You are on page 1of 64

1

1.1 Introduction
Hydrogels are three-dimensional high molecular weight networks composed of a
polymer backbone, water and a crosslinking agent. They are gaining tremendous
importance in a wide variety of applications in medical, pharmaceutical and related
elds, e.g., wound dressings [1], contact lenses [2], articial organs and drug-delivery
systems [3].
Hydrogels are polymeric materials that do not dissolve in water at physiological
temperature and pH. They swell considerably in an aqueous medium [4] and
demonstrate extraordinary capacity (> 20%) for imbibing water into the network
structure. Gels exhibiting a phase transition in response to change in external
conditions such as pH, ionic strength, temperature and electric currents are known
as stimuli-responsive or smart gels [5]. Being insoluble, these three-dimensional
hydrophilic networks can retain a large amount of water that not only contributes
to their good blood compatibility but also maintains a certain degree of structural
integrity and elasticity [6]. Hydrophilic functional groups such as -OH, -COOH,
-CONH
2
and -SO
3
H present in the hydrogel are capable of absorbing water without
undergoing dissolution.
Hydrogels can be prepared from natural or synthetic polymers [7]. Although hydrogels
made from natural polymers may not provide sufcient mechanical strength and may
contain pathogens or evoke immune/inammatory responses, they do offer several
advantageous properties such as inherent biocompatibility, biodegradability and
biologically recognisable moieties that support cellular activities. Synthetic hydrogels,
on the other hand, do not possess these inherent bioactive properties. Fortunately,
synthetic polymers usually have well-dened structures that can be modied to yield
tailorable degradability and functionality [8].
1.2 Responsive Stimuli-sensitive Materials
Hydrogels have been developed as stimuli-responsive materials, which can undergo
abrupt volume change in response to small changes in environmental parameters:
1
Introduction
2
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
temperature, pH, ionic strength, and so on (Figure 1.1). These unique characteristics of
hydrogels are of great interest in drug delivery, cell encapsulation and tissue engineering
[912]. Stimuli-responsive polymers play an important role in the development of
novel smart hydrogels [13].
Change in pH, Temperature
solvent, ionic strength etc.
Figure 1.1 A model depicting a stimuli-responsive drug-delivery system
The most important systems from a biomedical point of view are those sensitive to
temperature and/or pH of the surroundings. The human body exhibits variations of
pH along the gastrointestinal tract, and also in some specifc areas like certain tissues
(and tumoral areas) and sub-cellular compartments.
Polymerpolymer and polymersolvent interactions show an abrupt readjustment in
small ranges of pH or temperature. This is attributed to a chain transition between
extended and compacted coil states. In the case of pH-sensitive polymers, the key
element of the system is the presence of ionisable weak acidic or basic moieties
attached to a hydrophobic backbone. Upon ionisation, the coiled chains extend
dramatically, responding to the electrostatic repulsions of the generated charges
(anions or cations).
Thermosensitive polymers, like pH responsive systems, offer many possibilities in
biomedicine. They present a fne hydrophobichydrophilic balance in their structure;
and small temperature changes around a critical solution temperature (CST) make
the chains collapse or extend, responding to adjustments of the hydrophobic and
hydrophilic interactions between the polymer chains and the aqueous medium
[1416]. A CST may be defned as a temperature at which the polymer solution
undergoes separation from one phase to two phases. Thus, temperature-sensitive
3
Introduction
polymers undergo an abrupt change in volume as the temperature of the medium is
varied above or below the CST [17]. These unique characteristics make hydrogels
especially useful in biomedical applications such as controlled release of drugs and
in tissue engineering [1822].
Stimuli-responsive sensitive polymer gels offer potential economic alternatives
to conventional separation processes for industrial applications [23]. Controlled
permeability variations of responsive gels have also been used to achieve a variety
of size- or charge-selective separations. In addition to pH and temperature, other
stimuli-responsive hydrogels have been produced that exhibit dramatic changes in
their swelling behaviour, network structure, permeability and mechanical strength
in response to a number of external stimuli, including the presence of specifc solutes
and applied electrical or magnetic felds [24].
Polymers which are normally pH-sensitive are produced by adding pendant acidic
or basic functional groups to the polymer backbone; these either accept or release
protons in response to appropriate pH and ionic strength changes in aqueous media
[25]. The network porosity of these hydrogels changes with electrostatic repulsion.
Ionic hydrogels containing carboxylic or sulfonic acid groups show either sudden
or gradual changes in their dynamic or equilibrium swelling behaviour as a result
of changing the external pH. The degree of ionisation of these hydrogels depends
on the number of pendant acidic groups in the hydrogel, which results in increased
electrostatic repulsions between negatively charged carboxyl groups on different
chains. This, in turn, results in increased hydrophilicity of the network and greater
swelling ratio at high pH. Conversely, hydrogels containing basic pendant groups,
such as amines, ionise and show electrostatic repulsion at low pH [26].
Another novel type of responsive polymers results from surface modifcation of a
polymer matrix by attachment of responsive chains to produce responsive interfaces
showing different behaviour in response to small changes in environmental parameters.
Surfaces may change from hydrophobic to hydrophilic [27] or show a variation in
pore size [28].
The recent past has witnessed increasing interest in the use of polymer hydrogels
in biotechnological applications. Appropriate electric stimulus systems may modify
the swelling capacity of hydrogels, thus inducing specifc properties. Moreover, this
also affects their mechanical properties and morphology recognised parameters
that correlate hydrogel behaviour and end applications. For these reasons, particular
consideration has been given to polymers that respond to appropriate stimuli [29].
Hydrogels, in conjunction with conducting polymers, form materials capable of
undergoing chemical and/or physical transition in response to appropriate electrical
feld stimuli [30]. It is possible to produce environmentally sensitive or responsive
4
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
hydrogels, whose physicalchemical properties such as surface topography and
rheological properties, vary in response to specifc environmental stimuli [31].
1.2.1 Swelling-controlled Systems
Hydrogels consist of macromolecular chains crosslinked to create a tangled mesh
structure, providing a matrix for the entrapment of drugs. When such hydrogels
come into contact with a thermodynamically compatible solvent, the polymer chains
relax [32]. This happens when the characteristic glassrubber transition temperature
of the polymer is below the temperature of experiments. Swelling is the macroscopic
evidence of this transition. The dissolved drug diffuses into the external receiving
medium, crossing the swollen polymeric layer formed around the hydrogel. When
the hydrogel contacts the release medium, the penetrant water molecules invade
the hydrogel surface and thus a moving front is observed that clearly separates the
unsolvated glassy polymer region ahead of the front from the swollen and rubbery
hydrogel phase behind it. Just ahead of the front, the presence of solvent plasticises
the polymer and causes it to undergo a glass-to-rubber transition [33]. Now, the
following possibilities arise:
Case I If the glass transition temperature (T
g
) of polymer is well below the
experimental temperature, the polymer will be in the rubbery state and polymer
chains will have a high mobility that allows easier penetration of the solvent
into the loaded hydrogel and subsequent release of the drug molecules into the
release medium [34]. This clearly results in Fickian diffusion (Case I), which is
characterised by a solvent (or drug) diffusion rate R
diff
slower than the polymer
chain relaxation rate R
relax
(R
diff
<< R
relax
).
Case II If the experimental temperature is below T
g
, the polymer chains of
hydrogels are not suffciently mobile to permit immediate penetration of the
solvent into the polymer core.
The latter situation gives rise to a non-Fickian diffusion process which includes
Case II and anomalous diffusion, respectively, depending on the relative rates of
diffusion and chain relaxation (for Case II, R
diff
>> R
relax
, and for anomalous diffusion,
R
diff
~ R
relax
).
1.2.2 Magnetic-sensitive Release Systems
Magnetism has a profound infuence on living organisms. The haemoglobin in our
blood, an iron-containing protein, is magnetic. Magnetotactic bacteria are perhaps
5
Introduction
the frst living organisms to orient themselves with the Earths magnetic feld [35].
These bacteria are known to contain aligned chains of magnetic particles of various
shapes. There is now substantial evidence that all living organisms, including animals
and humans, contain magnetic particles that act as magnetic receptors [36]. It is an
established fact that the magnetism and magnetic materials have a strong role to
play in health care and biological applications [3740]. The combination of fne
particles and magnetism in the feld of biology and biomaterial has been found useful
in sophisticated biomedical applications such as cell separation [4143], gene and
drug delivery and magnetic intracellular hyperthermia treatment of cancer [44, 45].
A schematic presentation of stimuli-responsive drug-delivery systems is shown in
Figure 1.2.
Swelling Controlled
Temperature
Biomolecule Sensitive
Drug Release from Swollen Gel
pH
N
S
Figure 1.2 A schematic presentation of stimuli-responsive drug-delivery systems
1.3 Concept of Controlled Drug Delivery
In the mid 1960s Judah Folkman, MD, at Harvard was circulating rabbit blood
inside a Silastic (silicone rubber) arterio-venous shunt and discovered that if he
exposed the tubing to anaesthetic gases on the outside, the rabbits would fall asleep
[46]. He proposed that short, sealed segments of such tubing containing a drug could
6
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
be implanted, and if the silicone didnt change in dimensions or composition, the
implant would become a constant-rate drug-delivery device [47]. He also showed
that the rate decreased as the tubing thickness increased, which is obvious today,
but back then it was the frst suggestion of a zero-order controlled drug delivery
implant in vivo. The frst controlled drug delivery (CDD) devices that Alza designed
were macroscopic in scale. One was an ophthalmic insert called the Ocusert that
released the anti-glaucoma drug, pilocarpine, at a constant rate in the eye as shown
in the photograph in Figure 1.3.
An Ophthalmic insert that
provides a zero order release
Figure 1.3 Ocusert anti-glaucoma eye insert containing pilocarpine
Drug-delivery technology represents one of the broader areas of science, which
involves multidisciplinary scientifc approach, contributing to human health care.
The treatment of acute diseases or chronic illnesses has been achieved by delivery of
drugs to the patients for many years. These drug-delivery systems include tablets,
injectables, suspensions, creams, ointments, liquids and aerosols. Today these
conventional drug-delivery systems are widely used. The term drug delivery can be
defned as techniques that are used to get the therapeutic agents inside the human
body [48]. When administering a drug, there is an optimum concentration range to
gain the maximum therapeutic benefts. Ideally, the drug level in the body should
remain between the maximum and minimum effective levels. Beyond the maximum
level represents a toxic level and below the minimum the drug is no longer effective.
With conventional methods (such as immediate release tablets), a single dose will
result in the instant maximum release of the drug (perhaps above the maximum level),
which then drops back below the minimum effective level. Consequently multiple
doses are required to maintain the average drug level within the optimum range as
shown in Figure 1.4 [49].
7
Introduction
It has been a long-held aspiration of pharmacists to deliver a drug molecule to a
specifc site in the body and the concept was dreamed up by Paul Erhlich, who, in
the early twentieth century, coined the term magic bullets to describe such an entity
[50]. Today extensive pharmaceutical research has led to the development of drug-
delivery systems and strategies, which go some way to fulflling this idea, but few of
them could be described as magic bullets. Side effects and toxicities still affect these
approaches and, hence, Erhlichs visionary thinking has not yet been fully realised.
Conventional methods of delivering drugs such as dissolution from a tablet have their
own limitations and are far away from predictably being able to control the rate of
release or control the site of action.
Uncoated
Time (t)
C
o
n
c
e
n
t
r
a
t
i
o
n

(
c
)
Conventional profile
Controlled profile
(Sustained Release)
Coated
therapeutic
Concentration
Range
Figure 1.4 Comparison of conventional and controlled-release profles
The concept of drug targeting and controlled drug delivery is used to improve the
therapeutic index of drugs by increasing their localisation to specifc organs, tissues
or cells and by decreasing their potential toxic side effects at normal sensitive sites
[51]. As in the feld of cancer therapy, chemotherapeutic agents have toxic side effects
for tumour cells as well as for normal cells; the targeted delivery of these agents to
diseased sites would enable the use of higher doses for increasing therapeutic effcacy
[52]. Controlled drug delivery involves the association of a drug with a carrier system,
thereby allowing modulation of the pharmacokinetic properties and biodistribution
of the drug [53].
8
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Targeted drug delivery systems can control the site of action but are usually unable
to dictate the release kinetics of the drugs in a predictable manner. Controlled-release
systems generally can control the rate of drug release but cannot control the fate of
the drug once it has been released.
On moving forward into the twenty-frst century, it is apparent that modern medicines
still face many challenges. As can be seen in Figure 1.5a, the challenges faced in site-
specifc delivery of drugs are immense due to the numerous obstacles barricading
the drug along its desired route. Cellular structure and indeed the very components
of the cell itself will either prevent or act in some selective manner to hinder the
migration of drug from its point of administration to the intended destination site.
Nanotechnology appears to be the one area that may offer scientifc advances in
the coming decades and could lead to signifcant progress in the improvement of
therapeutic outcomes. In particular, the development of nanoparticulate drug-delivery
systems may enhance the probability of getting a drug to its target site [54]. Instead
of relying on the physiochemical properties of the drug to dictate its biodistribution,
the drug is incorporated as a payload into a particle resulting in a different transit
mechanism for the drug after administration [55]. This can be enhanced further by
virtue of the fexible nature of the nanoparticle scaffold, onto which subsections may
be chemically bolted, producing a tailor-made and multifaceted device.
This can be represented in Figure 1.5b, which gives the blueprint of an idealised
nanoparticulate delivery system that must make its way to the target cell. The particle
in Figure 1.5b has several properties, which are incorporated onto the particle, mostly
by covalent bonding to surface groups. A targeting system, such as a monoclonal
antibody, will recognise binding sites that are unique to the target cell and allow the
particle to dock onto the target cell [56]. This has been the focus of much current
research into developing strategies for targeting nanoparticles to the site of drug action
[5759]. This chapter focuses on the fundamentals of minimally invasive controlled
and targeted drug-delivery systems that use magnetism to administer repeatable and
patient-specifc dosages of therapeutic agents to specifc sites from the nanostructure
polymeric materials.
1.3.1 Controlled Drug Delivery
Controlled-release drug-delivery systems can be defned as chemical, mechanical or
electromechanical devices which deliver drugs to the systemic circulation or to specifc
target sites in the body at pre-determined and controlled rates (Figure 1.6). The ideal
controlled-release drug-delivery system would deliver the drug at a rate required by
the body during the treatment period and would also deliver the drug specifcally to
the site of action. In general, release rates are determined by the design of the system
9
Introduction
and are nearly independent of environmental conditions, such as pH. This system
can also deliver drugs for long time periods. Although vesicle or drug macromolecule
conjugate may prolong release, optimal control is afforded if the drug is placed in a
polymeric material.
Oral
administration
Intestinal
Wall
Targeting
System
(a)
(b)
(b)
Fusion
Molecule
Particle
permeation
Preventer of
phagocytosis
Connective
Tissue
Vascular
endothelium
Target Cell
Intravenous
administration
controlling
M
e
m
b
r
a
n
e
R
a
t
o Drug
Payload
Figure 1.5 (a) Simplifed overview depicting the barriers to drug localisation at a
target cell after oral administration into the gut, and (b) An idealised multifaceted
nanoparticulate system containing a drug payload that must also permeate in some
way to the target cell
While the last three decades have seen considerable advances in drug-delivery
technology, major unmet needs remain. Among these are the broad categories of:
Continuous release of therapeutic agents over extended time periods and in
accordance with a pre-determined temporal profle [6062].
10
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Localdeliveryofagentsatpre-determinedratestolocalsites,suchassolidtumours,
to overcome systemic drug toxicity and improve antitumour activity [63].
Improvedeaseofadministration,whichwouldincreasepatientcompliancewhile
minimising the need for intervention by health care personnel and decreasing the
length of hospital stays [64].
Success in addressing some or all of these challenges potentially would lead to
improvements in effcacy and patient compliance as well as minimisation of side
effects [65].
drug loading
uptake of drug
loaded
nanoparticles
eprubocin
cell death
Drug release
Figure 1.6 Schematic presentation of controlled drug-delivery system
1.3.2 Advantages of Controlled Drug Delivery
Controlled-release drug-delivery systems have the potential to provide continuous
drug release (i.e., zero-order kinetics), in which blood levels of drugs would remain
constant throughout the delivery period. By contrast, injected drugs are cleared by
frst-order kinetics, so that initial high levels of the drug after initial administration
drop exponentially over time [66]. The potential therapeutic advantages of continuous-
release drug-delivery systems are signifcant and encompass in vivo predictability
11
Introduction
of release rates; minimised peak plasma levels and thereby reduced risk of adverse
reactions; predictable and extended duration of action; and reduced inconvenience
of frequent dosing and thereby improved patient compliance [67, 68].
Some advantages of controlled-release systems are:
Controlled-releasepreparationsmaintainthedruginthedesiredtherapeuticrange
administration
Localiseddeliveryofthedrugtoaparticularbodycompartment,therebylowering
drug level. Preservation of medications that are rapidly destroyed by the body
(particularly for biologically sensitive molecules such as proteins)
Reducedneedforfollow-upcare
Increasedcomfort
Improvedcompliance
1.3.3 Types of Controlled Drug Delivery
There are three primary mechanisms by which active agents can be released from a
delivery system:
Diffusion-controlled
Swellingcontrolled
Chemicallycontrolled
The above three systems are depicted in Figure 1.7. Although each mechanism will
be discussed separately, most drug-delivery devices act by a combination of these
three mechanisms. The time frame in which the drug is released often defnes the
controlling mechanism. A drug can be included in a delivery system by entrapment
or covalent attachment to the matrix.
1.3.3.1 Diffusion-controlled System
Diffusion-controlled drug-delivery devices have been the subject of numerous studies
[69]. The release characteristics can be applied to such systems as long as the matrix
remains intact and its permeability remains unchanged until the drug it contains is
released.
12
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Diffusion
Controlled
Swelling
Controlled
ChemicallyControlled
(by aronion)
t
0
t
1
t
2
t
n
Time
H
2
O
H
2
O
H
2
O
H
2
O
H
2
O H
2
O
H
2
O H
2
O
H
2
O
H
2
O
H
2
O
H
2
O
H
2
O
Drug released at
equilibrium time
Drug released at
equilibrium time
Figure 1.7 A schematic diagram illustrating the three mechanisms for controlled
drug release from a polymer matrix
Three types of diffusion-controlled devices have been used in drug delivery. These
are matrix devices, reservoir devices and laminated matrix devices as shown in
Figure 1.8.
Dissolved Drug Dissolved Drug
Laminated Matrix
Matrix Device Reservoir Device
Dispersed Drug Dispersed Drug
Dissolved Drug Dispersed Drug
Figure 1.8 Schematic drawings of three types of polymer-based, diffusion-
controlled drug-delivery devices
13
Introduction
The drug component of each type of device can be dissolved or dispersed within the
device.Releaseofdisperseddrugfromapolymermatrixbydiffusionoccursinfour
steps:
Dissolutionofthedrugintothesurroundingpolymerorpores
Moleculardiffusionofthedrugacrossorthroughthepolymerbarrieralongits
concentration gradient
Drugdesorptionfromthepolymer
Diffusionintotheexternalmediumortissue
When a drug is dissolved in the delivery matrix and the mechanism for delivery is
diffusional, then the thermodynamic driving force is the concentration gradient [70]
and release predictions can be made based on Ficks law of diffusion. When the drug
is dispersed as particles rather than dissolved, an equation derived from Ficks law can
be used to predict release rates.
Diffusional release is dependent on the relative solubilities (or permeabilities) and
diffusivities of the drug in both the membrane and the surrounding medium. The
molecular size of the drug largely determines its diffusivity (D), which is proportional
to its molecular weight (M
w
) as shown in Equation 1.1:

1/ 3
1/
w
D M (1.1)
A drug diffusion-controlled-release system is illustrated in Figure 1.9.
Solid Matrix
Barrior Release
Norplant
System
Figure 1.9 A model depicting drug diffusion-controlled release
14
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
1.3.3.1.1 Reservoir Devices
In the reservoir systems shown in Figure 1.10, the drug-delivery rate can remain fairly
constant. In this design, a reservoir whether solid drug, dilute solution or highly
concentrated drug solution within a polymer matrix is surrounded by a flm or
membrane of a rate-controlling material. The only structure effectively limiting the
release of the drug is the polymer layer surrounding the reservoir. Since this polymer
coating is essentially uniform and of a non-changing thickness, the diffusion rate
of the active agent can be kept fairly stable throughout the lifetime of the delivery
system. The system shown in Figure 1.10a represents an implantable or oral reservoir
delivery system whereas the system shown in Figure 1.10b illustrates a transdermal
drug-delivery system in which only one side of the device will actually be delivering
the drug [71].
(a) (b)
Figure 1.10 Drug delivery from typical reservoir devices: (a) implantable or oral
systems and (b) transdermal system
1.3.3.1.2 Matrix Devices
Matrix devices have a major advantage over reservoir devices because they can not
undergo sudden dose dumping [72]. A matrix (or monolith) device is easy to formulate,
gives a higher initial release rate than a reservoir device and can be made to release at
a nearly constant rate. A monolithic solution device contains the drug as a solution
15
Introduction
within the polymer. A monolithic dispersion contains dispersed solid drug in a rate-
limiting polymer matrix, referred to as a matrix system.
Higuchi [73] showed that the release rate from a matrix, in which a drug is suspended,
is dependent on its diffusion coeffcient, its solubility in the polymer and the initial
drug loading. A system that releases according to the Higuchi equation is depicted
schematically in Figure 1.11. This model proposes that the surface region of the
polymer matrix is frst depleted of drug. The drug-depleted polymer layer gradually
becomes thicker requiring the remaining drug to diffuse further to be released from
the matrix. If the drug is dispersed in the matrix and exceeds its solubility, the release
rate is proportional to the square root of time [74]. This release can be described by
the approximation given in Equation 1.2.
Time
Figure 1.11 Drug delivery from a typical matrix drug delivery system

1/ 2
2
2
s
C Dt Mt
M Ah
, ]

, ]

]
(1.2)
16
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
The fraction of drug released (Mt/M) from a slab of thickness h, at time t, is
dependent on the drugs diffusion coeffcient (D), solubility in the matrix (C
S
) and
drug loading (A). Although such matrix devices generally release at a continuously
decreasingrate,devicegeometrycanbevariedtocompensateforthisdecline.Release
rate can also be made more constant by increasing the drug concentration near the
centre of the device, but the manufacture of such devices is technically diffcult. As
an alternative approach, matrix devices can be laminated to make them resemble a
reservoir device to achieve zero-order release.
Matrix systems for controlled-release applications are predominantly for delivery
ofsolublesubstances.Largemoleculessuchasinsulin,heparinandalbuminwere
released over an extended period of time owing to the slow diffusion through the
interconnected pore structure [75].
1.3.3.1.3 Laminated Matrix Devices
Laminated matrix devices belong to a class of drug-delivery systems which by
composition are a mixture of matrix and reservoir devices and represent combined
characteristics of the two drug-delivery systems.
1.3.3.2 Swelling-controlled Systems
It is also possible for a drug-delivery system to be designed so that it is incapable of
releasing its agent or agents until it is placed in an appropriate biological environment.
Swelling-controlled-release systems are initially dry and when placed in a body will
absorb water or other body fuids and swell. The swelling increases the aqueous
solvent content within the formulation as well as the polymer mesh size, enabling the
drug to diffuse through the swollen network into the external environment. Examples
of these types of devices are shown in Figure 1.10 and Figure 1.11 for reservoir and
matrix systems, respectively. Most of the materials used in swelling-controlled-release
systems are based on hydrogels, which are polymers that will swell without dissolving
when placed in water or other biological fuids. These hydrogels can absorb a great
deal of fuid and, at equilibrium, typically comprise 6090% fuid and only 1030%
polymer.
17
Introduction
1.3.3.3 Chemically Controlled Systems
1.3.3.3.1 Matrix with Covalently Attached Drug
Devices in which a drug is covalently bound to a polymer matrix have been reviewed
byLanger[76].Applicationsofpendantchainsystemshavegenerallycentredonshort
delivery times (hours), where the use of such devices can localise delivery reducing
systemic toxicity and increasing therapeutic effcacy. In these devices, the drug is
usually bound as a pendant group, e.g., poly(amino acids) with steroid pendant
groups. A drug can also be used as a crosslinking agent or as part of the polymers
backbone (Figure 1.12). Polypeptides containing the amino acid 3,4-dihydroxyl-
phenylalanine(L-DOPA)havebeenpreparedforuseintreatingParkinsonism.These
L-DOPAcontainingpolymersactasadrug-deliverysystemastheyarebiodegradable.
Polydrugs, or systems in which the drug is polymerised to form an insoluble matrix,
permit the drug loading of up to 10 wt%. Similar methods have been used to
prepare polymers of pesticides, herbicides and antifouling agents. Polydrugs are
simply insoluble prodrugs that can be converted to their active, soluble form through
hydrolysis.
Drug pendents on insoluble matrix Drug crosslinding insoluble matrix
Drug in backbone of insoluble matrix Prodruginsoluble drug polymer
Prodrugdrug attached to soluble polymer Prodrugsoluble drug polymer
Figure 1.12 Schematic drawings of four insoluble and two soluble drug-delivery
devices. The heavy dots represent drug delivery while fne lines
represent polymers
18
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
1.3.3.3.2 Devices with Entrapped Drug
Chemically controlled drug delivery has also involved devices containing entrapped
drug. These devices use hydrophobic polymers that are hydrolytically labile. Such a
device should ideally degrade in a heterogeneous fashion with no bulk or homogeneous
erosion. As the surface of such a device erodes, the included drug is released.
1.3.3.4 Other Delivery Systems
Ion-exchange polymers, which are biodegradable, can be used to deliver ionic drugs.
Releasefromthesedevicesisoftenunpredictableduetofuctuationsinthecomposition
of physiologic fuids, e.g., ionic strength and volume.
Drug release might also be controlled by the exploitation of endogenous enzyme.
Elastomeric poly(esters) and synthetic poly(amino) acids have been reportedly
degraded in vivo by enzymes [77]. Useful drug-delivery devices might result if this
enzymatic biodegradation can be controlled at predictable rates.
1.4 Targeted Drug Delivery
All of the controlled drug-delivery systems described so far have one notable
shortcoming: they do not affect drug distribution, i.e., while they ensure delivery of
drug to the body within a desired time frame, they do not restrict delivery to specifc
location(s) within the body. Drug targeting is the delivery of a drug to a specifc site
in the body where desirable effects can be achieved without exposing other sites to
possible drug toxicity. This is an important distinction from the basic targeting concept,
where the specifc drug receptor is the target and the objective is to improve ft, affnity
and binding to the specifc receptor that will ultimately trigger the pharmacological
activity. This distinction is made since the overall distribution of many drug receptors
does not follow the various diseases. Actually, most of the time, drug toxicity is
receptor related and receptor mediated; thus, improving intrinsic drug affnity and
activity, as well as receptor binding, does not improve the therapeutic index.
In principle, drug targeting can be achieved by physical, biological or molecular
systems that result in high concentrations of the pharmacologically active agent at
the patho-physiologically relevant site. If successful, the result of the targeting would
be a signifcant reduction in drug toxicity, reduction of the drug dose and increased
treatment effcacy. All in all, it is evident that with a biologically active agent of
reasonable activity at hand, targeting to the site of action should be superior to
molecular manipulations aimed at refning the receptor substrate interactions.
19
Introduction
1.4.1 Major Schemes of Targeted Drug Delivery
The following strategies may be adopted to achieve targeted drug delivery:
CaseADirectapplicationofthedrugintotheaffectedzone(organ,tissue).
CaseBPassiveaccumulationofthedrugthroughleakyvasculature(tumours,
infarcts, infammation).
Case C Physical targeting based on abnormal pH and/or temperature in the
target zone, such as tumour or infammation (pH- and temperature-sensitive drug
carriers).
CaseDMagnetictargetingofdrugsattachedtoparamagneticcarriersunder
the action of external magnetic feld.
Case E Use of vector molecules possessing high specifc affnity toward the
affected zone. In a certain sense, cases (c) and (d) can be considered together as
a physical targeting.
The goal of the drug-delivery systems is to put the medications to particular parts
of the body by means of either a physiological or a chemical trigger, such as smart
drug carriers. The smart drug carrier is synthesised by a polymer, which can carry or
release drugs in response to physiological conditions or external stimuli such as pH,
temperature, magnetic or electric feld (Figure 1.13).
drug carrier with drug
molecules in the blood
in the area with abnormal pH or
temperature, carrier disintegrates,
releasing free drug
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
Figure 1.13 Physical targeting of drugs
20
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Pharmaceutical carriers include soluble polymers, nanocapsules, nanoparticles, cells,
cell ghosts, lipoproteins, liposomes and micelles. All of them can be made targeted
in one way or another. The recognition of the target can occur:
onthelevelofawholeorgan
onthelevelofcertaincellsspecifcforagivenorgan
onthelevelofindividualcomponentscharacteristicofthesecells,suchascell
surface antigens
1.4.2 Types of Targeting Methods
There are three typical in vivo applications all requiring that nanoparticles should
accurately localise to therapeutic sites. All the targeting methods could be classifed
as physical, passive or active targeting.
1.4.2.1 Physical Targeting
Physical targeting allows distribution of the drugs and carrier system through external
infuences, such as in the presence of magnetic feld (Figure 1.14). However, with the
assistance only from outside of the body, the physical targeting has less capacity to
recognise specifc cells or tissues.
Tumor
Electromagnet
Drug loaded
magnetic
nanoparticles
Figure 1.14 A hypothetical model showing the possible applications of physical
targeting in treating brain tumours with an applied magnetic feld
21
Introduction
1.4.2.2 Passive Targeting
Passive targeting (Figure 1.15) is contributed to by the properties of drug and carrier,
and also the difference between therapeutic sites and other sites. In passive targeting the
distribution of the drugs within the body occurs through drug and carrier properties
that are unchanged. Only the disrupted endothelial lining of tumour tissues allows
the drug carrier of smaller size to pass through, so that drug-carried nanoparticles
could be localised by enhanced penetration and retention effect.
Enhanced penetration No penetration
large
small
vesicle
Endothelial cells
Blood vessels
Basal membrane
Normal tissue Tumour tissue Endothelial lining intact Endothelial lining disrupted
Perivascular tissue
Figure 1.15 Sketch of the process of passive drug targeting by enhanced
penetration and retention effect
1.4.2.3 Active Targeting
Active targeting (as shown in Figure 1.16) is achieved with mechanisms that allow
direct targeting of drugs and or carriers to specifc cells, tissues or organ systems
through specifc recognition mechanisms. The specifc targeting functional groups,
such as monoclonal antibodies, are immobilised on the particle surface to effciently
increase the chance of uptake by specifc cells. Compared to passive targeting, active
targeting works with more controllable particles [78, 79].
22
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Figure 1.16 The production of stealth liposomes, which are not rapidly cleared
by the liver and spleen, made the active targeting of liposomes a real possibility.
Various antibodies have thus been conjugated to the surface of stealth liposomes to
produce immunoliposomes (left and centre) for active targeting, as opposed to the
passively targeted species discussed above. Stealth liposomes bearing an antibody
that specifcally binds intracellular myosin in ischemic or necrotic cardiomyocyte
targetsinfractedareasinrabbits.ReproducedwithpermissionfromV.P.Torchilin,
V.S.Trubetskoy,A.M.Milshteyn,J.Canillo,G.L.Wolf,M.I.Papisov,A.A.
Bogdanov, J. Narula, B.A. Khaw and V.G. Omelyanenko, Journal of Controlled
Release, 1994, 28, 45. 1994, Elsevier [79]
1.5 Mathematical Modelling of Drug Delivery [80]
The recent past has witnessed serious attempts in modelling drug delivery and
predictability of drug release as a prior and proper optimisation of drug delivery can
be expected to signifcantly improve in accuracy and ease of application. In the not too
distant future mathematical programs will be routinely used to help in optimising the
design of novel dosage forms. Considering the type of administration, drug dose to be
incorporated and targeted drug-release profle, mathematical predictions will allow
good estimates of the required composition, geometry, dimensions and preparation
procedure of the respective dosage forms. Another major advantage of mathematical
modelling is that it may signifcantly control the number of required experimental
studies to develop a new drug and/or optimise an existing drug product which directly
translates into lower cost and reduced time. The history of mathematical theory for
drug-delivery systems dates back to Higuchi, who in 1961 published his well-known
equations[81].LaternumerousmathematicalmodelssuchasthosegivenbySiepmann
andPeppas,LinandMetters,Arifnandsoon,havebeenevaluatedanddocumented
in the literature [8284]. The models, which were based on empirical, semi-empirical
and realistic theories, have inherent limitations such as:
Mostofthemodelslackaccuracyandsimplicityinapplication.
23
Introduction
Themathematicaltreatmentsarepurelydescriptiveandnotbasedonrealphysical,
chemical and/or biological phenomena.
The models were not fully capable of gaining insights into the underlying
mechanisms.
Thepredictivepowersoftheproposedmodelswerenotveryhigh.
With the advancement of time and knowledge other mechanistic mathematical
theories were put forward, which were based on physicochemical events like
diffusion, dissolution, swelling, erosion, precipitation and degradation [85, 86]. These
models were capable of providing deeper insights into the underlying drug-release
mechanisms. For example, the parameters accompanying the drug-release process like
drug diffusion, polymer swelling and so on, could be estimated. In these theories the
dosage form was treated like a real drug-delivery system associated with a well-defned
release mechanism. The models also enabled one to allow for quantitative prediction
of the effects of formulation and processing parameters such as the initial tablet height
and radius on the resulting drug-release kinetics. In this way a desired dosage form
with predetermined properties can be predicted before the actual formulation comes
into existence. There are certain considerations which must be kept in mind prior to
designing a mathematical model for a drug-delivery system:
Themodelmustnotbetoocomplex.Althoughitisknownthatamorecomplex
model that takes into account many system-specifc parameters is more successful
in predicting the release behaviour, using a complex mathematical treatment is a
tedious task. Thus, while developing a model for a specifc drug-delivery system
great care must be taken to consider only dominant physical, chemical and
biological parameters.
There should be close agreement between the mathematical predictions and
trends of the experimental data. This may be achieved either by comparing the
theoretical calculations with experimental results or by ftting the data to the
evaluated formulae.
Itmaynotbepossibletoderiveamodelthatmaybecloselyappliedtoallkinds
of drug-delivery systems. In fact, a generalised scheme applies only to a limited
number of drug-delivery systems and, for different drug-delivery systems with
special features, certain other parameters must be taken into consideration while
structuring a mathematical model.
Atheoreticalschemealwayshassomeshortcomingsasitmaynotconsiderthose
real parameters which are hidden in a drug-release experiment but dictate the release
profle and kinetic nature of the drug-delivery process. Therefore, more weight should
be given to experimental facts rather than theoretical and mathematical logic.
24
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
1.5.1 Factors Operative in Release Mechanisms
The overall mechanism of drug release depends on many factors such as type of
drug, solubility of the drug, dimensions of the carrier, drugpolymer interactions,
swellability of the carriers, percentage loading of the drug, chemical composition of
the drug vehicle, experimental pH, temperature and ionic strength, external stimuli
such as electric feld, magnetic feld, light and so on [87, 88]. In addition some specifc
parameters like wetting of the systems surface with water, water permeation into the
device, creation of water-flled permeation channels, presence of cracks within the
device, amount and type of interaction between the drugdrug, polymerpolymer,
degradation, dissolution and erosion of the device layers and so on, are also responsible
for overall release behaviour of the drug and polymer matrix. It is important to
mention here that it is virtually impossible to accommodate all these parameters
while deriving the mathematical model for the drug release. More importantly, the
models mentioned so for are valid only up to the drug transport phenomenon and
they are not predictive about the fate of the drug inside the body or the effectiveness
of the administered drug [89, 90].
Recently,sometheorieshavebeenproposedtodescribethetransportofdrugsinthe
human body organisms [91], yet their predictions are not very conclusive or close to
the actual experimental fndings. The reason is that these models have considered the
human body as a simplifed system of two or more well-stirred liquid compartments
and ignored many more signifcant factors such as enzymatic degradation of the
drug, protein binding, intra-cellular drug movement, interactions of drug and various
biofuids, presence of salt ions, permeability of drug to membranes, active and passive
drug uptake into the cells, frst pass metabolism, transport across the bloodbrain
barrier and other similar potential factors.
1.5.2 Empirical and Semi-empirical Mathematical Models
As discussed above, empirical/semi-empirical models should generally not be used
if the underlying drug-release mechanisms are to be elucidated and/or quantitative
predictions of the effects of formulation and/or processing parameters on the resulting
drug-release profles are to be made. However, such a descriptive mathematical analysis
can be useful for a comparison of different drug-release profles (e.g., for experimental
design studies). Semi-empirical models might be realistic in certain, extreme cases
and give an indication for the underlying drug-release mechanism under very specifc
conditions. Nevertheless, care has to be taken and the potential violation of model
assumptions must be carefully verifed.
25
Introduction
1.5.2.1 Peppas Equation
A very frequently used and easy-to-apply model to describe drug release is the so-
called Peppas equation, or power law [92] see Equation 1.3:

n
Mt
kt
M

(1.3)
Here, Mt and M are the absolute cumulative amounts of drug released at time t
and infnite time, respectively; k is a constant incorporating structural and geometric
characteristic of the system, and n is the release exponent, which might be indicative
of the mechanism of drug release. Nicholas Peppas was the frst to introduce this
equation in the feld of drug delivery. Clearly, the classical Higuchi equation [93],
as well as the above-described short-time approximation of the exact solution of
Ficks second law for thin flms with initial drug concentrations, which are below
drug solubility, represent the special case of the Peppas equation where the release
exponent is equal to 0.5. Thus, a release exponent of 0.5 can serve as an indication
for diffusion-controlled-drug release, but only if all assumptions these particular
solutions are based on are fulflled, for example flm geometry with negligible edge
effects, time- and position-independent diffusion coeffcients in a non-swellable and
insoluble matrix former. For other device geometries and pure drug-diffusion control,
different release exponent values have been derived [94]. In contrast, if polymer
swelling is the sole release-rate controlling mechanism and in the case of a delivery
system with flm geometry, zero-order drug-release kinetics are observed (as discussed
above), corresponding to a release exponent of n = 1. But, again, none of the model
assumptions for this particular case must be violated. For other geometries than that
of thin flms with negligible edge effects, different n values can serve as indicators
forpurelyswelling-controlleddrugdelivery.Releaseexponentsthatareinbetween
these extreme values for the respective device geometry indicate so-called anomalous
transport, thus, an overlapping of different types of phenomena, potentially including
drug diffusion and polymer swelling.
1.5.2.2 Hopfenberg Model
Hopfenberg [95] proposed an interesting semi-empirical model allowing for a
quantitative description of drug release from degradable drug-delivery systems
exhibiting a release rate that is proportional to the (time-dependent) surface area of
the device.
All mass transfer processes that are involved in the control of drug release are assumed
to add up to a single zero-order process (characterised by a rate constant, k
0
), which is
26
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
confned to the surface area of the system. This zero-order process might correspond
to one single physical or chemical phenomenon, but it might also result from the
superposition of several processes, such as dissolution, swelling and/or polymer
chain cleavage. The Hopfenberg model can for instance be applied to surface-eroding
polymer matrices for which a zero-order surface detachment of the drug is the rate-
limiting release step. The general equation is as follows in Equation 1.4:

0
0
1 1
n
k t Mt
M c a
j \

, (

( ,
(1.4)
Here, Mt and M are the cumulative absolute amounts of drug released at time t and
at infnite time, respectively; c
0
denotes the uniform initial drug concentration within
the system; a is the radius of a cylinder or sphere or the half-thickness of a slab; and n
is a shape factor representing spherical (n = 3), cylindrical (n = 2) or slab geometry
(n = 1). The model ignores edge and end effects.
1.5.2.3 Cooney Model
A more detailed analysis for spheres and cylinders undergoing surface erosion was
presented by Cooney [96]. Also his model is based on the assumption that there is one
single zero-order kinetics process, which is confned to the surface of the drug-delivery
system. As in the Hopfenberg model the release rate is assumed to be proportional to
the surface area of the device, which is time dependent. For a cylinder with the initial
length L
0
and initial diameter D
0
, the following equation was derived quantifying the
drug release rate f as a function of time t, see Equation 1.5:

( ) ( )( )
2
0 0 0
2
0 0 0
2 2 2 2
2
D Kt D Kt L Kt
f
D D L
+

+
(1.5)
where K is a constant. Figure 1.17 illustrates the effects of the ratio initial
length : initial diameter (L
0
/D
0
) of a cylinder on the resulting relative drug release rate
versus time (= relative dissolution rate in this example). When L
0
/D
0
approaches zero
(flm geometry) the curves transform into a horizontal line with a constant relative
drug release rate of 1. It is interesting to note that for disc-like cylinders (ratios of
L
0
/D
0
< 1, curves numbered 0.1, 0.2 and 0.5), the relative drug release rate remains
fnite up to complete drug release. In contrast, for rod-like cylinders (L
0
/D
0
> 1,
curves numbered 1, 2, 5 and infnity), the relative drug release rate approaches zero
at late time points.
27
Introduction
1.0
0.8
0.6
0.4
0.2
0
0
1
2
5
8
0.2
0.2
0.5
0.1
0.4 0.6 0.8 1.0
DIMENSIONLESS TIME
R
E
L
A
T
I
V
E

D
I
S
S
O
L
U
T
I
O
N

R
A
T
E
Figure 1.17 Effects of the ratio initial length : initial diameter (L
0
/D
0
) of a cylinder
on the resulting relative dissolution rate (or relative drug release rate) versus time
according to the semi-empirical Cooney model. The numbers given at the curves
indicate the respective L
0
/D
0
ratios. The curve for L
0
/D
0
approaching zero (flm
geometry) is a horizontal line at relative dissolution rate
1.5.2.4 Artifcial Neural Networks
Artifcial neural networks (ANN) can also be used to model drug delivery [97100].
The basic principle of this type of mathematical analysis is illustrated in Figure 1.18. An
ANN consists of one input layer, one output layer and one or more hidden intermediate
layers. Each layer is composed of several units, corresponding to neurons. The input
layer encompasses n input values of causal factors, e.g., the drug loading, compression
force or excipients content.
The output layer can for instance consist of constants describing the drug release
profle. As illustrated, the units of neighbouring layers are interconnected, the links
corresponding to synapses. The strength of these links can vary; they are also
called weights. Upon defnition of the model structure a set of experimental results
28
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
(consisting of input and output values) is used to train the network that is to defne
the optimal equations and weights allowing for the calculation of the output values
based on the input values. Thus, ANN can be considered as nonlinear regression
analysis tools. Once the system is trained, it can be used to make quantitative
predictions for the output values based on new input values. This type of analysis
was for instance used by Takahara and co-workers [97] to simulate the effects of the
amounts of microcrystalline cellulose and hydroxypropyl methylcellulose as well as of
the compression pressure used to prepare trapidil-loaded matrix tablets on the resulting
drug release kinetics. Ibric and co-workers [101] applied ANN to study acetylsalicylic
acid release from Eudragit RS-based matrix tablets, whereas theophylline release
from coated pellets was analysed by Ghaffari and co-workers [102] using this type
of mathematical modelling approach. A further interesting application of neural
networks in drug delivery was presented by Shao and co-workers [103], predicting
drug release from and tablet tensile strength of immediate release formulations.
Input layer
Hidden layer
Output layer
m j
i n 2 1
Y Y Y Y
X X X
2 1
Figure 1.18 Basic principle of mathematic modelling using ANN: X
i
represents
the input value of casual factors, n is the number of casual factors, Y
j
denotes the
output value of responses and Y
m
the number of responses. Between the input and
outputlayer,oneormorehiddenlayersarelocated.Reproducedwithpermission
from J. Takahara, K. Takayama and T. Nagai, Journal of Controlled Release,
1997, 49, 11. 1997, Elsevier [97]
29
Introduction
1.5.3 Mechanistic Realistic Models
The mechanistic realistic models are based on certain real phenomena such as mass
transport by diffusion, dissolution of drug, transition of a polymer from glassy to
rubbery state and so on [104]. The models involve certain mathematical operations
like partial differentiation equations which require known boundary conditions for
their solution. For example, before a device is put in contact with the release medium
the distribution of drug within the device, maintenance of perfect sink conditions,
movements of specifc boundaries and so on, must be known [105].
1.5.3.1 Theories Based on Ficks Law of Diffusion
If drug release is purely diffusion controlled with constant diffusion coeffcients, the
mathematical treatment can be rather straightforward. As illustrated in Figure 1.19,
different types of systems can be distinguished, including:
reservoirdevicesconsistingofadrugdepot,whichissurroundedbyarelease-
rate-controlling barrier membrane (often polymer-based).
monolithic systems, also called one-block systems, because there is no local
separation between a drug reservoir and a release-rate-controlling barrier.
For both types of systems two subclasses can be distinguished: the initial drug
concentration is either below or above drug solubility in the device. In the case of
a reservoir device with an initial drug concentration below drug solubility (e.g., a
polymer-coated tablet or pellet with a low drug loading), released drug molecules are
not replaced and the drug concentration at the inner membranes surface continuously
decreases with time (= non-constant activity source). If the membrane does not swell
or dissolve, if perfect sink conditions are provided throughout the release period and if
the drug permeability through the barrier remains constant, frst-order release kinetics
result, irrespective of the geometry of the device [106], see Equation 1.6:

0 t t t
dM ADKc M M ADK
dt l l V

(1.6)
where M
t
represents the absolute cumulative amount of drug released at time t; c
t

denotes the concentration of the drug in the release medium at time t; M
0
is the initial
amount of drug within the dosage form; V is the volume of the drug reservoir, A is
the total surface area of the device, l is the thickness of the membrane; K represents
the partition coeffcient of the drug between the membrane and the reservoir, and D
is the diffusion coeffcient of the drug within the membrane.
30
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Reserveir systems
Release rate controlling
membrane
Constant activity source
Monolithic solution
Monolithic disperation
Drug and matrix former
Miscellaneous systems
Monolithic systems
Geometry dependent release
(e.g., Higuchi equation for thin flims)
Geometry dependent release
Zero order release
First order release Non-constant activity source
Drug
reserveir
Figure 1.19 Classifcation systems for primarily diffusion-controlled drug-delivery
systems. Stars represent individual drug modules, black circles represent drug
crystals and/or amorphous aggregates. Only spherical dosage forms are illustrated,
buttheclassifcationsystemisapplicabletoanytypeofgeometry.Reproduced
with permission from J. Siepmann and F. Siepmann, International Journal of
Pharmaceuticals, 2008, 364, 328. 2008, Elsevier [80]
In contrast, if the initial drug concentration exceeds the drug solubility in a reservoir
device, released molecules are replaced by the (partial) dissolution of drug crystals/
amorphous aggregates, resulting in constant drug concentrations (saturated solutions)
at the inner membranes surface (constant activity source, Figure 1.19). If the properties
of the release-rate-controlling barrier (including its thickness and permeability for
the drug) remain constant and if perfect sink conditions are provided throughout the
release period, zero-order release kinetics result as long as drug excess is provided,
irrespective of the geometry of the system [106], see Equation 1.7:

lim t s
dM ADKc AJ
dt l l
(1.7)
31
Introduction
where M
t
is the amount of drug release at time t; dM
t
/dt denotes the steady state release
rate at time t; A is the total surface area of the device, J
lim
is the membrane-limiting
fux, l is the thickness of the membrane, D is the diffusion coeffcient of the drug within
the membrane, K is the partition coeffcient of the drug between the membrane and
the reservoir, and c
s
is the solubility of the drug in the reservoir. However, in practice
often deviations from these ideal systems are observed, for instance the flm coatings
show crack formation due to signifcant hydrostatic pressure built up within the device
or due to membrane swelling and/or (partial) dissolution [107109]. This renders the
mathematical treatment much more complicated and yet there is a signifcant lack of
mechanistic realistic mathematical theories taking these phenomena appropriately
into account. In the case of monolithic devices (Figure 1.19), the system geometry
signifcantly affects the resulting drug-release kinetics. If the initial drug concentration
is below drug solubility, the drug molecules are individualised/dissolved within the
carrier material (monolithic solution). Otherwise, dissolved drug molecules co-exist
with amorphous aggregates and/or drug crystals (monolithic dispersions). In the
case of monolithic solutions and in the absence of signifcant changes in the carrier
matrix during drug release (e.g., constant porosity, no swelling, time-independent
permeability for the drug) and if perfect sink conditions are maintained throughout
the release period and if drug release is primarily controlled by diffusion through
the carrier matrix, the resulting release can be calculated as follows, depending on
the systems geometry:
(i) In the case of thin flms with negligible edge effects [110], in Equation 1.8:

( )
( )
2
2
2 2 2
0
2 1
8 1
1 exp
2 1
t
n
D n t
M
M L
n

j \
+
, (
, (
+
( ,

(1.8)
where M
t
and M

denote the absolute cumulative amounts of drug released at time


t and infnity, respectively; n is a dummy variable, D is the diffusion coeffcient of
the drug within the matrix former and L is the thickness of the flm. To avoid the
use of infnite series of exponential functions, the following early and late time
approximations have been proposed for Equations 1.19 and 1.10
(1.9)
(1.10)
32
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
(ii) In the case of spherical dosage form ssee Equation 1.11:

2 2
2 2 2
1
6 1
1 exp
t
n
M Dn t
M n R

j \

, (
( ,

(1.11)
where M
t
and M

denote the absolute cumulative amounts of drug released at time t


and infnity, respectively, n is a dummy variable, D is the diffusion coeffcient of the
drug within the matrix former and R is the radius of the sphere. This equation has
for example successfully been used to quantify drug release from non-degradable
controlled-release microparticles [111]. Based on the mathematical analysis, deeper
insight into the changes in the systems composition during drug release could be
gained. Figure 1.20 shows for instance the concentration profles of propranolol
hydrochloride within ammoniomethacrylate copolymer-based microparticles after 5
minutes, 1 hour and 8 hours exposure to phosphate buffer (pH 7.4).
(iii) In the case of cylinders (considering axial as well as radial mass transport) [112]
see Equation 1.12:

( )
( )
2
2 2 2
1
2
2
2 2
1
32 1
1 exp
2 1
1
exp
2 1
t n
n
n
n
M q
Dt
M q R
p
x Dt
H
p

j \

, (
( ,
j \
+
, (
, (
+
( ,

(1.12)
where M
t
and M

denote the absolute cumulative amounts of drug released at time


t and infnity, respectively, n and p denote dummy variables, the q
n
are the roots of
the Bessel function of the frst kind of zero order [J
0
(q
n
) = 0] and R and H denote the
radius and height of the cylinder. This equation can for instance be used to quantify
drug release from lipid implants [113]. As an example, the release of the protein drug
rh-interferon 2a (IFN-) from tristearin-based cylinders can successfully be described
[114]. Interestingly, the addition of poly(ethylene glycol) (PEG) (which is commonly
used as a pore former in inert matrices) results in protein precipitation/very limited
IFN- solubility within the water-flled pores of the implants and, thus, signifcant
deviations from Equation 1.9 [114]. When considering also potentially limited
local drug solubility, time- and position-dependent PEG concentrations and implant
porosity, resulting in time and position-dependent drug diffusion coeffcients, this more
comprehensive mathematical theory is able to quantitatively describe the resulting
protein release kinetics [115]. However, due to the complexity of the respective set of
partial differential equations, no analytical solution can be derived for this theory, but
33
Introduction
numerical analysis can be used for the implementation of the model. Importantly, this
type of mathematical theory is not only able to give deeper insight into the underlying
drug release mechanisms (e.g., relative importance of drug diffusion, limited solubility
and changes in local porosity), but also allows for quantitative predictions of the
resulting drug release kinetics as a function of the device design. Figure 1.21 shows
as an example the theoretically predicted and experimentally verifed release of IFN-
into phosphate buffer pH 7.4 from tristearin-based implants containing 10% IFN-/
hydroxypropyl--cyclodextrin (HP--CD) and 20% PEG. For monolithic dispersions
the mathematical description becomes even more complex.
(a) (c)
r
r
r
c
1
/c
0
c
1
/c
0
c
1
/c
0
r
1
z
z
z
z
0
1
1
0
0
(b) (d)
Figure 1.20 Calculated changes in the drug concentration gradients with spherical,
propranolol HCl-loaded microparticles upon exposure to phosphate buffer pH 7.4:
(a) illustration of the point of view, (b) concentration profle after 5 minutes, (c) 1
hour,and(d)8hours.ReproducedwithpermissionfromM.Hombreiro-Prez,J.
Siepmann,C.Zinutti,A.Lamprecht,N.Ubrich,M.Hoffman,R.BodmeierandP.
Maincent, Journal of Controlled Release, 2003, 88, 413. 2003, Elsevier [111]
34
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
75
experiment
theory
time, d
c
u
m
u
l
a
t
i
v
e

I
F
N


r
e
l
e
a
s
e
,

%
100
50
25
0
0 5 10 15
Figure 1.21 Theory and experiment: IFN- release into phosphate buffer pH 7.4
from tristearin-based implants: theoretical prediction (curve) and independent
experimental verifcation (symbols) in the case of implants initially containing
10% IFN-/HP--CD and 20% PEG (experimental results: average SD; n = 3).
ReproducedwithpermissionfromF.Siepmann,S.Herrmann,G.WinterandJ.
Siepmann, Journal of Controlled Release, 2008, 128, 233. 2008, Elsevier [115]
For the simplest geometry of thin flms with negligible edge effects, Higuchi published
the famous square root of time relationship between the amount of drug released
from a thin ointment flm with a large excess of drug (initial drug concentration drug
solubility in the carrier material) in 1961 [116], see Equation 1.13:

( )
0
2
s s
Mt
D c c c t
A
(1.13)
where Mt is the cumulative absolute amount of drug released at time t, A is the surface
area of the flm exposed to the release medium, D is the drug diffusivity in the carrier
35
Introduction
material and c
0
and c
s
represent the initial drug concentration and the solubility of
the drug in the carrier material, respectively. An important advantage of this equation
is its simplicity. However, when applying it to controlled drug-delivery systems, the
assumptions Higuchi based this equation on must be fulflled, including:
The initial drug concentration in the system must be much higher than drug
solubility. This aspect is crucial, because it provides the basis for the justifcation
of the applied pseudo-steady-state approach. The concentration profle of a drug
that is homogeneously suspended within an ointment is illustrated in Figure
1.22.
h
c0
cs
perfect sink
direction of drug release
Figure 1.22 Pseudo-steady-state approach applied for the derivation of the classical
Higuchi equation. Theoretical concentration profle existing in an ointment
containingsuspendeddrugandincontactwithaperfectsink.Reproduced
with permission from J. Siepmann and F. Siepmann, International Journal of
Pharmaceuticals, 2008, 364, 328. 2008, Elsevier [80]
The solid line represents the concentration profle after exposure of the ointment to
a perfect sink for a certain time t. Importantly, a sharp discontinuity is observed at
distance h from the surface/release medium. For this distance h the concentration
gradient is essentially constant, provided the initial drug concentration within the
system, c
0
, is much greater than the solubility of the drug (c
0
> c
s
) (pseudo-steady-
state). After an additional time interval, dt, the new concentration profle of the drug
36
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
is indicated by the dotted line. Again, a sharp discontinuity and otherwise linear
concentration profles result.
Thedevicegeometryisthatofathinflmwithnegligibleedgeeffects
Thesizeofthedrugparticlesismuchsmallerthanthethicknessoftheflm
Thecarriermaterialdoesnotswellordissolve
Thediffusivityofthedrugisconstant(notdependentontimeorposition)
Perfectsinkconditionsaremaintainedthroughouttheexperiment
Unfortunately, Equation 1.11 is often misused and applied to controlled drug-delivery
systems which do not fulfl all these assumptions. In these cases, any conclusion should
be viewed with great caution.
Even if the cumulative amount of drug that is released from a particular drug-delivery
system is proportional to the square root of time, this does not necessarily mean that
the underlying drug release mechanism is the same as in the ointment Higuchi studied.
For instance, the superposition of various other physicochemical phenomena (such as
polymer swelling, time- and position-dependent changes in the diffusion coeffcients of
water and drug) might result in an apparent square root of time kinetics. Furthermore,
as discussed previously, the cumulative amount of drug released is proportional to
the square root of time in the early time approximation for monolithic solutions with
flm geometry (Equation 1.7).
For monolithic dispersions with geometries other than that of a thin flm with
negligible edge effects, the reader is referred to the literature [117]. If both diffusion
through the inner device matrix as well as diffusion through a surrounding barrier
membrane are of importance for drug release (Figure 1.19, miscellaneous systems),
the mathematical modelling is also more complex and geometry dependent. Again,
the reader is referred to the literature for more details [118].
1.5.3.2 Theories Considering Polymer Swelling
If polymer swelling is of importance for the control of drug release, e.g., as in the
case of hydroxypropyl methylcellulose (HPMC)-based matrix tablets, the transition
of the macromolecules from the glassy (less mobile) to the rubbery (more mobile)
state has to be considered in the model [119]. The two most important consequences
of signifcant polymer swelling in a controlled-release matrix system are:
The length of the diffusion pathways increases, resulting in decreasing drug
37
Introduction
concentration gradients (being the driving forces for diffusion) and, thus,
potentially decreasing drug release rates.
Themobilityofthemacromoleculessignifcantlyincreases,resultinginincreased
drug mobility and, thus, potentially increasing drug release rates. In dry tablets,
diffusion is often negligible (diffusivities close to zero). In contrast, in a fully
swollen polymer matrix the diffusion coeffcient of the drug can be of the same
order of magnitude as in an aqueous solution.
Depending on the type of polymer and type of drug-delivery system, one of these effects
potentially dominates, resulting in decreasing or increasing drug release rates.
Figure 1.23 schematically shows the physical phenomena which can be involved in
the control of drug release from a swellable delivery system.
Bulk fluid
Dissolved drug only
Dissolved and
dispersed drug
Swollen matrix Non-swollen matrix
Diffusion front Swelling front Erosion front
Figure 1.23 Schematic presentation of a swelling-controlled drug-delivery system
containing dissolved and dispersed drug (stars and black circles, respectively),
exhibiting the following moving boundaries: (i) an erosion front, separating the
bulk fuid from the delivery system, (ii) a diffusion front, separating the swollen
matrix containing dissolved drug only and the swollen matrix containing dissolved
and dispersed drugs; and (iii) a swelling front, separating the swollen and non-
swollen matrix
38
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
This might represent a cross section through half of a matrix tablet which is exposed
to an aqueous bulk fuid in radial direction. On the right-hand side the inner tablet
core is still dry and in the glassy state (non-swollen); on the left-hand side the bulk
fuid is located. Upon contact with the release medium, water diffuses into the system.
With increasing water content, the mobility of the polymer chains (and, thus, also
drug molecules) increases. As soon as a certain polymer-specifc water concentration
is reached, the macromolecular mobility steeply increases. This phenomenon is
called polymer chain relaxation or glassy-to-rubbery phase-transition. The front
at which this process takes place is called the swelling front, which separates the
swollen from the non-swollen matrix. Importantly, this is not a stationary boundary,
but a moving one. If the initial drug concentration in the delivery system exceeds
drug solubility, dissolved and non-dissolved drug co-exist within the matrix. Due
to concentration gradients and the signifcantly increased mobility, dissolved drug
molecules diffuse out of the swollen matrix into the release medium. As long as a non-
dissolved excess of drug exists, the concentration of dissolved drug in this part of the
system is constant (drug molecules that are released are replaced by the dissolution
of non-dissolved drug, providing a saturated solution). But as soon as all excess drug
is dissolved, the concentration within the swollen matrix decreases. The front that
separates the swollen matrix containing only dissolved drug from the swollen matrix
that contains both dissolved and non-dissolved drug is called the diffusion front
(Figure 1.23). Importantly, this front is also moving. Furthermore, a third front can
be distinguished, which separates the drug-delivery system from the release medium
and which is also moving. In the case of water-soluble matrix formers, this front
is called the erosion front. If the polymer relaxation process is rate limiting (e.g.,
all other phenomena, such as diffusion and dissolution, are much faster) and if the
device has the geometry of a thin flm (with negligible edge effects) and an initial
homogeneous drug and polymer distribution, zero-order drug-release kinetics result,
because the rate at which the swelling front moves is independent of its position (and,
thus, constant). However, in the case of the geometry of a sphere or a cylinder, the
movement of a swelling front at a constant rate does not result in zero-order release
kinetics, but in a proportionality of the cumulative amount of drug released to the
time to the power of 0.85 and 0.89, respectively, (due to the change in the surface
area that is affected by the swelling with time) [120]. A very interesting, mechanistic
realistic mathematical theory allowing for the quantifcation of drug release from
swellable polymer flms has been proposed by Korsmeyer and co-workers [121]. It
allows for a simultaneous consideration of the diffusion of water into the device and
drug out of the system as well as of polymer swelling. To account for the increase
in water and drug mobility with increasing water content of the polymer matrix, a
Fujita-type exponential relationship was chosen [122] and shown to be appropriate
for the prediction of different types of transport behaviours. Dimensional changes in
the flms are accounted for by allowing each spatial increment to expand according to
the amount of water that diffused in. At early time points, the swelling is restricted to
39
Introduction
one-dimensional by the glassy core of the sample. At later time points, when enough
water has penetrated into the core of the system to plasticise it, the sample relaxes
to an isotropically swollen state.
Afterwards, swelling is three-dimensional. Under these conditions, water (subscript
1) diffusion can be described as follows in Equation 1.14:

1 1
1
c c
D

j \

, (

( ,
(1.14)
where D
1
is the diffusion coeffcient of water and c
1
is the normalised water
concentration in Equation 1.15:

1
,
w
w e
c
c
c
(1.15)
Here, c
w
is the water concentration in the flm at a particular position and c
w,e
is
the equilibrium water concentration in the system. Time t is scaled according to the
water diffusivity in the fully swollen system, D
1,s
, and the dry thickness of the flm,
L
0
, see Equation 1.16:

1,
2
0
s
tD
L
(1.16)
The spatial coordinate x is normalised with respect to the dry thickness of the thin
flm, see Equation 1.17:

0
x
L
(1.17)
To describe drug diffusion (subscript 2), the following Equations 1.18 and 1.19 are
used:

2 2
2
c c
D

j \

, (

( ,
(1.18)

2
,
s
s i
c
c
c
(1.19)
40
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Here, D
2
is the diffusion coeffcient of the drug and C
2
is the normalised drug
concentration; C
s
denotes the drug concentration in the flm and C
s,i
the initial drug
concentration in the system. The following boundary conditions are considered in
Equations 1.20 and 1.21:

( ) ( )
1 1
0, , 1 c c (1.20)

( ) ( )
2 2
0, , 0 c c (1.21)
where 0 and are the two surfaces of the thin flm. Note that describes the
continuously moving outside surface of the flm. The following initial conditions are
considered in Equations 1.22 and 1.23:

( )
1
, 0 0 c (1.22)

( )
2
, 0 1 c (1.23)
Due to the complexity of this set of partial differential equations, the latter was
solved numerically. As can be seen in Figure 1.24, good agreement between theory
and experiment was obtained when ftting this model to sets of experimentally
measure theophylline release kinetics from (hydroxyethyl methacrylate-co-N-vinyl-
2-pyrrolidone) copolymer-based flms.
1.5.3.3 Theories Considering Polymer Swelling and Polymer and Drug
Dissolution
In practice, often even more processes are simultaneously involved in the control of
drug release from oral controlled-release matrix tablets: generally, the matrix former
is water-soluble. Thus, polymer dissolution must also be taken into account. Different
comprehensive mathematical theories have been proposed aiming to describe this
type of drug-delivery system [123]. In the following only one example will briefy
be described. The reader is referred to the literature for more details. The so-called
sequential layer model takes into account the diffusion of water and drug with time-
and position-dependent diffusivities, moving boundary conditions, the swelling of
the system, polymer and drug dissolution and radial and axial mass transfer within
cylindrical tablets. The model was successfully ftted to drug-release kinetics from
matrices based on HPMC and HPMC derivatives, e.g., hydroxypropyl methylcellulose
41
Introduction
acetate succinate [124]. The theory is applicable to freely and poorly water-soluble
drugs and a wide range of initial drug loadings. Its practical usefulness could be
demonstrated via quantitative predictions of the effects of the design parameters
of HPMC-based controlled-release matrix tablets (including the size, shape and
composition of the systems) on the resulting drug release kinetics. Water and drug
diffusion are considered based on Ficks second law of diffusion for cylindrical
geometry, taking into account axial and radial mass transport and concentration-
dependent diffusivities see Equation 1.24:

1
k k k k k
k k
c c D c c
rD rD
t r r r r z z
j \ j \ j \
+ +

, ( , ( , (

( , ( , ( ,

(1.24)
0.8
1.0
0.6
M
t
2

/
M

0.4
0.2
0.0
0.0 0.25 0.50 0.75 1.0

1.25
Figure 1.24 Fit of the KorsmeyerPeppas model to experimentally determined
theophylline release kinetics from hydroxyethyl methacrylate-co-N-vinyl-2-
pyrrolidone copolymer-based flms (curve = theory, symbols = experiment)
42
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Here, c
k
and D
k
are the concentration and diffusion coeffcient of the diffusing species
(k = 1 for water, k = 2 for the drug), respectively, r denotes the radial coordinate, z is
the axial co-ordinate, is the angular coordinate (Figure 1.25a) and t represents time.
Analogous to the KorsmeyerPeppas model described above, a Fujita-type exponential
dependence of the water and drug diffusion coeffcients on the water content of the
system is taken into account in Equation 1.25:

1
1
exp 1
k kcrit k
crit
c
D D
c


j \


, (

( ,

(1.25)
where
1
and
2
are dimensionless constants characterising this concentration
dependence. Also D
1kcrit
and D
2crit
denote the diffusion coeffcients of water and drug
at the interface tablet/release medium, where polymer chain disentanglement occurs
[125]. Ideal mixing is assumed (no volume contraction upon mixing drug, polymer
and water), and the total volume of the system at any time point is given by the sum
of the volumes of the single components.
The calculation of the new tablet dimensions is based on a mass balance considering
drug, polymer and water. Polymer dissolution is taken into account using the reptation
theory [126]: Above a certain critical water concentration (c
1crit
), the polymer chains at
the surface of the tablet start to disentangle and diffuse through the liquid, unstirred
layer surrounding the device into the bulk fuid (release medium). A dissolution rate
constant, k
diss
, is considered characterising the polymer mass loss velocity, which is
normalised to the actual surface area of the system see Equation 1.25:

p p0 t diss t
M M k At (1.26)
Here, M
pt
and M
p0
are the dry polymer matrix mass at time t and t = 0, respectively;
A
t
denotes the surface area of the device at time t. The initial conditions refect the
fact that the matrix is dry and the drug uniformly distributed throughout the device
at t = 0. The boundary conditions are defned as follows: the water concentration at
the surface of the matrix, c
1crit
, is calculated from the critical polymer disentanglement
concentration [127]. The drug concentration at the surface of the tablet is assumed
to be equal to zero (perfect sink condition). In order to reduce computation time, the
origin of the coordinate system is placed at the centre of the cylinder, resulting in two
symmetry planes for the drug and water concentration profles (Figure 1.25b). Thus,
only the concentration profles within a quarter of the tablet need to be calculated.
Due to the complexity of the resulting set of partial differential equations, also
in this case a numerical solution is required. Figure 1.26 shows an example for a
practical application of this mathematical model: the theoretically predicted effects
of the initial radius of HPMC-based matrix tablets (with an initial height of 2.6 mm,
43
Introduction
composition: 50% drug and 50% HPMC) on the resulting relative and absolute
release of theophylline into phosphate buffer pH 7.4 is illustrated.
(a)
(b)
z

r
r = 0
z = 0
R
t
Z
t
Figure 1.25 Mathematic modelling of drug release from HPMC-based matrix
tables: (a) scheme of a cylindrical tablet for mathematical analysis, with (b)
symmetry planes in axial and radial direction for the water and drug concentration
profles(R
t
and Z
t
represent the time-dependent radius and half-height of the
cylinder, respectively)
44
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
(b)
0
25
time, h
50
d
r
u
g

r
e
l
e
a
s
e
d
,

%
75
100
2 0 4 6
1 mm
4 mm
6.5 mm
8
(a)
0
30
time, h
60
d
r
u
g

r
e
l
e
a
s
e
d
,

m
g
90
120
2 0 4 6 8
1 mm
4 mm
6.5 mm
Figure 1.26 Practical application of the sequential layer model: Theoretically
predicted effects of the initial tablet radius on the release patterns of theophylline
from HPMC-based matrix tablets in phosphate buffer pH 7.4 and experimental
verifcation: (a) relative amount of drug released and (b) absolute amount of drug
released versus time (37 C, initial table height = 2.6 mm, initial tablet radius
indicated in the fgures, 50% (w/w) initial drug loading) (curves: predicted values,
symbols: independent experimental date)
The curves show the theoretically predicted drug release profles. Then, in a second
step, the respective drug release rates were determined experimentally (symbols in
Figure 1.26). As it can be seen, good agreement between theory and experiment was
obtained in all cases.
45
Introduction
1.5.3.4 Theories Considering Polymer Erosion/Degradation
Unfortunately, the terms erosion and degradation are not uniformly used in
the literature. In this article, the following defnitions are applied [128]: polymer
degradation is the chain scission process by which polymer chains are cleaved into
oligomers and monomers. In contrast, erosion is defned as the process of material
loss from the polymer bulk. Such materials may be monomers, oligomers, parts of
the polymer backbone or even parts of the polymer bulk. Thus, the degradation of
water-insoluble polymers is part of their erosion process. Depending on the relative
rates of water penetration into such systems and of polymer chain cleavage, two
extreme types of erosion can be distinguished: surface (or heterogeneous) erosion
and bulk (or homogeneous) erosion [129]. In the frst case, the polymer chain
cleavage is much faster than the water penetration into the system. Consequently,
the degradation process is mostly restricted to the outermost polymer layers and the
erosion predominantly affects the surface, and not the inner parts of the device. In
contrast, if water penetration is much more rapid than polymer chain cleavage, the
entire system is rapidly wetted and degradation occurs throughout the device (bulk
erosion). Generally, drug-delivery systems which are based on polymers with highly
reactive bonds (e.g., polyanhydrides) in their backbone structure undergo surface
erosion, whereas devices that are based on polymers with less reactive functional
groups [e.g., poly(lactic-co-glycolicacid)(PLGA)]tendtobebulk eroding. However,
note that the dimensions of the drug-delivery system affect the relative water
penetrationrateintothedeviceandthatforinstanceaPLGA-basedsphereofthesize
of the moon would show surface erosion [130]. An interesting mathematical theory
for surface eroding drug-deliverysystemswithflmgeometrywasproposedbyLeein
1980 [131]. It is an analytical solution that is valid for different drug loading : drug
solubility ratios. As illustrated in Figure 1.27a, the movements of two fronts are
considered: a diffusion front and an erosion front.
Here, R denotes the time-dependent position of the diffusion front and S the time-
dependent position of the erosion front; A is the initial drug concentration within
the delivery system, which is above drug solubility, C
s
(monolithic dispersion); C
b

represents the drug concentration in the well-stirred release medium, and x the
position (with x = 0 at the centre and x = a at the surface of the flm). It is assumed
that the erosion front moves at a constant velocity, that edge effects are negligible
and that perfect sink conditions are maintained throughout the experiment. Under
these conditions, Lee derived the following equations (1.27-1.29) allowing for a
quantitative description of drug release:

3
1
2 6
t s
M C a Ba
M D A

j \
+ +
, (
( ,
(1.27)
46
Stimuli Responsive Drug Delivery Systems: From Introduction to Application

2
3
1 2
s s
A A
a h h h
C C

j \
+ +
, (
( ,
(1.28)

1
1
2
s
Ba A
h
D C
j \

, (
( ,
(1.29)
M
M
(a)
Diffusing
Front
Un-
dissolved
Solute
Eroding
Front
Initial
Surface
Planar Erodible Matrix
X
Ca
Cb
A
C

1.0
1.8
1.6
1.4
1.5 3
5
10
0.2
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
O R S a
(b)
= 1
A
Cs
1
Ba
D
Figure 1.27 Modelling drug release from surface-eroding monolithic dispersions
with flm geometry: (a) scheme of the drug concentration profle within the system
according to [131]. Two moving fronts are considered: a diffusion front and an
erosion front. (b) Calculated drug release profles as a function of the initial drug
loading : drug solubility ratio (A/C). The parameter of Ba/D serves as a measure
fortherelativecontributionoferosionanddiffusion.Reproducedwithpermission
fromP.I.Lee,Journal of Membrane Science, 1980, 7, 255. 1980, Elsevier [131]
Here, M
t
and M

are the cumulative absolute amounts of drug released at time


t and at infnite time, respectively, 1 denotes the relative separation between the
47
Introduction
diffusion and erosion fronts [1 = (S R)/a]; B is the surface erosion rate constant
with the dimensions of a velocity; a represents the half-thickness of the flm, D the
drug diffusivity within the system, and is the dimensionless time ( = Dt/a2). The
parameter Ba/D is a measure for the relative contribution of erosion and diffusion to
drug release. The calculated effects of the initial drug loading : drug solubility ratio
(A/C
s
) on the resulting drug-release patterns are illustrated in Figure 1.27b. In this
example, the relative contributions of erosion and diffusion (represented by the term
Ba/D) are kept constant (= 1). As can be seen, the relative drug-release rate decreases
with increasing A/C
s
ratio.
The model predicts that the release approaches zero-order kinetics when the initial
drug loading becomes much higher than drug solubility in the matrix. As polymer
chain cleavage is a random process, Monte Carlo simulations can effectively be used
tosimulatepolymerdegradation.Reference[132]wasthefrsttoproposethistype
of theory allowing for a quantitative description of drug release from surface-eroding
polymer matrices. The basic idea is to represent polymer matrix cross sections by
two-dimensional grids. Each pixel represents one of the systems components: drug,
polymer and potentially fller and pores. To simulate drug or polymer dissolution a
so-called life expectancy is defned for each type of pixel. As soon as a pixel comes
into contact with water, its lifetime starts to decrease. Once the lifetime expires, the
pixel is assumed to dissolve instantaneously. Importantly, different life expectancies
can be defned for the involved system compounds, taking into account differences in
their dissolution rates. However, diffusional mass transport is not taken into account.
The frst to combine Monte Carlo simulation to account for polymer degradation
and diffusional mass transport (based on Ficks second law) was Achim Goepferich
[133]. He developed theories that are applicable to surface-eroding systems, but also
models for bulk-eroding devices. Furthermore, drug-delivery systems containing
both surface- and bulk-eroding polymers can be considered, containing for instance
poly(d,l-lactic acid) and poly[1,3-bis(p carboxy phenoxy)propane-sebacic acid] [134].
In addition, the potential crystallisation of polymer degradation products and micro
environmentalpHeffectscanbetakenintoaccount.Lateron,asimilarapproach
(combining Monte Carlo simulations with diffusional mass transport) was used
toquantifydrugreleasefromsphericalPLGA-basedmicroparticles[135].Forthe
mathematical analysis the latter are divided into concentric rings of equal volume
(Figure 1.28, the rings are described upon rotation of the pixels shown in Figure
1.28b around the z-axis). Due to the symmetry planes at the r = 0 and z = 0 planes
(in the case of homogenous initial drug and polymer distribution), it is suffcient to
calculate the mass transport phenomena in only one quarter of the two-dimensional
circle shown in Figure 1.28b (Figure 1.29a). At t = 0 each ring represents either drug
or non-degraded polymer.
48
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
(a)
(b)
Z
X

Figure 1.28SchematicpresentationofasphericalPLGA-basedmicroparticlefor
mathematical analysis: (a) three-dimensional geometry, and (b) two-dimensional
cross section with two-dimensional pixel grids. Upon rotation of the latter around
the z-axis, rings of identical volume are described
Due to the identical volume of the polymer rings it is reasonable to assume that each
of them contains a similar number of cleavable ester bonds. Thus, the probability with
which a ring representing non-degraded polymer degrades upon its frst contact with
water is similar for all rings. As described previously, lifetime expectancies are assigned
to all polymer pixels (rings), refecting the degradation rate of the macromolecules.
Importantly, knowing the status of each pixel (ring) (non-eroded polymer or pore)
at each time point, the microparticle porosities in radial and axial direction (depending
on time and position) can be calculated (Figure 1.29b).
Based on these porosity values, the position- and direction-dependent drug diffusivities
within the spheres can be calculated as a function of the exposure time to the release
medium. This information is essential for the accurate calculation of the diffusional
49
Introduction
mass transport processes using Ficks second law see Equation 1.30:

1 c c D c c
rD rD
t r r r r z z

j \ j \ j \
+ +

, ( , ( , (

( , ( , ( ,

(1.30)
(a)
(b)
z
r
r
drug
non-degraded polymer
drug
pore
non-degraded polymer
z
J
0
0
11
11
12
1+1
2
1
1
r(1)
r(1)
r(1)
r(1)
radius(1)
radius(1)
1 1
0
11
11
12
1+1
2 1 1 1
2
0
1
2
J1
J+1
J1
J+1
J1
J2
J
J1
J2
H
e
i
g
h
t

(
J
)
H
e
i
g
h
t

(
J
)
Figure 1.29 Principle of a Monte Carlo-based approach to simulate polymer
degradationanddiffusiondrugreleasefromPLGA-basedmicroparticles.
Schematic structure of the system (one quarter of the two-dimensional grid shown
in Figure 1.28b): (a) at time t = 0 (before exposure to the release medium), and
(b) during drug release. Grey, dotted and white pixels represent non-degraded
polymer, drug and pores, respectively
50
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
Here, c and D are the concentration and diffusion coeffcient of the drug; r denotes
the radial coordinate, z the axial coordinate and the angle perpendicular to the rz-
plane. In addition, the limited solubility of the drug within the system is taken into
account: only drug which is soluble under the given conditions is considered to be
available for diffusion. Taking into account the given initial and boundary conditions
(initial homogeneous drug distribution and perfect sink conditions), the respective set
of partial differential equations is solved numerically, using fnite differences (since the
diffusion coeffcients are time- and position-dependent there is no analytical solution).
Importantly, good agreement between theory and experiment was obtained when
ftting this model to experimentally measured drug release from 5-fuouracil-loaded,
PLGA-basedmicroparticlesinphosphatebufferpH7.4(Figure 1.30).
100
experiment
theory
time, d
d
r
u
g

r
e
l
e
a
s
e
d
,

%
75
50
25
0
0 7 14 21
Figure 1.30 Fit of a mechanistic realistic mathematic theory based on Monte
Carlo-simulations and considering diffusional mass transport as well as limited
local drug solubility to experimentally determined drug release from 5-fuouracil-
loaded,PLGA-basedmicroparticlesinphosphatebufferpH7.4:experimental
results (symbols) and ftted theory (curve)
Based on these calculations, system-specifc parameters can be determined and the
dominant physical and chemical phenomena in each of the release periods can be
51
Introduction
identifed. For instance, it can be shown that in this specifc system the initial burst
release is primarily controlled by pure drug diffusion. Furthermore, the model allows
for quantitative predictions of the effects of formulation and processing parameters,
including the initial microparticle size and drug loading.
1.6 Some Milestones in the Fields of Controlled Drug Delivery
Here are shown some diagrams that represent milestones in the history of controlled
drug-delivery systems (Figures 1.31-1.34).
Sprayer
Hot air
Microspheres
Drug
Drug Loaded
Microsphere
Drug
Polymer
Solution
Figure 1.31Drug-loadedPLGAmicroparticleswerepreparedinaprocessknown
as the Prolease process, which was invented and patented by Gombotz, Healy
andBrown,andassignedtoEnzyrech,Inc.,acompanyfoundedbyBobLanger
[136](ReproducedwithalittlemodifcationfromthePatentfledbyGombotzand
coworkers [136]).
52
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
A rotary die cutting
press executes the
final manufacturing
step for an Alza skin
patch.
(photo taken from an Alza
annual report)
Figure 1.32 One of the earliest (if not the earliest) patents on the skin patch was a
1971patentissuedtoAlexZafforoniandassignedtoAlza[137].(Reproduced
with permission from the Patent fled by Zafforoni and co-workers [137]).
NCS
x y
NCS
(CH
2
CH ) ( CH OH )

O C

C O

OH
Styrene-Maleic Anhydride (SMA) conjugated
NeoCarcinoStatin (NCS) (SMANCS)
for treating liver cancer
Figure 1.33 Hiroshi Maeda synthesised the polymer-drug conjugated of
poly(styrene-co maleric anhydride) SMA and Neo Carcino Statin (NCS) and called
it SMANCS
53
Introduction
Drug (DOXO)
Lipid bilayer
membrane
PEG
corona
Figure 1.34 A PEGylated liposome containing the anticancer drug doxorubicin
wasdevelopedintheearly1990satLiposomeTechnology,Inc.,byM.Woodle
and F. Martin. It is called Doxil and was approved for clinical use in 1995. The
PEGylated carrier was called stealth because of its long circulating characteristic
1.7 Future Challenges and Scope
The concept of delivery of drugs from a suitable device induced by external signals
is not as simple as it looks. There are many considerations that have to be taken into
account while designing a desired drug carrier. It is, therefore, a joint responsibility
of a polymer chemist and a pharmaceutical expert to think of certain factors prior
to building up a scenario for responsive drug-delivery systems. These factors may
be, for example, biocompatibility of the device, cytotoxicity, in vivo studies, Food
and Drug Administration (FDA) approval, effciency, inconvenience caused to
patients, cost effectiveness, and so on. Only after these factors are examined can
a drug-delivery system be therapeutically acceptable. The delivery of a drug at a
predetermined rate over a specifed time to a selected target organ has been the ideal
requisite in drug-delivery technology and pharmacokinetics. Moreover, the need for
carriers that exhibit oscillatory behaviour of the releasing bioactive agent has also
emerged as a signifcant problem of drug design and formulation in recent years.
The traditional methods of drug administration in conventional forms, such as pills
and subcutaneous or intravenous injection, are still the predominant routes for drug
administration. But pills and injections offer limited control over the rate of drug
54
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
release into the body; usually they are associated with an immediate release of the
drug. Consequently, to achieve therapeutic levels that extend over time, the initial
concentration of the drug in the body must be high, causing peaks that gradually
diminish over time to an ineffective level. In this mode of delivery, the duration of the
therapeutic effect depends on the frequency of dose administration and the half-life
of the drug. This peak-and-valley delivery is known to cause toxicity in certain cases,
most frequently with chemotherapy for cancer. Thus, the design of a drug-delivery
system with optimum performance in specifc circumstances poses challenges. In an
overview of the whole scenario, the feld of drug-delivery systems has to confront
the following challenges:
Improvedeffcacy
Targeteddeliveryandreducedsideeffects
Optimumperformance
Interfacingandpacingwithmodernmethodologies
Guaranteesofsafeenvironment
Easeoffabricationandapplicationinreality
These benefts may be realised by adopting approaches that basically involve judicious
combination of highly specifc monomers and polymers of both synthetic and natural
origin. The use of smart materials in drug-delivery technologies has not only to focus
on the possible medical benefts but also must consider economic aspects of the
developed materials and/or technology. Furthermore, huge effort on synthetic polymer
chemistry must be undertaken to design tailor-made macromolecular systems that will
offer novelty in their operation and performance. Above all, the systems developed
must be acceptable to the patient community who are the end-users of any successful
research and technology. A logical consideration of the possibilities about bright
prospects for controlled drug delivery gives rise to positive signals and, therefore,
more effort deserves to be put into its growth and expansion. Since smart materials
have specifc modes of operability and are prone to typical experimental conditions,
there is large scope for synthetic polymer chemistry to design multiresponsive delivery
systems. Despite the tremendous research input that has been applied to achieve high-
performance technologies, a number of aspects still remain to be worked on:
Designofdrug-deliverysystemswithmultistimuli-responsivepotential
More precise synthetic routes for making responsive materials with greater
responsive sensitivity
55
Introduction
Assuranceofeconomicviabilitysoastopopularisedevicesonalargecommercial
and population scale
Designofmorelocaliseddrug-deliverysystems
Oral delivery of insulin using body-friendly natural polymers with enhanced
absorption in blood
Thus, it may be concluded that, although much advancement has been demonstrated
by untiring efforts of researchers worldwide, still there exist numerous challenges that
have to be addressed. Moreover, the feld of controlled drug delivery offers a wide
scope and future prospects to build up a technology that has high performance, is
economically viable is and potentially effcient.
References
S.V. Vinogradov, T.K. Bronich and A.V. Kabanov, 1. Advanced Drug Delivery
Reviews, 2002, 54, 135.
G.U. Ostrovidova, A.V. Makeev and M.M. Shamtsian, 2. Materials Science and
Engineering: C, 2003, 23, 545.
M.T.Razzak,D.DarwisandS.Zainuddin, 3. Radiation Physics and Chemistry,
2001, 62, 107.
N.A.Peppas,Y.Huang,M.Torres-Lugo,J.H.WardandJ.Zhang, 4. Annual
Review of Biomedical Engineering, 2000, 2, 9.
L.Chen,Z.TianandY.Du, 5. Biomaterials, 2004, 25, 3725.
Q.Li,J.Wang,S.Shahani,D.D.N.Sun,B.Sharma,J.H.ElisseeffandK.W. 6.
Leong,Biomaterials, 2006, 27, 1027.
K.A. Davis and K.S. Anseth, 7. Critical Reviews in Therapeutic Drug Carrier
Systems, 2002, 19, 385.
B. Jeong and A. Gutowska, 8. Trends in Biotechnology, 2002, 20, 305.
K.N.Plunkett,K.L.BerkowskiandJ.S.Moore, 9. Biomacromolecules, 2005, 6,
632.
N.Bhattarai,H.R.Ramay,J.Gunn,F.A.MatsenandM.Zhang, 10. Journal of
Controlled Release, 2005, 103, 609.
56
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
C. de las Heras Alarcn, S. Pennadam and C. Alexander, 11. Chemical Society
Reviews, 2005, 34, 276.
C.M.Schilli,M.Zhang,E.Rizzardo,S.H.Thang,Y.K.Chong,K.Edwards, 12.
G. Karlsson and A.H.E. Muller, Macromolecules, 2004, 37, 7861.
X-Z. Zhang, Y-Y. Yang, T-S. Chung and K-X. Ma, 13. Langmuir, 2001, 17,
6094.
M. Ebara, M. Yamato, M. Hirose, T. Aoyagi, A. Kikuchi, K. Sakai and T. 14.
Okano, Biomacromolecules, 2003, 4, 344.
D. Schmaljohann, J. Oswald, B. Jorgensen, M. Nitschke, D. Beyerlein and C. 15.
Werner, Biomacromolecules, 2003, 4, 1733.
S-Q.Liu,Y-Y.Yang,X-M.LiuandY-W.Tong, 16. Biomacromolecules, 2003, 4,
1784.
F-J. Xu, E-T. Kang and K-G. Neoh, 17. Biomaterials, 2006, 27, 2787.
J-Y.Wu,S-Q.Liu,P.W-S.HengandY-Y.Yang, 18. Journal of Controlled Release,
2005, 102, 361.
X-Z.Zhang,P.JoLewisandC-C.Chu, 19. Biomaterials, 2005, 26, 3299.
A.S. Hoffman, 20. Journal of Controlled Release, 1987, 6, 297.
C.S. Brazel and N.A. Peppas, 21. Journal of Controlled Release, 1996, 39, 57.
H.vonRecum,T.OkanoandS.WanKim, 22. Journal of Controlled Release,
1998, 55, 121.
I. Kucuk and A. Kuyulu in 23. Proceeding of the 8
th
Polymers for Advanced
Technologies Conference, Budapest, Hungary, 2005, p.1.
T. Shiga, Y. Hirose, A. Okada and T. Kurauchi, 24. Journal of Applied Polymer
Science, 1992, 46, 635.
R.LangerandN.A.Peppas, 25. AIChE Journal, 2003, 49, 2990.
R.Zhang,M.Tang,A.Bowyer,R.EisenthalandJ.Hubble, 26. Biomaterials,
2005, 26, 4677.
P. Uhlmann, N. Houbenov, S. Minko and M. Stamm, 27. e-Polymer, 2005, 75, 1.
57
Introduction
C. Geismann and M. Ulbricht, 28. Macromolecular Chemistry and Physics, 2005,
206, 268.
F.A.Aouada,M.R.Guilherme,G.M.Campese,E.M.Girotto,A.F.Rubiraand 29.
E.C. Muniz, Polymer Testing, 2006, 25, 158.
S.K.SiddhantaandR.Gangopadhyay, 30. Polymer, 2005, 46, 2993.
M.Micic,Y.Zheng,V.Moy,X-H.Zhang,F.M.AndreopoulosandR.M. 31.
Leblanc,Colloids and Surfaces B: Biointerfaces, 2003, 27, 147.
S. Shukla, A.K. Bajpai and J. Bajpai, 32. Macromolecular Research, 2003, 11,
273.
G.W.R.Davidson,IIIandN.A.Peppas, 33. Journal of Controlled Release, 1986,
3, 243.
R.A.Grinsted,L.ClarkandJ.L.Koenig, 34. Macromolecules, 2002, 25, 1235.
R.P.BlakemoreandR.B.Frankel, 35. Scientifc American, 1981, 245, 58.
D. Bahadur and J. Giri, 36. Sadhana, 2003, 28, 639.
O.Mykhaylyk,A.Cherchenko,A.Ilkin,N.Dudchenko,V.Ruditsa,M. 37.
Novoseletz and Y. Zozulya, Journal of Magnetism and Magnetic Materials,
2001, 225, 241.
D.K. Kim, Y. Zhang, J. Kehr, T. Klason, B. Bjelke and M. Muhammed, 38.
Journal of Magnetism and Magnetic Materials, 2001, 225, 256.
A.Jordan,R.Scholz,P.Wust,H.FhlingandF.Roland, 39. Journal of
Magnetism and Magnetic Materials, 1999, 201, 413.
C. Poeckler-Schoeniger, J. Koepke, F. Gueckel, J. Sturm and M. Georgi, 40.
Magnetic Resonance Imaging, 1999, 17, 383.
J. Chatterjee, Y. Haik and C-J. Chen, 41. Journal of Magnetism and Magnetic
Materials, 2001, 225, 21.
Z. Popovic and J. Sjstrand, 42. Vision Research, 2001, 41, 1313.
P.A.Liberti,C.G.RaoandL.W.M.M.Terstappen, 43. Journal of Magnetism and
Magnetic Materials, 2001, 225, 301.
58
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
M. Shinkai, M. Yanase, M. Suzuki, H. Hiroyuki, T. Wakabayashi, J. Yoshida 44.
and T. Kobayashi, Journal of Magnetism and Magnetic Materials, 1999, 194,
176.
S. Golui, D. Datta and D. Bohadur in the 45. Proceedings of the 8
th
International
Conference on Ferrites, Kyoto, Japan, 2009.
J.Folkman,D.M.Long,Jr.,andR.Rosenbaum, 46. Science, 1966, 154, 148.
J.FolkmanandD.M.Long, 47. Journal of Surgical Research, 1964, 4, 139.
T.W.LeeandJ.R.Robinson, 48. Remington: The Science and Practice of
Pharmacy, 21
st
Edition,LippincottWilliams&Wilkins,Philadelphia,PA,
USA, 2000.
C.L.BayerandN.A.Peppas, 49. Journal of Controlled Release, 2008, 132, 216.
R.Sykes, 50. International Journal of Antimicrobial Agents, 2000, 14, 1.
N.Dinauer,S.Balthasar,C.Weber,J.Kreuter,K.LangerandH.vonBriesen, 51.
Biomaterials, 2005, 26, 5898.
I. Brigger, C. Dubernet and P. Couvreur, 52. Advanced Drug Delivery Reviews,
2002, 54, 631.
J-C.Leroux,E.Doelker,R.Gurny,R.Cozens,J.L.Roesel,B.GalliandF. 53.
Kubel, Journal of Pharmaceutical Sciences, 1995, 84, 1387.
P.R.Lockman,R.J.Mumper,M.A.KhanandD.D.Allen, 54. Drug Development
and Industrial Pharmacy, 2002, 28, 1.
C. Damge, M. Aprahamian, G. Balboni, A. Hoeltzel, V. Andrieu and J.P. 55.
Devissaguet, International Journal of Pharmaceutics, 1987, 36, 121.
L.Illum,P.D.E.Jones,J.Kreuter,R.W.BaldwinandS.S.Davis, 56. International
Journal of Pharmaceutics, 1983, 17, 65.
S.S.Davis,L.Ilium,S.M.Moghimi,M.C.Davies,C.J.H.Porter,I.S.Muir,A. 57.
Brindley, N.M. Christy, M.E. Norman, P. Williams and S.E. Dunn, Journal of
Controlled Release, 1993, 24, 157.
J. Kreuter, 58. European Journal of Drug Metabolism and Pharmacokinetics,
1994, 3, 253.
S.M. Moghimi, 59. Advanced Drug Delivery Reviews, 1995, 17, 1.
59
Introduction
S. Sershen and J. West, 60. Advanced Drug Delivery Reviews, 2002, 54, 1225.
A. Kikuchi and T. Okano, 61. Advanced Drug Delivery Reviews, 2002, 54, 53.
A.B. Fleming and W.M. Saltzman, 62. Clinical Pharmacokinetics, 2002, 41, 403.
P. Boudes, 63. Controlled Clinical Trials, 1998, 19, 257.
J.R.LewisandM.Ferrarriin 64. Lab-on-a-Chip: Miniaturized Systems for (Bio)
Chemical Analysis and Synthesis, Eds., R.E. Oosterbroek and A. van den
Berg, Elsevier BV, New York, NY, USA, 2003.
S.L.TaoandT.A.Desai, 65. Advanced Drug Delivery Reviews, 2003, 55, 315.
R.K.Stoelting, 66. in Handbook of Pharmacology and Physiology in Anaesthetic
Practice, Eds., R.K. Stoelting and S. Hillier,LippincottWilliams&Wilkins,
Philadelphia, PA, USA, 1999, p.3.
D.D. Breimer, 67. Journal of Controlled Release, 1999, 62, 3.
E.A.Klausner,S.Eyal,E.Lavy,M.FriedmanandA.Hoffman, 68. Journal of
Controlled Release, 2003, 88, 117.
D.L.Wise, 69. Biopolymeric Controlled Release Systems,CRCPublishers,Boca
Raton,FL,USA,1984.
S. Yolles and M.F. Sartori in 70. Drug Delivery Systems: Characteristics and
Biomedical Applications,Ed.,R.L.Juliano,OxfordUniversityPress,Oxford,
UK, 1980, p.84.
L.Brannon-Peppas 71. Medical Plastics and Biomaterials, 1997.
D.B. Mirth, 72. Pharmacology and Therapeutics in Dentistry, 1980, 5, 59.
T. Higuchi, 73. Journal of Pharmaceutical Sciences, 1961, 50, 874.
I.M.BrookandR.vanNoort, 74. British Dental Journal, 1984, 157, 11.
R.A.SiegelandR.Langer, 75. Pharmaceutical Research, 1984, 1, 2.
R.Langer, 76. Science, 1990, 249, 1527.
Y.Lemmouchi,E.Schacht,P.Kageruka,R.DeDeken,B.Diarra,O.Dialland 77.
S. Geerts, Biomaterials, 1998, 19, 1827.
60
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
S.S. Davis, 78. Trends in Biotechnology, 1997, 15, 217.
V.P.Torchilin,V.S.Trubetskoy,A.M.Milshteyn,J.Canillo,G.L.Wolf,M.I. 79.
Papisov, A.A. Bogdanov, J. Narula, B.A. Khaw and V.G. Omelyanenko,
Journal of Controlled Release, 1994, 28, 45.
J. Siepmann and F. Siepmann, 80. International Journal of Pharmaceuticals,
2008, 364, 328.
T. Higuchi, 81. Journal of Society of Cosmetic Chemists, 1960, 11, 85.
J. Siepmann and N.A. Peppas, 82. Pharmaceutical Research, 2000, 17, 1290.
C-C.LinandA.T.Metters, 83. Advanced Drug Delivery Reviews, 2006, 58,
1379.
D.Y.Arifn,L.Y.LeeandC-H.Wang, 84. Advanced Drug Delivery Reviews,
2006, 58, 1274.
G. Frenning, . Tunn and G. Alderborn, 85. Journal of Controlled Release,
2003, 92, 113.
C.Raman,C.Berkland,K.KimandD.W.Pack, 86. Journal of Controlled
Release, 2005, 103, 149.
M.J. Abdekhodaie and X.Y. Wu, 87. Biomaterials, 2008, 29, 1654.
S.Chirico,A.Dalmoro,G.Lamberti,G.RussoandG.Titomanlio, 88. Journal of
Controlled Release, 2007, 122, 181.
N.Faisant,J.Siepmann,J.RichardandJ.P.Benoit, 89. European Journal of
Pharmaceutics and Biopharmaceutics, 2003, 56, 271.
M. Grassi, D. Voinovich, E. Franceschinis, B. Perissutti and J. Filipovic-Grcic, 90.
Journal of Controlled Release, 2003, 92, 275.
M.Geldof,J.I.Freijer,L.A.Peletier,L.vanBeijsterveldtandM.Danhof, 91.
European Journal of Pharmaceutical Sciences, 2008, 33, 217.
N.A. Peppas, 92. Pharmaceutica Acta Helvetiae, 1985, 60, 110.
T. Higuchi, 93. Journal of Pharmaceutical Sciences, 1961, 50, 874.
P.L.RitgerandN.A.Peppas, 94. Journal of Controlled Release, 1987, 5, 23.
61
Introduction
H.B. Hopfenberg, 95. Controlled Release Polymeric Formulations, 1976, 33, 26.
D.O. Cooney, 96. AIChE Journal, 1972, 18, 446.
J. Takahara, K. Takayama and T. Nagai, 97. Journal of Controlled Release, 1997,
49, 11.
Y.Chen,T.W.McCall,A.R.BaichwalandM.C.Meyer, 98. Journal of Controlled
Release, 1999, 59, 33.
K. Takayama, M. Fujikawa and T. Nagai, 99. Pharmaceutical Research, 1999,
16, 1.
T.Wu,W.Pan,J.ChenandR.Zhang, 100. Drug Development and Industrial
Pharmacy, 2000, 26, 211.
S.Ibric,M.Jovanovic,Z.Djuric,J.ParojcicandL.Solomun, 101. Journal of
Controlled Release, 2002, 82, 213.
A.Ghaffari,H.Abdollahi,M.R.Khoshayand,I.S.Bozchalooi,A.Dadgarand 102.
M.Rafee-Tehrani,International Journal of Pharmaceutics, 2006, 327, 126.
Q.Shao,R.C.RoweandP.York, 103. European Journal of Pharmaceutical
Sciences, 2007, 31, 137.
G. S. Park in 104. Synthetic Membranes: Science, Engineering and Applications,
Eds.,P.M.Bungay,H.K.LonsdaleandM.N.dePinho,D.ReidelPublishing
Company, Dordrecht, The Netherlands, 1986, p.57.
T.T.Wang,T.K.KweiandH.L.Frisch, 105. Journal of Polymer Science, Part B:
Polymer Physics Edition, 1969, 7, 2019.
Diffusion-controlledSystemsinControlledReleaseofBiologicallyActive 106.
Agents,Ed.,R.Baker,JohnWiley&Sons,NewYork,NY,USA,1987,p.39.
N.Bertrand,G.LeclairandP.Hildgen, 107. International Journal of
Pharmaceutics, 2007, 343, 196.
G. Frenning and M. Strmme, 108. International Journal of Pharmaceutics, 2003,
250, 137.
M.Marucci,G.Ragnarsson,U.NymanandA.Axelsson, 109. Journal of
Controlled Release, 2008, 127, 31.
62
Stimuli Responsive Drug Delivery Systems: From Introduction to Application
J. Crank, 110. The Mathematics of Diffusion, Oxford University Press, Oxford,
UK, 1975.
M.Hombreiro-Prez,J.Siepmann,C.Zinutti,A.Lamprecht,N.Ubrich,M. 111.
Hoffman,R.BodmeierandP.Maincent,Journal of Controlled Release, 2003,
88, 413.
J-M. Vergnaud, 112. Controlled Drug Release of Oral Dosage Forms Forms, Ellis
Horwood Ltd., Chichester, UK, 1993.
M.Grassi,L.Zema,M.E.Sangalli,A.Maroni,F.GiordanoandA. 113.
Gazzaniga, International Journal of Pharmaceutics, 2004, 276, 107.
S. Herrmann, G. Winter, S. Mohl, F. Siepmann and J. Siepmann, 114. Journal of
Controlled Release, 2007, 118, 161.
F. Siepmann, S. Herrmann, G. Winter and J. Siepmann, 115. Journal of Controlled
Release, 2008, 128, 233.
T. Higuchi, 116. Journal of Pharmaceutical Sciences, 1963, 52, 1145.
S.J. Desai, A.P. Simonelli and W.I. Higuchi, 117. Journal of Pharmaceutical
Sciences, 1965, 54, 1459.
L.T.FanandS.K.Singh, 118. Controlled Release: A Quantitative Treatment,
Polymers, Properties and Applications Series No.13, Springer-Verlag, Berlin,
Germany, 1989.
P. Colombo, 119. Advanced Drug Delivery Reviews, 1993, 11, 37.
N.A. Peppas and J.J. Sahlin, 120. International Journal of Pharmaceutics, 1989,
57, 169.
R.W.Korsmeyer,S.R.LustigandN.A.Peppas, 121. Journal of Polymer Science,
Part B: Polymer Physics Edition, 1986, 24, 395.
H. Fujita, 122. Fortschritte der Hochpolymeren-Forschung, 1961, 3, 1.
R.T.C.Ju,P.R.NixonandM.V.Patel, 123. Journal of Pharmaceutical Sciences,
1997, 86, 1293.
A.Streubel,J.Siepmann,N.A.PeppasandR.Bodmeier, 124. Journal of Controlled
Release, 2000, 69, 455.
63
Introduction
B. Narasimhan and N.A. Peppas, 125. Journal of Polymer Science, Part B: Polymer
Physics Edition, 1996, 34, 947.
B. Narasimhan and N.A. Peppas, 126. Macromolecules, 1996, 29, 3283.
R.T.C.Ju,PhillipR.NixonandM.V.Patel, 127. Journal of Pharmaceutical
Sciences, 1995, 84, 1455.
A. Goepferich, 128. European Journal of Pharmaceutics and Biopharmaceutics,
1996, 42, 1.
R.LangerandN.Peppas, 129. Journal of Macromolecular Science - Reviews in
Macromolecular Chemistry and Physics, 1983, C23, 61.
F.V. Burkersroda and A.M. Goepferich, 130. Polymeric Materials - Drugs, Delivery
and Devices, MRS Proceedings, 1999, 50, 17, Paper GG1.2.
P.I.Lee, 131. Journal of Membrane Science, 1980, 7, 255.
K. Zygourakis and P.A. Markenscoff, 132. Biomaterials, 1996, 17, 125.
A.GpferichandR.Langer, 133. Journal of Controlled Release, 1995, 33, 55.
A. Gpferich, 134. Journal of Controlled Release, 1997, 44, 271.
N. Faisant, J. Akiki, F. Siepmann, J.P. Benoit and J. Siepmann, 135. International
Journal of Pharmaceutics, 2006, 314, 189.
W.Gombotz,M.HealyandL.Brown,inventors;Enzytech,Inc.,assignee;US 136.
5,019,400, 1991.
A. Zaffaroni, inventor; Alza Corporation, assignee; US 3,598,122, 1971. 137.
64
Stimuli Responsive Drug Delivery Systems: From Introduction to Application