2-(Hydroxyimino)aldehydes: Photo- and Physicochemical Properties of a Versatile Functional Group for Monomer Design

Article abstract of DOI:10.1002/chem.201800059

In the context of our research on stimuli-responsive polymers bearing the 2-(hydroxyimino)aldehyde (HIA) group, we have explored the photochemical behavior and physicochemical properties of a number of HIAs. Interpretation of the experimental data is supp

Full text of DOI:10.1002/chem.201800059

A Journal of
Accepted Article
Title: 2-(Hydroxyimino)aldehydes: Photochemical and Physico-
Chemical Properties of a versatile functional group for monomer
design
Authors: Patrizia Gentili, Martina Nardi, Irene Antignano, Paolo
Cambise, Marco D'Abramo, Francesca D'Acunzo, Alessandro
Pinna, and Emanuele Ussia
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: Chem. Eur. J. 10.1002/chem.201800059
Link to VoR: http://dx.doi.org/10.1002/chem.201800059
Supported by

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
2-(Hydroxyimino)aldehydes: Photochemical and Physico-Chemical
Properties of a versatile functional group for monomer design
Patrizia Gentili*[ab], Martina Nardi[ab], Irene Antignano[a], Paolo Cambise[a], Marco D’Abramo[a], Francesca D'Acunzo*[b],
Alessandro Pinna[a], Emanuele Ussia[a]
1
2
3
4
5
6
7
8
9
Abstract: In the context of our research on stimuli-responsive polymers bearing the 2-(hydroxyimino)aldehyde (HIA) group, we explore
the photochemical behavior and physico-chemical properties of a number of HIAs. Interpretation of experimental data is supported by
quantum mechanical calculations. HIAs are expected to undergo photoisomerization, chelate metal ions, yield hydrogen-bonded dimers
or oligomers, exhibit relatively low pKas, and form >C=NO• radicals through OH hydrogen abstraction or oxidation of the oximate ion.
Besides the well-established E/Z oxime photoisomerism, we observe a Norrish-Yang cyclization resulting in cyclobutanol oximes, to our
knowledge not described in previous literature. Acidity, Bond Dissociation Enthalpies and electrochemical properties of HIAs are
compared to literature data on simple oximes. Results are discussed in relation to the many potential applications for HIAs, with
emphasis on the synthesis of novel HIA-containing responsive polymers.
10
Introduction
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
2-(hydroxyimino)aldehydes (HIAs; Figure 1a) are a class of organic
R= linear, cyclic or branched
compounds that has been sporadically investigated in the scientific
alkyl, up to C10; aryl;
R
CHO
methacrylic
literature, mainly in the context of organic synthesis. Convenient
methods for the synthesis of 2-(hydroxyimino)carbonyl compounds
are being continuously proposed: Sugamoto[2] resorted to reduction-
nitrosation with t-butyl nitrite and triethylsilane in the presence of
cobalt(II) porphyrin as a catalyst; 3-chloro-2-hydroxyiminopropanal
(Figure 1: R=CH2Cl) was prepared by Gilchrist and Roberts by
addition of nitrosyl chloride to acrylaldehyde at low temperature;[3]
Baidya and Yamamoto[4] obtained several HIAs through oxidation of
the corresponding aldehydes with aluminum nitrite. Finally, a
convenient -oximation reaction of aldehydes has been developed
in our laboratory,[5] which affords good to excellent yields of HIAs via
organo-SOMO catalysis. This reaction affords HIAs bearing
substituents of different bulk and nature (Figure 1), so that a
systematic investigation of their physico-chemical properties is now
accessible. Oximes in general are highly versatile molecules and
HIAs may exhibit yet more interesting properties, thanks to the
presence of an oxime and aldehyde group in adjacent position.
Oximes are building blocks for the synthesis of heterocyclic
compounds,[6] they can exhibit biological activity,[7] and they can be
used as handles for bioorthogonal conjugation[8] through the
formation of oxime ether bonds.[9] In this context, HIAs, which carry
a combination of electrophilic (CHO and C=NOH) and nucleophilic
sites (oxime N and O) can be viewed as multi-functional platforms
for bioconjugation. Oximate ions are of interest for the
C
N
pKa= 7.7-9.9
< 0.7 V to oxidize
conjugated base to NO.
OH
E/Z photoisomerization
hydrogen-bonded closed or linear oligomers; metal ligation
a) 2-(hydroxyimino)aldehydes (HIAs)
H C
3
H C
3
m
n
O
O
O
O
O
O
HO
7-8
O
N
H
CH
3
b) poly(HIAMBA-co-OEGMA475
)
Figure 1. a) General structure and properties of 2-(hydroxyimino)aldehydes
(HIAs); b) Structure of poly(HIAMBA-co-OEGMA475), an HIA-containing
copolymer from our recent investigation.[1] HIABMA= 4-
[(hydroxyimino)aldehyde]butyl methacrylate
decontamination from toxic metal ions[10] or from organophosphorous compounds.[11] Oximes are known to exist in two configurations,
38
39
40
41
where the residue with the highest priority is either E or Z with
respect to the oxime OH group. The photo-induced switch of the
E/Z configuration in imines, oximes and hydrazones has been
extensively investigated in the field of light-driven molecular
machines (see Ref[12] and references therein). In the case of HIAs,
the aldehyde group has the highest priority, so that E/Z isomerism
is referred to the relative positions of the oxime OH and aldehyde
groups about the oxime double bond (Figure 1a). The interaction
[a]
[b]
Prof. P Gentili, Dr. M Nardi, I Antignano, Dr. P Cambise, Prof. M
D’Abramo, Dr. A Pinna, Dr. E Ussia
Dipartimento di Chimica, Università degli Studi “La Sapienza” P.le A.
Moro 5, 00185 Rome, Italy
42
43
44
45
E-mail: patrizia.gentili@uniroma1.it
Prof. P Gentili, Dr. M Nardi,, Dr. F D’Acunzo
CNR, Istituto di Metodologie Chimiche, Sezione Meccanismi di
Reazione c/o Dipartimento di Chimica,Università degli Studi “La
Sapienza” P.le A. Moro 5, 00185 Roma, Italy.
E-mail: francesca.dacunzo@uniroma1.it
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
1
2
between the adjacent aldehyde and oxime groups in HIAs, mainly in terms of hydrogen bonding and conjugation of the oximate anion,
is likely to be different in the two configurations, thus resulting in distinct physico-chemical properties. Such interaction, which is also
influenced by the nature and bulk of the R group, has been studied in the case of other 2-substituted oximes, such as -keto-oximes,[13]
monoterpenoid oximes,[14] pyrrole-2-carbaldehyde oxime,[15] nitrosation products of phosphorylaldehydes[16] and -cyanoximes.[17] HIAs
are expected to possess a lower pKa than the corresponding unsubstituted oximes, so that it is of paramount importance to determine
their acidity in view of possible applications where oximate ions are involved. Knowledge of the redox properties and UV-Vis absorption
of HIAs is also relevant towards other technological applications, as in Dye-Sensitized Solar Cells, in which they may serve either as
sensitizing dyes, or as iodine-free shuttle systems.[18] In this context, the configurational behavior of HIAs upon exposure to light is likely
to be coupled with the physico-chemical properties of these molecules and of their metal complexes. In fact, the ability of oximes to
form dimers, trimers, tetramers or catamers has been associated to the combination of E/Z configuration on the C=N double bond, bulk
of the R group, and preferred conformations.[19] Metal coordination complexes and polymers have been obtained by using oximes as
ligands.[20] 2-D coordination polymers have been obtained using 3-hydroxyiminomethyl salicylic acid;[21] Croitor et al.[22] obtained a
variety of coordination networks through a mixed ligand approach, employing oximes with different coordination capacities and
preferences. Gerasimchuk and co-workers have been exploring the potential of 2-cyanoximes as ligands in coordination polymers for a
variety of applications.[23] Cai et al.[24] incorporated the oxime moiety as side group in oligo(ethylene glygol) monomethyl ether
methacrylate (OEGMA) random copolymers, thus obtaining water-soluble polymethacrylates that switch their pre-organization and
thermoresponsive behavior via light-tuned hydrogen bonding. Their work, which is possibly the first step towards exploiting the
versatility of oxime residues in multi-functional polymers, recently stimulated us to copolymerize an HIA-containing methacrylate with
OEGMA to obtain multi stimuli-responsive copolymers (Figure 1b).[1] Our first results elicit further insight into the dependence of some
physico-chemical properties of HIAs on their structure. Specifically, the oxime Bond Dissociation Enthalpies (BDEs) and the structure
and stability of the resulting >C=NO• radical may relate to reactions interfering with radical polymerizations, such as addition to the C=N
bond or H abstraction. We also observed that the E/Z switch at photostationary state,[25] is very limited in the case of the polymer in
Figure 1b and we surmise that it may be enhanced by inserting a different spacer between the HIA group and the polymer backbone.
The rate of thermal reversal to the pre-photostimulation E/Z ratio and the occurrence of further photoinduced reactions may be also
structure-dependent. Aiming at a rational design of an HIA-functionalized methacrylate for the synthesis of multi-stimuli responsive
polymers, in the present investigation we explore the photochemical and physico-chemical properties of HIAs bearing substituents of
different size and nature (Scheme 1). We focused on the R-group effect on oxime acidity and redox potential, UV-Vis spectra in solution,
photochemical behavior with subsequent thermal relaxation, and Bond Dissociation Enthalpies.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
pyrrolidine (0.2 equiv)
p-TsOH (0.2 equiv)
R
CHO
NOH
RCH2CHO
NaNO (1.0 equiv)
FeCl2 (1.0 equiv)
DMF, RT, 4.5h
R
R
1: CH3(CH2)7-
9: CH3(CH2)7-, 59%
2: [CH3(CH2)9][CH3(CH2)7]CH- 10: [CH3(CH2)9][CH3(CH2)7]CH-, 60%
3: (CH3)3C-
11: (CH3)3C-, 65%
4: CH=C(CH3)CO2(CH2)4-
12: CH=C(CH3)CO2(CH2)4-, 60%
5: Cyclohexyl
6: Ph-
13: Cyclohexyl, 93%
14: Ph-, 90%
7: PhCH-
8: Np-
15: PhCH-, 67%
16: Np-, 45%
Scheme 1. Synthesis of 2-(hydroxyimino)aldehydes
29
Results and Discussion
30
31
32
33
34
35
36
37
38
39
40
Synthesis of HIAs
All HIAs were obtained by -oximation of the precursor aldehydes according to a procedure developed in our laboratory (Scheme 1).
The -oximation reactions were performed at room temperature in dimethylformamide (DMF), with NaNO2 and FeCl3, in the presence
of pyrrolidine as organocatalyst, resulting in satisfactory (45%) to excellent yields (90%).[5] Compounds 9, 11, 13, 14 and 15 were
obtained from their commercially available precursor aldehydes as already described in a previous report.[5] HIA 12 was obtained in
good yields via a multi-step synthesis as we recently reported (Scheme 2).[1] Briefly, 1,6-hexanediol was monoacylated with methacrylic
anhydride, and the resulting 6-hydroxyhexyl methacrylate 17 was oxidized to the corresponding aldehyde 4 through Parikh and Doering
oxidation. -Oximation of 4 gave, finally, 60% of product 12 after purification (Scheme 1). 3-Octyl-2-(hydroxyimino)tridecanal (10) and
2-(hydroxyimino)(naphthalene-2-yl)ethanal (16) and their precursor aldehydes 2 and 8 were first synthesized in this investigation.
Scheme 3 outlines the synthesis of 2 and 8. Specifically, 3-octyltridecanal 2 was prepared by Parikh and Doering oxidation[26] of 2-
octyldodecanol 18 to aldehyde 19 in very good yield (90%). Wittig reaction of 2-octyldodecanal 19 with phosphonium ylide
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
1
2
CH3OCHPPh3 to enol ether 20 (yield 69%) and hydrolysis of this intermediate gave the 3-octyltridecanal 2 in excellent yield (94%).
O
O
O
O
O
a)
HO
+
+
OH
O
O
O
O
OH
17, 59%
a) Et3N (0.5 equiv), DMAP (0.1 equiv), THF, 0°C 1h, RT 2h
b) SO3.Pyr (3 equiv), Et3N (4 equiv), DMSO, CH2Cl2, RT 16h
b)
O
O
O
4, 98%
Scheme 2. Synthesis of HIABMA
4
(4-[(hydroxyimino)aldehyde]butyl methacrylate)[1]
c)
a) or b)
OCH3
RCH 2OH
RCHO
RCH CH
20: R = [CH3(CH2)9][CH3(CH2)7]CH-, 69%
23: R = Np-
19: R = [CH3(CH2)9][CH3(CH2)7]CH-, 90%
22: R = Np-, 72%
18: R = [CH3(CH2)9][CH3(CH2)7]CH-
21: R = Np-
d) or e)
R = [CH3(CH2)9][CH3(CH2)7]CH-:
a) SO3.Py (3 equiv), Et3N (4 equiv), DMSO, CH2Cl2, RT, 16 h
c) [CH OCH PPh ]+Cl- (3 equiv)
, t
-BuOK (3 equiv), THF, RT, 17 h
3
2
3
d) THF, HCl (2 equiv.), 75° C, 2h
RCH2CHO
R = Np-:
2: R = [CH3(CH2)9][CH3(CH2)7]CH-, 94%
8: R = Np-, 90%
b) TEMPO (0.1 equiv), BAIB (1.1 equiv), CH2Cl2, RT, 4 h
c) [CH OCH PPh ]+Cl- (3 equiv)
, t
-BuOK (3 equiv), THF, RT, 24 h
3
2
3
e) i) Cyclohexane, 80 °C; ii) (COCl)2 (7 equiv), RT, 1 h
Scheme 3. Synthesis of 3-octyltridecanal (2) and 2-(naphthalen-1-yl)acetaldeyde (8)
3
4
Finally, compound 10 was obtained in fair yield (60%) by -oximation of 2 (Scheme 1). As shown in Scheme 3, the oxidation of 1-
naphthylmethanol 21 with 2,2,6,6-tetramethylpiperidine-1-oxyl radical (TEMPO) in the presence of the stoichiometric oxidant
[bis(acetoxy)iodo]benzene (BAIB),[27] gave naphthalene-1-carbaldehyde 22 in good yield (72%). The next step was the Wittig reaction
of aldehyde 22 with phosphonium ylide CH3OCHPPh3 to obtain the enol ether 23.[28]
5
6
7
The crude product was extracted with cyclohexane, then thorough removal of phosphine oxide by-product was achieved by treatment
with oxalyl chloride to obtain a mixture of compound 23 and aldehyde 8.[28] Hydrolysis of this mixture in THF with concentrated
8
9
hydrogen chloride[29] gave 1-naphthylacetaldehyde 8 in quantitative yield (90% vs aldehyde 22) and high purity (GC: 95%). The final
oximation reaction of 8 gave a reasonable (45%) yield of product 16 after purification.
-
10
11
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
Z
E
H
H
syn
anti
O
O
N
N
OH
HO
HO
O
O
H
H
N
N
OH
Figure 2. Configurational (E/Z) and conformational (syn/anti) isomers
of 2-(hydroximino)aldehydes
1
2
Photochemical behavior of HIAs
Due to the presence of the C=N double bond, different configurational and geometrical isomers can be envisaged for HIAs (Figure 2).
Oximes exhibit a complex photochemistry involving isomerizations, rearrangements, dehydration to nitriles or loss of the hydroxylamine
function to form the parent carbonyl. The reader should refer to Roth’s comprehensive review[30] to appreciate the richness of oximes
photochemistry. Since it was first observed in 1890, E/Z photo- and thermal isomerism has been constantly an object of interest, with
the first NMR studies in the late 1950s, all the way to recent computational studies on isomers interconversion.[31] Concerning the
mechanism for E/Z isomerization, the transition state may be either linear, with the nitrogen atom turning its hybridization from sp2 to sp,
or a state where the imine double bond character is weakened, thus lowering the barrier for rotation about the C=N(OH) bond. Blanco
et al.[32] attribute the former to ground state interconversion and the latter to photoinduced isomerization. From literature inspection,[30] it
is quite clear that the propensity for photoisomerization, the relative stability of the configurational isomers and the complexity of the
resulting mixtures are peculiar to the specific substituent adjacent to the oxime group. -Oxo-oxime ethers were extensively
investigated by Cerfontain and co-workers (see Ref.[33] and references therein), who showed that, in addition to E/Z interconversion,
photolysis and/or a Norrish type II reaction can take place upon photoexcitation. The photoreactivity of -oxo-oxime esters has been
known for over a century and is still being investigated for the photoinitiation of acrylic monomers polymerizations.[34] In the present
work, we carried out an NMR-based investigation on photo- and thermal isomerism of several HIAs. The investigated compounds differ
by the bulk of the aliphatic R group (9: R= CH3(CH2)7-; 10: R= [CH3(CH2)9][CH3(CH2)7]CH-); 11: R= (CH3)3CH-) or by the type of
substituent (12: R= CH2=C(CH3)CO2(CH2)4-; 14: R= Ph). Compound 12 bears a remote methacrylate group, which is of interest for the
synthesis of polymeric materials. In order to investigate the photoisomerization of HIAs, we carried out a preliminary study on their UV-
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
1
visible absorption in several solvents. To follow the N-OH and CHO signals of the irradiated samples by H-NMR analysis, we chose
DMSO-d6 (UV cutoff: 265 nm) for our photostimulation experiments. HIA solutions in DMSO-d6 were then irradiated in a high pressure
mercury lamp with an emission maximum at 350 nm. In a first set of experiments, the samples were photoexcited for a relatively short
time, i.e. for up to 2h. No 1H-NMR-detectable difference was observed between 1h and 2h irradiation, so we assume that a
photostationary state had been reached. As detailed below, E/Z isomerization was observed within this timeframe. The samples were
then allowed to thermally equilibrate at 40°C in the dark and were monitored by NMR over time. In a second set of experiments,
photoexcitation was continued overnight (22h). In this case, significant, if not quantitative, formation of a single new product was
observed. Based on the evidence presented in this paper and on the behavior of -oxo-oximes documented by Buys et al.,[35] we
identified this product as a cyclobutanol oxime (CBO) derivative (Scheme 4) from the Norrish-Yang cyclization.[36]
R
R
R
1
1
1
R
R
1
R
2
H
O
R
2
H
O
C
R
2
C
OH
C
R
R
hn
2
3
OH
R
R
R
H
H
H
R
R
R
3
3
N
3
N
N
N
OH
OH
OH
OH
9: R= H, R1= CH3(CH2)5-, R2= R3= H
10: R= H, R1= CH3(CH2)8-, R2= CH3(CH2)7-, R3= H
R= H, R1= CH3(CH2)6-, R2= CH3(CH2)9-, R3= H
11: R= R1= H, R2= R3= (CH3)-
Scheme 4. Norrish-Yang cyclization of some HIAs.
28
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
1
2
UV-vis spectroscopy of HIAs
With the exception of the white or off-white solid HIAs 11, 14 and 10, all synthesized HIAs are colorless oils. The UV spectra at 0.1-0.5
mM concentration in several solvents are included in the Supporting Information (Table S1). A protic (i-PrOH) and three aprotic solvents
of different polarity (CHCl3, CH3CN, DMF) were selected so as to ensure solubility of most of the HIAs under investigation and to
highlight any band shifts caused by solvent effects. Consistently with literature on oximes and -oxo-oximes,[30] all non-conjugated
compounds show an allowed band at 230-240 nm ( ca. 10000 M-1cm-1) ascribable to the → * transition. For HIAs 14 and 16, having
an aromatic ring as substituent on C(=NOH)CHO group, the same absorption band is shifted to ca. 270 nm. Spectra of concentrated
(up to 1.2 mM) DMSO solutions of 10 and 12 were also recorded to check for the presence of a band ascribable to the forbidden n→ *
transition.[36] Indeed, both 10 and 12 exhibit a weak band ( 100 M-1cm-1) at 320-334 nm band, similarly to -oxo-oximes.[30] Following
deprotonation with (CH3)4NOH as the base, a new peak appears at 290-300 nm due to the → * absorption bands of HIA anions,
which delocalize the negative charge as depicted in Figure 3.[37] In contrast to amide-[38] and aryl-cyanoximates,[37a, 37b] the position of
these bands in the spectra does not show a significance solvent dependence, the absorption maxima being in all cases around 290 nm.
3
4
5
6
7
8
9
10
11
12
-
N O
N
O
O
R
R
-
O
H
H
Figure 3. Negative charge delocalization of 2-(hydroxyimino)aldehyde
anions
13
E/Z photo- and thermal isomerism
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
1H-NMR signals of C=NOH and CHO are diagnostic for E/Z isomerism of HIAs. In Figure 4, the expansions of spectra encompassing all
aldehyde and oxime signals are shown with their intensities. H-NMR spectra of as-synthesized 9 and 12, i.e. of the two compounds
1
that bear a primary carbon next to the oxime group, only exhibit, respectively, one CHO peak (1H) at 9.39 and 9.40 ppm and one NOH
peak (1H) at 12.89 and 12.92 ppm. After 2h light exposure of 9 or 12 in the photoreactor, two small additional peaks appear: 10.28 ppm
(CHO) and about 12.00 ppm (NOH). The chemical shifts of the aldehyde hydrogens computed by DFT methods for the different
isomers (Figure 2) of selected HIAs are reported in Table 1. The effect of DMSO was modeled using a continuum approach, i.e. by
means of the Polarizable Continuum Model (PCM).[39] The computed chemical shift values for the (Z)-syn isomers are at most 0.4 ppm
upfield of (E) isomers. Hence, it is not possible to assign any CHO aldehyde signals falling in the 9.1-9.7 ppm region of the spectrum to
either configuration based on chemical shift calculations alone. However, the downfield NOH signal of unhindered -oxo-oximes was
assigned by Baas and Cerfontain [37c] to the E-anti isomer, the most stable due to dipolar interactions. On these grounds, we assume
that signals around 12.9 ppm of unhindered HIAs 9 and 10 should be assigned to NOH in the E form. Hence, the signals around 9.4
ppm are attributed to the CHO of the E isomers. The calculated CHO chemical shifts for the (E)-syn/anti isomers (Table 1) are in good
agreement with the experimental signals (Figure 4). However, since rapid interconversion (on the NMR timescale) of E conformers
cannot be ruled out with such little hindrance to rotation of the CHO group, we are not assigning the (E)-9 and (E)-10 signals to either
conformation. The calculated CHO shift for the (Z)-anti isomer (10.4 ppm) lies very close to one of the two signals (b’: 10.28 ppm)
appearing after 2h irradiation. Since the intensity of the (Z)-NOH (12.00 ppm) signal matches that of (Z)-anti CHO, we surmise that only
the (Z)-anti isomer is formed upon photostimulation. The pristine secondary compound 10 is also in the E form only (CHO peak at 9.34
ppm and NOH peak at 12.83 ppm). Upon photoexcitation, two small signals appear at 12.04 and 10.32 ppm, which we assign to the Z
form in analogy with the primary compounds. The assignment of a further significantly intense signal at 10.15 ppm will be discussed
later. As for tertiary 11, similar arguments as for 9, 10 and 12 hold for the assignment of NMR signals to the E and Z isomers. However,
in contrast with these compounds, the Z isomer (NOH and CHO signals of equal intensity at 11.78 and 10.24 ppm, respectively) is
predominant over E (12.89 and 9.28 ppm of same intensity) prior to irradiation. In fact, (E)-11 may be destabilized by overcrowding of
the t-Bu group and N-OH group on the same side of the C=N bond. Photostimulation induces a complete switch in the configuration
ratio and the appearance of a further signal at 10.14 ppm, which will be discussed later. In the case of compound 14, the most relevant
changes upon irradiation are detected in the aromatic region, whereas evidence for E/Z isomerization is too weak for any interpretation.
Quantitative results for E/Z photoisomerization and subsequent thermal relaxation are summarized in Table 2. When HIAs bearing a
primary R group (compounds 9 and 12) are irradiated for two hours, limited inversion of configuration is observed (E/Z= 10/1). Thermal
relaxation to E-isomer is quantitative in 9 within 8 days, while with HIA 12 the E/Z ratio is essentially unchanged after about three weeks.
The remote methacrylate group, therefore, appears to slow down the thermal isomerization process. Also in the case of secondary 12
the E isomer is still predominant after 2h irradiation (E/Z~ 5/1) and thermal reversal to the E form is not completed after one week (E/Z~
7/1). The most remarkable effect of 2h photostimulation is that observed with compound 11. In fact, E/Z ratio switched from 1/10 to 5/1
and thermal relaxation was still incomplete after 40 days (E/Z= 1/2).
46
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
I= 0.6
I= 0.8
I =0.2
I=0.7
I= 0.7
I= 0.9
I=0.7
I= 0.7
I< 0.1 0.1
I< 0.1
I =0.4
I= 0.9
I< 0.1
I< 0.1
I= 0.3
I< 0.1
I= 0.5
I= 0.1
I= 0.8
I= 0.1
I=0.2
I= 0.5
I= 0.9
I= 0.5
I= 1.0
I< 0.1
I< 0.1
0.1
I= 0.8
I= 1.0
I= 0.7
I= 0.1
I= 0.9
I< 0.1
I= 0.9
I= 1.0
I= 0.1
I= 0.1
I< 0.1
I= 0.8
I= 0.9
I= 0.1
I= 1.0
I= 1.0
Figure 4. Expansion of the 1H-NMR spectra (DMSO-d6, : ppm) for the study of E/Z isomerism of compounds
2h photostimulation ( = 350 nm) followed by annealing at 40°C. Intensities are normalized to terminal –CH3 (9
9
,
,
10
,
11, 12 and 14 after
11) or -OCH2- (12).
10,
Broken lines are a pictorial of the calculated chemical shifts (Table 1) for different geometrical and conformational isomers of HIAs. Full
spectra in the Supporting Information.
1
Table 1. 1H-NMR computed chemical shifts[a] for the CHO of the different
configurational and conformational isomers of HIAs.
(Z)-anti (Z)-syn (E)-anti (E)-syn
RC(NOH)CHO
( )
( )
9.1
9.3
9.2
( )
9.1
9.2
9.2
( )
9.3
9.7
9.6
(9) CH3(CH2)7-
10.4
10.4
10.4
(11) (CH3)3C-
(12) CH2=C(CH3)CO2(CH2)4
-
[a]Geometry optimization in vacuum at the B3LYP/6-31+G(d, p) level;
Chemical shift calculations with WPO4/aug-cc-pVdZ; effect of the solvent
modeled by means of the PCM. : ppm.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
Table
2
E/Z photoisomerization[a] and thermal re-equilibration[b] of
selected HIAs.
RC(NOH)CHO
(9) CH3(CH2)7-
Time
t=0
E/Z
E only
2h, 350 nm[a]
8 days, 40°C[b]
10/1
E only
(10) [CH3(CH2)9][CH3(CH2)7]CH-
t=0
E only
~5/1
2h, 350 nm[a]
7 days, 40°C [b]
~7/1
(11) (CH3)3C-
t=0
1/10
5/1
2h, 350 nm[a]
40 days, 40°C [b]
1/2
(12) CH2=C(CH3)CO2(CH2)4
-
t=0
E only
10/1
2h, 350 nm[a]
23 days, 40°C [b]
10/1
[a] Irradiation ( = 350 nm) in DMSO-d6 followed by 1H-NMR analysis.
[b] After irradiation, the samples were stored in the dark at 40°C and
analyzed by 1H-NMR at suitable time intervals.
1
2
3
Photoisomerization to cyclobutanol oxime (CBO)
As anticipated in the previous paragraph, compounds 10 and 11 exhibit additional 1H-NMR peaks (10.15 and 10.14 ppm, respectively)
besides those ascribed to E/Z isomerization, already within 2h of photostimulation. Indeed, halving of the E-isomer CHO and NOH peak
intensities to the benefit of one single signal at 10.15 ppm is the most prominent effect of the 2h irradiation of compound 10 (Figure 4).
In order to better observe any photoreactions that may occur over longer timescales than E/Z interconversion, photostimulation of
compounds 9, 11 and 10 was continued overnight (22h). Some features can be observed in the 1H-NMR spectra of the resulting
solutions (Supporting Information S14-S22; Figure 5):
4
5
6
7
8
9
i) (Z)-anti signals are negligible; (E)-isomers signals are still present;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
ii) The main photoproduct formed exhibits a singlet at 10.1-10.2 ppm (1H) and a set of signals in the 5.5 (1H) and 4-5 ppm (1H) regions
of the spectra. D2O exchange shows that protons at 5.5 ppm exchange rapidly and couple with the 4-5 ppm protons (see Supporting
Information S15);
iii) A signal at 8.13 ppm peak is observed independently of the HIA substituent. This signal can be attributed to formic acid, i.e. the by-
product formed by decomposition of HIAs to nitriles.[30]
1
In order to identify the product originating the new set (see feature ii above) of H-NMR signals, we focused on 10, which shows the
least amount of residual HIA and of minor decomposition products. Full H- and 13C- NMR spectra of 10 after 22h irradiation, shown in
1
Figure 5, are compatible with the formation of a cyclobutanol oxime (CBO) from Norrish-Yang cyclization,[36] as depicted in Scheme 4.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
Figure 5. 1H-NMR and 13C-NMR (DMSO-d6) spectra of 10 after 22h
photostimulation (λ= 350 nm) followed by 4h at room temperature. (ppm):
12.8 (HIA, C=NOH, 0.1); 10.2 (He, 1.2); 9.34 (HIA, CHO, 0.1); 5.45 and 5.35
(Hd, 0.6 and 0.3); 4.15 and 4.02 (Ha, 0.3 and 0.7); 2.27 and 2.12 (Hc, 0.7 and
0.3); 1.65 (Hb, 1.0).
1
2
Unequivocal evidence is provided by D2O exchange and 2D NMR experiments (Supporting Information S15-S20). Specifically, the 1H-
NMR spectrum is simplified upon the addition of D2O due to the disappearance of two exchangeable protons (10.15 ppm and 5.45 ppm).
These signals can be attributed, respectively, to the C=NOH and CH-OH protons of the CBO structure. Further NMR experiments
(COSY, DEPT 135, NOESY, HSQC; Supporting Information S15-S20) contribute to the assignment of the CH ring signals (reported in
Figure 5), whereas the HMBC experiment clearly confirms the cyclic structure. In fact, the oxime carbon (159.8 ppm) shows long-range
3
4
5
2
6
correlation with C-Ha and C-Hc , while C-Hb has JCH with the alcoholic carbon (74.3 ppm). It is worth noting that the formation of the
7
cyclobutanol ring (Scheme 4) requires the HIA to be in the anti conformation, regardless of geometrical isomerism.[40] Upon thermal
annealing, occurring even as the NMR spectra are acquired at room temperature, E/Z equilibration of the oxime group in the CBO
structure takes place, as shown by the more complex NMR spectrum (Figure 5) with different CH-OH signals and splitting of its oxime
signal (Supporting Information S15, inset). When compound 9 is irradiated for 22 h (Supporting Information S14), the HIA oxime and
aldehyde (E isomer, 12.89 and 9.39 ppm) signals are only partially reduced and signals at 10.3 ppm, 5.64 and 5.46 ppm, 4.21 and 4.72
ppm are present, which are consistent with the formation of both E and Z isomers of CBO. 22h irradiation of compound 11 also yields
the CBO structure (Supporting Information S21; 10.15, 5.50, 4.61 ppm) along with residual HIA. However, a number of unidentified
minor signals are also detected. Table 3 shows the quantitative outcome of E/Z and Norrish-Yang photoisomerization of 9, 10 and 11.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
Table 3 E/Z vs. Norrish-Yang photoisomerization of selected HIAs.[a]
HIA
E/Z
CBO/HIA[b]
RC(NOH)CHO
Time
t=0
(CBO%)
nd[c]
(9) CH3(CH2)7-
E only
10/1
<1%
2h, 350 nm
22h, 350 nm
0.7/1
(40%)
E only
(10) [CH3(CH2)9][CH3(CH2)7]CH-
t=0
E only
nd[b]
0.5/1
(30%)
20/1
2h, 350 nm
~5/1
22h, 350 nm E only
(95%)
(11) (CH3)3C-
t=0
1/10
5/1
nd[b]
0.1/1
(9%)
12/1
2h, 350 nm
22h, 350 nm
6/1
(77%)
[a] Irradiation ( = 350 nm) in DMSO-d6 followed by 1H-NMR analysis.
[b]
Intensity of NOH signal of CBO/total CHO signal (both E and Z);
CBO% = 100xCBO/(CBO+HIA).
[c] No CBO detected by NMR.
1
2
3
4
5
6
The Z isomers of 9 and 10 are only observed upon short time irradiation (E/Z 10/1 and 5/1, respectively) and are no longer detected
after 22h exposure to light. After 22h, 10- and 11-CBO formation is almost quantitative (95 and 77%, respectively) and less remarkable
in 9 (40%). It should be noted that in these experimental conditions configuration inversion of HIA 10 appears to be associated with
sizable amounts of CBO at all irradiation times. In the case of 11, instead, it is possible to obtain inversion of configuration after 2h
photostimulation with less than 10% CBO formation.
7
8
Electrochemical properties, acidity and O-H bond dissociation enthalpies of HIAs
As outlined in the introduction, oximate ions find important applications related to their nucleophilic properties and to their ability to form
metal complexes. Particularly, oximate ions of low basicity (pKa< 9) exhibit an enhanced nucleophilicity relative to alkoxide anions, due
to the α-nucleophile effect. This property, which has drawn considerable attention towards oximes in the detoxification of
organophosphorous compounds, levels off as the acidity of the parent oxime decreases.[11] The capacity of oximes and their anions to
form metal coordination complexes can be used in multiple-analyte and/or electrochemical systems for detection of metal ions or other
organic species.[41] Redox properties and acidity of oximes are important properties to assess in view of these applications. In the case
of HIAs, the presence of the electron-withdrawing CHO substituent on the C(=NOH) group is expected to enhance their acidity and
affect the oxidation potential of their anion with respect to the corresponding unsubstituted oxime/oximate pairs. A further property of
HIAs that is worth investigating is their Bond Dissociation Enthalpy (BDE). The energy involved in the homolytic dissociation of the NO-
H is an important factor in the metabolic activity of oximes (Ref.[42] and references therein) and in evaluating the H-abstraction by other
radical species that may be present in a reaction environment, to yield -CHO iminoxyl radicals. For instance, in our previous
investigation on the RAFT copolymerization of an HIA-containing methacrylate,[1] we were left with the question of HIAs undergoing H-
abstraction as a possible interfering reaction. Potential H- acceptors are the 2-cyanopropyl radicals resulting from the decomposition of
the AIBN initiator or of the 2-cyanoprop-2-yl(4-fluoro)dithiobenzoate chain-transfer agent, as well as the carbon-centered radicals in the
propagating polymer chains. In fact, literature values for the BDE of 2-methylpropanenitrile are reported in the 85-92 kcal mol-1 range,
and those of unconjugated C-H bonds are likely to approach 100 kcal mol-1.[43] BDEs of simple oximes, instead, usually fall in the 77-90
kcal mol-1 range.[42, 44] However a plain comparison of gas-phase BDEs, which do not take into account solvent effects, is not conclusive
proof of the formation of >C=NO• species in the polymerization conditions. Anyway, assessing the relationship between BDEs and HIA
structures is a starting point in investigating some aspets of the radical polymerization of HIA-functionalized monomers.
As we are not aware of any previous investigation on the pKa, and oxidation potential of HIAs, we have determined their values for the
compounds included in the present work. As regards the BDEs, we compared the calculated gas-phase O-H bond dissociation
enthalpies of tertiary 11 and HIA-functionalized methacrylate 12. The latter also serves as a model compound for HIAs bearing a linear
R substituent.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Determination of oxidation potentials
HIA oxidation potentials were determined by cyclic voltammetry. Measurements were carried out at a scan rate of 0.5V/sec in DMF,
using 0.1M Bu4NBF4 as the supporting electrolyte, a glassy carbon working electrode and an Ag/AgCl/KCl 3M reference electrode.
Figure 6 shows representative voltammograms of 10, 12 and their anions. All other traces are reported in the SI. Potentials were first
swept to 2V/NHE. Similarly to simple ketoximes and aldoximes,[37c, 38] none of the HIAs were oxidized even at this potential. After
deprotonation with a 4-fold excess of Me4NOH, all 2-(hydroxyimino)aldehyde anions show two irreversible oxidation peaks (Table 24,
Figure 6 and SI), the first in the 0.40 - 0.65 V/NHE range, the second at E>1.1V/NHE. The latter is more pronounced in 10-, 11-, 13-.
When the potential was switched at 0.8V/NHE (scan rate of 0.5 V/s), oxidation peak of 10- (Ep = 0.52V/NHE) and 15- (Ep = 0.61V/NHE)
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
1
2
3
4
5
6
7
remained irreversible, while that of 12- (Ep = 0.64V/NHE) and all other anions became partially reversible (Figure 6, insets). A more
detailed investigation on the chemical processes causing this lack of reversibility is currently in progress. Anyway, these results indicate
that 2-(hydroxyimino)aldehyde anions, with the exception of compound 11- having a bulky substituent (Ep = 0.38V), are oxidized at
higher potentials than simple aldoximes (CH3CH=NOH, Ep= 0.137V/NHE; (E)-PhCH=NOH, Ep= 0.328V/NHE) and ketoximes
((CH3)2C=NOH Ep= 0.181 V/NHE; Ph2C=NOH, Ep= 0.225V/NHE).[40-41] Indeed, oxidation potentials of HIA anions are more similar to
those of -ketoximes (CH3COCH=NO-, Ep= 0.558V/NHE; PhCOCH=NO-, Ep= 0.540 V/NHE)[45] bearing a carbonyl group in conjugation
to the >C=NO- moiety.
10
-
10
-
10
0.6 0.4 0.2 0.0 -0.2
E (V vs Ag/AgCl/KCl 3M)
12
-
12
-
12
0.6 0.4 0.2 0.0 -0.2
E (V vs Ag/AgCl/KCl 3M)
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
E (vs Ag/AgCl/ KCl 3M)
Figure 6. Cyclic voltammetry of 2-(hydroxyimino)aldehydes 10, 12 and
their anions in DMF at 0.5 V/s
8
9
10
11
12
13
Determination of pKa
The pKa of HIAs 9-16 was determined by potentiometric titration. Due to their insolubility in pure water, all HIAs except 10 were
dissolved in DMSO/water mixtures (10/90, 20/80 and 30/70, v/v) and titrated with KOH 0.1 M in the presence of KCl (0.15 M) to
maintain a constant ionic strength. For each DMSO/water mixture, experiments were carried out at least twice and the corresponding
psKas were calculated using Henderson-Hasselbalch equation. Aqueous pKa values were then derived from psKa data through the
Yasuda-Shedlovsky equation (Table 4).42
14
Table 4. pKa of 2-(hydroxyimino)aldehydes and oxidation potential (Ep) of the
corresponding anions
Ep
[a]
RC(NOH)CHO
pKa
(V vs. NHE)[b]
(9) CH3(CH2)7-
8.8 ± 0.2
0.55 ± 0.01
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
8.7 ± 0.2[c]
9.90 ± 0.03
7.68 ± 0.07
8.7 ± 0.3
0.52 ± 0.02
0.38 ± 0.01
0.64 ± 0.02
0.51 ± 0.01
0.63 ± 0.01
0.61 ± 0.01
0.60 ± 0.02
(10) [CH3(CH2)9][CH3(CH2)7]CH-
(11) (CH3)3C-
(12) CH2=C(CH3)CO2(CH2)4-
(13) Cyclohexyl-
(14) Ph-
7.69 ± 0.02
8.05 ± 0.07
7.94 ± 0.06
(15) PhCH2-
(16) Np-
[a] In pKa units; psKa were determined by potentiometric titration in
DMSO/mixtures and converted to aqueous pKa by the Yasuda-Shedlovsky
equation:[46] psKa + log[H2O] = a + b/ε
[b] Oxidation potential were determined by CV in DMF at scan rate 0.5 V/s,
using a glass carbon working electrode and Ag/AgCl/KCl 3M reference
electrode; oxidation potential are referred to NHE by equation: E(V vs. NHE)
= E(V vs. Ag/AgCl/KCl 3M) + 0.21[45]
[c] Determined by linear regression of plot pKa versus Ep (see text)
1
2
3
Unfortunately, HIA 10 was not soluble in any DMSO/water proportion. Based on values in Table 4, however, we are able to show a
correlation of pKa and Ep values. A plot of pKa of HIAs vs. the Ep of the corresponding anionsis, indeed, fairly linear (Figure 7) and, from
linear regression of all data (r2= 0.96), it has been possible to estimate for compound 10 a pKa of 8.7.
8
7
11
6
9
5
4
3
13
15
14
12
16
6
8
10
12
14
16
p
-1
FE (kcal mol
)
Figure 7. Plot of pKa of 2-(hydroxyimino)aldehydes versus oxidation potential of the
corresponding anions
4
5
As shown in Table 4, the values of measured pKa for all HIAs fall within a narrow range (2.2 pKa units) around 7.7-9.9, so they are
weaker acids than -arylcyanoximes (pKa range 4.5-6.2).35 Their acidity is, instead, comparable to that of hydroxamic acids, ranging
45
6
approximately from 7.9 for C6H5C(O)N(OH)CH3 to 9.9 for (CH3)3C(O)N(OH)CH3.46 As highlighted by the linear correlation in Figure 7,
7
compound 11, i.e. the HIA bearing the bulkiest R group, is the least acidic among the investigated HIAs and its anion shows the lowest
oxidation potential. These properties could be related to poorer solvation of the anion, or to the steric effect of the t-Bu group.
8
9
10
11
12
13
14
15
16
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
Table 5. Calculated bond lenghts (Å) and angles (degrees) for HIAs 11 and 12 with their corresponding anions and radicals in DMF[a]
RC(NOH)CHO
r(R-CN)
r(C-CO)
r(C=N)
r(N-O)
r(C=O)
∠RCN
∠RCC
∠CCN
∠CNO
(Z)-11 syn/anti
(Z)-11 syn/anti
1.53/1.53
1.54/1.53
1.48/1.50
1.44/1.44
1.29/1.28
1.33/1.34
1.37/1.39
1.29/1.30
1.23/1.21
1.24/1.24
118.35/117.37
116.84/117.68
118.66/121.64
117.15/121.98
123.00/120.99
126.00/120.34
119.17/115.22
120.51/120.50
(Z)-11• syn/anti
(E)-12 syn/anti
(E)-12 syn/anti
(E)-12• syn/anti
1.52/1.52
1.50/1.50
1.51/1.50
1.51/1.50
1.48/1.48
1.50/1.48
1.44/1.43
1.48/1.47
1.29/1.29
1.28/1.29
1.33/1.33
1.29/1.30
1.22/1.22
1.38/1.38
1.29/1.29
1.21/1.22
1.21/1.21
1.21/1.21
1.24/1.24
1.21/1.22
120.89/120.33
126.28/127.04
123.02/124.17
121.52/122.41
122.41/124.70
118.63/121.23
119.58/122.45
122.40/123.52
116.71/114.95
115.09/111.73
117.40/113.38
116.07/114.07
132.55/135.02
112.86/113.18
117.55/117.98
133.73/133.94
[a] Geometry optimization at the B3LYP/6-311+G(2d,2p) level; effect of DMF solvent by means of the PCM
O
O
N
N
R
R
CHO
CHO
Figure 8. HIA radicals. Unpaired electron is localized on the N-O residue
and lies in the CNO plane.
2
3
Indeed, as reported in the literature,[42, 44a] the O-H bond dissociation enthalpy of (t-Bu)2C=NOH is about 5 kcal/mol lower than that of
linear oximes. This effect has been ascribed to the smaller repulsion between the two t-Bu groups, and between t-Bu and O, in the
iminoxyl radical than in the parent oxime. Similarly, one could expect a larger steric relief upon oxidation of 11- to the corresponding
adical than in the case of a linear, less hindered HIA-.
4
5
6
In order to assess the role of steric relief in lowering both the acidity of 11 and the energy requirement for the oxidation of its anion, we
computed the lowest-energy structures of 11, 12 and of their corresponding anions and radicals by QM calculations. The effect of DMF,
7
1
8
the solvent used for oxidation potential determinations, was modeled using the Polarizable Continuum Model (PCM). Since H-NMR
9
and computational analysis show that compound 11 exists mainly as isomer (Z)-anti (Z/E ≈10) while 12 is present exclusively as isomer
E (see previous section), calculations were limited to the (Z)-11 and (E)-12 isomers, both in syn and anti conformations (Figure 2), with
their corresponding anions and radicals. Discussion is focused on anti-(Z)-11 and syn/anti-(E)-12. Results in Table 5 highlight some
interesting structural features of HIA anions and radicals, regardless of substituent bulk. In anti-(Z)-11 the C-CO and N-O bonds are
shorter in the anion, compared to the parent compound (1.50 to 1.44 Å and 1.39 to 1.30 Å, respectively). Consistently, C=N and C=O
bonds are lengthened (1.28 to 1.34 Å and 1.21 to 1.24 Å, respectively). Similar observations can be made for the 12/12 pair.
Furthermore, the C(=NOH) and CHO groups in the optimized HIA/HIA structures are coplanar in both the syn and anti isomers. These
results suggest that the structure of HIA anions can be described as intermediate between the two resonance forms depicted in Figure
2. It is worth noting that neither 11, nor 12 exhibit any significant differences with their anions in terms of bond angles. Moreover, their
values are quite similar when comparing the two HIA/HIA pairs. In contrast with the anions, HIA radicals exhibit very similar bond
lengths as their parent compounds with the exception of the N-O bond, which is shorter by 0.15-0.17 Å, while ∠CNO angles increase by
13-20° for both 11 and 12. For the radical species, the calculation of spin densities, performed at the UB3LYP/6-311+G(2d,2p) level,
indicates a spin density of about 0.52-0.55 on O and about 0.43-0.47 on N atoms of the C(=NOH) moiety, while the oxygen of CHO
group and all carbons have almost null values of spin densitiy. In agreement with previous literature on oximes,[42, 44a] these results
suggest that the iminoxyl radical is of the type, i.e. the unpaired electron lies in the same plane as the local molecular framework
(Figure 8). In other words, in contrast with the negative charge of the corresponding anions, the unpaired electron is not delocalized on
the CHO substituent. Direct comparison of 11• and 12• with their respective anions shows that ∠CCN angles expand slightly more in the
11•/11 than the 12•/12 pair. The magnitude of these 11 vs. 12 differences, however, does not account for a significant steric effect in
the radical relative to the anion forms of 11.[44a] When all the above results are taken into account, it appears that the presence of the
bulky t-Bu group on 11 affects both its acidity and oxidation potential through poorer solvation of the anion, rather than through a
distortion of the anion or of the radical. This (lack of) solvation effect is expected to be particularly sizable in solvents such as water or
DMF, which are able to form hydrogen bonds with the C(=NOH) and CHO groups.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Calculation of O-H Bond Dissociation Enthalpies of some HIAs
The calculated BDEs of 11 and 12 are reported in Table 6 along with literature BDE values of some symmetric ketoximes bearing alkyl
substituents of different bulk.[44a] Firstly, results in Table 6 indicate that BDE values of HIAs are somewhat smaller than those of simple
oximes. Substituting the CHO group for one of the ketoxime alkyl substituents to form the HIA lowers the BDE by about 3-10 kcal mol-1.
As reported in the literature, BDEs in ketoximes depend on geometrical isomerism, steric effects, dipole-dipole interactions and
intramolecular hydrogen bonding. For instance, the BDEs of E-oximes are about 3-5 kcal mol-1 higher than those of the corresponding
Z-isomers.[42, 44a] Indeed, this also applies to HIA 12, even though the difference is limited to only 3 kcal mol-1. Interestingly, the hindered
HIA 11 exhibits an opposite trend (Table 6).
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
more destabilized
less destabilized
R
R
H
R
R
O
H
O
N
N
O
OH
H
HO N
HO N
O
OH
H
∠CCN angles
Z-syn
Z-anti
121.3°
123.7°
E-syn
110.7°
115.4°
E-anti
106.7°
112.2°
11
12
122.9°
123.3°
Figure 9. Dipolar repulsions in the different isomers of HIAs. 11 and 12. ∠CCN angles of from optimized geometries in the gas-phase at the B3LYP/6-
311+G(2d,2p) level.
Table 6. Gas-phase O-H BDEs of 2-(hydroxyimino)aldehydes[a] at 298 K in
kcal mol-1
RC(=NOH)CHO
Z-syn
Z-anti
E-syn
E-anti
(11) (CH3)3C-
74.8
75.5
72.7
74.4
72.9
75.7
72.5
77.5
(12) CH2=C(CH3)CO2(CH2)4-
t-Bu2C(N=OH)
75.6[b]
80.7[b]
81.8[b]
82.6[b]
i-Pr2C(N=OH)
Et2C(N=OH)
Me2C(N=OH)
[a] Calculated by DFT at the B3LYP/6-311+G(2d,2p) level of theory. [b] From
Ref.[44a]
2
3
A small effect of the syn/anti isomerism is also observed, in that the BDEs of (Z)-syn isomers (both 11 and 12) are about 2 kcal mol-1
larger than those of the corresponding (Z)-anti forms. Intramolecular hydrogen bond between the oxygen atom in CHO and the
hydrogen in C(N=OH) is likely to be accountable for this energy difference. The O-H--O distances in the optimized geometries of syn-
(Z)-11 and syn-(Z)-12, indeed, are 2.57 Å and 2.60 Å in 11 and 12, respectively, which is also indicative of hydrogen bonding. In
agreement with the NMR results discussed in the previous section, our calculations in the gas phase confirm the greater stability (5-7
kcal mol-1) of anti-(E)-12 relative to all other isomers of 12 and the lesser stability (about 1-5 kcal mol-1) of (E)-11 relative to (Z)-11. We
believe that this difference should be attributed to the combined effect of t-Bu bulk and dipole moments interactions, as explained next.
In Figure 9 the calculated ∠CCN angles are reported for all isomers, together with a pictorial of the dipole moments associated with the
C(N=OH) and CHO groups. While ∠CCN angles in syn/anti-(Z)-11 and 12 are quite similar, the C(N=OH) and CHO groups (and the
dipole moments associated with them) in E are brought closer together by the bulk of the t-Bu in 11 (∠CCN 107-111°) than by the less
bulky substituent in 12 (∠CCN 112-115°). Dipolar interactions would therefore be more destabilizing in (E)-11 than in (E)-12. As for the
gas-phase optimized structures of iminoxyl radicals, results are analogous to those obtained in DMF, in that the unpaired electron is of
the -type, lying in the molecular framework plane. In all radicals, moreover, ∠CNO angles are in the 132-134°, i.e. larger than in the
parent HIAs (∠CNO range: 112-119°), thus making dipole-dipole interactions less significant. The larger dipolar destabilization of HIA
11 relative to 12 and the corresponding larger relief in the radical optimized structures could rationalize the BDEs in (Z)-11 and (Z)-12
4
5
6
7
8
9
10
11
12
13
14
15
16
17
and
the
smaller
BDEs
of
(E)-11
with
respect
to
(E)-12
(Table
6).
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
1
Conclusions
2
3
We have presented a comprehensive investigation of the photochemistry and physico-chemical properties of a new class of organic
compounds, 2-(hydroxyimino)aldehydes. The interpretation of experimental NMR, acidity and redox properties is corroborated by
DFT calculations. UV irradiation ( = 350 nm) results in two main phenomena, i.e. E/Z photoisomerization, occurring within two hours,
and the slower, irreversible formation of cyclobutanol oxime (CBO) derivatives by a Norrish-Yang cyclization. To our knowledge, this
is the first time that CBOs are described in the literature. The predominant initial configuration, the extent of light-induced
isomerization and the rate of subsequent thermal annealing to the initial E/Z population are strongly dependent on HIA substituent.
HIAs bearing primary and secondary substituents yield limited configurational inversion, whereas the presence of a tertiary group
results in E/Z switch. Thermal relaxation in the dark to the initial configuration proceeds over weeks when HIA substituents are
secondary or tertiary and is complete within days with a linear alkyl group. However, our results with a primary substituent bearing a
remote methacrylate show that branching next to the HIA is not the only factor affecting the rate of thermal relaxation. Cyclization to
CBO begins at the same time as E/Z isomerization and becomes predominant over long irradiation times. However, it is possible to
obtain E/Z inversion with little CBO formation over short irradiation times of an HIA bearing a tertiary substituent.
As expected, the vicinity of the CHO and oxime groups causes the acidity of the latter to be stronger than that of simple oximes and
close to that of hydroxamic acids. The oxidation potentials (Ep) of HIA anions are in line with those of -oxo-oximes, i.e. higher than
simple oximes. BDEs of the O-H bond of C(=NOH) group, calculated by DFT methods in vacuum, are lower than in simple oximes.
BDEs of all isomers of a primary and a tertiary HIA are interpreted through differences in dipole-dipole interactions, which are found
to be affected, in some isomers, by the bulk of the tertiary substituent.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
This study confirms that the strong dependence of HIA photochemistry on the bulk and nature of substituents is of great interest
towards designing different HIA-functionalized monomers. In fact, it should be possible to obtain polymers with opposite E/Z
abundance, both at thermal- and photoequilibrium, and to exploit the differences in hydrogen bond-driven pre-organization in solution
between the two configurations.[24] Furthermore, the photoisomerization of some HIAs to the CBO structure would constitute a strong
response to UV light.
24
Experimental Section
25
26
Please refer to Supporting Information for details on chemicals, general instrumentation and yields and characterization of specific
compounds.
27
28
29
30
31
32
33
34
35
36
37
38
39
40
HIA syntheses. General procedure
HIAs were prepared according to a procedure optimized in our laboratory.[5] Generally, under argon atmosphere, pyrrolidine (0.4
mmol, 33 μL) is added to a solution of p-TsOH monohydrate (0.4 mmol, 76 mg) in 3 mL of anhydrous DMF, followed by addition of
distilled water (4.0 mmol, 72 μL) and aldehyde (2.0 mmol). After stirring for 5 minutes, NaNO2 (2.0 mmol, 0.14 g) is added and then
anhydrous FeCl3 (2.0 mmol, 0.32 g) is carefully introduced in aliquotes to avoid overheating (use caution: exothermic reaction).
Magnetic stirring is continued at room temperature and reaction progress is monitored by thin layer chromatography and GC or GC-
MS analysis. Note that in gaschromatographic analyses, HIAs exhibit a slight decomposition to the corresponding nitrile derivatives
as confirmed by comparison with authentic samples available in our laboratory. When no further reaction progress is observed (4.5 –
5.5 h), 5 mL of saturated NaCl solution is added and, after stirring for 30 minutes, the mixture is extracted with ethyl acetate (4x15
mL). The organic phase is washed with a saturated solution of NaCl and dried over Na2SO4. After evaporation of the solvent under
reduced pressure, all products are purified by silica gel chromatography using a hexane/ethyl acetate gradient and characterized by
spectroscopic analysis.
DFT calculations
41
42
43
44
45
46
47
48
49
The coordinates of the molecules were prepared using Molden software and all the calculations were performed with Gaussian 09
program.[47] For the calculation of the BDE the geometries of the neutral species and of the corresponding radicals were optimized by
means of DFT using the B3LYP functional, the 6-311+G(2d,2p) basis set. The enthalpy correction terms have been estimated from
the Hessian matrix.
To compare the effect of the substituents, the molecular geometries (here including also the corresponding anions) have been
optimized using the same level of theory and including the solvent effect by means of the Polarizable Continuum Model (solvent
DMF).
Concerning the NMR parameter estimates, the chemical shifts were computed using the procedure found in Pierens,[39] which
provides the best estimate of the experimental values of the chemical shifts;[48] that is, the optimizations were performed in vacuo and
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
1
2
3
the shielding tensors were calculated by means of the PCM (solvent DMSO) using the WPO4 functional and the aug-cc-pVdZ basis
set.
In all the optimizations described above, the computation of the Hessian matrix excluded the presence of negative frequencies.
4
5
6
Conflicts of interest
The authors declare no conflict of interest
Keywords: 2-(hydroxyimino)aldehydes • photoisomerism • acidity • bond energy • multi-stimuli responsive polymers
7
8
9
Note: § Bordwell et al.[49] measured irreversible oxidation potentials in DMSO (0.1 M Et4NBF4) by CV using a platinum working
electrode and Ep values are reported vs. ferrocinium/ferrocene couple. To compare with our results, these potential values are
referred to NHE electrode according to equation E(V vs.NHE) = E (V vs. Fc+/Fc) + 0.75.
10
11
12
13
14
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER 2-(hydroxyimino)aldehydes
Prof. Patrizia Gentili*, Dr. Martina Nardi,
Irene Antignano, Dr. Paolo Cambise,
Prof. Marco D’Abramo, Dr. Francesca
D’Acunzo*, Dr. Alessandro Pinna, Dr.
Emanuele Ussia
Versatile oxime derivatives.
2-(Hydroxyimino)aldehydes
(HIAs)
R
CHO
OH
C
N
Photochemical and physico-chemical
properties of a series of 2-
pKa
7.7 - 9.9
R
CHO
photoisomerism
h
C
N
100
50
(hydroxyimino)aldehydes (HIAs) were
explored as a function of structural
variations. Results show that HIAs are
highly versatile molecules for many
applications and that carefully-designed
HIA-containing monomers should impart
sensitivity to light, pH and oxidative
conditions to copolymers, with a broad
range of responses.
-
O
R
CHO
OH
R
CHO
0
-50
N
N
h
and/or
HO
-100
-150
-200
-250
N
R1
OH
R
CHO
O
R2
HO
C
N
Page No. – Page No.
-0.2
0.0
0.2
0.4
0.6
0.8
E
(V versus Ag/AgCl/KCl 3M)
H
C
H
O
C
3
3
m
n
2-(Hydroxyimino)aldehydes:
O
O
O
R
Photochemical and Physico-Chemical
Properties of a versatile functional
group for monomer design
HO
HIA-functionalized
responsive polymers
O
O
N
H
O
p
CH
3
[1] M. Nardi, F. D'Acunzo, M. Clemente, G. Proietti, P. Gentili, Polymer Chemistry 2017, 8, 4233.
[2] K. Sugamoto, Y. Hamasuna, Y. I. Matsushita, T. Matsui, Synlett 1998, 1270.
[3] T. L. Gilchrist, T. G. Roberts, Journal of the Chemical Society, Perkin Transactions 1 1983, 1283.
[4] M. Baidya, H. Yamamoto, Journal of the American Chemical Society 2011, 133, 13880.
[5] P. Gentili, S. Pedetti, Chemical Communications 2012, 48, 5358.
[6] E. Abele, R. Abele, Current Organic Synthesis 2014, 11, 403.
[7] Y. Ashani, I. Silman, Hydroxylamines and Oximes: Biological Properties and Potential uses as
Therapeutic Agents, in The Chemistry of Hydroxylamines, Oximes and Hydroxamic Acids, John Wiley
& Sons, Ltd, 2008, pp. 609.
[8] D. K. Kölmel, E. T. Kool, Chemical Reviews 2017, 117, 10358.
[9] K. K.-Y. Kung, K.-F. Wong, K.-C. Leung, M.-K. Wong, Chemical Communications 2013, 49, 6888.
[10] A. Balamurugan, H. I. Lee, Macromolecules 2015, 48, 3934.
[11] N. Singh, Y. Karpichev, A. K. Tiwari, K. Kuca, K. K. Ghosh, Journal of Molecular Liquids 2015,
208, 237.
[12] M. A. Gordillo, M. Soto-Monsalve, C. C. Carmona-Vargas, G. Gutiérrez, R. F. D'Vries, J. M. Lehn,
M. N. Chaur, Chemistry - A European Journal 2017, 23, 14872.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
[13] A. R. Hajipour, M. Karimzadeh, S. Jalilvand, H. Farrokhpour, A. N. Chermahini, Computational
and Theoretical Chemistry 2014, 1045, 10.
[14] Y. A. Gur’eva, O. A. Zalevskaya, L. L. Frolova, I. N. Alekseev, P. A. Slepukhin, A. V. Kuchin,
Russian Journal of General Chemistry 2014, 84, 137.
[15] A. V. Afonin, I. A. Ushakov, D. V. Pavlov, A. V. Ivanov, A. b. I. Mikhaleva, Magnetic Resonance
in Chemistry 2010, 48, 685.
[16] V. A. Pavlov, J. A. S. Smith, T. A. Zjablikova, Magnetic Resonance in Chemistry 1992, 30, 716.
[17] D. S. Bohle, Z. Chua, I. Perepichka, K. Rosadiuk, Chemistry - A European Journal 2013, 19, 4223.
[18] M. A. M. Al-Alwani, A. B. Mohamad, N. A. Ludin, A. A. H. Kadhum, K. Sopian, Renewable and
Sustainable Energy Reviews 2016, 65, 183.
[19] L. Infantes, S. Motherwell, Prediction of H-bonding motifs for pyrazoles and oximes using the
Cambridge structural database, in Science of Crystal Structures: Highlights in Crystallography, 2015,
pp. 269.
[20] D. S. Bolotin, N. A. Bokach, V. Y. Kukushkin, Coordination Chemistry Reviews 2016, 313, 62.
[21] A. S. Dinca, C. Maxim, B. Cojocaru, F. Lloret, M. Julve, M. Andruh, Inorganica Chimica Acta
2016, 440, 148.
[22] L. Croitor, E. B. Coropceanu, A. E. Masunov, H. J. Rivera-Jacquez, A. V. Siminel, V. I. Zelentsov,
T. Y. Datsko, M. S. Fonari, Crystal Growth and Design 2014, 14, 3935.
[23] a) D. R. Klaus, M. Keene, S. Silchenko, M. Berezin, N. Gerasimchuk, Inorganic Chemistry 2015,
54, 1890; b) A. Mann, N. Gerasimchuk, S. Silchenko, Inorganica Chimica Acta 2016, 440, 118.
[24] F. Wu, X. Tang, L. Guo, K. Yang, Y. Cai, Soft Matter 2013, 9, 4036.
[25] T. S. V. Buys, H. Cerfontain, J. A. J. Geenevasen, Recueil des Travaux Chimiques des Pays‐Bas
1990, 109, 531.
[26] G. A. Molander, J. A. C. Romero, Tetrahedron 2005, 61, 2631.
[27] A. De Mico, R. Margarita, L. Parlanti, A. Vescovi, G. Piancatelli, Journal of Organic Chemistry
1997, 62, 6974.
[28] J. Hu, D. L. Mattern, Journal of Organic Chemistry 2000, 65, 2277.
[29] M. G. Bell, R. A. Doti, M. J. Genin, P. A. Lander, T. Ma, P. R. Manninen, J. M. Ochoada, F. Qu, L.
S. Stelzer, R. E. Stites, Fxr agonists, Patent WO 2007140183 A1, 2007.
[30] H. D. Roth, Ligh-induced chemistry of oximes and derivatives, in PATAI'S Chemistry of
Functional Groups, John Wiley & Sons, Ltd, 2010.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
[31] M. Cyman, J. Wielińska, H. Myszka, D. Trzybiński, A. Sikorski, A. Nowacki, B. Liberek, Journal
of Molecular Structure 2016, 1125, 558.
[32] F. Blanco, I. Alkorta, J. Elguero, Croatica Chemica Acta 2009, 82, 173.
[33] F. Stunnenberg, H. Cerfontain, R. B. Rexwinkel, Recueil des Travaux Chimiques des Pays‐Bas
1992, 111, 438.
[34] D. E. Fast, A. Lauer, J. P. Menzel, A. M. Kelterer, G. Gescheidt, C. Barner-Kowollik,
Macromolecules 2017, 50, 1815.
[35] T. S. V. Buys, H. Cerfontain, J. A. J. Geenevasen, F. Stunnenberg, Recueil des Travaux Chimiques
des Pays‐Bas 1986, 105, 188.
[36] M. Oelgemöller, N. Hoffmann, Organic and Biomolecular Chemistry 2016, 14, 7392.
[37] a) D. Robertson, J. F. Cannon, N. Gerasimchuk, Inorganic Chemistry 2005, 44, 8326; b) O. T.
Ilkun, S. J. Archibald, C. L. Barnes, N. Gerasimchuk, R. Biagioni, S. Silchenko, O. A. Gerasimchuk, V.
N. Nemykin, Dalton Transactions 2008, 5715; c) P. Baas, H. Cerfontain, Journal of the Chemical
Society, Perkin Transactions 2 1979, 156.
[38] D. Eddings, C. Barnes, N. Gerasimchuk, P. Durham, K. Domasevich, Inorganic Chemistry 2004,
43, 3894.
[39] G. K. Pierens, Journal of Computational Chemistry 2014, 35, 1388.
[40] P. Baas, H. Cerfontain, Tetrahedron Letters 1978, 19, 1501.
[41] Y. Zhang, H.-b. Fa, B. He, C.-j. Hou, D.-q. Huo, T.-c. Xia, W. Yin, Journal of Solid State
Electrochemistry 2017, 21, 2117.
[42] S.-S. Chong, Y. Fu, L. Liu, Q.-X. Guo, The Journal of Physical Chemistry A 2007, 111, 13112.
[43] Y.-R. Luo, Handbook of Bond Dissociation Energies in Organic Compounds, CRC Press, 2007.
[44] a) D. A. Pratt, J. A. Blake, P. Mulder, J. C. Walton, H. G. Korth, K. U. Ingold, Journal of the
American Chemical Society 2004, 126, 10667; b) R. Dao, X. Wang, K. Chen, C. Zhao, J. Yao, H. Li,
Physical Chemistry Chemical Physics 2017, 19, 22309.
[45] E. P. Friis, J. E. T. Andersen, L. L. Madsen, N. Bonander, P. Moller, J. Ulstrup, Electrochimica
Acta 1998, 43, 1114.
[46] a) M. Yasuda, Bulletin of the Chemical Society of Japan 1959, 32, 429; b) T. Shedlovsky, R. Kay,
The behaviour of carboxylic acids in mixed solvents, in Electrolytes, Pergamon Press, New York, 1962,
pp. 146.
[47] G. W. T. M. J. Frisch, 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.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800059
Chemistry - A European Journal
FULL PAPER
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, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian 09, Revision
A.02 2009.
[48] a) M. W. Lodewyk, M. R. Siebert, D. J. Tantillo, Chemical Reviews 2012, 112, 1839; b) M. W.
Lodewyk, C. Soldi, P. B. Jones, M. M. Olmstead, J. Rita, J. T. Shaw, D. J. Tantillo, Journal of the
American Chemical Society 2012, 134, 18550; c) M. W. Lodewyk, D. J. Tantillo, Journal of Natural
Products 2011, 74, 1339; d) R. Jain, T. Bally, P. R. Rablen, The Journal of Organic Chemistry 2009, 74,
4017; e) P. R. Rablen, S. A. Pearlman, J. Finkbiner, The Journal of Physical Chemistry A 1999, 103,
7357.
[49] F. G. Bordwell, G. Z. Ji, Journal of Organic Chemistry 1992, 57, 3019.
This article is protected by copyright. All rights reserved.