Journal of Food Engineering 62 (2004) 143–150

www.elsevier.com/locate/jfoodeng

Effect of extrusion cooking of soy–sweet potato mixtures on available

lysine content and browning index of extrudates
a,*
M.O. Iwe , D.J. van Zuilichem b, W. Stolp b, P.O. Ngoddy c

a
Department of Food Science and Technology, University of Agriculture, P.M.B. 2373, Makurdi, Nigeria
b
Food & Bioprocess Engineering, Department of Food Science, Agricultural University, The Netherlands
c
Department of Food Science and Technology, University of Nigeria, Nsukka, Nigeria
Received 1 June 2000; received in revised form 12 September 2001; accepted 9 May 2003

Abstract
Effects of three processing variables: feed composition (% sweet potato), screw speed, and die diameter on available lysine and
browning index were investigated following extrusion cooking of mixtures of defatted soy flour and sweet potato flour. Response
surfaces for the parameters were generated using a second degree polynomial. Results show that increase in screw speed and a
reduction in die diameter enhanced lysine retention. Increase in feed composition, and screw speed increased browning index, but
decreases in die diameter and feed composition increased browning index. Optimum extrusion conditions resulting in maximum
available lysine and minimum browning index were estimated.
Ó 2003 Elsevier Ltd. All rights reserved.

Keywords: Extrusion; Soy–sweet potato; Lysine; Browning index

1. Introduction (1992) reported that discoloration could be a serious

problem during processing of sweet potatoes and stor-
Extrusion cooking, as a heat treatment affects and age of their products. Discoloration occurs as a result of
alters the nature of many food constituents, including surface darkening due to oxidation. Bouwkamp (1985)
starches and proteins, by changing physical, chemical reported that the darkening was due to polyphenol oxi-
and nutritional properties (Harper, 1979; Sgaramella & dase enzyme substrate complex, which is concentrated
Ames, 1993). in the cambial areas of the root. It was further proposed
Several reports had attempted to relate Maillard re- (Bouwkamp, 1985), that the darkening reaction is a two
action and discoloration or browning to loss of lysine step process, the first being the enzymatic oxidation of
(Asp & Bjorck, 1989; Bjorck & Asp, 1983; Cheftel, 1986; phenolic compounds to quinones. The second is the
Hurrel & Carpenter, 1977; Noguchi, Mooso, Aymard, nonenzymatic polymerization of quinones to melanin-
Jeunink, & Cheftel, 1982; OÕBrien & Morrissey, 1989). like compounds. The effect of processing leading to the
Furthermore, lysine loss has been related to extrusion loss of lysine and development of darkening and for-
process parameters such as raw material, feed moisture, mation of melanoidins seem therefore to be relevant in
screw speed, extrusion temperature, die diameter, feed sweet potato processing.
rate, screw compression ratio, torque and pressure, Extrusion of a soy–sweet potato system might favor
energy input and pH (Asp & Bjorck, 1989; Camire, Maillard reaction and of course lysine loss due to the
Camire, & Krumhar, 1990). presence of both reducing sugars and the epsilon-amino
Collins and Walter (1985) reported that the major group of lysine.
nutritional change due to heat processing in sweet po- The FDNB procedure has been considered a stan-
tato was the loss of lysine, probably via reaction with dard reference method for monitoring reactive lysine
reducing sugars. In addition, Kays (1985) and Wolfe (Hurrel & Carpenter, 1974; Hurrell & Carpenter, 1981),
however, Vigo, Malec, Gomez, and Llosa (1992) and
*
Corresponding author. Address: Department of Food Science and
Morales, Romero, and Jimenez-perez (1995) showed
Technology, Micheal Okpara University of Agriculture, Umudike, that the method is fairly complicated, time consuming,
P.M.B. 7267, Umuahia, Abia State, Nigeria. Tel.: +234-533-205-2047. and requires special precautions. Carbohydrate-rich
0260-8774/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0260-8774(03)00212-7
144 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150

samples need to be dialyzed (Tomarelli, Yuhas, Fisher, pilot mixer. Extrusion temperature was set at 34, 54,
& Weaber, 1985). To eliminate these inherent limita- 100, and 100 ± 5 °C along the feeding, compression,
tions, Goodnoo, Swaisgood, and Catignani (1981) de- metering, and die zones respectively.
vised a fluorometric assay (with a lot of advantages), Extrusion screw speed adopted were 80, 92, 110, 127,
using o-phthaldialdehyde (OPA) to estimate reactive 140 rpm respectively. Die diameter adopted was 6, 7, 8,
lysine in proteins and a number of workers (Morales 9, and 10 mm respectively. Each of the dies had an L=D
et al., 1995; Swaisgood & Ctagnani, 1985; Vigo et al., ratio of 2.
1992) have favorably adopted it. This method could be
considered advantageous in estimation of lysine in car- 2.4. Drying of extrudates
bohydrate-rich foods such as the sweet potato, and also
in industrial routine analysis which requires limited Cylindrical rods obtained were dried in a cabinet
time. dryer at 40 ± 1 °C to obtain dry extrudates of 5–7%
Response surface analysis (RSA) is a system for op- moisture (Iwe, 1997).
timizing variables which differs from the usual one-
variable-at-a-time experimental procedure (Henika,
2.5. Preparation of extrudates for analysis
1982; Joglekar & May, 1987). RSA tests several vari-
ables at a time, uses special experimental designs to cut
Preparation of extrudates for analysis followed the
costs, and measures several effects by objective tests. A
reports of Iwe (1997) and Iwe et al. (1998). The dry
computer takes the experimental results and calculates
extrudates were coarsely ground in the Condiux tooth
models, using Taylor second-order equations which
mill and milled to pass through a 1 mm sieve in a Retsch
define relationships between variables and responses
type ZM1 mill (Retsch B.V., The Netherlands). Reco-
(Dziezak, 1990; Henika, 1982). The relationships are
vered materials were used for analyses.
quantitative, cover the entire experimental range tested,
and include interactions if present. Thus the models can
then be used to calculate any and all combinations of 2.6. Proximate composition
variables and their effects within the test range.
The goal of this research was to investigate the effects Proximate composition, of selected raw samples, such
of extrusion process conditions on available lysine and as moisture, crude protein, ash, ether extract, carbohy-
browning index of extruded soy–sweet potato mixtures. drate were obtained by official methods of analysis
(AOAC, 1984). Energy was calculated based on Atwater
factors (Osborne & Voogt, 1978).
2. Materials and methods
2.7. Available lysine
2.1. Materials
Available lysine was determined according to Vigo
Chips of orange fleshed variety of sweet potato (Ipo- et al. (1992) and Morales et al. (1995), with some
moea batatas), and defatted soy flour were obtained modifications. Homogeneous solution was obtained
from commercial suppliers in The Netherlands. from samples intended for use in protein digestibility
index assay (American Oil ChemistÕ Society, 1979). A
2.2. Raw materials and preparation 1.5 ml sample of the solution was diluted with 5 ml of
distilled water, and 0.5 ml of the solution mixed with 1.5
Sweet potato chips were cleaned and coarsely crushed ml of 16% sodium dodecyl sulphate (SDS) and stored
in a Condiux tooth mill (Condiux, West Germany) and overnight at 4 °C in the refrigerator. Then 1.5 ml of
then milled to pass a 1 mm sieve in a Retsch type ZM1 freshly prepared o-phthaldialdehyde (OPA) solution was
mill (Retsch B.V., The Netherlands). The milled samples mixed with 50 ll of the sample solution and incubated at
were later extruded. 25 °C for 2 min, and the intensity read in a Perkin Elmer
Luminescence LS 50B Spectrometer (Perkin Elmer Ltd.,
2.3. Extrusion of samples UK). Relative fluorescence (RF) was read at 340 nm
(extinction) and 455 nm (emission) with error value of
The Almex-Battenfeld single screw extruder of a 2.5%.
screw length to diameter ratio (L=D) of 16:1 and com- A calibration curve was made using soy protein iso-
pression ratio of 1:1.15 was used for the runs (Iwe & late (Purina Protein Europe) which contained 6.3 g/100
Ngoddy, 1998; Iwe, Wolters, Gort, Stolp, & Van Zuili- g of lysine. Percentage available/unavailable lysine was
chem, 1998). calculated based on Eq. (1), taking into account that the
After stabilization of the extruder, runs were carried RF value of the unheated sample means 100% available
out on 4 kg samples moistened to 18% moisture in a lysine (Morales et al., 1995):
 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150 145

Table 1 2.9. Experimental design and statistical analysis

Experimental design for the response surface analysis (upper level: 10
maximum number of experiment: 23)
A second-order central composite exact rotatable and
Run Cause variables nearly orthogonal design was developed on a CA-
1 2 3 DEMO package (Rasch, Nurnberg, & Williams, 1993),
1 ) ) ) as shown in Table 1. Effects of the process variables
2 + ) ) (feed composition (fc), screw speed (ss) and die diameter
3 ) + )
(dd)), on the response variables: available lysine and BI
4 + + )
5 ) ) + were investigated according to Myers (1976). The
6 + ) + RSREG procedure (SAS, 1990) was used in fitting the
7 ) + + model (Iwe et al., 1998).
8 + + + The process was optimized for maximum values of
9 )a 0 0
available lysine and minimum browning.
10 +a 0 0
11 0 )a 0 Plots of the fitted response surfaces were made as
12 0 +a 0 contained in an earlier report (Iwe et al., 1998).
13 0 0 )a
14 0 0 +a
15 0 0 0
3. Results and discussion
Runs 1–8 are to be performed one time, runs 9–14 are to be performed
one time, run 15 is to be performed nine times.
The experimental design is exact rotatable and nearly orthogonal and 3.1. Proximate composition of raw samples
requires 23 experiments of which one is to be performed at each cube
point, one at each axial point and nine at each center point. The above The proximate composition (dry weight basis) of the
is the essential part of the design. Each row states the adjustment levels raw soy and sweet potato samples used in this research is
of the factors at one run.
shown in Table 2. The protein content of the soy flour
was about 11 times more than the sweet potato flour,
  hence mixing soy flour with sweet potato flour contri-
FRs
%Unavailablelysine ¼ 100
 100 ð1Þ buted about 72% protein in the raw mixture. As ex-
FRb
pected, the addition of soy flour raised the nutritional
where FRs is the sample fluorescence, and FRb is un- status of sweet potato, as for other low protein foods
processed sample fluorescence. (Asp & Bjorck, 1989).

3.2. Available lysine

2.8. Browning index
The percentage values of available lysine in the pro-
Browning index (BI), was determined according to
cessed samples ranged from 68% to 100%. It was ob-
Palombo, Gertler, and Saguy (1984) on 1 g sample. The
served that there is a limit within which soy addition
optical density of centrifuged, clear filtrate was read on a
affected lysine availability. The estimated regression
Cecil (CE 2020, Cecil Instruments, Cambridge England)
coefficients and ANOVA of the response function,
spectrophotometer at 420 and 550 nm. Water was used
available lysine, in terms of the studied variables are
as blank and browning index, was calculated as:
shown in Table 3.
BI ¼ A420 nm
A550 nm ð2Þ Results indicated that both linear and cross product
effects of screw speed and feed composition were sig-
For practical purposes BI was expressed as OD/g dry nificant (p 6 0:10). A re-computation of the responses
solids (Palombo et al., 1984). showed that the cross product of screw speed and die

Table 2
Proximate composition of raw samples of soy and sweet potato flours
Feed comp (% swt pot) Protein (%) Fat (%) Ash (%) Carb. (%) Energy (kJ)
100 4.55 0.5 2.3 80.12 1441.36
75 16.57 0.5 3.0 72.15 1509.4
50 28.83 0.5 4.0 60.20 1514.6
25 40.12 0.5 5.0 48.17 1502.17
0 50.92 0.5 7.0 34.85 1497.36
Feed comp ¼ feed composition; swt pot ¼ sweet potato.
Carb. ¼ carbohydrate.
Carbohydrate calculated by difference, and Energy by Atwater factor.
146 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150

Table 3
Estimated regression coefficients and ANOVA for available lysine (AL), using the variables: feed composition (% sweet potato) (fc), screw speed (ss)
and die diameter (dd)
Source Coefficient Std error df P value
Regression coefficients:
Regression on constant 133.6265 117.3687
fc 0.120110 1.310474 1 0.2771
ss )0.080358 0.872322 1 0.3751
dd )14.3197 13.89824 1 0.3232
fc*fc )0.00629 0.006034 1 0.3171
ss*fc 0.012070 0.005153 1 0.0372
ss*ss )0.00393 0.002682 1 0.1682
dd*fc 0.044328 0.091730 1 0.6376
dd*ss 0.124134 0.061325 1 0.0658
dd*dd )0.12009 0.603399 1 0.8456
R2 0.9390

ANOVA and canonical value (CV)

Factor df Sum of squares F value CV
fc 4 1285.6902 0.0000 98.49
ss 4 475.9105 0.0003 84.55
dd 4 56.4005 0.2691 2.26

the sweet potato raw material, since the losses were more
pronounced at increasing levels of soy addition (Table
4), which apparently has higher lysine content (Iwe, van
Zuilichem, Ngoddy, & Lammers, 2001).
Increase in screw speed increased lysine retention,
owing possibly to reduced residence time of the feed
mixture in the extruder, since operating temperature was
more or less kept constant. Olkku, Antila, Heikkinen,
and Linko (1980) and Bartels, Janssen, and van Zuili-
chem (1982) had reported that an increase in screw
speed could either reduce residence time or have no ef-
fect, and hence influence lysine retention. However, Asp
Fig. 1. Effect of screw speed and die diameter on available lysine and Bjorck (1981) found a correlation between lysine
in extrudate. loss and screw speed, owing possibly to indirect effect of
starch hydrolysis at higher shear. Even though increased
diameter was also significant (Eq. (3)). The resulting shear leads to more severe conditions, the corresponding
polynomial, after removal of the non-significant terms reduction in residence time, as a result of increase in
and re-computation, becomes: screw speed, limits the duration of heat treatment and
resulting in high lysine retention (Table 4). The apparent
AL ¼ 198:58
0:53fc
1:67ss
12:69dd positive effect of increasing screw speed due to reduced
þ 0:01fcss þ 0:12ssdd ð3Þ residence time are available in several reports (Asp &
Bjorck, 1989; Bounie & Cheftel, 1986; Noguchi et al.,
Feed composition showed a high significant effect, fol- 1982; Pham & Del Rosario, 1984b). The effect of screw
lowed by screw speed. The cross product effect of screw speed in the present research could therefore be ex-
speed and die diameter was however higher than that of plained in line with already published reports.
feed composition and die diameter (Eq. (3)). Canonical analysis of the response surface showed
The response surface for these values is shown in Fig that the critical values of the independent variables were
1. It was practically observed that the increase in the estimated at 98.49% sweet potato level, 84.55 rpm and
level of sweet potato increased lysine retention. Higher 2.26 mm die diameter.
screw speed also enhanced lysine retention, especially at It is observed in Fig. 1 that there was a slight increase
increasing levels of sweet potato in the mixtures (figure, in lysine retention with smaller die diameter and lower
not shown). screw speed. This effect could not be explained as Tsao,
High retention of lysine at increasing sweet potato Frey, and Harper (1978) had reported that increase in
levels could be attributed to the lower levels of lysine in die diameter increased lysine retention in single screw
 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150 147

Table 4
A description of the experimental model based on selected process variables and their response (independent) variables
fc swt pot (%) ss RPM dd (mm) BI (OD/g) s (s) AL (%) DG (%) PT (°C)
60 110 8 0.33 30.4 64.4 49.0 115.0
100 110 8 0.31 45.3 95.3 81.6 115.3
80 80 8 0.39 62.5 72.9 81.4 108.7
80 140 8 0.44 43.9 88.7 74.9 117.6
80 110 6 0.29 47.2 83.3 92.9 110.8
80 110 10 0.39 42.7 84.4 97.3 122.0
80 110 8 0.33 48.7 82.0 77.0 113.3
fc ¼ feed composition; swt pot ¼ sweet potato; ss ¼ screw speed; dd ¼ die diameter; BI ¼ browning index; s ¼ mean residence time; AL ¼ available
lysine; DG ¼ degree of gelatinization; PT ¼ product temperature.

Table 5
Estimated regression coefficients and ANOVA for browning index (BI), using the variables: feed composition (% sweet potato) (fc), screw speed (ss)
and die diameter (dd)
Source Coefficient Std error df P value
Regression coefficients:
Regression on constant 2.206583 0.781832
fc )0.00377 0.008929 1 0.6799
ss )0.01858 0.005811 1 0.0077
dd )0.19889 0.092581 1 0.0528
fc*fc )0.00002 0.000040 1 0.7130
ss*fc )0.00006 0.000034 1 0.0962
ss*ss 0.000099 0.000018 1 0.0001
dd*fc 0.001565 0.000611 1 0.0249
dd*ss 0.000281 0.000409 1 0.3266
dd*dd 0.004111 0.004019 1 0.2288
R2 0.8711

ANOVA and canonical value (CV)

Factor df Sum of squares F value CV
fc 4 0.0047 0.0769 67.37
ss 4 0.0157 0.0012 103.73
dd 4 0.0119 0.0037 7.82

extrusion of lysine-fortified rice. However elimination sweet potato content and screw speed; and decreasing
of the die effect from the estimated coefficients did not die diameter (10–16 mm) respectively.
affect the shape and magnitude of the plot, hence the The estimated regression coefficients of the indepen-
resulting polynomials after re-computation becomes: dent variables are shown in Table 5. Linear effects of
screw speed and die diameter were significant (p 6 0:10).
AL ¼ 97:0253
0:5329fc
0:6758ss þ 0:0121fcss ð4Þ The quadratic effect of screw speed, and cross product
Following the re-computation, lysine retention was effect of die diameter and feed composition were sig-
shown to be highly influenced by feed composition and nificant (p 6 0:10).
screw speed. It is also evident that available lysine relates Resulting polynomial after removing insignificant
to the browning of extruded sample, owing to Maillard terms is:
reaction. This relationship could be inferred from the BI ¼ 1:77
0:01fc
0:02ss
0:10dd þ 0:00ss2
fact that it was those extrusion variables which affected
browning index that affected available lysine also,
0:00fcss þ 0:00fcdd ð5Þ
however at differing magnitude (Eqs. (3) and (4)).
Eq. (5) accounts for 85.13% of the total variation in
browning index. The response surface for the significant
3.3. Browning index variables are shown in Figs. 2 and 3.
Changing screw speed affected browning index, al-
The BI of the extrudates varied from 0.3050 to 0.4370 though higher level of browning was achieved at higher
OD/g. Browning was observed to be higher at increasing screw speed and at increasing feed composition (Fig 3).
148 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150

expected that browning index should be higher at

higher soy addition due to the presence of e-amino acids
in the soy. The present situation might have occurred as
a result of the oxidation of polyphenols in the sweet
potato raw material during preliminary processing
(Bouwkamp, 1985; Kays, 1985), which was carried into
the extruder.
In relation to feed composition, die diameter affected
browning (Fig. 3). Smaller die diameter and less sweet
potato significantly raised browning index. This effect
should be directly attributed to die diameter which also
affects residence time and therefore increased interaction
between reducing sugars and amino acids. Berset (1989)
reporting the work of Maga and Cohen (1978) stated
that die opening affected color development because a
Fig. 2. Effect of feed composition (% sweet potato) and screw speed large expansion or puffing tends to spread out dark
on browning index (BI). pigments.
Canonical analysis of the response surfaces showed
that the critical values of the independent variables were
estimated at feed composition value of 67.36%, screw
speed of 103.72 rpm, and die diameter of 7.82 mm. The
optimum point was however a saddle point.

4. Conclusion

Available lysine retention during extrusion cooking

of soy–sweet potato mixtures was significantly influ-
enced by screw speed, feed composition and die diame-
ter respectively. Optimum available lysine was estimated
at a feed composition of 98.49%, screw speed of 118.98
rpm, and die diameter of 2.25 mm. Similarly, increase in
feed composition and screw speed and a decrease in die
diameter also increased browning index. Optimum val-
Fig. 3. Effect of feed composition (% sweet potato) and screw speed ues of the process variable for a minimum value of
on browning index (BI).
browning index was estimated at feed composition,
screw speed and die diameter values of 67.36%, 103.73
rpm and 7.82 mm.
Higher screw speed translates into higher shear,
although residence time is reduced. The increase in
browning due to increase in screw speed could be ex-
plained according to Areas (1992) and Mitchell and Acknowledgements
Areas (1992) who reported that only degraded poly-
saccharide can take part in the browning reaction. From The authors are grateful to Drs. F.J. Morales and
a previous report (Iwe et al., 1998), it was shown that A. A. Metwalli, Dairy and Food Physics Department,
most processed samples were up to 70% gelatinized. Our Wageningen Agricultural University, The Netherlands,
result as shown in Table 4 confirms this. At a high screw for their advice on lysine and color measurements; Dr.
speed, and moderate feed composition and die diameter, G. Gort of the Sub-Department of Mathematics, Wa-
a BI value of 0.44 OD/g was achieved. This value cor- geningen Agricultural University, The Netherlands, for
responded to a mean residence time of 43.89 s, available the generation of the response surfaces, the Nigerian
lysine of 88.67%, starch damage of 74.89% and product National Universities Commission/World Bank and the
temperature of 117.6 °C (Table 4). University of Agriculture Makurdi, Nigeria for pro-
The increase in browning index at lower levels of viding the sponsorship; and the Department of Food
soy addition might be due to the initial color of the raw Science Wageningen Agricultural University, The
sweet potato, as explained earlier. It should have been Netherlands, for research space and hospitality.
 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150 149

References Iwe, M. O., & Ngoddy, P. O. (1998). Proximate composition and some
functional properties of extrusion cooked and sweet potato blends.
AOAC (1984). Official Methods of Analysis. Washington DC: Asso- Plant Food for Human Nutrition, 53(2), 121–132.
ciation of Official Analytical Chemists. Iwe, M. O., Wolters, I., Gort, G., Stolp, W., & Van Zuilichem, D. J.
American Oil ChemistÕ Society (AOCS) (1979). Protein dispersibility (1998). Behavior of gelatinization and viscosity in soy–sweet potato
index (PDI), AOCS Official Method Ba 10-65. mixtures by single screw extrusion: a response surface analysis.
Asp, N.-G., & Bjorck, I. (1981). Influence of extrusion cooking on the Journal of Food Engineering, 38, 369–379.
nutritional value. Nordforsk-SIK-SNF Seminar: Extrusion cook- Iwe, M. O., van Zuilichem, D. J., Ngoddy, P. O., & Lammers, W.
ing of food, Gothenburg. (2001). Amino acid and protein dispersibility index (PDI) of
Asp, N. G., & Bjorck, I. (1989). Nutritional properties of Extruded mixtures of extruded soy and sweet potato flours. Lebensmittel-
foods. In C. Mercier, P. Linko, & J. M. Harper (Eds.), Extrusion Wissenschaft und-Technologie, 34, 71–75.
cooking (pp. 399–433). St. Paul: American Association of Cereal Joglekar, A. M., & May, A. T. (1987). Product excellence through
Chemists Inc. design of experiments. Cereal Foods World, 32(12), 857–868.
Areas, J. A. G. (1992). Extrusion of proteins. Critical Reviews in Food Kays, S. J. (1985). Formulated sweet potato products. In J. C.
Science and Nutrition, 32, 365–392. Bouwkamp (Ed.), Sweet potato products: a natural resource for the
Bartels, P. V., Janssen, L. P. B. M., & van Zuilichem, D. J. (1982). tropics (pp. 205–218). Boca Raton, FL: CRC Press.
Residence time distribution in a single screw cooking extruder. Maga, J. A., & Cohen, M. R. (1978). Effect of extrusion parameters
Seminar ZDS. Koch and Extrudier Technieken, Soligen-Grafrath. on certain sensory, physical, and nutritional properties of po-
Berset, C. (1989). Color. In C. Mercier, P. Linko, & J. M. Harper tato flakes. Lebensmittel-Wissenschaft und-Technologie, 11, 195–
(Eds.), Extrusion cooking (pp. 399–434). St. Paul: American 197.
Association of Cereal Chemists Inc. Mitchell, J. R., & Areas, J. A. (1992). Structural changes in
Bjorck, I., & Asp, N. G. (1983). The effects of extrusion cooking on biopolymers during extrusion. In J. L. Kokini, C. T. Ho, & M.
nutritional value––a literature review. Journal of Food Engineering, V. Karwe (Eds.), Food extrusion science and technology (pp. 345–
2, 281–308. 360). New York: Mercel Dekker.
Bounie, D., & Cheftel, J. C. (1986). The use of tracer and thermosen- Morales, F. J., Romero, C., & Jimenez-perez, S. (1995). New
sitive molecules to assess flow behavior and severity of heat methodologies for kinetic study of 5-(hydroxymethyl)-furfural
processing during extrusion-cooking. In C. Mercier, & C. Cantar- formation and reactive lysine blockage in heat-treated milk and
elli (Eds.), Pasta and extrusion cooked foods: some technological and model systems. Journal of Food Protection, 58(3), 310–315.
nutritional aspects (pp. 148–158). London, England: Elsevier Myers, R. H. (1976). Response surface methodology. Boston, MA:
Applied Science Publ. Ltd. Allyn and Bacon.
Bouwkamp, J. C. (1985). Processing of sweet potatoes––canning, Noguchi, A., Mooso, K., Aymard, C., Jeunink, J., & Cheftel, J. C.
freezing, dehydrating. In J. C. Bouwkamp (Ed.), Sweet potato (1982). Maillard reactions during extrusion-cooking of protein-
products: a natural resource for the tropics (pp. 185–203). Boca enriched biscuits. Lebensmittel-Wissenschaft und-Technologie, 15,
Raton, FL: CRC Press. 105–110.
Camire, M. E., Camire, A., & Krumhar, K. (1990). Chemical and OÕBrien, J., & Morrissey, P. A. (1989). Nutritional and toxicological
nutritional changes in foods during extrusion. Critical Reviews in aspects of the Maillard browning reaction in foods. Critical Review
Food Science and Nutrition, 29, 35–57. in Food Science and Nutrition, 28, 211–248.
Cheftel, J. C. (1986). Nutritional effects of extrusion-cooking. Food Olkku, J., Antila, J., Heikkinen, J., & Linko, P. (1980). Residence time
Chemistry, 20, 263–283. distribution in twin-screw extruder. In P. Linko, Y. Malkki, J.
Collins, W. W., & Walter, W. M., Jr. (1985). Fresh roots for human Olkku, & J. Larinkari (Eds.), Food processing systems: vol. 1. Food
consumption. In J. C. Bouwkamp (Ed.), Sweet potato: a natural process engineering (pp. 791–794). London: Applied Science Pub-
resource for the tropics. Boca Raton, FL: CRC Press. lishers.
Dziezak, J. D. (1990). Taking the gamble out of product development. Osborne, D. R., & Voogt, P. (1978). Calculation of calorific value.
Food Technology, 125, 110–117. The analysis of foods. London: Academic Press, pp. 107–110,
Goodnoo, C. C., Swaisgood, H. E., & Catignani, G. L. (1981). A 239.
fluorometric assay for available lysine in proteins. Analytical Palombo, R., Gertler, A., & Saguy, I. (1984). A simplified method for
Biochemistry, 115, 203–211. determination of browning in dairy powders. Journal of Food
Harper, J. M. (1979). Food extrusion. Critical Reviews in Food Science Science, 49, 1609–1613.
and Nutrition, 11, 155–215. Pham, C. B., & Del Rosario, R. R. (1984b). Studies on the
Henika, R. G. (1982). Use of response-surface methodology in sensory development of texturized vegetable products by the extrusion
evaluation. Food Technology, 36(11), 96–101. process. II. Effects of extrusion variables on the available lysine,
Hurrel, R. F., & Carpenter, K. J. (1974). Mechanisms of heat damage total and reducing sugars. Journal of Food Technology, 19, 549–
in proteins. IV. The reactive lysine content of heat-damaged 559.
material as measured in different ways. British Journal of Nutrition, Rasch, D., Nurnberg, G., & Williams, E., 1993. CADEMO manual.
32, 589–604. Biorat GmbH Rostock.
Hurrel, R. F., & Carpenter, K. J. (1977). Maillard reaction in foods. In SAS, 1990. The RSREG procedure. SAS/STAT UserÕs Guide, Version
T. Hoyem, & O. Kvale (Eds.), Physical, chemical, and biological 6, vol. 2. pp. 1351–1456, SAS Institute Inc. SAS Circle Cary, NC.
changes in food caused by thermal processing (pp. 168–184). Sgaramella, S., & Ames, J. M. (1993). The development of color in
London: Applied Science Publishers. extrusion cooked foods. Food Chemistry, 46, 129–132.
Hurrell, R. F., & Carpenter, K. J. (1981). The estimation of available Swaisgood, H. E., & Ctagnani, G. L. (1985). Digestibility of modified
lysine in foods after Maillard reactions. Progress in Food Nutri- milk proteins: nutritional implications. Journal of Dairy Science,
tional Science, 5, 159–176. 68, 2782–2790.
Iwe, M.O., 1997. Optimization of engineering, functional and nutri- Tomarelli, R. M., Yuhas, R. J., Fisher, A., & Weaber, J. R. (1985). An
tional properties of single-screw extruded soy–sweet potato blends HPLC method for the determination of reactive (available) lysine
for food use by response surface methodology. PhD thesis, in milk and infant formulas. Journal of Agircultural Food Chem-
University of Nigeria, Nsukka, Nigeria. istry, 33, 316–318.
150 M.O. Iwe et al. / Journal of Food Engineering 62 (2004) 143–150

Tsao, T.-F., Frey, A. L., & Harper, J. M. (1978). Available lysine in tion of reactive lysine in dairy products. Food Chemistry, 44, 363–
heated fortified rice meal. Journal of Food Science, 43, 1106–1108. 365.
Vigo, M. S., Malec, L. S., Gomez, R. G., & Llosa, R. A. (1992). Wolfe, J. A. (1992). Sweet potato: an untapped food resource.
Spectrophotometric assay using o-phthaldiadehyde for determina- Cambridge University Press.