Infrared identification of the -complex of Cl-C6H6 in the reaction of chlorine atom and
benzene in solid para-hydrogen
Mohammed Bahou, Henryk Witek, and Yuan-Pern Lee
Citation: The Journal of Chemical Physics 138, 074310 (2013); doi: 10.1063/1.4790860 View online: http://dx.doi.org/10.1063/1.4790860
View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/138/7?ver=pdfcov Published by the AIP Publishing
Articles you may be interested in
Reactions between atomic chlorine and pyridine in solid para-hydrogen: Infrared spectrum of the 1-chloropyridinyl (C5H5NCl) radical
J. Chem. Phys. 138, 054307 (2013); 10.1063/1.4789407
Infrared absorption of trans-1-chloromethylallyl and trans-1-methylallyl radicals produced in photochemical reactions of trans-1,3-butadiene and C2 in solid para-hydrogen
J. Chem. Phys. 137, 084310 (2012); 10.1063/1.4745075
Reactions between chlorine atom and acetylene in solid para-hydrogen: Infrared spectrum of the 1-chloroethyl radical
J. Chem. Phys. 135, 174302 (2011); 10.1063/1.3653988
Comparing the dynamical effects of symmetric and antisymmetric stretch excitation of methane in the Cl + CH 4 reaction
J. Chem. Phys. 120, 5096 (2004); 10.1063/1.1647533
Differential cross section polarization moments: Location of the D-atom transfer in the transition-state region for the reactions Cl+C 2 D 6 DCl (v =0,J =1)+ C 2 D 5 and Cl+CD 4 DCl (v =0,J =1)+ CD 3
Infrared identification of the
σ-complex of Cl-C
6H
6in the reaction
of chlorine atom and benzene in solid
para
-hydrogen
Mohammed Bahou,1Henryk Witek,1and Yuan-Pern Lee1,2,a)
1Department of Applied Chemistry and Institute of Molecular Science, National Chiao Tung University,
1001 Ta-Hsueh Road, Hsinchu 30010, Taiwan
2Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan
(Received 9 December 2012; accepted 25 January 2013; published online 19 February 2013)
The reaction of a chlorine atom with benzene (C6H6) is important in organic chemistry, especially
in site-selective chlorination reactions, but its product has been a subject of debate for five decades. Previous experimental and theoretical studies provide no concrete conclusion on whether the product is a π - or σ -form of the Cl-C6H6complex. We took advantage of the diminished cage effect of
para-hydrogen (p-H2) to produce Cl in situ to react with C6H6(or C6D6) upon photolysis of a Cl2/C6H6
(or C6D6)/p-H2matrix at 3.2 K. The infrared spectrum, showing intense lines at 1430.5, 833.6, 719.8,
617.0, and 577.4 cm−1, and several weaker ones for Cl-C6H6, and the deuterium shifts of observed
new lines unambiguously indicate that the product is a 6chlorocyclohexadienyl radical, i.e., the σ -complex of Cl-C6H6. Observation of the σ -complex rather than the π -complex indicates that the
σ-complex is more stable in solid p-H2 at 3.2 K. The spectral information is crucial for further
investigations of the Cl+ C6H6 reaction either in the gaseous or solution phase. © 2013 American
Institute of Physics. [http://dx.doi.org/10.1063/1.4790860]
I. INTRODUCTION
Rapid development of advanced experimental techniques and computational methods enables verification and detailed reinterpretation of many chemical reaction mechanisms, in-cluding identification and structural characterization of their transient intermediates1–3 and explicit demonstration of real-istic and not-so-fully-intuitive reaction pathways.4–7This velopment provides scientists a novel methodology for de-tailed investigations of fundamental organic reactions, which often seem to be a victim of oversimplified interpretations in organic chemistry textbooks. Chlorination reactions of aro-matic compounds constitute such a class of organic reactions, mechanism of which is not yet fully understood. In the rical mechanism, it is believed that the initial step is the ad-dition of a chlorine atom (Cl) to benzene (C6H6) to form a
Cl-C6H6 complex.8 However, it has been highly contentious
for more than 50 years whether the Cl-C6H6complex is a π
-or σ -complex, i.e., whether the reaction between Cl and C6H6
proceeds through a π - or σ -bonding.
In the 1950s, Russell and Brown postulated that Cl and C6H6 form a hexahapto (η6) π -complex (Figure1(a)) to
ac-count for the enhanced selectivity of Cl with respect to the ter-tiary C−H bond of 2,3-dimethylbutane in organic photochlo-rination carried out in the presence of benzene.9–12A detailed study performed by Skell et al. under a broad range of experi-mental conditions suggested that the highly selective interme-diate might be rather assigned to the 6-chlorocyclohexadienyl radical (the Cl-C6H6 σ-complex, Figure 1(c)), whereas the
low selective intermediates are best described as a mixture
a)Author to whom correspondence should be addressed. Electronic mail:
yplee@mail.nctu.edu.tw.
of Cl atoms and the Cl-C6H6π-complex.13,14Ingold and
co-workers employed laser flash photolysis to investigate the ki-netics of these reactions in solutions and reported that the se-lectivity can be quantitatively described by a scheme which involves hydrogen abstraction by just two species, free Cl atoms and the π -complex.15–18
Theoretical characterization of the Cl-C6H6 complexes
proved to be a complex task. Tsao et al. employed various ab
initio computational methods and density functional theory
(DFT) to find that the η6 π-complex and the η2 π-complex
(Figure1(b)) have imaginary vibrational frequencies, hence, should be considered as transition states.19 The stabilization energies of ∼23 and ∼16 kJ mol−1 were predicted for the
η1π-complex (Figure1(d)) and the σ -complex, respectively,
with the BH&HLYP/6-311++G** method and the barrier height for the η1 π-complex→ σ-complex conversion was
estimated as ∼9 kJ mol−1. The predicted values are within 5 kJ of those predicted with the CASPT2 method; the multi-reference character of η1 π-complex makes the MP4(SDQ),
CBS-QBH&H, and CCSD(T) results unreliable.19 The pro-nounced stability of the η1 π-complex of Cl-C6H6 was not
confirmed by the study of Croft and Howard-Jones, who em-ployed new density functionals designed particularly for non-covalent complexes and reported that the σ -complex is more stable than the η1 π-complex, with a stabilization energy of
28–32 and 35–43 kJ mol−1 predicted for the η1 π-complex
and the σ -complex, respectively, with the MPW1K method using various basis sets.20 In view of these results, it is dif-ficult to conclude unequivocally which of the Cl-C6H6
com-plexes is the most stable.
The nature of the bonding between Cl and benzene in these two types of complexes is distinctly different de-spite of their similar energetics. Graphical explanation of the
074310-2 Bahou, Witek, and Lee J. Chem. Phys. 138, 074310 (2013)
FIG. 1. Geometries of (a) η6π-complex, (b) η2π-complex, (c) σ -complex,
and (d) η1 π-complex predicted with the MPW1PW91/6-311++G(2d,2p)
method. Bond distances are given in Å, and angles in degrees. For (c) and (d), a graphical representation of the bonding mechanism is also shown in a form of the bonding orbital.
bonding mechanism in the σ - and π -complexes of Cl-C6H6—
represented in the form of bonding orbital between the Cl and C6H6moieties—is given in Figure1(c)and1(d). In the
π-bonding scheme, the bonding is more delocalized and in-volves predominantly the bonding π orbitals associating with three carbon atoms of the benzene ring. In the σ -bonding scheme, Cl forms a σ -bond with one of the carbon atoms and consequently disrupts its aromaticity. As a result, the H atom on the chlorine-bonded carbon moves out of the molec-ular plane of benzene and the Cl−C distance is shorter. The difference in structure and bonding should result in distinct spectra of these complexes.
Spectral evidence for the existence of the π -complex rests on a broad transient absorption band near 490 nm and an intense broad band near 320 nm observed in pulse radioly-sis of benzene in CCl4solvent by Bühler.21,22They suggested
that in the π -complex benzene acts as an electron donor and the chlorine atom as an electron acceptor; this feature resem-bles spectra of several other aromatic charge-transfer com-plexes. Bunce et al. also indicated that the 490 nm band, observed in their laser flash photolysis experiments, is un-likely due to a σ -compound, as usually σ -complexes absorb only weakly in the visible range.15In contrast, Bensen favors strongly the identification of this transient feature as belong-ing to the σ -complex.23The electron paramagnetic resonance (EPR) spectrum of the Cl-C6H6 complex observed in
irradi-ated frozen Cl2/C6H6 solid at 77 K indicates that this
com-plex is most probably a distorted σ -comcom-plex.24 Clearly, the available experimental (UV and EPR) and theoretical results do not permit for unambiguous assignment of the structure of the Cl-C6H6complex.
Infrared (IR) spectroscopy can provide more definitive information about molecular structures. Spectral
characteri-zation of distinct structural motifs for complexes of simple anions with arenes has been demonstrated clearly with IR spectroscopy.25Similarly, IR spectroscopy is expected to pro-vide definitive structural information for this Cl-C6H6
com-plex. If the σ -complex were formed, the characteristic C−H stretching mode below 3000 cm−1 and the ClCH bending mode near 1100 cm−1 for the newly formed ClCH moiety could be observed in the recorded IR spectrum, whereas if the π -complex were formed, the characteristic ring-stretching modes of the C6H6moiety near 1600 and 950 cm−1would be
activated due to perturbation caused by the Cl atom.
Matrix isolation technique has proven to be an excellent method for investigating many unstable intermediates. Unfor-tunately, conventional matrix isolation techniques cannot be readily applied for studying reactions involving a single chlo-rine atom due to the fact that the in situ photolysis of the most common precursor, Cl2, produces two Cl atoms, which cannot
easily migrate out of the original matrix cage. Consequently, the reaction products typically contain two chlorine atoms.26 An attractive alternative matrix host, which can be used to avoid this problem, is p-H2. This quantum solid has emerged
as a unique host for matrix isolation spectroscopy, with many excellent properties such as extremely narrow spectral width and a diminished matrix cage effect.27–29 We have demon-strated that free radicals that are difficult to be produced us-ing conventional inert-gas matrices can be readily produced in a p-H2matrix via photolysis in situ30or photo-induced
bi-molecular reactions.31,32 Here we took advantage of the di-minished cage effect of p-H2to produce chlorine atoms from
in situ photolysis of Cl2in a Cl2/C6H6/p-H2matrix at 3.2 K;
subsequent reaction between Cl and C6H6 produces the
Cl-C6H6 complex. As we demonstrate below, the observed IR
spectra and deuterium isotopic shifts clearly indicate that the Cl-C6H6produced in these experiments is a σ -complex.
II. EXPERIMENTS
The experimental setup has been described previously.26,31 In these experiments, a gold-plated cop-per plate served both as a cold substrate for the matrix sample and as a mirror to reflect the incident IR beam to the detector. The copper plate was maintained at 3.2 K with a closed-cycle helium refrigerator (Janis, SHI-415). Typically, gaseous mixtures of C6H6/p-H2 (1/750, 0.02 mol) and Cl2/p-H2
(1/750, 0.02 mol) were co-deposited over a period of 5 h. The 365 nm light from a light-emitting diode (375 mW) was employed to dissociate Cl2 to produce Cl atoms. IR
absorption spectra were recorded with a Fourier-transform infrared (FTIR) spectrometer (Bomem, DA8) equipped with a KBr beamsplitter and a Hg/Cd/Te detector (cooled to 77 K) to cover the spectral range 500–4000 cm−1. Typically, 400 scans at a resolution of 0.25 cm−1 were recorded at each stage of the experiment. It has been previously reported that infrared excitation of the solid p-H2bands (4000–5000 cm−1)
can induce Cl atoms to react with p-H2 to form HCl,33 and
therefore, when Cl atoms were present in the p-H2 matrix, a
2.4 μm infrared cutoff filter (Andover Corp.) was used when recording the infrared spectra.
Flow rates of C6H6/p-H2 and Cl2/p-H2 are typically
0.08–0.11 and 0.055–0.12 mmol h−1, respectively. C6H6
(99.5%, Fluka), C6D6 (listed isotopic purity 99.5%,
Cam-bridge Isotope Laboratories, Inc), and Cl2(99.9%, Air
Prod-ucts and Chemicals) were used without further purification. H2 (99.9999%, Scott Specialty Gases) was used after
pas-sage through a trap at 77 K before conversion to p-H2. The
p-H2 converter comprised a copper cell filled with hydrous
iron (III) oxide catalyst (Aldrich) and cooled with a closed-cycle refrigerator (Advanced Research Systems, DE204AF). The efficiency of conversion is controlled by the tempera-ture of the catalyst; at 11.5 K, the concentration of o-H2 is
<10 ppm.
III. RESULTS AND DISCUSSION
A. Cl+ C6H6reaction
The partial IR spectrum of a deposited sample of Cl2/C6H6/p-H2(1/1/1500) at 3.2 K (Figure2(a)) exhibits
ab-sorption lines of C6H6 and also complexes of C6H6-Cl2
(in-dicated with *). C6H6-Cl2 has been observed in solid N2.34
Spectra covering a wider spectral range and a comparison of the experimental and theoretical frequencies and relative intensities of C6H6-Cl2 are available in the supplementary
material.35 Upon irradiation with light at 365 nm for 5 h, the intensities of IR lines due to C6H6 and C6H6-Cl2complexes
decreased slightly, and several new features appeared. A dif-ference spectrum obtained by subtracting the spectrum of the deposited matrix from the spectrum recorded after irradiation at 365 nm is presented in Figure2(b); lines pointing upwards indicate production and those pointing downwards indicate destruction. Annealing of the irradiated matrix at 5 K for 2 min enhanced some of these new features significantly and further diminished the absorption lines of C6H6, as shown in
the difference spectrum in Figure2(c). Secondary photolysis
FIG. 2. (a) IR absorption spectra of a Cl2/C6H6/p-H2(1/1/1500) matrix
sam-ple after deposition at 3.2 K for 5 h. (b) Difference spectra after irradiation of the matrix with a LED at 365 nm for 1 h. (c) After annealing the sample at 5 K for 2 min. (d) After secondary photolysis at 3.2 K in the range 455– 700 nm from a Hg lamp for 1 h. Lines of the C6H6-Cl2complex are
indi-cated with asterisks (*). New lines attributable to Cl-C6H6are indicated with
arrows.
of this matrix with light in the 455–700 nm region diminished these features further and enabled us to identify clearly lines associated with a single species, as indicated with arrows in Figure 2(d). Intense features at 577.4, 617.0, 719.8, 833.6, and 1430.5 cm−1 and weaker ones at 876.8, 956.0, 1008.0, 1026.4, 1179.0, 1406.5, and 1509.4 cm−1 are clearly identi-fied. Common origin of these features is further confirmed by comparison of spectra recorded at various stages of the exper-iment and in several different experexper-iments.
The identification of the molecular species responsible for the observed additional IR signals is not a difficult puzzle. Upon irradiation of Cl2 at 365 nm, Cl atoms were produced
and subsequently reacted with C6H6, as indicated by the
de-creased absorption of C6H6and C6H6-Cl2. The new features
thus produced are expected to originate from the products of the Cl+ C6H6reaction. Further evidence comes from the
en-hancement of these features after annealing, which can be ex-plained by higher mobility of the Cl atoms at 5 K and lack of photolysis. At low temperatures and without irradiation, the hydrogen abstraction cannot occur because of a large barrier associated with this reaction channel. The only viable reaction channels lead hence to the formation of the Cl-C6H6
com-plexes. The secondary photolysis in the 455–700 nm region might dissociate some σ -complex of Cl-C6H6, which absorbs
in the visible region.14
The IR spectra of each specific Cl-C6H6 complex
sim-ulated with various theoretical models yield similar spectral patterns; the largest discrepancies concern only some small shifts in the positions of the bands. We present here only calculation results from the MPW1PW91/6-311++G(2d,2p) method; this method was chosen because it was designed particularly for non-covalent complexes.20 A comparison of the new spectral features observed in the experiment for the Cl-C6H6 complex (Figure 3(a)) with the stick IR spectra
(a) (b)
(c)
FIG. 3. (a) Difference IR absorption spectra of a Cl2/C6H6/p-H2(1/1/1500)
matrix sample after secondary photolysis in the range 455–700 nm from a Hg lamp at 3.2 K for 1 h; the matrix was deposited at 3.2 K for 5 h, followed by irradiation with a LED at 365 nm for 1 h and annealing the sample at 5 K for 2 min. Lines of C6H6are removed for clarity. IR spectra of the σ -complex (b)
and the η1π-complex (c) of Cl-C6H6simulated according to the anharmonic
vibrational frequencies and IR intensities predicted with the MPW1PW91/6-311++G(2d,2p) method. New lines attributable to Cl-C6H6 are indicated
074310-4 Bahou, Witek, and Lee J. Chem. Phys. 138, 074310 (2013)
of the σ -complex (Figure 3(b)) and the η1 π-complex
(Figure 3(c)), simulated according to anharmonic vibra-tional frequencies and IR intensities computed with the MPW1PW91/6-311++G(2d,2p) method, suggests that the best agreement in terms of relative intensities and positions is obtained for the predicted spectrum of the σ -complex. The
η1π-complex is predicted to have intense lines at 659 (44), 725 (88), 887 (13), 1186 (12), 1479 (22), 1468 (21), and 1578 (100) cm−1, whereas the σ -complex is predicted to have in-tense lines at 582 (13), 633 (22), 734 (100), 872 (7), and 1440 (19) cm−1; numbers in parentheses give relative IR in-tensities. Intense lines at 577.4 (19), 617.0 (26), 719.8 (100), 833.6 (11), and 1430.5 (25) cm−1 observed in this work are consistent with those predicted for the σ -complex. A list of wavenumbers and relative intensities for the observed lines and their comparison with calculations is given in Table I;
TABLE I. Comparison of experimentally observed line positions (cm−1) and intensities with the computed MPW1PW91/6-311++G(2d,2p) anhar-monic vibrational wavenumbers (cm−1) and IR intensities of the σ and π -complexes of Cl-C6H6.
Modea Sym. η
1π-complex σ-complex Experiment
ν1 a 3104 (7)b 3123 (3)b ν2 a 3089 (3) 3090 (7) 3070.9 (7.6)c ν3 a 3111 (10) 3067 (1) 3054.3 (0.5) ν4 a 3072 (0) 2999 (1) 2967.2 (2.3) ν5 a 1578 (100) 1591 (1) ν6 a 1468 (21) 1440 (19) 1430.5 (25) ν7 a 1186 (12) 1185 (2) 1179.0 (0.2) ν8 a 1025 (4) 1094 (4) 1112.5 (1.4) ν9 a 1019 (2) 1005 (3) 1008.0 (2.9) ν10 a 1009 (1) 998 (2) 993.5 (3.7) ν11 a 999 (2) 996 (2) 983.0 (1.4) ν12 a 949 (27) 891 (6) 876.8 (4.8) ν13 a 887 (13) 872 (7) 833.6 (11) ν14 a 725 (88) 734 (100) 719.8 (100) ν15 a 659 (44) 633 (22) 617.0 (26) ν16 a 602 (5) 582 (13) 577.4 (19) ν17 a 376 (4) 391 (19) ν18 a 94 (1) 224 (72) ν19 a 92 (4) 91 (0) ν20 a 3106 (9) 3076 (9) 3083.9 (10) ν21 a 3083 (4) 3079 (5) 3063.4 (7.9) ν22 a 1562 (0) 1523 (3) 1509.4 (4.0) ν23 a 1479 (22) 1425 (6) 1406.5 (2.9) ν24 a 1354 (0) 1344 (0) ν25 a 1332 (8) 1309 (0) ν26 a 1168 (1) 1162 (0) ν27 a 1166 (1) 1133 (1) 1118.5 (2.5) ν28 a 1036 (5) 1021 (5) 1026.4 (1.2) ν29 a 989 (0) 975 (0) 956.0 (3.0) ν30 a 828 (1) 779 (1) ν31 a 598 (0) 586 (1) ν32 a 389 (0) 404 (0) ν33 a 101 (1) 202 (1)
aMode numbers follow the order of the wavenumbers predicted for the σ -complex of Cl-C6H6.
bRelative IR intensities normalized to the most intense line. IR intensities for the most intense lines of η1π-complex and σ -complex are 71.2 and 100.3 km mol−1, respec-tively.
cIntegrated IR intensities relative to ν14are listed in parentheses.
nearly all lines of the σ -complex with computed IR inten-sity greater than 2 km mol−1 are observed in experiment. Although weak, the C−H stretching and the ClCH bending modes of the C(Cl)H moiety were observed at 2967.2 and 1112.5 cm−1, respectively, similar to the predicted values of 2999 and 1094 cm−1. In contrast, the intense ring-stretching mode predicted near 1578 cm−1 for the π -complex was un-observed.
We observed larger bandwidths for lines at 577.4, 617.0, 719.8, 833.6, and 876.8 cm−1 that are associated with vibra-tional modes involving significant change of the Cl−C dis-tance. The observed broadening might be because the increase in the Cl−C distance upon vibrational excitation leads to dis-sociation of the complex or because of the flat potential en-ergy surface along the C−Cl stretching coordinate that allows a variety of structures with slightly different vibrational fre-quencies for these modes. Considering the stabilization en-ergy of 19–43 kJ mol−1 predicted for the Cl−C6H6complex
and the low temperature, the latter explanation is more likely. Previous reports suggested that in the π -complex ben-zene acts as an electron donor and the chlorine atom as an electron acceptor. A transient broad absorption band near 490 nm observed in pulse radiolysis of benzene in CCl4
solvent by Bühler resembles spectra of several other aro-matic charge-transfer complexes.21,22Our calculations do not confirm the assumed sharp dichotomy between the charge-transfer characters of the σ - and π -complexes; electrostatic potentials computed for both complexes look very much alike, as shown in the supplementary material.35
Theoretical calculations using the BH&HLYP/6-311++ G**(5D) time-dependent density functional theory (TD-DFT) method predicted one intense band ( f = 0.208) near 456 nm and one medium band ( f = 0.050) near 335 nm for the π -complex, and one weak band
(a) (b)
(c)
FIG. 4. (a) Inverted difference IR absorption spectra of a Cl2/C6D6
/p-H2 (1/1/2000) matrix sample after secondary photolysis in the range 455–
700 nm from a Hg lamp at 3.2 K for 1 h; the matrix was deposited at 3.2 K for 5 h, followed by irradiation with a LED at 365 nm for 1 h and annealing the sample at 5 K for 2 min. (b) Lines of C6H6are removed for clarity. IR spectra
of the σ -complex. (c) The η1π-complex of Cl-C6D6simulated according to
the anharmonic vibrational frequencies and IR intensities predicted with the MPW1PW91/6-311++G(2d,2p) method. New lines attributable to Cl-C6H6
( f = 0.003) near 416 nm and a medium band ( f = 0.016) near 296 nm for the σ -complex. Our calculations using the TDDFT/MPW1PW91/6-311++G(2d,2p) method yield bands near 404 nm ( f= 0.118) and 307 nm ( f = 0.058) for the
π-complex and 478 nm ( f= 0.003) and 305 nm ( f = 0.047) for the σ -complex. Hence, the σ -complex is likely photolyzed with light in the region 455–700 nm so that we observed the decrease in intensity for lines of the σ -complex.
B. Cl+ C6D6reaction
A similar experiment was performed with Cl2/C6D6
/p-H2(1/1/1200). We show in Figure4(a)the inverted difference
spectrum of the matrix after 1 h of secondary irradiation with the 455–700 nm light; the matrix was initially irradiated with light at 365 nm for 1 h, followed by annealing at 5 K for 2 min
TABLE II. Comparison of experimentally observed line positions (cm−1) and intensities with the computed MPW1PW91/6-311++g(2d,2p) anhar-monic vibrational wavenumbers (cm−1) and IR intensities of the σ and π -complexes of Cl-C6D6.
Modea Sym. η
1π-complex σ-complex Experiment
ν1 a 2322 (5)b 2314 (3)b 2336.0 (3.8)c ν2 a 2294 (3) 2305 (14) 2308.6 (8.4) ν3 a 2297 (8) 2267 (2) 2255.3 (1.8)? ν4 a 2276 (0) 2224 (2) 2197.4 (9.3) ν5 a 1543 (100) 1521 (2) 1523.3 (8.0) ν6 a 1305 (11) 1263 (29) 1247.1 (46) ν7 a 868 (4) 855 (5) 849.4 (5.5) ν8 a 813 (4) 827 (27) 828.0 (25) ν9 a 966 (0) 903 (25) 892.3 (32) ν10 a 935 (4) 951 (5) 963.0 (2.5) ν11 a 872 (0) 828 (3) ν12 a 767 (8) 777 (2) 761.7 (1.3) ν13 a 672 (6) 700 (29) 683.6 (34) ν14 a 507 (50) 624 (100) 622.6 (100) ν15 a 596 (9) 480 (56) 474.1 (59) ν16 a 573 (10) 554 (53) 550.5 (51) ν17 a 333 (4) 341 (49) ν18 a 93 (1) 219 (159) ν19 a 88 (3) 84 (2) ν20 a 2305 (7) 2285 (17) 2293.2 (1.7) ν21 a 2294 (1) 2280 (5) 2269.3 (1.0) ν22 a 1497 (1) 1448 (0) 1433.8 (0.8) ν23 a 1343 (15) 1323 (8) 1308.4 (4.6) ν24 a 1060 (0) 1055 (0) ν25 a 1310 (3) 1257 (0) ν26 a 865 (1) 833 (0) ν27 a 834 (0) 853 (2) ν28 a 819 (4) 828 (8) 809.1 (19) ν29 a 802 (1) 781 (2) ν30 a 648 (1) 615 (2) ν31 a 571 (0) 562 (2) ν32 a 338 (0) 351 (0) ν33 a 89 (1) 187 (2)
aMode numbers follow the order of the wavenumbers predicted for the σ -complex of Cl-C6D6.
bRelative IR intensities normalized to the most intense line in our detection range. IR intensities for these lines of η1π-complex and σ -complex are 79.5 and 40.6 km mol−1, respectively.
cIntegrated IR intensities relative to ν14are listed in parentheses.
before this step. New features at 474.1, 550.5, 622.6, 683.6, 1247.1, 1308.4, 1433.8, and 1523.3 cm−1 are indicated with arrows. The spectrum is compared with the IR spectra of Cl-C6D6for the σ -complex (Figure4(b)) and the η1π-complex
(Figure 4(c)) simulated according to anharmonic vibra-tional frequencies and IR intensities predicted with the MPW1PW91/6-311++G(2d,2p) method. A comparison of the experimental and theoretical frequencies and relative in-tensities is given in TableII. The excellent agreement between the experimental and theoretical spectra for the σ -complex further supports our assignments.
IV. CONCLUSION
In summary, the work presented in this report clearly demonstrates that the product of the reaction of the Cl atom with the C6H6 molecule in solid p-H2 is the σ -complex
(6-chlorocyclohexadienyl radical) rather than the π -complex. This observation indicates that the σ -complex isolated in solid
p-H2at 3.2 K is likely lower in energy than the π -complex.
Although we expect that the matrix effect in solid p-H2 is
small, it is unclear if our conclusion on the relative stability of the complexes can be extended to the reaction of Cl+ C6H6in
solutions. Nevertheless, the IR absorption spectrum reported in this work provides definitive structural characterization of the Cl-C6H6σ-complex as the 6-chlorocyclohexadienyl
radi-cal, and is crucial for further investigations of the Cl+ C6H6
reaction either in the gaseous or solution phase.
ACKNOWLEDGMENTS
National Science Council of Taiwan (Grant Nos. NSC100-2745-M009-001-ASP and NSC99-2113-M-009-011-MY3) and the Ministry of Education, Taiwan (“Aim for the Top University Plan” of National Chiao Tung University) supported this work. The National Center for High-Performance Computing provided computer time.
1K. Suma, Y. Sumiyoshi, and Y. Endo,Science308, 1885 (2005). 2O. Asvany, P. Kumar P, B. Redlich, I. Hegemann, S. Schlemmer, and D.
Marx,Science309, 1219 (2005).
3S. Narra, S.-W. Chang, H. A. Witek, and S. Shigeto,Chem.-Eur. J.18, 2543
(2012).
4A. K. Mollner, S. Valluvadasan, L. Feng, M. K. Sprague, M. Okumura, D.
B. Milligan, W. J. Bloss, S. P. Sander, P. T. Martien, R. A. Harley, A. B. McCoy, and W. P. L. Carter,Science330, 646 (2010).
5P. R. Schreiner, H. P. Reisenauer, D. Ley, D. Gerbig, C.-H. Wu, and W. D.
Allen,Science332, 1300 (2011).
6M. P. Grubb, M. L. Warter, H. Xiao, S. Maeda, K. Morokuma, and S. W.
North,Science335, 1075 (2012).
7D. Townsend, S. A. Lahankar, S. K. Lee, S. D. Chambreau, A. G. Suits,
X. Zhang, J. Rheinecker, L. B. Harding, and J. M. Bowman,Science306, 1158 (2004).
8J. M. Tanko and N. K. Suleman, in Energetics of Organic Free Radicals,
edited by J. A. M. Simões, J. F. Liebman, and A. Greenberg (Chapman and Hall, New York, 1996), Chap. VIII.
9G. A. Russell and H. C. Brown,J. Am. Chem. Soc.77, 4031 (1955). 10G. A. Russell,J. Am. Chem. Soc.79, 2977 (1957).
11G. A. Russell,J. Am. Chem. Soc.80, 4987 (1958). 12G. A. Russell,J. Am. Chem. Soc.80, 4997 (1958).
13P. S. Skell, H. N. Baxter III, and C. K. Taylor,J. Am. Chem. Soc.105, 120
(1983).
14P. S. Skell, H. N. Baxter III, J. M. Tanko, and V. Chebolu,J. Am. Chem. Soc.108, 6300 (1986).
074310-6 Bahou, Witek, and Lee J. Chem. Phys. 138, 074310 (2013)
15N. J. Bunce, K. U. Ingold, J. P. Landers, J. Lusztyk, and J. C. Scaiano,J. Am. Chem. Soc.107, 5464 (1985).
16K. U. Ingold, J. Lusztyk, and K. D. Raner,Acc. Chem. Res.23, 219 (1990). 17K. D. Rander, J. Lusztyk, and K. U. Ingold, J. Phys. Chem.93, 564
(1989).
18K. D. Rander, J. Lusztyk, and K. U. Ingold,J. Am. Chem. Soc.110, 3519
(1988).
19M.-L. Tsao, C. M. Hadad, and M. S. Platz,J. Am. Chem. Soc.125, 8390
(2003).
20A. K. Croft and H. M. Howard-Jones,Phys. Chem. Chem. Phys.9, 5649
(2007).
21R. E. Bühler and M. Ebert,Nature (London)214, 1220 (1967). 22R. E. Bühler,Helv. Chim. Acta51, 1558 (1968).
23S. W. Benson,J. Am. Chem. Soc.115, 6969 (1993).
24G. B. Sergeev, A. V. Pukhovskii, and V. V. Smirov, Russ. J. Phys. Chem.
57, 589 (1983).
25B. Chiavarino, M. E. Crestoni, S. Fornarini, F. Lanucara, J. Lemaire, P.
Maître, and D. Scuderi,Chem.-Eur. J.15, 8185 (2009).
26C.-W. Huang, Y.-C. Lee, and Y.-P. Lee,J. Chem. Phys.132, 164303 (2010). 27T. Oka,Annu. Rev. Phys. Chem.44, 299 (1993).
28T. Momose and T. Shida,Bull. Chem. Soc. Jpn.71, 1 (1998). 29K. Yoshioka and D. T. Anderson,J. Chem. Phys.119, 4731 (2003). 30M. Bahou and Y.-P. Lee,J. Chem. Phys.133, 164316 (2010). 31J. Amicangelo and Y.-P. Lee,J. Phys. Chem. Lett.1, 2956 (2010). 32Y.-F. Lee and Y.-P. Lee,J. Chem. Phys.134, 124314 (2011).
33P. L. Raston and D. T. Anderson,Phys. Chem. Chem. Phys.8, 3124 (2006). 34L. Fredin and B. Nelander,Mol. Phys.27, 885 (1974).
35See supplementary material at http://dx.doi.org/10.1063/1.4790860 for
schematic representation of the electrostatic potential of the Cl-C6H6
com-plexes, experimental spectra covering a wider spectral range, and a compar-ison of the experimental and theoretical frequencies and relative intensities of C6H6and C6H6-Cl2.
