laser output was frequency-doubled in a KDPcrystal, and mixed with the fundamental output of the Nd:YAGlaser, to obtain an output in the range 230-236 nm. In all the experi-ments reported in this work, the same laser beam was employedas a pump as well as a probe, i.e., for both photodissociation of theparent molecule and ionization of the photoproducts, Cl(2P3/2)and Cl*(2P1/2) atoms. The laser beam was focused by a lens of

Figure 1. Optimized geometries for various isomers of fumaryl chloride.



200 mm focal length, and the distance of the lens from the centerof the molecular beam axis was varied, to obtain the best ratio ofon- and off-resonant signals. The spin-orbit ratio was deter-mined from the ion intensities for the corresponding transitions,after correction due to their two-photon oscillator strengths. Forthis purpose, two lines were chosen, one at 42492.5 cm-1 andanother at 42516.1 cm-1, corresponding to 4p 2D3/2r 3p 2P3/2and 4p 2P1/2 r3p 2P1/2 transitions, respectively. The polariza-tion of the resultant laser beam was rotated, using a doubleFresnel rhomb, and a polarizer ensured 99% polarization of thelaser beam entering the chamber. The laser power was mon-itored, using a power meter, and was typically 50-100 μJ/pulse.To obtain the TOFmass spectrum, the signal was sent to a digitaloscilloscope, which was interfaced to a PC. Subtracting the off-resonant signals from the on-resonant signals effectively removedthe minor pump-oil related background contribution to the TOFspectra, and also the contribution from a multiphotonic process.A digital delay/pulse generator, with pulse resolution of 20 ps,was employed as the master to trigger all the instruments for timesynchronization. The time delay between the trigger pulseapplied to the pulsed valve and its opening was obtained bymeasuring the delay between the trigger pulse and the fastionization gauge (FIG) signal, employing a digital oscilloscope.This delay is the sum of the time required to open the pulsedvalve from its trigger input and that for the molecular pulse toreach FIG from its position of generation, i.e., the nozzle exit. Bymeasuring these time delays for different FIG positions withrespect to the skimmer, we estimated the flow velocity of themolecular beam and used it to obtain the time required for themolecular beam to reach the extraction region of the TOFMS. APC was used to control the scan of the dye laser via an RS232interface, and collect data from SRS 245, through a GPIBinterface, using a control and data acquisition program.

The power dependence measurements were performed byintegrating the chlorine atom REMPI spectra, which wereobtained by sending the detector output directly to a boxcarintegrator, gated on the m/z 35 and m/z 37 peaks in the TOFmass spectrum. The laser power was varied, and the boxcaroutput was recorded at each power. The REMPI spectra andsurvey scans were taken by recording the boxcar output as afunction of the laser wavelength and archiving the spectra on aPC interfaced to the dye laser controller. TOF profiles were takenfor three different experimental configurations, vertical (laserpolarization II detection axis), horizontal (laser polarization ^detection axis), and magic angle (laser polarization at 54.7� todetection axis). Doppler broadening of the transitions was alwayswell within the laser bandwidth. For polarization experiments,TOF profiles were averaged for a minimum of 10000 laser shots.

III. RESULTS AND ANALYSIS

A. Analysis. We have measured TOF profiles of Cl and Cl* atdifferent laser polarizations. The translational energy distributionand the anisotropy parameter for both Cl and Cl* are determinedfrom these TOF profiles, using a commonly used forwardconvolution (FC) technique, as described in our earlierpublication.24 In this method, the knowledge of the instrumentalresponse function is required, and the same is determined bystudying the REMPI of aniline beam. The REMPI spectrum ofaniline recorded at 34029 cm-1 (293.77 nm) matches very wellwith that reported in the literature.26,27 The aniline molecularion signal was measured as a function of laser intensity, at the

resonance wavelength, and found to be quadratic dependent.This shows that, at 293.77 nm, the REMPI is 1þ 1 type becauseof one photon resonant transition, 1A1f

1B2, followed by pumpingto the ionization continuum by absorption of a second photon. Suchmeasurement showed the instrumental response function to be welldescribed by a fixed Gaussian function in the time domain, withfwhm of 27 ns at aniline mass. Under space focusing conditions, thisleads to a convolution function in the velocity domain, which islinearly dependent on the extractor voltage Vex.The forward convolution method that we have employed has

been discussed in our previous publication24 as well aselsewhere,28,29 and will be only briefly outlined. In the REMPI-TOF-MS technique, the component of the photofragment thatspeeds along the TOF-MS axis, which defines the lab frame Zaxis, is measured. This speed component results from theaveraging of the angular distribution, as described,30 over thephotofragment speed distribution, g(v), and is given as28,29,31,32

f ðvz;χÞ ¼Z ¥

jvzjgðvÞ2v

1þ βP2ðcos χÞP2 vzv

� �� dv ð1Þ

where vZ is the velocity component along the Z axis, v is the recoilspeed of the fragments, β is the anisotropy parameter given byβ = 2ÆP2(cos θm)æ. θm is the molecular frame angle between themolecular transition dipole moment and the photofragmentrecoil direction, P2(cos χ) is the second-order Legendre poly-nomial, and cos(χ) = ε 3 z in the above equation is the projectionof the pump laser electric field, ε, on the detector axis, z, which isalso defined as the angle between the dissociation laser polariza-tion and theZ axis. In the present work, we have used a procedureof noncore sampling data, in which it is assumed that the natureand the shape of the TOF profiles for chlorine photofragment isindependent of the probe polarization. In general, this assump-tion holds good. But, the presence of atomic v 3 j correlationmight make this assumption only approximate. However, thesecorrelations are weak enough to be neglected.33,34 In the case ofseveral contributing channels, eq 1 must be summed over thephotofragment speed distribution, gi(v), and anisotropy, βi, ofeach channel i. The dependence of f(vz,χ) on β can be eliminatedby either measuring the data with the magic angle of χ = 54.7�,where P2(cos χ) is zero, or coadding normalized profiles with χ =90� and χ = 0�, giving a 2:1 weightage. Both the approaches yieldisotropic f(vz,χ) profiles, and the total center of mass (c.m.)speed distribution is obtained by differentiation of eq 1 at χ =54.7� and is given as

gðvÞ ¼ - 2vddvz

f ðv;54:7�Þjvz ¼ v ð2Þ

Themeasurements in this work were recorded as TOF spectra,i.e., in the time domain I(t,χ). Under space focusing conditions, asimple linear transformation gives the signal in the velocity domain,I(vZ,χ), according to, vz = (qVex(t-t0))/m, where q and m are thecharge andmass of the photofragment, respectively,Vex is the electricfield in the extraction region, and t0 is the mean time-of-flight.

35

Application of eq 1 on themagic angle TOF profile I(vZ,54.7�)yields an estimate of the total c.m. speed distribution g(v) and, inturn, some indication of the form of the individual speed distribu-tions gi(v). These are usually modeled with the functional form

36,37

giðvÞ ¼ ðf TÞaii ð1- ðf TÞiÞbi ð3Þ

where (fT)i is the fraction of the available energy channeled intotranslational modes, Ei

trans/Eiavail, and ai and bi are adjustable



parameters. By taking into account an adjustable anisotropy para-meter βi and weight for each decay channel, we simultaneouslycalculate f(vZ,χ) for the geometries χ = 0�, 54.7�, and 90�.Convolutionwith the instrument response function yields simulatedTOF profiles, which can be compared with the experimental results.The parameters are then adjusted until a satisfactory agreementwiththe experimental data is achieved. Once the photofragment speeddistributions have been determined, thesemay be used to obtain thecorresponding translational energy distributions.Figure 2 shows the power dependence of the one-color

REMPI signals for transitions corresponding to Cl (2P3/2).The signal is linear in a log-log plot over the range of powersused in the present study. For both the types of chlorine atoms,the lines exhibit a slope of ∼3.1 ( 0.2, which is consistent withone-photon dissociation of fumaryl chloride, followed by 2 þ 1REMPI of the chlorine atoms, assuming that the ionization step issaturated. Apart from the power dependence studies, we alsosystematically monitored the shape and the width of TOFprofiles of Cl atoms at various laser intensities. All the experi-ments were performed in the intensities, which are much lowerthan the intensity at which the shape and width of the TOFprofiles were invariant. This experimental condition ensures that

the translational energy distributions and the anisotropy param-eters are invariant over the laser fluences used.B. Spin-Orbit Branching Ratio. Figure 3 shows typical

Doppler profiles of Cl(2P3/2) and Cl*(2P1/2) atoms producedin the dissociation of fumaryl chloride on excitation at 235 nm,which are wavelength scans in the region 235.336 and 235.205nm, respectively. The relative quantum yield of chlorine atomfragments in different spin-orbit states was determined bynormalizing the integrated intensity, i.e., peak areas, S(Cl) orS(Cl*), of the respective 2þ1 REMPI transitions with respect tothe laser intensity, and the ratio of the two-photon absorptioncoefficients. The measured area of 2þ1 REMPI lines is propor-tional to the actual product ratio, with a factor of k, which is therelative ionization probability for Cl and Cl*, f(Cl)/f(Cl*),

NðCl�ÞNðClÞ ¼ k

SðCl�ÞSðClÞ ð4Þ

whereN(Cl) andN(Cl*) designate the number density of Cl andCl* produced. S(Cl) and S(Cl*) are obtained by integrating themeasured ion signal intensity over the proper range containingthe Doppler width and the probe laser bandwidth. The measure-ments were repeated at different laser light intensities, givingsimilar relative signal intensities. For fumaryl chloride photo-dissociation, the ion integrated signal intensity ratio, S(Cl*)/S(Cl), has been measured to be 0.39( 0.06. From the measuredintegrated intensity ratio, one can easily obtain the product ratio,using the above eq 4, by taking a value of 0.85( 0.10 for k.38 Therelative quantum yields, Φ(Cl) and Φ(Cl*), can be determinedfrom the product ratio, and Φ(Cl*) can be expressed as

ΦðCl�Þ ¼ NðCl�ÞNðCl�Þ þNðClÞ ð5Þ

The value ofΦ(Cl*) calculated for fumaryl chloride dissociationis found to be 0.24 ( 0.03. On statistical grounds one expects aquantum yield,Φ(Cl*) = 0.33. As described in the later section,the low quantum yield for the Cl* can be attributed to the factthat the C-Cl bond fission via the lowest-recoil-kinetic energymechanism mainly produces Cl atoms in the Cl(2P3/2) state; thesame is true for the lowest-kinetic-energy C-Cl bond fissionchannel in photodissociation of many other alkyl chlorides.C. Translational Energy Distribution and Anisotropy Param-

eter. The measured Cl atom TOF profiles were analyzed todetermine the kinetic energy distribution and average kineticenergy of the fragments in the fumaryl chloride photodissocia-tion at ∼235 nm. The TOF profiles for the Cl and Cl* atomswere converted to the velocity domain. Figures 4 and 5 showbackground subtracted TOF spectra, recorded for the laserpolarization parallel, at the magic angle ∼54.78� and perpendi-cular to the detection axis, for the Cl and Cl* fragments,respectively. We analyzed the TOF data, using a forward con-volution procedure, as described in the earlier section. Here, aninitial photofragment speed distribution, g(v), and the anisotropyparameter β are assumed. As atomic v 3 j correlations are knownto be very weak, we can safely assume the TOF profiles to beindependent of the probe polarization. The TOF profiles arecalculated for the three experimental configurations, convolutedwith the instrumental response function determined, as de-scribed in an earlier section, and are compared with the experi-mental results. The translational energy distributions, P(ET),determined from the data in Figures 4 and 5, for the Cl and Cl*,are depicted in Figures 6 and 7. Inspection of Figures 6 and 7reveals that the P(ET) consists of two components. For the Cl

Figure 2. Dependence of the observed Cl(2P3/2) atom REMPI signalfrom fumaryl chloride photolysis on the laser intensity. The slope of thefitted linear log-log plot is 3.1 ( 0.2.

Figure 3. Profiles of Cl and Cl* atoms produced in the 235 nm laserphotolysis of fumaryl chloride used for the determination of their relativequantum yields.



atom, the faster component, centered at ∼14.4 kcal/mol, con-sists of 67 ( 5% of the total fragments while the slowercomponent, centered at∼2.7 kcal/mol, consists of the remaining33 ( 5%. Similarly, for Cl*, the faster component, centered at∼16.8 kcal/mol, consists of 87( 5% of the total fragments whilethe slower component, centered at∼2.4 kcal/mol, consists of theremaining 13 ( 5%. It was not possible to determine indepen-dently the anisotropy parameter for the two components, andhence, the anisotropies were assumed to be identical for eachchannel. The TOF profiles for both Cl and Cl* are independentof laser polarizations, implying the β is well characterized by avalue of ∼0.0, within the experimental uncertainties. The calcu-lated TOF profiles are displayed by the solid line, with eachcomponent shown by a dotted curve. The difference in the TOFprofile for Cl and Cl* in Figure 4 and 5 arises mainly due to thedifferent nature of fT, their nonstatistical branching ratio and theratio of fast to slow component.D. HCl Formation Channel. The multiphoton ionization

spectrum of fumaryl chloride at 235 nm clearly shows peaks atm/e 36 and 38, corresponding to H35Cl and H37Cl, respectively.To explore the origin of the HCl mass peak, the laser wavelengthwas scanned over its REMPI region in the range 236-237nm.39-41 In our experiments, we did find 2þ1 REMPI lines ofHCl in the TOF spectra in the range of 236-237 nm, for theQ(J) branch of the V 1Σþ (0þ) rr X 1Σþ (0þ) (12,0) bandsystem.39 Although the signal was weak due to the predissociativenature of this particular transition,39 we were able to carry out thepower dependence studies. The predissociative nature of thistransition was also established in our experiment, since weobserved the corresponding Cl and H atom ion signals at the

resonant wavelength corresponding to the REMPI lines of HClmolecules. A linear log-log plot for the power dependencestudies shows a slope of∼2.9( 0.2, which is consistent with one-photon dissociation of fumaryl chloride formingHCl followed by

Figure 4. REMPI-TOF profiles of Cl(2P3/2) produced from the 235 nmphotodissociation of fumaryl chloride. The circles are the experimentaldata and the solid line is a forward convolution fit. Three panels, namely,upper, middle, and lower panels, correspond to horizontal, magic angle,and vertical experimental geometries, respectively.

Figure 5. REMPI-TOFprofiles of Cl(2P1/2) produced from the 235 nmphotodissociation of fumaryl chloride. The circles are the experimentaldata and the solid line is a forward convolution fit. Three panels, namely,upper, middle, and lower panels, correspond to horizontal, magic angle,and vertical experimental geometries, respectively.

Figure 6. Center-of-mass recoil translational energy distribution de-rived from Figure 4 for Cl(2P3/2) in the photodissociation of fumarylchloride at 235 nm. The dashed lines indicate the speed distributions forthe fast and slow components of the chlorine atom formation channel;the solid line shows the sum. The vertical arrow indicates the maximumavailable energy for Cl(2P3/2) elimination channel.



its detection with the 2þ1 REMPI scheme, similar to thedynamics of Cl atom formation. Laser polarization dependentstudies on the HCl TOF profile to elucidate the translationalenergy distribution could not be done due to weak signal, asdiscussed earlier. The above set of experiments establishes theformation of HCl in the photodissociation of fumaryl chloride at235 nm as a primary channel.

IV. DISCUSSION

A. Nature of Excitation at 235 nm. As discussed earlier,fumaryl chloride contains one-CdC- and two-CdO groupsin conjugation, which make it bichromophoric, similar to fumaricacid (HOCO—CHdCH—COOH) and acryloyl chloride(H2CdCHCOCl). Hence, we expect that fumaryl chloride alsohas a similar type of absorption spectrum with two bandscorresponding to n-π* and π-π* transitions. Ab initio molec-ular orbital (MO) calculations were performed to investigate thenature of the excited electronic states of fumaryl chloride. Weoptimized ground state geometries of various isomers, as shownin Figure 1, employing time dependent (TD) density functiontheory (DFT) calculation, using the cc-pVDZ set of basis sets.There are three isomers, namely, cis-cis, cis-trans, and trans-trans, for fumaryl chloride. Theoretical calculations predict thatthe relative stabilities of all three isomers differ by less than 0.5kcal/mol. Thus, these isomers are indistinguishable, given theexperimental uncertainty of present study and the accuracy of thetheoretical methods employed here. Vertical excitation energieswere obtained for various transitions, to understand the nature ofexcitation at 235 nm. Although the calculated vertical transitionenergies are red-shifted compared to the experimental results incase of fumaric acid,42 the nature of transition and the orbitals areaccurately predicted using this method. Orbitals participating inthe different electronic transitions were visualized for betterunderstanding of the process. The vertical excitation of fumarylchloride is associated with a strong π-π* transition at higherenergy and a weaker n-π* at lower energy. The strong

transition, with oscillator strengths exceeding 0.4, is the firstπ-π* transition and is termed as S2. Topological analyses of theelectron density notably highlight the delocalized nature of theOdC—CdC—CdO chain, whereby the central CdC bondpossesses the largest electron density, although less than the pureπ-bonds. The C—C bonds consequently possess larger electronpopulations than the expected σ-bonds. The nature of π and π*orbitals in this transition is having a mixed character involvingmainly the CdC and the CdO π electrons. The weaker n-π*transitions at lower energy is termed as S1 and it mainly involvesthe transition from the nonbonding orbitals of both Cl and O tothe π* orbitals, which have a mixed character as discussed earlier.At 235 nm, we believe that fumaryl chloride is excited to the S2state. This S2 state adiabatically correlates only with highlyexcited photoproducts, and that is not feasible in a single-photonexcitation in the present case. Therefore, it is assumed thatfumaryl chloride from the S2 states crosses over to the nearbystates, mostly the n-σ* state, fromwhere the C-Cl bond cleavesforming Cl atoms. Also, it can undergo rapid internal conversionto the ground state, from where various other dissociationchannels can occur, in addition to the Cl atom elimination.B. Translational Energy Release and Anisotropy Param-

eter. Several processes have been proposed as primary dissocia-tion pathways in fumaryl chloride photoexcitation at 235 nm, andthese are given as follows:

ClOC—CHdCH—COCl f ClOC—CHdCH—COþCl

ΔH ¼ 83 kcal=mol ð6Þf ClOC—CHdCHþ COCl ΔH ¼ 100 kcal=mol

ð7Þf ClOC—CHdCdCOþ HCl ΔH ¼ 38 kcal=mol

ð8ÞReactions 6 and 7 involve the cleavage of single bonds C-Cl andC-C, respectively, while reaction 8 is the HCl molecularelimination channel. The Cl atom can be produced directly fromreaction 6 as a primary product and also from the subsequentdissociation of highly energetic COCl radical (reaction 9)formed in reaction 7 as a secondary product. The ΔH valuesare taken as similar to that of acryloyl chloride from refs 42 and43.

COCl f COþ Cl ΔH ¼ 8 kcal=mol ð9ÞIn fumaryl chloride, two types of P(ET) were observed for the C-Cl fission: one producing fragments with high kinetic recoilenergies and the other producing fragments with low recoilenergies. The P(ET) for Cl(2P3/2) derived from the forwardconvolution fit to the high-translational energy C-Cl fissionchannel peaks near 14.4 kcal/mol and extends to 30.0 kcal/mol.Similarly, the low translational energy C-Cl bond fission channelpeaks at 2.7 kcal/mol and extends to 10 kcal/mol. The presenceof two types of P(ET) in C-Cl bond fission is analogous to othersystems involving a π-π* transition, such as acryloylchloride,9,43,44 allyl chloride,45,46 and 2-chloropropene.47 The hightranslational energy chlorine atom channel, most likely, arises froman electronic predissociation via a state repulsive in the C-Clbond, as seen in the above systems. It is possible that the lowtranslational energy channel results from the C-Cl fission follow-ing internal conversion to the ground electronic state. Anotherpossibility of low energy Cl atom formation may arise fromreaction 9. However, the formation of the COCl radical cannot

Figure 7. Center-of-mass recoil translational energy distribution de-rived from figure 5 for Cl(2P1/2) in the photodissociation of fumarylchloride at 235 nm. The dashed lines indicate the speed distributions forthe fast and slow components of the chlorine atom formation channel;the solid line shows the sum. The vertical arrow indicates the maximumavailable energy for Cl(2P1/2) elimination channel.



compete with the Cl formation channel due to its high endother-micity on the ground potential energy surface. So the onlypossible route for the formation of COCl, and hence, formationof Cl from it, is from a higher excited state such as, S1, T2, and T1.But this route for the COCl formation is also ruled out in acryloylchloride, a molecule with photochemical behavior similar to thatof fumaryl chloride, on the basis of theoretical studies by Cuiet al.11 The nature of P(ET) of the slow Cl atom obtained in thepresent studies for fumaryl chloride also rules out this route onthe basis of energy consideration. So, the only formation pathwayfor the slowCl atom is attributed to the dissociation process fromthe ground electronic state, after internal conversion. Thisproposition is further supported by the fact that we did observethe molecular HCl elimination channel, which is conclusivelyfrom the ground state of fumaryl chloride.The partitioning of the available energy into various degrees of

freedom of the fragments is mainly governed by the nature of thedissociative potential energy surface. Very often, the dispositionof the available energy into the fragment translational and theinternal degrees of freedom can be predicted, using simplemodels. These models generally fall into two categories: im-pulsive and statistical. It is well-known that the energy partition-ing for a dissociative event on a repulsive surface is well describedby an impulsive model. So, an impulsive model24,48,49 has beenused in this case, to calculate theoretically the translationalenergy released to the products. In this model, the distributionof energy among the product states is governed by the dissocia-tive event, i.e., by the repulsive force acting during the breaking ofthe parent molecule into the products. For example, in thepresent case, by using only conservation of momentum andenergy, and the impulse assumption, one finds that the fraction ofthe available energy (Eavail) released as translational energy isgiven by

ET ¼ μC-Cl

μClCOCHCHCO-Cl

!Eavail; and f T ¼ ET=Eavail ð10Þ

where μC-Cl is the reduced mass of the C and Cl atoms,μClCOCHCHCO-Cl is the reduced mass of the ClCOCHCHCOand Cl, Eavail is the available energy, and fT is the fraction of theavailable energy going into the translational modes of thefragments. In the case of fumaryl chloride, the ratio of reducedmasses is 0.32. The available energy is given by

Eavail ¼ Ehν - D00ðClCOCHCHCO-ClÞ- ESO ð11Þ

where Ehν is the photon energy (122.0 kcal/mol), D00

(ClCOCHCHCO-Cl) is the C-Cl bond dissociation energy,and ESO is the spin-orbit energy of chlorine (2.4 kcal/mol). Thebond dissociation energy is taken as 83.0 kcal/mol, similar to thatfor acryloyl chloride.44 Thus, Eavail for the Cl andCl* channels are39.0 and 36.6 kcal/mol, respectively. The experimental averagetranslational energy (ET) for the fast component is found to be14.9 ( 1.6 and 16.8 ( 1.6 kcal/mol for Cl and Cl* channels,respectively, giving the fT values of 0.38 and 0.46 for the Cl andCl*, respectively. The experimental determined fT so obtained isslightly higher than the value of 0.32 predicted, using theimpulsive model.Now, coming to the slow component in the translational

energy distribution of the Cl atom, the average energies deter-mined are found to be 3.4 ( 0.8 and 3.1 ( 0.8 kcal/mol, for Cland Cl*, respectively. These values give the same fT of 0.08,for both Cl and Cl*, even though the available energies differ by

2.4 kcal/mol (spin-orbit energy) for these channels. A low fTvalue suggests that the kinetic energy released for the slowcomponent can be explained better in light of a statistical model.The statistical model assumes partitioning of the available energydemocratically throughout the molecule and therefore neglectsthe effects of specific dynamical interactions of the departingfragments. A statistical dissociation process is predominant for along-lived photoexcited parent molecule, allowing partitioning ofthe excess energy statistically among the available degrees offreedom of the products. This may be applicable in a processinvolving a rapid internal conversion to the ground electronicstate, followed by the subsequent slow dissociation. Under thesecircumstances, in a large molecule with many low frequencymodes, a relatively small amount of the excess energy is parti-tioned into translational motion of the products. For this kind ofdissociation process, a priori calculations50,51 were adopted,which give the fT value of 0.07, matching very well with theexperimental value of 0.08. This implies that the slow componentin the translational energy distribution is mainly due to the Clatom, which arises from the ground state potential energysurface, after internal conversion via some curve crossing me-chanism. This mechanism is further confirmed by observation ofthe molecular HCl elimination channel, as described earlier.The higher experimental fT value as compared to that calcu-

lated using the impulsive model obtained in the dissociation offumaryl chloride for the fast chlorine atom channel prompted usto apply the hybrid model, employed by North et al.52 forreactions with barrier. In this model, the Eavail for the productsis divided into two parts, namely, the excess energy above the exitbarrier (Estat) and at the exit barrier energy (Eimp). The partition-ing of Estat and Eimp is treated by the statistical and modifiedimpulsive models, respectively. The energy partitioned into eachfragment is then obtained by adding contributions from each ofthese two models. For this type of dissociation with barrier, ETdoes not changemuch with Eavail. Similar results were obtained inour earlier studies on various saturated and unsaturated car-boxylic acids, and also on the enone system.2,3 Photodissociationstudies on acryloyl chloride by Butler and co-worker9,44 at 193and 235 nm show similar translational energy release for the Clatom. At 193 nmdissociation, ETwas obtained to be 27 kcal/mol.Similarly, at 235 nm dissociation, a value of 23 kcal/mol wasobtained. Here, it can be clearly seen that the change in the Eavailfrom 65 kcal/mol (for 193 nm) to 39 kcal/mol (for 235 nm) didnot change the ET value considerably for the Cl atom dissociationchannel in acryloyl chloride. Also, the theoretical calculation byCui et al.11 clearly shows the presence of an exit barrier in theS1(

1n-π*) state, for the Cl atom dissociation channel. So, westrongly believe that the dissociation from the S1 state with anexit barrier cannot be ruled out in the photodissociation ofacryloyl chloride in addition to dissociation from the repulsivestate with (n-σ*) nature. With same analogy, we propose that infumaryl chloride the initially prepared S2(

1π-π*) state crossesover to various states, namely, S1(

1n-π*), (n-σ*), and theground state. Finally, the Cl atom can be formed from all thesestates with different values of ET.In the analysis of the recoil anisotropy, β, for the Cl fragment,

we note that in the limit of an instantaneous dissociation process,β = 2ÆP2(cos θm)æ, where θm is the molecular frame anglebetween the molecular transition dipole moment and the photo-fragment recoil direction, and P2(cos χ) is the second-orderLegendre polynomial. In a photodissociation process, followingexcitation via the electric transition dipole moments, namely,



parallel or perpendicular, would be expected to give a β value ofþ2 or-1, respectively. The anisotropy measured in the presentwork is ∼0.0, which indicates an isotropic dissociation process.This value of β does not give any indication of the nature of thetransition dipole moment in fumaryl chloride dissociation. Gen-erally, an impulsive dissociation is accompanied with an aniso-tropic distribution of the photoproducts, since the dissociationlifetime is much shorter than the rotation period of the molecule.However, the anisotropy in an impulsive dissociation can bereduced,53,33 or wiped out,54 due to several factors, such as mixedinitial transition with both parallel and perpendicular compo-nents, longer dissociation lifetime, dissociation not from a singlegeometry rather from a range of geometries etc. In fumarylchloride, for C2h geometry, the π-π* transition, which has Busymmetry, the transition dipole moment is mainly along the xaxis and y axis. The same is true for the n-π* transition. So, inthis case even if there is a crossover from the initially preparedπ-π* state to the n-π* state, the transition dipole moment isnot changed. However, during this crossover period themoleculemay rotate. The absence of recoil anisotropy in the presentstudies is expected mainly due to the mixed transition, asdiscussed, and the relatively longer dissociation lifetime, sincethe dissociation is not from the initially prepared state ratherfrom the crossover state. In addition, the different geometries offumaryl chloride also contribute to the β value of ∼0.0.

V. CONCLUSION

In summary, fumaryl chloride generates a chlorine atom uponexcitation at 235 nm, which prepares the molecule initially in itsπ-π* state. The nascent distribution of the photofragmentchlorine atom is measured by the 2 þ 1 REMPI with the TOFmass spectrometry technique. We have determined the photo-fragment speed distribution, the anisotropy parameter β, and thespin-orbit branching ratio for the chlorine atom eliminationchannels, to gain insight into the dynamics of the chlorine atomformation. Polarization-dependent and state-specific TOF profilesare deconvoluted to get translational energy distributions, using aforward convolution method with least-squares fitting, taking intoaccount the fragment anisotropies. The TOF profiles for Cl andCl* are independent of the laser polarization; i.e., the parameter βis well characterized by a value of 0.0, within the experimentaluncertainties. Two contributions, namely, the fast and the slowcomponents, are observed in the translational energy distribution,P(ET), of Cl and Cl* atoms formation channels. The averagetranslational energies for the Cl and Cl* channels for the fastcomponent are 14.9( 1.6 and 16.8( 1.6 kcal/mol, respectively.Similarly, for the slow component, the average translationalenergies for the Cl and Cl* channels are 3.4 ( 0.8 and 3.1 ( 0.8kcal/mol, respectively. The energy partitioning into the transla-tional mode is interpreted with the help of an impulsive model forthe fast component, and a statistical model for the slow compo-nent. While the experimental fT value for a slow chlorine atom isnicely described with the statistical model, that for the fast Cl atomchannel is slightly higher than the predicted value of 0.32, by theimpulsive model. Apart from the chlorine atom eliminationchannel, the molecular HCl elimination is also observed in thephotodissociation of fumaryl chloride. The observation of theHClmolecular elimination channel in the dissociation process and thebimodal translational energy distribution of the chlorine atomclearly indicate the existence of a crossover mechanism from theinitially prepared state to the ground state. Finally, it is proposed

that in fumaryl chloride the initially prepared S2(1π-π*) state

crosses over to various states, namely, S1(1n-π*), (n-σ*), and

the ground state, and the Cl atom can be formed from all thesestates, with different values of ET.

’AUTHOR INFORMATION

Corresponding Author*Electronic mail: [email protected].

’ACKNOWLEDGMENT

We thank Drs. S.K. Sarkar and T.Mukherjee for their constantguidance and keen interest throughout this work. We alsoacknowledge help rendered by Mr. Yogesh Indulkar during theinitial experiments.

’REFERENCES

(1) Schippers, P. H.; Dekkers, H. P. J. Am. Chem. Soc. 1983, 105, 145.(2) Upadhyaya, H. P.; Kumar, A.; Naik, P. D. J. Chem. Phys. 2003,

118, 2590.(3) Upadhyaya, H. P.; Kumar, A.; Naik, P. D.; Avinash, V. Sapre;

Mittal, J. P. J. Chem. Phys. 2002, 117, 10097.(4) Kumar, A.; Upadhyaya, H. P.; Naik, P. D.; Maity, D. K.; Mittal,

J. P. J. Phys. Chem. A 2002, 106, 11847.(5) Dhanya, S.; Maity, D. K.; Upadhyaya, H. P.; Kumar, A.; Naik,

P. D.; Saini, R. D. J. Chem. Phys. 2003, 118, 10093.(6) Tang, X.; Ratliff, B. J.; Fitzpatrick, B. L.; Butler, L. J. J. Phys. Chem.

B 2008, 112, 16050.(7) Person, M. D.; Kash, P. W.; Butler, L. J. J. Chem. Phys. 1992, 97,

355.(8) Wei, Z.-R.; Zhang, X.-P.; Lee, W.-B.; Zhang, B.; Lin, K.-C. J.

Chem. Phys. 2009, 130, 014307.(9) Szpunar, D. E.; Miller, J. L.; Butler, L. J.; Qi, F. J. Chem. Phys.

2004, 120, 4223.(10) Wu, W.; Liu, K.; Yang, C.; Zhao, H.; Wang, H.; Yu, Y.; Su, H. J.

Phys. Chem. A 2009, 113, 13892.(11) Cui, G.-L.; Li, Q.-S.; Zhang, F.; Fang, W.-H.; Yu, J.-G. J. Phys.

Chem. A 2006, 110, 11839.(12) Ahmed, M.; Blunt, D.; Chen, D.; Suits, A. G. J. Chem. Phys.

1997, 106, 7617.(13) Hemmi, N.; Suits, A. G. J. Phys. Chem. A 1997, 101, 6633.(14) Wu, C.-Y.; Lee, Y.-P.; Wang, N. S. J. Chem. Phys. 2004, 120,

6957.(15) Katon, J. E.; W. R. Feairheller, J. J. Chem. Phys. 1967, 47, 1248.(16) Koster, D. F.; Vasiloff, T. P. Spectrochim. Acta 1971, 27, 1633.(17) Pietri, N.; Jurca, B.; Monnier, M.; Hillebrand, M.; Aycard, J. P.

Spectrochim. Acta Part A 1999, 56, 157.(18) Hagen, K. J. Mol. Struct. 1985, 128, 139.(19) Landry, J. M.; Katon, J. E. Spectrochim. Acta 1984, 40A, 871.(20) Bizzocchi, L.; Esposti, C. D.; Dore, L. Astron. Astrophys. 2008,

492, 875.(21) Jabbari, E.; Gruetzmacher, J. A.; Lu, L.; Currier, B. L.; Yaszemski,

M. J. Development of a Novel Self-Crosslinkable poly(caprolactonefumarate) as a Biodegradable and Injectable Scaffold for Bone TissueEngineering. Proceedings of the 25th Annual International Conference of theIEEE EMBS, Cancun, Mexico, 2003.

(22) Mahmoudi, M.; Simchi, A.; Imani, M.; Hafeli, U. O. J. Phys.Chem. C 2009, 113, 8124.

(23) Maaref, S.; Roz, Z.; Sun, S.-S.; Seo, K.; Winston, K.; Bonner,C. E. J. Appl. Polym. Sci. 2004, 92, 317.

(24) Upadhyaya, H. P.; Saha, A.; Kumar, A.; Bandyopadhyay, T.;Naik, P. D.; Bajaj, P. N. J. Phys. Chem. A 2010, 114, 5271.

(25) Wiley, W. C.; McLaren, I. H. Rev. Sci. Instrum. 1955, 26, 1150.(26) Lubman, D. M.; Jordan, R. M. Rev. Sci. Instrum. 1985, 56, 373.(27) Brophy, J. H.; Rettner, C. T. Chem. Phys. Lett. 1979, 67, 351.



(28) McGivern, W. S.; Li, R.; Zou, P.; North, S. W. J. Chem. Phys.1999, 111, 5771.(29) Bergmann, K.; Carter, R. T.; Hall, G. E.; Huber, J. R. J. Chem.

Phys. 1998, 109, 474.(30) Zare, R. N. Angular Momentum; Wiley-Interscience: New York,

1988.(31) Vasudev, R.; Zare, R. N.; Dixon, R. N. J. Chem. Phys. 1984, 80,

4863.(32) Dubs, M.; Bruhlmann, U.; Huber, J. R. J. Chem. Phys. 1986, 84,

3106.(33) Wilson, R. J.; Mueller, J. A.; Houston, P. L. J. Phys. Chem. A

1997, 101, 7593.(34) Bracker, A. S.; Wouters, E. R.; Suits, A. G.; Lee, Y. T.;

Vasyutinskii, O. S. Phys. Rev. Lett. 1998, 80, 1626.(35) Syage, J. A. J. Chem. Phys. 1996, 105, 1007.(36) North, S. W.; Marr, A. J.; Furlan, A.; Hall, G. E. J. Phys. Chem. A

1997, 101, 9224.(37) Bracker, A. S.; North, S.W.; Suits, A. G.; Lee, Y. T. J. Chem. Phys.

1998, 109, 7238.(38) Liyanage, R.; Yang, Y. A.; Hashimoto, S.; Gordon, R.; Field,

R. W. J. Chem. Phys. 1995, 103, 6811.(39) Green, D. S.; Bickel, G. A.; Wallace, S. C. J. Mol. Spectrosc. 1991,

150, 388.(40) Arepalli, S.; Presser, N.; Robie, D.; Gordon, R. J. Chem. Phys.

Lett. 1985, 118, 88.(41) Kvaran, A.; Logadottir, A.; Wang, H. J. Chem. Phys. 1998, 109,

5856.(42) Boily, J.-F.; Seward, T. M. J. Solution Chem. 2005, 34, 1167.(43) Szpunar, D. E.; Miller, J. L.; Butler, L. J.; Qi, F. J. Chem. Phys.

2004, 120, 4223.(44) Lau, K.-C.; Liu, Y.; Butler, L. J. J. Chem. Phys. 2005, 123,

054322.(45) Morton, M. L.; Butler, L. J.; Stephenson, T. A.; Qi, F. J. Chem.

Phys. 2002, 116, 2763.(46) Liu, Y.; Butler, L. J. J. Chem. Phys. 2004, 121, 11016.(47) Mueller, J. A.; Parsons, B. F.; Butler, L. J.; Qi, F.; Sorkhabi, O.;

Suits, A. G. J. Chem. Phys. 2001, 114, 4505.(48) Tuck, A. F. J. Chem. Soc., Faraday T rans. 2 1977, 73, 689.(49) Riley, S. J.; Wilson, K. R. J. Chem. Soc., Faraday Discuss 1972, 53,

132.(50) Levine, R. D.; Kinsey, J. L. InAtom-Molecule Collision Theorys: A

Guide for the Experimentalist; Bernstein, R. B., Ed.; Plenum Press: NewYork, 1979.(51) Muckermann, J. T. J. Phys. Chem. 1989, 93, 179.(52) North, S. W.; Blank, D. A.; Gezelter, J. D.; Longfellow, C. A.;

Lee, Y. T. J. Chem. Phys. 1995, 102, 4447.(53) Carter, R. T.; Hallou, A.; Huber, J. R. Chem. Phys. Lett. 1999,

310, 166.(54) Hintsa, E. J.; Zhao, X.; Lee, Y. T. J. Chem. Phys. 1990, 92, 2280.

dynamics of cl ( p) atom formation in the ... · and 4p 2p 1/2 r3p 2p 1/2 transitions,...

Documents