EVALUATION OF THE FLAVOR QUALITY OF PEPPER

(PIPER NIGRUM L.) CULTIVARS BY GC–MS, ELECTRONIC NOSE

AND SENSORY ANALYSIS TECHNIQUES

B.S. MAMATHA1, M. PRAKASH1,3, S. NAGARAJAN2 and K.K. BHAT1

1
Department of Sensory Science
2
Plantation Products, Spices and Flavor Technology
Central Food Technological Research Institute
Mysore 570020, India

Accepted for Publication August 30, 2007

ABSTRACT

Pepper (Piper nigrum L.) is an extensively used spice, which has a char-
acteristic flavor and pungency. The properties of spices such as flavor, color,
pungency, etc., vary among cultivars and varieties. It is in this context that
pepper cultivars namely, Panniyur 1, Balankotta, Panniyur 5 and one commer-
cial sample were examined for flavor and odor profile using sensory, gas
chromatography–mass spectrometry (GC–MS) and electronic nose (E-nose)
analyses. The flavor profile of pepper powder dispersed in cornstarch gruel
clearly differentiated Balankotta samples from the other three samples; green
mango-like, turmeric-like and earthy notes were higher in Balankotta samples,
while the other samples had higher scores for pepper-like, pungent, spicy and
lingering herbaceous. The flavor profile of the essential oils of pepper samples
showed a higher intensity of pepper-like note in Panniyur 1, Panniyur 5 and
commercial sample, and turmeric-like and green mango-like characterized Bal-
ankotta. The odor profile of the essential oils further supported the flavor profile
data. Orthonasal olfaction (odor profile) provided more descriptive odor char-
acteristics for pepper powder than pepper essential oil. The orthonasal and
retronasal olfaction (flavor profile) showed an opposite trend when the flavor
profile of pepper essential oil samples was carried out in a starch-based carrier;
retronasal olfaction was more effective than the orthonasal. GC, GC–MS analy-
sis and E-nose aroma pattern complemented the sensory flavor profiling results.
The GC–MS of Balankotta pepper samples was different from Panniyur 1, Pan-
niyur 5 and the commercial sample, showing higher content of p-cymene. The
E-nose pattern matching further supported the sensory and instrumental data.

3
Corresponding author. TEL: +91-821-2515842; FAX: +91-821-2517233; EMAIL: mayaprksh@
yahoo.co.in

Journal of Sensory Studies 23 (2008) 498–513. All Rights Reserved.

498 © 2008, The Author(s)
Journal compilation © 2008, Wiley Periodicals, Inc.
 FLAVOR EVALUATION OF PEPPER CULTIVARS 499

PRACTICAL APPLICATIONS

The results of this study provided a protocol for the quality evaluation of
spices, in terms of sensory quality and aroma pattern as determined by gas
chromatography–mass spectrometry. Introducing the electronic nose tech-
nique for the rapid evaluation of spice aroma, as well as characterization of
spices, was an added information. The results of this study gave the odor
description of major compounds present in pepper essential oil and the
regional variations in odor profiles, which can help in designing spice blends
with specific flavor profiles.

INTRODUCTION

Black pepper, known as the king of spices, is the most popular and most
widely used spice in the world. It is extensively used for flavoring and pre-
serving processed foods and has medicinal properties. Western coastal regions
of south peninsular India is the traditional home of pepper. In India, pepper is
cultivated in an area of around 181,500 hectares with an annual production of
60–80,000 tons, which was reported by Ravindran and Johny (2000). The
cultivars of black pepper have originated from the wild types; more than
hundreds of cultivars are known and a few of them are popular (Ravindran
2000). Gopalkrishnan et al. (1993) described the odor evaluation for Panniyur
1, Panniyur 2, Panniyur 3 and Panniyur 4 of pepper cultivars. They depicted
the profile on a 4-point category scale and subjected the oils to ranking tests.
The aromagram developed by these authors has the desirable odors of pepper
in the upper quadrant and the undesirable odors in the lower quadrant depict-
ing the quality of pepper samples.
The quality of black pepper is as important as yield and depends on the
content of piperine and essential oil. The components of pepper contributing
to its value as a food additive are the essential oil for aroma and alkaloid
compounds for pungency. The characteristics of pepper alkaloids and the
chemistry of the volatile oil were reported by Govindarajan (1977). The author
reported that sensory evaluation with trained panels and reference standards is
the only technique available for the evaluation of the degree of superiority of
the aroma quality of pepper. More than 80 components have been reported in
pepper essential oil. Lewis et al. (1969) studied 17 cultivars from Kerala
(India) and found that the oil content ranged from 2.4 to 3.8%. Russel and Else
(1973) found a significant difference in the oil content and chemical compo-
sition of different varieties of black pepper. Richard et al. (1971) analyzed the
pepper samples from Lampong and Sarawak. Pangborn et al. (1970) studied
the sensory evaluation of Malabar pepper oil after column chromatographic
500 B.S. MAMATHA ET AL.

fractionation and indicated that the early fractions were pepper-like and floral
and the late fractions were pepper-like and woody. An aroma model for pepper
was developed based on the quantification of 19 odorants, and the calculation
of their odor activity values was studied by Jagella and Grosch (1999).
Narasimhan et al. (1992) conducted studies on the quality of powdered black
pepper during storage by gas chromatography (GC) analysis and sensory
analysis. Gopalkrishnan et al. (1993) depicted the odor profile of pepper
samples on a 4-point category scale and subjected the oil to ranking tests.
However, descriptive odor and flavor analysis by sensory methods are lacking.
Peryam and Swarty (1951) demonstrated by consumer-type preference
test that pepper was of considerable significance for enhancing the flavor
quality of different foods. The authors found significant differences among
genuine pepper, extracts and artificial pepper by the paired comparison tech-
nique. Aroma is an integrated response to a mixture of components and
thresholds of perception.
Reports were available on the effect of different dispersal media on the
aroma of the oil or fractions of the oil (Pangborn et al. 1970). Despite poor
dispersal in water, thin medium was found to be the best for reflecting quality.
The “musty” note was accentuated in the salt medium (Govindarajan 1977).
These studies emphasized the importance of selecting a suitable medium for
sensory evaluation trials. Electronic nose (E-nose) pattern matching was
carried out for discriminating the aroma of pepper cultivars. E-nose, like the
human nose, makes a global analysis of vapor emitted from a sample and
performs a classification process by comparing the samples with a database. It
performs a quick assessment of aroma quality and is being used by the food
industry to carry out quality control and product development. E-nose is
defined as an instrument that comprises an array of electronic chemical
sensors, appropriate pattern recognition system, capable of recognizing simple
and complex odors. E-nose is a tool that complements the sensory and instru-
mental data of flavors and aroma of foods. E-nose testing was used for
analyzing coffee aroma and instant coffee powder of different origins (Gretsch
et al. 1998) and quality of tea (Lucas et al. 1998). Madsen and Grypa (2000)
used E-nose to determine the origin of spices and compare formulae compo-
nents in the system. Vanilla flavor evaluation by sensory and E-nose techniques
was reported by Hariom et al. (2006). Korel et al. (2002) carried out on using
E-nose to discriminate ground red pepper samples by headspace volatiles. This
study showed that E-nose was able to group ground red pepper having differ-
ent capsaicin, dihydrocapsaicin and total capsaicinoids level, using the dis-
criminant function analysis as a pattern recognition technique. E-nose is used
by an appropriated pattern recognition technique to identify the odor through
comparison with previously obtained measurements of known samples
(Barnett 1999; Strike et al. 1999). In this context, quality evaluation through
 FLAVOR EVALUATION OF PEPPER CULTIVARS 501

aroma pattern matching was carried out for pepper samples. The objective of
the present study was to assess the odor and flavor quality characteristics of
different cultivars of pepper by sensory and instrumental analysis. The results
of the gas chromatography–mass spectrometry (GC–MS), E-nose and sensory
studies are presented here.

MATERIALS AND METHODS

Three samples of pepper cultivars namely, Panniyur 1, Panniyur 5 and

Balankotta were procured from the Indian Spice Research Institute, Madikeri,
Karnataka, India, and one sample was commercially bought from the local
market. The samples were packed in an airtight container and stored under
refrigerated condition (4–6C) during the course of the study.

Sample Preparation
The pepper samples were powdered in a laboratory dry grinder and
passed through a 60-mesh (Bureau of Standards, India) sieve. Essential oil
from pepper was extracted by Clevenger’s extraction method (ASTA 1985). In
this, 50 g of pepper powder was taken in a round-bottomed flask containing
500 mL of water, which is connected to Clevenger’s distillation unit and
heated on a mantle for about 4 h. The distilled essential oil collected in the side
arm was separated and dried over anhydrous sodium sulfate. The extracted oil
was stored in the refrigerator during the course of the study.

Panel Training
A group of 10–12 panelists aged 25–45 years was trained over three
sessions for psychometric studies and quantitative descriptive analysis (QDA)
test. The members of the panel were drawn from the scientific staff familiar
with sensory analysis techniques and who had earlier experiences in the
sensory evaluation of spices. The samples were evaluated in a sensory booth
room maintained at a temperature of 22 ⫾ 2C under fluorescent lighting
equivalent to daylight. Details of the panel training have been included
separately under the different sensory tests.

Time Intensity Study

Preliminary observation revealed differences in flavor intensity of the
four samples of pepper powder and essential oils. Flavor impact and its
persistence were determined by recording the changes in perceived flavor
intensity with respect to time (ASTM 1996). The panelists were trained to
502 B.S. MAMATHA ET AL.

mark the flavor intensity on a structured QDA scale of 15 cm anchored at 1.25

(recognition threshold) and 13.75 cm (saturation threshold). Pepper powder
(0.2%) and essential oil (0.1%) dispersion in cornstarch gruel were served in
25-mL beakers. The panelists were asked to mark the intensity of the perceived
sensation on the scorecard as soon as they take the samples in their mouths.
This marks the onset of flavor sensation. After holding the samples in their
mouths for 10 s, the panelists were asked to swallow and immediately mark
the intensity. Further, lingering flavor intensity scores were marked after every
10 s up to 60 s. Time intensity profile was obtained by plotting mean scores of
pepper flavor versus time in seconds. The study was taken up to evaluate the
lingering effect of pepper flavor with time, which gives an indication of the
quality of the spice with respect to its flavor principles.

Odor Analysis
Pepper powder (2 g) or essential oil (0.1 g) was taken in a 250-mL
Erlenmeyer flask. The sample was covered with 5-mm-thick cotton layer, and
the flask was stoppered, allowing the accumulation of volatiles in the head-
space. Odor profiling of the pepper powder and essential oil was carried out
using the QDA method (Stone and Sidel 1998). The panelists who were
involved in the earlier test participated in the evaluation. They were trained to
sniff the headspace and distinguish various odor notes. In the preliminary
session, the panelists were asked to list the odor descriptors perceived by
sniffing. The scorecard for odor analysis contained common descriptors
selected by at least one-third of the panel and a few important descriptors cited
in the literature.
The panelists were asked to mark the intensity of the attributes on QDA,
which consisted of a 15-cm line scale, wherein 1.25 cm was anchored as low
(recognition threshold) and 13.75 cm as high (saturation threshold). The pan-
elists were asked to mark a vertical line on the scale and write the code of the
sample close to the line. In between two evaluations, enough time (15 min)
was given for the accumulation of volatiles in the headspace. The scorecards
were decoded, and the mean values of the attribute were calculated.

Flavor Analysis
The samples for the sensory flavor evaluation were prepared in starch
gruel as a medium. The gruel was prepared by dispersing cornstarch (2 g) in
100 mL water to which 0.25 g table salt was added. The dispersion was heated
until a homogeneous gelatinized liquid was obtained. To this gruel, 0.05 g of
pepper powder (60 mesh) or 0.02 g of essential oil was added and thoroughly
stirred to get a uniform dispersion.
 FLAVOR EVALUATION OF PEPPER CULTIVARS 503

TABLE 1.
DESCRIPTORS AND REFERENCE COMPOUNDS USED FOR
THE SENSORY FLAVOR PROFILING OF PEPPER
ESSENTIAL OILS

Refreshing Menthol, myrcene

Herbaceous Isopropyl-quinoline
Spicy Piperine
Earthy Isopropyl-quinoline
Lemony Citral
Aromatic Benzaldehyde
Pepper-like Pepper powder
Turmeric Turmeric powder
Camphory DL-camphor
Citrussy Limonene
Green mango-like Iso-cyclocitral
Fruity Aldehyde

As a prerequisite for the flavor profiling of pepper, the panelists were

asked to give as many descriptors as applicable. Following this, an open
discussion was held to reach an agreement on appropriate descriptors, espe-
cially for aroma and flavor. For this, the guidelines described by Dravinieks
(1985) and Jellinek (1964) were used. The trained panelists who were involved
in the earlier tests participated in this evaluation. The descriptors used for the
profiling of pepper and the reference compounds used to train the panelists are
given in Table 1. Samples (15 mL) at 40C were served in a 25-mL beaker with
three-digit codes. The samples were served one by one in a random order. The
panelists were asked to mark the intensity of the attribute on the QDA scale.

GC
GC analysis of pepper essential oil volatiles was carried out by using a
Shimadzu GC-SB chromatography (Kyoto, Japan) with a capillary column
SPB-1 (Supelco Park, Bellefonte, PA; 30 m ¥ 0.32 mm id/film thickness
0.25 mm). The analysis was carried out using the program of oven temperature
at 60C followed by raising at 2C/min to 180C and was kept at 100C; the
detection temperature was at 200C. The carrier gas was nitrogen and the
detector was a flame ionization detector. Analysis was carried out with a
sample quantity of 0.1 mL dissolved in ethyl acetate.

GC–MS Analysis
Shimadzu GC-17A coupled with a QP5000 MS was used for the analysis
of volatiles. Volatile oil of 1 mL pepper (diluted 0.1–2 mL in acetone) was
504 B.S. MAMATHA ET AL.

FIG. 1. CHROMATOGRAM OF BALANKOTTA PEPPER ESSENTIAL OIL

injected to the SPB-1 column (Supelco; 30 m ¥ 0.32 mm id, film thickness

0.25 mm). The analysis was carried out under the following conditions: oven
temperature was programed at 60C followed by raising at 2C/min to 180C;
injection port temperature was at 100C and detector was at 220C; the carrier
gas was helium; the flow rate was 1 mL/min. The volatile constituents were
identified by computer matching of their mass spectral fragmentation pattern
of compounds in the NIST-MS library, along with Kovats indices by analyzing
standard hydrocarbons (Figs. 1 and 2) and published mass spectra (Adams
1989).

E-nose Analysis
E-nose (Alpha Fox 3000, Alpha M.O.S. SA, Toulouse, France),
equipped with six doped and six undoped metal oxide semiconducting
sensors, was used. Pepper powder (0.02 g) and pepper essential oil (0.01 g)
were taken in sample vials, and the volatiles were allowed to accumulate in
the headspace by holding the vials at 25C for 120 s. The volatiles were
carried by a stream of zero air (flow rate 150 mL/min) to the sensor
chamber. The injection time was 60 s, and the acquisition time was 120 s.
The sensors in the E-nose essentially measure the change in voltage because
of the presence of volatile odorous molecules, and the responses were then
analyzed by the software to get an olfactive picture of the product. E-nose
utilizes the data with a preprocessor, which is analogous to the olfactory
bulb in the human olfactory region.
 FLAVOR EVALUATION OF PEPPER CULTIVARS 505

50.23 55.32 19309

44.88

33.40 38.27

27.28

21.05

15.08
9.82
6.92
60.07

Time
34.25 44.25 54.25 64.25

FIG. 2. GAS CHROMATOGRAPHY–MASS SPECTROMETRY ANALYSIS OF

STANDARD HYDROCARBONS

Statistical Analysis
Time intensity study, odor profile and flavor profile studies were carried
out in duplicate, and the data were subjected for statistical significance by
Duncan’s multiple range test at a significance level of P ⱕ 0.05 (Duncan
1955).
506 B.S. MAMATHA ET AL.

RESULTS AND DISCUSSION

Time Intensity Profile

The time intensity aroma profile for pepper essential oil samples is shown
in Fig. 3. It can be seen from the figure that Panniyur 1 had the highest
lingering pepper aroma, followed by Panniyur 5 and the commercial pepper
oil. Panniyur 1 had an intensity score of 13 at 10, which reduced to 8 after 60.
Balankotta oil had the least intensity of 11 at 10 and reduced to 3.5 after 60,
indicating that Balankotta oil samples did not have intense pepper-like notes
and significantly differed from the other three samples (P ⱕ 0.05) up to 30 s.
By the end of 60 s, all four pepper samples differed significantly for the aroma
intensity (P ⱕ 0.05). Aroma intensity was not reported for 0 s as it was below
the threshold level.

Odor Profile
The odor profiling showed significant differences between the pepper
varieties. Balankotta had intense typical turmeric-like and green mango-like
notes, but it has less irritant, woody, herbal and pepper-like odor notes. Pan-
niyur 1, Panniyur 5 and the commercial sample had almost identical odor
profiles except for the citrussy note, which was more in the commercial sample
(Fig. 4). The Balankotta sample differed significantly (P ⱕ 0.05) from the
other samples for citrussy, turmeric-like, pepper-like, green and irritant/sharps

FIG. 3. TIME INTENSITY PROFILE OF PEPPER AROMA

Different superscripts differ significantly (P ⱕ 0.05).
 FLAVOR EVALUATION OF PEPPER CULTIVARS 507

FIG. 4. ODOR PROFILE OF PEPPER POWDER SAMPLES

* Differ significantly at P ⱕ 0.05.

shown in Fig. 4. However, Panniyur 1, Panniyur 5 and the commercial sample

had higher scores for overall quality compared with the Balankotta sample
because of the presence of the desirable characteristic pepper-like aroma.
The descriptors such as citrussy, turmeric-like, pepper-like, green-like and
irritant/sharp differed significantly between Balankotta and the other samples
(P ⱕ 0.05). The mean scores for each odor characteristics were plotted in
sequential order, showing the desirable odor like pepper-like, citrussy and
turmeric in the upper quadrants and the undesirable ones like green, herbal,
woody and rooty in the lower quadrant. The odor profile of the essential oil did
not generate much descriptors in describing odor characteristic of pepper and
hence, the data were not presented. It can be concluded that orthonasal olfac-
tion was not effective for pepper essential oil but more effective for pepper
powder.

Flavor Profile
Flavor profiling showed a significant difference (P ⱕ 0.05) between the
samples. The pepper-like, green mango-like, earthy and turmeric-like notes
were significantly (P ⱕ 0.05) differed for the Balankotta samples from other
samples (Panniyur 1, Panniyur 5 and commercial) as given in Table 2. The
characteristic pungency of pepper notes was significantly different between
the pepper cultivars (P ⱕ 0.05). Jagella and Grosch (1999) carried out some
studies on the flavor- and off-flavor-contributing components of black pepper
508 B.S. MAMATHA ET AL.

TABLE 2.
FLAVOR PROFILE OF PEPPER POWDER

Attribute Panniyur 1 Panniyur 5 Commercial Balankotta

Refreshing 9.0c 10.1d 7.0a 8.6b

Pepper-like 10.5b 10.0b 10.2b 8.5a
Green mango-like 8.5a 8.9a 8.2a 11.5b
Fruity 5.5b 6.5c 6.0bc 3.0a
Pungent 11.0d 9.0b 10.5c 5.5a
Spicy 10.3c 9.5b 10.2c 6.6a
Turmeric-like 4.8a 4.5a 5.5a 11.0c
Earthy 6.2a 6.5a 6.3a 10.4b
Lingering herbaceous 7.0b 7.2bc 7.8c 5.5a

Values with different superscripts in the same row differ significantly (P < 0.05).

TABLE 3.
FLAVOR PROFILE OF THE ESSENTIAL OIL OF PEPPER SAMPLES

Attribute Panniyur 1 Panniyur 5 Commercial Balankotta

Lemony 3.5a 4.0b 4.5c 5.2d

Aromatic 8.0ab 7.5a 8.2b 7.3a
Herbaceous 7.5b 7.2ab 6.8a 6.6a
Pepper-like 11.5bc 11.0b 11.8c 7.0a
Turmeric 5.0a 5.5a 4.6a 11.0b
Camphory 7.2b 7.2b 7.5b 5.5a
Citrussy 5.3a 5.2a 5.8a 6.5b
Balsamic 6.5a 6.5a 6.3a 6.8a
Green 4.0a 4.5a 4.2a 7.5b

Values with different superscripts in the same row differ significantly (P < 0.05).

and concluded that (+)-linalool, (+)-a-phellandrene, (-)-limonene, myrcene,

(-)-a-pinene, 3-methylbutanol and methyl propanol were the most odorant
of black pepper, and the moldy, musty off-flavor of Malaysian pepper was
shown to be because of 2-isopropyl-3-methoxypyrazine and 2,3-diethyl-5-
methylpyrazine.
When the flavor profiling was carried out for the pepper essential oil in
2% cornstarch gruel, different flavor descriptors were identified as shown in
Table 3. Retronasal olfaction gave a more descriptive olfaction than the ortho-
nasal olfaction for pepper essential oil samples. Pepper essential oil in corn-
starch gruel as a matrix effectively enhanced the flavor notes. Lemony and
pepper-like notes were significantly different among the pepper cultivars
(P ⱕ 0.05). The typical turmeric, citrussy and green mango-like notes were
found to be more in Balankotta, and no significant differences were seen for
 FLAVOR EVALUATION OF PEPPER CULTIVARS 509

TABLE 4.
GAS CHROMATOGRAPHY ANALYSIS OF THE PEPPER ESSENTIAL OIL

Compounds Panniyur 1 %* Panniyur 5 %* Balankotta %* Commercial %* KI

Thujene 2.96 ⫾ 0.01 2.78 ⫾ 0.01 – 2.35 ⫾ 0.01 925

a-Pinene 7.74 ⫾ 0.02 7.08 ⫾ 0.02 20.94 ⫾ 0.01 9.96 ⫾ 0.01 935
b-Pinene 21.18 ⫾ 0.01 22.34 ⫾ 0.01 – 24.42 ⫾ 0.02 971
Myrcene 13.79 ⫾ 0.03 12.32 ⫾ 0.01 13.46 ⫾ 0.01 15.18 ⫾ 0.01 976
Unidentified 3.06 ⫾ 0.03 2.52 ⫾ 0.01 2.62 ⫾ 0.02 – –
a-Phellandrene 1.26 ⫾ 0.01 – – – 1,002
a-3-Carene 3.38 ⫾ 0.01 2.32 ⫾ 0.01 11.67 ⫾ 0.02 – 1,013
p-Cymene – – 8.16 ⫾ 0.01 – 1,022
g-Terpinene 4.70 ⫾ 0.02 3.86 ⫾ 0.02 – – 1,031
Limonene 21.11 ⫾ 0.01 20.32 ⫾ 0.01 25.17 ⫾ 0.01 26.49 ⫾ 0.03 1,032
a-Terpinene 1.03 ⫾ 0.02 – – – 1,064
a-Cubebene 2.20 ⫾ 0.02 1.95 ⫾ 0.02 – 3.50 ⫾ 0.01 1,404
b-Caryophyllene 10.59 ⫾ 0.01 17.79 ⫾ 0.02 7.71 ⫾ 0.02 2.42 ⫾ 0.01 11,446
(E, E-a) farnesene 5.90 ⫾ 0.02 16.73 ⫾ 0.01 – 4.62 ⫾ 0.02 1,532
Caryophyllene oxide – – – 6.23 ⫾ 0.01 1,548
Unidentified 1.08 ⫾ 0.01 – – – –
Cedrol – – 3.24 ⫾ 0.02 – 1,604

* Values of triplicate samples.

KI, Kovats indices.

the other three pepper samples. Development in the physiology of perception

of flavor stimuli, psychophysical measurement of intensity and a better under-
standing of the sources of variation in subjective assessments through the use
of a trained panel and statistical analysis of data have led to sensory evaluation
procedures yielding reliable and valid data.

GC and GC–MS Analyses

GC–MS analysis identified 16 characteristic compounds present in
pepper varieties (Table 4). b-Pinene, limonene and b-caryophyllene have
contributed major composition. Literature survey indicates that p-cymene
levels in different pepper varieties were in the range of 0.74–2.3% (Govin-
darajan 1977). But in the present study, only Balankotta pepper showed 8.16%
of p-cymene. This compound is not found in other pepper samples such as
Panniyur 1, Panniyur 5 and commercial samples. Studies reported that Pan-
niyur varieties are superior to other common varieties of pepper (Ravindran
2000). Hence, it is observed that the high content of p-cymene might be
attributed to the different sensory perceptions in Balankotta pepper samples,
indicating green mango-like and turmeric-like notes. Some of the reports
indicated that the typical p-cymene has citrussy reminiscent to lemon-like top
510 B.S. MAMATHA ET AL.

FIG. 5. ELECTRONIC NOSE PATTERN MATCHING OF PEPPER POWDER SAMPLES

(A) Panniyur 1; (B) Panniyur 5; (C) Balankotta; and (D) market samples.
C1, component 1; C2, component 2.

note (Archander 1969) and of carrot (Furia and Bellanca 1975). The absence
of certain monoterpenic hydrocarbons and the presence of cedrol, which has a
cedar tree-like aroma (Furia and Bellanca 1975), in the Balankotta sample may
contribute to its different flavor profile. Although the instrumental and sensory
analysis can be correlated with constituents, it is reasonable to conclude that
the aroma is because of an integrated response to a mixture of components
with different characteristics and threshold of perception. The sensory and
instrumental analysis techniques have differentiated the flavor profile of dif-
ferent pepper cultivars including market sample.

E-nose Analysis
Data analysis in the E-nose was carried out based on the value of
maximum change in resistance of the sensors by the principal component
analysis (PCA) using the built-in software supplied by the manufacturers. The
results are shown as a two-dimensional representation of PCA with respect to
two axes. The E-nose pattern matching has discriminated the four pepper
samples, showing the discrimination between them. The PCA pattern of
E-nose data for the powder of the pepper cultivars is shown in Fig. 5. PC1
accounted for major differences (96.38%) in variances, while PC2 accounted
for minor differences (1.90%). As expected, the powder of the pepper cultivars
 FLAVOR EVALUATION OF PEPPER CULTIVARS 511

FIG. 6. ELECTRONIC NOSE PATTERN MATCHING OF PEPPER ESSENTIAL OIL SAMPLES

(A) Panniyur 1; (B) Panniyur 5; (C) Balankotta; and (D) market samples.
C1, component 1; C2, component 2.

showed discrimination among the four samples. The Balankotta pepper

variety, which had a different sensory flavor perception and GC-MS pattern,
was grouped in a separate quadrant, thus supporting the sensory and instru-
mental results. In the case of the pepper essential oil of the four cultivars as
given in Fig. 6, PC1 accounted for major differences as 99.25% in variances.
Similar discrimination for Balankotta was depicted. Therefore, E-nose analy-
sis can be used as a rapid method for testing the quality of spices. Reports also
indicated a similar discussion for ground red pepper samples using E-nose
analysis and concluded that E-nose has the potential to be used as a rapid and
easy method for determining the pungency of ground red pepper with no
sample preparation. The study gave an insight into the different odor notes
present in pepper essential oil, and the information generated can be used in
selecting the right type of pepper for use in a product to provide the required
flavor notes.

CONCLUSIONS

The flavor profile of pepper cultivars showed a clear difference between

Balankotta samples from the other three samples namely, Panniyur 1, Panniyur
5 and commercial. The odor profile of pepper essential oil supported the flavor
512 B.S. MAMATHA ET AL.

profile analysis. The GC–MS analysis of Balankotta samples showed a higher

content of p-cymene compared with the other pepper cultivars. The E-nose
pattern matching further supported the sensory and instrumental data showing
the discrimination between the samples. Thus, the present study gave an
insight into the different flavor compounds present in pepper cultivars, and the
information generated can be used in selecting the right type of pepper for use
in a product to provide the required flavor notes.

REFERENCES

ADAMS, R.P. 1989. Identification of Essential Oils by Ion Trap Mass

Spectroscopy, Academic Press, San Diego, CA.
ARCHANDER, S. 1969. Perfume and Flavor Chemicals, I, p. 819, Steffen
Archander, Montclair, NJ.
ASTA. 1985. Official Analytical Methods for Spices, 2nd Ed., American Spice
Trade Association, New York, NY.
ASTM. 1996. Sensory testing methods. In ASTM Manual 26, 2nd Ed. (E.
Chamber IV and M.B. Wolf, eds.) pp. 54–72, ASDTM, West Consho-
hocken, PA.
BARNETT, D. 1999. The electronic nose and food assessment. Food Aust. 51,
60–226.
DRAVINIEKS, A. 1985. Atlas of Odor Character Profiles, ASTM Data Series
DS 61, American Society for Testing and Materials, Philadelphia, PA.
DUNCAN, D.B. 1955. Multiple range test and multiple F tests. Biometrics 11,
1–42.
FURIA, T.E. and BELLANCA, N. 1975. Fenaroli’s Handbook of Flavor
Ingredients, 2nd Ed., CRC Press, Cleveland, OH.
GOPALKRISHNAN, M., MENON, N., PADMAKUMARI, K.P., JAYALAK-
SHMI, A. and NARAYAN, C.S. 1993. GC analysis and odor profile of
four new Indian genotypes of Piper nigrum L. J. Essent. Oil Res. 5,
247–253.
GOVINDARAJAN, V.S. 1977. Pepper-chemistry, technology, and quality
evaluation. Crit. Rev. Food Sci. Nutr. 9(2), 115–225.
GRETSCH, C., TOURY, A., ESTEBANAUZ, R. and WIARDON, R. 1998.
Sensitivity of metal oxide sensors towards coffee aroma. Semin. Food
Anal. 3(1), 37–42.
HARIOM, B.N., SHAYAMALA, M., PRAKASH, M. and BHAT, K.K. 2006.
Vanilla flavor evaluation by sensory and electronic nose techniques.
J. Sensory Studies 21, 228–239.
JAGELLA, T. and GROSCH, W. 1999. Desirable and undesirable odorants of
white pepper. Eur. Food Res. Technol. 209, 27–31.
 FLAVOR EVALUATION OF PEPPER CULTIVARS 513

JELLINEK, G. 1964. Introduction to critical and review of modern methods of

sensory analysis (odor, taste and flavor evaluation) with special emphasis
on descriptive analysis (flavor profile methods). J. Nutr. Diet. 1, 219.
KOREL, F., BAGDATLIOGLU, N., BALABAN, M.O. and HISIL, Y. 2002.
Ground red peppers: Capsaicinoids contents, Scoville scores and dis-
crimination by an electronic nose. J. Agric. Food Chem. 50, 3257–3261.
LEWIS, Y.S., NAMBODIRI, E.S. and KRISHNAMURTHY, N. 1969. Com-
position of pepper oil perfume. J. Essent. Oil Res. 60, 259–262.
LUCAS, Q., POLING, J. and BENINCASA, V. 1998. Analysis with a hybrid
system QCM/CP/MOS. Semin. Food Anal. 3(1), 53–58.
MADSEN, M.G. and GRYPA, R.D. 2000. Spices flavor system and the
electronic nose. Food Technol. 54(3), 44–46.
NARASIMHAN, S., RAJALAKSHMI, D. and CHAND, N. 1992. Quality of
powdered black pepper (Piper nigrum L.) during storage II. Principle
components analysis of GC and sensory profiles. J. Food Qual. 15, 67–83.
PANGBORN, R.M., JENNINGS, W.G. and NOELTING, C.E. 1970. Prelimi-
nary examination of odor quality of black pepper oil. Flavor Ind. 1, 763.
PERYAM, D.R. and SWARTY, V.N.W. 1951. Methodology for sensory evalu-
ation of imitation pepper. Food Technol. 5, 207.
RAVINDRAN, P.N. 2000. Introduction. In Black Pepper (P.N. Ravindran, ed.)
pp. 1–23, Harwood Academic, Amsterdam, the Netherlands.
RAVINDRAN, P.N. and JOHNY, A.K. 2000. Black pepper. In Handbook of
Herbs and Spices (K.V. Peter, ed.) pp. 62–110, CRC Press, Woodhead
Publishing Limited, Cambridge, England.
RICHARD, H.M., RUSSEL, G.F. and JENNINGS, W.G. 1971. The volatile
components of black pepper varieties. J. Chromatogr. Sci. 9, 560–566.
RUSSEL, G.F. and ELSE, J. 1973. Volatile compositional differences between
cultivars of black pepper (Piper nigrum L.). J. Assoc. Off. Anal. Chem.
56, 344–351.
STONE, H. and SIDEL, J. 1998. Quantitative descriptive analysis, develop-
ments, application and the future. Food Technol. 52(8), 48–52.
STRIKE, D.J., MEIJERINK, M.G.H. and KONDELDA-HEP, M. 1999.
Electronic noses – a mini review. J. Anal. Chem. 364, 499–505.