Antioxidant system for the preservation of vitamin A

in Ultra Rice

Yao Olive Li, Jane Lam, Levente L. Diosady, and Shirley Jankowski

Abstract of generally accepted food additives, improved vitamin A

stability while reducing the price, thus greatly improving
Background. Ultra Rice grains are micronutrient- the commercial viability of Ultra Rice grains for use as
fortified, extruded rice grains designed to address specific a rice fortificant.
nutritional deficiencies in populations where rice is a
staple food. Vitamin A and some of the B vitamins, as
well as iron and zinc, are target nutrients for fortifica- Key words: Antioxidants, rice fortification, stabilizer,
tion through Ultra Rice technology. Vitamin A is sensi- Ultra Rice, vitamin A, reformed rice grains
tive to degradation. Therefore, the original Ultra Rice
formulations included stabilizers, some of which were
not approved as food additives in all of the receiving Introduction
markets.
Objective. To develop a new antioxidant system Vitamin A deficiency (VAD) has been identified as a
for improving vitamin A storage stability in Ultra major health concern in many developing countries
Rice grains, while complying with international food [1]. It is estimated that 25% of people in the developing
regulations. world suffer from a subclinical deficiency of vitamin
Methods. Ten formulations were prepared containing A. If vitamin A deficiency were eliminated, more than
various combinations of hydrophilic and hydrophobic 1 million childhood deaths would be prevented annu-
antioxidants, as well as moisture stabilizers. Accelerated ally [2].
vitamin A storage stability tests were conducted at 25o, Fortification of suitable foods with vitamin A has
35o, and 45oC with 70% to 100% relative humidity. been well recognized as a cost-effective approach to
Results. The most stable samples contained one or combat vitamin A deficiency [3, 4]. Multiple food
more phenolic antioxidants, a water-soluble antioxidant, vehicles have been explored for vitamin A fortifica-
and stabilizing agents. The best results were obtained by tion, including monosodium glutamate, salt, sugar,
using butylated hydroxyanisole (BHA) in combination oil, bread, and cereals [3]. Among those, rice is an
with butylated hydroxytoluene (BHT) as the hydrophobic ideal candidate, since it is one of the most popular
antioxidants and ascorbic acid as the hydrophilic anti- cereals consumed worldwide, serving as a staple food
oxidant. Citric acid and sodium tripolyphosphate (STPP) for nearly half of the world’s population in some 39
were used to chelate metal ions and to stabilize moisture, countries [5]. Vitamin A deficiency is widespread in
respectively. The best formulations retained more than many rice-consuming populations.
85% and approximately 70% of the added vitamin A at Several approaches to rice fortification have been
25o and 45oC, respectively, after 24 weeks storage. explored, including powder enrichment and coat-
Conclusions. The best antioxidant system, composed ing of rice grains with moisture-resistant substances,
which applied the selected micronutrients by either
dry blending or wet coating [6, 7]. These methods have
Yao Olive Li, Jane Lam, and Levente L. Diosady are affiliated not proven successful because of significant vitamin
with the University of Toronto, Canada. Shirley Jankowski losses during common food-preparation processes
was formerly employed by PATH (Program for Appropriate and high costs.
Technology in Health), Seattle, Washington, USA. The scientific literature indicates that vitamin A deg-
Please direct queries to the corresponding author: Levente
Diosady, Department of Chemical Engineering and Applied radation generally follows first-order kinetics through
Chemistry, University of Toronto, 200 College St., Toronto, two mechanisms. Oxidative degradation is caused by
ON M5S 3E5, Canada; e-mail: [email protected]. free-radical attack on the vitamin A molecule. Since

82 Food and Nutrition Bulletin, vol. 30, no. 1 © 2009, The United Nations University.
Preservation of vitamin A in Ultra Rice 83

retinoids contain highly conjugated double bonds, they retained in the formulation after 6 months of storage
form a reactive, electron-rich system susceptible to at 35°C. Compared to the storage temperatures (23°C
reactions with electrophilic compounds, i.e., oxidation and 35°C), the relative humidity (55% and 80%) did not
[8]. The second mechanism is isomerization of vita- have significant impact on VAP stability. The authors
min A from its all-trans form to either 9-cis or 13-cis thus suggested that special sample handling might be
forms. This is mainly caused by exposure to light or required to avoid the significant loss of VAP in the
the presence of other light-sensitive substances, such fortified rice under high-temperature conditions.
as riboflavin, in a multivitamin mixture [9–12]. Because vitamin A retention at high temperature and
Vitamin A loss due to oxidative degradation appears humidity was insufficient, we proposed to investigate
to be the dominant mechanism in food systems. For the effect of various antioxidants on vitamin A degra-
example, Kim et al. [12] reported that vitamin A dation, with the goal of improving the storage stabil-
did not undergo significant isomerization during 6 ity of VAP in Ultra Rice grains using internationally
months of storage at two temperatures (23° and 45°C) approved food additives.
in a vitamin A palmitate (VAP)-fortified cornflake
product made by the Kellogg Company. The distribu-
tion of two isomers (5% of 13-cis and < 1% of 9-cis) in Materials and methods
VAP remained nearly constant throughout the stor-
age, regardless of sample types and storage conditions. Materials
The authors suggested that a free (or peroxy) radical
mechanism was responsible for the loss of more than Rice flour and consumer shortening (Crisco, Procter
90% of the VAP. and Gamble, Canada) were procured from a local
Ultra Rice (a registered trademark in the United supermarket. VAP was obtained from Roche Vitamins,
States of Bon Dente International) grains consist of BASF Corporation, Firmenich of Canada and Watson
highly fortified manufactured rice grains produced by Foods. The majority of antioxidants used in the project
extrusion, which are then blended into market rice. were obtained from Sigma-Aldrich chemical company,
This is a promising novel technology for rice fortifica- including butylated hydroxyanisole (BHA), butylated
tion, which has the unique advantage of protecting the hydroxytoluene (BHT), tert-butylhydroquinone
added micronutrients within the manufactured grains (TBHQ), sodium metabisulfite, ascorbic acid, a�����
scor-
and minimizes losses during storage, washing, and byl palmitate, citric acid, and sodium tripolyphosphate
cooking [13]. (STPP). D-Alpha-tocopherol (vitamin E) was obtained
The Ultra Rice technology was initially developed from Roche Vitamins. Sodium alginate was provided
in the late 1980s by Dr. James Cox and his son, Robert by ISP Alginates. All formulation components and ana-
Duffy Cox, of Lynden, Washington, USA. The Coxes lytical reagents were food grade or analytical grade.
transferred the patent to PATH (Program for Appropri-
ate Technology in Health) in 1997 [13]. Murphy et al. Ultra Rice grain production
[14] studied the stabilization of vitamin A in Ultra Rice
grains by a combination of saturated lipids and varied Ultra Rice grains are typically extruded in a forming
antioxidants, i.e., tocopherol and/or ascorbic acid. They extruder. The rice shape is achieved by a face cutter,
concluded that the types of saturated lipids and antioxi- which cuts the extruded dough parallel to the longitu-
dants used in the formulations played important roles dinal axis of the “rice” grain.
in the overall stability of vitamin A in Ultra Rice grains.
They also confirmed that VAP, the ester form of vita- Dough formation
min A, was much more stable than pure retinol. Flores Each formulation to be tested was made up in four
et al. [15] reported the bioavailability of vitamin A in 1-kg batches of dough with the use of a Kitchen Aid
rice fortified with Ultra Rice grains fed to 83 Brazilian stand mixer (Ultra Power Series, USA) and a Moulinex
children. The study proved the nontoxicity, sensory Mincer (Model HV6, Paris) during phase 1 of this
acceptability, and desired bioavailability of Ultra Rice project. The optimized formulations were later used
grains in the vitamin A–deficient children. Lee et al. during phase 2 to produce 4-kg batches of Ultra Rice
[16] further studied the cooking and storage stability grains with the use of a La Monferrina commercial
of VAP in the fortified rice containing Ultra Rice grains pasta extruder. The procedure described below was
made with an antioxidant system consisting of a com- followed for each formulation.
bination of butylated hydroxyanisole (BHA), butylated
hydroxytoluene (BHT), and tricalcium phosphate. They Dry ingredient mixture. The dry ingredients were com-
concluded the VAP in this formulation was quite stable bined in the mixer and homogenized through continu-
in different cooking tests (> 75% retention), whereas ous blending as follows:
its storage stability was greatly dependent on the stor- 1. The powder form of VAP and water-soluble ingredi-
age conditions, with less than 50% of the original VAP ents, including the water-soluble antioxidants, were
84 Y. O. Li et al.

combined in a beaker and mixed manually. bags containing approximately 150 g each. The 18
2. Approximately 50% of rice flour was placed into the bags of rice grains for each formulation were vacuum
mixer bowl. The contents of the beaker containing sealed, and 6 bags were placed in each of three ovens,
the minor ingredients were added along with an one CSZ Z8 Environmental Chamber (Cincinnati
alginate. The solids were homogenized by agitation Sub-Zero Co., USA) and two Precision Scientific Low
for 3 minutes. Temperature Incubators (Model 815, Chennai, India),
3. The remaining flour was then added to the mixer, and which were kept at 25°, 35°, and 45°C at 70% to 100%
the mixture was further blended for approximately relative humidity for each temperature condition. The
10 minutes to obtain a uniform composition. samples were analyzed at time zero (immediately after
production), and then monthly for 6 months. Vitamin A
Oil-based mixture. The shortening (Crisco) was heated retention was reported as the percentage of the original
in a beaker placed in a water bath until the temperature amount of vitamin A added to the formulations. The
reached 70° to 80°C. At this point, fat-soluble synthetic storage stability was tested on Ultra Rice formulations,
antioxidants (BHA, BHT, or TBHQ) were added to the not on rice samples fortified with 1% of Ultra Rice.
melted shortening and dissolved using mild agitation.
The oil-based mixture was then added to the dry Vitamin A determination
ingredients in the mixer and blended for approximately
5 minutes. Then water was slowly added to the mixture The Association of Official Analytical Chemists method
with vigorous agitation, which was continued for 5 43.003 (AOAC Fourteenth Edition), and Methods of
to 10 minutes. Water served as a processing aid, and Vitamin A Assays [17] were used to develop the method
the quantity used was based on achieving the correct of extracting vitamin A from the Ultra Rice grains. The
moisture for extrusion. The final product was typically vitamin A concentration in the extract was determined
a soft, creamy white-colored dough mass with a grainy by spectrofluorometry using an excitation wavelength
texture. of 330 nm and an emission wavelength of 480 nm. USP
Vitamin A Reference Solutions were used to calibrate
Ultra Rice grain formation and storage the method.
The dough was extruded immediately after dough
formation, using a modified die on either a Moulinex Vitamin A extraction from simulated rice premix samples
meat grinder (Moulinex HV6 Mincer, Paris) or a La Sample preparation. Each bag containing 150 g of grains
Monferrina Pasta Extruder (Model P12, Castell’ Alfero, was ground with a clean coffee grinder and sieved to
Italy), as shown in figure 1. The 8-mm-wide extruded obtain a uniform fine powder. Before each sample was
dough strips were cut into approximately 1.5-mm tested, it was well mixed in a Zip-Lock bag. Then the
pieces with a rotating knife to obtain the rice grain- material was randomly taken from every part of the
shaped particles. bag and accurately weighed with an analytical balance.
After extrusion, the Ultra Rice grains were coated Approximately 10 mg of the powder was weighed into a
with a crosslinking agent and then spread onto metal 50-mL Teflon-lined screw-capped glass centrifuge tube
trays and dried at 40°C overnight in a forced-air oven. containing 5 mL of distilled water. The tubes were foil
The dried grains (with a moisture content of 8% to wrapped prior to sample preparation. Four individual
10%) were then remixed by hand once again, sieved to samples were obtained from each bag. Thus, for each
remove broken pieces, and placed into 18 polyethylene formulation at each storage temperature, four replicate
results were obtained and averaged.

Saponification. To each of the tubes containing the

weighed, ground samples, 5 mL of potassium hydroxide
solution (50% w/w) and 8 mL of pyrogallol solution
(1%, w/v of 95% ethanol) were added. The tubes were
capped and stirred briefly with a vortex mixer to com-
bine all components. The tubes were then submerged
in a 60°C water bath for 30 minutes. After 30 minutes,
the tubes were cooled in a water bath.

Extraction of vitamin A. To each sample tube, 10 mL

of spectrophotometric grade hexane was added and
the contents were mixed for 1 minute with a vortex
mixer, after which the contents were allowed to settle.
Two distinct layers would normally form in the tube:
FIG. 1. Extruder die and cutter schematic
the upper clear layer consisted of hexane and vitamin
Preservation of vitamin A in Ultra Rice 85

A, while the bottom aqueous phase formed a brown contribute to oxidation in both polar and nonpolar
solution. With the use of a disposable pipette, the clear phases. Although vitamin A is known to be fat soluble,
upper layer was transferred to a 25-mL volumetric flask it was prudent to design a system to stabilize both the
wrapped in foil. An additional 10 mL of hexane was lipid and the aqueous phases. We investigated the use
added to the remaining aqueous layer once the first 10 of both fat- and water-soluble antioxidants, as well as
mL had been removed, and the contents were mixed combinations of primary and synergistic, secondary
again for 1 minute. After separation, the hexane layer antioxidants. A literature review was performed first
was removed as before, and the hexane solution was on the properties, food regulations, and applications of
added to the first extract in the volumetric flask. The various food antioxidants, and then a series of antioxi-
combined extracts were diluted to 25 mL with hexane dant combinations was proposed, as shown in table 1.
and capped. In all formulations, VAP was added at the level of
1,600 IU of retinol per gram of Ultra Rice, which was
Fluorescence measurement designed to provide one-half to one-third of the RDA
The fluorescence of the extract was quantitated with the when 200 g of the fortified rice containing 1% of Ultra
use of a Perkin Elmer Spectrofluorometer in 10-mm Rice was consumed daily. Phenolic antioxidants, i.e.,
matched quartz cuvettes. For each formulation, four BHA, BHT, and TBHQ, were added at 0.01%. Alpha-
replicate analyses were obtained, resulting in a mean tocopherol was used at 1.5%. Ascorbic acid and citric
value of vitamin A in each sample and a relative stand- acid were added at 0.5% and STPP at 0.03%. The quan-
ard deviation. tities of all additives complied with the regulations of
the US Food and Drug Administration (FDA) and the
Formulation design for antioxidant combinations Codex Alimentarius.

Ultra Rice grains are made up of a complex food Calculation of kinetic and thermodynamic parameters
matrix, including rice flour, VAP, selected antioxidants,
essential structural agents, and vegetable shortening. Vitamin A degradation was expected to follow first-
The major components are polar, and therefore water order reaction kinetics. The first-order rate constant (k)
was expected to play an important role in the forma- was calculated for each formulation by plotting linear
tion and transport of ions and free radicals that can regression of ln(% remaining) versus the time (days)

TABLE 1. Formulation design

Water-soluble
Formulation Fat-soluble antioxidant antioxidant Synergists Note
1 BHA, alpha-tocopherol Ascorbic acid, Original formulation*
NaHS
2 BHA, alpha-tocopherol Ascorbic acid Same as 1, without NaHS
3 BHT, alpha-tocopherol Ascorbic acid Same as 2, replaced BHA
with BHT
4 BHA, BHT, alpha-tocopherol Ascorbic acid Used both BHA and BHT
5 TBHQ, alpha-tocopherol Ascorbic acid Used TBHQ as primary
synthetic antioxidant
6 TBHQ Ascorbic acid Same as 5 without alpha-
tocopherol
7 TBHQ Ascorbic acid, Citric acid With extra citric acid as
ascorbyl palmitate metal chelator
8 BHT Ascorbic acid, Citric acid, STPP With extra STPP as
ascorbyl palmitate moisture-stabilizer
9 BHA, BHT Ascorbic acid Citric acid, STPP Use combination of BHA/
BHT and citric acid/
STPP
10 TBHQ Ascorbic acid Citric acid, STPP Same as 9, replaced BHA/
BHT with TBHQ
Note: BHA, butylated hydroxyanisole; BHT, butylated hydroxytoluene; NaHS, sodium metabisulfite; STPP, sodium tripolyphosphate; TBHQ,
tert-butylhydroquinone
* The original formulation was obtained from PATH (the Program for Appropriate Technology in Health), which was based on all previous
research on vitamin A fortified Ultra Rice.
86 Y. O. Li et al.

Regression analysis correlation coefficient

using the Origin Pro 7.5 program package. The stand- (confidence level)
ard deviation, correlation coefficient (R), and half-life 9 0.9423 (p < .01)

Ln [vitamin A palmitate remaining (%)]

(t1/2) (time required for 50% of VAP to degrade) were 4.7 10 0.9735 (p < .01)

calculated. Based on the temperature dependence

of the rate constant, the activation energy (∆Ea) and 4.6
frequency factor (lnA) were obtained by using the
Arrhenius equation: lnA = lnk + Ea/RT, where R is the
gas constant and T is the absolute temperature (Kelvin). 4.5
The enthalpy of activation (ΔH‡) was obtained by plot-
ting ln(k/T) vs. 1/T, and the entropy of activation (ΔS‡)
4.4
was obtained from the expression of rate constants
based on the transition state theory: ln(k/T) = ln(kb/h)
+ΔS‡/R -ΔH‡/RT, where kb is Boltzmann’s constant and 4.3
h is Planck’s constant. ΔG‡ was determined by differ- 0 30 60 90 120 150 180
ence: ΔG‡ =ΔH‡ – TΔS‡ [18]. Storage period (days)

FIG. 2. Apparent first-order degradation of vitamin A palmi-

Results and discussion tate (VAP) in formulations 9 and 10 during storage at 25°C
for 6 months

Vitamin A degradation kinetics

(day–1) depending on sample type, storage temperature
In our study, VAP followed an apparent first-order (45°, 55°, or 65°C) and water activity (aw) (0.11 or 0.75).
degradation in each of the formulations, as shown in In a later study of vitamin A–fortified Ultra Rice grains
figure 2. The degradation rate was dependent on the by Lee et al. [16], the first-order rate constant varied
storage temperatures and the antioxidant systems, in a relatively narrow range from 7.77 × 10–2 to 2.09
varying from 0.09 × 10−2 to 6.79 × 10−2 (day−1). With × 10–2 (day–1), where the samples were made with an
the use of the Origin Pro 7.5 program package, the antioxidant system consisting of BHA, BHT, and trical-
rate constants and half-lives were estimated for all cium phosphate and stored at different conditions (23°
10 formulations stored at three temperatures. The or 35°C and 55% or 80% relative humidity). Samples
half-life values obtained ranged from 10 to more than with improved antioxidant systems in our study have
800 days (table 2). No significant difference would be shown consistently lower rate constants for vitamin A
anticipated in the presence of dry market rice during degradation (0.09 × 10–2 to 6.79 × 10–2 [day–1]) in the
normal storage or shelf-life. Ultra Rice grain formulations.
These values were consistent with those reported in
previous studies. Murphy et al. [14] tested several Ultra Effects of temperature on vitamin A stability during
Rice grain formulations using antioxidant combina- storage
tions composed of alpha-tocopherol, BHA, BHT, and/or
ascorbic acid. In the study, the rate constant of vitamin In general, VAP degraded faster (with greater rate
A degradation varied from 16.84 × 10–2 to 0.59 × 10–2 constants) at higher storage temperatures, consistent

TABLE 2. Rate constants and half-lives estimated for all 10 formulations stored at three temperatures
First-order rate constant (x10 –2)(day –1) Half-life (days)
Formulation 25°C 35°C 45°C 25°C 35°C 45°C
1 2.32 ± 0.15 3.39 ± 0.89 NA 30 20 NA
2 3.50 ± 0.98 5.41 ± 0.75 6.79 ± 0.71 20 13 10
3 2.16 ± 0.36 4.24 ± 0.78 3.74 ± 0.55 32 16 19
4 1.80 ± 0.44 2.46 ± 0.56 2.59 ± 0.57 39 28 27
5 1.51 ± 0.44 2.07 ± 0.42 2.67 ± 0.33 46 34 26
6 0.73 ± 0.23 0.83 ± 0.20 0.87 ± 0.30 95 84 80
7 0.58 ± 0.08 0.76 ± 0.07 0.80 ± 0.09 120 91 87
8 0.57 ± 0.08 0.75 ± 0.09 0.80 ± 0.12 121 92 88
9 0.09 ± 0.01 0.23 ± 0.03 0.28 ± 0.05 805 298 252
10 0.10 ± 0.01 0.20 ± 0.03 0.22 ± 0.04 693 348 312
NA, not available
Preservation of vitamin A in Ultra Rice 87

with first-order kinetics. As expected, the best VAP more effective than either BHA or BHT (formulation
retention was observed at the lowest temperature, at 5). However, the combination of BHA and BHT had
25°C (table 3). This is in agreement with other findings similar stabilizing power to TBHQ, which indicated
[12, 14, 16]. However, the stability of VAP was greatly that they had a synergistic interaction in the system, as
affected by the antioxidants used, as discussed in the shown by formulations 4 vs. 5 and 9 vs. 10.
next section. The calculated rate constants and VAP retentions
(tables 2 and 3) suggested that formulations 9 and 10
Effect of antioxidants on vitamin A palmitate were most stable, with more than 80% of VAP retained
degradation after 6 months at 25°C and more than 60% at 35° and
45°C. Thus, synergy between both the synthetic anti-
As shown in table 3, the antioxidant systems played oxidants and between the water-soluble antioxidants
a significant role in stabilizing VAP in the Ultra Rice and STPP was evident.
grains. All formulations contained ascorbic acid (vita-
min C), which is a very effective oxygen scavenger Thermodynamic stability of the formulations
and a strong reducing agent and metal chelator. The
formulations were differentiated by the fat-soluble The thermodynamic properties of these formulations
antioxidants and secondary antioxidants used. were further examined by estimating the Arrhenius
In general, the performance of all 10 formulations activation energy (Ea), the enthalpy (ΔH‡), the entropy
can be categorized into two groups: formulations 1 to (ΔS‡), and the free energy of activation(ΔG‡). The
5 had greater degradation rate constants (> 1.0 × 10–2 Arrhenius plot for each formulation was obtained by
day–1), irrespective of storage conditions, whereas plotting the natural logarithm of the rate constant (ln
formulations 6 to 10 had much smaller rate constants k) versus the inverse of storage temperature (in Kelvin)
(< 1.0 × 10–2 day–1). Formulations 1 to 5 contained (T-1). The slopes of the plots were used to calculate the
alpha-tocopherol, whereas an extra metal chelator Arrhenius activation energies (Ea) of all formulations.
(citric acid), moisture stabilizer (STPP), or both were The Ea values for all formulations were positive, which
added to formulations 7 to 10. The observed rate con- confirmed that increasing temperature always leads
stants suggested that incorporation of alpha-tocopherol to an increase in the rate of vitamin A degradation
in the formulations (i.e., 1 to 5) may actually decrease (tables 2 and 3).
the stability of some systems stabilized with combina- Upon checking the values of E a and ΔH‡, it was
tions of phenolic antioxidants. On the other hand, citric found that formulations 9 and 10 had much higher Ea
acid and STPP showed a synergistic effect with other and ΔH‡ than the other formulations, which suggested
primary antioxidants in stabilizing VAP, as indicated that these two formulations required more energy for
by the much smaller rate constants for formulations VAP degradation to proceed. It further confirmed that
9 and 10. the combination of one of the phenolic antioxidants
When comparing the primary, phenolic antioxidants with ascorbic acid, citric acid, and STPP was most suc-
BHA, BHT, and TBHQ, we found that BHT was slightly cessful in preventing VAP degradation in the Ultra Rice
more effective than BHA in stabilizing the system grains. When formulations 9 and 10 were compared,
(formulation 3 vs. 2), whereas TBHQ was significantly the former showed greater values of Ea and ΔH‡, which

TABLE 3. VAP retention in all formulations after storage for 6 months at three temperatures
VAP retention (% of originally added)
after 6 mo storage
Formulation Antioxidant combination 25°C 35°C 45°C
1 BHA, tocopherol, ascorbic acid, NaHS 6.76 ± 1.98 0 0
2 BHA, tocopherol, ascorbic acid 1.47 ± 0.92 3.68 ± 0.31 2.14 ± 1.42
3 BHT, tocopherol, ascorbic acid 13.16 ± 2.70 9.93 ± 0.92 7.87 ± 1.20
4 BHA, BHT, tocopherol, ascorbic acid 18.42 ± 1.06 13.25 ± 1.25 8.40 ± 1.11
5 TBHQ, tocopherol, ascorbic acid 27.09 ± 0.84 15.12 ± 2.35 7.52 ± 1.12
6 TBHQ, ascorbic acid 50.60 ± 2.49 38.98 ± 2.73 37.17 ± 3.54
7 TBHQ, ascorbic acid, ascorbyl palmitate, citric acid 54.00 ± 0.56 35.83 ± 5.49 34.73 ± 2.79
8 BHT, ascorbic acid, ascorbyl palmitate, citric acid, STPP 52.91 ± 0.77 37.08 ± 0.56 40.27 ± 1.31
9 BHA, BHT, ascorbic acid, citric acid, STPP 85.38 ± 3.06 65.98 ± 3.01 63.11 ± 2.07
10 TBHQ, ascorbic acid, citric acid, STPP 82.32 ± 2.43 68.17 ± 3.77 69.58 ± 2.60
BHA, butylated hydroxyanisole; BHT, butylated hydroxytoluene; NaHS, sodium metabisulfite; STPP, sodium tripolyphosphate; TBHQ, tert-
butylhydroquinone; VAP, vitamin A palmitate
88 Y. O. Li et al.

suggested that a BHA/BHT combination (formulation VAP after 6 months of storage at 25° and 45°C, respec-
9) was more temperature sensitive than TBHQ (formu- tively, whereas the original formulation retained hardly
lation 10). This is consistent with the results of the rate any VAP. The first-order rate constants of the two best
constants and VAP retentions (tables 2 and 3). formulations were less than 0.3 × 10–2 day–1 under
Linear regression of all experimental data of ΔH‡ all three storage conditions, which was less than one-
and ΔS‡ indicated that there was a strong correlation tenth of the rate constant for the original formulation.
between the two values, with a correlation coefficient Accordingly, the best formulations have much longer
of 0.962 at p < .001. This suggested that the ΔH‡ and estimated half-lives, i.e., more than 600 days when
ΔS‡ values followed the compensation law for all stored at 25°C, as compared with a 30-day half-life for
the formulations so as to yield a relatively constant the original formulation.
ΔG‡ [19], which indicated that the VAP followed the The results of this study demonstrate the need for
same oxidative degradation mechanism in all studied both hydrophilic and hydrophobic antioxidants to
formulations [18]. Compared with the observations ensure the stability of vitamin A in Ultra Rice grains. As
of VAP stability in Ultra Rice grains reported earlier expected, the synthetic antioxidants TBHQ, BHA, and
[14, 16] (table 4), the formulations developed in our BHT were effective in preventing vitamin A degrada-
study showed consistently smaller rate constants and tion. The combination of BHA and BHT showed syn-
smaller activation energy values (Ea), which suggested ergistic effects, as indicated earlier in the literature. We
that our formulations were less temperature depend- also confirmed that ascorbic acid, citric acid, and STPP
ent and therefore more stable at higher temperatures. enhance the activity of primary antioxidants, probably
In addition, the consistently smaller values of enthalpy by stabilizing the moisture in the system. The incor-
(ΔH‡) and entropy (ΔS‡) in our formulations resulted poration of these water-soluble synergists is the major
in unchanged values of the free energy (ΔG‡) in the difference between the antioxidant systems used in
range of 26 to 28 kcal/mol, which was similar to the this study and those used in the earlier studies [14, 16].
values of ΔG‡ reported in previous studies. Antioxidant activity is due to the fact that by scaveng-
ing free radicals and their precursors in the polar phase,
their migration to the nonpolar phase is prevented, thus
Conclusions increasing the stability of the formulation.
All of the components of the proposed antioxidant
The results confirm that appropriately selected anti- systems are commonly used in food products and are
oxidant systems can greatly retard vitamin A palmitate well accepted by most jurisdictions. The quantities of
(VAP) degradation in Ultra Rice grains, which leads to these additives used in the present formulations are
an extended shelf-life. well below the prescribed limits in the regulations of
Specifically, the most stable antioxidant systems the FDA and the Codex Alimentarius, indicating that
consist of BHA/BHT or TBHQ as hydrophobic anti- the current literature has not found any long-term neg-
oxidants, ascorbic acid as hydrophilic antioxidant, and ative health implications at the levels allowed. Phenolic
citric acid/STPP as synergists. Formulations made antioxidants are much cheaper than alpha-tocopherol.
with these antioxidant systems (formulations 9 and There is no evidence that the uncontrolled mixture of
10) retained more than 80% and approximately 70% of phenolic compounds in natural antioxidants is safer

TABLE 4. Comparison of stability parameters of vitamin A palmitate in Ultra Rice grains

Parameter Murphy et al. 1992 [14] Lee et al. 2000 [16] This study
Studied material Ultra Rice grains Fortified rice with UR grains Ultra Rice grains
Storage conditions 45°C, 55°C, 65°C, 23°C or 35°C, 25°C, 35°C, 45°C,
aw 0.11 or 0.75 55% or 80% RH 70~100% RH
Rate constant k (× 10–2 day –1) 0.59 ~ 16.84 2.09 ~ 7.77 0.09 ~ 6.79
Activation energy—Ea 7.5 ~ 28.5 kcal/mol NA 7 ~ 45 kJ/mol
or 1.7 ~ 10.8 kcal/mol
Enthalpy of activation—ΔH‡ 6.8 ~ 27.9 kcal/mol NA 4.39 ~ 42.45 kJ/mol
or 1.05 ~ 10.05 kcal/mol
Entropy of activation—ΔS‡ -5 ~ –69 kcal/mol-Deg NA -0.26 ~ -0.37 kJ/mol-K
or -16.6 ~ -23.9 Kcal/
mol-Deg
Free Energy of activation—ΔG‡ 28 ~ 30 kcal/mol NA 110 ~ 118 kJ/mol
or 26 ~ 28 kcal/mol
NA, not available; RH, relative humidity
Preservation of vitamin A in Ultra Rice 89

than the pure synthetics—which are also cheaper and Acknowledgments

more effective. The proposed reformulation of the anti-
oxidant system will result in a significant improvement This study was financially and technically supported
over the original formulation in terms of vitamin A by PATH (Program for Appropriate Technology in
stability, nutritional quality, and ingredient costs. Health), Seattle, Washington, USA.

References
1. World Health Organization. Highlights of recent activi- 9. Ball GFM. Vitamin A: Retinoids and the provitamin A
ties in the context of the World Declaration and Plan of carotenoids. In: Ball GFM. Vitamins in foods: Analy-
Action for Nutrition. Geneva: WHO, 1995. sis, bioavailability, and stability. New York: CRC Press
2. Micronutrient Initiative/UNICEF. Vitamin and mineral 2006:39–92
deficiency: A global progress report. Ottawa, 2004. 10. Gaylord AM, Warthesen JJ, Smith DE. Effect of fluores-
Available at: http://micronutrient.org/CMFiles/PubLib/ cent light on the isomerization of retinyl palmitate in
VMd-GPR-English1KWW-3242008-4681.pdf skim milk. J Food Sci 1986;51:1456–8.
3. Lotfi M, Manner MGV, Merx RJHM, Naber-van den 11. Murphy PA, Engelhardt R, Smith SE. Isomerization of
Heuvel P. Micronutrient fortification of foods: Current retinyl palmitate in fortified skim milk under retail fluo-
practices, research, and opportunities. Ottawa: Micro- rescent lighting. J Agric Food Chem 1988;36:592–5.
nutrient Initiative/International Development Research 12. Kim Y-S, Strand E, Dickman R, Warthesen J. Degradation
Centre/International Agriculture Centre, 1996. of vitamin A palmitate in corn flakes during storage. J
4. Bishai D, Nalubola R. The history of food fortification in Food Sci 2000;65:1216–9.
the United States: Its relevance for current fortification 13. PATH. PATH’s Ultra Rice® Project, 2007. Available at:
efforts in developing countries. Chicago: University of http://www.path.org/files/MCHN_ultrarice_leaflet.pdf.
Chicago Press, 2002. Accessed 29 November 2008.
5. Juliano BO. Rice in human nutrition. Food and Agricul- 14. Murphy PA, Smith B, Hauck C, O’Connor K. Stabiliza-
ture Organization/International Rice Research Institute. tion of vitamin A in a synthetic rice premix. J Food Sci
FAO Food and Nutrition Series No. 26. Rome: Food and 1992;57:437–9.
Agriculture Organization, 1993. 15. Flores H, Guerra NB, Cavalcanti ACA, Campos FACS,
6. Hoffpauer DW. Rice enrichment for today. Cereal Foods Azevedo MCNA, Silva MBM. Bioavailability of vitamin A
World 1992;37:757–9. in a synthetic rice premix. J Food Sci 1994;59:371–2, 377.
7. Clarke R. Micronutrient fortification of food: Technol- 16. Lee J, Hamer ML, Eitenmiller RR. Stability of retinyl
ogy and quality control. Presented at FAO Technical palmitate during cooking and storage in rice fortified
Consultation on Food Fortification and Quality Control, with Ultra Rice fortification technology. J Food Sci 2000;
November 1995, Food and Agriculture Organization, 65: 915–9.
Rome FAO Food and Nutrition Paper, 1996. Accessed 17. Parrish DB, Moffitt RA, Noel RJ, Thompson JN. Vitamin
21 December 2008. Available at: http://www.fao.org/ A. In: Augustin J, Klein BP, Becker DA, Venugopal PB,
docrep/W2840E/w2840e0b.htm#annex%204%20%20 eds. Methods of vitamin assay, 4th ed. New York: John
%20micronutrient%20fortification%20of%20food:%20 Wiley and Sons, 1985:153–84.
technology%20and%20quality%20control 18. Kirk JR. Stability of vitamins in dehydrated foods. In:
8. Van den Berg H, Faulks R, Fernando Granado H, Hir- Rockland LB, Stewart GF, eds. Water activity: Influ-
schberg J, Olmedilla B, Sandmann G, Southon S, Stahl, ences on food quality. New York: Academic Press,
W. The potential for the improvement of carotenoid 1981:531–66.
levels in foods and the likely systemic effects. J Sci Food 19. Labuza TP. Enthalpy/entropy compensation in food
Agric 2000;80:880–912. reactions. Food Technol 1980;34:67–77.