Acylation of dimethyl maleate photocatalyzed by decatungstate anion: insights into the hydrogen atom transfer reaction mechanism

Article abstract of DOI:10.1007/s11164-017-3214-y

Polyoxometalates arise as significant catalysts in the field of organic chemistry due to their diverse properties and functions. Recent progress based on experimental evidence and density functional theory (DFT) calculations provides valuable information to demystify the chemistry of decatungstate anion, W₁₀O₃₂ ^4?. Particularly, functionalization of aldehydes by homolytic C–H bond cleavage can be efficiently achieved when it is catalyzed by this polyoxometalate. Two reaction mechanisms have been formulated to account for the role of W₁₀O₃₂ ^4? in organic chemical reactions: the single electron transfer and the hydrogen atom transfer (HAT) mechanisms. In this contribution, the HAT pathway for the acylation of dimethyl maleate is experimentally and quantum-chemically explored in detail. Results based on DFT calculations under the unrestricted formalism suggest that the acylation occurs in a barrierless process upon the formation of the lowest-in-energy triplet excited state of W₁₀O₃₂ ^4?. These outcomes agree well with the experimental evidence since the acylated adduct was produced at a 90% yield; in this regard, side reactions like radical couplings and decarbonylation resulted in less competitiveness. The current work may therefore help in the comprehension of the mechanistic details leading to the synthesis of organic compounds photocatalyzed by polyoxometalates, even under solar radiation.

Full text of DOI:10.1007/s11164-017-3214-y

Res Chem Intermed
https://doi.org/10.1007/s11164-017-3214-y
Acylation of dimethyl maleate photocatalyzed
by decatungstate anion: insights into the hydrogen atom
transfer reaction mechanism
1
Juan Pablo Martınez Ernesto Rivera-Avalos2
•
•
´
´
Sarai Vega-Rodrıguez Denisse de Loera2
2
•
Received: 25 September 2017 / Accepted: 17 November 2017
Ó Springer Science+Business Media B.V., part of Springer Nature 2017
Abstract Polyoxometalates arise as significant catalysts in the field of organic
chemistry due to their diverse properties and functions. Recent progress based on
experimental evidence and density functional theory (DFT) calculations provides
4-
valuable information to demystify the chemistry of decatungstate anion, W10O32
.
Particularly, functionalization of aldehydes by homolytic C–H bond cleavage can be
efficiently achieved when it is catalyzed by this polyoxometalate. Two reaction
4-
mechanisms have been formulated to account for the role of W10O32 in organic
chemical reactions: the single electron transfer and the hydrogen atom transfer
(HAT) mechanisms. In this contribution, the HAT pathway for the acylation of
dimethyl maleate is experimentally and quantum-chemically explored in detail.
Results based on DFT calculations under the unrestricted formalism suggest that the
acylation occurs in a barrierless process upon the formation of the lowest-in-energy
triplet excited state of W10O324-. These outcomes agree well with the experimental
evidence since the acylated adduct was produced at a 90% yield; in this regard, side
reactions like radical couplings and decarbonylation resulted in less competitive-
ness. The current work may therefore help in the comprehension of the mechanistic
details leading to the synthesis of organic compounds photocatalyzed by polyox-
ometalates, even under solar radiation.
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11164-
017-3214-y) contains supplementary material, which is available to authorized users.
&
Denisse de Loera
atenea.deloera@uaslp.mx
1
2
´
´
Escuela Superior de Fısica y Matematicas, Instituto Politecnico Nacional, Edificio 9, U.P.
A.L.M., Col. San Pedro Zacatenco, 07738 Ciudad de Mexico, Mexico
´
´
´
Facultad de Ciencias Quımicas, Universidad Autonoma de San Luis Potosı, Av. Dr. Manuel
Nava No. 6, Zona Universitaria, 78210 San Luis Potosı, Mexico
´
´
´
123

´
J. P. Martınez et al.
Keywords Polyoxometalates Á TBADT Á Acylation Á Photocatalyzed Á DFT
Introduction
Chemical functionalization of C–H bonds is an ambitious and challenging task.
Although radicals (R•) can be electrochemically [1] and thermally [2, 3] generated
[4], there is an alternative that provides an efficient strategy to form R• by means of
polyoxometalates [5–8]. Accordingly, the homogeneous photocatalyst tetrabuty-
lammonium decatungstate has been demonstrated to successfully catalyze the
alkylation of a,b-unsaturated ketones or alkenes, as well as other functional groups
[9–12]. Examples encompass the functionalization of inactivated C–H bonds of
organic compounds like alkanes, although more recent examples include the
functionalization of amides [13] and aldehydes [14, 15]. Moreover, carbon
nanostructures like fullerenes have been also used in these chemical transformations
as radical traps [16], thus shedding light on the increasing interest in decatungstate-
based photocatalysis in organic synthesis. Additionally, this strategy can be also
applied in the photochemical degradation of pollutants contained in water, resulting
in a growing interest in the subject [17].
Several studies provide mechanistic details for the catalytic transformation of C–
H bonds. The key steps are illustrated as follows [18–21]:
4À
3
W10O432À þ hv ! W10O Ã ! wO
ð1Þ
ð2Þ
ð3Þ
ð4Þ
32
RH þ wO ꢀ W10O532ÀHþ þ Rꢀ
Rꢀ þ C ¼ C ! R À C À Cꢀ
R À C À Cꢀ þ W10O35À2 Hþ ! R À C À C À H þ W10O432À
3
In the initial step, a reactive form of decatungstate anion is photoinduced since
4-
4-
W10O32 is chemically unreactive (Eq. 1). The excitation of W10O32 by light
absorption generates an excited state characterized by a ligand-to-metal charge
transfer (LMCT) transition, that is W10O324-*, which is also chemically inactive. In
this electronic transition, electron density initially localized at bridging oxygen
4-
atoms is displaced to tungsten atoms. The reactive form of W10O32 was recently
characterized to be the lowest-in-energy triplet excited state (3wO), which is
accessed from the LMCT excited state via intersystem crossing [22]. After the
3
formation of wO, the activation of the C–H bond may be reached via a hydrogen
atom transfer (HAT) mechanism, generating radicals and a reduced form of dec-
atungstate (Eq. 2), W10O325-H? [23]. Alkenes react with the generated R• to
produce another radical species with a new C–C bond, R–C–C• (Eq. 3). This latter
species recovers the hydrogen atom previously abstracted by decatungstate,
resulting in the formation of a new adduct and the reoxidation of the catalyst, thus
completing the catalytic cycle. Nevertheless, the radical addition shown in Eq. 3
competes with reverse HAT (see Eq. 2), as well as the coupling of radicals or
123

Acylation of dimethyl maleate photocatalyzed by…
disproportionation to alkenes; therefore, the yield to produce R–C–C–H also
depends upon the kinetics of the radical addition [17]. Moreover, the single electron
transfer (SET) mechanism represents another pathway leading to the transformation
of C–H bonds. Within this framework, charge transfer occurs between the reactive
species of W10O324- and RH, giving rise to the complexes W10O325- and RH•?; the
latter eventually dissociates into R• and H? [23]. The SET mechanism is beyond the
scope of the current work.
Within the photocatalysis with aldehydes involving W10O324-, the acylation of
electrophilic alkenes can be efficiently achieved [14]. In this regard, the key role
of W10O32 is the generation of acyl radicals (ROC•) so that the OC–H bond is
4-
activated. The main function of electrophilic alkenes is the trapping of the
photogenerated ROC• leading to the formation of unsymmetrical ketones in
regular to good yields. However, ROC• may undergo decarbonylation, resulting in
the corresponding alkyl radical. This side reaction represents a drawback since it
directly competes with the formation of acylated adducts. Nonetheless, radical
decarbonylation of secondary aldehydes can be disfavored as compared to the
acylation reaction by decreasing the temperature. In this contribution, therefore,
we aimed to experimentally and quantum-chemically explore the acylation of
4-
dimethyl maleate with heptanal photocatalyzed by W10O32 (see Scheme 1). To
our best knowledge, detailed mechanistic information leading to the formation of
acylated adducts has not been provided to date, even though previous studies
based on time-dependent (TD) density functional theory (DFT) demonstrate the
4-
key features of the reactive state of W10O32 [22, 24]. In this regard, TD-DFT
calculations represent a powerful tool to elucidate electronic, magnetic, and
structural properties of polyoxometalates, as well as chemical reactivity and
biological interactions, among others [25].
Scheme 1 Acylation of
electrophilic alkenes
photocatalyzed by decatungstate
anion
123

´
J. P. Martınez et al.
Results and discussion
Since tetrabutylammonium decatungstate was studied in activation of aldehydes and
alkylations [14], we decided to select one reported synthesis as a model reaction in
order to evaluate the reaction mechanism.
Photochemical acylation of dimethyl maleate. Tetrabutylammonium decatung-
state was used as photocatalyst in the synthesis of dimethyl-2-heptanoylsuccinate,
using heptanal and dimethyl maleate as reagents (Scheme 2). The reaction was
induced using either a 365-nm lamp or solar radiation; in both cases, the product
was obtained in 90% yield at similar rates.
To confirm the formation of an acylated adduct, a comparison between the
starting materials and the resulting product was done using infrared spectroscopy.
The spectrum of the product showed absence of aldehyde sp2 C–H stretching at
2713 cm-1, and sp2 C=C stretching at 1646 cm-1. As expected, it corroborated the
presence of the corresponding sp3 C–H stretching, C=O stretching at 1736 cm-1 for
ester carbonyl and at 1718 cm-1 for ketone carbonyl, and symmetric and
asymmetric C–O stretching (Figure S1 of electronic supplementary).
The reaction efficiency was determined by gas chromatography coupled to mass
spectrometry (GC–MS; Table 1) and the product was confirmed by analysis of the
obtained mass spectra (Figure S2–S4 of electronic supplementary). A blank
experiment was performed (absence of photocatalyst), in this case the reaction does
not proceed.
Computational investigations. In a previous work [22] based on TD-DFT, it is
demonstrated that the lowest-in-energy triplet excited state is the chemically
reactive species of decatungstate anion, 3wO; thus, also alluding to the Kasha’s rule.
3
The excitation energy leading to wO was calculated to be 73.8 kcal/mol. Our
U-DFT results predict that 3wO in its equilibrium geometry is 74.2 kcal/mol
(DG = 72.8 kcal/mol) above the ground state, in agreement with the previous TD-
DFT results.
3
In Fig. 1a, the structure of wO is schematized with the aim of identifying the
different non-equivalent W and O atoms; see shaded regions. Observing the upper
3
half of wO, in the apical region, an O atom monocoordinated (MC) to a W atom
can be distinguished. In the lateral region, four W atoms are localized in the same
plane as other eight mono- and dicoordinated (DC) O atoms, and there are four
remaining DC O atoms situated outside the plane. The bridging region is constituted
by other four DC O atoms.
Scheme 2 Synthesis of dimethyl-2-heptanoylsuccinate
123

Acylation of dimethyl maleate photocatalyzed by…
Table 1 GC–MS analysis of
Compound
Retention time (min)
the crude reaction
Heptanal
4.7
7.5
8.9
Heptanal self-condensation
Dimethyl-2-heptanoylsuccinate
Fig. 1 a Structure of the
lowest-in-energy triplet excited
state 3wO along with
nomenclature used in this
contribution. b Frontier
molecular orbitals available in
3wO. Labels stand for
irreducible representations
corresponding to the D4h point
group
The electronic structure of 3wO is schematized in Fig. 1b along with irreducible
representations resulting from the D4h symmetry point group. Since the chemical
reaction under study involves radicals, we focus our attention on the electrons that
3
give rise to wO. However, it is convenient to introduce the parameter c, which
accounts for the difference between a-spin and b-spin electron densities for either a
specific atom or a group of atoms. In an acyl radical, for instance, the carbon atom
lacking a hydrogen atom results in c = 1e- because of the excess of one a-spin
electron; on the other hand, c = 2e- in triplet states due to the excess of two a-spin
electrons. The a-spin and b-spin molecular orbitals in 3wO are reported in Table 2.
Observe that the symmetry of the a-spin highest occupied molecular orbital (a-
HOMO) is A1.u and contributions mainly come from d-type orbitals of lateral W
atoms in the plane with minor contributions from p-type orbitals of lateral DC O
atoms outside the plane (see Fig. 1b). Besides, Table S1 shows that one a-spin
3
electron forming the triplet state of wO is delocalized through the eight lateral W
atoms in the plane; indeed, c = 0.8e- for such a group of atoms due to the absence
123

´
J. P. Martınez et al.
Table 2 Frontier highest
a-spin
b-spin
occupied and lowest unoccupied
molecular orbitals (H and L,
respectively) in the lowest-in-
energy triplet excited state 3wO.
Energies are in eV and
Orbital
9I
A1.u
Energy
Orbital
9I
A1.u
Energy
H
- 5.29
- 8.54
- 8.59
- 8.62
- 8.62
- 8.62
- 8.67
- 8.67
- 8.68
- 8.68
- 8.85
L ? 1
L
- 3.76
- 7.08
- 8.22
- 8.23
- 8.23
- 8.28
- 8.28
- 8.51
- 8.56
- 8.57
- 8.62
- 8.62
H-1
H-2
H-3
H-4
H-5
H-6
H-7
H-8
H-9
H-10
A1.g
B2.g
E1.u
E1.u
A2.u
E1.g
E1.g
E1.u
E1.u
A2.g
A2.g
A1.u
E1.g
E1.g
E1.u
E1.u
A1.g
B2.g
A2.u
E1.u
E1.u
irreducible representations of 9I
corresponding to the D4h point
group are also reported
H
H-1
H-2
H-3
H-4
H-5
H-6
H-7
H-8
H-9
of one b-spin electron. In addition, Table 2 shows several frontier molecular orbitals
from which every a-HOMO with a specific symmetry has an analogous b-HOMO,
except for A2.g. Molecular orbitals represented as A2.g are mainly delocalized
through the eight lateral DC O atoms outside the plane with minor contributions
from bridging DC O atoms, like a-HOMO-10. As a matter of fact, the other triplet
3
a-spin electron of wO is distributed in the eight lateral DC O atoms outside the
plane because c = 1.4e-, which is confirmed by the lack of the analogous A2.g b-
spin electron [b-lowest unoccupied molecular orbital (LUMO)]. To sum up, the two
triplet electrons that give rise to wO are delocalized along lateral DC O atoms
3
outside the plane and lateral W atoms in the plane, both groups adding up to
c = 2.2e-. The acylation of dimethyl maleate photocatalyzed by decatungstate is
therefore conducted around the lateral region, although we also analyzed the apical
region.
Gibbs energy profiles. The overall reaction pathway for the acylation is
illustrated in Fig. 2 where changes in Gibbs energy (DG) are given with respect to
3
3
the energy of wO and heptanal. After the formation of wO, a reactant complex
(RC) is formed with a slight release of energy (DG = - 1.0 kcal/mol). The
˚
hydrogen atom of the heptanal reactant is positioned at d1 = 3.12 A from the
closest lateral DC O atom outside the plane. Linear transit (LT) calculations, where
the C–H distance in aldehyde is varied and every structure is optimized along the
path (with the exception of the frozen coordinate), suggest that the HAT evolves
under a barrierless process (see Figure S5 of electronic supplementary). After the
HAT is completed, an intermediate is characterized as a local minimum in the
potential energy surface (PES) and it corresponds to the formation of an acyl radical
complex, A•1. The formation of A•1 occurs in an exergonic process releasing
•
˚
27.9 kcal/mol. In A1, the acyl-radical carbon atom is localized at d2 = 1.95 A from
the transferred hydrogen atom. In wOH•, c = 1.3e- with the main contribution
coming from the lateral W atoms (c = 1.0e-) in the plane because the lateral DC O
123

Acylation of dimethyl maleate photocatalyzed by…
Fig. 2 Gibbs energy profile (in kcal/mol) for the acylation of the trans isomer of dimethyl maleate
(DM), dimethyl fumarate, photocatalyzed by decatungstate anion. Some geometrical parameters are also
provided (details in the text)
atoms reacted with a hydrogen atom. Another stationary point A•2 similar to A1• is
˚
identified at d2 = 3.42 A and DG = - 29.4 kcal/mol. In addition, it could be
argued that the quantum-chemical analysis of the HAT can become a challenging
task; however, interesting descriptions based on TD-DFT related to the abstraction
of hydrogen have been attained previously. For instance, Phillips et al. studied the
formation of arylphenyl ketyl radical species via the proton-coupled electron
transfer and HAT mechanisms. Based on experimental evidence and TD-DFT
calculations, the authors described the excited triplet np* and pp* intermediates of
some aromatic carbonyl compounds. The triplet np* is mainly localized at the
oxygen of the carbonyl so that this is the most favored reaction site for the HAT. On
the other hand, the triplet pp* is delocalized through the carbonyl group and the
aromatic ring, thus allowing a proton-coupled electron transfer mechanism. Even
though both mechanisms are feasible, a higher energy barrier for the HAT is
calculated in aqueous solutions; this is not the case for the proton-coupled electron
transfer mechanism where the triplet excited state is more delocalized [26].
Additionally, the DFT-based study by Firme et al. related to the photochemical
abstraction by triplet aliphatic carbonyls provides another comprehensive descrip-
tion about the HAT. In the case of hydrogen abstraction in acetonitrile from
methylamine by the triplets of formaldehyde, acetaldehyde, and acetone, the HAT
mechanism was a barrierless process, a fact that was attributed to the increased
reactivity of the triplet np* mainly localized at the oxygen atom in the carbonyl
group, even though they did calculate energy barriers with other hydrogen donors
3
[27]. In this regard, our results show that the triplet wO is delocalized through
several lateral W and O atoms; therefore, it gains stability so that the HAT pathway
can be completed.
123

´
J. P. Martınez et al.
The reaction between the acyl radical and the trans isomer of dimethyl maleate
(DM) occurs with no energy barrier as determined by LT calculations (see
Figure S6 of electronic supplementary), where the distance between the acyl radical
and the closest alkene carbon atom, d3, was monitored. Since photochemical
maleate-fumarate isomerization has been reported [28], the reaction between the
acyl radical and dimethyl maleate was modeled with the trans isomer, considering
that both isomers conduct to the same results (we studied the reaction with dimethyl
maleate and dimethyl fumarate). The initial pre-acylated radical complex, A•DM,
was characterized as a local minimum in the PES. In A•DM, the acyl-radical carbon
˚
atom results in c = 1.1e- and d3 = 3.16 A. The formation of the acylated radical
adduct, ADM•, is thermodynamically favored by 51.0 kcal/mol. This latter is also a
˚
local minimum in which d3 = 1.58 A and the radical is located at the non-bonded
alkene carbon atom (c = 1.1e-).
The final step leading to the acylated adduct, ADM, and the recovery of the
catalyst also occurs in a barrierless, exergonic process as illustrated in Figure S7 of
electronic supplementary; in which an energy of 79.8 kcal/mol is released. It is
worth mentioning that this stage of the reaction was modeled with the open-shell
singlet state where one electron is situated at wOH• and the other one at ADM•,
thus modeling a radical coupling rather than a spin-flip transition. As a matter of
fact, we infer that the triplet state 3wO decayed at the moment it reacted to form the
acyl radical A•1.
Fig. 3 Linear transit from the reactant complex formed between heptanal and 3wO toward the formation
of the acyl radical species A1•. Observe that the hydrogen atom transfer occurs in a barrierless process.
The distance C–H, d2, corresponding to the aldehyde group is monitored. Calculations were done for the
triplet state at the (COSMO:acetonitrile)UBP86(D3-BJ)/DZP (TZP for tungsten) level of theory including
frozen core approximation (FCA) and zero order regular approximation (ZORA)
123

Acylation of dimethyl maleate photocatalyzed by…
The HAT mechanism depicted in Fig. 2 was formulated by taking into
consideration the localization of the triplet electrons as determined via U-DFT.
However, based on TD-DFT, Ravelli et al. [24] proposed that the most likely
reaction site is the apical position. Figure 3 shows that the HAT in the apical
position also occurs in a barrierless, downhill process and transition states could not
be characterized. As a matter of fact, LT calculations illustrated in Figure S7 also
demonstrate a barrierless process for the product release along with recovering the
catalyst. The increased reactivity of O atoms in both apical and lateral positions is
attributed to the triplet excited state in which more spin density can be localized at
such atoms.
The lack of energy barriers resembles our experimental results since ADM was
produced in a remarkable high yield. Nevertheless, a 100% yield for the acylation is
not achieved since the radical coupling may compete because of the absence of
energy barriers in such a process (see Figure S8 of electronic supplementary).
Additionally, the decarbonylation takes place upon overcoming an energy barrier of
* 15 kcal/mol, which may be feasible at room temperature (see Figure S9 of
electronic supplementary). At least these two side reactions compete with the
photocatalyzed acylation of dimethyl maleate, even though decarbonylation is less
competitive.
As a final remark, solar-driven decatungstate-based photocatalysis can be also
attained for the alkylation of organic compounds [29]. Our results show that the
3
rate-determining step is the generation of wO. As a matter of fact, we carried out
the acylation of dimethyl maleate in the presence of solar light, and the reaction also
evolves to the acylated adduct, thus suggesting that excited states can be accessed
3
by the absorption of solar photons followed by interconversion toward wO.
Conclusions
An acylated adduct generated from heptanal and dimethyl maleate was photocat-
alytically synthesized with a high yield. The reaction was successfully completed
under solar-light and UV-lamp radiation, thus highlighting the versatility of such a
transformation. Based on density functional theory calculations under the
unrestricted formalism, the hydrogen atom transfer mechanism turns out to be a
proper reaction pathway for the acylation of organic compounds catalyzed by
decatungstate anion, wherein the lowest-in-energy triplet excited state is the reactive
form of this polyoxometalate, able to abstract a hydrogen atom from aldehydes,
forming acyl radicals in a barrierless, downhill process. Additionally, side reactions
like acyl radical coupling and decarbonylation may compete with the acylation of
dimethyl maleate, although this latter is experimentally and quantum-chemically
demonstrated to be less competitive. Overall, the current work provides mechanistic
details based on quantum chemistry methods which may be useful for the systematic
understanding of organic synthesis photocatalyzed by polyoxometalates.
123

´
J. P. Martınez et al.
Experimental section
General
All reagents and solvents were purchased from Sigma-Aldrich and used as received.
Infrared spectra were measured with a Thermo Scientific NICOLET iS10 FTIR
spectrometer with ATR (SMART iTR). UV–Vis spectra were obtained with
JENWAY 7310 spectrometer. GC–MS analyses were performed with HP 6890 GC
system fitted with a 5973 Network mass selective detector (MSD); the column used
was a capillary HP-5 ms. Photochemical reactions were set up in a reactor system
consisting of 4 15-W mercury lamps placed equidistantly around the reactor. The
major peak of spectrum wavelength occurs at 365 nm [30].
Synthesis of tetrabutylammonium catalyst
Photocatalyst tetrabutylammonium decatungstate (TBADT) was synthesized
according to the literature [29] method with variation to make eco-friendlier the
preparation by reducing purification steps.
In separate flasks, tetrabutylammonium bromide (2.5 g) and sodium tungstate
dihydrate (5.0 g) were dissolved in 150 mL of deionized water, and solutions were
kept at 90 °C for 30 min with vigorous stirring. The pH was adjusted to 2 with
concentrated hydrochloric acid. Then, the two solutions were mixed and maintained
at 90 °C for 30 min with stirring. The white suspension of TBADT formed was
cooled to room temperature and filtered in a Buchner funnel. The solid was washed
with water and then dried in an oven at 120 °C for 3 h.
Photochemical acylation of dimethyl maleate
The reaction was adapted from the literature [29].
Solutions of heptanal (10 mmol, 0.1M) and dimethyl maleate (10 mmol, 0.1M)
were mixed in a solution of 660 mg of TBADT (2 9 10-3 M) in 50 mL of
acetonitrile. The solution was purged for 10 min with argon. The Pyrex flask was
introduced into a reactor system and irradiated at 365 nm at room temperature with
stirring. The reaction was monitored by thin-layer chromatography (TLC). After
reaction completion, the solvent was evaporated, and 15 mL of ethyl acetate was
added to the obtained solid to remove the TBADT by filtration. The resulting
product was characterized by GC–MS. The compounds were purified by column
chromatography and characterized by IR.
The photochemical acylation under solar irradiation was done as mentioned
above, but in this case the solution was placed in a Pyrex condenser tube and
exposed to sunlight on the laboratory roof.
Dimethyl-2-heptanoylsuccinate IR (cm-1): 2997–2854 (C–H sp3), 1740 (C=O
ester), 1718 (C=O ketone), 1429 and 1370 (CH3), 1160 (C–O); GC–MS 71 eV (m/
z): 258 [M?], 227 [M?- CH3O], 199 [M?- C2H3O2], 188 [M?- C5H10], 156
123

Acylation of dimethyl maleate photocatalyzed by…
[M?- C7H10O2], 145 [M?- C7H13O], 113 [M?- C6H9O4], 85 [M?- C7H9O5],
55 [M?- C10H19O4], 43 [M?- C10H15O5].
3-hydroxy-2-pentylnonanal GC–MS 71 eV (m/z): 228 [M?], 210 [M?- OH], 153
[M?- C4H10O], 139 [M?- C6H12O], 125 [M?- C7H15O], 97 [M?- C9H19O],
83 [M?- C10H21O], 69 [M?- C11H23O], 55 [M?- C12H25O], 43 [M?-
C11H20O2].
2-pentyl-2-nonenal GC–MS 71 eV (m/z): 210 [M?], 153 [M?- C4H9], 139
[M?- C5H11], 125 [M?- C6H14], 97 [M?- C7H16], 83 [M?- C8H18], 69 [M?-
C9H20], 55 [M?- C10H22], 43 [M?- C11H19O].
Computational details
Calculations of structural and electronic properties of the chemical species under
study were completed with the Amsterdam density functional (ADF) quantum
chemistry code [31–33]. The methodology adopted in this contribution is similar to
the one previously reported by Ravelli et al. [22, 24] which turns out to provide
4-
consistent results in the description of optoelectronic properties of W10O32
.
Geometries were optimized with no restrictions via the local density approxi-
mation (Slater exchange) with nonlocal corrections for exchange (Becke88) [34]
and correlation (Perdew86) [35], that is the BP86 functional. Besides, dispersion
energy corrections as developed by Grimme et al., including the so-called Becke
and Johnson damping (D3-BJ), were added during the geometry optimization
procedure [36, 37]. It has been shown that dispersion corrections are essential for a
proper description of van der Waals interactions, systems involving non-repulsive
steric contact, and thermochemistry [38, 39]. Uncontracted sets of Slater-type
orbitals (STOs) of double-f and triple-f quality containing one set of polarization
and diffuse functions (DZP and TZP, respectively) were used to expand molecular
orbitals. DZP orbitals were centered at the hydrogen, carbon, and oxygen atoms;
and the TZP ones at tungsten. STOs correctly describe the asymptotic, long-range,
and near-the-nucleus (cusp) behaviors. However, their evaluation for systems with
many tens of atoms or several polyelectronic atoms requires large-scale compu-
tational resources. Therefore, the frozen-core approximation (FCA) was included to
solve such a predicament, which is an approximation that freezes the core orbitals
(1 s for the carbon and oxygen atoms, and up to 4d for W) during the self-consistent
field procedure [31]. It was shown that the FCA has a negligible effect on the
geometry optimization procedure [40, 41]. The size of the selected basis set is
appropriate since geometries optimized with an all-electron TZ2P basis set (the ones
used by Ravelli et al.) [22, 24] have very similar results to the one obtained in this
contribution, as illustrated in Figure S10 of electronic supplementary. Structures
were equilibrated in the presence of acetonitrile, which was modeled via the
conductor-like screening model (COSMO) [42] as implemented in ADF. Moreover,
scalar relativistic corrections were added within the zero order regular approxima-
tion (ZORA) [43].
All geometry optimizations were performed by the quantum regions intercon-
nected by local descriptions (QUILD) code [44]. This latter works as a wrapper
around the ADF program since it creates input files for ADF, then executes this
123

´
J. P. Martınez et al.
program and collects energies and gradients. QUILD uses adapted, delocalized
coordinates and constructs model Hessians with the appropriate number of
eigenvalues, which results in an advantage in the determination of transition state
structures [45]. All the stationary points were analyzed through analytical
vibrational frequencies as calculated at the COSMO-BP86(D3-BJ)/DZP (TZP for
tungsten) level of theory including FCA and ZORA. Low frequencies were
recalculated via the scanfreq keyword [31], since frequencies resulting in a small
force constant are more sensitive to the numerical noise. This keyword is
particularly useful to either confirm or discard the existence of small imaginary
frequencies in stationary points.
Electronic energies were corrected at the COSMO-PBE0 [46–48] (D3-BJ)/TZP//
COSMO-BP86(D3-BJ)/DZP (TZP for tungsten) level of theory including FCA and
ZORA. Gibbs energies were determined by adding to the corrected electronic
energies the contributions of the zero-point energy, thermal corrections to the
internal energy, and the entropy term computed at 298.15 K, considering an ideal
gas in standard conditions. Finally, excited states and radicals were modeled via the
unrestricted formalism; in this regard, the modifystartpotential keyword was
thoroughly required. Results are reported in Tables S1–S3.
´
y Tecnologıa
(CONACYT) grant number CB-2013-01-220844 and doctoral fellowship no. 304251 to Ernesto
Acknowledgements This work was supported by Consejo Nacional de Ciencia
´
´
Rivera-Avalos, and Universidad Autonoma de San Luis Potosı (UASLP) grant number C15-FRC-04-
12.12. Access and excellent service by the Dutch National Supercomputer (LISA-SURFsara system) is
gratefully acknowledged.
References
1. S. Suga, S. Suzuki, J. Yoshida, J. Am. Chem. Soc. 124, 30 (2002)
2. T. Yoshimitsu, Y. Arano, H. Nagaoka, J. Am. Chem. Soc. 127, 11610 (2005)
3. A. Citterio, A. Gentile, F. Minisci, M. Serravalle, S. Ventura, J. Org. Chem. 49, 3364 (1984)
4. C. Chatgilialoglu, D. Crich, M. Komatsu, I. Ryu, Chem. Rev. 99, 1991 (1999)
5. C.L. Hill, Synlett 2, 127 (1995)
6. N. Mizuno, K. Yamaguchi, Chem. Rec. 6, 12 (2006)
7. A. Proust, R. Thouvenot, P. Gouzerh, Chem. Commun. 16, 1837 (2008)
8. D.-L. Long, R. Tsunashima, L. Cronin, Angew. Chem. Int. Ed. 49, 1736 (2010)
9. D. Dondi, M. Fagnoni, A. Molinari, A. Maldotti, A. Albini, Chem. Eur. J. 10, 142 (2004)
10. D. Dondi, M. Fagnoni, A. Albini, Chem. Eur. J. 12, 4153 (2006)
11. I. Ryu, A. Tani, T. Fukuyama, D. Ravelli, M. Fagnoni, A. Albini, Angew. Chem. Int. Ed. 50, 1869
(2011)
12. S. Montanaro, D. Ravelli, D. Merli, M. Fagnoni, A. Albini, Org. Lett. 14, 4218 (2012)
13. S. Angioni, D. Ravelli, D. Emma, D. Dondi, M. Fagnoni, A. Albini, Adv. Synth. Catal. 350, 2209
(2008)
14. S. Esposti, D. Dondi, M. Fagnoni, A. Albini, Angew. Chem. Int. Ed. 46, 2531 (2007)
15. M. Okada, K. Yamada, T. Fukuyama, D. Ravelli, M. Fagnoni, I. Ryu, J. Org. Chem. 80, 9365 (2015)
16. M.D. Tzirakis, M. Orfanopoulos, J. Am. Chem. Soc. 131, 4063 (2009)
17. M.D. Tzirakis, I.N. Lykakis, M. Orfanopoulos, Chem. Soc. Rev. 38, 2609 (2009)
18. C. Tanielian, R. Seghrouchni, C. Schweitzer, J. Phys. Chem. A 107, 1102 (2003)
19. I. Texier, J.A. Delaire, C. Giannotti, Phys. Chem. Chem. Phys. 2, 1205 (2000)
20. H. Duclusaud, S.A. Borshch, Chem. Phys. Lett. 290, 526 (1998)
21. D.C. Duncan, M.A. Fox, J. Phys. Chem. A 102, 4559 (1998)
123

Acylation of dimethyl maleate photocatalyzed by…
22. V. De Waele, O. Poizat, M. Fagnoni, A. Bagno, D. Ravelli, ACS Catal. 6, 7174 (2016)
23. C. Tanielian, Coord. Chem. Rev. 178–180, 1165 (1998)
24. D. Ravelli, D. Dondi, M. Fagnoni, A. Albini, A. Bagno, Phys. Chem. Chem. Phys. 15, 2890 (2013)
´
´
25. X. Lopez, J.J. Carbo, C. Bo, J.M. Poblet, Chem. Soc. Rev. 41, 7537 (2012)
26. X. Zhang, J. Ma, S. Li, M.-D. Li, X. Guan, X. Lan, R. Zhu, D.L. Phillips, J. Org. Chem. 81, 5330
(2016)
27. C.L. Firme, S.J. Garden, N.C. de Lucas, D.E. Nicodem, R.J. Correa, J. Phys. Chem. A 117, 439
(2013)
28. G.S. Hammond, J. Saltiel, A.A. Lamola, N.J. Turro, J.S. Bradshaw, D.O. Cowan, R.C. Counsell, V.
Vogt, C. Dalton, J. Am. Chem. Soc. 86, 3197 (1964)
29. S. Protti, D. Ravelli, M. Fagnoni, A. Albini, Chem. Commun. 7351 (2009)
´ ´
30. J. Medina-Valtierra, E. Moctezuma, M. Sanchez-Cardenas, C. Frausto-Reyes, J. Photochem. Pho-
tobiol. A 174, 246 (2005)
31. G. te Velde, F.M. Bickelhaupt, E.J. Baerends, C. Fonseca Guerra, S.J.A. van Gisbergen, J.G. Sni-
jders, T. Ziegler, J. Comput. Chem. 22, 931 (2001)
32. S.J.A. van Gisbergen, J.G. Snijders, E.J. Baerends, Comput. Phys. Commun. 118, 119 (1999)
33. C. Fonseca Guerra, J.G. Snijders, G. te Velde, E. Baerends, J. Theor. Chem. Acc. 99, 391 (1998)
34. A.D. Becke, Phys. Rev. A 38, 3098 (1988)
35. J. Perdew, Phys. Rev. B 33, 8822 (1986)
36. S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 132, 154104 (2010)
37. S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 32, 1456 (2011)
38. S. Grimme, WIREs Comput. Mol. Sci. 1, 211 (2011)
`
39. S. Osuna, M. Swart, M. Sola, J. Phys. Chem. A 115, 3491 (2011)
40. J.G. Snijders, M. Swart, Theor. Chem. Acc. 110, 34 (2003)
41. M. Swart, J.G. Snijders, Theor. Chem. Acc. 111, 56 (2004)
42. A. Klamt, J. Phys. Chem. 99, 2224 (1995)
43. E. van Lenthe, E.J. Baerends, J.G. Snijders, J. Chem. Phys. 99, 4597 (1993)
44. M. Swart, F.M. Bickelhaupt, J. Comput. Chem. 29, 724 (2008)
45. M. Swart, F.M. Bickelhaupt, Int. J. Quantum Chem. 106, 2536 (2006)
46. C. Adamo, V. Barone, J. Chem. Phys. 110, 6158 (1999)
47. J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77, 3865 (1996)
48. J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78, 1396 (1997)
123