Kinetics of the Esterifieation of Palmitie Acid

with Isobutyl Alcohol

SHIGEO GOZO*, TOMOHIKO TAGAWA, and ASMA YUSOFF

Depart ment of Chemical Engi neeni ng, Nagoya Uni nersit y, Chik usa, Nagoya, 464-01 Japan

Abstract
Kinetics of the sulfuric acid catalyzed esterification of palmitic acid with isobutyl alcohol
has been studied and reviewed critically. A theoretical rate equation is derived by the
following re- action mechanism: (1) The reaction of sulfuric acid and isobutyl alcohol
produces isobutyl sul- furic acid which acts as a catalyst; (2) Palmitic acid is protonated by
the catalyst to form the reaction intermediate; and (3) The esterification proceeds
reversibly. The experimental results are consistent with this rate equation.

Introduction
The reaction between alcohol and carboxyl acid with acid catalyst to
form ester and water is of industrial importance. The reaction is
represented by the following scheme.
(1) RCOOH + R’OH " ’“ RCOOR’ + HCO
(A) (B) (E) (W)
Many kinetic studies on this reaction have been done and several
empiri- cal rate equations were presented. Some studies suggested that the
reaction was of second order with respect to acid. The rate constant has
been corre- lated with reaction temperature, concentrations of catalyst, and
the initial ratio of reactants [1]. Dhanuka, Malshe, and Chandalia [2]
summarized the rate equation into general form and commented on its
limitations in the in- troductory part of their article.
(2) Rate = /z C'2
(3) # = [a + bCc — 8( BO// A0 )Cm] exp(—£//t, 7’)
Dhanuka et al. [2] also proposed that the sulfuric acid catalyst was con-
verted to a mono alkyl sulfuric acid ester which acted as a monovalent
acid catalyst during the esterification. They concluded that the reaction
could be expressed by a simple second order reversible kinetics with the
mono alkyl sulfuric acid catalyst.
(4) Rate = bCc(CA B — CJ Cv/K)
with R'OSO3H as a catalyst.

* To whom correspondence should be addressed.

International Journal of Chemical Kinetics, Vol. 23, 17—26 (1991)

0 1991 John Wiley & Sons, Inc. CCC 0538-8066/91/0l00l7-l0$04.00
18 GOTO, TAOAWA, AND YUSOFF

In the case of esterification of palmitic acid with isobutyl alcohol cata-

lyzed by sulfuric acid, an empirical first order kinetics was reported by
Sreeramulu and Rao [3].
(5) Rate = b EA
(6)
(7) #/ = (9.12 — 0.264( ao/ 80)} x 10" 5 exp(18.32 — 6992/Z’)
(8) lab —— 1.17 x 10" s exp(18.32 — 6992/7’)
Many data were summarized according to Goldschmidt’s equation derived
from following scheme where a protonated alcohol was assumed as a reac-
tion intermediate [4—6];
(9) RCOOH + R'OHz“ ¥ DOOR' ’F H3O+
(10) R’OHS’ + H2O R'OH + H3O’
Thus, the reaction rate was derived assuming CB being constant;
(11) Rate = AC p GB CC/( B + KBC )
where Kj is the adsorption equilibrium constant of water.
Recent advances in separation technology enabled a selective separation of
water using various methods such as pervaporation technique. This would
lead the esterification beyond the equilibrium. Thus, it is necessary to for-
mulate a theoretical rate equation which can deal with the reversibility [7,8].
In the previous article [9], the kinetics of the reaction between sulfuric acid
and isobutyl alcohol was examined and found to be of first order of sul- furic
acid. Especially, the reaction proceeded very fast at higher tempera- tures and
was almost complete within 5 min above 360 K. In this article, the
esterification of palmitic acid with isobutyl alcohol has been systematically
studied by using sulfuric acid as homogeneous catalyst. Equilibrium con-
stants were determined under various reaction conditions. A rate equation
was derived from the latest reaction mechanism and compared with previ-
ously proposed rate equations.

Derivation of Rate Equations

Recent studies in reaction mechanism concluded that a protonated car-
boxyl acid is a possible reaction intermediate [10,11];
(12) RCOOH + H* RCO Hz+
(A) (C) (Aff)
(13) RCOOHz’ + R'OH RCOOR’H’+ H,O
(Aff) (B) (EH) (W)
(14) RCOOR’H“ RCOOR’ + H’
(A//) (E)
(C)
A theoretical rate equation is derived according to the latest reaction
mechanism of eqs. (12)—(14) as follows; Prom eq. (12),
 ESTERIFICATION OF PALMITIC ACID 19

Assuming the steady states of AH and EH ions,

(16) dC,ui/dt —- — -I AH 2 AH B + k -2 C yfi C -- 0

(17) d EM/d t -— b y AH B 2 EH lY 3 EH + —3 E —- 0

Eliminating C from eqs. (16) and (17),

(18) Rate = c( i 2 3 A B

By assuming #- $3 < b2 3 B <• 2- # -sCw, eq. (18) can be reduced to

(19) Rate = b, C (/ B fi Cw/K) [ B + W -1 -2/( 2 3)]
where
(20) K -— # 2 3/( -› —2 —3)
Blank tests indicate a small quantity of conversions even if no catalysts
exist. Then, the rate constant without catalyst As should be considered.
Total rate constant can be defined by
(21) A =A,Cc+A,
Thus, the final form of the rate equation is
(22) Rate = b(C B " C fi]C /fi:)f(Cy + Kv Co)
where
(23) Jr —— # -1 -2/( 2 3)

Equation (22) is described with conversion Zy.

(24) Rate = Cxo dip ldt

where
(25) — ao I o
(26) \yo = WO AO
If the reaction is assumed to be irreversible, that is, the value of K iz
infinitive, eq. (22) can be reduced to eq. (11) derived from Goldschmidt’s
mechanism where a protonated alcohol exists [4—6].
Then, although the mechanisms are different, the final expression of rate
equation becomes identical.
Also, if Kg —— 0 and K -- infinitive, eq. (22) can be reduced to eq. (5)
which is of first order with respect to acid [3).

Integrated Forms of Rate Equations

Equation (4) for elementary reversible reaction of second order can be in-
tegrated as follows;
(27) (27)
ln[(1 — 2o t Xg/(If + 1 + z))/(1 2a t Ag/( + 1 Oz))] ' Oz # C'zo
where

(28) a, — 1 — 1/Jr
(29) O z — ((If + 1)2 4o t 3f] " 2
20 GOTO, TAGAWA, AND YUSOFF
The integrated form of eq. (5) for the irreversible reaction of first order is
given by

(30) —ln(1 — EA) - b

Equation (24) based on protonated carboxyl acid can also be integrated into
(31) = [#(1 — 1/Jr)] —'

where

(32) (Ay, A ) - [(1 + M + Wo/ K)

+ (1 -b + o/ K) — 43f(1 — 1/Jf)]/[2(1 — 1/Jf) ]
where +sign indicates Am and —sign X .
When the values of /z, K, and Kg are specified, the relation t and EA can be
calculated from eqs. (27), (30), and (32) for three reaction rates.

Experimental
In order to identify catalytic species for sulfuric acid, two starting meth- ods
were adopted in a batch reactor, that is, direct-mixed and pre-mixed methods.
Di rect-mi xed method. Palmitic acid was placed in a flask equipped with
a stirrer, a condenser, and a thermometer. It was heated in an oil bath to the
reaction temperature. Pre-heated isobutyl alcohol was then added. The
reac- tion was started by adding sulfuric acid (97%) with stirring at 300
rpm.
Pre-mixed method. Sulfuric acid was added to isobutyl alcohol at the re-
action temperature and allowed to stand for 2 h to complete the formation of
mono-isobutyl sulfuric acid. Then, the esterification was started by adding
palmitic acid to this mixture.
A 2 ml of the reaction mixture was sampled at various reaction time and ti-
trated by 0.1 N NaOH aqueous solution using phenolphthalein as an indicator.
The equilibrium constants were determined by the reaction in a sealed tube
(reactor content; 50 ml) for more than 3 days under various reaction
conditions.
The boiling point of isobutyl alcohol is 381 K and the melting point of
palmitic acid is 336 K. Then, the reaction temperature in this work was
within 380 K and 340 K.
 Results and Discussions

Equilibrium Constant
As is well known, the equilibrium of esterification may be affected by re-
action conditions [6,10,12]. In many cases, the composition of the reactants,
the kind and amount of the catalyst as well as reaction temperature af-
fected the equilibrium constant K. Therefore, practical equilibrium con-
stants must be measured for various reaction conditions to determine the
reversible kinetics.
Experimental equilibrium constants Jfexp Can be calculated from the con-
versions at the equilibrium Xp as follows;

(33)
where ester and water are not contained in the feed.
In Table I, equilibrium constants were summarized along with reaction
conditions. It is seen from Run No. 1, 2, 3, 8, and 13 at constant values of
Cz —— 17.0 mol/m3 and M = 5 that Jfp%p is increased by changing the reaction
temperature from 340 to 380 K.
Arrhenius plot for x is shown in Figure 1 and the following equation is
given from the straight line.
(34)
K,z -- 6500 exp(—24100/ft 7') at Cc —— 17.0 mol/m3 and M=5
where fi is the gas constant (8.31 I/(mol K)).

TABLE I. Equitibrium constants.

R un ñ o . Te.r pe re t ure Cp
ex ca 1
p

j7.0 86.4 1.32 1.28

35O 88.2 i . 60

17.0 91.1 240 2 . 06

3?0’ I/.O 70.0 3. t8

87.7 3.85 3 30

6 6.3 93 . 0 3.0 2 . 36

10.3 92.3 2.72 2 . 56

92.7 2.89 2.J6

9 26 . 2 93.0 3.04 2 . 56

10 33. 5 2.80 2.â6

17.O 7 3. 0 6.95

88.4
i3 17 .0 93.1 3.09 3.lñ
 k•, 1O^ [-s ] , K (—]

5 1 1 I
2.6 2.7 2.8 2.9 3.O
1OOO/T [K -']
Figure 1. Arrhenius plots of reaction rate constant l„ equilibrium constant of reaction
K, and equilibrium constant of water adsorption Kg.

Catalyst concentrations Cz were varied at constant values of 7’ = 370 K

and 3f = 5 in Run No. 6—10. The effect of Cz on x exp 1s not so significant.
Initial ratio of reactant concentration M was changed from 1 to 5 in Run
No. 4—6 at 370 K and Run No. 11—13 at 380 K. As the value of M increases,
Jfexp IS decreased. The effect of 3f at 370 K and 380 K could be summarized
into following equations, respectively:
(35) K e —— 1.25 + 7.86/(1 + M j at 370 K
(36) K,o —— 1.25 + 11.4/(1 + 3f) at 380 K
The same expression, K -— 1.31 + 4.86/(1 + ñf) was found for the esteri-
fication of acetic acid with ethyl alcohol in the literature [6].
These calculated values by eqs. (34)—(36) are also presented in Table I
and were used as the equilibrium constants in the following calculations.

Behavior of Slulfuric Acid Catal yst

Dhanuka et al. [2] insisted on that kinetics of esterification might be in-
terfered by the slow formation of alkyl sulfuric acid and re-interpreted pub-
lished data. On the other hand, in our prevous study [9], we have shown that
the reaction of sulfuric acid with isobutyl alcohol irreversibly proceeded
very fast and was completed within one min above 380 K. The produced
isobutyl sulfuric acid showed mono-valent acidic nature. Therefore, it is
possible that mono-isobutyl sulfuric acid exists and is a catalyst, during the
esterification of palmitic acid with isobutyl alcohol in the presence of sul-
furic acid.
Figure 2 compares two starting methods for the esterification at two tem-
perature levels. There are no differences between direct-mixed and pre- mised
methods. It is confirmed from these results that monoisobutyl sulfuric acid
should be quickly formed and it behaves as a catalyst from the initial stage of
the esterification. Therefore, the real concentration of cata-
 36OK. C = 15 mo\

0.8
340K. C,= 21 mot/m
0.6

0.4
Pre-mixed method
0.2 Direct-mixed method

t [h]
Figure 2. Comparison between pre-mixed and direct-mixed methods; (C = 1270 mol/
m', M —- 5, and lY = 0).

lyst (H’) for the esterification should be regarded to half as much as the ini-
tial concentration of proton in sulfuric acid.
The following data were obtained by the direct-mixed method which was
simpler than the pre-mixed method.

Effect of Reactant Composition

Figure 3 shows the effect of ratio of initial concentrations of isobutyl alco-
hol to palmitic acid (M = CBO// ). As the value of M iz increased, the con-
version at equilibrium is increased and the reaction may be regarded as
irreversible.
The effect of the initial water content ( o = +o/C ) is shown in
Figure 4. As the value of o is varied from 0 to 0.445, the conversion
becomes lower at the same reaction time. Then, strong retardation effect
was observed.

0.4 key
0.2 750
7.5250
1750

t (h]
Figure 3. Effect of initial ratio of isobutyl alcohol to palmitic acid 3f (7’ = 370 K and
0 —— 0)
 0.8

0.6’
(—]
W = 0.445 "
”0.223 _
0.4-
X/

0.2- -

0 1 2 3 4 5
I [h]
Figure 4. Effect of initial water content lY (T = 380 K, Cc —— 3.7 mol/m', Cg —
1250 mol/m’, and M = 5).

E//ect of Catal yst Amount

Figure 5 shows the effect of amount of catalyst. Even if no sulfuric acid

catalyst exists, the reaction proceeded although the reaction rate was very
low. The reaction rate is increased with the increase of the concentration
of catalyst.

Selection of Rate Equation

To check the validity for the simple second order reversible kinetics eq. (4)
adopted by Dhanuka et al. [2], our experimental data in Figure 5 were re-
plotted with the reaction time as abscissa against the left hand side of
eq. (27) as ordinates. Figure 6 shows that the relations are not linear for any
concentrations of catalyst. Therefore, eq. (4) is not valid for our data.
Dotted lines in Figure 5 indicate the results calculated by the integrated
form of the irreversible reaction of first order eq. (30) with the rate constant

Figure S. Effect of catalyst amount; (T = 380 K, AO = 1250 mol/m°, M —- 5, and

 Left hand sideo f eq.(27)

t [hl
Figure 6. Check for validity of the elementary reversible reaction, eq. (27) (Z' = 380 K,
Cq = 1250 mol/m’, M = 5, and lY = 0).

# in eqs. (6), (7), and (8) reported by Sreeramulu and Rao [2]. Our experi-
mental data did not fit well.
Finally, eq. (22) was tested by curve fitting of eq. (31) in which parameters
#, in eq. (21) and Kg in eq. (31) were adjusted. The value of Ozs could be deter-
mined as 1.05 x 10*’ s*' at 380 K from experimental data without catalyst
in Figure S.
Solid lines in Figures 2—5 indicate the results of curve fitting for eq. (31).
All experimental data are in excellent agreements with the calculated re-
sults. Then, eq. (22) is valid in this work.
Arrhenius plots of parameters #, and Kg determined by curve fitting
from all experimental data are also shown in Figure 1 and the following
equations can be obtained from straight lines.
(37) /z, = 2.98 x 105 exp(—67400/Its 7’)
(38) K -— 3.09 x 103 exp(—18100/R 7’)
By assuming the same value of activation energy for A› and #„ the rate con-
stant without catalyst â, may be expressed as
(39) #t = 1.96 x 10‘ exp(—67400/a T’)
The dotted line in Figure 1 indicates the relation of eq. (7) in Sreeramulu
and Rao [3]. The activation energy for /z is almost the same as that for #, in
this work although the rate equations are different.

Conclusion
The reaction of palmitic acid with isobutyl alcohol by using sulfuric acid
as a catalyst was studied in details. A theoretical rate equation was derived
based on the following reaction mechanisms: (1) the reaction of isobutyl al-
cohol and sulfuric acid gives a mono-isobutyl sulfuric acid which acts as the
real catalyst species; (2) the carboxy group of the palmitic acid is then proto-
nated to give a reaction intermediate; (3) the reversible reaction of proto-
nated acid and isobutyl alcohol gives the protonated ester and water; and
26 GOTO,TAGAYA,ANDYUSOFF

(4) the removal of proton gives the ester and regenerates the catalyst. The
final form of rate equation could be expressed by eq. (22). This rate equation
gave the best fitting to all the experimental results among the reported rate
equations. Kinetic parameters were given by eqs. (37)—(39). Equilibrium
constants were varied by the composition of reactants and temperature and
could be correlated by eqs. (34)—(36).

Bibliography
[Ij D. J. McCracken and P.F. Dickson, Ind. Eng. C/tem. Proc. Des. Den., 6, 286 (1969).
[2) 4. R. Dhanuka, V C. Malshe, and S. B. Chandalia, Chem. Eng. S!ci., 32, 551 (1977); and ref-
erences therein.
[31 V Sreeramulu and P.B. Rao, Ind. Eng. Chem. Proc. Des. Den., 12, 483 (1973).
[4] A. T. Williamson and C.N. Hinshelwood, Rons. ''araday Soc., 30, 1145 (1934).
[5] H.A. Smith, J. Am. Chem. Soc., 61, 254 (1939).
[6] C. Leyes, Esteri fication, in Kirk-Othmer Encyclopedia of Chem. Tech., 2nd ed., 8, 313;
and references therein.
H. Sawistowski and P.A. Pilavakis, Chem. Eng. Sci., 43, 355 (1988).
[8] H. Kita, S. Sasaki, K. Tanaka, K. Okamoto, and M. Yamamoto, Chemistry Letters, 2025
(1988).
[9) S. Goto, T. Tagawa, and Y. Fukuta, in1. ‹f. Chem. Ki net., 21, 729 (1988).
[10] E.G. Zey, Esteri fication, in Kirk-Othmer Encyclopedia of Chem. Tech., 3rd ed., 0, 291;
and references therein.
(11) I. Roberts and H.C. Urey, J. Am. Chem. Soc., 61, 2584 (19S8); D. S. Kemp and F. Vellaccio,
Organic Chemistr y, Worth Publ., 1980, p. 331.
[12j A. Darlington and W. B. Guenther, J. C/tem. Brig. Data., 12, 605 (1967).

Received February 2, 1990

Accepted July 20, 1990