575

The Effect of Added Solvents on Soy Oil Lutein Adsorption

by SUicic Acid
Jia Mingyu and Andrew Proctor*
Department of Food Science and Technology, The Ohio State University, Columbus, Ohio 43210-1096
I t h a s b e e n r e p o r t e d t h a t a d d i t i o n o f i s o p r o p a n o l t o a s o y
oi l mi s c e l l a i nhi bi t s t h e b i n d i n g o f s o y l u t e i n t o a d d e d
s i l i c i c a c i d b y c o mp e t i t i v e a d s o r p t i o n . I t w a s s u g g e s t e d
t h a t t h e c o mp e t i t i o n w a s b a s e d o n t h e p o l a r i t y o f t h e
mi s c e l l a c o ns t i t ue nt s . Thi s i n v e s t i g a t i o n s t udi e d t h e e f f e c t s
o f a h o mo l o g o u s s e r i e s o f l o we r a l c o h o l s t o c o mp e t i t i v e l y
i nhi bi t l u t e i n bi ndi ng t o s i l i c i c ac i d f r o m a s o y oi l h e x a n e
mi s c e l l a . Lu t e i n bi ndi ng i nhi bi t i o n b y mo l e c u l e s o f c a r bo n
c hai ns wi t h t h e s a me l e ng t hs , but wi t h di f f e r e nt f unc t i o na l
g r o ups , w a s a l s o e x a mi n e d . Mi n o r d i f f e r e n c e s we r e f o u n d
b e t we e n me mb e r s o f a h o mo l o g o u s s e r i e s o f a l c o h o l s . A
s i mi l a r r e s u l t w a s f o u n d wi t h s h o r t - c h a l n f a t t y ac i ds . Th e
abi l i t y o f v a r i o u s f u n c t i o n a l g r o ups t o di s pl a c e l ut e i n f r o m
s i l i ci c ac i d w a s de pe nde nt o n t h e mo l e c ul e s ' a bi l i t y t o f o r m
h y d r o g e n bo nds , r a t he r t h a n o n po l a r i t y .
KEY WORDS: Adsorption, lutein, miscellas, silicic acid, solvents,
soy oil.
Oil is extracted from soy flakes with hexane, which is then
evaporated to produce the crude oil. Crude soy oil contains
a number of substances, such as pigments, phospholipids
and free fatty acids, which must be removed to produce a
bland, light-colored oil that is acceptable to consumers. The
commercial removal of pigments is achieved by an adsorp-
tion process at 100C on bleaching clays under reduced
pressures (1). Soy oil pigment is almost exclusively the
carotenoid lutein (2}. Hassler and Hagberg (3) showed that
the adsorption of soy oil pigment on bleaching clay occurs
according to a Freundlich isothern~
Adsorption of phospholipids (4) and lutein (5) onto silicic
acid from soy oil/hexane misceUas also conforms to a
Freundlich isotherm. An advantage of this bleaching tech-
nique, relative to conventional methods, is that it is con-
ducted at ambient temperatures, and the binding of these
oil components is modified by the addition of a polar sol-
vent to the miscell~ One percent isopropanol in the misceUa
promotes adsorption of phospholipid, which presumably oc-
curred by removal of triglyceride from adsorption site~ This
then facilitates phospholipid binding (4). Free fatty acid ad-
sorption to amorphous cristobalite silica was also promoted
by isopropanol in a similar system (5). In contrast, iso-
propanol inhibited adsorption of lutein to silicic acid (5). This
inhibition was explained in terms of competition between
misceUa constituents for silanol sites. Polarity was sug-
gested to be a basis for competition because of the is~
propanol effect. However, reducing triglyceride concentra-
tion promoted lutein adsorption, suggesting that concen-
tration and/or molelcular weight of competing species may
also be factors that determine adsorption.
The objective of this investigation is to study the effect
of alcohol molecular weight on lutein binding from a soy
oil miscella and the adsorption isotherm. The effect on the
*To whom correspondence should be addressed at Department of Food
Science, University of Arkansas, 272 Young Avenue, FayetteviUe,
AR 72703.
lutein isotherm of C3 and C4 compounds with different
functional groups was also examined.
MATERIALS AND METHODS
Oil and sol vent s. Commercially extracted alkali-refined
soy oil was stored at 4C and used throughout the in-
vestigations. Soy oil miscellas were prepared by diluting
soy oil with hexane
Water and the following alcohols were added to modify
miscella polarity (0.1 M concentration} prior to lutein ad-
sorption: methanol, ethanol, n-propanol, isopropanol, n-
butanol, isobutanol, 2-octanol, 1-nonanol and 1-decanol.
The following aldehydes, ketones, acids and esters were
also used: propanal, acetone, 2-butanone, acetic acid, pro-
pionic acid, n-butyric acid, isobutyric acid, octanoic acid,
methyl acetate and ethyl acetate.
Ads or bent s . The adsorbent used was silicic acid (Bio-
Sil A., 100-200 mesh, Bio-Rad Laboratories, Richmond,
CA). Silicic acid was heated in a drying oven to remove
moisture and was stored in a desiccator until used. The
term "silica" is used as a synonym for silicic acid.
Lut e i n meas ur ement . Pigment concentration in the
miscellas was measured as lutein by reading optical ab-
sorbance at 445 nm, according to the method of Proctor
and Snyder (5).
Lut e i n i sot herms. Lutein isotherms were determined by
preparing 100-mL volumes of 2.5, 5, 10, 20, 30 and 40%
(vol/vol) concentrations of soy oil miscellas in hexane The
lutein content of each miscella was measured before 0.5
g of silica was added. The misceUas were agitated with
a magnetic stirrer in a closed vessel at 22C for 15 min.
The concentration of residual lutein remaining unadsorbed
was then found, and the amount of lutein adsorbed was
calculated by difference Isotherms were plotted as the
amount of lutein adsorbed, per gram of silica, vs. the
residual concentration of lutein. Duplicate determinations
were made. This was the control experiment.
Mi scel l a polarity. Isotherms were prepared as described
above but with 0.1 M concentration of additional solvent
present in the miscella.
RESULTS AND DISCUSSION
Figure l a shows the effect of adding water and low molecu-
lar weight alcohols (C1-C3), to the misceUa on pigment
binding. Lutein adsorption followed a Freundlich
isotherm, and added solvent reduced lutein adsorption
relative to the control in each case There is little difference
in the isotherms obtained with ethanol, n-propanol and
isopropanol, which were more effective t han water in in-
hibiting lutein binding. Methanol was slightly less effec-
tive in reducing lutein adsorption. The differences in the
results obtained with different solvent systems were best
seen at high residual lutein levels.
Water was the least effective solvent, but it did signifi-
cantly reduce lutein adsorption relative to the control.
Water can hydrogen bond to the silica or to other water
molecules. In this lipid system it is probably more thermo-
Copyright 1993 by the American Oil Chemists' Society JAOCS, Vol. 70, no. 6 (June 1993)
576
J. MINGYU AND A. P ROCTOR
O
9
O
E
v
JD
O
"O
c-
D
2.50
2. 00
1. 50
1. 00
0, 50
0. 00
a
6 1 2 16
Resi dual l ut ei n ( pM)
2. 50
"~ 2.00
~ 1 . 5 0
"O
. ~ ' 1. 00
C 0 . 5 0
0 . 0 0
0
b
I I
6 12 18
Resi dual l ut ei n ( pM)
F I G. 1. Lu t e i n i s o t h e r ms we r e o b t a i n e d b y i n c u b a t i n g 0. 5 g o f s i l i c i c
a c i d w i t h 100 mL o f 2. 5, 5, 10, 20, 30 a n d 40% ( vol / vol ) a l ka l i - r e f i ne d
s o y o i l / h e x a n e mi s c e l l a f o r 15 mi n a t 2 2 C i n t h e p r e s e n c e o f 0. 1 M
c o n c e n t r a t i o n o f (a) wa t e r {~,~, me t h a n o l (O), e t h a n o l ( ) , n- pr opanol
(A) or i s o p r o p a n o l (A). A c o nt r o l { +) w a s pr e par e d w i t h o u t a dde d s ol -
v e n t ; (b) n - b u t a n o l (D), i s o b u t a n o l ( . ) , 2 - o c t a n o l (Y), 1- nonanol { ~ )
a n d 1- decanol ([7). A c ont r ol ( + ) w a s pr e par e d wi t h o u t a dde d s o l v e n t .
dynamically stable for water molecules to associate to-
gether. The effect of added water in this system may be
due to the strength of water/water hydrogen bonding and
water's lipophobic nature. To enable water molecules to
bind to silica, water hydrogen bonding has to be disrupted,
and individual molecules migrate to the adsorption sur-
face. This is probably not as energetically favorable as
disruption of alcohol hydrogen bonding. Furthermore, an
alkane structure would increase solubility in a lipid system
and permit binding to an adsorbent. Therefore, competi-
tive adsorption would be expected to be improved by ad-
2. 50
.9
~ 2.00 +
O)
(D 1. 50
o
E
" 0
(D 1. 00
O
" o
c- o. 5o
' $
_1
0. 00 i i I i ~ I I J i I I I I I i I I
6 1 2
Resi dual l ut ei n ( pM)
F I G. 2. Lut e i n i s o t h e r ms we r e o b t a i n e d b y i n c u b a t i n g 0. 5 g o f s i l i c i c
a c i d w i t h 100 mL o f 2. 5, 5, 10, 20, 30 a n d 40% ( vol / vol ) a l ka l i - r e f i ne d
s o y o i l / he x a ne mi s c e l l a f o r 15 rai n a t 2 2 C i n t h e pr e s e nc e o f 0. 1 M
c o n c e n t r a t i o n o f a c e t i c a c i d (A); p r o p i o n i c a c i d (A), n - b u t y r i c a c i d
([7), i s o b u t y r i c a c i d ( B ) a n d o c t a n o i c a c i d (O). A c o n t r o l ( + ) w a s
pr e pa r e d w i t h o u t a d d e d s o l v e n t .
18
dition of an alkyl group. Methanol is slightly less effec-
tive than ethanol at reducing lutein binding, but the ef-
fect of ethanol is similar to that of propanol isomers.
Therefore, the methyl groups bound to the hydroxyl car-
bon of isopropanol do not sterically hinder adsorption
relative to the primary alcohol.
Butanol isomers, octanol and nonanol also produced
similar isotherms to those obtained wi th propanol (Fig.
lb). These data indicate that with lower alcohols there are
several small differences in adsorption on the basis of
isomerism or molecular weight. Molecular shape and size
are reported to be important factors preventing hydrogen
bonding to silica due to steric hinderance (6). In studies
of long-chain species, Hau and Newar (7) reported that the
number of moles adsorbed to silica decreases as chain-
length increases. The results of studyi ng the competitive
adsorption of lutein suggest that wi th alcohols (C0-C10)
(Fig. l a and lb) there is no large change in the isotherm,
as alcohol length is increased but small differences are
seen.
These isotherm studies give indirect evidence that lower
alcohols bind largely independently of chainlength. This
study is complicated by the presence of triglyceride, which
is the major species bound overall (5).
Figure 2 shows the effect on lutein adsorption of adding
members of a homologous series of fatty acids to the soy
oil miscellas. Al though lutein adsorption is reduced rela-
tive to the control, there is little difference between iso-
therms. Therefore, the ability of free fatty acids to com-
pete wi th lutein for binding sites is independent of
chainlength. The data conform to Wu and Mead's (8) find-
ings that fatty acid adsorption to silica is independent
of chainlength. This shows the importance of extracting
free fatty acids from soy oil before pigment adsorption.
JAOCS, Vol. 70, no. 6 (June 1993)
EFFECT OF SOLVENTS ON LUTEIN ADSORPTION
577
2. 50
0
'~ 2 . 0 0
(~ 1. 50
0
E
'10
(D 1. 00
, . 0
L.,
0
m
t = 0, 50
3
. . 1
0. 00
6 12 18
Residual lutein (pM)
FI G. 3. Lut e i n i s o t h e r ms were o bt a i ne d b y i nc uba t i ng 0. 5 g o f s i l i ci c
a c i d wi t h 100 mL o f 2. 5, 5, 10, 20, 30 a nd 40% (vol / vol ) al kal i - ref i ned
s o y oi l / hexane mi s c e l l a f or 15 mi n a t 22 C i n t he pr e s e nc e o f 0. 1 M
c o nc e nt r a t i o n o f n- propanol (A), i s o pr o pa no l (A), pr opi oni c a c i d ( I ) ,
a c e t o ne (C:]), pr opanal ( ) a nd me t h y l a c e t a t e (O). A cont rol ( +) wa s
prepared wi t h o u t a dde d s o l v e nt .
2, 50
0
" ~ 2. 00
(D 1. 50
"6
E
" O
~D 1 . 0 0
. O
0)
" 0
e- 0, 50
_J
0.00
0 6 12 18
Residual lutein (pM)
FI G. 4. Lut e i n i s o t he r ms were obt ai ne d by i nc uba t i ng 0. 5 g o f s i l i ci c
ac i d w i t h 100 mL o f 2. 5, 5, 10, 20, 30 a nd 40% (vol / vol ) al kal i - ref i ned
s o y oi l / he xane mi s c e l l a f or 15 mi n at 22 C i n t h e pr e s e nc e o f 0.1 M
c onc e nt r at i on o f n- but anol , ([:]), i s o b u t a n o l ( I ) , n- but yr i c ac i d (A},
i s o but y r i c ac i d (A}, 2- but anone ( } a nd e t h y l a c e t a t e (O). A c ont r ol
(Jr) wa s prepared wi t h o u t a dde d s o l v e nt .
The effect of an added Ca alcohol, aldehyde, ketone,
acid and ester on lutein binding is illustrated in Figure
3. Freundlich isotherms were observed in the presence of
each solvent, but there were differences in the solvent' s
ability to inhibit lutein binding. The alcohols were the
most effective in inhibiting lutein binding, which is prob-
ably because they are best able to form hydrogen bonds
with silanol groups. Hau and Newar (7) reported t hat the
greater the tendency for hydrogen bonding, the stronger
the adsorption. This st udy supports t hat premise, with
the alcohols being most effective. A silanol hydroxyl is
capable of forming two hydrogen bonds with a single
alcohol hydroxyl group, or can hydrogen bond to two dif-
ferent alcohols (9).
Propionic acid was the most effective of the nonalcohol
solvents, despite the fact t hat it has a lower polarity than
the aldehydes and ketones used. This is most likely due
to its ability to hydrogen bond. Each molecule is capable
of forming two hydrogen bonds to a silanol group (9).
Organic acids are not bound to silica to the same degree
as alcohols, but the acid binding strength is greater (7).
At low miscella residual concentrations (<9 ~M), the lu-
tein isotherm, obtained with the remaining solvents, did
not differ from the control or from each other. However,
differences were evident in the 40% miscella. Acetone
reduced lutein binding slightly more than propanal, prob-
ably because of the greater polarity in a similar system
(10).
The isotherm obtained with methyl acetate was similar
to t hat of the control. This indicates t hat this ester was
not any more effective t han triglyceride esters in com-
peting with lutein for adsorption sites. This could be
because the amount of added ester is negligible compared
to the amount of triglyceride esters in the system. In any
case, the position of the carbonyl group on an ester may
make hydrogen bonding to a surface sterically difficult.
Nevertheless, in this system triglyceride is the major oil
component, and therefore, concentration is the overriding
consideration (5).
The nonalcohol solvents were practically ineffective at
reducing lutein binding at most miscella concentrations.
Therefore, ability to hydrogen bond is more important
t han polarity in determining a molecule's ability to com-
pete with lutein for adsorption sites.
The effectiveness of the solvents as competitors for lu-
tein adsorption was alcohol > acid > ketone > aldehyde
> ester. This is supported by Hau and Newar (7}, who
found t hat when comparing different classes of com-
pounds of the same chalnlength, the amount adsorbed
was alcohol > acid > ester.
Figure 4 shows the effect of selected C4 compounds on
lutein adsorption by silic& The pat t ern of adsorption was
similar to t hat obtained with C3 compounds, i . e . , alcohol
> acid > ketone > ester. In contrast to the isotherms ob-
tained with C8 compounds, at all residual lutein levels,
ketones and acids produced isotherms t hat were dissimilar
from the control but similar to each other.
In summary, alcohols can compete more effectively with
lutein for adsorption sites on silica t han other solvents
because of their ability to form hydrogen bonds, rather
t han their polarity. Water is not as effective as alcohols
because of the energy needed to overcome hydrogen bond-
ing between water molecules in a hydrophobic environ-
ment. There is little difference between isotherms obtained
within a homologous series of small molecular weight
alcohols or fat t y acid. Functional groups, other than alc~
hols, inhibit lutein binding to a lesser extent, even if
polarity is greater t han the corresponding alcohol.
JAOCS, Vol. 70, no. 6 (June 1993)
578
J. MI NGYU AND A. PROCTOR
ACKNOWLEDGMENTS
Sal ari es and research suppor t provi ded by st at e and federal f unds
appr opr i at ed t o t he Ohio Agr i cul t ur al Research and Devel opment
Center, The Ohio St at e Uni versi t y, were gr eat l y appreci at ed. Thi s
is J our nal Ar t i cl e number 64-93. Thanks is expressed t o Dr. J ohn
Lowbri dge for his useful comment s on t he data.
REFERENCES
1. Brekke, O.L., in Handbook of Soybean Processing and Utiliza-
tion, edi t ed by D.R. Eri ekson, E. H. Pryde, O.L. Brekke, T.L.
Mount s and R.A. Falb, Ameri can Soybean Associatior~ St. Louis,
and Amer i can Oil Chemi st s' Society, Champai gn, 1980, pp.
71-88.
2. Vogel, P.V., Fette Seifen Anstrichm. 79.97 (1977).
3. Hassler, J.W., and R.A. Hagber g, Oil and Soap 16:188 (1946).
4. Brown, H.G., and H. E. Snyder, J. Am. Oil Chem. Soa 62:753
(1985).
5. Proctor, A. , and H. E. Snyder, Ibid. 64:1163 (1987).
6. Iler, R.K., in The Chemistry of Silica, Chapt er 6, J ohn Wiley &
Sons Co., New York, 1979.
7. Hau, L.-B., and W~. Nawar, J. Am. Oil Chem. Soa 62:1596 (1985).
8. Wu, G.S., and J.F. Mead, Lipids 12:965 (1977).
9. Marshall, K., and C.H. Rochester, J. Chem. Soc. Faraday Trans.
I 71:1754 (1975).
10. McClellan, A.L., Tables of Experimental Dipole Moments, W.H.
Fr eeman and Co., San Fr anci sco and London, 1963.
[Received November 5, 1992; accept ed Apr i l 8, 1993]
JAOCS, Vol. 70, no. 6 ( June 1993)