Microwave-accelerated direct synthesis of 3-picoline from glycerol through a liquid phase reaction pathway

Article abstract of DOI:10.1039/c5nj02735g

A novel route for the synthesis of 3-picoline from glycerol via liquid phase reaction under mild condition, using microwave irradiation, had been well established. The heating mode had a profound effect on the yield of 3-picoline. The formation of 3-picoline could be promoted significantly by microwave heating, however, only very little amount of 3-picoline was generated by conventional heating under the reaction conditions employed in this study. Influencing factors were systemically investigated on the basis of a HAc-catalyzed under microwave irradiation. Additionally, a lot of heterogeneous catalysts were screened. It was found that as high as a 71% yield of 3-picoline was obtained with the mass ratio of pure glycerol/ammonium acetate/acetic acid/TiO₂ = 1/3/10/0.2 at 373 K, after only 20 min of microwave irradiation. A catalyst pair (HAc and TiO₂) exhibited better catalytic performance relative to other catalysts in this work. Accordingly, microwave assistance together with the catalysts achieved effective transformation of glycerol to 3-picoline under mild conditions. The related plausible reaction mechanisms were also proposed.

Full text of DOI:10.1039/c5nj02735g

View Article Online
View Journal
NJC
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: C. Luo, X. Feng
and Z. Chao, New J. Chem., 2016, DOI: 10.1039/C5NJ02735G.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/njc

Page 1 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
Microwave-accelerated Direct Synthesis of
Glycerol to 3-picoline under Liquid-phase
Pathway
Cai-Wu Luo 1, 2, Xiao-Yan Feng 2, Zi-Sheng Chao 2, *
1School of Environmental Protection and and Safety Engineering,
University of South China, Hengyang, Hunan 421001, China
2
College of Chemistry and Chemical Engineering, State Key
Laboratory of Chem/Biosensing and Chemometrics, Hunan
University, Changsha, Hunan 410082, China
* Corresponding author: Prof. ZiꢀSheng Chao;
Tel: +86ꢀ731ꢀ88713257;
Eꢀmail address: zschao@hnu.edu.cn or zschao@yahoo.com
1

New Journal of Chemistry
Page 2 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
Abstract
The mild liquidꢀphase synthesis of glycerol to 3ꢀpicoline had been
developed under microwave irradiation. The heating mode had a
profound effect on the yield of 3ꢀpicoline. The catalytic performance
evaluation showed that a great amount of 3ꢀpicoline was generated using
microwave irradiation. For comparison, under similar reaction conditions
as above but employing conventional heating mode, 3ꢀpicoline was
hardly produced. Some factors were systemically investigated on the
basis of the HAcꢀcatalyzed system. Additionally, a lot of heterogeneous
catalysts were screened. It is found that as high as 71 % yield to
3ꢀpicoline was obtained with the mass ratio of pure glycerol / ammonium
acetate / acetic acid / TiO2 = 1 / 3 / 10 / 0.2 for only 20 min at 373 K. The
catalyst pair (HAc and TiO2) exhibited better catalytic performance
relative to other catalysts in this work, and the plausible reaction
mechanisms were proposed. Accordingly, microwave assistance together
with the catalysts achieves the effective transformation of glycerol to
3ꢀpicoline under mild conditions.
Keywords: Glycerol; 3ꢀpicoline; Microwave; HAc; TiO2.
2

Page 3 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
1. Introduction
Pyridine bases, including mainly pyridine and its alkyl derivatives, e.g.,
2ꢀ, 3ꢀ, and 4ꢀpicolines, have found versatile application in a broad
spectrum of areas. Among various pyridine bases, 3ꢀpicoline is of much
more high economic interest, due to the important applications in the
1ꢀ4
manufactures of such like pharmaceuticals and pesticides,
and the
demand to 3ꢀpicoline has been steadily increasing in recently years. The
Chichibabin condensation between formaldehyde, acetaldehyde and
ammonia has constituted currently the predominant route for the
5ꢀ8
commercial manufacture of 3ꢀpicoline.
However, this process is
accompanied with a few issues, for examples, they are expensive, toxic,
derived from oil resources, prone to polymerization, and particularly, the
low yield of 3ꢀpicoline. In the literature, it is also reported that 3ꢀpicoline
9
could be synthesized from the reaction of acrolein and ammonia, and
10
the addition of a third component such as acetaldehyde
or
propionaldehyde 11 increases to some extent the yield of 3ꢀpicoline. The
most important interest associated with this process is that an appreciably
high selectivity to 3ꢀpicoline, which is even 100% for acrolein as reactant,
can be achieved. However, the employment for acrolein is still, and even
to a much more large extent, suffers from the issues as mentioned above
for the Chichibabin process. In addition, the aboveꢀmentioned reactions
are usually occurred at high temperature, leading to the decomposition of
3

New Journal of Chemistry
Page 4 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
3ꢀpicoline. Moreover, the use of the acidic catalysts to accelerate the
polymerization of the reactants and absorb the products can be
responsible for the generation of the coke, resulting in the fast
deactivation of catalyst. Finally, 3ꢀpicoline is produced by the reaction of
12ꢀ15
aldehydes and ammonium salt in the liquidꢀphase pathway,
but
generally under rigorous reaction conditions. Besides, the aldehydes have
to be constantly injected into the reactor during reaction, avoiding serious
polymerization. It increases the difficulty of operation and reduces the
production capacity. For example, the yield of 3ꢀpicoline was 68 %
(formaldehyde calculated) at 508 K and 3.8 ~ 4.0 MPa while using
formaldehyde, acetaldehyde and ammonium dihydrogen phosphate as
12
raw materials.
There is no doubt that the development of a green,
sustainable, and lowꢀcost pathway to manufacture 3ꢀpicoline under mild
conditions is highly desired.
Biomass, as an efficient substitute for traditional fossil fuel resource
(coal, oil and natural gas), is cheap, abundant, renewable and
environmentally friendly, and its application in the manufacture of useful
2, 16ꢀ18
chemicals, is of extremely large interest.
For example, our recent
study presented that a ca. 62% total yield of pyridine and 3ꢀpicoline was
obtained from the reaction of glycerol and ammonia in the gasꢀphase
route. 2 Although the high yield of acrolein is acquired in the liquidꢀphase
dehydration of glycerol, it needs relatively high temperature and/or high
4

Page 5 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
16ꢀ18
16
pressures.
For example, Yue et al.
reported that the yield of
acrolein over Cu/HZSMꢀ5 was 91.2% in the liquidꢀphase glycerol
dehydration reaction at 523 K in a sealed reactor. In this respect, it is
much difficult to achieve the oneꢀpot synthesis of glycerol to 3ꢀpicoline
with high yield under mild conditions. One of the promising strategies to
solve the problem is to alter the heating mode. Microwave irradiation is
19ꢀ21
an internal heat method,
which the heat is used for the reaction
system as much as possible, and thus reaction conditions such as
temperature or time become lower or shorter. Contrastively, traditional
heating is an external heat source, which transfers energy to reaction
system with the slow and inefficient properties. Some studies about
microwave irradiation involving glycerol have been well documented in
the literature. 22, 23 For example, Calvinoꢀcasilda et al. 22 had demonstrated
the microwaveꢀactivated direct synthesis of glycerol to acrylonitrile under
mild conditions. Very few catalysts, including mainly homogeneous
catalysts such as HAc, are employed and promote the formation of
4, 12ꢀ15
3ꢀpicoline in the liquidꢀphase route over the past decades.
It is
hoped to apply HAc as catalyst in the effective transformation of glycerol
to 3ꢀpicoline under mild conditions. Additionally, only
SO4/ZrO2/FeꢀZSMꢀ5 as heterogeneous catalyst in the HAcꢀbased system
4
is reported for the synthesis of 3ꢀpicoline up to now. Obviously, it is
much necessary to develop more heterogeneous catalysts.
5

New Journal of Chemistry
Page 6 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
In this work, we report the oneꢀpot synthesis of glycerol and
ammonium acetate to 3ꢀpicoline using microwave irradiation in the liquid
media. The main objective is to study the influence of process parameters.
The high yield of 3ꢀpicoline is obtained under mild conditions. The
catalyst pair (HAc + TiO2) is highly efficient for this reaction, and the
plausible reaction mechanisms are proposed. It provides many useful
references for the synthesis of 3ꢀpicoline or other pyridine bases in the
liquidꢀphase pathway.
2. Experimental
2.1. Materials
All the chemicals with analytic purity were commercially purchased.
2.2. Synthesis of 3-picoline
The conventional reaction was conducted under reflux condition at
atmospheric pressure. In brief, 0.5 g pure glycerol, 1.5 g ammonium
acetate (or 2.2 g ammonium dihydrogen phosphate) and 5.0 g acetic acid
(or 8.2 g phosphoric acid) were firstly mixed in a roundꢀbottomed flask
and then they were vigorously stirred for 10 min. The reaction (using oil
bath heating) was carried out at 373 K (or 503 K) for a period of 20 min
(or 2 h).
The microwave irradiation was conducted under reflux condition at
373 K and atmospheric pressure. They are as follows: (I) 0.5 g pure
6

Page 7 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
glycerol, a certain amount of ammonium acetate (or ammonium
hydroxide or urea or ammonium dihydrogen phosphate) and acetic acid
(or propionic acid) were firstly mixed in a roundꢀbottomed flask and then
they were vigorously stirred for 10 min; Or, (II) containing a suitable
amount of impurities such as NaCl or methanol in glycerol (pure glycerol,
0.5 g), 1.5 g ammonium acetate and 5.0 g acetic acid were firstly mixed
in a roundꢀbottomed flask and then they were vigorously stirred for 10
min; Or, (III) 0.5 g pure glycerol, 1.5 g ammonium acetate, 5.0 g acetic
acid and 0.1 g solid catalyst were firstly mixed in a roundꢀbottomed flask
and then they were vigorously stirred for 10 min. The above mixture ((I)
or (II) or (III)) was placed in the microwave equipment. A long condenser
pipe through the small hole on the microwave equipment was connected
to the upper of the roundꢀbottomed flask to play the role of the reflux. All
the reaction was carried out for a certain time (5 to 30 min).
After the reaction, the products were cooled to room temperature. The
qualitative and quantitative analysis of liquid products mixture were
carried out on a Varian Saturn 2200/CP 3800 GC/MS spectrometer,
equipped with a flame ionization detector (FID) and two CPꢀWax 52CB
fused silica capillary columns (15 m × 0.32 mm). In the qualitative
analysis of liquid products mixture, isopropyl alcohol had been added and
employed as internal standard. The glycerol conversion and the yield of
3ꢀpicoline were calculated as follows:
7

New Journal of Chemistry
Page 8 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
3. Results and discussion
3.1. Comparison with microwave heating and conventional heating for
the synthesis of 3-picoline
Fig. 1 shows a comparison between microwave heating (MWH) and
conventional heating (CH) on the yield of 3ꢀpicoline. Under conventional
heating, it provides only 24.6 % glycerol conversion, and no 3ꢀpicoline is
generated from the reaction of glycerol and ammonium acetate in acetic
acid for 2 h at 373 K. The glycerol is completely transformed but the
yield of 3ꢀpicoline is only 2.3 % while the reaction of glycerol and
ammonium dihydrogen phosphate in phosphoric acid is conducted for 2 h
at 503 K. Under microwave heating, the glycerol conversion reaches
about 92 % and the selectivity of 3ꢀpicoline is as high as 65.6 %, arriving
a yield of 60.3 % to 3ꢀpicoline from the reaction of glycerol and
ammonium acetate in acetic acid for 20 min at 373 K. Obviously, it
possesses a remarkably high catalytic performance under microwave
heating. This is due to the fact that the highly efficient heat is utilized and
thus it needs lower temperature to achieve the reaction under similar
conditions, relative to conventional heating. Gajengi et al. [20] considered
8

Page 9 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
that microwave irradiation resulted in the fast initial heating, and also, the
generation of some high pressure zones. Thus, it could significantly
improve the reaction kinetics compared with conventional heating.
Besides, the literature described that carbon sources such as acrolein were
initially diluted by solvent such as water or acetic acid and then they were
4, 12ꢀ15
injected into the reactor during reaction.
Therefore, the use of
microwave heating is helpful for improving the yield of 3ꢀpicoline and
dramatically reducing reaction conditions.
Much literature 24ꢀ26 reported that the interaction between aldehyde and
ammonia was a key procedure in the formation of pyridine bases.
Accordingly, the accepted reaction mechanism from the reaction of
glycerol and ammonium acetate to 3ꢀpicoline under microwave heating
involves two steps: step (1) glycerol dehydrates to acrolein as major
product; Step (2): the condensation of acrolein with ammonia from the
liberation of ammonium acetate generates 3ꢀpicoline. Meanwhile,
acrolein and/or its intermediate selfꢀpolymerizes into heavy byproducts
by a parallel reaction. These heavy byproducts from acrolein have been
preliminarily identified by our research groups in detail. 4 A lot of works 4,
15, 17, 18, 27, 28
demonstrated that the process of step (1) generally happened
at higher temperature than that of step (2). For example, the optimal
temperatures in the liquidꢀphase dehydration of glycerol to acrolein over
17, 18, 27, 28
various catalysts were usually at 473 ~ 673 K,
and the optimal
9

New Journal of Chemistry
Page 10 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
temperature from the condensation of acrolein and ammonium salt was at
403 K. 4, 15 Clearly, microwave heating can achieve the direct synthesis of
glycerol to 3ꢀpicoline under mild conditions. In other words, the heating
mode is one of key factors for the formation of 3ꢀpicoline in this reaction.
3.2. Effect of reaction time
Fig. S1 (Support information) shows the effect of the reaction time on
the yield of 3ꢀpicoline under microwave irradiation. One can see that with
increasing the reaction time, the yield of 3ꢀpicoline increases, achieving
about 60 % as the largest value at the reaction time = 20 min, and then
almost keeps unchanged, implying that ammonia is completely consumed
after this time.
3.3. Effect of ammonia sources
Table 1 lists the effect of various ammonia sources on the yield of
3ꢀpicoline under microwave irradiation. One can see that the glycerol
conversion over various ammonia sources decreases in the following
order: ammonium acetate > ammonia solution > urea > ammonium
dihydrogen phosphate. Likewise, the selectivity of 3ꢀpicoline is displayed
the same order of glycerol conversion. The differences about their yields
using the former three ammonia sources are relatively small. While
ammonia solution is added into the reaction mixture, it immediately
reacts with acetic acid to generate ammonium acetate. Urea is initially
decomposed during reaction and thus generates carbon dioxide and
10

Page 11 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
ammonia. Then, ammonia is easily trapped by acetic acid in liquidꢀphase
and vaporꢀphase environment, finally, ammonium acetate is produced.
And meanwhile, carbon dioxide and water provide weak acid
environment, which is favor for suppressing the polymerization of
acrolein. Accordingly, they play similar roles in providing ammonia in
three cases. However, ammonia is not produced from the decomposition
of ammonium dihydrogen phosphate at low temperature (373 K) and
therefore 3ꢀpicoline is not obtained in this route. It provides a good
reference for the solventꢀfree synthesis of 3ꢀpicoline.
Fig. 2 shows the effect of the usage of ammonium acetate (AA) on the
yield of 3ꢀpicoline under microwave irradiation. One can see that with
increasing the mass ratio of ammonium acetate/glycerol in the range of
1.0 ~ 2.5, the yield of 3ꢀpicoline increases slowly. With increasing the
mass ratio of ammonium acetate/glycerol, the yield of 3ꢀpicoline reaches
about 60 % as its maximum at a mass ratio of ammonium acetate/glycerol
= 3, and then decreases. These results show that the reactant ratios have a
great influence on the yield of 3ꢀpicoline. It is well accepted that imine
species are important intermediates in the formation of pyridine and
24, 25
picolines in the gas and liquid conditions.
are shown as follows (Eqs. (1ꢀ6)):
The reaction procedures
(1)
11

New Journal of Chemistry
Page 12 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
(2)
(3)
(4)
(5)
(6)
From the view of these equations, the increase the amount of
ammonium acetate provides more ammonia, resulting in producing more
3ꢀpicoline, however, a too much ammonium acetate increases the amount
of HAc and the basicity. In this case, the polymerizations of acrolein and
propyleneimine are absolutely increased (Eqs. (4) to (6)). Accordingly,
the yield of 3ꢀpicoline is decreased, and also, it increases the entire cost
of reaction system, because ammonium acetate generates acetamide (Eq.
(2)).
3.4. Effect of the usage of HAc
12

Page 13 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
Fig. 3 shows the effect of the usage of acetic acid (HAc) on the yield of
3ꢀpicoline under microwave irradiation. When no HAc is added into the
mixture of glycerol and ammonium acetate, the highly low yield of
3ꢀpicoline is obtained. It is mainly attributed to the insufficient acid active
sites and the incompletely contact between the reactants. While the mass
ratio of HAc/glycerol is increased up to 5, the yield of 3ꢀpicoline is
slowly increased. With increasing the usage of HAc, the yield of
3ꢀpicoline increases sharply, reaching about 60 % as its maximum at a
mass ratio of glycerol / HAc = 1 / 10, and then decreases. It indicates that
the appropriate usage of HAc is in favor of improving the yield of
3ꢀpicoline. This is due to the fact that HAc can provide protonic acid sites
for catalyzing the dehydration of glycerol, and also the condensation of
4, 29
the dehydrated products and ammonia.
We reported that more heavy
byproducts were formed while adding too much the usage of HAc. It was
attributed to the polymerization of acrolein and/or its related intermediate,
4
as observed from the Eq. (7). Therefore, the increase the usage of HAc
promotes the yield of 3ꢀpicoline, however, a too much the usage of HAc
decreases the yield of 3ꢀpicoline. Besides, the loss of ammonia is
effectively suppressed, because ammonium acetate is formed by the
reaction of HAc and ammonia under bulk solution and vapor phase,
thereby, making this reaction to conduct under atmospheric pressure. In
addition, microwave irradiation can directly induce the dehydration of
13

New Journal of Chemistry
Page 14 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
glycerol. It is reported that 38.2 % conversion of glycerol and 83.3 %
selectivity to acrolein were obtained at 373 K, Power = 0 – 100 W and
reaction time = 1 h under microwave irradiation in blank study. 22 Clearly,
HAc together with microwave irradiation results in the formation of
3ꢀpicoline under mild conditions. To check HAc involving this reaction,
the reaction of glycerol and HAc is carried out under microwave
irradiation. The result shows the glycerol conversion is only ca. 14 %,
and the related esterification are hardly detected. This is due to the fact
that the esterification of glycerol and HAc is a reversible process, and
water as one of the products is not removed in time in this study. Thus,
few esterification products are generated. It is agreement with the
30
literature.
It is reasonable that HAc is an effectively homogeneous
catalyst in the reaction of glycerol and ammonium acetate toward
3ꢀpicoline under microwave irradiation.
(7)
To confirm the unique property of HAc, propionic acid is employed for
this reaction. Fig. S2 displays a comparison for the production of
3ꢀpicoline using acetic acid or propionic acid as both the solvent and
catalyst under microwave irradiation. One can see that a 12.3 % yield of
3ꢀpicoline over the latter is obtained, which is remarkably lower than that
14

Page 15 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
of the former under the similar conditions. This is due to the fact that the
generated ammonium propionate between propionic acid and ammonia
from ammonium acetate providing reduces the utilization of ammonia to
3ꢀpicoline, since the decomposed temperature of ammonium propionate
is higher than that of ammonium acetate. It further confirms that HAc as
catalyst is very suitable for this reaction.
3.5. Effect of the impurities in glycerol
Table 2 lists the effect of the impurities in glycerol on the yield of
3ꢀpicoline under microwave irradiation. One can see that a 43 ~ 47 %
yield of 3ꢀpicoline is obtained while water or methanol is added into the
HAcꢀbased system. Contrastively, the blank experiment runs as high as
about 60 % yield of 3ꢀpicoline. Apparently, by the addition of solvent to
the mixture of glycerol, ammonium acetate and acetic acid has a negative
influence on the yield of 3ꢀpicoline, especially the coꢀaddition of water
and methanol. Nevertheless, the selectivity of 3ꢀpicoline is increased in
these cases. It indicates that these solvents can restrain the generation of
the byproducts from acrolein and accelerate the reaction rate of Eq. (3)
process, thereby promoting the formation of 3ꢀpicoline. The most
important role of water or methanol is to dilute reaction system, even it
reacts with the active intermediate such as acrolein. For example, one
acetal is generated from the condensation of acrolein and methanol, as
shown in Eq. (8). By this way, the large local concentration of acrolein is
15

New Journal of Chemistry
Page 16 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
decreased and thus the possibility of side reactions becomes less.
However, the dilutions reduce the HAcꢀcatalyzed glycerol conversion due
to the competition reaction among glycerol, water and methanol.
Accordingly, the glycerol conversion is sharply decreased, further
reducing the yield of 3ꢀpicoline. The composition of glycerol with the
close proximity to the crude glycerol is investigated, and the result shows
that the glycerol conversion and the yield of 3ꢀpicoline are at 76.7 % and
39.6 % after adding three components to glycerol. Interestingly, this
result is still higher than the yield of 3ꢀpicoline from the condensation of
acrolein and ammonium acetate.15 The negative effect is ascribed to the
inhibition of glycerol dehydration to acrolein and accelerating the
acrolein selfꢀpolymerization in the presence of Na+ and Clꢀ. This
deduction is proved that the yield of 3ꢀpicoline is at 17.8 % after adding
NaCl to the reaction mixture. The above results show that this novel
pathway can resist common impurities in glycerol, which is in favor of
decreasing the cost.
(8)
3.6. Effect of the addition of the solid catalysts in HAc solution
Fig. 4 shows the effect of HSiW and HZSMꢀ5 on the yield of
3ꢀpicoline under microwave irradiation. One can see that the glycerol
conversion and the selectivity of 3ꢀpicoline (0 #) are 91.9 % and 65.6 %
16

Page 17 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
in the mixture of glycerol, ammonium acetate and acetic acid,
respectively. After HSiW and HZSMꢀ5 (1 # and 2 #) are respectively
introduced into the reaction mixture, the glycerol conversion, and the
selectivity of 3ꢀpicoline are 92.6 %, 99.2 %, and 33.9 %, 21.6 %. It is
found that the glycerol conversion is increased, relative to the blank
experiment (0 #), which is attributed to the strong acidity for the catalysts.
16, 28 And also, the declined degree is much more serious in HZSMꢀ5. It is
known that HZSMꢀ5 is the most studied for the gasꢀphase synthesis of
pyridine bases. However, its catalytic performance is very low in this
reaction. This question is answered by the pore structure and acidity of
the catalyst. Table S1 lists the textual and acidity properties of HZSMꢀ5.
It is observed that the ratio between micropore surface area and specific
surface area is as high as 90 %. And meanwhile, the pore size is relatively
small. Thus, it increases the mass transfer resistance and accelerates the
polymerization of acrolein and/or its related intermediate. Besides, it
appears two desorption peaks of ammonia, which is the corresponding to
weak and strong acid sites, respectively. And also, desorption
temperatures of acid sites are both above 373 K. In this case, it is much
more difficult for the products to desorb on the acid sites at low
temperature, because the target product possesses the alkalinity. As a
consequence, these factors result in a weaker catalytic performance after
adding HZSMꢀ5 in the reaction mixture.
17

New Journal of Chemistry
Page 18 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
Fig. S3 shows the effect of various organic acid solid catalysts on the
yield of 3ꢀpicoline under microwave irradiation. One can see that the
glycerol conversion (3ꢀ7 #) has an order stearic acid > oxalic acid >
adipic acid > sulfosalicylic acid > edetic acid. Compared with the blank
experiment, the glycerol conversion is only increased in the former two (3
# and 4 #). This is the most probably due to the fact that they provide the
stronger acidity relative to HAc, thereby promoting the glycerol
conversion. And also, a too strong acidity enhances the esterification and
thus suppresses the dehydration of glycerol to acrolein, thereby, leading
to decreasing the glycerol conversion. Accordingly, the glycerol
conversion is decreased while adding the adipic acid and sulfosalicylic
acid (5 # and 6 #). The lowest glycerol conversion over edetic acid is
obtained, because the decomposition is occurred during reaction, and thus
the competitive reaction consumes some HAc, resulting in decreasing the
glycerol conversion. It is clearly observed that the selectivity of
3ꢀpicoline over all the cases (3ꢀ7 #) is decreased, resulting in a lower
yield of 3ꢀpicoline. It is attributed to the insufficient ammonia for
generating 3ꢀpicoline. Namely, ammonia from the decomposition of
ammonium acetate reacts with these containing carboxylic acid catalysts
to generate ammonium salt with the higher decomposition temperature.
Obviously, it is difficult to decompose these newly formed salts at low
temperature in this study.
18

Page 19 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
Fig. S4 shows the effect of various ionꢀexchanged resin on the yield of
3ꢀpicoline under microwave irradiation. One can see that the yields of
3ꢀpicoline over all the resin catalysts (8ꢀ10 #) are no more than 7 %. It is
found that the glycerol conversion and the selectivity of 3ꢀpicoline are
both shown an order D 001 H > D 113 III > D 402 Na. The results are
proportional to the acidity of the catalysts. Accordingly, the increasing
acidity promotes this reaction. However, compared without the addition
of solid catalysts, the selectivity of 3ꢀpicoline is sharply decreased due to
the strong adsorption. Although the positive effect of the macrospores by
accelerating the mass transfer rate does not play in this reaction, the
separation and recovery is an obvious merit.
Fig. S5 shows the effect of KF and MgF2 on the yield of 3ꢀpicoline
under microwave irradiation. One can see that the glycerol conversion
over KF and MgF2 (11 # and 12 #) is 60 ~ 75 %, and the selectivity of
3ꢀpicoline is both no more than 1 %. It is well known that strong base
catalyst can largely accelerate the polymerization of acrolein. And also, it
reduces the acidity of reaction system and therefore affects the
HAcꢀcatalyzed performance. In these cases, the dehydration of glycerol
to acrolein is to some extent suppressed and therefore the glycerol
conversion and the selectivity of 3ꢀpicoline are remarkably decreased.
Fig. 5 shows the effect of various metal oxides on the yield of
3ꢀpicoline under microwave irradiation. One can see that a 71% yield of
19

New Journal of Chemistry
Page 20 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
3ꢀpicoline is obtained using TiO2 as catalyst (14 #), which is significantly
higher than that of other two catalysts (13 # and 15 #), and also this value
is the highest in this study. Compared with HAc, the glycerol conversion
is both increased while ZnO and TiO2 (13 # and 14 #) are added,
respectively. It shows that they exhibit higher catalytic performance in the
glycerol conversion. The selectivity of 3ꢀpicoline is apparently increased
while TiO2 and ZrO2 (14 # and 15 #) are introduced, respectively. These
results indicate that these catalysts enhance the condensation of the
dehydrated products and ammonia to form 3ꢀpicoline. In a word, TiO2 is
an excellent heterogeneous catalyst in this reaction. Additionally,
3ꢀpicoline is not detected with or without the addition of TiO2 while the
reaction
of
glycerol
and
urea
is
conducted
using
ethylene glycol monobutyl ether as solvent. It is reasonable that HAc
plays a greater catalytic role in this reaction, relative to TiO2. Accordingly,
HAc and TiO2 as homogeneous and heterogeneous catalysts are both
responsible for the formation of 3ꢀpicoline with high yield. Moreover,
HAc not only acts as solvent but also adsorbs ammonia. In addition,
commercial TiO2 catalyst should be extreme stability under mild
conditions in this work. It is not necessary to check the reusability. The
aboveꢀmentioned results offer many good ideas for the preparation of
highly effective heterogeneous catalysts in the synthesis of 3ꢀpicoline.
3.7. A comparison of the yield of 3-picoline
20

Page 21 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
Table 3 compares the reactions of various reactants and ammonia to
3ꢀpicoline under liquidꢀphase pathway. As is known, the reaction of
acetaldehyde/formaldehyde and ammonia constitutes the predominant
route for the synthesis of 3ꢀpicoline in industry. In this route, the yield of
12
3ꢀpicoline was 68 % in the liquid media, and phosphoric acid acts as
catalyst, deriving from the decomposition of salt. The main drawbacks
associated with this route consist of the oil dependence and toxicity of
aldehydes. As the substitutes of acetaldehyde/formaldehyde, acrolein was
4
employed to react with ammonia, affording a 52% 14 or 60% yield of
3ꢀpicoline without or with the addition of solid catalyst. However, the
polymerization and toxicity of acrolein constitute the main challenges for
its industrial application in the production of 3ꢀpicoline. In our work,
glycerol is employed to react with ammonia under microwave irradiation.
Compared with the above various raw materials, glycerol is green,
nontoxic, cheaper, oilꢀindependent, and difficult to polymerization. Over
the catalyst pairs (HAc + TiO2), a 71% yield of 3ꢀpicoline is obtained,
with the milder conditions. The above merits indicate that this pathway
for the synthesis of 3ꢀpicoline developed in this work is very promising
for commercial application.
3.8. Plausible reaction mechanisms
According to aboveꢀmentioned results and the literature, the related
reaction mechanisms are proposed for the synthesis of 3ꢀpicoline in the
21

New Journal of Chemistry
Page 22 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
homogeneous and heterogeneous reactions, respectively. In the
HAcꢀbased homogeneous reaction system (Eq. (9)), the interaction
between glycerol and a protonic acid site from the HAc catalyst first
results in the protonation of the hydroxyl group, locating the middle
29
position in the glycerol molecule,
subsequently, it undergoes a
dehydration process. With the elimination of water, one acrolein molecule
is formed and also it generates HAc. Besides, the glycerol dehydration to
acrolein can achieve under microwave irradiation (Eq. (10)); second,
25
Golunski et al.
proposed a common reaction mechanism in the
synthesis of various pyridine bases. Namely, the imine species were
initially formed from the reaction of aldehyde group and ammonia
through the dehydration by a protonic acid site, and thus pyridine bases
were generated the condensation from imine species. In our reaction
system, propyleneimine is generated by the direct reaction between
acrolein and ammonia via dehydration, and then 3ꢀpicoline is formed (Eq.
(3)). Thus, the catalyst must possess protonic acid sites to active aldehyde
group in reagents. Accordingly, it is speculated that protonic acid sites
can activate not only the hydroxyl groups from glycerol but also aldehyde
group in acrolein. Obviously, HAc can promote the glycerol dehydration
and the condensation of the dehydrated products such as acrolein with
ammonia from the decomposition of ammonium salt to finally produce
3ꢀpicoline.
22

Page 23 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
(9)
(10)
In the HAcꢀbased heterogeneous reaction system, another different
reaction mechanism is proposed because the heterogeneous catalysts
(metal oxides as references) involve this reaction. On the one hand, it is
well accepted that metal oxides (TiO2) are typical Lewis acid catalysts.
29
According to the theory from the literature,
more acetol would be
generated from the dehydration of glycerol using pure TiO2 as catalyst. In
this case, the yield of 3ꢀpicoline is absolutely decreased. Accordingly, it
31
can not account for the related result (Fig. 5, 14 #). Buhler et al.
demonstrated that two reaction competition pathways were appeared in
the liquidꢀphase dehydration of glycerol to acrolein. One pathway was an
ionic process: a glycerol molecule was initially protonized to form the
carbocation, and then underwent the dehydration step to generate acrolein.
Another pathway was a pyrolysis process: hydroxyl radical (·OH)
attacked the terminal hydroxyl group in glycerol molecule and then
underwent the hydrogenation process to produce acrolein (Eq. (11)).
23

New Journal of Chemistry
Page 24 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
18
Watanabe et al. further confirmed that the ionic process was happened
at low temperature whereas the pyrolysis process was beneficial to
conduct at high temperature. TiO2 is one of the most common
semiconductors, and it can generate hydroxyl radicals (·OH) in aqueous
solution under microwave irradiation (Eqs. (12 to 19)). 32 Accordingly, the
aboveꢀmentioned reaction pathways are occurred basing on the collected
results from the glycerol conversion (Fig. 5, 13 # and 14 #), although
reaction conditions are different between this work and the literature.18, 31
From the analysis it can be concluded that the ionic process is accelerated
by HAc and the radical process is enhanced by TiO2 under microwave
irradiation, thereby, resulting in a higher glycerol conversion relative to
the HAc alone.
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
24

Page 25 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
(19)
On the other hand, the increased selectivity of 3ꢀpicoline can be
26, 33
explained by the similar literature in the gasꢀphase method.
For
33
2ꢀ
example, Zenkovets et al.
reported that the SO4 /TiO2 catalyst
exhibited high activity and selectivity from the condensation of acrolein
26
and ammonia to pyridine bases. Singh et al.
investigated that
(Al+Zr)ꢀPILC montmorillonite clay as catalyst was applied for the
synthesis of pyridine bases from the condensation of formaldehyde,
acetaldehyde and ammonia. They proposed a different reaction
mechanism (Eq. (20)): acrolein, was initially produced from the
condensation of formaldehyde and acetaldehyde, and then formed a
carbonium ion via the protonation. This ion condensed with another
acrolein to obtain active ion complex. Finally, this complex reacting with
ammonia generated 3ꢀpicoline. These cases show that containing TiO2
and ZrO2 materials are both excellent catalysts for the synthesis of
3ꢀpicoline. This is due to the fact that they can accelerate the reaction
process because of their dehydrogenation ability. Obviously, TiO2 and
ZrO2 promote the selectivity of 3ꢀpicoline.
(20)
In terms of the reaction results, we consider in the presence of free
radical mechanism. However, it needs be confirmed by suitable
25

New Journal of Chemistry
Page 26 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
experiment. In a word, microwave irradiation and the catalysts are two
key factors to obtain the high yield of 3ꢀpicoline with mild conditions.
4. Conclusions
We reported a mild route for the synthesis of 3ꢀpicoline under
liquidꢀphase conditions. The heating mode had a great impact on the yield
of the target product. 3ꢀpicoline with high yield was obtained under
microwave irradiation whereas very few 3ꢀpicoline was produced using
traditional heating. A near 93 % glycerol conversion and 71 % yield of
3ꢀpicoline were obtained with the mass ratio of pure glycerol /
ammonium acetate / acetic acid / TiO2 = 1 / 3 / 10 / 0.2 for 20 min at 373
K under microwave irradiation. Among various catalysts, the catalyst pair
(HAc + TiO2) exhibit higher catalytic performance in this reaction. Both
acidꢀcatalytic and free radical mechanisms were occurred in the
heterogeneous system.
26

Page 27 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
Acknowledgments
This work was supported by the National Natural Science Foundation
of China (Grant 21376068), Program for New Century Excellent Talents
in University, the Ministry of Education of P. R. China, and the Program
for FuꢀRong Scholar in Hunan Province, P. R. China.
27

New Journal of Chemistry
Page 28 of 40
View Article Online
DOI: 10.1039/C5NJ02735G
References
(1) X. Zhang, Z. Wu, Z. S. Chao, J. Mol. Catal. A, 2016, 411, 19–26.
(2) C. W. Luo, C. Huang, A. Li, W. J. Yi, X. Y. Feng, Z. J. Xu, Z. S.
Chao, Ind. Eng. Chem. Res., 2016, 55, 893–911.
(3) S. Shimizu, N. Abe, A. Iguchi, M. Dohba, H. Sato, K. Hirose,
Microporous Mesoporous Mater., 1998, 21, 447–451.
(4) X. Zhang, C. W. Luo, C. Huang, B. H. Chen, D. G. Huang, J. G. Pan,
Z. S. Chao, Chem. Eng. J., 2014, 253, 544–553.
(5) J. I. Darragh. US Pat., 4089863, 1978.
(6) R. R. Rao, S. J. Kulkarn, M. Subrahmanyam, A. V. R. Rao, React.
Kinet. Catal. Lett., 1995, 56, 301–309.
(7) K. R. S. K. Reddy, I. Sreedhar, K. V. Raghavan, Appl. Catal., A.,
2008, 339, 15–20.
(8) F. Jin, Y. G. Cui, Y. D. Li, Appl. Catal., A., 2008, 350, 71ꢀ78.
(9) H. Beschke, H. Schaefer, G. Schreyer, W. A. Schuler, W. Weigert, US
Pat., 3960766, 1976.
(10) H. Beschke, H. Friedrich, US Pat., 4147874, 1979.
(11) H. Beschke, H. Friedrich, US Pat., 4163854, 1979.
(12) R. Dinkel. US Pat., 4337342, 1982.
(13) R. Dinkel, H. Roedel, J. I. Grayson, US Pat., 4482717, 1984,.
(14) J. I. Grayson, R. Dinkel, US Pat., 4421921, 1983.
(15) A. Nicolson, GB Pat., 1240928, 1971.
28

Page 29 of 40
New Journal of Chemistry
View Article Online
DOI: 10.1039/C5NJ02735G
(16) C. J. Yue, M. M. Gan, L. P. Gu, Y. F. Zhuang, J. Taiwan. Inst. Chem.
Eng., 2014, 45, 1443–1448.
(17) A. May, J. Salvadó, C. Torras, D. Montané, Chem. Eng. J, 2010,
160, 751–759.
(18) M. Watanabe, T. Iida, Y. Aizawa, T. M. Aida, H. Inomata, Bio.
Technol., 2007, 98, 1285–1290.
(19) S. Horikoshi, J. Tsuzuki, M. Kajitani, M. Abe, N. Serpone, New J.
Chem., 2008, 32, 2257–2262.
(20) A. L. Gajengi, T. Sasaki, B. M. Bhanage, Catal. Comm., 2015, 72,
174–179.
(21) H. Saggadi, D. Luart, N. Thiebault, I. Polaert, L. Estelb, C. Len,
RSC Adv., 2014, 4, 21456–21464.
(22) V. CalvinoꢀCasilda, M. O. GuerreroꢀPérez, M. A. Baňares, Appl.
Catal., B, 2010, 95, 192–196.
(23) H. Saggadi, D. Luart, N. Thiebault, I. Polaert, L. Estel, C. Len,
Catal. Comm., 2014, 44, 15–18.
(24) J. R. Calvin, R. D. Davis, C. H. McAteer, Appl. Catal., A, 2005,
285, 1–23.
(25) S. E. Golunski, D. Jackson, Appl. Catal., 1986, 23, l–14.
(26) B. Singh, S. K. Roy, K. P. Sharma, T. K. Goswami, J. Chem.
Technol. Bio., 1998, 71, 246–252.
(27) A. S. de Oliveira, S. J. S. Vasconcelos, J. R. de Sousa, F. F. de
29