UV Photochemistry of Acetylacetaldehyde Trapped in Cryogenic Matrices

Article abstract of DOI:10.1021/acs.jpca.0c02512

The broad band UV photochemistry of acetylacetaldehyde, the hybrid form between malonaldehyde and acetylacetone (the two other most simple molecules exhibiting an intramolecular proton transfer), trapped in four cryogenic matrices, neon, nitrogen, argon, and xenon, has been studied by IRTF spectroscopy. These experimental results have been supported by B3LYP/6-311G++(2d,2p) calculations in order to get S0 minima together with their harmonic frequencies. On those minima, we have also calculated their vibrationally resolved UV absorption spectra at the time-dependent DFT ωB97XD/6-311++G(2d,2p) level. After deposition, only the two chelated forms are observed while they isomerize upon UV irradiation toward nonchelated species. From UV irradiation effects we have identified several nonchelated isomers, capable, in turn, of isomerizing and fragmenting, even if this last phenomenon seems to be most unlikely due to cryogenic cages confinement. On the basis of these findings, we have attempted a first approach to the reaction path of electronic relaxation. It appeared that, as with acetylacetone, the path of electronic relaxation seems to involve triplet states.

Full text of DOI:10.1021/acs.jpca.0c02512

Subscriber access provided by Uppsala universitetsbibliotek
A: Kinetics, Dynamics, Photochemistry, and Excited States
UV Photochemistry of Acetylacetaldehyde Trapped in Cryogenic Matrices
Pierre Rousselot-Pailley, Sophie Sobanska, Nicolas Ferré, and Stephane Coussan
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.0c02512 • Publication Date (Web): 22 May 2020
Downloaded from pubs.acs.org on May 24, 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 57
The Journal of Physical Chemistry
H
O
O
O
O
1
2
3
4
5
R1
R2
R1
R2
6
7
8
9
1
1
1
1
1
1
1
1
1
1
H
H
H
0
1
2
3
4
5
6
7
8
9
Keto
Enol
R1 : H or CH3
ACS Paragon Plus Environment
R2 : H or CH
3

The Journal of Physical Chemistry
Page 2 of 57
CCTOH/CO
TTC
OH/CO
TTTOH/CO
R1
R2
R1
R2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
R1
R2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
CTTOH/CO
CTCOH/CO
R2
R2
CCCOH/CO
R1
R1
TCCOH/CO
TCTOH/CO
R1
R1
R2
R1
R2
R2
R = H or CH3
R = H or CH
1
ACS Paragon Plus Environment
2
3
R ≠ R
1
2

Page 3 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 4 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
1
60
40
20
HF
TCTco
TTTco
CTTco/CCTco
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MF
LF
CTCo
H
TCTo
H
100
TTTo
H
CTTo
H
/CCTo
H
TTCo
TCCco
H
8
6
4
0
0
0
TTCco/CTCco
TCCo
H
20
0
3
860
3850
3840
3830
3820
3810
3800
3790
-1
Wavenumbers (cm )
ACS Paragon Plus Environment

Page 5 of 57
The Journal of Physical Chemistry
7
00
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
CXX
6
00
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
TXX
500
4
3
00
00
TCCco
CXX
200
1
00
0
1
780
1760
1740
1720
1700
1680
1
1660
1640
1620
ACS Paragon Plus Environment
-
Wavenumbers (cm )

OH
C=O/C=C
The Journal of Physical Chemistry
Page 6 of 57
0
.4
Ar
a
Ar
HF
LF
0.5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
0.3
0.2
0.1
0
a
0
b
b
c
-0.5
c
-
0.1
0.2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
-
-1
0
0
0
.4
0
0
.6
.4
.3 N2
.2
N2
a
a
0.2
0
0.1
0
b
b
-0.2
c
-0.1
-
0.4
-0.6
0.8
-0.2 c
-
-0.3
0.3
Ne
a
Ne
a
0
0
.5
0
0
.2
.1
0
b
c
b
c
-0.5
-1
-0.1
1
0
0
0
0
.6
.5
.4
.3
Xe
a
Xe
a
0.5
0
b
c
b
c
0.2
-0.5
0
.1
ACS Paragon Plus Environment
3
680 3660 3640 3620 3600 3580 3560
1720 1700 1680 1660 1640 1620 1600 1580 1560
-
1
Wavenumbers (cm-1)
Wavenumbers (cm )

Page 7 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1.1
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
0
0
.9
.8
.7
0.6
Xe
Ar
0
0
0
0
.5
.4
.3
.2
Ne
N2
400
0.1
0
50
100
150
200
250
300
350
450
ACS Par
T
ago
i
n
m
Plus
e
Envi
(
ro
min)
nmen
t

The Journal of Physical Chemistry
Page 8 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
OD
Ne
N2
*
*
0
.5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
*
Ar
Xe
1700
1600
1500
1400
1300
ACS Paragon Plus Environment
Wavenumbers (cm )
1200
1100
1000
900
800
-
1

Page 9 of 57
The Journal of Physical Chemistry
Relative Energy
CCTOH
1
-1
kcal mol
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
+ 14.63
0
+
14
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
CCTCO
+
13.90
TCTCO
TCCOH
+
+
13
11
TTTOH
TCCCO
TCTOH
CTTOH
+11.92
+
11.61
+11.49
+
11.40
+11.10
CTCOH
+10.85
TTTCO
+
10.30
+
10
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CTCCO
TTCCO
CTTCO
TTCOH
+9.55
+9.58
+9.32
+9.16
+
9
Diketo
7.89
CCCOH
CCCCO
+
+0.13
0
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 10 of 57
PRP03-109BIS.1.fid
PROTON CDCl3 /opt/topspin3.2 pailley 30
9
8
0000
5000
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
AAD
80000
75000
AAD
AAD
7
6
6
5
0000
5000
0000
5000
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
50000
4
4
3
3
2
5000
0000
5000
0000
5000
20000
1
1
5
0
5000
0000
000
-5000
ACS Paragon Plus Environment
1
0.0
9.5
9.0
8.5
8.0
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
4
5
6
f1 (ppm)

Page 11 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 12 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 13 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
UV Photochemistry of Acetylacetaldehyde
Trapped in Cryogenic Matrices
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
‡
¶
∗,§
P. Rousselot-Pailley, S. Sobanska, N. Ferr ´e , and S. Coussan
†
Aix-Marseille Univ, Centrale Marseille, CNRS, iSm2 UMR 7313, Marseille, France
‡
Institut des Sciences Mol ´e culaires, Universit ´e de Bordeaux 1, CNRS UMR 5255, Talence,
France.
¶
Aix-Marseille Univ, CNRS, ICR, Marseille, France
§
Aix-Marseille Univ, CNRS, PIIM, Marseille, France
E-mail: stephane.coussan@univ-amu.fr
Abstract
The broad band UV photochemistry of acetylacetaldehyde, the hybrid form between
malonaldehyde and acetylacetone (the two other most simple molecules exhibiting an
intramolecular proton transfer), trapped in four cryogenic matrices, neon, nitrogen,
argon and xenon, has been studied by IRTF spectroscopy. These experimental re-
sults have been supported by B3LYP/6-311G++(2d,2p) calculations in order to get
S0 minima together with their harmonic frequencies. On those minima we have also
calculated their vibrationnally-resolved UV absorption spectra at the time-dependent
DFT ωB97XD/ 6-311++G(2d,2p) level. After deposition, only the two chelated forms
are observed while they isomerize upon UV irradiation toward nonchelated species.
From UV irradiation effects we have identified several nonchelated isomers, capable,
in turn, of isomerizing and fragmenting, even if this last phenomenon seems to be
most unlikely due to cryogenic cages confinement. Based on these findings, we have
attempted a first approach to the reaction path of electronic relaxation. It appeared
1
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 14 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
that, as with acetylacetone, the path of electronic relaxation seems to involve triplet
states.
Introduction
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Acetylacetadehyde (hereafter referred as AAD), is one of the most simple molecules, together
with malonaldehyde and acetylacetone (hereafter referred as MA and AA, respectively), ex-
hibiting an intramolecular proton transfer. From our past series of works in cryogenic matri-
ces1–4 and by comparison with supersonic jet works,5–7 we put in evidence that in cryogenic
matrix, the OH proton is localized, and that the main form present after deposition is the
enol one (Figure S1 and Tables 3 and S1). This form is stabilized by a strong internal H-bond,
−
1
8
calculated to be of ≈ 12 kcal mol , in the case of acetylacetone. In ”classical” cryogenic
matrices, such as nitrogen, argon, neon or xenon, one does not observe the tautomerization
9
between keto and enol forms (Figure 1), while Cr ´e pin et al induced it by UV (KrF laser)
excitation at 248 nm in parahydrogen. Due to the electronic delocalization, the enol forms
are rather rigid. However, upon UV irradiation, these forms, also called chelated forms or
CCC (see Figure 2), mainly isomerize toward open forms, CXX and TXX (see Figure S2,
X = C or T) in matrices,1–4 while they fragment in molecular beam.10–13 In Figure S2, the
1
6 possible AAD isomers are diplayed, where C and T characters stand for Cis or Trans
configurations with respect to C-C, C=C and C-O bonds, respectively. However, due to
the assymetry (compared to AA and MA) of AAD, the complexity increases because to one
MA/AA isomer corresponds two AAD’s, depending on H and -CH positions. OH and CO
3
underscripts stand for -CH position with respect to C=O or OH ”side” (Figure S2). AAD’s
3
litterature is really scarce. We have found two theoretical works, one at low level of the-
ory deals with isomers geometries and energies,14 while a second, a lot more sophisticated,
and uniquely dedicated to AAD, addresses also tunneling between the two chelated forms,
and tautomerization phenomena.15 Experimentally, a first study of AAD trapped in argon
2
ACS Paragon Plus Environment

Page 15 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
and xenon matrices has been published by our laboratory many years ago,16 and two NMR
ones.17,18 The first goal of the present work is to identify if both AAD CCCOH and CCCCO
forms are observed, whatever the matrix, after deposition, then, which are the nonchelated
isomers present after UV irradiation. Is there a clear difference between the matrices, i.e.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
can we assess if there is a control of the UV photochemistry by the matrix? The second goal
of this work is to approach electronic relaxation pathway, like we did for AA and MA.1–4
On the theoretical point of view, in the case of MA, namely, if numerous theoretical studies
have been published on its ground state tunneling effect,19–21 a lot fewer have addressed that
of AA,22 or MA and AA photochemistries.23–25 Comparing those latter studies with our
experimental results1–4 we showed that AA electronic relaxation pathway proceeds through
T and T states, and that nonchelated forms are those which fragment. On the contrary, in
2
1
MA case, electronic relaxation proceeds uniquely through singlet states and no fragmenta-
tion is observed. So, even the methyl rotators coupling is really weak in AA, this is precisely
what induces this electronic relaxation pathways difference with MA. The last goal of this
work was thus to guess what is the electronic relaxation pathway of this ”hybrid” molecule,
in other words, is it the methyl −methyl coupling that governs a relaxation process through
triplet states, or is it just the presence of a methyl group that induces this phenomenon?
Experimental
Acetylacetaldehyde Synthesis
The Acetylacetaldehyde was prepared as follow. To 10g of 4,4-Dimethoxy-2-butanone (Acros
organics 92%) (75.6 mmole) was added 100mL of HCL 0.5N. The solution was stirred for
72h. After 72h the solution was extracted with 3 × 25mL of methylene chloride. The
organic layer was dried over MgSO and evaporated under reduced pressure to afford 3g of
4
1
Acetylacetaldehyde as a yellow oil. The H NMR was recorded on BRUKER Avance III
nanobay - 300MHz equipped with a BBFO+ probe. Solvant was CDCl (Figure S3). The
3
3
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 16 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
−1
UV spectrum was recorded at 298 K (Figure 3), in dichloromethane ([AAD] ≈ 0.1 mM L ),
with an Agilent Technologies Cary 60 UV/Vis spectrometer, at 4800 nm/min, with a data
interval of 1 nm and an average time of 0.0125 s.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Matrix Experiments
For matrix isolation experiments, AAD was subjected to multiple freeze-pump-thaw cycles
under primary vacuum to remove dissolved gases. AAD was then mixed with matrix gases,
Ne (N50 grade, Air Liquide), Ar and N (N60 grade, Air Liquide), Xe (N50 grade, Linde),
2
in partial ratio AAD/MG (MG, matrix gas) of ≈ 7/1000. Mixtures were deposited onto a
gold-platted copper cube cooled to 4.3 K in Ne, 20 K in Ar and N , and 40 K in Xe, by a
2
closed-cycle cryogenerator PT-405 from Cryomech compagny. Cryostat and sample-carrier
were protected from thermal background radiation by a chrome-platted brass shield kept at
≈
32 K. Controlled heating (Lakeshore Model 336) of the sample carrier was carried out
with a 50 Ω resistor, while the background pressure in the vacuum chamber was kept at 10 7
−
mbar by a turbomolecular pump. Fourier transform IR (FTIR) spectra were all recorded at
−
1
4
.3 K in the reflection mode using a Bruker IFS 66/S spectrometer (resolution: 0.12 cm )
equipped with a MCT detector. Broad band UV irradiation were carried out with an Oriel
high-pressure 500 W mercury lamp (average power: 300 mW, no optical filter, no water
filter. For kinetics an average power of 100 mW was used).
Theoretical Calculations
We have carried out two types of theoretical calculations and compared them to those we
got in MA and AA cases:1–4 UV vertical transitions and DFT ground state minima together
with their harmonic frequencies sets. Both are presented below.
4
ACS Paragon Plus Environment

Page 17 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
UV Vertical Transitions
For each isomer, we start by optimizing its geometry in gas phase, using Kohn-Sham Density
Functional Theory (DFT), the ωB97XD range-separated exchange-correlation functional26
and the 6-311++G(2d,2p) basis set. Then, a linear-response Time-Dependent DFT (TD-
DFT) calculation is performed to select the lowest electronic excited state with a significant
oscillator strength, corresponding to a bright allowed transition which can be described by a
π∗ ← π single excitation. Vertical energy excitation is calculated as the difference between
the excited and the ground state electronic energies. The approximate vibrationally-resolved
absorption spectrum is obtained using the selected excited state nuclear gradient only, in the
Franck-Condon approximation.27 In other words, both the ground and the selected excited
states are assumed to share the same potential energy surface shape. Note that less crude
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
∗
approximations could not be used since the π ← π excited states do not feature a stable
energy minimum geometry (but rather a conical intersection or an avoided crossing with
a twisted C=C bond). The resulted absorption spectra have been plotted assuming a 500
−
1
cm half-width at half-maximum for the convolution.
Theoretical UV Spectra
All the theoretical UV spectra, including vertical excitation and 0-0 transition wavelenghts,
are displayed in Figures S4 and S5 and listed in Table 1 (we did not calculate vertical
transition of diketo form (displayed in Figure S2), because this latter would just present
∗
∗
forbidden π ← n ones. All the first allowed π ← π transitions are S ← S . As expected,
2
0
CCCCO/OH species present the highest wavelenghts because they are the most conjugated
ones. In the case of the non-chelated species there is a clear gap between the XXXCO
and XXXOH character. Indeed, as clearly displayed in Figures S4 and S5, the XXXCO
species present systematically a vertical absorption wavelenghts ≈ 10 nm lower than those
of XXXOH. It is due to the fact that -CH is electro-donor while -C=O is electro-attractor.
3
As a result when -CH is in close vicinity with -C=O (XXXCO species), this electro-donor
3
5
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 18 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
effect is reduced by the attractivity of -C=O, reducing the conjugation and then inducing
a shorter vertical excitation value for those latter. It is then convenient to discriminate
between three groups:
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
. CCCCO and CCCOH with allowed transitions at 241 and 248 nm, respectively.
. XXXCO nonchelated species with allowed transitions ranging between 209 and 221 nm.
. XXXOH nonchelated species with allowed transitions ranging between 215 and 232 nm.
Even the theoretical method used in this work is not the one we used for MA and AA,1,3
we get comparable vertical excitation values for AAD XXXOH, MA and AA isomers. It is
due to the fact they present comparable conjugated systems. We can then use confidently
the same UV broad band irradiation experimental process to discriminate between all the
XXXCO/OH acetylacetaldehyde isomers, because all these species should present close UV
experimental spectra.
DFT Calculations
Calculations have been performed at the B3LYP/6-311++G(2d,2p) level of calculation, us-
ing the GAUSSIAN 16 computer code28 and the basis set 6-311++G(2d,2p) of Pople et
al.29,30 Becke’s31,32 three-parameters functional was used, including the gradient-exchange
correction and the non local correlation function of Lee, Yang and Parr.33
In Figure S2 is displayed all the minima with their relative energies compared to that of
CCCOH one. To each MA or AA nonchelated isomers correspond two AAD ones, depending
on terminal -H and -CH group positions. We have chosen to call them XXX and XXXOH
3
CO
with respect to the methyl position, C=O side or O-H one. If we compare those relative
1
34
energies with those found, at the same level of calculation, for MA or AA, we notice that
AAD presents an intermediate situation between MA, for which TTC is located 6.7 kcal
−
1
−1
mol above CCC, and AA for which CTC is located 11.1 kcal mol above CCC (TTC and
6
ACS Paragon Plus Environment

Page 19 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
CTC are the most stable nonchelated isomers with respect to their relative CCC MA and
AA parents molecules, respectively). In the same trend, the highest nonchelated energy is
seen to be those of CCT(MA) at 12.8, CCTOH(AAD) at 14.6 and TTT(AA) at 16.6 kcal
−1
mol , respectively. Is it so surprising? It seems that the energy gap with respect to the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
chelated forms, increases with the number of methyl groups. Those latter induce a larger
distortion in AA case, namely for TTT and TTC which exhibit a syn-1,3 methyl-methyl
2
3
interaction, while CCT, because of sp /sp oxygen lone pairs interaction is the most ener-
getically disfavoured isomer in both MA and AAD cases.
Considering our past works on AA and MA,1,2 we should be able to observe all the nonchelated
2
3
forms excepted CCTOH/CO. Indeed, those two latters display a sp /sp oxygen lone pairs
interaction which avoids their observation, at our recording timescale (s), before they relax
towards more stable ones. However, from calculated frequencies, namely in the νOH and
νC=O/νC=C regions (those in which isomers present clear spectral differences) (Table 2), we
can guess that many bands will overlap, rendering uneasy an unambiguous vibrational as-
signment to one or another isomer. A first step will be to identify groups of bands belonging
to CXX or TXX species.
Theoretical IR Spectra
As in the case of MA and AA,1,2 we have chosen νOH and νC=O/νC=C (Table 2) regions
because non-chelated isomers present, in those ones, typical sets of bands. Studying those
regions would allow us to discriminate between CXX and TXX species.
νOH Region (Figure 4). CCCOH/CO νOH modes are calculated at 3218.3 and 3215.1
−1
cm , respectively, with normalized intensities given to be 0.0. Adding the fact, that as in
the case of MA and AA,1,2 due the strength of the internal H-bond, these bands should be
spread over hundreds of wavenumbers, we do not expect to observe them. With regards to
nonchelated isomers, we can conveniently discriminate between three groups:
8
1
. A high frequency (HF) group containing TCTCO, TTTCO, CTTCO, CCTCO and TCTOH,
7
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 20 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
−1
with bands centered at 3858.0, 3854.8, 3849.4, 3848.7 and 3840.3 cm , respectively.
2
. A medium frequency (MF) group containing TTTOH, CTTOH and CCTOH, with bands
−
1
centered at 3827.4, 3821.1 and 3821.8 cm , respectively.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3. A low frequency (LF) group containing TTCOH, TTCCO, TCCOH, TCCCO, CTCOH
and CTCCO with bands centered at 3797.5, 3793.8, 3791.8, 3800.1, 3795.1 and 3794.5
−
1
cm , respectively.
It is interesting to note that couples TTCOH/CO, TTTOH/CO, CTTOH/CO and CCTOH/CO are
spread over HF and MF regions, CO character being only HF while OH is MF. The three
last couples TTCOH/CO, TCCOH/CO and CTCOH/CO, are uniquely LF. Those latters present
frequency splits (frequency gap between OH and CO νOH modes of each AAD isomer) of ≈
−
1
4
(TTC), 8 (TCC) and 1 (CTC) cm , while between HF and MF bands, the splits are of
≈
27 (TTT), 18 (TCT), 28 (CTT) and 27 (CCT). As a result it could be uneasy to identify
unambiguously LF species because of the experimental overlapping.
νC=O/νC=C Region (Figure 5). CCCOH and CCCCO νC=O/νC=C modes are calculated
−
1
−1
−1
−1
at 1646.2 and 1679.5 cm (∆ν = 33.3 cm ), and 1630.0 and 1683.4 cm (∆ν = 53.4 cm ),
respectively. As in the case of MA and AA,1,2 νC=C mode is calculated to be higher than
that νC=O. It is a direct consequence of the strong internal H-bond which induces this strong
νC=C blue shift. The average normalized intensity of these modes is ≈ 0.45. They should be
the most intense bands of the spectrum. For nonchelated species, this region is one of the
most salient to discriminate among themselves. As for CCC, those bands should be the most
intense of their spectra. However, due to the existence of possibly non less than 16 isomers,
−
1
i.e 32 bands, spread over 130 cm , it could result a strong experimental overlapping. As in
the νOH region we can conveniently separate them by their ∆ν = |νC=O − νC=C| (Table 2)
and the relative positions of their bands.
1
. The first groups contains CTTOH/CO, CTCOH/CO and CCTOH/CO couples which should
−1
−1
display ∆ν ranging from 87.0 cm to 116.4 cm . As displayed in Figure 5, and in
8
ACS Paragon Plus Environment

Page 21 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Table 2, the bands of this group are supposed to surround those of the others non-
chelated isomers.
2
. The second group contains TCCOH/CO and TTCOH/CO couples which should display
−
1
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
∆
ν of 55.2, 39.1, 63.9 and 54.6 cm , respectively.
3. The third group contains TCTOH/CO and TTTOH/CO couples which should display ∆ν
−
1
of 24.4, 12.1, 28.5 and 9.2 cm , respectively.
Even the study of these regions should be enough to discriminate between non-chelated
isomers, others typical sets of bands could be used to support the vibrational assignments,
−
1
namely in the 1500-900 cm range (for example CTCCO should display a normalized inten-
−
1
sity (on CTCOH νC=C mode) 0.5 band centered at 1140.9 cm , and TTT a 0.5 normalized
OH
−
1
intensity band centered at 1207.8 cm ).
Experimental Results and Discussion
UV spectroscopy
The experimental spectrum of AAD in dichloromethane at 298 K is displayed in Figure
3
. One is able to observe only one band centered at 269 nm, with a shoulder at 260 nm,
which should be those of CCCOH and CCCCO species, respectively, almost the only forms
present before UV irradiation, in cryogenic matrices1,2 and in solution. However, we cannot
discard the presence of a few amount of diketo form even this form is energetically strongly
−1
disfavoured with respect to the enol one. Considering that 1 kcal mol ≈ 503.2 K, the
thermal bath at 298 K does not bring enough energy to isomerize CCC species toward
diketo form or the nonchelated ones, the closest being calculated to range between 9 and 10
−1
kcal mol above CCC species (Figure 2). Unfortunately we have not been able to record
this spectrum at 4 K, in the four matrices. As a result we are not able to present nonchelated
species one. However, one can retrieve some insights from this spectrum. From theoretical
9
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 22 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
results, one could expect for CCC species two bands of almost the same intensity separated by
about 6 nm. Indeed, the two CCC species are almost the same energy (Figure 2), and present
∗
almost the same oscillator strength for their two first allowed S ← S /π ← π transitions
2
0
(
Table 1), i.e. 0.2353 against 0.2717 for CCCCO and CCCOH, respectively. Experimentally
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
we observe a 9 nm difference between the top of the band and the shoulder. Is it so surprising
that expecting two isomers, we observe a quite symetrical band? MA and AA are known
to exchange the enol proton in gas phase and in liquid, leading to a splitting of the ground
state.5,6,19–21,35,36 It could be what happens here: we observe a kind of average situation
in which proton is exchanged and concomitantly it remains enough forms of each CCC
isomer to structure the band with this shoulder. It should also be noted that even this
UV spectrum has been recorded in dichloroethane, at room temperature, the 269 nm band
for CCC matches with those of MA and AA chelated forms, at 269-275 and 265-269 nm,
respectively.1,2
IR spectroscopy - Broad band UV irradiations
The effects of UV broad band irradiation in the four cryogenic matrices are displayed in
Figure 6. The full vibrational spectrum of CCCXX chelated isomers in the four matrices is
given in Figure S1. Neither νOH region is displayed because these bands are not observable
−
1
(
see text below), nor νCH which presents weak bands ranging between 3090 and 2650 cm .
Experimental frequencies of the two chelated forms are gathered in Tables 3 and S1.
Argon Matrix. Before irradiation, one observes no νOH CCC band, because these bands,
−
1
calculated at 3218.3 and 3215.1 cm , for CCC
and CCCCO, respectively, are certainly
OH
spread over hundreds of wavenumbers,37 and their normalized intensities are calculated to
−
1
be 0 (Table 2). It is due to the internal H-bond strength, calculated to be 12 kcal mol , in
8
the AA case. In the other hand, broad multiplets are observed at 1648.7-1647.0-1641.5 and
−
1
1
608.2-1602.8-1594.3-1591.1 cm . If we consider the most intense band of each multiplet,
10
ACS Paragon Plus Environment

Page 23 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
−1
−1
i.e. 1647.0 and 1591.1 cm , it leads to a ∆ν of ≈ 56 cm . Comparing this latter ∆ν, with
those theoretical (Table 2), we can assign it to CCCCO, while those at 1641.5 and 1608.2
−1
−1
cm (i.e. ∆ν ≈ 33 cm ), which are surrounded by the two formers, could be assigned
to CCCOH ((Table 3). After 500 min of irradiation, one observes new band groups in the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−1
νOH region: one, HF-MF, presents bands ranging between 3660 and 3630 cm , the other,
−1
LF, between 3605 and 3590 cm . These bands are due to nonchelated isomers. In the
νC=O/C=C region, one observes multiplets growing with strongest bands located at 1711.9,
−1
1
707.8, 1695.3, 1688.5, 1625.4, 1617.0 and 1615.2 cm (Tables 4 and 5). At this stage,
it is impossible to discriminate between nonchelated isomers candidates for these spectral
features. However, if we consider the position of these multiplets we can confidently assume
the presence of CXX and TXX species. But which ones? Are we able to go further into the
identification of those nonchelated isomers? One observes that after 60 min irradiation, some
nonchelated bands start to decrease in favor of other nonchelated ones (there could also be a
back conversion toward CCC, but as it continues to be consumed, we are unable to identify
it). In Figure 6, Ar (trace (c)) shows difference between spectrum after 500 min irradiation
minus spectrum after deposition, when photostationnary state is reached, and spectrum after
5
00 min irradiation minus after 60 min irradiation (trace (b)), i.e. time from which inter
nonchelated isomerization is observed. In the νOH region 3640.0, 3636.2, 3602.4 and 3597.3
−1
−1
cm bands decrease while 3649.8, 3649.1, 3600.3, 3599.4 and 3598.3 cm increase. In
the νC=O/C=C region, two groups of bands decrease, the first group contains 1712.2, 1707.8,
−1
1
1
695.5 and 1691.4 cm bands and the second one contains those centered at 1625.2, 1621.1,
−1
616.9 and 1615.3 cm . These decreases are counterbalanced by four sets of increasing bands
at 1689.3, 1688.6, 1686.6, 1685.2, then 1676.8, 1675.0, then 1667.7, 1666.3, then 1640.0 and
−1
1
638.4 cm . Therefore, decreasing and increasing nonchelated isomers present following
spectral features:
1. Decreasing νOH bands are MF and LF while decreasing νC=O/C=C ones present ∆ν
−
1
ranging between 68 and 97 cm . Moreover, those latters surround increasing ones.
11
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 24 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
. Increasing νOH bands are HF and LF while νC=O/C=C ones present multiplets, whose
−1
average positions are peaked at 1687, 1676, 1667 and 1639 cm , i.e. separated by ∆ν
−1
ranging from 9 to 48 cm . This distribution fits well with that theoretical displayed
in Figure 5. From Table 2 and Figure 5, we can confidently assume that these νOH
and νC=O/C=C increasing bands are due to TXX species.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The combination of MF and LF νOH decreasing bands with νC=O/C=C ∆ν ranging between 68
−
1
and 97 cm leads us to assign those signals to CXX, TCCOH and TTTOH species. Indeed,
with ∆ν ranging between 68 and 97 cm 1 with bands surrounding the other nonchelated
ones, the decreasing species are unambiguously CXX with the exception of TCCOH and
TTTOH (Table 2). However, we can discard the presence of CCT species because of the
−
3
2
strong sp − sp electron lone pairs repulsion (Figure S2), and by the fact we do not observe
νC=O bands blue-shifted by ≈ 3 and 9 cm− with respect to that at 1712.2 cm (tenta-
tively assigned to CTTCO, see below). This assumption is highly supported by the fact CCT
species have never been observed in our past works on AA and MA.1,2 We assign the two
decreasing bands at 1695.5 and 1691.5 cm 1 to TCCOH and TTTOH, respectively, because
they belong to typical TXXOH multiplets, and because of the two decreasing OH bands at
1
−1
−
−
1
3
640.0 (MF) and 3597.3 (LF) cm which are also TXX. In one hand, there is only one TXX
MF species (which should display a blue-shifted νOH band with respect to CTTOH, ∆ν ≈ 6
−
1
−1 −1
frequency at 1691.5 cm , red-shifted by about 4 cm with respect to
cm ), with a ν
C=O
the highest νC=O TXX band, it is TTTOH. In the other hand, the lowest calculated νOH
frequency is that of TCCOH which also presents the highest (TXX) νC=O one. This is why
we have assigned the 3597.3 and 1695.5 cm 1 bands to TCCOH. In the same way, CTTCO
−
species should present a HF νOH band that we do not observe because the TXX HF bands
−
1
overlap it, but the presence of the band at 1712.2 cm , the highest we observe in the ν
C=O
region clearly indicates its presence. Another strong argument in favor of the identification
of those six isomers is the fact they are calculated to absorb at roughly the same wave-
lenght (Table 1), c.a. 232, 221, 230, 219, 231 and 227 nm for CTTOH, CTTCO, CTCOH
,
12
ACS Paragon Plus Environment

Page 25 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
CTCCO, TCCOH and TTTOH, respectively. It makes thus sense, that if these species absorb
at roughly the same wavelenght, they should display the same behaviour upon UV irradia-
tion. What about TCTOH and TTCOH which are calculated to absorb both at 221 nm? We
should also see them decrease. From Table 2, they present almost degenerated νCO mode
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−1
with TTTOH (1729.7 and 1728.4 cm , respectively) while they should be HF and LF in
the νOH region, respectively. If one considers that XXXOH species (mainly) are consumed
from 60 min irradiation, and the fact XXXCO ones present UV transitions lower than those
former (excepted CCTCO, CTCCO and CTTCO, Table 1), we should mainly observe inter
XXXOH → XXXOH and XXXOH → XXXCO isomerizations. Moreover, at the same time
CCCOH/CO is still consumed giving XXXOH/CO isomers. One relevant example is the obser-
−1
vation of multiplet growing at 1689.3, 1688.6 and 1686.6 cm , which is typical of TXX
OH
isomers. This is why we can assign this multiplet to TCTOH, which is νOH HF (calculated
to be blue-shifted by ≈ 19 cm−1 with respect to CTTOH, against 13 cm experimentally)
and to TTCOH which is νOH LF (calculated to be blue-shifted of ≈ 6 cm−1 with respect to
−1
−1
TCCOH, against 3 cm experimentally). The two growing multiplets peaked at 1676 and
−1
1
667 cm should be associated to TTTCO and TCTCO while TTCCO and TCCCO present
−1
degenerated νC=O modes with those two latters (multiplet at 1676 cm ). The last multiplet
−1
peaked at 1639 cm can be associated with TTCCO and TTCOH. To summarize the UV
broad band irradiation effects observed in Ar matrix, CCCOH/CO chelated forms are the only
forms present after deposition, while upon broad band UV irradiation, those latters isomerize
toward all nonchelated forms excepted CCTOH/CO species which are not identified (at our
timescale). We could suggest some vibrational assignments, keeping in minds that in argon
like in nitrogen there could be also site splittings, enhancing spectral congestion (even if we
did not encounter those typical spectral changes in the present study), but to go deeper into
this identification, selective UV and IR irradiations are needed. At least, it should be noted
that we did not observe a back isomerization toward CCCOH/CO by matrix annealing.
13
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 26 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N matrix. If UV broad band effects, displayed in Figure 6, are comparable to those
2
observed in argon matrix, i.e. CCC species are isomerized toward nonchelated ones, vi-
brationnal assignment of those latters is more subtle because of the well-known trapping
site multiplicity (especially we did not anneal the sample after irradiation). However, by
analogy with argon matrix, after deposition there are only CCC isomers in nitrogen matrix.
Following the same reasoning than in argon, we can confidently assign bands at 1650.0 and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−1
−1
1
597.4 cm (∆ν ≈ 55.3 cm−1 against 53.4 cm theoretically, Table 2) to CCCCO and those
−1
(∆ν ≈ 41.1 cm−1 against 33.3 cm−1
at 1641.7-1673.9-1641.7-1604.7-1602.8 and 1600.7 cm
theoretically, Table 2) to CCCOH. After broad band UV irradiation, one observes CCCOH/CO
bands decrease counterbalanced by the increase of those nonchelated. In the νOH region a
−1
broad HF-MF multiplet ranging between 3638 and 3598 cm and a LF one between 3590
−1
and 3568 cm , grow. Both multiplets present the same shape, a high frequency side with a
discrete structure while the low frequency part presents a kind of broad unstructured foot.
Unlike in argon, one observes nonchelated consumption from 25 min of irradiation (against
from 60 min in argon), but one observes also site exchanges rendering unambiguous vibra-
tionnal assignments more difficult. Nevertheless, keeping in minds that excepted CCTCO
,
CTCCO and CTTCO, it is the XXXOH species that are likely to interconvert, we can suggest
some identifications. So, after 25 min irradiation, one observes one LF decrease at 3580.9
−1
−1
cm while two LF bands increase at 3585.8 and 3583.5 cm , and, two HF decreases at
−1
3
3
629.1 and 3624.9 cm against five increases at 3633.9, 3632.9, 3631.0, 3628.3, 3627.1 and
−1
623.1 cm . Concomitantly in the ν
region one observes decrease of the multiplets
C=O/C=C
−1
at 1715.0-1711.0-1708.2 and 1619.6-1618.1-1616.8 (i.e ∆ν ≈ 93 cm ) and of one band cen-
−1
−1
tered at 1635.7 cm (i.e ∆ν ≈ 76 cm ). We also observe increases of multiplets peaked at
−1
−1
1
685.5 and 1672.9 cm (i.e ∆ν ≈ 13 cm ). Considering this last ∆ν is typical of TTT
CO
and TCTCO species (Table 2) coupled to the observation of HF increases at 3633.9, 3632.9
−1
and 3631.0 cm , i.e. the most ”blue” ones, also compatible with these two species, we can
−
1
assign those signals to them. As for the two HF decreasing bands at 3629.1 and 3624.9 cm
14
ACS Paragon Plus Environment

Page 27 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
−1
coupled to the observation of the decrease at 1715.0 cm , they only can be associated with
CTTCO isomer. Concerning the LF assignments, by analogy with argon matrix the 3580.9
−1
cm decreasing band, being the most ”red” of the LF ones could be assigned to TCCOH
−1
especially since we observe weak decreases at 1691.7 cm , in the neighborhood of the one
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−1
we assigned to this isomer in argon (1695.5 cm , Table 2). However, in this last region,
−1
one observes other decreasing band ∆ν of ≈ 93 cm , what is only compatible with CXX
species. There is unfortunetly no decisive help from the other spectral regions, we are thus
led to tentatively suggest two options: considering the νC=O/C=C ∆ν, and the fact decreasing
bands are surrounding the other nonchelated ones, it is impossible to discard the presence
of CTCCO and/or CTCOH isomers but the decrease of their νOH bands could be hidden by
the increases. At the same time, seeing no CXX increases, it means that the 3585.8 and
−1
3
583 cm bands are due to TXX isomers. Since, in HF region we identified the decrease
−1
of CTTCO, and that the 3580.9 cm is potentially that of TCCOH, we could assign 3585.8
−1
and 3583.5 cm bands to TCCCO, TTCOH or TTCOH, or to two sites of the same isomer.
Ne matrix. In this matrix, which is the softest among the four we used, we observe the
same scenario, i.e. upon UV broad band irradiation, CCCOH/CO isomers isomerize toward
−1
nonchelated ones (Figure 6). As in the two former matrices, bands at 1653.3 and 1597.7 cm
−1
(
∆ν ≈ 55.6 cm−1 against 53.4 cm theoretically, Table 2) are assigned to CCCCO and those
at 1647.9-1643.7 and the broad one at 1609.9 cm (∆ν ≈ 35.1 cm−1 against 33.3 cm the-
−1
−1
oretically, Table 2) to CCCOH. Upon broad band UV irradiation one observes grows, in the
−1
νOH region, of weak and broad HF-MF bands centered at 3678.9, 3670.9, 3655.6 cm and of
−1
a LF multiplet ranging between 3630 and 3610 cm with two maxima at 3623.0 and 3619.7
−1
cm . In the ν
region, one observes the grow of bands roughly centered at 1717.0,
C=O/C=C
1
700.0, 1697.0, 1694.2, 1690.5, 1686.1, 1678.4, 1668.9, 1642.1, 1640.2, 1634.9, 1626.7-1625.3
−1
and 1618.0 cm . A closer look to these bands reveal they exhibit more complex structures.
The fact is also that neon matrix contains water pollution which allows us to discard, for
15
ACS Paragon Plus Environment

The Journal of Physical Chemistry
Page 28 of 57
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
−1
example, the band at 1630.7 cm as possible AAD isomer candidate. It is a water polymer
band.38 As in the other matrices, one observes nonchelated band decreases after 45 min
irradiation (Figure 6), i.e. isomerizations among nonchelated species. In the νOH region the
−1
most clear decrease is observed at 3617.9 cm , i.e. the red foot of LF multiplet. By anal-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ogy with Ar and N matrices, we assign this band to TCCOH. This decrease is correlated to
2
−1
decrease at 1702.0 cm in the νC=O/C=C region. By analogy with Ar matrix we also suggest
−1
to assign the decreasing band at 1699.6 cm to TTTOH. In the νOH region, after 45 min
−1
irradiation, one observes weak increases in the LF bands at 3622.4 cm , and at 3678.5 and
−1
3
670.7 cm , in the HF ones. At the same time, the only nonchelated bands (excepted those
already assigned to TTTOH and TCCOH) we observe decreasing in the νC=O/C=C region at
−1
1
716.9, 1626.7-1625.3 and 1618.0 cm are spaced by ≈ 92 and 100 cm−1 respectively, i.e.
are typical of CXXOH/CO species. As a result, even we do not clearly observe CXXOH/CO
νOH band decreases, the fact is that like in Ar matrix, those latter species are consumed. We
suggest, by analogy with this former matrix, to assign the increasing LF band to TTCOH
−1
(
3622.4 cm ). Concerning the increasing HF and MF ones, at 3678.6, 3670.9 and 3655.6
−1
cm , even their intensities are really weak, they are already observable after 45 min irradi-
ation, i.e. they are TXX, because CXX are consumed. By comparison with Ar matrix, and
from Figure 4 and Table 2, we suggest to (tentatively) assign them to TCTCO and TTTCO
−1
−1
−1
(3655.6 cm ), respectively.
OH
(
3685.7-3678.6 cm ), TCTOH (3670.9 cm ) and TTT
Xe matrix. As in the three other matrices, upon UV broad band irradiation one observes
CCCOH/CO toward nonchelated species isomerization. To the contrary of what we observe in
the former matrices, whether in νOH or νC=O/C=C region, spectra are more structured, bands
are very sharp and then less in overlapping. Moreover, about 95 % of photochemical effect
−1
(∆ν ≈ 57.7 cm−1
is achieved after only 5 min irradiation. Bands at 1643.9 and 1586.2 cm
−1
against 53.4 cm theoretically, Table 2) should be associated with CCCCO while those at
−
1
−1
1
636.6 and 1611.0-1601.0 cm (∆ν ≈ 25.6−35.6 cm−1 against 33.3 cm theoretically, Table
16
ACS Paragon Plus Environment

Page 29 of 57
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
2
3
) should be assigned to CCCOH. In the νOH domain, grow of LF bands at 3581.4, 3578.6,
−1
574.6, 3569.4, 3566.9 and 3563.2 cm is almost saturated after 5 min irradiation while those
−1
HF at 3612.9 and 3610.5 cm grow during all the photochemical process. By observation of
νC=O/C=C region, one observes again isomerization between nonchelated species from 25 min
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
irradiation. In this last domain one observes main grows at 1704.8-1703.7, 1685.6, 1683.8,
−1
1
674.4, 1662.6, 1656.0-1654.7, 1639.0-1637.3, 1611.2, 1607.6 and 1586.2 cm . The main
−1
−1
−1
decreases are at 1704.8 and 1611.2 cm (∆ν ≈ 93.6 cm ), and 1607.6 cm (∆ν ≈ 97.2
−1
cm ). Those last ∆ν are typical of CXX species, while the HF ν
increasing bands are
OH
−1
typical of TXX ones. Decreases are also observed at 1683.8, 1655.8-1654.7 and 1639.0 cm ,
−1
and increases at 1685.6, 1674.4-1673.2 and 1662.4 cm which are TXX species (with the
−1
exception of the 1639.0 cm band which could be CTTOH νC=C mode). Concomitantly,
−1
in the νOH region, one observes increases at 3612.9, 3610.5 and 3581.4 cm . Considering
that the only nonchelated species which could be consumed upon this UV irradiation are
CTTCO (HF, 221 nm), TCTOH (HF, 231 nm), TTTOH (MF, 231 nm), CTTOH (MF, 232
nm), CTCOH (LF, 230 nm) and TCCOH (LF, 227 nm), we suggest to assign 3612.9, 1674.3
−1
and 1662.6 cm bands to TCTCO, and 3610.5, 1673.2-1671.0 bands to TTTCO. We also
−1
suggest to assign decreasing bands at 1683.8 and 1654.7 cm bands to TTTOH. The 1639.0
−1
−1
cm band does not match a TXX one while it is located 28 cm above CXX one at 1611.3
−1
−1
cm . It could be associated with CTTOH. The increasing bands at 3581.4 and 1685.6 cm
could only be associated with TTCOH, because this is the only TXX isomer which presents a
νC=O mode is this region, a LF νOH band and should not be consumed upon UV broad band
irradiation (215 nm). Concerning the LF νOH decreasing bands at 3574.8, 3573.3, 3569.0,
−1
3
566.7 and 3563.2 cm they can only be associated with CTCOH and TCCOH. Indeed,
as mentioned above, they are the only LF species able to isomerize upon UV broad band
irradiation. This fact, in the case of CTCOH, is comforted by the observation of ∆ν ≈ 95
−1
cm in the νC=O/C=C region.
17
ACS Paragon Plus Environment