View Article Online View Journal
PCCP Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: E. MartinezNunez and S. Vazquez, Phys. Chem. Chem. Phys., 2015, DOI: 10.1039/C4CP05626D.
This is an Accepted Manuscript, which has been through the Royal Society of Chemistry peer review process and has been accepted for publication. Accepted Manuscripts are published online shortly after acceptance, before technical editing, formatting and proof reading. Using this free service, authors can make their results available to the community, in citable form, before we publish the edited article. We will replace this Accepted Manuscript with the edited and formatted Advance Article as soon as it is available. You can find more information about Accepted Manuscripts in the Information for Authors. Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains.
www.rsc.org/pccp
Page 1 of 8
Physical Chemistry Chemical Physics View Article Online
Journal Name
RSCPublishing
ARTICLE
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
Cite this: DOI: 10.1039/x0xx00000x
HCN elimination from vinyl cyanide: Product energy partitioning, the role of hydrogen-deuterium exchange reactions and a new pathway Saulo A. Vázquez,a and Emilio Martínez-Núñez*a
Received 00th January 2012, Accepted 00th January 2012 DOI: 10.1039/x0xx00000x www.rsc.org/
The different HCN elimination pathways from vinyl cyanide (VCN) are studied in this paper using RRKM, Kinetic Monte Carlo (KMC), and quasi-classical trajectory (QCT) calculations. A new HCN elimination pathway proves to be very competitive with the traditional 3-center and 4center mechanisms, particularly at low excitation energies. However, low excitation energies have never been experimentally explored, and the high and low excitation regions are dynamically different. The KMC simulations carried out using singly deuterated VCN (CH2=CD-CN) at 148 kcal/mol show the importance of hydrogen-deuterium exchange reactions: both DCN and HCN will be produced in any of the 1,1 and 1,2 elimination pathways. The QCT simulation results obtained for the 3-center pathway are in agreement with the available experimental results, with the 4-center results showing much more excitation of the products. In general, our results seem to be consistent with a photodissociation mechanism at 193 nm, where the molecule dissociates (at least the HCN elimination pathways) in the ground electronic state. However, our simulations assume that internal conversion is a fully statistical process, i.e., that the HCN elimination channels proceed on the ground electronic state according to RRKM theory, which might not be the case. In future studies it would be of interest to include the photoprepared electronically excited state(s) in the dynamics simulations.
four HNC elimination pathways were detected using high level ab initio calculations.35
Introduction The photodissociation dynamics of haloethylenes (CH2=CHX, X = F, Cl, Br) has been extensively studied in the past.1-26 Among the different channels taking place in the decomposition of these systems, the study of the HX eliminations has been central in many investigations. This is because there is a competition between 1,1 and 1,2 HX elimination pathways, which proceed, respectively, via 3-center and 4-center transition states on the ground electronic state of the system.12, 14, 15, 18, 19 Although alternative routes for HX elimination were found for these systems, the 1,1 (or 3-center) elimination channel is the most important route, followed by the 1,2 (or 4-center) channel. The 3-center/4-center branching ratio ranges from 2 to 7 for the typical laser wavelength of 193 nm (i.e. 148 kcal/mol).2, 18, 27 This leads, in general, to highly rovibrationally excited photoproducts, which complicates the spectroscopic measures of energy partitioning. Another substituted analogue of ethylene, whose photodissociation dynamics has been the subject of a number of recent studies is vinyl cyanide (VCN).28-36 In a previous ab initio/kinetic study of our group, three different HCN elimination pathways were found: 3center, 4-center and 5-center in the terminology of the present paper (see Figure 1), or paths I-III in the previous study.35 Additionally,
This journal is © The Royal Society of Chemistry 2013
The ab initio/kinetic study indicates that, among the HCN elimination channels, the 3-center is the preferred one, and the HCN/HNC ratio at 193 nm is 1.9,35 in very good agreement with the experimental result.34 Although the previous theoretical study gives a satisfactory explanation for the HCN/HNC elimination mechanisms from VCN, some further issues still need to be clarified. Very recently, one of us proposed an automated method for finding transition states of chemical reactions using dynamics simulations; the method has been termed Transition State Search using Chemical Dynamics Simulations (TSSCDS).37 One of the test cases employed to study the efficiency and efficacy of the method was VCN, for which a total of 83 transition states were found using density function theory (DFT) calculations. Among all these transition states, many H2 elimination pathways, as well as H- and CNmigration pathways were found. Most importantly, an additional HCN elimination channel (green route in Figure 1), involving three steps has been found. The transition state (TS) for the last step of this new pathway (CCdiss) has an energy lower than that of the 3-center mechanism. In addition, the experimental translational energy distribution for the HCN elimination channel was recorded employing photofragment
J. Name., 2013, 00, 1-3 | 1
Physical Chemistry Chemical Physics Accepted Manuscript
DOI: 10.1039/C4CP05626D
Physical Chemistry Chemical Physics
Page 2 of 8 View Article Online
Journal Name DOI: 10.1039/C4CP05626D
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
translational spectroscopy,28 and the average internal energies of the HCN elimination products were provided in a later time-resolved Fourier transform infrared emission spectroscopy study.34 However, there is a lack of simulation studies that would provide further support for the reaction mechanism. The aim of this paper is twofold. On the one hand, the relative importance of the new HCN elimination mechanism (CCdiss) will be assessed using the combination of Rice-Ramsperger-KasselMarcus (RRKM) theory calculations and Kinetic Monte Carlo (KMC) simulations. The importance of the H- and CN-migration mechanisms will be studied in a separate set of RRKM+KMC calculations, where singly deuterated VCN (CH2=CD-CN) is employed. If the efficiency of hydrogen-deuterium exchange (H/D scrambling) was negligible, then, DCN would be exclusively produced by the 1,1 elimination pathways (3-center and the new CCdiss pathways), whereas HCN would only be obtained from the 1,2 elimination channels (4-center and 5-center channels). By contrast, if H/D scrambling is 100% efficient, a 2:1 ratio would be always obtained for HCN:DCN, irrespective of the H(D)CN elimination route followed by the system. TS 4-center r
2
r1
118.0
Additionally, quasi-classical trajectory (QCT) simulations will be performed to calculate the product energy distributions (PEDs) obtained at excitation energies that correspond to 193 nm, to compare with available experimental data. If intramolecular vibrational redistribution (IVR) is fast, which is the assumption made for the initialization of the QCT, then, an agreement between QCT and experiment would support the current consensus of the photodissociation mechanism at 193 nm: the dissociation channels occur on the ground electronic state following internal conversion (IC) from the initially optically prepared state.28
TS 5-center
115.3
r1
100.6
95.8
r2 TS 3-center
r1 TS cyc
81.9
HCN + H2C=C:
TS CCdiss r2 86.4 81.9
HCN + H2C=C:
TS iso 57.4
cyc 50.1
39.6
39.6
HCN + HCCH
HCN + HCCH 21.1 VisoCN 0.0 VCN
Figure 1. Different HCN elimination pathways found in our previous work (3-center, 4-center and 5-center channels),35 and the new pathway (green line) via the CCdiss transition state.37 The numbers are relative energies in kcal/mol (including the zero-point vibrational energy) with respect to VCN, calculated at the CCSD(T)/6-311++G(3df,3pd)//CCSD/6-311+G(2d,2p) level of theory with the vibrational frequencies obtained using CCSD/6-311+G(2d,2p) numerical Hessians.
Computational details Kinetic calculations. The branching ratios of the different elimination pathways from VCN can be efficiently computed using a combination of both RRKM calculations and KMC simulations. As a first step, the microcanonical ki(E) rate coefficients for each elementary step i were calculated using RRKM theory following the procedure of previous work.35, 38 The density of states were
2 | J. Name., 2012, 00, 1-3
evaluated by direct count of the harmonic vibrational states using the Beyer-Swinehart algorithm. Then, the time evolution of reactants, intermediates and products was simulated by KMC,39, 40 which is a very useful Monte Carlo simulation for modeling the transient behavior of various molecular species that participate in many highly coupled chemical reactions. The above two-step (RRKM+KMC) procedure was employed in this study with a twofold purpose. On the one hand, to re-compute the
This journal is © The Royal Society of Chemistry 2012
Physical Chemistry Chemical Physics Accepted Manuscript
ARTICLE
Physical Chemistry Chemical Physics View Article Online
Journal Name
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
branching ratios of the different HCN/HNC elimination pathways, including the new HCN dissociation pathway (green line of Figure 1) found recently.37 Secondly, to assess the importance of the H- and CN- migration channels, shown in the Supporting Information (SI). Therefore, two different sets of computations were performed. In the first set of kinetic calculations, the four channels depicted in Figure 1 are considered, as well as the four HNC elimination channels computed previously.35 These simulations were carried out for excitation energies in the range 110-200 kcal/mol. The barrier heights and vibrational frequencies of the elementary steps involved in the new (CCdiss) pathway, needed in the RRKM analysis, were calculated here using the CCSD and CCSD(T) methods with the 6311+G(2d,2p) and 6-311+G(2d,2p) basis sets, as it was done for the other pathways.35 For the second set of computations the α-hydrogen was substituted by deuterium (CH2=CD-CN). The excitation energy in this case was 148 kcal/mol, i.e., the energy that corresponds to a photon of 193 nm. The aim of this simulation is an assessment of the importance of H/D scrambling. For this reason, besides the HCN elimination channels of Figure 1, several H- and CN- migration channels37 are also incorporated in the simulations. This set of calculations involves two approximations. Firstly, to avoid increasing substantially the total number of elementary steps included in the simulation, the four HNC channels, as well as the corresponding migration channels from vinyl isocyanide (VisoCN) are neglected. One has to bear in mind that, since deuterium can be in any of the three different positions, the total number of elementary steps, despite the above simplification, is 51. Secondly, since the discussion on H/D scrambling is done in qualitative terms, the cheap but reliable MPWB1K/6-31+G(d,p) level of theory41 was selected to compute the barrier heights and vibrational frequencies. For a detailed explanation of these simulations the reader is referred to the SI. Quasi-classical trajectory (QCT) simulations. The QCT simulations are carried out starting from the 3-center, 4-center and CCdiss HCN elimination transition states (see Figure 1) of VCN. For a photon wavelength of 193 nm (148 kcal/mol), the 3-center, 4center and CCdiss mechanisms of Figure 1 account for 97% of the total amount of HCN formed (vide infra), with only 3% of VCN dissociating via the 5-center mechanism. For this reason only these three channels are considered in the QCT dynamics. Since the trajectories are propagated from the transition state to the products, the most relevant feature of the potential energy surface is the reverse barrier, i.e., the energy difference between the transition state and the products. The accurate energies calculated in our previous ab initio study35 are employed here as a benchmark to select a suitable level of theory for the direct dynamics calculations. The new pathway CCdiss was also characterized at the same level of theory as indicated in Figure 1. A method that usually provides rather good energy barriers at an affordable computational cost is MP2. The benchmark reverse barriers (see Figure 1) are 18.7, 78.4 and 4.5 kcal/mol, for the 3-center, 4-center and CCdiss pathways, respectively. These values can be compared with the MP2(fc) results using the 6-31G(d) basis set: 21.6, 88.1, and 4.4 kcal/mol, respectively. Although the use of MP2/(fc) calculations with larger basis sets improves the agreement with the benchmark calculations,
This journal is © The Royal Society of Chemistry 2012
ARTICLE DOI: 10.1039/C4CP05626D overall MP2(fc)/6-31G(d) provides a reasonable good balance between accuracy and computational cost. Therefore, this level of theory was our choice to perform the ab initio direct dynamics simulations of the present study. The general chemical dynamics program VENUS,42 interfaced with the electronic structure theory software package NWCHEM-5.1,43 was used for the direct dynamics simulations.44-46 The trajectories were initiated at the 3-center, 4-center and CCdiss transition states of Figure 1, using a quasi-classical normal mode sampling, which leads to a microcanonical ensemble of vibrational states;44, 47 the initial rotational angular momentum is zero. Since the aim is to simulate the photodissociation dynamics at 148 kcal/mol, an excess of vibrational energy of 47.4, 30, and 61.6 kcal/mol (see Figure 1) is placed at the TS 3-center, TS 4-center and TS CCdiss, respectively. Ensembles of 2000 trajectories were integrated using a Hessianbased predictor corrector algorithm with a fixed step size of 0.05 fs.48 Since the initial direction of the reaction coordinate velocity is random,44, 47 only approximately half the trajectories end in the product side. The trajectories are integrated until r1 and r2 (defined in Figure 1 for each TS) reached 10 Å. At this point, the relative translational energy, as well as the vibrational and rotational energies are calculated. Additionally, the rotational quantum number J of each fragment is obtained by equating the modulus of the classical rotational angular momentum to [J(J + 1)]1/2ℏ. The vibrational states of hydrogen cyanide (v1,v2,lv3), where v1, v2, v3 and l refer to the CH stretch, bend, CN stretch, and vibrational angular momentum, respectively, are also calculated here. For the vibrational quantum numbers, the normal mode analysis approach, including anharmonicity and Coriolis coupling terms, of Espinosa-Garcia is employed.49 The vibrational angular momentum l of HCN is computed from ja = lℏ, where ja is the projection of the rotational angular momentum of the molecule j onto the molecular axis. The (real) values of the quantum numbers thus obtained are rounded to the nearest integers. Trajectories that do not conserve the zero-point vibrational energy of the products are discarded. Additionally, the simulation distributions have been scaled by a factor of 0.9 to match the experimental available energy.34
Results and discussion The new HCN elimination pathway. Very recently one of us proposed an automated method to find transition states of chemical reactions called TSSCDS.37 Using this methodology, we have been able to find 83 transition states involved in the unimolecular decomposition of VCN at the MPWB1K/6-31+G(d,p) level of theory. Among all these structures, the TSs involved in the green pathway of Figure 1 were found (TS cyc and TS CCdiss). Intrinsic reaction coordinate (IRC) calculations carried out in this paper confirm the paths depicted in Figure 1. The first step of the path involves the VCN→VisoCN isomerization, which is not new.35 Then, there is another isomerization between VisoCN and another isomer of VCN called here cyc. The last step connects cyc with the HCN + H2C=C: elimination channel via the CCdiss transition state. The new channel went unnoticed in our previous ab initio/RRKM study.35 However, as will be shown below, our kinetic simulations carried out on the S0 ground electronic state show that this channel is competitive with the major channel for HCN production (the 3center channel) at low excitation energies. Also, at 193 nm the new channel is at least as important as the 4-center channel.
J. Name ., 2012, 00 , 1-3 | 3
Physical Chemistry Chemical Physics Accepted Manuscript
Page 3 of 8
Physical Chemistry Chemical Physics
Journal Name DOI: 10.1039/C4CP05626D
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
As mentioned in the Introduction, although alternative routes for HX elimination were also found for CH2=CHX systems, those pathways involved high-energy transition states and were not competitive with the conventional 3-center and 4-center elimination pathways. This is the first time a new HX elimination pathway competes with the wellknown 3-center and 4-center processes in CH2=CHX ethylene analogues. Relative importance of the different HCN and HNC elimination pathways from VCN. Figure 2a shows the relative abundances (product yields) of HCN produced via the different elimination mechanisms of Figure 1, as a function of the excitation energy. The 3-center pathway clearly dominates at all excitation energies, representing, at 135 kcal/mol, as much as 81% of the total number of HCN eliminations. However, the new pathway CCdiss, and the 4center pathways are competitive with (and even a bit more important than) the 3-center mechanism at low (110 kcal/mol) and high (200 kcal/mol) excitation energies, respectively. Since in the QCT simulations carried out in this work the excitation energy is 148 kcal/mol, it is also important to provide the product yields at this energy: 76.4%, 9.7%, 3.0% and 10.9% for the 3-center, 4-center, 5-center, and CCdiss pathways, respectively.
(a)
Product yields
80
60
40
at 193 nm, which compares very well with the 79% yield obtained in the present work. H/D scrambling. Figure 3 shows the DCN:HCN branching ratios obtained for the different elimination channels of Figure 1, when 148 kcal/mol are placed in singly deuterated VCN (CH2=CD-CN). The figure compares the results obtained in our KMC simulations with two hypothetical limiting situations. One limiting case corresponds to 0% of H/D scrambling, i.e., to a neglect of the H- and CNmigration channels. The other limiting case corresponds to a 100% efficiency of H/D scrambling, i.e., the assumption is that the H- and CN- migration channels are extremely efficient and present negligible barrier heights. In the case of 0% H/D scrambling, the 1,1 elimination channels (3-center and CCdiss) would lead to DCN exclusively, whereas the 1,2 elimination pathways (4-center and 5center) would only lead to HCN. By contrast, if H/D scrambling is 100% efficient, a 1:2 ratio is expected for DCN:HCN, irrespective of the pathway. As seen in the figure, the KMC simulation results fall halfway between the two limiting cases, which points out the importance of H/D scrambling in the ground state of VCN.
3-center
(a)
4-center
(b)
64 (b) HCN/HNC ratio 32
5-center
16 3-center 4-center 5-center CCdiss
(c)
CCdiss
(d)
8 4
20
0
2
120 140 160 180
Energy (kcal/mol)
1
120 140 160 180
Energy (kcal/mol)
Figure 2. KMC results for (a) the product yields (given in percentages) of the different HCN elimination channels of Figure 1, and (b) the HCN/HNC ratios obtained as a function of the excitation energy. These KMC results have been obtained including the four HNC simulation channels found in our previous study,35 and therefore, the HCN/HNC branching ratio can also be calculated; this result is shown in Figure 2b. Inclusion of the new HCN elimination pathway has a marked effect on this ratio at the lowest energies. In particular, at 120 kcal/mol, the HCN/HNC ratio computed previously (in the absence of the new CCdiss pathway) was 13.7,35 compared to a value of 18.5 obtained here. However, at 148 kcal/mol the value obtained without the new channel was 1.9,35 which compares very well with a value of 2.1 obtained in the present study. Additionally, Dai and co-workers,33, 34 obtained a 77% yield for vinylidene (vs 23% for acetylene) in the photodissociation of VCN
4 | J. Name ., 2012, 00 , 1-3
Figure 3. Relative abundances of HCN and DCN obtained via the different HCN elimination pathways when the –H and –CN migration channels are included in the KMC simulations. The KMC results are compared with two limiting situations of inefficient H/D scrambling and 100% efficient H/D scrambling. The above result can be explained from a simple inspection of Figure 1S of the SI, where the potential energy profiles for the different H- and CN- migration channels involved in this part of the study are depicted. The barrier heights of these isomerization pathways are substantially lower than those of the HCN elimination mechanisms (see Figure 1S of the SI). The H- and CN- migration processes occur in the picoseconds timescale, whereas HCN elimination does not take place significantly until several tenths of nanoseconds. QCT simulations. Figures 4 and 5 show the product energy distributions (PEDs) obtained in our simulations for the 3-center, 4center and CCdiss elimination channels. Overall, the PEDs of the 4center channel are hotter than those of the 3-center and CCdiss elimination pathways. This is not surprising since acetylene, which is formed in the 4-center channel, is substantially more stable than vinylidene, formed in the 3-center and CCdiss channels. In other words, the available energy for the 4-center mechanism is
This journal is © The Royal Society of Chemistry 2012
Physical Chemistry Chemical Physics Accepted Manuscript
ARTICLE
100
Page 4 of 8 View Article Online
Page 5 of 8
Physical Chemistry Chemical Physics View Article Online
ARTICLE DOI: 10.1039/C4CP05626D
substantially higher (104 kcal/mol) compared to that of the 3-center and CCdiss mechanisms (62 kcal/mol).34
P(Etr) distributions QCT Rescaled
QCT Simulated
3-center (76%) 4-center (10%) CCdiss (11%)
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
3-center 4-center CCdiss
20 40 60 80 100
20 40 60 80 100
QCT Total (rescaled) Expt
20 40 60 80 100
Energy (kcal/mol) Figure 4. Translational P(Etr) distributions obtained in our QCT simulations (left) for the 3-center, 4-center, and CCdiss channels. The middle plot shows the scaled QCT distributions using the percentages calculated in the KMC analysis. The right plot displays the total QCT distribution compared with the experimental one.28 The average values of the translational energies obtained in this study for the 3-center, 4-center and CCdiss channels are 20.0, 30.8, and 22.2 kcal/mol, respectively, which can be compared with the experimental value of 15 kcal/mol.28 The simulation P(Etr) distribution obtained for the 3-center channel resembles the experimental one more closely than the other two distributions (4center and CCdiss) do (see Figure 4). Although the total QCT distribution (including the contributions from all the channels) compares reasonably well the experimental one,28 remarkable good agreement can be obtained if the simulated QCT distribution is shifted by −5 kcal/mol.
C2H2 distributions 3-center
4-center
CCdiss
Vibration Rotation
20 40 60 80 100
20 40 60 80 100
20 40 60 80 100
Energy (kcal/mol)
Vibration Rotation
20 40 60 80 100
20 40 60 80 100
Energy (kcal/mol) Figure 5. Internal energy distributions of the C2H2 and HCN fragments, obtained in our simulations for the 3-center (left), 4center (middle), and CCdiss (right) channels.
This journal is © The Royal Society of Chemistry 2012
Additionally, the average internal energies of HCN produced via the CCdiss (16.2 kcal/mol) and 3-center pathways (24.0 kcal/mol) are much smaller than that obtained for the 4-center mechanism (55.9 kcal/mol). The value obtained in this study for the 3-center channel compares very well with the experimental result of 25.2 kcal/mol.34 Hydrogen cyanide produced in the 4-center channel is so vibrationally excited that approximately 2% of the HCN molecules experience an isomerization to HNC in the very short time dynamics of this study. The QCT simulation results will be further discussed below, in terms of the vibrational quantum number distributions. An interesting observation of a previous QCT study of the HCl elimination from vinyl chloride (VCl) was a very fast (of the order of 40-80 fs) vinylidene↔acetylene isomerization.4, 6 The dynamics calculations of the present paper point in the same direction: as much as 92% (93%) of the vinylidene formed in the 3-center (CCdiss) process isomerizes to acetylene in less than 476 fs, which is the maximum duration of the trajectories. By contrast, the acetylene molecules formed in the 4-center channel do not have enough energy to isomerize to vinylidene within the simulation time. Moreover, the average lifetime of vinylidene is ∼60 fs for both elimination pathways. The result for the average lifetime obtained in this study for VCN is in agreement with the above QCT result for VCl. The study of the relaxation rate from vinylidene to acetylene has been studied by several research groups,36, 50-54 showing a complex and interesting behavior. The vinylidene↔acetylene isomerization path is through extreme acetylene local bender vibrational levels,51, 52 that is, via discrete eigenstates on the acetylene side of the barrier. The nature of these eigenstates is very different depending on the rotational excitation of the molecule,53 leading to different limiting behaviors at low and high-J. In the rotationally cold limit54 one will have few-level quantum beats, whereas for highly rotationally excited C2H216, 36 the many-eigenstate dephasing will be fast and irreversible. The rotational quantum number distributions P(J) for C2H2 and HCN, shown in Figure 6, can be modeled reasonably well by Boltzmann distributions with rotational temperatures Trot ranging from 2331 to 6725 K.
HCN distributions
20 40 60 80 100
The experimental result for the average internal energy of vinylidene, 21.8 kcal/mol,34 falls between the simulation results obtained in this study for the 3-center and CCdiss channels: 18.0 and 23.6 kcal/mol, respectively.
The HCN vibrational quantum number distributions P(v), with v = (v1,v2,v3), are also shown in Figure 6 and the average values of the quantum numbers are collected in Table 1. The figure shows that the HCN P(v) distributions obtained for the 3-center and CCdiss pathways are similar to each other, and completely different from the distribution obtained in the 4-center mechanism. However, the bend mode (v2) is highly excited in all cases, with average quanta of ~5 and ~6 and ∼4, for the 3-center, 4-center, and CCdiss channels, respectively (see Table 1).
J. Name ., 2012, 00 , 1-3 | 5
Physical Chemistry Chemical Physics Accepted Manuscript
Journal Name
Physical Chemistry Chemical Physics
Journal Name DOI: 10.1039/C4CP05626D
Trot = 2331 K
50
100
Trot = 4749 K
Trot = 4401 K
150
50
100
Regarding the excitation of vibrational angular momentum, in general the average values of l are high in all cases, with the 3-center channel being slightly more prone to populate higher vibrational angular momentum states than the 4-center channel.
CCdiss
4-center 3-center C2H2 P(J) distributions
50
150
100
150
J HCN P(J) distributions Trot = 6725 K
50
100
Trot = 3133 K
Trot = 5372 K
150
50
100
50
150
100
150
J HCN P(v) distributions
(0,4,0)
(0,4,0) (0,6,0) (0,2,0)
(0,2,0) (0,6,0)
Photodissociation of vinyl cyanide at 193 nm. The results obtained in this paper are consistent with a mechanism where the molecule, after IC from the excited electronic state S1, dissociates in the ground electronic state S0 primarily via the 3-center pathway. This mechanism is supported by the good agreement found between the translational and internal energy distributions obtained in this paper and the experimental results obtained in the photodissociation of the molecule at 193 nm. However, the features of the vibrational quantum number distributions of H(D)CN obtained in recent experiments by Prozument et al.36, 55, where singly deuterated vinyl cyanide (CH2=CD-CN) is photoexcited at 193 nm, are very difficult to explain with the above mechanism. In these experiments, only the J = 0−1 rotational transition can be observed by Chirped-Pulse millimitre-Wave (CPmmW) spectroscopy for every vibrational level. Thus, the vibrational population is obtained from the measured transition intensity assuming that rotation is fully thermalized.
(0,8,0) (0,8,0)
(0,0,0) (4,0,0) (2,0,3)
0
The results of the CPmmW experiments are summarized as follows:
(0,0,0)
5000 10000 15000 0
10000
(6,0,0) (4,0,2)
20000
0
5000 10000 15000
-1
Energy (cm ) Figure 6. Rotational and vibrational quantum numbers distributions of the C2H2 and HCN fragments, obtained in our simulations for the 3-center (left), 4-center (middle), and CCdiss (right) channels. The solid blue lines are Boltzmann fits to the P(J) trajectory distribution. Table 1: Average values of the CH stretch , bend , CN stretch , and vibrational angular momentum quantum numbers obtained in our direct dynamics study. Channel
3-center
0
5
1
2.8
4-center
3
6
2
2.5
CCdiss
0
4
1
2.5
Figure 6 shows that the highest populated vibrational levels of HCN are (0,2,0), (0,4,0), and (0,6,0), when the molecule is formed via the 3-channel and CCdiss pathways. By contrast, when the molecules are initialized at the 4-center TS, the most likely vibrational state corresponds to six quanta in the CH stretch (v1). These results can be interpreted from the geometry of the transition states. While in the 3center and CCdiss transition states, the CH distance is 1.330 Å and 1.068 Å, respectively, the corresponding value in the 4-center transition state becomes 1.821 Å. This explains why the CH stretching mode is not excited in the HCN molecules produced via the 3-center and CCdiss dissociation pathways, whereas = 3 for the HCN molecules formed via the 4-center mechanism. The CN stretch mode is vibrationally excited in the HCN molecules produced through any of the different pathways, but again the 4-center pathway leads to more vibrational excitation in the this mode, probably due to coupling with the other modes of the molecule.
6 | J. Name ., 2012, 00 , 1-3
(a) The HCN vibrational distributions obtained from nondeuterated VCN and from CH2=CD-CN are very similar.36 (b) The vibrational distribution for DCN, which is consistent with that obtained here for the 3-center channel in Figure 5, peaks at the (0,4,0) state.55 (c) Overall, the HCN signal is about five times stronger than that of DCN.55 (d) The HCN vibrational quantum number distribution displays an exponential decay, i.e., peaking at (0,0,0).36, 55 In the first place, the exponential decay found for HCN does not correspond to any of the P(v) distributions found in the present study. Secondly, for the HCN signal to be substantially higher than that of DCN, the molecule has to proceed via the 4-center or 5-center paths (Figure 3), which only contribute to 9.7% and 0.3%, respectively of the total amount of HCN at 193 nm (see Figure 2), if the elimination takes place on S0. Thus, the experimental results36, 55 are only compatible with an alternative mechanism of HCN elimination. One alternative mechanism could be roaming of the cyano group, which has, indeed, been detected in a recent chemical dynamics study of the system.37 However, the HCN vibrational distribution obtained in such case would not be consistent with an exponential decay, it would rather be a highly excited distribution.56 Thus, the explanation for the recent CPmmW study36, 55 could be either an inefficient IVR after IC from the excited state and/or incomplete rotational thermalization of HCN and HNC products. Clearly, some further experimental and theoretical efforts are needed to fully understand the photodissociation dynamics of VCN at 193 nm.
Conclusions The main conclusions of this paper are summarized as follows:
This journal is © The Royal Society of Chemistry 2012
Physical Chemistry Chemical Physics Accepted Manuscript
ARTICLE
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
Page 6 of 8 View Article Online
Physical Chemistry Chemical Physics View Article Online
Journal Name 1.
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
2.
3.
4.
5.
6.
7.
8.
A new HCN elimination pathway (CCdiss) involving three elementary steps is characterized in this work. In contrast with other ethylene analogues (CH2=CHX), where the 3-center and 4-center HX elimination account for the vast majority of reactions, the CCdiss HCN elimination channel from VCN is very competitive. In fact, the CCdiss pathway is more important than the 4-center channel for excitation energies up to 150 kcal/mol, and accounts for 50% of the HCN eliminations at 110 kcal/mol. The computed HCN/HNC branching ratios increase significantly at low excitation energies when the new channel is incorporated in the kinetic calculations, but remains almost unaltered for energies greater than 150 kcal/mol. The experimental HCN/HNC ratio at 148 kcal/mol is very close to our theoretical result. Hydrogen/deuterium scrambling in singly deuterated VCN (CH2=CD-CN) is significant at 148 kcal/mol. For instance, instead of obtaining 100% of DCN for the 3-center elimination channel, our kinetic simulations predict an approximate DCN/HCN ratio of 2. The PEDs obtained in our QCT simulation for the 3center pathway are in general agreement with experiment. The PEDs obtained for the 1,1 elimination pathways (3-center and CCdiss) are rather similar to each other, whereas the (1,2 elimination) 4-center PEDs are much hotter. HCN molecules produced in any of the elimination mechanisms present large amounts of energy in the bending mode. Additionally, those produced via the 4center pathway have an average of three quanta in the CH stretch. The above results are consistent with a photodissociation mechanism at 193 nm, where the molecule dissociates in the S0 state after IC. However, recent CPmmW results may suggest inefficient IVR after IC into S0 and/or incompleteness of rotational thermalization in HCN (HNC) products.
Acknowledgements
ARTICLE DOI: 10.1039/C4CP05626D 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.
We thank Kirill Prozument and Robert W. Field for sharing their unpublished results, which motivated the present work. The authors thank “Centro de Supercomputación de Galicia (CESGA)” for the use of their facilities.
Notes and references a
Departamento de Química Física and Centro Singular de Investigación en Química Biológica y Materiales Moleculares, Campus Vida, Universidade de Santiago de Compostela, 15782, Santiago de Compostela, Spain Electronic Supplementary Information (ESI) available: Electronic energies, optimized geometries and vibrational frequencies obtained at the MPWB1K/6-31+G(d,p) level of theory for the stationary points of the pathways involved in the second set of kinetic calculations. See DOI: 10.1039/b000000x/
This journal is © The Royal Society of Chemistry 2012
29. 30. 31. 32. 33. 34. 35. 36.
37. 38. 39. 40. 41. 42.
T. Tarrazo-Antelo, E. Martinez-Nunez and S. A. Vazquez, Chem. Phys. Lett., 2007, 435, 176. E. Martinez-Nunez and S. Vazquez, Chem. Phys. Lett., 2006, 425, 22. E. Martinez-Nunez and S. Vazquez, J. Chem. Phys., 2005, 122, 1. E. Martinez-Nunez, S. A. Vazquez, F. J. Aoiz, L. Banares and J. F. Castillo, Chem. Phys. Lett., 2004, 386, 225. E. Martinez-Nunez and S. Vazquez, J. Chem. Phys., 2004, 121, 5179. E. Martínez-Núñez, A. Fernández-Ramos, S. A. Vázquez, F. J. Javier and L. Bañares, J. Phys. Chem. A, 2003, 107, 7611. J. Gonzalez-Vazquez, E. Martinez-Nunez, A. Fernandez-Ramos and S. A. Vazquez, J. Phys. Chem. A, 2003, 107, 1398. J. Gonzalez-Vazquez, A. Fernandez-Ramos, E. Martinez-Nunez and S. A. Vazquez, J. Phys. Chem. A, 2003, 107, 1389. E. Martinez-Nunez, C. M. Estevez, J. R. Flores and S. A. Vazquez, Chem. Phys. Lett., 2001, 348, 81. E. Martinez-Nunez and S. A. Vazquez, Chem. Phys. Lett., 2000, 332, 583. W. A. Jalenak and N. S. Nogar, J. Chem. Phys., 1983, 79, 816. M. Umemoto, K. Seki, H. Shinohara, U. Nagashima, N. Nishi, M. Kinoshita and R. Shimada, J. Chem. Phys., 1985, 83, 1657. T. R. Fletcher and S. R. Leone, J. Chem. Phys., 1988, 88, 4720. Y. Huang, Y. Yang, G. He, S. Hashimoto and R. J. Gordon, J. Chem. Phys., 1995, 103, 5476. G. He, Y. Yang, Y. Huang, S. Hashimoto and R. J. Gordon, J. Chem. Phys., 1995, 103, 5488. D. A. Blank, S. Weizhong, A. G. Suits, Y. T. Lee, S. W. North and G. E. Hall, J. Chem. Phys., 1998, 108, 5414. Y. R. Lee, L. D. Wang, Y. T. Lee and S. M. Lin, J. Chem. Phys., 2000, 113, 5331. S. R. Lin, S. C. Lin, Y. C. Lee, Y. C. Chou, I. C. Chen and Y. P. Lee, J. Chem. Phys., 2001, 114, 160. S. R. Lin, S. C. Lin, Y. C. Lee, Y. C. Chou, I. C. Chen and Y. P. Lee, J. Chem. Phys., 2001, 114, 7396. H. S. Ko, Y. R. Lee, C. C. Chen, L. D. Wang and S. M. Lin, J. Chem. Phys., 2002, 117, 6038. M. Bahou and Y. P. Lee, Aust. J. Chem., 2004, 57, 1161. J.-F. Riehl and K. Morokuma, J. Chem. Phys., 1994, 100, 8976. S. Kato and K. Morokuma, J. Chem. Phys., 1981, 74, 6285. T. Takayanagi and A. Yokoyama, Bull. Chem. Soc. Japan, 1995, 68, 245. S. A. Abrash, R. W. Zehner, G. J. Mains and L. M. Raff, J. Phys. Chem., 1995, 99, 2959. M. Barbatti, A. J. A. Aquino and H. Lischka, J. Phys. Chem. A, 2005, 109, 5168. E. Martínez-Núñez and S. A. Vázquez, Structural Chemistry, 2001, 12, 95. D. A. Blank, A. G. Suits, Y. T. Lee, S. W. North and G. E. Hall, J. Chem. Phys., 1998, 108, 5784. S. W. North and G. E. Hall, Chem. Phys. Lett., 1996, 263, 143. C. A. Bird and D. J. Donaldson, Chem. Phys. Lett., 1996, 249, 40. A. Fahr and A. H. Laufer, J. Phys. Chem., 1992, 96, 4217. A. Derecskei-Kovacs and S. W. North, J. Chem. Phys., 1999, 110, 2862. L. Letendre and H.-L. Dai, J. Phys. Chem. A, 2002, 106, 12035. M. J. Wilhelm, M. Nikow, L. Letendre and H.-L. Dai, J. Chem. Phys., 2009, 130, 044307. Z. Homayoon, S. A. Vazquez, R. Rodriguez-Fernandez and E. Martinez-Nunez, J. Phys. Chem. A, 2011, 115, 979. K. Prozument, R. G. Shaver, M. A. Ciuba, J. S. Muenter, G. B. Park, J. F. Stanton, H. Guo, B. M. Wong, D. S. Perry and R. W. Field, Faraday Discussions, 2013, 163, 33. E. Martínez-Núñez, J. Comput. Chem., 2014, 36, 222. W. H. Miller, J. Am. Chem. Soc., 1979, 101, 6810. A. B. Bortz, M. H. Kalos and J. L. Lebowitz, J. Comput. Phys., 1975, 17, 10. D. T. Gillespie, J. Comput. Phys., 1976, 22, 403. Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2004, 108, 6908. W. L. Hase, R. J. Duchovic, X. Hu, A. Komornicki, K. F. Lim, D.-H. Lu, G. H. Peslherbe, K. N. Swamy, S. R. Vande Linde, A. Varandas, H. Wang and R. J. Wolf, Quantum Chemistry Program Exchange (QCPE) Bulletin, 1996, 16, 671.
J. Name ., 2012, 00 , 1-3 | 7
Physical Chemistry Chemical Physics Accepted Manuscript
Page 7 of 8
Physical Chemistry Chemical Physics ARTICLE
44.
45. 46. 47.
Published on 05 February 2015. Downloaded by Selcuk University on 09/02/2015 06:13:27.
48. 49. 50. 51. 52. 53. 54. 55. 56.
Journal Name DOI: 10.1039/C4CP05626D M. Valiev, E. J. Bylaska, N. Govind, K. Kowalski, T. P. Straatsma, H. J. J. Van Dam, D. Wang, J. Nieplocha, E. Apra, T. L. Windus and W. A. de Jong, Computer Physics Communications, 181, 1477. K. Bolton, W. L. Hase and G. H. Peslherbe, Modern Methods for Multidimensional Dynamics Computations in Chemistry, World Scientific, Singapore, 1998. L. Sun and W. L. Hase, Reviews in Computational Chemistry, Wiley, New York, 2003. W. L. Hase, K. Song and M. S. Gordon, Comput. Sci. Eng., 2003, 5, 36. C. Doubleday Jr, K. Bolton, G. H. Peslherbe and W. L. Hase, J. Am. Chem. Soc., 1996, 118, 9922. U. Lourderaj, K. Song, T. L. Windus, Y. Zhuang and W. L. Hase, J. Chem. Phys., 2007, 126, 044105. J. C. Corchado and J. Espinosa-Garcia, Phys. Chem. Chem. Phys., 2009, 11, 10157. K. M. Ervin, J. Ho and W. C. Lineberger, J. Chem. Phys., 1989, 91, 5974. B. M. Wong, A. H. Steeves and R. W. Field, J. Phys. Chem. B, 2006, 110, 18912. B. M. Wong, R. L. Thom and R. W. Field, J. Phys. Chem. A, 2006, 110, 7406. D. S. Perry, J. Martens, B. Amyay and M. Herman, Mol. Phys. , 2012, 110, 2687. L. J., H. Feldman, A. Baer, D. Ben-Hamu, O. Heber, D. Zajfam and Z. Vager, Phys. Rev. Let.. 1998, 81, 3347. K. Prozument and R. W. Field, (unpublished results). D. Townsend, S. A. Lahankar, S. K. Lee, S. D. Chambreau, A. G. Suits, X. Zhang, J. Rheinecker, L. B. Harding and J. M. Bowman, Science, 2004, 306, 1158.
8 | J. Name ., 2012, 00 , 1-3
Physical Chemistry Chemical Physics Accepted Manuscript
43.
Page 8 of 8 View Article Online
This journal is © The Royal Society of Chemistry 2012