Microsoft Word - 55castellanos.docx


 CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 64, 2018 

A publication of 

 
The Italian Association 

of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Enrico Bardone, Antonio Marzocchella, Tajalli Keshavarz
Copyright © 2018, AIDIC Servizi S.r.l. 
ISBN 978-88-95608- 56-3; ISSN 2283-9216 

Selective Catalytic Hydrogenation of Triglycerides: Activity 
and Selectivity Towards C18:1. 

Umberto Pasqual Laverduraa*, Francesco Ferellaa, Marco Creatia, Mattia 
Giampaoloa, Katia Galluccia, Claire Coursonb, Antonio Zarlic, Leucio Rossid 
a
Università degli studi dell’Aquila DIIIE, via G. Gronchi 18 Nucleo Industriale di Pile, 67100 L’Aquila, Italy  

b
Université de Strasbourg, ICPEES, 25 rue Becquerel , 67087 Strasbourg cedex 2, France 

c
Processi Innovativi Srl, via di Vannina 88, 00156 Rome, Italy 

d
Università degli studi dell’Aquila DSFC, via Vetoio, 67100 L’Aquila, Italy

 

umberto.pasquallaverdura@graduate.univaq.it 

Aim of this research is the implementation of a batch lab scale plant for the hydrogenation of vegetable oils. In 
this preliminary paper, we report the results obtained in the hydrogenation reaction of rapeseed oil carried out 
with a well-known Pd-based catalyst (Lindlar catalyst) to evaluate the best process conditions. Tests are 
performed in a batch reactor at two different levels of temperature (60 °C and 180 °C) and pressure (0.4 MPa 
and 1.2 MPa), a medium point at 120 °C and 0.8 MPa is taken into account in order to estimate the best 
operative conditions. Best results were observed during the first 1.5 hours of the test carried out at high 
temperature and low pressure conditions; a relatively high amount of oleic acid is obtained. However, an 
uncompleted conversion of dienes and trienes compounds has been observed. After 6h the oleic acid reacts 
producing trans-isomers and stearic acid undesired hydrogenation by-products. 

1. Introduction 
In the last few years, it has been seen a growing interest, mainly in the developed countries, towards the 
valorisation of vegetable oils and their products, as an alternative to petrochemicals and as building block in 
industrial application. Also in the past, vegetable oils were studied for industrial use, but only with incremented 
oil price and more interest in products from renewable sources, industry considers them a viable raw materials 
for many common industrial products. 
Since XIX century, vegetable oils found widespread application in industry and transports, but since the 
increasing of oil drilling their use decreases and only in the end of XX century, because of global crisis oil price 
and oil shortening due to political instability and consumption of reserves, a renewed interest in vegetable oils 
firstly in diesel transport and then as a raw material for production of fine chemicals, is exhibited. Currently, the 
main oil application is biodiesel production, but especially in implementation of a wider range of vegetables as 
Sapium tree oil (Cogollo-Herrera e al. 2015) or safflower oil (Ilkılıç et al. 2011) or other kind of biofuels from 
deoxygenation and decarboxylation of oils (Romero et al. 2014). 
These oils main interesting properties are biodegradability, production from renewable resources, the crop is 
well spread globally and, also, can be used as reagents although their composition can vary between plants 
species and with climatic conditions. These raw materials have an intrinsically variable composition, but the 
components usually observed in triglycerides are: oleic acid (C18:1), linoleic acid (C18:2) and linolenic acid 
(C18:3), other compounds could be arachidonic acid (C14:0), palmitic acid (C16:0), stearic acid (C18:0) and 
sometimes, for example in non-canola rapeseed oil, also erucic acid (C22:0 and C22:1). Some of these oils as 
rapeseed oil, soybean oil, sunflower oil are also used in food industry. 
The main problem connected to the use of this raw material is the high contents of polyunsaturated acids 
(C18:2 and C18:3 compounds) which limits their use (Behr et al. 2008). Decreasing the linoleic acid and 
linolenic acid concentrations increase the stability of the oils towards oxidation and the oleic acid content 
(C18:1) that because of cis configuration of the double bond is still liquid at -15 °C, maintaining the fluidity of 
the mixture (Schneider 2006). To adjust the composition of these vegetable oils and reduce the concentration 

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1864020

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Pasqual Laverdura U., Gallucci K., Courson C., Zarli A., Rossi L., Foscolo P.U., 2018, Selective catalytic 
hydrogenation of triglycerides activity and selectivity towards c18:1, Chemical Engineering Transactions, 64, 115-120   
DOI: 10.3303/CET1864020 

115



of poly-unsaturated fatty acids, several solutions are viable: use of solvents to enhance solubility of hydrogen, 
use of critical or near critical solvents, use of some additives in order to reduce monoene hydrogenation, use 
of heterogeneous catalysis principally noble catalysts or copper catalysts (Dijkstra 2006). Using 
heterogeneous catalyst is an interesting solution, the catalyst can be easily recovered from the oil without 
complex operations and reutilized in further batch. Usually, the catalyst used in industrial hydrogenation 
contains, as active phase, nickel over silica or alumina, load of Ni around 22 %w/w. The principal obstacle is 
the achievement of a high concentration of trans-isomers and the formation of saturated fatty acid (C18:0 
stearic acid). Research in this field proceeds following two different paths: the use of noble metal catalysts, 
mainly Pd and Pt, or the use of Cu, that is less noble and cheaper, but with a lower activity towards complete 
saturation (List et al. 2015). 

2. Selective hydrogenation experiments and analysis 
Two different experiments were carried out using commercial Lindlar catalyst (Sigma-Aldrich®). Lindlar 
catalyst is composed by Palladium (5%w/w), deposed on calcium carbonate deactivated with lead oxides; 
sometimes quinoline is used to reduce activity and enhance selectivity. Lindlar catalyst was developed for 
selective hydrogenation of alkynes to alkenes, but was also used in hydrogenation of polyolefins. Since fatty 
acids and olefins similarity, Lindlar catalyst was chosen to conduct preliminary oil hydrogenation tests. 
The first hydrogenation test was carried out in a 25 mL dead-end reactor using 5 mL of purified sunflower oil 
(Sigma-Aldrich®) in “mild conditions”, nominally at 150 °C, maintained with a silicon bath oil, and endogenous 
pressure in complete atmosphere of hydrogen and 0.2%w/w of active phase. This test is realized in order to 
evaluate the effective activity of the catalyst, therefore sunflower oil is chosen for its higher unsaturation 
degree; especially, a high concentration of C18:2: 68%w/w of linoleic acid. The reactor is charged with oil and 
the catalyst; then, the system is purged with argon to eliminate oxygen. The reactor is filled with hydrogen, 
sealed and heated up to 150 °C. Any 2 hours, a sample is collected and H2 atmosphere is restored. Cause of 
the mild experimental conditions (0.15 MPa of H2), the test is performed for 6 h to properly evaluate changes 
in oil composition after reacting with H2. 
After these, tests with commercial Canola oil, with high starting concentration of oleic acid (63%w/w) were 
performed in more severe conditions in a lab-scale plant, as shown in Figure 1. 

 

Figure 1: Lab-scale plant reactor for hydrogenation of oils and their derivatives. 

The reactor is a Parr Instruments 600 mL reactor model 4560, charged with 200 mL of rapeseed oil (Lidl 
Canola oil), about 180 g and with 800 mg of Lindlar catalyst. A different oil was used for these tests because it 
is interesting to see the catalytic selectivity of the Lindlar with a high oleic content and its activity towards 
complete saturation of monenes. The reactor is purged in N2 and then conducted to the temperature set point 

116



and maintained under quite isothermal conditions during each test run. For the tests in lab-scale plant, two 
different values of temperature (60 and 180°C) and pressure (0.4 MPa and 1.2 MPa) were chosen to obtain a 
response curve for the catalyst. A fifth test in intermediated conditions (120 °C and 0.8 MPa) was carried out. 
Low temperature tests are carried out to estimate the lower limit of the catalyst activity towards hydrogenation. 
Tests are carried out for 6 hours; during this time twelve samples were collected at interval of 30 min. 
Pressure is maintained by means of a back-pressure regulator (Bronkhorst®). 
The samples taken from the reactor are trans esterified using IUPAC Standards methods for the analysis of 
oils, fats and derivatives. The transesterification reaction is performed by means of BF3, methanol and KOH 
solution as reagents, the trans esterified sample is extracted with n-hexane and dried with sodium sulphate. 
All reactants are from Sigma-Aldrich®. 
Fatty acid composition in the oil was determined by Gas Chromatography. The trans-esterified samples are 
analyzed with a Varian 3400 GC with a Flame-Ionization detector (FID) equipped with a Supelco® SP-2380 
GC column (30mx25μm). Identification of fatty acid methyl esters is achieved with commercially available 
standards and purified sample of oil (sunflower and rapeseed Canola Oil) and confirmed with literature data. 
The analysis for each sample is repeated at least twice to decrease measure error. 
Iodine value is calculated from GC analysis in accordance to ISO-EN 3961 (2013) following the Eq (1): 
 ( : ∗ 0.950) + ( : ∗ 0.860) + ( : ∗ 1.732) + ( : ∗ 2.616) + ( : ∗ 0.785) + ( : ∗ 0.723)  
 

(1) 

where w16:1 is the relative composition of palmitoleic acid, w18:1 of oleic acid, w18:2 of linoleic acid, w18:3 of 
linolenic acid, w20:1 of gadoleic acid, w22:1 of erucic acid. 

3. Results and discussion 
3.1 “Mild condition” test 

The results for mild condition test (150 °C and 0.15 MPa) are shown in Table 1. After 6 h, the hydrogenation of 
C18:2 linolenic acid is not complete. At this pressure and temperature conditions the formation of trans 
isomers is low but part of the dienes is converted in their geometric isomers respectively. No relevant 
concentration of positional isomers is detected. 

Table 1:  Results for Lindlar catalyst at 150 °C and endogenous pressure with sunflower oil (Sigma-Aldrich), 
concentrations are expressed in relative composition %, trans isomers of oleic acid and CLA of linoleic acid. 

Time (h) C16:0 C18:0 C18:1 C18:2 C18:3 trans CLA 
0 (oil as received) 3 2 25 68 - - - 

2 3 3 26 64 - 2 1 
4 3 3 28 60 - 2 2 
6 3 3 31 57 - 3 3 

 
The reaction for diene is not complete because of the low pressure. In fact, low pressure in the gas phase 
leads to low concentrations of hydrogen in the liquid phase and then at the surface of the catalyst, solubility of 
hydrogen depending from of oil pressure and temperature (Andersson et al. 1974). 

3.2 Lab-scale batch reactor tests 

Firstly, blank tests are carried out in the same conditions to verify if there is any homogeneous phase reaction 
between oil and dissolved hydrogen. Only the 6 h sample has been considered to individuate an effective 
conversion of linoleic and linolenic to oleic and, also, the formation of trans-isomers. After the blank sample 
analysis, it is not noticed an effective change of composition, compared to the rapeseed oil charge (any 
formation of the above compounds is found and the composition is the same as the rapeseed oil charged in 
the reactor, only traces, less than 0.05 % of elaidic acid, are detected from the GC and were not taken in 
account for test with the catalyst). 
Using harsh reaction conditions in respect of “mild conditions” test, the catalyst exhibits a higher activity as 
shown in the following tests (Figure 2 and Figure 3). 
The results of the upper limit conditions test (180°C and 1.2 MPa) are reported in Figure 2: the activity of the 
catalyst is high and in two hours a quasi-complete conversion of linoleic acid and linolenic acid occurs, but 
high concentration of trans elaidic isomer is detected. Concentration of saturated compounds enhances in 
time, since Pd is active also for the complete saturation of the double bonds in fatty acids. 

117



0 1 2 3 4 5
0

10

20

30

40

50

60

70

80

90

100

 C18:0
 cis-C18:1
 trans-C18:1
 C18:2
 C18:3

re
la

tiv
e
 c

o
n
c
e
n
tr

a
ti
o
n
 %

 w
/w

Time (h)

 

Figure 2: Results tests at 180 °C and under 1.2 MPa. 

The formation of elaidic acid is due to isomerization reaction. Thermodynamically the most stable compound is 
trans isomers and the mechanism of reaction (Dijkstra 2009) is affected by the degree of unsaturation 
presented in the triglycerides. Likely, poly-unsaturated compounds adsorb on the catalyst surface more rapidly 
than mono-unsaturated ones, but when the concentration of dienes and trienes decreases, oleic and stearic 
acids are formed. The affinity with the surface decreases with the addition of a hydrogen molecule to the 
double bond and then more surface is available for isomerization reaction. 
The results of the test carried out at 180°C and 0.4 MPa are reported in Figure 3, the concentration of palmitic 
acid and other saturated compounds is stable around 4%w/w. The concentration of positional isomers cis-11 
oleic at the end of test (less than 1%w/w), and for CLA acids (1.5%w/w) are not reported. 
At the same temperature under a lower pressure (0.4 MPa) the smaller formation of saturate compounds can 
be correlated to a lower concentration of H2 adsorbed on the catalyst surface. After 1.5 h, a complete 
conversion of linolenic acid is observed, less than 1 %w/w of linoleic acid remains in the oil, and elaidic is lower 
than oleic. However, the same trend which has been observed at higher pressure is shown in cis-C18:1 
conversion to trans-isomers and stearic acid, and at the end of the test high amount of these two compounds 
are found in the oil mixture (Figure 3). 

0 1 2 3 4 5 6
0

10

20

30

40

50

60

70

80

90

100

 C18:0
 cis-C18:1
 trans-C18:1
 C18:2
 C18:3

re
la

tiv
e

 c
o
n
c
e
n

tr
a
tio

n
 %

 w
/w

Time (h)

 

Figure 3: Results tests for at 180 °C under P 0.4 MPa. 

118



The low temperature test present displays low conversion of polyunsaturated acids in the triglycerides. The 
temperature is low, and the kinetic of reaction is depressed. The concentration of elaidic acid is under the 
threshold value of detection of the GC, the variations of oleic acid and stearic are below 2%w/w. 

0 1 2 3 4 5 6
0

2

4

6

8

10

12

14

16

18

20

 C18:2 P 1,2 MPa
 C18:2 P 0,4 MPa
 C18:3 P 1,2 MPa
 C18:3 P 0,4 MPa

re
la

tiv
e
 c

o
m

p
o
s
iti

o
n
 %

 w
/w

Time (h)

 

Figure 4: Comparison between concentration of linoleic and linolenic acid under 0.4 MPa and 1.2 MPa tests 
performed at 60 °C. 

In the end, the middle point test, Figure 5, shows that the catalyst reaction starts but activity towards 
conversion of C18:2 and C18:3 is very low and only at the end of 5 h the two fatty acids are completely 
converted. Under these conditions, reaction yields more elaidic acid in the final product. This is due to the 
reversibility of isomerisation reaction: at high temperature (180 °C) the partially hydrogenated double bond is 
attached on the surface and can rotate and so some elaidic formed can reform oleic acid. At lower 
temperature (120 °C) after the formation of intermediate the tendency is that less elaidic re-forms oleic acid. 

0 1 2 3 4 5
0

10

20

30

40

50

60

70

80

90

100

 C18:0
 cis-C18:1
 trans-C18:1
 C18:2
 C18:3

re
la

tiv
e

 c
o
m

p
o
s
iti

o
n

 %
 w

/w

Time (h)

 

Figure 5: Results test for at 120 °C under 0.8 MPa. 

Iodine value (IV) trends are reported for all tests in Figure 6, the high temperatures tests produced a high 
saturation degree in oil and then also a low IV, otherwise low temperature conditions practically do not induce 
any relevant modifications in the oil. The middle point test produces a less saturate product but a higher 
concentration of trans-isomers. 

119



0 1 2 3 4 5 6
0

20

40

60

80

100

120

 T 60 °C-P 0,4 MPa
 T 60 °C-P 1,2 MPa
 T 120 °C-P 0,8 MPa
 T 180 °C-P 0,4 MPa
 T 180 °C-P 1,2 MPa

C
a
lc

u
la

te
d
 I

o
d
in

e
 V

a
lu

e

Time (h)

 

Figure 6: Calculated Iodine value variation during hydrogenation tests. 

4. Conclusions 
Tests help to clarify the behaviour of the Lindlar catalyst in vegetable oils hydrogenation: Lindlar catalyst 
demonstrates a high activity towards hydrogenation of double bonds typical of palladium based catalysts, but 
the selectivity, shown in other hydrogenation processes, such as in conversion of alkynes to alkenes), is never 
obtained under investigated conditions.  
In order to complete this work with Lindlar catalyst, a factorial experiment will be done and the best conditions 
test will be performed to verify the activity and selectivity of the catalyst. The test at 180 °C and 0.4 MPa will 
be furthermore investigated for the duration of 1-1.5 h, C18:0 and trans C18:1 concentrations are low and 
increment in C18:1 are not negligible, short time reaction are always of industrial interest. 
However, from this study many important information can be extract which are transferable in future work with 
different materials. Future work will be carried out on new synthesized catalysts families, following the 
analogous factorial experiments scheme. 

Acknowledgments  

This work was financially supported by Processi Innovativi S.r.l. and by M.I.U.R. under P.O.N. 2014-2020 
Project within the PhD code DOT13LHQ8Y. 

Reference  

Andersson, K., Hell M., Löwendahl L., N.-H. Schöön, 1974, Kinetics of hydrogenation of rapeseed oil: I. 
Influence of transport steps in kinetic study, JAOCS, 52 Iss. 6, 182-187, DOI: 10.1007/BF02672166 

Behr A., Westfechtel A., Pérez Gomes J., 2008, Catalytic Processes for the Technical Use of Natural Fats and 
Oils, Chem. Eng. & Tech., 31 Iss. 5, 700-714, DOI: 10.1002/ceat.200800035 

Cogollo-Herrera K., Lombana-Carmona S., Bonilla-Correa D., Peralta-Ruiz Y.Y., 2015, Evaluation of biodiesel 
production process from sapium tree oil sebiferum using exergy analysis methodology, CET, 43, 463-468 
DOI: 10.3303/CET1543078 

Dijkstra A. J., 2006, Revisiting the formation of trans isomers during partial hydrogenation of triacylglycerol 
oils, Eu.J.Lipid Sci.Tech. 108 (2006) 249–264, DOI: 10.1002/ejlt.200500335 

Dijkstra A. J., 2010, Selectivities in Partial Hydrogenation, JAOCS, 87 Iss.1, 115-117, DOI: 10.1007/s11746-
009-1507-z 

Ilkılıç C., Aydın S., Behcet R., Aydinb H., 2011, Biodiesel from safflower oil and its application in a diesel 
engine, Fuel Proc. Thec. 92 Iss.3, 356-362, DOI: 10.1016/j.fuproc.2010.09.028 

International Standard ISO-EN 3961, 2013, Animal and vegetable fats and oils-determination of Iodine Value, 
5

th Ed., UK 
List G.R. and King J.W., 2015, Hydrogenation of fats and oils: theory and practice, 2nd Ed., Academics Press 

and AOCSPress, UK 
Romero Rivas M.J.A., Pizzi A., Toscano G., Bosio B., Arato E., 2014, Study of an innovative process for the 

production of biofuels using non-edible vegetable oils, CET, 37, 883-888 DOI: 10.3303/CET1437148 
Schneider M.P, 2006, Plant-oil-based lubricants and hydraulic fluids, J.S.F.A., 86 Iss.12, 1769-1780, DOI: 

10.1002/jsfa.2559 

120