Social Media

An extensive investigation of reactions involved in the nitrogen trifluoride dissociation

Description
An extensive investigation of reactions involved in the nitrogen trifluoride dissociation
Categories
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
  3244  NewJ.Chem.,  2013,  37 , 3244--3251  This journal is  c  The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013 Cite this:  NewJ.Chem., 2013, 37 , 3244 An extensive investigation of reactions involved in thenitrogen trifluoride dissociation † Simone S. Ramalho, a Wiliam F. da Cunha, b Alessandra F. Albernaz, b Pedro H. O. Neto, b Geraldo Magela e Silva b and Ricardo Gargano* b In the course of our studies on nitrogen trifluoride dissociation, we consider in this paper two differentsets of reactive systems: the first one consists of the abstraction and the exchange channels of NF 3  + Fand the other is composed of the unimolecular and the abstraction production of N 2  + F. Accurateelectronic properties are determined and applied in the scope of the transition state theory (TST) toobtain thermal rate constants for these systems. An extensive investigation in terms of minimum energypaths and intrinsic reaction coordinates was previously carried out in order to ensure the good qualityof our TST results. We apply Wigner corrections to consider tunneling effects whenever theirimportance is numerically verified. The obtained results for the abstraction channel thermal rateconstants are in good agreement with experimental data which indicates that this kind of study is ofpotential use to help to clarify the decomposition mechanism of NF 3 . I Introduction In the last decades the environmental issue has continuously gained importance due to the major industrial development observed. New production lines are usually subjected to moresevere environmental statutes and governments throughout the world are led to follow several pro-environmental policies. An important example of the aforementioned policies is theper-fluorocarbons (PFC) emission reduction goal. It is a wellknown fact that the microelectronic industry is an important consumer of PFC, including tetrafluoromethane (CF 4 ), as etchgases for chamber cleaning processes. As a candidate for acleaner substitute to CF 4  we have focused attention on nitrogentrifluoride(NF 3 )which,besidesavoidingcontaminationwithcarbonresiduals and providing a process of near zero PFC emission, alsopresents advantages in terms of energy consumption and longertool lifetime. 1  Added to this is the fact that nitrogen trifluoridedecomposition is a source of fluorine atoms and NF 2  radicals,essential structures for a number of experimental physiochemicalinvestigations. 2–4 These evidences led to a substantial rise in overallNF 3  consumption which, together with the fact that NF 3  is animportant greenhouse gas, provides a necessity to investigate thedecomposition process of this compound.Despite all recent efforts to understand this important mechanism, the kinetics of NF 3  decomposition remain con-troversial, particularly when it comes to the removal of thispotent greenhouse gas from the atmosphere. 5,6  While theliterature is filled with experimentally dependent processes with high quality and reliable data, the phenomenologicalunderstanding of a more general qualitative mechanism, albeit less accurate, is missing. As an instructive example, we canmention the successful use of plasma technology for removing pollutants from gas streams resulted from catalyst degradationrecently reported. 7 In their work, Chen  et al.  propose a NF 3 breakdown mechanism based on a cylindrical dielectric barrierdischarge reactor constructed for the bench-scale experimentsthat yield an excellent comparison between experimental andtheoretically predicted data. Another experimental study by Wang  et al. 8 investigates the effects of different experimental parameterson the NF 3  decomposition process. We are interested in a moregeneral intermediate path of the NF 3  gas phase dissociationmechanism, a class of work with little presence in the literature.In this sense we have previously proposed a simple kineticmechanism composed of several elementary reactions taking part as intermediary steps to reach the global reaction. 9 Ourgoal was to gain a first understanding of the NF 3  decompositionprocess as a whole,  i.e. , without concerning with the specificity of each particular experimental situation. These limitations were accepted as a reasonable compromise between qualitativeunderstanding and simplicity of the proposed mechanism,although important effects and deviations of our results fromexperimental evidences can be partly attributed to this procedure. a Goias Federal Institute of Education, Science and Technology, Brazil  b  Institute of Physics, University of Brasilia, Brasilia, 70.919-970, Brazil. E-mail: gargano@unb.br  † Electronic supplementary information (ESI) available. See DOI: 10.1039/c3nj00553d Received (in Porto Alegre, Brazil)23rd May 2013,Accepted 1st August 2013DOI: 10.1039/c3nj00553d www.rsc.org/njc NJC PAPER     P  u   b   l   i  s   h  e   d  o  n   0   2   A  u  g  u  s   t   2   0   1   3 .   D  o  w  n   l  o  a   d  e   d   b  y   C  e  n   t  r  o   F  e   d  e  r  a   l   d  e   E   d  u  c  a  c  a  o   T  e  c  n  o   l  o  g   i  c  a   d  e   G  o   i  a  s  o  n   1   2   /   0   3   /   2   0   1   4   1   3  :   1   0  :   1   3 . View Article Online View Journal | View Issue  This journal is  c  The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013  NewJ. Chem.,  2013,  37 , 3244--3251  3245 Besidesunimoleculartypesystems(NF=N+F,NF 2 =NF+F,NF 3 =NF 2  + F, N 2 F = N 2  + F, N 2 F 3  = NF 2  + NF), the reactions involved inour mechanism are either of abstraction (NF 2  + F = NF + F 2 , NF 3  +F = NF 2  + F 2 , NF + N = N 2  + F, NF 3  + N = NF 2  + NF) or of exchange(NF + F = NF + F, NF 2  + F = NF 2  + F, NF 3  + F = NF 3  + F). The successachieved by our methodology in treating the systems involving NF 2  + F 9 and NF + F 10 leads us to proceed in our route to describethe kinetics of the 14 reactions that compose the proposed gasphase mechanism of NF 3  dissociation.Our procedure is to first perform accurate electronic structurecalculations in order to obtain energies, geometries and frequenciesfor reactants, products and transition states. We also obtain theminimum energy path and intrinsic reaction coordinates, featuresthat allow us to better understand each reaction mechanism. Wemake use of these properties to apply the transition state theory (TST) method 11 and thus to be able to calculate the thermal rateconstant (TRC) for each one of these reactions. We also included astandard Wigner correction of the theory  12 in order to contemplatetunneling effects.In this work, we treat two sets of different systems—onerelated to the reaction between NF 3  + F and the other producing N 2  + F—each one consisting of two different channels. By starting with the abstraction channel of the former set, NF 3  +F = NF 2  + F 2 , we obtained TRC in agreement with the experi-mental data present in literature. Endorsed by these results wethen applied the same methodology to tackle the other channel of theNF 3  +Fsystem,  i.e. , the exchangechannel (NF 3  +F=NF 3  +F) as well as the unimolecular and abstraction reactions of the N 2  + Fsystem, N 2 F = N 2  + F and NF + N = N 2  + F, respectively.The idea is to provide the literature with reliable theoreticaldata concerning the reaction rate of these three latter systems.Since no experimental data is available to this date, our resultsare of fundamental importance for those trying to elucidate thenitrogen trifluoride mechanism. This paper is organized asfollows: Section II describes the main features of the electronicstructure calculations performed and also briefly discuss theTST; our results are presented and discussed in Section III andits subsections, and finally summarized in Section IV. Furtherdetails on the electronic structure results can be found on theESI† of this work. II Methodology TST method is a mixed formalism that considers thermo-dynamic, kinetic theory and statistical mechanics treatments.It includes the best features of each one of these theories interms of precision without giving up the simplicity achieved by focusing attention on activated complexes which are assumedto be in ‘‘quasi-equilibrium’’ with reactants and their rate of transformation. 13 Due to its accuracy and simplicity, TST stillis, nearly 80 years after its development, the standard methodfor obtaining reaction rate constants from the theoretical point of view. 14 Provided geometries, energies, frequencies of reactants,transitionstates(TS)andproducts,we canmakeuseof thistheory to compute the thermal rate constant of the respective system inan elegant, accurate, reliable and simple fashion. 15 In this work  we managed to obtain the aforementioned quantities for NF 3 ,NF 3 F, NF 2 , F 2 , N 2 F, NF, N 2  and N 2 F by performing accurateelectronic structure calculations with GAUSSIAN09 16 package with different basis sets and levels of theory. We determined the systems TS by applying full (all electronsincluded in the correlation calculation) second order Møller-Plesset (MP2) level of theory with the 6-31G(d) and cc-pVDZ basis sets. Asfor the other species—reactants and products—we also made useofextendedbasissetsathigherlevelsoftheory.Twosetsofenergies were obtained for these reactions. By starting from MP2/cc-pVDZoptimized geometries, the first set was determined using theaug-cc-pVDZ, cc-pVTZ and aug-cc-pVTZ Dunning basis sets at MP4(SDQ), MP4(SDTQ), QCISD, QCISD(T), CCSD and CCSD(T)levels. The second set was determined starting from MP2/6-31G(d) optimized geometries at the same six levels of theory and the following Pople basis sets: 6-31++G(d,p), 6-311++G(d,p),6-311++G(df,pd) and 6-311++G(3df,3pd). To determine an accuratethermal rate constant, the frequencies are scaled by a factor to takeinto account the known deficiencies of the MP2 level. This factor iscalculated as follows: ScaleFactor ¼ P N i  ¼ 1 w i  ð exp : Þ = w i  ð theor : Þ N   ;  (1) where  N   is the number of frequencies,  w i  (exp.), and  w i  (theor.)are the experimental and theoretical vibrational frequencies. It obtained scale factors of 0.977884 and 0.962846 for the cc-pVDZand 6-31G(d) basis sets, respectively.In the scope of the TST formalism, the thermal rate constant of a general bimolecular reaction such as A + BC - X  ‡ - C + ABis given by: 15,17–19 k TST ¼ k B T hQ X z Q A Q BC exp   V  Ga RT     (2) where  Q X  ‡ ,  Q  A  , and  Q BC  are the partition functions of transitionstate (TS) and reactants A and BC, respectively.  k  B  is theBoltzmann constant;  h  is Planck’s constant;  T   is the tempera-ture;  R  is the universal gas constant and  V  Ga  is the potentialbarrier given by  V  Ga  =  V  MEP  +  e ZPE . (3)In this expression  e ZPE  is the harmonic zero-point energy (ZPE)and  V  MEP  is the Eckart classical potential energy  20 point mea-sured from the overall zero energy of the reactants: V  MEP  ¼  ay 1 þ  y þ  by ð 1 þ  y Þ 2 ;  (4) where  y  = e a ( s – s 0 ) ,  a  =  D  H  00  =  V  G ‡ a  ( s  = +inf)    V  G ‡ a  ( s  =   inf), b  = (2 V  G ‡ a    a ) + 2( V  G ‡ a  ( V  G ‡ a    a )) 1/2 ,  s 0  ¼ 1 a ln  a þ bb  a   , a 2 ¼  m o z ð Þ 2 b 2 V  z V  z  a ð Þ , and  m  is the reduced mass.  a  and  b  dependon the reactants ( V  G ‡ a  ( s  =   inf)), products ( V  G ‡ a  ( s  = +inf)) andTS ( V  G ‡ a  ) energies, as well as the imaginary frequency on theTS ( o ‡ ), 21,22 and  y  is related to the reaction coordinate. Paper NJC    P  u   b   l   i  s   h  e   d  o  n   0   2   A  u  g  u  s   t   2   0   1   3 .   D  o  w  n   l  o  a   d  e   d   b  y   C  e  n   t  r  o   F  e   d  e  r  a   l   d  e   E   d  u  c  a  c  a  o   T  e  c  n  o   l  o  g   i  c  a   d  e   G  o   i  a  s  o  n   1   2   /   0   3   /   2   0   1   4   1   3  :   1   0  :   1   3 . View Article Online  3246  NewJ.Chem.,  2013,  37 , 3244--3251  This journal is  c  The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013 In order to take tunneling effects along the reaction coordinateinto account, we introduce the transmission coefficient   k  W  ( T  ),thus obtaining   k   =  k  W  ( T  ) k  TST ( T  ), where  k  W  ( T  ) stands for Wignercorrection. The Wigner correction ( k  W  ( T  )) for tunneling assumes aparabolic potential for the nuclear motion near the TS. Thepotential used takes the form: 17,23 k W ð T  Þ¼ 1 þ  124  h o z k B T   2 :  (5)The characteristics of the MEP and the TRC, with the Wignertunneling corrections, are determined using our own code, which is described in the literature. 18,19,24 The TRC are thenexpressed in the Arrhenius form as  k ð T  Þ¼ AT  N  exp   E  a RT    , where  A  is the pre-exponential factor,  N   is the temperaturepower factor, and  E  a  is the activation energy. III Results and discussions  As we consider a TST approach in this work, we first need toanalyze the values of geometries, energies and frequencies of reactants, products and transition state species. Table 1 ismeant to deal with the first of these tasks for reactants andproducts. We performed geometry calculations at MP2/cc-pVDZand MP2/6-31G(d) levels. The table lists a summary of calcu-lated inter-atomic distances and bond angles for each leveltogether with a standard value obtained in the literature (whenavailable) for comparison purposes. The good agreement achieved for both bond distances and bond angles for all thementioned calculations is readily seen and can be quantified by calculating the maximum absolute error for each case which was found to be no larger than 0.04 Å and around 1.0 1 ,respectively. Also, the theoretical references available inTable 1 provide an efficient criterion when deciding which values, or in other words, which basis set, is to be used in theTST formalism.The accuracy obtained for the geometrical parameters allow us to rely on the levels of theory and basis sets used for thepresent calculations and proceed in the task of obtaining thequantities required by TST with the frequency calculation forreactants and products, summarized in Table 2. Comparing todata present in the literature, our calculations presented smallabsolute error values, which is an indication of their accuracy. As we would expect, it is noted that a tendency of the higherlevel of theory results in calculated frequency values nearer tothe experimental ones. This fact also indicates the consistency of our calculations. An interesting feature present in the last column of Table 2 is the zero point energy (ZPE) correction.The idea is to express all the calculated energies referring tothe same energy level, and the correction should be made as Table 1  Geometrical parameters for reactants and products of the unimolecular,abstractionandexchangereactionscalculatedatMP2/cc-pVDZandMP2/6-31G(d)levels Species BasesInteratomic distances (Å) Bond angles ( 1 )  R NF  R FF  R NN  A FNF  A NNF F 2  cc-pVDZ — 1.424 — — —6-31G(d) — 1.421 — — —Exp. — 1.412 25,26 — — —— 1.435 27 — — —Theor. — 1.421 28 — — —N 2  cc-pVDZ — — 1.129 — —6-31G(d) — — 1.130 — —Exp. — — 1.094 27 , 1.098 26 — —Theor. — — 1.131 28 — —NF cc-pVDZ 1.317 — — — —6-31G(d) 1.330 — — — —Exp. 1.317 26,29 — — — —Theor. 1.330 28 — — — —N 2 F cc-pVDZ 1.579 — 1.318 — 65.36-31G(d) 1.578 — 1.321 — 65.2NF 2  cc-pVDZ 1.349 — — 103.7 —6-31G(d) 1.359 — — 103.3 —Exp. 1.370 25,30 — — 104.2 25,30 —Theor. 1.359 28 — — 103.3 28 —NF 3  cc-pVDZ 1.377 — — 102.0 —6-31G(d) 1.385 — — 101.7 —Exp. 1.371 25 — — 102.9 25 —Theor. 1.385 31 — — 101.7 31 —1.380 28 — — 101.7 28 — Table 2  Harmonic vibrational frequencies (cm  1 ) and zero-point energy(kcal mol  1 ) for reactants and products of the unimolecular, abstraction andexchange reactions at MP2/cc-pVDZ and MP2/6-31G(d) levels Species Bases  n  1  n  2  n  3  n  4  e ZPE F 2  cc-pVDZ 933.4 — — — 1.3056-31G(d) 1007.8 — — — 1.387Exp. 892.0 27,32 — — — —916.6 26 — — — —Theor. 957.0 28 — — — —N 2  cc-pVDZ 2176.2 — — — 3.0426-31G(d) 2178.7 — — — 2.999Exp. 2359.6 27,32 — — — —2358.6 26 — — — —Theor. 2118.7 28 — — — —NF cc-pVDZ 1175.3 — — — 1.6436-31G(d) 1192.0 — — — 1.641Exp. 1115.0 25 — — — —1141.4 26,29 — — — —Theor. 1138.5 29 — — — —1104.6 28 N 2 F cc-pVDZ 814.7 939.3 1382.0 — —6-31G(d) 862.0 932.5 1386.8 — —NF 2  cc-pVDZ 586.0 982.8 1126.6 — 3.7686-31G(d) 574.0 1026.1 1147.0 — 3.781Exp. 573.0 25 931.0 25 1074.0 25 — —573.4 30 930.7 30 1069.5 30 — —Theor. 569.7 28 908.1 28 1078.6 28 — —NF 3  cc-pVDZ 496.8 656.1 915.8 1043.8 6.3266-31G(d) 489.3 653.7 959.9 1062.0 6.351Exp. 492.0 25 642.0 25 906.0 25 1032.0 25 —497.0 33 648.0 33 898.0 33 1027.0 33 —Theor. 489.3 31 653.7 31 959.3 31 1061.6 31 —484.6 28 644.2 28 860.6 28 1032.8 28 — NJC Paper     P  u   b   l   i  s   h  e   d  o  n   0   2   A  u  g  u  s   t   2   0   1   3 .   D  o  w  n   l  o  a   d  e   d   b  y   C  e  n   t  r  o   F  e   d  e  r  a   l   d  e   E   d  u  c  a  c  a  o   T  e  c  n  o   l  o  g   i  c  a   d  e   G  o   i  a  s  o  n   1   2   /   0   3   /   2   0   1   4   1   3  :   1   0  :   1   3 . View Article Online  This journal is  c  The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013  NewJ. Chem.,  2013,  37 , 3244--3251  3247 e ZPEcorr  =  e ZPEprod    e ZPEreact  , whenever the minimum of thepotential energy curve is not the minimum energy of the vibrational ground state. We make use of this definitionthroughout the present study.Table 3 presents the TS optimized geometries, harmonic vibrationalfrequencies,andzero-pointenergyfortheabstractionof NF 3  + F and N 2  + F systems (TS a  and TS d , respectively) and also forthe exchange channel of NF 3  + F (TS b ) and the unimolecularreaction N 2 F = N 2  + F (TS c ). The notation is clarified in Fig. 1, which represents the transition structures for all the studiedsystems. The imaginary frequency   n  i   is also reported. FromTable 3, one can see that our results are in good agreement withthe experimental and theoretical data available in the literature. We present the reactant and product formation enthalpy  with ZPE correction for the species considered in this work inTable 4. We also add theoretical and experimental referencesfor comparison with our results. One can readily note a widerange of results for the formation enthalpies reported in theliterature for the different reactions. This is due to the different levels of theory and basis sets employed in each case. We cansee that the results systematically present good agreement withboth experimental and theoretical data. The complete set of  ab initio  properties, total electronic energies, forward barrier,reverse barrier, and calculated and experimental  D  H  0  values at 298 K are available in the ESI.† All these properties werecalculated using several basis sets (cc-pVDZ, aug-cc-pVDZ,cc-pVTZ, aug-cc-pVTZ, 6-31G(d), 6-31++G(d,p), 6-311++G(d,p),6-311++G(df,pd) and 6-311++G(3df,3pd)) and at MP2, MP4(SDQ),MP4(SDTQ), QCISD, QCISD(T), CCSD, CCSD(T) levels of theory forreactant, product and TS for abstraction (NF 3  + F = NF 2  + F 2  andNF + N = N 2  + F), exchange (NF 3  + F = NF 3  + F) and unimolecular(N 2 F = N 2  + F) reactions. The final TRCs for all reactions werecalculated using the MP4/cc-pVTZ level. This level was chosenby comparing the  ab initio  and experimental  D f   H  0  values forunimolecular, abstraction and exchange reactions and concluding that it yielded the smallest difference between these data. With all the static properties needed in a TST treatment available, the remainder of this section is divided into twosubsections, to accomplish the TRC calculation for each one of the desired systems. In order to gather a qualitative feeling of how each reaction takes place it is worth first studying thebehavior of the intrinsic reaction coordinate of each system. After that, we present the minimum energy path (MEP) resultsfor the reactive systems and finally discuss the obtained TRCsfor both reactions. We chose to explicitly present and discusshere our main results of TRC plots together with the discussionof the main features of IRC and MEP plots, that we left in theESI† in favor of succinctness. Table 3  Transition state geometrical parameters (inter-atomic distances in Å and bond angles in degrees), harmonic vibrational frequencies (cm  1 ), and zero-pointenergy (kcal mol  1 ) calculated at MP2/cc-pVDZ and MP2/6-31G(d) levels forNF 3  + F = NF 2  + F 2  (TS a ), NF 3  + F = NF 3  + F (TS b ), N 2 F = N 2  + F (TS c ) and NF + N = N 2  + F (TS d )reactions. The imaginary frequency ( n  i ) was also reported NF 3  + F = NF 2  + F 2  (TS a ) NF 3  + F=NF 3  + F (TS b ) N 2 F = N 2  + F (TS c ) NF + N = N 2  + F (TS d )cc-pVDZ 6-31G(d) cc-pVDZ 6-31G(d) cc-pVDZ 6-31G(d) cc-pVDZ 6-31G(d)  R NF  1.347 1.355 1.338 1.350 1.985 2.006 1.329 1.340  R NF 0  2.607 2.353 1.507 1.499 — — — —  R NF 00  — — 1.507 1.499 — — — —  R F 00 F 0 0 0  1.586 1.661 — — — — — —  R N 0 F 0  — — — — 1.440 1.437 — —  R NN 0  — — — — 1.270 1.266 1.971 1.983  A FNF 0  103.76 103.37 105.67 105.31 — — — —  A F 0 NF 00  107.08 101.17 100.13 99.75 — — — —  A NF 0 F 00  173.70 173.70 — — — — — —  A F 00 NF 0  — — 146.14 147.58 — — — —  A N 0 NF  — — — — 93.98 95.60 — —  A NF 0 N 0  — — — — — — 100.95 100.14 n  1  25.42 39.91 300.40 302.32 268.85 260.54 1166.18 1174.32 n  2  28.55 45.06 402.23 400.94 1116.03 1140.73 1478.56 1493.54 n  3  55.97 80.48 453.69 452.34 — — — — n  4  77.76 116.18 580.54 590.03 — — — — n  5  122.98 170.68 651.05 644.47 — — — — n  6  587.89 578.12 738.13 733.50 — — — — n  7  987.05 1031.38 1037.16 1071.56 — — — — n  8  1129.01 1148.61 1055.63 1100.62 — — — — n  i   570.73 i   399.00 i   1441.55 i   1309.92 i   471.45 i   462.76 i   509.36 i   539.34 i  e ZPE  4.214 4.419 7.296 7.345 1.936 1.929 3.698 3.672 Fig. 1  Schematic representation of transition structures for (a) abstraction (TS a ),(b) exchange (TS b ), (c) unimolecular (TS c ), and (d) abstraction (TS d ) reactions. Paper NJC    P  u   b   l   i  s   h  e   d  o  n   0   2   A  u  g  u  s   t   2   0   1   3 .   D  o  w  n   l  o  a   d  e   d   b  y   C  e  n   t  r  o   F  e   d  e  r  a   l   d  e   E   d  u  c  a  c  a  o   T  e  c  n  o   l  o  g   i  c  a   d  e   G  o   i  a  s  o  n   1   2   /   0   3   /   2   0   1   4   1   3  :   1   0  :   1   3 . View Article Online  3248  NewJ.Chem.,  2013,  37 , 3244--3251  This journal is  c  The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013  A. NF 3  + F system The channels considered in the NF 3  + F system are those of abstraction and exchange. As already mentioned, the IRCs of allreactive systems are placed in the ESI,† and the abstractionchannel of this set is represented in Fig. 1. We can note thebreaking of a F atom from the NF 3  molecule expressed by thecontinuous deviation of the upper line from the others. Thesubsequent formation of the F 2  molecule is manifested in theconvergence of the remaining two lines. The exchange channelpresented in Fig. S2 of the ESI† on the other hand is representedby a characteristic symmetrical IRC plot, consistent with theinterchange of two indistinguishable fluorine atoms: one fromthe molecule and the other from the atomic species. We then analyze the minimum energy path for both reactions,figures that are also presented in the ESI.† Fig. 3 consists of theMEP representation for the abstraction channel. We present twoplots in the figure:  V  MEP  associated with the standard representa-tion of the minimum energy path, and V  aG , which consists of   V  MEP added to  e ZPE  in each region. We obtained an exoergecity of 21.8863 kcal mol  1 and the experimentally expected absence of complexes, reactants and products. The minimum energy pathof the exchange channel presented in Fig. S4 of the ESI† shows aresult in strikingly good accordance with the expected lack of energy gain or loss from reactants to products. The reasoning should be obvious from the indistinctness of the fluorine speciesmanifested in Fig. 2. Another important feature is the relatively lower barrier this channel presents when compared to the formerreaction. Conventional and Wigner-corrected TRCs are plottedagainst reciprocal temperature for the abstraction channel inFig. 2 of this work. We can see that in the high temperature regime (3000–4000 K),the two curves are almost coincident. This is due to the low importance tunneling has in this regime, since thermal excitationprevails in promoting the reaction.In the low temperature regime, tunneling effects were found tobe of intermediate importance. In this case, the coincidencebetweenWigner-correctedandconventionalTRCisquiteinteresting and can be attributed to the influence of the harmonic potential. Also, the intermediate importance of tunneling effects is consistent  with the intermediate value of 52.76 1  for the skew angle obtained.These features are more readily observed when describing theobtained TRC in Arrhenius form as: k  TST = 2.8692    10 12 T  1.7636 e  56982/  RT  k  TST W   = 1.6373    10 12 T  1.828 e  56699/  RT  , where  R  = 1.9859 kcal  1 mol  1 is the universal gas constant. It is important to note that in this work, we expressed all theactivation energy quantities (  E  a ) in the Arrhenius fitting in unitsof cal mol  1 so as to better present our results. The valuesobtained for the coefficients provide a useful way to compareour results with others from the literature. The main result obtained for this system is expressed in the inset of Fig. 2. Fromthis figure and Table 5, one can see that our conventional and Table 4  Reactant and product formation enthalpy (kcal mol  1 ) with ZPE correction for all species involved in the unimolecular, abstraction and exchange reactions Species This work Experimental references Theoretical referencesF 2  0.403 0 25,34,35 0.3 a , 0.3 b , 0.688 i  , 1.288   j  0.980 c , 0.056 d  , 0.686 e  2.027 c , 0.927 d  ,   2.142 e N 2  1.786 0 25,34,35 1.3 b , 2.004 i  , 1.994   j   2.027 c , 0.927 d  ,   2.142 e NF 53.757 55.688    0.72, 34 55.6    0.5 36 54.9 a , 54   f   , 53.9  g  , 56.18 h 59.501    7.89 25 46.698 c , 54.635 d  , 51.903 e N 2 F 146.344 — —NF 2  7.234 10.7    1.91 25 , 8.8    1.20 34 6.6 a , 8   f   , 8.5  g  , 8.67 h 8    1, 32 8.3    0.5 36  3.411 c , 5.845 d  , 5.852 e NF 3   31.600   30.20    0.27 25,34  33.8 b ,   26.5  g  ,   30.2   f    41.310 c ,   35.271 d  ,   32.321 ea  At G2 level. 37  b  At G2 level. 38  c  At B3LYP/6-311++(3df,3pd) level. 28  d   At G2 level. 28  e  At G3 level. 28  f   For G3 level. 39  g   At BAC-MP4(SDTQ) level. 40 h  At CCSD(T) level. 41  i  Estimated from  D   f    H  0 (298) at G3 level. 42,43  j  Estimated from  D   f    H  0 (298) at G3 level. 42,43 Fig. 2  Conventional ( k  TST ), and Wigner ( k  TST k  TSTW  ) plots of thermal rate constantin the temperature range of 200–4000 K for the abstraction channel NF 3  + F =NF 2  + F 2 . NJC Paper     P  u   b   l   i  s   h  e   d  o  n   0   2   A  u  g  u  s   t   2   0   1   3 .   D  o  w  n   l  o  a   d  e   d   b  y   C  e  n   t  r  o   F  e   d  e  r  a   l   d  e   E   d  u  c  a  c  a  o   T  e  c  n  o   l  o  g   i  c  a   d  e   G  o   i  a  s  o  n   1   2   /   0   3   /   2   0   1   4   1   3  :   1   0  :   1   3 . View Article Online
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x