Preparation of dry honey by microwave–vacuum drying

Zheng-Wei Cui

a,*

, Li-Juan Sun

a

, Wei Chen

b

, Da-Wen Sun

c,*

a

Food Engineering & Machinery Group, School of Mechanical Engineering, Jiangnan University, Wuxi, Jiangsu 214122, PR China

b

School of Food Science and Technology, Jiangnan University, Wuxi, Jiangsu 214122, PR China

c

Department of Biosystems Engineering, University College Dublin, National University of Ireland, Earlsfort Terrace, Dublin 2, Ireland

Received 15 November 2006; received in revised form 26 June 2007; accepted 29 June 2007

Available online 20 July 2007

Abstract

Microwave–vacuum (MWV) drying was investigated as a potential method for obtaining high-quality dried honey. Liquid honey was

heated and dehydrated in a MWV dryer to a moisture content less than 2.5% within about 10 min. The drying curves and the temper-
ature changes of samples were tested during MWV drying at a different of microwave power, vacuum pressure levels and sample thick-
nesses. Fructose, glucose, maltose and sucrose contents in the liquid and dry honey were determined by high-performance liquid
chromatography (HPLC). The volatiles in the liquid and dry honey were concentrated by solid-phase microextraction (SPME), separated
and identified by gas chromatography–mass spectrometry (GC–MS). A sample thickness of less than 8 mm and a vacuum pressure of
30 mbar were identified as the better parameters for the MWV drying. The core temperatures of the sample were about the same as the
surface temperatures, the temperature changes were from 30 to 50

°C with higher dehydration rates while no darkening of the honey took

place during MWV drying. There were no significant changes on the contents of fructose, glucose, maltose and sucrose in the honey after
MWV drying. The volatile acids, alcohols, aldehydes and esters made up the bulk of the identified aroma compounds of the used liquid
honey and the content of alcohols and the esters changed slightly. The acids decreased markedly whereas the aldehydes and the ketones
increased remarkably in the honey dehydrated by MWV drying.
Ó 2007 Elsevier Ltd. All rights reserved.

Keywords: Microwave–vacuum drying; Honey; Dry honey; Fructose; Flavors

1. Introduction

Honey is a sweet viscous yellowish liquid with tempting

flavors, which is elaborated by the honeybee from the nec-
tar of plants. It contains fructose and glucose (60–85%) as
the predominant monosaccharides, maltose and sucrose
(7–10%) as the most important disaccharides, melezitose
as the main trisaccharide and other low molecular weight
oligosaccharides (

Doner, 1977; Doner & Hicks, 1982; Laz-

aridou, Biliaderis, Bacandritsos, & Sabatini, 2004

). Beside

those, antioxidants (such as pinocembrin, pinobanksin,
chrysin and galagin), acids (primarily gluconic acid), pro-

tein, minerals, flavonoids, vitamins and enzymes among
others are also found in honey (

Bouseta, Scheirman, &

Collin, 1996; Sabatier, Amiot, Tacchini, & Aubert, 1992;
Wang, Gheldof, & Engeseth, 2004

). Therefore, honey is

often eaten as a hygienic food that is good for health. Most
honeys are supersaturated solutions of fructose and glucose
with low pH (3.4–6.1), which have a tendency to crystallize
spontaneously at room temperature, making them less
appealing to the consumer. Moreover, in many cases, crys-
tallization of honey results in increased moisture of the
liquid phase which can allow naturally occurring yeast cells
to multiply causing fermentation of the product (

Doner,

1977

). It also causes metal containers to corrode easily.

All these characteristics lead to the inconvenience for stor-
age and transportation of honey.

Honey, in its liquid and natural state, presents signifi-

cant handling problems in mass production operations or

0260-8774/$ - see front matter

Ó 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.jfoodeng.2007.06.027

*

Corresponding authors. Tel./fax: [phone] (Z.-W. Cui).
E-mail

addresses:

[e-mail]

(Z.-W.

Cui),

[e-mail]

(D.-W. Sun).

URLs:

www.ucd.ie/refrig

,

www.ucd.ie/sun

(D.-W. Sun).

www.elsevier.com/locate/jfoodeng

Journal of Food Engineering 84 (2008) 582–590

consumption due to its viscosity and stickiness. There is a
strong and constant consumer demand for dried honey that
is convenient to be consumed or used in the food industry.
The available dry honey products are derived from pure
liquid honey, which have been dried to low moisture (not
more than 2.5%). The dry honeys can be divided into
two groups. One group is the honey that is solidified into
blocks or flakes by crystallizing the components able to
crystallize in lower moisture contents. This group of honey
is usually consumed as honey candy. In this case evapora-
tion in a vacuum has been carried out (

Taizo, 1994

). The

other group is the honey powders comprising 50–70%
honey with or without other sweeter solids such as high
fructose corn syrup or glucose syrup and other processing
aids and/or ingredients. This kind of commercial dried
honey product, such as ‘‘ADM Honi-Bake Dry Honey
Powder”, ‘‘ADM Honi-Bake 705 Honey Powder” available
from ADM (Archer Daniels Midland) Company with its
headquarter in Decatur (Illinois, USA) are formulated
and processed to be free-flowing. As the honey powder
has very low moisture content, it can be directly added into
dry mixes, seasonings or dry coatings, be easily blended
with other dry ingredients maintaining a full honey flavor,
and be used in commercial technologies or other areas
where process and product constraints previously pre-
vented the use of liquid honey (

Ferriola & Stone, 1998

).

Other advantages of honey powder may include: conve-
nience, free-flow, ease of handing and weighing, reduced
storage space, ease of cleaning and sanitary aspects.

For preparation of the honey powder, special drying

processes such as spray drying, tunnel drying and drum
drying are usually used (

White, 1978

). In fact, various con-

cerns arise during these processes due to the higher viscos-
ity of honey. Hitherto, the preparation of dry honey in
large scale is still a challenging issue and the output of
dry honey in the world is very limited.

MWV drying has been investigated as a potential

method for obtaining high-quality dried foodstuffs, includ-
ing fruits, vegetables and grains (

Cui, Xu, & Sun, 2003,

2004a; Drouzas & Schubert, 1996; Kaensup, Chutima, &

Wongwises,

2002;

Yongsawatdigul

&

Gunasekaran,

1996b; Yousif, Durance, Scaman, & Girard, 2000

). It com-

bines the advantages of both vacuum drying and micro-
wave drying with higher drying rates, lower temperatures
(25–50

°C) and more uniform energy efficient compared

to other drying methods (

Decareau, 1985; Durance &

Wang, 2002; Yongsawatdigul & Gunasekaran, 1996a

). It

dissipates energy throughout a product, and is able to
automatically level any moisture variation within it. Spe-
cific product features such as aroma and flavor compo-
nents, and color are conserved.

The objectives of this study were (1) to optimize the

operating parameters of microwave–vacuum drying of
honey and (2) to evaluate the quality of dry honey by the
current drying methods.

2. Materials and methods

2.1. Samples

Liquid acacia honey samples were provided by Nanjing

Lao Shan Honey Co. Ltd., Nanjing, PR China. It was
already processed through separation from the comb by
centrifugal force, gravity, straining or other means. It was
minimally processed and can be approximately regarded
as botanic origin.

2.2. Drying equipment

A lab scale MWV dryer in which the materials to be

dried can be rotated in the cavity was developed and
described in detail elsewhere (

Cui et al., 2003

). The rotation

speed of the turntable is 5 rpm. In our MWV dryer, the
magnetron and voltage changer can be sufficiently cooled
by two fans with higher power to ensure the consistent
power output of magnetron. The Sliding vane rotary vac-
uum pump (Model: 4XZ, Suction Capacity: 4 L/s, Wuxi
Vacuum Pump Works, Jiangsu, China) is used to pump
the air in the cavity to desired vacuum pressure within 50 s.

Nomenclature

C

p

specific heat capacity of sample (kJ/kg K)

m

mass of sample (kg)

n

i

number of replicates of the experimental point i

N

0

number of experimental points for each experi-
ment

Q

abs

energy absorbed by sample per unit time (W)

Q

microwave power output in magnetrons (W)

s

standard experimental error at experiment
point i

S

E

total standard experimental error

t

microwave drying time (s)

DT

temperature rise in sample (

°C)

T

(i,j)

moisture content at experiment point i and at
replicate j (

°C)

T

ðiÞ

mean moisture content at experiment point i
(

°C)

M

(i,j)

moisture content at experiment point i and at
replicate j (% w.b.)

M

ðiÞ

mean moisture content at experiment point i(%
w.b.)

Subscripts
i

experiment point

j

replicate

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

583

2.3. Microwave power output measurement

In this study, the measurement of the MWV dryer

power output was determined calorimetrically, which
was to measure the change of temperature of a known
mass of water (1000 g) for a known period of time.
The increase in temperature of water per unit time could
be given by

Q

abs

¼ Q ¼

mC

p

DT

t

¼

4187

DT

t

ð1Þ

where Q = microwave power output in magnetrons (W),
Q

abs

= energy absorbed by sample per unit time (W),

m = mass of sample (kg), C

p

= specific heat of sample (J/

kg K), DT = temperature rise in sample (

°C), t = micro-

wave heating time (s).

The standard procedure described by

Schiffmamn

(1987)

was used to determine the power output. In the

current study, the power output for full power and
80% full power were 330.0 ± 1.5 W and 290.0 ± 2.3 W,
respectively.

2.4. Drying experiments and procedure

2.4.1. Microwave–vacuum drying experiments

For preparation of dry honey, it is important to control

the drying temperature and drying time. Over-higher tem-
perature and long drying time will damage the nutrition,
color and flavors of dry honey. In order to investigate
the drying curves and the temperature distribution and
changes of samples during MWV drying, different micro-
wave powers, vacuum pressure and sample thickness were
concerned. The drying experiments methods are described
as below:

MWV drying liquid honey in different thickness (6, 8,

10 mm) to moisture content less than 3% to exam the
temperature gradient along the thickness throughout
the drying process (microwave output power, 330 W;
vacuum pressure, 30 mbar);

MWV drying liquid honey at the different vacuum pres-

sure (30 mbar, 50 mbar) and microwave output power
(330 W, 290 W) to moisture content less than 3% to
study the drying curves and the temperature changes
of samples throughout the drying process (the thickness
of sample, 8 mm).

2.4.2. Experimental procedure

The initial moisture content of the honey was 20.83%

(wet basis), which was measured according to the vacuum
oven method (

AOAC, 1995

). During drying, the sample

was spread to a thickness of 8, 12 and 16 mm respec-
tively, in a cylinder dish made of tetrafluoroethylene with
the diameter of 155 mm and rotated with the turntable
and then the appropriate experimental conditions (vac-

uum and microwave power) were imposed. For each
experiment, the vacuum was interrupted and the sample
was taken out and its core and surface temperatures were
measured at three locations along radius using an auto-
matic check rig with 16 K-type thermocouple probes
(Model

XMD-16,

thermocouple:

Platinum–Rhodium

10–Platinum, accuracy = ±0.25%T, diameter of probe in
ball = 1 mm, response time 6 10 s, Shanghai Automatic
Instrument Co. Ltd., Shanghai, China), then weighed
by electronic balance (Model MP2000D, accuracy =
±0.01 g, Shanghai Electronic Balance Instrument Co.
Ltd., Shanghai, China) every 2 min and watered the
probes. The sample was dried until the moisture content
was less than 2.5% (wet basis). All the measurements
were taken within 0.5 min. The moisture of the dried
sample at the end of every drying period was calculated
according to the loss of weight and the value of the ini-
tial moisture content. Compared to the evaporation heat,
the sensible heat lost due to the above interruption was
small and could be neglected (continuous drying experi-
ment on similar weight was conducted to examine the
effect of this interruption during drying on weight loss
and it is found the effect was negligible). Each experiment
was done in triplicate.

The MWV drying experiments were carried out for two

levels of microwave power (330.0 W and 290.0 W) and two
levels of vacuum pressure (30 mbar and 50 mbar). The
lower power levels were obtained from a magnetron that
was cycled between on and off.

The experimental data points and the process conditions

are presented in

Figs. 1 and 2

. In these figures, Measure-

ments were taken at each experimental data point and
the averages and standard errors (s) for each experimental
data point are reported and the total standard errors (S

E

) is

also calculated and shown. The equations for calculating s
and S

E

are given below

s

¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1

3n

i

1

X

3n

i

i

¼1

ðT

ðjÞ

T

ðjÞ

Þ

2

v

u

u

t

ð2Þ

s

¼

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1

n

i

1

X

n

i

i

¼1

ðM

ðjÞ

M

ðjÞ

Þ

2

s

ð3Þ

S

E

¼

P

N

0

i

¼1

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1

3n

i

1

P

3n

i

j

¼1

ðT

ði;jÞ

T

ðiÞ

Þ

2

s

N

0

ð4Þ

S

E

¼

P

N

0

i

¼1

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1

n

i

1

P

n

i

j

¼1

ðM

ði;jÞ

M

ðiÞ

Þ

2

s

N

0

ð5Þ

where n

i

is the number of replicates of the experimental

point i, N

0

is the number of experimental points for each

experiment, T

(i,j)

and M

(i,j)

are the temperature and

moisture content at experiment point i respectively at
replicate j, T

ðiÞ

and M

ðiÞ

are the mean temperature and

moisture content at experiment point i, respectively.

584

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

2.5. Color measurement

Color was measured by a colorimeter (WSC-S system,

Shanghai Precision Instrument Co. Ltd., Shanghai, China).
The cylindrical plastic dishes (58 mm in diameter and
15 mm in depth) containing the same quantity of liquid
honey and dried honey (rehydrated to original moisture
content) samples were placed at the light port (50 mm in
diameter), respectively. Each sample was measured for
color values three times. The instrument was initially cali-

brated with a white standard plate (L

*

= 96.22, a

*

= 6.11,

b

*

= 15.05). The information given by L, a and b is gener-

ally expressed as the total color of the samples, with L rep-
resenting the brightness or dullness, a for redness to
greenness and b for yellowness to blueness.

2.6. Determinations of the main sugar compositions in liquid
and dry honey

High-performance

liquid

chromatography

(HPLC)

analysis was achieved by reference to the method of

GB/

T 2491-2004

(2004, National Standard of PR China). The

20

30

40

50

60

0

4

Time (min)

Time (min)

Time (min)

Temperature (

ο

C )

Temperature (

ο

C )

Temperature (

ο

C )

0

5

10

15

20

25

T

co

=T

s

+ 2.6

ο

C

S

E

(T

s

)=0.95

S

E

(T

co

)=0.93

S

E

(M)=0.42

Moisture content (% w.b.)

Moisture content (% w.b.)

Moisture content (% w.b.)

Surface Temperature, Ts

Core temperature, Tco

Moisture content, M

8

6

2

0

4

8

6

2

0

4

8

6

2

20

30

40

50

60

0

5

10

15

20

25

T

co

=T

s

+ 5.6

ο

C

S

E

(T

s

)=1.53

S

E

(T

co

)=1.62

S

E

(M)=0.38

Surface Temperature, Ts

Core temperature, Tco

Moisture content, M

Surface Temperature, Ts

Core temperature, Tco

Moisture content, M

20

30

40

50

60

0

5

10

15

20

25

T

co

=T

s

+ 9.2

ο

C

S

E

(T

s

)=1.46

S

E

(T

co

)=1.06

S

E

(M)=0.41

Fig. 1. Temperature changes at core and surface points for three different
sample thickness: microwave output power = 330 W, vacuum pressure
= 30 mbar (a) Initial sample weight = 186.4 g, initial sample thick-
ness = 8 mm; (b) initial sample weight = 234.5 g, initial sample thick-
ness = 12 mm;

(c)

initial

sample

weight = 281.2 g,

initial

sample

thickness = 16 mm.

20

30

40

50

60

0

6

Time (min)

Time (min)

Time (min)

Temperature (

ο

C)

Temperature (

ο

C)

0

5

10

15

20

25

S

E(

T,300W)=0.94

S

E

(T,290W)=1.28

S

E

(M,330W)=0.52

S

E

(M,290W)=0.46

Moisture content (% w.b.)

Moisture content (% w.b.)

Temperature (T, 330 W)

Temperature (T, 290 W)

Moisture content (M, 330 W)
Moisture content (M, 290 W)

Temperature (T, 330 W)

Temperature (T, 290 W)

Moisture content (M, 330 W)
Moisture content (M, 290 W)

20

30

40

50

60

0

5

10

15

20

25

S

E

(T, 330W)=1.03

S

E

(T, 290W)=1.15

S

E

(M, 330W)=0.49

S

E

(M, 290W)=0.45

0

5

10

15

20

25

Moisture content (% w.b.)

Moisture content
(330 W, 30 mbar)

Moisture content
(330 W, 50 mbar)

Moisture content
(290 W, 30 mbar)

Moisture content
(290 W, 50 mbar)

8

4

2

0

6

8

4

2

0

6

8

4

2

Fig. 2. Temperature and moisture content changes during drying in
samples at the microwave output power 330.0 W and 290.0 W: initial
sample weight = 187.1 g, initial sample thickness = 8 mm. (a) P = 30
mbar and (b) P = 50 mbar.

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

585

HPLC equipment was composed of a solvent delivery sys-
tem (Waters 600, Waters Co., Milford, MA, USA), work
station (M32, Waters Co.), and a refractive index detector
(Waters 2410). Two milliliters of liquid honey was added
into a 100 mL volumetric flask and diluted to the volume.
Then, 15 lL dilute honey was injected into HPLC column
(Water Co.). For analysis of the glucose and fructose, the
Waters Sugar-Pak

TM

-I column (6.5

300 mm) was used

with water (0.4 mL/min) as the mobile phase at 85

°C.

Sucrose and maltose were separated on a Waters–NH

2

col-

umn (4.6

250 mm) with 80% acetonitrile solution (1 mL/

min) as mobile phase at temperature of 30

°C. They were

detected using a refractive index detector system. The con-
tents of the sugars in honey were calculated based on the
ratio of integrated peak areas using commercial glucose,
fructose, sucrose and maltose (Sigma–Aldrich, St. Loius,
MO, USA) as the standard compounds.

For the determination of main sugar composition in the

dry honey, 20 g dry honey was rehydrated to its original
moisture content. After stirring for about half an hour, it
was completely dissolved followed by the rest of the proce-
dure being similar to that for the liquid honey. The sugar
analysis for each sample was replicated twice.

2.7. Determinations of the flavor of liquid and dry honey

Headspace–solid-phase microextraction–gas chroma-

tography–mass

spectrometry

(HS–SPME–GC–MS)

is

widely used to determine volatile organic compounds in
fruits, vegetables, botanic lipin and honey (

Josep & Franc-

esc, 2003; Ngassoum, Jirovetz, & Buchbauer, 2001; Rosa,
Consuel, Rosa, & Jose, 2002; Song, Fan, & Beaudry,
1998

).

Headspace–solid-phase

microextraction

(HS–

SPME) was used as the sample preparation technique
before the determination of the volatile organic com-
pounds by GC–MS. Ten milliliters of diluted liquid honey
and the reconstituted dry honey samples were introduced
into a 15 mL headspace vial with a magneto-stirrer. Extrac-
tion was carried out at 60

°C and an equilibration time of

1 h. After extraction, the samples were introduced into
the GC injection liner and desorbed at 250

°C for 2 min.

Volatile composition analysis i.e. SPME desorption

analysis was carried out on a Finnigan Trace MS (GC–
MS). One microliters (1 lL) of the samples were injected
into the chromatographic system and the volatile com-
pounds were separated using a 30 m

25 mm PEG-20 m

column with film thickness 0.25 lm. Helium, at a flow rate
of 0.8 mL/min, was used as the carrier with a spilt ratio of
10:1. The following programmed temperature was applied:
initial temperature of the column at 40

°C was held for

3 min and then increased to 120

°C at a heating rate of

4

°C/min. From this point the temperature was increased

to 230

°C at a heating rate of 10 °C/min and kept at

230

°C for 8 min. The temperature of vaporizing chamber

was 250

°C. The mass spectrograph was operated in elec-

tron-impact (EI) mode. The emission current was 200 lA
and ionization voltage was 70 eV. The ion source tempera-

ture and the interface temperature were 200

°C and 250 °C,

respectively while the detector voltage was 350 V.

3. Results and discussion

3.1. Influence of sample thickness on temperature
distribution

Fig. 1

shows that the core temperature of the sample is

2.6

°C (average value) higher than that of the surface tem-

perature for samples with 8 mm thickness, indicating that
a less internal or external mass transfer resistance exists for
liquid honey with thickness less than 8 mm during MWV
drying. The temperature gradient develops as the sample
thickness increases to more than 8 mm, and the thicker the
sample is, the greater the temperature gradient will develop.

For high-viscosity liquid honey, the controlling resis-

tance and moisture transport are strongly related to the
sample thickness and the drying periods and the drying
process can be described by models in one dimension for
samples in which the thickness is normally much less than
the other two dimensions. If sample thickness is less than
8 mm then a homogeneous temperature distribution in
the sample will take place, with almost no temperature gra-
dient or pressure gradient generated across the sample due
to very little internal mass transfer resistance. This internal
mass transfer resistance will become the controlling factor
when the sample is more than 8 mm thick. For thicker sam-
ples, although external mass transfer resistance due to vac-
uum does not apply, the temperature gradient or pressure
gradient will develop along the dimension of thickness as
confirmed by

Koumoutsakos, Avramidis, and Hatzikiria-

kos (2001a, 2001b)

. In order to avoid the over-higher tem-

perature which probably make the product darker in the
center of samples. Usually, the thickness, 8 mm, of liquid
honey during MWV drying is the better choice.

3.2. Drying curves and temperature changes during drying of
honey

In MWV drying, electromagnetic energy is directly con-

verted into the kinetic energy of the water molecules, thus
generating heat within the product, and energy transport is
not affected by conductivity barriers, especially in high-vis-
cosity materials (

Cui, Xu, Sun, & Chen, 2006

). The amount

of heat generated depends on the strength of the electro-
magnetic field and the dielectric properties of the material
being heated. The energy absorbed by the material initiates
moisture evaporation, which increases the internal pressure
and drives the moisture from the interior to the surface. An
absolute pressure of 20–70 mbar, usually applied during
MWV drying, corresponds to a water evaporation temper-
ature and, consequently, product temperature, of approxi-
mately 23–45

°C. The saturation temperatures of water are

28.6

°C and 37.5 °C at the vacuum pressure of 30 mbar and

50 mbar, respectively.

Fig. 2

a and b shows that the temper-

atures of honey being dried are very close to the saturation

586

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

temperature of water corresponding to the used vacuum at
the beginning of the drying period when much water need
to be evaporated. In the later drying stages, a little amount
of water is available, and the energy needed for moisture
vaporization is much less than the thermal energy con-
verted from the microwave power, resulting in the temper-
ature

of sample being higher

than

the saturation

temperature of water. The lower the vacuum pressure,
the lower of drying temperature, but the electrical dis-
charge will be easily caused in the cavity if the vacuum
pressure is less than 20 mbar. Therefore, the vacuum pres-
sure in the range of 25–30 mbar is a better choice.

It can also be seen from

Fig. 2

a and b that the more the

microwave power density is, the higher the drying rate is.
In the later drying stages, the temperature of the sample
with little moisture available will rise rapidly if the micro-
wave power is not properly supplied. Therefore, sophisti-
cated process control will be needed to obtain high
evaporation rates at gentle conditions with minimized dete-
rioration of temperature-sensitive compounds in the mate-
rials being dried and/or to conserve the specific product
features such as aroma, flavor components and color.

Fig. 2

c shows that the effect of vacuum pressure on the dry-

ing curves. The drying curves of 50 mbar are almost the
same with those of 30 mbar. It is because of the drying rate
is related to the value of latent heat of evaporation of water
at vacuum pressure (

Cui, Xu, & Sun, 2004b

), while the

value of latent heat of evaporation of water at 30 mbar is
a little different from that of 50 mbar. In order to reserve
the flavors that are easily volatilized by higher temperature
and lower pressure, the microwave output power (density)
should be carefully chosen so that the total drying time is
shorter as possible and drying temperature is mild.

3.3. Changes in product color

The L value is designated by the Hunter Colorimeter to

measure the degree of whiteness and blackness in the prod-
uct. Together with the amount of the yellowness measured
(b), an interpretation on the degree of browning can be
interpreted.

Table 1

shows that the values of L and b of

liquid honey are not significantly different from those of
the rehydrated dry honey, indicating that the product did
not become darker as the honey was being dried by
MWV drying and little or no Maillard browning occurred.

3.4. Changes of main sugar composition

The largest portion of the dry matter in the honey con-

sists of sugars. In general, the sugars are responsible for

much of the physical nature of honey, its viscosity, hygro-
scopicity, granulation properties and energy values (

Crane,

1975

). In nearly all honeys, fructose predominates in the

form of levulose. However, some few varieties contain a
higher percentage of glucose in the form of dextrose. These
two sugars together account for 85–95% of the honey car-
bohydrates. More complex sugars (oligosaccharides) con-
stitute the remainder except for a trace amount of
polysaccharides. Other dissacharides have been identified
including maltose, isomaltose, nigerose, turanose, maltu-
lose, kojibiose, leucrose, neotrehalose, gentiobiose and
laminaribiose.

Theoretically, Maillard browning reactions would occur

during the period of the drying of liquid honey when the
existence of sugars (such as fructose), amino acids and pro-
teins in honey are heated together. The end result is a prod-
uct with a brown color and a decrease of the sugars.
However from

Table 2

, it is calculated that the contents

of fructose and glucose increased by 1.71% and 2.45%,
respectively. Moreover, the content of maltose and sucrose
decreased by 8.60% and 4.07%, respectively. As no darken-
ing took place after the MWV drying of honey, it can be
deduced that the decreases of maltose and sucrose did
not result from the Maillard browning reaction and are
probably the effects of invertase in the honey on sucrose
and maltose. The mild temperature (30–50

°C) and micro-

wave radiation enhances the activity of invertase with the
result that sucrose and maltose are converted into fructose
and glucose during MWV drying (which is desirable
because of the eat of metabolism of the monosaccharides).

3.5. Retention of volatile aromatic components in honey

Honey varies highly in color, flavor, moisture content

and sugar composition. These attributes depend on the cli-
mate, the floral type and of course, individual bee keeping
practices. Perhaps the most attractive feature of honey is its
characteristic flavor. Moreover, the volatile aromatic mate-
rials are the most important components for its flavor.
Therefore the maximum retention of volatile aromas in
honey is very important for the quality of dry honey.
HS–SPME–GC–MS was applied to examine the volatile
aroma characteristics in the original liquid and the dry
honey.

Figs. 3 and 4

show that over 100 compounds were

separated in 30 m

25 mm PEG-20m column. The acids

(29.33%), alcohols (45.5%), aldehydes (10.98%) and esters

Table 1
Mean color values (±standard error) of the liquid and the dry honey

L

a

b

Liquid honey

37.10 ± 0.51

5.82 ± 0.47

127.1 ± 0.64

Dry honey

33.03 ± 0.36

7.77 ± 0.40

134.8 ± 0.71

(p < 0.05).

Table 2
Monosaccharides and disaccharides values in the liquid and the dry honey
and changes following drying

Liquid honey
(mg/mL)

Dry honey
(mg/mL)

Loss or increase (%)

Fructose

399.5 ± 1.10

406.35 ± 1.40

+1.71

Glucose

325.90 ± 1.23

333.90 ± 1.15

+2.45

Maltose

18.48 ± 0.36

16.89 ± 0.41

8.60

Sucrose

2.70 ± 0.18

2.59 ± 0.24

4.07

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

587

Fig. 3. Flavor of liquid honey.

Fig. 4. Flavor of dry honey (microwave output = 330.0 W, initial sample weight = 187.0 g, initial sample thickness = 8 mm, vacuum pressure = 30 mbar).

Table 3
Volatile alcohols in the liquid and the dry honey

Composition

Liquid honey

Dry honey

Composition loss
or increase

a

(%)

Peak area (A)

% area (calculated
in-house)

Peak area (A

0

)

% Area (calculated
in-house)

Ethanol

85 978 956

26.21

72 789 688

20.28

15.34

1-Butanol, 3-methyl

2 778 334

0.85

725 256

0.20

73.90

1-propanol, 2-methyl

7 645 033

2.13

1-Octanol

793 672

0.24

4 966 969

1.38

525.82

2,3-Butanediol

16 878 345

5.15

26 518 754

7.39

57.12

1,2-Propanediol

1 123 775

0.34

1 952 041

0.54

73.70

Linalool oxide

1 502 454

0.46

3 278 006

0.91

118.18

2-Furanme thanol

1 486 461

0.45

5 425 205

1.51

264.97

Benzyl alcohol

13 651 701

4.16

13 883 699

3.87

1.70

Phenylethyl alcohol

21 240 007

6.48

5 682 176

1.58

73.25

Benzenemethanol, 4-methoxy

1 582 676

0.48

Isosorbide

2 229 046

0.68

A = Peak area of composition in liquid honey.
A

0

= Peak area of composition in dry honey.

a

Composition loss or increase

¼

A

A

0

A

100%.

588

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

(1.86%) made up the bulk of the identified compounds in
liquid honey. The main volatile alcohols, acids, aldehydes,
esters and other volatiles in the liquid and dry honey (rela-
tive percentage > 0.2) are listed in

Tables 3–7

. From

Table

4

, it is clear that the kinds and total content of the acids

decreased markedly, while that the contents of the alde-

hydes and the ketones increased remarkably in the dry
honey after MWV drying as shown in

Table 5

and Table

7. Moreover, the alcohols and the esters are changed
slightly in the dry honey as shown in

Table 3

and Table

6. Upon drying, increases in alcohols, aldehydes and
ketones suggested that macromolecules such as esters and

Table 4
Volatile acids in the liquid and the dry honey

Composition

Liquid honey

Dry honey

Composition loss
or increase (%)

Peak area (A)

% Area (calculated in-house)

Peak area (A

0

)

% Area (calculated in-house)

Acetic acid

82 142 486

25.04

11 613 572

3.23

85.86

Propanoic acid

1 316 909

0.40

100

Butyric acid

1 494 240

0.46

100

Butanoic acid

1 739 945

0.53

100

Caproic acid

1 059 862

0.32

960 299

0.27

9.40

Octanoic acid

1 419 443

0.43

100

Nonanoic acid

1 041 518

0.32

100

Geranic acid

4 679 341

1.43

804 567

0.22

82.81

Benzoic acid

1 325 304

0.40

100

Table 5
Volatile aldehydes in the liquid and the dry honey

Composition

Liquid honey

Dry honey

Composition loss
or increase (%)

Peak area (A)

% Area (calculated in-house)

Peak area (A

0

)

% Area (calculated in-house)

Butyraldehyde

2 725 319

0.76

Hexanal

1 260 924

0.35

Nonaldehyde

1 249 220

0.38

100

Furfural

14 402 354

4.39

13 504 987

3.76

6.23

Benzaldehyde

2 728 870

0.83

7 980 983

2.22

192.46

Benzeneacetaldehyde

16 616 539

5.07

46 801 943

13.04

181.66

Benzaldehyde, 4-methosy-

1 010 415

0.31

100

Table 6
Volatile Esters in the liquid and the dry honey

Composition

Liquid Honey

Dry Honey

Composition loss or
increase, %

Peak Area
(A)

%Area
(calculated in-house)

Peak Area
(A

0

)

%Area
(calculated in-house)

ethyl acetate

1854121

0.56

2146364

0.60

15.76

Ethyl lactate

1307860

0.40

1251991

0.35

4.27

2(3H)-Furanone, dihydro-(cas)

1499050

0.46

1307860

0.47

12.75

2-Hydroxy-2-cyclohexane

827475

0.25

100

3, 7- Dimethyl - Octanoic acid,

ethyl ester

753486

0.23

100

1,2-Benzenedicarboxylic acid,

dibutyl ester

695294

0.21

100

Table 7
Other volatile compositions in liquid honey and dry honey

Composition

Liquid Honey

Dry Honey

Composition loss or
increase, %

Peak Area
(A)

%Area
(calculated in-house)

Peak Area
(A

0

)

%Area
(calculated in-house)

2-Butanone, 3-hydroxy

2533595

0.77

6065465

1.69

139.40

Pyridinetrimethyl

1438491

0.40

2-Propanone, 1-hydroxy

1808997

0.55

2146364

0.60

18.65

Acetophenone

23123726

6.43

4,5-Dimethyl-2-formylfuran

1153737

0.35

719245

0.20

37.66

2-Furancarbixaldehyde,

5-(hydroxymethyl)-

817725

0.25

856596

0.24

4.75

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590

589

acids probably decomposed into low molecular alcohols,
aldehydes and ketones caused by the heating, microwave
radiation and the native enzymes. The decrease in alcohol
is clear in

Table 3

possibly due to its lower boiling point.

The contents of the cooking aroma such as furfural, pyri-
dine and furan in the liquid and dry honey had very limited
and little change, it also indicted that no Maillard brown-
ing reaction occurred during the MWV drying. Although
some changes of the volatile aromatic components and
content in the honey treated with MWV drying were diffi-
cult to avoid, the flavors did not change much due to short
heating/radiation time in our MWV drying process.

4. Conclusions

This study has shown that a thickness less than 8 mm

and a vacuum pressure of 30 mbar were the better param-
eters for the MWV drying of honey. The core temperatures
of the sample were about the same as its surface tempera-
tures, and the temperature changes were from 30 to 50

°C

with higher dehydration rate during MWV drying when
the dryer was done at those parameters. The color of dry
honey is not significantly different from that of the liquid
honey, and the contents of main sugars (fructose, glucose,
maltose and sucrose) are slighted changed. The volatile
acids, alcohols, aldehydes and esters made up the bulk of
the identified aroma compounds of current used liquid
honey, and the content of alcohols and the esters changed
slightly, the acids decreased markedly while the aldehydes
and the ketones increased remarkably in the honey dehy-
drated by MWV drying.

Acknowledgement

This work is supported by National Natural Science

Foundation of China (20436020).

References

AOAC (1995). Official methods of analysis of AOAC International (16th

ed.). Washington, USA: Association of Official Analytical Chemists.

Bouseta, A., Scheirman, V., & Collin, S. (1996). Flavor and free amino

acid composition of lavender and eucalyptus honeys. Journal of Food
Science, 61(4), 683–687, 694.

Crane, E. (1975). In E. Crane (Ed.), Honey: A comprehensive survey

(pp. 160–200). London: Heinemann.

Cui, Z. W., Xu, S. Y., & Sun, D. W. (2003). Dehydration of garlic slices by

combined microwave–vacuum and air drying. Drying Technology,
21(7), 1173–1185.

Cui, Z. W., Xu, S. Y., & Sun, D. W. (2004a). Effect of microwave–

vacuum drying on the carotenoids retention of carrot slices and
chlorophyll retention of Chinese chive leaves. Drying Technology,
22(3), 561–574.

Cui, Z. W., Xu, S. Y., & Sun, D. W. (2004b). Microwave–vacuum drying

kinetics of carrot slices. Journal of Food Engineering, 65(2), 157–164.

Cui, Z. W., Xu, S. Y., Sun, D. W., & Chen, W. (2006). Dehydration of

concentrated Ganoderma lucidum extraction by combined microwave–
vacuum and conventional vacuum drying. Drying Technology, 24(5),
595–599.

Decareau, R. V. (1985). Microwave in the food processing industry. New

York: Academic Press.

Doner, L. W. (1977). The sugars of honey – A review. Journal of the

Science of Food Agricultural, 28, 443–456.

Doner, L. W., & Hicks, K. B. (1982). Lactose and the sugars of honey and

maple: Reactions, properties, and analysis. In D. R. Lineback & G. E.
Inglett (Eds.), Food carbohydrates (pp. 74–112). West Port, CT: AVI
Publishing Company.

Drouzas, A. E., & Schubert, H. (1996). Microwave application in vacuum

drying of fruits. Journal of Food Engineering, 28, 203–209.

Durance, T. D., & Wang, J. H. (2002). Energy consumption, density, and

rehydration rate of vacuum microwave- and hot-air convection-
dehydrated tomatoes. Journal of Food Science, 67(6), 2212–2216.

Ferriola, D., & Stone, M. (1998). Sweetener effects on flaked millet

breakfast cereals. Journal of Food Science, 63(4), 726–729.

GB/T 2491-2004. Isomaltooligosccharide. National Development and

Reform Commission Press. Beijing, PR China.

Josep, S. B., & Francesc, V. C. (2003). Flavor index and aroma profiles of

fresh and processed honeys. Journal of the Science of Food and
Agriculture, 83, 275–282.

Kaensup, W., Chutima, S., & Wongwises, S. (2002). Experimental study

on drying of chilli in a combined microwave–vacuum-rotary drum
dryer. Drying Technology, 20(10), 2067–2079.

Koumoutsakos, A., Avramidis, S., & Hatzikiriakos, S. G. (2001a). Radio

frequency vacuum drying of wood-I mathematical model. Drying
Technology, 19(1), 65–84.

Koumoutsakos, A., Avramidis, S., & Hatzikiriakos, S. G. (2001b). Radio

frequency vacuum drying of wood-II experimental model evaluation.
Drying Technology, 19(1), 85–98.

Lazaridou, A., Biliaderis, C. G., Bacandritsos, N., & Sabatini, A. G.

(2004). Composition, thermal and rheological behaviour of selected
Greek honeys. Journal of Food Engineering, 64, 9–21.

Ngassoum, M., Jirovetz, L., & Buchbauer, G. (2001). SPME/GC/MS

analysis of headspace aroma compounds of the Cameroonian fruit
Tetrapleura tetraptera (Thonn.) Taub. European Food Research and
Technology, 213(1), 18–21.

Rosa, A. P., Consuel, S. B., Rosa, M. C., & Jose, L. T. (2002). Analysis of

volatiles from Spanish honey by solid-phase microextraction and gas
chromatography- mass spectrometry. Journal of Agricultural and Food
Chemistry, 50, 2633–2637.

Sabatier, S., Amiot, M. J., Tacchini, M., & Aubert, S. (1992). Identifi-

cation of flavonoids in sunflower honey. Journal of Food Science, 57(3),
773–774, 777.

Schiffmamn, R. F. (1987). Microwave and dielectric drying. In A. S.

Mujunmdar (Ed.), Handbook of industrial drying (pp. 345–372). New
York: Marcel Dekker.

Song, J., Fan, L., & Beaudry, R. M. (1998). Application of solid-phase

microextraction and gas chromatography/times-of-flight mass spec-
trometry for rapid analysis of flavor volatiles in tomato and strawberry
fruits. Journal of Agricultural and Food Chemistry, 46(9), 3721–3726.

Taizo, K. (1994). Process for production solid honey. US Patent

5,356,650.

Wang, X. H., Gheldof, N., & Engeseth, N. J. (2004). Effect of processing

and storage on antioxidant capacity of honey. Journal of Food Science,
69(2), C96–C101.

White, J. W. Jr., (1978). Honey. In C. O. Chichester, E. M. Mark, & G. F.

Stewart (Eds.), Advances in Food Research (pp. 287–374). New York:
Academic Press.

Yongsawatdigul, J., & Gunasekaran, S. (1996a). Pulsed microwave–

vacuum drying of cranberries, part I. Energy use and efficiency.
Journal of Food Processing and Preservation, 20(2), 121–143.

Yongsawatdigul, J., & Gunasekaran, S. (1996b). Pulsed Microwave–

vacuum drying of cranberries, part II. quality evaluation. Journal of
Food Processing and Preservation, 20(3), 145–156.

Yousif, A. N., Durance, T. D., Scaman, C. H., & Girard, B. (2000).

Headspace volatiles and physical characteristics of vacuum–micro-
wave, air, and freeze-dried oregano (Lippia berlandieri Schauer).
Journal of Food Science, 65(6), 926–930.

590

Z.-W. Cui et al. / Journal of Food Engineering 84 (2008) 582–590