Thermal and Photochemical Switching of Chiral Sugar Azoalkenes: A Mechanistic Interrogation

Article abstract of DOI:10.1002/ejoc.202000029

This manuscript describes a thorough study on the thermal and photochemical evolution of carbohydrate-based azoalkenes, which have been proposed as putative intermediates en route to other heterocyclic derivatives and nucleoside analogues. These substances represent new protagonists in the azo chemical space, under intense study due to reversible photoswitching of this functional group suitable for designing artificial nanomachines and responsive materials. Although structurally simple azadienes derived from monosaccharides have long been known, the dynamics of such species in solution and solid phase remains poorly characterized. Herein, we show that such azoalkenes undergo 3E→3Z thermal isomerization, which is accelerated by the presence of Br?nsted acids. On the other hand, they undergo 1E→1Z isomerization when irradiated by sunlight, while the reverse 1Z→1E isomerization occurs thermally in the dark or under acid catalysis. As a result, sugar monoazadienes not only exhibit inherent chirality, but also dual on-off isomerization under external stimuli. Both experiments and DFT-based computational analyses provide a consistent and unifying mechanistic picture of these transformations.

Full text of DOI:10.1002/ejoc.202000029

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: Thermal and Photochemical Switching of Chiral Sugar
Azoalkenes. A Mechanistic Interrogation
Authors: Ana María Sánchez León, Pedro Cintas, Mark E. Light, and
Juan Carlos Palacios
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.202000029
Link to VoR: https://doi.org/10.1002/ejoc.202000029
01/2020

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Thermal and Photochemical Switching of Chiral Sugar
Azoalkenes. A Mechanistic Interrogation
Ana María Sánchez-León,*[a] Pedro Cintas,[a] Mark E. Light,[b] and Juan Carlos Palacios*[a]
Dedicated to the memory of María Dolores Méndez
[a]
Dr. A. M. Sánchez-León, Dr. P. Cintas, Dr. J. C. Palacios
Departamento de Química Orgánica e Inorgánica, Facultad de Ciencias, and
IACYS-Unidad de Química Verde y Desarrollo Sostenible,
Universidad de Extremadura,
E-06006 Badajoz, Spain,
E-mail: anasanchez@unex.es; palacios@unex.es
Dr. M. E. Light
[b]
Department of Chemistry, Faculty of Natural and Environmental Sciences,
University of Southampton,
Southampton SO17 1BJ, U.K.
Supporting information and the ORCID identification number(s) for the author(s) of this article can be found via a link at the end of the document.
Abstract: This manuscript describes a thorough study on the
thermal and photochemical evolution of carbohydrate-based
azoalkenes, which have been proposed as putative intermediates en
route to other heterocyclic derivatives and nucleoside analogs.
These substances represent new protagonists in the azo chemical
space, under intense study due to reversible photoswitching of this
functional group suitable for designing artificial nanomachines and
responsive materials. Although structurally simple azadienes derived
from monosaccharides have long been known, the dynamics of such
species in solution and solid phase remains poorly characterized.
Here, we show that such azoalkenes undergo 3E→3Z thermal
isomerization, which is accelerated by the presence of Brönsted
acids. On the other hand, they undergo 1E→1Z isomerization when
irradiated by sunlight, while the reverse 1Z→1E isomerization occurs
thermally in the dark or under acid catalysis. As a result, sugar
monoazadienes not only exhibit inherent chirality, but also dual on-
off isomerization under external stimuli. Both experiments and DFT-
based computational analyses provide a consistent and unifying
mechanistic picture of these transformations.
Introduction
The azo functional group has been a versatile structural motif in
synthetic chemistry, whose importance dates back to the late
19th century, ranging from the preparation of the first azo-dyes
to the seminal discovery of aromatic azo compounds as in-vivo
prodrugs of sulfa drugs. The facile cis-trans isomerization of the
azo group, usually achieved upon photoirradiation, thereby
leading to salient luminescent changes, has attracted
considerable attention in recent years. Substituted azobenzenes
can thus work as photoswitches that can further be harnessed
as artificial molecular machines,[1] photoresponsive materials,[2]
as well as (bio)sensors and actuators.[3] Moreover, the
photoswitchable double bonds can be incorporated into active
catalysts,[4] and the azo group has shown a directing effect in
organocatalysis.[5] In addition, the term azolog space has been
introduced to denote azologation of chemical groups (i.e.
azoisosteres) for optical control of biological targets.[6]
Scheme 1. Synthesis of carbohydrate-derived azoalkenes
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Scheme 2. Thermal and photochemical behavior of sugar 1,2-diaza-1,3-butadienes
The putative existence of sugar azoalkenes (sugar 1,2-diaza-
1,3-butadienes) should be traced back to pioneering studies by
Wolfrom and associates during the acetylation of sugar
hydrazones,[7,8] although they proved to be elusive intermediates
for several decades, and either postulated as transient
intermediates or occasionally isolated in diverse reactions of
unsaturated carbohydrates.[9] A convenient two-step synthetic
procedure involves the condensation of unprotected
monosaccharides and aryl hydrazines followed by O-acetylation
and mild heating, thereby promoting elimination from a non-
isolable monoazo-alkene to the corresponding diazo-diene
(Scheme 1).[10] Within the usual pattern of azoderivatives
susceptible of photoactivation, these monosaccharidic
azadienes (4-9) can be regarded as modified azobenzenes,[11]
lacking a second aromatic ring. This arrangement is unusual as
most carbohydrate-containing azo-photoswitches incorporate
the common azobenzene unit into the framework.[12] The
preparation of azobenzene-based glycoamphiphiles opens the
door to novel materials that resemble glycolipids.[13]
Results and Discussion
Structure and stereochemistry of monosaccharide
azoalkenes
As mentioned, the synthesis of azoalkenes has been reported
elsewhere and involves the formation of aldose arylhydrazones
followed by subsequent acetylation and elimination reaction.[10]
According to this procedure, (1E,3E)-4-(tetra-O-acetyl-D-
arabino-tetritol-1-yl)-1-aryl-1,2-diaza-1,3-butadienes 4-8 can be
obtained from D-mannose and their D-lyxo isomer 9 from D-
galactose, which can be isolated as colored compounds in the
solid state and are stable enough, so long as they are protected
from heat sources and light. Their initial (1E,3E) configurations
(4a-9a) can be inferred from FT-IR and Raman spectra showing
the stretching band for the C=C bond at ~1650 cm-1. The intense
Raman band within 1450-1435 cm-1 has been assigned to the
stretching band of the azo group. The other active band in the
Raman spectrum, characteristic of the azo group, lying in the
range 1160-1140 cm-1, can be attributed to the N=N-Ar
symmetric bending (Figures S1-S10, SI).[14]
Here, we look back at monosaccharidic 1,2-diaza-1,3-
butadienes, describing structural and configurational aspects in
detail and focusing on their unexplored photophysical properties.
The O-acylated sugar chain imparts chirality but also
conformational
flexibility,
accompanied
by
enhanced
hydrophobicity. It is well established that reversible trans-to-cis
photoisomerization of the azobenzene motif can be tuned by
shifts at shorter and longer wavelengths, respectively (the latter
induced by thermal relaxation too). We show the dual lability of
the C=C and N=N bonds in thermal and photochemical
isomerizations (Scheme 2), often under mild conditions (sunlight
suffices), and assessing the mechanistic aspects by both
experiment and theoretical calculations. Overall, this dynamic
behavior paves the way to a new class of chiral switches derived
from cheap and abundant natural products.
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
For clarity with the remaining discussion, some NMR data
(gathered in Table 1, see also SI) should be mentioned as they
exhibit a diagnostic value in monitoring the fate of bond
isomerisation (Figures S11-S29). The lack of any other
significant signals in the NMR spectra of 4a-9a indicates that in
the dark these substances do exist in only one isomeric form,
which can be attributed to the (1E,3E) configuration. This
stereochemistry is based on spectroscopic data and thus, the E-
configuration for the carbon-carbon double bond is supported by
the chemical shifts of H-1 (~7.32 ppm) and H-2 (~6.75 ppm)
protons and large coupling constants (~13.5 Hz).[15] The E-
configuration attributed to the diazo moiety is also supported by
the shift of the H-2 signal. It has been reported that both (Z)-
olefinic[15] and (Z)-arylazo[16] compounds are unstable and
isomerize to the corresponding (E,E)-compounds. Protons H-2
and H-3 consistently show a chemical shift difference of ~0.75-
0.85 ppm (ΔδH = δH2-δH3), while for the olefinic carbons that
difference is ~ 13-16 ppm (ΔδC = δC1-δC2) (Table 1).
Figure 1. ORTEP diagram for the crystal structure of 5a with thermal ellipsoids
drawn at the 35% probability. Hydrogen atoms have been omitted for clarity.
Moreover, the unequivocal ground configuration could be
corroborated by single-crystal X-ray analyses of compounds
5a[17] and 6a.[18] As shown in Figure 1 (see Supporting
information), the unit cell of compound 5a consists of two
molecules having each the (E,E) configuration and s-trans
disposition between the azo and ethylene groups. To determine
the robustness of the stereochemical arrangement, we
optimized the structure of 5a in the gas phase with DFT
calculations (Figure 2), and Table 2 compares the experimental
distances and bond angles with the calculated ones, which
match each other to a significant extent. Both the experimental
and calculated data indicate that the azodienic system in 5a is
essentially flat or slightly puckered. The same conclusion can be
inferred from the relative dispositions of the azo group and the
benzene ring.
Figure 2. Optimized structure of 5a (monomer) at the M06-2X/6-311G(d,p)
level in the gas phase.
Table 1. Selected NMR data for azoalkenes 4a-9a[a]
[b]
[c]
Isomer
4a
H-1
7.33
7.28
7.30
7.30
7.30
7.38
H-2
6.78
6.64
6.71
6.78
6.79
6.78
H-3
5.93
5.88
5.91
5.91
5.91
ΔδH
J1,2
C-1
C-2
ΔδC
0.85
0.76
0.81
0.86
0.87
0.87
13.5
13.5
13.6
13.5
13.5
13.4
149.9
150.1
150.2
149.9
149.9
151.2
136.4
134.2
135.4
137.0
137.2
135.8
13.5
15.9
14.6
12.8
12.7
15.4
5a
6a
7a
8a
9a
5.71
[a]Recorded in CDCl3 at 500 MHz and 33 ºC. [b]ΔδH = δH2-δH3
.
[c]ΔδC = δC1-δC2
.
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Table 2. Experimental and calculated bond distances (Å) and angles (º) of 5a.
Calculated[a]
1.246
Experimental (X-ray data)[b]
N=N
N-C
1.274
1.414
1.264
1.402
1.406
C=C
1.331
1.321
1.333
C-C
1.494
1.499
1.503
C-N=N
N=N-C
N-C=C
C=C-C
Car=Car-N=N
N=N-C=C
114.981
112.219
119.488
122.798
-178.825
-178.21
113.052
112.525
118.669
121.372
-165.471
-170.556
112.702
111.888
118.168
121.55
-161.492
-171.503
[a]At the M06-2X/6-311G(d,p) level in the gas phase. [b]Each column denotes data for the two monomers present in the unit cell.
Dynamics of chiral azoalkenes in solution
Remarkably, azoalkenes 4a-9a undergo further evolution in
solution, even under mild conditions. Most transformations are
driven by an initial E/Z isomerization of the C=C bond. Thus,
azoalkenes 4a-9a isomerize (in part) into the corresponding
(1E,3Z) derivatives (4b-9b) in CDCl3 solution to give after 24 h a
~5:1 equilibrium mixture (Scheme 3), as inferred from 1H and 13
C
NMR monitoring. This evolution is exemplified by compounds 5a
and 9a (Figures S30-S37) and depicted in Figure 3. All attempts
to isolate single isomers failed and led invariably to mixtures of
both azoalkenes in variable ratios. The H-2 and H-3 protons are
markedly different in both stereoisomers (compare Tables 1 and
3) and their chemical shifts are interchanged, thereby having a
diagnostic value as mentioned above. These variations in
chemical shift (H-2 ~ 0.7 ppm and H-3 ~ -1.0 ppm) agree with
data for other azoalkenes.[15] Now, the H-2 and H-3 protons
show a chemical shift difference (ΔδH) of approximately -0.7 to -
0.9 ppm; that is, the absolute value of Δδ is similar to that for 4a-
9a but negative (Table 3). Moreover, one of the acetate groups
(presumably located on C-3) appears appreciably shielded at
~1.90 ppm.
Scheme 3. E/Z isomerization of azoalkenes 4a-9a in CDCl3 solution.
Table 3. Population of the minor isomer at the equilibrium and selected NMR data for azoalkenes 4b-9b[a]
[c]
[d]
[e]
Isomer
4b
[3Z][b]
18
19
14
12
8
ΔG0
0.94
0.90
1.12
1.23
1.51
1.03
H-1
7.21
7.14
7.17
7.18
7.18
7.25
H-2
6.10
5.98
6.04
6.11
6.12
6.18
H-3
6.89
6.86
6.87
6.85
6.75
6.82
ΔδH
J1,2
7.7
7.7
8.1
7.6
7.6
8.0
C-1
C-2
ΔδC
-0.80
-0.88
-0.84
-0.74
-0.72
-0.72
147.6
147.0
147.6
147.6
147.6
148.4
131.9
129.8
130.9
132.4
132.6
131.8
15.7
17.2
16.7
15.2
15.0
16.6
5b
6b
7b
8b
9b
16
[a]In CDCl3 solution at 500 MHz and 33 ºC. [b]In %. [c]In kcal/mol. [d]ΔδH = δH2-δH3 [e]ΔδC = δC1-δC2
. .
4
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
H-2
H-1
H-3
b
H-1
H-6, 6’
H-2
H-4
H-3
H-5
a
8.00
ppm
7.50
7.00
6.50
6.00
5.50
5.00
4.50
Figure 3. Evolution of compound 5a in solution (1H NMR spectrum in CDCl3 at 500 MHz): (a) just-dissolved sample; (b) after equilibration with its isomer 5b.
As expected, the coupling constants are also useful in assessing
this configurational change, because for the olefinic protons J
varies from ~13.5 Hz (1E,3E configuration) to 8 Hz in the
isomerized (1E,3Z) azoalkene. In addition, the upfield chemical
shifts observed for the C-1 and C-2 signals are instrumental
when one moves from a (1E,3E) configuration to its isomeric
(1E,3Z) form (|C-1| ~ 2.5-3 ppm and |C-2| ~ 4-5 ppm).
However, the difference of the olefinic carbons is similar to that
shown by 4a-9a (15-17 ppm) (see Table 3). These results in
CDCl3 agree with the pattern observed for other
azoalkenes,[15,16] although they disagree with those of Jørgensen
and Pedersen, who in a further study reported that azoalkene 4a
in CDCl3 solution undergoes Z/E isomerization around the N=N
bond as both isomers show trans-2,3-couplings of 13.5 Hz.[19]
However, the present reinvestigation of 4a points clearly to the
same behaviour as compounds 5a-9a, in terms of both chemical
shifts and coupling constants (data collected in Tables 1 and 3).
Photoisomerization of azoalkenes
Scheme 4. Photoisomerization of azoalkenes 4a-9a in benzene solution.
(1E,3E)-Azoalkenes 4a-9a undergo partial isomerization when
their benzene solutions are irradiated with sunlight (Scheme 4).
After a few minutes of light exposure a photostationary state is
attained, constituted by a mixture of two isomers (~10:1, Table
4). UV-V spectra of 4a-9a show absorptions at ~340-300 nm
corresponding to the ππ* transition and responsible for their
photoisomerization.[10a] Consistent with spectroscopic data, the
emerging isomers 4c-9c should have the (1Z,3E) configuration
because their large coupling constants (J1,2 ~13.5 Hz) are
coincidental with those of the (1E,3E) isomers (Table 4). In fact,
all coupling constants show an identical pattern. Furthermore,
the olefinic protons and H-3 exhibit small chemical shift
variations with respect to 4a-9a (ΔH-1 and ΔH-2 < 0.2 ppm, and
IΔH-3I < 0.3 ppm), whereas there is a strong shielding effect for
In the subsequent 24 h after sunlight exposure (for 1 h), another
stereoisomer could be detected in the reaction mixture, whose
configuration at the azoalkene moiety should be (1E,3Z) as
supported by
a lower J1,2 coupling (~8.0 Hz) and the
corresponding chemical shifts (the H-2 proton resonates at ~6.8
ppm and H-3 at ~5.9 ppm). When the azoalkene-containing
mixture is stored in the dark the (1Z,3E) isomer gradually
disappears, thus evidencing its lower thermodynamic stability
relative to the other azoalkenes. This behavior is well illustrated
by the evolutionary conversion of 5a in benzene solution (Figure
4). After exposing the flask to sunlight radiation, a mixture of
5a/5c in 94:6 ratio can be identified. After 24 h in the dark, the
resulting stereoisomeric mixture contains a distribution of 5a
(1E,3E), 5b (1E,3Z), and 5c (1Z,3E) in a 92:2:6 ratio.
1
the ortho protons at the aromatic ring (|ΔHorto| ~ 1.0 ppm). H
NMR, COSY and HMQC spectra of 4a-9a after 1 h of light
exposure is gathered in Figures S38 and S53 (see SI).
5
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Table 4. Selected NMR data for (1Z,3E)-configured azoalkenes 4c-9c[a]
[c]
[d]
Isomer
4c
[1Z][b]
H-1
7.23
7.38
7.32
7.11
7.10
7.28
H-2
6.93
6.87
6.79
6.92
6.91
6.96
H-3
5.74
5.79
5.76
5.75
5.75
ΔδH
J1,2
C-1
C-2
ΔδC
3.2
6.6
4.7
2.5
2.4
4.5
10
6
1.19
1.08
1.15
1.17
1.16
1.47
13.5
13.5
13.5
13.5
13.5
13.5
141.7
142.5
142.1
141.3
141.6
142.6
138.5
135.9
137.4
138.8
139.2
138.1
5c
6c
7
7c
9
8c
9
9c
10
5.49
[a]Recorded in C6D6 at 500 MHz and 33 ºC. [b]In %. [c]ΔδH = δH1-δH2
.
[d]ΔδC = δC1-δC2
.
d
c
H-2
H-4
H-3
H-4
H-5
b
H-2
H-3
H-6
H-5
H-6’
a
7.75
7.50
7.25
7.00
6.75
6.50
6.25
6.00
5.75
5.50
5.25
5.00
4.75
4.50
4.25
ppm
Figure 4. Evolution of compound 5a in C6D6 solution: (a) just dissolved; (b) after 1 h of solar irradiation; (c) after 24 h in the dark; (d) after 5 days in the dark.
The identification of 5b (1E,3Z) demands a further rationale, as
its formation in the dark without acid catalysis is not immediately
obvious. Such acid conditions cannot be completely excluded
and minor degradation at the azoalkene moiety could take place.
The (1Z,3E) isomer might also be involved, albeit this surmise
should be taken with caution in the absence of clear-cut
evidences. The fact that the (1Z,3E) isomer relaxes into (1E,3E)
form in the dark, even without any acid catalysis, does suggest
that the Z-azo bond might relax faster in the presence of acid
traces. Overall, formation of the (1E,3Z) isomer would proceed
through the same intermediate (vide infra, acid-catalyzed
mechanism).
After prolonged storing in the dark (5 days) the isomeric
composition moves to 96:3:1. The (1Z,3E) isomer has virtually
disappeared, while the (1E,3Z)-configured azoalkene remains
unaffected; a fact supporting the slow conversion of 5c into 5a in
the dark. The same 1E→1Z photoisomerization can be observed
in DMSO-d6 (~7%) and perdeuterated acetone (~8%). However,
under solar irradiation in CDCl3 only the (1E,3Z) isomer could be
detected (7%). As rule of thumb, the azo group of azoalkenes is
prone to 1E→1Z photoisomerization when irradiated with
sunlight whereas the opposite 1Z→1E isomerization occurs
thermally in the dark.
Stability of azoalkene stereoisomers
Table 3 collects the equilibrium composition of (1E,3E)/(1E,3Z)-
configured azoalkenes. At the experimentally measured
temperature (33 ºC; 306 K) their difference in stability can easily
be assessed by ΔGo = RTlnKe, where Ke = [E]/[Z]. On using the
above tabulated data, we found energy differences ranging from
0.9 to 1.5 kcal/mol, favoring invariably the (1E,3E) stereoisomer.
A computational study has been undertaken to determine the
relative stability of all stereoisomers involving double bond
configurations; namely (1E,3E) (a), (1E,3Z) (b) and (1Z,3E) (c),
which have been observed experimentally, and also (1Z,3Z)
azoalkenes (d).
6
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Initially, gas-phase theoretical calculations with complete
geometry optimization at the M06-2X/6-311G(d,p) level,[20,21] as
implemented in the Gaussian09 package of programs,[22] were
performed to ascertain the relative stability of diastereomeric
structures for model compounds 10-12, as well as the electronic
influence exerted by the aliphatic substituents and their
propensity to interconvert. Further calculations of vibrational
frequencies have also been performed to estimate the
isomerization energies in terms of free energies. Bulk solvation
(through dielectric constants for C6H6, CHCl3, and DMSO) has
also been assessed with the usual solvation model based on
density (SMD).[23] The relative stabilities of the different
diastereomers of 10-12 are summarized in Table 5, which reflect
the greater stability of the (1E,3E) isomer in the cases of 10 and
11 and approximately equal stability between the (1E,3E) and
(1E,3Z) isomers of 12. Azoalkenes (1Z,3E) and (1Z,3Z),
showing similar stability, are however appreciably less stable
than the corresponding (1E,3E) and (1E,3Z) diastereomers by
more than 10 kcal/mol.
The gas-phase calculations for compounds 4, 5, 7, and 9 predict
a greater stability of (1E,3E) and (1E,3Z) diastereomers than
their (1Z,3E) and (1Z,3Z) counterparts. Isomerization around the
double carbon-carbon bond causes a minor change on moving
from (1E,3E) to (1E,3Z) or from (1Z,3E) to (1Z,3Z) configurations
[ΔG1E,3E↔1E,3Z and ΔG1Z,3E↔1Z,3Z < |2.3| kcal/mol]. In stark contrast,
the isomerization around the N=N bond leads to much more
significant variations [ΔG1E,3E↔1Z,3E
>
|8| kcal/mol and
ΔG1E,3Z↔1Z,3Z > |12| kcal/mol]. Such a destabilization could be
ascribed on the one hand to the steric repulsion between the
ortho hydrogens, relative to the azo group, and the vinyl
hydrogen at C-1, and on the other to the steric repulsion
between lone pairs on the azo functional group when they adopt
an eclipsed arrangement in the (1Z) configuration.[24] Both
repulsive effects disappear on adopting the (1E) configuration.
Table 5. Relative stability of diastereomeric azoalkene structures[a,b]
Configurations
Compound
a(1E,3E) b(1E,3Z)
c(1Z,3E)
d(1Z,3Z)
X
H
Ɛ
1
ΔE
ΔG
ΔE
ΔG
ΔE
ΔG
ΔE
ΔG
4[c]
4[d]
4[e]
4[f]
2.08
2.02
1.32
0.27
2.14
2.01
1.49
0.78
1.80
1.72
1.16
0.65
0.00
0.00
0.00
0.00
0.00
0.00
0.00
2.17
1.66
0.50
0.00
1.45
2.28
2.18
0.20
1.38
2.33
2.11
2.14
0.00
0.00
0.00
0.00
0.00
0.00
0.10
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.29
0.53
0.63
0.79
0.36
0.41
0.46
0.00
0.00
0.00
0.13
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.43
1.37
1.01
1.86
0.46
1.10
0.00
11.41
11.87
10.49
9.18
12.54
11.57
9.87
11.90
11.92
11.26
10.48
11.59
11.77
11.09
10.18
10.94
11.24
10.46
9.60
12.06
11.16
10.28
10.90
12.20
13.42
12.62
11.27
12.17
12.16
13.01
12.46
13.97
13.98
13.85
11.68
12.29
12.33
11.74
H
2.3
4.8
46.7
1
H
H
8.82
5[c]
5[d]
5[e]
5[f]
OMe
OMe
OMe
OMe
Cl
12.74
13.06
11.62
10.10
11.30
11.67
10.26
9.13
13.08
13.93
12.94
11.42
11.33
12.89
11.99
10.59
11.85
12.25
11.55
10.08
12.25
11.33
11.44
2.3
4.8
46.7
1
7[c]
7[d]
7[e]
7[f]
Cl
2.3
4.8
46.7
1
Cl
Cl
9[c]
9[d]
9[e]
9[f]
H
12.52
12.20
11.27
10.18
12.84
11.80
11.53
14.27
13.63
12.69
11.23
12.76
11.88
11.82
H
2.3
4.8
46.7
1
H
H
10[c]
11[c]
12[c]
OMe
H
1
NO2
1
[a]In kcal/mol. [b]At the M06-2X/6-311G(d,p) level. [c]In the gas phase. [d]Including the solvent effect (SMD model, benzene as solvent, ε = 2.3). [e]Including the
solvent effect (SMD model, CHCl3 as solvent, ε = 4.8). [f]Including the solvent effect (SMD model, DMSO as solvent, ε = 46.7).
7
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Inclusion of solvent effects does not alter appreciably the results
found for compounds 4, 5, 7, and 9 in the gas phase.
Computation reveals that the (1E,3E) and (1E,3Z) diastereomers
have nearly identical energies. However, an increase in the
polarity of the solvent also increases the stability of the isomer
(1E,3E). This agrees relatively well with experimental data in
CHCl3 solution for 4-9, because only (1E,3E)- and (1E,3Z)-
configured azoalkenes could be observed. Despite the high
energy gap between (1Z,3E) and (1E,3E) isomers, sunlight
irradiation provides enough activation to generate the former.
This pronounced difference in stability justifies otherwise the
transformation in the dark into the more stable all-(E)-configured
azoalkenes.
The conformational profiles for 11a/11b and 11c/11d are almost
coincidental, with the s-trans arrangement (θN1-N2-C1-C2 ~180°)
being the most stable conformer (> 3.4 kcal/mol). Twisted s-cis
conformers show a dihedral angle θN1-N2-C1-C2 of ~30-35º for
azoalkenes featuring both (1E,3E) and (1E,3Z) configurations,
while this angle is greater (~55-60º) for 11c (1Z,3E) and 11d
(1Z,3Z) due to an increase of repulsive effects at θ ~0º (moving
from
9
to 17 kcal/mol). However, those twisted s-cis
conformations should be adopted for hetero-Diels-Alder to occur
and show low interconversion barriers.[25,26] For comparative
effects, the conformational behavior of the most stable
stereoisomers 5a and 5b has been evaluated and the resulting
profiles are similar to those of 11a and 11b (Figure 7).
-458.285
-458.290
-458.295
Lastly, it is pertinent to emphasize that model compounds 10-12
evidence the enhanced stability of (1E,3E) isomers, a trend
verified in part for azoalkenes bearing a polyacetoxyl side chain.
The observed discrepancy may be due in part to the presence of
this chain, which can adopt a large number of conformations that
have not been calculated to reduce the high computational cost
involved. In addition, the model used to simulate solvent effects
(bulk solvation) might not be appropriate as no discrete
interactions with a given number of solvent molecules were
taken into account either.
-458.300
11c
11d
-458.305
-458.310
-458.315
Conformational analysis
-180
-120
-60
0
60
120
180
q
N1-N2-C1-C2 (º)
Given the acyclic nature of the sugar side chain, to gain further
insights into azoalkene stability, a conformational study has
been performed on the four diastereomers of 11 (Figure 8),
structure devoid of substitution at the aromatic ring, together
with the (1E,3E) and (1E,3Z) isomers of 5.
Figure 6. Conformational analyses of 11c (1Z,3E isomer) and 11d (1Z,3Z
isomer) by rotation around the dihedral angle θN1-N2-C1-C2. Minima and maxima
(labeled in red) have been calculated without geometrical restrictions.
-1602.140
To this end the rotational energy, calculated through 15º-steps,
around the dihedral angle θN1-N2-C1-C2 has been determined.
Conformational minima and maxima have been estimated
without any geometrical restriction; with maxima identified by
frequency analysis (they show one and only one imaginary
frequency). The resulting conformational profiles are collected in
Figures 5 and 6 and Table 6 shows the energy values for both
minima and maxima.
-1602.145
5a
5b
-1602.150
-1602.155
-1602.160
-458.315
-458.320
-180
-120
-60
0
60
120
180
q
N1-N2-C1-C2 (ꢀ)
11a
11b
Figure 7. Conformational analysis of 5a (1E,3E isomer) and 5b (1E,3Z
isomer) by rotation around the diedral angle θN1-N2-C1-C2. Minima and maxima
-458.325
(labeled in red) have been calculated without geometrical restrictions.
-458.330
-458.335
-180
-120
-60
0
60
120
180
q
N1-N2-C1-C2 (º)
Figure 5. Conformational analyses of 11a (1E,3E isomer) and 11b (1E,3Z
isomer) by rotation around the dihedral angle θN1-N2-C1-C2. Minima and maxima
(labeled in red) have been calculated without geometrical restrictions.
8
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Figure 8. Optimized structures for geometrical isomers of 11 in the gas phase.
Table 6. Relative stability of conformers derived from azoalkenes 11a, 11b, 5a and 5b[a]
θ[b]
ΔE[c]
ΔG[c]
θ [b]
0.0°
4.45
5.07
0.0°
29.4°
4.16
4.27
36.6°
4.94
4.87
34.7°
3.71
2.67
30.6°
1.32
1.57
97.1°
180.0°
262.9°
8.77
330.6°
4.16
11a (1E, 3E)
11b (1E, 3Z)
5a (1E, 3E)
5b (1E, 3Z)
8.77
0.00
8.69
0.00
8.69
4.28
95.8°
8.01
179.9°
0.00
264.2°
8.01
323.4°
4.94
ΔE[c]
ΔG[c]
θ[b]
5.57
6.58
10.2°
3.97
4.53
7.40
0.00
7.40
4.87
100.0°
6.84
181.3°
0.00
261.0°
5.91
322.5°
1.73
ΔE[c]
ΔG[c]
θ[b]
6.67
0.00
5.78
2.35
100.3°
5.38
174.8°
0.00
257.0°
6.95
ΔE[c]
ΔG[c]
5.67
0.00
6.09
[a]At the M06-2X/6-311G(d,p) level in the gas phase. [b]Dihedral angle θN1-N2-C1-C2 [c]In kcal/mol.
.
1H NMR data for stereoisomers a(1E,3E), b(1E,3Z) and
c(1Z,3E) of compounds 4-9 allow us to elucidate the
conformational preference in solution. Previous work dealing
with simple azoalkenes indicated that the photochemical
transformation of (1E,3Z)-phenylazocyclohexene into its (1Z,3Z)-
stereoisomer causes a significant variation of the olefinic
resonance (ΔδH-2 ~1.5 ppm), a fact rightly attributed to the
effective shielding generated by the aromatic ring placed in non-
planar arrangement with the N=N bond in the twisted s-cis
conformation (Scheme 5).[15] The latter, however, contrasts with
the photoisomerization of 4a-9a, for which chemical shift
variations experienced by the H-2 proton are negligible (ΔδH-2
≤
0.27 ppm). Accordingly, the preferential conformation adopted
by 4c-9c in solution should be s-trans; as the opposite s-cis
disposition would have caused a substantial shielding of the H-2
signal (δH-2 ~ 5.4 ppm) (Scheme 6). Such conclusive statements
agree with the theoretical calculations collected in Tables 5 and
Scheme 5. Proton shift variation of the olefinic resonance in the
photochemical transformation of phenylazocyclohexene, taken from Ref. [15].
6
pointing to s-trans conformations as the most stable
arrangement. The remaining framework should have the same
conformation as that of (1E,3E) isomers because of identical
coupling constants.
9
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
bearing the aromatic ring. Further conrotatory ring-opening
accounts for equilibration between (1E,3E) and (1E,3Z)
azoalkenes. An overall picture is shown in Scheme 7 based on
azoalkene 11 as simplified model.
Scheme 7. Mechanistic proposal applied to the thermal isomerization of
azoalkene 11, according to Ref. [16].
Figure 9 and Table 7 collect the relative electronic and free
energies of all stationary points involved in the potential energy
hypersurface of 11a↔11b isomerization. The gas-phase
computation shows that, despite the low activation energy (TS15
<4 kcal/mol) involving the cis-trans isomerization of
diazocyclobutenes 14 and 15, their formation from the
corresponding azoalkenes proceeds through high barriers (TS14
and TS16 >39 kcal/mol), a fact reflecting presumably the strain of
the four-membered ring (Figure 10). Related ring closures of
substituted 1,3-butadienes to cyclobutenes show similar
activation energies (a high energy barrier, ΔG≠ ~32 kcal/mol, has
Scheme 6. Dynamic equilibria involving s-trans and s-cis conformations for
diazadienes showing different E- and Z-configurations.
On the mechanism of azoalkene stereomutation
The conrotatory ring-closure mechanism
been determined for cyclobutene formation).[27]
Inclusion of
A
mechanistic proposal for the thermal isomerization of
azoalkenes was introduced by Schantl and Hebeisen,[16] where
s-cis conformers arising from both (1E,3E) and (1E,3Z)
azoalkenes would lead to trans- and cis-diazocyclobutenes,
respectively, by a conrotatory ring-closure. Their interconversion
takes place by configurational inversion at the nitrogen atom
solvent effects (CHCl3, benzene, or DMSO) leads to similar
results, which are also nearly coincidental with those of gas-
phase calculations (Table 7). This should otherwise be expected
for electrocyclic reactions with an isopolar transition state, which
are largely unaffected by solvent variations.[28]
Vaccum
Chloroform
Benzene
DMSO
ΔG = 44.33
ΔG = 39.21
ΔG = 15.35
ΔG = 12.74
ΔG = 12.19
ΔG = 4.28
ΔG = 8.51
ΔG = 8.69
14
15
ΔG = 5.97
16
13
ΔG = 1.01
ΔG = 0.00
11b
11a
Figure 9. Location of stationary points along the 11a (1E,3E) to 11b (1E,3Z) isomerization pathways. Color codes are employed to highlight the energy profile in
the gas phase and different solvent systems.
10
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Figure 10. Optimized structures for transition states along the 11a (1E,3E) to 11b (1E,3Z) isomerization path.
Table 7. Relative stability of stationary states for the 11a/11b isomerization[a,b]
11a
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
TS13
8.77
8.69
8.72
9.57
8.65
8.70
8.79
8.95
13
TS14
14
TS15
13.01
15.35
10.70
14.08
9.63
15
TS16
16
TS11b
8.42
8.51
8.36
10.23
8.30
7.61
8.37
8.09
11b
0.41
1.10
0.68
1.92
0.76
1.14
0.53
1.00
ΔE[c]
ΔG[c]
ΔE[d]
ΔG[d]
ΔE[e]
ΔG[e]
ΔE[f]
4.16
4.28
4.27
5.32
4.27
4.58
4.22
4.63
39.69
39.21
38.24
38.88
37.42
37.12
38.77
38.60
10.03
12.19
7.79
10.34
12.74
8.21
43.80
44.33
42.84
43.82
41.93
42.40
43.25
43.34
5.35
5.97
5.46
5.17
5.45
6.10
5.32
5.98
10.60
6.65
11.53
7.07
8.62
12.14
11.15
13.82
9.33
8.50
8.91
ΔG[f]
10.50
11.46
[a]In kcal/mol. [b]At the M06-2X/6-311G(d,p) level. [c]In the gas phase. [d]Including the solvent effect (SMD model: CHCl3). [e]Including the solvent effect (SMD model:
DMSO). [f]Including the solvent effect (SMD model: C6H6).
The acid-catalyzed mechanism
involve either protonation-deprotonation of the azo group,
thereby enabling facile rotation around the C1-C2 bond, or [1,4]-
addition of HCl to the azoalkene moiety, followed by rotation and
subsequent [1,4]-elimination, as depicted in Scheme 8 for the
interconversion of 4a/4b. Similar proposals have been
suggested to account for the cis-trans isomerization of diethyl
maleate[31] or 3-aryl propenylideniminium salts.[32]
It is worth noting however that isomerizations of compounds 4a-
9a into 4b-9b, respectively, are not observed, at least
appreciably, in non-halogenated solvents like benzene, toluene
or DMSO. Variable-temperature 1H NMR experiments show that,
although an increase in temperature does not produce any
change in a solution of 6a in benzene-d6 (ε = 2.3), the use of a
more polar solvent like DMSO-d6, (ε = 47), leads to equilibration
with its 6b isomer (Figures S54-S55).This behavior disagrees
with the above-mentioned mechanistic scheme,[16] because of
similar energy profiles regardless of the solvent system. On the
other hand, the intermediacy of diazetine derivatives has been
proposed for azoalkene isomerizations occurring under
photochemical or pyrolytic protocols,[29] which deviate from our
experimental conditions.
The plausibility of such transformations has now been assessed
by theoretical methods and data found for 4a are shown in Table
8. The specific chirality of the stereoisomeric intermediates 17
and 18 at the first stereogenic center has been computed too. A
mechanism involving addition-elimination of HCl seems unlikely
as the addition products are significantly more stable than the
corresponding azoalkenes (Figure 11). The concerted addition
of HCl through the s-cis conformation of 4a leading to 17 (R/S
mixture) appears to be improbable, because the corresponding
transition structures could not be located either.
The isomerization is strongly inhibited in CHCl3 after passing this
solvent through a column packed with basic alumina, thus
removing acid impurities, mainly HCl (generated by
photochemical decomposition). After one week however,
equilibration between (1E,3E) and (1E,3Z) isomers of 4 and 9 is
observed again. Clearly, the isomerization should be triggered
by such acidic side products. The lack of azoalkene stability in
CHCl3 solutions has been sometimes described.[15a,30]
Mechanistic hypotheses, compatible with the above facts, could
On the other hand, an alternative protonation mechanism
followed by rotation should also be unlikely in the gas phase as
it involves polar intermediates and high-energy transition
structures (ΔΔG >110 kcal/mol). Obviously, things may be
dramatically different when solvation is taken into account,
making it possible the isomerization in CHCl3 (ΔΔG < 31
kcal/mol).
11
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Scheme 8. Mechanistic hypotheses based on stepwise sequences of protonation-deprotonation, bond rotation, [1,4]-addition of HCl/bond rotation, and [1,4]-
elimination.
Table 8. Relative stability of stationary states for the 4a/4b isomerization[a,b]
4a + HCl
0.00
17 R
-23.50
-7.77
17 S
-23.68
-7.07
18 R
-19.04
-4.61
18 S
-24.59
-7.54
19 + Cl-
106.06
113.61
18.88
20 + Cl-
116.53
121.55
22.64
1b + HCl
-2.08
ΔE[c]
ΔG[c]
ΔE[d]
ΔG[d]
0.00
-2.17
0.00
-22.50
-21.65
-19.47
-23.37
-1.32
0.00
-7.44
-4.67
-4.87
-6.75
27.77
30.76
-0.50
[a]In kcal/mol. [b]At the M06-2X/6-311G(d,p) level. [c]In the gas phase. [d]Including the solvent effect (SMD model: CHCl3).
Figure 11. Energy profiles in the gas phase for the different isomerization pathways of (1E,3E) and (1E,3Z)-configured azoalkenes (values in parentheses refer to
CHCl3, SMD model).
12
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
1H NMR monitoring shows that addition of trifluoroacetic acid-d1
to a CDCl3 solution of either 4a or 9a (ca. 5:1 molar ratio, see SI
for details) triggers a fast equilibration between (1E,3E) and
(1E,3Z) isomers yielding a ratio coincidental with that obtained in
the absence of acid (Figures S56-S58). Fast equilibria are also
reached after addition of acetic acid-d4 to a solution of 5a in
CDCl3 (5b ~14%) or of D2SO4 (one drop) to a solution of 5a in
DMSO-d6. For the latter the isomeric ratio (5b ~16%) is similar to
that found in CDCl3. Further addition of acid does not change
such equilibria. Equilibration does not occur upon addition of
acetic acid-d4 to a solution of 5a in C6D6, which is however
initiated by adding D2SO4 (~13% of the minor isomer).
kcal mol-1 less stable than those with stereochemistry (1E,3E)
and (1E,3Z) (see Table 5), only the latter could be detected as
minor component in the final equilibrium mixture.
These results can be exploited to inhibit the photoisomerization
of azoalkenes in the presence of an appropriate acid. In fact,
when a solution of 6a in DMSO-d6, to which trifluoroacetic acid-
d1 had been previously added, was irradiated, only the (1E,3Z)
isomer (6b) was detected (Figure 61S).
Moreover, such experimental findings allow us to conclude that
under photochemical conditions and in the absence of acids,
only the 1E1Z isomerization takes place, whereas the
presence of acids leads exclusively to 3E3Z isomerization.
Notably, under no circumstances will both isomerizations be
occurring at the same time. We have found no data supporting
the presence of the (1Z,3Z) isomers, as their formation should
actually be difficult. The generation of such an isomer from its
(1E,3E) counterpart is not feasible, because both photochemical
isomerization plus acid catalysis acting simultaneously or
sequentially would be required.
These observations are more consistent with the protonation
mechanism than that of acid addition, as one would expect
marked reactivity differences of Brønsted acids in the addition-
elimination mechanism. In DMSO, acetic acid (pKa ~12.6)[33] is
weaker than trifluoroacetic acid (pKa ~0.23),[34] HCl (pKa
~
1.8),[35] and H2SO4 (pKa ~1.99)[36] with acetate anion being a
poorer leaving group than trifluoroacetate, chloride and bisulfate.
One wonders which mechanisms follow the opposite pathway
from the photoisomers (1Z,3E) 4c-9c to the starting isomers
(1E,3E) 4a-9a. When solutions of the latter in acetone-d6 or
DMSO-d6 are exposed to sunlight, they isomerize to the (1Z,3E)
isomers only. With the subsequent addition of trifluoroacetic
acid-d1, the isomer (1Z,3E) instantly disappears and the (1E,3Z)
derivative is formed. These results demonstrate that 1Z1E
isomerization can occur easily and quickly under acid catalysis.
This behavior is illustrated for 6a in Figures S59-S60 and
Scheme 9 shows the plausible mechanism accounting for these
transformations.
When solutions of compounds 4-9 were left to stand in CHCl3
solutions for several days, progressive browning was observed,
indicative of azoalkene degradation as inferred from TLC
analyses and NMR spectra.
Conclusions
The preceding results show for the first time the fate of chiral
azoalkenes derived from acyclic sugars, which undergo double
bond isomerizations. Both the structure and conformational
profile of such species in solution can be determined with
confidence through diagnostic spectroscopic signals and
computational analysis. Visible-light irradiation of 1E-configured
azoalkenes causes their isomerization into 1Z-isomers, while the
opposite 1Z→1E isomerization occurs thermally in the dark or
under acid catalysis. The mechanism of thermal stereomutation
has been investigated thoroughly by experimental and
theoretical methods. While
a
conrotatory ring-closure
mechanism, once proposed for simpler derivatives, should be
ruled out, acid-catalyzed transformations represent the most
plausible scenarios. However, protonation rather than addition-
elimination pathways, appears to be consistent with our
experimental findings. Thus, the 3E→3Z and 1Z→1E
isomerizations proceeds in chloroform solution and in the
presence of acids.
Experimental Section
Computational analyses
Scheme 9. Mechanistic hypothesis accounting for acid-catalyzed
isomerisation of (1Z,3E)-azoalkenes.
Cartesian coordinates for all computed species have been
included in the Supporting material. The DFT study has been
conducted using the M06-2X hybrid density functional[20] in
conjuction with the 6-311G(d,p) basis set[21] as implemented in
the Gaussian09 package.[22] The M06-2X functional was chosen
on the basis of previous studies showing its accuracy in
Thus, azoalkene 6a is photoisomerized to 6c, which in the
presence of acid is protonated (21). The resulting azonium ion
easily isomerizes via rotation since protonation produces a
decrease in the double bond character, not only between the
carbon atoms of the C=C bond but also between the nitrogen
atoms of the azo linkage,[37] thereby enabling their rotations (22
and 23). Given the fact that stereoisomers (1Z,3E) are ~10-12
determining
conformational
energies
of
non-covalent
interactions.[38] Frequency analysis were performed to confirm
the existence of true minima on the potential energy surface. All
13
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
after equilibration (%) = 81:19. 1H NMR (500 MHz, CDCl3)  7.80
(d, J = 8.8 Hz, 2H, Z), 7.76 (d, J = 9.8 Hz, 2H, E), 7.29 (dd, J =
13.5, J = 1.2 Hz, 1H, E), 7.14 (dd, J = 7.7, J = 1.0 Hz, 1H, Z),
6.98-6.89 (m, 2H E+2H Z), 6.869 (ddd, J = 8.1, J = 4.6, J = 0.8
Hz, 1H, Z), 6. 65 (dd, J = 13.6, J = 6.2 Hz, 1H, E), 5.98 (t, J = 8.1
Hz, 1H, Z), 5. 89 (ddd, J = 6.0, J = 3.4, J = 1.2 Hz, 1H, E), 5.52
(m, 1H, Z ), 5.50 (dd, J = 8.3, J = 3.5 Hz, 1H, E), 5.27 (ddd, J =
7.6, J = 4.4, J = 2.7 Hz, 1H, E), 5.27 (m, 1H, Z), 4.31 (m, 1H, Z),
4.29 (dd, J = 12.4, J = 2.6 Hz, 1H, E), 4.21 (m, 1H, Z), 4.20 (dd,
J = 12.6, J = 5.0 Hz, 1H, E), 3.87 (s, 3H, Z), 3.85 (s, 3H, E), 2.11
(s, 3H, E), 2.08 (s, 3H, E), 2.05 (s, 3H, E), 1.94 (s, 3H, Z); 13C
NMR (125 MHz, CDCl3)  170.4, 169.7, 169.6, 169.5, 169.4,
162.5, 162.3, 150.1, 147.6, 147.0, 146.6, 134.1, 129.8, 125.0,
124.7, 114.1, 71.7, 70.0, 69.3, 69.0, 68.3, 67.3, 61.8, 61.7, 55.4,
20.8, 20.6, 20.5, 20.5.
thermal corrections were calculated at the standard values of 1
atm at 298.15 K. Solvent effects were estimated by the
continuum solvation model, based on the quantum mechanical
charge density (SMD), and comprising the bulk electrostatic
contribution from a self-consistent reaction field treatment, plus a
second contribution from short-range solute-solvent interactions
in the first solvation cell (cavity-dispersion-solvent-structure
term) that takes into account different contribution such as long-
range electrostatic polarization (bulk solvent effects).[23,39]
Spectroscopic methods
Crystallographic data (acquisition method, refinement, and
structural parameters) as well as spectroscopic monitoring of
sugar azoalkenes in solution are collected in the Supporting
information. FT-IR and Raman spectra were recorded in the
range of 4000-600 cm-1 using KBr (Merck) pellets on a Thermo
spectrophotometer and a Nicolet Almega XR (Thermo scientific)
apparatus, respectively. NMR spectra were recorded on a
Bruker instrument operating at 400 and 500 MHz for 1H, and 100
and 125 MHz for 13C nuclei, respectively. Hydrogen and carbon
atom assignments were facilitated by DEPT (Distortionless
Enhancement by Polarization Transfer), COSY (Correlation
Spectroscopy) and HMQC (Heteronuclear Multiple-Quantum
Correlation) experiments and isotopic exchange after addition of
D2O for proton resonances.
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-methylphenyl)-1,2-diaza-1,3-butadiene (6a)
and
(1E,3Z)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-(4-
methoxyphenyl)-1,2-diaza-1,3-butadiene (6b): E/Z ratio after
equilibration (%) = 86:14. 1H NMR (500 MHz, CDCl3)  7.72 (d, J
= 8.3 Hz, 2H, Z), 7.59 (d, J = 8.3 Hz, 2H, E), 7.23 (dd, J = 13.6, J
= 1.2 Hz, 1H, E), 7.17 (d, J = 8.1 Hz, 2H E+2H Z), 7.10 (d, J =
7.9, 1H, Z), 6.87 (ddd, J = 8.4, J = 4.8 Hz, J = 1.0 Hz, 1H, Z),
6.65 (dd, J = 13.6, J = 6.0 Hz, 1H, E), 6.04 (t, J = 8.1 Hz, 1H, Z),
5.84 (ddd, J = 6.0, J = 3.3, J = 1.0 Hz, 1H, E), 5.53 (dd, J = 8.1,
J = 4.8, 1H, Z), 5.44 (dd, J = 8.5, J = 3.2, 1H, E), 5.21 (ddd, J =
8.0, J = 4.8, J = 2.7 Hz, 1H, E), 5.21 (m, 1H, Z), 4.31 (dd, J =
12.3, J = 2.6 Hz, 1H, Z), 4.21 (dd, J = 12.5, J = 2.7 Hz, 1H, E),
4.14 (m, 1H, Z), 4.12 (dd, J = 12.5, J = 4.9 Hz, 1H, E), 2.42 (s,
3H, Z), 2.31 (s, 3H, E), 2.06 (s, 3H, E), 2,01 (s, 3H, E), 2,00 (s,
3H, E), 1.85 (s, 3H, Z); 13C NMR (125 MHz, CDCl3)  170.4,
169.7, 169.6, 169.5, 169.4, 162.5, 150.7, 150.3, 150.0, 147.6,
142.3, 142.0, 135.4, 131.0, 129.6, 123.1, 122.8, 71.7, 70.0, 69.3,
69.0, 68.3, 67.3, 61.8, 61.7, 21.3, 20.8, 20.7, 20.6, 20.5.
NMR studies on the Isomerization of (1E,3E)-1,2-diaza-1,3-
butadienes (4a-9a) into (1E,3Z)-1,2-diaza-1,3-butadienes (4b-
9b). All samples were prepared by dissolving (1E,3E)-
azoalkenes (0.1 g) in CDCl3 (0.5 mL). Their evolution was
1
monitored by H NMR spectroscopy for 36 h. Formation of the
corresponding (1E,3Z) isomers (ca. 5:1) could be observed;
structural assignment and peak identification of each isomer
were performed as follows:
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-phenyl-1,2-diaza-1,3-butadiene
(4a)
and
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-chlorophenyl)-1,2-diaza-1,3-butadiene (7a)
(1E,3Z)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-phenyl-1,2-
diaza-1,3-butadiene (4b): E/Z ratio after equilibration (%) =
and
(1E,3Z)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-(4-
1
82:18. H NMR (500 MHz, CDCl3)  7.84-7.82 (m, 2H, Z), 7.78-
chlorophenyl)-1,2-diaza-1,3-butadiene (7b): E/Z ratio after
equilibration (%) = 88:12. 1H NMR (500 MHz, CDCl3)  7.76 (d, J
= 8.7, Hz, 2H, Z), 7.63 (m, 2H, E), 7.49-7.43 (m, 2H E+2H Z),
7.24 (dd, J = 13.5, J = 1.6 Hz, 1H E), 7.18 (dd, J = 7.6, J = 1.2
Hz, 1H, Z), 6.85 (ddd, J = 9.6, J = 5.2 Hz, J = 1.2 Hz, 1H, Z),
6.71 (dd, J = 13.5, J = 5.9 Hz, 1H, E), 6.11 (t, J = 8.4 Hz, 1H, Z),
5.85 (ddd, J = 5.5, J = 3.1, J = 1.3 Hz, 1H, E), 5.52 (m, 1H, Z),
5.45 (dd, J = 8.5, J = 3,3, 1H, E), 5.21 (ddd, J = 8,0, J = 4.7, J =
2.8 Hz, 1H, E), 5.31 (m, 1H, Z), 4.23 (dd, J = 12.3, J = 2.9 Hz,
1H, Z), 4.22 (dd, J = 12.5, J = 2.8 Hz, 1H, E), 4.14 (m, 1H, Z),
4.13 (dd, J = 12.5, J = 4.9 Hz, 1H, E), 2.07 (s, 3H, E), 2.03 (s,
3H, E), 2.01 (s, 3H, E), 1.89 (s, 3H, Z); 13C NMR (125 MHz,
CDCl3)  170.4, 169.8, 169.7, 169.6, 169.5, 150.9, 150.6, 149.8,
147.4, 137.6, 137.2, 137.0, 132.3, 129.3, 129.2, 124.3, 124.0,
71.6, 69.9, 69.2, 68.9, 68.3, 67.3, 61.7, 20.7, 20.6, 20.5.
7.76 (m, 2H, E), 7.50-7.45 (m, 3H, E+Z), 7.33 (dd, J = 13.5, J =
1.2 Hz, 1H, E), 7.21 (dd, J = 7.7, J = 0.9 Hz, 1H, Z), 6.89 (ddd, J
= 8.1, J = 4.6, J = 0.8 Hz, 1H, Z), 6.78 (dd, J = 13.6, J = 5.8 Hz,
1H, E), 6.10 (t, J = 8.0 Hz, 1H, Z), 5.94 (ddd, J = 5.8, J = 3.2, J =
1.2 Hz, 1H, E), 5.55 (m, 1H, Z ), 5.53 (dd, J = 8.4, J = 3.4 Hz,
1H, E), 5.29 (ddd, J = 8.0, J = 4.8, J = 2.7 Hz, 1H, E), 5.29 (m,
1H, Z), 4.31 (dd, J = 12.4, J = 2.8 Hz, 1H, Z), 4.30 (dd, J = 12.5,
J = 2.6 Hz, 1H, E), 4.23 (dd, J = 12.1, J = 4,6 Hz, 1H, Z), 4.21
(dd, J = 12.6, J = 4.6 Hz, 1H, E), 2.15 (s, 3H, E), 2.10 (s, 3H, E),
2.08 (s, 3H, E), 1.94 (s, 3H, Z); 13C NMR (125 MHz, CDCl3) 
170.6 (E+Z), 169.8 (E+Z), 169.7 (E+Z), 169.6 (E+Z), 152.6 (Z),
152.3 (E), 150.0 (E), 147.6 (Z), 136.5 (E), 131.9 (Z), 131.7 (Z),
131.4 (E), 129.1 (E+Z), 123.2 (Z), 122.9 (E), 71.7 (Z), 70.0 (E),
69.3 (E), 69.0 (Z), 68.3 (E), 67.4 (Z), 61.8 (E+Z), 20.8 (E+Z),
20.7 (E+Z), 20.6 (E+Z), 20.4 (E+Z).
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-bromophenyl)-1,2-diaza-1,3-butadiene (8a)
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-methoxyphenyl)-1,2-diaza-1,3-butadiene
(5a) and (1E,3Z)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-
(4-methoxyphenyl)-1,2-diaza-1,3-butadiene (5b): E/Z ratio
and
(1E,3Z)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-(4-
bromophenyl)-1,2-diaza-1,3-butadiene (8b): E/Z ratio after
1
equilibration (%) = 92:8. H NMR (500 MHz, CDCl3)  7.63 (d, J
14
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
= 8.6 Hz, 2H, Z), 7.57 (d, J = 8.8 Hz, 2H, Z), 7.57 (d, J = 8.8 Hz,
2H, E), 7.52 (d, J = 8.8 Hz, 2H, E), 7.24 (dd, J = 13.5, J = 1.1
Hz, 1H, E), 7.18 (dd, J = 7.6, J = 1.2 Hz, 1H, Z), 6.85 (ddd, J =
9.2, J = 5.0 Hz, J = 1.2 Hz, 1H, Z), 6.72 (dd, J = 13.5, J = 4.9 Hz,
1H, E), 6.12 (t, J = 8.2 Hz, 1H, Z), 5.85 (ddd, J = 4.9, J = 2.8, J =
1.1 Hz, 1H, E), 5.84 (dd, J =7.5 Hz, J = 4.0 Hz, 1H, Z), 5.41 (dd,
J = 8.5, J = 3.2, 1H, E), 5.21 (ddd, J = 7.9, J = 4.7, J = 2.7 Hz,
1H, E), 5.39 (dd, J = 12.5 Hz, J = 3.0 Hz, 1H, Z), 4.32 (dd, J =
12.5, J = 5.5 Hz, 1H, Z), 4.23 (dd, J = 12.5, J = 2.7 Hz, 1H, E),
4.13 (dd, J = 12.5, J = 4.8 Hz, 1H, E), 4.13 (m, 1H, Z), 2.07 (s,
3H, E), 2.03 (s, 3H, E), 2.00 (s, 3H, E), 1.89 (s, 3H, Z); 13C NMR
(125 MHz, CDCl3)  170.4, 169.8, 169.7, 169.5, 151.3, 150.9,
149.8, 147.4, 137.1, 132.6, 132.4, 132.3, 132.2, 131.9, 129.0,
126.2, 125.8, 124.5, 124.3, 123.2, 71.6, 69.9, 69.2, 68.9, 68.3,
67.3, 61.8, 61.7, 20.8, 20.7, 20.6, 20.5.
= 5.0, J(5,6) = 3.0 Hz, 1H, H-5, Z), 4.27 (dd, J(6,6’) = 12.5, J(5,6)
= 3.0 Hz, 1H, H-6, E), 4.21 (dd, J(6,6’) = 12.0, J(5,6) = 3.0 Hz,
1H, H-6, Z), 4.18 (dd, J(6,6’) = 12.5, J(5,6’) = 5,0 Hz, 1H, H-6’,
E), 4.11 (dd, J(6,6’) = 12.0, J(5,6’) = 5.0 Hz, 1H, H-6’, Z), 1.78 (s,
3H, CH3, E), 1.70 (s, 3H, CH3, E), 1.56 (s, 3H, CH3, E); 13C NMR
(125 MHz, C6D6)  169.9 (E), 169.5 (E), 169.4 (E), 169.2 (Z),
154.1 (Z), 152.9 (E), 150.5 (E), 141.7 (Z), 138.5 (Z), 137.4 (E),
131.5 (E), 129.3 (E), 129.2 (Z), 127.3 (Z), 123.5 (Z), 123.3 (E),
119.5 (Z), 72.1 (Z), 70.3 (E), 70.2 (Z), 70.1 (Z), 69.6 (Z), 69.5
(E), 68.7 (E), 62.0 (E), 61.9 (Z) 20.4 (E), 20.3 (Z), 20.2 (E), 20.1
(E), 20.0 (Z), 19.9 (E).
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-methoxyphenyl)-1,2-diaza-1,3-butadiene
(5a) and (1Z,3E)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-
(4-methoxyphenyl)-1,2-diaza-1,3-butadiene (5c): E/Z ratio
1
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-lyxo-
after equilibration (%) = 94:6. H NMR (500 MHz, C6D6)  7.89
tetritol-1-yl)-1-phenyl-1,2-diaza-1,3-butadiene
(9a)
and
(d, J = 9.0 Hz, 2H, arom, E), 7.65 (dd, J(1,2) = 14.0, J(1,3) = 1.0
Hz, 1H, H-1, E), 7.38 (dd, J(1,2) = 13.5, J(1,3) = 1.0 Hz, 1H, H-1,
Z), 6.87 (dd, J(1,2) = 13.5, J(2,3) = 6.5 Hz, 1H, H-2, Z), 6.86 (m,
1H, arom, Z), 6.84 (dd, J(1,2) = 14.0, J(2,3) = 6.0 Hz, 1H, H-2,
E), 6.66 (d, J = 9.0 Hz, 2H, arom, E), 6.60 (m, 1H, arom, Z), 6.06
(ddd, J(2,3) = 6.0, J(3,4) = 3.0, J(1,3) = 1.5 Hz, 1H, H-3, E), 5.79
(ddd, J(2,3) = 6.5, J(3,4) = 3.5, J(1,3) = 1.0 Hz, 1H, H-3, Z), 5.64
(dd, J(4,5) = 9.0, J(3,4) = 3.0 Hz, 1H, H-4, E), 5.56 (dd, J(4,5) =
8.5, J(3,4) = 3.5 Hz, 1H, H-4, Z), 5.47 (ddd, J(4,5) = 9.0, J(5,6’) =
5.0, J(5,6) = 3.0 Hz, 1H, H-5, E), 5.41 (ddd, J(4,5) = 8.5, J(5,6’)
= 5.0, J(5,6) = 2.5, Hz, 1H, H-5, Z), 4.28 (dd, J(6,6’) = 12.5,
J(5,6) = 3.0 Hz, 1H, H-6, E), 4.24 (dd, J(6,6’) = 12.5, J(5,6) = 3.0
Hz, 1H, H-6, Z), 4.19 (dd, J(6,6’) = 12.5, J(5,6’) = 5.0 Hz, 1H, H-
6’, E), 4.14 (dd, J(6,6’) = 12.5, J(5,6’) = 6.0 Hz, 1H, H-6’, Z), 3.17
(s, 3H, CH3, Z), 3.10 (s, 3H, CH3, E), 1.78 (s, 3H, CH3, E), 1.73
(s, 3H, CH3, E), 1.72 (s, 3H, CH3, Z), 1.71 (s, 3H, CH3, E), 1.69
(s, 3H, CH3, Z), 1.63 (s, 3H, CH3, Z), 1.62 (s, 3H, CH3, Z) , 1.59
(s, 3H, CH3, E); 13C NMR (125 MHz, C6D6)  169.4 (E), 169.5
(E), 169.4 (E), 162.8 (E) 150.8 (E), 147.4 (E), 142.5 (Z), 135.9
(Z), 135.0 (E), 125.6 (Z), 125.3 (E), 122.7 (Z), 114.6 (Z), 114.5
(E), 114.2 (Z), 70.4 (E), 69.9 (Z), 69.7 (E), 68.8 (E), 62.1 (E),
61.9 (Z), 55.0 (E), 54.9 (Z), 20.4 (E), 20.3 (Z), 20.2 (E), 20.1 (E),
19.9 (E).
(1E,3Z)-4-(tetra-O-acetyl-D-lyxo-tetritol-1-yl)-1-phenyl-1,2-
diaza-1,3-butadiene (9b): E/Z ratio after equilibration (%) =
84:16. 1H NMR (500 MHz, CDCl3)  7.85 (d, J = 8.0 Hz, 2H, Z),
7.78 (m, 2H, E), 7.49-7.46 (m, 3H E+3H Z), 7.38 (d, J = 13.4 Hz,
1H, E), 7.25 (dd, J = 8.2 Hz, J = 1.0 Hz, 1H, Z), 6.82 (ddd, J =
9.0 Hz, J = 6.0 Hz, J = 1.0 Hz, 1H, Z), 6.78 (dd, J = 13.7, J = 7.8
Hz, 1H, E), 6.18 (dd, J = 9.0 Hz, J = 8.2 Hz, 1H, Z), 5.72 (t, J =
7.2 Hz, 1H, E), 5.56 (dd, J = 6.0, J = 4.5 Hz, 1H, Z), 5.50-5.47
(m, 2H, E), 5.41 (ddd, J = 6.5, J = 5.0 Hz, J = 4.5 Hz, 1H, Z),
4.29 (dd, J = 11.8, J = 5.0 Hz, 1H, E), 4.23 (dd, J = 11.5, J = 5.0
Hz, 1H, Z), 4.13 (dd, J = 11.5, J = 6.5 Hz, 1H, Z), 4.04 (dd, J =
11.9, J = 6.2 Hz, 1H, E), 2.11 (s, 3H, E), 2.10 (s, 3H, E), 2.09 (s,
3H, E), 2.08 (s, 3H, Z), 2.07 (s, 3H, E), 2.04 (s, 3H, Z), 2.02 (s,
3H, Z); 13C NMR (125 MHz, CDCl3)  170.4 (E), 170.0 (E), 169.9
(Z), 169.8 (Z), 169.4 (Z), 152.5 (Z), 152.3 (E), 151.2 (E), 148.4
(Z), 135.8 (E), 131.8 (Z), 131.5 (E), 129.1 (E+Z), 123.3 (Z),
122.9 (E), 70.8 (E+Z), 69.1 (E), 68.5 (Z), 68.1 (E), 67.1 (Z), 61.9
(Z), 61.8 (E), 20.9 (E+Z), 20.8 (E+Z), 20.7 (E+Z).
Photoisomerization of (1E,3E)-1,2-diaza-1,3-butadienes (4a-
9a) into (1Z,3E)-1,2-diaza-1,3-butadienes (4c-9c): All samples
were prepared by dissolving (1E,3E)-azoalkenes (0.1 g) in
1
benzene (0.5 mL). Their evolution was monitored by H NMR
spectroscopy after 1 h of irradiation with sunlight. Formation of
the corresponding (1Z,3E) isomers could be observed; structural
assignment and peak identification of each isomer were
performed as follows:
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-methylphenyl)-1,2-diaza-1,3-butadiene (6a)
and
(1Z,3E)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-(4-
methylphenyl)-1,2-diaza-1,3-butadiene (6c): E/Z ratio after
equilibration (%) = 93:7. 1H NMR (500 MHz, C6D6)  7.83 (d, J =
8.5 Hz, 2H, arom, E), 7.63 (dd, J(1,2) = 13.5, J(1,3) = 1.5 Hz,
1H, H-1, E), 7.32 (dd, J(1,2) = 13.5, J(1,3) = 1.0 Hz, 1H, H-1, Z),
6.91 (d, J = 8.5 Hz, 2H, arom, E), 6.86 (dd, J(1,2) = 13.5, J(2,3)
= 6.0 Hz, 1H, H-2, E), 6.79 (dd, J(1,2) = 13.5, J(2,3) = 6.0 Hz,
1H, H-2, Z), 6.05 (ddd, J(2,3) = 5.5, J(3,4) = 3.0, J(1,3) = 1.5 Hz,
1H, H-3, E), 5.76 (ddd, J(2,3) = 5.5, J(3,4) = 3.0, J(1,3) = 1.5 Hz,
1H, H-3, Z), 5.63 (dd, J(4,5) = 9.0, J(3,4) = 3.0, 1H, H-4, E), 5.52
(dd, J(4,5) = 9.0, J(3,4) = 3.0, 1H, H-4, Z), 5.47 (ddd, J(4,5) =
9.0, J(5,6’) = 6.0, J(5,6) = 3.0 Hz, 1H, H-5, E), 5.39 (ddd, J(4,5)
= 9.0, J(5,6’) = 6.0, J(5,6) = 3.0 Hz, 1H, H-5, Z), 4.27 (dd, J(6,6’)
= 12.5, J(5,6) = 3.0 Hz, 1H, H-6, E), 4.22 (dd, J(6,6’) = 12.5,
J(5,6) = 3.0 Hz, 1H, H-6, Z), 4.17 (dd, J(6,6’) = 12.5, J(5,6’) = 5.0
Hz, 1H, H-6’, E), 4.12 (dd, J(6,6’) = 12.5, J(5,6’) = 5.0 Hz, 1H, H-
6’, Z), 1.97 (s, 3H, CH3, E), 1.78 (s, 3H, CH3, E), 1.71 (s, 3H,
CH3, E), 1.70 (s, 3H, CH3, E), 1.57 (s, 3H, CH3, E); 13C NMR
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-phenyl-1,2-diaza-1,3-butadiene
(4a)
and
(1Z,3E)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-phenyl-1,2-
diaza-1,3-butadiene (4c): E/Z ratio after equilibration (%) =
1
90:10. H NMR (500 MHz, C6D6)  7.86 (m, 2H, arom, E), 7.60
(dd, J(1,2) = 13.5, J(1,3) = 1.5 Hz, 1H, H-1, E), 7.23 (dd, J(1,2) =
13.5, J(1,3) = 1.5 Hz, 1H, H-1, Z), 7.08 (m, 2H, arom, E), 7.03
(m, 1H, E), 6.95 (m, 2H, arom, Z), 6.93 (dd, J(1,2) = 13.5, J(1,3)
= 6.0 Hz, 1H, H-2, Z), 6.88 (dd, J(1,2) = 13.5, J(1,3) = 5.5 Hz,
1H, H-2, E), 6.81 (m, 1H, arom, Z), 6.68 (m, 1H, arom, Z), 6.04
(ddd, J(2,3) = 5.5, J(3,4) = 3.0, J(1,3) = 1.5 Hz, 1H, H-3, E), 5.74
(ddd, J(2,3) = 6.0, J(3,4) = 3.0, J(1,3) = 1.5 Hz, 1H, H-3, Z), 5.62
(dd, J(4,5) = 9.0, J(3,4) = 3.0 Hz, 1H, H-4, E), 5.50 (dd, J(4,5) =
8.5, J(3,4) = 3.0 Hz, 1H, H-4, Z), 5.46 (ddd, J(4,5) = 9.0, J(5,6’) =
5.0, J(5,6) = 3.0 Hz, 1H, H-5, E), 5.37 (ddd, J(4,5) = 9.0, J(5,6’)
15
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
(125 MHz, C6D6)  169.9 (E), 169.5(E), 169.4 (E), 169.3 (Z),
151.7 (Z), 151.3 (E), 150.7 (E), 142.1 (Z), 142.0 (E),137.5 (Z),
137.4 (Z), 136.4 (E), 130.0 (E), 129.7 (Z), 123.4 (Z), 120.1 (Z),
70.3 (E), 69.7 (Z), 69.5 (E), 68.7 (E), 62.0 (E), 61.9 (Z), 21.2 (E),
20.8 (Z), 20.4 (E), 20.3 (Z), 20.2 (E), 20.1 (E), 20.0 (Z), 19.9 (E).
90:10. 1H NMR (500 MHz, C6D6)  7.83 (d, J = 8.5 Hz, 2H, arom,
E), 7.61 (dd, J(1,2) = 13.5, J(1,3) = 1.0 Hz, 1H, H-1, E), 7.28 (dd,
J(1,2) = 13.5, J(1,3) = 1.0 Hz, 1H, H-1, Z), 7.07 (m, 2H, arom,
E), 7.03 (d, J = 7.0 Hz, 1H, arom, E), 6.94 (dd, J(1,2) = 13.5,
J(1,2) = 7.5 Hz, 1H, H-2, E), 6.82 (m, 1H, arom, Z), 6.69 (m, 2H,
arom, Z), 5.94 (t, J = 7.5 Hz, 1H, H-3, E), 5.64 (ddd, J(5,6’) =
7.0, J(5,6) = 5.5, J(4,5) = 3.0 Hz, 1H, H-5, E), 5.60 (dd, J(3,4) =
7.5, J(4,5) = 3.0 Hz, 1H, H-4, E), 5.52 (ddd, J(5,6’) = 7.0, J(5,6)
= 5.5, J(4,5) = 3.0 Hz, 1H, H-5, Z), 5.48 (dd, J(3,4) = 7.5, J(4,5)
= 3.0 Hz, 1H, H-4, Z), 4.33 (dd, J(6,6’) = 11.5, J(5,6) = 5.5 Hz,
1H, H-6, E), 4.28 (dd, J(6,6’) = 11.5, J(5,6) = 5.5 Hz, 1H, H-6, Z),
3.98 (dd, J(6,6’) = 11.5, J(5,6’) = 7.0 Hz, 1H, H-6’, E), 3.90 (dd,
J(6,6’) = 11.5, J(5,6’) = 7.0 Hz, 1H, H-6’, Z), 1.74 (s, 3H, CH3, E),
1.68 (s, 3H, CH3, E), 1.67 (s, 3H, CH3, E), 1.56 (s, 3H, CH3, E),;
13C NMR (125 MHz, C6D6)  169.8 (E), 169.7 (E), 169.4 (E),
169.1 (E), 153.0 (E), 151.7 (E), 142.6 (Z), 138.1 (Z), 136.8 (E),
131.6 (E), 129.3 (E), 126.3 (Z), 123.6 (Z), 123.3 (E), 119.4 (Z),
106.9 (Z), 71.2 (E), 70.9 (Z), 69.2 (E), 69.0 (Z), 68.5 (E), 62.0
(E), 20.3 (E), 20.2 (E), 20.1 (Z), 19.9 (E).
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-chlorophenyl)-1,2-diaza-1,3-butadiene (7a)
and
(1Z,3E)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-(4-
chlorophenyl)-1,2-diaza-1,3-butadiene (7c): E/Z ratio after
equilibration (%) = 91:9. 1H NMR (500 MHz, C6D6)  7.55 (d, J =
6.5 Hz, 2H, arom, E), 7.54 (dd, J(1,2) = 13.5, J(1,3) = 1.5 Hz,
1H, H-1, E), 7.11 (dd, J(1,2) = 13.5, J(1,3) = 1.0 Hz, 1H, H-1, Z),
7.01 (d, J = 6.5 Hz, 2H, arom, E), 6.92 (dd, J(1,2) = 13.5, J(1,3)
= 6.5 Hz, 1H, H-2, Z), 6.90 (m, 2H, Z), 6.87 (dd, J(1,2) = 13.5,
J(1,3) = 5.5 Hz, 1H, H-2, E), 6.42 (m, 2H, arom, Z), 6.04 (ddd,
J(2,3) = 5.5, J(3,4) = 3.0, J(1,3) = 1.5 Hz, 1H, H-3, E), 5.75 (ddd,
J(2,3) = 6.5, J(3,4) = 3.0, J(1,3) = 1.0 Hz, 1H, H-3, Z), 5.62 (dd,
J(4,5) = 9.0, J(3,4) = 3,0, 1H, H-4, E), 5.51 (dd, J(4,5) = 8.5,
J(3,4) = 3.0 Hz 1H, H-4, Z), 5.45 (ddd, J(4,5) = 8.5, J(5,6’) = 5.0,
J(5,6) = 3.0 Hz, 1H, H-5, E), 5.37 (ddd, J(4,5) = 8.5, J(5,6’) =
5.0, J(5,6) = 3.0 Hz, 1H, H-5, Z), 4.26 (dd, J(6,6’) = 12.5, J(5,6) =
3.0 Hz, 1H, H-6, E), 4.21 (dd, J(6,6’) = 12.5, J(5,6) = 3.0 Hz, 1H,
H-6, Z) 4.17 (dd, J(6,6’) = 12.5, J(5,6’) = 5.0 Hz, 1H, H-6’, E),
4.12 (dd, J(6,6’) = 12.5, J(5,6’) = 5.0 Hz, 1H, H-6’, Z), 1.78 (s,
3H, CH3, E), 1.71 (s, 3H, CH3, E), 1.70 (s, 3H, CH3, E), 1.58 (s,
3H, CH3, E); 13C NMR (125 MHz, C6D6)  169.6 (E), 169.1 (E),
169.0 (E), 150.9 (E), 150.0 (E), 141.3 (Z), 138.8 (Z), 137.6 (E),
137.1 (E), 129.2 (E), 129.0 (Z), 124.2 (E), 120.8 (Z), 70.0 (E),
69.2 (E), 68.4 (E), 61.7 (E), 61.5 (Z), 20.1 (E), 20.0 (Z), 19.9 (E),
19.8 (E), 19.6 (E).
Acknowledgments
Financial support from the Junta de Extremadura and Fondo
Europeo de Desarrollo Regional (Grant GR18015 and Project
IB16167) are greatly appreciated. The authors thank both Cenits
and Foundation COMPUTAEX for allowing us the use of
supercomputer LUSITANIA.
Conflicts of interest
The authors declare no conflict of interest.
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-arabino-
tetritol-1-yl)-1-(4-bromophenyl)-1,2-diaza-1,3-butadiene (8a)
Keywords: azoalkene
photoswitching · thermal isomerization
·
carbohydrates
·
chirality
·
and
(1Z,3E)-4-(tetra-O-acetyl-D-arabino-tetritol-1-yl)-1-(4-
bromophenyl)-1,2-diaza-1,3-butadiene (8c): E/Z ratio after
1
equilibration (%) = 91:9. H NMR (500 MHz, C6D6)  7.54 (dd,
J(1,2) = 13.5, J(1,3) = 1.5 Hz, 1H, H-1, E), 7.48 (d, J = 8.5 Hz,
2H, arom, E), 7.18 (d, J = 8.5 Hz, 2H, arom, E), 7.10 (dd, J(1,2)
= 13.5, J(1,3) = 1.5 Hz, 1H, H-1, Z), 7.05 (m, 2H, arom, Z), 6.91
(dd, J(1,2) = 13.5, J(2,3) = 6.0 Hz, 1H, H-2, Z), 6.87 (dd, J(1,2) =
13.5, J(2,3) = 5.5 Hz, 1H, H-2, E), 6.03 (ddd, J(2,3) = 5.5, J(3,4)
= 3.0, J(1,3) = 1.5 Hz, 1H, H-3, E), 5.75 (ddd, J(2,3) = 6.0, J(3,4)
= 3.0, J(1,3) = 1.5 Hz, 1H, H-3, Z), 5.62 (dd, J(4,5) = 9.0, J(3,4)
= 3.0 Hz, 1H, H-4, E), 5.51 (dd, J(4,5) = 8.5, J(3,4) = 3.0, 1H, H-
4, Z), 5.46 (ddd, J(4,5) = 8.5, J(5,6’) = 5.0, J(5,6) = 3.0 Hz, 1H,
H-5, E), 5.36 (ddd, J(4,5) = 8.5, J(5,6’) = 5.0, J(5,6) = 3.0 Hz,
1H, H-5, Z), 4.27 (dd, J(6,6’) = 12.5, J(5,6) = 3.0 Hz, 1H, H-6, E),
4.21 (dd, J(6,6’) = 12.5, J(5,6) = 3.0 Hz, 1H, H-6, Z), 4.17 (dd,
J(6,6’) = 12.5, J(5,6’) = 5.0 Hz, 1H, H-6’, E), 4.11 (dd, J(6,6’) =
12.5, J(5,6’) = 5.0 Hz, 1H, H-6’, Z), 1.77 (s, 3H, CH3, E), 1.71 (s,
3H, CH3, E), 1.70 (s, 3H, CH3, E), 1.57 (s, 3H, CH3, E),; 13C
NMR (125 MHz, C6D6)  169.8 (E), 169.4 (E), 169.3 (E) 151.6
(E), 150.3 (E), 141.6 (Z), 139.2 (Z), 138.0 (E), 132.5 (E), 132.3
(Z), 126.0 (Z), 124.7 (E), 121.3 (Z), 70.3 (E), 69.5 (Z), 69.5 (E),
68.7 (E), 62.0 (E), 61.9 (Z), 20.4 (E), 20.2 (E), 20.1 (E), 19.9 (E).
[1] a) S. Erbas-Cakmak, D. A. Leigh, C. T. McTernan, A. L. Nussbaumer,
Chem. Rev. 2015, 115, 10081-10206. b) J. D. Harris, M. J. Moran, I.
Aprahamian, Proc. Natl. Acad. Sci. USA 2018, 115, 9414-9422.
[2] a) W.-C. Xu S. Sun, S. Wu, Angew. Chem. Int. Ed. 2019, 58, 9712-9740;
Angew. Chem. 2019, 131, 9814-9843. b) A. H. Heindl, J. Becker, H. A.
Wegner, Chem. Sci. 2019, 10, 7418-7425.
[3] For recent examples: a) A. Chevalier, P. Y. Renard, A. Romieu, Chem.
Asian J. 2017, 12, 2008-2028. b) M. Dong, A. Babalhavaeji, C. V. Collins, K.
Jarrah, O. Sadovski, Q. Dai, G. A. Woolley, J. Am. Chem. Soc. 2017, 139,
13483-13486. c) C.-Y. Lai, G. Raj, I. Liepuoniute, M. Chiesa, P. Naumov,
Cryst. Growth Des. 2017, 17, 3306-3312. d) A. Adam, G. Haberhauer, J. Am.
Chem. Soc. 2017, 139, 9708-9713. e) P. S. Addy, S. B. Erickson, J. S. Italia,
A. Chatterjee, J. Am. Chem. Soc. 2017, 139, 11670-11673. f) Y. Ma, K.
Promthaveepong, N. Li, ACS Sensors 2017, 2, 135-139. g) M. Poutanen, Z.
Ahmed, L. Rautkari, O. Ikkala, A. Priimagi, ACS Macro Lett. 2018, 7, 381-386.
h) W.-C. Lin, R. H. Kramer, Bioconjugate Chem. 2018, 29, 861-869. i) T. Lutz,
T. Wein, G. Höfner, J. Pabel, M. Eder, J. Dine, K. T. Wanner, J. Med. Chem.
2018, 61, 6211-6235.
[4] R. Dorel, B. L. Feringa, Chem. Commun. 2019, 55, 6477-6486.
[5] L.-W. Qi, J.-H. Mao, J. Zhang, B. Tan, Nat. Chem. 2018, 10, 58-64.
[6] J. Morstein, M. Awale, J.-L. Reymond, D. Trauner, ACS Cent. Sci. 2019, 5,
607-618.
Mixture of Compounds (1E,3E)-4-(tetra-O-acetyl-D-lyxo-
tetritol-1-yl)-1-phenyl-1,2-diaza-1,3-butadiene
(9a)
and
(1Z,3E)-4-(tetra-O-acetyl-D-lyxo-tetritol-1-yl)-1-phenyl-1,2-
diaza-1,3-butadiene (9c): E/Z ratio after equilibration (%) =
16
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
[7] a) M. L. Wolfrom, M. Blair, J. Am. Chem. Soc. 1946, 68, 2110-2111. b) L.
Mester, A. Major, J. Am. Chem. Soc. 1955, 77, 4297-4300. c) L. Mester, H. El
Khadem in The Carbohydrates: Chemistry and Biochemistry, Vol. IB (Eds.: W.
Pigman, D. Horton), Academic Press, San Diego, 1980, pp. 929-988, and
references therein.
B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Revision A.1,
Gaussian Inc., Wallingford, CT, 2009.
[23] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378-6396.
[24] P. Deslongchamps, Stereoelectronic Effects in Organic Chemistry,
Pergamon Press, New York, 1985.
[8] a) M. C. Wolfrom, A. Thompson, D. R. Linebak, J. Org. Chem. 1962, 27,
2563-2567. b) M. C. Wolfrom, G. Fraenkel, D. R. Linebak, F. Komitsky, J. Org.
Chem. 1964, 29, 457-461.
[25] For some reviews on azadienes: a) O. A. Attanasi, L. Caglioti, Org. Prep.
Proced. Int. 1986, 18, 301-327. b) O. A. Attanasi, P. Filippone, Synlett 1997,
1128-1140. c) O. A. Attanasi, L. De Crescentini, G. Favi, P. Filippone, F.
Mantellini, F. R. Perrulli, S. Santeusanio, Eur. J. Org. Chem. 2009, 3109-3127.
d) A. Lemos, Molecules 2009, 14, 4098-4119.
[9] a) P. M. Collins, S. Kumar, W. G. Overend, Carbohydr. Res. 1972, 22, 187-
192. b) A. Kraus, H. Simon, Chem. Ber. 1972, 105, 954-968. c) V. Zsoldos, A.
Messmer, I. Pintér, A. Neszmèlyi, Carbohydr. Res. 1978, 62, 105-116. d) L.
Somogyi, Carbohydr. Res. 1988, 175, 183-192. e) O. Han, Y. Shih, L. Lin, H.
Lin, J. Org. Chem. 1988, 53, 2105-2108. f) H. S. El Khadem Jr., A. Crossman,
Carbohydr. Res. 1992, 228, 451-454. g) V. Zsoldos-Mády, I. Pintér, M.
Peredy-Kajtár, A. Perczel, Carbohydr. Res. 2011, 346, 1534-1540.
[10] a) M. Ávalos, R. Babiano, P. Cintas, J. L. Jiménez, J. C. Palacios, J. B.
Sánchez, Tetrahedron: Asymmetry 1995, 6, 945-956. b) The original
numbering of starting sugars D-mannose and D-galactose is maintained in the
resulting azadienes to clarify the discussion and comparative analysis through
the text. The correct nomenclature is given in the Experimental section.
[11] E. Merino, M. Ribagorda, Beilstein J. Org. Chem. 2012, 8, 1071-1090.
[12] a) O. Srinivas, N. Mitra, A. Surolia, N. Jayaraman, Glycobiology 2005, 15,
861-873. B) For a perspective: F. Hamon, F. Djedaini-Pilard, F. Barbot, C.
Len, Tetrahedron 2009, 65, 10105-10123. c) For a recent example on a
glycomacrocycle: J. Hain, G. Despras, Chem. Commun. 2018, 54, 8563-8566.
[13] a) Y. Ogawa, C. Yoshiyama, T. Kitaoka, Langmuir 2012, 28, 4404-4412.
b) R. F. Tabor, D. D. Tan, S. S. Han, S. A. Young, Z. L. E. Seeger, M. J.
Pottage, C. J. Garvey, B. L. Wilkinson, Chem. Eur. J. 2014, 20, 13881-13884.
c) F. Reise, J E. Warias, K. Chatterjee, N. R. Krekiehn, O. Magnussen, B. M.
Murphy, T. K. Lindhorst, Chem. Eur. J. 2018, 24, 17497-17505.
[26] For cycloadditive chemistry using chiral azoalkenes derived from
monosaccharides: a) M. Ávalos, R. Babiano, P. Cintas, F. R. Clemente, J. L.
Jiménez, J. C. Palacios, J. B. Sánchez, J. Org. Chem. 1999, 64, 6297-6305.
b) M. Ávalos, R. Babiano, P. Cintas, F. R. Clemente, R. Gordillo, J. L. Jiménez,
J. C. Palacios, P. Raithby, J. Org. Chem. 2000, 65, 5089-5097. c) M. Ávalos,
R. Babiano, P. Cintas, F. R. Clemente, R. Gordillo, J. L. Jiménez, J. C.
Palacios, J. Org. Chem. 2001, 66, 5139-5145. d) M. Ávalos, R. Babiano, P.
Cintas, F. R. Clemente, R. Gordillo, J. L. Jiménez, J. C. Palacios, J. Org.
Chem. 2002, 67, 2241-2251.
[27] G. A. Doorakian, H. H. Freedman, J. Am. Chem. Soc. 1968, 90, 5310-
5311.
[28] C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, 3rd Ed.,
Wiley-VCH, Weinheim, 2003, pp. 197-198.
[29] a) D. S. Malament, N. Levi, J. Org. Chem. 1975, 40, 3285-3287. b) A. A.
Reid, H. R. Sood, J. T. Sharp, J. Chem. Soc. Perkin Trans 1 1976, 362-366. c)
H. McNab, E. A. Murray, J. Chem. Soc. Perkin Trans 1 1981, 122-123. d) U.
Ziegler, G. Zimmermann, B. Ondruschka, V. V. Volkova, L. E. Gusel’nikov, V. I.
Smirnov, N. S. Nametkin, Tetrahedron Lett. 1986, 27, 5825-5828.
[30] N. P. Belskaia, T. G. Deryabina, A. V. Koksharov, M. I. Kodess, W.
Dehaen, A. T. Lebedev, V. A. Bakulev, Tetrahedron Lett. 2007, 48, 9128-9131.
[31] R. C. Fahey, H. J. Schneider, J. Am. Chem. Soc. 1970, 92, 6885-6893.
[32] R. F. Childs, B. D. Dickie, J. Am. Chem. Soc. 1983, 105, 5041-5046.
[33] F. G. Bordwell, D. Algrim, J. Org. Chem. 1976, 41, 2507-2508.
[34] J. F. J. Dippy, S. R. C. Hughes, A. Rozanski, J. Chem. Soc. 1959, 2492-
2498.
[14] D. R. Armstrong, J. Clarkson, W. E. Smith, J. Phys. Chem. 1995, 99,
17825-17831.
[15] a) J. G. Schantl, Org. Magn. Reson. 1979, 12, 652-654. b) J. G. Schantl,
P. Margaretha, Helv. Chim. Acta 1981, 64, 2492-2494.
[16] J. G. Schantl, P. Hebeisen, Tetrahedron 1990, 46, 395-406.
[17] Crystal data for compound 5a have been deposited with the Cambridge
Crystallographic Data Centre (CCDC-1828935).
[35] F. G. Bordwell, Acc. Chem. Res. 1988, 21, 456-463.
[18] Crystal data for compound 6a[10] have been reported in a separate study:
M. J. Diánez, M. D. Estrada, S. Pérez-Garrido, Acta Cryst. 2006, E62, 2811-
2812.
[36] I. M. Kolthoff, P. J. Elving, Treatise on Analytical Chemistry, Interscience
Encyclopedia, Inc., New York, 1959.
[37] C. Dugave, L. Demange, Chem. Rev. 2003, 103, 2475-2532.
[38] Y. Zhao, D. G. Truhlar, Acc. Chem. Res. 2008, 41, 157-167.
[39] A. V. Marenich, C. J. Cramer, D. J. Truhlar, J. Phys. Chem. B 2009, 113,
4538-4543.
[19] C. Jørgensen, C. Pedersen, Carbohydr. Res. 1997, 299, 307-310.
[20] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215-241.
[21] a) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639-5648. b)
K. Raghavachari, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980,
72, 650-654.
[22] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J.
R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G.
Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J.
Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.
A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E.
Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K.
Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N.
Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J.
17
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000029
European Journal of Organic Chemistry
FULL PAPER
Entry for the Table of Contents
Insert graphic for Table of Contents here.
Magic Azo: Incorporation of the photolabile azo group into an unsaturated sugar chain affords a class of glycomimetics susceptible
of thermal and photo-induced switches under mild conditions (sunlight irradiation). Here such stereomutations and mechanistic routes
are disentangled by experiment and DFT calculations.
18
This article is protected by copyright. All rights reserved.