�������� ����� ��

Compression profiles of different molecular weight chitosans

Faisal Alakayleh, Iyad Rashid, Mahmoud Al-Omari, Khaldoun Al-Sou¡!–[INS][’]–¿’¡!–[/INS]–¿od, Babur Z. Chowdhry, Adnan A. Badwan

PII: S0032-5910(16)30253-4DOI: doi: 10.1016/j.powtec.2016.05.019Reference: PTEC 11659

To appear in: Powder Technology

Received date: 16 November 2015Revised date: 7 May 2016Accepted date: 14 May 2016

Please cite this article as: Faisal Alakayleh, Iyad Rashid, Mahmoud Al-Omari, Khal-doun Al-Sou¡!–[INS][’]–¿’¡!–[/INS]–¿od, Babur Z. Chowdhry, Adnan A. Badwan, Com-pression profiles of different molecular weight chitosans, Powder Technology (2016), doi:10.1016/j.powtec.2016.05.019

This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

1

Compression profiles of different molecular weight

chitosans

Faisal Alakayleh1, Iyad Rashid

2, Mahmoud Al-Omari

2, Khaldoun Al-Sou’od

3, Babur Z.

Chowdhry4, Adnan A. Badwan

2*

1 Department of Pharmaceutics and Pharmaceutical Technology, College of Pharmacy,

Petra University, Amman, Jordan

2 The Jordanian Pharmaceutical Manufacturing Co., Naor, Jordan

3 Department of Chemistry, Al-Albait University, Irbid, Jordan

4 Faculty of Engineering & Science, University of Greenwich, Medway Campus,

Chatham Maritime, Kent, UK

*Corresponding author: Dr Adnan Badwan, e-mail: jpm@go.com.jo

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

2

Abstract

The compression behavior of hydrophilic chitosan has been investigated as a function of

molecular weight (MW), ranging from 8-100 kDa. Powder packing, plasticity, elasticity,

brittle-fracture nature were evaluated using force displacement curves and

Heckel/Kawakita analysis upon compression of different MW chitosan powders. The

compacts produced were subjected to ejection work determination and tensile strengths

analysis, the later using Leuenberger equation. Molecular modeling was used to correlate

the MW-binding energy dependence on the physical strength of the chitosan compacts.

The high work of compression of the low MW (≤30 kDa) chitosans is accounted for their

brittle nature upon compression. At high compression forces (> 3kN), the low MW

chitosan compacts exhibited slightly lower tensile strength due to their increasing extents

of elastic recovery and ejection work. Lower particle size of chitosan enhances lower

fragmentation, packing, and frictional tendency that can be related to an increase in

plastic deformation and compacts tensile strength. In molecular modeling, the calculated

higher binding energies of the high MW chitosan (88 kDa) than that of the low MW

chitosan (8 kDa) may further explain its rigid, complex nature when compared to low

MW chitosan. Upon compression, increasing MW of chitosan is manifested as higher

energies for densification which in turn determines the extent of helical packing from

intermingled molecular to open structure as the MW is decreased. Such changes in

structural configuration are suggested to contribute to the transfer from plastic to brittle-

fracture nature with high packing extent of chitosan upon compression when the MW is

reduced. The foregoing may justify the slow diclofenac Na drug release upon dissolution

testing when low MW chitosan is used in tablet preparation.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

3

Keywords: chitosan, molecular weight, compression force, molecular modeling

1. INTRODUCTION

Chitosan powders have been the subject of various investigations as matrices for

immediate/controlled release solid dosage forms [1-9]. Since chitosan is prepared from

chitin [10,11], the final molecular weight (MW) of chitosan depends on the source or

initial MW of chitin and the deacetylation time and temperature [12]. However, Hwang

KT et al [13] demonstrated the complexity of variables, e.g. temperature and time,

involved to control the depolymerization and/or the deacetylation of chitosan. Many of

these variables were found to manipulate the effective surface charge, electrokinetic

mobility, and colloidal stability of chitosan especially upon interfacial interaction inside

an aqueous environment [14,15]. Unfortunately, the literature itself lacks sufficient

handling of the physicochemical investigations as a function of chitosan MW.

Investigations were mostly concerned with applications with few justifications, if present,

on the role of chitosan MW at the molecular and morphological level [16-25]. These

justifications were confined to structure-MW dependence and supported by the fact that

the MW is equivalent to the number of repeating units of D-glucosamine. Onto each of

these units, interactive amine groups (NH2) are bonded. The general observation

conclusively illustrates that the longer the chain length of chitosan, the more NH2 binding

sites are available. Consequently chitosan binding affinity, uptake capacity, rheological

properties, particle morphology and subsequently surface area, are all strongly dependent

on the number of NH2 groups in a specific polymer mixture, i.e. average MW.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

4

The scientific inputs become even shallower when chitosan is present in the powder state.

In this context, the compression behavior of chitosan powder was not understood to be

MW related [1,5,6,26-27]. In order to correlate such compression dependency, the

particle morphology and density have to be introduced for analysis when the MW of

chitosan is varied. These two parameters play a significant role in the packing and

arrangement upon powder compression [28]. Fortunately, some attention has been given

to try to understand the complex relationship between the MW of chitosan and the

structural changes and consequently the powder physical properties, all as a function of

chain length (or MW). For example, Goycoolea et al pointed out that chitosan of longer

chain lengths tends to form ‘loops’ giving rise to the creation of more expanded

structures and to a denser packing [7]. Ofori-Kwakye et al related the change in the

intrinsic viscosity of different MW chitosans to the change in particle shape and size [29].

Aggarwal et al highlighted the dependence of chitosan molecular architecture and MW

on the mechanical strength of branched chitosans synthesized by grafting low MW

chitosan chains [30]. Furthermore, bulk density and flow of powders were found to be

dependent on the MW of chitosan when the later is spray-dried with lactose [31]. In this

regard, it was noted that the lower the MW of chitosan the higher the bulk density and

better flow of their processed mixture. However, the poor powder flow and thereby

compressibility have never been investigated as a function of chitosan MW.

Consequently, an insight into the behavior of different MW chitosan powders upon

compression is of significant interest.

In light of the foregoing remarks which are important parameters in characterizing

powder compression properties for direct compression pharmaceutical applications, the

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

5

work reported herein addresses the mechanistic densification behavior of depolymerized

chitosan powders. The properties of the compressed tablets were assessed in the context

of testing tablet mechanical properties as a function of MW.

2. MATERIALS AND METHODS

2.1. Materials

Chitosan (Hong Ju Ginseng Co., Ltd., Dalian Liaoning, China), particle size <150 μm,

MW 100 kDa, (93% degree of deacetylation) of 100 kDa MW was supplied as

pharmaceutical grade. The 8, 13, 18, and 30 MW chitosan powders were prepared by the

Jordanian Pharmaceutical Manufacturing Company (JPM, Naor, Jordan) using the acid

hydrolysis of the 100 kDa chitosan. This technique was described in details by Qandil et

al [32], who illustrated the de-polymerization of chitosan by acid hydrolysis, the

determination of viscosity average molecular weight using the relative and reduced

viscosities of the Mark-Houwink technique, and finally the determination of the degree of

deacetylation which was approximately 100% for all chitosan samples according to the

first derivative UV spectroscopic method [33]. Diclofenac sodium was supplied by

CHEM. Co., Ltd (CCS. Xiaoshan Hangzhou, China).

2.2. METHODS

2.2.1. Compression of chitosan samples

Prior to compression the chitosan samples were grinded and sieved using an oscillating

granulator with a mesh size 250 µm and collected on a 90 µm mesh. 100 mg of each MW

(8, 13, 18, 30, 100 kDa) chitosan sample was compressed using a Gamlen Tablet Press

(Gamlen Tableting Ltd. Biocity Nottingham, UK). Initially, the powders were

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

6

compressed at a punch size of 6.0 mm diameter, enough to accommodate the chitosan

samples. The die was manually filled and the machine operated under an automatic

compression force or displacement control. Ejection was carried out in the same direction

as compression i.e., the tablets were pushed through the bottom of the die into a tablet

holder which was removed from the machine. The test speed was set at 60 mm/min. The

applied loads used were; 100, 200, 300, 400, and 500 kg equivalent to 098-4.9 kN

compression force or to 34.6-173.2 MPa compression pressure (compression force/punch

surface area) applied on a 6 mm diameter circular punch. The Gamlen Tablet Press

(GTP) software displayed a full data capture of the work of compression and work elastic

recovery on a PC screen. After ejection the thickness of the tablets were immediately

measured using a caliber. Other tablets produced in the same manner were directly

subjected to tensile strength tests using a hardness tester (Copley, Nottm Ltd.,

Therwil,Switzerland). The average tensile strength of 10 tablets was recorded at each

compression force for each MW chitosan. From the measured tablet hardness, the tensile

strength was calculated according to the equation [34]:

σ = 2F/(πDh) (1)

where σ is the tensile strength (Pa), F is the tablet hardness (N), D the tablet diameter

(m), and h the tablet thickness (m).

2.2.2. Work of compression, elastic recovery and tablet ejection, and extent of

powder packing

A typical illustration of the force displacement curve is presented in Fig. 1 for the 100

kDa chitosan (100 mg) whereby the input load was set at 100 kg (4.91 kN). The upper

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

7

punch, at a distance of 27.379 cm from the base, displays an automatic displacement

starting from the zero position (point A), followed by point B where the upper punch

makes the first contact with the powder. The powder then undergoes compression

whereby the recorded curve represents an increase in the amount of force as a function of

the upper punch displacement. The maximum recorded force terminates the punch

movement in the downward direction and is presented by point C of the displacement

value D. However, the powder’s elasticity exerts an upward force to the punch in the

form of a decreasing force profile until it reaches zero at point E.

From Fig. 1, the area under the curve BCD is the work of compression in J (N.m). On the

other hand, the area under the curve ECD represents the work of elastic recovery. In both

cases, the data are tabulated by the software of the GTP into two columns, one for the

position of the upper punch (mm) and the other for the load applied (kg). The total area

was numerically calculated by approximation using areas of rectangles. The width and

height of such rectangles were taken as the difference between two successive

displacement points and their corresponding ‘load value’ length respectively. The

summation of all rectangles areas represents the area under the curve (AUC).

The packing extent can be estimated from the final volume of the powder after being

compressed compared to its initial bed volume before compression. Empirically, the

packing extent can be estimated from Fig. 1 using the following equation;

Packing extent = %100B27.379

BD

(2)

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

8

where (D-B) and (27.329-B) represent the final (after compression) and initial (before

compression) heights of the powder bed respectively, mm.

Generally, the die-wall friction can be usually estimated by either of the following

methods; by calculating the difference in the work of compression of unlubricated and

lubricated granules [45], or by calculating the difference in the work of compression for

the upper punch forces and that for the lower punch [46]. However, both methods were

not adopted in this work as lubricants, in the first case, may disrupt the arrangement of

particles through agglomeration [47]. The extent of such disruption may vary between

different MW of chitosans. In the second case, the GTP does not provide a lower punch

for compression, it can, however, measure the work for tablet ejection by the upper punch

from within the die. A typical force-displacement profile for the ejection of the 100 kDa

chitosan tablets compressed at the input load of 500 kg (4.9 kN) is presented in Fig. 2.

The AUC, which represents the work for tablet ejection, was calculated for all force-

displacement curves for each MW of chitosan and then plotted at each compression force.

Accordingly, after the powder was being compressed into a tablet, the GTP setting was

turned into the ejection mode using the same upper punch and inverting the closed die

base position into an open position.

2.2.3. Analysis of compression data

Heckel, Kawakita, and Leuenberger analysis were used to characterize the compression

properties of different MW chitosan samples. The Heckel equation (Eq. 3) describes the

change of powder porosity as a function of the applied pressure upon powder

densification [38,39].

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

9

AKPD

)1(

1ln (3)

where D (compact density/true density) is the relative density of a powder compact at

pressure P (MPa), whereby compact density is the tablet weight (100 mg)/volume of 6

mm diameter tablet. The true densities, measured using a helium pycnometer

(Ultrapycnometer 1000 v. 2.2, Quatachrome Co., Odelzhausen, Germany) of the 100, 30,

18, 13, and 8 kDa chitosans were 1.351, 1.549, 1.592, 1.758, and 1.853 g/cm3

respectively. K (MPa-1

) is the slope and it is a measure of the plasticity of a material. The

inverse of the slope is the yield pressure (Py) of the materials (MPa) [40], which describes

the tendency of the material to deform either by plastic flow or fragmentation. Py is

inversely related to the ability of the material to deform plastically under pressure. The

constant A is related to the die filling and particle rearrangement before deformation and

bonding of the discrete particles.

For comparison purpose, the compression behavior of the samples can be also evaluated

using Kawakita analysis of powder compression data. Such a model is usually used to

linearize the non-linear pressure-porosity relationship provided by the Heckel model into

a linear pressure-volume reduction relationship. Therefore such a model is dependent

upon the initial powder bulk density instead of the true density of the powder, as in the

case of the Heckel model. The Kawakita equation (Eq. 4) is used to study powder

compression via the degree of volume reduction, C (Eq. 5) under an applied pressure, P

(MPa).

aba

P

C

P 1 (4)

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

10

(5)

where V0 is the initial bulk volume and V is the volume at pressure of P. a and b are

constants which are obtained from the slope and intercept of the P/C versus P plots.

The constant a is the initial porosity of the material before compression, while the

constant b relates to the degree of plasticity of the material. The reciprocal of b or Pk

defines the pressure required to reduce the powder bed by 50% [41,41].

In addition to the previous particle size of chitosan adopted for compression analysis, i.e.

between 250 µm and 90 µm, all of the aforementioned compression tests were repeated

for the 100 kDa and the 8 kDa chitosans at two additional particle sizes. The first particle

size represents the coarse grade whereby the chitosan powder was passed over a mesh

size 500 µm and collected on a 250 µm mesh. The second particle size represents the fine

particles that pass through the 90 µm mesh.

The last compression analysis carried out in this work is the Leuenberger analysis. In this

model, the correlation of the tablets tensile strength with the compression pressure is

found to be further dependant on the tablets density. Blattner and Leuenberger analysis

describe such correlation whereby there is a non linear relationship between the tensile

strength (σ) and the product of applied pressure (P) with the tablets relative density (ρR)

as Eq. 6 implies [43].

)(1σσ maxRP

e

(6)

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

11

where σmax is the maximum tensile strength when P is infinite and ρR (the relative density

of the tablet) = 1, γ is the compression susceptibility which represents the decrease in

porosity under pressure. γ is related to the how fast (high γ ) or how slow (low γ) is the

transition state between a powder and tablet when compacts are formed under

compression pressure [44]. The obtained data were analyzed by using GraphPad Prism

(GraphPad Software Inc., La Jolla, CA).

2.2.4. Molecular modeling

Computations were performed with Hyperchem® (6.03 professional, Hyperchem Inc.,

Waterloo, Canada), using the Amber Force field implemented in Hyperchem. All

molecular calculations were carried out using MM+ force fields (bond dipoles). The

partial atomic charges were calculated using AM1 semi-empirical methods. Energy

minimizations were performed using the Polak-Ribiere algorithm (0.01 and 0.1

kcal/mol.Å for D-glucosamine and chitosan, respectively) in order to obtain the stable

conformations of chitosan. For simplicity, chitosan polymers were built up from amine

groups (100% deacetylated). Chitosan monomer (D-glucosamine) was also built up from

natural bond angles, as defined in the Hyperchem software and then optimized at the HF-

ab initio level uisng the 3-21G* basis set (structures I). A set of named selections has

been made to define the HEAD and TAIL for building the polymer. An additional set of

named selections was made to define the atoms involved in the torsion angles used to

assemble the monomers. The HEAD is assumed to be connected to x which in turn is

connected to x´. The TAIL is assumed to be connected to y which in turn is connected to

y´. The torsional angles connecting the monomers are therefore x´–x–y–y´, with retention

of the internal structure of the monomer. The torsion angle used to bring the monomers

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

12

together may be specified to have a random value rather than the constant value described

earlier. The chitosan polymer was constructed with a suitable length from the monomer

unit [45].

2.2.5. Dissolution of diclofenac sodium

Three preparations of diclofenac sodium tablets were formulated so that each tablet (150

mg) contained 50 mg diclofenac sodium and 100 mg of the 100, or 8, or the 30 kDa

chitosan powders. The foregoing MWs represent the insoluble, the lowest MW soluble

and the highest MW soluble chitosans respectively. All components were physically

mixed for each preparation and then directly compressed at pressures of 30-35 kN using a

single punch tabletting machine (Manesty, Merseyside, UK) in which a 10-mm flat

circular punch was fitted. Compression was carried out at different forces (30-35 kN) to

maintain similar tablet tensile strength at 3.0 (±0.1) MPa for the purpose of having a fair

comparison in drug release data. The tablets had an average tablet density of 6.25 (±

0.01) g/cm3. For drug dissolution studies, eight tablets were used according to USP

specifications (USP30-NF25, 2007) for the dissolution media and apparatus.

3. RESULTS

3.1. Extent of powder packing and compression, elastic recovery, and tablet ejection

works

For each MW, the extent of powder packing, the work of compression, elastic recovery,

and tablet ejection were calculated for each chitosan powder from the force-displacement

curves measured using the Gamlen tablet press. Figures 3, 4, 5, and 6 illustrate the plots

of powder packing extent and the works for powder compression, elastic recovery, and

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

13

tablet ejection, respectively as a function of the five applied forces; 1-5 kN which are

equivalent to the GTP compression loads from 100-500 kg on100 mg of the chitosan

samples.

In Figures 3 and 4 the increase in powder packing extent and compression work as the

compression force increases is clearly evident for all the different MW chitosan samples

(8 kDa-100 kDa). Moreover, decreasing the MW of chitosan from 100 kDa to 8 kDa

shows an increase in both the powder packing extent (at compression forces > 2.94 kN)

and the work for compression.

The work for elastic recovery calculated from the force-displacement curves is illustrated

in Fig. 5 for the same chitosan samples. It is clear from the data in Fig. 5 that increasing

the compression force for all samples results in an increase in the elastic recovery work.

Such behavior manifests the high amount of elastic energy stored at higher compression

forces which is then recovered upon termination of the applied force. In addition, the data

in Fig. 5 shows that all of the low MW chitosans (8, 13, 30 kDa) show lower elastic

recovery work than the 100 kDa chitosan for all compression forces below 2.94 kN.

Higher than 2.94 kN, the elastic work of all the low MW chitosans approach that of the

100 kDa. The foregoing suggests that although higher compression work is displayed for

chitosan samples of lower MW at all compression forces (Fig. 4), there appear to be

force-related factors that contribute to the observed elastic recovery trend when compared

with 100 kDa chitosan.

Since tablet ejection depends on the frictional forces between the tablet and the die, the

work of tablet ejection measured using the GTP was used to indicate variations with

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

14

compression forces and chitosan MW. Fig. 6 illustrates the foregoing criteria. In this

regard, with the exception of the 100 kDa chitosan, the work increases with increasing

the compression force and decreasing the MW. In contrast, for the 100 kDa chitosan,

ejection work decreases with increasing the compression force.

3.2. Heckel, Kawakita, and Leuenberger compression analysis

The properties and behavior, e.g. plastic/elastic deformation, brittle fracture, and extent of

particles arrangement, of powder compacts upon compression may justify the changes

obtained in the work for compression and elastic recovery when increasing the applied

force on one hand and decreasing the MW of chitosan on the other. Heckel and Kawakita

analysis of the compression data are considered herein in an attempt to evaluate the

behavior of chitosan powders .when subjected to different compression forces.

Fig. 7 illustrates the Heckel plots of the samples whereby non-linear trends are obtained

over all the compression pressure data range (i.e. 34.6-173.2 MPa). It is clear from the

Figure that powder porosity decreases when the compression pressure increases and

when the MW of chitosan deceases. The lowest porosity is thus obtained for chitosan of 8

kDa MW. From the same Figure, linear portions can be drawn through three successive

points (porosity-pressure points). For example, for all the samples, except for chitosan of

8 kDa, the first three points were considered whereas the 3rd

, 4th

, and the 5th

points were

considered for chitosan of 8 kDa. The slopes and the intercepts of the straight line

portions are tabulated in Table 1.

Kawakita analysis was carried out using the same compression data obtained by the GTP.

In this model, bulk densities were considered for calculations of the powder volume

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

15

reduction upon compression instead of the true densities used in Heckel analysis when

calculating porosity changes upon compression. Kawakita parameters; a and 1/b (PK)

were calculated from the slopes and intercepts of Kawakita equation (Eq. 4). Results are

tabulated in Table 2.

From the compression analysis using Leuenberger equation, the radial tensile strength at

each compression force, calculated using Eq.1, was plotted against the product of the

pressure of compression and the relative density of the different MW chitosans as Figure

8 illustrates. By choosing a logarithmic regression for curve fitting of the data, the

unknown parameters representing the compression susceptibility and the maximum

tensile strength can be displayed. Values of these parameters are presented in Table 3.

The first two high (100 and 30 kDa) and the lowest (8 kDa) MW chitosans had low

compression susceptibility, whilst values of the 18 and the 13 kDa chitosans were the

highest. Thus the maximum tensile strengths (Table 3) of the 18 and the 13 kDa are

reached faster than all the other MW chitosans. However, the σmax of the 18 amd 13 kDa

chitosans were the lowest indicating poor bonding at higher compression pressures.

The tensile strength of chitosan compacts are rearranged in Table 4 at each compression

force for the whole MW range of the samples. Generally, increasing the compression

force, allows closer packing of the powder bed, thus minimizing the porosity and forming

a more coherent compact. The tighter less porous structure of the compacts, the higher

their tensile strengths when compressed at increasing force, as the data in Table 4

illustrates. In addition, the highest tensile strengths were obtained for the highest MW

chitosan, i.e. 100 kDa at the 4.9 kN compression force. In contrast, the same MW

manifested the lowest compacts tensile strengths when compressed at 0.98 and 1.96 kN

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

16

compression forces. It can be further noticed that, the tensile strengths of the 18 and the

13 kDa chitosans started to form stronger compacts than the other MW chitosans at the

compression pressure of 2.94 MPa. Further increase in the compression force (>2.94

MPa), however, resulted in weaker compacts for the same foregoing MWs of chitosan.

On the other hand, all the other MWs encountered a tensile strength increase with

pressures higher than 2.94 MPa.

3.3. Particle size effect on compression properties

All compression properties were found to be affected by chitosan particle size (Table 5)

using the highest (100 kDa) and lowest (8 kDa) MWs. Although the packing extent

decreases as the particle size is decreased, the work of compression is higher for smaller

particle size chitosans. Such behavior is contradictory to Figures 3 and 4 which indicate

that the recorded high compression work belongs to high chitosan packing when low

MW chitosans are compressed at 4.9 kN. On the other hand, the work of elastic recovery

and tablet ejection are reduced when lower chitosan particle size is used. Heckel analysis

of the three particle size chitosans indicates the decrease in the yield pressure (PY) and

rearrangement values when the particle size is reduced. Finally, high tensile strength of

the compacts is achieved for the compression of low particle size chitosans.

3.4. Molecular modeling

In molecular modeling using Hyperchem, the chemical structure of D-glucosamine

(chitosan monomer) built up from natural default bond angles is shown in Fig. 9(A).

Building chitosan polymers (high MW chitosan and low MW chitosan) were constructed

with a suitable length from the monomer unit. The high MW chitosan (Fig. 9(B)) was

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

17

constructed using 365 monomer units (MW is 66 kDa), while the low MW (Fig. 9(C))

chitosan from 45 D-glucosamine units (MW is 8 kDa). Accordingly, both high and low

MW chitosans exist in two different orientations owing to the chain length. Compared to

the low MW chitosan, the high MW chitosan, Fig. 9(B), shows a longer and more

extended structure (more extensive H-bonding). The interior space interactions within

polymeric species were studied manually by allowing two chitosan polymers of the same

MW to approach each other, and then system optimization was carried out under the

conditions described in the experimental section.

It is noteworthy that the high MW chitosan polymer units approach more closely to each

other than the low MW chitosan, allowing more extensive intermolecular H-bonds when

compared to the less extended low MW chitosan polymers (Figures 10(A) and (B)). The

calculated binding energy per monomer unit, Ecal, is determined from the following

equation:

Ecal = {Esystem –(Ep1 + Ep2)}/n (7)

Where Esystem is the energy of the optimized system, Ep1 and Ep2 are the energies of

the isolated gas phase polymers in similar conformations, with N being the number of D-

glucosamine units in the polymer.

Dissolution data for diclofenac sodium tablets containing different MW chitosan powders

is presented in Fig. 11. The 100 kDa chitosan tablets showed fast drug release whereas

slow drug release was the case for the 30 kDa and the 13 kDa chitosan tablets. The extent

of powder packing was tested for the three chitosan MWs used in the diclofenac product

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

18

preparation (Fig. 12). There is clearly an increasing extent of packing the higher the MW

of chitosan.

4. DISCUSSION

The increase in compression work as the compression force increases is clearly evident

for all the chitosan samples. Generally, upon powder compression, increasing the applied

force by the GTP allows either more folding for plastically deforming materials or high

extent of particle fragmentation for brittle materials. Thus, in the course of compression

by the upper punch there will either be a change in the fracture behavior of the materials,

or extra displacement to be undertaken when achieving a complete input load, or the

establishment of both of the foregoing changes. In the current investigation, the extent of

powder packing using force-displacement curve has been used to give measurable

estimates of the amount of these changes taking place as a function of compression force

and chitosan MW. In this regard, the effect of compression force on the powder packing

was found to be more significant than chitosan MW whereby, as expected, higher

compression forces induce extensive packing (Fig. 3). Examining the effect of chitosan

MW, the extent of powder packing was found to increase with increasing the MW and

decrease with increasing the MW for the lowest and highest tested compression forces

respectively. In other words, decreasing the chitosan MW implies more packing extent

attained with increasing the compression force. Such finding explains the obtained data in

Fig. 4 whereby there is more displayed work of compression when lower MW chitosans

are used. Since the work required for compression is the sum of work necessary to

rearrange the particles, deform, and finally to fragment them [46], it is suggested that

decreasing the MW of chitosan implies that the compacts undergo excessive folding

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

19

and/or fragmentation at the micro level upon compression. Furthermore, such particles

behavior will have an impact on the resulted profiles of elastic recovery work for each

MW (Fig. 5). In this regard, the profiles of the elastic recovery work of all MW chitosans

except for the 100 kDa are similar, with values lower than the 100 kDa chitosan for

compression forces less than 2.94 kN. Above 2.94 kN, the aforementioned profiles start

to approach that of the 100 kDa chitosan. Such observation is in harmony with the

packing extent as a function of MW (Fig. 3) whereby above 2.94 kN the packing extent is

clearly in increasing order with chitosan MW. This indicates that higher powder packing

induced by higher compression forces manifests higher elastic recovery which increases

with MW. Since elastic recovery is related to the amount of stored energy during

compression [47] and is released during decompression, then the apparent high

packing/elastic recovery extents are thought to be correlated to higher ability of the lower

MW to store more energy than the higher MW chitosans. The foregoing will be given a

further attention in the molecular modeling work. In the meanwhile, compression data

analysis using Heckel and Kawakita equations may assist to elucidate the impact of the

MW of chitosan on powder properties upon compression.

Heckel and Kawakita analysis determine optimum tabletting properties usually obtained

with plastically deforming materials. Generally, plastic materials exhibit greater degree of

folding resulting at higher compression pressures thus allowing the appearance of new

surfaces for binding to take place. With regard to Heckel analysis, the lower the yield

pressure of a material is an indication of higher extent of plastic deformation. Thus the

data in Table 1 shows that the lower the MW of chitosan (e.g. 8 kDa), the lower the

extent of plastic deformation. On the contrary, the 100 kDa chitosan displays the highest

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

20

plastically deforming sample. In fact, there is a tendency for the lower MW chitosan

samples to undergo brittle fracture upon compression. Such a conclusion is supported by

the fact that lower MW chitosan samples showed higher intercept values (A) of the

Heckel analysis implying that there is greater tendency of the lower MW chitosan to

undergo fragmentation and to fill the die.

The results for the Kawakita analysis of the experimental data are shown in Table 2. It is

clear from the data in Table 2 that the highest porosity (highest ‘a’ Kawakita parameter)

was obtained for the highest MW chitosan, i.e. 100 kDa, whereas the lowest porosity

(lowest ‘a’ Kawakita parameter) was obtained for the 8 kDa chitosan. The foregoing is

agreement with the results of Fig. 3 whereby the extent of packing increases, and

subsequently porosity is decreased, when lower MW chitosans are used. On the other

hand On the other hand, the PK values are lower for the higher MW chitosan samples

indicating that they readily deform plastically under pressure. This MW-plasticity

relationship is in agreement with the yield pressure obtained by Heckel analysis of

compression data.

Optimal tabletting properties are generally attained for powders with maximum amount

of plastic deformation and/or with minimum amount of elastic recovery [48,49]. The

highest MW chitosan (100 kDa) showed the highest compacts tensile strength (Table 4)

as its total amount of plastic deformation was the highest revealed by its lowest yield

pressure (Table 1). However, despite their higher work of compression, the low MW

chitosan samples exhibited less plastic behavior and therefore slightly lower tensile

strength of the compacts than the 100 kDa chitosan. This result contradicts the findings

by Picker-Freyer et al who investigated the slope of the Heckel plot as a function of

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

21

different MW chitosans [50]. They found that chitosan with a MW of 87.2 kDa behaves

in a similar way to microcrystalline cellulose as a highly plastically deforming excipient.

The other higher MW chitosans ((173.3 and 210.5 kDa) showed a lower Heckel slope and

thus higher resistance against deformation. They suggested that the higher Heckel slope -

thus a higher and easier deformation- of the 87.2 kDa chitosan are correlated to its low

MW. In re-evaluating such behavior, the aforementioned judgment did not consider

differences in the physical properties of the chitosans used. For example all their

investigated MW chitosans differ in the degree of N-deacetylation, bulk density, and

particle size. All these parameters affect the packing and extent of deformation upon

compression. In addition, the determination of the Heckel slopes includes plastic and

elastic deformation and therefore the contribution of the undetermined elastic recovery

might have added up to the Heckel slope estimation [51-55]. The foregoing is justified in

the current work as the high plastically deforming character of the 100 kDa chitosan

presented the highest work of elastic recovery (Fig. 5). All the other lower MW chitosans

(8, 13, 18, 30 kDa) had lower elastic recovery due to their lower extent of plastic

deforming or to their brittle-fracture character. Such behavior is, however, valid up to a

compression force of 2.94 kN (Fig. 5) which is almost equivalent to the linear portion

range considered for the Heckel plots (Fig. 7). Higher than that, the elastic recovery work

seems to be almost the same for all MW chitosans. It is suggested that, under high

compression forces (>2.94 kN), the low MW chitosans deviate away from brittle-fracture

deformation and start to exhibit plastically deforming character as their elastic recovery

work becomes equivalent to the 100 kDa chitosan. Patel et al highlighted such behavior

for dicalcium phosphate where they noted that as the pressure is increased, more and

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

22

more primary particles fragment to smaller particles, which have higher resistance against

deformation due to the appearance of elastic character [56].

In summary, the recorded increase in the work for compression is due to appearance of

brittle-fracture characteristics. The foregoing implies the appearance of new smaller

granules that require extra work to compress and exhibit less elastic recovery work. The

increased (A) values of the Heckel analysis for chitosan MW<100 kDa implies more

dense packing due to high fragmentation which again justifies the brittle-fracture

behavior upon compression.

Changes in the compacts tensile strength with increasing the compression pressure (Table

4) are interestingly in agreement with the deduced parameters using Leuenberger analysis

of the compression data (Fig. 8/Table 3). For example, the high compression

susceptibility parameters (γ) for the 18 and 13 kDa chitosans (Table 3) indicate that high

tensile strengths are attained at lower pressure of compression compared to the other MW

chitosans. The foregoing is in agreement with Table 4 which indicates that the 18 and the

13 kDa reach a higher tensile strength values at 2.94 MPa than the other MW chitosans.

The 100, 30, and 8 kDa chitosans showed low γ values, thus higher compression

pressures are required to produce stronger tablets. However, their σmax values (Table 3)

are higher than that for the 13 and 8 kDa chitosans indicating stronger compacts at higher

pressures. Table 4 confirms the foregoing finding as the highest tensile strengths are

achieved by the 100, 30, and 8 kDa chitosans.

In order to rationalize differences in tablets strength with pressure for different MW

chitosans, the powder elastic recovery and compacts ejection work have to be considered

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

23

hereafter. Initially, when the tensile strength of the compacts in the whole applied force

range (Table 4) is examined for chitosan lower than 100 kDa, up to an applied force of

2.94 kN, almost all samples exhibited higher tensile strength than the 100 kDa chitosan.

Increasing the applied force above 2.94 kN the tensile strengths for the same samples

were lower than that for the 100 kDa. Examining the profiles of the elastic recovery work

in Fig. 5 for forces lower and higher than 2.94 kN, shows that for chitosan < 100 kDa, the

elastic recovery was much lower than the 100 kDa chitosan up to 2.94 kN. Above 2.94

kN, the elastic recovery work of the low MW chitosans started to approach the elastic

recovery work value of the 100 kDa. This trend correlates well with the tensile strength

trend obtained below and above 2.94 kN. In other words, compacts with lower tendency

for elastic recovery work have higher tensile strength values. On the other hand, when the

tablets ejection work is considered (Fig. 6), at the lowest compression force (0.98 kN),

the compact of the 100 kDa chitosan displayed the highest ejection work indicating the

highest frictional forces with the die wall. Such high forces may have further contributed

to the lowest tensile strength of the 100 kDa at 0.98 kN. Higher compression forces

render smoother ejection of the 100 kDa compacts and thus higher tensile strengths

compared to the other MW chitosans. In contrast, lower MW chitosan manifest higher

ejection work when the compression pressure is increased thus lower compacts tensile

strength.

The high mechanical strength of the high MW chitosan has been previously reported and

attributed to the entanglement network of the high MW chitosan [57]. The 100 kDa

sample shows a β conformation whereby the individual chains are arranged in a parallel

manner. Hence giving rise to the appearance of ordered structure and thereby to a semi-

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

24

crystalline lattice [58]. Acharya et al reported that lowering the MW of chitosan results in

a decrease in crystallinity due to a larger d-spacing as a result of an increase in the

dimensions of the unit cell. Such a change results in a weaker network of hydrogen-bonds

between chitosan chains as evidenced by FT-IR experiments. The later technique

indicated a change in the environment of primary -OH groups and thus in the overall

hydrogen bonding network [59].

When lower chitosan particle size is used (Table 5), the reduced packing extent indicates

that the powder bed is of less porous structure resulting in lower volume reduction upon

compression of both MWs at 4.9 kN. Such reduced powder porous structure takes place

concurrently with enhanced powder deformation resulting in higher work of compression

for lower particle size of chitosan. The powders are found to undergo increased plastic

deformation (Lower PY) when smaller particle size chitosan is used. Such increased

plasticity, and thereby extent of particle deformation, results in the appearance of fresh

surfaces for binding. Therefore more coherent and stronger compacts are produced when

lower chitosan particle size is compressed. The lower A Heckel parameter for lower

particle size indicates the lower particle rearrangement and thereby explains the lower

volume reduction obtained. Consequently, having less porous powders, more coherent

compacts are produced with less frictional forces with the die wall as the work of tablet

ejection is lower for lower particle size.

The molecular structure of the high MW chitosan has been previously shown to be of

open helical orientation whereas the low MW chitosan has been shown to be of closed

and highly packed helical complex [45]. Such finding was established when chitosan

molecular modeling was carried out in water. In contrast, when chitosan molecular

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

25

modeling was carried out in vacuum, as in the current work, the high MW chitosan

showed high extended and closely packed molecular structure. This result is reasonable

with regard to the fact that high MW chitosan has more hydrogen bonds across parallel

polymeric chains, and consequently, more stable as compared to low MW chitosan.

Therefore, it can be postulated that low MW chitosan requires higher energy to yield

compacted helical structures where molecules can be intermingled. For high MW

chitosan of highly packed structure and, accordingly low d-spacing, requires lower

energy to bind. This was evident in the current work (Fig. 9) using molecular modeling

when calculating the binding energies between two polymeric species of two different

MW chitosans (88 and 8 kDa). The binding energy of high MW chitosan (-2.12 kCal/mol

per unit) is higher than that for the low MW chitosan (-1.56 kCal/mol per unit). Such high

binding energy leads to a more rigid structure of the high MW chitosan. Consequently

high MW chitosan requires less applied compression forces to bind and allows less

energy (as measured herein from the force displacement curve) to get the two polymeric

species closer. On the other hand, a greater compression force and energy are required to

get the small helical strands closer for the low MW chitosan.

The powder compression and tablet properties dependence on chitosan MW will clearly

impart variations in drug dissolution profiles when the MW is changed. For diclofenac

sodium (Fig. 11) the relatively fast drug release for the 100 kDa chitosan tablets reflects

appropriate application of such chitosan immediate release tablet preparations. On the

contrary, the slow drug release for the 30 kDa and the 13 kDa chitosan tablets reflects

suitability to be used in controlled release drug applications.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

26

The high MW chitosan in its own is not generally used for sustained release applications;

in contrast, it is used as an immediate disintegrating additive [60]. The reason behind that

is due to the fact that high MW chitosan, if not modified, does not undergo gelling which

represents a perquisite for sustained action to take place [61]. On the contrary, the high

water uptake of chitosan makes it a good candidate for immediate release applications

[60]. Thus, in order to prolong the release, chitosan is generally combined with anionic

polymers which form swellable non-erodible gels that are capable of prolonging the drug

release [62-65]. On the other hand, low MW chitosan, in its own, has not been

investigated thoroughly and compared with higher MW chitosans in regards to drug

release behavior. However, the low MW chitosan was noticed to cause prolonged release

action on some drugs [66]. In the current investigation, tablet swelling, as a factor in

hindering the drug release, is excluded in having a contribution since the drug dissolution

media was not acidic (phosphate buffer pH 6.8 according to USP specifications) [67].

Thus the apparent slow drug release using low MW chitosans is suggested to be

attributed to other factors such as the slow erosion rate. On the other hand, both modes of

drug release, i.e. immediate and sustained, are preliminary attributed to the behavior of

packing extent in response to increasing the MW. Such behavior is valid irrespective of

the absence (Fig. 3) or the presence (Fig. 12) of diclofenac Na with chitosan whereby

lower MW chitosans induce higher packing extents. The net result, therefore,

conclusively demonstrates the suitability of chitosan to be used in immediate and

sustained release applications for the high and low MW chitosan powders respectively.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

27

5. CONCLUSIONS

The change from high to low MW chitosan brings about changes in the mechanical

properties. Above specific compression force limits, low MW (≤30 kDa) chitosans

manifest higher levels of; fragmentation and brittle-fracture tendency, compression work,

packing, and, to some extent, compacts ejection work, relative to the 100 kDa chitosan.

All of the aforementioned changes depend on two main factors; i.e. the extent of

compression force applied and the chitosan particle size. In the first factor, Leuenberger

analysis revealed that stronger compacts of high MW chitosans (≥30 kDa) are reached

slower, upon increasing the compression forces, than the other lower MW chitosans. The

foregoing has been attributed to the increasing elastic recovery and ejection work of the

low MW chitosans upon increasing the compression pressure thus contributing to weaker

compacts. In the second factor, lowering the chitosan particle size, for the highest (100

kDa) and the lowest (8 kDa) MWs, reduces the particle rearrangement and volume

reduction upon compression resulting in less porous powders and more coherent

compacts. Modeling high MW chitosan interacting polymeric pairs appears more

extended and shows more extensive hydrogen bonding parallel chains leading to more

compact and stable structure than the low MW chitosan. The aforementioned results

explain the low energy required to compress the high MW chitosan as indicated by the

force displacement curves. Differences in compression properties may further correlate to

the immediate release profile of diclofenac with the 100 kDa chitosan and the sustained

release profile with the 30 and 8 kDa chitosans upon dissolution of the drug.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

28

CONFLICT OF INTEREST

The authors have declared that there is no conflict of interest.

REFERENCES

[1] M. El-Barghouthi, A. Eftaiha, I. Rashid, M. Al-Remawi, A. Badwan, A Novel

Superdisintegrating Agent Made from Physically Modified Chitosan with Silicon

Dioxide, Drug Dev. Indust. Pharm. 34 (2008) 373–383.

[2] R. Ruiz-Caro, MD. Veiga-Ochoa, Characterization and Dissolution Study of

Chitosan FreezeDried Systems for Drug Controlled Release, Molecules 14 (2009)

4370-4386.

[3] L. Sun, Y. Wang, T. Jiang, X. Zheng, J. Zhang, J. Sun, C. Sun, S. Wang, Novel

Chitosan-Functionalized Spherical Nanosilica Matrix As an Oral Sustained Drug

Delivery System for Poorly Water-Soluble Drug Carvedilol, ACS Appl. Mater.

Interfaces 5 (2013) 103-13.

[4] C. Tarirai, G.M. Enslin, J.H. Steenekamp, J.H. Hamman, Moulded cross-linked

chitosan matrix systems for controlled drug release, Drug Develop. Technol. 16

(2011) 295-301.

[5] A.F. Eftaiha, N. Qinna, I.S. Rashid, M.M. Al Remawi, M.R. Al Shami, T.A.

Arafat, A.A. Badwan, Bioadhesive Controlled Metronidazole Release Matrix

Based on Chitosan and Xanthan Gum. Mar. Drugs 8 (2010) 1716-1730.

[6] F. Al-Akayleh, M. Al Remawi, I. Rashid, A. Badwan, Formulation and In Vitro

Assessment of Sustained Release Terbutaline Sulfate Tablet Made from Binary

Hydrophilic Polymer Mixtures, Pharm. Dev. Technol. 18 (2013) 1204-1212.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

29

[7] F.M. Goycoolea, A. Valle-Gallego, R. Stefani, B. Mencicchi, L. David, C.

Rochas, M.J. Santander-Ortega, M.J. Alonso, Chitosan-based nanocapsules:

physical characterization, stability in biological media and capsaicin

encapsulation, Colloid Polym. Sci. 290 (2012) 1423-1434.

[8] C. Prego, D. Torres, M.J. Alonso, Chitosan nanocapsules: a new carrier for nasal

peptide delivery, J. Drug Del. Sci. Tech. 16 (2006) 331–337.

[9] M.V. Lozano, D. Torrecilla, D. Torres, A. Vidal, F. Dominguez, M.J. Alonso,

Highly efficient system to deliver taxanes into tumor cells: docetaxel-loaded

chitosan oligomer colloidal carriers, Biomacromolecules 9 (2008) 2186–2193.

[10] H.K. No, J.W. Nah, S.P. Meyers, Effect of time/ temperature treatment

parameters on depolymerization of chitosan, J. Appl. Polym. Sci. 87 (2003a)

1890-1894.

[11] P.R. Rege, L.H. Block, Chitosan processing:Influence of process parameters

during acetic acid and alkaline hydrolysis and effect of the processing sequence

on the resultant chitosan’s properties, Carbohyd. Res. 321 (1999) 223-245.

[12] Q. Li, E.T. Dunn, E.W. Grandmaison, M.F. Goosen, Applications and properties

of chitosan, J. Bioact. Compat. Pol. 7 (1992) 370-397.

[13] K.T. Hwang, S.T. Jung, G.D. Lee, M.S. Chinnan, Y.S. Park, H.J. Park,

Controlling Molecular Weight and Degree of Deacetylation of Chitosan by

Response Surface Methodology, J. Agric. Food Chem. 50 (2002) 1876-82.

[14] C. Schatz, C. Viton, T. Delair, C. Pichot, A. Domard, Typical physicochemical

behaviors of chitosan in aqueous solution, Biomacromolecules 4 (2003) 641–648.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

30

[15] G. Lamarque, J.M. Lucas, C. Viton, A. Domard, Physicochernical behavior of

homogeneous series of acetylated chitosans in aqueous solution: role of various

structural parameters, Biomacromolecules 6 (2005) 131–142.

[16] W. Pasanphan, T. Rattanawongwiboon, S. Choofong, O. Güvenc, K.K. Katti,

Irradiated chitosan nanoparticle as a water-based antioxidant and reducing agent

for a green synthesis of gold nanoplatforms, Radiat. Phys. Chem. 106 (2015) 360–

370.

[17] Almalik, R. Donno, C.J. Cadman, F. Cellesi, P.J. Day, N. Tirelli, Hyaluronic acid-

coated chitosan nanoparticles: Molecular weightdependent effects on morphology

and hyaluronic acid presentation, J. Control Release 172 (2013) 1142-1150.

[18] D. Meia, S. Maoa, W. Suna, Y. Wanga, T. Kissel, Effect of chitosan structure

properties and molecular weight on the intranasal absorption of

tetramethylpyrazine phosphate in rats, Eur. J. Pharm. Biopharm. 70 (2008) 874–

881.

[19] A.I. Ruiz Matute, A. Cardelle-Cobas, A.B. García-Bermejo, A. Montilla, A.

Olano, N. Corzo, Synthesis, characterization and functional properties of

galactosylated derivatives of chitosan through amide formation, Food

Hydrocolloid. 33 (2013) 245-255.

[20] E. Dini, S. Alexandridou, C. Kiparissides, Synthesis and characterization of cross-

linked chitosan microspheres for drug delivery applications, J. Microencapsul. 20

(2003) 375-385.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

31

[21] M. Huang, E. Khor, L.Y. Lim, Uptake and cytotoxicity of chitosan molecules and

nanoparticles: Effects of Molecular Weight and Degree of Deacetylation, Pharm.

Res. 21 (2004) 344-353.

[22] K.A. Janes, M.J. Alonso, Depolymerized chitosan nanoparticles for protein

delivery: Preparation and Characterization, J. Appl. Polym. Sci. 88 (2003) 2769–

2776.

[23] K.G.H. Desai, H.J. Park, Preparation and characterization of drug-loaded

chitosan–tripolyphosphate microspheres by spray drying, Drug Dev. Res. 64

(2005) 114–128.

[24] V.B. Morris, S. Neethu, T.E. Abraham, C.K. Pillai, C.P. Sharma, Studies on the

condensation of depolymerized chitosans with DNA for preparing chitosan-DNA

nanoparticles for gene delivery applications, J. Biomed. Mater. Res. B Appl.

Biomater. 89 (2009) 282-292.

[25] V.G. Mir, J. Heinämäki, O. Antikainen, O.B. Revoredo, A.I. Colarte, O.M. Nieto,

J. Yliruusi, Direct compression properties of chitin and chitosan, Eur. J. Pharm.

Biopharm. 69 (2008) 964-968.

[26] C. Le Tien, M. Lacroix, P. Ispas-Szabo, M.A. Mateescu, N-acylated chitosan:

hydrophobic matrices for controlled drug release, J. Control. Release 93 (2003) 1-

13.

[27] N. Zerrouk, N. Mennini, F. Maestrelli, C. Chemtob, P. Mura, Comparison of the

effect of chitosan and polyvinylpyrrolidone on dissolution properties and

analgesic effect of naproxen, Eur. J. Pharm. Biopharm. 57 (2004) 93-99.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

32

[28] X. Fu, M. Dutt, A.C. Bentham, B.C. Hancock, R.E. Cameron, J.A. Elliott,

Investigation of particle packing in model pharmaceutical powders using X-ray

microtomography and discrete element method, Powder Technol. 167 (2006)

134–140.

[29] K. Ofori-Kwakye, J.T. Fell, S.L. Kipo, Effects of pH of medium and molecular

weight on polyelectrolyte complex formation between pectin and chitosan, J. Sci.

Technol. 26 (2006) 66-73.

[30] D. Aggarwal, H.W. Matthew, Branched chitosans: effects of branching

parameters on rheological and mechanical properties, J. Biomed. Mater. Res. A

82 (2007) 201-212.

[31] D.D. Chinta, R.A. Graves, S. Pamujula, N. Praetorius, L.A. Bostanian, T.K.

Mandal, Spray-dried chitosan as a direct compression tableting excipient, Drug

Dev. Ind. Pharm. 35 (2009) 43-48.

[32] A.M. Qandil, A.A. Obaidat, M.A.M. Ali, B.M. Al-Taani, B.M. Tashtoush, N.D.

Al-Jbour, M.M. Al Remawi, K.A. Al-Sou’od, A.A. Badwan, Investigation of the

interactions in complexes of low molecular weight chitosan with ibuprofen, J.

Solution Chem. 38 (2009) 695-712.

[33] British Pharmacopoeia (BP), Norwich: Stationery Office Books, 2007, Vol I, p

456.

[34] J.T. Fell, J.M. Newton, Determination of tablet strength by diametral-compression

test, J. Pharm. Sci. 59 (1970) 688–691.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

33

[35] E. Nelson, L.W. Busse, T. Higuchi, The physics of tablet compression. VII.

Determination of energy expenditure in the tablet compression process, J. Am.

Assoc. Sci. 44 (1955) 223-225.

[36] C.J. De Blaey, J. Polderman, The quantitative interpretation of forcedisplacement

curves, Pharm. Weekblad. 105 (1970) 241-250.

[37] I. Rashid, N. Daraghmeh, M. Al-Remawi, S.A. Leharne, B.Z. Chowdhry, A.

Badwan, Characterization of the impact of magnesium stearate lubrication on the

tableting properties of chitin-Mg silicate as a superdisintegrating binder when

compared to Avicel® 200, Powder Technol. 203 (2010) 609–619.

[38] R.W. Heckel, Density-pressure relationships in powder compaction, T. Metall.

Soc. AIME 221 (1961) 671–675.

[39] R.W. Heckel, An analysis of powder compaction phenomena, T. Metall. Soc.

AIME 221 (1961) 1001–1008.

[40] Krycer, D.G. Pope, J.A. Hersey, An evaluation of the techniques employed to

investigate powder compaction behavior, Int. J. Pharm. 12 (1982) 113-134.

[41] P. Shivanand, O.L. Sprockel, Compaction behavior of cellulose polymers, Powder

Technol. 69 (1992) 177–184.

[42] C. Lin, T. Cham, Compression behavior and tensile strength of heat-treated

polyethylene glycols, Int. J. Pharm. 118 (1995) 169–179.

[43] D. Blattner, M. Kolb, H. Leuenberger, Percolation theory and compactibility of

binary powder systems, Pharm. Res. 7 (1990) 113–117.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

34

[44] M. Kuentz, H. Leuenberger, A new theoretical approach to tablet strength of a

binary mixture consisting of a well and a poorly compactable substance, Eur. J.

Pharm. Biopharm. 49 (2000) 151–159.

[45] R. Obaidat, N. Al-Jbour, K. Al-Sou’d, K. Sweidan, M. Al-Remawi, A. Badwan,

Some physico-chemical properties of low molecular weight chitosans and their

Relationship to conformation in aqueous solution, J. Solution Chem. 39 (2010)

575–588.

[46] R.J. Oates, AG. Mitchell, Calculation of punch displacement and work of powder

compaction on a rotary tablet press, J. Pharm. Pharmacol. 41 (1989) 517–23.

[47] C.C. Sun, D. J.W. Grant, Influence of crystal shape on the tabletting performance

of L-lysine monohydrochloride dehydrate, J. Pharm. Sci. 90 (2001) 567–577.

[48] C. Nyström, G. Alderborn, M. Duberg, P-G. Karehill, Bonding surface area and

bonding mechanism-two important factors for the understanding of powder

compactability, Drug Dev. Ind. Pharm. 19 (1993) 2143–2196.

[49] G. Shlieout, T. Laich, G. Zessin, Evaluating the elastic behavior of

pharmaceutical excipients and binary mixtures using the modified Fraser-Suzuki

function, Pharm. Tech. Eur. 14 (2002) 21–28.

[50] K.M.P. Freyer, D. Brink, Evaluation of powder and tableting properties of

chitosan, AAPS PharmSciTech. 7 (2006; (3):75.

[51] P. Paronen, Heckel plots as indicators of elastic properties of pharmaceuticals,

Drug Dev. Ind. Pharm. 12 (1986) 1903–1912.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

35

[52] S. Patel, A.M. Kaushal, A.K. Bansal, Effect of particle size and compression force

on compaction behavior and derived mathematical parameters of compressibility,

Pharm. Res. 24 (2007) 111–124.

[53] J.M. Sonnergaard, A critical evaluation of the Heckel equation, Int. J. Pharm. 193

(1999) 63–71.

[54] J.M. Sonnergaard, Impact of particle density and initial volume on mathematical

compression models, Eur. J. Pharm. Sci. 11 (2000) 307–315.

[55] C. Sun, D.J. Grant, Influence of elastic deformation of particles on Heckel

analysis, Pharm. Dev. Technol. 6 (2001b) 193–200.

[56] S. Patel, A.M. Kaushal, A.K. Bansal, Mechanistic investigation on pressure

dependency of Heckel parameter, Int. J. Pharm. 389 (2010) 66–73.

[57] J. Nunthanid, S. Puttipipatkhachorn, K. Yamamoto, G.E. Peck, Physical

properties and molecular behavior of chitosan films, Drug Dev. Ind. Pharm. 27

(2001) 143-57.

[58] B. Focher, A. Naggi, G. Tori, A. Cosani, M. Terbojevich, Chitosans from

Euphausia superba. 2: Characterization of solid-state structure, Carbohydr. Polym.

18 (1992) 43–49.

[59] A.B. Kumar, L.R. Gowda, R.N. Tharanathan, Non-specific depolymerization of

chitosan by pronase and characterization of the resultant products, Eur. J.

Biochem. 271 (2004) 713–723.

[60] K. Huanbutta, K. Cheewatanakornkool, K. Terada, J. Nunthanid, P. Sriamornsak,

Impact of salt form and molecular weight of chitosan on swelling and drug

release from chitosan matrix tablets, Carbohydr. Polym. 97 (2013) 26-33.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

36

[61] S.K. Kim, Chitin and Chitosan Derivatives: Advances in Drug Discovery and

Developments; CRC Press, Boca Raton, FL, USA, 2013.

[62] J.S. Maciel, H.C.B. Paula, M.A.R. Miranda, J.M. Sasaki, R.C.M. de Paula,

Reacetylated chitosan/cashew gum gel: Preliminary study for potential

utilization as drug release matrix, J. Appl. Polym. Sci. 99 (2006) 326-34.

[63] T.V. Rao, G.K. Kumar, M.G. Ahmed, V. Joshi, Development and evaluation of

chitosan based oral controlled release matrix tablets of losartan potassium, Int. J.

Pharm. Investig. 2 (2012) 157-61.

[64] S.H. Park, M.K. Chun, H.K. Choi, Preparation of an extended release matrix tablet

using chitosan/carbopol interpolymer complex, Int. J. Pharm. 347 (2008) 39–44.

[65] L. Li, L. Wang, Y. Shao, R. Ni, T. Zhang, S. Mao, Drug release characteristics from

chitosan-alginate matrix tablets based on the theory of self assembled film, Int. J.

Pharm. 450 (2013) 197–207.

[66] Y. Shao, L. Li, X. Gu, L. Wang, S. Mao, Evaluation of chitosan-anionic polymers

based tablets for extended-release of highly water-soluble drugs, Asian J pharm.

Sci. 10 (2015) 24-30.

[67] M.K. Singh, S.K. Prajapati, A. Mahor, N. Rajput, R. Singh, Chitosan: A novel

excipient in pharmaceutical formulation: a review, Int. J. Pharm. Sci. Res. 2

(2011) 2266-2277.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

37

Captions:

Tables

Table 1: Heckel parameters; yield pressure (PY) and the intercept (A), of different MW

chitosan powders upon compression using the GTP.

Table 2: Kawakita parameters; compressibility (a) and PK, of different MW of chitosan

powder upon compression using the GTP.

Table 3: Compression analysis using Leuenberger equation of different MW chitosans

compressed using the GTP. The tensile strength and density of the 6 mm tablets were

calculated from their hardness and tablet thickness values.

Table 4: Tensile strengths (MPa) of the different MW chitosan samples compressed at a

force range 0.98-4.91 kN (equivalent to 100-500 kg load applied by the GTP).

Table 5: Compression and tablet properties of different particle size chitosans for the 100

kDa and the 8 kDa MWs.

Figures

Fig. 1.: A typical force-displacement curve displayed by the GTP for the compression of

100 kDa chitosan (100 mg sample) at 100 kg (4.91 kN) compression force.

Fig. 2: A typical force-displacement curve displayed by the GTP for the tablet ejection

compressed at 100 kg (4.91 kN) compression force for the 100 kDa chitosan (100 mg

sample).

Fig. 3.: Extent of powder packing after compression relative to the initial bed volume for

each chitosan MW at different compression forces.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

38

Fig. 4.: Compression work of different MW chitosan powders as a function of

compression force

Fig. 5.: Elastic recovery work of different MW chitosan powders as a function of

compression force.

Fig. 6.: Tablet ejection work of different MW chitosan powders as a function of

compression force.

Fig. 7.: Heckel plots of different MW chitosan samples compressed at a pressure range

34.6-173.2 MPa.

Fig. 8.: The tensile strength versus the product of the pressure of compression and the

relative density of different MW chitosan powders. The powders were compressed using

the GTP.

Fig. 9.: Chemical structure of D-Glucosamine (A), high MW (B) and low MW (C)

chitosans.

Fig. 10.: A section of the high MW/more extended (A) and low MW/less extended (B)

chitosans showing extensive and less extensive intermolecular hydrogen bonds.

Fig. 11.: Dissolution of diclofenac sodium tablets containing different MW chitosan

powders; 100 kDa, 30 kDa and 13 kDa chitosans.

Fig. 12.: Extent of powder packing of diclofenac/chitosan preparations after compression

relative to the initial bed volume for each chitosan MW at different compression forces.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

39

Figure 1

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

40

Figure 2

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

41

Figure 3

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

42

Figure 4

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

43

Figure 5

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

44

Figure 6

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

45

Figure 7

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

46

Figure 8

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

47

Figure 9

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

48

Figure 10

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

49

Figure 11

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

50

Figure 12

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

51

Table 1

MW

(kDa)

Yield pressure (1/K), PY (MPa) A

100 72.46 0.41

30 98.03 0.45

18 106.38 0.46

13 120.48 0.52

8 153.84 0.59

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

52

Table 2

Chitosan

MW (kDa)

Kawakita

slope

Kawakita

intercept a 1/b (PK)

100 1.32 10.83 0.75 8.15

30 1.86 15.13 0.54 8.15

18 1.58 11.87 0.63 7.50

13 1.45 55.67 0.69 38.32

8 1.91 41.16 0.52 21.55

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

53

Table 3

Chitosan MW (kDa) Compression susceptibility

(MPa-1)

Maximum tensile strength,

σmax (MPa)

100 0.003751±0.001 14.23±1.1

30 0.003009±0.001 14.98±0.9

18 0.04288±0.005 3.853±0.5

13 0.02535±0.003 4.149±0.8

8 0.006552±0.001 9.446±1.0

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

54

Table 4

MW of Chitosan (kDa)

100 30 18 13 8

Force (kN) Tensile strength (MPa)

0.98 0.80±0.11 2.26±0.33 2.22±0.24 1.41±0.18 2.78±0.14

1.96 1.53±0.20 2.60±0.17 3.59±0.31 2.37±0.24 2.50±0.30

2.94 3.42±0.29 2.88±0.18 3.63±0.28 4.27±0.21 3.19±0.19

3.92 4.90±0.12 4.73±0.17 3.44±0.38 3.78±0.26 4.85±0.14

4.91 6.53±0.31 6.11±0.24 4.20±0.14 3.75±0.23 5.76±0.19

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

55

Table 5

Test MW (kDa)

Test values

at chitosan particle size

500-250 µm 250-90 µm <90 µm

Packing extent (%) 100 83.46 81.21 79.27

8 84.37 82.82 80.94

AUC*/compression 100 341.23 379.13 417.19

8 495.68 501.98 557.25

AUC/Elastic

recovery

100 59.45 57.14 56.79

8 57.65 55.51 53.06

AUC/Tablet

ejection

100 48.72 37.56 29.63

8 69.43 59.19 48.05

Heckel parameter,

PY (MPa)

100 91.65 72.46 60.71

8 790.74 153.84 130.76

Heckel parameter,

A

100 0.58 0.41 0.38

8 0.67 0.59 0.54

Tablet tensile

strength (MPa)

100 3.71 6.53 7.14

8 3.29 5.76 6.42

*AUC: area under the curve which is equal to work (x 10-2

J)

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

56

Graphical abstract

Chemical structure of D-Glucosamine (A), high MW (B) and low MW (C) chitosans.

ACC

EPTE

D M

ANU

SCR

IPT

ACCEPTED MANUSCRIPT

57

Highlights

Tensile strength of chitosan compacts is associated with chitosan molecular

weight.

Low chitosan compacts tensile strength is attributed to high elastic recovery.

High MW chitosan has a more stable structure than the low MW chitosan.

High MW chitosan provides immediate release properties in solid form

preparations.

Low MW chitosan provides sustained release properties