Introduction
Deep-fat frying is a commonly used practice in the preparation and manufacture of foods. In convenience and restaurant derived foods, such as chicken, French fries and potato chips, deep fat-frying is commonly used. It has been estimated that more than 230 tons of edible fats and oils are used annually for the manufacture of potato chips in the United States alone (Chang et al., 1978).
In commercial deep-fat frying, the oil is continuously exposed to air and light for extended periods of temperatures approaching 180°C (Augustin and Berry, 1983). Under these conditions, both thermal and oxidative decompositions of oil occur (Yang et al., 1983). It is well established that these reactions result in rapid deteriorations of the frying oil which ultimately effect significantly the functional, sensory and nutritive values, as well as safety, of the derived fried food product (Che Man and Irwandi, 2000). Numerous studies have shown that lipid oxidation products not only influence the quality and safety of the oil (Izaki et al., 1984), but also affect the acceptability of the fried product (Jacobson 1991). In many of these studies, relationships between sensory and chemical parameters of frying oils have been reported (Idris et al., 1992; Augustin et al., 1987). There are however, very few studies that have examined the specific correlations between physicochemical analyses of a frying oil and the related sensory properties of the fried product. Idris et al. (1992), reported some correlations between overall quality ratings of palm olein with chemical changes of the oil that involved free fatty acids, peroxide value and anisidine value. Other workers have shown that frying oils with low quality will produce lower acceptability scores for sensory properties of deep fat-fried processed foods (Singh, 1995).
This study reports the relationships between 5 organoleptic characteristics of potato chips that were deep-fat fried for 5 consecutive days in refined, bleached and deodorized (RBD) palm olein with 11 specific indices of oil deterioration attributed to deep-fat frying processing.
Materials and Methods
Materials
RBD palm olein used in this study was obtained from a local refinery. Oleoresin
rosemary (OR; Herbalox Brand, Type 0) and sage (OS; Herbalox seasoning,
Type S-0) extracts, were donated by Kalsec Inc., USA (Gulf Chemical Sdn. Bhd.,
Selangor, Malaysia). Citric acid (CA, food grade) was supplied by Bumi Sains
Sdn.Bhd, Selangor, Malaysia. Fresh potatoes of the same variety (Russed var.)
used throughout the experiment and sodium chloride were purchased from a local
supermarket. All reagents were of analytical grade and supplied by local suppliers.
Experimental Design
To obtain data of frying oil with wide range of values used for calculation
of linear regression equations, the frying procedure was set using 15 different
combinations of natural antioxidants (Table 1). The antioxidants
were added into the RBD palm olein prior to frying at day 1 (Irwandi et al.,
2000a).
Frying Experiment
Fresh potatoes were peeled by hand and sliced to a thickness of 1.5 mm before
frying. The sliced potatoes were then soaked in a 2.5% NaCl solution for 5 min,
filtered and dried using paper napkins before frying. OR, OS and CA were added
into RBD palm olein just before frying. The addition of the antioxidants was
performed once at the beginning of the study (Che Man et al., 2000; Irwandi
et al., 2000b; Irwandi et al., 2000c).
Frying experiments were performed in two replicates, similar to that reported
by Che Man and Irwandi (2000). Frying was conducted in batch fryers (Berto’s,
Model ELT 8B, Italy), with 3 kg of oil heated up to 60°C. The oil was then
heated up to 180±5°C in 10 min. Frying was started 30 min after the
temperature of the oil reached 180°C.
| Table 1: |
Combinations of Oleoresin Rosemary Extract, Sage Extract and
Citric Acid Added into RBD1 Palm Olein before Frying |
 |
| 1RBD=Refined, bleached and deodorized |
One hundred grams of fresh sliced potato were fried for 2.5 min and then the
oil temperature was returned to 180°C within 30 min. There were 10 frying
conducted every day for 5 consecutive days. No fresh oil was added during the
frying. The fryers were left uncovered during the frying operations. For the
physico-chemical analyses, 200 g oil at temperature 60°C was removed from
the fryer at the end of each day, flushed with nitrogen gas and kept in a cold
room at 5°C. Fryer lids were then replaced and the fryer was left overnight
for continuation of the experiment the following day.
After frying, the potato chips were removed from the fryer and surplus oil was drained. Both the ninth and tenth batches of potato chips were taken each day, cooled, labeled and packaged in low-density polyethylene plastic bags for sensory evaluation. The evaluation was conducted on the same day the frying was conducted.
Sensory Evaluation
Hedonic sensory attributes evaluated in this study included acceptability
in appearance, taste, odor, crispiness and overall acceptance of the potato
chips (Larmond, 1977). All these attributes were evaluated using a 7-point scale
(1=dislike extremely, 4=moderate, 7=like extremely) by 20 experienced panelists
selected from staff and graduate students of the Faculty of Food Science and
Biotechnology, Universiti Putra Malaysia. All panelists regularly participated
in sensory evaluation and are also regular consumers of potato chips. Sensory
evaluation procedures were explained to the panelists before testing commenced.
Panelists were asked to read through the instructions and the questions on the
sensory form and a consensus understanding of each descriptor chosen was ensured
to avoid misinterpretation. All panelists were given time to ask for clarification
of the sensory evaluation procedure when uncertain or unclear about the process.
All analysis was conducted in triplicates.
Analyses of Oil
Peroxide value (PV), free fatty acid (FFA) and iodine value (IV) were all
determined using PORIM test methods (PORIM 1995). The oil color was measured
in a one-inch cell in a Lovibond Tintometer (Salibury, United Kingdom) (PORIM
1995). Viscosity was monitored by using a Brookfield viscometer (Stoughton,
MA) (PORIM 1995). The oil polymer content was analyzed according to the method
of Peled et al. (1975). The absorbances at 232 and 268 mm and the anisidine
value (AnV) were obtained using IUPAC methods (IUPAC 1979). The fatty acid profile
of the oil was determined by gas chromatography (Hewlett Packard gas chromatography
Type 5890) as reported by Berry (1980) using a 15m X 0.53 mm capillary column
and a flame ionization detector. The temperature of the column was 140°C,
programmed to increase at 4°C/min to 200°C. The temperature of both
the injector and detector was 250°C. Flow rates for carrier gas nitrogen,
hydrogen and air were 65 mL min-1, 44 mL min-1 and 440
mL min-1, respectively. Each reported value represents the mean of
three replicates.
Statistical analyses
Data from physico-chemical analyses of oil and sensory evaluation of potato
chips were statistically analyzed by one-way analysis of variance procedure
using a SAS (1989) software package to determine the effect of frying time on
the oil and the fried product. Significant differences (P<0.05) between means
were further determined by Duncan’s multiple-range test. In addition, the
SAS programme was also used to determine linear regression relationship between
the oil data and sensory responses of potato chips.
Results and Discussion
The changes in panelist sensory scores of fried potato chips (Table
2) and the physico-chemical properties of RBD palm olein (Table
3), during 5 days of deep-fat frying have been reported previously from
studies that examined the effects of natural antioxidants on sensory characteristics
of fried potato chips (Irwandi et al., 1999) and oil composition (1Irwandi
et al., 2000a; Irwandi et al., 2000d). What was not described
or discussed from our former studies was the correlation between the inter-related
parameters of oil quality and the acceptability of the processed product. For
example, all sensory scores derived from descriptive analysis of potato chip
appearance after deep-fried potato chips decreased significantly (P<0.05)
from day 1 to day 5 of frying. The lack of significant differences in appearance
ratings of fried potato chips over time of use of the frying oil, could be explained
by the choice of the preference test used in this study. For example, in samples
where fried products possessed darker colors as a result of exposure to oils
for 5 days of frying, higher scores reflected an actual preference by some panelists,
while others showed a different preference for lighter colour product which
originated from samples exposed to a shorter period (e.g. one day) of deep-fat
frying.
During the 5 days of frying, all chemical quality parameters derived from the
RBD palm olein decreased (Table 3). Notable increases in PV,
AnV, FFA, oil colour (e.g. redness and yellowness), polymer content, viscosity
and the C18:2/C16:0 ratio were observed.
| Table 2: |
Effect of Oleoresin Rosemary Extracts, Sage Extracts and
Citric Acid on Sensory Scoresa of Potato Chips |
 |
| a Mean of 20 Panelists |
| Table 3: |
Physico-Chemical Changes of RBD Palm Olein during Deep-Fat
Frying |
 |
|
Similarly, estimates of the degree of primary oxidation (e.g. absorbance readings
at 323 nm) and the presence of diethylenic ketones (e.g. absorbance readings
at 268nm), also indicated that the longer frying time resulted in marked changes
in lipid oxidation. A decrease in IV also provided a good estimate of the gradual
lipid deterioration rate of the frying oil over the 5 day period examined.
Relationship between Sensory Characteristics of Fried Potato Chip and Quality
Parameters of Frying Oil
Assessment of sensory characteristics of processed potato chips and associated
changes in physico-chemical parameters of RBD palm olein during frying was performed
on a data collected from 15 different samples over a 5 day period of frying
and analyzed by linear regression analysis. Higher probabilities of significant
correlations between sensory scores and physico-chemical changes of the frying
oil were found in samples processed with 5 day-old frying oil, compared to relatively
fresh (e.g. 1 day old frying) oil. For example, the hedonic scores for overall
acceptability collected from samples processed in the oil used for 5 days of
frying , had a lower range of values (e.g. 3.8-5.6) than samples fried in the
oil used for only one day (e.g. 4.3-5.8). Similar differences in the range of
C18:2/C16:0 ratio values in the cooking oil, after 5 days of frying (range=
0.135-0.183), versus only 1 day of frying (range= 0.190-0.217) were observed.
Tables 4-8 show the different correlation coefficients, associated P-values and related equations for predicted relationships between sensory response of potato chips and quality parameters of the RBC palm olein. Linear regression equations are only presented for the significant (P<0.05) relationships. In general, highly significant (P<0.01) linear relationships were obtained between the appearance of fried samples and physico-chemical quality parameters for the RBD palm olein after day 5 of frying (Table 4). Of interest was the finding that the correlation between potato chip appearance and viscosity changes of the frying oil were only significantly correlated (P<0.05) when data in individual frying days 1 and 5 were assessed separately. In contrast, no correlations were obtained when the data was combined for all days of frying, thus signifying that correlations for changes in potato chip appearance and frying oil viscosity were both dependent on a critical number of frying procedures for the oil. This explanation identifies the importance of thermal oxidation reactions that carry on sufficiently long enough to produce polymerization products that contribute to both increased viscosity of the frying oil and color changes of the fried product.
The significant correlations obtained between potato chip taste and the quality
indices of the RBD palm olein used in frying (Table 5), support
the findings of Idris et al. (1992), who also reported that flavor intensity
scores of RBD palm olein were closely (r= -0.92) related to oil quality indices
comprising of FFA, PV and AnV. In our study, the only exception to the significant
correlations found between taste of fried potato chip and associated estimates
oil composition changes occurred with the C18:2/C16:0 parameter obtained from
oil that was used for only 1 day of frying. Deterioration of oils, common to
the process of deep-fat frying, results in many thermally oxidative reaction
products which in turn contribute to the flavor of the oil (Idris et al.,
1992; Augustin et al., 1987). However, the findings of this study indicate
that this is not necessarily an issue with oil that has undergone only 1 day
of frying. Thus nutritional implications of deep-fat frying (e.g. stability
of essential fatty acids such as C18:2) are not a concern with oil that has
undergone thermal treatment for a relatively short period of time (e.g. 1 day).
| Table 4: |
Relationship between Appearance of Potato Chips and Quality
Parameters of Palm Olein |
 |
| *** highly significant (P<0.001); ** highly significant
(P<0.01); * significant (P<0.05) |
This observation does not hold true for oil that was used for 5 frying periods.
Since the final fat content of the potato chip approached 40% after frying (Guillaumin,
1988), changes in fatty acid content of the frying oil related to peroxidation
and degradation reactions would contribute to the reduced taste acceptance scores
of the processed potato chips.
Potato chip odor quality parameters were significantly (P<0.05) correlated
with chemical changes in the frying oil for both short (1 day) and long (5 days)
deep-fat frying periods. No correlations were obtained after only 1 day of frying
(Table 6). Unlike the taste perception scores, these results
indicate that oil deterioration of the frying oil was correlated less with odor
perceptions of acceptance. Odors are primarily attributed to volatile substances
which are perceived quickly by the olfactory sensory system.
| Table 5: |
Relationship between Taste of Potato Chips and Quality Parameters
of Palm Olein |
 |
| *** highly significant (P<0.001); ** highly significant
(P<0.01); * significant (P<0.05) |
Thermal treatment of oil resulting in the production of volatile oxidation
products such as ketone, aldehydes or alcohols would be expected to be greater
after 5 frying periods than after …
Significant (P<0.05) relationships were also obtained between texture parameters
(e.g. crispiness scores of potato chips) and all of the different physico-chemical
parameters of the RBD quality (Table 7). After 1 day of frying,
potato chip crispiness was significantly (P<0.05) correlated with peroxide,
anisidine and iodine values, as well as absorbance measurements taken at both
232 and 268 nm on days 1 and 5 of frying. Oil polymer content and viscosity
character of the oil after both frequencies of frying were also correlated (P<0.05)
with fried potato chip texture measurement (Table 7). These
results indicate that textural changes in fried potato chips are more closely
related to all parameters of oxidation and degradation of the cooking oil than
the perception of taste and odor changes.
| Table 6: |
Relationship between Odor of Potato Chips and Quality Parameters
of Palm Olein |
 |
| *** highly significant (P<0.001); ** highly significant
(P<0.01); * significant (P<0.05) |
As was the case with both the observed taste and odor qualities of the fried
potato chips, indices of nutritional quality of the oil, such as the C18:2/C16:0
ratio, was found to correlate only to the textural changes of crispiness after
day 5 of frying. Thus, a relationship between the nutritional quality of the
cooking oil and textural changes of potato chips could only be predicted after
repeat fryings of the oil.
Finally, the significant (P<0.05) relationships noted between the overall
acceptability of potato chips and the physico-chemical characteristics of RBD
palm olein during frying is given in Table 8. It is clear,
that with exception to the yellow colour parameter of the oil, all RBD quality
parameters were significantly (P<0.05) related to the overall hedonic acceptability
scores for fried potato chips.
| Table 7: |
Relationship between Crispiness of Potato Chips and Quality
Parameters of Palm Olein |
 |
| *** highly significant (P<0.001); ** highly significant
(P<0.01); * significant (P<0.05) |
After only 1 day of frying, a linear correlation (P<0.05) between overall
acceptability of fried product and most physico-chemical parameters, with exception
to the C18:2/C16:0 ratio, was observed. This result confirms our other findings
that nutritional quality of the oil, based on the essential fatty acid composition
and sensory parameters are not directly related early on in the frying process
of potatoes. A possible explanation for this observation , is the fact that
potatoes represent a relatively low fat containing food and thus very little
endogenous lipid would be released from the potatoes into the frying fat. As
a consequence to this, endogenous lipids of the potatoes would have contributed
little to the fatty acid composition of the RBD after only 1 day of frying.
Changes in the essential fatty acid ratio after 5 days of frying therefore represent
thermal oxidation and decomposition of the cooking oil and thus loss of nutritional
value that corresponded to similar losses in hedonic scores.
| Table 8: |
Relationship between Overall Acceptability of Potato Chips
and Quality Parameters of Palm Olein |
 |
| *** highly significant (P<0.001); ** highly significant
(P<0.01); * significant (P<0.05) |
This explanation may not hold true if for example the food source that was
been fried was relatively rich in fat (e.g. sardines) and would have contributed
endogenous lipid to the cooking oil. The significant correlations obtained between
peroxide value (inverse linear {r= -0.69; P=0.001}) and absorbance measures
at 268 nm (inverse linear {r=-0.67; P=0.0001}) with hedonic acceptance score,
reflects changes in both primary and secondary stages of lipid oxidation, respectively.
In conclusion, results from this study indicate that significant correlations between 5 sensory attributes of potato chips and 11 physico-chemical quality parameters of the RBD palm olein existed with prolonged deep-fat frying. The data collected from day 5 of frying produced significant results for correlations between appearance, taste, odour, texture and overall acceptance for the processed potato chips and the physico-chemical quality estimates of the heat treated RBD palm olein. Lipid thermo-oxidative decomposition products, as indicated especially by peroxide value and absorbance changes at 268, were found to be good predictors of the acceptability of potato chips fried in this oil. The nutritional quality of the frying oil and the sensory scores of the processed potato chips were not related with 1 day frying oil, but did show a significant relationship with 5 day frying oil. We conclude from this observation, that changes in essential fatty acid composition of the thermally processed RBD palm olein are related closely to both oxidation and degradation of the oil and that these associated changes corresponded to reduce organoleptic quality of the fried product.
Acknowledgements
This study was a collaboration between IRPA Project No. 03-02-04-003 at Universiti Putra Malaysia and the Natural Sciences and Engineering Council of Canada at University of British Columbia. The authors also thank the Research Centre of the International Islamic University Malaysia for their support in producing this manuscript.
Subscribe Today
