Hydrogen Abstraction Reaction Mechanisms and Rate Constants for Isoflurane with a Cl Atom at 200～2000 K: A Theoretical Investigation①

REN Hong-Jing YANG Xio-Hui LI Xio-Jun LI Jing-To YANG Ju-Xing

?

Hydrogen Abstraction Reaction Mechanisms and Rate Constants for Isoflurane with a Cl Atom at 200～2000 K: A Theoretical Investigation①

REN Hong-Jianga, b②YANG Xiao-HuiaLI Xiao-JunaLI Jiang-TaoaYANG Ju-Xianga

a(710065)b(100875)

The kinetics and mechanisms of H abstraction reaction between isoflurane and a Cl atom have been investigated using DFT and G3(MP2) methods of theory. The geometrical structures of all species were optimized by the wB97XD/6-311++G** method. Intrinsic reaction coordinate (IRC) analysis has been carried out for the reaction channels. Thermochemistry data have been obtained by utilizing the high accurate model chemistry method G3(MP2) combined with the standard statistical thermodynamic calculations. Gibbs free energies were used for reaction channels analysis. Two channels were obtained, which correspond to P(1) and P(2). The rate constants for the two channels over a wide temperature range of 200～2000 K were also obtained. The results show that the barriers of P(1) and P(2)reaction channels are 50.36 and 50.34 kJ/mol, respectively, predicting that it exists two competitive channels. The calculated rate constant is in good agreement with the experiment value. Additionally, the results also show that the rate constants also increase from 1.85×10-16to 2.16×10-12cm3·molecule-1·s-1from 200 to 2000K.

isoflurane, Cl atom, reaction mechanisms, rate constants,, G3(MP2);

1 INTRODUCTION

Partially halogenated ethers have been widely used both as potential replacements for chlorofluo- rocarbons[1](CFCs) in a number of applications and as used anaesthetics[2, 3]. The inhalation anaesthetics isoflurane (2-chloro-2-(difluoromethoxy)-1,1,1-tri-fluoroethane, CF3CHClOCHF2) is one of the halo- genated methyl ethers family, which is clinically administered as synthetic racemic mixtures[4, 5]. It is also widely used for the medical settings worldwide for induction and maintenance of general anesthesia and is a successful second generation halogenated anaesthetic, too[4]. It possesses the desirable pro- perties of good anaesthetic syndrome, nonflam- mability, negligible metabolism, and fast recovery time[6], which is fairly well understood in terms of reactivity and atmospheric fate[7].

However, it is reported by Ref. [8] that the rest ones of all the clinically used volatile anesthetics can be released into the atmosphere circumstances. The report from Ref. [9] also shows that the global emission of isoflurane is likely in the range of several ktons/year based on consumption figures from a large US hospital. The increasing atmos- pheric concentrations of these halogenated com- pounds are partially responsible for the global climate changing. The presence of these compounds has also resulted in the extensive concerns for the increasing risk to human health[10]. And for the moment, it is an extremely immediate problem how to degrade these compounds including isoflurane, which brings us a new challenge. So, it is very important for us to have a good understanding of the atmospheric degradation mechanisms of the com- pound isoflurane.

Due to the higher reactivity of a chlorine atom towards most organic compounds[11],a Cl atom may be also very significant to the degradation of isoflurane. Furthermore, it is known that a Cl atom is used widely in the laboratory to mimic the OH radical for the investigation of oxidation mecha- nisms of organic compounds[12, 13]. Moreover, we have known that a Cl atom can effectively degrade the anaesthetic sevoflurane in the Ref. [14]. Obviously, the reaction mechanism investigation of isoflurane with a Cl atom is extremely essential, and the obtained results might be available for understanding the atmospheric chemistry property, atmospheric lifetime and degradation process.

So far, many investigations were reported focu- sing on the configurations analysis, thermodynamic stabilities and reaction processes for isoflurane in the experiments[15, 16]and theoretical calcula- tions[17-20]. In these reports, Beach et al.[15]deter- mined the rate constant for the reaction of a Cl atom with the isoflurane using discharge-flow techniques and resonance fluorescence method at 318, 362 and 399 K. Andersen.[16]used the LP/LIF technique to determine the rate coefficient of atmospheric H abstraction reaction mechanisms and the overall rate coefficient for the degradation reaction of isoflurane with a Cl atom.

In the theoretical aspects, Biedermann et al.[17]investigated the six conformers and their VCD spectra using hybrid density functional theory B3LYP and GIAO basis functions at the 6- 311++G** level of theory. Polavarapu et al.[18]reported the absolute configuration of isoflurane for understanding the stereospecific interactions at the molecular level. Henin et al.[19]have constructed the atomistic model of isoflurane and described its conformational preferences and intermolecular interactions, which is compatible with the CHARMM force field for biomolecules at the B3LYP method. Flyunt et al.[20]have carried out the theoretical investigations on the kinetics reaction of halogenated hydroperoxides, peroxyl and alkoxyl radicals with isoflurane in the aqueous solution in the presence and absence of methionine (MetS) for evaluating the reactive fates.

Although there are previous reports on the H abstraction reaction for isoflurane with a Cl atom in the experiments aspects. To the best of our knowledge, there are no detailed theoretical inves- tigations for the titled reactions, especially using a proper method for Cl atom and weak interactions for the complexes including hydrogen bonds. Thus, this work will bring out a careful theoretical investi- gation onthe H abstraction reactions between isoflurane and a Cl atom, especially using a proper method for Cl atom and diffusion functions for the complexes.

2 COMPUTATIONAL METHOD

All the geometrical structures of reactants, transi- tion states, products and radicals were optimized using density functional theory DFT-wB97XD method[21, 22]combined with the 6-311++G** basis set[23, 24]. The wB97XD method can exhibit better behavior when used to describe cases of strong inhomogeneity and strong non-dynamic correlation because it has involved occupied Kohn-Sham orbitals as additional non-local variables, which are evasive for the global hybrids at the GGA and meta-GGA levels of the theory.This functional is also fourth-rung functional.

And the high accurate model chemistry method G3(MP2)[25, 26]was adopted in the energy calcula- tions. Many reports[27-29]have shown that the DFT-wB97XD functional method expresses excee- dingly well in the calculations of structures and stabilization energies of the chlorine atom com- plexes. We have also carried out vibrational fre- quency calculations to confirm the stationary spe- cies and saddle point properties at the same theory level. The intrinsic reaction coordinates (IRC) for the forward and reverse directions beginning from the TS locations were also searched to ascertain the correct connections with stable species[30-33].

In the energy calculations of G3(MP2), this must be illustrated that the G3(MP2)//wB97XD method is a composite process based on three single-point energy calculations, including QCISD(T)/6-31G*, MP2/6-31G* and MP2/G3MPT large. Thecorrections are obtained at the wB97XD/6- 311++G** level of theory. The standard enthalpies, entropies and Gibbs free energies of all species over the temperature range of 200～2000 K were achieved throughout applying the standard statistical thermodynamic calculations of the self-composed program provided by Su Ke-He group. In calcula- ting the Cl atom electronic partition function, in view of spin orbital coupling resulting to energy splitting, the splitting energy is chosen as 881 cm-1. For obtaining the tunneling coefficients due to the H abstraction reaction, Eckart tunneling correction method has been used employing Zhangrate 1.0 program[34]. All the computations were carried out with GAUSSIAN 09 set of programs[35].The NBO analysis was carried out at the wB97XD/6- 311++G** level of theory using version 3.1 of NBO package[36]included in the GAUSSIAN 09 program package because it can characterize hydrogen bonding and reveal the origin ofbond strength changes in either type of H-bonding for the com- plexes in this work.

3 RESULTES AND DISCUSSION

The optimized two isoflurane conformer struc- tures, geometrical parameters and X-ray crystal experimental data from Ref. [37] are shown in Fig. 1.The reactant complexes, transition states, product complexes structures as well as some related bond lengths and bond angles are given in Fig. 2. The reaction potential energy profiles for P(1) and P(2)as well as the relative Gibbs free energies of all species are depicted in Fig. 3. The profile of the isomerization reaction for the two conformers ISO-1 and ISO-2 is shown in Fig. 4. Zero point energies, relative thermodynamic energies, relative Gibbs free energies and relative enthalpies for all species in the titled reactions are listed in Table 1. Thermodynamic energies, entropies, Gibbs free energiesand enthal- piesat 298.15 K are shown in Supplement 1. The NBO analysis values are given in Table 2. The activation Gibbs free energies, the calculated rate constants and the overall rate constants over a wide temperature range of 200～2000 K as well as1/(1+2)includingEckart quantum tunneling effects are listed in Table 3. The tunneling correction coef- ficients estimated with Eckart tunneling correction method and rate constants without tunne- ling correction at 200～2000 K for P(1)and P(2) channels are given in Supplement 2.

3. 1 Structures and energies

As can be seen from Fig. 1, ISO-1 and ISO-2 are the conformers linked throughout a transition state TS2 with an imaginary vibrational frequency of 76.1 cm-1. The calculated results have shown thatthe two conformers have1point group symmetry, in which the chemical surrounding of the H12 atom is obviously different.

C(1)-C(2)=1.536(1.536)[1.536]{1.591} C(2)-O(3)=1.393(1.380) [1.379]{1.402} O(3)-C(4)=1.365(1.376) [1.381] {1.372}

C(1)-F(5)=1.337(1.335) [1.336] C(1)-F(6)=1.334(1.334) [1.335] C(1)-F(7)=1.329(1.327) [1.328]

C(2)-Cl(8)=1.783(1.799) [1.796] C(2)-H(9)=1.090(1.091) [1.089] C(4)-F(10)=1.351(1.356) [1.349]

C(4)-F(11)=1.354(1.330) [1.341] C(4)-H(12)=1.088(1.092) [1.090] C(2)O(3)C(4)=117.2(116.4) [120.5]{116.2}

C(1)C(2)O(3)=107.1(106.7) [106.7] {106.0} C(1)C(2)O(3)C(4)=137.3(165.2) [152.8]

H(12)C(4)O(3)C(2)=178.6(60.3) [118.0] O(3)C(2)C(1)F(6)= -179.4(-179.4) [0.87]

Fig. 1. Structures and parameters of two isoflurane conformersISO-1 and ISO-2 as well as transition state TS2 obtained at the wB97XD/6-311++G** level of theory. The bond lengths are given in angstrom (?) and the bond angles in degree (°). The dada in parentheses are given for ISO-2, those in square brackets for TS2 and the ones in braces from X-ray crystal experiments. Other data are given for ISO-1

Fig. 2. Geometrical structures and the parameters of all isoflurane complexes and transition statesobtained at the wB97XD/6-311++G** level of theory. The bond lengths are given in angstrom (?) and the bond angles in degree (°)

Fig. 3. Scheme and the relative energy (in kJ·mol-1) profiles of the two channels P(1) and P(2) for the hydrogen abstraction reactions between isoflurane (CF3CHClOCHF2) and a Cl atom. Energies are the Gibbs free energy of 298.15 K obtained at the G3(MP2)//wB97XD/6-311++G** level of theory

Fig. 4. Scheme and the relative energy (in kJ·mol-1) profiles of the isomerization reaction between the two conformers ISO-1 and ISO-2. Energies are the Gibbs free energy of 298.15 K obtained at the G3(MP2)//wB97XD/6-311++G** level of theory

Table 1. Zero Point Energy (ZPE, a.u.) Obtained at the wB97XD/6-311++G** Level of Theory, Relative Thermodynamic Energy (ΔU0K,kJ·mol-1) at Both wB97XD/6-311++G** and G3(MP2) Level, Relative Gibbs Free Energy (ΔGθ298.15k,kJ·mol-1) and Relative Enthalpy (ΔHθ298.15k,kJ·mol-1) for all the Species of the Titled Reactions

aThe relative Gibbs free energy of all the complexes are based on the summary of isolated ISO-1 and Cl atom energies.

bThe relative energies of ISO-2 and TS2 are based on that of the radical ISO-1.

cThe relative energies of the radical IM2 are based on that of the radical IM1.

Table 2. NBO Analysis Data Calculated at the wB97XD/6-311++G** Level of Theory

Table 3. Activation Gibbs Free Energies (kJ·mol-1), Calculated Rate Constants k (cm3·molecule-1·s-1) of the Two Channels P(1) and P(2) and the Overall Reaction Rate Constant Over the Wide Range of Temperature 200~2000 K as well as k1/(k1+k2)Including Eckart Quantum Tunneling Effects and the Reported Experiment Data at 298K.

In comparison of the corresponding bond lengths and bond angles calculated in this work with the reported ones of Ref.[17], it is found that ISO-1 is the most stable conformer with three dihedral angles C(1)–C(2)–O(3)–C(4), H(12)–C(4)–O(3)–C(2) and O(3)–C(2)–C(1)–F(6) of 137.3°, 178.6° and –179.4°, and these angles are almost the same with those of Ref.[17], implying that the chosen structure ISO-1 is the most possible one existing in the atmosphere. In Fig. 1, ISO-2 is a conformer of ISO-1 which can isomerize to ISO-2an O(3)–C(4) single bond rotation. ISO-2 will be found in the P(2) channel.

As can be seen from Table 1, the relative Gibbs free energies of the two reaction complexes R1 and R2 are 7.87 and 16.09 kJ·mol-1as a zero of the total energy of isolated ISO-1 and a Cl atom at the G3(MP2) level, which indicates the obvious entropy decreasing effects.

As can be seen from Fig. 2, in complex R1, the Cl(13)···H(9) distance between ISO-1 and a Cl atom is 2.916 ?. In complex R2, the Cl(13)···H(9) and Cl(13)…H(12) bond distances are 3.073 and 2.901 ?, respectively. As also can be seen from Table 3, in R1 the second order perturbation lpCl(13)→*C(2)–H(9) interaction is 1.17 kJ·mol-1and in R2 and the lpCl(13)→*C(4)–H(12) interaction is 3.60 kJ·mol-1, implying that there are two hydrogen bonds and the H-bonding interaction in R2 is stronger than that in R1, which is according with the H-bond distance. As can be seen from Table 1, the relative Gibbs free energy is –1.59 kJ·mol-1for P1, however, P2 has higher energy than P1. The transition state TS1 and TS3 structures are shown in Fig. 2 and they have only imaginary vibrational frequency (1118.5and 783.6for TS1 and TS3). The angles Cl(13)–H(9)?C(2) in TS1 and Cl(13)?H(9)?C(4) in TS3 are 174.9° and 172.1°, respectively, showing that these two bonds are almost linear during the H abstraction reaction. In TS1, the angles H(9)?C(2)?C(1), H(9)?C(2)?O(3) and H(9)?C(2)?Cl(8) are 103.0°, 112.4° and 103.9°, indicating the existence of tetrahedral character. In TS3, this character also exists in the angles of H(12)?C(4)?F(10), H(12)?C(4)?O(3) and H(12)? C(4)?F(11), which is helpful for the H dissociation.

3.2 Reaction channels

The potential energy profiles are shown in Fig. 3 for the two reaction channelsP(1)and P(2). The channel P(1) is marked with little dots and it corresponds to the abstraction of the H(9) atom, and another channel P(2) is marked with short lines corresponding to the abstraction of atomH(12). TS1 possesses a larger imaginary frequency, 1118.5, which indicates that the width of the potential barrier might be narrower and the tunneling effect may be more important in the calculation of the rate constant. This prediction will be testified by the next calculated tunneling coefficientsin Supplement 2. The reaction mechanisms can be decomposed as a three-step process. In the first channel P(1), the hydrogen atom H(9) is abstracted and the process is below.

CF3CHClOCHF2+ Cl·CF3CHClOCHF2···Cl· (1)

CF3CHClOCHF2···Cl·TS1[CF3CClOCHF2]···HCl (2)

[CF3CClOCHF2]···HCl[CF3CClOCHF2]· + HCl (3)

To our interest, when identifying another reaction channel for the H(12) abstraction, a transition state TS3 is found and it is composed by ISO-2 and a Cl atom. ISO-2 can be isomerized by ISO-1, that is to say, the H(12) abstraction reaction can not directly occurISO-1 and a Cl atom. Before H(12) was abstracted, an additional isomerization reaction could proceed. Thus, Fig. 4 shows the profile of the isomerization reaction from ISO-1 to ISO-2. After carrying out IRC analysis, TS2 was linked to ISO-1 and ISO-2, and TS3 was linked to R2 and P2. In R2, a Cl atom attacked the H(12) atom of the conformer ISO-2, and ISO-2 can be producedTS2 from ISO-1 only requiring the activation energy of 15.34 kJ·mol-1. The low activation Gibbs free energy can make this isomerization reaction occur at room temperature. Therefore, the second channel P(2) is listed below, where the hydrogen atom H(12) is abstracted.

CF3CHClOCHF2+ Cl·CF3CHClOCHF2···Cl· (4)

CF3CHClOCHF2···Cl·TS3[CF3CHClOCF2] ···HCl (5)

[CF3CHClOCF2] ···HCl[CF3CHClOCF2]· + HCl (6)

Each step of the whole process studied will be discussed in detail in the following contents.

3. 2. 1 Formation of isoflurane··· Cl complex

As can be seen from Fig. 3, the obtained results on the potential energy surface (PES) have shown the presence of two stationary complexes R1 and R2. Firstly, R1 and R2 are produced accompanied with the increase of Gibbs free energies (G3(MP2)) by 7.87 and 15.11 kJ·mol-1at 298.15K, respectively. However, seen from Table 1, the relative thermo- dynamic energies of the two complexes are –13.96 and –12.23 kJ·mol-1at 0 K at the same level, respec- tively. Obviously, after the addition of a Cl atom, the stability of the complexes can be improved and the relative Gibbs free energies of the complexes are increased mainly coming from an obvious entropy effect decrease. The absolute value of the Gibbs free energy of R2 is larger than that of R1 by 8.22 kJ·mol-1, respectively.

As can be seen from Fig. 2, the Cl atom in com- plex R1 interacts with the H(9) atom of isoflurane. The Cl(13)···H(9) bond distance is 2.916?. Com- paring R1 withthe isolated ISO-1,both their C(2)? O(3) and C(2)?Cl(8) distances are dramatically lengthened by0.008and 0.015 ?, respectively; However, other bond distances are not nearly chan- ged, implying the large effects on isoflurane mole- cular structuresuffered from the attacking of a Cl atom. The weak interaction between the Cl(13) and H(9) atoms can be confirmed by the decrease of the C(2)?H(9) stretching frequency by only 4.88 cm-1.

3. 2. 2 Formation of isoflurane radical and HCl complex

In the second step, two isoflurane radicals and HCl molecule complexes are produced throughout two transition states TS1 and TS3, corresponding to the P(1) and P(2) channels. In TS1 and TS3, the C–H bond is elongated and the hydrogen atom H(9) or H(12) shifts to the Cl atom, and at last, aHCl molecule is formed. As can be seen from Fig. 3, the activation Gibbs energy barriers≠are 50.36 and 50.34 kJ·mol-1for P(1) andP(2), respectively, and it is the rate-determining reaction step of all. The structures of the transition states were carefully diagnosed by the analysis of the vibrational frequen- cies. Only one imaginary frequency has been obtained with the 1118.5and 783.59cm-1for TS1 and TS3, respectively. Each of the imaginary frequencies corresponds to the normal mode, which can be described as the vibrational motion of the shifted H atom between the C and Cl atoms.

As can be seen from Fig. 3, a Cl atom firstly attacks the H(9) atom of TS1, which will be abstracted and the C(2)···H(9) bond distance is elongated from 1.091 to 1.336 ? by 0.245 ?. In the transition state TS3, the Cl(13) atom attacks the H(12) atom, which will be also abstracted and the C(4)···H(12) bond distance is elongated by 0.317 ? from 1.093 to 1.410 ?. In TS1, an elongation of the C(2)–H(9) bond accompanies with the shortness of the H(9)–Cl(13) bond by 1.405 ?. In TS3, an elongation of the C(4)–H(12) bond causes the con- traction of the H(12)–Cl(13) bond by1.442 ? and the shortness of the C(4)–F(11) bond by0.043 ?. Following this step, the two product complexes P(1) and P(2) are formed. The calculated barriers are 42.49 and 35.23 kcal·mol-1for channels P(1) and P(2) at the G3(MP2) level of theory, respectively.

It can be seen from Fig. 3 thatin the product complex P1, the H(9)···F(11) bond distance is 2.188 ?, and the F(11) →*Cl(13)?H(9) interaction is 2.47 kJ·mol-1, indicting that the hydrogen bond is formed.In P2, the H(9)···Cl(13) and H(12)···F(5) bond distances are 2.852 and 2.524 ?, respectively, and the second order perturbation lpF(5) →*Cl(13)?H(12) interaction is 0.29 kJ·mol-1and the lpCl(13)→*C(2)?H(9) interaction is 2.13 kJ·mol-1, implying weaker hydrogen bonds appearing between the HCl and isoflurane radical IM2 than P1.

In Table 1, the enthalpy changes in the second step of hydrogen abstraction reactions between isoflurane and HCl at 298.15K are obtained. From Table 1, it reveals that the ΔH298.15kis decreased by 9.15 kJ·mol-1for channelP(1)and increased by 11.58 kJ·mol-1for P(2),showing that the P(1)channel is exothermic and the P(2)channel is endothermic. This result will be proved by the next rate constants.

3. 2. 3 Dissociation of HCl molecule

The last step of the title reaction corresponds to the dissociation of a HCl molecule of the isoflurane product complexes to produce the radicals IM1, IM2 and an isolated HCl molecule. Seen from Fig. 3, the Gibbs free energies are lowered by 22.76 and 18.4 kJ·mol-1for P(1) andP(2) channels in the disso- ciation reactions, respectively, which can mainly result from the increased entropy effects. The thermodynamic energies at 0 K, Gibbs free energies and enthalpies of the two radicals IM1, IM2 and H2O molecule at 298.15K are given in Supplement 1 and the relative energies are also shown in Table 1. According to these values, the Gibbs free energy of IM2 is much lower by 25.55 kJ·mol-1than that of IM1, indicating that the radical IM2 is more stable than IM1 in the gas phase reaction from the view of thermodynamics.

Spin contamination was not an issue for the CF3CHClOCHF2radicals, IM1 and IM2, formed during the progress of reaction channels P(1) and P(2) because <2> for the radicals were found to be 0.7542, 0.7522 and 0.7520 for IM1, IM2 and Cl radicals before annihilation, which is only slightly larger than the expected value of <2>= 0.7500 for doublets.

3. 3 Rate constants

The rate constants of the title reactions are cal- culated using canonical transition state theory(CTST)[38]. It includes a semi-classical one-dimen- sional multiplicative tunneling correction factor given by the following equation:

where,,(),Bandare respectively the symmetry number, the tunneling correction factor at temperature, the Boltzmann constant with the data of 1.380649×10-23J·K-1andthe Planck constant with 6.6260696×10-34J·s. In the equation,0refers to the standard concentration with 2.68719×1019molecules·cm-3at standard temperature and pressure, Δ≠is the activation Gibbs free energies between the transition state and the reactants (the sum of isolated isoflurane and the Cl atom) andis the gas constant with 8.314 J·mol-1·K-1, respectively. Supplement 2 shows the tunneling correction coeffi- cients calculated throughout Eckart tunneling correction method and the rate constants with no tunneling effect correction at 200～2000 K for the P(1) and P(2)channelsZhangrate 1.0 program package[34].

As shown in Table 3 and Fig. 3, the overall Δ≠are 50.36 and 50.34 kJ·mol-1for the P(1) andP(2) channels at 298.15K using dual-level G3(MP2)//wB97XD/6-311++G** method, respec- tively. The calculated rate constants are 1.32×10-15cm3·molecule-1·s-1for the P(1) channel and 7.50×10-16cm3·molecule-1·s-1for the P(2)channelat 298.15 Kequation Eq.1. The reaction rate of P(1) channel is 1.76 times that of the P(2) channel at 298.15K, implying that P(1) is the dominant reaction channel and P(2)is also a competitive reaction channel at room temperature. But P(2) is slightly endergonic and reversible, so the P(1) channel is the only possible H abstraction reaction.

It can also be seen from Supplement 2 that P(1) has a larger tunneling coefficient, which is illustrated with the larger frequency of TS1 than TS2. And we also calculated the1/(1+2) specific value to be 0.638, predicting that P(1) occupies 63.8% possibility for the main reaction channel and P(2) is 36.2%. Hence, we calculated the overall reaction rate constant (1+2) and it is 2.07×10-15cm3·molecule-1·s-1at 298.15 K. To our interest, the experiment data are(3.5±2.2)×10-15cm3·mole- cule-1·s-1at 298 K reported by Beach et al[14]. Andersen and co-workers[15]have recently also determinedthe rate constant of the titled reaction for isofluraneand the Cl atom experimentally, and it equals to (4.5±0.8)×10-15cm3·molecule-1·s-1at 296K. The calculated result in this work is in excellent agreement with the experiment data. Hence, the H(9) abstraction reaction is the main reaction channel for isoflurane hydrogen abstraction reactions in the whole chlorine-atom-initiated degradation reaction process and the H(12) is also a competitive channel having a lower possibility.

Table 3 also gives out activation Gibbs free energy barriers and the rate constants from 200 to 2000 K with a 100 K increment including 298.15K. It can be seen that all the rate constants are increased with the temperature increasing for the P(1) and P(2)channels. It is also found that the Gibbs free energies obviously increase along with the temperature change for the two channels. These results are in good agreement with the report by Beach et al.[15]thatthe rate constants increase with the temperaturechange from 318to 399 K. Overall, the P(1)channel has larger reaction rates and the reaction is easy to occurP(1) at room tem- perature.

3. 4 Comparison of the reactions for isoflurane and sevoflurane with a Cl atom

In the previous report[14], we have carried out the detailed theoretical investigation between sevo- flurane and a Cl atom. In the sevoflurane, the chemistry formula is (CF3)2CHOCH2F, which shows that there are two positions of -CHO- and -CH2F for H abstraction reactions. The calculated barriers are 55.01 and 32.50 kJ·mol-1, which corresponds to H abstraction reactions of -CHO- and -CH2F groups, respectively. However, in this work, the abstracted functional groups for isoflurane (CF3CHClOCHF2) are -CHCl- and -CHF2, which needs almost the same barriers of 50.36 and 50.34 kJ·mol-1, respec- tively. It implies that the Cl withdrawing electronic effect of isoflurane (CF3CHClOCHF2) is similar with a -CF3functional group, leading to similar bond strengths for the C?H bond. In (CF3)2CHOCH2F, the Mulliken atomic charge of C atom in the group -CHO- is ?0.837 e, however, it is ?0.717 e for C atom in the -CHCl group of iso- flurane (CF3CHClOCHF2). The C charge of -CH2F in (CF3)2CHOCH2F is ?0.066 e, but 0.1036 e for the C charge of -CHF2, which mainly results from the difference of withdrawing electronic effects for single F atom and double F atoms. This can demonstrate why the barrier (32.50 kJ·mol-1) of another channel for sevoflurane is much lower by about 18 kJ·mol-1than that of isoflurane. That is to say, the less withdrawing electronic atoms, the lager the probability of the C?H bonds breaking, implying that this inhalation anaesthetics will be degraded more easily.

4 CONCLUSION

The mechanism of hydrogen abstraction reactions between isoflurane (CF3CHClOCHF2) and a Cl atom was investigated at the G3(MP2)//wB97XD/6- 311++G** levelof theory. The geometrical struc- tures of all the species were optimized at the wB97XD/6-311++G** level of theory. Reaction channels were confirmed with the intrinsic reaction coordinates (IRC) analysis. Thermochemistry data were determined with the accurate model chemistry method G3(MP2) as well as the standard statistical thermodynamics. Gibbs free energies were used for the energy analysis of the reaction channels. We find that the barriers of both P(1) and P(2)channels are 50.36 and 50.34 kJ/mol, respectively, and very interestingly, two reaction channels result from two different conformer complexes, which can be easily isomerized at room temperature. However, P(2)is slightly endergonic and reversible, and hence P(1) is the only possible main reaction channel. The calculated rate constant is in good agreement with the experiment value. And we also predict that the rate constants can increase from 1.85×10-16to 2.16×10-12cm3·molecule-1·s-1along the temperature increment from 200 to 2000K. Additionally,the comparison of isoflurane and sevoflurane with Cl atom reactions is included in the end.

(1) Sekiya, A.; Misaki, S. The potential of hydrofluoroethers to replace CFCs, HCFCs and PFCs.. 2000, 101, 215-221.

(2) Terrell, R. C.; Speers, L.; Szur, A. J.; Treadwell, J.; Ucciardi, T. R. General anesthetics. 1. Halogenated methyl ethyl ethers as anesthetic agents.. 1971, 14, 517-519.

(3) Koblin, D. D.; Laster, M. J.; Ionescu, P.; Gong, D.; Eger, E. I.; Halsey, M. J.; Hudlicky, T. Polyhalogenated methyl ethyl ethers: solubilities and anesthetic properties.1999, 88, 1161-1167.

(4) Halpern, D. F.. Elsevier, Amsterdam 1993.

(5) Halpern, D. F.. Manchester 1994.

(6) Young, J. W.; Brandt, S. (Sepracor, Inc.),1992-05-19.

(7) Young, J. W.; Brandt, S. (Sepracor, Inc.),1992-05-19.

(8) Shiraishi, Y.; Ikeda, K. Uptake and biotransformation of sevoflurane in humans: a comparative study of sevoflurane with halothane, enflurane, and isoflurane.1990, 2, 381-386.

(9) Sulbaek Andersen, M. P.; Sander, S. P.; Nielsen, O. J.; Wagner, D. S.; Sanford, T. J.; Wallington,T. J. Inhalation anaesthetics and climate change.. 2010, 105, 760-766.

(10) Ryan, S. M.; Nielsen, C. J.Global warming potential of inhaled anesthetics: application to clinical use.. 2010, 111, 92-98.

(11) Demore, W. B.; Sander, S. P.; Golden, D. M.. JPL Publication, Pasadena, CA 1997.

(12) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P.An FTIR study of the Cl atom-initiated oxidation of CH2Cl2and CH3Cl.. 1980, 12, 1001-1012.

(13) Orlando, J. J.; Tyndall, G. S.; Wallington, T. J. Atmospheric oxidation of CH3Br: chemistry of the CH2BrO radical.. 1996, 100, 7026-7033.

(14) Ren, H.J.; Yang, X.H.; Li, X.J.; Liu, Y.; Wei, X. The hydrogen abstraction reaction mechanism and rate constants from 200 K to 2000 K between sevoflurane and chlorine atom: a theoretical investigation.. 2014, 605-606, 28-34.

(15) Beach, S. D.; Hickson, K. M.; Smith, I. W. M.; Tuckett, R. P. Rate constants and arrhenius parameters for the reactions of OH radicals and Cl atoms with CF3CH2OCHF2, CF3CHClOCHF2and CF3CH2OCClF2, using the discharge-flowresonance fluorescence method.2001, 3, 3064-3069.

(16) Sulbaek Andersen, M. P.; Nielsen, O. J.; Karpichev, B.; Wallington, T. J.; Sander, S. P. Atmospheric chemistry of isoflurane, desflurane, and sevoflurane: kinetics and mechanisms of reactions with chlorine atoms and OH radicals and global warming potentials.2012, 116, 5806-5820.

(17) Biedermann, P. U.; Cheeseman, J. R.; Frisch, M. J.; Schurig, V.; Gutman, I.; Agranat, I. Conformational spaces and absolute configurations of chiral fluorinated inhalation anaesthetics. A theoretical study.1999, 64, 3878-3884.

(18) Polavarapu,P. L.; Cholli,A. L.; Vernice, G. Absolute configuration of isoflurane.1992, 114, 10953-10955.

(19) Henin, J.; Brannigan, G.; Dailey, W. P.; REckenhoff, M. L. An atomistic model for simulations of the general anesthetic isoflurane.2010, 114, 604-612.

(20) Flyunt,R.; Makogon,O.; Naumov,S.; Sch?neich,C.; Asmus,K. D. Reactions of halogenated hydroperoxides and peroxyl and alkoxyl radicals from isofluranein aqueous solution.2007,111, 11618-11625.

(21) Chai, J. D.; Head-Gordon, M. Systematic optimization of long-range corrected hybrid density functionals.2008, 128, 084106-084121.

(22) Chai, J. D.; Head-Gordon, M. Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections.. 2008,10, 6615-6620.

(23) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. Self-consistent molecular orbital methods. XX. A basis set for correlated wave functions.. 1980, 72, 650-654.

(24) Frisch, M. J.; Pople, A. J.; Binkley, J. S. Self-consistent molecular orbital methods 25. Supplementary functions for gaussian basis sets.. 1984, 80, 3265-3269.

(25) Baboul, A. G.; Curtiss, L. A.; Redfern, P. C.; Raghavachari, K. Gaussian-3 theory using density functional geometries and zero-point energies.. 1999, 110, 7650-7657.

(26) Curtiss, L. A.; Redfern, P. C.; Raghavachari, K.; Rassolov, V.; Pople. J. A. Gaussian-3 theory using reduced Moller-Plesset order.. 1999, 110, 4703-4709.

(27) Croft, A. K.; Howard-Jones, H. M. Chlorine-benzene complexes—the reliability of density functionals for non-covalent radical complexes.. 2007, 9, 5649-5655.

(28) Croft, A. K.; Howard-Jones, H. M.; Skates, C. E.; Wood, C. C. Controlling the action of chlorine radical: from lab to environment.. 2011, 21, 7439-7447.

(29) Zierkiewicz. W. Reaction of volatile anaesthetic desflurane with chlorine atom. Theoretical investigation.. 2013, 555, 72-78.

(30) Fukui, K. The path of chemical reactions-the IRC approach.. 1981, 14, 363-368.

(31) Page, M.; McIver, J. W. On evaluating thereaction path Hamiltonian.. 1988, 88, 922-935.

(32) Gonzalez, C.; Schlegel, H. B. An improved algorithm for reaction path following.. 1989, 90, 2154-2161.

(33) Gonzalez, C.; Schlegel, H. B. Reaction path following in mass-weighted internal coordinates.. 1990, 94, 5523-5527.

(34) Zhang, S.W.; Truong, T. N., University of Utah2001.

(35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G.A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A. Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Orokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowsk, J.; Fox, D. J., Revision A.02, Gaussian In., WallingfordCT2009.

(36) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.NBO Version 3.1 Gaussian 09, Revision A.02, Gaussian In., Wallingford CT2009.

(37) Schurig, V.; Jung, M.; Green, B. S.; Horakh, J.; Simon, A. Absolute configuration of the inhalation anesthetics isoflurane and desflurane.. 1996, 35, 1680-1682.

(38) Truhlar, D. G.; Garrett, B. C.; Klippenstein, S. J. Current status of transition-state theory.. 1996, 100, 12771-12800.

①The work was financed by the Natural Science Foundation of Shaanxi Province (2014JM2046，2013JQ2027), the Special Natural Science Foundation of Science and Technology Bureau of Xi’an City Government (CXY1443WL03, CXY1352WL19 and CXY1352WL20), National Science Foundation of China (21303135),and the Industrial research project of Science and Technology Department of Shaanxi Province (2013K09-25). We are also thankful for standard statistical thermodynamics calculation mini-program provided by Su Ke-He groups of School ofNatural and Applied Sciences in Northwestern Polytechnical University

②Corresponding author. Ren Hong-Jiang, born in 1979, associate professor, majoring in physical chemistry. E-mail: hjren@xawl.edu.cn

10.14102/j.cnki.0254-5861.2011-0656

28 January 2015; accepted 2 April 2015