Food Chemistry 167 (2015) 264271

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Determination of sucrose content in sugar beet by portable visible and

near-infrared spectroscopy
Leiqing Pan a, Qibing Zhu b, Renfu Lu c,, J. Mitchell McGrath d
a

College of Food Science and Technology, Nanjing Agricultural University, No. 1 Weigang Road, 210095 Nanjing, China
School of Communication and Control Engineering, Jiangnan University, 1800 Lihu Road, 214122 Wuxi, China
c
USDA Agricultural Research Service, Sugarbeet and Bean Research Unit, Michigan State University, 524 S. Shaw Lane, East Lansing, MI 48824, USA
d
USDA Agricultural Research Service, Sugarbeet and Bean Research Unit, Michigan State University, 1066 Bogue Street, East Lansing, MI 48824, USA
b

a r t i c l e

i n f o

Article history:
Received 26 February 2014
Received in revised form 10 June 2014
Accepted 29 June 2014
Available online 5 July 2014
Keywords:
Near-infrared spectroscopy
Sugar beet
Sucrose content
Detection
High-performance liquid chromatography

a b s t r a c t
Visible and near-infrared spectra in interactance mode were acquired for intact and sliced beet samples,
using two portable spectrometers for the spectral regions of 4001100 nm and 9001600 nm, respectively. Sucrose prediction models for intact and sliced beets were developed and then validated. The spectrometer for 4001100 nm was able to predict the sucrose content with correlations of prediction (rp) of
0.80 and 0.88 and standard errors of prediction (SEPs) of 0.89% and 0.70%, for intact beets and beet slices,
respectively. The spectrometer for 9001600 nm had rp values of 0.74 and 0.88 and SEPs of 1.02% and
0.69% for intact beets and beet slices. These results showed the feasibility of using the portable spectrometer to predict the sucrose content of beet slices. Using simple correlation analysis, the study also identied important wavelengths that had strong correlation with the sucrose content.
Published by Elsevier Ltd.

1. Introduction
Sugar beet is grown in a wide range of climatic conditions and
in about 50 countries worldwide, including North America (United
States and Canada), South America (Chile), Asia, North Africa (Morocco and Egypt), and most of Europe (Mosen, 2007). Sucrose content is the most important trait in sugar beet production, and it
is made up of more than 99.5% in the nal white crystalline sugar.
Hence breeders and researchers are striving to achieve high content of sucrose in beet production, using genetic, molecular, and
conventional breeding approaches.
Rapid measurement of sucrose content in sugar beets can assist
breeders in selecting promising germplasms and help sugar beet
growers and processors in determining the yield and quality of
beets after harvest and during storage and processing. Many methods have been used to measure the sucrose content of beets,
including polarimetry, enzyme-based spectroscopic assays, and
high-performance liquid chromatography (HPLC) (McGrath &
Fugate, 2012). Polarimetry is the generally accepted method used
in commercial sugar beet processing factories. Newer generation
polarimetric instruments can measure sucrose from dark and
coloured samples of molasses without juice clarication
Corresponding author. Address: 524 S. Shaw Lane, Room 224, Michigan State
University, East Lansing, MI 48824, USA. Tel.: +1 517 432 8062.
E-mail address: renfu.lu@ars.usda.gov (R. Lu).
http://dx.doi.org/10.1016/j.foodchem.2014.06.117
0308-8146/Published by Elsevier Ltd.

(Singleton, Horn, Bucke, & Adlard, 2002). Enzyme-based spectroscopic assays utilise a series of enzyme-catalysed reactions to
quantitatively couple sucrose reaction to the synthesis of a spectrally detectable compound (Spackman & Cobb, 2002). A relatively
rapid and inexpensive enzymatic-uorometric microtitre plate
assay was developed for sucrose quantication (Trebbi &
McGrath, 2004). The method provided accurate and sensitive
sucrose measurements from the tissues of young sugar beet roots
with a coefcient of determination (r2) of 0.976, but it was less
accurate for older, eld-grown root tissues (r2 = 0.605). Owing to
its high sensitivity and specicity, HPLC is also used in the sucrose
analysis of sugar beet. But the technique is time-consuming and
labour intensive in sample preparation and sequential analysis
(12 min per sample) (Mulcock, Moore, Barnes, & Hickey, 1985).
Numerous studies have been reported in recent years on using
visible and near-infrared (Vis/NIR) spectroscopy for the spectral
region of 4002500 nm for fast measurement of soluble solids content and other quality attributes of apple (Fan, Zha, Du, & Gao,
2009; Mendoza, Lu, & Cen, 2012), peach (Carlomagno, Capozzo,
Attolico, & Distante, 2004), pear (Xu, Qi, Sun, Fu, & Ying, 2012),
pineapple (Chia, Abdul Rahim, & Abdul Rahim, 2012), Satsuma
mandarin (Gmez, He, & Pereira, 2006), sweet cherry (Lu, 2001),
and many other fruits (Camps & Christen, 2009; Cayuela &
Weiland, 2010; McGlone, Jordan, Seelye, & Martinsen, 2002; Paz,
Snchez, Prez-Marn, Guerrero, & Garrido-Varo, 2009; Wang,
Nakano, & Ohashi, 2011). NIR reectance spectroscopy was used

L. Pan et al. / Food Chemistry 167 (2015) 264271

to determine sucrose content in ground soybean samples (Sato,

Zahlner, Berghofer, Lok, & Vollmann, 2012). NIR spectroscopy
was reported for achieving excellent predictions of sucrose in chocolate, with a correlation coefcient of 0.997 or higher (Da Costa
Filho, 2009). Vis/NIR spectroscopy was also reported for measuring
the sucrose content of sugarcane. Mat Nawi, Chen, Jensen, and
Mehdizadeh (2013) applied visible and shortwave near-infrared
(Vis/SWNIR) spectroscopy to predict the sugar content of sugarcane based on skin scanning. They reported an r2 value of 0.91
and a root mean square error of prediction of 0.721 Brix. Roggo,
Duponchel, and Huvenne (2004) measured the sucrose content of
more than 2700 homogenised beet brei samples from 15 different
factories using Vis/NIR (4002500 nm) and reported excellent
results with a standard error of prediction of 0.10%.
While previous Vis/NIR studies have showed promising results
for sucrose measurement in beets, they needed to destroy and
homogenise beet samples. In addition, these studies were carried
out using expensive laboratory Vis/NIR instruments that cover
the entire visible and near-infrared region (i.e., 4002500 nm).
No research has been reported on measuring the sucrose content
of intact and sliced beets using spectroscopic instruments that only
cover the Vis/SWNIR region of 4001100 nm or a portion of the NIR
region (i.e., 9001600 nm). Spectral measurements for these two
spectral regions offer considerable advantages over the full Vis/
NIR region because they can be done more quickly and conveniently, using low-cost portable or handheld CCD-based (chargecoupled device) (4001100 nm) or InGaAs-based (indium gallium
arsenide) (9001600 nm) spectrometers.
Therefore, the objective of this research was to study the feasibility of Vis/SWNIR and NIR spectroscopy for predicting the
sucrose content of intact and sliced sugar beets. The specic objectives were to: (1) measure spectra of intact and sliced sugar beets
using two portable spectrometers operated in interactance mode,
for the spectral regions of 4001100 nm and 9001600 nm, respectively; (2) develop statistical models for the spectral data to predict
the sucrose content of intact beets and beet slices; and (3) identify
important wavelengths that have strong contributions to the
sucrose content prediction.

2. Materials and methods

2.1. Samples
Sugar beets used for this experiment were harvested from the
Michigan Sugar Companys ofcial variety trials in the experimental eld of Michigan State Universitys Saginaw Valley Research
and Extension Center at Frankenmuth, Michigan during the 2012
harvest season. The beet samples came from different commercial
hybrid varieties, and they were stored in refrigerated storage at
4 C for about 2 months prior to the experiment. Only those beets
that did not show physiological deterioration (e.g., rot) and physical damage (i.e., cuts and bruises) were selected for the study. The
samples were moved from cold storage within 24 h prior to the
experiment and were washed to remove adhered soil. Spectral
measurements were rst made on intact beets using two portable
spectrometers for the spectral regions of 4001100 nm and 900
1600 nm, respectively (see details in the following section). Thereafter, the beet samples were cut into two sections in the direction
that is approximately perpendicular to the crown-root axis. Spectral measurements were immediately made at the centre of the
lower section of the beet. After the spectral measurements had
been completed, wet lab chemistry analysis (i.e., HPLC) was performed to measure the sucrose content of each beet sample. A total
of 398 beet samples were tested in the study.

265

2.2. Visible and near-infrared spectroscopic measurement

Two portable spectrometers were used for this research: a Vis/
SWNIR spectrometer (Model LOE-USB; tec5USA Inc., Plainview,
NY) for the spectral region of 4001100 nm, and an NIR spectrometer (Model NIR 512L-1.7T1; Control Development Inc., South
Bend, IN) for the spectral region of 9001600 nm. Both spectrometers were operated in interactance mode (Fig. 1a). The sugar beet
sample was illuminated with a broadband light source that was
delivered from the 25-mm diameter ring of a lighting/detection
probe, which was in contact with the beet sample. In the centre
of the probe was a light detector covering an area of 11 mm diameter to receive the light that had travelled through the esh tissue.
A separating distance of 3.5 mm between the light illuminating
ring and the detector ensured that only the light that travelled
through the beet tissue would be received by the probe (Fig. 1b).
A soft rubber sealing ring was used between the illuminating ring
and the detector to block the illuminating light from entering the
detecting area directly. In addition, a sponge ring (5 mm thick)
was attached to the periphery of the probe to block ambient light.
For Vis/SWNIR measurements, the lamp power supply was set
to 100 W and the integration time of the spectrometer was set to
575 ms. For NIR measurements, the lamp power supply was set
to 200 W and the integration time was set to 4 s. Spectral measurements were made at a location approximately 10 mm from the
crown end of each intact beet. For both spectrometers measurements for intact beets, three scans were acquired from the same
location of each beet sample, and they were then averaged for further analysis.
After spectral measurements for the intact sample had been
made, the beet was sliced in the transverse direction (i.e., perpendicular to the crown-root axis) into two sections using a stainless
steel kitchen knife. Spectral measurements were then made, using
each spectrometer, from the lower section (toward the root side) of
the beet. Two scans were taken at two locations that were approximately equidistant from the centre and the outside edge of each
beet slice, and these values were then averaged. The integration
times were 0.2 s and 1.5 s for Vis/SWNIR and NIR measurements,
respectively. Thereafter, one cylindrical specimen of 50 mm in
diameter and 10 mm in height (without skin) was taken from
one of the two detection locations for the beet slice and subsequently used for sugar content analysis by HPLC. Reference spectra
were acquired from a white Teon disk in order to calculate the
relative interactance of each beet sample.
2.3. Sucrose measurement by HPLC
Sucrose content was measured by means of high-performance
liquid chromatography (HPLC) (Trebbi & McGrath, 2004). Each beet
specimen of 30 g fresh weight was lyophilised until the pressure
was <1 mTorr for at least 3 h and then ground to ne powder with
mortar and pestle. Pulverised dried tissue (100 mg) was resuspended in 4 mL of 80% ethanol in a 5-mL uted-cap tube (USA Scientic, Inc., Ocala, FL) and placed horizontally on an orbital shaker
(50 rpm) at 40 C for 16 h. The suspension was centrifuged at
3000g for 10 min to obtain the claried ethanol sugar extract. An
aliquot (1.0 mL) of claried ethanol sugar extract was vacuum
dried, the pellet was resuspended in 1.0 mL of high-resistivity
water (18 MX cm1), and the solution was passed through a
0.22-lm nylon lter (Spin-X Centrifuge Tube Filter; Corning, New
York, NY). The aliquot of water-resuspended sugar extract
(1.0 mL) was used for HPLC analysis with a 6.5 mm  300 mm
Waters Sugar-Pak I carbohydrate column (WAT085188; Waters
Co., Milford, MA). The mobile phase was 134 lM Na2CaEDTA at a
constant ow of 0.5 mL min1, 90 C, 12 min run time, and quantied with a Waters 410 differential refractometer held at 35 C,

266

L. Pan et al. / Food Chemistry 167 (2015) 264271

Fig. 1. Schematic of the visible and near-infrared spectral measurement system (a), and the conguration of the lighting/detection probe (b).

according to the manufacturers instructions (Dorschel, 1984).

Concentration standards for sucrose (2.9246.74 mM) were used
to generate standard curves. Fig. 2 shows a typical HPLC chromatogram, in which sucrose is clearly separated from other compounds
present in the beet. The linearity for sucrose was established by
plotting the peak area (y) versus concentration (x), which was
expressed as: y = 661321x + 335005 (r2 = 0.9985). System control
and data management were accomplished using Empower Chromatography Manager software (Waters Co., Milford, MA).
2.4. Spectral treatment and model development
Relative interactance was calculated from the original sample,
dark and reference spectral data, using the following equation:

RI IS  ID =IR  ID

where RI = relative interactance; IS = measured light intensity from

sample; IR = measured light intensity from the white Teon reference; and ID = dark current from the spectrometer. Like in reectance or transmittance measurement, absorbance for interactance
may be calculated as log (1/RI). However, our preliminary analysis
showed that there was no advantage of using absorbance for developing calibration models. Hence, this paper only presents the procedure of, and results from, analysing relative interactance.
Two standard data preprocessing methods, i.e., standard normal
variate (SNV) (Herrero Latorre, Pea Crecente, Garca Martn, &
Barciela Garca, 2013; Huang, Zhao, Chen, & Zhang, 2014) and the
rst order SavitzkyGolay derivative (SG1) (Sirisomboon, Tanaka,
Fujita, & Kojima, 2007), were applied to reduce multi-collinearity
and the baseline offset arising from scattering effects, thus enhancing the information related to chemical constituents. A partial least
squares (PLS) model was developed by using MATLAB 7.5.0 (The

MathWorks, Inc., Natick, MA) with the PLS Toolbox (Eigenvector

Research, Inc., Wenatchee, WA).
All 398 sugar beet samples were rst sorted for sucrose content
in ascending order. The samples were then systematically separated into two sets with 75% for calibration and the remaining
25% for prediction (i.e., every fourth sample was taken out for prediction). A calibration model for sucrose prediction was developed,
using partial least squares, for the calibration samples only. The
optimum number of latent variables was determined by selecting
the rst minimum from the predicted residual sum of squares
(or PRESS) curve, calculated using the leave-one-out cross validation method. After the calibration model had been developed, it
was then used to predict the sucrose content of beet samples in
the prediction set that were not used in the model calibration. Statistical parameters, including correlation coefcient of calibration
(rc) and prediction (rp), the standard error for the calibration
(SEC) and prediction (SEP) data sets and the ratio of sample standard deviation to standard error of prediction (RPD), were calculated. These parameters were used to assess the performance of
each calibration model for predicting the sucrose content of beets.
Since calibration and prediction results are affected, to some
extent, by the sampling procedure, the above model calibration
and prediction procedure was repeated four times to ensure that
the reported results more accurately reect the actual performance
of the two spectrometers. As thus, after completion of the rst run
of calibration and prediction, another 25% samples were taken out
from the calibration set and put aside as a new set of prediction
samples in the second run, while the prediction samples that were
used in the rst run were returned to the calibration set. The same
procedure of model calibration and prediction as that used in the
rst run was then followed in the second run. This procedure
was repeated four times until each sample in the dataset was

Fig. 2. HPLC chromatogram for analysis of sucrose in sugar beet with the peak at 8.010 min representing sucrose.

L. Pan et al. / Food Chemistry 167 (2015) 264271

eventually used once for prediction (Cen, Lu, Mendoza, & Ariana,
2012). Finally, values of the statistics (i.e., rc, rp, SEC, SEP and
RPD) from the four runs were averaged and reported.

3. Results and discussion

3.1. Spectral features for sugar beets
Fig. 3 shows the relative interactance spectra of beet slices for
4001100 nm and 9001600 nm, respectively. Spectra for intact
beets were similar in pattern to those for beet slices.
Absorption in the visible spectral region (400700 nm) was
weaker than that in the NIR region, since relative interactance
generally decreased with wavelength over the spectral region of
7001000 nm (Fig. 2a and b) and over the spectral region of
11001400 nm (Fig. 2b). This could be due to the fact that the colour
of skin and esh for sugar beets was close to white, meaning that
less light was absorbed by the tissue in the visible region. Vis/NIR
spectra are sensitive to the organic compounds that are composed
of the molecular bonds of CH, OH, and NH. Sucrose is composed
of CH, OH, CC, and CO. Two relatively small absorption peaks

Fig. 3. Relative interactance spectra of beet slices obtained using the visible and
shortwave near-infrared spectrometer for 4001100 nm (a) and the near-infrared
spectrometer for 9001600 nm (b).

267

at 480 nm and 637 nm, which are associated with chlorophyll b,

were observed from the interactance spectra (Fig. 3a). For the spectral region of 700 nm1100 nm, absorption mainly came from the
third overtone and second overtone of both CH and OH. An
absorption peak around 980 nm (Fig. 3a) was likely due to the existence of the OH stretching vibration and water (Vanoli et al.,
2011). Similar spectral proles between 900 and 1100 nm were
observed for both Vis/SWNIR and NIR instruments. Lower energy
for the spectral region of 11501400 nm could be due to the existence of the second overtone and combination of rst overtones of
CH. Little energy was detected for the region of 14001600 nm,
resulting from strong absorption of light by the beet tissue, due
to the combination of the rst overtones of bond CH and OH
in H2O.
3.2. Sucrose content prediction
Sucrose content values for the 398 beet samples obtained from
the HPLC analyses were approximately normally distributed
around the mean value of 20.43% (standard deviation = 1.48%)
and ranged between 15.08% and 25.36%.
Different spectra pretreatment methods affected the prediction
performance of PLS models for sliced samples, as measured by rc
and SEC, rp and SEP, and RPD (Table 1). For the Vis/SWNIR spectra,
SNV showed the best prediction results among the four spectral
pretreatments (i.e., no pretreatment, SNV, SG1, and SNV + SG1),
For the NIR spectral data, the three pretreatment methods produced better results compared to those for no pretreatment.
SNV + SG1 gave better predictions over the other pretreatments
(Table 1). Both Vis/SWNIR and NIR spectra, after SNV and
SNV + SG1 pretreatments respectively, provided acceptable predictions for the sucrose content of beet slices, even though NIR generally performed slightly better (Table 1 and Fig. 3). For beet slices,
the best correlation for the calibration model was 0.93 and the
SEC was 0.55 for Vis/SWNIR, compared with the best correlation
of 0.92 and the SEC of 0.59 for NIR. In predicting sucrose content
for the prediction set of beet slices, the correlations (i.e., rp) were
0.88 and 0.88, the SEPs were 0.70 and 0.69, and RPDs were 2.13
and 2.14, for Vis/SWNIR (Fig. 4a) and NIR (Fig. 4b), respectively.
For intact beets, the three spectra pretreatment methods for
both Vis/SWNIR and NIR spectra produced mixed results for
sucrose prediction, compared to that for no pretreatment (Table 2).
For the Vis/SWNIR spectra, SNV + SG1 achieved better results than
no pretreatment, SG1 and SNV, whereas for the NIR spectra, SNV
was the best among the four pretreatments. The results in Tables
1 and 2 demonstrate that although spectra pretreatment generally
improved the PLS model prediction, no single pretreatment
method performed consistently better than the others. Furthermore, the effectiveness of a pretreatment method was inuenced
by type of instrument (i.e., Vis/SWNIR versus NIR) and sample condition (intact versus sliced).
The results for intact beets are not as good as those for beet
slices (Table 1 versus Table 2). With the selected spectra pretreatment method of SNV for NIR and of SNV + SG1 for Vis/SWNIR,
the values of rc and SEC were 0.87 and 0.73 for Vis/SWNIR, which
are better than that for NIR (rc = 0.85 and SEC = 0.78). When the
model was used to predict sucrose content for intact beets, rp,
SEP and RPD deteriorated for both Vis/SWNIR and NIR; the correlations were 0.80 and 0.74, the SEPs were 0.89 and 1.02, and the RPDs
were 1.69 and 1.47 for Vis/SWNIR and NIR, respectively. Consistent
with the calibration results, Vis/SWNIR performed better than NIR
in predicting sucrose content for intact beets. One possible explanation is that beet peel had more inuence on the NIR measurement of intact beets than Vis/SWNIR measurement, as a result of
stronger absorption by the beet tissue in the NIR region than in
the Vis/SWNIR region. Alternatively, since the concentration of

268

L. Pan et al. / Food Chemistry 167 (2015) 264271

Table 1
Calibration and prediction results of sucrose content for beet slices using visible and shortwave near-infrared (Vis/SWNIR) and near-infrared (NIR) instruments with different
spectra pretreatments.*
Instrument

Treatment

rc

SEC (%)

rp

SEP (%)

RPD

Vis/SWNIR

No
SNV
SG1
SNV + SG1

0.93
0.93
0.92
0.92

0.55
0.55
0.57
0.58

0.87
0.88
0.85
0.86

0.76
0.70
0.80
0.76

1.98
2.13
1.89
1.95

NIR

No
SNV
SG1
SNV + SG1

0.91
0.93
0.91
0.92

0.60
0.55
0.60
0.59

0.79
0.86
0.86
0.88

0.96
0.76
0.76
0.69

1.77
1.98
2.00
2.14

No stands for no pretreatment for spectra; SNV, standard normal variate; SG1, the rst order SavitzkyGolay derivative.
The statistical values presented in the table are the average of four calibration and prediction runs.
*
rc and rp are the correlation coefcients of calibration and prediction, respectively; SEC and SEP, standard errors for calibration and prediction, respectively; RPD, the ratio
of sample standard deviation to standard error of calibration and prediction.

Table 2
Calibration and prediction results of sucrose content for intact beets using the visible and shortwave near-infrared (Vis/SWNIR) and near-infrared (NIR) instruments with different
spectra pretreatments.*
Instrument

Treatment

rc

SEC (%)

rp

SEP (%)

RPD

Vis/SWNIR

No
SNV
SG1
SNV + SG1

0.88
0.87
0.87
0.87

0.70
0.72
0.73
0.73

0.74
0.64
0.78
0.80

1.04
1.37
0.96
0.89

1.45
1.25
1.56
1.69

NIR

No
SNV
SG1
SNV + SG1

0.86
0.85
0.82
0.85

0.75
0.78
0.84
0.78

0.72
0.74
0.71
0.73

1.05
1.02
1.09
1.03

1.42
1.47
1.37
1.45

No stands for no pretreatment for spectra; SNV, standard normal variate; SG1, the rst order SavitzkyGolay derivative.
The statistical values presented in the table are the average of four calibration and prediction runs.
*
rc and rp are the correlation coefcients of calibration and prediction, respectively; SEC and SEP, standard errors for calibration and prediction, respectively; RPD, the ratio
of sample standard deviation to standard error of calibration and prediction.

sucrose in the beet is not uniform and generally is higher in the

centre of the beet, deeper illumination via Vis/SWNIR might have
returned more representative spectra. It was difcult to achieve
satisfactory sucrose content prediction for intact beets using either
Vis/SWNIR or NIR measurement in interactance mode. In addition
to the presence of beet peel, the surface of intact beets was
unsmooth and irregular, which could, in turn, have affected
consistent measurements of interactance spectra, thus yielding
poorer Vis/SWNIR and NIR prediction of sucrose content. In some
cases such as to assess sucrose content of in-ground sugar beets
prior to harvesting, it would be more desirable to measure sucrose
content in intact beets. To meet such application needs, further
improvement to the lighting/detection conguration is needed.
It should be pointed out that during postharvest storage, physiological activities continue in beets, which would change some of
the chemical constituents as well as the cellulosic structure. These
physiological changes would, in turn, affect Vis/NIR spectral measurements. Hence the calibration models developed in this study
for beets that had been kept in cold storage for 2 months may
not be suitable for predicting freshly harvested beets or beets that
have undergone different postharvest storage treatments. To
develop a more robust sucrose prediction model, further research
is needed for quantifying the effect of storage condition (i.e., temperature and time) on the sucrose prediction model.

3.3. Spectra correlation analysis

In the above results, contributions of individual wavelengths to
the prediction accuracy were not considered. Further analysis on
the correlation between the sucrose content and interactance for
individual wavelengths would enable us to ascertain the most use-

ful wavelengths for determining the sucrose content and thus gain
a better understanding of light interaction with the beet tissue.
Fig. 5 shows the correlation between the sucrose content and each
single wavelength in the Vis/SWNIR region of 4001100 nm using
the optimum spectra pretreatment for beet slices and intact beets,
as determined from simple correlation analysis. For beet slices, the
sucrose content was negatively correlated with interactance over
400650 nm and was positively correlated with interactance over
6501100 nm (Fig. 5a). The correlations were relatively constant
and high, ranging between 0.3 and 0.5 for the spectral region of
6501100 nm, and between 0.5 and 0.3 for the spectral region
of 400650 nm. No particular wavelengths showed a distinctly
higher correlation (Fig. 5a). This is not totally surprising, in view
of reported studies for other plant products. For instance, Lu
(2001) reported that the correlation between soluble solids content and individual wavelength for sweet cherry was relatively
smooth and stable within the spectral region of 8001300 nm,
without showing a distinctly higher correlation at any specic
wavelengths. However, the pattern of correlation over the NIR
region was quite different from that for the Vis/SWNIR region. In
the NIR region of 9001600 nm, several single wavelengths were
strongly correlated with sucrose content (Fig. 5c). The highest correlation of 0.6 occurred at 1080 nm, followed by the negative at
980 nm. For the region between 1380 nm and 1600 nm, correlations were relatively constant and low, around 0.15, probably
due to the exceptionally low signal in this region resulting from
strong light absorption by the peel (as shown in Fig. 3b). It should
be pointed out that different correlation patterns over the same
spectral region of 9001100 nm were noticed for the two spectrometers, which indicates the varying effects of spectra pretreatment
methods and optical instruments on the spectral measurement
and subsequent prediction of sucrose content for beets.

L. Pan et al. / Food Chemistry 167 (2015) 264271

269

Fig. 4. Sucrose content predictions for beet slices using the visible and shortwave near-infrared (4001100 nm) (a) and near-infrared (9001600 nm) (b) spectrometers.

For intact beets, simple correlation spectra were quite different

from those of beet slices. There were several wavelengths that
showed a strong correlation with sucrose content. The correlations
were higher at the wavelengths of 460 nm (r = 0.28), 660 nm
(r = 0.36), and 1000 nm (r = 0.18) for Vis/SWNIR (Fig. 5b). For
NIR, there were two high correlation peaks of 0.2 at 930 nm and
0.33 around 1200 nm, and several lower peaks at other wavelengths. The fact that some wavelengths had relatively strong correlations with the sucrose content suggested the possibility of
using a few selected wavelengths to predict the sucrose content
of intact beets. This should be investigated in further research.
Correlation analysis demonstrated that type of instrument and
different spectral region directly affected the pattern of correlation
between sucrose content and wavelength. Furthermore, the pattern of correlation was also dependent on whether spectral measurements were made on sliced or intact beets. These ndings
are not totally unexpected, because plant materials are a complex
biological entity composed of multiple chemical constituents, with
heterogeneous structural characteristics, all of which would have
various degrees of inuence on spectral measurements and, hence,

the correlation pattern. These results also show the challenge and
need of developing a general, reliable calibration model for accurate prediction of the sucrose content of beets that have undergone
different postharvest storage conditions.

4. Conclusion
The potential of two portable spectrometers for 4001100 nm
and 9001600 nm was evaluated for sucrose content assessment
of intact and sliced sugar beets. Results showed good correlation
between sucrose content and interactance spectra for beet slices,
with a correlation (rp) of 0.88, standard error of prediction (SEP)
of 0.70, and a ratio of sample standard deviation to standard error
of prediction (RPD) of 2.13 for Vis/SWNIR, and rp = 0.88, SEP = 0.69,
and RPD of 2.14 for NIR. PLS models for Vis/SWNIR gave better predictions of sucrose content for intact beets with rp = 0.80,
SEP = 0.91 and RPD = 1.69, compared with rc = 0.72, SEP = 1.02 and
RPD = 1.47 for NIR. Hence Vis/SWNIR and NIR in interactance mode
can be used to predict sucrose content for beet slices with acceptable

270

L. Pan et al. / Food Chemistry 167 (2015) 264271

Fig. 5. Simple correlation between the sucrose content and relative interactance for individual wavelengths of the visible and shortwave near-infrared for beet slices (a) and
intact beets (b), near infrared for beet slices (c) and intact beets (d) after spectra pretreatment.

accuracy. However, further improvement in the sensing conguration and calibration methods is needed, in order to achieve satisfactory sucrose measurement for intact beets. Several important
wavelengths were identied, which showed relatively strong correlations with sucrose content, for both Vis/SWNIR and NIR measurement. Different spectral preprocessing and modelling
methods and postharvest storage effect need to be considered in
order to improve the sucrose content prediction accuracy for beet
slices and intact beets.

Acknowledgements
The authors would like to thank Thomas Goodwill and R. Scott
Shaw in the USDA Agricultural Research Service (ARS) Sugarbeet &
Bean Research Unit at East Lansing, Michigan, for their technical
support for the sucrose measurement by HPLC. The experiment
was carried out when the rst and second authors were visiting
scientists to the USDA/ARS research unit at Michigan State University, East Lansing, Michigan. The research was also supported by
the Chinese National Foundation of Natural Science (31101282)
and Special Fund for Agro-scientic Research in the Public Interest
(201303088).

References
Camps, C., & Christen, D. (2009). Non-destructive assessment of apricot fruit quality
by portable visible-near infrared spectroscopy. LWT Food Science and
Technology, 42(6), 11251131.

Carlomagno, G., Capozzo, L., Attolico, G., & Distante, A. (2004). Non-destructive
grading of peaches by near-infrared spectrometry. Infrared Physics & Technology,
46(1), 2329.
Cayuela, J. A., & Weiland, C. (2010). Intact orange quality prediction with two
portable NIR spectrometers. Postharvest Biology and Technology, 58(2), 113120.
Cen, H., Lu, R., Mendoza, F. A., & Ariana, D. P. (2012). Assessing multiple quality
attributes of peaches using optical absorption and scattering properties.
Transactions of the ASABE, 55(2), 647657.
Chia, K. S., Abdul Rahim, H., & Abdul Rahim, R. (2012). Prediction of soluble solids
content of pineapple via non-invasive low cost visible and shortwave near
infrared spectroscopy and articial neural network. Biosystems Engineering,
113(2), 158165.
Da Costa Filho, P. A. (2009). Rapid determination of sucrose in chocolate mass using
near infrared spectroscopy. Analytica Chimica Acta, 631(2), 206211.
Dorschel, C. (1984). Analysis of sugars 1: Retention times on Sugar-PakI. In Waters
Lab Highlight; LAH 0210, 2. Milford, MA: Waters Co..
Fan, G., Zha, J., Du, R., & Gao, L. (2009). Determination of soluble solids and rmness
of apples by Vis/NIR transmittance. Journal of Food Engineering, 93(4), 416420.
Gmez, A. H., He, Y., & Pereira, A. G. (2006). Non-destructive measurement of
acidity, soluble solids and rmness of Satsuma mandarin using Vis/NIRspectroscopy techniques. Journal of Food Engineering, 77(2), 313319.
Herrero Latorre, C., Pea Crecente, R. M., Garca Martn, S., & Barciela Garca, J.
(2013). A fast chemometric procedure based on NIR data for authentication of
honey with protected geographical indication. Food Chemistry, 141(4),
35593565.
Huang, L., Zhao, J., Chen, Q., & Zhang, Y. (2014). Nondestructive measurement of
total volatile basic nitrogen (TVB-N) in pork meat by integrating near infrared
spectroscopy, computer vision and electronic nose techniques. Food Chemistry,
145, 228236.
Lu, R. (2001). Predicting rmness and sugar content of sweet cherries using nearinfrared diffuse reectance spectroscopy. Transactions of the ASAE, 44(5),
12651274.
Mat Nawi, N., Chen, G., Jensen, T., & Mehdizadeh, S. A. (2013). Prediction and
classication of sugar content of sugarcane based on skin scanning using visible
and shortwave near infrared. Biosystems Engineering, 115(2), 154161.
McGlone, V. A., Jordan, R. B., Seelye, R., & Martinsen, P. J. (2002). Comparing density
and NIR methods for measurement of Kiwifruit dry matter and soluble solids
content. Postharvest Biology and Technology, 26(2), 191198.

L. Pan et al. / Food Chemistry 167 (2015) 264271

McGrath, J. M., & Fugate, K. K. (2012). Chapter 30: Analysis of sucrose from sugar
beet. In V. R. Preedy (Ed.), Dietary sugars: Chemistry, analysis, function and effects
(pp. 526545). Cambridge, UK: Royal Society of Chemistry Publishing.
Mendoza, F. A., Lu, R., & Cen, H. (2012). Comparison and fusion of four
nondestructive sensors for predicting apple fruit rmness and soluble solids
content. Postharvest Biology and Technology, 73, 8998.
Mosen, A. (2007). Beet-sugar handbook. New Jersey: John Wiley & Sons.
Mulcock, A. P., Moore, S., Barnes, F., & Hickey, B. (1985). The analysis of sugars in
beet, part III: HPLC analysis. International Sugar Journal, 87, 203207.
Paz, P., Snchez, M., Prez-Marn, D., Guerrero, J. E., & Garrido-Varo, A. (2009).
Instantaneous quantitative and qualitative assessment of pear quality using
near infrared spectroscopy. Computers and Electronics in Agriculture, 69(1),
2432.
Roggo, Y., Duponchel, L., & Huvenne, J. (2004). Quality evaluation of sugar beet (Beta
vulgaris) by near-infrared spectroscopy. Journal of Agricultural and Food
Chemistry, 52(5), 10551061.
Sato, T., Zahlner, V., Berghofer, E., Lok, T., & Vollmann, J. (2012). Near-infrared
reectance calibrations for determining sucrose content in soybean breeding
using articial reference samples. Plant Breeding, 131(4), 531534.
Singleton, V., Horn, J., Bucke, C., & Adlard, M. (2002). A new polarimetric method for
the analysis of dextran and sucrose. Journal of the American Society of Sugarcane
Technologists, 22, 112119.

271

Sirisomboon, P., Tanaka, M., Fujita, S., & Kojima, T. (2007). Evaluation of pectin
constituents of Japanese pear by near infrared spectroscopy. Journal of Food
Engineering, 78(2), 701707.
Spackman, V. M. T., & Cobb, A. H. (2002). An enzyme-based method for the rapid
determination of sucrose, glucose and fructose in sugar beet roots and the
effects of impact damage and postharvest storage in clamps. Journal of the
Science of Food and Agriculture, 82(1), 8086.
Trebbi, D., & McGrath, J. M. (2004). Fluorometric sucrose evaluation for sugar beet.
Journal of Agricultural and Food Chemistry, 52(23), 68626867.
Vanoli, M., Rizzolo, A., Grassi, M., Farina, A., Pifferi, A., Spinelli, L., et al. (2011). Timeresolved reectance spectroscopy nondestructively reveals structural changes
in Pink Lady apples during storage. Procedia Food Science, 1, 8189.
Wang, J., Nakano, K., & Ohashi, S. (2011). Nondestructive evaluation of jujube
quality by visible and near-infrared spectroscopy. LWT Food Science and
Technology, 44(4), 11191125.
Xu, H., Qi, B., Sun, T., Fu, X., & Ying, Y. (2012). Variable selection in visible and nearinfrared spectra: Application to on-line determination of sugar content in pears.
Journal of Food Engineering, 109(1), 142147.