Turmeric Powder and Starch: Selected Physical,

Physicochemical, and Microstructural Properties

Dhanalakshmi Kuttigounder, Jaganmohan Rao Lingamallu, and Suvendu Bhattacharya
C: Food Chemistry

Abstract: Turmeric powder and its starch were characterized for physical, physicochemical, and microstructural char-
acteristics. X-ray diffractogram indicated that turmeric starch to be of B type. Dried and cured-dried turmeric powder
samples showed higher water-holding capacity (3.62 and 4.78 g/g, respectively) compared to isolated starch (1.07 g/g)
at 30 ◦ C. Non-Newtonian shear-thinning characteristics were observed with turmeric powder dispersion containing
10% (w/w) solids. A power law model fitted well to correlate the shear-rate and shear-stress data (r = 0.993 to 0.999,
P ≤ 0.01) for both samples. Apparent viscosities of isolated turmeric starch and cured-dried turmeric powder dispersion
containing 10% (w/w) solids were 1.29 ± 0.03 and 7.57 ± 0.39 mPa s, respectively. Microstructure of starch particles
showed a smooth flat outer surface. The approximate length and breadth of isolated elliptical starches were 25 and 10 μm
while the thickness was about 5 μm.
Keywords: Curcuma longa, microstructure, rheological properties, turmeric starch, X-ray diffraction

Practical Application: Isolation and characterization of starch from an unconventional source like turmeric rhizome
indicate a potential application as a functional ingredient in foods and pharmaceutical industries including agglomerated
products.

Introduction predominantly starch. Hence, there exists a scope to isolate and

Starch is a valued component of many foods and has an im-
portant role both as a macronutrient and structural component.
Starch, a biopolymer of glucose, is the major storage component
of many economically important crops such as cereals, legumes,
tubers, and rhizomes. The structure and physicochemical proper-
ties of starch differ with its botanical origin. Starch is frequently
isolated and is used in food industries to impart the desirable
functional properties, and modify food texture and consistency.
The amount of starch as well as the starch type is important for
the texture of a given product (Khatoon and others 2009). Thus,
starch has many applications in both food (for example, thickener,
binder, bulking, water retention and gelling agent, texture mod-
ifier, and adhesive) and nonfood (for example, paper, cardboard,
pharmaceutical) industries (Manzan and others 2003; Braga and
others 2006).
Turmeric is derived from the rhizome of the plant Curcuma longa
L. (turmeric); it is a tropical herb of the Zingiberaceae family, which
is native to southern Asia. It is primarily consumed in the form
of powdered rhizomes for coloring and flavoring foods. The spice
extract is used in the form of liquid concentrate or encapsulated on
a carrier matrix (Govindarajan 1980). Turmeric oleoresin usually
contains 30% to 45% curcuminoids and 15% to 20% of volatile
oil. After the extraction of the oleoresin for its primary coloring
components (curcuminoids), the major portion of the residue is
MS 20110528 Submitted 4/27/2011, Accepted 8/11/2011. Authors are with ceutical products. Hence, the objective of this study was to isolate
Central Food Technological Research Inst. (Council of Scientific and Industrial Re-
search), Mysore 570020, India. Direct inquiries to author Bhattacharya (E-mail: starch from turmeric rhizome and characterize the same with par-
suvendu@cftri.res.in).
characterize the starch obtained from turmeric rhizome and to
compare its properties with that of dried and cured-dried turmeric
rhizome powders to identify their similarities and differences after
processing.
Studies on the isolation of starch from turmeric rhizome and
its characterization are limited. Jyothi and others (2003) have ex-
tracted starch from 2 species of curcuma such as Curcuma zedoaria
and Curcuma malabarica; no significant differences have been re-
ported between these 2 species in respect of granule shape, size,
and amylose content. Elliptical granule shape with an average gran-
ule size of about 35 μm has been reported. Leonel and others
(2003) have extracted starch from 2 curcuma species (C. longa and
C. zedoaria). Microscopic examination has revealed a flat triangu-
lar shape while the viscoamylographic investigation has indicated
the pasting (gelatinization) temperature to be 81 and 78 ◦ C for
C. longa and C. zedoaria, respectively. Braga and others (2006)
have observed that turmeric rhizomes contain about 40% (w/w)
starch, isolated after the extraction of oil from turmeric by using
super critical fluid extraction. The general conclusion that was de-
rived from all of these studies is that starch from curcuma species
is different from conventional sources like cereal, pulse, and tuber
crops with respect to physical and physicochemical characteristics.
Scope thus exists to have a detailed characterization of turmeric
starch and processed powder with particular reference to mi-
crostructural/structural features and rheological characterization.
The possible applications of these studies lie in the use of turmeric
starch as a functional ingredient in different foods and pharma-
ticular reference to physical, physicochemical, and morphologi-
cal characteristics. In addition, the characterization of dried and

C 2011 Institute of Food Technologists

 R

C1284 Journal of Food Science r Vol. 76, Nr. 9, 2011 doi: 10.1111/j.1750-3841.2011.02403.x
Further reproduction without permission is prohibited
Characteristics of turmeric starch . . .

cured-dried turmeric powders has been conducted and compared Proximate composition
with the isolated turmeric starch. The fresh, dried, and cured-dried turmeric rhizomes were
characterized for their proximate composition including moisture
Materials and Methods content, total ash, crude fat, crude fiber, protein (N∗ 6.25), and
carbohydrate content (by difference) by following the AOAC
Materials methods (2005) on triplicate samples.
Fresh turmeric (C. longa) rhizomes were procured from a local

C: Food Chemistry
cultivator near Mysore, Karnataka, India. The matured rhizomes Amylose content
from 3 different batches were harvested when the plants were
The amylose content of isolated starch was determined on tripli-
about 10 months old. The rhizomes were manually cleaned with
cate samples following the method of Sowbhagya and Bhattacharya
water to remove the adhering soil and extraneous matter, if any.
(1971). Isolated starch samples (approximately 100 mg) were dis-
Surface water was removed with a dry cloth. The moisture content
solved in 1 N sodium hydroxide solution followed by incubation
of fresh rhizome was determined on 6 samples by employing the
for 20 h at 25 ◦ C. The resulting solution was neutralized with
ASTA method 2.0 (1997). The moisture content of the fresh
0.05 N HCl. Two milliliter of 0.2% (w/v) iodine reagent was
rhizome sample was 80.7% ± 1.2% (v/w) (wet basis). The fresh
added and allowed to stand for 20 min. The absorbance of
rhizomes were grated and dried at 40 ◦ C in a tray dryer for 8 h
the amylose–iodine complex was measured at 620 nm using
followed by powdering in a laboratory mill to obtain fine powder
an UV–visible spectrophotometer (model: UV-160A, Shimadzu,
having a moisture content of 8.3% ± 0.1% (v/w); this sample
Corp., Kyoto, Japan). The standard calibration curve was plot-
has been referred to as the “dried” rhizome powder sample. The
ted using different concentrations of standard amylose and amy-
fresh rhizomes were cured by employing the conventional method
lopectin. The amylose content was reported as the average apparent
by cooking in excess of boiling water for about an hour; excess
amylose content.
water was discarded and the rhizomes were dried in the shade
for a week. These dried rhizomes were powdered to obtain fine
Water-holding capacity of starch
powder having a moisture content of 10.7% ± 0.2% (v/w); this
The water-holding capacity (WHC) of isolated starches was de-
sample is referred to as the “cured-dried” sample. The process was
termined employing the AACC method 56-20 (1983). Approxi-
repeated twice. Standard amylose and amylopectin were obtained
mately 1 g of sample was dispersed in 30 mL of distilled water in a
from Sigma, St. Louis, Mo., U.S.A. All chemicals used were of
preweighed centrifuge tube. The centrifuge tubes were mixed in a
analytical grade and were procured from local chemical suppliers.
vortex mixer and were kept in a water bath maintained at different
temperatures (30, 40, 50, 60, 70, and 80 ◦ C) for 30 min. Later,
Isolation of starch
the centrifuge tubes containing the dispersion were centrifuged at
Starch was isolated from fresh rhizome following the method
2670 × g for 20 min. The supernatant was decanted and the cen-
(Figure 1) suggested by Badenhuizen (1964). This process was
trifuge tubes along with sediment were weighed and WHC was
repeated in triplicate and the isolated starch was dried in a hot air
◦ calculated from Equation 1. WHC was determined on triplicate
oven maintained at 50 C followed by storage at room temperature
samples and was expressed as g of water absorbed per g of starch:
(about 25 ◦ C) for further analysis.
(Mass of sediment − Mass of sample)
WHC = (1)
Mass of sample
Fresh grated turmeric rhizome
1% cold ammonium
Blending
oxalate solution/water Rheological characterization
A controlled stress rheometer (Model # RS6000 Haake
Thin slurry
RheoWin, Thermo Scientific, Karlsruhe, Germany) was used
Filtration through cloth to determine the flow behavior of the dispersions containing
turmeric powder (cured-dried and dried rhizomes) and starch (iso-
Filtrate
lated from fresh turmeric rhizome). All rheological measurements
were conducted at 25 ◦ C on triplicate samples by employing a cir-
Centrifugation at 2670 x g culatory water bath for control of temperature by ±0.1 ◦ C. The
rheological experiments were performed using a coaxial cylin-
Starch cake der (Z41) attachment (Bhattacharya 1999). Dispersions containing
10% (w/w) turmeric powder (dry solid basis) were prepared and
Washing and suspending in distilled water
analyzed for their flow characterization. These samples were ini-
tially presheared at a shear rate of 10 s−1 for 30 s followed by gradual
Crude starch
increase in shear rate up to 500 s−1 in 60 s. The flow characteristics
Washing with methanol of turmeric starch, isolated by using ammonium oxalate and water,
were also measured by preparing dispersions containing 10%, 15%,
Cleaned starch cake and 20% (w/w) solids. Apparent viscosity was taken as the ratio
of shear stress and shear rate when the latter was taken as 100 s−1 .
Drying in hot air oven (50°C) Yield stress was determined by shearing at 5 s−1 for 30 s followed
by stopping the spindle and allowing the sample to relax for 120 s.
Isolated starch powder However, we have not observed any detectable yield stress for
the samples. Hence, shear-stress and shear-rate data were fitted
Figure 1–Flow chart for the isolation of starch from turmeric rhizomes. to a power law model (Equation 2), and the model parameters

Vol. 76, Nr. 9, 2011 r Journal of Food Science C1285

 Characteristics of turmeric starch . . .
(k and n) were calculated by using the software provided by the
instrument manufacturer. In the power law model (Equation 2), σ
is the shear stress (Pa),γ̇ is the shear rate (s−1 ), k is the consistency
index (Pa sn ), and n is the flow behavior index (dimensionless). The
suitability of the power law model was examined by determining
the variance (χ 2 ) and correlation coefficient (r); the significance
of r-values was judged at P ≤ 0.01
C: Food Chemistry
σ = k γ̇ n
Fourier-transformation infrared spectroscopy
Fourier-transformation infrared (FTIR) spectra of isolated
starches from fresh rhizome were obtained by scanning in a FTIR
spectrophotometer (Model # Nicolet 5700, Thermo Electron
Corp., Madison, Wis., U.S.A.) at room temperature. The starch
granules were blended with potassium bromide to obtain a fine
powder mix and then pressed into thin transparent pellets before
analysis. The sample was scanned in the region between 4000 and
400 cm−1 (Liu and others 2002).
X-ray diffraction
The X-ray diffraction pattern of isolated starch and turmeric
powder was obtained using an X-ray diffractometer (Model #
Rigaku miniflux II benchtop XRD system, Tokyo, Japan)
equipped with a monochromator attachment that selects Kα ra-
diation from a copper target generated at a target voltage of
40 kV and a tube current of 30 mA. The diffraction intensity of
the powder sample was measured from 10◦ to 40◦ , which covers
all the significant peaks of starch as a function of 2θ at a scanning
speed of 4◦ /min (Nara and Komiy 1983). Percentage crystallinity
of the samples was calculated as suggested by Hayakawa and others
(1997). The area of crystalline reflection was divided by the sum
of the areas of crystalline reflection and amorphous reflection, and
expressed as percentage crystallinity.
Microstructure, particle size, and shape
The microstructure of isolated starch and turmeric powder sam-
ples was examined by employing a scanning electron microscope
(Model # 435 VP, Leo Electron Microscopy, Cambridge, U.K.).
The sample was mounted on aluminum stubs with double adhesive
conducting tape and was coated with a thin film of gold employ-
ing a sputter coater. Microscopical examinations of the samples
were conducted at an accelerating voltage of 15 kV at magnifica-
tions of 200×, 1000×, and 5000× to observe the morphology,
shape, and size of the samples. Representative photomicrographs
are presented in this study.
The geometric foundation of the concept of sphericity rests
upon the isoperimetric property of a sphere. A practical 3-
dimensional expression can be stated for determining the spheric-
ity of an object (Equation 3) (Mohsenin 1986). Ten samples were
examined each time:
1
(a b c ) 3
Sphericity =
a Fresh
Here, a is the longest intercept, b is the longest intercept normal
to a, and c is the longest intercept normal to both dimensions of a
and b.
Statistical analysis
All measurements were carried out in triplicates and data were
expressed as means ± standard deviations. One-way analysis of
C1286 Journal of Food Science r Vol. 76, Nr. 9, 2011
variance and least significant difference (LSD) were conducted
using the software Statistica 99 (StatSoft, Tulsa, Ohio, U.S.A.) to
ascertain the significant differences among the mean scores at a
probability of P ≤ 0.05 (Snedecor and Cochran 1968).
Results and Discussion
Isolation of starch
(2) The fresh turmeric rhizome had a moisture content of around
80% (v/w) and starch constituted the major portion (67%, w/w)
of the remaining dry material. The isolated starch was pale yellow
in color at the initial phase of extraction due to the presence of
curcumin. This color disappeared when the extracted crude starch
was washed repeatedly with methanol (Figure 1). The starch was
finally washed with distilled water followed by drying at 50 ◦ C to
a moisture content of 9.2% ± 0.3% (w/w). Yield of starch was
around 56% (w/w) (dry basis) whereas the earlier reported value
was around 47% (w/w) (Leonel and others 2003).
Proximate composition
The proximate composition of turmeric rhizomes (fresh,
dried, and cured-dried) was significantly different (at P ≤ 0.05)
(Table 1). The carbohydrate content of fresh rhizome was about
67.6% (w/w) (dry basis), which compares well with the earlier re-
ported values of Braga and others (2006) and Govindarajan (1980)
who had reported it in the range of 70% to 80% (w/w).
Amylose content
The amylose content for isolated turmeric starch was 48.4% ±
2.8% (w/w). The amylose content of turmeric starch, as deter-
mined employing the iodometric method, was appreciably higher
than that of potato starch (20.1% to 31.0%, w/w) (Kim and others
1995), starch from Indica and Japonica rice varieties (18% to 32%
and 10% to 22%, w/w, respectively) (Juliano 1985). The amylose
content of isolated turmeric starch had been reported to be 22%
and 48% (w/w) by Leonel and others (2003) and Braga and others
(2006), respectively. The high amylose content is beneficial for the
manufacture of extruded and fried snack foods particularly where
low expansion, crispiness, and reduced fat uptake upon frying are
desired (Jobling 2004).
Water holding capacity
The WHC of isolated turmeric starch (Figure 2) showed a
nonsignificant change (at P ≤ 0.05) up to 60 ◦ C but increased
thereafter with an increase in temperature of measurement. It is
interesting to note that WHC drastically increased at 80 ◦ C, which
may be attributed to the phenomenon of pasting or gelatinization.
Hence, the starch granules gelatinize around this temperature. It
is obvious that the WHC of gelatinized starch is more than that
of native uncooked starch. This phenomenon might be attributed
to thermal processing wherein the native structure of starch had
(3) Table 1–Proximate composition of different turmeric rhizomes.
Dried Cured-dried
Moisture 80.02 ± 0.30 8.26 ± 0.03 10.66 ± 0.18b
c a
Protein (N∗ 6.25) 2.07 ± 0.11a 9.27 ± 0.02c 9.12 ± 0.03b
Fat 0.80 ± 0.11a 3.62 ± 0.04b 4.70 ± 0.04c
Crude fiber 1.79 ± 0.05a 3.41 ± 0.06b 4.10 ± 0.13c
Total ash 1.83 ± 0.04a 7.87 ± 0.06b 6.95 ± 0.15c
Carbohydrate (by difference) 13.51 ± 0.17a 67.54 ± 0.04c 64.48 ± 0.23b
Mean score with different superscripts within the same row differ significantly at P ≤
0.05 according to LSD.
Characteristics of turmeric starch . . .

been disturbed in the presence of excess water, exhibiting swelling
in addition to volume increase along with high uptake of water.
For a cured-dried sample, cooking of rhizome in boiling water
is an integral step for processing and thus the starch present in it
appears to be partially or completely gelatinized to exhibit high
WHC values at all temperatures. The grating and drying pro-
cesses possibly might have modified the native structure of starch.
ence of yield stress (by employing the method of stress relaxation).
The flow behavior index (n) of isolated native starch dispersion was
1.6 indicating shear-thickening characteristics. Though not com-
mon in the food system, cornstarch dispersions had been reported
to exhibit shear-thickening behavior (Okechukwu and Rao 1995).
The apparent viscosity of starch dispersion from fresh samples
was low (1.29 mPa s) possibly due to low WHC. An increase in

However, WHC also depended on the presence of protein (9.3%,
w/w) and crude fiber (3.6% to 4.7%, w/w) (Table 1). The net
result was the higher WHC values of 3.62 and 4.78 g/g for dried
and cured-dried samples, respectively, compared to that of isolated
starch (1.07 g/g) at 30 ◦ C.
Rheological characterization
The parameters for the flow characterization of isolated starch,
dried, and cured-dried turmeric powder dispersions containing
10% solids (w/w) are shown in Table 2. The sample rheogram
(Figure 3) for 10% (w/w) isolated starch dispersion indicated a
shear-thickening behavior, and we were unable to detect the pres-
C: Food Chemistry
the concentration of solid (starch) from 10% to 20% (w/w) in
the dispersions increased apparent viscosity and consistency index
(Table 3). The flow behavior index (n) of these starch dispersions
decreased with an increase in solid concentration. All these disper-
sions exhibited shear-thickening phenomenon (n > 1), and these
samples obeyed the power law model (0.977 ≤ r ≤ 0.998).
3

Shear stress (Pa)

2
10
WHC (g of water/g of sample)

8
1
6
C
4
B

2 0
A
0 100 200 300 400 500
0
20 40 60 80 -1
Shear rate ( s )
o
Temperature ( C)
Figure 3–Sample rheograms for dispersions containing 10% cured-dried
Figure 2–WHC of (A) isolated turmeric starch, (B) dried turmeric powder, turmeric powder (hollow square), and starch from fresh rhizome (solid
and (C) cured-dried turmeric powder at different temperatures. triangle).

Table 2–Rheological parameters of turmeric powder and starch dispersions containing 10% dry solid.
Power law model parameters
Apparent Flow Consistency Correlation
viscosity∗ behaviour index coefficient
Sample (mPa s) index (n) (–) (k) (Pa sn ) Variance (–) (r) (–)
Isolated starch 1.29 ± 0.03a 1.64 ± 0.01c 4.98E − 05 ± 1.29E − 05a 0.73 0.998
Dried turmeric powder 4.15 ± 0.23b 1.00 ± 0.04b 0.004 ± 0.001a 0.01 0.953
Cured-dried turmeric powder 7.57 ± 0.39c 0.81 ± 0.01a 0.018 ± 0.004b 0.07 0.999
∗
Reported at a shear rate of 100 s−1 .
Mean score with different superscripts within the same column differ significantly at P ≤ 0.05 according to LSD.

Table 3– Rheological parameters of turmeric starch dispersions at different concentrations of solids isolated using ammonium
oxalate and water.
Power law model parameters
Apparent Flow Consistency Correlation
Concentration viscosity∗ behavior index coefficient
of solid (%) (mPa s) index (n) (–) (k) (Pa sn ) Variance (–) (r) (–)
Ammonium oxalate extracted
10 1.20 ± 0.03a 1.62 ± 0.03b 4.98E − 05 ± 1.03E − 05a 0.46 0.977
15 1.50 ± 0.07b 1.59 ± 0.05b 6.44E − 05 ± 1.68E − 05a 0.53 0.996
20 1.90 ± 0.01c 1.45 ± 0.02a 2.18E − 04 ± 8.25E − 05a 0.39 0.992
Water extracted
10 1.21 ± 0.03a 1.61 ± 0.01b 5.31E − 05 ± 2.43E − 05a 0.03 0.998
15 1.58 ± 0.01b 1.58 ± 0.01b 7.09E − 05 ± 6.80E − 05a 0.08 0.996
20 1.90 ± 0.01c 1.39 ± 0.02a 2.38E − 04 ± 3.48E − 05b 0.15 0.993
∗
Reported at a shear rate of 100 s−1 .
Mean score with different superscripts within the same column differ significantly at P ≤ 0.05 according to LSD.

Vol. 76, Nr. 9, 2011 r Journal of Food Science C1287

 Characteristics of turmeric starch . . .

On the other hand, the dried and cured-dried turmeric pow-
der samples, appearing to be partially gelatinized, showed higher
viscosity compared to the isolated starch sample; the cured-dried
sample exhibited the highest viscosity among all the 3 samples (Ta-
ble 2). Turmeric powder dispersions showed a common trend of
shear-thinning behavior as 0.81 ≤ n ≤ 1.00. The power law model
(Equation 2) could correlate the shear-stress and shear-rate data
C: Food Chemistry
cess as well as moisture content (van Soest and others 1995). The
FTIR spectrum of isolated starch showed characteristic peaks at
1020, 1080, and 1156 cm−1 (Figure 5). The peak at 1020 cm−1
was assigned to the vibration of C–O–H deformation and was
related to changes in the amorphous and crystalline parts of the
granule (Liu and others 2002). The intensity of transmittance was
less (15%) at 1020 cm−1 showing that the crystallinity of starch had

well (Figure 4) as the variance (χ 2 ) values were low (0.01 to 0.73) not significantly altered due to the isolation process. The starch
while the correlation coefficients (r) were high (0.953 to 0.999).

Structural features of isolated starch from the infrared 6000

A
spectrum
The IR spectrum of starch indicated it to be sensitive to changes 4000
in molecular structure (short-range order), such as starch chain
conformation, helicity, crystallinity, and the retrogradation pro- 2000

0
2.0
6000
B

Intensity (cps)
4000
Shear stress (Pa)

1.5

2000

1.0 0

6000
C
0.5 4000

2000

0
0
00 100
100 200
200 300
300 400
400 500
500
-1
10 15 20 25 30 35 40
Shear
Áinrate
1/s (s ) Diffraction angle (2θ)

Figure 4–Suitability of the power law model for isolated turmeric starch Figure 6–X-ray diffraction spectrum of (A) isolated starch, (B) dried
dispersion at 10% concentration. turmeric powder, and (C) cured-dried turmeric powder.

starch Figure 5–FTIR

95
spectrum of
90 starch isolated
2135.3

85 from turmeric
rhizome.
859.7

80

75

70
929.8

65
710.7
1651.1

60
1248.0

55
764.7
% Transmittance

1368.8

50
527.9
1208.0

440.2
1416.0

45
2930.9

576.7

40

35

30

25
1156.2

20
3345.6

15
1019.7

10
1081.0

4000 3500 3000 2500 2000 1500 1000 500

-1
Wave number ( cm )

C1288 Journal of Food Science r Vol. 76, Nr. 9, 2011

Characteristics of turmeric starch . . .

was in its native form and absorbed most of the radiations emitted.
Peaks at 1080 and 1156 cm−1 could be attributed to the coupling
of C–O, C–C, and O–H bond stretching, bending and asym-
metric stretching of the C–O–C glycosidic bridge (van Soest and
others 1995). The changes at 1020 and 1640 cm−1 indicate that
gelatinization was primarily a hydration process (Liu and others
2002). The wide band at 3300 to 3600 cm−1 could be assigned
The X-ray diffractogram pattern of isolated turmeric starch
in the present study resembled B-type (Figure 6A) showing a
strong peak at 17◦ with a maximum intensity of 5312 cycles
per second (cps), and weak peaks at around 20◦ , 22◦ , and 24◦ .
Our observations were similar to that of earlier reported type on
starches from curcuma species (C. zedoaria, C. malabarica, and C.
amad) (Jyothi and others 2003; Policegoudra and Aradhya 2008)

to the O–H stretching, and the bands at 2931 cm−1 to the asym-
metric stretching of C–H. The band at 1651 cm−1 was attributed
to the absorbed water, and the band at 1416 and 1369 cm−1 , re-
spectively, to the angular deformation of C–H (Policegoudra and
Aradhya 2008). The IR spectra of the isolated starch revealed that
the starch existed in the native pure form.
X-ray diffraction
The X-ray diffraction technique is a nondestructive method to
reveal the presence and characteristics of the crystalline structure
of the starch granules (Hoover 2001). Starch crystallinity patterns
can be classified into A, B, and C forms (Zobel 1988). The as-
signment of a particular starch to one of the groups is gener-
ally based on the direct comparison of the X-ray diffractograms
with those of starches having established and characterized pat- Microstructure, shape, and particle size
terns. The crystalline nature of starch depends on the botanical
source.
C: Food Chemistry
The X-ray diffraction pattern of dried and cured-dried turmeric
powders (Figure 6B and 6C) resembled that of the isolated native
starch (Figure 6) but with lower intensity; the maximum intensity
for dried powder was 3166 cps. For cured-dried turmeric powder,
the X-ray diffraction pattern did not show any prominent peak.
The percentage crystallinity was 44.6%, 26.3%, and 19.2%
for isolated starch, dried turmeric powder, and cured-dried
turmeric powder, respectively. The curing and/or drying process
on turmeric rhizome might have partially gelatinized the starch
present in them resulting in lower percentage crystallinity values
for dried and cured-dried turmeric powder samples. The thermal
treatment provided to cured-dried turmeric powder has possibly
gelatinized the starch granules resulting in the loss of crystallinity.
The microstructure of freeze-dried turmeric rhizome, turmeric
powders (freeze dried, dried in tray dryer and cured-dried), and

Figure 7–Microstructure of grated freeze-dried turmeric rhizome at (A) 200× and (B) 1000× magnifications, respectively.

Figure 8–Microstructure of isolated starch from fresh turmeric rhizome at (A) 1000× and (B) 5000× magnifications, respectively.

Vol. 76, Nr. 9, 2011 r Journal of Food Science C1289

 Characteristics of turmeric starch . . .
the isolated starch from fresh turmeric rhizome was observed
(Figure 7 to 10). The rhizome showed a porous cellular struc-
ture (Figure 7A) in which the starch granules appeared to be
arrested (Figure 7B) in a matrix. The dimensions of pores are
between 25 and 100 μm. These pores appeared to be irregular
closed spheroids in shape and the cell wall thickness was about
2 μm (Figure 7B). The isolated starch appeared to have a smooth
C: Food Chemistry
surface and was mostly irregular ellipsoid in shape and flaky in
nature (Figure 8A and 8B) similar to that of the mango ginger
Figure 9–Microstructure of freeze-dried (A and B), dried (C and D), and cured-dried (E and F) turmeric powder. Samples A, C, and E: at a magnification
of 1000×. Samples B, D, and F: at a magnification of 5000×.
C1290 Journal of Food Science r Vol. 76, Nr. 9, 2011
(Curcuma amada) and banana starches (Eggleston and others 1992;
Policegoudra and Aradhya 2008). It is interesting to note that this
flaky starch can pack well during the development of compacted
products because of the structural similarity. These starch gran-
ules existed as the packs of several particles, while a few scattered
nonpacked starch particles were also visible.
The dried turmeric powders (Figure 9) appeared to be of ir-
regular shape (Figure 9A and 9C), and consisted mainly of starch
while other components like protein, crude fiber, and fat were also
Characteristics of turmeric starch . . .
Figure 10–Simulated diagram of an isolated starch particle showing char-
acteristic dimensions (a, b, and c).
visible but difficult to identify (Figure 9B and 9D). On the other
hand, the cured-dried turmeric powder underwent the process
of boiling in water such that there was an increase in the size of
particles (Figure 9E and 9F). The outer surfaces appeared to be
uneven due to partial gelatinization and subsequent retrogradation
of starch, and other thermal changes associated with protein and
fat. The expanded granules appeared to be glued together by the
leached out amylose due to the phenomenon of gelatinization.
In addition, the high amylose content (about 48%, w/w) in this
sample also helped this phenomenon.
The application of scanning electron microscope thus helped
in understanding the physical characteristics and morphology of
turmeric rhizome, processed powder, and isolated starch. Police-
goudra and Aradhya (2008) studied the structural features of mango
ginger (C. amada Roxb.) rhizome, a biologically similar rhizome
belonging to the same family (Zingiberaceae) as that of turmeric.
These research works observed a wide variation in the shape and
size of starch particles; the reported size was 3 to 20 μm for small
and 20 to 48 for large size particles. Granule structure, size, and
their distribution are important as they can affect the functional
properties of starch.
The approximate simulated shape of isolated starch is shown in
(Figure 10) which indicated the 3 major dimensions (Equation 3).
The longest dimension observed in the present study was between
15.7 and 24.1 μm, which was much higher than the second largest
dimension (7.3 to 15.1 μm) and the thickness (3.2 to 5.2 μm). The
range of sphericity was between 0.45 and 0.64 having an average
value of 0.53 ± 0.06. These sphericity values thus indicated that
turmeric starch granules were far away from an ideal spherical
shape of 1.00. Braga and others (2006) have reported an elliptical
shape for turmeric starch granules and a size between 10 and
33 μm along the major axes. Jyothi and others (2003) reported it
to be 35 μm and an elliptical shape for the species of C. zedoaria
and C. malabarica.
Conclusions
The turmeric rhizomes can be considered as a rich uncon-
ventional source of starch of commercial interest for specialty
applications. Low-viscosity and shear-thickening behavior was ob-
served for isolated turmeric starch dispersions. X-ray diffractogram
showed the starch to be of B-type. The microstructure of starch
revealed it to possess a smooth surface with regular elliptical shape.
C: Food Chemistry
The high amylose content in turmeric starch is beneficial for the
manufacturing of selected products requiring high crispness and
low expansion. Curcuma starch can be used in agglomerated prod-
ucts as they are flaky in nature and can pack well in a stack due to
structural similarity. These starch granules may be used as a matrix
for encapsulating biomolecules possessing health benefits.
Acknowledgments
The first author (Dhanalakshmi, K.) wishes to thank the Council
of Scientific and Industrial Research (CSIR), New Delhi, India,
for awarding the Senior Research Fellowship (SRF). The help of
Mr. A. Chakkaravarthi is acknowledged for simulation diagram of
starch particles. The work is also partially funded by CSIR under
the Network project (SIP 002).
References
AACC Intl. 1983. Method 56-20. Approved methods of the American Association of Cereal
Chemists, 8th ed. St. Paul, Minn.: AACC Intl.
AOAC Intl. 2005. Official methods of analysis. In: Horwitz W. editor. Official methods of
analysis of Association of Official Analytical Chemists. 18th ed. Gaithersburg, Md.: AOAC
Intl.
ASTA. 1997. Method 2.0. Official analytical methods of the American Spice Trade Association,
4th ed. Washington, D.C.: American Spice Trade Assoc.
Badenhuizen NP. 1964. General method for starch isolation. In: Whistler RL, Smith RJ, Bemiller
JN, Wolform, ML, editors. Methods in carbohydrate chemistry. Vol. 4. New York: Academic
Press. p. 14–5.
Bhattacharya S. 1999. Yield stress and time-dependent rheological properties of mango pulp.
J Food Sci 64:1029–33.
Braga MEM, Moreschi SRM, Meireles MAA. 2006. Effects of supercritical fluid extraction on
Curcuma longa L. and Zingiber officinale R. starches. Carbohyd Polym 63:340–6.
Eggleston G, Swennen R, Akoni S. 1992. Physicochemical studies on starches isolated from
plantain cultivars, plantain hybrids and cooking bananas. Starke 44:121–8.
Govindarajan VS. 1980. Turmeric—chemistry, technology and quality. CRC Crit Rev Food
Sci Nutr 12:199–301.
Hayakawa K, Tanaka K, Nakamura T, Endo S, Hoshino T. 1997. Quality characteristics of waxy
hexaploid wheat (Triticum aestivum L.): properties of starch gelatinization and retrogradation.
Cereal Chem 74:576–80.
Hoover R. 2001. Composition, molecular structure, and physicochemical properties of tuber
and root starches: a review. Carbohyd Polym 45:253–67.
Jobling S. 2004. Improving starch for food and industrial applications. Curr Opin Plant Biol
7:210–8.
Juliano BO. 1985. Rice chemistry and technology. St. Paul, Minn.: American Association of
Cereal Chemists. 443p.
Jyothi AN, Moorthy SN, Vimala B. 2003. Physicochemical and functional properties of starch
from two species of Curcuma. Int J Food Prop 6:135–45.
Khatoon S, Sreerama YN, Raghavendra D, Bhattacharya S, Bhat KK. 2009. Properties of enzyme
modified corn, rice and tapioca starches. Food Res Int 42:1426–33.
Kim SY, Wiesenborn DP, Orr PH, Grant LA. 1995. Screening potato starch for novel properties
using differential scanning calorimetry. J Food Sci 60:1060–5.
Leonel M, Sarmento SBS, Cereda MP. 2003. New starches for the food industry: Curcuma longa
and Curcuma zedoaria. Carbohyd Polym 54:385–8.
Liu Q, Charlet G, Yelle S, Arul J. 2002. Phase transition in potato starch-water system I. Starch
gelatinization at high moisture level. Food Res Int 35:397–407.
Manzan ACCM, Toniolo FS, Bredow E, Povh NP. 2003. Extraction of essential oil and pigments
from Curcuma longa [L.] by steam distillation and extraction with volatile solvents. J Agric Food
Chem 51:6802–7.
Mohsenin NN. 1986. Physical properties of plant and animal materials. New York: Gordon and
Breach Science Publishers. 79 p.
Nara S, Komiy T. 1983. Studies on the relationship between water-saturated state and crystallinity
by the diffraction method for moistened potato starch. Starke 35:407–10.
Okechukwu PE, Rao MA. 1995. Influence of granule size on viscosity of corn starch suspension.
J Texture Stud 26:501–16.
Policegoudra RS, Aradhya SM. 2008. Structure and biochemical properties of starch from
unconventional source-mango ginger (Curcuma amada Roxb.) rhizome. Food Hydrocolloids
22:513–9.
Snedecor GW, Cochran WG. 1968. Statistical methods. 6th ed. Calcutta, India: Oxford and
IBH Publishers. 258 p.
Sowbhagya CM, Bhattacharya KR. 1971. A simplified colorimetric method for determination
of amylose content in rice. Starke 23:53–6.
van Soest JJG, Tournois H, deWit D, Vliegenthart JFG. 1995. Short-range structure in (partially)
crystalline potato starch determined with attenuated total reflectance Fourier-transfrom IR
spectroscopy. Carbohyd Res 279:201–14.
Zobel HF. 1988. Starch crystal transformations and their industrial importance. Starke 40:1–7.
Vol. 76, Nr. 9, 2011 r Journal of Food Science C1291