Food and Chemical Toxicology 43 (2005) 1413–1422

www.elsevier.com/locate/foodchemtox

Effect of storage period under variable conditions on

the chemical and physical composition and colour of Spanish
refrigerated orange juices
M.J. Esteve a, A. Fr´ıgola a,*, C. Rodrigo b, D. Rodrigo c
A´ rea de Nutricio´ n y Bromatolog´ıa, Facultat de Farma` cia, Universitat de Vale`ncia, Avda. Vicent Andre´s Estelle´s, s/n. 46100 Burjassot, Valencia,
a

Spain
b
CTT Universidad Polite´cnica de Valencia, Camino de Vera s/n. 46027 Valencia, Spain
c
Instituto de Agroqu´ımica y Tecnologı´a de Alimentos, C.S.I.C. Apartado de Correos 73, 46100 Burjassot, Valencia, Spain

Received 14 September 2004; accepted 28 March 2005

Abstract

The effects of the physicochemical and quality characteristics of various minimally pasteurized refrigerated orange Spanish
juices and their changes with storage time and temperature were investigated. Essential oils, acidity, conductivity, diacetyl index,
hydroxy- methylfurfural, formol index, viscosity and ascorbic acid varied with storage time more significantly at 10 °C than at 4 °C.
Density, colour and pectinmethylesterase did not vary at 4 °C. Some of the parameters could be used as indicators of quality loss or
spoilage of the juices. The degradation kinetics of the concentration of remaining ascorbic acid against time follows a straight line
whose slope indicates the degradation rate. A period of at least 42 days at 4 °C and 35 days at 10 °C was established as the shelf life
of the juices.
© 2005 Elsevier Ltd. All rights reserved.

Keywords: Orange juice; Composition; Colour; Physicochemical characteristics; Storage

1. Introduction the USA already exceeds the production of concentrate;

There is growing interest in consuming minimally
processed foods with characteristics closer to those of
fresh foods (Oliveira and Oliveira, 1999; Mart´ınez et
al., 1999). Juices are no exception, and there is there-
fore a strong tendency towards consumption of refriger-
ated juices (not obtained from concentrates), freshly
squeezed or with very mild pasteurization and distrib-
uted refrigerated, with a relatively short shelf life. The
consumption of refrigerated juices in the USA is cur-
rently of the order of $4.4 billion (Jago, 2004), and in
the case of orange juice manufacture of direct juice in
*
Corresponding author. Tel.: +34 9635 44955; fax: +34 9635 44954.
E-mail address: ana.frigola@uv.es (A. Fr´ıgola).
in recent years it has increased from 30% to 60%. In
Spain the annual production of orange and mandarin
juice not obtained from concentrate (fresh juice, slightly
pasteurized and frozen fresh) was of the order of 20 mil-
lion litres in 2002.
As a result of the problem that has arisen because of
the development of pathogens in some (unpasteurized)
fresh orange juices (Parish, 1996; Parish et al., 1997;
Par- ish, 1998), the image and safety of these juices has
been damaged and the FDA is recommending
manufacturers of these juices to increase safety
measures by introducing the Hazard Analysis and
Critical Control Points (HACCP) system in their
processes and applying a pas- teurization or treatment
that will ensure five decimal reductions of Escherichia
coli (Food and Drug Adminis- tration, 1998). This has
produced strong opposition

0278-6915/$ - see front matter © 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.fct.2005.03.016
141 M.J. Esteve et al. / Food and Chemical Toxicology 43 (2005) 1413–1422
4
among manufacturers, based on the loss of the ‘‘fresh
food’’ characteristics of their juices as a consequence
of thermal treatment, which is the technique available
at present.
Currently, refrigerated juices which are not obtained
from concentrates and have been subjected to mild pas-
teurization partly satisfy the requirements of higher
qual- ity demanded by consumers. The shelf life of these
juices ranges between 28 and 45 days in refrigeration
and their quality approaches that of freshly squeezed
juices.
The characteristics of these orange juices basically
depend on the raw material that they use and the condi-
tions of the process of manufacture, transport and stor-
age of the product (Parish, 1996; Kaanane et al., 1988;
Decio and Gherardi, 1992; Arena et al., 2001; Manso et
al., 2001; Johnston and Bowling, 2002). The final
characteristics of these juices are also affected by the
conditions of the mixing and pre-storage of the raw
material used.
During the production season the manufacturer can
predominantly use the varieties that attain suitable ripe-
ness in the field as raw material, depending on harvest-
ing time. This is precisely what happens during the
season when the consumer buys oranges in the market
in order to prepare freshly squeezed juice. In these con-
ditions it is clear that the quality characteristics of these
juices will vary throughout the season. Precisely in order
to obtain more uniform final qualities the manufacturers
use a properly balanced mixture of freshly squeezed
juice and fresh juice frozen or stored aseptically, with
known characteristics, so that they can be selected and
mixed appropriately.
Many studies have been carried out on the quality
and stability of pasteurized orange juices (Kaanane
et al., 1988; Graumlich et al., 1986; Mart´ın et al., 1995).
But in some cases the juices are obtained from
concentrates (Fellers and Carter, 1993), in other cases
(Graumlich et al., 1986; Mart´ın et al., 1995) the pasteur-
ization conditions applied are fairly intense (from
90 °C—15 s to 110 °C—15 s), anticipating storage at
ambient temperature or a very long shelf life in refriger-
ation. In these latter cases the impact of pasteurization
on quality is clearly appreciable. In other cases high
storage temperatures have been studied so as to be able
to observe clearly the effects of temperature on certain
factors (browning, development of hydroxymethylfur-
fural, loss of Vitamin C, etc.), and to deduce the kinetic
models that define these changes (Manso et al., 2001).
And, finally, in other cases there are studies on the effect
of certain processes, the conditions of processing, can-
ning or storage, on one or more specific quality para-
meters (Ayhan et al., 2001; Nienaber and Shellhamer,
2001; Decio and Gherardi, 1992; Manso et al., 2001;
Johnston and Bowling, 2002; Trammell et al., 1986;
Yeom et al., 2000; Farnworth et al., 1992; Choi et al.,
2002; Fan et al., 2002; Sa´nchez-Moreno et al., 2003).
In view of the absence of data for the characterization
of refrigerated orange juices obtained from fresh juices
with mild pasteurization, the aim of the present work
is to study the main physicochemical quality, nutritional
and sensory characteristics of the Spanish juices that can
be found on the market and their evolution with time (1–
6 weeks) and storage temperature in refrigeration (4
°C and 10 °C).
2. Materials and methods
2.1. Juice samples
Four samples (A, B, C and D) of refrigerated orange
juice, each from a different company, were received in
the laboratory less than 24 h after completion of their
manufacture by the companies involved. The general
manufacturing process for the orange juice was as fol-
lows: after suitable washing and hygienization of the
fruits, they were subjected to an extraction process
(FMC juice extractors with a 2-mm diameter sieve)
and the juice was introduced into a tank. The juice
was immediately subjected to a mild pasteurization
(77 °C, 20 s), rapidly cooled to a temperature of 0–
2 °C, packed in 1-L cartons and stored at 0–3 °C. Four
different batches of samples (A, B, C and D) were
re-
ceived, with 50 cartons per batch, and each batch was
di- vided into three groups. The first group was stored
at 4 °C for six weeks, the second at 10 °C for five
weeks, and the third
— at 40 °C as a control.
The parameters tested were determined in the freshly
manufactured samples of each batch, after 1, 2, 3, 4, 5
and 6 weeks in the samples stored at 4 °C and at
10 °C. The purpose of storage at 10 °C was to observe
the changes that can take place in certain physicochem-
ical parameters in cases of cold chain breaking during
the shelf life of the product. The results for each param-
eter in each of the three batches, as shown in the corre-
sponding tables and figures, are the mean values of three
different cartons, each analysed in triplicate.
2.2. Methods of analysis
Density was measured at 20 °C with a pycnometer
(FIPJF, 1968).
The pH was determined with a Crison model 2001
micro pH meter.
Total acidity was determined by means of a potentio-
metric titration of the acidity of the juice, with a solu-
tion of 0.1 N NaOH up to pH = 8.1. The results were
expressed as g/100 mL with reference to citric acid,
and converted to g/100 g with the aid of tables (FIPJF,
1968; BOE, 1988; Redd et al., 1986).
The conductivity measurements were made with a
Crison conductivity meter, with manual temperature
compensation at 25 °C.
 Viscosity was determined with a rotational viscome-
ter (Haake, VT5R Viscotester) with the spindle R1 to
200 rpm.
The pectinmethylesterase (PME) method was based
on measuring the carboxylic groups released from pectin
by the effect of the enzymatic activity of the juice
(Rouse and Atkins, 1955).
The diacetyl index was determined using the method
proposed by Kimball (1986).
The formol index was analysed by a potentiometric
evaluation of the acidity (FIPJF, 1984).
Hydroxymethylfurfural (HMF) was measured spec-
trophotometrically at 550 nm (FIPJF, 1972).
Essential oils were measured by the method proposed
by Scott and Veldhuis (1966).
Ascorbic acid was determined by polarography
(Aparicio et al., 1992).
The colour was determined using a Hunter Labscan
II spectrophotometric colorimeter controlled by a com-
puter that calculated colour ordinates from the reflec-
tance spectrum (Calvo and Dura´n, 1997). The
results were expressed in accordance with the CIELAB
system with reference to illuminant D65 and with a
visual angle of 10°.
For the sensory evaluations of colour and appear-
ance, 25 assessors without any training were randomly
selected from volunteer personnel working at the Agro-
Chemistry and Food Technology Institute. When the test
was done, freshly made orange juice, kept in the freezer
( 40 °C) and—unfrozen when the test was about to be
made, was used as a control sample. The juice samples
were presented in glasses with a capacity of 100 mL,
and the colour and appearance were assessed in a booth
of a neutral grey colour (Munsell value 7) with a lighting
that corresponded to illuminant D65. For both
characteristics (colour and appearance) the assessors
tested the similarity of the sample to the con- trol. For
colour they rated from 10 to 0 (10 typical col- our–0
degraded) and for appearance they rated from 10 to 0
(10 turbid–0 clarified).
2.3. Statistical methods
An analysis of variance was made with the results ob-
tained for each parameter, and Tukey’s test was used to
reveal any differences.
3. Results
Table 1 shows the results of the physicochemical
characteristics and colour of the four batches of juices
(A, B, C and D).
An analysis of variance (ANOVA) was applied to the
results obtained during the storage study with the four
juices for each of the characteristics in order to deter-
mine whether there were significant differences, in which
case an LSD test was applied.
3.1. Essential oils
The essential oils of oranges are formed as products
of secretion or excretion in metabolic processes. They
are found in secreting glands located in the flavedo,
from which they are incorporated into orange juices
during extraction in quantities that depend on the raw
material used and the extraction conditions applied.
Essential oils contribute aroma to juices, but an excess
produces a burning sensation in the throat. The maxi-
mum concentration of essential oils permitted interna-
tionally for orange juice (AIJN Reference Guidelines)
is 0.3 mL/L, and in the case of the USDA, for Grade
A pasteurized orange juice it is 0.35 mL/L (Redd et al.,
1986; USDA, 1985). The concentrations in juices B, C
and D can be considered normal (0.144–0.235 mL/L),
but the concentration of juice A is very close to the max-
imum permitted (0.331 mL/L). When the juices were
compared with one another, there were significant
differ- ences between all of them except between C and
D (Table 1).
During storage, a statistically significant (p 6 0.05)
decrease in oils was only observed in juice A, the juice
with the highest content, reaching 0.27 mL/L in the
fourth week of storage at 4 °C. The variations observed
in the other juices, both at 4 °C and at 10 °C, were not
significant.
3.2. Total acidity
Organic acids contribute to the particular flavour and
palatability of orange juice and are found as a result of
biochemical processes, or, in the case of fermentations,
through the development of certain spoilage microor-
ganisms. To a large extent, acidity protects against the
development of pathogens. In orange juice, citric acid
is the most abundant, followed by malic, both being
present mostly as free acids, although in limited quanti-
ties they are also combined as citrates or malates, which
gives orange juice its buffer effect. Other non-volatile
free acids (oxalic, tartaric and galacturonic, quinic and
many others) are found in much lower quantities
(Lo´ - pez, 1995). The total acidity in the four juices
studied was significantly different (p 6 0.05) (Table 1),
but in all cases it was within the recommended values
(0.6 to
1.6 g/100 g) (Redd et al., 1986). With storage, a signifi-
cant increase in acidity (p 6 0.05) was observed in sam-
ple D at 4 °C (1.37 g/100 g) and at 10 °C (1.36 g/100 g),
after 4 and 3 weeks respectively, and in sample C at
4 °C and 10 °C after 2 weeks (1.06 and 1.09 g/100 g)
(Figs. 1 and 2). This increase in acidity indicates the
start of spoilage or fermentation of the sample.
Table 1
Physicochemical characteristics of orange juices A, B, C and D prior to storage
Parameter Juice A Juice B Juice C Juice D
a b d
Ascorbic acid (mg/100 mL) 42.46 ± 2.59 46.62 ± 0.95 73.45 ± 1.91 48.60 ± 0.07c
Colour
a* 5.17 ± 0.12a 5.24 ± 0.07a 11.74 ± 0.13c 8.25 ± 0.11b
b* 64.64 ± 1.05c 62.89 ± 0.71b 62.33 ± 0.19a 65.42 ± 0.11d
h = a tan(b*/a*) 85.43 ± 0.04d 85.24 ± 0.03c 79.33 ± 0.01a 82.82 ± 0.03b
L* 59.58 ± 0.11c 60.73 ± 0.12d 57.00 ± 0.24a 58.83 ± 0.19b
Conductivity (S/m) 0.307 ± 0.007b 0.310 ± 0.005b 0.243 ± 0.010a 0.331 ± 0.009c
Density (g/L) 1.048 ± 0.001a 1.049 ± 0.015a 1.061 ± 0.001c 1.052 ± 0.006b
Diacetyl index (mg/L) N.D. 0.051 ± 0.059a 0.771 ± 0.103c 0.490 ± 0.287b
Essential oils (mL/L) 0.331 ± 0.046c 0.235 ± 0.023b 0.152 ± 0.017a 0.144 ± 0.02a
Formol index (meq/100 mL) 22.3 ± 3.2c 21.9 ± 2.7b,c 19.0 ± 1.7a 20.3 ± 0.6a,b
HMF (mg/L) 0.54 ± 0.18b 0.50 ± 0.03b N.D. 0.08 ± 0.03a
pH 3.52 ± 0.01c 3.48 ± 0.11c 3.36 ± 0.03b 3.23 ± 0.03a
PME (PEU · 104) 0.63 ± 0.01c 1.33 ± 0.05d 0.45 ± 0.12b 0.35 ± 0.06a
Total acidity (g citric/100 g) 0.78 ± 0.03a 0.86 ± 0.04b 0.93 ± 0.01c 1.26 ± 0.02d
Viscosity (mPa/s) 24.3 ± 1.4a 27.3 ± 1.4c 31.3 ± 1.4d 28.0 ± 2.5b
N.D.: not detectable.
a,b,c,d
Different superscripts indicate statistically significant differences (p < 0.05).

1.6 1.6

1.4
1.4

1.2
1.2
g citric/100g
g citric/100g

1
0.8

0.8
0.6

0.6
0.4
0 1 2 3 4 5 6
0.4 time (weeks)
0 1
2 3 4 5 6 A B C D
time (weeks)

ABCD
Fig. 2. Orange juice total acidity evolution with storage time at 10 °C.

Fig. 1. Orange juice total acidity evolution with storage time at 4 °C.

case of orange juice the CIELAB hue angle value h

3.3. Colour
The colour of a food is the first quality factor that
the consumer appreciates and has a remarkable
influence on its acceptance. Colour is also an indicator
of the natural transformation of a fresh food
(ripeness) or of changes that occur during its storage
or processing (Calvo and Dura´n, 1997). In the case
of orange juices, the typical colour is generally due
to a mixture of carotenoids and, in some varieties
(blood), also of anthocyan. Instrumental measurement
of the CIELAB parameters L*, a* and b* with the
Hunter Labscan spectrophoto- metric colorimeter is
the method most used, and in the
(h = a tan (b*/a*)) is also used to express the colour. Pre-
viously the Hunter a/b quotient was used to quantify
colour objectively, as this index is also correlated with
the sensory appreciation of colour in orange juice and
its acceptance by the consume. In the measurements of
the colour of the juices analysed it was observed that
in each of the indices (CIEL*, a*, b* and h) there were
significant differences between the four juices (Table
1). The juices with the highest values of (+a*) (reddest
hue) were C and D, and the highest values of L* (lumi-
nosity) corresponded to B and A. During storage at 4 °C
there were slight decreases in L* and variations in (+a*)
and (+b*) which were not significant in any of the three
indices. With storage at 10 °C there was a significant in-
crease (p 6 0.05) in L* (luminosity) from the initial value
(59.58, 57.0 and 58.83) in juices A, C and D, respec-
tively, at the end of the first or second week (60.23,
62.86 and 65.33). In B, however, no significant varia-
tions were observed. At 10 °C there were also significant
increases in (+a*) in juice A, from its original value
(5.17) to the value after three weeks of storage (6.53),
and significant reductions in (+b*) in juice D, from its
original value (65.42) to the value after two weeks
(60.34). The variations in the other parameters and
juices at 10 °C were not significant (data not shown).
When a sensory comparison was made between the
colour of samples from the same batch but stored in dif-
ferent conditions and a freshly manufactured control
sample stored at 40 — °C, a correlation was found
between the sensory and instrumental values. Samples
A, C and D experienced the greatest degradation of col-
our at each of the two temperatures. Sample B main-
tained an adequate colour rating even when stored at
10 °C. In each batch, the colour evaluations were always
higher in the samples stored at 4 °C than in those stored
at 10 °C.
The results for sensory acceptance on the basis of
appearance (turbid or clarified juice) were very similar
to those for colour acceptance. All the juices stored at
4 °C achieved higher ratings than those stored at
10 °C. The worst appearances were those of juices A,
C and D at 10 °C and the best appearance (most turbid)
corresponded to juices B and D stored at 4 °C.
Trifiro et al. (1995) evaluated the influence of three
different storage conditions on the quality of blood
orange juice and found that colour degraded according
to first-order kinetics. Choi et al. (2002) studied the
effect of ascorbic acid retention on juice colour in blood
orange juice during refrigerated storage and concluded
that ascorbic acid degradation was highly correlated
(r > 0.93) with red colour intensity (CIE a*) in the juice.
In line with these results, in our study the best colour
(objective and sensorial) and appearance rating corre-
sponded to juice B, which also had the lowest ascorbic
acid degradation rates (1.421 and 2.225 at 4 and 10 °C
respectively), as commented below.
3.4. Conductivity
The electrical conductivity of a food represents the
facility that it offers to the passage of an electrical con-
duction current. Conductivity increases with tempera-
ture, is affected by the nature of the ions, ionic
movement and viscosity of the liquid and decreases with
the content and size of solids or particles and as a result
of the presence of non-ionic constituents (oils and sug-
ars) (Palaniappan and Sastry, 1991). The conductivity
of the four juices studied varied between 0.243 and
0.331 S/m. The differences between the conductivities
of juices A and B were not significant (0.307 and
0.310 S/m respectively), but there were significant
differ- ences between the conductivities of these two
juices and that of C (0.243 S/m), the lowest value, and
between each of these conductivities and that of D
(0.331 S/m) (Table 1). These values are somewhat
lower than the
0.427 S/m reported by Barbosa-Ca´novas et al.
(1999) and the 0.36 of Raso et al. (1999), although they
are within the customary range for orange juices,
0.3–
0.5 S/m. During storage an increase in conductivity
was only observed in juice C after two weeks at 4 °C
and after one week at 10 °C. This might be due to the
increase in acidity as a result of the onset of microbio-
logical spoilage. In the other juices the variations with
temperature and time were not significant.
3.5. Density
With the exception of juice C, which had a density of
1.061 g/L, the values obtained in the juices analysed are
within the recommended range (1.045–1.055 g/L)
(R.S.K. Regulations in Redd et al., 1986). There were
no significant differences (p 6 0.05) between juices A
and B (1.048 and 1.049 g/L), but the differences
between these two juices and C (1.061 g/L) or D (1.052
g/L) were
significant (Table 1). During storage at 4 °C the density
of the samples remained constant. At 10 °C the density
of sample C decreased significantly (p 6 0.05) in the
sec-
ond week of storage and in samples A and D the
decrease in density was significant in the third week
(data not shown). These decreases might be due to the
onset of spoilage or fermentation.
3.6. Diacetyl index
The diacetyl index is used for rapid detection of
microbiological growth in orange juice and for control-
ling hygiene conditions in the production line (Hill and
Wenzel, 1957; Murdock, 1966). Murdock (1966) pro-
posed a scale for interpreting diacetyl concentration val-
ues: between 0.040 and 0.350 ppm, safe range; between
0.375 and 0.450 ppm, caution range; between 0.470
and 0.575 ppm, danger range; and above 0.600 ppm,
risk of spoilage. According to this classification, juice
B (0.051) is in the safe range, juice C (0.771) is in the
risk of spoilage range and D (0.490) is in the danger
range (Table 1). In all cases the increase in the diacetyl
index at the end of storage was significant (data not
shown), and was more marked at 10 °C, this increase
might indi- cate the start of spoilage or fermentation of
the sample.
3.7. Hydroxymethylfurfural
The control of furanic aldehydes is important in the
evaluation of non-enzymatic browning,
adulterations, heating, incorrect storage, and
sensory characteristics
of a food. The main decomposition product of the
hydrolysis of sugars catalyzed by acid is 5-(hydroxy-
methyl)-2-furfuraldehyde (HMF). The formation of
HMF is directly proportional to storage time and tem-
perature (Mart´ın et al., 1995; Naim et al., 1993; Solo-
mon and Svanberg, 1995). In the four juices analysed,
the highest values correspond to juices A and B (0.54
and 0.50 mg/L), which do not differ significantly
(p 6 0.05) from one another. The values corresponding
to juices C and D (N.D. and 0.08 mg/L) are significantly
lower than the other values (Table 1). Nevertheless, the
values are very low in all the juices, below the tolerable
limit for orange juices subjected to mild heat treatment
(5 mg/L of HMF). This indicates that the heat treatment
to which the juices had been subjected was very mild, as
is confirmed by the high ascorbic acid content. During
storage there was an increase in the content of HMF,
which became pronounced in the final week at both tem-
peratures, except in the case of juice D (Table 2).
During storage at 4 °C significant reductions
(p 6 0.05) were only seen in sample A after six weeks
of storage. When the juice was stored at 10 °C, however,
the formol index only remained constant with time in
sample B, whereas in the other samples it decreased sig-
nificantly from week 2 onwards (Figs. 3 and 4). This
decrease can be attributed, once again, to consumption
of amino acids by microorganisms causing the start of
fermentation. However, when Trifiro et al. (1995) stud-
ied the effect of storage time and temperature on the
quality of fresh juices of pigmented oranges (Moro, San-
guinello and Tarocco varieties), they observed an in-
crease in the formal index which they attributed to an
effect of proteolysis, connected with the acid nature of
the juice and storage temperature. Villamiel et al. (1998)
studied the influence of heat treatment (conven- tional
and by microwaves) on orange juice and reported that
there were no differences in the content of amino

3.8. Formol index 24

Although the formal index is not very specific, it is 22

used to estimate the total content of amino acids in a

juice, and partly to estimate its purity. In the four juices 20
meq/100mL

analysed the value is higher than the recommended

18
value of 18 and lies within the accepted range of 15–
26, which depends on the raw material used (especially
16
variety and ripeness) and the manufacturing conditions.
Unripe oranges give low values and very ripe oranges 14
may even exceed the limits mentioned. During refining
of the juice there is a reduction in the formol index, 12
due to the elimination of pulp and rind, which are richer 0 1 2 3 4 5 6
in amino acids. The results obtained (Table 1) show that time (weeks)

there are significant differences between juices A and C, A

BC D

between A and D and between B and C. Fig. 3. Orange juice formol index evolution with storage time at 4 °C.

Table 2
Variation of HMF (mg/L) concentration with storage time at 4 and 10 °C
Storage temperature (°C) Time (weeks) Juices
A B C D
4 0 0.54 ± 0.18b 0.50 ± 0.03b N.D. 0.08 ± 0.03a
1 0.50 ± 0.19b 0.56 ± 0.10b N.D. 0.06 ± 0.02a
2 0.48 ± 0.19b 0.59 ± 0.19b N.D. 0.12 ± 0.06a
3 0.50 ± 0.08c 0.57 ± 0.15c 0.01 ± 0.00a 0.10 ± 0.03b
4 0.52 ± 0.05b 0.58 ± 0.09b 0.03 ± 0.01a 0.04 ± 0.02a
5 0.68 ± 0.10b 0.65 ± 0.08b 0.03 ± 0.01a 0.03 ± 0.01a
6 0.99 ± 0.19b 0.80 ± 0.19b 0.03 ± 0.01a 0.01 ± 0.01a
10 1 0.50 ± 0.11b 0.58 ± 0.03b N.D. 0.05 ± 0.02a
2 0.55 ± 0.12b 0.58 ± 0.07b 0.01 ± 0.01a 0.04 ± 0.01a
3 0.62 ± 0.15b 0.59 ± 0.12b 0.01 ± 0.01a 0.08 ± 0.05a
4 0.70 ± 0.20b 0.75 ± 0.15b 0.02 ± 0.01a 0.07 ± 0.02a
5 0.98 ± 0.21b 0.81 ± 0.18b 0.01 ± 0.01a 0.02 ± 0.01a
6 1.20 ± 0.30b 0.99 ± 0.20b 0.02 ± 0.01a 0.03 ± 0.01a
N.D.: not detectable.
a,b,c
Different superscripts indicate statistically significant differences (p < 0.05).
 25
23
21
19
meq/100mL
17
15
13
11
9
7
0 1 2
3 4 5 6
time (weeks)
A B C D
Fig. 4. Orange juice formol index evolution with storage time at 10 °C.
acids between the control and the juice treated by micro-
waves, but they observed a decrease in some amino
acids in the juice treated conventionally. They attributed
these differences to the presence of hot surfaces in the
conven- tional treatment.
3.9. pH
The pH values of the juices analysed are within the
normal range (3.1–4) and the differences between them
are significant (p 6 0.05), with the exception of the dif-
ferences between juices A and B. As one would expect,
the pH values are lower in the juices with more acidity.
Owing to the presence of a natural buffer medium in or-
ange juices (mainly potassium citrates and malates), the
variations in pH with storage are less pronounced than
the variations in acidity. During the six weeks of storage
at 4 and 10 °C the variations in pH observed in the juices
studied did not become statistically significant (data not
shown). The same conclusion was reached by other
authors (Kaanane et al., 1988; Mart´ın et al., 1995; Souci
et al., 1987) who studied the effect of storage time and
temperature on the pH of juices.
3.10. PME
The pectins present in orange juices may degrade if
inactivation of the natural enzymes in the juices, such
as PME, is not achieved. These enzymes catalyze the
demethoxylation of the pectins, causing an increase in
the free carboxyl groups which favours clarification of
the juices. Clarified orange juices lose their commercial
value. The values of residual PME activity in the four
juices studied are significantly different (p 6 0.05) from
one another. The levels of activity in juices A and B
(0.63 and 1.33 PEU · 104 respectively) are higher
than
the recommended levels (PEU · 104 < 0.5) and
juice C
is almost at the limit (0.45 PEU · 104) (Table 1).
During storage there was a decrease in PME activity,
which ap-
0.95
0.85
0.75
4
0.65
PEU*10
0.55
0.45
0.35
0.25
0 1 2 3 4 5 6
time (weeks)
4ºC 10°C
Fig. 5. Orange juice C PME evolution with storage time at 4 and
10 °C.
peared earlier at 10 °C than at 4 °C, but it was not sig-
nificant. In the case of juice C there was a reactivation
of PME activity in week 3, which was significant and
greater at 10 °C than at 4 °C (Fig. 5).
3.11. Viscosity
Orange juices consist of a dispersing phase or serum,
in which sugars, acids, soluble pectins, proteins, salts,
etc., are dissolved, and a dispersed phase made up of
particles of different sizes and volumes which come
from the tissues of the orange (Jime´nez and Dura
´n, 1979). Irrespective of their pulp content, orange
juices are con- sidered Newtonian fluids. The viscosity
of the juices is an important physical characteristic
because it affects the manufacturing process (whether
conventional, by heat, or by pulsed electric fields), the
conditions applied in the operation of stabilization, and
acceptance by the consumer. Freshly squeezed orange
juice has a higher viscosity than juice prepared by
reconstituting frozen concentrate (Race, 1991). As for
the viscosity values of the four juices analysed, there are
no significant differ- ences between juices B and D, but
there are between these juices and juices A and C (Table
1). During storage at 4 °C there were no significant
changes in the viscosi- ties of juices A, B and D, and
only in juice C was a sig- nificant loss observed after six
weeks of storage. At 10 °C there was a decrease in
viscosity in all the juices towards the end of the storage
period, which was not significant in juices B and D but
was significant in juice A from week 3 onwards, and in
juice C it was more rapid (week 1) and more significant
(Fig. 6). This may be connected with the onset of
spoilage or the enzymatic reaction described earlier.
3.12. Ascorbic acid
Vitamin C (ascorbic and L-dehydroascorbic acid) is
an essential nutrient for the human being and has a high
 M.J. Esteve et al. / Food and Chemical Toxicology 43 (2005) 1413–1422
34
32
30
28
mPa/s
26
24
22
20
18
0 1 2 3 4 5
time (weeks)
A B C D
Fig. 6. Orange juice viscosity evolution with storage time at 10 °C.
antioxidant power, providing protection against the
presence of free radicals and consequently participating
in the prevention of many degenerative diseases. Be-
cause of the instability of vitamin C and its nutritional
importance, its content guarantees the presence of other
nutrients and is considered an indicator of the nutri-
tional quality of processed foods (Bo¨ ttcher, 1993;
Wil- liams et al., 1995; Canet, 1996). The main source of
vitamin C for consumers is citrus fruits. The values of
ascorbic acid present at the start of storage in the four
juices analysed were significantly different from one an-
other (Table 1) and always above the minimum values
recommended for processed orange juice (40 mg/
100 mL). This leads one to suppose that ascorbic acid
had been added, especially in juice C. The ascorbic acid
content of the juices decreased during storage, faster at
10 than at 4 °C. The degradation kinetics of the concen-
tration (C) of remaining ascorbic acid in the orange juice
against time follows a straight line whose slope indicates
the degradation rate. The values of these slopes at 4 °C
are 2.273, 1.421, 2.468 and 1.845 (mg/100 mL) day—1
for juices A, B, C and D respectively.
At 10 °C, with the exception of juice D, the values of
the slopes are practically double: 4.595, 2.225 and 4.452
(mg/100 mL) day—1 for juices A, B and C respectively.
If we consider that the mean life of ascorbic acid
establishes the shelf life of the juice, a period longer
than 42 days at 4 °C and equal to 42 days at 10 °C
would be recom- mended for juices B, C and D. A shelf
life of 35 days would be recommended for juice A
because 20 mg/L was attained after five weeks at 10 °C.
This would ensure that if the storage temperature
increased (a break in the cold chain), the juice would
conserve its nutritive charac- teristics during its mean
life. Choi et al. (2002) studied the retention of ascorbic
acid with storage in blood oranges, whether fortified
with ascorbic acid or not, and they, too, observed a
linear reduction in concentration with time. More than
50% was lost within three weeks of refriger- ated (4 °C)
storage, and it was completely degraded after
1421
five weeks’ storage in the control juice. Over the seven-
week period, the rate of ascorbic acid decline was about
10.7% per week for the fortified juice and 18.6% for the
control juice, higher than the 1.5–2% per day previously
reported as the decomposition rate for ascorbic acid in
single-strength orange juice in refrigerated conditions. In
line with these results, the values of the ascorbic acid
degradation slopes obtained in our study and quoted
above show once again that the juices studied had been
fortified with ascorbic acid. Fan et al. (2002) also ob-
6 served a linear degradation of ascorbic and dehydroa-
scorbic acid with time in orange juices, whether
irradiated or not. J o h n s t o n a n d B o w l i n g ( 2 0 0 2 )
studied
the stability of ascorbic acid in commercially available
orange juices and beverages obtained from concentrate
in sealed and open containers and also found linear
regressions of the concentrations of ascorbic acid against
time, with fairly similar rates in the various cases. In the
juices obtained from frozen concentrate, the ascorbic
acid content reduced to 50% of the original after four
weeks at 4 °C. However, the degradation of ascorbic
acid observed by other authors in pasteurized orange
juice (Fellers, 1988; Pino et al., 1987) and in a mixed
pasteur- ized orange–carrot juice (Rodrigo et al., 2003)
followed a first-order kinetics, and was also a function of
time and temperature.
4. Conclusions
For all the characteristics studied there were statisti-
cally significant differences (p 6 0.05) between juices A,
B, C and D, attributable to differences in the raw mate-
rial and the processes used. Variations in the character-
istics with storage time were always more marked at
10 °C than at 4 °C. Except for pH, all the characteristics
studied (essential oils, acidity, colour, conductivity,
density, diacetyl index, hydroxymethylfurfural, formol
index, pectinmethylesterase, viscosity and ascorbic acid)
varied with storage time at 4 °C and 10 °C, but the col-
our, density and pectinmethylesterase variations were
not statistically significant at 4 °C. The shelf life of the
juices, based on the mean life of ascorbic acid, ranged
between 35 days at 10 °C and more than 42 days at
4 °C. Juice B was the one with the lowest diacetyl index
and the highest stability because parameters such as
essential oils, acidity, colour, acceptance, conductivity,
density, formol index, pectinmethylesterase and viscos-
ity remained stable with storage time and temperature.
These differences might be attributable to a greater
degree of hygienization in the process.
Acknowledgments
This study was carried out with funds from the FED-
ER Project (1FD97-0575-C03). And this work was
financed by a PETRI research and technological devel-
opment project (PTR 1995-0592-OP-03-03) and Ayudas
a Grupos de Investigacio´n de la Generalitat Valenciana
(Grupos 03/147). D. R. holds an award from CSIC.
References
Aparicio, P., Farre´, R., Fr´ıgola, A., 1992. Determinacio´ n del a
´cido asco´ rbico en vegetales por polarograf´ıa diferencial de
impulsos. Anales de Bromatolog´ıa XLI, 257–261.
Arena, E., Fallico, B., Maccarone, E., 2001. Thermal damage in blood
orange juice: kinetics of 5-hydroxymethyl-2-furancarboxaldehyde
formation. International Journal of Food Science and Technology
36/2, 145–151.
Ayhan, Z., Yeom, H.W., Zhang, Q.H., Min, D.B., 2001. Flavor, color
and vitamin C retention of pulsed electric field processed orange
juice in different packaging materials. Journal of Agricultural and
Food Chemistry 49, 669–674.
Barbosa-Ca´novas, G.V., Gongora-Nieto, M., Pothakamury, U.R.,
Swanson, B.G., 1999. Preservation of foods with pulsed electric
fields. In: Food Science and Technology Series. Academic Press,
San Diego.
BOE, 1988, Orden de 29 de Enero de 1988. Me´todos Oficiales
de Ana´lisis de Zumos de Frutas y otros Vegetales. Boletin Oficial
del Estado, Spain.
Bo¨ ttcher, H., 1993. Zur Bewertung des Na¨ hrwertes von Gemu¨ se.
Die Nahrung 37, 20–27.
Calvo, C., Dura´n, L., 1997. Propiedades f´ısicas II. O´ pticas y color.
En:
Temas en Tecnolog´ıa de Alimentos. Programa Iberoamericano de
Ciencia y Tecnolog´ıa para el Desarrollo (CYTED). Instituto Polite
´cnico Nacional, Mexico, DF.
Canet, W., 1996. Estabilidad e importancia de la vitamina C en
vegetales congelados. Alimentacio´n, Equipos y Tecnolog´ıa 5,
75– 87.
Choi, M.H., Kim, G.H., Lee, H.S., 2002. Effects of ascorbic acid
retention on juice color and pigment stability in blood orange
(Citrus sinensis) juice during refrigerated storage. Food Research
International 35, 753–759.
Decio, P., Gherardi, S., 1992. Freshly squeezed orange juice. Confr-
ucta-Studien 36, 162–167.
Fan, X., Thayer, D.W., Handel, A.P., 2002. Nutritional quality of
irradiated orange juice. Journal of Food Processing & Preservation
26, 195–211.
Farnworth, E.R., Lagace, M., Couture, R., Yaylayan, V., Stewart, B.,
1992. Thermal processing storage conditions and the composition
and physical properties of orange juice. Food Research Interna-
tional 34/1, 25–30.
Fellers, P.J., 1988. Shelf life and quality of freshly squeezed, unpas-
teurized, polyethylene-bottle citrus juice. Journal of Food Science
53, 1699–1702.
Fellers, P.J., Carter, R.D., 1993. Effect of thermal processing and
storage of chilled orange juice on flavor quality. Fluessiges-Obst
60/ 12: Fruit Processing 3/12, 436–441.
FIPJF, 1968. Me´thodes d’Analyses. Fe´de´ration Internationale
des Producteurs de Jus de Fruits (FIPJF), c/o Schweiz
Obstverband, CH-6300 Zug 2, Switzerland.
FIPJF, 1972. Me´thodes d’Analyses. Fe´de´ration Internationale
des Producteurs de Jus de Fruits (FIPJF) number 12, c/o
Schweiz Obstverband, CH-6300 Zug 2, Switzerland.
FIPJF, 1984. Me´thodes d’Analyses. Fe´de´ration Internationale
des Producteurs de Jus de Fruits (FIPJF), c/o Schweiz
Obstverband, CH-6300 Zug 2, Switzerland.
Food and Drug Administration, 1998. Food labelling: warning and
notice statements; labelling of juice products. Federal Register, 63,
20486–20493.
Graumlich, T.R., Marcy, J.E., Adams, J.P., 1986. Aseptically pack-
aged orange juice and concentrate: a review of the influence of
processing and packaging conditions on quality. Journal of
Agricultural and Food Chemistry 34, 402–405.
Hill, E.F., Wenzel, F.W., 1957. The diacetyl test as an aid for quality
control of citrus product. I. Detection of bacterial growth in orange
juice during concentration. Food Technology 11, 240–243.
Jago, D., 2004. New food products around the world. Food &
Beverage International 3/2, 11–15.
Jime´nez, G., Dura´n, L., 1979. Propiedades reolo´ gicas de
productos derivados de frutas y de tomate. Revisio´ n bibliogra
´fica. Revista de Agroquimica y Tecnolog´ıa de los Alimentos 19/2,
203–216.
Johnston, C.S., Bowling, D.L., 2002. Stability of ascorbic acid in
commercially available orange juices. Journal of the American
Dietetic Association 102/4, 525–529.
Kaanane, A., Kane, D., Labuza, T.P., 1988. Time and temperature
effect on stability of Moroccan processed orange juice during
storage. Journal of Food Science 53, 1470–1473, and 1489.
Kimball, D., 1986. In: California Citrus Producers Inc. (Ed.), Citrus
Processing. Quality Control and Technology. Van Nostrand-
Reinold, New York.
Lo´pez, J.F., 1995. In: Conselleria de Agricultura y Medio
Ambiente (Ed.), La naranja, composicio´n y cualidades de
sus zumos y esencias. Generalitat Valenciana, Valencia, Espan˜ a.
Manso, M.C., Oliveira, F.A.R., Oliveira, J.C., Frias, J.M., 2001.
Modelling ascorbic acid thermal degradation and browning in
orange juice under aerobic conditions. International Journal of
Food Science & Technology 36/3, 303–312.
Mart´ın, J.J., Solanes, E., Bota, E., Sancho, J., 1995. Chemical and
organoleptic changes in pasteurised orange juice. Alimentaria 261,
59–63.
Mart´ınez, A., Rodrigo, C., Ferna´ndez, P.S., Ocio, M.J., Rodrigo, F.,
Rodrigo, M., 1999. Microbial and quality assessment in thermal
processing. In: Oliveira, F.A.R., Oliveira, J.C. (Eds.), Processing
Foods: Quality Optimization and Process Assessment. CRC Press,
New York.
Murdock, D.I., 1966. Diacetyl test as a quality control in processing
frozen concentrated orange juice. The Proceedings of the Florida
State Horticultural Society 79, 312.
Naim, M., Wainish, S., Zehavi, U., Peleg, H., Rouseff, H.P., Nagy, S.,
1993. Inhibition by thiol compounds of off-flavor formation in
stored orange juice. 1. Effect of L-cysteine and N-acetyl-L-cysteine
on 2,5-dimethyl-4-hydroxy-3(2h)-furanone formation. Journal of
Agricultural and Food Chemistry 41, 1355–1358.
Nienaber, U., Shellhamer, T.H., 2001. High pressure processing of
orange juice: combination treatments and a shelf life study. Journal
of Food Science 66, 332–336.
Oliveira, F.A.R., Oliveira, J.C., 1999. In: Oliveira, F.A.R., Oliveira,
J.C. (Eds.), Processing Foods: Quality Optimization and Process
Assessment. CRC Press, New York.
Palaniappan, S., Sastry, S.K., 1991. Electrical conductivity of selected
juices: influences of temperature, solids content, applied voltage and
particle size. Journal of Food Process Engineering 14, 247–260. Parish,
M.E., 1996. Incidence of coliforms, E. coli and Salmonella associated
with a citrus processing facility. IFT Annual Meeting
Book of Abstracts, 44.
Parish, M.E., 1998. Coliforms, E. coli and Salmonella serovars
associated with a citrus processing facility implicated in a salmo-
nellosis outbreak. Journal of Food Protection 61, 280–284.
Parish, M.E., Narciso, J.A., Friedrich, L.M., 1997. Survival of
Salmonellae in orange juice. Journal of Food Safety 17, 273–281.
Pino, J.A., Ranivoelisoa, V., Torrecilla, R., 1987. Influencia de la
temperatura de almacenamiento sobre la calidad del zumo de naranja
envasado en frascos de vidrio. Revista de Agroqu´ımica y
Tecnolog´ıa de los Alimenos 27, 261–269.
Race, S.W., 1991. Improved product quality through viscosity
measurement. Food Technology 45/7, 86–88.
Raso, J., A´ lvarez, I., Condo´ n, S., Sala, F.J., 1999. La conservacio´ n
de alimentos mediante pulsos ele´ctricos de alto voltaje.
Aspectos te´cnicos. Alimentacio´ n, Equipos y Tecnolog´ıa 8, 115–
124.
Redd, J.B., Hendrix, C.M., Hendrix, D.L., 1986. Quality Control
Manual for Citrus Processing Plants. Intercit, Inc., Florida, USA.
Rodrigo, D., Arranz, J.I., Koch, S., Fr´ıgola, A., Rodrigo, M.C.,
Esteve, M.J., Calvo, C., Rodrigo, M., 2003. Physicochemical
characteristics and quality of refrigerated Spanish orange–carrot
juices and influence of storage conditions. Journal of Food Science
68, 2111–2116.
Rouse, A.H., Atkins, C.D., 1955. Pectinesterase and pectin in
commercial juices as determined by methods used at the Citrus
Experiment Station. Agric. Exp. St. Univ. Fla., Bulletin no. 570.
Sa´nchez-Moreno, C., Plaza, L., De Ancos, B., Cano, M.P.,
2003. Vitamin C, provitamin A, and other carotenoids in high-
pressur- ized orange juice during refrigerated storage. Journal of
Agricul- tural and Food Chemistry 51, 647–653.
Scott, W.C., Veldhuis, M.K., 1966. Rapid estimation of recoverable oil
in citrus juices by bromate titration. Journal of AOAC Interna-
tional 49, 628–633.
Solomon, O., Svanberg, U., 1995. Effect of oxygen and fluorescent
light on the quality of orange juice during storage at 8 °C. Food
Chemistry 53/4, 363–368.
Souci, S.W., Fachmann, W., Kraut, H., 1987. Food Composition and
Nutrition Tables 1986/87. Wissenschaftliche Verlagsgesellschaft,
Stuttgart.
Trammell, D.J., Dalsis, D.E., Malone, C.T., 1986. Effect of oxygen on
taste, ascorbic acid loss and browning for HTST-pasteurised,
single-strength orange juice. Journal of Food Science 51, 1021–
1023.
Trifiro, A., Gherardi, S., Calza, M., 1995. Effetti della temperatura e
del tempo di magazzinaggio sulla qualita` di succhi freschi di
arance pigmentate. Industria conserve 70, 243–251.
USDA, 1985. Standards for Grade for Citrus Products.
USDA/AMS, Standardization Branch, USA.
Villamiel, M., Castillo, M.D., San Mart´ın, C., Corzo, N., 1998.
Assessment of the thermal treatment of orange juice during
continuous microwave and conventional heating. Journal of
Science Food and Agricultural 78, 196–200.
Williams, P.G., Ross, H., Brand, J.C., 1995. Ascorbic acid and 5-
methyltetrahydrofolate losses in vegetables with cook/chill or cook/
hot-hold foodservice systems. Journal of Food Science 60, 541–546.
Yeom, H.W., Streaker, C.B., Zhang, Q.H., Min, D.B., 2000. Effects of
pulsed electric fields on the activities of microorganisms and
pectinmethylesterase in orange juice. Journal of Food Science 65,
1359–1363.