Tunable Transport of Glucose Through
Ionically-Crosslinked Alginate Gels: Effect
of Alginate and Calcium Concentration
Mari-Kate E. McEntee,1 Sujata K. Bhatia,2 Ling Tao,1 Susan C. Roberts,1 Surita R. Bhatia1
1Department of Chemical Engineering, University of Massachusetts Amherst, 159 Goessmann Laboratory,
Amherst, Massachusetts 01003-9303
2Biochemical Sciences and Engineering, DuPont Central Research and Development, Wilmington, Delaware 19880-0328
Received 24 May 2007; accepted 25 July 2007
DOI 10.1002/app.27478
Published online 26 November 2007 in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Alginate beads have numerous biomedical
applications, ranging from cell encapsulation to drug
release. The present study focuses on the controlled
release of glucose from calcium-alginate beads. The
effects of alginate concentrations (1–6 wt %) and calcium
chloride concentrations (0.1–1.0M) on glucose release
from beads were examined. It was found that the time
required for complete glucose release from beads could
be tuned from 15 min to over 2 h, simply by varying algi-
nate and calcium chloride concentrations in beads. For
calcium-alginate beads with sodium alginate concentra-
tions of 1–4 wt %, higher sodium alginate concentrations
lead to more prolonged release of glucose and thus a
smaller value of a rate constant k, a parameter shown to
be proportional to the diffusion coefficient of glucose in
the alginate gel. For beads with sodium alginate concen-
trations of 4–6 wt %, there was no statistically significant
difference in k values, indicating a lower limit for glucose
release from calcium-alginate beads. Similarly, higher cal-
cium chloride concentrations appear to extend glucose
release, however, no conclusive trend can be drawn from
the data. In a 50 : 50 mixture of calcium-alginate beads of
two different alginate concentrations (1 and 4 wt %), glu-
cose release showed a two-step profile over the time
range of 20–50 min, indicating that the pattern and time
of glucose release from beads can be tuned by making
combinations of beads with varying alginate and/or cal-
cium chloride concentrations. � 2007 Wiley Periodicals, Inc.
J Appl Polym Sci 107: 2956–2962, 2008
Key words: diffusion; hydrogels; polysaccharides; biopoly-
mers; biomaterials
INTRODUCTION
Alginic acid, a linear block copolymer polysaccha-
ride derived from algae, consists of b-D-mannuronic
acid (M) and a-L-guluronic acid (G) residues joined
by 1,3-glycosidic linkages, and forms in a non-regu-
lar, block-wise pattern along the chain. The block-
wise pattern contains three types of polymer seg-
ments: one consisting essentially of D-mannuronic
acid units, the second of L-guluronic acid units, and
the third of alternating D-mannuronic acid and L-
guluronic acid residues.1 The proportion of each
block and the arrangement of blocks along the mole-
cules vary depending on the algal source. When a
solution of sodium alginate is added drop-wise to a
solution containing divalent metal ions such as Ca21
or Cu21, water-insoluble cation-alginate gel beads
are formed in aqueous solution.2 Depending on the
composition of the two residues and their sequential
distribution within the molecules, ionically cross-
linked complexes form either precipitants or hydro-
gels.3 Guluronic acid blocks are known to form a
rigid buckled structure, the so-called ‘‘egg-box’’
array, in which chelating calcium ions are nestled in
the aqueous environment of an ordered gel structure
due to the spatial arrangements of the oxygen atoms
of carboxyl and hydroxyl groups in the guluronic
block.3 This interaction is not only based upon elec-
trostatic interactions, which neutralize acidic groups,
but also on the coordinating function of the calcium
ions as the chelating center.3
Alginate gel beads are used in many applications as
matrices for cell immobilization and drug delivery.
Alginate has long been investigated as a material for
transplantable cell encapsulation systems because it
protects transported cells from the host’s immune sys-
tem, promotes normal cell function, and does not pro-
duce a fibrotic response in vivo.4–6 Rat islets were first
encapsulated in alginate to create a bioartificial endo-
crine pancreas.7 Hepatocytes have also been encapsu-
lated in alginate to create extracorporeal liver assist
Correspondence to: S. R. Bhatia (sbhatia@ecs.umass.edu).
Contract grant sponsor: National Science Foundation;
contract grant number: DMI-0531171.
Contract grant sponsor: NSF CAREER Award (REU Sup-
plement); contract grant number: CTS-0238873.
Contract grant sponsor: Glass Foundation.
Journal of Applied Polymer Science, Vol. 107, 2956–2962 (2008)
VVC 2007 Wiley Periodicals, Inc.
devices.5 Also, alginate has been used to encapsulate
microsomes or whole cells for the production of glu-
curonides in response to drug dosages and insulin.8,9
Alginate has also been utilized to create tissue engi-
neered thyroid tissue and parathyroid tissue.10,11
In addition to promising applications in cell
encapsulation and tissue engineering, alginate gels
also show potential for delivery structures because
gel beads can be formed very easily in aqueous solu-
tions at room temperature, without the use of any
organic solvents. Calcium-alginate beads have a
wide range of applications, including delivery of in-
sulin, pesticides, proteins, drugs, and preservatives.5
Some encapsulated proteins appear to crosslink with
the alginate themselves, further extending the sus-
tained release.12 Alginate gels are not thermoreversi-
ble, but they will dissolve in the presence of a
cation-sequestering agent such as EDTA.13 Moreover,
calcium-alginate gel beads shrink at acidic pH and
erode in an alkaline environment such as the intes-
tine, so they therefore serve as a potential oral deliv-
ery system.14 Alginate is also a mucoadhesive and is
likely to stick to intestinal mucosa for prolonged
periods of time.15 Alginate gel beads have been stud-
ied for the development of oral drug delivery sys-
tems for entrapment in both in vitro and in vivo stud-
ies of controlled release of proteins and drugs.16
Despite widespread interest in alginate-based bio-
materials, there are relatively few studies on control-
ling the transport of small molecules such as glucose
through alginate matrices. Such information is im-
portant in designing alginate gels for cell encapsula-
tion and tissue engineering.
MATERIALS AND METHODS
Materials
Sodium alginate was obtained as a dry powder from
Sigma Chemical Company (St. Louis, MO). The
molecular weight of this product has been reported
as 269 kDa.17 Additional materials were obtained
from Sigma-Aldrich.
Preparation of glucose-loaded
calcium-alginate gel beads
Sodium alginate was dissolved in deionized and dis-
tilled water at the following concentrations: 1, 2, 3, 4,
and 6 wt %. D-glucose was then dissolved com-
pletely in the alginate solution to achieve a concen-
tration of 100 g/L. Calcium chloride solutions were
made at concentrations of 1.0, 0.5, and 0.1M in dis-
tilled water with a glucose concentration of 100 g/L.
The same concentration of glucose was used in both
the alginate and calcium chloride to prevent pre-
experimental mass transfer. The glucose-alginate
solution (10 mL) was added dropwise into 150 mL
of mildly agitated CaCl2 solution using a syringe
pump through a 20-gauge needle at a drop rate of
1.5 mL/min. A peristaltic pump was used to ensure
reproducibility. The drops of alginate solution began
crosslinking with Ca21 ions upon contact to form
calcium alginate gel beads. The beads were cured in
the CaCl2 solution at room temperature for 16–18 h,
protected from light. After the gelling time was com-
plete, the calcium-alginate beads were collected. The
average bead diameter was 2 mm.
Glucose release studies from
calcium-alginate beads
The following procedure was used to determine the
glucose release profile for the alginate gel beads.
Once cured and collected, sixty alginate beads were
washed three times with nano-pure water to remove
residual, nonencapsulated glucose from bead surfa-
ces. The beads were added into 150 mL PBS solution
(pH 5 7.4) at room temperature (258C) with a mag-
netic stirrer. Samples were collected every 30 s for
the first 5 min, then every 5 min until the experi-
ment reached 30 min, and every 15 min thereafter.
The release experiment was carried for 2–3 h with
mild agitation. Once the run was complete, the glu-
cose concentration for each sample was measured to
determine the release profile.
Glucose assay
Glucose concentrations were determined with an en-
zymatic assay, using a uQuant – Universal Microplate
Spectrophotometer with a 96-well plate assay format.
Samples from release studies were diluted 1 : 11 with
distilled water. A 30 lL aliquot of diluted sample or
glucose standard was added to each well of a 96-well
plate. A 200 lL quantity of color reagent, PGO
enzymes (glucose oxide and peroxide mixture), was
then added to each sample on the plate. Protected
from light, the plate was incubated at 378C for 30 min
to activate the color reagent and absorbance was then
read on the uQuant spectrophotometer at 450 nm.
RESULTS AND DISCUSSION
Diffusion analysis
The following mechanism was fitted to the measured
glucose concentrations for each release experiment:
% Release ¼ 100 � ð1� e�k�tÞ (1)
where t is time, and k is a fitting parameter (time21).
The parameter k, with units of inverse time, is a
‘‘rate constant’’ for glucose release and is related to
the diffusion coefficient of glucose in the alginate gel
GLUCOSE TRANSPORT THROUGH CROSSLINKED ALGINATE GELS 2957
Journal of Applied Polymer Science DOI 10.1002/app
beads. To obtain eq. (1), a diffusion model is applied
to the transport of glucose from beads into a well-
stirred solution of limited volume. The increase in
solute concentration in the surrounding liquid is
measured over time, and the diffusion coefficient
can be calculated using an unsteady-state diffusion
model,18 which yields:
Ct
C‘
¼ 1�
X‘
n¼1
6aðaþ 1Þ
9þ 9aþ q2n
exp �Dq
2
n
r2
t
� �
(2)
where r is the radius of calcium-alginate beads in
the PBS solution and qn
2 are the positive, nonzero
roots of the equation:
tanðqnÞ ¼ 3qn
3þ aq2n
(3)
a ¼ Kp Vð4=3Þpr3 (4)
TABLE I
High and Low Extreme Values Chosen for Each Variable Studied in a Design of
Experiments (25–2 Factorial Design)
Alginate
concentration
CaCl2
concentration (M)
Glucose
concentration (g/L)
Gelling
time (h)
Sodium citrate
concentration
1% (w/v) 0.5 100 0.5 0% (w/v)
3% (w/v) 0.1 80 5 0.25% (w/v)
Kp ¼ Concentration of glucose in alginate beads at equilibrium
Concentration of glucose in bulk ðPBSÞ at equilibrium (5)
Kp is a partition coefficient for equilibrium between
glucose in the calcium-alginate gel beads and glu-
cose in the PBS solution with a total volume V (mL).
Assuming there is no partitioning of glucose
between alginate beads and the PBS solution, Kp is 1.
Therefore, both a: and qn
2 are functions of the total
bulk volume, V, and the radius of the beads, r:
a ¼ Vð4=3Þpr3 (6)
If V and r are kept constant for all runs, k is propor-
tional to the diffusion coefficient:
k ¼ q
2
n
r2
D (7)
where D is the diffusion coefficient (in this case a
function of sodium alginate and calcium chloride
concentration), and Ct=C‘ is the percentage of
release. Therefore, eq. (1) can be rewritten as:
% Release ¼ A1 þ A2 � expð�k � tÞ (8)
where parameter A1 is function of final concentration
of glucose in PBS and parameter A2 is function of qn
2
and a. Under all of the above assumptions, eq. (1)
can be correlated to eq. (2), making k a correlative
diffusion coefficient.
In experiments run with multiple varied parame-
ters, data was analyzed using one-way ANOVA fol-
lowed by Tukey’s comparison test. Differences were
considered statistically significant when P < 0.05.
Reported data results are the average of three sam-
ples, with the standard deviation taken to be the
error.
Preliminary screening by design of experiments
In a preliminary study, design of experiments was
utilized to identify the most relevant physical pa-
rameters for glucose release. Previous studies have
shown that the gel microstructure and mechanical
properties depend on calcium chloride concentra-
tion, gelling time, intrinsic viscosity and molecular
weight of the alginate, and alginate concentration.19
Also, sodium citrate has been reported to aid cal-
cium chloride in supplying calcium ions for effective
gelation.20 Therefore, a 25–2 fractional factorial design
of experiments was initially conducted to determine
which factors most significantly affect the release
rate of glucose from the calcium-alginate gel beads,
varying the following: sodium alginate weight con-
centration, calcium chloride concentration, glucose
2958 MCENTEE ET AL.
Journal of Applied Polymer Science DOI 10.1002/app
content, gelling time, and effect of sodium citrate.
The high and low extremes for each parameter
tested are given in Table I. The value of k, the rate
constant, was analyzed as the response variable.
Results from these preliminary experiments indi-
cated that increasing the concentration of alginate
slows down the release rate of glucose. Increasing
the concentration of calcium chloride has the same
effect, but less significantly than alginate. The rate of
glucose release also increased with higher glucose
contents, but this result was expected due to a larger
concentration gradient driving mass transfer. No sig-
nificant differences in the release rates were found
upon varying gelling time or the presence of sodium
citrate (data not shown). Therefore, a more detailed
study focused solely on the effects of alginate con-
centration and calcium chloride concentration on the
rate of glucose release.
Effect of sodium alginate concentration and
calcium chloride concentration
Alginate concentrations of 1, 2, 3, 4, and 6 wt % and
calcium chloride concentrations of 0.1, 0.5, and 1.0M
were investigated in glucose release experiments. All
runs were performed in triplicate with 100 g/L glu-
cose content in alginate beads and 17.5 h gelling
time. Table II summarizes the value of k obtained in
these experiments.
The release data for beads with a constant CaCl2
concentration of 0.1M and varying sodium alginate
concentrations are shown in Figure 1. For these sys-
tems, we obtained complete release of glucose from
the beads in roughly 30–50 min. As seen in Table II,
the rate constant, k, is a function of alginate concen-
tration and decreases from 0.13 to 0.07 min21 as the
TABLE II
Rate Constant, k, Values for a Full Factorial Design
(min21)
Alginate
conc. (wt %)
CaCl2 conc. (M)
0.1 0.5 1.0
1 0.134 6 0.008 0.105 6 0.010 0.332 6 0.041
2 0.106 6 0.008 0.084 6 0.007 0.201 6 0.019
3 0.087 6 0.007 0.050 6 0.006 0.057 6 0.006
4 0.074 6 0.010 0.032 6 0.004 0.011 6 0.007
Figure 1 Percent glucose release from calcium-alginate gel beads prepared with 0.1M CaCl2 and (a) 1 wt % alginate,
(b) 2 wt % alginate, (c) 3 wt % alginate, and (d) 4 wt % sodium alginate.
GLUCOSE TRANSPORT THROUGH CROSSLINKED ALGINATE GELS 2959
Journal of Applied Polymer Science DOI 10.1002/app
sodium alginate concentration increases from 1 to 4
wt %. Extended release can thus be obtained by
increasing the sodium alginate concentration, while
faster release can be obtained using a lower concen-
tration of alginate. Similar results are obtained using
0.5M CaCl2 (Fig. 2), and 1M CaCl2 (Fig. 3), although
in these cases, release occurred over longer time
scales of 100–180 min. Alginate beads prepared with
6 wt % sodium alginate were also explored in this
study, but the difference in release rate using these
beads compared to the 4 wt % alginate beads was
not statistically significant (data not shown). Figure 4
shows the rate constant k as a function of alginate
and CaCl2 concentration. As expected, k decreases
as the alginate concentration increases, correspond-
ing to a slower release. The dependence of k on
CaCl2 concentration, however, is nonmontonic.
This may be due to difficulties in forming beads
with a homogenous microstructure at high CaCl2
concentration.
Multiple step release profiles
By mixing calcium-alginate gel beads prepared with
different concentrations of alginate and/or CaCl2,
release patterns can be manipulated to accommodate
the needs of different applications. In the case where
a ‘‘cascade’’ type of release pattern is desired, it can
now be attained by designing a system using a mix-
ture of different types of beads. In this study, such a
two-step system was created by mixing (50 : 50) cal-
cium-alginate beads prepared with 1 wt % alginate
in 0.5M CaCl2 with an equal number of beads pre-
pared with 4 wt % alginate in 0.5M CaCl2. The
resulting release pattern is shown in Figure 5, and
portrays a cascade, or pulsed, controlled release pat-
tern. Such a release pattern may be useful in deliver-
ing glucose to diabetic individuals and/or endur-
ance athletes.
CONCLUSION
In this study, the effect of CaCl2 concentration and
alginate concentration on the transport of glucose
from alginate beads was investigated. Results show
that it is possible to tune the transport of glucose
and create systems with controlled glucose release
over nearly 3 h. Release time can be extended by
increasing either the sodium alginate concentration
or calcium chloride concentration used in bead
Figure 2 Percent glucose release from calcium-alginate gel beads prepared with 0.5M CaCl2 and (a) 1 wt % alginate,
(b) 2 wt % alginate, (c) 3 wt % alginate, and (d) 4 wt % sodium alginate.
2960 MCENTEE ET AL.
Journal of Applied Polymer Science DOI 10.1002/app
formation. This study has also shown that beads
with different individual profiles can be mixed het-
erogeneously to result in a more complex release
profile that contains multiple steps.
The results lend insight into physicochemical
properties of solutes in alginate gels. While diffusion
coefficients of glucose in alginate have been
reported, there is little information in the literature
on the effect of Ca21 concentration and alginate con-
centration on these coefficients.17,21–23 The findings
of this study may have relevance for other solutes in
alginate gels, as well as other hydrogel systems.
These calcium-alginate gel controlled release sys-
tems are biocompatible and thus have potential uses
in tissue engineering, cell encapsulation, and drug
Figure 4 Comparison of the k value trends with alginate
concentration among different calcium chloride concentra-
tions.
Figure 3 Percent glucose release from calcium-alginate gel beads prepared with 1.0M CaCl2 and (a) 1 wt % alginate,
(b) 2 wt % alginate, (c) 3 wt % alginate, and (d) 4 wt % sodium alginate.
Figure 5 Glucose release profile of a mixed system: 50 :
50 mixture of calcium-alginate gel beads prepared with 1
wt % alginate and 0.5M CaCl2 and calcium-alginate gel
beads prepared with 4 wt % alginate and 0.5M CaCl2.
GLUCOSE TRANSPORT THROUGH CROSSLINKED ALGINATE GELS 2961
Journal of Applied Polymer Science DOI 10.1002/app
delivery. An alginate-based gel could also potentially
be used for controlled delivery of glucose to diabetic
patients and endurance athletes over extended peri-
ods of time. The present study is significant for the
design of such glucose delivery systems. For clinical
use of alginate gels, it is essential that glucose deliv-
ery be finely tuned, to avoid adverse effects of hypo-
glycemia and hyperglycemia. The present study
demonstrates that delivery can be tuned based on
calcium and alginate concentrations. In addition,
because combinations of different beads can be
mixed to provide a multistep release profile, one
can envision creating a cocktail of beads for specific
clinical targets. Future work on this subject should
include in vivo studies, to determine the effects of
alginate and calcium chloride concentrations in
beads on in vivo glucose release, and provide a cor-
relation between in