Infrared Spectra of Monohydrogenated Aniline, ortho- And para-HC6H5NH2, Generated in Solid para-Hydrogen

Article abstract of DOI:10.1021/acs.jpca.0c06079

The isomers of monohydrogenated aniline (HC6H5NH2) are regarded as important intermediates in reduction reactions of aniline, but their spectral identification has been limited to electron paramagnetic resonance in an adamantane matrix. We report here infrared (IR) spectra of two least-energy isomers of HC6H5NH2, produced on electron bombardment during the deposition of a matrix of aniline and para-hydrogen at 3.2 K. The intensities of IR lines of HC6H5NH2 increased during maintenance of the electron-bombarded matrix in darkness for a prolonged period because of the neutralization of protonated aniline, H+C6H5NH2, by trapped electrons and further reactions between aniline and the unreacted hydrogen atoms that were produced during electron bombardment. The observed lines were grouped according to their behaviors on secondary photolysis with light at 520, 465, and 375 nm. On comparison of experimental spectra with quantum chemically predicted spectra for four possible isomers of HC6H5NH2, lines in one group were assigned to the most stable ortho-HC6H5NH2 and those in the other group were assigned to the secondmost stable para-HC6H5NH2. Their photolytic behaviors at varied wavelengths are consistent with predicted ultraviolet absorption bands. The mechanisms of formation of these isomers are discussed according to semiquantitative analysis.

Full text of DOI:10.1021/acs.jpca.0c06079

Subscriber access provided by UNIVERSITY OF BRIGHTON
A: Spectroscopy, Molecular Structure, and Quantum Chemistry
Infrared Spectra of Mono-hydrogenated Aniline, ortho-
6
5
2
and para-HCHNH, Generated in Solid para-Hydrogen
Masashi Tsuge, Yu-Hsuan Chen, and Yuan-Pern Lee
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.0c06079 • Publication Date (Web): 18 Aug 2020
Downloaded from pubs.acs.org on August 21, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 39
The Journal of Physical Chemistry
1
2
3
4
5
6
7
Infrared Spectra of Mono-hydrogenated Aniline, ortho- and para-
HC6H5NH2, Generated in Solid para-Hydrogen
8
9
Masashi Tsuge, *,†,‡ Yu-Hsuan Chen,† and Yuan-Pern Lee*,†,§,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
Department of Applied Chemistry and Institute of Molecular Science, National Chiao Tung
University, Hsinchu 300093, Taiwan
‡ Institute of Low Temperature Sciences, Hokkaido University, Sapporo 060-0819, Japan
§ Center for Emergent Functional Matter Science, National Chiao Tung University, Hsinchu 300093,
Taiwan
 Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 106319, Taiwan
* Corresponding authors
E-mail address: tsuge@lowtem.hokudai.ac.jp (M. Tsuge).
E-mail address: yplee@nctu.edu.tw (Y.-P. Lee).
Full postal address: 1001 Ta-Hsueh Road, Hsinchu 300093, Taiwan (Y.-P. Lee).
KEYWORDS
Infrared spectroscopy, matrix isolation, para-hydrogen, hydrogenated aniline
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 2 of 39
1
2
3
4
5
6
7
8
9
ABSTRACT: The isomers of mono-hydrogenated aniline (HC6H5NH2) are regarded as important
intermediates in reduction reactions of aniline, but their spectral identification has been limited to
electron paramagnetic resonance in an adamantane matrix. We report here infrared spectra of two least-
energy isomers of HC6H5NH2, produced on electron bombardment during deposition of a matrix of
aniline and para-hydrogen at 3.2 K. The intensities of IR lines of HC6H5NH2 increased during
maintenance of the electron-bombarded matrix in darkness for a prolonged period because of the
neutralization of protonated aniline, H+C6H5NH2, by trapped electrons and further reactions between
aniline and the unreacted hydrogen atoms that were produced during electron bombardment. The
observed lines were grouped according to their behaviors on secondary photolysis with light at 520,
465, and 375 nm. On comparison of experimental spectra with quantum-chemically predicated spectra
for four possible isomers of HC6H5NH2, lines in one group were assigned to the most stable ortho-
HC6H5NH2 and those in the other group were assigned to the second most stable para-HC6H5NH2.
Their photolytic behaviors at varied wavelengths are consistent with predicted UV-absorption bands.
The mechanisms of formation of these isomers are discussed according to semi-quantitative analysis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
ACS Paragon Plus Environment

Page 3 of 39
The Journal of Physical Chemistry
1
2
3
1. INTRODUCTION
4
5
6
7
8
9
Aniline (C6H5NH2), the simplest amino-substituted benzene, is widely used in chemical synthesis
of rubber materials, dyeing agents, and pharmaceuticals. An important chemical reaction of substituted
benzene (C6H5X) is the Birch reduction,1-3 in which the benzene ring is partially reduced to give
substituted, unconjugated 1,4-cyclohexadiene, in contrast to catalytic reductions that typically generate
substituted cyclohexane. The Birch rule,4 which predicts a kinetic preference for that regioisomers
with the maximum number of substituents on the residual double bonds, has been quite successful. In
the Birch reduction of substituted benzene, a substituted cyclohexadienyl radical (3) appears as an
intermediate (Scheme 1). The preferred site of protonation in the rate-determining second step (2 →
3) has been debated;5-8 even though the final product (5) would be the same, whether the intermediate
(3) is hydrogenated at the ortho- or meta-position with respect to the substituent is unclear. Early
computational studies indicated that the intermediate could be hydrogenated at either the ortho- or the
meta-position.5-7 However, Zimmerman and Wang reported that, according to experiments, the ortho-
position is much preferred for hydrogenation in the Birch reduction of anisole (X = OCH3), toluene (X
= CH3), and other derivatives.8 In situ detection of reaction intermediates will give strong evidence for
the confirmation of mechanism, but spectral information on them (i.e., hydrogenated substituted
benzene) are quite limited.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The hydrogenation of benzene to form C6H7 has been well studied. The infrared spectra of C6H7
and C6H6D in solid p-H2 clearly indicated that the observed species is a cyclohexadienyl radical, the
so-called -complex.9 Spectral studies of the structural identification of mono-hydrogenated
substituted benzene radical are, however, few. To our knowledge, for mono-hydrogenated aniline
(HC6H5NH2), only an electron-paramagnetic resonance (EPR) spectrum of the amino-cyclohexadienyl
radical was reported by DiGregorio et al., who produced this radical through X-ray irradiation of
aniline in an adamantane matrix.10 These authors found that C6H5NH2 is hydrogenated dominantly at
the ortho-site and with a small contribution at the para-site. Hydrogenation of aniline might occur at
four distinct positions-- ipso-, ortho-, meta-, and para-sites, designated as ipso-, ortho-, meta-, and
para-HC6H5NH2 in this paper. Obtaining further spectral information about these isomers is important
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 4 of 39
1
2
3
in understanding the related reaction mechanism.
4
5
6
7
8
9
Para-hydrogen (p-H2) has emerged as a new matrix host suitable for producing hydrogenated
species.11-12 Two techniques employed in generating hydrogenated species in solid p-H2 are photolysis
in situ of precursors producing hydrogen atoms and the electron bombardment of a p-H2 matrix during
its deposition. In the former method, UV photolysis of a p-H2 matrix containing trace of Cl2 was
commonly used to produce isolated Cl atoms that further reacted with vibrationally excited H2
molecule upon IR irradiation (> 4000 cm−1) to produce hydrogen atoms.13 The hydrogen atoms thus
produced diffused in solid p-H2 through the breaking and formation of H‒H bonds so that they have
better chance to react with embedded molecules than with only nearest neighbors. Even if the
hydrogenation has a small barrier, it might occur via quantum-mechanical tunneling. Hence, an isomer
with the smallest barrier tends to be produced in this method. In the latter method, electron
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
bombardment of p-H2 during matrix deposition produces proton (H3 ) and hydrogen atom, which react
with the guest molecule to produce protonated and hydrogenated species. The neutralization of
protonated species is expected also to contribute to the production of hydrogenated species. A greater
variety of hydrogenated isomers might hence be produced, compared with the photolytic method. The
spectral lines of protonated and hydrogenated species could be readily distinguished because reactions
such as trapped electrons with the protonated species destroyed the protonated species and produced
the hydrogenated species during maintenance of the matrix in darkness for a prolonged period. For
both methods, the hydrogenation typically stops at the first step because, with a limited amount of H
atoms and with the mixing ratio of the parent much greater than that of the mono-hydrogenated radicals,
further hydrogenation cannot compete with the first hydrogenation step. The electron-bombardment
method has been used to identify several hydrogenated aromatic species, including benzene
derivatives,9, 14 heterocyclic compounds,12, 15 and polycyclic aromatic hydrocarbons. 16-20
We have reported the IR spectra of three isomers of protonated aniline, para-, amino-, and ortho-
H+C6H5NH2, produced on electron bombardment of an aniline/p-H2 matrix;21 these spectral lines
decreased in intensity during maintenance of the matrix in darkness for a prolonged period. Here we
report the assignments of those lines produced on electron bombardment of an aniline/p-H2 matrix of
4
ACS Paragon Plus Environment

Page 5 of 39
1
The Journal of Physical Chemistry
2
3
which the intensities increased during maintenance of the matrix in darkness to the two most stable
4
5
isomers of hydrogenated aniline, ortho- and para-HC6H5NH2.
6
7
8
9
2. EXPERIMTNAL AND COMPUTATIONAL METHODS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The experimental setup for infrared spectrometry and p-H2 matrix isolation coupled with electron
bombardment has been described elsewhere.9 A gold-plated copper block, cooled to 3.2 K with a
closed-cycle helium refrigerator, served as a matrix substrate and as a mirror to reflect the incident IR
beam to the detector for reflective absorption measurements. Infrared absorption spectra were recorded
with a Fourier-transform infrared (FTIR) spectrometer equipped with a KBr beam splitter and a
HgCdTe detector cooled to 77 K. Two hundred scans for spectral range 5000–450 cm−1 at resolution
0.25 cm−1 were typically acquired at each stage of the experiment.
The gaseous mixtures of C6H5NH2 and p-H2 were bombarded with electrons during deposition at
3.2 K. A beam of electrons with kinetic energy 200 eV and current 30 μA was provided by an electron
gun (Kimball Physics, EFG-7). We used a mixture C6H5NH2/p-H2 with typical mixing ratio of 70 ppm.
This mixture was deposited at a rate of 13 mmol h−1 over a period of 8–10 h. After deposition, the
matrix was maintained in darkness for a prolonged period to differentiate various products. Secondary
photolysis experiments were performed subsequently using light at 520±10, 465±10, and 375±10 nm
from light-emitting diodes.
Preparation of a pure p-H2 sample has been described previously.9, 22 Conversion temperatures 12–
13 K yielded p-H2 samples that contain less than 100 ppm ortho-H2. Pure p-H2 gas thus obtained was
mixed with C6H5NH2 vapor in a vacuum manifold. Before mixing, the commercially obtained
C6H5NH2 (Aldrich, 99.5 %) was degassed with several freeze-pump-thaw cycles.
Quantum-chemical calculations were performed with programs Gaussian 09 (Rev. E.01) and
Gaussian 16 (Rev. B.01).23-24 The density-functional theory with the B3LYP functional25-26 and
Dunning’s correlation-consistent basis set, cc-pVTZ,27 were employed to perform geometry
optimizations and harmonic vibrational analyses. Anharmonic vibrational wavenumbers were also
obtained with the second-order vibrational perturbation theory (VPT2) implemented in the Gaussian
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 6 of 39
1
2
3
4
5
6
7
8
09 program.28 For the calculation of single-point energies, the zero-point vibrational energy (ZPVE)
was corrected with harmonic vibrational wavenumbers without scaling. Calculations of UV-Vis
spectra were performed with time-dependent density-functional theory (TD-DFT)29 with CAM-
B3LYP functionals.30
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3. COMPUTATIONAL RESULTS
3.1 Geometry and relative energies
As mentioned above, aniline has four possible sites for hydrogenation: para-, meta-, ortho-, and
ipso-positions. The geometries of these isomers of mono-hydrogenated aniline optimized with the
B3LYP/cc-pVTZ method are presented in Figure 1; the carbon atoms are numbered. para-HC6H5NH2
possesses point-group symmetry Cs; other isomers belong to point group C1. Hydrogenation distorts
the planar phenyl ring slightly, as evident from the changes in inner angles (for example, C(1)C(2)C(3)
= 113.6° in ortho-HC6H5NH2); the out-of-plane distortion is negligible. The C(1)–N bond length in
ortho- and meta-HC6H5NH2 varied by ~0.01 Å from that (1.395 Å) of aniline, whereas that in ipso-
HC6H5NH2 increased by ~0.1 Å due to the change in hybridization of the carbon atom. The NH2 group
is slightly tilted towards the hydrogenated side in ortho- and meta-HC6H5NH2.
As summarized in Table 1, ortho-HC6H5NH2 has the least energy and para-, meta-, and ipso-
HC6H5NH2 have energies greater by 9.8, 14.2, and 42.0 kJ mol−1, respectively; these energies are
corrected for ZPVE. The transition states (TS) for hydrogenation and isomerization reactions were
also calculated; the potential-energy scheme for the H + C6H5NH2 system is depicted in Figure 2. The
hydrogenation to form ipso-HC6H5NH2 has the largest barrier, 38.3 kJ mol−1, whereas those paths to
form other isomers have barriers in a much smaller range, 9.8‒15.0 kJ mol−1. The energies of the
transition states (TS1–TS3) for isomerization, 195‒225 kJ mol−1, are much greater than that of H +
C6H5NH2, indicating that these isomerization reactions are unlikely to occur.
3.2 Predicted vibrational spectra
Scaled harmonic vibrational wavenumbers and IR intensities predicted with the B3LYP/cc-pVTZ
6
ACS Paragon Plus Environment

Page 7 of 39
1
The Journal of Physical Chemistry
2
3
4
5
6
7
8
9
method are listed in Tables 2, 3, S-1, and S-2 for ortho-, para-, meta-, and ipso-HC6H5NH2,
respectively. The calculated harmonic vibrational wavenumbers are scaled according to equations y =
(0.9212 ± 0.031) x + (138 ± 101) and y = (0.9716 ± 0.004) x + (8.5 ± 4.6) for values above and below
2000 cm−1, respectively, in which y is the scaled values and x is the calculated harmonic vibrational
wavenumber. These equations, derived according to a linear relation between predicted harmonic
vibrational wavenumbers and observed wavenumbers of C6H5NH2 isolated in a p-H2 matrix,21 are
employed for other species discussed in this work. Anharmonic vibrational wavenumbers and IR
intensities are also listed in these tables for comparison.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Several intense vibrational lines are predicted to be associated with the NH2 and CH2 groups of
HC6H5NH2. We describe here the characteristic vibrational modes of the two most stable isomers,
ortho- and para-HC6H5NH2, and list scaled harmonic vibrational wavenumbers unless otherwise
indicated. For ortho-HC6H5NH2, antisymmetric (ν1) and symmetric (ν2) NH2 stretching modes are
predicted at 3495 and 3399 cm−1, which are slightly red-shifted from those of aniline, predicted at 3504
and 3415 cm−1. Further red shifts are predicted for para-HC6H5NH2, of which the wavenumbers of
antisymmetric (ν23) and symmetric (ν1) NH2-stretching modes are 3489 and 3397 cm−1; respectively.
The lines associated with the NH2-scissoring modes (δsNH2) are predicted at 1617 (ν9) and 1585 (ν10)
cm−1 for ortho-HC6H5NH2, and at 1611 (ν6) and 1601 (ν7) cm−1 for para-HC6H5NH2. In both cases,
the former is a local NH2-scissoring mode and the latter is coupled with a ring-deformation mode (δR).
The lines associated with the NH2-rocking modes (ρNH2) are predicted to be weak with IR intensity <
10 km mol−1. Although the NH2-wagging (inversion) modes (ωNH2) are predicted to be intense, this
result is likely a computational artifact, as discussed previously.21 Because of a small inversion barrier
for the NH2 group, the NH2-wagging mode is markedly anharmonic; its wavenumber and IR intensity
cannot be predicted accurately with either harmonic or anharmonic (VPT2) vibrational analysis.
The lines of symmetric and antisymmetric CH2-stretching modes (νCH2) are predicted at 2820 (ν7)
and 2791 (ν8) cm−1, respectively, for ortho-HC6H5NH2, and at 2819 (ν4) and 2796 (ν5) cm−1 for para-
HC6H5NH2; the predicted IR intensities of these lines are > 30 km mol−1. The lines associated with the
CH2-wagging mode (ωCH2), 1252 (ν17) cm−1 for ortho-HC6H5NH2 and 1284 (ν10) and 1165 (ν30) cm−1
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 8 of 39
1
2
3
4
5
6
7
8
9
for para-HC6H5NH2, are predicted to be also intense. The line associated with the CH2-rocking modes
(ρCH2, ν25) is weak in ortho-HC6H5NH2 (with IR intensity ~2 km mol−1) but is intense in para-
HC6H5NH2 (ν29 with IR intensity 63 km mol−1). Other intense vibrational modes include the CH-
stretching (νCH) and out-of-plane CH-bending (oop δCH) modes. The spectral ranges for the CH-
stretching modes are 3033–3082 and 3028–3060 cm−1 for ortho- and para-HC6H5NH2, respectively,
which are near the range predicted for aniline 3044–3080 cm−1. The intense lines associated with the
out-of-plane CH-bending modes are predicted at 654 (ν30) cm−1 for ortho-HC6H5NH2 and at 949 (ν33)
cm−1 for para-HC6H5NH2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The characteristic vibrational modes mentioned above were predicted to be intense also for meta-
and ipso-HC6H5NH2, as listed in Tables S-1 and S-2. Overall, although similar vibrational modes are
predicted to be intense among the four isomers, the predicted spectral patterns (vibrational
wavenumbers and relative intensities) are sufficiently distinct for spectral identification.
4. EXPERIMETNAL RESULTS
Figure 3 shows representative IR spectra of an electron-bombarded C6H5NH2/p-H2 matrix in
spectral ranges 1700–1100 and 1100–500 cm−1; spectra in the NH- and CH2-stretching regions, 3550
–3340 and 2840–2680 cm−1, are shown in Figure S-1. Figure 3(a) represents the spectrum measured
after matrix deposition for 10 h; for clarity, the lines of aniline were subtracted using a spectrum of a
C6H5NH2/p-H2 matrix deposited without electron bombardment. For some intense lines of aniline or
lines slightly shifted upon electron bombardment, the subtraction was not clean; these regions are
shaded with grey. After subtraction of lines of aniline, the remaining lines in this spectrum were
induced upon electron bombardment. Figure 3(b) is a difference spectrum recorded after maintaining
the electron-bombarded matrix in darkness for 19 h; the difference spectrum was obtained on
subtracting the spectrum recorded before the experimental step from that recorded after the step so that
positive lines indicate generation and negative ones indicate destruction. After this experimental step,
the matrix was irradiated with light at 520 nm for 2 h, followed by irradiation with light at 465 nm for
10 min and at 375 nm for 10 min. Traces (c)–(e) show the difference spectra after each step of the
8
ACS Paragon Plus Environment

Page 9 of 39
1
The Journal of Physical Chemistry
2
3
secondary photolysis.
4
5
+
An electron bombardment during matrix deposition of p-H2 is known to produce H3 and H atoms;
6
7
+
proton transfer from H3 and hydrogen addition to a guest molecule produce protonated and
8
9
hydrogenated products, respectively.11-12 The protonated products become slowly neutralized with
electrons trapped in solid p-H2. When we measured a spectrum after maintaining the matrix in darkness
for a prolonged period, the intensities of lines of protonated products consequently decreased. The
lines indicated with blue dots in trace (a) and showing corresponding negative lines in trace (b) have
been assigned to three isomers of protonated aniline H+C6H5NH2; see Ref. 21 for details. In contrast,
hydrogenated products are expected to increase after maintaining the electron-bombarded matrix in
darkness for a prolonged period, because neutralization of protonated products produced their neutral
counterparts and unreacted hydrogen atoms might further diffuse in solid p-H2 to react with guest
molecules. As a result, in the difference spectrum recorded after this step, trace (b), the positive lines
might be associated with hydrogenated species. We identified two groups of lines that increased in this
experimental step and labeled them A and B in Figure 3. The classification of these lines to groups A
and B was established according to secondary photolysis experiments, in which distinct changes in
intensity were observed for each group of lines. These lines in groups A and B are assigned to ortho-
HC6H5NH2 and para-HC6H5NH2, respectively, to be discussed in Sections 5.1 and 5.2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As shown in traces (b)‒(e) of Figure 3 and summarized in Table 4, the intensities of lines in group
A increased by (43±17) % during maintenance of the matrix in darkness for 19 h and decreased by
(35±15), (13±8), and (75±10) % upon irradiation at 520, 465, and 375 nm, respectively; the percentage
variations refer to the intensities recorded after deposition; the error limits represent one standard
deviation for measurements of intense lines of group A at 1614.9, 1431.2, 1259.1, and 650.9 cm−1 and
weaker ones at 1516.9 and 987.1 cm−1. The intensities of lines in group B increased by (50±17) %
during maintenance of the matrix in darkness for 19 h and decreased by (36±19), (30±19), and (19±5)
% upon irradiation at 520, 465, and 375 nm, respectively. Intense lines of group B at 1606.0, 1455.7,
1290.4, 1263.8, 1035.8, and 521.9 cm−1 and weaker ones at 1588.3, 962.5, and 957.8 cm−1 were used
in averaging. In the NH2- and CH2-stretching regions, four lines at 3482.3, 3406.7, 2770.8, and 2713.7
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 10 of 39
1
2
3
4
5
6
7
8
cm−1 were identified to belong to group A; four lines at 3486.0, 3405.0, 2775.1, and 2706.1 cm−1 are
identified to belong to group B. The correlations in intensity variations of these lines were, however,
less satisfactory because of spectral interference from aniline and, in some cases, overlapping with
lines in the other group.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5. DISCUSSION
5.1 Assignments of lines in group A to ortho-HC6H5NH2
As discussed previously, the IR lines that were induced by electron bombardment and gained
intensity during maintenance of the matrix in darkness are most likely due to hydrogenated species.
To enable assignments of lines in groups A and B, we compared the experimental spectra in a
representative region with the predicted spectra of four possible isomers of HC6H5NH2 in Figure 4.
The difference spectrum after secondary photolysis at 375 nm, which depleted lines in group A by ~75
% (and those in group B by only ~19 %), is inverted and presented in trace (a) so that the lines in group
A, marked with blue squares, are pointing upward. The difference spectrum after the secondary
photolysis at 465 nm, which depleted lines in group B by ~30 % (and those in group A by only ~13 %),
is inverted and presented as trace (c), in which lines in group B are marked with red circles. Traces
(b), (d), (e), and (f) are the stick IR spectra of ortho-, para-, meta-, and ipso-HC6H5NH2, respectively,
simulated according to the scaled harmonic vibrational wavenumbers and IR intensities predicted with
the B3LYP/cc-pVTZ method.
In the spectral region below 1650 cm−1, lines in group A were observed at 1614.9, 1516.9, 1431.2,
1259.1, 987.1, and 650.9 cm−1 (Figure 4(a)), in satisfactory agreement with the predicted spectrum of
ortho-HC6H5NH2, which has intense lines at 1617, (1585), 1518, 1428, 1252, 986, and 654 cm−1
(Figure 4(b)); see also Table 2. The line calculated at 1585 cm−1 (ν10) is predicted to be intense enough
(28 km mol−1) but was unidentified in the experiment, likely because of interference from an
unassigned line near 1583 cm−1, of which the intensity varied differently from those in groups A and
B; this line might be a perturbed line of the parent after electron bombardment. Observed and predicted
wavenumbers for the NH2-stretching mode also agree satisfactorily; the lines observed at 3482.3 and
10
ACS Paragon Plus Environment

Page 11 of 39
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
3406.7 cm−1 correspond to those predicted at 3495 and 3399 cm−1 for the antisymmetric and symmetric
NH2-stretching modes. For the symmetric and antisymmetric CH2-stretching modes, the lines observed
at 2770.8 and 2713.7 cm−1 correspond to lines predicted at 2820 and 2791 cm−1. These deviations are
significantly greater than those of other lines but are characteristic of the harmonic vibrational
calculations on CH2-stretching modes of hydrogenated species. The anharmonic vibrational
calculations indicate 2735 and 2703 cm1 for these two modes of ortho-HC6H5NH2, much smaller than
those predicted from scaled harmonic vibrational calculations (Table 2); these deviations indicate that
these modes are strongly coupled with other modes. We compared experimental values with scaled
harmonic and anharmonic vibrational calculations of these two CH2-stretching modes of several
protonated aromatic compounds that we have studied in Table S-3; they all show a similar trend.
Based on the expected chemistry and the satisfactory agreement between predicted and observed
spectra in terms of wavenumbers and relative intensities (Table 2), we assigned these ten lines in group
A to ortho-HC6H5NH2. The mean absolute deviation between scaled harmonic wavenumbers and
observed wavenumbers is 16.5±24.5 cm−1, in which the error represents one standard deviation in
averaging; the average deviation decreases to 4.8±3.8 cm−1 when the lines of CH2-stretching modes
are excluded. The corresponding deviations from the anharmonic wavenumbers are 9.2±9.7 cm−1 and
5.7±4.1 cm−1, respectively.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5.2 Assignment of lines in group B to para-HC6H5NH2
In the spectral region below 1650 cm−1, intense lines in group B were observed at 1606.0, 1455.7,
1290.4, 1263.8, and 521.9 cm−1 with intensities greater than 60 % of the most intense line at 1455.7
cm−1; weaker lines were observed at 1588.3, 1035.8, 962.5, and 957.8 cm−1 (Figure 4(c)). These line
positions and relative intensities are in satisfactory agreement with the spectrum predicted for para-
HC6H5NH2 (Figure 4(d)), of which intense lines are predicted at 1611, 1456, 1284, 1265, and 529 cm−1
with IR intensities > 45 km mol−1 and weaker ones are at 1601, 1037, 938, and 949 cm−1. In this
spectral region, all lines with predicted IR intensity > 7 km mol−1 were observed. In the NH2-stretching
region, the observed wavenumbers of lines in group B at 3486.0 and 3405.0 cm−1 agree satisfactorily
11
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 12 of 39
1
2
3
4
5
6
7
8
with those predicted at 3489 and 3397 cm−1 for the antisymmetric and symmetric NH2-stretching
modes. In the CH2-stretching region, lines in group B were observed at 2775.1 and 2706.1 cm−1,
corresponding to the predicted scaled harmonic vibrational wavenumbers of 2819 and 2796 cm−1 for
the symmetric and antisymmetric CH2-stretching modes of para-HC6H5NH2; similarly to ortho-
HC6H5NH2, the anharmonic vibrational wavenumbers 2714 and 2705 cm1 for these two modes are
much smaller than the scaled harmonic vibrational wavenumbers (Table 2).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on the expected chemistry and the satisfactory agreement between predicted and observed
spectra in terms of wavenumbers and relative intensities (Table 3), we assigned these 14 lines in group
B to para-HC6H5NH2. The mean absolute deviation between scaled harmonic vibrational
wavenumbers and observed wavenumbers is 15.7±23.4 cm−1; the deviation decreases to 7.1±6.3 cm−1
when lines of CH2-stretching modes are excluded. The corresponding values for the anharmonic
vibrational wavenumbers are 13.6±15.2 cm−1 and 10.7±7.7 cm1, respectively
5.3 Behavior in secondary photolysis
To understand the behavior of secondary photolysis of these hydrogenated radicals HC6H5NH2,
we calculated their ultraviolet and visible absorption spectra with the TD-DFT method employing
CAM-B3LYP/cc-pVTZ. The predicted vertical transitions, convoluted with Gaussian profiles of full
width at half maximum (FWHM) 0.25 eV, are presented in Figure 5; actual absorption bands are
expected to be red-shifted. In this figure, the wavelength regions used for secondary photolysis,
375±10, 465±10, and 520±10 nm, are indicated with violet, blue, and green rectangles, respectively.
ortho-HC6H5NH2 has intense absorption bands near 335 and 205 nm and weaker ones near 450 and
280 nm. The predicted spectrum of para-HC6H5NH2 shows two intense absorptions near 280 and 220
nm and one moderately intense absorption centered near 515 nm. The predicted spectrum of meta-
HC6H5NH2 has three intense absorption bands near 210, 270, and 360 nm and one moderately intense
absorption centered at 520 nm. ipso-HC6H5NH2 has intense absorption bands near 225 and 310 nm
with weaker ones near 225, 385, and 455 nm. Similarities among these spectra of various isomers of
HC6H5NH2 indicate that a wavelength that can dissociate only one isomer is unlikely to occur.
12
ACS Paragon Plus Environment

Page 13 of 39
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
As summarized in Table 4, the irradiation at 375 nm significantly (by ~75 %) decreased the
intensities of lines in group A (ortho-HC6H5NH2), but only slightly (by ~19 %) the lines in group B
(para-HC6H5NH2), consistent with the predicted UV absorbance of ortho-HC6H5NH2 much greater
than that of para-HC6H5NH2 near this region (Figure 5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the other two wavelengths, the photolytic behaviors of these two species are less distinct.
Irradiation at 465 nm diminished ortho-HC6H5NH2 by 13 %, but para-HC6H5NH2 by 30%, consistent
with a greater absorption cross section predicted for the latter in this region. However, irradiation at
520 nm decreased ortho- and para-HC6H5NH2 by a similar fraction (~35%), even though the
absorbance of para-HC6H5NH2 was predicted to be about ten times that of ortho-HC6H5NH2 at this
wavelength. Considering that we are using only the predicted vertical excitations and convoluted with
a fixed width, this deviation might be acceptable. Another possibility is that the excitation did not
result into dissociation effectively. The key point is that the 375-nm wavelength is near the predicted
main absorption band of ortho-HC6H5NH2, which is mainly contributed by SOMO→LUMO and
HOMO→LUMO transitions, whereas the corresponding absorption band of para-HC6H5NH2 was
predicted to have a smaller wavelength, so that significantly decreased IR lines of ortho-HC6H5NH2
was observed near 375 nm.
5.4 Mechanism of formation
The mechanism for the formation of hydrogenated species upon electron bombardment of mixtures
of p-H2 and guest molecules has been discussed in the literature.11-12 Briefly, electron bombardment
+
+
ionizes H2 molecule to produce H2 , which readily reacts with another H2 molecule to produce H3 and
+
H atom. The H3 and H are mobile in solid p-H2 through exchange reactions with the nearby H2
molecules. Proton transfer occurs to produce a protonated guest molecule if the proton affinity of the
guest molecule is greater than that of H2 (422 kJ mol−1).31 As aniline has a proton affinity of 882.5 kJ
+
mol−1,31 H3 readily transfers a proton to aniline to form protonated aniline and H2. The protonated
aniline becomes neutralized by trapped electrons, which slowly diffuse in solid p-H2, to produce
hydrogenated species. The other process responsible for the generation of hydrogenated species is
13
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 14 of 39
1
2
3
4
5
6
7
8
9
hydrogen addition to aniline. The hydrogen additions to form ortho-, meta-, and para-HC6H5NH2 have
small barriers (Table 1), but these reactions can occur even at 3.2 K via quantum-mechanical tunneling.
Mixing ratios of products in solid p-H2 are estimable according to the method developed by Tam
and Fajardo.32 Their method is based on Beer’s law; the mixing ratio x (in ppm) in solid p-H2 is
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
represented as
10 (퐼/퐼0)푑휈
휀(cm mol ―1)푙(cm)
푥 = 2.303 ∫log
3
―1
6
(1)
,
× 푉푚(cm mol ) × 10
in which ε is the absorption coefficient,
is the observed integrated absorbance, l is
∫log10 (퐼/퐼0)푑휈
the path length of IR light through solid p-H2, and Vm (= 23.16 cm3 mol−1)33 is the molar volume of
solid p-H2. Infrared intensities calculated with the B3LYP/cc-pVTZ method were used for ε. The IR
path length l was estimated from the integrated band area of transition S1(0) + S0(0) of the p-H2
matrix;34-35 l is typically ~2 mm in these experiments. The major error in an estimate of this type is the
error in calculated IR intensities, which might be as large as a factor of two. To minimize this error,
we typically selected several lines of a species and averaged the estimated mixing ratio. Furthermore,
on selecting similar vibrational modes of two species, we envisage that the relative error between these
two species might cancel. To derive a definitive conclusion from quantitative analysis of the mixing
ratios is difficult because of the associated large errors. However, the relative variations of each species
in each experimental step are reliable.
The mixing ratios of observed protonated aniline isomers (ortho-, para-, and amino-H+C6H5NH2)
and hydrogenated aniline isomers (ortho- and para-HC6H5NH2) after matrix deposition and after
maintenance in darkness for 19 h were estimated using several intense lines and the predicted IR
intensities of each species. The estimated mixing ratios (in ppm) are summarized in Table 5. Just after
the matrix deposition, the mixing ratios of C6H5NH2, ortho- and para-HC6H5NH2 are 53±5, 2.0±0.9
and 2.5±0.9, respectively; the error limits represent one standard deviation in averaging values derived
from various absorption lines. The mixing ratios of ortho-, para-, and amino-H+C6H5NH2 are 0.6±0.1,
1.0±0.1, and 1.0±0.5, respectively, as reported previously;21 because para-H+C6H5NH2 has a lower
energy than ortho-H+C6H5NH2 by 18.4 kJ mol−1 according to the B3LYP/cc-pVTZ method, the
barrierless formation of para-H+C6H5NH2 over ortho-H+C6H5NH2 is preferred. For the formation of
14
ACS Paragon Plus Environment

Page 15 of 39
1
The Journal of Physical Chemistry
2
3
4
5
6
7
8
9
HC6H5NH2 from H + C6H5NH2, small barriers exist; the barriers to produce ortho- and para-
HC6H5NH2 are 9.8 and 11.3 kJ mol−1, respectively; the energy of para-HC6H5NH2 is greater than that
of ortho-HC6H5NH2 by 9.8 kJ mol−1. Considering the barriers and relative energies, one would expect
that ortho-HC6H5NH2 might be slightly preferentially produced during deposition, from either
neutralization of H+C6H5NH2 or hydrogenation of C6H5NH2, but our observation indicated that the
formation of para-HC6H5NH2 might be slightly favored, 2.5±0.9 over 2.0±0.9, although within
experimental uncertainties. This condition might also imply that the mixing ratio of para-HC6H5NH2
was overestimated.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
During maintenance of the matrix in darkness, hydrogenation can occur only via quantum-
mechanical tunneling, so that the formation of ortho-HC6H5NH2 is expected to be more efficient
because of the smaller barrier. In contrast, formation of hydrogenated aniline via neutralization of
protonated aniline is expected to depend only on the original mixing ratio of protonated species, if the
neutralization involves no isomerization or dissociation; the formation of para-HC6H5NH2 from
neutralization is hence slightly preferred because of a greater mixing ratio of para-H+C6H5NH2.
After maintenance of the matrix in darkness for 19 h, we observed that the mixing ratios of ortho-
and para-HC6H5NH2 increased by 0.8±0.4 and 1.3±0.6 ppm, respectively; these values are ~40 and 52
% of the original ones. Even though the two uncertainty ranges overlap, the formation of para-
HC6H5NH2 appeared to be slightly favored. Furthermore, during this process, the estimated total loss
of para- and ortho-H+C6H5NH2, 0.5±0.2 ppm, is smaller than the observed total production of ortho-
and para-HC6H5NH2, 2.1±1.4 ppm; the loss of aniline was estimated to be (5.4±2.6) ppm. Even though
quantitative consistency was not achievable, these observations are consistent with an expectation that
both neutralization of protonated aniline and hydrogenation of aniline occurred during the maintenance
of the matrix in darkness. Even though the formation of meta-HC6H5NH2 from H + C6H5NH2 also has
a small barrier and the energy of meta-HC6H5NH2 is also small, we did not observe meta-HC6H5NH2;
this result seems to indicate that the hydrogenation is less important than neutralization during
maintenance of the matrix in darkness, because meta-H+C6H5NH2 was not produced after electron
bombardment due to its high energy, or the barrier for the formation of meta-HC6H5NH2 from H +
15
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 16 of 39
1
2
3
4
5
6
7
8
C6H5NH2 is actually larger than predicted. The absence of meta-H+C6H5NH2 might indicate that
protonation on the meta-site is not the major channel of the Birch mechanism in this case.
Finally, we observed amino-H+C6H5NH2 but not its neutral counterpart C6H5NH3, even
though we expect the neutralization of amino-H+C6H5NH2 to occur. The reason is that, as C6H5NH3 is
unstable, neutralization of amino-H+C6H5NH2 might lead to a dissociation to C6H5NH2 and H.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6. CONCLUSION
The two most stable isomers of hydrogenated aniline, ortho- and para-HC6H5NH2, generated upon
electron bombardment of a mixture of C6H5NH2 and p-H2 during matrix deposition, were identified
with infrared spectroscopy. The IR spectra of these isomers of HC6H5NH2 are new. The intensity of
these IR lines increased during maintenance of the electron-bombarded matrix in darkness; they were
classified into two groups (A and B) according to their intensity variations upon secondary photolysis
at 520, 465, and 375 nm. The spectral assignments were made on comparison of observed
wavenumbers and relative intensities with the vibrational wavenumbers and IR intensities predicted
with the B3LYP/cc-pVTZ method. Ten lines in group A, of which the intensity decreased significantly
upon irradiation at 375 nm and moderately at 520 and 465 nm, were assigned to the most stable ortho-
HC6H5NH2. Fourteen lines in group B, of which the intensities decreased moderately at all
wavelengths employed, were assigned to para-HC6H5NH2, which is predicted to have energy greater
than ortho-HC6H5NH2 by ~10 kJ mol−1. The behavior of these species upon secondary photolysis was
explained according to the UV-Vis spectra predicted with the TD-DFT method. Characteristic
absorption lines of the CH2-symmetric and CH2-antisymmetric stretching modes were observed at
2770.8 and 2713.7 cm1 for ortho-HC6H5NH2 and 2775.1 and 2706.1 cm1 for para-HC6H5NH2.
The formation of these isomers is due partly to the neutralization of H+C6H5NH2 and, likely to a
lesser extent, to the hydrogenation of aniline, according to the estimated variations of mixing ratios of
isomers of HC6H5NH2 and H+C6H5NH2.
ASSOCIAATED CONTENT
16
ACS Paragon Plus Environment

Page 17 of 39
1
The Journal of Physical Chemistry
2
3
4
5
6
7
8
9
Supporting Information. Full lists of vibrational wavenumbers and IR intensities of meta-
HC6H5NH2 (Table S1) and ipso- HC6H5NH2 (Table S2); comparison of experimental wavenumbers of
CH2-stretching modes in hydrogenated species (Table S3); and infrared spectra of an electron-
bombarded C6H5NH2/p-H2 matrix in the region 2680‒3550 cm1 (Figure S1)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
AUTHOR INFORMATION
Corresponding authors
* E-mail (MT): tsuge@lowtem.hokudai.ac.jp.
* E-mail (YPL): yplee@nctu.edu.tw.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
This work was supported by Ministry of Science and Technology, Taiwan (grant numbers MOST108-
2639-M-009-001-ASP and MOST108-3017-F009-004) and the Center for Emergent Functional
Matter Science of National Chiao Tung University from The Featured Areas Research Center Program
within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in
Taiwan. Japan Society for the Promotion of Science (JSPS KAKENHI grant number JP18K03717)
partially supported this work. The National Center for High-Performance Computation of Taiwan and
the Information Initiative Center of Hokkaido University provided computational resources.
References
1.
2.
Birch, A. J., Reduction by Dissolving Metals. Part I. J. Chem. Soc. 1944, 430–436.
Zimmerman, H. E., A Mechanistic Analysis of the Birch Reduction. Acc. Chem. Res. 2012,
45, 164–170.
3.
Watt, G. W., Reactions of Organic and Organometallic Compounds with Solutions of Metals
in Liquid Ammonia. Chem. Rev. 1950, 46, 317–379.
4.
5.
Raphael, R. A., Advances in Organic Chemistry. Interscience Publishers: New York, 1972.
Birch, A. J.; Hinde, A. L.; Radom, L., A theoretical Approach to the Birch Reduction.
Structures and Stabilities of Cyclohexadienyl Radicals. J. Am. Chem. Soc. 1980, 102, 4074–4080.
17
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 18 of 39
1
2
6.
Birch, A. J.; Hinde, A. L.; Radom, L., A theoretical Approach to the Birch reduction.
3
4
5
6
7
8
Structures and Stabilities of Cyclohexadienyl Anions. J. Am. Chem. Soc. 1980, 102, 6430–6437.
7.
Zimmerman, H. E., Orientation in Metal Ammonia Reductions. Tetrahedron 1961, 16, 169–
Zimmerman, H. E.; Wang, P. A., The Regioselectivity of the Birch Reduction. J. Am. Chem.
176.
8.
9
Soc. 1993, 115, 2205–2216.
9. Bahou, M.; Wu, Y.-J.; Lee, Y.-P., A New Method for Investigating Infrared Spectra of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
Protonated Benzene (C6H7 ) and Cyclohexadienyl Radical (c-C6H7) Using para-Hydrogen. J. Chem.
Phys. 2012, 136, 154304.
10.
DiGregorio, S.; Yim, M. B.; Wood, D. E., Electron Paramagnetic Resonance Study of Free
Radicals in an Adamantane Matrix. VII. Aromatic Hydrogen Addition Radicals. J. Am. Chem. Soc.
1973, 95, 8455–8457.
11.
Bahou, M.; Das, P.; Lee, Y.-F.; Wu, Y.-J.; Lee, Y.-P., Infrared Spectra of Free Radicals and
Protonated Species Produced in para-Hydrogen Matrices. Phys. Chem. Chem. Phys. 2014, 16, 2200–
2210.
12.
Radicals Isolated in para-Hydrogen Matrices. Phys. Chem. Chem. Phys. 2018, 20, 5344–5358.
13. Raston, P. L.; Anderson, D. T., Infrared-induced Reaction of Cl Atoms Trapped in Solid
Parahydrogen. Phys. Chem. Chem. Phys. 2006, 8, 3124–3129.
14. Tsuge, M.; Lai, C.-P.; Lee, Y.-P., Infrared Spectra of 3-hydroxy-(1H)-pyridinium Cation and
3-hydroxy-(1H)-pyridinyl Radical Isolated in Solid para-Hydrogen. J. Chem. Phys. 2018, 149, 014306.
15.
Golec, B.; Das, P.; Bahou, M.; Lee, Y.-P., Infrared Spectra of the 1-Pyridinium (C5H5NH+)
Tsuge, M.; Tseng, C.-Y.; Lee, Y.-P., Spectroscopy of Prospective Interstellar Ions and
Cation and Pyridinyl (C5H5NH and 4-C5H6N) Radicals Isolated in Solid para-Hydrogen. J. Phys.
Chem. A 2013, 117, 13680–13690.
16.
Bahou, M.; Wu, Y.-J.; Lee, Y.-P., Formation and Infrared Absorption of Protonated
+
+
Naphthalenes (1-C10H9 and 2-C10H9 ) and their Neutral Counterparts in Solid para-Hydrogen. Phys.
Chem. Chem. Phys. 2013, 15, 1907–1917.
17.
Counterpart in Solid para-Hydrogen. J. Phys. Chem. Lett. 2013, 4, 1989–1993.
18. Bahou, M.; Wu, Y.-J.; Lee, Y.-P., Infrared Spectra of Protonated Coronene and its Neutral
Bahou, M.; Wu, Y.-J.; Lee, Y.-P., Infrared Spectra of Protonated Pyrene and its Neutral
Counterpart in Solid Parahydrogen: Implications for Unidentified Interstellar Infrared Emission Bands.
Angew. Chem. Int. Ed. 2014, 53, 1021–1024.
19.
Tsuge, M.; Bahou, M.; Wu, Y.-J.; Allamandola, L.; Lee, Y.-P., Infrared Spectra of Ovalene
(C32H14) and Hydrogenated Ovalene (C32H15•) in Solid para-Hydrogen. Phys. Chem. Chem. Phys.
2016, 18, 28864–28871.
20.
Sundararajan, P.; Tsuge, M.; Baba, M.; Sakurai, H.; Lee, Y.-P., Infrared Spectrum of
Hydrogenated Corannulene rim-HC20H10 Isolated in Solid para-hydrogen. J. Chem. Phys. 2019, 151,
044304.
21.
Tsuge, M.; Chen, Y.-H.; Lee, Y.-P., Infrared Spectra of Isomers of Protonated Aniline in Solid
para-Hydrogen. J. Phys. Chem. A 2020, 124, 2253-2263.
18
ACS Paragon Plus Environment

Page 19 of 39
1
2
3
The Journal of Physical Chemistry
22.
Bahou, M.; Huang, C.-W.; Huang, Y.-L.; Glatthaar, J.; Lee, Y.-P., Advances in Use of p-H2
as a Novel Host for Matrix IR Spectroscopy. J. Chin. Chem. Soc. 2010, 57, 771–782.
4
5
23.
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J.
6
7
8
R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09 Rev. E.01,
Wallingford CT, 2009.
9
24.
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; et al. Gaussian 16 Rev. B.01,
Wallingford, CT, 2016.
25.
Becke, A. D., Density-functional Thermochemistry. III. the Role of Exact Exchange. J. Chem.
Phys. 1993, 98, 5648–5652.
26.
Lee, C. T.; Yang, W. T.; Parr, R. G., Development of the Colle-Salvetti Correlation-energy
Formula into a Funcitional of the Electron-density. Phys. Rev. B 1988, 37, 785–789.
27.
Atoms Boron through Neon and Hydrogen. J. Chem. Phys. 1989, 90, 1007–1023.
28. Bloino, J.; Barone, V., A Second-order Perturbation Theory Route to Vibrational Averages
Dunning Jr., T. H., Gaussian-basis Sets for Use in Correlated Molecular Calculations. 1. The
and Transition Properties of Molecules: General Formulation and Application to Infrared and
Vibrational Circular Dichroism Spectroscopies. J. Chem. Phys. 2012, 136, 124108.
29.
Bauernschmitt, R.; Ahlrichs, R., Treatment of Electronic Excitations within the Adiabatic
Approximation of Time Dependent Density Functional Theory. Chem. Phys. Lett. 1996, 256, 454–
464.
30.
the Coulomb-attenuating Method (CAM-B3LYP). Chem. Phys. Lett. 2004, 393, 51–57.
31. Hunter, E. P. L.; Lias, S. G., Evaluated Gas Phase Basicities and Proton Affinities of
Molecules: An Update. J. Phys. Chem. Ref. Data 1998, 27, 413–656.
32. Tam, S.; Fajardo, M. E., Single and Double Infrared Transitions in Rapid-Vapor-Deposited
Yanai, T.; Tew, D. P.; Handy, N. C., A New Hybrid Exchange–correlation Functional Using
Parahydrogen Solids: Application to Sample Thickness Determination and Quantitative Infrared
Absorption Spectroscopy. Appl. Spectrosc. 2001, 55, 1634–1644.
33.
Static Properties. Reb. Mod. Phys. 1980, 52, 393–452.
34. Fajardo, M. E., Matrix Isolation Spectroscopy in Solid Parahydrogen: A Primer. In Physics
Silvera, I. F., The Solid Molecular Hydrogens in the Condensed Phase: Fundamentals and
and Chemistry at Low Temepratures, Khriachtchev, L., Ed. Pan Stanford Publishing Pte. Ltd.:
Singapore, 2011; pp 167–202.
35.
Fajardo, M. E., Solid Parahydrogen Thickness Revisited. Appl. Spectrosc. 2019, 73, 1403–
1408.
19
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 20 of 39
1
2
Table 1. Relative energies of isomers of mono-hydrogenated aniline HC6H5NH2 and transition states
for formation and isomerization reactions calculated with the B3LYP/cc-pVTZ method.a
3
4
5
6
7
8
9
relative energy
/kJ mol−1
0.0
barrier for formation c
isomerb
/kJ mol−1
ortho-
(9.8)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
para-
9.8
(11.3)
(15.0)
(38.3)
meta-
14.2
ipso-
42.0
H + C6H5NH2
TS1 (ortho-meta)
TS2 (meta-para)
TS3 (ipso-ortho)
103.8
195.8
204.8
224.5
a Zero-point vibrational energies are corrected using unscaled harmonic vibrational wavenumbers.
b See Figure 1 for hydrogenation sites.
c Energy of the transition state relative to that of H + C6H5NH2.
20
ACS Paragon Plus Environment

Page 21 of 39
1
2
3
The Journal of Physical Chemistry
Table 2. Comparison of observed vibrational wavenumbers (cm−1) and IR intensities of lines in group
A with scaled harmonic and anharmonic vibrational wavenumbers and relative IR intensities of ortho-
HC6H5NH2 predicted with the B3LYP/cc-pVTZ method.
4
5
6
7
8
9
calculations
scaled harmonica,b
experiment
νi
mode descriptione
d
anharmonicc
3479 (13)
3393 (11)
3060 (34)
3033 (54)
3029 (17)
3021 (42)
p-H2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
3495 (10)
3399 (10)
3082 (15)
3063 (24)
3045 (8)
3033 (22)
2820 (62)
2791 (54)
1617 (100)
1585 (28)
1518 (16)
1428 (30)
1420 (7)
1411 (2)
1339 (0)
1291 (6)
1252 (63)
1168 (1)
1160 (0)
1123 (4)
1076 (1)
986 (9)
3482.3 (–)f a-νNH2
3406.7 (–)f s-νNH2
2
3
g
–
–
–
–
νC5H + s-νC3H/νC4H/νC6H
s-νC3H/νC4H + νC5H
a-νC3H/νC4H + νC5H
νC6H
g
g
g
4
5
6
7
2735 (133) 2770.8 (–)f s-νCH2
8
2703 (69) 2713.7 (7) a-νCH2
9
1617 (100) 1614.9 (100) δsNH2
g
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
1573 (55)
1512 (16)
1421 (16)
1412 (26)
1409 (2)
1339 (1)
1277 (7)
1252 (28)
1172 (17)
1167 (0)
1126 (16)
1076 (2)
989 (10)
960 (1)
–
δsNH2 + s-νC3C4/νC1C6
1516.9 (6) a-νC3C4/νC1C6
1431.2 (39) a-νC1C2/νC5C6 + δC6H + δsCH2
s-νC1C2/νC4C5 + δsCH2 + δC4H
a-νC2C3/νC4C5 + δC3H/δC4H + δsCH2
δC3H/δC4H/δC5H/δC6H + ωCH2
δR + ωCH2
1259.1 (39) νC1N + δR + ωCH2
tCH2
a-δC3H/δC4H
a-δC5H/δC6H
ρNH2 + s-νC1C2/νC4C5
987.1 (4) δR
961 (1)
δR
940 (1)
940 (1)
oop (δC4H + s-δC3H/δC5H
oop δC3H + ρCH2
δR
)
934 (2)
929 (5)
915 (6)
918 (8)
862 (11)
767 (1)
871 (7)
oop (δC5H + δC6H
ring breathing
oop a-δC3H/δC5H
)
769 (3)
725 (1)
742 (3)
654 (84)
582 (4)
653 (85)
580 (9765)
426 (688)
530 (65)
491 (69)
650.9 (44) oop s-δC3H/δC4H/δC5H
δR
577 (231)
531 (9)
ωNH2 (inversion)
oop δR
492 (11)
δR
21
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 22 of 39
1
2
3
4
5
6
7
8
35
36
37
38
39
459 (39)
381 (5)
346 (32)
275 (4)
135 (0)
453 (160)
366 (5)
oop δR + ωNH2
δCCN
317 (103)
268 (6)
τCN
oop δR
oop δR
118 (1)
9
a Scaled with equations y = 0.9212 x + 138 and y = 0.9716 x + 8.5 for values above and below 2000
cm−1, respectively; y is the scaled vibrational wavenumber and x is the calculated harmonic vibrational
wavenumber; see text for details.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
b Predicted percentage IR intensities relative to the intensity of the ν9 line, 106 km mol−1, are listed in
parentheses.
c Predicted percentage IR intensities relative to the intensity of the ν9 line, 61 km mol−1, are listed in
parentheses
d. Percentage integrated IR intensities relative to the most intense line at 1614.9 cm−1 (ν9) are listed in
parentheses.
e Approximate mode descriptions. ν: stretch, δ: bend or deformation, δR: ring deformation, δs: scissor,
ρ: rock, ω: wag, t: twist, τ: torsion, a: antisymmetric, s: symmetric, oop: out-of-plane.
f Indeterminate because of significant interference from absorption of the parent or other lines.
g Unidentifiable definitively because of significant interference from absorption of the parent or other
lines.
22
ACS Paragon Plus Environment

Page 23 of 39
1
2
3
The Journal of Physical Chemistry
Table 3. Comparison of observed vibrational wavenumbers (cm−1) and IR intensities in group B with
scaled harmonic and anharmonic vibrational wavenumbers and relative IR intensities of para-
HC6H5NH2 predicted with the B3LYP/cc-pVTZ method.
4
5
6
7
8
9
calculations
scaled harmonica,b anharmonicc
experiment
i sym.
mode descriptione

d
p-H2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1 a′
2 a′
3397 (18)
3060 (5)
3029 (45)
2819 (181)
2796 (75)
1611 (157)
1601 (17)
1456 (100)
1424 (1)
1284 (113)
1174 (11)
967 (2)
3383 (10) 3405.0 (27) s-νNH2
3037 (35)
2983 (73)
s-(νC3H/νC5H + νC2H/νC6H
)
f
3 a′
–
s-(νC2H/νC6H) ‒ s-(νC3H/νC5H)
4 a′
2714 (282) 2775.1 (–)g s-νCH2
2705 (158) 2706.1 (9) a-νCH2
1608 (130) 1606.0 (64) δsNH2
5 a′
6 a′
7 a′
1587 (70) 1588.3 (17) s-νC2C3/νC5C6+ δsNH2
8 a′
1451 (100) 1455.7 (100) δR + δsCH2 + δC2H/δC3H/δC5H/δC6H
1414 (1) δsCH2
1286 (71) 1290.4 (71) νC1N +δR
9 a′
10 a′
11 a′
12 a′
13 a′
14 a′
15 a′
16 a′
17 a′
18 a′
19 a′
20 a′
21 a′
22 a′
23 a″
24 a″
25 a″
26 a″
27 a″
28 a″
29 a″
30 a″
31 a″
32 a″
33 a″
34 a″
1175 (27)
978 (5)
a-δC2H/δC3H + a-δC5H/δC6H
δR
938 (11)
919 (20)
865 (9)
937 (9)
962.5 (8) ρCH2 + oop-δC3H/δC5H
911.7 (12) oop (s-δC2H/δC3H ‒ s-δC5H/δC6H
δR
924 (34)
868 (22)
761 (32)
708 (90)
505 (812)
)
767 (5)
δR (breathing)
707 (122)
619 (641)
529 (116)
495 (1)
oop δC3H/δC5H
ωNH2 (inversion)
534 (234) 521.9 (66) oop δC2H/δC3H/δC5H/δC6H + ρCH2
493 (8)
282 (23)
119 (3)
δR
277 (17)
119 (3)
oop δR
oop δR + ρCH2
3489 (23)
3058 (69)
3028 (33)
1516 (9)
1392 (7)
1336 (0)
1265 (138)
1165 (0)
1152 (3)
1037 (47)
949 (56)
926 (1)
3466 (22) 3486.0 (29) a-νNH2
f
3020 (123)
3021 (88)
1507 (7)
–
–
a-νC3H/νC5H + a-νC2H/νC6H
a-(νC2H/νC6H) ‒ a-(νC3H/νC5H
a-νC2C3/νC5C6
f
)
1388 (13)
1333 (0)
δR + ωCH2
δC2H/δC3H/δC5H/δC6H + ωCH2
1254 (161) 1263.8 (70) δR + ωCH2
1159 (0)
1158 (4)
tCH2
a-δC2H/δC3H/δC5H/δC6H + ωCH2
1043 (51) 1035.8 (34) ρNH2
951 (59)
932 (0)
957.8 (18) δR + ωCH2
oop δC2H/δC3H/δC5H/δC6H
23
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 24 of 39
1
2
3
4
5
6
7
8
35 a″
36 a″
37 a″
38 a″
39 a″
699 (0)
604 (2)
389 (0)
362 (3)
258 (32)
705 (0)
606 (3)
386 (0)
364 (2)
295 (45)
oop δC2H/δC3H/δC5H/δC6H
δR
oop δR
δCCN
τCN
9
a Scaled with equations y = 0.9212 x + 138 and y = 0.9716 x + 8.5 for values above and below 2000
cm−1, respectively; y is the scaled vibrational wavenumber and x is the calculated harmonic vibrational
wavenumber; see text for details.
b Predicted percentage IR intensities relative to the intensity of the ν8 line, 46 km mol−1.
c Predicted percentage IR intensities relative to the intensity of the ν8 line, 30 km mol−1.
d Percentage integrated IR intensities relative to the most intense line at 1455.7 cm−1 (ν8) are listed in
parentheses.
e Approximate mode descriptions. ν: stretch, δ: bend or deformation, δR: ring deformation, δs: scissor,
ρ: rock, ω: wag, t: twist, τ: torsion, a: antisymmetric, s: symmetric, oop: out-of-plane.
f Unidentifiable definitively because of significant interference from absorption of the parent or other
lines.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
g Indeterminate because of significant interference from absorption of a line in group A.
24
ACS Paragon Plus Environment

Page 25 of 39
1
The Journal of Physical Chemistry
Table 4. Variation of spectral intensities of lines in groups A and B at each stage of experiments.
2
3
4
5
6
7
8
9
in darkness
19 h
secondary photolysis
465 nm
group
520 nm
375 nm
A
B
+(43±17)a % −(35±15)a % −(13±8)a % −(75±10)a %
+(50±17) % −(36±19) % −(30±19) % −(19±5) %
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a The percentage variations are the integrated intensities of lines relative to those in the IR spectrum
recorded just after matrix deposition (Figure 3(a)); values from several representative lines in each
group are averaged, as described in the text. The errors represent one standard deviation in averaging
several representative lines in each group.
25
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 26 of 39
1
2
Table 5. Estimated mixing ratios (in ppm) of isomers of protonated and hydrogenated aniline.
3
4
5
hydrogenated aniline
protonated aniline
6
7
8
9
isomer a
after
depositionb
2.0±0.9
2.5±0.9
–
19 h in
darknessc
+(0.8±0.4)
+(1.3±0.6)
–
after
19 h in
depositionb
0.6±0.1
1.0±0.1
1.0±0.5
2.6±0.7
darknessc
−(0.2±0.1)
−(0.3±0.1)
−(0.3±0.2)
−(0.9±0.4)
ortho-
para-
amino-
sumd
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.5±1.9
+(2.1±1.0)
a The mixing ratio of aniline was estimated to be (53±5) ppm after deposition; it was diminished by
(5.4±2.6) ppm after maintenance of the matrix in darkness for 19 h.
b Estimated from the spectra shown in Figure 3(a) and predicted harmonic IR intensities according to
the method of Tam and Fajardo;32 see the text for details. The errors represent one standard deviation
in averaging several representative lines of each species.
c Variations in mixing ratios after this experimental process are shown.
d Sum of mixing ratios of all observed isomers.
26
ACS Paragon Plus Environment

Page 27 of 39
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
Scheme 1. Mechanisms of the Birch reduction of substituted benzene; the substituent (X) is an
electron-donating group.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
27
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 28 of 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Geometries of (a) ortho-HC6H5NH2, (b) para-HC6H5NH2, (c) meta-HC6H5NH2, and (d)
ipso-HC6H5NH2 optimized with the B3LYP/cc-pVTZ method. The ZPVE-corrected energies (in kJ
mol−1) are listed in parentheses. For each species, bond lengths in Å are indicated on the left and bond
angles in degree on the right. Dihedral angles are indicated with Newman diagrams. para-HC6H5NH2
belongs to symmetry point group Cs and others belong to symmetry point group C1.
(double column)
28
ACS Paragon Plus Environment

Page 29 of 39
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Figure 2. Energy diagram representing the relative energies of hydrogenated aniline isomers (ortho-,
para-, meta-, and ipso-HC6H5NH2) and transition states for formation and isomerization. Blue lines
indicate hydrogen-addition reactions with barriers indicated in parentheses, and red lines indicate
isomerization reactions. Calculations were performed with the B3LYP/cc-pVTZ method; ZPVE were
corrected with unscaled harmonic wavenumbers. The relative energies are summarized in Table 1.
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(1.5 column)
29
ACS Paragon Plus Environment