J Am Oil Chem Soc (2014) 91:1643–1650
1649
with the mass transfer coefficient calculable with a variety
of correlations.
activation energies for both reactions have been estimated.
The activation energy of the first reaction was 63.7 kJ/mol
and the activation energy of the second reaction was
45.6 kJ/mol. From these values, the conversion of octade-
canoic acid and the selectivity to the desired product as
functions of temperature, space velocity, reactor pressure,
and inlet octadecanoic acid concentration were presented.
The model predicts maximum productivity of 1-octadeca-
nol per mass of octadecanoic acid occurs at higher tem-
peratures with short residence times.
The above argument supporting the fact that hydrogen,
treated either as a first order or pseudo-zero order com-
ponent, manifests itself as an excess reactant and permits
one to strongly examine the kinetic expression describing
IsoTherming with respect to the acid feed concentration.
Based on the Biot number calculations and Arrhenius
dependency of the data, it is likely that the Axens catalyst
was operated under conditions that were not diffusion
limited since no break in the slopes of Figs. 2 and 3 were
Acknowledgments The authors acknowledge Mr. John Coleman of
Process Dynamics who provided training, manpower, and technical
support during the processing and Dr. Glen Ackridge of Process
Dynamics who assisted with analysis of the reaction products.
Funding was provided by the Ralph E. Martin Department of
Chemical Engineering and the Ford Foundation.
observed. The values of k determined by the model are the
i
observed reaction rate constant, the product of the intrinsic
value times and the interparticle effectiveness factor
(kintrinsic g). In the absence of catalyst physical property
data, the effectiveness factor cannot be calculated. Never-
theless, the k values can be used to examine the sensitivity
i
of the system to various combinations of temperature,
pressure and feed since the value of g for a first-order
reaction is independent of external concentration.
References
1
2
. Costa J, Cunill F, Izquierdo JF (1984) Kinetic study of gas phase
catalytic hydrogenation of furfural. An Quim Ser A 80:586–592
. Soner E, Dogu T, Yorulmaz Y (1996) Kinetics of hydrogenation
of furfural to furfuryl alcohol over copper chromite catalyst and
optimum regeneration procedures. Arabian J Sci Eng 21:321–332
. Adkins H, Connor R (1931) Catalytic hydrogenation of organic
compounds over copper chromite. J Am Chem Soc 53:1091–1095
. Willemart A, Loury M, Everaerts P (1948) The preparation of
dodecyl alcohol by catalytic hydrogenation of the total esters of
coconut butter. Oleagineux 3:68–71
The 1-octadecanol is the desired product and maximi-
zation of this product can be studied with parametric plots
based on the developed model. One such parametric plot is
shown in Fig. 7. The trends shown in Fig. 7 indicate that
better productivity was achieved at higher temperatures
with higher space velocities, i.e., shorter residence times in
the reactor. Although our experimental work found com-
plete conversion to the undesired product octadecane at
temperatures above 351 °F, this was at relatively long
residence times. The trend analysis presented in Fig. 7
predicts that optimum productivity in this reactor for
3
4
5
6
7
. Richardson AS (1945) Catalytic hydrogenation of fatty acids to
form alcohols, US Patent 2375495
. Aly M, Baumgarten E (2001) Hydrogenation of hexanoic acid
with different catalysts. Appl Catal A 210:1–12
. Hoffmann AN, Montgomery JB (1955) Tall-oil alcohols, US
Patent 2727885
1
-octadecanol would occur at 566 °F with a residence time
of 0.1 h. Further investigations will explore the data space
around this set of conditions to confirm that the model
developed can predict conditions of high productivity.
Minor modifications to the hardware will enable these
investigations. Thus, we have shown that by understanding
the reaction kinetics of both reactions one can achieve high
productivity without having to change catalysts.
8. Nakaoka H, Mototani M (2003) Liquid vegetable unsaturated
alcohol and process for producing the same, JP Patent
2
003089393
9
. Manyar HG, Paun C, Pilus R, Rooney DW, Thompson JM,
Hardacre C (2010) Highly selective and efficient hydrogenation
of carboxylic acids to alcohols using titania supported Pt cata-
lysts. Chem Commun (Cambridge, UK) 46:6279–6281
1
1
0. Deshpande VM, Ramnarayan K, Narasimhan CS (1990) Studies
on ruthenium-tin boride catalysts. II. Hydrogenation of fatty acid
esters to fatty alcohols. J Catal 121:174–182
1. Sanchez MA, Mazzieri VA, Sad MR, Grau R, Pieck CL (2011)
Influence of preparation method and boron addition on the metal
function properties of Ru-Sn catalysts for selective carbonyl
hydrogenation. J Chem Technol Biotechnol 86:447–453
2. Miyake T, Makino T, Taniguchi S, Watanuki H, Niki T, Shimizu
S, Kojima Y, Sano M (2009) Alcohol synthesis by hydrogenation
of fatty acid methyl esters on supported Ru-Sn and Rh-Sn cata-
lysts. Appl Catal A 364:108–112
Conclusions
1
-Octadecanol was made in a liquid-phase trickle-bed
1
catalytic reactor by the catalytic hydrogenation of octade-
canoic acid. A model of the parallel-series reactions was
generated and correlated to experimental values of hydro-
genation conducted in a plug flow reactor at pressures of
1
3. Ackerson MD, Byars MS (1998) Two phase hydroprocessing. US
Patent 9859019
7
00, 1,400 and 2,000 psig (13.8 MPa) hydrogen and at
14. Ackerson MD, Byars MS (2005) Two-phase hydroprocessing
with an eliminated hydrogen recirculation, US Patent
varying temperatures and space velocities. The model most
useful for this system was found to be a series–parallel
reaction first order in octadecanoic acid and pseudo-zero
order in hydrogen. The Arrhenius frequency factors and the
2
0050082202
1
5. Patterson HBW (2011) Hydrogen process techniques. In: List
GR, King JW (eds) Hydrogenation of fats and oils: theory and
practice. AOCS Press, Urbana, IL, pp 33–48
1
23