Continuous niobium phosphate catalysed Skraup reaction for quinoline synthesis from solketal

Article abstract of DOI:10.1039/c6gc03140d

Solketal is derived from the reaction of acetone with glycerol, a by-product of the biodiesel industry. We report here the continuous reaction of solketal with anilines over a solid acid niobium phosphate (NbP), for the continuous generation of quinolines in the well-established Skraup reaction. This study shows that NbP can catalyse all the stages of this multistep reaction at 250 °C and 10 MPa pressure, with a selectivity for quinoline of up to 60%. We found that the catalyst eventually deactivates, most probably via a combination of coking and reduction processes but nevertheless we show the promise of this approach. We demonstrate here the application of our approach to synthesize both mono- and bis-quinolines from the commodity chemical, 4,4′-methylenedianiline.

Full text of DOI:10.1039/c6gc03140d

Green Chemistry
View Article Online
View Journal
PAPER
Continuous niobium phosphate catalysed Skraup
reaction for quinoline synthesis from solketal†
Cite this: DOI: 10.1039/c6gc03140d
b
Jing Jin,a Sandro Guidi,a,b Zahra Abada, a Zacharias Amara,a,c Maurizio Selva,
Michael W. Georgea,d and Martyn Poliakoff*a
Solketal is derived from the reaction of acetone with glycerol, a by-product of the biodiesel industry. We
report here the continuous reaction of solketal with anilines over a solid acid niobium phosphate (NbP),
for the continuous generation of quinolines in the well-established Skraup reaction. This study shows that
NbP can catalyse all the stages of this multistep reaction at 250 °C and 10 MPa pressure, with a selectivity
for quinoline of up to 60%. We found that the catalyst eventually deactivates, most probably via a combi-
nation of coking and reduction processes but nevertheless we show the promise of this approach. We
demonstrate here the application of our approach to synthesize both mono- and bis-quinolines from the
commodity chemical, 4,4’-methylenedianiline.
Received 14th November 2016,
Accepted 6th April 2017
DOI: 10.1039/c6gc03140d
rsc.li/greenchem
an important class of hetero-aromatic compounds with large
scale industrial applications.12
Introduction
Significant advances in the production of fuels, fine chemi-
The Skraup reaction is relatively old; it was first described13
cals, and highly valuable products from renewable feedstocks in 1880 but remains somewhat under-exploited. It relies on the
have been achieved over the past decade.1 Biomass has dehydration of glycerol to form acrolein, a highly reactive inter-
emerged as one of the most important sustainable sources for mediate, which subsequently condenses with aniline.
renewable organic carbon in line with the Principles of Green Typically, it requires stoichiometric quantities of harsh and/or
Chemistry.2,3 In this context, renewable energy sources derived toxic soluble reagents such as concentrated sulfuric acid, nitro-
from biomass such as biodiesels, offer opportunities to access benzene and a range of metal salts and strong oxidants
glycerol, an inexpensive by-product representing 10 wt% of the (Scheme 1, top).14
overall production.4,5 The current increased production of bio-
More recently, a range of catalytic routes have been deve-
diesel generates an ever greater abundance of glycerol to the loped; however, these procedures generate considerable
point where supply exceeds demand.6
amounts of waste and have to be performed under relatively
The possibility of using glycerol as a precursor to other
chemicals has therefore attracted increasing attention in
recent years.7,8 The chemistry of glycerol is widely developed
as summarised in a number of major reviews.9–11 Among the
varied transformations, the Skraup reaction stands out as one
of the few that enables the direct transformation of glycerol
into heteroaromatic compounds such as quinolines, which are
aSchool of Chemistry, University of Nottingham, University Park,
NG7 2RD Nottingham, UK. E-mail: martyn.poliakoff@nottingham.ac.uk
bDepartment of Molecular Sciences and Nanosystems, Università Ca’ Foscari Venezia
Via Torino 155, 30172 Venezia-Mestre, Italy
cLaboratoire Chimie Moléculaire Génie des procédés chimiques et énergétiques,
CNAM, 2 rue Conté, 75003 Paris, France
dDepartment of Chemical and Environmental Engineering, University of Nottingham
Ningbo China, 199 Taikang East Road, Ningbo 315100, China
†Electronic supplementary information (ESI) available.CCDC 1517275. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c6gc03140d
Scheme 1 Original Skraup reaction13 and the continuous flow Skraup
reaction of solketal 1b and anilines 2 over niobium phosphate (NbP)
described in this paper.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
harsh conditions, involving tedious and lengthy workup.15–18
Given the previous reports43,44 of the dehydration of glycerol
The large quantities of waste, harsh reagents and conditions and other biomass derived compounds45–47 using niobium
mean that this reaction is often overlooked as a viable route based catalysts, we have investigated how these catalysts might
despite one of the main reagents being glycerol. Even though be used in a continuous flow system to develop a Skraup reac-
modified conditions have been developed,19,20 the use of an tion that provides a greener, cleaner and continuous synthesis
oxidant and corrosive acid or expensive ionic liquids21 remains of quinoline-derived products (Scheme 1, bottom).
necessary.
From a green point of view, routes such as the Skraup reac-
tions, which involve homogenous acids, are associated with
the problems of toxicity, corrosion, use of stoichiometric quan-
Results and discussion
tities, separation and recovery. Heterogeneous catalysts poten- The small scale of our equipment means that pure glycerol is
tially offer a greener alternative as they can be easily separated too viscous to be pumped through the narrow-bore of the pipe-
at the end of a reaction and can often be recycled, reducing work. Therefore, glycerol needs to be diluted with a solvent to
the environmental cost of a chemical process. Furthermore, reduce its viscosity. Unfortunately the choice of solvent is
solid acid catalysts can combine the aspects of both Brønsted limited. Thus, the starting point was to react glycerol 1a and
and Lewis acidity offering a versatile alternative to homo- aniline 2a with water as the solvent. In addition to its obvious
geneous acids, particularly in terms of acid strength, number lack of toxicity, water allows a high concentration of glycerol to
of active sites and support morphology (e.g. surface area and be used and it is also compatible with niobium catalysts.48
porosity). Several heterogeneous catalytic systems have already However, whilst glycerol is soluble in water, aniline on the
shown to be active in the dehydration of glycerol to acrolein; other hand has only limited miscibility to a maximum of
these include zeolites,22,23 heteropolyacids,24–27 and metal ca. 0.4 M.49 The highest yield of quinoline 3a that could be
oxides.28,29
achieved was only 10% despite our testing of a range of para-
In this report we use niobium phosphate (NbP) as a hetero- meters (temperature, pressure, flow rate, and reagent stoichio-
geneous acid catalyst that exhibits strong surface acidity and metry). Even though the yield was low, it was an encouraging
has been successfully used in the past to catalyse several reac- proof of concept. The remaining reaction products were often
tions,30 including the dehydration of several bio-derived com- a complicated mixture. We have reported in our previous work
pounds such as xylose,31 fructose,32 cellulose33 and gly- that aniline reacts with polar solvents, such as small alcohols,
cerol.28,34 NbP has acidic catalytic properties similar to in the presence of NbP resulting in alkylation reactions.30
niobium oxide.35,36,40 However, its acidic properties are more Since glycerol is only soluble in water or very polar protic sol-
stable than those of the oxide at higher temperatures37,38 and vents, we decided to switch to an alternative reagent, solketal
frequently NbP limits side reactions as a result of additional 1b, which is the isopropylidene acetal-protected form of gly-
Brønsted sites.39 Moreover, NbP has been demonstrated to be cerol. It is produced by the condensation of acetone with gly-
an effective catalyst for reactions in which water molecules are cerol.50 As two of its hydroxyl groups are protected, solketal is
present or generated.40
Inorganic solid acid catalysts are generally highly insoluble a wider range of solvents.
less prone to hydrogen bonding than glycerol and is soluble in
in organic solvents which make them suitable for applications
We envisaged that reaction with the acidic NbP would
in continuous flow processes. Tubular reactors can be easily deprotect 1b to regenerate glycerol allowing its subsequent de-
filled with a solid acid to make a fixed-bed system that can be hydration to acrolein. This should allow the Skraup reaction to
coupled to a flow setup allowing the continuous catalytic pro- proceed in a slower and more controlled manner (Scheme 2).
cessing of reagents. Such reactors provide enhanced heat and It was anticipated that this approach might avoid the large
mass transfer while offering opportunities for optimization quantities of by-products observed in our preliminary experi-
and therefore simpler scale-up and greater safety.41 ments. With a larger choice of solvents available, the problem
Furthermore, enhanced reaction rates are often observed of poor aniline solubility could also be overcome. To minimize
because of high local concentrations.42
potential alkylation reactions protic solvents were avoided and
Scheme 2 The multiple roles of niobium phosphate in the synthesis of quinoline 3a from solketal 1b. The three-step transformation is envisaged as
a single continuous process.
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
Table 1 Physical properties of the reagents and products involved in
the reaction
a more inert solvent was chosen; we ultimately settled on
toluene as the solvent for the reaction.
Our strategy has been to investigate the reaction of solketal
with aniline to identify the best conditions in terms of conver-
sion and selectivity as a function of pressure, temperature,
reactor volume and ratio of aniline to solketal. Then we
applied the reaction to two other anilines 3-methoxyaniline
(MA) and 4,4′-methylenedianiline (MDA). Of course, the use of
solketal represents unnecessary derivatization (in contradic-
tion to Green Chemistry Principle No. 8) but on a larger scale
the viscosity of neat glycerol would be less of a problem and its
replacement by solketal would not be required. However, solke-
tal does allow these continuous reactions to be explored on a
laboratory scale.
Compounds
Boiling point (°C)
Tc (°C)
Pc (MPa)
Toluene
Glycerol
Solketal
Acrolein
110.6
290.0
188.0–190.0
53.0
184.2
246.0
398.0
237.0
265.5
318.6
492.2
—
254.0
425.6
—
—
—
—
4.11
4.31
—
5.07
5.31
—
—
—
—
Aniline
4-Methoxyanilinea
4,4′-Methylenedianilineb
Quinoline 3a
3-methylindole 3b
a Melting point 57 °C. b Melting point 89 °C.
Solketal 1b and aniline 2a were delivered via independent
HPLC pumps; the reactor was filled with NbP and different
parameters were explored to optimise the reaction. The initial
temperature was 250 °C as this had previously been identified
as the best for the dehydration step of glycerol.51 The reaction
was carried out at low pressure with a ratio of aniline to solke-
tal (2a : 1b) of 1 : 3 to allow for the possible loss of acrolein by
polymerisation. The flow rate for solutions of both 2a and 1b
Table 2 Influence of the pressure on the conversion of aniline, and
selectivity and productivity of quinoline 3a over NbP
Pressure
1 MPa
5 MPa
7.5 MPa
10 MPa
Aniline conversiona
Quinoline selectivitya
71%
44%
90%
53%
90%
49%
80%
57%
Quinoline productivity 0.067 g h−1 0.1 g h−1 0.093 g h−1 0.1 g h−1
a The conversion, selectivity and productivity were averaged between
was set at 0.1 mL min−1
.
2
nd and 10th hours and calculated based on GC analysis. Reaction con-
ditions were 2a : 1b ratio of 1 : 3 and 0.2 ml min−1 total flow rate,
The results are shown in Fig. 1 and it is clear that (i) the
conversion of aniline, as measured by GC, decreased very
rapidly with time, at a pressure of 0.1 MPa; (ii) as the pressure
was increased the rate of drop off was slower; (iii) increasing
the pressure above 5 MPa had no further effect; (iv) the selecti-
vity for quinoline 3 was never above 60% and reduced over
time; (v) the reduction in selectivity appears to be greater at
lower pressures.
250 °C at 10 MPa.
The major by-product was isolated by column chromato-
graphy and was identified via NMR and mass spectrometry as
3-methylindole 3b. Its formation can be understood as arising
from the reaction of hydroxyacetone with aniline (Scheme 3).54
Hydroxyacetone has previously been reported55 as being
formed from acrolein in the presence of Lewis acid sites.56,57
Over Lewis acid sites, coordination of the glycerol intermediate
takes place, leading to the removal of water from a primary OH
group of glycerol as reported previously.58 This results in an
unstable enol intermediate, which undergoes rapid tautomeri-
sation to hydroxyacetone. Similar mechanisms have been
observed over La2CuO4 as a catalyst explaining the occurrence
of hydroxyacetone 1d.55 Once glycerol is converted to hydroxya-
cetone, the latter is transformed to 2-hydroxy-1-propanal via a
The effect of pressure in fact can be rationalised, at least
qualitatively, by looking at the physical data in toluene
(Table 1).
Apart from glycerol, all the reaction components will be in
the vapour phase at 250 °C and a pressure of 0.1 MPa.
Increasing the pressure will create a two-phase system and
most probably the entire reactor is filled with liquid at
pressure >5 MPa.52,53 This switch from the vapour to the liquid
phase will give higher volumetric concentrations of reactants
and longer residence times for a given flow rate of reactants
into the system (Table 2).54
Fig. 1 Effect of the system pressure on the (a) aniline conversion and (b) quinoline selectivity over NbP. The different experiments were performed
under the following conditions: 250 °C, 0.2 mL min−1 of total flow rate for a 1 : 3 ratio of 2a : 1b in toluene.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
Scheme 3 Main by-product observed from the continuous flow reaction of solketal 1b and aniline 2a over niobium phosphate (NbP). (a) Synthesis
of quinoline 3a through acrolein 1c over Brønsted acid sites and (b) synthesis of 3-methylindole 3b through the hydroxyacetone 1d intermediate
over Lewis acid sites.
Fig. 2 Plots illustrating that the ratio reagent has relatively little effect on the (a) conversion of aniline and (b) quinoline selectivity over NbP. The
different experiments were performed under the following conditions: 250 °C, NbP catalyst, 0.2 mL min−1 of total flow rate and 10 MPa for 1 : 1
(blue), 1 : 2 (red), 1 : 3 (green) and 2 : 1 (yellow) ratios of 2a : 1b.
keto–enol tautomerism. The aldehyde undergoes condensation lower amount of solketal has been fed (ratio 2 : 1) to minimize
with aniline to form a α-hydroxyimine intermediate. Then, the acrolein degradation with a result similar to the stoichiometric
loss of a second molecule of water generates a secondary car- ratio for both conversion and selectivity over 15 hours.
bocation able to trigger a ring closure which provides the final
indole derivative.
The effect of temperature was more striking. As shown in
Fig. 3, the conversion of aniline increases with temperature
The effect of increasing the ratio of 2a : 1b from 1 : 1 to 1 : 3 from 53% at 225 °C up to 78% at 275 °C and, although not
is shown in Fig. 2 and Table 3. As expected, increasing the illustrated, it increases to 85%, at 300 °C (Table 4). Selectivity
ratio has a positive effect on the conversion of aniline. however drops to 20–30% at 275 °C and only 10% at 300 °C.
However, after a few hours, there is very little difference in the This observation is possibly due to the different effects of
selectivity for quinoline 3a between the different experiments. temperature on the relative rates of generation of acrolein from
By contrast, the lifetime of the catalyst appears to be reduced solketal and its subsequent polymerization.
at higher ratios, presumably because of the coking process via
A modest scale-up of the reaction was attempted using
the polymerisation of acrolein. The higher concentration of three reactors filled with NbP in series with a fourth reactor
water, coproduced with acrolein may also have an effect. A placed upstream and filled with sand to act as a pre-mixer (Rb
– 20 cm length, Ø = 1/4″ each and ∼4.5 mL) as described in the
Experimental section and in previous work.59 The flow rate
was initially set at the same value (0.1 mL min−1) of each
reagent. The triple reactor ran for 60 hours before any fall off
Table 3 Influence of the reagent ratio effect on the conversion of
aniline, and the selectivity and productivity of quinoline 3a over NbP
in conversion was detected. By comparison, the single reactor
showed significant fall off in less than 20 hours (one reactor
2a : 1b ratio
1 : 1
1 : 2
1 : 3
2 : 1
Ra, 0.1 mL min−1). When the flow rate through the triple
reactor was increased by a factor of three, its temporal behav-
iour was similar to that of the single reactor; indeed the fall off
in both conversion and selectivity was more dramatic (Fig. 4).
This suggests that there may be a relationship between the
amount of substrate passing through the reactor and the active
lifetime of the NbP (Table 5).
Aniline conversiona
63%
73%
52%
72%
53%
61%
52
Quinoline selectivitya 51%
Quinoline productivity 0.068 g h−1 0.08 g h−1 0.08 g h−1 0.065 g h−1
a The conversion, selectivity and productivity were averaged values of
the reaction from 2nd to 15th hours and calculated based on GC ana-
lysis. Reaction conditions were 2a : 1b ratio of 1 : 1 and 0.2 ml min−1
total flow rate, 250 °C at 10 MPa.
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
Fig. 3 Effect of the temperature on (a) aniline conversion and (b) quinoline selectivity of four sets of experiments were carried out isothermally at
225, 250 and 275 °C at 10 MPa with a constant reagent 2a : 1b ratio of 1 : 1. Two 0.275 M solutions of solketal and aniline were set to a total flow rate
of 0.2 mL min−1. The dotted lines are purely to aid visualization.
Table 4 Aniline conversion, quinoline selectivity and productivity
measured as a function of temperature
catalytic steps (i.e. the ketal deprotection and the condensation
of acrolein with aniline), the overall yield will be low.
Two of the steps in Scheme 4 might be causing the decline
Temperature 200 °C
225 °C
53%
250 °C
64%
275 °C
78%
300 °C
85%
in catalytic steps. The first is coking caused by polymerization
of the free acrolein. The other is the final oxidation step
leading to the formation of 3a. The literature suggests that the
acidity61 strength of the most selective catalysts for the de-
Aniline
45%
16%
conversiona
Quinoline
selectivitya
Quinoline
productivity
42%
52%
24%
10%
0.015 g h−1 0.49 g h−1 0.68 g h−1 0.04 g h−1 0.05 g h−1
hydration of glycerol to acrolein lies between 8.2 ≤ H0
≤
−3.0.23 Systems having very strong acid activities (H0 ≤ −8.2),
such as NbP, may offer low acrolein selectivity (40–50%) due to
the onset coke deposition during the reaction. The dehydration
process usually requires high temperatures (250–300 °C) and
acrolein can easily undergo undesired polymerization and coke
formation.62 Various approaches have been proposed to limit
the deactivation of dehydration catalysts.24 These include the
suppression of the rate of coking by adding O2 or H2 to the feed
flow63–65 and modifying the catalysts with promoters to change
the acid properties of the catalysts. We have found that the de-
activation of NbP is related to the amount of solketal (and thus,
of glycerol) used in the reaction. The more the solketal is deli-
vered to the reactor, the more the acrolein is formed and conse-
quently, the larger the coke deposition over the catalyst active
phase. The TGA traces shown in the ESI† for the used catalyst
are consistent with the presence of organic coking.
a The conversion, selectivity and productivity were averaged values of
the reaction from 2nd to 15th hours and calculated based on GC ana-
lysis. Reaction conditions: 2a : 1b ratio of 1 : 1, total flow rate of
0.2 mL min−1 at 10 MPa.
Scheme 4 sketches a possible mechanism for the Skraup
reaction catalysed by NbP. Even under our best conditions, the
conversion of aniline and the selectivity towards quinoline did
not exceed 80% and 60%, respectively. These moderate yields
can be understood by considering the overall reaction which
occurs through three consecutive catalytic steps. The available
data indicate that the dehydration of glycerol over NbP or
mesoporous siliconiobium phosphate as a (doped) catalyst
proceeds with conversion and selectivity from 68 to 100% and
from 70 to 75%, respectively.34,45,46 It is therefore likely that,
once the dehydration of glycerol is coupled to the other two
There seems to be no obvious reason why polymerization
should give rise to the near stoichiometric behavior which we
Fig. 4 Scaled-up continuous flow quinoline synthesis in series. (a) Aniline conversion and (b) selectivity of quinoline studied under the same con-
ditions as those shown in Fig. 3 (250 °C, 10 MPa). Equimolar solutions of both reactants 2a and 1b were fed at flow rates of 0.1 and 0.3 mL min−1
The yellow profile is the same as that shown in Fig. 3 and it is reported here for comparison.
.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
Table 5 Scale-up continuous flow Skraup reaction productivity comparison for the one reactor Ra versus the triple reactor Rb setup at various flow
rates
1.2 g NbP Cat.
1.2 g 3 NbP Cat.
Aniline
Total
flow rate
Aniline
conv.a
Quinoline
Quinoline
productivity
TON
Quinoline
productivity
TON
sel.a
51%
(TOF, s−1 b
)
conv.a
Quinoline sel.a
(TOF, s−1 b
)
0.2 ml min−1
0.6 ml min−1
63%
0.068 g h−1
4.33(0.57 × 10−3
)
79%
74%
41%
41%
0.07 g h−1
0.19 g h−1
1.45(0.24 × 10−3
4.09(0.67 × 10−3
)
)
a The conversion, selectivity and productivity were averaged values of the reaction from over 15 hours and calculated based on GC analysis. The
different experiments were performed under the following conditions: 2a : 1b ratio of 1 : 1 at 250 °C and 10 MPa. b Turnover frequency (TOF) has
been calculated relative to the number of acid sites as follows: TOF = number of quinolines formed/number of NbP acid sites. The number of
acid sites is based on pyridine desorption.61
Scheme 4 One possible acid-catalyzed mechanism for the formation of quinoline, based on papers cited in our introduction, starting from solketal
1b and aniline 2a over niobium phosphate (NbP).60 None of the postulated intermediates nor any other were detected presumably because whatever
intermediates formed are unstable under our conditions.
observed. On the other hand, the oxidation step must give rise solution under the same reaction conditions as shown in
to a corresponding reduction of the surface. Such a reduction Fig. 3 at 250 °C. Less than 8% yield of two products 2,4-di-
would be expected to be close to the stoichiometric. Whichever methylquinoline 4a and 2,2,4-trimethyl-1,2-dihydroquinoline
of the explanations is correct, the route to regenerating the
spent NbP may well be heating under a stream of air. We
found that heating up to 300 °C for 20 hours was unsuccessful
and, unfortunately, our reactors could not be heated to higher
temperatures without compromising their pressure rating.32
Control experiments were carried out to establish whether
the acetone liberated from the solketal might be reacting with
aniline as this compound 2,2,4-trimethyl-1,2-dihydroquinoline
4b has been reported66 from the reaction of aniline with
acetone over zeolite catalysts but also halide acid clusters in
Scheme 5 Condensation of acetone with aniline in the absence of
solketal over NbP to give compounds 4a and 4b. The same reaction pro-
the vapour phase above 200 °C.67 Therefore, a 0.275 M solution
feed of acetone in toluene was used instead of the solketal ducts have been reported over a zeolite catalyst.67
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
Scheme 6 Products observed from the continuous flow reaction of 4-methoxyaniline 2b and 4,4’-methylenedianiline 2c with solketal over
niobium phosphate (NbP).
4b was obtained (Scheme 5). The products were isolated and terized by NMR and MS (see the ESI†). The process was not
characterized by MS and NMR analysis (see the ESI†). further optimised.
However, these products were not detected in any of our reac-
For the toxic 4,4′-methylenedianiline 2c (MDA), a higher
tions carried out with solketal, demonstrating clearly that temperature of 300 °C was necessary to increase the aniline
acetone is unlikely to take part in any side reactions in our conversion with a 1 : 1 mixture of 2c : 1b. Both dissymmetric 6a
system under these conditions.
and symmetric 6b products were observed with a GC-yield of
In order to demonstrate that NbP can be extended to cata- 20 and 12% respectively. A crystal of compound 6a was grown
lyse the Skraup reaction more widely, two anilines, 4-methoxy- and the structure was determined by crystallographic analysis
aniline and 4,4′-methylenedianiline (2b and 2c in Scheme 6) for the first time. The compounds were isolated and character-
were used to react with solketal. The expected heterocyclic ized by MS and NMR analysis (see the Experimental section).
quinoline product was formed as the major component in It is interesting that the major product is the desymmetrized
both the cases.
compound 6a.
We began with the best reaction conditions found for
aniline itself i.e. 1 : 1 ratio, 250 °C and 10 MPa. With 4-methoxy-
aniline, 6-methoxyquinoline 5a was obtained with a GC-yield
of 30%. The two main by-products observed were 4-methoxy-N-
methylaniline 5b and N-ethyl-4-methoxyaniline 5c (total
amount <10%). All these compounds were isolated and charac-
Conclusions
The experiments described here are a useful proof of concept.
Solketal can be used as a feedstock for the continuous Skraup
reaction promoted by a niobium phosphate solid acid catalyst.
No liquid acids or added oxidants are needed. It therefore
adds to the range of transformations that can be carried out
using glycerol, a by-product of the energy industry. Although
the reaction contravenes the sixth Principle of Green
Chemistry in that it requires high temperature and pressure, it
should be noted that (i) the high pressure may not be a sub-
stantial hindrance since the reactants are liquids and therefore
relatively incompressible requiring little energy to be pressur-
ized; and (ii) technologies for the integrated heat recovery in
modern biorefineries allow cheap access to temperatures up to
300 °C.
The major problem still to be addressed is how best the
niobium catalyst can be regenerated. Heating the catalyst to
300 °C in air had little effect. Our TGA experiment suggests
that at much higher temperatures calcination may be effective.
Even at the current stage of development, however, the reac-
tion could provide a convenient route to less common quino-
lines such as the desymmetrized 6a. From a sustainability
point of view the key points are that a continuous Skraup reac-
tion is possible and can be catalysed by niobium phosphate a
relatively inexpensive and abundant metal in a continuous
flow process (Fig. 5).
As explained above, we have used solketal because the limit-
ations of our small-scale equipment prevent the efficient
Fig. 5 Schematic apparatus for the continuous flow niobium phosphate
catalysed quinoline synthesis.
This journal is © The Royal Society of Chemistry 2017
Green Chem.

View Article Online
Paper
Green Chemistry
pumping of glycerol. Those experiments which we did perform
with glycerol gave poor selectivity. We attributed the better
selectivity which we observed with solketal as being due to the
better control of equipment which we could achieve with this
substrate. However, there remains the intriguing possibility
that solketal may give inherently better selectivity than glycerol
itself under these reaction conditions. If this were the case,
solketal would be the preferred starting material for this
process of valorisation of the glycerol by-product.
Unfortunately we have insufficient data to test this possibility.
Nevertheless, the point should be explored before dismissing
the use of solketal as unnecessary derivatization, in contraven-
tion of the eighth Principle of Green Chemistry (i.e. “unnecess-
ary derivatization [blocking group, protection/deprotection,
temporary modification of physical/chemical processes]
should be avoided whenever possible”). For the moment, it is
clear that solketal is potentially an alternative to glycerol for
smaller scale reactions, even if glycerol may eventually turn out
Notes and references
1 C.-H. Zhou, J. N. Beltramini, Y.-X. Fan and G. Q. Lu, Chem.
Soc. Rev., 2008, 37, 527–549.
2 C. B. Field, M. J. Behrenfeld, J. T. Randerson and
P. Falkowski, Science, 1998, 281, 237–239.
3 H. F. Sneddon, Green Chem., 2016, 18, 5082–5085.
4 C. A. G. Quispe, C. J. R. Coronado and J. A. Carvalho,
Renewable Sustainable Energy Rev., 2013, 27, 475–493.
5 W. Borzani, Braz. J. Microbiol., 2006, 37, 87–89.
6 A. Behr, J. Eilting, K. Irawadi, J. Leschinski and F. Lindner,
Green Chem., 2008, 10, 13–30.
7 M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and
C. D. Pina, Angew. Chem., Int. Ed., 2007, 46, 4434.
8 F. Jérôme, Y. Pouilloux and J. Barrault, ChemSusChem,
2008, 1, 586.
9 A. Behr, J. Eilting, E. Pérez, A. Garrido and J. Peinado,
Energy Fuels, 2008, 22, 4274–4280.
to be preferable from a Green Chemistry point of view for 10 C. Len and R. Luque, Sustainable Chem. Processes, 2014, 2,
larger scale processes.
1.
11 B. Katryniok, S. Paul and F. Dumeignil, ACS Catal., 2013, 3,
1819–1834.
12 M. B. Yadav, S. Kulkarni, R. A. Joshi and A. A. Kulkarni,
Org. Process Res. Dev., 2016, 10, 1621.
Experimental
13 Z. H. Skraup, Monatsh. Chem., 1880, 1, 316–318.
14 Z. Wang, Skraup Reaction in Comprehensive Organic Name
Reactions and Reagents, John Wiley & Sons, Inc., 2010.
15 G. A. Ramann and B. J. Cowen, Tetrahedron Lett., 2015, 56,
6436–6439.
16 X.-G. Li, X. Cheng and Q.-L. Zhou, Synth. Commun., 2002,
32, 2477–2481.
17 X. Zhang and X. Xu, Chem. – Asian J., 2014, 9, 3089–3093.
18 J. Horn, S. P. Marsden, A. Nelson and G. G. Weingarten,
Org. Lett., 2008, 10, 4117–4120.
19 Y. Hamada and I. Takeuchi, Chem. Pharm. Bull., 1971, 19,
1857.
20 M. C. Kozlowski, et al., Org. Lett., 2000, 2, 875.
21 G. A. Ramann and B. J. Cowen, Molecules, 2016, 21, 986.
22 C.-J. Jia, Y. Liu, W. Schmidt, A.-H. Lu and F. Schuth,
J. Catal., 2010, 269, 71–79.
All the continuous flow experiments were carried out using a
simple system consisting of a stainless steel tubular reactor (1/4
in. OD) or 4 tubular reactors in series for scale-up as detailed
in the ESI.† The reactors were filled with the niobium phos-
phate catalyst without any treatment (1.2 g). A tubular pre-
mixer was filled with sand and attached before the reactor.
The reagents were delivered using HPLC pumps and were
flowed through the reactor in a downward direction. A back
pressure regulator was used to maintain and adjust the system
pressure. The reaction requires specific sensors linked to a trip
box to ensure safety requirements for high temperature and
pressure conditions. An in-line GC/FID, placed between the
reactor and the BPR, was used to automatically sample and
analyse the reaction controlled by MathLab software. Niobium
phosphate was supplied in powder form by CBMM and 1.2 g
(unless specified otherwise) was packed into the reactor
without any treatment. The system was set to the studied para-
meter for approximately 10 to 25 h. The conversion and selecti-
vity were determined at regular intervals, by analyzing samples
taken online GC from the reaction medium for calculation of
the conversion and selectivity.
23 K. Pathak, K. M. Reddy, N. N. Bakhshi and A. K. Dalai,
Appl. Catal., A, 2010, 372, 224–238.
24 B. Katryniok, S. Paul, V. Bellière-Baca, P. Rey and
F. Dumeignil, Green Chem., 2010, 12, 2079.
25 B. Katryniok, S. Paul, M. Capron, C. Lancelot, P. Rey and
F. Dumeignil, Green Chem., 2010, 12, 1922.
26 E. Tsukuda, S. Sato, R. Takahashi and T. Sodesawa, Catal.
Commun., 2007, 8, 1349–1353.
27 P. Lauriol-Garbay, J. M. M. Millet, S. Loridant, V. Bellière-
Baca and P. Rey, J. Catal., 2011, 280, 68–76.
Acknowledgements
Financial support for JJ’s PhD by the Companhia Brasileira de 28 L.-Z. Tao, S.-H. Chai, Y. Zuo, W.-T. Zheng, Y. Liang and
Metalurgia e Mineração (CBMM) is gratefully acknowledged as B.-Q. Xu, Catal. Today, 2010, 158, 310–316.
well as the supply of niobium catalysts. We thank Richard 29 N. R. Shiju, D. R. Brown, K. Wilson and G. Rothenberg,
Wilson, Peter Field and Mark Guyler for technical support. We Top. Catal., 2010, 53, 1217–1223.
also thank William Lewis for the crystallographic analysis of 30 Z. Amara, E. S. Streng, R. A. Skilton, J. Jin, M. W. George
6a and Shazad Asalam for NMR assistance.
and M. Poliakoff, Eur. J. Org. Chem., 2015, 2015, 6141–6145.
Green Chem.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Green Chemistry
Paper
31 M. J. C. Molina, M. L. Granados, A. Gervasini and 50 D. Royon, S. Locatelli and E. E. Gonzo, J. Supercrit. Fluids,
P. Carniti, Catal. Today, 2015, 254, 90–98. 2011, 58, 88–89.
32 Y. Zhang, J. Wang, J. Ren, X. Liu, X. Li, Y. Xia, G. Lu and 51 S. Bagheri, N. Muhd Julkapli and W. A. Yehye, Renewable
Y. Wang, Catal. Sci. Technol., 2012, 2, 2485. Sustainable Energy Rev., 2015, 41, 113–127.
33 J. Xi, Y. Zhang, D. Ding, Q. Xia, J. Wang, X. Liu, G. Lu and 52 M. Selva, S. Guidi and M. Noè, Green Chem., 2015, 17,
Y. Wang, Appl. Catal., A, 2014, 469, 108–115. 1008–1023.
34 L. C. A. Oliveira, M. F. Portilho, A. C. Silva, H. A. Tarocoa 53 S. Guidi, R. Calmanti, M. Noè, A. Perosa and M. Selva, ACS
and P. P. Souza, App. Catal., B, 2012, 117-118, 29–35. Sustainable Chem. Eng., 2016, 4, 6144–6151.
35 T. Iizuka, K. Ogasawara and K. Tanabe, Bull. Chem. Soc. 54 Y. Cui, X. Zhou, Q. Sun, et al., J. Mol. Catal. A: Chem., 2013,
Jpn., 1983, 56, 2927–2931.
378, 238–245.
36 M. Ziolek, Catal. Today, 2003, 78, 47–64.
37 C. Guo and Z. Qian, Catal. Today, 1993, 16, 379.
55 M. Velasquez, A. Santamaria and C. Batiot-Dupeyrat, Appl.
Catal., B, 2014, 160–161, 606–613.
38 T. Ushikubo, T. Iizuka, H. Hattori and K. Tanabe, Catal. 56 S. Sato, D. Sakai, F. Sato and Y. Yamada, Chem. Lett., 2012,
Today, 1993, 16, 291. 41, 965–966.
39 A. Florentino, P. Cartraud, P. Magnoux and M. Guisnet, 57 C.-W. Chiu, M. A. Dasari, G. J. Suppes and W. R. Sutterlin,
Appl. Catal., A, 1992, 89, 143–153. AIChE J., 2006, 52, 3543–3548.
40 M. Moraes, W. D. S. F. Pinto, W. A. Gonzalez, L. M. P. 58 A. Alhanash, E. F. Khozevnikova and I. V. Khozevnikov,
M. Carmo, N. M. R. Pastura and E. R. Lachter, Appl. Catal.,
A, 1996, 138, L7–L12.
41 I. Rossetti and M. Compagnoni, Chem. Eng. J., 2016, 296,
56–70.
Appl. Catal., 2010, 378, 11.
59 D. S. Lee, Z. Amara, M. Poliakoff, T. Harman, G. Reid,
B. Rhodes, S. Brough, T. McInally and S. Woodward, Org.
Process Res. Dev., 2015, 19, 831–840.
42 C. G. Frost and L. Mutton, Green Chem., 2010, 12, 1687.
43 V. V. Ordomsky, V. L. Sushkevich, J. C. Schouten, J. Van Der
Schaaf and T. A. Nijhuis, J. Catal., 2013, 300, 37–46.
44 S. Okazaki and A. Kurosaki, Catal. Today, 1990, 8, 113–12;
60 N. R. Chernonukov, N. P. Egonov and I. R. Mochalova Russ,
J. Inorg. Chem., 1978, 23, 1627.
61 P. Carniti, A. Gervasini, S. Biella and A. Auroux, Chem.
Mater., 2005, 17, 6128–6136.
Y. Zhang, J. J. Wang, X. C. Li, X. H. Liu, Y. J. Xia, B. C. Hu, 62 A. Corma, G. Huber, L. Sauvanaud and P. O’Connor,
G. Z. Lu and Y. Q. Wang, Fuel, 2015, 139, 301–307. J. Catal., 2008, 257, 163–171.
45 Y. Choi, D. S. Park, H. J. Yun, J. Baek, D. Yun and J. Yi, 63 A. Ulgen and W. Hoelderich, Catal. Lett., 2009, 131, 122–
ChemSusChem, 2012, 5, 2460–2468. 128.
46 Y. Y. Lee, K. A. Lee, N. C. Park and Y. C. Kim, Catal. Today, 64 F. Wang, J.-L. Dubois and W. Ueda, J. Catal., 2009, 268,
2014, 232, 114–118.
260–267.
47 S. Okazaki and N. Wada, Catal. Today, 1993, 16, 349–359.
48 K. Nakajima, Y. Baba, R. Noma, M. Kitano, J. N. Kondo,
65 F. Wang, J. Xu, J. L. Dubois and W. Ueda, ChemSusChem,
2010, 3, 1383–1389.
S. Hayashi and M. Hara, J. Am. Chem. Soc., 2011, 133, 4224– 66 A. Hegedüs, Z. Hell, T. Vargadi, A. Potor and I. Gresits,
4227. Catal. Lett., 2007, 117, 99–101.
49 J. C. Smith and R. E. Drexel, Ind. Eng. Chem., 1945, 37, 601– 67 S. Kamiguchi, I. Takahashi, H. Kurokawa, H. Miura and
602.
T. Chihara, Appl. Catal., A, 2006, 309, 70–75.
This journal is © The Royal Society of Chemistry 2017
Green Chem.