Exploring Pd/Al2O3 Catalysed Redox Isomerisation of Allyl Alcohol as a Platform to Create Structural Diversity

Article abstract of DOI:10.1007/s10562-017-2087-4

Abstract: We report our results on exploiting the different reactivities present in the catalytic cycle of the Pd/Al₂O₃ catalyzed redox isomerization of allyl alcohol. We show that the reactivity of allyl alcohol derived acrolein and enol can be involved in further cascade reactions leading to a diverse set of products. While the oxidation product acrolein can react via Michael and oxa-Michael reactions, the isomerization product enol can be readily involved in aldol condensation processes. Salicylaldehydes, that are able to react on their electrophilic carbonyl and nucleophilic OH-groups with allyl alcohol derived enol and acrolein, respectively, are used to explore conditions where the structure of the product heterocycles can be controlled. Graphical Abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-017-2087-4

Catal Lett (2017) 147:1834–1843
DOI 10.1007/s10562-017-2087-4
Exploring Pd/Al O Catalysed Redox Isomerisation of Allyl
2
3
Alcohol as a Platform to Create Structural Diversity
1
2
2
2
2
Attila Dékány · Enikő Lázár · Bálint Szabó · Viktor Havasi · Gyula Halasi ·
2
2,3
2,4
5
5,6
András Sápi · Ákos Kukovecz · Zoltán Kónya · Kornél Szőri · Gábor London
Received: 28 February 2017 / Accepted: 19 May 2017 / Published online: 26 May 2017
©
Springer Science+Business Media New York 2017
Abstract We report our results on exploiting the different
isomerization product enol can be readily involved in aldol
condensation processes. Salicylaldehydes, that are able to
react on their electrophilic carbonyl and nucleophilic OH-
groups with allyl alcohol derived enol and acrolein, respec-
tively, are used to explore conditions where the structure of
the product heterocycles can be controlled.
reactivities present in the catalytic cycle of the Pd/Al O
2
3
catalyzed redox isomerization of allyl alcohol. We show
that the reactivity of allyl alcohol derived acrolein and enol
can be involved in further cascade reactions leading to a
diverse set of products. While the oxidation product acr-
olein can react via Michael and oxa-Michael reactions, the
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-017-2087-4) contains supplementary
material, which is available to authorized users.
*
Gábor London
london.gabor@ttk.mta.hu
1
2
Department of Organic Chemistry, University of Szeged,
Szeged, Dóm tér 8, 6720, Hungary
Department of Applied and Environmental Chemistry,
University of Szeged, Szeged, Rerrich Béla tér 1, 6720,
Hungary
3
4
5
6
MTA-SZTE “Lendület” Porous Nanocomposites Research
Group, Szeged, Rerrich Béla tér 1, 6720, Hungary
MTA-SZTE Reaction Kinetics and Surface Chemistry
Research Group, Szeged, Rerrich Béla tér 1, 6720, Hungary
MTA-SZTE Stereochemistry Research Group, Szeged, Dóm
tér 8, 6720, Hungary
Present Address: Institute of Organic Chemistry, Research
Centre for Natural Sciences, Hungarian Academy
of Sciences, Budapest, Magyar tudósok körútja 2., 1117,
Hungary
Vol:.(1234567890)
1
3

Exploring Pd/Al O Catalysed Redox Isomerisation of Allyl Alcohol as a Platform to Create…
1835
2
3
Graphical Abstract
Keywords Allyl alcohol · Isomerization · Pd/Al O · Pt/
starting material that could be exploited in different reac-
tions. Such catalyst/reactant systems with tuneable reac-
tivity profiles would be particularly interesting to develop
one-pot multistep processes for producing diverse molec-
ular structures [8].
2
3
Al O · Michael addition · Heterocycles
2
3
1
Introduction
To further develop this concept an interesting reac-
tion is the transition metal-catalysed redox isomerisa-
tion of allylic alcohols to carbonyl compounds (Fig. 1a)
[10–13]. This reaction has been widely studied because of
its atom-economical nature, and also for its use in tandem
isomerisation/C–C or C–heteroatom bond formation pro-
cesses [14]. Regarding the tandem processes, allyl alcohol
has been almost exclusively applied as an enolate equiva-
lent (Fig. 1b).
Major efforts in organic chemistry are directed towards
synthesizing increasingly complex and diverse molecu-
lar structures in operationally simple and environmen-
tally benign ways [1]. In this context several synthesis
concepts have emerged in the last decades such as atom
economical reactions [2], click chemistry [3] or one-pot
strategies [4]. Especially attractive are the methodologies
that realize those concepts under heterogeneous cata-
lytic conditions [5–7]. Applying multi-site solid catalysts
for one-pot sequential reactions is an emerging area of
research with clearly foreseeable environmental and eco-
nomic benefit [8, 9]. Solid catalysts represent environ-
mentally more benign alternatives of many homogene-
ous systems as they can be recovered and reused several
times that can ultimately reduce the energy need and the
amount of waste produced during chemical processes.
Multi-site catalysts, due to their isolated catalytically
active centres, are potentially able to produce differ-
ent intermediates with distinct reactivities from a single
However, considering the proposed mechanisms of
1
2
allylic alcohol isomerisation there are at least three inter-
mediates where different reactivities could be extracted
(Fig. 2). Acrolein (α,β-unsaturated carbonyls, generally) A
with an electrophilic reactivity at the β-carbon, enolate B
with a nucleophilic reactivity and aldehyde C with an elec-
trophilic carbonyl carbon could be involved selectively in
further reactions with properly chosen reagents.
Recently, we have expanded the scope of the available
heterogeneous Pd-catalyzed allyl alcohol isomerization
processes [16–20] and showed that reactivities B and C
can be selectively exploited in combination with aldol
condensation and oxidative heterocyclization reactions
accessing a diverse set of structures with a single cata-
lyst/reactant system (Fig. 3) [21]. Using simple Pd/Al O
2
3
1
Note that the simplified mechanism presented here is based on
recent mechanistic studies on Ru and Rh based catalysts and only
aimed at showing the key intermediates whose reactivity could be
exploited. For more mechanistic insights see refs. [10–13] and refer-
ences therein.
Fig. 1 Transition metal catalysed redox isomerisation of allyl alco-
hol (M=Ru, Rh, Fe, Co, Ni, Mo, Ir, Pd, Pt, Os, etc.). The enolate
intermediate could either tautomerise to the corresponding saturated
carbonyl compound (a) or sequentially react with an electrophile (b)
to form a new C–C or C–heteroatom bond
2
Note that allyl alcohol itself could be used for allylation also in
Tsuji-Trost type reactions. For an overview see Muzart [15].
1
3

1
836
A. Dékány et al.
Fig. 2 Schematic representa-
tion of a plausible mechanism
of the transition metal catalysed
isomerisation of allyl alcohol.
Structures A, B and C show
possible reactivity profiles that
could be individually exploited
to create structurally divers
products in a one-pot fashion
using a single catalyst
catalyst in combination with allyl alcohol we were able to
access α,β-unsaturated carbonyl compounds via an isom-
erisation/aldol condensation sequence and cyclic lactones
via an isomerisation/aldol condensation/cyclic hemi-
acetal formation/oxidation sequence (Fig. 3 Reactivity
B). While these reactions made use of the enolate reac-
tivity derived from allyl alcohol, the carbonyl reactivity
of the isomerisation product propionaldehyde could be
exploited in the formation of a benzimidazole derivative
Here we report our results on expanding the accessible
structural diversity based on the heterogeneous catalytic
redox isomerisation of allyl alcohol with a focus on the
involvement of acrolein reactivity.
2 Experimental
2.1 General Remarks
(
Fig. 3 Reactivity C). The initial step of this latter reac-
tion is an imine formation between the aromatic amine
and the carbonyl group of propanal.
Commercial reagents, solvents and catalysts (Aldrich,
Alfa Aesar, Fluka, VWR) were purchased as reagent-
grade and used without further purification. 5% Pd/Al O
Although using allyl alcohol as a carbonyl precursor in
tandem reactions has scarcely been described, this reactiv-
ity is not independent from the enolate form as the two spe-
cies are in equilibrium with each other. A more challeng-
ing task is to selectively exploit reactivity A involving the
α,β-unsaturated compound acrolein. Acrolein forms in the
first stage of the isomerisation process upon the oxidative
dehydrogenation of ally alcohol and it is not in equilibrium
with any of the following species in the cycle. Thus, any
reaction involving acrolein has to take place before it gets
reduced to the enolate form. Although the Pd-catalyzed
oxidation chemistry of allylic alcohols have been investi-
gated extensively, in most cases the isomerization pathway
is not pronounced [22–33]. On the other hand it is a com-
monly observed side-reaction during the hydrogenation of
the same type of compounds [16–19, 34–36]. In these cases
isomerisation occurs only in the presence of hydrogen gas
over heterogeneous Pd-catalysts, which is likely involved in
the formation of Pd-H species necessary for the reduction
step that leads to enolate formation.
2
3
and 5% Pt/Al O were Engelhard 40,692 and Engelhard
2
3
4759 type, respectively. Catalysts were used without any
3
pre-treatment, or otherwise noted. Reactions were per-
formed without the exclusion of air, or otherwise noted.
Solvents for extraction or column chromatography were
of technical quality. Organic solutions were concentrated
by rotary evaporation at 25–40°C. Thin layer chromatog-
raphy was carried out on SiO –layered aluminium plates
2
(60778-25EA, Fluka). Flash column chromatography
was performed using SiO –60 (230–400 mesh ASTM,
2
0.040–0.063 mm from Merck) at 25°C.
3
Our earlier results (see ref. [21]) showed that reductive pretreat-
ment of the catalyst results in a decreased activity, hence we used the
as-received catalysts throughout the present study.
1
3

Exploring Pd/Al O Catalysed Redox Isomerisation of Allyl Alcohol as a Platform to Create…
1837
2
3
Fig. 3 Examples of the use
of enol-type reactivity (B) and
carbonyl-type reactivity (C)
for the synthesis of different
products based on the Pd/allyl
alcohol catalyst/reactant system
2
.2 Instrumentation
mode kit and gas flushing accessory. The surface reac-
tions on different catalysts were followed in diffuse
1
13
−1
H NMR and C NMR spectra were recorded on a Bruker
AVANCE DRX 400 or a Varian 500 spectrometer. All
spectra were recorded at 25°C. The residual solvent peaks
were used as the internal reference (CDCl : δ =7.26 ppm,
reflectance (DRIFT) mode, with 512 scans and 2 cm
resolution.
3
H
1
δ =77.16 ppm). H NMR spectra are reported as follows:
2.3 Allyl Alcohol Oxidation/Isomerization Over Pd/
Al O and Pt/Al O in the Gas Phase
C
chemical shift δ in ppm relative to TMS (δ=0 ppm), mul-
tiplicity, coupling constant (J in Hz), number of protons.
The resonance multiplicity is described as s (singlet), d
2
3
2
3
Catalytic transformation of allyl-alcohol to acrolein and
propanal was performed in a continuous flow reactor
system at 80 °C, where N :O with a 9:1 ratio and a total
(
doublet), t (triplet), q (quartet), combinations thereof, or
m (multiplet). Broad signals are described with br. (broad).
2
2
1
3
−1
C NMR spectra are reported as follows: chemical shift
flow rate of 100 ml min at 1 bar were bubbled through
a saturator filled with allyl alcohol at 23 °C. The reaction
products were analyzed with an Agilent 4890 gas chro-
matograph equipped with PORAPAK 1/2Q + PORA-
PAK 1/2S packed and Equity-1 capillary column. The
sampling loop of the GC was 225 μl. The conversion of
the allyl-alcohol was calculated taking into account the
amount of reactant consumed. In a typical catalytic test,
5 mg of the catalyst were mixed with 1 g of inert Stöber
silica and pretreated for 30 min in Ar with a flow rate of
δ in ppm relative to TMS (δ=0 ppm) (number of carbons
if greater than 1). Mass spectra (MS) of the products were
recorded on an Agilent 6980N-5973 GC-MSD.
Imaging of the Pt/Al O and the Pd/Al O catalysts was
2
3
2
3
2
performed by a FEI TECNAI G 20 X-Twin high-resolution
transmission electron microscope (equipped with electron
diffraction) operating at an accelerating voltage of 200 kV.
The samples were drop-casted onto carbon film coated cop-
per grids from ethanol suspension.
−
1
The surface reactions were followed by FTIR spectros-
copy on a Bruker Vertex 70 FTIR spectrometer with MIR
light source. In-situ reactions were performed in a Specac
HPHT reaction chamber with ZnSe windows. The cell
was upgraded with temperature controller, reflectance
10 ml min at 80 °C before the reaction. The conversion
were <14% during the catalytic tests. Calculation of the
metallic active sites was based on the average size of the
nanoparticles assuming that every surface atom is active
in the reactions.
1
3

1
838
A. Dékány et al.
Fig. 4 TEM images of 5% Pt/
Al O (a) and 5% Pd/Al O (b)
2
3
2
3
catalysts
Fig. 5 Activity (a) (steady state
conditions) and selectivity (b)
of allyl alcohol transformations
over 5% Pt/Al O and 5% Pd/
2
3
Al O catalysts in the gas phase
2
3
at 80°C
2
.4 In‑Situ Infrared Spectroscopic Study
2.5.1 General Procedure for the Solution Phase Cascade
Reactions
In a typical procedure, 5 mg of the catalyst (Pt/Al O ; Pd/
2
3
Al O ; Al O ) was deposited by pressuring to thin lay-
The ketone, ketoester, aldehyde or N-tosyl imine
2
3
2
3
ers onto the 12 mm diameter, highly reflective aluminium
foil surfaces. Catalysts were deposited in order to perform
Diffuse Reflectance Infrared Fourier Transform (DRIFT)
measurements in a High Temperature-High Pressure
(0.5 mmol), 5% Pd/Al O or Pt/Al O (5 mol% metal),
2
3
2
3
K CO (0.5 mmol) were suspended in toluene (1 ml) and
2
3
subsequently allyl alcohol was added (1 mmol). The mix-
ture was heated at 100°C for 24 h then filtered. The resi-
due was washed several times with EtOAc. The organic
phase was concentrated and purified by flash column
chromatography.
(
HTHP) cell. For the static state reactions, the HTHP cell
was first purged by N to avoid contaminants and water.
2
The actual catalyst compositions were backgrounded under
N atmosphere. The allyl alcohol reactant was injected to
2
the chamber as saturated vapour by N flow through a gas
2
saturator at RT. After 5 min of the allyl alcohol saturation,
the chamber was closed and heated to the reaction tempera-
ture (80°C). After 24 h, the chamber was purged for 15 min
3 Results and Discussions
3.1 Gas‑Phase Transformation and In‑Situ FTIR Study
with pure N in order to determine the adsorbed species on
2
the surface of the catalysts.
Initially, we investigated the capacity of the applied Pd/
Al O catalyst to transform allyl alcohol into the oxida-
2
3
tion product acrolein and isomerisation product propanal.
This was expected to give information on the conversion
and selectivity values that can be obtained during the trans-
formation. As a control catalyst, we used Pt/Al O which
2
.5 Synthesis and Characterization
The detailed synthetic procedures and characterization of
compounds described in this article can be found in the
Supporting Information.
2
3
was expected to be active primarily in the oxidation of allyl
alcohol to acrolein [37].
1
3

Exploring Pd/Al O Catalysed Redox Isomerisation of Allyl Alcohol as a Platform to Create…
1839
2
3
Fig. 7 Transformation of acetylacetone to 2-acetylphenol with acr-
olein generated from allyl alcohol. The sequence, apart from the oxi-
dation of allyl alcohol, involves a Michael addition, an intramolecular
aldol-condensation and an aromatization step
Fig. 6 In-situ IR spectra of the surface of the Pt/Al O and Pd/Al O
3
2
3
2
after 24 h allyl alcohol exposure at 80°C
In the case of both 5% Pt/Al O and 5% Pd/Al O cata-
2
3
2
3
lysts, the metal components were present as nanoparti-
cles highly dispersed on the surface of the Al O support
2
3
(
Fig. 4). The average diameter of the catalyst particles
was 3.4± 1.4 and 3.8± 1.1 nm for Pt/Al O and Pd/Al O ,
2
3
2
3
respectively.
We tested the activity of both catalysts in the transfor-
mation of allyl alcohol in a flow mode gas reactor at 80°C.
The catalytic activity of Pd and Pt on Al O was mark-
2
3
edly different (Fig. 5). While with Pd/Al O an over-
2
3
−
1
−1
all Turnover frequency of 0.57 molecules site
s
was
obtained, Pt/Al O was poorly active delivering only
2
3
Fig. 8 Transformation of ethyl 2-methylacetoacetate to a cyclohex-
enone derivative with acrolein generated from allyl alcolhol. The
sequence, apart from the oxidation of allyl alcohol, involves a
Michael addition, and an intramolecular aldol-condensation step
−
1 −1
0
.09 molecules site
s
(Fig. 5a). Not only had the activ-
ity showed differences, but also the obtained selectivities
(
Fig. 5b). As expected, the Pt containing catalyst was active
only in the oxidation step and only acrolein was detected
in the product mixture. On the other hand, over Pd/Al O
2
3
both the oxidation product acrolein and isomerisation
product propanal was formed with 78 and 22% selectiv-
ity, respectively. These results are in line with our previous
findings on the activity of these catalyst in the reaction of
allyl alcohol with substituted benzaldehydes leading to α,β-
unsaturated carbonyls via reactivity B [21].
We investigated the transformation pathways of allyl
alcohol over the Pd- and Pt-based catalyst by in-situ DRIFT
spectroscopy to get more insight into the processes occur
on the catalyst surfaces. Generally, the results reflect the
more rich chemistry occur on the surface of Pd/Al O
2
3
compared to Pt/Al O , similarly to that observed in the gas
2
3
The results point out that under Pd catalysis both the
oxidation and the isomerisation processes are operating.
Moreover, the presence of considerable amount of acrolein
suggests that its reduction to the enol is not so fast that
would prevent the possibility of its involvement in other
reactions (e.g. Michael addition). It has to be noted, that
the gas-phase measurements cannot be directly paralleled
to the results obtained in solution phase reactions where a
complex multicomponent mixture is surrounding the cata-
lyst, however, they give the basis for further exploring solu-
tion phase cascade reactions.
phase reactions.
Figure 6 shows the assumed adsorbed species in depend-
ence of catalyst composition. Due to intensive purging of
the reaction chamber only adsorbed species appeared on
the spectra. Minimal deposition of allyl alcohol appeared
on the reference aluminium foil, with hardly shifted broad
peaks compared to its gas phase spectrum (see SI). Spec-
tra for Al O and Pt/Al O samples show similar absorp-
2
3
2
3
tion patterns, indicating the same type of surface associ-
ated bonding of the allyl-alcohol intermediates. In contrast,
Pd/Al O showed differently shifted, broader and even
2
3
1
3

1
840
A. Dékány et al.
oxidation
O
O
O
O
intramolecular
aldol condensation
R
R
O
O
O
oxa-Michael-product
2H-chromene derivatives
OH
2
Pd/Al O3
R
+
OH
hemiacetal formation/
oxidation
R
R
OH
O
O
O
OH
aldol condensation product
coumarin derivatives
isomerization
Fig. 9 Salicylaldehydes are able to react on their electrophilic carbonyl and nucleophilic –OH groups with allyl alcohol derived enol and acr-
olein, respectively. Exploiting the different reactivites could be a handle to control the product structure
splitted peaks, presumably influenced by surface associa-
tion bonding.
condensation reaction. The amount of the allyl alcohol
was not affecting the reaction as both 1 and 2 equiv. gave
the same results. Performing the reaction under identical
conditions but using acrolein directly, led to a somewhat
increased yield of 17%. This suggests that in the acety-
lacetone/allyl alcohol reaction mixture other reaction
pathways are also operational, such as the transfer hydro-
genation of the diketone with the alcohol as the hydro-
gen source. This would decrease the concentration of the
active nucleophile and ultimately reduce the yield. It is
noted, that using Pt/Al O catalyst or without Pd/Al O
−
1
The peaks in the 3800–3400 and 1400–1200 cm
regions indicate the additional Al–OH groups and allyl
alcohol OH groups on the catalyst surfaces. In the case of
the Pd/Al O catalyst, peaks are more intense and broader,
2
3
indicating multiple ways of association of the allyl alcohol.
In addition, the associated M–O and M–C bonds related
−
1
2
050–1770 cm region shows a blue shift compared to
the Pt/Al O and Al O samples indicating the presence of
2
3
2
3
favourable surface adsorbed propanal as well as the allyl
2
3
2
3
alcohol associated on multiple points on Pd/Al O due to
the reaction did not result in product formation.
2
3
the high affinity of the carbon–carbon double bond to the
metallic Pd. In the case of the Pd-based catalyst, C=C
−1
stretching at 1630 cm showing the stable presence of
acrolein or enol form on the surface. We suggest that the
occurred selectivity differences of Pd/Al O and Pt/Al O
O
R
O
2
3
2
3
5
2
% Pd/Al O3
O
O
can be related to association affinity difference of carbon
K CO
2
3
P1
P2
OH
R
+
double bond to Pd and Pt on alumina support. High selec-
toluene,
100 C, 24 h
OH
tivity of Pt/Al O towards acrolein seems to be related to
o
2
3
(2 eq.)
low affinity of carbon double bond to the Pt metal. In the
case of Pd/Al O catalyst, the surface intermediated enol
R
O
H
2
3
form are more stable resulted in the high amount of propa-
nal in the gas phase reaction products.
O
O
O
MeO
Cl
H
OH
H
3
.2 Solution Phase Cascade Reactions
OH
OH
Initially, we tested carbon nucleophiles under the previ-
ously established reaction conditions [21] to exploit acr-
olein reactivity by means of a Michael addition.
P1 - < 5%
P2 - 50%
P1 - < 5%
P2 - 12%
P1 - 19%
P2 - 5%
O
O
Acetylacetone could be converted to 2-acetylphenol
O N
2
H
H
(
Fig. 7), a common motif in bioactive compounds [38],
although the conversion of the starting material was poor.
This reaction is suggested to proceed through a Robin-
son annulation/aromatization sequence, where the metal
component of the catalyst has a role in the allyl alcohol
OH
N
OH
P1 - < 5%
P2 - < 5%
P1 - < 5%
P2 - < 5%
oxidation and the last aromatization step, while the Al O
2
3
Fig. 10 Reactivity of salicylaldehydes towards allyl alcohol derived
promotes the Michael step and the intramolecular aldol
acrolein and enol
1
3

Exploring Pd/Al O Catalysed Redox Isomerisation of Allyl Alcohol as a Platform to Create…
1841
2
3
O
Cl
Cl
X= O
O
X
Pd/Al O - 19%
Pt/Al O - 13%
Cl
2
3
OH
+
2
3
OH
X= NTs
O
O
Pd/Al O - 22%
2
3
Pt/Al O - 30%
2
3
Fig. 12 Switching the reactivity of 5-cholorsalicylaldehyde towards
allyl alcohol derived acrolein and enol by N-tosylimine formation
Previously, we found that unsubstituted salicylaldehyde
undergo reaction with the enol derived from allyl alcohol
yielding the corresponding coumarin (Fig. 3).
We tested salicylaldehydes with different substituents,
both electron donor and acceptor types, to explore their
reactivity towards the species derived from allyl alcohol
(Fig. 10).
Fig. 11 Reactivity of N-tosylimine derivatives of salicylaldehydes
towards allyl alcohol derived acrolein and enol
We found that within the examined series of salicylalde-
hydes the substituents had a major role in determining the
reactivity. Salicylaldehyde with no substituent yielded the
corresponding coumarin (P2) and also provided the high-
est yield among the compounds. Introduction of the donor
methoxy substituent, although the yield was comparably
lower, still resulted in the coumarin derivative as the major
product. This tendency was reversed when a Cl-substituent
was present in the molecule. Unlike in the case of salicylal-
dehyde and 5-methoxysalicylaldehyde, the formation of the
2H-chromene derivative (P1) was preferred in the reaction.
This implies that while in the first two cases the electro-
philic carbonyl group was the reactive function, in the case
of the Cl-substituted salicylaldehyde the oxa-Michael reac-
tion/aldol condensation was the preferred pathway. When
strong acceptor (–NO ) or strong donor (–NEt ) group was
Interestingly, a somewhat weaker nucleophile, ethyl
2
-methylacetoacetate, was converted more efficiently
(
Fig. 8). This substrate is of interest as no aromatization
is possible in this case and the transformation leads to the
formation of 2-cyclohexenone ring structure containing
an all-carbon quaternary centre. Similarly to the previ-
ous case, product formation was not observed when the
reaction was performed without the presence of metal
catalyst.
Heteroatom nucleophiles such as phenol and thiophe-
nol, did not participate in conjugate addition with acrolein
under the reaction conditions. In case of phenol no conver-
sion was obtained, while from thiophenol the correspond-
ing disulphide was formed selectively.
2
2
Salicylaldehydes, comprising both electrophilic car-
bonyl function and nucleophilic phenolic OH-group,
are especially interesting as they are potentially able to
undergo reaction with both the elecrophilic acrolein and
nucleophilic enolate derived from allyl alcohol (Fig. 9).
Reaction of salicylaldehydes with acrolein would lead to
present, essentially no desired product could be isolated.
These starting materials led to a complex (polymeric) prod-
uct mixture. The lack of metal catalyst in the reaction mix-
ture resulted in no product formation.
As an attempt to improve the results we converted the
aldehydes to the corresponding N-tosylimines (Fig. 11). We
expected that this transformation will shift the reactivity
towards the aldol chemistry on the carbonyl group.
2
H-chromene derivatives through an oxa-Michael addition/
aldol condensation sequence. Reaction with the enol would
give coumarin derivatives via an aldol condensation/hemi-
acetal formation/oxidation pathway. Both 2H-chromene
and coumarin derivatives are common motifs in natural
products. The ambident reactivity character of salicylalde-
hydes has scarcely been considered in controlling product
structure through controlling reactivity pattern [39–42].
The transformation did not affect the reactivity of the
nitro- and diethylamino-substituted starting materials.
However, in line with the expectations, for 5-methoxy and
5-chlorosalicylaldehyde it improved the yield of the cou-
marin product (P2). This also means that in the case of
5-chlorosalicylaldehyde the derivatization is switching the
1
3

1
842
A. Dékány et al.
initial reactivity of the compound from primarily nucleo-
philic to primarily electrophilic (Fig. 12).
condensation sequences with allyl alcohol derived acrolein
leading to industrially relevant molecular building blocks.
Salicilyladehydes could be involved in reactions both on
the electrophilic carbonyl and on the nucleophilic phenolic
–OH sites. Their reactivity, however, was strongly influ-
ence by the nature of substituents. Transforming the salicy-
laldehydes to the corresponding N-tosyl imine derivatives
resulted in switching initial reactivity in case of 5-chloro-
salicylaldehyde from primarily nucleophilic to primarily
electrophilic.
Unexpectedly, Pt/Al O was also found to be active in
2
3
the transformation of 5-cholorsalicylaldehyde delivering
comparable results to that obtained with Pd/Al O regard-
2
3
ing both yields and the product selectivity. This is in con-
trast to the previously discussed reactions, where the Pt-
based catalyst showed no activity. We do not have a firm
explanation to this change in catalytic activity, however,
the formation of acylplatinum species [43, 44] of some sort
could be involved leading to soluble Pt-complexes. Homo-
geneous Pt-complexes have been described to have activity
in ally alcohol isomerisation processes [45].
The possibility to exploit the different reactivities
involved in the catalytic allyl alcohol isomerization cycle
by using heterogeneous transition metal catalyst gives new
perspectives for this chemistry in a greener and sustainable
context.
We conducted control experiments in order to get fur-
ther insights into the latter process. Isomerization of allyl
alcohol in toluene in the presence of Pt/Al O catalyst gave
2
3
Acknowledgements Financial support by the János Bolyai Research
Scholarship of the Hungarian Academy of Sciences (AS, GL), the
ÚNKP-ÚNKP-16-4 New National Excellence Program of the Min-
istry of Human Capacities (AS) and the National Research, Devel-
opment and Innovation Office, Hungary (NKFIH OTKA Grants PD
similar results to that obtained in the gas-phase measure-
ments regarding catalyst activity. When the reaction mix-
ture contained only allyl alcohol and the catalyst in toluene
after 4 h at 100°C about 98% of the allyl alcohol remained
untouched and only about 1% of each product (acrolein
and propanal) could be detected. The addition of K CO
1
20877 (AS), PD 115436 (GL) and K 109278 (KS, GL)) is gratefully
acknowledged. This collaborative research was partially supported by
the “Széchenyi 2020” program in the framework of GINOP-2.3.2-15-
2
3
did not improve considerably the conversion of allyl alco-
hol (about 3% conversion to 1:1 mixture of the products).
These results point towards the involvement of 5-chlorosal-
icylaldehyde in the catalytic process when Pt is the cata-
lytic metal. In comparison, under the same conditions the
transformation of allyl alcohol over Pd/Al O catalyst was
2
016-00013 “Intelligent materials based on functional surfaces – from
syntheses to applications” project and the NKFIH (OTKA) K112531
grant.
References
2
3
complete in 4 h to propanal. When K CO was added to
2
3
1
.
Sheldon RA, Arends I, Hanefeld U (2007) In: Green chemistry
and catalysis. Wiley-VCH, Weinheim
the reaction mixture the formation of 2-methyl-2-pentenal,
the self aldol condensation product of pentanal was also
detected by GC-MS supporting the role of the base in the
aldol chemistry following the isomerization step.
2
3
.
.
Trost BM (1991) Science 254:1471
Kolb HC, Finn MG, Sharpless KB (2001) Angew Chem Int Ed
40:2004
4
.
Broadwater SJ, Roth SL, Price KE, Kobaslija M, McQuade DT
When 5-Chlorosalicylaldehyde was submitted to the
reaction conditions (see Sect. 2.5) using Pt/Al O as cata-
(
2005) Org Biomol Chem 3:2899
2
3
5
6
.
.
Felpin FX, Fouquet E (2008) ChemSusChem 1:718
lyst, after 12 h GC-MS analysis showed the starting alde-
hyde and the chromene derivative in about 1:1 ratio and
trace amount of the coumarin product. (Note that this does
not mean 50% conversion to the desired product as uni-
dentified polymeric side products could not be measured.)
After 12 h the solid material was filtered and the filtrate
was heated for an additional 12 h. The results after the 24 h
reaction time were practically unchanged.
Climent MJ, Corma A, Iborra S (2011) Chem Rev 111:1072
7. Daştan A, Kulkarni A, Török B (2014) Green Chem 14:17
8
.
Climent MJ, Corma A, Iborra S, Sabater MJ (2014) ACS Catal
:870
4
9
.
Climent MJ, Corma A, Iborra S (2009) ChemSusChem 2:500
1
0. van der Drift RC, Bouwman E, Drent E (2002) J Organomet
Chem 650:1
1
1
1
1
1. Uma R, Crévisy C, Grée R (2003) Chem Rev 103:27
2. Cadierno V, Crochet P, Gimeno J (2008) Synlett 1105
3. Mantilli L, Mazet C (2011) Chem Lett 40:341
4. Ahlsten N, Bartoszewicz A, Martín-Matute B (2012) Dalton
Trans 41:1660
1
1
5. Muzart J (2005) Tetraherdon 61:4179.
4
Conclusions
6. Zharmagambetova AK, Ergozhin EE, Sheludyakov YL,
Mukhamedzhanova SG, Kurmanbayeva IA, Selenova BA,
Utkelov BA (2001) J Mol Catal A 177:165
In summary, we have shown that allyl alcohol can not
only be considered as an enolate (or carbonyl) equivalent
but it is possible to use it to generate acrolein, which can
be involved in further cascade reaction with appropriately
chosen reactants. Carbon nucleophiles (acetylacetone, ethyl
17. Musolino MG, Cutrupi CMS, Donato A, Pietropaolo D, Pie-
tropaolo R (2003) Appl Catal A 243:333
1
8. Musolino MG, De Maio P, Donato A, Pietropaolo R (2004) J
Mol Catal A 208:219
1
9. Musolino MG, Caia CV, Mauriello F, Pietropaolo R (2010) Appl
Catal A 390:141
2
-methylacetoacetate) did undergo Michael addition/aldol
1
3

Exploring Pd/Al O Catalysed Redox Isomerisation of Allyl Alcohol as a Platform to Create…
1843
2
3
2
2
2
2
2
2
0. Sadeghmoghaddam E, Gaieb K, Shon YS (2011) Appl Catal A
33. Falletta E, Della Pina C, Rossi M, He Q, Kiely CJ, Hutchings GJ
(2011) Faraday Discuss 152:367
4
05:137
1. Zsolnai D, Mayer P, Szőri K, London G (2016) Catal Sci Tech-
nol 6:3814
34. Sadeghmoghaddam E, Gu H, Shon Y-S (2012) ACS Catal
2:1838
2. Lee AF, Hackett SFJ, Hargreaves JSJ, Wilson K (2006) Green
Chem 8:549
35. Di Pietrantonio K, Coccia F, Tonucci L, d’Alessandro N, Bressan
M (2015) RSC Adv 5:68493
3. Hackett SFJ, Brydson RM, Gass MH, Harvey I, Newman AD,
Wilson K, Lee AF (2007) Angew Chem Int Ed 46:8593
4. Parlett CMA, Bruce DW, Hondow NS, Lee AF, Wilson K (2011)
ACS Catal 1:636
36. Maung MS, Dinh T, Salazar C, Shon Y-S (2017) Coll Surf A
513:367
37. Kon S, Siddiki SMAH, Shimizu KI (2013) J Catal 304:63
38. Mo F, Trzepkowski LJ, Dong G (2012) Angew Chem Int Ed
51:13075
5. Parlett CMA, Durndell LJ, Wilson K, Bruce DW, Hondow NS,
Lee AF (2014) Catal Commun 44:40
39. Kaye PT, Musa MA, Nocanda XW, Robinson RS (2003) Org
Biomol Chem 1:1133
2
2
6. Muzart J (2003) Tetrahedon 59:5789
7. Keresszegi C, Bürgi T, Mallat T, Baiker A (2002) J Catal
40. Lesch B, Toräng J, Vanderheiden S, Bräse S (2005) Adv Synth
Catal 347:555
2
11:244
2
2
8. Mallat T, Baiker A (2004) Chem Rev 104:3037
9. Enache DI, Edwards JK, Landon P, Solsona-Espriu B, Carley
AF, Herzing AA, Watanabe M, Kiely CJ, Knight DW, Hutchings
GJ (2006) Science 211:362
41. Toräng J, Vanderheiden S, Nieger M, Bräse S (2007) Eur J Org
Chem 943
42. Gross U, Gross PJ, Shi M, Bräse S (2011) Synlett 635
43. Motschi H, Pregosin PS (1979) J Organomet Chem 171:C37
44. Motschi H, Pregosin PS, Ruegger H (1980) J Organomet Chem
193:397
3
3
3
0. aa
1. bb
2. Morad M, Sankar M, Cao E, Nowicka E, Davies TE, Miedziak
PJ, Morgan DJ, Knight DW, Bethell D, Gavriilidis A, Hutchings
GJ (2014) Catal Sci Technol 4:3120
45. Clark HC, Kurosawa H (1972) J Chem Soc Chem Commun 150
1
3