Hydrazo coupling: The efficient transition-metal-free C-H functionalization of 8-hydroxyquinoline and phenol through base catalysis

Article abstract of DOI:10.1039/c9gc02824b

Azodicarboxylate esters are common reagents in organic synthesis laboratories due to their utility in the Mitsunobu reaction. They can also be regarded as possible starting compounds for C-H functionalization, which up until now has been mainly achieved by transition-metal-catalyzed reactions. We have developed a novel reaction involving the quantitative coupling of 8-hydroxyquinoline or phenol with azodicarboxylate esters. The functionalization proceeds under mild base-catalyzed conditions selectively, and either the ortho-position of 8-hydroxyquinoline or para-position of the phenol/naphthol is involved in the reaction. This type of transformation can be considered as "hydrazo coupling" (by analogy with azo coupling). Herein, we discuss a plausible mechanism for this catalyzed substitution, backing up our findings with deuterium NMR experiments and by varying the starting compounds and bases. Using Boc-NN-Boc as a substrate, we have developed the convenient and efficient synthesis of (8-hydroxyquinolin-7-yl)hydrazines, as well as demonstrating a new stereoselective route for the synthesis of medicinally important 4-hydroxyphenylhydrazine for laboratory use, which almost doubles the yield of the common industrial process and reduces the number of synthetic steps. A new "one-pot" procedure for the synthesis of aromatic 8-hydroxyquinolin-7-yl hydrazones was applied.

Full text of DOI:10.1039/c9gc02824b

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Hydrazo coupling: the efficient transition-metal-
free C–H functionalization of 8-hydroxyquinoline
and phenol through base catalysis†
Cite this: Green Chem., 2019, 21,
6381
Pavel N. Solyev, * Daria K. Sherman,
and Sergey N. Kochetkov
Roman A. Novikov,
Eugenia A. Levina
Azodicarboxylate esters are common reagents in organic synthesis laboratories due to their utility in the
Mitsunobu reaction. They can also be regarded as possible starting compounds for C–H functionalization,
which up until now has been mainly achieved by transition-metal-catalyzed reactions. We have developed
a novel reaction involving the quantitative coupling of 8-hydroxyquinoline or phenol with azodicarboxy-
late esters. The functionalization proceeds under mild base-catalyzed conditions selectively, and either
the ortho-position of 8-hydroxyquinoline or para-position of the phenol/naphthol is involved in the reac-
tion. This type of transformation can be considered as “hydrazo coupling” (by analogy with azo coupling).
Herein, we discuss a plausible mechanism for this catalyzed substitution, backing up our findings with
deuterium NMR experiments and by varying the starting compounds and bases. Using Boc-NvN-Boc as
a substrate, we have developed the convenient and efficient synthesis of (8-hydroxyquinolin-7-yl)hydra-
zines, as well as demonstrating a new stereoselective route for the synthesis of medicinally important
4-hydroxyphenylhydrazine for laboratory use, which almost doubles the yield of the common industrial
process and reduces the number of synthetic steps. A new “one-pot” procedure for the synthesis of aro-
matic 8-hydroxyquinolin-7-yl hydrazones was applied.
Received 10th August 2019,
Accepted 15th October 2019
DOI: 10.1039/c9gc02824b
rsc.li/greenchem
organic synthesis for several decades. In early research, such
coupling was postulated to be a radical process.2 Intensive
Introduction
Selective functionalization of C(sp2)–H bonds has been a hot research over the last two decades has resulted in a collection
topic in organic synthesis for the last decade. Aromatic struc- of studies concerning the use of ADs for the amination of aro-
tures prevail in complex natural products and simple organic matic substrates, such as benzene derivatives and phenols. In
molecules, so it is no surprise that many strategies have been general, these approaches include either acidic conditions,
developed to functionalize heterocycles and drug precursors.1 Lewis acids, or radical initiators, and they are mainly based on
Every method presents challenges in terms of yield, site transition-metal-catalyzed hydrazination via electrophilic sub-
selectivity, and others. As a result, experimental techniques are stitution at 40–70 °C (Table 1).
becoming more complex, catalysts are becoming more exotic
and expensive, and harsh conditions are often required.
However, acidic conditions are harmful to many functiona-
lized derivatives, and moreover, expensive and complex transition
Azodicarboxylate (AD) esters can be regarded as possible metal catalysts need to be completely removed from the products,
starting compounds for feasible C–H functionalization. Their especially when dealing with the pharmaceutical industry.
use in the Mitsunobu reaction makes them common reagents
in organic synthesis laboratories, although their utility is not ADs in the 1990s, due to studies on the reductive addition of
limited to activators alone.
DEAD to allyl fragments,13 Buchwald–Hartwig amination6 and
There was a spike in the number of novel publications on
Without the assistance of transition metal catalysis, azodi- B. Sharpless’s group’s discovery of azodicarboxylates as con-
carboxylate coupling has demonstrated narrow perspectives in venient substrates for unique “on-water” reactions and 2σ +
2σ + 2π cycloadditions to double bonds.14
Research into reactions of phenol-type compounds with
Engelhardt Institute of Molecular Biology, Russian Academy of Sciences,
azodicarboxylates is scarce. A direct reaction of phenol with
32 Vavilov St., 119991 Moscow, Russian Federation. E-mail: solyev@gmail.com
azodicarboxylate esters gives an approximately equimolar
†Electronic supplementary information (ESI) available: Materials and methods,
procedures for synthesis, and 1H, 13C and 2D NMR and HRMS spectra of the pro-
ducts. See DOI: 10.1039/c9gc02824b
mixture of oxidation and substitution products; an excess of
DEAD leads to further oxidation of 3,3′,5,5′-tetraalkyl-4,4′-
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6381–6389 | 6381

View Article Online
Paper
Green Chemistry
Table 1 Transition-metal-catalyzed approaches for the hydrazination
of arenes
act through the same mechanism with phenol compounds.
That assumption lasted for more than 25 years, until the
topic was reborn with catalysts introduced into the reaction.
The use of activated bis(2,2,2-trichloroethyl) azodicarboxylate
(BTCEAD) in triflic acid and/or LiClO4 medium conditions
was described by Y. Leblanc and colleagues, suggesting a syn-
chronous mechanism between the HOMO of phenol and the
LUMO of BTCEAD.17
Later, W. J. Kinart and C. M. Kinart proposed two mecha-
nisms of phenol/β-naphthol hydrazination: via N: → Sn-co-
ordinated metallo-ene mechanism or via electrophilic substi-
tution through
(Scheme 1C).
a
stabilized Wheland intermediate18
Z
Lewis acid or catalyst
Sc(OTf)3 3 or 3 mol% RuCl3,4 or 1 mol% AuCl3
5
H
Br
There is an example of atropisomer synthesis described via
cinchona-alkaloid-catalyzed amination of β-naphthol with Di-
tBu-AD (Scheme 1D). The authors mentioned that a quinolin-
6-ol system of dihydrocupreidine was also affected and could
be aminated at the 5th position under more severe reaction
conditions.19
2–4 mol% Pd(dba)2 + 1.5 eq. Cs2CO3 + 3–6 mol%
TrixiePhos-type ligand (100 °C)6
Br, I
10 mol% CuI(I) + PPh3 + 20 mol% phen7
8
H, Cl
B(OH)2
OTf
ZrCl4
η1-Hydrazinato(phen)Pd9 or 5 mol% Cu(OAc)2
10
10 mol% Cu2O + NaH + LiI (70 °C)11
–N+uN
FeCl2
12
Quite recently, a SET mechanism of DIAD substitution with
AgOAc catalyst was proposed (Scheme 1e).20 However, this reac-
biphenyldiol
into
diphenylquinone
(Scheme
1A).15 tion demonstrates the necessity of the pyridine-containing
Previously, these products were obtained only via radical oxi- directing group and the α-aminonaphthalene moiety, as no
dation with K3Fe(CN)6 and peroxides,16 which is why reaction occurs in the corresponding aniline derivative.
S. H. Schroeter suggested that azodicarboxylate esters might
Thus, there are five different mechanisms and catalytic
approaches described for a one-type reaction, and every
method requires additional steps of purification from the
single-use catalyst.
8-Hydroxyquinoline (8-HQ) has gained a great deal of atten-
tion in medicinal chemistry, photochemistry and agrochemis-
try because of its strong coordinative nature. It is the most
widely produced quinoline derivative due to its cheap energy-
effective and simple synthesis. It is worth noting that levofloxa-
cin, a fluoroquinolone derivative containing an 8-HQ scaffold
with 7-amino fragments in its structure, comprised almost
30% of all the fluoroquinolone antibiotics on the market in
1991–2013.21 8-HQ is a framework for new generations of anti-
HIV agents, as integrase or Rev protein inhibitors.22 Activities
against key targets in Alzheimer’s disease have also been
found.23 However, the lack of known functionalization
methods described for 8-HQ restricts the variety of product
structures that can be obtained. Mainly, the Betti reaction (a
type of Mannich base synthesis),24 esterification and etherifi-
cation of hydroxyl groups,25 electrophilic substitution giving
the 5-substituted 8-HQ derivatives24,26 and some transition-
metal-catalyzed approaches are used. Most of the processes
used to access quinoline functionalization are performed by
substitution using several major methods (some of them
under harsh reaction conditions).24,27
Finally, the Mitsunobu reaction has hardly ever been a
method of choice for phenolic substrates. Although the
Mitsunobu reaction can be set with phenols and various
hydroxyquinolines, the products are formed with poor selecti-
vity and in moderate yields.28 It has been stated that salicylal-
dehyde tends to form hydrazine derivatives under Mitsunobu
reaction conditions.29
Scheme 1 The hydrazination of phenolic substrates and other arenes
with directing groups.
6382 | Green Chem., 2019, 21, 6381–6389
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
In fact, dialkyl azodicarboxylate by itself is a good substrate However, in the case of 8-HQ, the acidic medium converted
for the reaction with 8-hydroxyquinoline, and this is the the starting compound into a salt that was insoluble in THF/
subject of the discussion below.
dioxane, and no reaction was observed. However, there have
The aim of this work was to develop a new and facile syn- been reports on direct reactions of phenol with azodicarboxy-
thetic method for obtaining 7-hydrazino-8-hydroxyquinoline late ester, which have given moderate yields of 50–60%,15 and
derivatives and to explore the scope of base-catalyzed reactions some later works that employed harsher conditions with
of azodicarboxylate ester coupling to other substrates (with LiClO4 or triflic acid catalysis.17 In contrast to these already
phenolic OH as a directing group).
studied conditions, we conducted our syntheses with base cat-
Up until now, there were no 8-HQ hydrazine derivatives alysts. The same base catalysis was applied for phenol
described. Here, we report the first synthesis of 7-hydrazino-8- functionalization.
hydroxyquinoline derivatives and a “one-pot” method for their
We found that the desired coupling reaction can be pro-
transformation to aromatic hydrazones, using cost-effective vided by many bases in excellent yields, with sodium hydride
and transition-metal-free mild conditions. The proposed proving to be the best. In fact, it facilitates the full conver-
method was also applied to the synthesis of 4-hydroxyphenyl- sion of 8-HQ in several gram scale, and the product can be
hydrazine in our laboratory, resulting in double the yield com- easily isolated by partitioning in CH2Cl2/water and can be
pared to the commonly used industrial process. Access to this used without further purification. We assume that NaH acts
type of compound could reveal a new chemistry of azo reac- as a base initiator of the activation, since the reaction con-
tions and products with potential biological activity. A mecha- tinues with adding new portions of the starting reagents
nism discussed below resembles azo coupling, and that is why until it is quenched with a saturated NaHCO3 solution.
we suggest the use of the term “hydrazo coupling” for this Using more than 5 mol% NaH is not necessary: the excess of
kind of reaction.
the hydride remains uninvolved at the bottom of the flask. In
the model reaction with DIAD, we reached the 5,7-bis-adduct
after full conversion of the first DIAD equivalent into 1a
(Scheme 2).
Results and discussion
Other alkali metal hydrides and mild bases were tested and
found to effectively catalyze the functionalization. Nevertheless,
only with NaH was it possible to exclude the mixture of mono-
and bis-products and to obtain pure 1a and 1b. Interestingly,
phenol gained a higher yield of 2a with the less reactive Li-con-
taining bases, LiH and LiNH2. The latter gives almost the same
excellent yield as LiH, and has practically the same commercial
price as NaH for a slightly slower reaction rate, but LiNH2 sig-
nificantly lowers the contamination of the product with oxi-
dation by-products.31 In contrast, phenol and 8-hydroxyquino-
line remained unreactive when relatively weaker bases such as
CaH2 and alkylamines were used. Amines were found to be the
least effective bases in the reaction because they can form ion
pairs with ADs and convert the latter into nucleophiles, leading
to AD self-condensation by-products and preventing AD from
coupling with 8-HQ or phenol.32 However, this is not observed
when using DBU as the base.
Our initial investigation into the reaction of 8-hydroxyquino-
line with diisopropyl azodicarboxylate (DIAD) as a model reac-
tion resulted in moderate conversion of the starting 8-HQ. The
amination proceeded smoothly at room temperature and the
sole product obtained was purified and identified using 1H,
13C, 15N and 2D NMR and HRMS spectra, which confirmed its
structure to be diisopropyl 1-(8-hydroxyquinolin-7-yl)hydrazi-
nyl-1,2-dicarboxylate 1a. The formation of 1a proceeded both
in THF and dioxane with similar efficacy and the solvents did
not need to be freshly dried, though the reaction stopped each
time when the conversion reached 55–65%. No by-products
were found. Increasing the temperature up to 40 °C did not
lead to changes, but the yield significantly decreased above
that temperature level. We also studied phenol utilization as
an active part of the 8-HQ molecule in the reaction; in this
case, the para-adduct was formed. Furthermore, we proceeded
to investigate this reaction in order to facilitate the coupling of
8-HQ/phenol and azodicarboxylate esters and reveal the reac-
tion mechanism.
Solvent selection is limited to base-capable solvents, so
DMSO and DMF were excluded. Instead, replacing THF or
dioxane with greener industrial sulfolane led to a similar
efficacy but a three-fold decrease in the reaction rate, while 2-
methyltetrahydrofuran did not lead any decrease in the reac-
Finding the appropriate catalyst
We varied the conditions and determined the optimal ones for tion rate. Elevated temperature did not facilitate the reaction:
the reaction of 8-HQ with DIAD (Table 2). One of the principle the yield of the non-catalyzed reaction significantly dropped
restrictions in this search was not to use transition metal cata- while it was heated above 45 °C, and the starting material and
lysts, which would otherwise lead to a different mechanism of by-products prevailed in the reaction mixture. One would
the process. Furthermore, transition metal catalysts would form expect to see the opposite effect, bearing in mind the
products in a less environmentally friendly manner, requiring Arrhenius equation, but it can be speculated that 8-HQ mole-
additional chromatography to remove these catalysts.
cules are capable of stable self-dimerization, and intra- and
Based on previously known studies of azodicarboxylate intermolecular proton-transfer reactions become the major
ester substitution to form benzene hydrazines, we tried using processes in further molecule excitation.33 In this case, C–H
TFA and a combination of TFA and TfOH as catalysts.30 activation is suppressed, since the phenol part of the molecule
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6381–6389 | 6383

View Article Online
Paper
Green Chemistry
Table 2 Screening for reaction conditions
Yield of the major
quinolin-8-ol product (1a)
Yield of the major
phenol product (2a)
Conditionsa
Catalyst eq.
Time
Without a catalyst
Without a catalyst, >45 °C
TFA
NaH
NaH
NaH
NaOH
KHMDS
DBU
Cs2CO3
NEti-Pr2
LiH
—
—
5
1
0.05
<0.04
1
1.1
1.1
0.5
1.1
1
1
0.05
1
0.05
1
0.05
1
3 days
3 days
1.5 h
1.5 h
1.5 h
1.5 h
12 h
12 h
12 h
12 h
12 h
65%
Traces
5%
99%
99%
<90%
0%
70%
92%c
90%
32%
45%
90%b
87%b
60%b
60%b
20%b
90%b
87%c
80%b
60%
40%
24%
—
70%
90%
<70%
90%
63%
90%
77%
0%
95%
90%d
90%d
10%
5%
12 h
KH
KH
1.5 h
1.5 h
1.5 h
1.5 h
1.5 h
1.5 h
1.5 h
1.5 h
1.5 h
CaH2
CaH2
tBuOK
tBuOK
LiNH2
LiNH2
NBu4+OH−
90%
83%
95%
70%
15%
0.05
1
a Standard reaction conditions: 0.50 mmol of the reagent, 0.50 mmol of azodicarboxylate ester, in 2 ml of THF under an air atmosphere while stir-
ring at rt (if not stated otherwise). b +Bis-hydrazinated product and starting compound. c +Bis-hydrazinated product. d +Biphenyldiol.
Scheme 2 Synthesis of the bis-adduct of DIAD coupling to 8-HQ.
loses the proton in favor of the nitrogen of the neighboring
cycle.
To summarize, sodium hydride and lithium amide are two
environmentally friendly, cheap and affordable initiators for
this functionalization reaction. Although not officially claimed
as “green” reagents, they are involved in sufficiently clean reac-
tions and are used in small amounts. NaH and LiNH2 can be
easily quenched and isolated during the extraction of the
Scheme 3 The mechanism for azodicarboxylate esters to couple with
product, thus, reducing purification steps to a minimum. This
simplification cannot be achieved by the known methods
employing transition metal catalysts.
8-HQ under NaH-catalyzed conditions.
a catalytic amount of NaH induces phenolate formation. The
Mechanism discussion
charged quinoxide shows a higher +M effect for electron delo-
On the basis of these results and the related literature,18 calization to the o-position than the neutral OH group. Then
herein, we propose the following mechanism (Scheme 3). First, (step 2), the electrophilic carbon attacks the diazene molecule.
6384 | Green Chem., 2019, 21, 6381–6389
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
At this step the electron density migrates to the nitrogen atom no reaction of NaH/LiD with the azodicarboxylate ester is
from the –NvN– double bond. Further (step 3), a hydrogen observed at 20 °C in the 1H and 2H NMR spectra, as well as the
shift occurs and the aromaticity of the phenol ring is restored.
It is common knowledge that in electrophilic substitutions,
HRMS spectra.
We also conducted reactions to check whether a radical
the highest rate of substitution is achieved by the interaction mechanism takes place (or 1H+/1e− transfer) instead of the
of the most acidic equilibrium form of the electrophilic proposed ionic species. For this purpose, several test reactions
reagent (azodicarboxylate ester) and the most basic form of with and without base catalysts were carried out with additives
the nucleophilic substrate (phenolic substrate). Historically, to facilitate/quench hypothetical radical species. Irradiation
the first evidence for this was found in azo coupling reac- should facilitate radical generation. Therefore, we tried UV
tions,34 which are closely related to this topic.
lamp conditions (254 nm and 360 nm, 6 W, quartz test-tubes,
One might suggest a different explanation for the surpris- stirring for 6 hours) and incandescent lamp irradiation (150 W
ing selectivity of the NaH catalytic reaction, which is slightly and 300 W at 30 cm distance, in the closed and specially
superior to Cs2CO3 and DBU catalysis in the product outcome cooled reflecting tank, in quartz test-tubes, for 1 h), but the
(see Table 1). Diazene compounds are known to be good sub- yield of the catalyzed process did not increase compared to the
strates for cyclometallation reactions,35 and at the same time, non-catalyzed reaction. Darkroom conditions also did not
there is a class of nucleophilic substitution/addition reactions decrease the yield. Second, if the reaction follows a radical
of anionic N-sodio hydrazine derivatives.36 Hence, a different mechanism, such as SET or PCET, then the addition of radical
mechanism of coupling might have been assumed, involving scavengers or initiators would affect the yield and rate of the
the initial addition of NaH to the –NvN– bond and further reaction. We used TEMPO as an active catalyst for hydrogen
nucleophilic attack by the formed in situ hydrazinide deriva- transfer and as a radical scavenger, and found no sign of its
tive. However, it seems unlikely for the unstable hydrazinide influence on the yield or product ratio. 1,4-Dinitrobenzene was
intermediate to be formed at room temperature.
also tested as a SET inhibitor, but again, it did not affect the
Therefore, we suggest the electrophilic amination mecha- reaction. Thus, the possible SET (single electron transfer) and
nism instead. To clarify the issue, we conducted NMR experi- PCET (proton-coupled electron transfer) mechanisms were
1
2
ments: H and H NMR spectroscopy was conducted with par- excluded. The plausible reactive species are ions, and this can
tially deuterated substrates to assess the mechanism in situ be supported by the closely related reactions with phenol.
(Fig. 1). 8-Deuteroxyquinoline reacts with NaH (0.5 eq.)
forming the phenolate and deuterohydrogen. The reaction
between LiD and 8-HQ is hardly observed in this short time
period. That makes us consider the metallic nature of the Li–D
Replacing 8-hydroxyquinoline with phenol
It was interesting to explore this transition-metal-free approach
to obtain closely related structures. As shown in Table 2,
bond and the lower basicity (which is also required for acti-
phenol amination shows less selectivity towards base catalysts.
vation) to be diminishing factors for the reaction. As expected,
Both NaOH and NaH act equally well. The phenol OH group
acts as a directing group in this coupling.
The reaction of azodicarboxylate esters with phenol results
in para-substitution. This mechanism resembles the well-
known azo-coupling reaction, offering the same product regio-
selectivity, except that the azodicarboxylate ester and the
product possess one less π-bond (Scheme 4). Thus, the whole
conception of the reaction can be changed for the simple
general base catalysis, and the process itself can be regarded
as a coupling reaction. Using di(tert-butyl)azodicarboxylate (or
di-Boc-diazene) and degrading Boc-groups of the product, it is
possible to obtain 4-hydroxyphenylhydrazine in excellent yield
(Scheme 5).
Fig. 1 Deuterium NMR spectra (solvent: dioxane) from different steps
Scheme 4 The mechanisms for azodicarboxylate esters to couple with
of the possible activation mechanism.
phenol and for azo-coupling.
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6381–6389 | 6385

View Article Online
Paper
Green Chemistry
phenone.42 The drawbacks of the industrial route are that
nitration of chlorobenzene gives a mixture containing one-
third ortho-substituted products and two-thirds para-substi-
tuted products and the opposite ratio for the nitrosation of
phenol.
They are separated by careful fractionation through an
efficient column. The ecological impact of these industrial and
previously known laboratory methods is that they generate oxi-
dizers or free radicals as gas or liquid by-products, whereas
our new method generates non-hazardous by-product gases
during Boc-deprotection.
The proposed hydrazo coupling approach proves to be an
efficient gram-scale protocol that eliminates two steps of syn-
thesis with harsh reagents and doubles the overall yield. This
makes it perfect for laboratory synthesis and for screening in
medicinal chemistry. Boc-diazene is a commercially available,
shelf-stable and easy-to-handle solid. It would be a promising
addition to the well-established methods.
Scheme 5 4-Hydroxyphenylhydrazine synthesis.
Although 4-hydroxyphenylhydrazine has a history of more
than 120 years,37 not many methods have been described for
its synthesis. As a result, modern chemists seldom use this
compound as a starting reagent or during screening in medic-
inal chemistry (in contrast to phenyl- and 2,4-dinitrophenylhy-
drazine). Another issue is that currently, there are only a few
commercial suppliers for laboratory synthesis, selling at a
price of more than $800 per 1 gram.38 To compare, the same
amount of money spent on all the reagents (phenol, Di-tBu-
AD, LiNH2, BF3·Et2O and solvents) required for the synthesis
of 4-hydroxyphenylhydrazine described in the current work
would result in about 65 grams of product in a 1–2-day routine
procedure.38 To check whether the new method is reliable, we
scaled up the synthesis to a 5-gram PhOH loading and
achieved (for 2 steps) 90% yield with 0.05 eq. NaH or 95% with
an equimolar amount of LiNH2. Previously, the most reliable
laboratory methods involved diazobenzene formation via dia-
zotization, followed by hydrogenation (a well-established proto-
col in the Fischer indole synthesis), or substitution of halo-
derivatives with hydrazine.39
Several important medicines are synthesized from 4-hydro-
xyphenylhydrazine as an intermediate product, e.g. the well-
known non-steroidal anti-inflammatory drug (NSAID) indome-
tacin,40 which has now been used for 50 years in practice and
has inspired drug design towards a new triptan class of
NSAIDs. 4-Hydroxyphenylhydrazine is a key intermediate in
the synthesis of a new and still expensive drug, bazedoxifene,
the latest generation selective estrogen receptor modulator (by
Pfizer). It is prepared in 59% yield, thus being the lowest-yield
product in the total synthesis of bazedoxifene.41 In industry,
4-hydroxyphenylhydrazine is synthesized from 4-aminophenol,
a precursor of antipyretic agents such as paracetamol (acet-
aminophen), acetanilide and phenacetin. Nowadays, bulk
4-aminophenol for industrial paracetamol synthesis is pro-
duced through different manufacturing routes requiring
mineral acids: (1) from 4-nitrochlorobenzene; (2) from phenol;
(3) through the electrolytic reduction of nitrobenzene (also,
catalytic hydrogenation is possible); (4) from 4-hydroxyaceto-
Scope of the reaction
To explore the scope of the reaction, we investigated the possi-
bility of selective functionalization of substituted quinolines,
8-HQ N-oxide, α- and β-naphthol (Table 3). 8-HQ N-oxide can
be regarded as a masked form of 8-HQ, which can be easily
reduced back to the 8-HQ moiety at any stage via enzymatic
reduction by oxidoreductases (e.g., from baker’s yeast43).
N-Oxide is capable of the same coupling reaction as 8-HQ.
However, we obtained the same product 1c through another
pathway, oxidizing 1a with m-CPBA, in 57% yield. The oxi-
dation is feasible at the N1 atom regardless of the 7-hydrazinyl
substituent and this N-oxide can be used for further
functionalization at the 2nd position.44 5-Substituted 8-hydro-
xyquinolines with EDG react smoothly in the same way as
8-HQ, although some by-products are formed as well.
It is worth noting that 5-fluoro-8-HQ barely reacts with AD
esters without NaH catalysis. We investigated the H NMR and
1
19F NMR spectra during this reaction in the NMR tube in
dioxane-d8, with and without 1 eq. NaH, without stirring of the
heterogeneous reaction mixture. We observed an increase of
several orders of magnitude in the rate of the catalytic process.
In the base-initiated reaction, the product yield reached ∼50%
in 6 h and conversion continued to increase linearly, while
non-catalyzed conditions gave barely 2% of the product in 12 h.
α-Naphthol is substituted at the para-position during the
reaction, if the para-position is not occupied by another substi-
tuent. In the case of β-naphthol, the coupling takes place pre-
sumably both at the α- and γ-positions, and the conversion is
drastically lower. Quinolines without an 8-OH group do not
interact with DIAD, because there is no directing group in
their structure.
Even with several equivalents of NaH added as an initiator
or other bases, 8-hydroxyquinoline-5-sulfonic acid barely reacts
with ADs, with insufficient traces of the product (detected by
NMR), and 5-nitro-8-HQ does not react at all. These two EWG-
substituted hydroxyquinolines possess strong deactivating
groups (–I, –M), which decrease the nucleophilicity of the aro-
6386 | Green Chem., 2019, 21, 6381–6389
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
Table 3 The substrate scope for substituted substrates. The reaction of
arenes with DIAD or Di-tBu-AD in the presence of 5 mol% NaH in THF
Hydrazone synthesis
One of the most efficient and fastest ways of obtaining 7-hydra-
zino-8-hydroxyquinoline derivatives is the synthesis of hydra-
zones. Diethyl or diisopropyl hydrazinocarboxylate esters are
stable under highly acidic or basic conditions, even when
heated to 80 °C, and the only previously reported achievable
transformation involves the reduction to amines; however, we
could not achieve the product following the described
method47 with 1a. At the same time, the di-tert-butyl ester of
azodicarboxylate (i.e., Boc-diazene) is suitable for carboxylate
elimination. The 7-(N,N′-di-Boc-hydrazino)8-HQ derivative 1b
has been readily hydrolyzed in 3 M HCl solution, giving a
maroon-colored precipitate of 7-hydrazino-8-hydroxyquinoline
4 (Scheme 6). Surprisingly, even after acid neutralization, fil-
tration and complete drying, this compound remains unreac-
tive with aldehydes, possibly owing to the formation of hydro-
chloride salt obtained during the reaction.
Conversion
(obtained
yield)
Reagent
Product
0%
—
99% (83%)
95% (70%)
97% (97%)
90% (80%)
We tested some additives described earlier for facilitating
hydrazone synthesis, such as anthranilic acid derivatives.48 We
could not bypass this problem of unreactive 4 except for to use
another method of Boc cleavage for 1b. Oxidative methods,
such as treatment with CAN, were unsuitable and resulted in
quinoline residue destruction and self-condensation.49 We also
tried KI-catalyzed basic conditions50 that appeared to be incap-
able of the cleavage. Finally, the most convenient and steady
Boc degradation was achieved with BF3·Et2O.51 This procedure
allows us not only to obtain the desired reactive 4 but also to
use it in situ with initially added aldehydes and to make this
reaction a “one-pot” process. The co-products formed by Boc
and boron trifluoride etherate are totally evaporated in vacuo, so
there is no need for laborious isolation procedure.
86% (84%)
99% (95%)
60% (a)
0%
A mixture of α- and
γ-hydrazinyldicarboxylates
of β-naphthol
—
The synthesized hydrazones possess great potential for che-
lation of metals with different selectivities depending on the
substituent.
97% (93%)
92% (90%)
95% (88%)
Isolation issues with 8-HQ derivatives
One of the major difficulties while working with 8-HQ deriva-
tives is their isolation or purification with chromatography.
a β-Naphthol products were not isolated and were identified only by
HPLC-HRMS.
matic ring. 2-Nitroso-1-naphthol is also unreactive with its
prevalent –M effect of the nitroso group. All this clearly sup-
ports the proposed mechanism.
Thiophenol and 8-mercaptoquinoline oxidation to disulfides
Thio-analogues of phenol and 8-hydroxyquinoline do not form
hydrazo coupling products; instead, they tend to be almost
quantitatively oxidized by ADs to disulfides. This phenomenon
is a known process for related classes, such as diazenecarboxa-
mides45 and phenylazocarboxylate esters,46 but has not been
associated with symmetrical AD esters yet.
Scheme 6 Synthesis of hydrazones.
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6381–6389 | 6387

View Article Online
Paper
Green Chemistry
The compounds show vast “tailing” on columns and TLC pounds. In addition to the practical application of this method
plates in the common eluting systems consisting of DCM, for laboratory use, 4-hydroxyphenylhydrazine, an industrial
ethanol, EtOAc or hexane due to the zwitterionic nature of the intermediate product for total synthesis in medicinal chem-
N1 – 8-OH motif. Not only are the products discussed in this istry, has been obtained through a straightforward and fast
manuscript affected, but the whole class of 8-HQ derivatives gram-scale protocol that allows the elimination of two steps
possesses this problem, lowering the yield of the final products used in industrial synthesis with harsh reagents, while also
and exacerbating the separation procedure. We thoroughly doubling the overall yield.
adjusted the eluting system and found that the crucial additive
was 1% AcOH; as a result, CCl4–dioxane (4 : 1) + 1% AcOH
appeared to be the optimal system for our derivatives. In the
case of the hydrazine derivatives 5a–f, we found that the pro-
Conflicts of interest
ducts are insoluble in DCM. To make product isolation easier, There are no conflicts to declare.
the excess aldehyde in the reaction mass can be first washed
out through the filter in chloroalkanes, and then, the sole
clear product can be washed with ethanol.
Acknowledgements
There is another approach to solve the problem: masking
the OH group in 1b prevents peak tailing in chromatography. This work was supported by the Russian Science Foundation
For this purpose, we tried three protective groups. Benzoyl and (RSF grant No. 18-74-00094). The recording of NMR and HRMS
silyl ester protection proceeded smoothly, but the following de- spectra was supported by the Program of fundamental
protection of derivatives 5g–i revealed some peculiarities of research for state academies for 2013–2020 (No. 01201363818).
stability. [Si]-cleavage with TBAF was less effective even with
excess reagent, and at the same time the [Si]–F by-products
were difficult to separate from the target products. Inorganic
References
fluorides were among the best in this cleavage reaction:52 we
tried KHF2 and found that the products were partially affected,
but ammonium fluoride appeared to be better (and it was
almost insoluble in acetone, making it easy to isolate the
excess reagent). Basic conditions (Cs2CO3 with imidazole, or
NaH) did not lead to the cleavage of the [Si] groups from the
8-HQ moiety,53 whereas benzoyl protection led to extremely
stable derivatives 1k and 5i, which were not affected in 3 M
NaOH solution or even 33% ammonia. This surprising effect
may be due to the intermolecular hydrogen bonding of the
7-hydrazinyl fragment.
1 P. Wedi and M. van Gemmeren, Angew. Chem., Int. Ed.,
2018, 57(40), 13016.
2 R. Huisgen, F. Jakob, W. Siegel and A. Cadus, Justus Liebigs
Ann. Chem., 1954, 590(1), 1; L. Horner and W. Naumann,
Justus Liebigs Ann. Chem., 1954, 587(2), 81.
3 J. Yadav, B. Reddy, G. Veerendhar, R. Srinivasa and
K. Nagaiah, Chem. Lett., 2002, 31(3), 318.
4 S. M. Inamdar, V. K. More and S. K. Mandal, Tetrahedron
Lett., 2013, 54(6), 530.
5 L. Gu, B. S. Neo and Y. Zhang, Org. Lett., 2011, 13(7), 1872.
6 F. Ma, Z. Peng, W. Li, X. Xie and Z. Zhang, Synlett, 2011,
(17), 2555.
7 I. Yavari, M. Ghazanfarpour-Darjani, Y. Solgi and
S. Ahmadian, Synlett, 2011, (12), 1745.
Conclusions
A new reaction involving the catalyzed C–H functionalization
of 8-hydroxyquinoline, phenol and α-naphthol derivatives has
been developed. It can be carried out under mild conditions
and room temperature using commercially available in-
expensive materials: NaH as an initiator, and THF or dioxane
8 S. Bombek, R. Lenaršič, M. Kočevar, L. Saint-Jalmes,
J. Desmursd and S. Polanc, Chem. Commun., 2002, (14),
1494.
9 K. Muñiz and A. Iglesias, Angew. Chem., Int. Ed., 2007,
46(33), 6350.
as a solvent. We have also proposed and confirmed the mecha- 10 T. Uemura and N. Chatani, J. Org. Chem., 2005, 70(21), 8631.
nism of this reaction, backing up our suggestions with NMR 11 I. Yavari, M. Ghazanfarpour-Darjani, M. Bayat and
data. The participation of the phenolate as a directing group
results in excellent yields of the coupling reaction, and the 12 M. Khalaj and M. Ghazanfarpour-Darjani, RSC Adv., 2015,
transformation can be regarded as “hydrazo coupling” (by 5, 80698.
analogy with azo coupling). Moreover, Boc-protected 7-hydra- 13 Y. Leblanc, R. Zamboni and M. A. Bernstein, J. Org. Chem.,
zino-8-hydroxyquinoline was then employed in the synthesis of 1991, 56, 1971.
aromatic hydrazones using a novel efficient “one-pot” protocol. 14 S. Narayan, J. Muldoon, M. Finn, V. Fokin, H. Kolb and
Several approaches were investigated to overcome technical K. Sharpless, Angew. Chem., Int. Ed., 2005, 44, 3275.
A. Malekafzali, Synlett, 2015, 26(07), 942.
problems in the chromatography procedures of 8-hydroxyqui- 15 S. H. Schroeter, J. Org. Chem., 1969, 34(12), 4012.
noline compounds as zwitterionic structures. To summarize, 16 E. Müller, R. Mayer, B. Narr, A. Rieker and K. Scheffler,
we have synthesized previously undescribed products bearing
Justus Liebigs Ann. Chem., 1961, 645(1), 25.
potential as new building blocks and biologically active com- 17 Y. Leblanc and N. Boudreault, J. Org. Chem., 1995, 60, 4268.
6388 | Green Chem., 2019, 21, 6381–6389
This journal is © The Royal Society of Chemistry 2019

View Article Online
Green Chemistry
Paper
18 W. Kinart and C. Kinart, J. Organomet. Chem., 2006, 691(8),
1441.
19 S. Brandes, M. Bella, A. Kjærsgaard and K. Jørgensen,
Angew. Chem., Int. Ed., 2006, 45(7), 1147.
product-detail/?H950165, Alichem (USA), https://www.
alichem.com/en/products/23197-23-1.html,
vendors found on the reference from https://pubchem.ncbi.
nlm.nih.gov/compound/4-Hydrazinylphenol and web
and
other
20 H. Zhu, S. Sun, H. Qiao, F. Yang, J. Kang, Y. Wu and Y. Wu,
Org. Lett., 2018, 20(3), 620.
21 Z. S. Almalki, X. Yue, Y. Xia, P. R. Wigle and J. J. Guo,
PharmacoEconomics, 2017, 1(2), 123.
22 P. Zhan, C. Pannecouque, E. De Clercq and X. Liu, J. Med.
Chem., 2016, 59(7), 2849.
23 F. Prati, C. Bergamini, R. Fato, O. Soukup, J. Korabecny,
search engines. Prices for other readily available reagents
were checked on “Sigma-Aldrich” website at the date of the
paper submission.
39 J. Chen, Y. Zhang, W. Hao, R. Zhang and F. Yi, Tetrahedron,
2013, 69, 613.
40 A. R. Maguire, S. J. Plunkett, S. Papot, M. Clynes, R. O’Connor
and S. Touhey, Bioorg. Med. Chem., 2001, 9(3), 745.
V. Andrisano, M. Bartolini and M. L. Bolognesi, 41 H. S. Kwan, P. C. Ha, S. R. Seok, C. S. Hee, K. K. Deok and
ChemMedChem, 2016, 11(12), 1284.
K. D. Hwa, Methods of preparing Bazoedoxifene, KR10-
2018-0095239(A), 2017; H. Lizhi, Preparation method of
selective estrogen receptor modulator bazedoxifene and key
intermediate thereof, CN107935906(A), 2018.
24 C. C. Thinnes, A. Tumber, C. Yapp, G. Scozzafava, T. Yeh,
M. C. Chan, T. A. Tran, K. Hsu, H. Tarhonskaya,
L. J. Walport, S. E. Wilkins, E. D. Martinez, S. Müller,
C. W. Pugh, P. J. Ratcliffe, P. E. Brennan, A. Kawamura and 42 On the electrochemical reduction of aromatic nitro com-
C. J. Schofield, Chem. Commun., 2015, 51, 15458.
25 N. Muruganantham, R. Sivakumar, N. Anbalagan,
V. Gunasekaran and J. T. Leonard, Biol. Pharm. Bull., 2004,
27(10), 1683; L. E. Da Silva, P. T. De Sousa Jr. and
A. C. Joussef, Synth. Commun., 2009, 39(8), 1378.
26 D. R. Motati, D. Uredia and E. B. Watkins, Chem. Sci., 2018,
9(7), 1782.
27 H. Gershon, D. Clarke and M. Gershon, Trends Heterocycl.
Chem., 1995, 4, 1.
28 R. E. Hartung, M. C. Wall, S. Lebreton, M. Smrcina and
M. Patek, Heterocycles, 2017, 94(7), 1305.
pounds: S. H. Cadle, P. R. Tice and J. Q. Chambers, J. Phys.
Chem., 1967, 71(11), 3517; Y. Sun, W. Xu and K. Scott,
Electrochim. Acta, 1993, 38(13), 1753. On the catalytic
hydration of p-nitrochlorobenzene: C. V. Rode, M. J. Vaidya
and R. V. Chaudhari, Org. Process Res. Dev., 1999, 3(6), 465.
Nitrosation route: P. J. Hermsen and P. Poechlauer,
Method of nitrosation of phenol, EP2975018A1, 2014.
Fractionation of the products: H. E. Fierz-David and
L. Blangey, in Fundamental Processes Of Dye Chemistry,
Hardpress Publishing, Miami, 2012. Details on all of the
methods: R. Joncour, N. Duguet, E. Métay, A. Ferreira and
M. Lemaire, Green Chem., 2014, 16(6), 2997.
29 M. Girard, P. Murphy and N. N. Tsou, Tetrahedron Lett.,
2005, 46(14), 2449.
30 K. Y. Lee, Y. J. Im, T. H. Kim and J. N. Kim, Bull. Korean
Chem. Soc., 2001, 22(2), 131.
43 W. Baik, D. I. Kim, S. Koo, J. U. Rhee, S. H. Shin and
B. H. Kim, Tetrahedron Lett., 1997, 38(5), 845.
44 T. Storz, R. Marti, R. Meier, P. Nury, M. Roeder and
K. Zhang, Org. Process Res. Dev., 2004, 8(4), 663.
31 The commercially available price was calculated according
to the online resources: https://www.chemicalbook.com/ 45 J. Košmrlj, M. Kočevar and S. Polanc, J. Chem. Soc., Perkin
Price/Sodium-hydride.htm and https://www.chemicalbook.
com/Price/Lithium-amide.htm.
32 X. Xu and X. Li, Org. Lett., 2009, 11(4), 1027.
33 E. Bardez, I. Devol, B. Larrey and B. Valeur, J. Phys. Chem.
Trans. 1, 1998, 3917.
46 E. M. Kosower and N. S. Kosower, Nature, 1969, 224, 117.
47 A. Shamsabadi, J. Rena and V. Chudasama, RSC Adv., 2017,
7, 27608.
B, 1997, 101, 7786; C. M. Cheatum, M. M. Heckscher and 48 P. Crisalli and E. T. Kool, J. Org. Chem., 2013, 78(3), 1184.
F. F. Crim, Chem. Phys. Lett., 2001, 349(1–2), 37.
49 J. R. Hwu, M. L. Jain, S. Tsay and G. H. Hakimelahi,
34 I. Szele and H. Zollinger, in Preparative Organic Chemistry.
Tetrahedron Lett., 1996, 37(12), 2035.
Topics in Current Chemistry, Springer, Berlin, Heidelberg, 50 A. Mastitski, A. Abramov, A. Kruve and J. Järv, Proc. Est.
1983, vol. 112.
Acad. Sci., 2017, 66(1), 10.
35 M. I. Bruce, Angew. Chem., Int. Ed. Engl., 1977, 16(2), 73.
36 T. Kauffmann, C. Kosel and D. Wolf, Chem. Ber., 1962,
51 E. F. Evans and N. J. Lewis, Synth. Commun., 1997, 27(11),
1819.
95(6), 1540; T. Kauffmann, K. Lötzsch, E. Rauch and 52 M. K. Lakshman, F. A. Tine, T. A. Khandaker, V. Basava,
W. Schoeneck, Chem. Ber., 1965, 98, 904.
37 J. Altschul, J. Prakt. Chem., 1897, 57(1), 201.
38 4-Hydroxyphenylhydrazine price was declared as
N. B. Agyemang, M. S. A. Benavidez, M. Gaši, L. Guerrera
and B. Zajc, Synlett, 2017, 28(3), 381.
53 R. A. Fernandes, S. P. Gholapa and S. V. Mulay, RSC Adv.,
2014, 4, 16438; D. K. Mohapatra and K. A. Durugkar,
ARKIVOC, 2005, 14, 20.
a
minimum per 1 gram among the suppliers: Toronto
research chemicals (Canada), https://www.trc-canada.com/
This journal is © The Royal Society of Chemistry 2019
Green Chem., 2019, 21, 6381–6389 | 6389