JFS E: Food Engineering and Physical Properties

Comparisons of Chemical and Physical

Properties of Catfish Oils Prepared from
Different Extracting Processes
S. SATHIVEL, H. YIN, W. PRINYAWIWATKUL, AND J.M. KING

E: Food Engineering &

Physical Properties

ABSTRACT: Four different catfish oil extraction processes were used to extract oil from catfish viscera: process
CF1 involved a mixture of ground catfish viscera and water, no heat treatment, and centrifugation; process CF2
involved ground catfish viscera (no added water), heat treatment, and centrifugation; process CF3 involved a mixture of ground catfish viscera and water, heat treatment, and centrifugation; process CF4 involved ground catfish
viscera, enzymatic hydrolysis, and centrifugation. Chemical and physical properties of the resulting of catfish oils
were evaluated. The CF4 process recovered significantly higher amounts of crude oil from catfish viscera than the
other 3 extraction methods. The CF4 oil contained a higher percent of free fatty acid and peroxide values than CF1,
CF2, and CF3 oils. Oleic acid in catfish oil was the predominant fatty acid accounting for about 50% of total fatty
acids. Weight loss of oils increased with increasing temperatures between 250 and 500 C. All the catfish oil samples melted around 32 C regardless of the extraction methods. The flow behavior index of all the oil samples was
less than 1, which indicated that the catfish oils exhibited non-Newtonian fluid behavior. The apparent viscosity at
5 and 0 C was significantly higher (P < 0.05) than those at 5, 10, 15, 20, 25, and 30 C. The average magnitude of
activation energy for apparent viscosity of the oil was higher for CF2 than CF1, CF3, and CF4.
Keywords: catfish oil, extraction methods, rheological properties, thermal properties

Introduction

iscera from Channel catfish (Ictalurus punctatus) is an abundant and underutilized by-product that can be used as a
unique lipid source. Channel catfish is the 4th-most popular fish
product consumed in the United States. The 4 major commercial
catfish producing states in the United States are Alabama, Arkansas,
Louisiana, and Mississippi. In 2005, these states produced over
272000 metric tons of catfish with a stable monthly production of
about 22700 tons (NASS 2006). The by-products of catfish processing consists of heads, frames, skin, and viscera, which often ends up
in landfills or rendering plants. The yield of catfish when processed
as whole fillets is around 45%, generating about 55% waste. The average weight of viscera is about 265 g, which is about 10% by weight
of a live whole catfish. Much of the oil in the catfish is found in the
viscera, which contains approximately 33% lipid (Sathivel 2001).
The viscera can be used for recovering of catfish oil that could be
converted into edible oil or biodiesel products.
Extracting oil from viscera may add value to catfish viscera,
which is currently a processing waste. For the last 2 decades, interest in dietary effects of marine omega-3 fatty acids has increased
because they play a major role in human health (Kronhout and
others 1985). Natural fish oils have been claimed to help maintain heart and vascular health in humans (Haglund and others
1998). Producing and purifying catfish oil from catfish viscera for
the growing fish oil market can benefit the catfish industry. Small
fish oil processors and entrepreneurs are interested in establishing
small scale, cost effective oil extraction, clarification, and stabiliza-

tion methods for catfish oil destined for human consumption. Fish
oil can be extracted using a number of methods including rendering, enzymatic hydrolysis, chemical extraction, mechanical pressing, and use of centrifugal force. However, information related to
quality of fish oil prepared from the different extraction methods is
not readily available. The unpurified catfish oil contains free fatty
acids, primary oxidation products, minerals, pigments, moisture,
phospholipids, and insoluble impurities that reduce the oil quality. The amount of these impurities present in the oil may depend
on the fish oil extraction methods. The longer these components
remain in the oil, the greater their negative effects on oil quality.
Finding a suitable method to extract oil from catfish viscera may
reduce some of the impurities in the oil and thus may cut down
the multiple fish oil purification steps to 1 or 2 steps. The amount
of impurities present in the oil influences the rheological and thermal properties of fish oil (Sathivel and others 2008a, 2008b). Knowledge of thermal, rheological, and oxidation properties of the catfish
oil is essential for the design of purifying processes, the analysis of
production cost, and the final quality evaluation. The objective of
study was to evaluate the effects of different extraction processes
on chemical and physical properties of catfish oils.

Materials and Methods

Catfish oil production

Catfish viscera was obtained from a local seafood store in

Baton Rouge (La., U.S.A.), and stored at 40 C until further processed. The viscera was thawed overnight at 4 C and ground
MS 20080545 Submitted 6/21/2008, Accepted 11/14/2008. Authors are with using a Hobart grinder (K5SS, Hobart Corp., Troy, Ohio, U.S.A.)
the Dept. of Food Science, Louisiana State Univ. Agricultural Center, Baton through a 7-cm diameter plate having 12-mm diameter openings,
Rouge, LA 70803-4300, U.S.A. Direct inquiries to author Sathivel (E-mail:
and subsequently ground through a plate with 6-mm diameter
ssathivel@agcenter.lsu.edu).
openings. The ground viscera was used to produce catfish oil by the
E70

JOURNAL OF FOOD SCIENCEVol. 74, Nr. 2, 2009

R
Institute of Food Technologists
doi: 10.1111/j.1750-3841.2009.01050.x


C 2009

Further reproduction without permission is prohibited

Chemical and physical properties of catfish oils . . .

The crucible was dried at 101 C until constant weight was obtained. Insoluble impurities value in the oil was calculated as: insoluble impurities (%) = [gain in mass of crucible/mass of the residue]
100 triplicate determinations were performed.

Thermal properties of catfish oils

Thermal stability of the unpurified catfish oil was analyzed using the Thermogravimetric Analyzer (Model Q50, TA Instruments,
New Castle, Del., U.S.A.). Approximately 1 to 1.2 mg of the oil sample were added to an aluminum pan, the pan was placed in the
furnace, and the exact sample weight was determined. The sample
was heated to 700 C under air atmosphere at an increasing rate of
5 C/min. Sample weight differences were automatically recorded
every 0.5 s. Collected data were analyzed and plotted using the TA
Universal Analyzer Software (TA Instruments). The graph was normalized based on the sample weight basis.
The differential scanning calorimetric experiments were conducted using a differential scanning calorimetery (Model DSC 2920,
TA Instruments). Approximately 0.5 to 1 mg of the unpurified catfish oil sample was placed in an aluminum sample vessel. An empty
aluminum vessel was placed on the reference platform. To determine the phase transition of the unpurified catfish oil sample, a
linear heating rate of 5 C/min over a temperature range of 75 to
100 C was used. The thermogram peak was used to provide an estimate of enthalpy (H). The thermogram peak points of the repProximate composition of catfish viscera
resentative of the thermogram were used to determine the melting
Catfish viscera samples were analyzed in triplicate for moisture points.
and ash contents using the AOAC standard methods 930.15 and
942.05, respectively (AOAC 1999). Fat content was determined us- Rheological properties of catfish oils
ing dichloromethyl ether as solvent in an automated ASE-200 fat
Rheological properties of the unpurified catfish oil samples were
extractor (Dionex Corp., Sunnyvale, Calif., U.S.A.). Nitrogen content measured in triplicate using an AR 2000 Rheometer (TA Instruwas determined in triplicate using the Leco FP-2000 Nitrogen An- ments) fitted with a 2 cone-angle plate geometry (acrylic plates
alyzer (LECO Corp., Mich., U.S.A.). The protein content was calcu- with a 20-mm diameter, having a 200 m gap between the 2 plates).
lated as percent nitrogen times 6.25.
Each sample was placed in the temperature-controlled parallel
plate and allowed to equilibrate to 5, 0, 5, 10, 15, 20, 25, or 30 C.
Analysis of PV, FFA, and fatty acid
The shear stress was measured at 5, 0, 5, 10, 15, 20, 25, and 30 C
composition of catfish oil
at varying shear rates from 0 to 500/s. The mean values of triplicate
The peroxide value (PV) of the catfish oil samples was mea- samples were reported.
sured by titration according to AOAC (1999). Free fatty acids (FFA)
The power law (Eq. 1) was used to analyze the flow behavior inwere determined using the titration procedure according to AOCS dex of the unpurified catfish oil samples.
(1998). Fatty acid composition of catfish oils was determined at the
POS Pilot Plant Corp Lab., Saskatchewan, Canada. The fatty acid
= Kn
(1)
methyl esters (FAME) were prepared according to the AOAC procedure 969.33 (AOAC 1999) and fatty acid in the catfish oil samples where = shear stress (Pa.s), = shear rate (s1 ), K = consistency
was determined according to AOAC procedure 996.06 (AOAC 1999). index (Pa.sn ), and n = flow behavior index. The logarithms were
Triplicate determinations were performed.
taken on both sides of Eq. 1, and a plot of log against log was
constructed. The resulting straight line yielded the magnitude of
Analysis of mineral, tocopherols, insoluble
log K (that is, intercept) and n (that is, slope).
impurities, and moisture content of catfish oils
The effect of temperature on apparent viscosity was described
Mineral, tocopherols, insoluble impurities, and moisture con- through the Arrhenius relationship as described in Eq. 2 (Rao 1999).
tent of oil samples were determined at the POS Pilot Plant Corp
Lab. Mineral content of catfish oil samples was determined ack = Ae(E a /RT)
(2)
cording to AOCS Ca17-01 and AOCS Ca 20-99 (1998) and reported
as parts per million. Levels of tocopherols in the oil samples (al- where k is the reaction rate constant, A is the frequency factor, Ea is
pha, beta, gamma, and delta) were determined according to AOCS the activation energy (J/mol), R is the gas constant (8.314 J/mol/K),
Ce8-89 (1998) and reported as milligrams per gram of oil. Moisture and T is the temperature (K).
content of catfish oil samples was measured using the Karl Fisher
Apparent viscosity of the unpurified catfish oil samples was meatitration method (AOAC Method 984.20 (1999). Insoluble impuri- sured at 5, 0, 5, 10, 15, 20, 25, and 30 C at a shear rate of 500/s
ties of catfish oil samples were measured according to AOCS Ca- using the AR 2000 Rheometer. A plot of ln apparent viscosity com46 (1998). Moisture and volatile matters were removed from the oil pared with 1/T (that is, 1/absolute temperature) was constructed
sample according to AOCS Ca 2b-38 (1998). The oil residue was dis- for each catfish oil sample. The slope of the straight line, the intersolved in kerosene and filtered through a crucible. The remaining cept and the regression coefficient were calculated using the trend
kerosene in the crucible was removed by washing with petroleum. line of the plot. The magnitude of Ea was calculated as the slope of
Vol. 74, Nr. 2, 2009JOURNAL OF FOOD SCIENCE

E71

E: Food Engineering &

Physical Properties

following 4 different processing methods: (1) ground catfish viscera was added with distilled water (water: ground viscera, 1:1 v/w)
and homogenized in a Waring blender (Waring Products Div., New
Hartford, Conn., U.S.A.) for 2 min. The homogenized mixture was
centrifuged at 7520 g for 30 min at 23 C and catfish oil separated;
(2) ground catfish viscera (no added water) was heated at 75 C
for 30 min, centrifuged at 7520 g for 30 min at 23 C and the
catfish oil separated; (3) ground catfish viscera was added with distilled water (water: ground viscera, 1:1 v/w) and homogenized. The
homogenized mixture was heated at 75 C for 30 min, centrifuged at
7520 g for 30 min at 23 C and the catfish oil removed; (4) ground
catfish viscera was enzymatically hydrolyzed (no added water),
centrifuged, and the visceral oil separated. Hydrolysis conditions
were similar to those documented by Hoyle and Merritt (1994).
The alcalase enzyme with the activity of 2.4 AU/g (Anson units per
gram) (Novo Nordisk, Franklinton, N.C., U.S.A.) was added to the
ground catfish viscera at 0.5% w/w. The mixture was stirred for
75 min at 50 C and then the temperature was increased to above
85 C for 15 min to inactivate the enzyme. The hydrolysate was centrifuged at 7520 g for 30 min and the catfish oil removed. The resulting unrefined catfish oils were collected and stored at 70 C
until analyzed. The oil extracted from methods 1, 2, 3, and 4 was
referred to as CF1, CF2, CF3, and CF4, respectively, throughout this
study.

Chemical and physical properties of catfish oils . . .

the plot multiply by the gas constant, and A was an exponential of CF3, and CF4 were 13.6, 13.8, 14, and 13.4 (milligrams per gram of
the intercept.
oil), respectively. The predominance of omega-6 and omega-9 in
catfish oils may have been attributed to the feed fed to the catfish,
Statistical analysis
especially if it was made from soy products. Diet has a major efMeans and standard deviations of the data were reported. Anal- fect on FA composition of lipid (Satoh and others 1989). All the oil
ysis of variance (ANOVA) comparison was performed at the signif- samples had similar fatty acids composition, which demonstrated
icant level of P < 0.05 using SAS version 8.2 (SAS 2002). Tukeys that extraction procedures had no affect on fatty acid composition.
studentized range tests were performed to locate differences
Mineral, tocopherols, insoluble impurities,
among the different treatments.

and moisture content of catfish oils

Results and Discussion

E: Food Engineering &

Physical Properties

P, Na, Mg, Ca, and K were abundant minerals found in the

catfish oil samples (Table 3). Phosphorus levels ranged from
Yield, PV, FFA, and fatty acid composition
179 ppm in CF2 to 2.76 ppm in CF3. The phosphorus present in catof catfish fish oil
fish oil samples may be attributed to the phospholipids in the oil
The catfish viscera contained 60.9% 3.8% moisture, 10.2% sample and calciumphosphate complexes (Young 1986a). All oil
1.5% protein, 0.8% 0.01% ash, and 27.9% 3.8% fat. The fat content of catfish viscera from our study was higher than that reported
Table 2 --- Fatty acid composition of catfish oil (mg/g of oil)
by Belal and Assem (1995). CF4 process recovered a significantly from different extraction methods.a
higher amount of crude oil (14.3%) from catfish viscera than did
Fatty acids
CF1
CF2
CF3
CF4
CF1, CF2, and CF3 (Table 1). CF1 had lowest oil recovery (5.9%)
C14 (myristic)
7.6
7.6
7.8
7.7
while CF2 and CF3 had similar oil yield (9% to 10.8%).
0.3
0.3
0.3
0.3
The CF4 oil had higher percent FFA, while CF1 and CF3 had low- C14:1n5 (tetradecenoic)
C15 (pentadecanoic)
0.7
0.8
0.8
0.9
est percent FFA. Acceptable levels of FFA content in refined fish oil C16 (palmitic)
149.3
156.8
159.5
159.2
ranges between 1.8% and 3.5% (Young 1986a). All catfish oils had C16:1n7 (palmitoleic)
25.3
24.4
25.5
25.4
1.6
1.7
1.8
1.8
an acceptable level of FFA, except CF4, which had a slightly higher C17 (margaric)
57.5
62.3
60.3
60.8
percent FFA than recommended. Young (1986a) reported that ap- C18 (stearic)
C18:1n9 (oleic)
455.3
446.7
446.9
446.4
propriate fish oil processing conditions may reduce FFA up to 50%.
C18:1 (octadecenoic)
8.7
6.3
7.8
6.8
The PV value for the oil samples indicated the amount of primary C18:2n6 (linoleic)
139.9
134.6
138.9
137.6
oxidation products produced during the oil extraction. Enzymatic C18:2 (cla isomers)
0.3
0.4
0.3
0.4
2.9
3.0
2.7
3.1
extraction produced a significant amount of PV than that of the C18:3n6 (gamma-linolenic)
8.1
8.0
8.5
8.2
other extraction methods. The enzymatic extraction process (CF4) C18:3n3 (alpha-linolenic)
C18:4n3 (octadecatetraenoic)
0.4
0.4
0.4
0.4
recovered the highest amount of crude oil from catfish viscera com- C20 (arachidic)
1.7
1.8
1.8
1.8
pared to other extraction methods but the oil had higher FFA and C20:1n9 (eicosenoic)
19.4
19.3
19.1
18.9
9.6
9.3
9.3
9.2
PV values. CF1 process recovered a lower amount of crude oil from C20:2n6 (eicosadienoic)
8.0
7.9
7.5
7.7
catfish viscera but the oil had lower FFA (except for CF3) and PV val- C20:3n6 (homo-g-linolenic)
C20:4n6 (arachidonic)
3.9
4.4
3.4
3.7
ues than did other oils. This demonstrated that extraction methods
C20:3n3 (meads acid)
0.4
0.3
0.4
0.4
affected recovery of the crude oil from catfish viscera, and the FFA C20:4n3 (eicosatetraenoic)
0.5
0.5
0.5
0.5
and PV values of the oil.
C20:5n3 (eicosapentaenoic)
1.1
1.0
1.1
1.0
0.7
0.7
0.7
0.7
The fatty acid compositions of CF1, CF2, CF3, and CF4 are given C22 (behenic)
0.7
0.8
0.7
0.7
in Table 2. Saturated fatty acids from CF1, CF2, CF3, and CF4 oils C22:1n9 (erucic)
C22:2n6 (docosadienoic)
1.9
1.8
1.8
1.8
accounted for 219.3, 231.6, 232.7, and 232.8 (milligrams per gram C22:4n6 (docosatetraenoic)
0.9
0.7
0.7
0.8
of oil), respectively. Oleic acid was the predominant fatty acid in C22:5n6 (docosapentaenoic)
2.0
1.9
1.7
1.7
1.3
1.4
1.3
1.3
catfish oil accounting for about 50% of the total fatty acids. Total C22:5n3 (docosapentaenoic)
1.9
2.1
1.8
1.7
polyunsaturated fatty acid present in catfish oil was slightly lower C22:6n3 (docosahexaenoic)
C24:1n9 (nervonic)
0.3
0.3
0.3
0.3
than total saturated fatty acid and amounted to 183.1, 177.8, 180.4,
Others
16.1
16.3
15.6
15.4
and 179.4 (milligrams per gram of oil) for CF1, CF2, CF3, and CF4, Total fatty acids
928.7
923.7
929.4
926.5
respectively. Among unsaturated fatty acids, oleic acid and linoleic Saturates
219.3
231.6
232.7
232.8
510.2
498.0
500.6
498.9
acid were predominant fatty acids accounting for almost 86% of all Monounsaturates
183.1
177.8
180.4
179.4
unsaturated fatty acids. The CF1, CF2, CF3, and CF4 contained 1.9, Polyunsaturates
Omega 3
13.6
13.8
14.0
13.4
2.1, 1.8, and 1.7 mg of DHA/g of oil, respectively. All oil samples had Omega 6
169.1
163.6
166.1
165.6
slightly lower EPA content than DHA in the samples. The polyun- Omega 9
475.8
467.0
467.0
466.4
saturated fatty acids are considered to be of major importance in a
Values are means of triplicates determinations. See Table 1 for brief description
terms of human health. Total omega-3 fatty acids in the CF1, CF2, of CF1, CF2, CF3, and CF4.
Table 1 --- Yield, FFA, and PV of catfish oil from different extraction methods.a

Yield (%)
FFA (%)
PV (meq/kg oil)

CF1

CF2

CF3

CF4

5.9 1.3c
1.17 0.06c
4.9 0.46d

10.8 0.3b
2.8 0.04b
11.4 0.18b

9.0 1.0b
1.04 0.01c
7.2 0.5c

14.3 1.4a
3.55 0.09a
36.12 0.18a

Values are means SD of triplicate determinations. abcd Means with the same letters in each row are not significantly different (P > 0.05). CF1 = process involved
a mixture of ground catfish viscera and water, no heat treatment, and centrifugation; CF2 = process involved ground catfish viscera (no added water), heat
treatment, and centrifugation; CF3 = process involved a mixture of ground catfish viscera and water, heat treatment, and centrifugation; CF4 = process involved
ground catfish viscera, enzymatic hydrolysis, and centrifugation.

E72

JOURNAL OF FOOD SCIENCEVol. 74, Nr. 2, 2009

Table 3 --- Mineral content of catfish oil from different ex- samples had copper and iron contents below the acceptable maxitraction methods.a
mum level for copper of 0.23 ppm and iron of 8 ppm (Young 1986a;
Bimbo 1998). Metal, such as copper and iron, present in the oil catMineral (ppm)
CF1
CF2
CF3
CF4
Aluminum
<0.20
<0.20
<0.20
<0.20 alyzes oxidation (Young 1986b). CF2 and CF4 had higher iron and
Calcium
1.73
9.91
0.49
1.96 copper contents than other samples, which might be a reason that
Copper
<0.05
0.12
<0.05
0.05 the PV values of CF2 and CF4 was higher than other oil samples.
Iron
0.02
0.39
0.02
0.17
Lunde (1971) reported that unpurified edible oils are expected
Potassium
2.70
14.20
2.86
3.15
to contain a certain amount of minerals because phospholipids
Magnesium
1.37
8.26
0.66
1.69
Phosphorus
25.70
179.00
2.76
43.20 are reported to bind minerals in the oil. Minerals such as phosSodium
1.99
6.83
0.34
1.36 phorous, iron, magnesium, and calcium in the oil can be reduced
Zinc
0.02
0.18
<0.02
0.04 to trace levels by neutralization (Hvolby 1989). Neutralization may
Arsenic
<0.20
<0.20
<0.20
<0.20 also remove phospholipids during the washing step. All oil samples
Cadmium
0.03
0.06
<0.03
0.03
Mercury
<0.20
<0.20
<0.20
<0.20 had trace levels of mercury, lead, arsenic, and selenium content
Lead
<0.20
<0.20
<0.20
<0.20 (Table 3).
Silicon
<0.05
<0.05
<0.05
<0.05
Tocopherols play a role as a vitamin and natural antioxidant. The
Selenium
<0.50
<0.50
<0.50
<0.50 amounts of tocopherols in the oils are listed in Table 4. Total tocoa
Values are means of triplicates determinations. See Table 1 for brief description pherol content of the CF2 (0.14 mg/g) was slightly lower than CF1,
of CF1, CF2, CF3, and CF4.
CF3, and CF4 (0.20 - 0.21 mg/g). Insoluble impurities value of CF2,
CF1, CF3, and CF4 was <0.01%, 0.01%, 0.02%, and 0.01%, respectively. The moisture content was 0.26%, 0.42%, 0.43%, and 0.48%
Table 4 --- Tocopherols content of catfish oil (mg/g of oil) for CF4, CF1, CF3, and CF2, respectively.
from different extraction methods.a
Tocopherol (mg/g)
Alpha
Beta
Gamma
Delta
Total

CF1

CF2

CF3

0.18
<0.01
0.02
<0.01
0.20

0.11
<0.01
0.03
<0.01
0.14

0.17
<0.01
0.04
<0.01
0.21

CF4

Thermal properties

0.17
<0.01
0.04
<0.01
0.21

The thermogravimetric (TG) curves (Figure 1) show the thermal behaviors of catfish oils prepared from 4 extraction methods.
Between 250 and 500 C, weight loss of oils drastically increased
with increasing heating temperature, regardless of the extraction
methods. In an oxygen atmosphere, oil is oxidized by oxygen,
a
Values are means of triplicates determinations. See Table 1 for brief description
leading to a formation of oxidation products known as peroxides.
of CF1, CF2, CF3, and CF4.

120

100

Weight (%)

80

60

CF1
CF2
CF3
CF4

40

20

0
0

100

200

300

Temperature (C)

400

500

600
Universal V4.3A TA Instruments

Figure 1 --- Thermal degradation of catfish oils from different extraction methods. See Table 1 for brief description of
CF1, CF2, CF3, and CF4.
Vol. 74, Nr. 2, 2009JOURNAL OF FOOD SCIENCE

E73

E: Food Engineering &

Physical Properties

Chemical and physical properties of catfish oils . . .

Chemical and physical properties of catfish oils . . .

others 2008a), respectively. The negative melting points of catfish
oil may be attributed to triacylglycerol, which contained unsaturated fatty acids (Tan and Che Man 2002). The 1st peak at the
melting point of 32 C may be attributed to the presence of
polyunsaturated fatty acids (Sathivel 2001). Oil samples with a
higher degree of unsaturated fatty acids melt at lower temperatures,
whereas those with a higher degree of saturated fatty acids melt at
higher temperatures. The 2nd, 3rd, and 4th peaks appeared around
22 and 11 C; this may be attributed to the presence of linolenic
acid (C18:3) and linoleic acid (C18:2), whose melting points range
from 4 to 21 C (Sathivel 2001). The 3rd peak appeared closer
to 0 C, which might be attributed to the moisture present in the
sample (Berjak and others 1992; Connor and Bonner 2001). Physical properties of triglycerides are more complex than those of the
individual fatty acids because they contain 3 fatty acid groups. In
this study, we did not analyze the triglyceride composition of the
catfish oil. However, the trends in melting points observed for catfish oils would likely reflect its fatty acid composition. Fatty acids
found in catfish visceral oil were C14:0, C16:0, C16:1, C18:0, C18:0,
C18:1, C18:2, C18:3, C20:0, C20:1, C20:2, C20:4, and C22:6, and the
catfish visceral oil contained more than 77% of total unsaturated
fatty acids (Table 2). All 5 peaks for the unpurified catfish oils were
not sharp; this might be attributed to the presence of impurities,
such as phospholipids, ketones, and other materials in the unpurified fish oil (Sathivel 2005). A similar DSC thermogram was reported
for pollock oil by Sathivel and others (2008a). Those impurities melt
uncharacteristically compared to pure fatty acids; therefore, sharp
peaks were not observed for the oil. Sathivel (2001) reported that
the DSC thermograms of purified catfish oil showed sharper and
narrower peaks than unpurified fish oil.

0.5

0.0

Heat Flow (W/g)

E: Food Engineering &

Physical Properties

This reaction is usually identified by an increase in the initial sample mass (Hassel 1976). In this study, no weight gain was observed
in the TG curves for all catfish oils that were analyzed under air
atmosphere, indicating that thermal decomposition of catfish oils
was not related to oxygen absorption. The weight loss between 250
and 550 C was 88.8%, 87.3%, 86.4%, and 84.6% for CF3, CF1, CF4,
and CF2, respectively. Sathivel and others (2003a) reported that the
weight loss of fish oils due to thermal decomposition was higher in
refined oils than crude or unrefined oils. The higher thermal decomposition implies higher quality of oil. Unrefined oils contain
impurities such as phospholipids, complexed metals and minerals,
free fatty acids, and peroxides and their breakdown products, that
are highly interactive with the oil (Wiedermann 1981). This study
demonstrated the extraction process affected the amount of impurities and mineral contents in the oil samples. The presence of
impurities reduces effectiveness of heat transfer to unrefined oils,
and, therefore, resulting in less energy available to evaporate the
volatiles (Wesolowski and Erecinska 1998). Based on this study, the
TG analysis can be used to determine the quality of fish oils.
Figure 2 shows typical melting curves of CF1, CF2, CF3, and CF4
catfish oils. The melting points of catfish oil ranged from 32.7 to
9.9 C for CF1, 32.7 to 11.4 C for CF2, 33 to 11.6 C for CF3,
and 33.4 to 12 C for CF4. This indicated that extraction procedures affected melting points of crude oils from catfish viscera. Five
endothermic peaks, considered as melting points, were observed
in the DSC thermogram for all unpurified catfish oils (Figure 2).
The melting point of the unpurified catfish oil was relatively narrow
compared to the reported melting point range of 69.6 to 0.36 C,
64.7 to 20.8 C, and 64.5 to 14.86 C for red salmon and
pink salmon oils (Sathivel 2005), and pollock oil (Sathivel and

-0.5

-1.0

CF1.001
CF2.001
CF3.001
CF4.001

-1.5

-2.0
-50

-40

-30

-20

-10

Exo Up

Temperature (C)

10

20

30

40

50

Universal V4.3A TA Instruments

Figure 2 --- DSC thermogram of catfish oil from different extraction methods. See Table 1 for brief description of CF1,
CF2, CF3, and CF4.
E74

JOURNAL OF FOOD SCIENCEVol. 74, Nr. 2, 2009

Chemical and physical properties of catfish oils . . .

Table 5 --- Apparent viscosity (Pa.s) at different temperatures for catfish oils.
Temperature ( C)
5
0
5
10
15
20
25
30

CF1

CF2

CF3

CF4

0.64 0.02aD
0.35 0.05bD
0.16 0.00cD
0.12 0.00cdB
0.10 0.00deB
0.08 0.00deB
0.07 0.00deB
0.06 0.00eA

2.70 0.01aA
2.15 0.06bA
0.32 0.01cB
0.12 0.00dB
0.10 0.00dB
0.08 0.00dB
0.07 0.00dB
0.06 0.00dA

1.55 0.01aB
1.09 0.02bB
0.78 0.01cA
0.23 0.06dA
0.14 0.00eA
0.11 0.00efA
0.08 0.00efA
0.06 0.00fA

0.93 0.07aC
0.58 0.01bC
0.20 0.02cC
0.12 0.00dB
0.09 0.00dC
0.08 0.00dC
0.07 0.00dC
0.06 0.00dA

Values are means SD of triplicate determinations. ABCD Means with the same letters in each row are not significantly different (P > 0.05). abcdef Means with the
same letters in each column are not significantly different (P > 0.05). See Table 1 for brief description of CF1, CF2, CF3, and CF4.

Temperature ( C)
5
0
5
10
15
20
25
30

CF1

CF2

CF3

CF4

0.84 0.01abA
0.79 0.01bcAB
0.75 0.07cA
0.90 0.01aA
0.89 0.01aA
0.87 0.00aA
0.86 0.00abA
0.84 0.01abAB

0.71 0.01bcB
0.61 0.12cB
0.78 0.13abcA
0.90 0.00aA
0.89 0.00abA
0.88 0.00abA
0.85 0.01abA
0.84 0.00abAB

0.80 0.04abAB
0.74 0.07abAB
0.58 0.23bA
0.81 0.08abA
0.83 0.01abB
0.78 0.01abB
0.80 0.01abB
0.85 0.00aA

0.73 0.03bB
0.80 0.01abA
0.80 0.15abA
0.91 0.01aA
0.89 0.00aA
0.87 0.00abA
0.85 0.01abA
0.83 0.00abB

Values are means SD of triplicate determinations. AB Means with the same letters in each row are not significantly different (P > 0.05). abc Means with the same
letters in each column are not significantly different (P > 0.05). See Table 1 for brief description of CF1, CF2, CF3, and CF4.

Table 7 --- Consistency index (K) at different temperatures for catfish oils.
Temperature ( C)
5
0
5
10
15
20
25
30

CF1

CF2

CF3

1.44 0.04aC
0.92 0.06bC
0.68 0.01cB
0.20 0.01dA
0.17 0.01dB
0.15 0.00dB
0.14 0.00dB
0.13 0.00dAB

12.35 0.76aA
7.80 0.04bA
1.43 0.43cB
0.21 0.00dA
0.18 0.00dB
0.15 0.00dB
0.15 0.01dB
0.13 0.00dAB

4.39 0.91aB
4.42 1.85aAB
3.70 1.06aA
0.68 0.49bA
0.34 0.01bA
0.35 0.01bA
0.23 0.02bA
0.13 0.00bA

CF4
3.67 0.42aB
1.58 0.05bBC
0.62 0.36cB
0.19 0.01cA
0.16 0.00cB
0.15 0.00cB
0.14 0.00cB
0.13 0.00cB

Values are means SD of triplicate determinations. ABC Means with the same letters in each row are not significantly different (P > 0.05). abcd Means with the same
letters in column row are not significantly different (P > 0.05). See Table 1 for brief description of CF1, CF2, CF3, and CF4.

Table 8 --- Activation energy (Ea) and frequency factor ( )

The power law parameters for the catfish oils at 5, 0, 5, 10, 15,
for different catfish oil extraction methods.
20, 25, and 30 C are given in Table 6 and 7. The flow behavior index
(n) of the catfish oil samples ranged from 0.58 to 0.91. The flow beSample

Ea (J/mol)
havior index (n) of the oil samples was lower than 1 indicating that
CF1
5.4E-09 3.8E-10a
45955.64 29.39d
CF2
8.1E-16 6.1E-17c
78869.38 187.01a the oils were non-Newtonian fluids. The n-value for all the samCF3
1.1E-13 2.9E-15c
67587.28 97.1b
ples at 5 to 5 C behaved as pseudoplastic fluids regardless of the
CF4
2.1E-11 7.5E-12b
53941.23 928.7c
catfish oil extraction methods (Table 5). The consistency index (K )
value for the catfish oils was higher at lower temperatures. CF2 had
Values are means SD of triplicate determinations. abc Means with the same
letters in each column are not significantly different (P > 0.05). See Table 1 for
higher K values at 5 and 0 C than the CF1, CF3, and CF4.
brief description of CF1, CF2, CF3, and CF4.
The Arrhenius equation (Eq. 2) was employed to calculate the
average magnitude of activation energy of the catfish oil. CF2 had a
higher magnitude of Ea than that CF1, CF3, and CF4 (Table 8). The
Rheological properties
Ea indicates the energy barrier that must be overcome before the
Apparent viscosity for all of catfish oils increased significantly elementary flow process can occur (Rao 1999). This study showed
(P < 0.05) with decreased temperature (Table 5). CF2 had higher that the magnitude of the apparent viscosity of the catfish oil was
apparent viscosity (P < 0.05) at 5 to 0 C than CF1, CF3, and greatly influenced by temperatures, regardless of the oil extraction
CF4 samples. Wiedermann (1981) reported that unpurified fish oil methods.
contained a large amount of impurities including Mg, Ca, Fe, P,
free fatty acids, and peroxides, insoluble impurities, and moisture,
Conclusions
which were present in catfish oil; these impurities are highly inhis
study
demonstrated
the effect of the extraction methteractive with the oil, while Sathivel and others (2003b) reported
ods on chemical, nutritional, thermal, and rheological
changes of catfish oil rheological properties during the purification processes. The interface between impurities and oil may be properties of catfish oil, which is useful for designing the
attributed to the development of an aggregated colloidal dispersion purification process of the oil. The CF4 oil had higher percent
system, which might be a reason for different apparent viscosity ex- FFA and PV value than other oil samples. All oil samples had
slightly lower EPA content than DHA. CF2 and CF4 had higher iron
hibited in the oil samples at lower temperatures.

Vol. 74, Nr. 2, 2009JOURNAL OF FOOD SCIENCE

E75

E: Food Engineering &

Physical Properties

Table 6 --- Flow behavior index (n) at different temperatures for catfish oils.

Chemical and physical properties of catfish oils . . .

and copper contents. All oil samples had slightly different thermal
properties. The catfish oils exhibited non-Newtonian fluid behavior at lower temperature regardless of extraction methods. The
apparent viscosity of all oil samples was significantly increased
with decreased temperature. Information on thermal, rheological,
and oxidation properties from this study can be used for the design
of a purification process to produce catfish oil destined for human
consumption.

References

E: Food Engineering &

Physical Properties

AOAC. 1999. Official methods of analysis. 16th ed. Arlington, Va.: Assn. of Official Analytical Chemists.
AOCS. 1998. Official methods and recommended practices of the American Oil
Chemists Society. Champaign, Ill.: American Oil Chemists Society.
Belal IEH, Assem H. 1995. Substitution of soybean meal and oil for fishmeal in practical diets fed to channel catfish Ictalurus punctatus (Raflnesque): effects on body
composition. Aquac Res 26:1415.
Berjak P, Pammenter NW, Vertucci C. 1992. Homoiohydrous (recalcitrant) seeds: development status, desiccation sensitivity and the state of water in axes of Landolphia kirkii dyer. Planta 186:24661.
Bimbo AP. 1998. Guidelines for characterizing food-grade fish oil. INFORM 9:47383.
Connor K, Bonner FT. 2001. The effects of desiccation on seeds of acer saccharinum
and Aesculus pavia: recalcitrance in temperate tree seeds. Tress 15:1316.
Haglund O, Wallin R, Wretling S, Hultberg B, Saldeen T. 1998. Effects of fish oil alone
and combined with long chain (n-6) fatty acids on some coronary risk factors in
male subjects. J Nutr Biochem 9:62935.
Hassel RL. 1976. Thermal analysis: an alternative method of measuring oil stability.
J Am Oil Chem Soc 53:17981.
Hoyle NT, Merritt JH. 1994. Quality of fish protein hydrolysates from herring (Clupea
harengus). J Food Sci 59:769.
Hvolby A. 1989. Removal of nonhydratable phospholipids from soybean oil. J Am Oil
Chem Soc 48:5039.
Kronhout D, Bosschieter EB, Coulander C. 1985. The inverse relation between fish
consumption and 20-year mortality from coronary heart disease. N Engl J Med
312:12059.

E76

JOURNAL OF FOOD SCIENCEVol. 74, Nr. 2, 2009

Lunde G. 1971. Activation analysis of trace elements in lipids with emphasis on marine oils. J Am Oil Chem Soc 48:51722.
NASS. 2006. National Agricultural Statistics Service. Catfish processing. Released
August 21, 2006.
Rao MA. 1999. Rheological of fluids and semisolids. Principal and applications.
Gaitherburg, Md.: Aspen Publishers Inc.
SAS. 2002. SAS/STAT users guide, version 8.2. Cary, N.C.: SAS Inst.
Sathivel S. 2001. Production, process design and quality characterization of catfish
visceral oil [PhD diss]. Louisiana State Univ., Baton Rouge, La..
Sathivel S. 2005. Thermal and flow properties of oils from salmon head. J Am Oil Chem
Soc 82:14751.
Sathivel S, Prinyawiwatkul W, Negulescu II, King JM, Basnayake BFA. 2003a. Thermal
degradation of fatty acids and catfish and menhaden oils at different purification
steps. J Am Oil Chem Soc 80:11314.
Sathivel S, Prinyawiwatkul W, Negulescu II, King JM, Basnayake BFA. 2003b. Effects of
purification process on the rheological properties of catfish oil. J Am Oil Chem Soc
80:82932.
Sathivel S, Huang J, Prinyawiwatkul W. 2008a. Thermal properties and applications
of the Arrhenius equation for evaluating viscosity and oxidation rates of unrefined
pollock oil. J Food Eng 84:18793.
Sathivel S, Prinyawiwatkul W, Negulescu II, King JM. 2008b. Determination of melting
points, specific heat capacity and enthalpy of catfish oil during different purification steps. J Am Oil Chem Soc 85:2916.
Satoh S, Poe WE, Wilson RP. 1989. Effect of dietary n-3 fatty acids on weight gain and
liver polar fatty acid composition of fingerling channel catfish. J Nutr 119:23.
Tan CP, Che Man YB. 2002. Differential scanning calorimetric analysis of palm oil,
palm oil based products and coconut oil: effects of scanning rate variation. Food
Chem 76:89102.
Wesolowski M, Erecinska J. 1998. Thermal analysis in quality assessment of rapeseed
oils. Thermochim Acta 323:13743.
Wiedermann LH. 1981. Degumming, refining and beaching soybean oil. J Am Oil
Chem Soc 15966.
Young FVK. 1986a. The Chemical and physical properties of crude fish oils for refiners
and hydrogenators. In: Fish Oil Bulletin No. 18. Herefordshire, U.K.: Fish oil International Association of Fish Meal Manufactures.
Young FVK. 1986b. The refining and hydrogenation of fish oil. In: Fish Oil Bulletin
No. 17, p 127. Herefordshire, U.K.: Fish oil International Association of Fish Meal
Manufacture.