Coursehero >> New York >> Cornell >> MSC 2

Course Hero has millions of student submitted documents similar to the one below including study guides, homework solutions, papers, exam answer keys and textbook solutions.

JOURNAL THE OF CHEMICAL PHYSICS 122, 174304 2005 Photodissociation channels for N2O near 130 nm studied by product imaging H. M. Lambert, E. W. Davis, O. Tokel, A. A. Dixit, and P. L. Houstona Department of Chemistry and Chemical Biology, Cornell University, Ithaca, New York 14853-1301 Received 14 December 2004; accepted 18 February 2005; published online 2 May 2005 The photodissociation of N2O at wavelengths near 130 nm has been investigated by velocity-mapped product imaging. In all, ve dissociation channels have been detected, leading to the following products: O 1S + N2 X 1 , N 2D + NO X 2 , N 2P + NO X 2 , O 3P + N2 A 3 +u , and O 3P + N2 B 3 g . The most signi cant channel is to the products O 1S + N2 X1 , with strong vibrational excitation in the N2. The O 3P + N2 A , B : N 2D , 2P + NO branching ratio is measured to be 1.4 0.5, while the N2 A + O 3PJ : N2 B + O 3PJ branching ratio is determined to be 0.84 0.09. The spin-orbit distributions for the O 3PJ , N 2PJ , and N 2DJ products were also determined. The angular distributions of the products are in qualitative agreement with excitation to the N2O D 1 + state, with participation as well by the 3 v state. 2005 American Institute of Physics. DOI: 10.1063/1.1888579 INTRODUCTION N 2O + h O 1S + N 2 X 1 N 2D + NO X 2 N 2P + NO X 2 O 3P + N2 A 3 O 3P + N2 B 3 + g 4 5 6 Nitrous oxide is an important component of the earth s natural atmosphere, produced primarily by biological processes in soils and oceans. Mostly inert in the troposphere, it is transported to the stratosphere where it is destroyed both by photodissociation and by reaction. Its mixing ratio falls from about 300 parts per 109 by volume ppbv at the tropopause to about 20 ppbv at 40 km, with a sharp decline beginning at about 25 km. This decline is partially due to the photodissociation N 2O + h N 2 X 1 + O 1D . 1 + u 7 . 8 g This process occurs at wavelengths below about 220 nm, but is most effective in the atmospheric window at about 200 nm. N2O is also destroyed in the stratosphere by reaction with O 1D produced either in 1 or from the dissociation of ozone N 2O + O 1D N 2 + O 2 NO + NO. 2 3 NO produced in 3 is the primary catalytic agent destroying the stratospheric ozone in the natural atmosphere, as noted by Crutzen.1 At shorter wavelengths, N2O can be excited to the C 1 state near 145 nm or to the D 1 + state near 130 nm. The absorption coef cient for the latter transition is high,2 4 about 80 Mb 1 Mb= 10 18 cm2 . Although the dissociation process 1 near 200 nm has been extensively studied,5 13 that at shorter wavelengths is both more complex and less well understood. At 130 nm, the following channels are thought to be important: a Electronic mail: plh2@cornell.edu Black et al.14 investigated the O 1S , N 2D , and N2 A 3 +u products in the photodissociation of N2O from 110 to 150 nm. While the O 1S was detected directly from its emission at 557.7 nm, the N 2D was detected indirectly from its reaction with N2O to give NO B 2 , which decays radiatively. Similarly, N2 A 3 +u was detected indirectly from its energy transfer to NO, yielding NO A 2 + , which decays radiatively. They report that the O 1S quantum yield is near unity in the 128 138-nm range, that the N 2D quantum yield is nearly zero at 130 nm, and that the quantum yield for N2 A 3 +u is less than 10% at 130 nm. No information is available from these experiments on the dynamics of the dissociation, i.e., on what the kinetic energy release is for any of the channels studied. Kinetic-energy release data is available, although at low resolution, from the experiments of Gilpin and Welge GW 15 and of Stone, Lawrence, and Fairchild SLF .16 In both studies, a molecular beam of N2O was dissociated with a ash lamp operating near 130 nm, and the arrival time of metastable fragments at an ionizer was recorded. In the SLF study, data for the 130-nm photodissociation indicate that the product extends vibrational distribution for the N2 X 1 from = 0 to = 6 with a peak at = 3.16 None of the other channels was investigated. In the GW study, processes 4 and 7 were observed, but the limited resolution precluded detailed energy analysis. 2005 American Institute of Physics 0021-9606/2005/122 17 /174304/7/$22.50 122, 174304-1 Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp 174304-2 Lambert et al. J. Chem. Phys. 122, 174304 2005 A deeper understanding of the dissociation process is provided by the theoretical work of Hopper, who performed extensive multicon guration self-consistent- eld/con guration interaction MCSCF/CI calculations identifying some 30 electronic states of nitrous oxide.17 In particular, the correlation diagrams of Figs. 9 and 10 in Ref. 17 show that the N2O D 1 + state reached at 130 nm correlates with dissociation channels 4 and 6 , while excitation to a nearby 3 v state correlates to channels 7 and 5 . Channel 8 cor relates with a bent geometry of the 3 v state. The D 1 + state is of A symmetry when the molecule is bent away from a linear geometry, while the 3 v state separates into two con gurations, one of A symmetry and of A symmetry. In the results described below, we have investigated channels 4 8 by using product imaging to detect the angular distributions for the ve channels as well as the kinetic-energy release for each of the product atoms or diatoms. The atomic products were probed by resonant 1 + 1 ionization using vacuum-ultraviolet excitation for the rst step, while the diatomic products were probed by nonresonant ionization. Our work is part of a larger effort at Cornell to characterize this photodissociation process. Witinzki, Ortiz-Su rez, and Davis have used oxygen Rydberg time-of ight spectroscopy to study channels 7 and 8 , with results that are in reasonable agreement with ours.18 EXPERIMENT A skimmed molecular beam of N2O was introduced along the axis of a time-of- ight mass spectrometer TOFMS , through the central holes in a set of three parallel, circular electrodes. Midway between the high-voltage repeller and accelerator plates, the beam was crossed at right angle by three or more copropagating laser beams. Along the intersection cylinder, the laser radiation photolyzed N2O and ionized the recoiling dissociation products. The resulting ion cloud was accelerated through the opening in the third plate, held at ground, into a eld-free region, all the while expanding at the velocities from the dissociation. The voltages and spacing of the electrodes were chosen to focus ions born with the same velocity onto a position-sensitive, dual microchannel plate assembly coupled to a fast phosphor screen at the end of the time-of- ight TOF tube. Gating the detector allowed a fast charge-coupled device CCD camera to record the resulting two-dimensional image of the ion cloud on the detector for a particular ion mass. This ion-imaging technique has been described in detail elsewhere.19 21 Vacuum-ultraviolet VUV laser radiation was used both to photodissociate N2O and to detect the products by multiphoton ionization techniques. VUV photons were generated by a nonlinear four-wave mixing scheme at the difference frequency 2 1 2 between a two-photon resonance in krypton and a tunable photon in the visible to infrared reat 212.55 nm gion. Ultraviolet laser radiation 1 0.8 mJ/ pulse was employed to excite the 5p 0 , 1 / 2 , 0 4p6 transition in Kr by frequency doubling with a betabarium borate BBO crystal the output of a neodymium: yttrium aluminum garnet Nd:YAG third harmonic pumped Scanmate dye laser. The tunable 2 photon em- ployed in the VUV generation was produced by a second dye laser, pumped by the second harmonic of the same Nd:YAG laser. The 1 and 2 lasers were spatially and temporally overlapped by means of dichroic mirrors and a delay line, respectively, and focused by an achromatic lens into a stainless steel tube containing 10 20-Torr Kr. The resulting VUV and the residual incident laser light entered the TOFMS through a MgF2 collimating lens, which served as the exit window of the Kr cell. With horizontal and vertical polarizations of 1 and 2, respectively, the resulting polarization of the VUV was vertical and parallel to the plane of the detector. N2O was dissociated near the peak of the absorption curve at wavelengths around 130 nm, which were convenient for detection of atomic products by 1 + 1 resonanceenhanced multiphoton ionization REMPI , namely, O 3P2,1,0 at 130.2, 130.5, and 130.6 nm and N 2P3/2 at 131.05 and 131.09 nm, respectively. In each case, the VUV dissociation wavelength was coincident with an atomic resonance to an excited electronic state, which could then be ionized by absorption of the residual 212.55-nm UV laser light. The detection of atomic products at dissociation wavelengths other than an atomic resonance required the generation of a second VUV wavelength. This was accomplished by sending a second visible laser into the Kr cell, coincident in time and spatially overlapped with the previously mentioned UV and visible lasers, which generated the rst VUV wavelength. The scheme has been described more fully in a previous report.22 With the photolysis laser tuned just off the O 3P2 resonance near 130.2 nm, N 2D and O 1S photofragments could be detected with the second VUV laser by 1 + 1 REMPI schemes at 124.3 and 121.76 nm, respectively. The 2 photons required to generate the VUV wavelengths were 578 nm for O 3P2 , 561 nm for N 2P , 731 nm for N 2D , and 835 nm for O 1S . The latter wavelength was obtained as the rst Stokes line from Raman shifting 620 nm in H2. Typical laser pulse energies were 2 5 mJ for the visible light and 1 2 mJ for the Raman-shifted infrared light. Due to appreciable absorption by N2O at 124.3 and 121.76 nm, product images recorded by two VUV color schemes were a composite of dissociation processes at both pump and probe wavelengths. The contribution to the dissociation at the detection wavelength was minimized by optimizing the Kr pressure for more ef cient VUV generation at the dissociation wavelength. The remaining contribution at the detection wavelength was removed by a background subtraction scheme, involving the use of electronic shutters in each 2 laser. This allowed images to be collected on alternate sequences of laser shots, with dissociation and detection wavelengths together and then each separately. Several experiments were undertaken with a photomultiplier tube replacing the camera and without gating the detector. The TOF mass spectra were recorded in this way for the mass peaks at N+, O+, N+, and NO+ as a function of the 2 laser power for each of the 1 and 2 lasers. This was repeated for NO and N2 gases to determine the origin of the observed molecular ions. The scans of the 2 laser wavelength were useful in determining the spin-orbit branching ratios by comparison of the integrated ion signals at each of Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp 174304-3 Photodissociation channels for N2O J. Chem. Phys. 122, 174304 2005 TABLE I. Product branching ratios: Spin-orbit populations and dissociation channels. Atomic spin-orbit population ratios 2 N 2D5/2 : N 2D3/2 N 2P3/2 : N 2P1/2 O 3P 2 : O 3P 1 : O 3P 0 1.37 0.11 1.9 0.51 0.09 : 0.41 0.15 : 0.08 0.02 FIG. 1. O + N2 product images from N2O dissociation at 130.2 nm: a O 1S , b O 3P2 , and c N+. 2 Dissociation product channel ratios 2 O 3P 2 + N 2 A : O 3P 2 + N 2 B O 3P 1 + N 2 A : O 3P 1 + N 2 B O 3P 0 + N 2 A : O 3P 0 + N 2 B N 2 A + O 3P J : N 2 B + O 3P J O 3P + N2 A , B : N 2D , 2P + NO N 2DJ + NO: N 2PJ + NO 0.71 0.16 0.95 0.06 1.32 0.25 0.84 0.09 1.4 0.5 3 the ne structure levels for O 3PJ , N 2DJ , and N 2PJ . The relative sensitivity for VUV detection of O 3P2 and NO X , = 0 , 1 at 130.2 nm was established by dissociating NO2 at 355 nm and was used to calibrate the corresponding O:NO branching ratio obtained from N2O dissociation. The composition of the molecular beam was nominally 100% N2O 99% min, Matheson C.P. grade , except for the experiments detecting the major O 1S product channel for which a 10% N2O mixture in Ar 99.999 5%, Spectra Gases was prepared. In a diagnostic experiment, a premade 10% NO/ He mixture 9.97%, Matheson was used. The Krypton was research grade 99.998 5%, Praxair . RESULTS Following the dissociation of N2O at 130.2 nm, the TOF mass spectrum showed product peaks at N+ or O+ whenever the 2 laser was tuned to the appropriate 1 + 1 REMPI transition for detection of N 2DJ , N 2PJ , O 3PJ , or O 1S . Spectral scans over the ne structure levels yielded relative spin-orbit populations for O 3PJ , N 2DJ , and N 2PJ after normalization for variations in the laser powers. These are given in Table I. Product peaks also were observed for the molecular ions, N+, NO+, and N2O+, due to nonresonant 2 multiphoton ionization processes. Power dependence experiments established that NO+ was produced by sequential absorption of UV and visible photons by the neutral photoproduct and not by predissociation of N2O+. Experiments with a N2 beam con rmed that N+ could not be produced optically 2 from the ground state, inferring that the signal was due to the ionization of an excited state, likely by VUV absorption. The ratio of mass peaks for O+ and NO+ can be converted to the branching ratio between the dissociation channels, leading to O 3P + N2 and N + NO by calibrating against the O+ : NO+ mass peak ratio obtained from the dissociation of NO2 at 355 nm. The ion signals were normalized for variations in laser intensities, using the measured power dependences, and corrected for the fraction of the spin-orbit population appearing as O 3P2 .23 The resulting branching ratio is given in Table I. Velocity-mapped ion images were recorded for the dissociation products of N2O near 130 nm. The O + N2 product channels are represented by the O 1S , O 3P2 , and N+ im2 ages in Fig. 1. The N + NO product channels are represented by the N 2P3/2 , N 2D5/2 , and NO+ images in Fig. 2. Inverse Abel transformation of the images recovers a slice through the original three-dimensional velocity distribution. Integration over angles yields the speed distributions. The magni cation due to the camera lens and ion optics was determined by a calibration of the system with the dissociation of O2. Conversion to total kinetic-energy released TKER allows for comparison between product channels and reveals the internal energy distributions in the unobserved dissociation coproduct. The O + N2 TKER distributions derived from the images in Fig. 1 are shown in Fig. 3. The O 3P0 TKER distribution was included to show the effect of the spin-orbit level. Similarly, the N + NO TKER distributions derived from the images in Fig. 2 are shown in Fig. 4. The energy combs in Figs. 3 and 4 were derived from the photon energy, bond energies, and spectral constants. Inspection of the TKER distributions for the O + N2 dissociation channels shows that the observed N+ results from 2 the ionization of N2 in both the A 3 +u and B 3 g electronically excited states, as coproducts with O 3PJ . O 1S is coproduced with unobserved N2 in the ground electronic state. Electronically excited atoms/diatoms are coupled with ground state diatoms/atoms, respectively. The N2 A : B branching ratio is easily determined by integration over the two peaks in the O atom TKER, and appears to differ with the atomic ne-structure level. Assuming that the ionization steps are equally ef cient, the same information is obtained from the peak areas in the N+ TKER. The N + NO data indi2 cate that the NO+ image is a composite of two dissociation channels, leading to the two electronically excited states of atomic nitrogen, with the N 2P correlating with slower NO and the N 2D with faster NO. The N 2P products were shown to be bimodal with a minor contribution at very low kinetic energy. The N 2D : N 2P branching ratio may be determined from scaling the TKER distributions from the nitrogen atoms to t the NO distribution, and accounting for the spin-orbit populations. These branching ratios are also given in Table I. For electronically excited atomic nitrogen and oxygen, the peak kinetic energy is substantially lower FIG. 2. N + NO product images from N2O dissociation at 130.2 nm: a N 2D5/2 , b N 2P3/2 , and c NO+. The dissociation wavelength for b was 131.05 nm. Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp 174304-4 Lambert et al. J. Chem. Phys. 122, 174304 2005 FIG. 5. Angular distributions and t of 1 + P2 cos : a = 1.9, b O 3P2 + N2 A , = 0 , N+ outer ring; b = 1.3, O 1S + N2 X , = 0 , 2 3 2 2 O P0 + N2 B , = 0 ; c = 0.5, N P1/2 + NO X , N D5/2 + NO X ; and d = 0.14, NO+. The zero of the scale is shifted so that all the data can be displayed on one plot. I =c 1+ P2 cos , 9 FIG. 3. Total kinetic energy released for O + N2 products from the dissociation of N2O at 130.2 nm: a O 1S and b O 3P2 top solid curver, O 3P0 top dashed curve, and N+ bottom curve. 2 than the maximum allowed by conservation of energy, indicating that the diatoms have signi cant internal excitation. Product angular distributions were obtained from the inverse Abel-transformed image by tting the following function to annuli of eight to ten pixels in width: where c is a scaling constant, is the anisotropy parameter, which has the limiting values of 2 for a parallel transition and 1 for a perpendicular transition, and P2 x is the second-order Legendre polynomial, 1 / 2 3x2 1 . For the O + N2 dissociation channels, the tted anisotropy parameters are large and positive, indicating that the dissociating state is reached by a parallel transition. With increasing vibrational excitation in the N2 coproduct, however, the anisotropy parameter decreases substantially. The good agreement of the O 3P2 and O 3P0 angular distributions indicates that there is no alignment of the atomic angular momentum. The N + NO dissociation channels yielded angular distributions for the N 2D , 2P products, which are characterized by positive values of , whereas the NO+ image appeared to be very nearly isotropic. An alignment of the N 2PJ fragments could not be con rmed by comparison of the angular distributions since both J = 1 / 2 and J = 3 / 2 fragments gave similar s within the fairly large 2 uncertainties. Angular distributions together with the resulting t of Eq. 9 to the data are displayed in Fig. 5, while the determined anisotropy parameters are given in Table II. DISCUSSION Branching ratios A product branching ratio for O 3P : NO X of 1.4 0.5 was obtained for the 130.2-nm dissociation of N2O. The VUV detection scheme for O and NO was calibrated by using it with the dissociation of NO2, which has O + NO as the sole product channel. This required the normalization of the O+ and NO+ ion signals for variations in the laser powers, using measured power dependences both for N2O and NO2 photodissociations. The O+ : NO+ ratio was taken as O+/NO+ = SO/SNO f SO O 3P / NO , FIG. 4. Total kinetic energy released for N + NO products from the dissociation of N2O at 130.2 nm: N 2P3/2 top solid curve at 131.054 nm , N 2D5/2 top dashed curve, and bottom NO+ curve. where SO/SNO is the experimental sensitivity relating the ion signal to number densities, refers to the number densities, and f SO is the spin-orbit fraction of O 3PJ with J = 2. For the Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp 174304-5 Photodissociation channels for N2O J. Chem. Phys. 122, 174304 2005 TABLE II. Anisotropy parameters determined from tting Eq. 9 to the angular distributions of various N2O dissociation product channels. Product channel O 1S + N 2 X , O 3P 2 + N 2 A , O 3P 0 + N 2 A , N 2 A , + O 3P J O 3P 2 + N 2 B , O 3P 0 + N 2 B , N 2 B , + O 3P J N 2D5/2 + NO X N P3/2 + NO X N 2P1/2 + NO X NO X + N 2D2P 2 Anisotropy parameter = 0 1.3 0.2; 0.9 0.1 = 0 2.0 0.2 = 0 1.8 0.2; = 0 1.8 0.2 = 0 1.3 0.2; = 0 1.3 0.1; = 0 1.4 0.1; 0.48 0.15 2 = 2 1.1 0.1; =3 = 1 1.3 0.2; = 1 1.6 0.3 = 1 0.7 0.1 = 1 0.9 0.1 = 1 0.9 0.1 0.68 0.12 fast ; 0.03 0.31 slow 0.51 0.12 fast ; 0.39 0.26 slow 0.14 0.13 dissociation of NO2 at 355 nm, the relative sensitivity factor is determined because O 3P / NO is unity and f SO has been previously measured to be 0.82.23 With the relative sensitivity factor known and the spin-orbit fraction measured as 0.51 for N2O dissociation, the above relationship yields O 3P / NO . The 2 uncertainty includes contributions from f SO and the relative ion signals from more than 50 TOF mass spectra recorded at various laser powers. The resulting branching ratio relies on the assumption that the relative sensitivity for O + NO measured with NO = 0 , 1 in the NO2 system does not change signi cantly over the NO = 0 9 range of vibrational levels accessible in the N2O system. That this assumption is reasonable is demonstrated by the good agreement of the TKER distributions of NO+ with N 2D , 2P , for which the detection ef ciency is independent of the vibrational level of the NO partner. A product branching ratio for N 2D : N 2P of 3 was obtained for the 130.2-nm dissociation of N2O from comparison of the TKER distributions of the N atoms with that of NO+. Beginning with TKER distributions normalized to unit area, the N 2D TKER was scaled to t the NO+ TKER at energies between 1 and 2 eV. Below 1 eV, N 2P begins to make a contribution, especially near 0.5 eV. Scaling the N 2P TKER to make up the difference between the results for NO+ and N 2D yields the branching ratio. The sum of the contributions from the N atoms accounts for about 90% of the NO+ TKER. The regions of mismatch fall in the highenergy wing of the N 2D distribution and in the valley of the bimodal N 2P distribution this bimodality appears to be detection ef ciency bereal . It is unlikely that the NO haves strangely at these particular internal energies. It is more likely that the dif culties in determining and subtracting the background contribution to the weak NO+ image are the source of the problem. Considering the poor signal-tonoise ratio of both the NO+ and the N 2D TKER distributions, the agreement of the t is very reasonable. While no uncertainty for the branching ratio is given in Table I, the t was recognizably worse with N 2D /N 2P ratios of 2 or 4. The t assumes that the TKER distributions do not differ between atomic spin-orbit levels. Although this is not the case for O 3PJ , analysis of N 2P1/2,3/2 images gave insigni cantly different TKER distributions. Despite the branching ratio favoring N 2D production, the N 2P signal was stronger due to the relatively inef cient VUV generation at 124.3 nm compared to 131.05 nm. As well, the ionization step for N 2D relied solely on the residual 1 ultraviolet laser light, whereas the ionization step for N 2P could be effected by the 2 visible laser light, which was at least a factor of 5 greater in power. Product branching ratios for N2 A 3 +u : N2 B 3 g were determined from the analysis of the O 3PJ and N+ images. 2 The ratio of the integrated peaks in the TKER distributions was on the order of unity, although the ratio did depend strongly on the O atom spin-orbit level detected, favoring the N2 A 3 +u state for O 3P0 and the N2 B 3 g state with O 3P2,1 . Witinski et al. also observed this difference with spin-orbit level and our branching ratio results are in reasonable agreement.18 The branching ratio obtained from the N+ 2 data incorporates contributions from each of the O 3P spinorbit levels and is consistent with the weighted average 0.86 0.18 of the O 3P2,1,0 branching ratios, where the weights are the measured relative abundances. The good agreement indicates that the ionization ef ciencies of the N2A 3 +u and B 3 g states must not be very different at 130.2 nm. As well, the effect of dissociation at O 3P2,1,0 detection wavelengths differing by 0.3 nm is shown to be insigni cant. Previous investigators3,14 have detected visible N2 B3 g-A3 +u uorescence, following N2O dissociation only at shorter wavelengths where the emission is from higher vibrational levels. Their experiments were not sensitive to infrared emission from the low vibrational levels of the B state populated at dissociation wavelengths near 130 nm. As a result, they grossly underestimated the B state population, reporting an A : B branching ratio of 99 at 123.6 nm. Despite the uncertainty in how the relative ionization ef ciencies of the two excited electronic states of N2 change with wavelength, it is clear from the N+ images that the order 2 of magnitude of the A / B ratio is closer to unity than to 100, with values estimated to be 0.3, 0.84, and 3.3 at dissociation wavelengths of 124.3, 130.2, and 131.05 nm, respectively. Because of uncertainties in the relative ionization ef ciency mentioned above, we cannot say whether or not there is a systematic trend in the ratio with dissociation wavelength. Combining the three branching ratios relating O 3P to NO, N 2P to N 2D , and N2 A 3 +u to N2 B 3 g , the branching between the dissociation channels 5 8 at 130.2 nm is O 3P + N 2 A 3 + NO X 2 + u :O 3P + N2 B 3 g :N 2D :N 2P + NO X 2 = 2.6:3:3:1, with 2 uncertainties about 35%. Black et al. have determined the O 1S quantum yield from N2O dissociation at 129 nm to be 0.95 0.05 by direct observation of the O 1S emission and by comparing the signal level with that from the photodissociation of CO2 at 112 nm where the quantum yield is unity.14 They then xed the quantum yield to be 0.93 Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp 174304-6 Lambert et al. J. Chem. Phys. 122, 174304 2005 because they observed a 0.07 quantum yield for N2 A3 +u . Nee et al. also measured the O 1S quantum yield at dissociation wavelengths close to 130 nm by observing O 1S emission enhanced by Xenon buffer gas.2 From the data in their Fig. 3, the average quantum yield is 0.88 0.05 for wavelengths of 128, 129, 132, and 133 nm. It is reasonable and convenient to x the O 1S quantum yield to be 0.90 at 130.2 nm and to treat our combined branching ratios as % quantum yields. At 130 nm, the quantum yield for N2 A3 +u is given by Black et al. as 6 1%. Under their conditions, the N2 B3 g would have been completely quenched to N2 A3 +u at the start of the experiment,24 so that 6% is the sum total of O 3P + N2 A , B . This agrees well with our sum of 2.6% + 3%. Their quantum yield of 3 3% for N 2D at 130 nm is also in agreement with our value. A comparison of our O 1S and O 3P signal levels after normalizing for laser powers and oscillator strengths for the atomic transitions con rmed the order of magnitude difference in quantum yields for the two electronic states. We cannot give a better estimate without quantifying the relative ef ciencies for VUV generation at 121.76 and 130.2 nm and the relative ionization ef ciencies of the resonantly excited O atoms. Vibrational distributions of the N2 and NO products are qualitatively consistent with the structures of the N2O excited states determined by Hopper,17 who found both the N NO distance and the N NO distance to be 1.3 . Because the N N distance in the N2 products is 1.094, 1.293, and 1.2123 for the X, A, and B states, respectively,25 it is clear that the N2 product will be vibrationally excited for all three channels. Similarly, the N O bond distance is 1.1508 for the NO X product,25 so this product also is likely to be vibrationally excited. The vibrational combs in Figs. 3 and 4 provide an indication of the excited vibrational distributions for these diatomic fragments. Angular distributions The electronic con guration of the 1 + ground state of N2O is 7 2 1 4 2 4. Ab initio calculations by Hopper17 have con rmed that the strong absorption near 130 nm is mainly due to the electronic excitation to the linear D 1 + state the electronic con guration of which is 7 2 1 4 2 3 4 . The - * excitation weakens both the N O and the N N bonds. In addition to the D 1 + state, there is also a 3 v state in this same energy region, just slightly higher in energy than the D 1 + state at the linear geometry. Its electronic con guration is 7 1 4 2 4 3 , where the last electron is in an antibonding orbital on the N N and N O bonds. Let us rst consider the dissociation from these states to N2 + O. For dissociation along the NN O coordinate, the D 1 + state correlates in the linear con guration to N2 X 1 +g + O 1S , but in the bent con guration it becomes a 1A state and correlates to N2 A 3 +u + O 3P . Thus, there is an avoided seam along the dissociation surface. Along this same NN O coordinate, the 3 v state correlates in the linear con guration to N2 A 3 +u + O 3P , but in the bent con guration, the 3 v state becomes two components of a Renner Teller pair, a lower 3A state and an upper 3A state. The lower-energy 3A component correlates to N2 B 3 g + O 3P , while the higher-energy 3A component correlates to N2 A 3 +u + O 3P . Thus, there is also a seam of crossing between the 3A and 3A surface, but this crossing is allowed because the surfaces are of different symmetries. Next, con sider the dissociation from the D 1 + and 3 v states to N + NO. For dissociation along the N NO coordinate, the D 1 + state correlates in the linear con guration to N 2P + NO X 2 , and in the bent con guration it correlates to N 2D + NO X 2 . Thus, there is another avoided seam along this dissociation surface. Along this N NO coordinate, the 3 v state correlates in the linear con guration to N 2D + NO X 2 , but in the bent con guration, again, there are two Renner Teller components. Both the 3A component and the 3A component correlate to N 2D + NO X 2 . Note that because, in the linear con guration, the D 1 + state, which is 3 lower in energy than the v state, correlates to the more energetic N 2P + NO X 2 channel, while the 3 v state correlates to the less energetic N 2D + NO X 2 , there must be a crossing between these surfaces. In the absence of spinorbit coupling, the crossing is allowed, but in its presence the 3 A and 1A components may have an avoided crossing. The angular distributions of the products can help to determine which of these dissociation paths are the most important. In a linear molecule for prompt axial recoil, the expected values of the anisotropy parameter are, using linearly polarized light, 2 for a parallel transition and 1 for a perpendicular transition. For the dissociation of a bent triatomic molecule, again in the axial recoil limit and for linear polarization, the anisotropy parameter is related to the angle between the recoil direction and the transition dipole mois the second Legendre ment = 2P2 cos , where P2 polynomial. We now enumerate the various product channels, note how each may be reached, and comment on the expected and measured anisotropy parameters. The N2 X 1 +g + O 1S channel is accessed by dissociation only from the D 1 + state in the linear con guration. For a parallel transition, such as from the N2O ground state 1 + to the optically allowed X D 1 + state, the limiting value of the anisotropy parameter is well-known to be 2 for dissociation by linearly polarized light or 1 for dissociation by unpolarized light.26 The angular distribution of the O 1S fragment from the 130-nm photodissociation measured by Stone et al. using an unpolarized light source is consistent with = 1, indicating a parallel transition.16 Similarly, our own measurement using linearly polarized light gave = 1.3 for the O 1S , consistent with a parallel transition, but perhaps not quite as high as we might have expected. The N2 A 3 +u + O 3P channel is accessed a from the 1+ state after it bends to become 1A , b from the 3 v D state in the linear con guration, or c from the upper 3A Renner Teller component of the 3 v state in the bent con guration. The basic transition is again a parallel one, but two of the three possibilities come from dissociation of a bent N2O. Thus, we might expect that could be reduced from its limiting value of 2.0. Somewhat surprisingly, the Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp 174304-7 Photodissociation channels for N2O J. Chem. Phys. 122, 174304 2005 data indicate = 2.0, so that it must be that dissociation takes place from a nearly linear con guration. The correlations calculated by Hopper would suggest that dissociation proceeds following crossing to the 3 v state. The N2 B 3 g + O 3P channel can come only from the lower 3A component of the 3 v, state in the bent con gu ration assuming that only the D 1 + and 3 v states to be involved . The data indicate that = 1.3, in general agreement with a parallel excitation followed by dissociation from a somewhat bent geometry. Witinski et al. also reported a similar value of = 1.5 for this channel.18 The N 2D + NO X 2 channel can come a from the 3 state in a linear con guration, b from either of the v Renner Teller components of the 3 v state in a bent con gu ration, or c from the bent 1A con guration of the D 1 + 2 state. The N D image gives a of 0.5, suggesting that the bent con gurations dominate. channel comes only from Finally, the N 2P + NO X2 1 + state. We expect = 2, the linear con guration of the D but the data suggest something lower, = 0.5 0.7. The value of the anisotropy parameter, based on measurement of the channel, is also somewhat lower than expected NO X 2 from the measured parameters for the N 2P and N 2D , but this determination is rendered less certain due to the necessity for background subtraction in the measurement. The variations of the anisotropy parameter with internal energy of the diatomic fragment indicated in Table I can be understood qualitatively based on the model described by Demyanenko et al.27 in which the transverse recoil component is calculated based on the angular momentum calculation. The principal result is given by the equation sin2 = diatomr 2 A BCRC 2 tions from 5 8 , all of the quantum yields between 1% and 3%. Both of the diatomic products are vibrationally excited in all accessible channels. The spin-orbit ratios, relative branching ratios, and anisotropy parameters have been determined, as listed in Tables I and II. The dynamics of the dissociation appear to be consistent with the calculations of Hopper; dissociation takes place predominantly through the D1 + state, with likely participation of the 3 v state. ACKNOWLEDGMENTS This work was supported by the Department of Energy under Grant No. DE-FG02-88ER13934 and in part by the National Science Foundation under Grant No. CHE0239903. We have bene tted from conversations with M. F. Witinski, M. Ortiz-su rez, and H. F. Davic, who are thanked for making a copy of their paper available to us prior to publication. P. J. Crutzen, Q. J. R. Meteorol. Soc. 96, 320 1970 ; P. J. Crutzen, J. Geophys. Res. 76, 7311 1971 . 2 J. B. Nee, J. C. Yang, P. C. Lee, X. Y. Wang, and C. T. Juo, Chin. J. Phys. Taipei 37, 172 1999 . 3 L. C. Lee and M. Suto, J. Chem. Phys. 80, 4718 1984 . 4 M. Zelikoff, K. Watanabe, and E. C. Y. Inn, J. Chem. Phys. 21 1643 1953 . 5 P. Felder, B.-M. Haas, and J. R. Huber, Chem. Phys. Lett. 186, 177 1991 . 6 N. Shafer, K. Tonokura, Y. Matsumi, S. Tasaki, and M. Kawasaki, J. Chem. Phys. 95, 6218 1991 . 7 L. L. Springsteen, S. Satyapal, Y. Matsumi, L. M. Dobeck, and P. L. Houston, J. Phys. Chem. 97, 7239 1993 . 8 T. F. Hanisco and A. C. Kummel, J. Phys. Chem. 97, 7242 1993 . 9 T. Suzuki, H. Katayanagi, Y. Mo, and K. Tonokura, Chem. Phys. Lett. 256, 90 1996 . 10 D. W. Neyer, A. J. R. Heck, and D. W. Chandler, J. Chem. Phys. 110, 3411 1999 . 11 D. W. Neyer, A. J. R. Heck, D. W. Chandler, J. M. Teule, and M. H. M. Janssen, J. Phys. Chem. A 103, 10388 1999 . 12 A. Brown, P. Jimeno, and G. G. Balint-Kurti, J. Phys. Chem. A 103, 11089 1999 . 13 S. Nishida, K. Takahashi, Y. Matsumi, N. Taniguchi, and S. Hayashida, J. Phys. Chem. A 108, 2451 2004 . 14 G. Black, R. L. Sharpless, T. G. Slanger, and D. C. Lorents, J. Chem. Phys. 62, 4266 1975 . 15 R. Gilpin and K. H. Welge, J. Chem. Phys. 55, 975 1971 16 E. J. Stone, G. M. Lawrence, and C. E. Fairchild, J. Chem. Phys. 65, 5083 1976 . 17 D. G. Hopper, J. Chem. Phys. 80, 4290 1984 . 18 M. F. Witinski, M. Ortiz-Su rez and H. F. Davis, J. Chem. Phys. 122, 174303 2005 , preceding paper. 19 R. J. Wilson, J. A. Mueller, and P. L. Houston, J. Phys. Chem. A 101, 7593 1997 . 20 B.-Y. Chang, R. C. Hoetzlein, J. A. Mueller, J. D. Geiser, and P. L. Houston, Rev. Sci. Instrum. 69, 1665 1998 . 21 B. R. Cosofret, H. M. Lambert, and P. L. Houston, J. Chem. Phys. 117, 8787 2002 . 22 H. M. Lambert, A. A. Dixit, E. W. Davis, and P. L. Houston, J. Chem. Phys. 121, 10437 2004 . 23 J. Miyawaki, T. Tsuchizawa, K. Yamanouchi, and S. Tsuchiya, Chem. Phys. Lett. 165, 168 1990 . 24 R. A. Young, G. Black, and T. G. Slanger, J. Chem. Phys. 50, 303 1969 . 25 G. Herzberg, Spectra of Diatomic Molecules Van Nostrand, New York, 1950 . 26 R. N. Zare, Mol. Photochem. 4, 1 1972 . 27 A. B. Demyanenko, B. Dribinski, H. Reisler, H. Meyer, and C. X. W. Qian, J. Chem. Phys. 111, 7384 1999 . 1 Eavl 1 Erot , is the angle between the actual and axial recoil where directions, Eavl = Etrans + Erot is the energy available to the fragments, exclusive of internal vibrational and electronic energy, r is the diatomic bond length, and Rc is a critical distance parameter at which the angular momenta have reached their nal values. Because we do not resolve the rotational energy of the fragment, we cannot perform a quantitative analysis, but it is clear from the formula that as Eavl decreases, assuming comparably sampled values for Erot, the denominator will decrease, leading to larger predictions for . Because, in this theory, = 2P2 cos , it is clear that should decrease with decreasing Eavl. This behavior is obchannel, the O 3P0 served for the O 1S + N2 X , + N2 A , channel, the O + N2 B , channel, and the N 2P + NO X channels. CONCLUSIONS Dissociation of N2O near 130 nm leads to at least ve product channels, 4 8 . Based on this and previous work, it appears that channel 4 dominates, with minor contribu- Downloaded 20 Jun 2005 to 128.253.219.80. Redistribution subject to AIP license or copyright, see http://jcp.aip.org/jcp/copyright.jsp

Find millions of documents here - Study Guides, Homework Solutions, Papers, Exam Answer Keys and more. Course Hero has millions of course related materials that will enable you to learn better, faster and get an A in all your courses.
Below is a small sample set of documents: