N-Alkylation of Alkylolamines with Alcohols Over Mesoporous Solid Acid–Base Cs–B–Zr Catalyst

Article abstract of DOI:10.1007/s10562-016-1737-2

Abstract: The mesoporous solid acid–base Cs–B–Zr mixed oxides were synthesized using the co-precipitation method followed by a subsequent thermal treatment. The catalytic activity of solid Cs–B–Zr mixed oxide was tested for solvent free acid–base catalysed direct alkylolamines with alcohols as green alkylating agent. The effects of Cs/B/Zr ratio, calcination temperature, reaction conditions, and reaction substrate on the catalytic performance of the catalysts were investigated. The XRD, N₂ adsorption–desorption, ICP-OES, FT-IR and NH₃/CO₂-TPD results showed that the mesoporous structure and acid–base properties of the catalysts play important roles in the reaction. A suitable number of acid and basic sites on the catalyst lead to a high activity for the N-alkylation reaction. Graphical Abstract: A direct N-alkylation of amino alcohol with alcohols has been developed using mixed oxide Cs–B–Zr as an acid–base bifunctionalized catalyst.[Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-016-1737-2

Catal Lett
DOI 10.1007/s10562-016-1737-2
N-Alkylation of Alkylolamines with Alcohols Over Mesoporous
Solid Acid–Base Cs–B–Zr Catalyst
1
1
1
1
1
Aimin Chen
•
Houyong Wang
•
Rui Liu
•
Yingying Bo
•
Jun Hu
Received: 24 February 2016 / Accepted: 23 March 2016
Ó Springer Science+Business Media New York 2016
Abstract The mesoporous solid acid–base Cs–B–Zr
mixed oxides were synthesized using the co-precipitation
method followed by a subsequent thermal treatment. The
catalytic activity of solid Cs–B–Zr mixed oxide was tested
for solvent free acid–base catalysed direct alkylolamines
with alcohols as green alkylating agent. The effects of Cs/
B/Zr ratio, calcination temperature, reaction conditions,
and reaction substrate on the catalytic performance of the
catalysts were investigated. The XRD, N2 adsorption–
desorption, ICP-OES, FT-IR and NH /CO -TPD results
3
2
showed that the mesoporous structure and acid–base
properties of the catalysts play important roles in the
reaction. A suitable number of acid and basic sites on the
catalyst lead to a high activity for the N-alkylation reaction.
Graphical Abstract A direct N-alkylation of amino
alcohol with alcohols has been developed using mixed
oxide Cs–B–Zr as an acid–base bifunctionalized catalyst.
Keywords N-Alkylation Á Mesoporous Á Acid–base
catalyst Á Cs–B–Zr mixed oxide
1
Introduction
N-alkylaminoalcohols are important intermediates in fine
chemical industries, manufacturing, pharmaceuticals,
agrochemicals, polymers and dyes, especially as desulfur-
izer and/or decarbonizer in environmental protection [1–3].
The production of N-alkylaminoalcohols is increasing with
years. Traditionally, N-alkylaminoalcohols are mainly
produced by direct alkylation by halide [4], reductive
amination of methyl-carbonate/dimethyl sulfate [5–7] and
ethylene oxide method [8, 9]. Nevertheless, all these pro-
cesses have their own drawbacks, such as expensive reac-
tants, low selectivity, poor yields, harsh reaction conditions
and large amounts of harmful wastes polluting the
&
&
Aimin Chen
amchen@zjut.edu.cn
Jun Hu
hjzjut@zjut.edu.cn
1
College of Chemical Engineering, Zhejiang University of
Technology, Hangzhou 310014, China
123

A. Chen et al.
environment. Therefore, much attention has been paid to
explore a new ecologically benign synthetic route.
treatment. A typical procedure for alcohol amines N-
alkylation reaction with alcohols was investigated to
evaluate the catalytic performance of Cs–B–Zr catalyst.
The activity and selectivity of these new catalysts were
correlated with the results of NH /CO -TPD measure-
In the recent years, the direct N-alkylation of amines
with alcohols has been reported as an example of ‘‘green’’
technology which is an alternative route through a bor-
rowing-hydrogen (B-H) [10–12]. The advantage of this
type of amination is the high atom efficiency of the reac-
tion sequence, which forms water as the only by-product.
Through a B-H process, imines are usually formed as the
intermediates, and can be further reduced to alcohol amine
products by the hydrogen species generated on the catalyst
3
2
ments. In addition, the mechanism of Cs–B–Zr catalysts for
direct N-alkylation of N-alkylaminoalcohol with alcohols
was discussed.
2 Experimental Section
[
13]. Since Grigg et al. and Watanabe et al. had introduced
homogeneous catalysts for N-alkylation of amines with
alcohols in 1981 for the first time [14, 15], transition metal
complexes and salts such as Ru, Pd, Pt, Au and Ir catalysts
have been developed as the homogeneous catalysts in the
presence of solvent, promoters and bases have been
reported as efficient homogeneous catalytic systems for
direct N-alkylation of amines with alcohols [16–22].
Unfortunately, these homogeneous catalytic systems have
disadvantages related to the recovery, high cost, special
handling of metal complex and/or the indispensable use of
co-catalysts such as bases and stabilizing agents. So, the
exploration of a ligand free, base and promoter free, and
solvent free heterogeneous catalytic system is highly
desirable for direct N-alkylation of amines.
2.1 Preparation of the Catalysts
The Cs–B–Zr mixed oxide catalyst was prepared by a two-
step process method with the following procedure. Firstly,
22.5 g ZrOCl Á8H O was dissolved in 100 ml distilled
2
2
water to which aqueous ammonia (25 %) was added drop-
wise with constant stirring till the solution became alkaline
(pH 10) and zirconium hydroxide precipitated as white
solid. A mixed solution containing 3.7 g H BO , 11.7 g
3
3
CsNO and 5.1 g CTAB in 100 ml distilled water were
3
then added to the solution obtained above with constant
stirring to obtain a slurry. The resulting slurry was aged
overnight in the mother liquor. Then, the resultant solution
was filtered and the residue was washed with distilled
water till free from chloride ions. The obtained powders
were dried in an oven at 100 °C for 5 h and calcined at
various temperatures (in the range of 450–750 °C) for 6 h
In addition to the borrowing hydrogen strategy, another
approach for N-alkylation of amines is direct alkylation by
nucleophilic substitution of alcohols using solid acids cat-
alysts such as nanosized zeolite beta [23], Al-MCM-41
Á
in ambient air. The molar ratio of ZrOCl 8H O:H BO :-
2
2
3
3
2
?
[
24] and Co -zeolites [25] and sulfated tungstate [26]
CsNO :CTAB is 1:X:1:0.2, where X represents the added
3
which have been studied for solvent free direct N-alkyla-
tion amines using alcohols to establish an efficient and
greener catalytic synthesis route. Recently, acid–base bi-
functionalized catalysts [27–29] have attracted a great deal
of attention in the field of catalysis, such as Cs–P–Si mixed
oxide as a cooperative acid–base catalyst for the direct N-
alkylation reaction with alcohols and obtained good yields
in synthesizing different alkylaminoalcohols [30]. It is also
found that weak acid–base pair sites played a vital role in
the reaction; however, high catalytic activity and selectiv-
ity were obtained only under the supercritical fluids state.
amount of H BO , it was taken as 0.5, 1.0, 1.5, and
3
3
2.0,respectively. The synthesized samples using molar ratio
of Cs:B:Zr = 1:0.5:1, 1:1:1, 1:1.5:1 and 1:2:1 are denoted
as Cs–B–Zr/0.5, Cs–B–Zr/1, Cs–B–Zr/1.5 and Cs–B–Zr/2,
respectively.
2.2 Characterizations
XRD patterns were recorded on an X’Pert PRO X-ray
diffractometer produced by Dutch PANalytical company
(CuKa radiation k = 0.1541 nm, 40 kV, 40 mA) with a
scan speed of 8°/min and a scan range of 1°–10° and 17°–
80°. The nitrogen adsorption–desorption was performed by
using an ASAP 2020 Micromeritics, USA, after the degas-
sing of samples under vacuum (-0.1 MPa) at 200 °C for
3 h, measurements at -196 °C. Surface area was deter-
mined by Brunauer–Emmett–Teller (BET) equation. The
pore-size distribution, based on the desorption branch of the
isotherm, was estimated using the method developed by
Barret, Joyner, and Halender (BJH), assuming a cylindrical
As known to all, ZrO with the multiphase crystal struc-
2
ture, high thermal stability and amphoteric properties is
widely applied in catalysis [31]. Furthermore, the acid–
base properties of catalyst could be varied by doping
acidic/alkaline oxide to facilitate the ability of adsorbing
organic molecules [32, 33].
In this work, the solid acid–base Cs–B–Zr mixed oxides
with mesoporous structure were synthesized using the co-
precipitation method followed by a subsequent thermal
1
23

N-Alkylation of Alkylolamines with Alcohols Over Mesoporous Solid Acid–Base Cs–B–Zr Catalyst
pore modal. The inductively coupled plasma-optical emis-
sion (ICP-OES) spectrophotometer (Optima 2000DV, Per-
kin–Elmer, Eden Prarie, MN) was used to determine the
composition and the doping amount of element in the syn-
thesized materials. The sample material was digested in
concentrated HF and diluted with water. A scanning elec-
tron microscope (SEM) equipped with HIT S-4700 was used
to detect the morphology of samples. Transmission elec-
tronic microscope (TEM) images were acquired by using a
Philips-FEI Tecnai G2F30 S-Twin microscope. FT-IR was
detected by Nicolet205 type Fourier transform infrared
spectrometer. NH /CO -TPD measurements were con-
3 Results and Discussion
3.1 XRD of Catalysts
Figure 1a showed wide-angle XRD patterns of ZrO , B–Zr
2
and Cs–B–Zr/1. The pattern of ZrO which calcined at
2
550 °C, presented a pure tetragonal crystalline phase
(JCPDS, CAS No. 00-014-0534). With the introduction of
borate, the peaks in the B–Zr and Cs–B–Zr/1 were lower
and broader than pure ZrO2 which indicated that
3
2
ducted by using a Micromeritics Au-tochem-II instrument,
and for this 200 mg of Catalyst sample was placed in a U
shaped sample tube. The sample was pre-treated in helium
(a)
Cs-B-Zr
3
-1
(
40 cm min ) for 0.5 h at 250 °C then cooled down to
00 °C. NH and CO gases were used to study acidic and
1
3
2
basic strength respectively. Corresponding gas was adsor-
bed on the sample for 1 h under 1.33 MPa. The physisorbed
gas was removed by flashing the samples with helium for
B-Zr
ZrO2
3
0 min. TPD measurements were conducted in the tem-
perature range from 100 to 700 °C. The graph was recorded
using a TCD detector.
20
30
40
50
60
70
80
2
θ/degree
2
.3 Catalytic Activity Measurements
m-ZrO
(
b)
2
t-ZrO
2
N-Alkylation reaction was performed in a volume of
Cs2O
o
750 C
5
2
00 ml stainless-steel autoclave batch reactor. A mixture of
-aminoethanol (0.5 mol) with methanol (1.5 mol) was
o
6
50 C
added to the reactor containing 4 wt% of catalyst (with
respect to 2-aminoethanol). The N-alkylation reaction was
conducted under the constant stirred with a speed
o
5
50 C
-
1
2
00 r min
pressure in the reaction kettle by adding nitrogen) for
5 min. The reaction products were identified by GC–MS
at 220 °C and 4.3 MPa (To maintain the
o
4
50 C
3
20
30
40
2
50
θ/degree
60
70
80
analysis (Agilent 5973N-6890N, Agilent Technologies).
The selectivity and chemical conversions of the products
were determined by GC analysis (GC-17A, Shimadzu Co;
FID detector and DB-1 ms capillary column, J&W). The
conversion of amino alcohol was calculated on the basis of
its weight percent as follows:
(c)
Cs-B-Zr/1.5
Cs-B-Zr/1
Conversionð%Þ ¼ 100
Initial wt% of amino alcohol À Final wt% of amino alcohol
Cs-B-Zr/0.5
B-Zr
Â
:
Initial wt% of amino alcohol
ZrO2
The selectivity of the product was calculated as:
2
4
6
8
Selectivityð%Þ ¼ 100 %
2θ/degree
Peak area of product
Total Peak area for all the products
 P
:
Fig. 1 a Wide angle XRD patterns for the synthesized samples
calcined at 550 °C. b Wide angle XRD patterns for the synthesized
Cs–B–Zr/1 mixed oxide calcined at various temperatures. c Small
angle XRD patterns for the synthesized samples calcined at 550 °C
for 6 h
Qualitative analysis was compared with standard chro-
matographic and mass spectrometry of the product.
123

A. Chen et al.
amorphous state and low crystal morphology were formed
in the framework of B–Zr and Cs–B–Zr/1, respectively. It
could be attributed to the smaller particles and grain size in
the structure of B–Zr and Cs–B–Zr/1. Previous studies
found that with incorporation of the borate ion in the ZrO2
system suppressed the crystallization and led to the for-
mation of poor crystal structure and smaller particle size
materials [32]. Figure 1b presented the XRD patterns of
Cs–B–Z/1 sample at different calcined temperatures. The
broad nature of these diffraction spectra at 450 and 550 °C
suggested the low crystalline nature in the structure of
samples. With the rising of temperature, the crystallinity of
the sample was increased and typical diffraction peaks for
isotherms for these materials could be categorized as type I
with borderline type V characteristics and a type H3 hys-
teresis loop, indicating the presence of mesoporous struc-
ture [34, 35] with a pore size distribution at 5–50 nm.
Furthermore, type H3 hysteresis were usually only
observed on solids containing aggregates or agglomerates
of particles, which could lead to slit-shaped pores that were
non-uniform in size and shape. The composition and tex-
tural properties of the different catalysts were summarized
in Table 1. From the ICP results, the molar ratios of
Cs:B:Zr in the catalysts determined by chemical analysis
were close to those added in the reactant mixtures during
the synthesis. The incorporation of B and Cs into the mixed
oxide catalyst system also resulted in significant improve-
ments in surface area. The Cs–B–Zr/1 showed the largest
tetragonal metastable t-ZrO were appeared. The diffrac-
2
tion peaks of Cs–B–Zr/1 at 750 °C were added with several
new peaks at 25.93°, 28.22°, 42.52° and 45.42°, which
2
-1
surface area (ABET = 227 m g ). When increasing
doping amount of boron, Cs–B–Zr/1.5 and Cs–B–Zr/2
samples showed smaller surface areas (189 and
were assigned to hexagonal phase of Cs O (JCPDS, CAS
2
No. 00-09-0104). While the new peaks at 23.98°, 28.08°
2
-1
and 31.32° were assigned to monoclinic m-ZrO crystalline
152 m g ) than Cs–B–Zr/1 and the average pore diam-
eter decreased from 21 to 31 nm. It is believed that the
excess boric acid decomposed to form boron oxide parti-
cles, and the pores were blocked by the boron oxide par-
ticles which led to the decreasing of the surface and the
pore size of the catalyst [36, 37].
2
phase, namely, a polycrystalline phase was formed in the
Cs–B–Zr/1 sample. Figure 1c showed small angle XRD
diffraction patterns of mixed oxide samples and ZrO . The
2
Cs–B–Zr/1 sample calcined at 550 °C showed an obvious
single broad peak at around 2h = 1.9°–3.5° which implied
the formation of mesoporous in the Cs–B–Zr mixed oxide.
3.3 Textural and Surface Properties of Catalysts
3
.2 ICP and BET Analysis of Catalysts
Figure 3 showed the SEM and TEM images of Cs–B–Zr/1
mixed oxide calcined at 550 °C. As shown in Fig. 3a, the
SEM of the Cs–B–Zr/1 sample revealed the presence of
uniform spherical particles structure on the surface of the
catalyst with an average diameter in the range of
30–40 nm. The TEM image in Fig. 3b demonstrated that
the existence of worm-like pores in the spherical particles
and uniformly dispersed on the catalyst surface.
Figure 2 showed the N adsorption–desorption isotherm
2
and pore size distribution of Cs–B–Zr mixed oxide cal-
cined at 550 °C. The N isomer for Cs–B–Zr/1 presented
2
three adsorption stages (P/P \ 0.2, 0.2 \ P/P \ 0.7 and
0
0
P/P [ 0.7). Based on the IUPAC classification, the
0
2
1
1
40
80
20
0
0
0
0
0
.4
.3
.2
.1
.0
3
.4 FT-IR Spectra of Synthesized Oxides
Figure 4 showed the FT-IR spectra of ZrO , Cs–B–Zr/1
2
and B–Zr calcined at 550 °C. The IR spectra of samples
Cs–B–Zr/1 and B–Zr showed strong absorption bands in
-
the range of 1290–1440 cm with a peak at 1379 cm
1
-1
,
0
20
40
60
80
100
corresponding to tetrahedral [BO ] unit. Simultaneously,
4
-1
the peaks at 1190 cm were assigned to trigonal [BO3]
Pore Diameter (nm)
6
0
0
unit [38, 39], which confirmed that two units (tetrahedral
BO and trigonal BO ) existing in the synthesized Cs–B–
4
3
-
1
Zr/1 [40]. The peaks at 1005 and 900 cm in the syn-
thesized Cs–B–Zr/1, which was absent in the spectrum of
0
.0
0.2
0.4
0.6
0.8
1.0
pure ZrO and B–Zr mixed oxide were assigned to char-
2
Reletive Pressure (P/P0)
acteristic peaks of Cs O [33]. In addition, the peak at 3424
2
-
1
and 1626 cm
in all samples were assigned to the
Fig. 2 N
sponding pore-size distribution of Cs–B–Zr/1 mixed oxide calcined at
50 °C
2
adsorption–desorption isotherm. The inset is the corre-
stretching and bending vibration of hydroxyl groups and
water molecules. In the spectrum of ZrO2, two
5
1
23

N-Alkylation of Alkylolamines with Alcohols Over Mesoporous Solid Acid–Base Cs–B–Zr Catalyst
Table 1 The composition and
textural properties of the
different catalysts
a
2
-1 b
)
3
V (m /g)
c
d
Entry
Samples
ZrO
Cs:B:Zr
A
BET (m g
D (nm)
1
2
3
4
5
6
a
2
–
106
224
207
227
189
152
0.14
0.12
0.15
0.18
0.16
0.08
52.83
21.43
28.99
31.72
23.86
21.05
B–Zr
0.91:1
Cs–B–Zr/0.5
Cs–B–Zr/1
Cs–B–Zr/1.5
Cs–B–Zr/2
0.97:0.47:1
1.03:0.97:1
1.07:1.43:1
1.00:1.80:1
The ratio of Cs:B:Zr by ICP-OES detection under the conditions of calcination temperature at 550°
BET surface area
b
c
d
Single point adsorption total pore volume at P/P
Average pore diameter (4 V/A by BET)
0
= 0.97
Fig. 3 Images of Cs–B–Zr/1 mixed oxide calcined at 550 °C. a SEM and b TEM
-
626 cm , assigned to physisorbed and coordinated
1
1
water, were also observed.
Cs-B-Zr/1
B-Zr
3.5 Catalytic Activity
[
BO3]
[BO4]
3.5.1 Catalytic Activity of Different Catalyst for N-
alkylation of 2-Amino Ethanol with Methanol
ZrO
2
The catalytic performance of the reaction of 2-aminoethanol
with methanol was tested using 4 wt% of catalyst at 220 °C
and 4.3 MPa in Scheme 1. The screening tests of the effec-
tive catalysts were shown in Fig. 5. It presented the results of
2-aminoethanol reacting with methanol over various cata-
3
500
3000
2500
2000
1500
1000
500
lysts. ZrO catalyst provided poor catalytic activity with
2
Wavenumber/nm-1
7.3 % selectivity of N-methylated products at 12.1 % con-
version. B–Zr and Cs–Zr catalysts also exerted unsatisfac-
tory selectivity (11.9 and 3.4 %, respectively)and a moderate
conversion (34.1 and 52.4 %, respectively). Nevertheless,
Cs–B–Zr catalyst gave the corresponding N-alkylation
products in moderate to excellent yields for 2-aminoethanol
reaction with methanol. Surprisingly, Cs–B–Zr/1 was the
best catalyst with the selectivity in the products reached up to
Fig. 4 FT-IR spectra of the synthesized samples calcined at 550 °C
-
1
characteristic peaks at 750 and 620 cm were assigned to
the monoclinic phase ZrO [41]. However, the intensities
2
of two bands in Cs–B–Zr/1 and B–Zr corresponding to
ZrO were decreased because of the existence of [BO ] and
2
4
96.3 % with 93.2 % conversion.
[
BO ] units. Additionally, the bands at 3450 and
3
123

A. Chen et al.
Scheme 1 N-Alkylation reaction of 2-aminoethanol with methanol over synthesized catalysts
1
00
100
removed completely after calcination at 550 °C. Never-
theless, the catalytic activity sharply decreased with the
temperature further increased over 550 °C. It is believed
that with calcination temperature increasing, the formation
of the mesoporous structure will collapse and result in
lower specific surface area which is responsible for
reducing the catalytic activity [42]. XRD and BET results
confirmed that the doped amount of boron had an impact
on the specific surface area and pore size, which directly
affected the catalytic performance of the catalysts [43].
Selectivity (%)
Conversion(%)
8
6
4
2
0
0
0
0
0
80
60
40
20
0
2
/
0.5
r/
1
/
1.5
/
2
ZrO
B-Zr
3.5.3 Effect of Reaction Temperature and Reaction
Pressure
Cs-Zr
-B-Zr
Cs
Cs-B-Z Cs-B-Zr
Cs-B-Zr
Fig. 5 Reaction of HO (CH
2
)
2
NH
2
with CH
3
OH over various solid
catalysts. Conditions: The reaction was conducted at 220 °C and
.3 MPa in a bath reactor with HO (CH NH to CH OH molar ratio
of 1:3 for 35 min and 4 wt% of catalyst with respect to amine was
added
Figure 7 showed the catalyst performance of the ternary
Cs–B–Zr/1 mixed oxide catalyst which was strongly
affected by reaction temperature and reaction pressure. As
shown in Fig. 7a, the conversion and selectivity were
substantially improved when the temperature was above
4
2
)
2
2
3
3
.5.2 Effect of the Calcined Temperature of Cs–B–Zr/1
100 °C, which could reach the highest selectivity of
93.2 % at 220 °C and achieve maximum conversion of
98.4 % at 240 °C. It is well known that the reaction acti-
Figure 6 showed the effect of calcination temperature on
Cs–B–Zr/1 catalytic performance. At the range of
vation energy of the substrate molecular was beyond reach
of when the temperature was low. Therefore, the reactions
proceed slowly at low temperature, which led to a low
conversion. When the temperature further increased
3
00–550 °C, with the calcined temperature increased, the
catalytic activity was gradually improved and achieved the
maximum activity at 550 °C. Less activity with sample
calcined at temperature less than 550 °C was possibly due
to incorporation of CTAB in the sample leading to no
formation of a large number of pore structure, which
(
T [ 220 °C), the selectivity was sharply declined, illus-
trating side reactions following with the reaction. At higher
reaction temperature, the decrease in selectivity was
mainly caused by the formation of poly-alkylated products,
such as, 2-methoxyethylamine; the dehydration of alcohols
or the isomerization of products including the intra- or
intermolecular dehydration compounds as byproducts [30].
As shown in Fig. 7b, the N-alkylation product yield was
also strongly influenced by the reaction pressure. At
1
00
Selectivity (%) 100
Conversion(%)
8
6
4
2
0
0
0
0
0
80
60
40
20
0
0.1–2 MPa, the pressure had little effect on selectivity and
conversion (both below 10 %). Increasing the pressure
from 2.0 to 4.5 MPa, the conversion and selectivity of the
N-alkylated products were markedly increased. Noticeably,
pressure around at 4.5 MPa, Cs–B–Zr/1 catalyst gave the
optimum catalytic activity for N-alkylated products with
the best conversion of 98.4 % and a total yield of 90.0 %.
The conversion and selectivity of N-alkylation was
observed with a significantly decrease with the pressure
further increased over 4.5 MPa. High pressure is helpful
for liquefying the reactants (which are low boiling point
3
00
400
500
600
700
800
900
Temperature/ºC
Fig. 6 Effect of the calcined temperature of Cs–B–Zr/1 mixed oxide
on the conversion and selectivity under 220 °C and 4.3 MPa for
3
5 min. Cs-B-Zr/1 was added 4 wt% of catalyst with respect to amine
1
23

N-Alkylation of Alkylolamines with Alcohols Over Mesoporous Solid Acid–Base Cs–B–Zr Catalyst
100
a)
2-ethanolamine (less than 1:3) caused a serious drop in
conversion (Table 2, entries 10–11) although excess
amount of the alcohol was used as a reactant. It could be
seen that the N-alkylated product selectivity enhanced with
decreasing the ratio of amines to methanol from 1:1 to 1:3
and then reached a higher level about 96.3 %. The reason
was that high quantity of 2-ethanolamine was responsible
for the reduction of selectivity leading to predominantly
intermolecular dehydration products, such as piperazine or
triethylenediamine [31], which gave poor selectivity in N-
alkylated product.
(
80
60
40
20
0
Conversion(%)
Selectivity(%)
0
50
100
150
200
250
300
350
3.5.5 Effect of Reaction Substrate
Temperature/ºC
The substrate scope of N-alkylation over the Cs–B–Zr
mixed-oxide catalysts was investigated at 220 °C and
1
00
(b)
4.3 MPa with an amine:alcohol molar ratio of 1:3. Table 3
8
6
4
2
0
0
0
0
0
showed the effect of the substitution in the methanol
reaction with different functionalized amines in the opti-
mized reaction conditions. Various functionalized amines
proceeded to provide the corresponding N-alkylated prod-
ucts with different selectivity. Noticeably, it was found that
the product yield was influenced by the structures of the
amines. An increase in the number of the methylene units
in the amino alcohols caused remarkable decrease in the
reactivity following the order of 2-aminoethanol [ 3-
aminopropanol [ 5-aminopentanol. Although 4-aminobu-
tanol exhibited good conversion (Table 3, entry 3), it gave
predominantly the cyclization product, pyrrolidine, owing
to thermodynamic reasons that a high reaction temperature
favor the formation of thermodynamically stable six
membered cyclization products. It is well-known that
molecular dehydration cyclization reaction would be
occurred for the alcohol amine to form a ring compound,
which made N-alkylation product selectivity declined
greatly. In sharp contrast to the reactivity of bifunctional-
ized amines, simple amines such as n-butylamine (Table 3,
entry 8) or aniline (Table 3, entry 9) were less reactive. It
gave the N-alkylated products with excellent selectivity
regardless of their structures, but with low conversion
(10–20 % for the amines). For the linear bifunctionalized
amines, the highest conversion and selectivity achieved
93.2 and 96.3 %, respectively (Table 3, entry 1). The
results strongly suggested that the hydroxyl group in the
functionalized amines was a crucial structural factor to
attain excellent reactivity and selectivity.
Conversion(%)
Selectivity (%)
0
2
4
6
8
10
12
Pressure/MPa
Fig. 7 The conversion and N-alkylation selectivity under different
conditions: a 4.3 MPa for 35 min. b 220 °C for 30 min. Conditions:
Cs–B–Zr/1 mixed oxide calcined at 550 °C, 4 wt% of catalyst with
2 2 2 3
respect to amine was added, HO(CH ) NH :CH OH = 1:3
liquids) at high temperature reaction (220 °C) which pro-
vide more contact time with catalyst and for adsorption and
reaction. Further increase in pressure i.e., after optimum
pressure, the reaction was hindered due to detrimental
effect on product desorption.
3
.5.4 Effect of the Molar Ratio of the Reaction Substrate
The molar ratio of substrate molecules had also a direct
impact on the reaction rate, conversion and selectivity for
ethanolamine N-alkylation. Table 2 summarized the data of
the investigation of different molar ratio of amine alcohol
reaction with alcohol. The selectivity of N-alkylated pro-
duct was improved by increasing the 2-aminoethanol:
methanol molar ratio from 1:1 to 1:20 (73.3–97.4 %,
Table 2, entries 1–11). The best conversion result was
obtained when the molar ratio of 2-ethanolamine: metha-
nol = 1:3. Cs–B–Zr/1 Catalyst afforded a maximum value
of 93.2 % (Table 2, entry 8). Low molar ratio of
Table 4 showed the effect of the substitution in the
2-aminoethanol reaction with different alcohols in the
optimized reaction conditions. A wider range of alcohols,
including primary, secondary and tertiary had been sub-
stituted with 2-aminoethanol over Cs–B–Zr/1 catalyst. As
can be seen from the Table 4, Cs–B–Zr/1 showed the
diversity of catalytic effect on different structure alcohol.
123

A. Chen et al.
Table 2 Effect of the molar
ratio of of reaction substrate
over the Cs–B–Zr catalyst
Entry
Amine:alcohol (molar ratio)
Conv. (%)
N-Alkylation products sel. (%)
1
2
3
4
5
6
7
8
9
1:1
53.6
64.9
69.1
76.7
83.6
89.3
90.7
93.2
90.7
74.7
66.5
73.3
75.1
79.8
85.2
88.4
89.7
92.5
96.3
96.9
97.4
97.3
1:1.1
1:1.2
1:1.3
1:1.5
1:2
1:2.5
1:3
1:5
1
1
0
1
1:10
1:20
Conditions: The reaction was conducted at 220 °C and 4.3 MPa in a bath reactor for 35 min and 4 wt% of a
catalyst with respect to amine was added over the Cs–B–Zr/1 Catalyst
Table 3 N-Alkylation of
various amines over the Cs–B–
Entry
1
Amine
Conv. (%)
93.2
N-Alkylation products sel. (%)
3
Zr Catalyst with CH OH
OH
96.3
NH2
NH2
2
3
92.6
91.7
95.0
17.4
OH
OH
NH2
NH2
4
5
83.9
71.3
92.6
87.8
OH
OH
NH2
NH2
6
7
52.6
57.4
92.7
92.3
OH
NH2
OH
8
9
19.4
15.3
94.0
93.6
NH2
NH2
Conditions: The reaction was conducted over the Cs–B–Zr catalyst at 220 °C and 4.3 MPa in a bath reactor
with an amine to CH
added
3
OH molar ratio of 1:3 for 30 min and 4 wt% of catalyst with respect to amine was
The conversion and selectivity of linear alcohols could
reach up around 90 % (Table 4, entries 1–6). While the
branched alcohols were relatively low exhibiting compar-
atively good reactivity and selectivity (77.3 and 89.8 %
respectively, Table 4, entry 7). It was found that tertiary
alcohols had the worst reactivity over Cs–B–Zr/1 catalyst
with poor conversion (32.0 %, Table 4, entry 8). But, the
secondary alcohols as the source of the electrophile could
give the corresponding tertiary aminoethanol with moder-
ate conversion (77.3 %, Table 4, entry 7). Primary alcohols
such as methanol were performed under the same
conditions giving the preferable reactive activity with
96.3 % selectivity and at 93.2 % conversion (Table 4,
entry 1). Moreover, steric hindrance also had great effect
on the N-alkylation reaction process. The marked steric
effect of the alkyl substituent on the amino group was
demonstrated by the decrease in the reactivity in the order
of–H [ –CH [ –C H [ –CH (CH ) [ –C (CH ) . The
catalytic effect on aromatic alcohols such as phenol, benzyl
alcohol and cyclohexanol (Table 4, entries 10–12) were far
less than aliphatic alcohols, wherein the conversion and
selectivity of cyclohexanol were only 15.9 and 21.5 %
3
2
5
3 2
3 3
1
23

N-Alkylation of Alkylolamines with Alcohols Over Mesoporous Solid Acid–Base Cs–B–Zr Catalyst
Table 4 N-Alkylation of
Entry
Alcohol
HO
HO
Conv. (%)
N-Alkylation products sel. (%)
various alcohols over Cs–B–Zr
Catalyst with HO(CH NH
2
)
2
2
1
2
93.2
96.3
95.4
94.0
3
4
94.3
92.6
95.7
92.4
HO
HO
HO
HO
5
6
7
93.8
89.5
77.3
89.1
91.2
89.8
HO
8
9
32.0
77.4
HO
45.6
24.6
33.2
82.3
46.2
63.8
HO
1
1
0
1
HO
HO
HO
a
2
1
15.9
21.5
Conditions: The reaction was conducted over the Cs–B–Zr catalyst at 220 °C and 4.3 MPa in a bath reactor
with HO (CH NH to an alcohol molar ratio of 1:3 for 30 min and 4 wt% of catalyst with respect to
amine was added
)
2 2
2
a
The molar ratio of HO(CH
2
)
2
NH
2
to alcohol is 1:2
respectively (Table 4, entry 12). Following the above
results, the N-alkylation process had been greatly affected
by structure of substrate alcohol.
basic sites. In contrast, the MT desorption signals of Cs–B–
Zr/1 sharply increased with adding impregnated of boron
and cesium suggesting the presence of weak basic sites on
the surface of Cs–B–Zr/1. Meanwhile, the peak area of Cs–
3
.6 Acid–Base Property of the Catalysts
B–Zr/1 was much higher than ZrO in the same tempera-
2
ture region, which revealed the incorporation of cesium in
the mixed oxides system enhancing the total basicity of
Cs–B–Zr/1 (Table 5). Borate species do not have any
contribution to the basicity except for the fact that
increased surface area leads to an increase in basicity [37].
The NH -TPD profiles of the oxide catalyst ZrO and Cs–
The CO /NH -TPD was performed to investigate the
3
2
impact of the catalyst on the N-alkylation reaction process.
The CO -TPD profiles of the oxide catalyst ZrO and Cs–
2
2
B–Zr/1 were shown in Fig. 8a. ZrO showed a broad CO
2
2
desorption peak, including four peaks centered at 250, 425,
3
2
5
10, and 540 °C, which indicated that the surface basic
strengths were not homogeneously distributed. In general,
B–Zr/1 were shown in Fig. 8b. ZrO showed two NH
2 3
desorption peaks at 210 and 410 °C, respectively, which
illustrated both the weak acid sites and moderate acid sites
on the surface of ZrO . Cs–B–Zr/1 showed only a NH
the broad CO peaks were divided into two types of des-
2
orption peaks,the first broad medium temperature (MT)
2
3
desorption signal in the range 100–300 °C, corresponding
desorption peak at 200 °C suggesting only the weak acid
sites on the surface of Cs–B–Zr/1. Furthermore, the peak
to CO adsorbed on the basic sites at weak strengths and
2
the second broad high-temperature (HT) peak in the range
area of Cs–B–Zr/1 was significantly greater than ZrO in
2
of 300–600 °C, suggesting the presence of very strong
the 170–500 °C region indicating the total acidity of Cs–
123

A. Chen et al.
negative charge of boron is diffused into boron oxide bulk
by the resonance between the lone pair of oxygen and the
empty orbital of boron enhancing the total acidity of the
catalyst responsible for better catalytic activity [44].
Table 5 showed total acid amount and basic amount of the
synthesized catalysts. It revealed that the total acidity and
basicity of the Cs–B–Zr/1 catalyst was improved compared
(a)
Cs-B-Zr/1
to other catalyst samples. The acidity increased from
-2
ZrO2
0
.0025 to 0.0875 mmol m and basicity from 0.0153 to
-
.0850 mmol m . The surface area of the samples also
2
0
2
-1
increased from 106 to 227 m g because of borate and
cesium modification (vide supra).The acid–base character
of Cs–B–Zr catalyst was directly correlated to the surface
area. It increased with surface area because of the incre-
ment of accessible active sites, which is responsible for
improving the catalytic activity. According to Fig. 5 the
1
00
200
300
400
500
600
Temperature/ºC
(b)
results, it could be found that ZrO , B–Zr and Cs–Zr
2
exerted unsatisfactory selectivity catalytic activity, how-
ever, higher catalytic activity could be obtained over Cs–
B–Zr/1 catalyst, which demonstrated that the weak strength
acid-basic sites of Cs–B–Zr/1 was much more than other
catalysts having a great impact on the conversion and
selectivity, which was evidenced by its excellent catalytic
performance in the corresponding N-alkylation reaction.
Cs-B-Zr/1
ZrO2
3.7 Roles of Cs–B–Zr in N-Alkylation Synthesis
Process
1
00
200
300
400
500
600
It is reported that the traditional N-alkylation process was
employing the borrowing hydrogen strategy [45–47], and
some intermediate products, such as aldehyde or imine
would appear in the reaction process. Surprisingly, we did
not detect any similar intermediate products in our reaction
process; thus the reaction mechanism might differ from ‘‘B-
H’’ process. Plausible mechanism of the reaction was also
illustrated in Scheme 2. (i) The N-alkylation reaction pro-
ceeded through selective adsorption of alkylamines by the
aid of hydroxyl group on the acid–base pair site of the cat-
alyst to lead to the aminoalkyl ester species [48], and alco-
hols formed its alkyl ester on the neighboring basic and
acidic sites in a similar manner; (ii) The amino group in the
aminoalkyl ester moieties would exclusively attack the alkyl
carbon of the neighboring alkyl ester species on the surface
because of the high concentration of alcohols on the catalyst
surface. The solvation or local clustering around amino
functional group in alcoholamine with alcohol molecules
might hinder the intermolecular condensation, leading to the
other byproducts [49, 50]; (iii) N-alkylation was performed
on the aminogroup with alkyl moieties through deprotona-
tion of the amino groups by aid of the basic sites, and the
proton on the amino combined with hydroxyl to generate
neutral water; (iv) The deprotonation of aminoalkyl ester
specie rapidly desorbed from the surface of catalyst in
Temperature/ºC
Fig. 8 Patterns of the synthesized materials calcined at 550 °C.
a CO -TPD; b NH -TPD
3
2
Table 5 Total acid amount and basic amount of the synthesized
materials
-
Quantity of desorbed gas (910 mmol/m )
4
2
Entry Samples
CO
2
NH
3
1
2
3
4
5
6
ZrO
2
153
235
25
B–Zr
51.5
46.2
87.5
67.1
51.9
Cs–B–Zr/0.5 468
Cs–B–Zr/1 850
Cs–B–Zr/1.5 672
Cs–B–Zr/2 539
The synthesized materials were calcined at 550 °C and determined by
CO /NH -TPD experiment
2
3
B–Zr/1 was improved (Table 5) by reason of impregnating
of boron. This is because that the [BO ] unit is responsible
3
for the enhancement of acidity, which has an empty orbital
and pulls electron cloud from the oxygen of ZrO . The
2
1
23

N-Alkylation of Alkylolamines with Alcohols Over Mesoporous Solid Acid–Base Cs–B–Zr Catalyst
Scheme 2 Proposed reaction
pathway of the acid–base
synergetic effect for
dehydration on the Cs–B–Zr
catalyst surface
conjunction with proton to form alkylamine alcohol. The
acid–base pair site could facilitate the first step leading to a
better activation of the substrates by the facilitation of the
nucleophilic attack at the alkyl ester species in an acid–base
cooperative effect. It was the reason why the highest yield of
N-alkylation of products appeared in the case of Cs–B–Zr
mixed oxide (Table 3, entries 1–2). In sharp contrast to the
reactivity of bifunctional amines, amine bearing no free
hydroxyl group (Table 3, entries 8–9) was less reactive;
suggesting that the hydroxyl group of the amino alcohols
was a structural factor crucial to determining the high
reactivity and selectivity. The amines without hydroxyl as
an anchor on the catalyst surface resulted in great differ-
ences in the conversion. It is worth noting that higher con-
version was obtained from aliphatic amines (Table 3,
entries 1–2), not aromatic amine (Table 3, entries 6–7),
which both bore amino and hydroxyl groups. It was due to
the existence of steric effect in the structure of aromatic
amine. Meanwhile, the aromatic group substituents on the
amino showed poor nucleophilic substitution performance
ordinarily. It was also found that the structure of substrate
alcohol (Table 4) was important factor affecting the cat-
alytic activity of the Cs–B–Zr catalyst for the N-alkylation
of alkylolamines with alcohols. It is known that a nucle-
entries 1–6). So, secondary and tertiary alcohol (Table 4,
entries 7–8) was shown less reactivity. Further studies are
essential to gain a better and clearer insight into the mech-
anism on the overall reaction in our later work.
4
Conclusions
We have demonstrated that Cs–B–Zr mixed oxide catalyst
works as an efficient catalyst for the N-alkylation of
alkylolamines with alcohols providing a very clean pro-
cess. Cs/B/Zr ratio, calcination temperature, reaction con-
ditions, and reaction substrate had great influence on the
catalytic activity of N-alkylation of alkylolamines with
alcohols. The mechanistic model adequately rationalized
the observed reactivity and selectivity. It was revealed that
the suitable surface weak acid–base sites, strong interaction
between hydroxyl group and the Cs–B–Zr were critical
factors leading to excellent catalytic behavior and activity
for the N-alkylation reaction. The ternary Cs–B–Zr mixed
oxide is an attractive alternative catalyst as a cheap and
commercially available catalyst in environmentally benign
chemical process.
ophilic substitution of –OH of alcohol by –NH of amines
2
Acknowledgments The authors are grateful for the National Nat-
ural Science Foundation of China (Nos. 51002137, 10804099) and the
Zhejiang Basic Research Development Program (LQ12E02009).
substrate via SN2 mechanism, and SN2 transition state
would be favored if alcohol is of primary nature (Table 4
123

A. Chen et al.
2
5. Sun Y, Lu X, Wei X, Zhou D, Xia Q (2014) Catal Commun
4
References
3:213
2
2
2
6. Pathare S, Akamanchi K (2013) Appl Catal A Gen 452:29
7. Lee YY, Kevin WCW (2012) Phys Chem Chem Phys 14:13914
8. Peng WH, Lee YY, Connie W, Kevin WCW (2012) J Mater
Chem 22:23181
9. Ivanova I, Knyazeva EE (2013) Chem Soc Rev 42:3671
0. Oku T, Arita Y, Tsuneki H, Ikariya T (2004) J Am Chem Soc
1
2
3
4
5
6
. Lawrence S (2004) Amines: synthesis, properties and application.
Cambridge University Press, Cambridge
. Seayad A, Ahmed M, Klein H, Jackstell R, Gross T, Beller M
(2002) Science 297:1676
2
3
. Brown B (1994) The organic chemistry of aliphatic nitrogen
compounds. Oxford University Press, New York, pp 350–352
. Yin H, Jin M, Chen W, Chen C, Zheng L, Wei P, Han S (2012)
Tetrahedron Lett 53:1265
. Lombardo M, Mosconi E, Pasi F, Petrini M, Trombini C (2007) J
Org Chem 72:1834
. Nakano H, Yokoyama J, Fujita R, Hongo H (2002) Tetrahedron
Lett 43:7761
. Kashima C, Harada K, Omote Y (1985) Can J Chem 63:288
. Chakraborti A, Kondaskar A (2003) Tetrahedron Lett 44:8315
. Pujala B, Rana S, Chakraborti A (2011) J Org Chem 76:8768
1
26:7368
3
3
3
3
3
3
3
1. Qiu C, Lin T, Zhang Q, Xu HD, Chen Y, Gong M (2011) Chin J
Catal 32:1227
2. Malshe K, Patil P, Umbarkar S, Dongare M (2004) J Mol Catal A
Chem 212:337
3. Kong X, Wang G, Du X, Lu L, Li L, Chen L (2012) Catal
Commun 27:26–29
4. Liang Q, Hu Q, Miao G, Yuan B, Chen X (2015) Mater Lett
7
8
9
1
48:45
5. Dayananda D, Sarva V, Prasad S, Arunachalam J, Parameswaran
P, Ghosh N (2015) Appl Surf Sci 329:1
6. Panda A, Sinhamahapatra A, Sutradhar N et al (2011) Appl Catal
A Gen 402:87
7. Panda A, Sinhamahapatra A, Pal P et al (2013) Chem Cat Chem
1
0. He W, Wang L, Sun C, Wu K, He S, Chen J, Wu P, Yu Z (2011)
Chem Eur J 17:13308
1. Cui X, Dai X, Deng Y, Shi F (2013) Chem Eur J 19:3665
2. Jumde V, Gonsalvi L, Guerriero A, Peruzzini M, Taddei M
1
1
(2015) Eur J Org Chem 2015:1829–1833
5
:331
1
1
3. Edwards M, Jazzar F, Paine B, Shermer D, Whittlesey M, Wil-
liams M, Edney D (2004) Chem Commun Camb 1:90
4. Grigg R, Mitchell T, Sutthivaiyakit S, Tongpenyai N (1981) J
Chem Soc Chem Commun 12:611
5. Watanabe Y, Tsuji Y, Ohsugi Y (1981) Tetrahedron Lett 22:2667
6. Hamid M, Allen C, Lamb G, Maxwell A, Maytum H, Watson A,
Williams J (2009) J Am Chem Soc 131:1766
7. Dang T, Ramalingam B, Shan SP, Seayad A (2013) ACS Catal
3
8. Ohshima T, Miyamoto Y, Ipposhi J, Nakahara Y, Utsunomiya M,
Mashima K (2009) J Am Chem Soc 131:14317
9. Yang H, Mao R, Luo C, Lu C, Cheng G (2014) Tetrahedron
3
3
4
8. Egili K, Oraby A (1996) J Phys Condens Matter 8:8959
9. Motke S, Yawale S, Yawale S (2002) Bull Mater Sci 25:75
0. Xu B, Cheng S (1999) S J, Zhu Q. Appl Catal A Gen
1
1. Srivastava A, Dongare M (1987) Mater Lett 5:111
88:361–368
1
1
4
4
2. Shen W, Tang X, Yong Li et al (2006) Appl Catal B Environ
6
2:265
3. Xu S, Hong Y, Chen C, Li S, Xiao L, Fan J (2013) J Mater Chem
A 1:6191
1
1
1
4
:2536
4
4
4
4. Dongare M, Madje B, Patil P et al (2004) Catal Commun 5:353
5. Xu Q, Li Q, Zhu X, Chen J (2013) Adv Synth Catal 355:73
6. Lu X, Sun Y, Wei X, Peng C, Zhou D, Xia Q (2014) Catal
Commun 55:78
7
0:8829
2
2
0. Gunanathan C, Milstein D (2008) Angew Chem Int Ed 47:8661
1. Gnanamgari D, Sauer E, Schley N, Butler C, Incarvito C, Crab-
tree R (2009) Organometallics 28:321
4
7. Bahn S, Imm S, Neubert L, Zhang M, Neumann H, Beller M
(
2011) ChemCatChem 3:1853–1864
4
4
8. Tsuneki H (2001) Appl Catal A Gen 221:209
9. Wang W, Seiler M, Ivanova I, Sternberg U, Weitkamp J, Hunger
M (2002) J Am Chem Soc 124:7548
0. Wang W, Buchholz A, Seiler M, Hunger M (2003) J Am Chem
Soc 125:15260
2
2
2. Kawahara R, Fujita K, Yamaguchi R (2010) J Am Chem Soc
1
32:15108
3. Reddy M, Kumar M, Swamy P, Naresh M, Srujana K, Satya-
narayana L, Venugopala A, Narender N (2013) Green Chem
5
1
5:3474
2
4. Tayade K, Mishra M, Munusamy K, Somani R (2014) J Mol
Catal A Chem 390:91
1
23