Note: Descriptions are shown in the official language in which they were submitted.
POLYGLYCEROL ESTER SYNTHESIS
Paul Seiden
Ricky A. Woo
Technical Field
5This invention relates to the synthesis of polyol esters, in
particular, polyglycerol fatty acid esters.
Backqround of the Invention
There are a number of processes available for synthesizing
polyol esters. These include esterification with fatty acids with
and without catalysis, as well 3S transesterification using
triglycerides. Stenzel et al., Die Nah~ , 21, 429-441 ~1977)
reviews a number of literature references on the synthesis of
polyglycerol fatty acid esters and their use. Phosphoric acid was
~ound to be a favorable catalyst for esterification, as was
potassium or sodium hydroxide.
One method for synthesizing polyol esters uses an ion exchange
resin as a catalyst, see Canadian Patent 834,214, issued to Kuhrt
et al. ~1970). This process is said to yield pure monoesters of
various polyglycerols. However, not all ion exchange resins are
~0 approved for use in foods or for pr paring food additives.
Moreover, ion exchange resins are more expensive than alkaline
catalysts. ~
` One of the problems wi~l polyol esters, in particular
polyglycerol esters, is the inconsistent quality of the product.
The polyol starting material is a mixture of dimers, trimers,
tetramers, etc. of the starting alcohol. When this material is
esterified with the fatty acid, the fatty acid will react with any
of the free hydroxyl groups within the molecule. Thus, the
esterified polyol is a mixture of monoesters, diesters, triesters,
etc. Upon cooling in the presence of a catalyst or moisture,
these esters can rearrange r,ot only by changing the position of
.` ' `~
the ester group within the molecule, but actual equilibration
between diesters forming monoesters and triesters, etc. This
equilibration chan3es the composition to a lower monoester content
and increases the concentration of higher esters.
The degree of shift caused by interesterification is
variableO The process herein eliminates the shift în composition
and the batch-to-batch variations.
In general, the more hydrophilic type of polyglycerol esters
are more functional. These polyglycerol esters have a greater
number of free hydroxyl groups relative to the number of
esterified hydroxyl groups.
The esterification reaction is carried out under an inert
atmosphere to obtain good color7 filavor and odor characteristics.
However, unless the mixture is properly neu-tralized to remove the
esterification catalyst, rearrangement will occur. That is, the
polyglycerol ester will revert to polyol and higher esters
(rearrangement).
.
After the esterification is completed the catalyst is
neutralized with a mild acid at~~ Z5C. During the acid addition
or the cooling step which follows, rearrangement occurs. The
composition shiFts to a more lipophilic and therefore less
functional product. The rearrangement releases free polyol which
separates from the reaction mixture. The free polyol, if recycled
into the product, contaminates the product and causes darkening of
2s the color. In particular, the recycle of polyglycerol results in
additional undesirable polymerization of the glycerol. The polyol
to fatty acid ratio also affects the ester composition but it is
difficult to compensate for the shift caused by
interesterification (i.e. rearrangement).
The process of this invention eliminates the uncontrolled
interesterification and eliminates the need for polyol recycle.
The process permits one to produce the most hydrophilic, most
functional esters at the lowest polyol to fatty acid ratio. Since
polyol is always more expensive than fatty acid, this process is
also more economical.
5~
Moreover, it has been found that the water from the reaction
must be substantially removed before the neutrallzation of the
catalyst to insure a product without reversion or random
rearrangement.
S Therefore, it is an object of the present invention to provide
a process for producing polyol esters of consistently good color,
flavor and odor characteristics. It is a further object of this
invention to produce polyol esters which do not revert or
rearrange upon standing.
It is an additional object of this invention to provide a
process which utilizes inexpenslve materials which are approved
~or preparation o~ food grade materîals.
These and other ob~ects will become apparent from reading the
speci~lcation.
Unless specified herein, all percentages are by weight.
Brief SummarY of the Invention
A process for esterifying a polyol comprising the stPps of:
reactihg fatty acid and polyol under an inert atmosphere
of temperatures from about 220C to about 260C and an
; 20 absolute pressure of ~rom about 500 mm to about 900 mm o~
mercury, in the presence of an alkaline catalyst;
(2) reducing the pressure to less than 127~of mercury when
the free fatty acid level is less than 1~ to remove
~ubstantially all the water;
(3) neutralizing th~ alkaline catalyst with a weak acid under
reduced pressure; and
(4) rapidly cooling the reaction mixture to less than 177C.
.
Detail~d Description of Invention
The process herein is useful for the esterification nf many
different polyols. Polyols are organic compounds oontaining more
than one hydroxy group on the molecule. These include
dihydroxyalkanes, glycerine, carbohydrates; sugar alcohols,
polyglycerols, etc.
The most preferred polyol is polyglycerol which is prepared by
the polymerization o~ glycerine in the presence of either acid or
base. The polyglycerols can contain from 2 to 20 gl-ycerol
moieties. Preferably, the polyglycerols will be those having from
~ to 15 glycerol moieties.
The polyglycerol compounds can be made by any synthetic
method9 the method of their preparation is not critical to the
esterification process. See for example, U.S. 3,968,169 issued to
Seiden and Martin (1976). However, the purer the starting
10 polyglycerol? the purer the polyglycerol ester will be.
Additional polyols that may be used in the esterification
reaction described herein are glycerine, sorbitol, xylitol,
erythritol and pentaerythritol. These materials may be reacted
with fatty acids to produce mono- or diglycerides9 sorbitol
esters, xylitol esters, etc. The reaction is especially useful
for polyols which are not soluble in the fatty acid and which are
`~ not good solvents themselves.
; The polyols can be esterified with any fatty acid or
interesterified with a fatty acid ester or fatty acid
20 triglyceride. The fatty acids can be saturated, unsaturated, or
polyunsaturated. In particularg those having from 8 to 24 carbon
atoms are preferred for use herein. For the preparation of
emulsifiers, those having from about 10 carbons to about 22
carbons are preferred. These include decanoic acid, dodeca~oic
2s acid, stearic acid, palmitic acid, oleic acîd, behenic acid, and
others. Triglycerides of these acids can also be used. Lower
alkyl esters, in particular, methyl and ethyl esters of the fatty
acids, can be used as fatty acid sources.
The polyol and fatty acid are mixed together with an alkaline
30 catalyst. The alkaline catalyst can be the hydroxide of any of
the alkaline earth metals, for example sodium, potassium or
lithium hydroxide. The order of addition is not critical to the
process.
The mole ratio of the polyol to fatty acid can range ~rom
35 about 0.1:1 to about 3:1.
3s~
An amount of the alkaline catalyst effective to catalyze the
reaction is used. Preferably, an amount in the range of from
about û.01 to about 0.2 mole per mole of fatty acid used3 most
preferably from about 0.04 to about 0.10 mole per mole of fatty
acid.
The fatty acid, the polyol and the catalyst are added to a
reaction vessel which is held under an inert atmosphere. The
inert atmosphere can be maintained by sparging the mixture, or by
simply passin3 a non-reactive gas through the vessel. Inert
(non-reactive) gases include nitrogen, helium, argon, etc. Under
some conditions, water or carbon dioxide can act as the inert gas.
In order to produce a final product which has a good color and
less off-odor, effective agitation is necessary~ The agitation
can be accomplished by any conventional means. This includes the
use of a mechanical mixer as well as by inert gas spar3ing.
The esterification mixture is then heated to a temperature of
from about 220C to about 260C, preferably from 225C to 235C.
The vessel is held under atmospheric pressure or at a slight
vacuum (508 mm to 76û mm of mercury absolute) or at a slight
positive pressure (760 to 900 mm of mercury absolute). During the
esterification reaction, the water is distilled off. The water is
not returned to the reaction vessel. The reaction is monitored by
measuring the free fatty acid level in the reaction mixture.
The free fatty acid level can be determined by titrating the
free acid in aliquot portions.
When the free fatty acid level has been reduced to less than
1%, and preferably less than 0.5%, the pressure in the mixing
vessel is reduced to less than about 127 mm of mercury, preferably
less than 12 mm of mercury. The temperature is maintained at
about 204C to about 238C. This reduction in pressure further
strips the water from the mixture. The removal of the water also
deodorizes the polyol ester.
The reaction mixture is held at full vacuum until
substantially ail of the water has been removed. The amount of
time necessary to do this will, of course, depend on the size of
. . .
the reaction vessel and the vacuum system ~e.g. pump or ejector
systems). It can vary from 10 minutes to 4 hours. For example, a
68 kg. reaction mixture would be heid at less than 127 mm of
mercury for about 15 minutes. It is extre,~ely important to
eliminate the water or minimize the water content to control the
interesterification and thus reversion to the more lipophilic
estersO
The catalyst neutralization and the subsequent cooling steps
also represent an improvement over the prior art.
After removal of the water, the catalyst is neutralized with a
mild acid. For example3 phosphoric acid, sodium di-hydrogen
phosphate, acetic acid, citric acid, and other carboxylic acids
can be used~ Water can be present in the acid as water is readily
removed because of the high temperatures and the vacuum.
The acid is added slowly to the reaction vessel. For examplet
a rate of from about lOû g/minute is suitable for a 45 kg.
reaction batch. The pressure of less than 127 mm of H3 and the
elevated temperatures (~04C to 238C) are maintained during the
neutralization to remove any water present. For maximum
20 efficiency, the acid is added through the bottom oF the reaction
vessel. The rate of acid addition is also a~fected by the water
content in the acid and the effectiveness of the vacuum system.
Following the acid addition, the reaction mixture is cooled
rapidly to less than 177C. ~referably, the mixture is cooled to
25 less than 160C. The rate of cooling should be at least 3C to
15C per minute, preferably 5C to 1~C/minute. Once the
temperature has reached about 145C, the rate of cooling is no
longer critical to the stability of the-product.
Rapid cooling can be accomplished using a heat exchanger,
cooling coils or a water sparge. After the reaction mixture is
cooled, any excess water is removed to minimize product
deterioration during storage.
The minimizing of the water content before and during
neutralization is critical to the formation of polyol esters which
have good color odor and flavor characteristics and which are
consistent in compos;tion on a batch to batch basis. The rapid
cooling of the reaction mixture minimizes the interesterification
nf the polyol esters
Reversion, i.e. random rearrangement shifts in the
composition, during and after processing can be detected by
measuring the refractive index of the products or by analyzing the
; differential scanning calorimetry curve.
If a compositional shift has occurred in the product due to
improper process control, the final refractive index reading will
be sîgnificantly lower than the refractive index prior to
neutralization and the peak height ratio of the differential
scanning calorimeter curve will change. (Peak height ratio is
defined below.)
Table 1 illustrates the changes that can occur in the peak
height ratio measurements and in the refractive indexO The seven
samples, A through E, are all palmitic and stearic acid esters of
hexapolyglycerol. They were prepared using about the same ratios
of fatty acid, polyglycerol, sodium hydroxide, and phosphoric acid
as in Example I. The scale of the reaction varied.
Samples C, D, E and G were prepared according to the process
of this invention. Samples A and B were prepared using a process
which was similar to that of Example I except that the pressure
; was not reduced prior to neutralization. The rearrangement that
occurred in Samples A and B was a result of the high moisture
conditions during this step of the process.
Sample F was prepared in the same manner as Sample G except
that the reaction was cooled at the rate of about 1C per minute.
Because of this slow cooling, rearrangement occurred.
TABLE 1
FREE FATTY REFRACTIVE INDEX* PEAK HEIGHT RATX0**
ACID ~e~ S~ple 2 Sample 1 Sample 2
A less than 0.3 58.9 53.3 0.55 0.41
a less than 0.5 55.4 49.9 0.62 0.36
C less than 0.2 59.1 58.9 0.79 0.78
D less than 0.1 59.0 58.7 0.68 0.71
E less than 0.1 59.9 59.8 1.28 1.25
F less than 0.2 60.2 52.0 - -
G less than 0.5 60.9 60.0 - -
*the butyro scale was used to measure the refractive index.
Sample 1 is the sample before neutralization, Sample 2 is the
final sample.
**Ratio of peak heights from the differential scaling
calorimeter. Sample 1 is taken before neutralization7 Sample
2 is taken at`ter final cooling.
A. Refractive Index Measurement
Any refractometer can be used.
A Zeiss refractometer was used to measure the re-fractive
index of the process samples. The refractometer is preheated and
maintained at 6û'' + û.1C using a constant temperature water
bath. Two samples are obtained from the reactor:
1. A sample prior to neutralization.
2. A sample 2fter the reaction has cooled rapidly to 60C to
82C.
Sarnple 1 which is taken from the reactor before
neutralization is rapidly cooled to 60C and then the refractive
index measured. Sample 2 is measured at~60C. The standard
refractive index method for the instrument is used.
B. Differential Scannin~ Calorimetry
Any differential scanning calorimeter can be used.
A duPont ~odel 990 thermal analyzer connected to a Model 910
dif~erential scanning calorimeter was used to measure the peak
height ratios of the process samples.
3s 10.0 + 0.1 mg of solid ester is placed in the sample cup.
Most samples are more easily handled in their solid formO The
lid is placed on the sample cup making sure that no material is
on the sealing lip. The sample cup and lid are hermetically
crimped using the sample crimper. ~ reference cell is prepared
by crimping two lids on one empty sample cup.
The instrument controls are set as follows:
Control Setting
X-Axis Zero Shift 0
X-Axis Scale 5C/in.
Y-Axis Zero Shift as required
lo Y-Axis Scale (Sensitivity) 2
Base line slope 0
Program mode isothermal
Temperature rate 5C/min~
Starting temp. lOûC
Allow tho instrument to heat to 110C to remove all the
moisture in the cell. Reset the starting temperature to 20C and
allow the instrument to cool to 20C ~changing to isothermal
control cools).^ Remove the cell cover and the silver lid from the
cell. Place the sample pan on the rear thermocouple and the
20 reference pan on the front thermocouple. The same reference is
left in place for all the tests. Allow the cell to equiliorate
for approximately 2 to 3 minutes. Using the Y-Axis Zero control,
position the pen to the starting point on the chart paper and move
the pen to the down position to begin recording. Switch the t
25 control program mode to Heat. It should take about 10 minutes to
reach 70C. When the temperature reaches 70C, move the pen to
the up position to stop recording and reset the starting
temperature to 100C and change to the isothermal mode. When the
temperature reaches 100C, remove the sample from the cell.
30 Repeat these same steps for the next sample.
Determination of the Peak Height Ratio
The determination of the ratio of the two peaks produced by
the instrument is done by first determining a base line. The base
line is determined by extrapolating the line back from the
35 horizontal section of the curve tracing around the 60C to 70C
--10-
mark. This extrapolated line, going back to the 30C mark should
; be as straight and hor~zontal as possible. The height (number o~
blocks from base line to the top o~ the peak) o~ the first peak is
divlded by the height of the second peak to arrive at the ratio.
Detennination of the peak teme~a ures
The determination of the differential scanning calorimeter
peak temperatures i5 done by reviewing the heating mode
differential scanning calorimeter scale cell of the polyol ester.
The maximum point oF each peak apex is extrapolated vertically to
10 the temperature ax;s (X-Axis). These temperature points are the
peak temperature for each curve, respectively.
EXAMPLE I
Synthesis of a mixed ester of hexapolyglycerol.
; A reaction vessel, sufficient to hold 50 kg. of reactant, is
15 used. The reaction vessel is fitted with a nitrogen spar~e and a
propeller mixer, is adapted to run under vacuum, and is also
equipped with a condenser to collect the water removed during the
reaction. To t~is vessel is added 27.67 kg. of polyglycerol
; having an average chain length of 6. The polyglycerol is mixed
20 with 0.43 kg. of 50~ sodium hydroxide. The reaction mixture is
then heated to about 115C under f`ull vacuum for about 15 minutes
to remove the water from the sodium hydroxide polyglycerol
mixture. Palmitic acid at a weight of 11.4 kg. (45.2 moles) and
6.12 kg. of stearic acid (21.S moles) are added to the reaction
25 mixture at atmospheric pressure. A partial vacuum is then pulled
` on the reaction vessel (about 5û8 rmm of mercury) and tne vessel
heated to about 220C over the period of about an hour. The
percent free fatty acid after an hour is about 6.3~. The vessel
is then maintained at this partial pressure and at a temperature
30 of about 230C for an additional 20 minutes when the free fatty
acid drops to less than 0.3~.
The pressure is then lowered to about 12 mm of mercury and
held there about ~or 15 minutes. This removes substantially all
of the water from the reaction mixture. The temperature is then
3s maintained at about 224C, the pressure maintained at 12 mm, and
aeid is added from the bottom Over the period of about 5
minutes, 0.68 kg. ~f 75~ phosphoric acid is added. The mixture is
then cooled by using a water sparge and a 4.5C eooling coil to a
temperature of about 157C over a period of about 7 minutes.
During this cooling, the water distills out of the reaction
mixture. The water sparge is then stopped, and the mixture cooled
by the cooling coil to about 93C. The product is an opaque
liquid which solidifies on further cooling.
The total yield of the ester of the hexapolyglycerol is
40.6 kg. The weight of the distillate, including water collected
during the reaction is 1.8 kg.9 approximately 0.9 kg. of this is
water. The remainder is low molecular weight polyol. The
saponification value is 98.74, and the hydroxyl value is 441.
; When this reaction is repeated using 27.66 kg. of the
polyglycerol having an average chain length of 6, 11.6 kg.
palmitic acid, 6 kg. of stearic acid, essentially the sa~e results
are obtained. This product has a saponification value of 99.98
and a hydroxyl yalue of 430.
EXA~PLE ll
A reaction vessel sufficient to hold 680 kg of reactant is
used. The reaction vessel is fitted with nitrogen sparge a
propeller mixer is adapted to run under vacuum, and is
- equipped with a condenser to collect the water removed during
; the reaction. To this ves;sel is added 394. 6 kg of polyglycerol
having an average chain length of 6. l he polyglycerol is mixed
with 6.3 kg of sodium hydroxide catalyst ~50% aqueous solution).
The reaction mixture is then heated to 110-116C under full
vacuum for approximately lO to 15 minutes.
The vacuum is released and 285.8 kg of stearic and palmitic
acid is added. The ratio of palmitic to stearic acid is 65:35.
~litrogen sparging is then started. The reaction is heated to
about 229C.
Samples are removed from the reac-tion vessel periodically
and titrated for free fatty acid. When the free fatty acid level is
below l~, the pressure is lowered to a full vacuum ~about 12 mm.
of mercury) and held there for one hour. The vacuum is pulled
slowly (to avoid foaming ancl excessive temperature drop).
Substantially all the water is removed from the reaction mixture.
Twenty pounds of phosphoric acid Is then slowly added to the
5 reaction vessel at the rate of about one pound/minute.
Two to three minutes after the addition of the phosphoric
ackl, the vacuum is released and the system pressure reaches
atmospheric. Nitrogen sparging is continuous. Water is injected
into the vessel at the rate of 6.35 kg/minute over a 10 minute
10 period. This allows the reactor temperature to rapidly drop from
235C to about 1 49C. The cooling rate is about 8. 5C/minute.
Once the product reaches 300~F, a vacuum (12 mmHg) is then
pulled on the system and the system continues to cool at a
reduced rate to 82 . 2C .
The product has the following characteristics:
Refractive index6û . 5 ~butyro scale at 60C)
Saponi~ication value101
Hydroxyl value 418
pH 7
Free fatty acid content 296
