1375
ANALYSIS OF SOYBEAN OIL BY CGC
9. Nohl, A., Fast Capillary Chromatography, Spectra-Physics,
3333 North First Street, San Jose,CA (no date}.
10. D'Alonzo, R.P., W.J. Kozarekand H.W. Wharton, J. Amer.
Oil Chem. Soc. 58:215 (1981).
11. Devinat, G., J.L. Coustille, J.L. Perrin and A. Prevot, Rev. Fr.
Corps Gras 30:463 (1983).
Transesterification Kinetics of S o y b e a n Oil1
Bernard Freedman*, R o y d e n O. Butterfield a n d Evereff H. Pryde~
Northern Regional Research Center, Agricultural Research Service, U.S. Department of Agriculture, Peoria, IL 61604
Transesterification of s o y b e a n oil (SBO) a n d o t h e r
triglycerides w i t h alcohols, in t h e presence of a catalyst,
y i e l d s f a t t y e s t e r s a n d glycerol. Di- a n d monoglycerides
are i n t e r m e d i a t e s . R e a c t i o n s are consecutive a n d
reversible. Rate c o n s t a n t s have been determined for
each reaction with a computerized kinetic program. The
e f f e c t s of t h e type of alcohol, 1-butanol or m e t h a n o l
(MeOH); m o l a r r a t i o of alcohol to SBO; type a n d a m o u n t
of catalyst; a n d r e a c t i o n temperature o n rate c o n s t a n t s
a n d kinetic o r d e r were examined. Forward reactions
appear to be pseudo-first o r d e r or s e c o n d o r d e r
d e p e n d i n g upon c o n d i t i o n s used. Reverse r e a c t i o n s
a p p e a r to be s e c o n d order. A t a m o l a r r a t i o of
MeOH/SBO o f 6:1, a s h u n t reaction was observed.
Energy of activation was determined for all forward a n d
r e v e r s e r e a c t i o n s u n d e r a variety of e x p e r i m e n t a l
c o n d i t i o n s from plots of log k v s 1/T. Values r a n g e d
from 8-20 kcal/mol.
T r a n s e s t e r i f i c a t i o n o f v e g e t a b l e oils with simple
alcohols has long been a preferred m e t h o d for p r e p a r i n g
f a t t y esters. These e s t e r s have good p o t e n t i a l as an
a l t e r n a t i v e o r emergency fuel t o replace #2 diesel oil in
farm t r a c t o r s . Because of this potential, we have focused
our a t t e n t i o n on variables t h a t affect the yield of these
e s t e r s {1,2} as well as analysis of intermediates and
products as determined b y a n I a t r o s c a n a n a l y z e r (3}. In
t h e present p a p e r we report on a s t u d y of the kinetics of
t h e transesterification of soybean oil (SBO). A capillary
gas chromatographic (CGC} m e t h o d used to support this
s t u d y has been described (4). One of o u r objectives was
to determine how variations in type of alcohol, molar
ratio of alcohol t o SBO, c a t a l y s t type and reaction
temperature affected kinetic order, reaction rates and
energies of activation. Another objective was to gain a
more f u n d a m e n t a l u n d e r s t a n d i n g o f t h e chemistry
involved in the transesterification of vegetable oils.
Various mechanisms have been proposed for both
acid- and alkaline-catalyzed transesterification {5-7}. A
n u m b e r o f kinetic studies have been r e p o r t e d for the
transesterification o f simple {non-fatty) e s t e r s with
alcohols {6,8-10} a n d d i m e t h y l t e r e p h t h a l a t e with
ethylene g l y c o l {11,12). Only a few kinetic studies have
dealt with the transesterification of vegetable oil or f a t t y
esters. These include t h e conversion of c a s t o r oil t o
*To whom correspondence should be addressed.
'Presented at the AOCS meetingin Philadelphia, PA, May, 1985.
2Deceased.
12. See, J.B., Fette, Seifen, Anstrichm. 85:72 (1983).
13. Goh, E.M., and R.E. Timms, J. Amer. Oil Chem. Soc. 62:730
(1985).
14. Haefelfinger,P., J. Chromatogr. 218:73 (1981}.
[Received D e c e m b e r 13, 1985]
m e t h y l ricinoleate {7), t h e glycerolysis of m e t h y l o l e a t e
(13) a n d t h e t r a n s e s t e r i f i c a t i o n o f t h e e s t e r s o f
9(10)-carboxystearic acid (14} and m e t h y l 9(10}-car-
b o x y m e t h y l s t e a r a t e {15}. T h e p r e s e n t s t u d y was
u n d e r t a k e n with a commercially important vegetable
oil, SBO, utilizing a computerized kinetic p r o g r a m (16)
t o shed additional light on the kinetics o f transesterifica-
tion reactions.
EXPERIMENTAL
Materials. Refined SBO was o b t a i n e d from C e n t r a l
S o y a , Chattanooga, Tennessee. T h e molecular w e i g h t
was a s s u m e d t o be 872.4. M e t h a n o l (MeOH) a n d
1-butanol (BuOH) were MCB Omnisolv (spectrograde)
a n d were stored over molecular sieves 4A. A solution o f
s o d i u m butoxide (NaOBu) was p r e p a r e d b y the r e a c t i o n
o f 240 mg o f s o d i u m with 10 ml o f BuOH a t 59-99 C with
stirring. Sulfuric acid was purchased from B&A, Allied
Chemical, Morristown, New J e r s e y . S o d i u m methoxide
(NaOCH3) was obtained from Aldrich Chemical Com-
p a n y , Milwaukee, Wisconsin.
Reaction conditions employed in kinetic studies. The
scope of our kinetic studies is outlined in Table 1.
Experiments were designed t o determine kinetic o r d e r
and r a t e constants u s i n g two simple alcohols, two molar
r a t i o s o f alcohol t o SBO, acidic vs alkaline catalysis, two
w e i g h t percentages o f NaOBu and two temperature
r a n g e s each containing five temperatures. All reactions
studied were conducted a t atmospheric pressure.
Transesterification reaction and sampling. T h e f i r s t
two of five systems shown in Table 1 utilized a 250-ml
round-bottomed four-necked flask, equipped with a
m e c h a n i c a l stirrer, thermometer, condenser a n d / o r
d r y i n g tube and stopper (for sample removal}. T h e
remaining s y s t e m s in Table 1 e m p l o y e d a 400-ml
c u s t o m - m a d e r e a c t i o n f l a s k with five n e c k s . T h e
a d d i t i o n a l neck was equipped with a thermocouple
connected t o a d i g i t a l pyrometer. This p y r o m e t e r
provided rapid temperature readings t h a t were essential
t o follow t h e very f a s t reactions. The l a r g e r size f l a s k
e n a b l e d u s t o scale u p t h e r e a c t i o n and p o s s i b l y
permitted more vigorous agitation.
The reaction f l a s k was immersed in a Polyscience
Series 9000 refrigerated constant temperature b a t h a n d
circulator. The o p e r a t i n g r a n g e of this b a t h was - 1 5 t o
+150 C, with a control accuracy o f +_0.02 C. T h e
temperature c o n t r o l o f t h e r e a c t i o n m i x t u r e was
generally ca. 0.1 C.
JAOCS, Vol. 63, no. 10 ( O c t o b e r 1986)
1376

B. FREEDMAN, R.O. BUTTERFIELD AND E.H. PRYDE

TABLE 1
R e a c t i o n Conditions Employed i n K i n e t i c Studies
Molar ratio Weight of reactants
of
Alcohol used alcohol/SBO Alcohol (g) SBO (g) Catalyst Temperature (C)a

1-Butanol 30:1 51 20 1% H~SO4 77,87,97, 107, 117

1-Butanol 30:1 51 20 1% NaOBu 20, 30, 40, 50, 60
1-Butanol 30:1 102 40 0.5% NaOBu 20, 30, 40, 50, 60
1-Butanol 6:1 51 100 1% NaOBu 20, 30, 40, 50, 60
Methanol 6:1 22 100 0.5% NaOCH3 20, 30, 40, 50, 60

aTemperature control: 0.1C.

The r e a c t i o n f l a s k and its associated e q u i p m e n t and Overall Reaction:

o t h e r g l a s s a p p a r a t u s used f o r h a n d l i n g s t a r t i n g catalyst
Triglyceride (TG) + 3ROH 3R'CO~R + glycerol (GL)
materials were f i r s t h e a t e d t o expel moisture. Calculated
a m o u n t s o f SBO and alcohol were a d d e d t o t h e reaction
S t e p w i s e Reactlons:
f l a s k t h a t was b r o u g h t t o a predetermined temperature
TG-DG(k,)
u s i n g t h e c o n s t a n t temperature b a t h . A calculated 1. T G + R O H _ diglceride (DG) + R'CO2R
a m o u n t o f c a t a l y s t was quickly added to the reactants, DG-TG(k,)
and timing o f the reaction was begun.
Samples (0.5 ml each} were r e m o v e d from the reaction DG-MG(k2)
mixture a t various intervals, depending upon r e a c t i o n 2, DG + ROH monoglyceride (MG) + R'C%R
MG DG(ks)
conditions. F o r example, alkali-catalyzed reactions a t
60 C were s a m p l e d initially a t 15-sec intervals and then MG-GL(k3)
over l o n g e r intervals d u r i n g the course of one hr. F o r the 3, MG + ROH- ~ glycerol + R'CO2R
GL-MG(k~)
acid-catalyzed reaction a t 117 C, early samples were
t a k e n a t 6-min intervals, and l a t e r samples were t a k e n
a t 30-min intervals d u r i n g t h e course o f t h r e e hr. FIG. 1. Chemistry of the t r a n s e s t e r i f i c a t i o n of v e g e t a b l e o i l s .
Samples were immediately quenched in 0.5 ml water,
thus stopping the reaction. Because the oil l a y e r did not
always cleanly s e p a r a t e from the water, s a l t i n g out and
c e n t r i f u g i n g were used t o enhance separation. The 100
r e s u l t i n g oil was used f o r both thin l a y e r chromatog-
r a p h y (TLC) and CGC analysis. 80 . *x"/ , . , - / " /
TLC was used initially to determine t h e q u a l i t a t i v e
composition of each sample of the r e a c t i o n mixture and
t o determine w h i c h samples were t o be quantitatively
analyzed. This TLC p r o c e d u r e has been described ~ 4o
previously {2). The r e a c t i o n was stopped when TLC
indicated high conversion to the ester. F o u r t e e n samples /..-- / .," .,/
were t a k e n , of w h i c h nine were quantitatively analyzed
b y the CGC. T h e CGC m e t h o d provided a q u a n t i t a t i v e 0 1 , . ~ , , i l l . . . . . . . . I

analysis of t h e triglyceride (TG), diglyceride (DG), 0.1 0.2 0.5 1 2 5 10 20

monoglyceride (MG) and e s t e r composition needed for
t h e kinetic study.
Preparation o f CGC samples and their analysis.
Samples from t h e oil layers o b t a i n e d a b o v e were
accurately weighed and silylated, i n t e r n a l s t a n d a r d was
added, and t h e concentration was adjusted and then
a n a l y z e d b y CGC as p r e v i o u s l y d e s c r i b e d (4). A
computerized p r o g r a m was used to analyze CGC d a t a .
The o u t p u t from this p r o g r a m gave wt %, ~mol or mmol,
s t a n d a r d deviation and relative s t a n d a r d deviation f o r
each component.
Computerized kinetic program. A computerized
kinetic program, described previously 06), was used to
determine w h e t h e r a proposed kinetic o r d e r appeared
possible. The p r o g r a m requires t h a t a specific kinetic
scheme be proposed for each reaction b e i n g studied. The
reactions studied are shown in Figure 1. Additional
Time, hr
FIG. 2. Effect of time and temperature o n ester formation.
B u O H / S B O (30:D, 1% H~SO4, 77-117 C. X, 117 C; o, 107 C; a , 97 C;
m , 87 C; + , 77 C .
i n p u t consists o f the concentration of each component
with time. Concentration u n i t s were e i t h e r mmol o r
~mol/ml depending on requirements of the program. T h e
p r o g r a m utilizes t h e s e d a t a t o p r o d u c e p l o t s o f
concentration vs time, w h i c h are shown later. A s e p a r a t e
plot is produced f o r each t e m p e r a t u r e . B a s e d upon t h e
kinetic scheme b e i n g tested, t h e p r o g r a m a t t e m p t s t o
draw a line t h r o u g h the points. A close fit of the lines t o
t h e p o i n t s suggests t h a t the kinetic scheme e m p l o y e d is
adequate, and from this a kinetic o r d e r can be proposed.
On t h e o t h e r h a n d , a poor fit of the lines to the p o i n t s

JAOCS, Vol, 63, no, 1 0 ( O c t o b e r 1986)

 ]377

TRANSESTERIFICATION KINETICS OF SBO

suggests t h a t an incorrect scheme was proposed and/or 60

t h e d a t a are f a u l t y .
T h e p r o g r a m also r e p o r t s two t y p e s o f errors
50 J
associated with t h e fit o f t h e lines to the points a t each
t e m p e r a t u r e . The f i r s t is t h e " s u m m e d squared error,"
w h i c h is the s u m o f all v e r t i c a l distances squared 4O J

b e t w e e n p o i n t s a n d lines. The g r e a t e r t h e coincidence

b e t w e e n p o i n t s and lines, the smaller t h e error. The
second e r r o r r e p o r t e d is t h e " a v e r a g e error," w h i c h is the
s q u a r e root o f t h e summed squared e r r o r divided by the
n u m b e r of d a t a points (concentrations). By comparing
t h e t o t a l of e i t h e r t h e summed squared or a v e r a g e errors
o f t h e various schemes b e i n g t e s t e d a t all temperatures, ~ - ~ - - -Li- . . . . . . . . I
one can determine w h i c h scheme gives the best fit. O ~ ~- J I I ~ L - ~l ~ - - i-~

Finally, t h e c o m p u t e r p r o g r a m r e p o r t s r a t e constants 0 25 50 75 100 125 150 175 200

f o r each r e a c t i o n a t each t e m p e r a t u r e .
RESULTS AND DISCUSSION
Chemistry of the transesterification of vegetable oils.
T h e stoichiometry o f t h e overall transesterification
r e a c t i o n requires 1 mol of TG for 3 mol o f alcohol to give
3 mol of e s t e r a n d 1 mol of glycerol (GL) (Fig. 1). This
reversible reaction, w h i c h is e i t h e r acid- or alkaline-
catalyzed, involves stepwise conversions of TG to DG to
M G t o GL producing 3 mol of e s t e r in the process. One
o b j e c t i v e of this s t u d y was to determine t h e r a t e
constants for all forward and r e v e r s e reactions. R a t h e r
t h a n use k, t o designate the r a t e constant for the
conversion o f the TG t o DG, we have chosen to use the
designation TG-DG. Similarly, designations for the
o t h e r r a t e constants are shown in Figure 1. It should be
n o t e d t h a t the transesterification r e a c t i o n consists of
consecutive as well as reversible reactions t h a t m u s t be
accounted f o r in any kinetic study.
E f f e c t o f time and temperature on ester formation. To
determine the effect of temperature on e s t e r formation
for t h e reaction of BuOH with SBO a t a molar ratio of
30:1 catalyzed b y 1% H2SO4, five temperatures r a n g i n g
from 77-117 C were examined. The r e s u l t s of this s t u d y
showed t h a t e s t e r formation is essentially complete in
3 h r a t 117 C, c o m p a r e d t o 20 h r a t 77 C (Fig. 2). It will be
n o t e d t h a t the curves have spacing w h i c h indicates a
d i s t i n c t r a t e change for every 10 C temperature r i s e . The
c u r v e s also i n d i c a t e t h e r e a c t i o n begins a t a slow r a t e ,
proceeds a t a f a s t e r r a t e and then slows again as the
r e a c t i o n n e a r s completion. The shapes of the curves a t
two adjacent temperatures enabled u s to p r e d i c t the
approximate shapes o f the curves f o r t h e remaining
t h r e e temperatures. This information was valuable for
determining both when samples should be t a k e n and
when to stop the reaction. Similarly shaped curves for
t h e esters were observed for alkaline-catalyzed reac-
tions.
Kinetics of acid-catalyzed transesterification reac-
tions. W e u n d e r t o o k a kinetic s t u d y of t h e r e a c t i o n of
BuOH with SBO u s i n g a 30:1 molar ratio with 1%
H2SO4 a t five temperatures. A 30:1 molar ratio was
c h o s e n b e c a u s e a t m o l a r r a t i o s of 6:1 a n d 20:1,
conversion to e s t e r was unsatisfactory (1). The r e s u l t s
obtained a t 117 C are shown in Figure 3. Because we
used a large excess of alcohol over t h a t r e q u i r e d by the
stoichiometry, we e x p e c t e d t h e forward reactions to
follow pseudo-first-order kinetics. The r e v e r s e reac-
Time, min
FIG. 3. Change of component concentration with time for a m o l a r
r a t i o of B u O H / S B O (30:1), 1% H2SO4, 117 C . . , S o y b e a n b u t y l
esters; A , s o y b e a n oil; X , diglyceride; I , monoglyceride.
100
f i , ~ A r . . ~ ~--~- - . . . . . . . . . . . . . . . . . . . .
BO
t~,?n
- j !
~ 4[1
20
L i I i I I
2 64 8 10 12
Time, rain
FIG. 4. Change of component concentration with time for a m o l a r
r a t i o of B u O H / S B O (6:2), 1% NaOBu, 3 0 C. e, S o y b e a n butyl esters;
I , monoglyceride X 10; A , s o y b e a n oil; X , diglyceride.
tions would be e x p e c t e d t o be second order. The fit o f the
lines to t h e p o i n t s was excellent for all four components.
Similarly, good fits were obtained for t h e o t h e r four
temperatures s t u d i e d . T h e s e f i n d i n g s s u p p o r t o u r
kinetic a s s i g n m e n t f o r t h e forward a n d r e v e r s e
reactions. The curves for all four components s h o w n in
Figure 3 are characteristic of consecutive reactions
a n d confirm t h a t t h e t r a n s e s t e r i f i c a t i o n r e a c t i o n
consists of a series of consecutive reactions. It should be
n o t e d t h a t b o t h t h e forward and reverse reactions are
consecutive. The r a t e constants o b t a i n e d f o r t h e s e
reactions are discussed later.
K i n e t i c s of alkaline-catalyzed transesterification
reactions. Alkaline-catalyzed transesterifications pro-
ceed a t considerably f a s t e r rates than acid-catalyzed
transesterifications (2,17). Partly for this r e a s o n and
partly because alkaline c a t a l y s t s are less corrosive t o
i n d u s t r i a l e q u i p m e n t t h a n acid catalysts, m o s t commer-
cial transesterifications are conducted with alkaline
c a t a l y s t s . S o d i u m alkoxides are a m o n g t h e m o s t
efficient c a t a l y s t s used for this purpose (5,6,18) a n d were
therefore used in o u r study.
The molar ratio of alcohol t o SBO has a n important
b e a r i n g on the kinetic o r d e r of the reaction. W e have

JAOCS, Vol. 63, no. 10 (October 1986)

 1378
B. FREEDMAN, R.O. BUTTERFIELD AND E.H. PRYDE
s h o w n t h a t acid catalysis o f t h e forward r e a c t i o n
employing a 30:1 molar ratio o f BuOH to SBO followed
pseudo-first o r d e r kinetics. This was e x p e c t e d because
o f t h e large molar e x c e s s o f alcohol. Similarly, we
e x p e c t e d alkaline-catalyzed transesterifications u s i n g a
30:1 molar ratio o f alcohol t o SBO t o follow pseudo-first
o r d e r kinetics. W e conducted two studies, both u s i n g
m o l a r r a t i o s o f BuOH t o SBO o f 3 0 : 1 , a t five
temperatures b e t w e e n 20 a n d 60 C in w h i c h t h e
catalysts were 1% and 0.5% NaOBu b y weight. The d a t a
obtained from t h e s e experiments were used in compu-
terized p l o t s of concentration vs time as previously
discussed. These p l o t s strongly supported o u r assump-
tion t h a t t h e NaOBu-catalyzed forward reactions a t
30:1 followed pseudo-first o r d e r kinetics.
W e next studied the effect of lowering t h e molar ratio
from 30:1 t o 6:1 with 1% NaOBu as catalyst. It has
previously been shown t h a t with alkaline catalysis, a 6:1
molar ratio o f alcohol to vegetable oil results in high
conversions (93-98%) o f vegetable oil to e s t e r , whereas
below t h e 6:1 ratio conversion decreases (1). A 6:1 molar
ratio would therefore p r o b a b l y b e t h e preferred ratio in
commercial operations. A t this ratio a second-order
reaction appeared more likely than a pseudo-first-order
reaction because the e x c e s s alcohol was only twice t h a t
of theory c o m p a r e d t o 10 times t h a t of t h e o r y for a 30:1
r a t i o . B o t h schemes, however, were t e s t e d u s i n g t h e
computerized kinetic program. A g a i n it was a s s u m e d
t h a t t h e consecutive, r e v e r s e reactions were second
order. This p r o g r a m indicated a b e t t e r fit of lines t o
p o i n t s f o r t h e second-order r e a c t i o n t h a n f o r t h e
p s e u d o - f i r s t - o r d e r r e a c t i o n . T h e summed s q u a r e d
errors for the second-order reactions a t 2 0 - 6 0 C were
46.5, 3.4, 17.4, 27.7 and 22.7, for a t o t a l of 117.7. The
summed s q u a r e d errors for t h e pseudo-first-order
reactions a t 2 0 - 6 0 C were 235.2, 7.1, 33.2, 44.2 and 25.6,
f o r a t o t a l of 345.3. T h e a v e r a g e errors f o r t h e
second-order reactions were 1.3, 0.4, 0.8, 1.0 and 0.9, for
a t o t a l of 4.4. The average errors for the pseudo-first-
o r d e r reactions were 2.9, 0.5, 1.1, 1.3 and 1.0, for a t o t a l
o f 6.8. Thus both the t o t a l of t h e summed squared errors
and t h e a v e r a g e errors were l o w e r for the second-order
r e a c t i o n c o m p a r e d to t h e pseudo-first-order reaction.
These differences a p p e a r t o be significant and were the
basis f o r choosing the second-order reaction. A plot o f
data f o r component concentration vs time for a 6:1 molar
ratio of BuOH to SBO with 1% NaOBu a t 30 C,
a s s u m i n g a second-order reaction, is shown in Figure 4.
The fit of the lines t o the p o i n t s in this case was
excellent. Good fits were also observed a t the o t h e r four
temperatures. These r e s u l t s suggest t h a t a t a 6:1 molar
ratio with 1% NaOBu, t h e r e a c t i o n of BuOH with SBO
follows consecutive, second-order kinetics.
W e also expected the reactions of M e O H with SBO a t
a 6:1 molar ratio with 0.5% NaOCH3 a t 2 0 - 6 0 C t o b e
second order. MeOH, however, behaved differently from
BuOH in this r e g a r d . Although a second-order scheme
was evaluated by our c o m p u t e r p r o g r a m for the r e a c t i o n
of MeOH/SBO a t 6:1, t h e average e r r o r obtained was
s o m e w h a t l a r g e r t h a n t h a t observed with BuOH/SBO a t
6:1. This increase in e r r o r is because o f the rapid
appearance of e s t e r w i t h o u t a corresponding rapid
i n c r e a s e a n d d e c r e a s e o f D G a n d MG. B e c a u s e
consecutive r e a c t i o n s could n o t explain t h e rapid
JAOCS, Vol. 63, no. 10 ( O c t o b e r 1986)
appearance of e s t e r w i t h o u t t h e appearance of inter-
mediate D G and MG, a shunt-reaction scheme was
tested. This shunt-reaction scheme c o m p r i s e d both the
second-order reactions and a s h u n t r e a c t i o n in w h i c h 3
mol of M e O H directly a t t a c k e d 1 mol o f TG. By u s i n g t h e
shunt-reaction scheme, we obtained m u c h b e t t e r fits of
t h e d a t a . Thus, t h e kinetics w h i c h best describe t h e
r e a c t i o n o f MeOH/SBO a t 6:1 molar ratio consist of a
combination of second-order consecutive and f o u r t h -
o r d e r s h u n t reactions.
The shunt-reaction scheme was f u r t h e r supported by
comparing certain r e s u l t s obtained from kinetic studies
o f both t h e s h u n t a n d second-order schemes. Two
criteria were used in this comparison. The f i r s t was a
comparison of summed squared and a v e r a g e errors. The
summed s q u a r e d errors f o r t h e second-order scheme a t
2 0 - 6 0 C were 1.5, 3.4, 2.9, 6.7 and 4.1, all values X 10-5,
for a t o t a l of 18.6 10-5. The summed squared errors f o r
t h e shunt-reaction scheme a t 2 0 - 6 0 C were 1.5, 0.4, 1.9,
1.8 and 1.5, all values 10-5, f o r a t o t a l of 7.1 X 10-5. The
average errors for t h e second-order scheme were 0.8,
1.1, 1.0, 1.6 and 1.3, all values 10-3, f o r a t o t a l of 5.8
10-3. The average errors for the shunt-reaction scheme
were 0.8, 0.4, 0.8, 0.8 and 0.8, all v a l u e s 10-3, for a t o t a l
of 3.6 10-3. Thus b o t h the t o t a l of the summed squared
errors a n d t h e a v e r a g e errors were l o w e r f o r t h e
shunt-reaction scheme c o m p a r e d t o the second-order
scheme. A plot o f d a t a for component concentration vs
time for MeOH/SBO a t a 6:1 molar ratio with 0.5%
NaOCH3 a t 40 C, a s s u m i n g t h e shunt-reaction scheme,
is shown in Figure 5. The close fit of lines t o points, a t
this temperature and t h e o t h e r four temperatures
studied, strongly supports t h e validity o f this scheme.
T h e s e c o n d criterion was correlation coefficients
obtained b y linear regression from p l o t s o f log k vs 1/T.
These p l o t s were used to determine e n e r g y o f activation,
E, and are discussed later. The correlation coefficients
obtained with t h e shunt-reaction scheme were generally
h i g h e r t h a n those from the second-order scheme. Thus,
t h e s h u n t r e a c t i o n m u s t be used in conjunction with the
second-order r e a c t i o n t o fully explain the kinetics for
t h e reaction of M e O H with SBO a t a 6:1 molar r a t i o .
S h u n t reactions have been observed in kinetic studies o f
t h e hydrogenation o f conjugated d i e n e (19) and SBO and
linseed oils (20).
W e next reexamined the r e a c t i o n of BuOH/SBO a t 6:1
0.12
4 ~ f t r
0.09 -of
0.06
L~_
0.03 --m,
/
0.0 1 I I I I
5 10 15 20 25 30
Time, min
FIG. 5 . Change of component concentration with time for a m o l a r
r a t i o of MeOH/SBO (6:2), 0.5% N a O C H 3 , 4 0 C. s , S o y b e a n methyl
esters; X , diglyceride X 10; l , monoglyceride X 10; b , s o y b e a n oil.
 1379

TRANSESTERIFICATION KINETICS OF SBO

t o determine if the shunt-reaction scheme was applic-
able. Our computer p r o g r a m calculated a zero reaction
r a t e , clearly indicating that the s h u n t reaction did not
e x i s t for the BuOH reaction. Why was the transester-
ification of SBO w i t h MeOH different from that w i t h
BuOH at the 6:1 m o l a r ratio? Two possible explanations
can be proposed. The smaller size of the MeOH molecule
relative t o that of BuOH could facilitate the simultane-
ous a t t a c k of MeOH on all t h r e e chains of the TG. Also,
the greater polarity of the methoxy anion compared t o the
b u t o x y a n i o n m i g h t promote its a t t a c k on the ester
carbonyl according t o established mechanisms.
R a t e constants. Rate constants were calculated by our
computer p r o g r a m for all reactions included in our
kinetic s t u d y . The rate constants determined for the
reaction of BuOH w i t h SBO at a 30:1 m o l a r r a t i o u s i n g
various catalysts are shown in T a b l e 2. Because the
forward and reverse reactions are not of the same kinetic
o r d e r at a 30:1 m o l a r ratio, a direct comparison of t h e s e
rate constants c a n n o t be m a d e . The alkaline-catalyzed
reactions had much l a r g e r rate constants than the acid-
catalyzed r e a c t i o n s for b o t h f o r w a r d and r e v e r s e
reactions, as expected. The h i g h e r rate constants for the
TABLE 2
Effect of Various C a t a l y s t Systems on Rate C o n s t a n t s
0.5% NaOBu- vs 1% NaOBu-catalyzed reactions was
unexpected. This difference probably resulted because
the 0.5%- and l%-catalyzed reactions were run on
different scales (see T a b l e 1) and w i t h different reactors
(see Experimental). The l a r g e r size of the 400-ml f l a s k
may have permitted more vigorous agitation, w h i c h
could have influenced reaction rates. Had the reactions
been run u n d e r identical conditions, r a t e s for the
l%-catalyzed reactions should have been h i g h e r than
those for the 0.5%-catalyzed reactions. A n u m b e r of
kinetic studies have shown that reaction r a t e s increase
w i t h increasing catalyst concentration (7,8,14).
E n e r g y o f activation. The relationship b e t w e e n
reaction rate constant, k, and temperature is g i v e n by
the integrated form of the Arrhenius equation: log,ok =
(-Eo/2.303R)/T + C w h e r e E, is the e n e r g y of activation,
R the gas constant in calories mo1-1 d e g r e e-1, T the
absolute temperature and C a constant. From a plot of
log k vs l/T, the slope can be determined. This slope is
e q u a l t o (-E,/2.303R). Thus, E, = - 4 . 5 8 (slope).
We determined e n e r g y of activation in this m a n n e r for
most of the reactions. An e x a m p l e is s h o w n in F i g u r e 6
for the acid-catalyzed butanolysis of TG t o DG at
77-117 C. L i n e a r regression analysis of t h e s e data gave
a slope of - 3 2 5 8 . 1 w i t h a correlation coefficient of

BuOH/SBO, 30:1, rate constants X 10~ -1.5 117 ~ ' ~ 1

Rate 1% H~SO4, 1% NaOBu, 0.5% NaOBu, -1.7 07 C
designation 77 C 60 C 60 C \
-1.9 ~97 C
\
TG-DG 3 3,822 26,626
DG-MG 8 1,215 3,584 -2.~ \
MG-GL 7 792 2,373
DG-TG 0.02 121 439 -2.3
MG-DG 0.05 7 8
GL-MG 0.03 11 7 -2.5 L__ L ___~- J _ _ _
0.25 0.26 0.27 0.28 0.29
For the forward reactions units are min-L For the reverse reactions lIT x 10~
units are ml mmol-~ rain-' for the acid-catalyzed reaction, and ml
~no1-1 rain-' for the alkali-catalyzed reaction. TG, triglyceride; DG, FIG. 6. Plot of log k v s l i T for the b u t a n o l y s i s of triglyceride to
diglyceride; MG, monoglyceride; GL, glycerol. diglyceride with 1% H2804 at 77-117 C.

TABLE 3
Energy of A c t i v a t i o n (E.) Under Studied Conditions
E. (caYmol): reaction conditions

BuOH/SBO, 30:1 BuOH/SBO, 6:1 MeOH/SBO,6:1

1% H2SO4 1% NaOBu 0.5% NaOBu 1% NaOBu 0.5% NaOCH3

Rate designation 77-117 C 20-60 C 20-60 C 20-60 C 20-60 C

TG-DG 14,922 15,360 15,662 17,092 16,062

DG-MG 16,435 11,199 13,053 12,137 17,247
MG-GL 15,067 11,621 13,395 10,693 --
DG-TG 19,895 17,195 15,587 15,925 15,843
MG-DG 16,885 -- 13,336 15,816 13,571
GL-MG 12,196 -- 13,110 8,181 --
TG-GL . . . . 20,022

TG, triglyceride; DG, diglyceride; MG, monoglyceride; GL, glycerol.

JAOCS, Vol. 63, no. 10 ( O c t o b e r 1986)

 ]380
B. FREEDMAN, R.O. BUTTERFIELD AND E.H. PRYDE
--0.9964. By u s i n g this slope in the equation E, = - 4 . 5 8
(slope), an Ea of 14,922 cal/mol was calculated for this
reaction. Correlation coefficients for the o t h e r five
reactions of BuOH/SBO, 30:1 catalyzed by 1% H2SO4
were - 0 . 9 9 1 8 , - 0 . 9 9 6 5 , - 0 . 8 9 1 8 , - 0 . 9 7 9 5 and
-0.9905.
E n e r g i e s of activation determined for o t h e r reactions
in our study r a n g e d from 8,000-20,000 cal/mol (Table 3).
O t h e r investigators have reported E ov a l u e s within this
r a n g e for other transesterification reactions (6,8,9-12).
Eo for the s h u n t reaction TG-GL had a v a l u e of 20
kcal/mol. The spread of values seen in T a b l e 3 for Ea is
due partly t o experimental error. In addition, the
simplified m o d e l used in the e q u a t i o n log,oK =
(-Eo/2.30R)/T + C may not be adequate t o account for
all the variables involved. This m i g h t also explain why
E o for some reverse reactions are g r e a t e r than those of
the corresponding forward reactions. F u r t h e r research is
n e e d e d t o develop a m o d e l that m i g h t provide a narrower
r a n g e of E o values.
ACKNOWLEDGMENTS
Haifa Khoury assisted in the laboratory and in manuscript
preparation; Wilma Bailey gave statistical assistance.
REFERENCES
1. Freedman, B., E.H. Pryde and T.L. Mounts, J. Amer. Oil
Chem. Soc. 61:1638 (1984}.
2. Freedman, B., and E.H. Pryde, in Vegetable Oil Fuels,
.Study on the Oxidative Rate and ProoxJdant Activity
of Free Fatty Acids
Kazuo Miyashita a n d Tom Takagi*
Department of Chemistry, Faculty of Fisheries, Hokkaido University, Hakodate, Japan 041
Oleic, linoleic a n d linolenic acids were autoxidized more
r a p i d l y t h a n their corresponding m e t h y l esters. Addition
of s t e a r i c a c i d accelerated the rate of a u t o x i d a t i o n of
m e t h y l linoleate a n d t h e d e c o m p o s i t i o n of m e t h y l linole-
ate hydroperoxides. Therefore, the higher oxidative rate
of F F A ' s than their m e t h y l e s t e r s c o u l d be due to t h e
c a t a l y t i c e f f e c t of t h e carboxyl groups o n the f o r m a t i o n
of free radicals by t h e d e c o m p o s i t i o n of hydroperoxides.
Addition of stearic acid also accelerated the oxidative rate
of s o y b e a n oil. This result s u g g e s t s t h a t particular atten-
tion s h o u l d be p a i d to t h e F F A c o n t e n t t h a t a f f e c t s t h e
oxidative s t a b i l i t y of o i l s .
A few papers (1,2) have been published on the comparison
of the oxidative rate of free f a t t y acids (FFA) and their
esters. H o l m a n et al. (2,3) reported that F F A ' s were ox-
idized more rapidly than their esters, and he suggested
that this effect probably was due t o participation of the
*To whom correspondence should be addressed.
JAOCS, Vol, 63, no, ]O ( O c t o b e r ]986)
Proceedings o f the International Conference on P l a n t and
Vegetable Oils as Fuels, American Society of Agricultural
Engineers, St. Joseph, MI, 1982, p. 117.
3. Freedman, B., E.H. Pryde and W.F. Kwolek, J. Amer. Oil
Chem. Soc. 61:1215 {1984}.
4. Freedman, B., W.F. Kwolek and E.H. Pryde, Ibid. 62:663
{1985}.
5. Markley, K.S., in F a t t y A c i d s , T h e i r Chemistry, Properties,
Production and Uses, edited by K.S. Markley, 2nd ed.,
Interscience Publishers, New York, 1961, Part 2, p. 862.
6. Sridharan, R., and I.M. Mathai, J. Sci. Ind. Res. 33:183 {1974}.
7. Cotes, A.C., and L.A. Wanzel, Lat. A m . J. Chem. E n g . Appl.
Chem. 4:126 11974).
8. Davies, B., and G.V. Jeffreys, Trans. I n s t . Chem. E n g . 51:271
{1973).
9. Farkas, L., O. Schachter and B.H. Vromen, J. A m . Chem. Soc.
71:1991 11949}.
10. Baker, G.E., and R.R. White, Chem. E n g . Prog. S y m p . Ser. No.
4 48:75 (1952).
11. Ravindranath, K., and R.A. Mashelkar, J. Polym. Sci. Polym.
Chem. E d . 20:3447 {1982}.
12. Tomita, K., and H. Ida, Polymer 14:55 (1973}.
13. Pore, J., J.P. Houis and I. Rasori, Rev. Tech. Ind. Cuir 73:164
(1981).
14. Dufek, E.J., R.O. Butterfield and E.N. Frankel, J. A m e r . Oil
Chem. Soc. 49:302 {1972}.
15. Dufek, E.J., Ibid. 55:337 (1978}.
16. Butterfield, R.O., Ibid. 46:429 (1969}.
17. Reid, E.E., A m . Chem. J. 45:506 {1911}.
18. Formo, M.W., J. Amer. Oil Chem. Soc. 31:548 {1954}.
19. Koritala, S., R.O. Butterfield and H.J. Dutton, Ibid. 50:317
{1973}.
20. Vigeron, P.Y., S. Koritala, R.O. Butterfield and H.J. Dutton,
Ibid. 49:371 ~1972).
[Received December 17, 1985]
carboxyl groups in the decomposition of peroxides. H e
also said in this review (3) that this idea has been sup-
ported by the work of Privett et al., in w h i c h addition of
linoleic acid t o methyl linoleate peroxide accelerated its
decomposition. However, it was not c l e a r whether its
decomposition was promoted by the catalytic action of
the carboxyl group in linoleic acid or oxidized products
of this acid, and the effect of F F A ' s on the autoxidation
of esters has not been investigated.
In this paper, w e report the detailed data for the dif-
ference in oxidative rates between F F A ' s and their methyl
esters w i t h periodic measurement of the unoxidized
substrate content by GLC and POV, and elucidate the
catalytic action of the carboxyl group in F F A ' s on the
autoxidation of oils by the use of stearic acid as a catalyst
which is not autoxidized u n d e r the conditions of the pres-
ent experiment.
MATERIALS AND METHODS
P r e p a r a t i o n o f m a t e r i a l s . Oleic, linoleic and linolenic acids