|Title of Invention||
"METHOD TO DECREASE THE ALDEHYDE CONTENT OF POLYESTERS"
|Abstract||The present invention relates to a method to decrease an aldehyde content of a polyester that comprises incorporating into the molten polyester an effective amount of an additive that is capable of catalyzing a hydride-transfer reaction between an organic donor molecule and said aldehyde, said additive being disposed substantially throughout said polyester.|
|Full Text||The present invention relates to a method to decrease an aldehyde content of a polyester.
Background of the Invention
 Polyesters, especially poly(ethylene terephthalate) (PET) are versatile
polymers that enjoy wide applicability as fibers, films, and three-dimensional structures. A particularly important application for PET is for containers, especially for food and beverages. This application has seen enormous growth over the last 20 years, and continues to enjoy increasing popularity. Despite this growth, PET has some fundamental limitations that restrict its .application in these markets. One such limitation is its tendency to generate acetaldehyde (AA) when it is melt processed. Because AA is a small molecule, AA generated during melt processing can migrate through the PET. When PET is processed into a container, AA will migrate over time to the interior of the container. Although AA is a naturally occurring flavorant in a number of beverages and food products, in many instances the taste imparted by AA is considered undesirable. For instance, AA will impart a fruity flavor to water, which detracts from the clean taste preferred for this product.
 A second limitation in the use of PET for packaging applications is its
tendency to become more yellow with increased or more severe processing. This tendency to become more yellow has been associated with the presence of various aldehydes present in the polyester. One such aldehyde is the aforementioned acetaldehyde. A second aldehyde known to promote yellowing in polyesters is 4-carboxybenzaldehyde (4-CBA). 4-CBA is an impurity formed during terephthalic acid manufacture, and while methods exist to decrease the 4-CBA level of terephthalic acid to 25-100 ppm, reducing the 4-CBA content of terephthalic acid below these levels is difficult to achieve. Even at these low concentrations, 4-CBA can adversely affect the color of PET, and is thought to be a contributor to the increased yellowing of PET during recycling or after melt blending with polyamides.
 PET is traditionally produced by the transesterification or esterification
of a terephthalate precursor (either dimethyl terephthalate or terephthalic acid, respectively) and ethylene glycol, followed by melt polycondensation. If the end use
 In addition to the afore-mentioned process-related methods, other
methods to minimize AA content of polyesters include modification of the polymer
itself through the use of lower intrinsic viscosity (IV) resins or the use of lower
melting resins. However, lower IV resins produce containers that are less resistant to
environmental factors such as stress crack failure. Lower melting resins are achieved
by increasing the copolymer content the PET resin, but increasing the copolymer
content also increases the natural stretch ratio of the polymer, which translates into
decreased productivity in injection molding and blow molding.
 Methods to reduce the impact of yellowing from 4-CBA include the
aforementioned purification of the terephthalic acid feedstock. Other methods include
the addition of toners (especially cobalt salts, or blue and red dyes) to mask the
yellowness. However, these approaches also have inherent costs, and do not
completely address the issue of increasing yellow discoloration in polyesters with
increasing or more severe processing, especially for recycled PET.
 Another approach to minimize the AA content of polyesters has been
to incorporate additives into the polyester that will selectively react with, or scavenge,
the acetaldehyde that is present. Thus Igarashi (US 4,837,115) discloses the use of
amine-group terminated polyamides and amine-group containing small molecules as
AA scavengers. Igarashi teaches that the amine groups are effective because they can
react with AA to form imines, where the amine nitrogen forms a double bond with the
AA moiety. Igarashi teaches that essentially any amine is effective. Mills (US
5,258,233; 5,650,469; and 5,340,884) and Long (US 5,266,416) disclose the use of
various polyamides as AA scavengers, especially low molecular weight polyamides.
Turner and Nicely (WO 97/28218)Disclose the use of polyesteramides. These
polyamides and polyesteramides are believed to react with AA in the manner
described by Igarashi. Rule et. al. (US 6,274,212) discloses the use of heteroatomcontaining
organic additives that can react with acetaldehyde to form unbridged 5- or
6-member rings, with anthranilamide being a preferred organic additive.
 While these AA scavengers are effective at reducing the AA content of
polyesters, they suffer from their own drawbacks. For example, relatively high
loadings of polyamides or polyesteramides are needed to effect significant AA
reductions, and very significant yellowing of PET can occur on incorporation of these
amine-containing additives. The use of anthranilamide also results in some degree of
discoloration of PET. This color formation inherently restricts the use of these
additives to packaging where the PET can be tinted to mask the color. However, most
PET packages in use today are clear and uncolored.
 Another drawback of these approaches for controlling the AA content
of PET is related to their mechanism of action in that they all depend on incorporation
of an additive that reacts stoichiometrically with acetaldehyde. Consequently, the
amount of AA that can be sequestered in a polyester by these additives is inherently
limited to the amount of additive incorporated. Moreover, because the reaction
between these additives and AA is thermodynamically reversible, the amount of
additive incorporated must be substantially greater than the amount of AA to be
sequestered. This limitation is especially important if relatively large amounts of AA
need to be scavenged from the polyester, such as in polyesters that have been
subjected to very severe processing, or in polyesters that have not had their meltphase
AA content reduced via solid-state polymerization.
 A final drawback of the additives disclosed in the above references is
that, to a greater or lesser degree, they all are extractable, and therefore can
themselves affect the taste of food or beverages packaged in containers made from
polyesters incorporating these additives.
 A different method of decreasing the AA content of polyesters is
disclosed by Rule (US 6,569,479) wherein acetaldehyde present in melt-processed
PET issoxidized to acetic acid by the action of an active oxidation catalyst and
molecular oxygen. While this method is catalytic, and is therefore capable of
removing greater than stoichiometric amounts of acetaldehyde, it suffers from the
drawback that the active oxidation catalysts useful for this invention are relatively
unselective and are themselves active for generating acetaldehyde under meltprocessing
conditions, thus limiting their effectiveness. Thus, the greatest amount of
decrease in the beverage AA content disclosed by this invention is only 32%.
Summary of the Invention
 The present invention provides a method to decrease an aldehyde
content of a polyester hy incorporating into the polyester an effective amount of an
additive that is capable of catalyzing a hydride-transfer reaction between an organic
donor molecule arid an aldehyde. The hydride transfer reactions contemplated in the
invention can be aMeerwein-Ponndorf-Verley, Oppenauer, Cannizzaro, or
Tishchenko reaction. Typically., the organic donor molecule will be the same or
another aldehyde, or it may be an alcohol or glycol. Exemplary additives are hydrous
metal oxides such as hydrous zirconium oxide. The additive can be incorporated into
a molten polyester such as poly(ethylene terephthalate) honiopolymer or copolymer.
In a preferred embodiment, the additive is present in the polyester at a concentration
between about 1 and 2000 ppm, more preferably about 10 and 500 ppm. Exemplary
additives have a particle size less than about 30 microns, and have a surface area of
about 200-500 m2/g. The treated polyester can be advantageously molded into a solid
article, such as a container for food or beverage. The invention is similarly directed to
articles produced from the inventive method.
Detailed Description of the mvention
 The present invention relates to a method which substantially
decreases the aldehyde content of polyesters, especially polyesters that are made from
ethylene glycol and aromatic diacids or diesters. These polyesters are especially
prone to contain aldehydes derived from the thermal degradation of the ethylene
linkages, or from impurities in the aromatic diacids. The present invention is
particularly directed toward PET, but is also applicable to other polyesters that
contain aldehydes either as impurities or as reaction byproducts. Examples of other
polyesters contemplated by this invention include but are not limited to poly(ethylene
naphthalate), poly(cyclohexylenedimethylene terephthalate), poly(ethylene
isophthalate), and copolymers of these polyesters.
 In the present invention., the aldehydes present in these polyesters are
reduced to alcohols by contact with an additive capable of catalyzing a hydride (ET)
transfer from an organic donor molecule to the aldehyde. Reactions of this type are
known collectively as Meerwein-Ponndorf-Verley, Oppenauer, Cannizzaro, or
Tishchenko reactions. In the Meerwein-Ponndorf-Verley and Oppenauer reactions,
the hydride donor molecule is an alcohol, while in the Cannizzaro and Tishchenko
reactions the hydride donor molecule is the same or another aldehyde. Thus, in the
Meenvein-Ponndorf-Verley and Oppenauer reactions, the net effect of the reaction is
the reduction of an aldehyde or ketone to an alcohol, with the simultaneous oxidation
of a different alcohol to an aldehyde or ketone; while in the Cannizzaro and
Tishchenko reactions the net effect is the disproportionation of two aldehyde (or
ketone) molecules to an alcohol and an acid, with possible condensation of the alcohol
and acid to form an ester. Preferably, the hydride donor molecule is naturally present
in the polyester, but it is within the scope of this invention for the hydride donor
molecule to be intentionally added.
 These hydride-transfer reactions are known to occur both in liquid
medium and in the vapor phase. When these reactions occur in a liquid medium, the
reaction conditions typically employed involve the use of high loadings of catalyst,
long reaction times at elevated temperatures, and use of high concentrations of the
reactants. Often, the hydride donor or the hydride acceptor is also the reaction
solvent. Although the selectivity for these reactions is high, conversions are often not
quantitative, with conversions of 30-90% being typical. The reaction times required
to effect sufficient conversion range between 2 and 200 hours. Reaction temperatures
are usually between 80 and 300 deg C. In a liquid medium, the preferred catalyst is
often an aluminum alkoxide, although a number of other catalysts have been disclosed
in the literature. Thus, Mizusaki (US 4,877,909) discloses the use of a hydrous
zirconium oxide as a catalyst for the Meenvein-Ponndorf-Verley reaction. Matsushita
(US 4, 910,177) discloses the use of partially dehydrated metal oxides selected from
the group consisting of titanium, tin, iron, aluminum, cerium, and niobium. Heveling
(US 6,075,145) discloses the use of a partially dehydrated hydrous zirconium oxide
that has been modified by the addition of copper or nickel salts.
 When these reactions are conducted in the vapor phase, the reactants
are evaporated and passed over a solid catalyst at elevated temperatures. In the case
of vapor phase reactions, the catalyst loadings are very high relative to the amount of
reactants present at any given time. Temperatures employed range from 100 deg C to
300 deg C, and conversions are invariably less than quantitative. Suitable catalysts
disclosed for the vapor phase reaction are similar to those disclosed above as effective
for the liquid phase reaction. Additional catalysts disclosed to be effective for the
vapor phase reaction include hydrous hafnium oxide, hydrous vanadium pentoxide,
hydrous titanium dioxide, hydrous niobium oxide, and hydrous tantalum oxide
(Reichle, US 5,354,915).
 While these reactions are effective for the reduction of aldehydes, it is
surprising that these hydride-transfer reactions would be effective for the removal of
aldehydes from polyesters. For example, compared to the high concentrations of
reactants necessary to achieve reasonable conversions in the liquid or vapor phase,
aldehydes are present in polyesters at very low concentrations, typically at levels of 1-
100 ppm: In contrast to the high loadings of catalyst required for achieving
reasonable reaction rates in the liquid or vapor phase, only low concentrations of the
hydride-transfer catalyst can be tolerated in a polyester, since catalyst loadings greater
than approximately 0.25 wt% may adversely affect other properties of the polyester,
such as clarity. Furthermore, because most of the acetaldehyde present in a polyester
container sidewall is formed via the room-temperature hydrolysis of vinyl esters and
methyl dioxolane, the catalyst must be active at room temperature. Moreover, many
of the catalysts employed for these reactions are deactivated by the presence of
moisture, whereas moisture is an unavoidable component of polyesters under normal
use conditions. Finally, for catalyst to be active at room temperature, it must be active
when the polyester is in the solid state, where the diffusional rates for the reactants are
many orders of magnitude lower than in the liquid or gas phase.
 However, as will be seen in the examples presented below, the
hydride-transfer reactions disclosed in the present specification do occur in polyesters
at room temperature, even with very low loadings of catalyst and at very low
concentrations of aldehydes. That the reaction is so effective under these conditions
is both surprising and highly useful, because it provides an efficient method to
catalytically decrease the aldehyde content of polyesters.
 While not being bound to any particular theory, it is believed that the
hydride-transfer reaction of the present invention is much more effective than would
be expected because of the nature of the catalyst The catalysts effective for the
hydride-transfer reactions of the present invention are heterogeneous catalysts; that is,
the catalytic action of these materials is provided by a solid surface where the
organization of the atoms at the material's internal surface is critical to the catalytic
activity. For example, most of the catalysts disclosed to be effective for the liquidphase
or vapor phase hydride-transfer reaction are insoluble hydrous metal oxides.
Even aluminum alkoxides, which might be considered to be soluble catalysts, are
believed to function because, under the liquid-phase reaction conditions employed,
they are present as polymeric complexes (see, for example, Whittaker, J., J. Am.
Chem. Soc. 1969, 91, 394). In the field of heterogeneous catalysis, it is known that
the catalytic activity of a surface can be adversely affected by the presence of too
much reactant or reaction product; that is, the activity of the catalyst depends on the
presence of a substantial number of unoccupied sites on the catalyst surface. In the
process of the present invention, the amount of aldehyde present is low enough that
the heterogeneous catalyst surface is not folly occupied by the reactants or reaction
products, and consequently, the catalytic activity exhibited by these materials is in the
present environment far greater than that previously disclosed, where high levels of
reactants have been employed.
 Organic donor molecules suitable as hydride transfer agents for use in
the present invention may be intentionally added to the polyester, or may be naturally
present in the polyester. For food-contact applications, it is preferred that the organic
.donor molecule be naturally present in the polyester. In particular, for removal of
acetaldehyde it is preferred that the organic donor molecule is acetaldehyde, and the
reaction that occurs is the disproportionation of acetaldehyde to ethanol, acetic acid,
and ethyl acetate. As will be seen in the examples below, this appears to be the
predominant reaction that occurs in PET containing catalysts of the present invention.
 Thus, unlike previous methods to sequester acetaldehyde which
depend on preventing the migration of acetaldehyde by binding it to a larger
molecule, in the present invention, acetaldehyde is converted catalytically into
ethanol, acetic acid, and ethyl acetate. These molecules are of similar molecular
weight to acetaldehyde, and consequently are capable of migrating from the polyester
into the package interior. However, these molecules all possess taste thresholds that
me as much as 1000 times greater than that of acetaldehyde; consequently, migration
of these reaction products to the package interior does not pose a taste or odor issue.
In addition, these compounds are all generally recognized as safe (GRAS) and are
frequently used as direct food additives: therefore they do not pose an issue regarding
food safety. Finally, because the catalysts of the present invention are solid, insoluble
materials, these additives are largely incapable being extracted from the polymer
matrix, and therefore, have no potential to directly impact the taste of products.
 Specific catalysts effective for the reduction of aldehydes in polyesters
can be selected from those catalysts effective at catalyzing the Meerwein-Ponndorf-
Verley, Oppenauer, Cannizzaro or Tishchenko reactions in the liquid or vapor phase.
Those catalysts, for example, include the hydrous oxides of magnesium, calcium,
strontium, barium, scandium, titanium, vanadium, chromium, manganese, iron,
cobalt, nickel, zinc, aluminum, gallium, indium, tin, cadmium, niobium, zirconium,
yttrium, hafnium, tantalum, lanthanum, and the rare earths. Because the most
effective method for incorporation of these catalysts into a polyester requires adding
the catalyst to the molten polymer or polymer precursors, preferred catalysts are ones
that are substantially stable to hydrolytic degradation at the temperatures that they will
be exposed to. For this reason, aluminum alkoxides are not preferred as catalysts for
the reaction of the present invention, because they are substantially decomposed to
monomeric aluminates under the conditions required to process high molecular
weight polyesters. Similarly, the carboxylates, carbonates, and alkoxides of other
metals are not preferred, since they will also readily decompose under the reaction
conditions to dissolved monomeric species.
 Catalysts effective for the reduction of aldehydes in polyesters also
need to exhibit sufficient stability to thermal degradation. For this reason, hydrous
aluminum oxides are somewhat less preferred as catalysts, since they are more prone
to undergo dehydration to aluminum oxide under the conditions found in the polyester
melt. In contrast, the hydrous oxides of zirconium, hafnium, niobium, and tantalum
do not exhibit similar propensities for dehydration, and consequently possess
substantial activity, even after exposure to the high temperatures found in the molten
polyester. Therefore, these hydrous oxides are preferred catalysts for the present
invention. Of these, hydrous zirconium oxide is especially preferred due to its low
cost, ready availability, and high activity.
 Catalysts of the present invention can be employed as single materials,
or as physical mixtures of two or more catalytically active materials. In addition, the
catalysts employed may be pure substances, or may be deposited on other materials.
For example, hydrous zirconium oxide may be used as is, or may be deposited onto a
silica or alumina support. Optionally, the activity of the catalysts may be enhanced
further by absorbing other materials onto their surfaces. Thus, the activity of hydrous
zirconium oxide may be further enhanced by absorbing the oxides of lithium, sodium,
potassium, magnesium, calcium, strontium, yttrium, lanthanum, cerium, neodynrum,
copper, iron or nickel onto its surface. In addition, the hydrous metal oxides may be
partially dehydrated to improve their catalytic activity and/or stabilize them against
further dehydration when added to a molten polyester. For purposes of the present
disclosure, it is to be understood that the term hydrous metal oxides includes partially
dehydrated hydrous metal oxides.
 Because the additives of the present invention are insoluble,
heterogeneous catalysts, the specific surface area, pore radius, and particle size may
influence the effectiveness of the catalyst. In general, higher surface areas and greater
pore radius will correspond with higher specific activities; therefore, a material that
possesses a higher specific surface area and a given pore radius may be preferred over
a material with the same nominal chemical composition that possesses a lower
specific surface area and the same pore radius. For the hydrous zirconium oxides,
increasing calcination temperature correlates with lower specific surface area but
higher pore radius; for this reason, partially dehydrated hydrous zirconium oxides
calcined at about 300 deg C possess higher activity than hydrous zirconium oxides
calcined at lower or higher temperatures. Typical surface areas for active hydrous
zirconium oxides are in the 200-500 m2/g range, while surface areas for relatively
inactive zirconium oxides are in the 2-5 m2/g range. Similarly, smaller particle sizes
correlate with higher activity, since the diffusion path to the catalyst will be less. This
is especially true at temperatures below about 80 deg C, where the diffusion of
acetaldehyde may be the rate limiting step for reaction. In addition, the smaller the
particle size, the less likely will the catalyst affect the processability and clarity of the
polyester, and therefore smaller mean particle size materials are preferred relative to
larger particle size material of the same chemical composition and specific surface
area. In the present invention, high activity is obtained by additives where the
average particle size is about 15 microns, while relatively lower effectiveness is
obtained by additives where the average particle size is >30 microns. There is no
specific lower limit to the preferred particle size, except as dictated by the cost,
availability, and processability of the materials.
 Because the additives of the present invention are catalytic in their
action, the quantity of aldehydes that can be removed from polyesters is not
dependent on the amount of additive incorporated. However, for a given additive,
higher loadings will result in a higher rate of reaction of the aldehyde. Higher
loadings are therefore relatively preferred over lower loadings. The upper limit of the
amount of additive to be incorporated is dictated by the rate of aldehyde removal
desired, and by the impact of higher loadings on other factors, such as degree of
participate haze, processability, and cost. As will be seen in the examples, loadings of
100-1000 ppm are usually sufficient to achieve the technical effect desired for most
 The compositions of the polyesters disclosed in the present invention
are not critical, and essentially any monomer or co-monomer can be utilized without
adversely affecting the performance of the additives in reducing the aldehyde content.
Because of their economic importance, polyesters based on terephthalic acid and
ethylene glycol are especially important.
 The point of addition of the additives of the present invention is
relatively unimportant, as long as they are added prior to forming the final article.
However, it is important to maximize the degree of dispersion of the additives within
the polyester matrix. For this reason, it is preferred to add these catalysts where
sufficient melt mixing can occur. For most applications, it is sufficient to add the
catalysts as powders or as dispersions immediately prior to the injection molding
process. However, it is possible to add the catalysts before or during the meltpolymerization
process. Addition of the catalysts of the present invention early in
polymerization process is preferred when removal of aldehydes present as impurities
in the raw materials (such as 4-CBA) is desired. Addition of these catalysts at the end
of melt-polymerization is preferred when the object is to decrease the time required to
remove AA or other aldehydes in the solid-state polymerization process, or when the
object is to eliminate the need for a solid-state polymerization process altogether.
 The method of incorporation of the disclosed additives into polyesters
is not critical. The additives can be dispersed in a solid or liquid carrier, and mixed
with the polyester pellets immediately before injection molding. They may also be
incorporated by spraying a slurry of the additive onto the polymer pellets prior to
drying. They may be incorporated by injection of a dispersion of the additive into
pre-melted polyester. They may also be incorporated by making a masterbatch of the
additive with the polyester, and then mixing the masterbatch pellets with the polymer
pellets at the desired level before drying and injection molding or extrusion. In
addition to the use of slurries or dispersions, the additives of the present invention
may be incorporated as dry powders.
 Because the additives of the present invention are effective at greatly
reducing the acetaldehyde content of polyesters, where low AA levels are important
they are useful for achieving very low preform and beverage AA levels in polyester
containers. However, the additives of the present invention are also useful for
enabling the practice of modes of polyester container production that are now
precluded because of the issue of acetaldehyde. Thus, the additives of the present
invention can enable the use of high-activity melt-polymerization catalysts, which
heretofore have been avoided because of the issue of AA. They can also enable the
use of higher melting polyesters which have desirable physical properties, but
concomitantly higher AA content because of the higher melt-processing temperatures
required. They can also enable a revision of the design of injection molding
equipment, since careful control of AA can now be less of a design factor. And
finally, they can enable totally new methods of manufacturing of polyester containers,
such as direct conversion of polyester melts into preforms without prior solidification
and AA removal.
 The following examples illustrate the use of the disclosed additives for
decreasing the aldehyde content of polyesters. The examples are provided to more
fully describe the invention and are not intended to represent any limitation as to the
scope thereof. In these examples, the effectiveness of the additives in reducing the
aldehyde content was determined by measuring the AA content of PET in the
presence of the additive, relative to the AA content of identically processed PET
without the additive. The AA content was determined by taking a representative
portion of the melt-processed polyester, grinding it to pass a 20 mesh (850 micron)
screen, and desorbing the contained AA from 0.1 grams of the ground polyester by
heating at the specified time and temperature in a sealed 20 mL vial. The desorbed
AA in the headspace of the vial was then analyzed using a gas chromatograph
equipped with a flame ionization detector. Beverage AA levels were determined by
removing a 5 gram aliquot of the beverage, placing the aliquot into a 20 mL vial,
adding 2.5 grams of sodium chloride, and desorbing the contained AA at 80 deg C for
30 minutes, followed by analysis of the AA desorbed into the headspace of the vial
using a gas chromatograph equipped with a flame ionization detector.
 In the following examples, amorphous melt-polymerized 0.80 IV PET
pellets were dried in a vacuum oven at 80 deg C for 3 days. After drying, the pellets
had a residual moisture content of 3 kilograms of the dried resin was added 3 grams of mineral oil and the indicated
amount of hydrous zirconium oxide. (The hydrous zirconium oxide used in examples
2-4 had an average particle size of about 15-20 microns.) The additive was dispersed
onto the pellets by tumbling, and then the pellets were injection molded into 27 gram
preforms using an Arburg unit cavity press. The barrel temperature utilized was 280
deg C, and the cycle time was 30 seconds. The AA content of the molded preforms
was measured after heating the ground preform at 150 deg C for 30 minutes. The
results tabulated below demonstrate the effectiveness of the hydrous zirconium oxide
at reducing the AA content of PET, even when the PET initially had a very high AA
 The gas cliromatographic traces from Examples 1-4 were examined.
For Example 1, the only GC peaks observed had the same retention times as
acetaldehyde and 2-methyl-l,3-dioxolane. For Examples 2-4, new peaks were
observed that were not present in the control, and the GC integration area of the new
peaks increased in proportion to the amount of hydrous zirconium oxide added. GCMS
analysis of the gases desorbed into the headspace of the vials confirmed that
ethanol, acetic acid, and ethyl acetate were formed in the polyester samples containing
hydrous zirconium oxide, but not in the control. This result demonstrates that the
hydrous zirconium oxide decreased the acetaldehyde content of the polyester by
catalytically converting the acetaldehyde to ethanol, acetic acid, and ethyl acetate,
rather than by merely sequestering the acetaldehyde.
 The impact of the hydrous zirconium oxide on the color of the
polyester in examples 1-4 was evaluated by grinding the preforms to a fine powder,
and then measuring the L, a, and b color values using a Hunter color meter.
[003 6] These results show that the presence of hydrous zirconium oxide
actually improves the color of polyesters, in marked contrast to previously disclosed
acetaldehyde-reducing additives, which invariably caused an increase in the amount
of color (especially yellowness, which is reflected in increased b* values for
polyesters containing those previously disclosed additives).
 The AA content test using ground preforms from Examples 1-4 above
was repeated, except that the ground preforms were held at room temperature (22-24
deg C) for the indicated number of days. The vial headspace was sampled without
heating; therefore, the AA content measured was the amount that diffused from the
ground polymer at room temperature. The results tabulated below demonstrate that
for the control, the amount of AA continues to increase with time, consistent with the
continual hydrolysis of methyl dioxolane and vinyl esters in the ground polymer. In
contrast, for the ground resins containing hydrous zirconium oxide, the initial amount
of AA released is decreased in proportion to the amount of added hydrous zirconium
oxide, and the rate of release over time is negligible. These results are consistent with
the hydrous zirconium oxide catalytically consuming acetaldehyde at room
 The gas chroniatographic traces from days 3, 10, and 17 for Examples
9-12 were examined. For Example 9, the only GC peaks observed for all three days
had the same retention times as acetaldehyde and 2-methyl-l,3-dioxolane. For
Examples 10-12, a new peak was observed with the same retention time as ethanol.
The integration area of the new peak increased in proportion to the amount of hydrous
zirconium oxide added and with time. This result demonstrates that the hydrous
zirconium oxide incorporated into PET is catalytically active for converting
acetaldehyde to ethanol at room temperature.
 Preforms from Example 1-4 were blow molded into 20 oz. containers
using standard blow molding conditions. Immediately after blow molding., the bottles
were filled with carbonated water (containing 4 volumes of CO2), capped, and stored
at room temperature. Aliquots were removed at the time intervals specified in the
table below and analyzed for beverage AA content.
 These results demonstrate that use of the catalyst of the present
invention also results in markedly lower beverage AA levels. This observation is
particularly important, since this result demonstrates that catalysts of the present
invention are active at room temperature even in the presence of carbon dioxide and
 In the following examples, crystallized, solid-state polymerized 0.84
IV PET pellets were dried overnight at 150 deg C. After drying, the pellets had a
residual moisture content of kilograms of the dried resin was added 3 grams of mineral oil and the indicated
amount of hydrous metal oxide. All of the hydrous metal oxides utilized in Examples
18-22 had an average particle size of approximately 1-2 microns. The additives were
dispersed onto the pellets by tumbling, and then the pellets were injection molded into
27 gram preforms using a Husky unit cavity press. The barrel temperature utilized
was 270 deg C, and the cycle time was 30 seconds. The AA content of the molded
preforms were measured after heating the ground preform at 150 deg C for 30
minutes. The results tabulated below demonstrate the activity of the hydrous oxides
of other metals for catalyzing the reduction of acetaldehyde in PET via a hydridetransfer
 In the following example, crystallized, solid-state polymerized 0.84 IV
PET pellets were dried overnight at 150 deg C. After drying, the pellets had a
residual moisture content of kilograms of the dried resin was added 3 grams of mineral oil and the indicated
amount of hydrous zirconium oxide. The pellets were then melt extruded at 280 deg
C, and the extradate quenched in a water bath and chopped to form pellets. The
amorphous pellets were crystallized at 160 deg C for two hours, and were then
subjected to solid-state polymerization at 200 deg C. The AA content of the pellets
were measured before crystallization, after crystallization, and once an hour during
solid-state polymerization. The AA content of the pellets with and without added
hydrous zirconium oxide are tabulated below:
1 hour SSP
2 hours SSP
3 hours SSP
4 hours SSP
5 hours SSP
6 hours SSP
AA Content of PET Control
AA Content of PET with 500 ppm
Hydrous Zirconium Oxide (ppm)
 As can be seen from this example, the addition of a catalyst of the
present invention to molten PET greatly reduced the initial AA content of the meltprocessed
polymer. In addition, further reductions in AA content of the polymer to
levels less than 0.4 ppm were achieved much more rapidly and under much milder
conditions than were possible in the absence of the added catalyst.
 The importance of the structural nature of the hydride-transfer catalyst
was tested by evaluating the effectiveness of different compounds for reducing the
AA content of melt-processed PET.
Comparative Examples 1-13
 In the following comparative examples, crystalline, solid-state
polymerized 0.84 IV PET pellets were dried overnight in a vacuum oven at 150 deg
C. After drying, the pellets had a residual moisture content of residual AA content of of mineral oil and the indicated amount of the specified zirconium compound. The
additives were dispersed onto the pellets by tumbling, and then the pellets were
injection molded into 27 gram preforms using a Husky unit cavity press. The barrel
temperature utilized was 270 deg C, and the cycle time was 30 seconds. The AA
content of the molded preforms was measured after heating the ground preform at 150
deg C for 30 minutes. The results tabulated below demonstrate that soluble zirconium
compounds such as zirconium 2-ethylhexanoate, zirconium acetate, zirconium
benzoate, zirconium carbonate, and zirconium tartrate are not effective catalysts for
the hydride-transfer reaction disclosed in the present invention.
 The GC traces for the Comparative Examples 1-13 were examined for
the presence of peaks with retention times corresponding to ethanol, acetic acid, or
ethyl acetate. In no case were these peaks observed.
Comparative Examples 14-16
 In the following comparative examples, crystalline, solid-state
polymerized 0.84 IV PET pellets were dried overnight in a vacuum oven at 150 deg
C. After drying, the pellets had a residual moisture content of residual AA content of of mineral oil and the indicated amount of aluminum isopropoxide, a catalyst active
for the Meerwein-Pondorf-Verley reaction in the liquid phase at low temperature, but
that would be expected to decompose to monomeric aluminate species under the meltprocessing
conditions employed for polyesters. The additive was dispersed onto the
pellets by tumbling, and then the pellets were melted and extruded. The barrel
temperature utilized was 270 deg C, with a residence time of 90 seconds. The AA
content of the extrudate was measured after heating the ground extrudate at 150 deg C
for 30 minutes. The results tabulated below demonstrate that compounds capable of
dissolving under the melt-processing conditions employed for polyesters are not
effective catalysts for the hydride-transfer reaction disclosed in the present invention.
 The invention has been described with' reference to a preferred
embodiment. Modifications and alternatives will be apparent to the skilled artisan
upon reading and understanding the preceding detailed description. It is intended that
the invention be construed as including all such modifications and alternatives that
fall within the scope of the appended claims or equivalents thereof.
1. A method to decrease an aldehyde content of a polyester that comprises incorporating into the molten polyester an effective amount of an additive that is capable of catalyzing a hydride-transfer reaction between an organic donor molecule and said aldehyde, said additive being disposed substantially throughout said polyester.
2. The method as claimed in claim 1, wherein the organic donor molecule in said hydride-transfer reaction is an alcohol.
3. The method as claimed in claim 1, wherein the organic donor molecule in said hydride transfer reaction is at least one of an aldehyde and a ketone.
4. The method as claimed in claim 1, wherein the additive is a hydrous metal oxide.
5. The method as claimed in claim 4, wherein the additive is a hydrous zirconium oxide.
6. The method as claimed in claim 1, wherein the additive is incorporated into molten polyester.
7. The method as claimed in claim 1, wherein the polyester is a poly(ethylene terephthalate) homopolymer or copolymer.
8. The method as claimed in claim 1, wherein the additive is present in the polyester at a concentration between 1 and 2000 ppm.
9. The method as claimed in claim 1, wherein the additive is present in the polyester at a concentration between 1 and 500 ppm.
10. The method as claimed in claim 1, wherein the additive has a particle size less than 30 microns.
11. The method as claimed in claim 1, wherein the additive has a particle size less than 5 microns.
12. The method as claimed in claim 1, further including molding said polyester into a solid article.
13. The method as claimed in claim 12, wherein the solid article is molded into a container.
14. The method as claimed in claim 1, wherein the additive has a surface area of 200-500 m2/g.
15. The method as claimed in claim 1 wherein the organic donor molecule is naturally present in the polyester.
16. The method as claimed in claim 1 wherein the additive is a hydrous metal oxide further comprising lithium, sodium, potassium, rubidium, magnesium, calcium, strontium, yttrium, lanthanum, cerium, neodymium, nickel, copper, or iron ions.
17. The method as claimed in claim 1, further including forming a polyester container for storing food or beverage wherein said additive is selected from hydrous metal oxides and a molten poly(ethylene terephthalate) homopolymer or copolymer to form a treated material and molding said treated material to form said container.
18. The method as claimed in claim 1 wherein said additive is a hydrous zirconium oxide.
19. The method as claimed in claim 17 wherein said additive is present in the poly(ethylene terephthalate) at a concentration between 10 and 2000 ppm.
20. The method as claimed in claim 1, wherein the polyester composition has an improved flavor retaining property, and is comprised of dicarboxylic acid units and diol units, wherein said additive is selected from hydrous zirconium oxide, hydrous niobium oxide, hydrous tantalum oxide, hydrous tin oxide, hydrous aluminum oxide, and hydrous titanium oxide, said additive being present at a concentration between 10 and 2000 ppm.
21. The method as claimed in claim 1, wherein said polyester is molded into a container for food or beverage products, wherein said additive is selected from hydrous zirconium oxide, hydrous niobium oxide, and hydrous tantalum oxide, and being present at a concentration between 10 and 2000 ppm.
|Indian Patent Application Number||236/DELNP/2006|
|PG Journal Number||16/2010|
|Date of Filing||13-Jan-2006|
|Name of Patentee||COLORMATRIX CORPORATION|
|Applicant Address||3005 CHESTER AVENUE, CLEVELAND, OHIO 44114 USA|
|PCT International Classification Number||C08K 3/22|
|PCT International Application Number||PCT/US2004/022680|
|PCT International Filing date||2004-07-14|