Full text of DOI:10.1111/j.1365-2621.2002.tb08725.x

JFS: Food Engineering and Physical Properties
Co-crystallization of Sucrose at
High Concentration in the
Presence of Glucose and Fructose
B.R. BHANDARI AND R.W. HARTEL
o
ABSTRACT: Co-crystallization of sucrose from a highly concentrated sucrose syrup ( 7% moisture, w/w) at 131 C
with 0, 5, 10, 15, and 20% of fructose, glucose, or a mixture of fructose and glucose was investigated. The crystalliza-
tion of sucrose was delayed in presence of these lower molecular weight sugars. The DSC melting endotherm of co-
crystallized samples exhibited a decrease in crystalline sucrose in the sample as a function of increased level of
glucose and fructose. The mechanical strength of co-crystallized granules was found to be related to the moisture
content and the amount of glucose or fructose content in the sample. The samples containing 10, 15, and 20%
glucose in co-crystallized product demonstrated crystallization of glucose in its monohydrate form during 1 mo of
storage.
Keywords: co-crystallization, sucrose, glucose, fructose, melting enthalpy, X-ray diffraction
Introduction
mixture generally decrease the glass transition temperature
RYSTALLIZATION OF SUCROSE IS AN IMPORTANT ASPECT OF MANY (Roos 1995). Addition of glucose or fructose (or a mixture of glu-
C
foods (Hartel 2001). In some foods, crystallization is con- cose and fructose) decreases T (Roos 1995), whereas addition of
g
trolled to give the desired texture (fondant, fudge, panned can- commercial corn syrup causes an increase in T (Gabarra and
g
dies, and so on) or appearance (frosted cereals, icing, and so Hartel 1998).
on). However, in other foods crystallization of sugars is prevent-
Even though the presence of these additives affects the ther-
ed (that is, hard candies, spray dried juices, and so on) and modynamic driving force for crystallization, their effect on the ki-
crystallization upon storage is considered a defect. Sucrose may netics of crystallization may be the dominant effect. Although
be crystallized from a supersaturated solution (70 to 80% solids there is not an abundance of quantitative work on the effects of
such as in sugar refinery situation) or from a highly concentrat- food constituents on sucrose crystallization rates, it is widely rec-
ed solution (85 to 90% solids in fondant manufacturing). Su- ognized that different components inhibit crystallization to dif-
crose is also crystallized at even higher concentration (> 90% ferent extents (Smythe 1967; Bamberger and others 1980; Hartel
solids) in the co-crytallization process (Chen 1994). It is also and Shastry 1991; Hartel 2001). Inhibition of sucrose crystalliza-
crystallized from a nearly anhydrous melt in, for example, the tion may arise from mass transfer influences, where the diffusivi-
production of grained mints. In virtually all of these foods, su- ty of sucrose is decreased in the presence of other components,
crose is crystallized in the presence of other compounds, in- or from specific adsorption effects, where the impurity molecule
cluding other sugars, proteins, starch, or fats. Although the interacts with an existing sucrose crystal surface and impedes
presence of many of these other food constituents is known to further molecular deposition.
impede sucrose crystallization (Bamberger and others 1980;
Hartel and Shastry 1991; Gabarra and Hartel 1998; Hartel netics effects may combine to influence sucrose crystallization
001), our understanding of exactly when and where this inhi- and it is sometimes difficult to specify the exact contribution of
bition occurs is still lacking.
each mechanism. In this research, a co-crystallization process
In processing of complex foods, both thermodynamic and ki-
2
The crystallization of sucrose in a highly concentrated syrup is was studied in which the sucrose was spontaneously crystallized
o
affected by the temperature at which the crystallization is initiat- at high temperature (130 to 135 C). The co-crystallization pro-
ed, water content, and the presence of other ingredients. Other cess has been applied in the encapsulation and granulation of
sugars such as glucose, fructose, invert sugar (an equimolar mix- various food products such as flavor, honey, fruit juices, peanut
ture of glucose and fructose caused by the hydrolysis of sucrose), butter, and others (Chen and others 1988; Chen 1994; Bhandari
and corn syrup are the primary components of concern in many and others 1998). Many of these ingredients (honey, fruit juices)
food products. Thermodynamically, these other components contain glucose, fructose, or both. It is important to know the ef-
may affect both the solubility curve of sucrose and the glass tran- fect of these sugars on the crystallization behavior of sucrose at
sition curve (Gabarra and Hartel 1998). Generally, the addition of these conditions. The quality characteristics of the product (me-
these other components decreases the amount of sucrose that is chanical strength, effectiveness of encapsulation) and process
soluble in water (Smythe 1967; Horn 1977; Pancoast and Junk parameters are governed by the crystallization behavior of su-
1
980; Tjuradi and Hartel 1995) since they compete with sucrose crose. Therefore, the objective of this research was to provide an
for the available water. However, the effect on glass transition extended understanding of the influence of glucose and fructose
temperature (T ) depends primarily on molecular weight. Com- on the crystallization of sucrose under conditions important to
g
ponents that decrease the average molecular weight of the sugar co-crystallization and granulation.
©
2002 Institute of Food Technologists
Vol. 67, Nr. 5, 2002—JOURNAL OF FOOD SCIENCE 1797

Co-crystallization of sucrose . . .
Materials and Methods
Table 1—Crystallization parameters at various addition lev-
els of glucose, fructose, and mixture of glucose and fruc-
tose (31:38).
Raw materials
The raw materials used were crystalline sucrose (Domino,
Tate & Lyle American Sugars Inc., Brooklyn, N.Y., U.S.A.), fructose
Approx
Approx
Final
Approx total
processing
time
Addition moisture beating Nucleation
temp
o
level
%)
at 131 C*
(%)
time
(min)
temp of granules
(
A.E. Staley Mfg. Co., Decatur, Illinois, U.S.A.), and glucose
o
o
(
( C)
( C)
(min)
monohydrate (Roquette Freres, Lestrem, One Broadway, N.Y.,
U.S.A.).
0
5
1
1
6.0
6.3
6.6
6.9
7.2
2.0
3.0
4.0
5.0
5.5
125 ± 2
120 ± 2
115 ± 2
110 ± 2
107 ± 2
100 ± 2
90 ± 2
85 ± 2
80 ± 2
80 ± 2
7
8
9
12
14
0
5
Co-crystallization
Sucrose (500 g) was mixed with water (17 g water/100 g of su- 20
crose), as previously described by Bhandari and others (1998), in *Moisture at a boiling temperature of 131 oC of glucose, fructose, or mixture
of two assumed 12% (extrapolated values from Pancoast and Junk 1980)
a copper kettle and heated over a gas flame to dissolve the sugar.
The temperature of the syrup was monitored continuously. Once
the temperature of the syrup reached 120 °C, a point above the
melting temperatures of both glucose monohydrate and fruc-
tose, the required amounts of glucose or fructose were added di-
rectly to the hot syrup. The addition levels of glucose, fructose, Corp., Wilton, Conn., U.S.A.) connected to a low-temperature re-
and a mixture of glucose and fructose in the sucrose syrup were frigeration system (Intracooler II Perkin Elmer, Wilton, Conn.,
0
, 5, 10, 15, and 20% (solid basis). The mixture of glucose and U.S.A.) was used. Dry nitrogen was used to purge the sample
fructose was added at a ratio of 31:38 (glu:fru), a similar ratio as head and the dry box. The instrument was calibrated using mer-
o
o
present in honey. Each preparation was undertaken in triplicate. cury (-38.87 C) and indium (156.60 C) at the same temperature
Once the syrup reached 131 ± 1 oC, the copper kettle was re- scanning rate of sample. On average 8 to 12 mg sample was
moved from the heating source. The time taken to reach this weighed in hermetically sealed, aluminum sample pans for anal-
temperature from the initial point (room temperature) was 4.5 to ysis. The heat of fusion (DH) was obtained by integration of
5
.0 min in all the experiments indicating that minimum sucrose melting endotherm. Each sample was analyzed in triplicate.
inversion occurred during cooking. The mixture was then vigor-
ously agitated with a whipping mixer. Within a few min, a slight X-ray diffraction
turbidity was observed at the initiation point of spontaneous nu-
To identify the crystalline sugars present, the samples were
cleation for crystallization. When the content was fully crystal- evaluated by powder X-ray diffraction. A Bragg-Brentano X-ray
lized (indicated by solidification), the crystallized product was diffractometer (STOE, Darmstadt, Germany) in reflection mode
placed in a wide bowl and allowed to cool to room temperature. with radiation of wavelength 1.54056A provided by a copper K␣1
o
o
Samples were then placed in open glass jars and kept for 24 hrs filter was used. The samples were scanned from 2␪ with 4 to 50
o
at 40 ± 1 C to equilibrate. Further crystallization and moisture (total 741 steps). Data acquisition was done every 2.5 s with a
loss continued during this time. After this, the granular products scan time of 52 min.
were removed from the temperature control cabinet and sealed
Results and Discussion
with airtight lids. Some samples were taken in 20 mL glass vials
o
and stored in freezer (-34 C) prior to freeze-drying at a condens-
o
Crystallization of sucrose
er temperature of -55 C and pressure of 100 mbar. The moisture
The vigorous agitation caused rapid cooling and incorpora-
tion of air into the syrup. Rapid nucleation was noted at a point
when the syrup exhibited a sudden increase in turbidity and de-
veloped a creamy consistency. The viscosity of the syrup in-
creased rapidly due to the crystal formation and growth. Agita-
content of the co-crystallized product was analyzed using a vacu-
um oven (Lab-line, Barnstead International, Dubuque, Iowa,
U.S.A.) at 68 C and < 6 kPa (abs) vacuum for 48 h.
o
Texture analysis
To compare the relative stiffness of the granules, the samples tion was stopped when the agitator became ineffective at
were compressed using a texture analyzer (TA-XT2; Stable Micro moving the highly viscous slurry/solid granules. It should be not-
Systems, Godalming, Surrey GU7 1YL, England). The samples ed that nucleation might have been initiated by the combined
were sieved prior to texture analysis to a size range of 840 to 1180 effect of increased supersaturation of sucrose due to cooling and
mm and loosely packed to a level of 18 mm in a cylinder of 22 mm whipped air during agitation. The approximate time and temper-
height. The compression probe was flat-faced with a dia of 25.4 ature at the point of nucleation were recorded for each sample
mm. The samples were compacted to one-third of their initial (Table 1).
height (6-mm depth). The peak force required to compact the
Crystallization of pure sucrose occurred very quickly (within 2
sample was recorded. Three measurements were taken for each min) and rapid evaporation of moisture occurred during crystalli-
sample.
zation. Due to the low level of amorphous portion in the crystal-
lized granules of pure sucrose, moisture was easily removed to a
low level dependent on the amount of water needed to equili-
DSC melting endotherms
The melting endotherms of freeze-dried samples were ob- brate with the surrounding environment. In contrast, crystalliza-
tained by differential scanning calorimetry (DSC), using a scan- tion of sucrose in the presence of glucose or fructose led to slower
o
o
crystallization at lower temperatures (Table 1) and caused mois-
ning rate of 10 C/min to the final temperature of 200 C. To min-
imize the effect of moisture on melting, all the powder samples ture to remain at a higher level in the granules (Table 2). Greater
were freeze dried and maintained in a dry desiccator containing extent of cooling was needed to initiate nucleation as the level of
CaSO4 (Drierite WA Hammond Drierite Co. Ltd., Xenia, Ohio, glucose and fructose was increased due to the inhibition effect of
U.S.A.) before DSC analysis. A Pyris series 1 DSC (Perkin Elmer these sugars. Higher moisture content in the granules made with
1
798 JOURNAL OF FOOD SCIENCE—Vol. 67, Nr. 5, 2002

Co-crystallization of sucrose . . .
Table 2—Moisture and mechanical property of sucrose Table 3—Melting endotherm property of co-crystallized
o
o
o
granules as influenced by the addition of glucose and fruc- samples (heat scanning at 10 C/min from 50 C to 200 C)
tose
Addition
level
T *
T *
Tend*
on
( C)
p
o
o
o
Moisture
content
Force for
( C)
( C)
rd
Addition
level
1/3
Sucrose
of granules* compaction*
1
00%
185 ± 1.9
189 ± 2.1
194±2.3
(
% TS)
(%, w/w)
(N)
Fructose
5%
10%
15%
20%
Glucose
5%
10%
15%
20%
Sucrose
Glucose
100
5
0.67 (0.11)
0.79(0.10)
1.31(0.23)
1.85(0.13)
2.69(0.08)
0.95(0.27)
1.55(0.07)
2.25(0.13)
3.22(0.18)
0.69(0.07)
1.49(0.11)
2.05(0.16)
3.02(0.28)
434(44)
384(21)
170(24)
154(25)
73(5)
361(38)
124(18)
78(5)
143 ± 0.6
140 ± 2.0
137 ± 1.5
133 ± 2.1
173±3.3
171±1.5
166±0.5
159±0.5
191±0.3
190±0.8
190±0.4
190±0.2
1
1
2
5
1
1
2
5
0
5
0
Fructose
142 ± 1.6
142 ± 0.7
137 ± 1.7
129 ± 0.1
174±1.1
170±1.5
165±0.9
157±0.1
190±0.2
189±0.8
189±0.5
188±0.7
0
5
0
20(4)
Glucose+Fructose
388(40)
143(17)
132(13)
34(7)
Gluc+Fruc
5%
10%
(
31:38)
10
143 ± 1.6
141 ± 1.1
140 ± 2.0
133 ± 1.4
176±1.1
175±1.1
171±2.9
164±0.5
193±1.2
194±1.4
193±1.4
193±0.3
1
2
5
0
15%
2
0%
*
Values under the bracket are standard deviations.
*
T
, T and T
are the onset, peak, and endset temperatures of melting
end
on
p,
endotherm respectively.
fructose and glucose is related to the amount of water need to
keep these sugars solubilized after complete crystallization of and others 1980). However, under our specific experimental condi-
the sucrose. The continuous syrup phase surrounding the su- tion, fructose and glucose exhibited similar effect on nucleation
crose crystals contained dissolved sucrose, glucose, and fructose, temperature. In summary, there was a strong molecular interfer-
which would eventually equilibrate with the surrounding envi- ence by glucose and fructose on the nucleation and crystal growth
ronment.
In the process of co-crystallization under these conditions, fect on solubility and Tg.
primary heterogeneous nucleation was initiated by the com-
bined effect of agitation and increased supersaturation by cool- Moisture
of sucrose even at very high temperature condition despite the ef-
ing. The decrease in solubility of sucrose in the presence of im-
The moisture content (average of 3 replicates) for all samples
purities (including invert sugars) suggests an increase in the after crystallization and equilibration (24 h at 40 °C) are present-
supersaturation, which would potentially increase the nucle- ed in Table 2. The moisture content of the crystallized sucrose
ation rate. In actual fact, there was an inhibition effect on the for- with no additive was very low, indicating the presence of higher
mation of nuclei. Greater cooling and a longer period of agitation amount of crystalline fraction. It should be noted that at a given
were required in the presence of increased amounts of lower mo- condition of temperature and relative humidity the crystalline
lecular weight sugars (Table 1).
form of sugars have a significantly lower hygroscopicity than
At moisture content below 4%, Saleki-Gerhardt and Zografi their amorphous form. As the amount of glucose or fructose in-
1994) found that the critical crystallization temperature of su- creased, the moisture level also increased, indicating the pres-
(
crose was lowered as the moisture content was increased. In addi- ence of increased amount of amorphous fraction in the sample
tion, they reported that at moisture levels below 4%, the critical (Saleki-Gerhardt and others 1994). At the same level of additions
crystallization temperature (T ) occurred at the midpoint between of fructose and glucose, the granules containing fructose carried
c
the glass transition (T ) and melting point (T ) temperatures of higher moisture indicating higher hygroscopicity of fructose than
g
m
sucrose. In this case, the crystallization behavior of sucrose was re- that of glucose (Table 2).
lated to the increased molecular mobility of the sucrose-water sys-
tem due to lowering of T by the presence of water. This meant Texture
g
that the additives, which increase the Tg of the system, should
The resistance to compaction (stiffness) of a powder is related
augment the T of sucrose due to the reduction in molecular mo- to the relative phase volumes of crystalline and noncrystalline
c
bility. However, in the same work it was found that an addition of a phases, and to the nature (water content, viscosity, and so on) of
small amount of other sugars (trehalose, lactose, and raffinose) in- the noncrystalline phase. In principle, an increased amount of
creased the T much more significantly than the increase in the T crystal phase volume (lower water content) in the granular prod-
c
g
of the mixture due to these additives. This indicated that factors uct will have a greater resistance to the deformation. From Table
other than glass transition would have been interfering with the 2, the addition of glucose and fructose reduced the stiffness of
crystallization of sucrose. A similar result was reported by Gabarra the granules compare to the pure sucrose product. For the same
and Hartel (1998) for crystallization of sucrose in the presence of level of addition, fructose reduced the stiffness of the granules
corn syrup solids. In our research, the addition of fructose and glu- more than glucose. It should also be noted that at increased level
cose would reduce the T of the mixture. It can be seen from the of glucose or fructose, there was also an increase in the moisture
g
result that the nucleation was obviously delayed by these sugars content as the noncrystalline phase increased. Figure 1 shows a
(
Table 1). It should also be noted that the T of fructose is much direct relationship between compaction force and moisture con-
g
lower than glucose. Although the T is depressed more by fructose tent of the granules. This result can also be correlated to the
g
than glucose, other reports have indicated that the crystallization amount of crystalline sucrose present in the mixture. The re-
of sucrose is more inhibited by fructose than glucose (Bamberger duced stiffness, therefore, can be explained by a combined ef-
Vol. 67, Nr. 5, 2002—JOURNAL OF FOOD SCIENCE 1799

Co-crystallization of sucrose . . .
Table 4—Effect of anhydrous glucose and fructose melt peak was also observed just before the main melting endotherm,
on the melting endotherm of sucrose (heat scanning at
which might be due to the presence of some amorphous fraction
o
o
o
1
00 C/min to 145 C for fructose and 180 C for glucose,
o o
(highly concentrated mother liquor) in the granules and this small
peak can be a devitrification peak (result not presented). This
small peak also could be due to the solubilization (heat of solu-
tion) of the surface of the crystals by the residual moisture. As the
content of fructose or glucose increased in the mixture, the main
melting endotherm of sucrose became distorted and broader. The
primary effects of glucose or fructose addition were a decrease in
the onset and peak temperatures of the melting endotherm, with
only a small change in endset temperatures (Table 3). In addition,
the total endotherm developed a trimodal pattern (one major
peak between 2 minor peaks) over the whole range of melting. The
and immediate cooling at 100 C/min to 50 C prior to heat
scanning at 10 oC/min to 200 oC)
Addition
level
H
T *
T *
Tend
*
(J/g sucrose)
on
p
Sucrose
1
00%
185 ± 1.9
189 ± 2.1 194 ± 2.3
126.5 ± 2.5
Fructose
1
2
0%
0%
169 ± 0.8
164 ± 0.3
185 ± 1.0 192 ± 0.4
179 ± 0.1 191 ± 0.4
123.0 ± 0.4
117.0 ± 1.0
Glucose
1
2
0%
0%
174 ± 3.0
168 ± 2.5
186 ± 0.3 192 ± 0.5
184 ± 2.0 190 ± 0.8
119.5 ± 1.2
117.0 ± 1.8
peak value (T ) presented in Table 3 is for the major peak. One po-
p
*
T
, T and T
are the onset, peak, and endset temperatures of melting
end
on
p,
endotherm respectively.
tential cause for the appearance of 3 peaks might be due to melt-
ing of crystals of different size at various stages during heating. It
should also be noted that in an individual scanning of fructose,
glucose, and the mixture of the two over the same range of tem-
perature did not show any degradation peaks, thus there was no
fect of reduced amount of crystalline fraction of sucrose, in- possibility of interference from endothermic degradation peaks.
creased level of amorphous fructose/glucose and increased In addition, the enthalpy change associated with the sucrose-
moisture in the product. For all the samples, there was a high lev- melting peak decreased as more fructose and/or glucose was add-
el of standard deviation within the replicates; however, the val- ed (Figure 2). The effect of impurities on the broadening of peak
ues were still significantly different among the treatments.
and reducing melting enthalpy has been reported previously by
Roos and Karel (1991) and Roos (1993). Saleki-Gerhardt and
Zografi (1994) reported that the melting temperature was reduced
Melting characteristics
The melting enthalpy (⌬H/g of sucrose) of freeze-dried gran- as the moisture content was increased (0 to 3.13%) but did not
ules at various addition levels of glucose and fructose is present- elaborate on the broadening of the melting endotherm due to
ed in Figure 2. The total amount of energy needed to melt the moisture. The crystallization conditions used in these experiments
crystals present in the granule decreased as the addition level of would produce irregular and imperfect crystals with some inclu-
these sugars increased. This suggests that there is increased lev- sion of amorphous material, which might lead to changes in melt-
el of inhibition of sucrose crystallization due to the presence of ing endotherm characteristics. In addition, the presence of the
these additives. The effect of the 2 sugars was equivalent up to amorphous fraction tightly interacting at the crystal surface may
an addition level of 10%, whereas above this level glucose inhib- also affect the melting characteristics. The broadening of the melt-
ited sucrose crystallization more than fructose.
ing peak and decreased melting enthalpy of sucrose in the pres-
Onset temperature of melting (Ton), melting peak temperature ence of other sugars indicate that the determination of crystallini-
Tp), and endset temperature of melting (Tend) are 3 important ty of sugar on the basis of melting enthalpy during DSC scanning
(
characteristics of a melting endotherm curve. For pure crystalline can be erroneous. This can only be considered as a very approxi-
sucrose, the endotherm had a single peak at about 190 °C with a mate value.
narrow melting range from 185 to 194 °C (Table 3). A very small
In order to investigate the effect of amorphous glucose and
Figure 1—Effect of moisture content on the mechanical
property of granules (ꢀ = 100% sucrose, = sucrose with
fructose ꢁ = sucrose with glucose ꢂ = sucrose with
glucose+fructose)
Figure 2—Effect of the addition levels of glucose and fruc-
tose on the DSC melting enthalpy of co-crystallized sample
1
800 JOURNAL OF FOOD SCIENCE—Vol. 67, Nr. 5, 2002

Co-crystallization of sucrose . . .
fructose on the melting properties of sucrose, further studies we presume that the fructose and glucose melts may act as sol-
were performed to clarify the interactions between the sugars vent for sucrose (same as water) and probably some exothermic
during melting in the DSC. About 5 to 6 mg of predried fructose heat of solution of sucrose in fructose and glucose melt counter-
o
or anhydrous glucose crystals were rapidly (heating rate 100 C/ balanced endothermic heat of melting (endothermic). This re-
o
o
min) heated to 145 C and 185 C, respectively, in the DSC and sulted in an overall reduction in melting enthalpy. Further analy-
then rapidly cooled (100 °C/min) to room temperature to pro- sis using more rigorous conditions is required to investigate this
duce sugar melt (glass). A specified amount of sucrose was rapid- phenomenon.
ly weighed on top of the fructose or glucose melt at a ratio of
Gluc/Fruc:Suc 10:90 and 20:80 by weight. The pan was then Changes during storage
sealed and heated at 10 °C/min to determine the melting charac-
The objective of this study was to identify the crystalline su-
teristic of the sucrose crystals. The melting endotherm of sucrose crose and other added sugars during co-crystallization and stor-
was broadened in all cases (Table 4), but the endotherms were age. To this end, X-ray diffraction analysis and DSC analysis were
not distorted as observed in co-crystallized samples. The melting undertaken on product stored for one mo. Physically, it was evi-
enthalpies were also reduced by about 3 and 8% at 10 and 20% dent that the glucose crystallized in the samples containing 10,
level of additives, respectively, although this was significantly 15, and 20% glucose, because the products were crumbly, milky
less than observed in the co-crystallized samples (Figure 2). and appeared dry as compared to the fresh product. The DSC
Since the fructose or glucose melt was prepared separately from scanning clearly showed a glucose monohydrate melting peak in
the sucrose crystals, and the sucrose crystals were of high purity, these stored samples. Figure 3 shows the X-ray diffractograms of
stored co-crystallized sample (20% glucose) in comparison to
pure sucrose and glucose monohydrate crystals. However, with
the X-ray diffraction procedure adopted, it was not possible to
detect the presence of glucose monohydrate crystals in the
stored granules. This might be because of overlapping of many
of the characteristic peaks of both sugars. In an enclosed condi-
tion, the water from the mixture would have been reabsorbed by
the glucose, which crystallized in the monohydrate form. This re-
sulted in drier and tougher product. At the same time period, it
was interesting to note that the product at 10 and 15% glucose
level were relatively softer than that of sample with 20% glucose
(
opposite to the property when fresh). No crystallization of glu-
cose or fructose was detected in DSC scanning in any other sam-
ples. The above results imply that the crystallization of glucose
continues during storage, and that this crystallization is mainly
assisted by the presence of moisture in the sample.
Conclusions
HE CRYSTALLIZATION OF HIGHLY CONCENTRATED SUCROSE WAS IN-
T
hibited by the presence of fructose and glucose. Both rate of
crystallization and total amount of sucrose crystal phase volume
were decreased by the addition of fructose and glucose or the
combination of fructose and glucose. No crystallization of glu-
cose or fructose occurred during the co-crystallization process, al-
though crystallization of glucose in monohydrate form was ob-
served during storage of co-crystallized samples with glucose
content above 10% after one mo of storage.
References
Bamberger M, Segall S, Lee CM. 1980. Factors affecting crystallization of sugars
th
in multi-component systems. Proceedings of the 34 Pennsylvania Manufac-
turing Confectioners Association Production Conference; April 8-10, Hershey,
Pennsylvania, US: PMCA, Center Valley, Pennsylvania.
Bhandari BR, Datta N, Rintoul GB, D’Arcy BR. 1998. Co-crystallization of honey
with sucrose. Lebensm.-wiss. u.-Technol. 31:138-142.
Chen AC. 1994. Ingredient technology by the sugar co-crystallization process.
Int Sugar J 96(1152):493-495.
Chen AC, Veiga MF, Rizzuto AB. 1988. Co-crystallization: an encapsulation pro-
cess. Food Technol 42(11):87-90.
Gabarra P, Hartel RW. 1998. Corn syrup solids and their saccharide fractions
affect crystallization of amorphous sucrose. J Food Sci 63(3):523-528.
Hartel RW. 2001. Crystallization in Foods. Gaithersburg, Md.: Aspen Publishing.
3
25 P.
Hartel RW, Shastry AV. 1991. Sugar crystallization in food products. Critical Rev
in Food Sci and Nutrit 1(1):49-112.
Horn HE. 1977. An alternative to sucrose in panned confections. The Manufac-
tur Confection June 57(6):79-86.
nd
Pancoast HM, Junk WR.1980. Handbook of Sugars. 2 edition. Westport, Conn.:
AVI Publishing Co., Inc. 327 P.
Roos Y. 1993. Melting and glass transitions of low molecular weight carbohy-
drates. Carbohyd Res 238:39-48.
Roos Y. 1995. Characterization of food polymers using state diagrams. J Food Eng
Figure 3—X–ray diffraction pattern of sucrose, co-crystal-
lized sucrose + glucose and glucose monohydrate
2
4(3):339-360.
Vol. 67, Nr. 5, 2002—JOURNAL OF FOOD SCIENCE 1801

Co-crystallization of sucrose . . .
Roos Y, Karel M. 1991. Plasticizing effect of water on thermal behavior and crys-
tallization of amorphous food models. J Food Sci 56:38-43.
Saleki-Gerhardt A, Ahlneck C, Zografi G. 1994. Assessment of disorder in crys-
talline solids. Int J of Pharm 101:237-247.
This was part of the collaborative research between Univ. of Queensland (UQ), Australia,
and Univ. of Wisconsin-Madison (UW), US. This work was supported by UQ under the
Overseas Travel Grant for Collaborative Research. This research was undertaken at UW. The
authors thank Mr Baomin Liang (UW) for his assistance with X-ray diffraction study and
Saleki-Gerhardt A, Zografi G. 1994. Nonisothermal and isothermal crystalliza- other technical matters.
tion of sucrose from the amorphous state. Pharm Res 11:166-1173.
Smythe BM. 1967. Sucrose crystal growth: II. Rate of crystal growth in presence Author Hartel is with the Department of Food Science, University ofWiscon-
of impurities. Aust J Chem 20:1097-1114.
sin, Madison, Wis. Author Bhandari is with Food Science and Technology,
Tjuradi P, Hartel RW. 1995. Corn syrup oligosacharide effects on sucrose crystal- School of Land and Food Sciences, Univ. of Queensland, St. Lucia QLD 4072
lization. J Food Sci 60:1353-1356.
Australia. Direct inquiries to author Bhandari (E-mail: bb@fst.uq.edu.au).
MS 20010288 Submitted 6/5/01, Accepted 7/9/01, Received 7/17/01
1
802 JOURNAL OF FOOD SCIENCE—Vol. 67, Nr. 5, 2002