Zirconium phenyl phosphonate phosphite as a highly active, reusable, solid acid catalyst for producing fatty acid polyol esters

Article abstract of DOI:10.1016/j.apcata.2013.04.043

The application of zirconium phenyl phosphonate phosphite (ZrPP) as a solid acid catalyst for producing polyol esters by esterification of glycerol or trimethylolpropane with a fatty acid (C₈-C_18.1) is reported for the first time. ZrPP exhibits high catalytic activity and in particular, (di + tri) esters selectivity (92.3 mol%). These esters of polyols are known for their application as biolubricants. The catalyst prepared using phosphorous acid to phenyl phosphonic acid molar ratio of 3:1 was found superior. The influence of process parameters on activity and selectivity of the catalyst was investigated. ZrPP was reusable in at least three recycling experiments. Hydrophobicity due to exposed phenyl groups on the surface is the possible cause for superior esterification activity of this novel, solid catalyst.

Full text of DOI:10.1016/j.apcata.2013.04.043

Applied Catalysis A: General 462–463 (2013) 129–136
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Zirconium phenyl phosphonate phosphite as a highly active, reusable,
solid acid catalyst for producing fatty acid polyol esters
Poonam Varhadi, Mehejabeen Kotwal, D. Srinivas∗
Catalysis Division, National Chemical Laboratory, Pune 411 008, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The application of zirconium phenyl phosphonate phosphite (ZrPP) as a solid acid catalyst for producing
polyol esters by esterification of glycerol or trimethylolpropane with a fatty acid (C8–C18.1) is reported for
the first time. ZrPP exhibits high catalytic activity and in particular, (di + tri) esters selectivity (92.3 mol%).
These esters of polyols are known for their application as biolubricants. The catalyst prepared using
phosphorous acid to phenyl phosphonic acid molar ratio of 3:1 was found superior. The influence of
process parameters on activity and selectivity of the catalyst was investigated. ZrPP was reusable in at
least three recycling experiments. Hydrophobicity due to exposed phenyl groups on the surface is the
possible cause for superior esterification activity of this novel, solid catalyst.
Received 5 March 2013
Received in revised form 23 April 2013
Accepted 27 April 2013
Available online xxx
Keywords:
Polyol esters
Biolubricants
Esterification of fatty acid
Zirconium phenyl phosphonate phosphite
Hydrophobic solid acid catalyst
Fatty acid glycerides
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
acid-based process generates salt by-product and also huge amount
of waste-water. Replacement of mineral acids with solid acid cata-
Polyol esters are a family of synthetic lubricants used in high
temperature operations such as industrial oven chains, tenter
frames, stationary turbine engines, jet engine lubricants, high tem-
perature greases, fire resistant transformer coolants, fire resistant
hydraulic fluids and textile lubricants [1,2]. They are also used
with the new generation of chlorine-free hydrofluorocarbon refrig-
erants, such as R-420A [3]. Unlike mineral oils, these esters are
wax-free and biodegradable. They can be designed to meet the
lubricity requirement equivalent to those of mineral oils. They are
prepared by reacting monobasic acids with polyhydric alcohols.
Glycerol (3-OH; G), trimethylolpropane (3-OH; TMP), pentaery-
thritol (4-OH; PE) and dipentaerythritol (6-OH; DiPE) are a few
examples of polyhydric alcohols. Elimination of ␤-hydrogen (in
glycerol) with an alkyl group elevates the thermal stability of
polyol esters and allows them to be used at higher tempera-
tures. The reaction of a trihydroxy alcohol (glycerol or TMP) with
fatty acid results in mono- (ME), di- (DE) and tri- (TE) esters
(Scheme 1). Among these, DE and TE are compatible for biol-
ubricant applications. Conventionally, these esters are prepared
using mineral acid catalysts [2]. Separation of these homoge-
neous catalysts from the product mixture needs additional process
steps making the process economically non-attractive. The mineral
lysts makes the process economical and eco-friendly. Several solid
acid catalysts for fatty acid esterification reaction including sulfated
zirconia/titania/tungsten oxide, ion-exchange resins, zeolites, sul-
fonic acid functionalized mesoporous silica, H3PW12O40 supported
on ZrO2 and Ta2O3, sulfonated-C, WO3/AlPO4, WO3/ZrO2, silica-
bonded N-propyl sulfamic acid and silica-supported tin oxide, have
been reported [4–13]. Water is the by-product in esterification
reactions. It competes with reactant molecules for adsorption and
suppresses esterification activity. It also reacts and deactivates the
active acid sites on the solid catalyst and thereby reducing the
life of the catalyst. Recently, we have reported the application of
hydrophobic Fe-Zn double-metal cyanide and three-dimensional
mesoporous titanosilicates for this transformation [14–16]. With
all these solid catalysts, selectivity for DE and TE (components used
in synthetic lubricant applications) is low; monoester is the major
product. Hence, development of a hydrophobic and more efficient
and selective solid acid catalyst that yields high amount of DE + TE
is a desirable objective.
Metal organophosphonates are an interesting class of layered
inorganic materials that have attracted much attention as novel
proton conductors, luminescent and magnetic materials and corro-
sive resistive agents [17]. They have also been used to encapsulate
photoactive molecules [17]. However, there are only a few stud-
ies of their application as catalysts [18–20]. We report here, for
the first time, the use of zirconium phenyl phosphonate phosphite
(ZrPP) as a solid acid catalyst for the synthesis of polyol esters. Fig. 1
∗
Corresponding author. Tel.: +91 20 2590 2018; fax: +91 20 2590 2633.
E-mail address: d.srinivas@ncl.res.in (D. Srinivas).
0926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.04.043

130
P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
CH2OH
CH2OCOR
CH2OH
CH2OH
CH2OH
CH2OH
CH2OCOR
CH2OH
CH2OH
CH3CH2C
2 CH3CH2C
+
CH3CH2C
CH3CH2C
2 H2O
2 H2O
2 RCOOH
+
+
Trimethylolpropane (TMP)
Monoester (ME)
CH2OCOR
CH2OH
CH2OCOR
CH2OCOR
CH2OCOR
CH2OH
2 ME
2 RCOOH
2 RCOOH
+
+
+
CH3CH2C
+
Diester (DE)
CH2OCOR
CH2OCOR
CH2OCOR
2 CH3CH2C
2 DE
2 H2O
+
Triester (TE)
Scheme 1. Esterification of fatty acid with TMP.
shows tentative structures of ZrPP wherein several orientations of
phenyl phosphonate and phosphite groups are possible [21–23].
The organic phosphonate moiety imparts hydrophobic nature to
the structure. Interestingly, ZrPP catalysts of the present study are
found highly selective for (di + tri) esters than the reported solid
catalysts for this reaction.
ZrPP-2 and ZrPP-3 with phosphorous acid/phenyl phosphonic
acid molar ratio of 2 and 3 were prepared in a similar manner taking
1.04 and 1.56 g of phosphorous acid, respectively. Color: white.
2.2. Characterization procedure
The percentage composition of C and H in ZrPP was deter-
mined using a Carlo-Erba 1106 elemental analyzer. X-ray powder
diffraction patterns (XRD) of the catalyst samples were recorded
on a Philips X’pert Pro diffractometer in the 2ꢀ range of 10–80◦
and at a scan rate of 4◦/min using Cu-K␣ radiation and a propor-
tional counter detector. Fourier transform infrared (FTIR) spectra
of the samples (as KBr pellets) were recorded using a Shimadzu
8201PC spectrophotometer in the region 400–4000 cm−1. Solid
state 31P magic angle spin nuclear magnetic resonance (MAS
NMR) spectra were measured at 121.49 MHz on a Bruker AV300
NMR spectrometer using 4 mm sample rotors. Samples were
spun at a speed of 8 kHz. A single pulse sequence was applied
and the NMR measurements were performed at 25 ◦C. Ther-
mogravimetric and differential thermal analyses of the samples
exposed to water vapor were done under nitrogen atmosphere
(50 ml/min) on a Perkin Elmer Diamond TG-DTA instrument in the
2. Experimental
2.1. Catalyst preparation
Phenyl phosphonic acid (1 g; Spectrochem Pvt. Ltd.) and phos-
phorous acid (0.52 g; Loba Chem.) were dissolved in 20 ml of
double distilled water [20–23]. To it, 2.04 g of zirconium oxychlo-
ride (ZrOCl2·8H2O; Loba Chem.) dissolved in 5 ml water was added.
While stirring, the solution was heated to dryness at 90 ◦C. The solid
formed was recovered, washed with water (500 ml) and dried at
90 ◦C overnight. Zirconium phenyl phosphonate phosphite, thus
formed, was labeled as ZrPP-1, where 1 denotes the molar ratio
phosphorous acid/phenyl phosphonic acid.
H
P
P
O
O
P
P
O
O
O
O Zr
O
O Zr
P
P
O
O
O
O
O
O
Zr
Zr
O
O
O
O
Zr
Zr
O
Zr
Zr
O
Zr
Zr
Zr
O
P
Zr
O
O
O
Zr
O
O
Zr
O
O
O
O
Zr
O
O
O
P
O
Zr
O
O
P
P
H
P
H
P
H
Interlayer spacing
H
Interlayer spacing
H
P
H
P
H
O
H
P
O
O
O
O Zr
P
O
O
P
O Zr
O
O
P
O
Zr
O
O
O
O
Zr
O
O
Zr
O
Zr
Zr
Zr
Zr
Zr
Zr
Zr
Zr
Zr
Zr
Zr
(a)
(b)
Fig. 1. Tentative structure of zirconium phenyl phosphonate phosphite.

P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
131
temperature range of 25–1000 ◦C and with a ramp rate of 10 ◦C/min.
11.6
Surface area (SBET) of the samples was determined from nitrogen-
adsorption measurements carried out at −196 ◦C using a NOVA
1200 Quanta Chrome equipment. Acidity of the samples was
quantified by temperature-programmed desorption of ammo-
nia (NH3-TPD) technique (Micromeritics AutoChem 2910). About
100 mg of the sample was placed in a U-shaped, flow-through,
quartz sample tube. Before the TPD experiments, the catalyst was
activated at 300 ◦C under a flow of He (30 ml/min). The sample was
cooled to 50 ◦C and NH3 was adsorbed for 30 min. Desorption of
NH3 was followed by raising the temperature from 50 to 400 ◦C at
the ramp rate of 10 ◦C/min.
ZrPP-3
4.4
2.7
3.5
5.3
1.74
1.33
ZrPP-2
2.3. Reaction procedure
Known quantities of oleic acid (OA; C18.1; cis-9-octadecenoic
acid) and polyol (glycerol or TMP) were taken in a glass, round
bottom flask placed in a temperature-controlled oil bath, fitted
with a water-cooled reflux condenser. To it, 5 wt% of catalyst with
respect to OA was added. Temperature of the reaction mixture was
raised to a desired value and the reaction was conducted for a spe-
cific period of time. At the end of the reaction, the catalyst was
separated by centrifugation. The liquid product was treated with
petroleum ether (20–30 ml). Two layers formed with the bottom
being glycerol and the top being OA-glycerides and unconverted
OA. Glycerol was separated out by decantation. Petroleum ether in
the top layer portion was removed using a rotary evaporator. OA
conversion was estimated by titration with 0.1 N NaOH solution and
using 1, 10-phenolphthalein as indicator. The composition of OA-
glycerides was determined using a PerkinElmer (Series 200) high
performance liquid chromatograph fitted with an ELSD detector
(from Gilson) and a reverse-phase PerkinElmer Brownlee column
(C-18 Spheri-5, 250 × 4.6 mm with a 5 ␮m particle size). In kinetic
studies, samples were withdrawn at regular intervals of time and
analyzed as described above. Esterification reactions of TMP were
also done with fatty acids of varying chain length [valeric acid (C5;
ZrPP-1
10
20
30
40
50
60
2θ (degree)
Fig. 2. X-ray powder diffractograms of zirconium phenyl phosphonate phosphites
(ZrPP-1, ZrPP-2 and ZrPP-3) prepared with phosphorous acid/phenyl phosphonic
acid input molar ratios of 1, 2 and 3; Interplanar spacing (d-values) corresponding
to different reflections are indicated for ZrPP-1.
region 2310–2447 cm−1 is assigned to P H stretching vibrations
[20–24]. Several small bands were also observed in the range
1400–1700 cm−1 which could be assigned to aromatic C H bending
vibrational modes [20–24].
pentanoic acid), caprylic acid (C8; octanoic acid), capric acid (C10
n-decanoic acid) and lauric acid (C12; n-dodecanoic acid)].
;
Evidence for the presence of phosphite and phosphonate groups
in the catalysts was derived also from 31P MAS NMR spectroscopy
(Fig. 3, right panel). Two sets of 31P NMR signals at chemical shift
(ı) values of −17.2 and −7.6 were observed. While the former
is attributed to phosphite groups, the latter is corresponded to
phosphorous in phenyl phosphonate moieties. The signals corre-
sponding to ZrPP-3 are more intense than those of ZrPP-1 and
ZrPP-2 (Fig. 3, right panel). This could be due to difference in crys-
tallinity of these compounds. XRD peaks of ZrPP-3 are more intense
than those of ZrPP-1 and ZrPP-2 (Fig. 2)
The zirconium samples showed two stages of weight loss: a
minor loss in the temperature range 60–100 ◦C due to weakly
bound water (ZrPP-1, 2.1 wt%; ZrPP-2, 0.8 wt% and ZrPP-3, 0.5 wt%)
and a major loss at 400–500 ◦C (ZrPP-1, 18.8 wt%; ZrPP-2, 18.4 wt%
and ZrPP-3, 17.9 wt%; Fig. 4) due to decomposition of phenyl phos-
phonate and conversion of ZrPP into zirconium pyrophosphate
(ZrP2O7) [22]. Based on the weight loss, molecular formula of ZrPP
was determined and listed in Table 1. The %C and %H values deter-
mined using the proposed molecular formula and those determined
from microanalysis (Table 1) are in good agreement with each other
(Table 1). The experimentally observed %C and %H values for ZrPP
samples are as follows: 21.3 and 1.9 (for ZrPP-1), 21.3 and 1.8 (for
ZrPP-2) and 20.1 and 1.6 (for ZrPP-3). As the weight loss due to
water is very low [0.5 wt% (for ZrPP-3), 0.8 wt% (for ZrPP-2) and
2.1 wt% (for ZrPP-1)] and as it occurs at a temperature below 100 ◦C
unlike in the case of aluminosilicate zeolites (∼200 ◦C), the cata-
lysts of the present study are hydrophobic. It can be stated that
the hydrophobicity of ZrPP is much higher than most of the known
3. Results and discussion
3.1. Catalyst characterization
Zirconium phenyl phosphonate phosphite catalysts ZrPP-1,
ZrPP-2 and ZrPP-3 were prepared with phosphorous acid to phenyl
phosphonic acid molar ratio of 1:1, 2:1 and 3:1, respectively. These
compounds showed XRD patterns (Fig. 2) typical of a layered struc-
ture [20,21]. The XRD peaks were broad and indicate amorphous
nature in the structure. ZrPP-3 showed peaks at d-values of 11.6,
˚
5.3, 4.4, 3.5, 2.7, 1.74, and 1.33 A. With decreasing molar ratio of
phosphorous acid/phenyl phosphonic acid from 3 to 2 and 1, the
interlayer spacing of ZrPP [(0 0 1) reflection] had increased from
˚
11.6 to 11.8 and 15 A, respectively. This difference in basal spacing
of these materials is due to differences in the amount of hydrated
water present in their composition (Table 1) [24]. According to
elemental and thermogravimetric analyses, one mole of ZrPP-1 is
associated with 0.4 moles of hydrated water molecules while ZrPP-
2 and ZrPP-3 contained 0.15 and 0.1 moles of water per molecular
formula (Table 1).
ZrPP samples showed intense FTIR bands at 1015–1073 cm−1
and weak bands at 692–730 cm−1 and 2310–2447 cm−1 (Fig. 3,
left panel). While the intense bands at 1015–1073 cm−1 are cor-
responded to P O stretching vibrations of PO3 groups, those
at 692–730 cm−1 are attributed to out-of-plane bending vibra-
tions of mono substituted phenyl groups. The band in the

132
P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
Table 1
Chemical composition, molecular formula and textural properties of zirconium phenyl phosphonate phosphite.
Catalyst
Molecular formula
Elemental
SBET (m2/g)
Acidity (mmol/g)
Catalytic activitya
analysis (wt%)
C
H
<150 ◦
C
150–350 ◦
C
Total
OA conversion
(mol%)
Product selectivity (mol%)
ME
DE
TE
2.1
17.0
26.3
DE + TE
ZrPP-1
ZrPP-2
ZrPP-3
Zr(C6H5PO3)0.99(HPO3)1.01.0.4 H2O
Zr(C6H5PO3)0.97(HPO3)1.03.0.15 H2O
Zr(C6H5PO3)0.95(HPO3)1.05.0.1 H2O
21.3
21.3
20.1
1.9
1.8
1.6
439
260
268
1.59
0.23
0.30
–
0.04
0.07
1.59
0.27
0.36
39.6
47.6
48.9
58.1
11.8
7.7
39.8
71.7
66.1
41.9
88.7
92.4
a
Reaction conditions: Oleic acid (OA) = 4 g, glycerol = 0.32 g, OA: glycerol (molar ratio) = 4: 1, reaction temperature = 180 ◦C, reaction time = 1 h. ME = mono ester, DE = diester
and TE = triester.
ZrPP-1
ZrPP-2
ZrPP-3
-7.6
-17.2
ZrPP-3
2447
2310
ZrPP-2
ZrPP-1
1073
600 800 1000120014001600
1015
2320 2400 2480 2560
-1
-60
-40
-20
0
δ (ppm)
20
40
Wavenumber (cm
)
Fig. 3. FTIR (left) and 31P MAS NMR (right) spectra of zirconium phenyl phosphonate phosphite compounds.
solid acid catalysts (water adsorption capacity of solid acid cata-
lysts = 5–15 wt% while that of ZrPP = 0.5–2.1 wt% only).
and acidity of these complexes. Surface area and overall acidity of
ZrPP decreased with increasing input molar ratios of phosphous
acid/phenyl phosphonic acid (Table 1). Ammonia was used as a
probe molecule to determine the acidity of the catalysts. Two NH3
The input molar ratio of phosphorous acid/phenyl phospho-
nic acid has a marked effect on the specific surface area (SBET
)
ZrPP-1
ZrPP-2
100
96
92
88
84
80
ZrPP-3
0
200
400
600
0
800
1000
Temperature ( C)
Fig. 4. Thermogravimetric analysis of zirconium phenyl phosphonate phosphite compounds.

P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
133
87
90
ZrPP-1
ZrPP-2
248
0
70
140
210
280
350 0
70
140
210
280
350
Temperature
77
(
0C
)
ZrPP-3
257
280
0
70
140
210
350
Temperature ( )
0C
Fig. 5. NH3-TPD plots of zirconium phenyl phosphonate phosphites.
desorption peaks [below 150 ◦C (peak 1) and 150–350 ◦C (peak 2)]
were observed for the ZrPP compounds (Fig. 5). While the former
is corresponded to weak acid sites, the latter is attributed to strong
acid sites (Table 1). Peak 2 is more apparent for ZrPP-2 and ZrPP-
3 than for ZrPP-1 (Fig. 5). The input molar ratio of phosphorous
acid/phenyl phosphonic acid has a little effect on phenyl phos-
phonate/Zr and HPO32−/Zr molar ratios in the catalyst samples.
These ratios for different samples are as follows: 0.99 and 1.01 (for
ZrPP-1), 0.97 and 1.03 (for ZrPP-2) and 0.95 and 1.05 (for ZrPP-
3), respectively. However, it may be noted that the phosphorous
acid/phenyl phosphonic acid molar ratio has a marked effect on
adsorbed H2O/Zr molar ratio which is 0.4, 0.15 and 0.1 for ZrPP-1,
ZrPP-2 and ZrPP-3 respectively. The amount of adsorbed water per
unit specific surface area of the catalyst also varies in the same
order: 2.658 × 10−6, 1.709 × 10−6 and 1.036 × 10−6 moles/m2 for
ZrPP-1, ZrPP-2 and ZrPP-3, respectively. These studies reveal that
phosphorous acid/phenol phoshonic acid ratio used in the prepa-
ration did not alter phenyl/Zr and P/Zr molar ratio, but affected the
surface hydrophobicity of the catalysts significantly with ZrPP-3
being the highest hydrophobic one.
mole of fatty acid resulting in one mole of monoester (ME) which
in the second-step reacts with another mole of fatty acid and yields
a diester (DE). In the third-step, DE gets converted into triester (TE).
In each step, one mole of water is formed as by-product. Excess of
polyol or fatty acid shifts the equilibrium toward right side. The
by-product water can compete for adsorption with polyols over
hydrophilic catalysts. This issue is less significant with the present
set of hydrophobic ZrPP catalysts. This is one of the superior fea-
tures of the catalysts of the present study. As seen from Table 1,
catalytic activity (OA conversion) of ZrPP-1 is lower than that of
ZrPP-2 and ZrPP-3. Most importantly, the selectivity for DE + TE is
higher over ZrPP-3 than on ZrPP-2 and ZrPP-1. Higher hydrophobic-
ity of ZrPP-3 is possibly responsible for its higher selectivity for the
desired esters (DE + TE). Remember that water adsorption capacity
of ZrPP-3 is only 0.5 wt% while those of ZrPP-2 and ZrPP-3 were 0.8
and 2.1 wt%, respectively. In view of its higher activity and selectiv-
ity, all the subsequent catalytic studies were conducted only with
ZrPP-3.
The molar ratio of reactants (OA: polyol) has a major influ-
ence on OA conversion and product ester selectivity (Table 2). With
increasing ratio of OA/polyol from 0.5 to 4, a decrease in the appar-
ent conversion of OA from 83.0 to 49.0 mol% was observed. But then,
more and more of ME and DE got converted to TE. The total selec-
tivity of DE + TE increased with increasing OA/polyol molar ratio. At
OA: polyol molar ratio of 4:1, reaction temperature of 180 ◦C and
reaction time of 1 h, the selectivity for DE + TE was 92.3 mol% in the
case of glycerol and 86.0 mol% in the case of TMP. Polyol conver-
sion was 100% at these conditions. This is the highest selectivity
achieved so far over a solid catalyst. Hydrophobicity of ZrPP-3 is
the cause driving the equilibrium toward DE + TE esters.
3.2. Catalytic activity
In this study, trihydroxy alcohols (glycerol and trimethylol-
propane, TMP) were chosen as representative polyols. Oleic acid
(OA) was considered as a representative fatty acid. A few reac-
tions were conducted also with valeric acid (C5), caprylic acid (C8),
capric acid (C10) and lauric acid (C12) acids. Esterification of tri-
hydroxy alcohols is a three-step consecutive equilibrium reaction
(Scheme 1). In the first-step, one mole of polyol reacts with one

134
P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
Table 2
Esterification of oleic acid with polyol over ZrPP-3: effect OA:polyol molar ratioa.
OA: polyol molar ratio
Polyol = glycerol
Polyol = trimethylolpropane (TMP)
OA conversion (mol%)
Product selectivity (mol%)
OA conversion (mol%)
Product selectivity (mol%)
ME
DE
TE
0.8
0.3
14.5
12.1
26.2
DE + TE
ME
DE
3.5
32.9
74.8
74.6
68.4
TE
1.6
2.0
9.4
16.4
17.6
DE + TE
0.5:1
1:1
2:1
3:1
4:1
83.0
79.0
61.0
54.0
49.0
68.0
68.9
28.3
20.3
7.7
31.2
30.7
57.2
67.6
66.1
32.0
31.0
71.7
79.7
92.3
64.3
60.0
54.2
51.6
45.2
94.9
65.1
15.8
9.0
5.1
34.9
84.2
91.0
86.0
14.0
a
Reaction conditions: Polyol = 0.32 g (glycerol) or 0.48 g (TMP), catalyst = 5 wt% of OA, reaction temperature = 180 ◦C, reaction time = 1 h. ME = mono ester, DE = diester and
TE = triester.
OA conversio
n
OA conversio
n
MG
DG
TG
90
60
30
0
ME
DE
TE
80
60
40
20
0
140
160
180
200
14
0
160
18
0
200
Reaction temperature (ºC)
Reaction temperature (ºC)
Fig. 6. Effect of reaction temperature on esterification oleic acid with glycerol (left) and trimethylolprapane (TMP; right) over ZrPP-3 catalyst. Reaction conditions: oleic acid
(OA) = 4 g, polyol = 0.32 g (glycerol) or 0.48 g (TMP), OA: polyol molar ratio = 4: 1 catalyst = 5 wt% of OA, reaction time = 1 h. ME = mono ester, DE = diester and TE = triester.
As the reaction was prolonged from 1 to 4 h, significant changes
in product selectivity were observed. Selectivity for TE increased
at longer reaction time. The influence of reaction temperature on
the esterification of OA was evaluated by varying the reaction tem-
perature from 140 to 200 ◦C (Fig. 6). As expected, the conversion
of OA increased with temperature. At the same time, the selectiv-
ity also varied. The selectivity to DE + TE increased with increase
in temperature. The present catalyst showed considerable activity
even at as low reaction temperature as 140 ◦C. When the catalyst
amount was increased from 3 to 9 wt%, only a moderate increase
in OA conversion was observed. Thus, it is clear that the increase in
catalyst loading above 3 wt% did not help to improve the initial rate
of the reaction. These results suggest that a small amount catalyst
is sufficient to attain maximum conversion.
acid (C18.1) instead of valeric acid (C5). Addition of water (up to
0.5 wt %) did not alter the esters yield.
ZrPP-3 being hydrophobic is quite stable and reusable in ester-
ification reactions (Fig. 7). OA conversion was almost the same
in four recycling experiments. At the end of each cycle, the cata-
lyst was separated from the reaction mixture by filtration, washed
with methanol, dried at 373 K for 3 h and then reused in sub-
sequent recycle. The catalyst retained its activity even after its
OA-GLYCEROL
50
40
30
20
10
0
OA-TMP
Chain length of fatty acid has a notable effect on conversion
and product selectivity (Table 3). While the conversion of fatty acid
and individual selectivity of TE had decreased, selectivity of DE + TE
increased with increasing chain length of the acid i.e., by using oleic
Table 3
Catalytic data of ZrPP-3 in the esterification of fatty acids with TMP.a
Fatty acid
Fatty acid
conversion
(mol%)
Product selectivity (mol%)
ME
DE
TE
DE + TE
Valeric acid (C5)
Caprylic acid (C8)
82.3
75.1
65.1
55.4
45.3
36.9
35.5
32.3
29.8
14.0
30.4
41.9
46.6
51.1
68.4
32.7
22.6
21.1
19.1
17.6
63.1
64.5
67.7
70.2
86.0
0
1
3
4
Number of2recycles
Capric acid (C10
Lauric acid (C12
)
)
Oleic acid(C18.1
)
Fig. 7. Catalyst reusability study in esterification of OA with glycerol and TMP.
Reaction conditions: oleic acid (OA) = 4 g, polyol = 0.32 g (glycerol) or 0.48 g (TMP),
OA: polyol molar ratio = 4: 1, catalyst (ZrPP-3) = 5 wt% of OA, reaction tempera-
ture = 180 ◦C, reaction time = 1 h. ME = mono ester, DE = diester and TE = triester.
a
Reaction conditions: Fatty acid = 4 g, fatty acid: trimethylolpropane (TMP) molar
ratio = 4: 1, catalyst = 5 wt% of fatty acid, reaction temperature = 180 ◦C, reaction
time = 1 h. TMP conversion = 100%. ME = mono ester, DE = diester and TE = triester.

P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
135
11.6
(a)
Fresh ZrPP-3
Spent ZrPP-3
4.4
2447
2.7
3.5
5.3
1.74
1.33
60
(b)
Fresh ZrPP-3
Spent ZrPP-3
1015
1042
20
40
500 1000 1500 2000 2500 3000 3500 4000
Wavenumber ( )
cm-1
2
θ
(degree)
Fig. 8. (a) XRD and (b) FTIR of fresh and spent ZrPP-3 catalysts.
(b) ZrPP-3
100
100
80
60
40
20
0
(a) SZ
80
60
40
20
0
OA conversion
ME
DE
TE
DE + TE
1
2
3
4
5
1
2
3
4
5
Reaction time (h)
Reaction time (h)
Fig. 9. Kinetics of esterification of OA with TMP over (a) SZ and (b) ZrPP-3 catalysts. Reaction conditions: oleic acid (OA) = 4 g, trimethylolpropane (TMP) = 0.48 g, OA: polyol
molar ratio = 4: 1, catalyst = 5 wt% of OA, reaction temperature = 180 ◦C. ME = mono ester, DE = diester and TE = triester.
use. To establish the structural integrity, the spent catalyst was
characterized by XRD and FTIR spectroscopy. As shown in Fig. 8,
the fresh and spent catalysts have the same finger print fea-
tures in both XRD and FTIR indicating that the structure of the
used catalyst is intact. Compositional integrity was established by
microanalysis and energy-dispersive X-ray spectroscopy (EDAS).
The %C and %H of the fresh and spent catalysts were nearly the
same (20.1 and 1.6 for the fresh catalyst and 21.5 and 1.7 for
the spent ZrPP-3 catalyst). EDAX revealed that Zr:P:O:C composi-
tion (wt%) of the fresh (51.23:18.77:17.39:12.61) and spent ZrPP-3
(47.34:20.36:18.67:13.63) is also the same. Analysis of the product
samples did not show any traces of Zr and P elements in their com-
position. All these characterization studies of spent catalyst and
product mixture, thereby, reveal that ZrPP-3 is a stable and usable
solid catalyst.
With a view to establish that surface hydrophobicity of ZrPP-3
is the cause for its high selectivity for DE + TE esters, we have car-
ried out experiments also with a commercial hydrophilic sulfated
zirconia (SZ; Saint-Gobain, Norpro) catalyst which has both Brön-
sted and Lewis acidic active sites. The reactions were conducted
at 180 ◦C taking OA: TMP molar ratio of 4: 1 and catalyst of 5 wt%
of OA. Reactions were followed as a function of time for 5 h. As
expected, conversion of OA was higher over Brönsted and Lewis
acidic SZ than on ZrPP-3 (Fig. 9). But then, DE + TE esters selectiv-
ity was significantly higher over ZrPP-3 than on SZ catalyst (Fig. 9).
These studies thus confirm that hydrophobicity plays a major role
on product selectivity in reactions with polyhydric alcohols. The
by-product water can compete for adsorption with polyols and
OA. On hydrophobic surfaces as that of ZrPP-3, only the reactant
molecules, OA and polyol adsorb and undergo reaction to form
polyol esters. But, on hydrophilic surfaces, competitive adsorption
and reverse reactions (hydrolysis) would also be favored reducing
the selectivity for DE + TE esters.
4. Conclusions
Zirconium phenyl phosphonate phosphites were prepared with
varying phosphorous acid to phenyl phosphonic acid input molar
ratio and evaluated as catalysts in esterification reactions of
oleic acid with glycerol and TMP. The catalyst with phospho-
rous acid/phenyl phosphonic acid molar ratio of 3 showed high
catalytic activity and most importantly the DE + TE selectivity.
Hydrophobicity of ZrPP and strong acidity are possibly the key
features for their high activity and selectivity in polyesterifi-
cation reactions. ZrPP is a reusable catalyst. It is highly active
irrespective of the type of fatty acid and polyol. Synthetic esters
of varying lubricity properties could be prepared with this cata-
lyst by changing process parameters or by varying fatty acid and
polyol.

136
P. Varhadi et al. / Applied Catalysis A: General 462–463 (2013) 129–136
Acknowledgment
[10] Y.-Y. Linko, T. Tervakangas, P. Lämsä, P. Linko, Biotech. Tech. 11 (1997)
889–892.
[11] Y.-M. Park, S.-H. Chung, H.J. Eorn, J.-S. Lee, K.-Y. Lee, Bioresour. Technol. 101
(2010) 6589–6593.
[12] W. Xie, D. Yang, Bioresour. Technol. 119 (2012) 60–65.
[13] W. Xie, H. Wang, H. Li, Ind. Eng. Chem. Res. 51 (2012) 225–231.
[14] P.S. Sreeprasanth, R. Srivastava, D. Srinivas, P. Ratnasamy, Appl. Catal. A: Gen.
314 (2006) 148–159.
M.K. acknowledges Council of Scientific and Industrial Research
(CSIR), New Delhi for the award of Senior Research Fellowship.
References
[15] M. Kotwal, S.S. Deshpande, D. Srinivas, Catal. Commun. 12 (2011)
1302–1306.
[16] M. Kotwal, A. Kumar, D. Srinivas, J. Mol. Catal. (2013), http://dx.doi.org/10.
1016/j.molcata.2013.04.029.
[17] A. Clearfield, K.D. Demadis (Eds.), Metal Phosphate Chemistry: From Synthesis
to Applications, RSC Publishing, Cambridge, UK, 2012.
[18] D. Lanari, F. Montanari, F. Marmottini, O. Piermatti, M. Orrù, L. Vaccaro, J. Catal.
277 (2011) 80–87.
[1] L.R. Rudnick, Synthetics, Mineral Oils, and Bio-Based Lubricants: Chemistry and
Technology, Taylor & Francis, CRC, Press, Boca Raton, USA, 2006.
[2] L. Honary, E. Richter, Biobased Lubricants and Greases: Technology and Prod-
ucts, John-Wiley & Sons Ltd, West Sussex, UK, 2011.
[3] W. Qin, X. Ying-jie, C. Xi, G. Zan-jun, H. Shuai, H. Xiao-hong, C. Guang-ming,
Fluid Phase Equilib. 293 (2010) 110–116.
[4] P. Bondioli, Top. Catal. 27 (2004) 77–82.
[5] H. Masood, R. Yunus, T.S.Y. Choong, U. Rashid, Y.H. Taufiq Yap, Appl. Catal. A:
Gen. 425–426 (2012) 184–190.
[19] S. Calogero, D. Lanari, M. Orrù, O. Piermatti, F. Pizzo, L. Vaccaro, J. Catal. 282
(2011) 112–119.
[6] S. Gryglewicz, W. Piechocki, G. Gryglewicz, Bioresour. Technol. 87 (2003)
35–39.
[7] C.O. Åkerman, Y. Gaber, N.A. Ghani, M. Lämsa, R. Hatti-Kaul, J. Mol. Catal. B:
Enzym. 72 (2011) 263–269.
[8] J. Oh, S. Yang, C. Kim, I. Choi, J.H. Kim, H. Lee, Appl. Catal. A: Gen. 455 (2013)
164–171.
[9] E. Santacesaria, G. Martinez Vicente, M. Di Serio, R. Tesser, Catal. Today 195
(2012) 2–13.
[20] X.-Z. Lin, Z.-Y. Yuan, Eur. J. Inorg. Chem. (2012) 2661–2664.
[21] J.L. Colón, D.S. Thakur, C.-Y. Yang, A. Clearfield, C.R. Martin, J. Catal. 124 (1990)
148–159.
[22] A. Clearfield, J.D. Wang, Y. Tian, E. Stein, C. Bhardwaj, J. Solid State Chem. 117
(1995) 275–289.
[23] X. Lei, L. Wang, Z. Cui, S. Xu, F. Zhang, Thin Solid Films 519 (2011)
3552–3556.
[24] J.D. Wang, A. Clearfield, Mater. Chem. Phys. 35 (1993) 208–216.