Vous êtes sur la page 1sur 10

E L S E VI E R International Journal of Pharmaceutics 116 (1995) 159-168

i n t e r n a t i o n a l
j o u r n a l o f
p h a r m a c e u t i c s
Matrix swelling: A simple model describing extent of swelling of
HPMC matrices
L u c y Sa i Ch e o n g Wa n *, P a u l Wa n Si a He n g , L e e F u n Wo n g
Department of Pharmacy, National University of Singapore, Singapore, Singapore
Received 13 May 1994; accepted 20 August 1994
Abstract
The swelling of HPMC matrices was measured by determining the vertical displacement with time using a dial
indicator. Curve fitting of the data of swelling was carried out by using the coupled Case I-Case II model. The
thickness of the swollen layer formed around the matrix core was greater in HPMC compacts of higher viscosity
grade. The rates of Case I (a~ coefficient) and Case II (/3 S coefficient) swelling mechanisms increased with
concentration and/ or viscosity grade of HPMC.
Keywords: Hydroxypropylmethylcellulose; Swelling; Case I diffusion; Case II relaxation; Coupled Case I; Case II
model; Viscosity
1. I nt r oduc t i on
Swelling me a s ur e me nt s of cel l ul ose and t he
r el at i on bet ween t he r heol ogi cal and swelling
pr oper t i es of cel l ul ose wer e ext ensi vel y di scussed
by We s t ma n and Li nds t r om (1981a,b,c,d). The
swelling of cel l ul ose was f ound t o be i nf l uenced
by cross-l i nki ng, heat t r e a t me nt and p H of t he
liquid used. The swelling of cor e t abl et s dur i ng
aqueous coat i ng and a model descr i bi ng t he ex-
t ent of swelling and wat er penet r at i on f or insolu-
* Corresponding author. National University of Singapore,
Dept of Pharmacy, 10 Kent Ridge Crescent, 0511 Singapore,
Singapore.
bl e t abl et s cont ai ni ng a super - di si nt egr ant was
de mons t r a t e d by Far oongs ar ng and Peck (1991,
1992). Mi t chel l et al. (1991) showed t hat t he
swelling behavi our of cel l ul ose et her mat r i ces
cont ai ni ng HPMC depends on t he t emper at ur e,
dr ug and pol ymer subst i t ut i on type.
The swelling of HPMC- i bupr of e n mat r i ces
coul d be descr i bed by a f i r st - or der ki net i cs ( Wan
et al., 1993a). Mat r i x swelling was af f ect ed by t he
concent r at i on and viscosity gr ade of t he pol ymer .
On f ur t her exami nat i on, it was f ound t hat mat r i x
swelling of HPMC- pr opr a nol ol mat r i ces devi at es
significantly f r om this ~/time rel at i onshi p. In this
study, t he ki net i cs of mat r i x swelling was exam-
i ned usi ng t he coupl ed Case I - Case I I equat i on as
was used for liquid upt ake i nt o t hese mat r i ces
( Wan et al., 1993b,c).
0378-5173/95/$09.50 1995 Elsevier Science B.V. All rights reserved
SSDI 0378-5173(94)00285-1
160 L.S. Cheong Wan et aL / International Journal of Pharmaceutics 116 (1995) 159-168
2. Ma t e r i a l s a n d me t h o d s
2.1. Materials
Pr opr anol ol hydr ochl or i de ( USP XXI I ) and
i bupr of en ( phar maceut i cal grade, Italy) wer e cho-
sen as model drugs t o r epr es ent wat er-sol ubl e
and poorl y wat er-sol ubl e drugs, respectively.
Hydroxypropyl met hyl cel l ul oses, HPMC (Shin-
Et su Chemi cal Co., Japan) of di f f er ent viscosity
grades: 4000 (K4), 15000 (K15), 30000 (K30),
50 000 (K50) and 100 000 (K100) cps wer e used as
mat ri x pol ymers. The appar ent viscosities of 2%
( w/ v) aqueous sol ut i ons of t he HPMC wer e 4380,
18200, 35800, 44400 and 100000 cps, respec-
tively. The viscosity was det er mi ned by t he USP
XXI I met hod.
devi at i on of t he i ncrease in thickness of t hr ee
mat ri ces wer e det er mi ned.
2.2.3. Treatment of data
All cal cul at i on was per f or med using S AS TM
soft ware on a per sonal comput er .
3. Re s u l t s a n d d i s c u s s i o n
Taki ng a as t he original t hi ckness of t he dry
matrix, s as t he t hi ckness of t he swollen matrix,
t he nor mal i sed i ncrease in t he mat ri x thickness,
expressed as a per cent age, ~, can be used as a
swelling i ndex where:
s - - a
6 = 100% (1)
a
2.2. Methods
2. 2.1. Preparation of ma trices
The dr ug and HPMC wer e t hor oughl y mi xed
in a mixing bag f or 10 min. A wei ghed amount of
t he mi xt ure was f ed manual l y i nt o t he di e of a
si ngl e-punch t abl et t i ng machi ne ( Manest y F3,
UK) to pr oduce a mat ri x t abl et of 600 mg and
porosi t y of 0.10 + 0.01 using fiat punches of di-
amet er 14 mm f or HPMC- i bupr of en mat ri ces,
and 300 mg and porosi t y of 0.15 + 0.01 using fiat
punches of di amet er 9.5 mm f or HPMC- pr o-
pr anol ol mat ri ces. Mat ri ces wer e pr epar ed using
t he vari ous viscosity grades of HPMC and with
di f f er ent concent r at i ons of t he drug.
2. 2. 2. Swelling studies
In a ci rcul ar dish measur i ng 6.5 cm in di ame-
t er and 3.4 cm deep, t he mat ri x was mount ed
bet ween two fi l t er paper s ( What man 54) and a
glass cover slip was pl aced bet ween t he tip of t he
dial and t he upper fi l t er paper ( Wan et al., 1993a).
The whol e assembl y was pl aced in a wat er - bat h
t her most at i cal l y mai nt ai ned at 37 + 0.5C. A fixed
vol ume (50 ml) of t he liquid, pr e- equi l i br at ed at
37 _+ 0.5C, was pour ed i nt o t he dish. The vert i cal
i ncrease in t he mat ri x t hi ckness was r ecor ded
with a dial i ndi cat or (Mi t ut oyo, 2118-50, Japan).
For each f or mul at i on, t he mean and st andar d
3.1. Kinetic models of the swelling mechanisms
The dynami c swelling of t he matrix, r epr e-
sent ed by 3, as a f unct i on of time, t, was anal yzed
accordi ng to:
log 6 = ns l og t + e2 ( 2)
wher e 3 is t he swelling i ndex def i ned in Eq. 1, n S
is t he exponent descri bi ng Fi cki an or anomal ous
swelling mechani sm and c 2 is a const ant . The
val ues of c 2 and exponent s n s f or t he vari ous
f or mul at i ons anal yzed are given in Tabl es 1 and
2.
The swelling mechani sm of HPMC- i bupr of en
mat ri ces in pH 7.2 buf f er sol ut i on was approxi -
mat el y Case I (Fi cki an) wi t h n s cl ust eri ng ar ound
0.5 ( Tabl e 1). At ver y low HPMC concent r at i on
( 5- 10%) , t he est i mat i on of ns was less accurat e,
pr obabl y due t o t he l i mi t ed number of dat a poi nt s
col l ect ed f or t hese mat ri ces, in compar i son with
ot hers. Thi s was also t r ue for n~ val ues obser ved
in HPMC- pr opr anol ol mat r i ces cont ai ni ng similar
amount of HPMC ( Tabl e 2). Thi s was because
t he amount of i ncrease in mat ri x swelling in t hese
mat r i ces wer e ver y low and in some instances, t he
mat r i ces col l apsed af t er a maxi mum i ncrease in
t hi ckness was at t ai ned. For HPMC- pr opr anol ol
mat ri ces, t he swelling mechani sm in wat er be-
came non- Fi cki an wi t h i ncreasi ng amount s of
L.S. Cheong Wan et al. ~I nternational J ournal of Pharmaceutics 116 (1995) 159-168 161
Table 1
Values of kinetic exponent ( n s ) , c 2 and correlation coefficient
( r 2) following linear regression analysis of data from swelling
studies of HPMC-ibuprofen matrices at p H 7. 2
Viscosity Percent Log 6 = n s log t + c 2
grade HP MC
H s C 2
r 2
K4 5 0. 12_+0. 01 - 2 . 4 5 0. 8511
10 0. 30 _+ 0. 00 - 1.54 0. 9853
20 0. 40 _+ 0.01 - 0. 88 0. 9725
30 0.41 -+0.01 - 0 . 7 6 0. 9867
40 0. 48 _+ 0. 00 - 0. 99 0. 9968
50 0. 50 _+ 0. 00 - 0. 89 0. 9968
100 0. 56_+0. 01 - 1 . 4 4 0. 9952
K15 5 0. 16_+0. 01 - 4 . 3 6 0. 8473
10 0. 32 + 0.01 - 1.86 0. 9664
20 0. 42 _+ 0.01 - 0. 89 0. 9834
30 0. 43 _+ 0.01 - 0. 82 0. 9876
40 0. 47 _+ 0. 00 - 0. 82 0. 9970
50 0. 50 _+ 0. 00 - 1.05 0. 9985
100 0. 58 _+ 0. 00 - 1.60 0. 9977
K30 5 0. 22 _+ 0.01 - 2. 25 0. 9675
10 0. 37_+0. 01 - 2 . 2 0 0. 9535
20 0. 47 _+ 0. 00 - 1.28 0. 9940
30 0. 45 _+ 0. 00 - 0. 88 0. 9944
40 0. 47 -+ 0. 00 - 0. 70 0. 9984
50 0. 48 + 0. 00 - 0. 79 0. 9992
100 0. 58 -+ 0. 00 - 1.45 0. 9980
K50 5 0. 52 -+ 0. 02 - 3. 82 0. 9435
10 0. 44 _+ 0.01 - 2. 47 0. 9884
20 0. 42 + 0. 00 - 0. 88 0. 9942
30 0. 43 _+ 0. 00 - 0. 69 0. 9933
40 0. 47 _+ 0. 00 - 0. 82 0. 9970
50 0. 48 -+ 0. 00 - 0. 72 0. 9958
100 0. 60_+0. 01 - 1.46 0. 9956
HP MC ( Tabl e 2). Thi s i ndi c a t e d t he i nc r e as i ng
i mpo r t a nc e o f Cas e II r e l axat i onal me c h a n i s m t o
overal l mat ri x s wel l i ng.
In or de r t o as c e r t ai n t he i mpo r t a nc e o f t he
t wo s we l l i ng me c h a n i s ms t o ove r al l mat ri x
s wel l i ng, t he ki net i cs o f s we l l i ng was r e - e v a l ua t e d
by c o ns i de r i ng t hat t he s we l l i ng o f mat ri ces de -
p e n d s o n t wo proces s es : l i qui d di f f us i on i nt o t h e
mat ri x and pol yme r s we l l i ng d u e t o t h e p e n e -
trant. Ca l c ul a t i o n o f t he appr oxi mat e c ont r i bu-
t i ons o f t he t wo me c h a n i s ms t o overal l s we l l i ng
pr oc e s s was carri ed o u t by f i t t i ng t h e dat a t o t he
s e c o n d or de r e q u a t i o n a n a l o g o u s t o t hat pr o-
p o s e d by Pe ppa s and Sahl i n ( 1989) f or r e l e as e
f rom s we l l abl e mat ri ces. Th e e q u a t i o n o f t he
mo d e l is:
t~ = a s t l / 2 + f i s t + c 3 (3)
whe r e t he first t erm o n t he r i ght - hand s i de repre-
s e nt s Cas e I ( Fi c ki an) c o nt r i but i o n and t he sec-
o n d t erm is t he Cas e II ( r e l axat i onal ) c ont r i bu-
t i on, ~5 is t he s wel l i ng i nde x o f matri x, % and fls
c o r r e s po nd t o s wel l i ng rat es o f t he Cas e I and
Cas e II me c ha ni s ms res pect i vel y, c 3 is a c o ns t a nt
and t is t i me. Th e g o o dne s s - o f - f i t o f t he s wel l i ng
dat a was al s o t e s t e d ac c or di ng t o t wo o t he r mo d-
el s r e pr e s e nt i ng Cas e II a l o ne ( Eq. 4) and Cas e I
a l o ne ( Eq. 5):
= / ~ ' s t + C 4 ( 4 )
= a,st 1/2 + cs ( 5 )
Table 2
Values of kinetic exponent ( n s ) c 2 and correlation coefficient
( r 2) following linear regression analysis of data from swelling
studies of HPMC-propranolol matrices in water
Viscosity Percent Log t5 = n s log t + c .
g r a d e HP MC
/ / s C2
K4 5 0. 25 _+ 0. 03 0. 43 0. 7739
10 0. 06 + 0.01 1.28 0. 6502
25 0. 79 _+ 0. 03 - 3. 78 0. 9344
50 0. 64_+0. 02 - 1.96 0. 9_+49
75 1.58 _+ 0. 08 - 8. 74 0. 9036
K15 5 0. 69_+0. 18 - 3 . 1 9 0. 4999
10 0. 50 _+ 0. 02 - 2. 16 0. 9327
25 0. 39 -+ 0. 00 - 0. 55 (I.9948
50 0. 89 _+ 0. 03 - 3. 77 0. 9502
75 0. 89 +_ 0. 02 - 3. 37 0. 9787
K30 5 0. 11_+0. 04 0. 57 0. 2716
10 0. 16_+0. 00 0. 74 0. 9518
25 0. 36- +0. 00 - 0 . 1 7 0. 9898
50 0. 52 -+ 0. 00 - 0.81 0. 9955
75 0. 63 + 0. 00 - 1.45 0. 9993
K50 5 0. 63 -+ 0. 00 - 1.45 0. 9993
10 0. 40 -+ 0. 05 - 0. 58 0. 6735
25 0. 32 + 0. 00 - 0. 22 0. 9977
50 0. 52 _+ 0. 00 - 0. 73 0. 9981
75 1. 80_+0. 07 - 10.21 0. 9258
K100 5 1. 16_+0. 12 - 3 . 8 0 0. 8807
10 0. 44 _+ 0. 05 - 1.05 0. 6291
25 0. 29 _+ 0. 00 0.51 0. 9969
50 1.17 _+ 0. 03 - 5. 53 0. 9528
75 1.10 + 0. 03 - 4. 73 0. 9647
162 L . S . Cheong Wan et al. /I nternational J ournal of Pharmaceutics 1 1 6 ( 1 9 9 5 ) 1 5 9 - 1 6 8
wh e r e / 3 ' S and a' s are t he rat e c o ns t a nt s o f Cas e II
and Cas e I t rans port , res pect i vel y, and c 4 and c 5
are c ons t ant s .
Tabl e s 3 a nd 4 gi ve t he g o o dne s s - o f - f i t o f t he
t hr e e mo d e l s bas e d o n Eq. 3 - 5 . As c a n be s e e n
f rom t he s quare o f t he c or r e l at i on c oe f f i c i e nt s ,
t he Cas e I mo de l ( Eq. 5) ge ne r al l y fits t he s wel l i ng
dat a be t t e r t han t he Cas e II mo d e l ( Eq. 4). Ho w-
ever, o n f urt her e x a mi na t i o n us i ng t he po l y no -
mi al mo d e l ( Eq. 3), t he s e c o n d - o r d e r po l y no mi a l
e q u a t i o n gave t he be s t dat a fit. As an e xampl e o f
t he dat a o bt a i ne d, pl ot t i ng t he s wel l i ng dat a as a
f unc t i o n o f t i me ( Fi g. 1) a g o o d fit o f t he po l y no -
mi al mo d e l t o t he act ual dat a c ol l e c t e d. Typi cal l y,
t he Cas e I mo d e l t e nds t o o v e r e s t i ma t e t he
s we l l i ng i nde x i n t he earl y s t age and t o unde r e s t i -
ma t e t he s we l l i ng i nde x at a mu c h l at er s t age o f
t he st udy. Th e c onve r s e is t rue f or t he Cas e II
nl ode l . Ba s e d o n t he s e resul t s, Eq. 3 is us e d
s ubs e que nt l y f or t he anal ysi s o f al l dat a o n matri x
s wel l i ng.
3.2. Swelling of HPMC
HP MC is a hydr ophi l i c pol ymer. It s wel l s o n
c ont ac t wi t h wat er. Th e s wel l i ng o f HP MC c om-
pact s o f di f f e r e nt vi s cos i t y grade we r e de t e r mi ne d
i n p H 7. 2. Th e t hi c kne s s o f s wol l e n l ayer f or me d
a r o und t he matri x c or e was great er i n mat ri ces
c o nt a i ni ng HP MC o f hi g he r vi s cos i t y grade. Th e
hy dr o dy na mi c v o l u me o c c u p i e d by t he hydr at e d
pol yme r c hai ns is l arger i n hi gh vi s cos i t y grade
pol ymer. Co ns e que nt l y , great er s wol l e n mas s o f
t he mat ri ces was f or me d ( Tabl e 5).
Tabl e 3
Val ues of correlation coeffi ci ent ( r 2) fol l owi ng linear regression analysis o f data from swel l i ng studies of HPMC- i buprof en matrices
at pH 7.2 usi ng Eq. 3- 5
Vi scosi ty Percent 6 = f l ~t + c 4 ~ = a ~ t 1/ 2 + c 5 ~ = as t l / 2 + fist + c 3
grade HPMC
K4 5 0.3344 0.5535 0.8202
10 0.9079 0.9875 0.9879
20 0.9491 0.9940 0.9967
30 0.9505 0.9963 0.9988
40 0.9439 0.9987 0.9998
50 0.9442 0.9985 0.9996
100 0.9794 0.9831 0.9996
K15 5 0.4434 0.6712 0.8828
10 0.9509 0.9863 0.9910
20 0.9431 0.9974 0.9985
30 0.9420 0.9985 0.9993
40 0.9400 0.9992 0.9998
50 0.9472 0.9983 0.9999
100 0.9785 0.9841 0.9997
K30 5 0.8582 0.9683 0.9747
10 0.9747 0.9776 0.9944
20 0.9453 0.9983 0.9996
30 0.9878 0.9976 0.9993
40 0.9312 0.9998 0.9999
50 0.9367 0.9996 0.9999
100 0.9727 0.9887 0.9999
K50 5 0.7435 0.9242 0.9809
10 0.9375 0.9977 0.9982
20 0.9377 0.9989 0.9993
30 0.9295 0.9987 0.9988
40 0.9414 0.9989 0.9997
50 0.9606 0.9936 0.9991
100 0.9863 0.9749 0.9993
L. S. Ch e o n g Wa n et al. / I nt e r nat i onal J our nal o f Ph a r ma c e u t i c s 116 ( 1995) 1 5 9 - 1 6 8 163
Table 4
Val ues of cor r el at i on coef f i ci ent ( r 2) fol l owi ng l i near r egr essi on
mat r i ces in di st i l l ed wat er us i ng Eq. 3 - 5
Vi scosi t y Per cent 8 =/3' st + c 4
gr ade HP MC
anal ysi s of dat a f r om swel l i ng st udi es of HPMC- pr opr anol ol
6 = a~t 1/~ + c s 6 = a s t I / 2 + ~s t + c 3
K4 5 0.4416 0.6870 0.9822
10 0.1377 0.2897 0.5921
25 0.8523 0.9815 0.9981
50 0.8812 0.9912 0.9867
75 0.9782 0.9555 0.9867
K15 5 0.3759 0.5869 0.7921
10 0.7984 0.9444 0.9894
25 0.8146 0.9696 0.9991
50 0.8923 0.9821 0.9835
75 0.9536 0.9917 0.9958
K30 5 0.1165 0.3253 0.9004
10 0.5382 0.7253 0.9135
25 0.7623 0.9433 0.9978
50 0.8768 0.9919 0.9993
75 0.9535 0.9944 0.9979
K50 5 0.8171 0.8819 0.8820
10 0.4719 0.6928 0.9372
25 0.8091 0.9659 0.9970
50 0.9059 0.9959 0.9981
75 0.9866 0.9967 0.9903
K100 5 0.7809 0.8696 0. 87l l
10 0.5550 0.7640 0.9451
25 0.7923 0.9559 0.9933
50 0.9314 0.9891 0.9898
75 0.9654 0.9870 0.9960
The swelling profiles of HPMC compacts could
be accurately described by the polynomial model
(Eq. 3), as shown by the results of linear regres-
sion analysis of HPMC compacts in Table 4. The
coefficients of the Case I and Case II mecha-
nisms on the HPMC compacts of various viscosity
grades showed that the coefficients of the Case I
mechanism were rather similar while that of the
Case II mechani sm was slightly l arger f or t he
HPMC K50 compact .
3.3. Swelling studies of matrices containing HPMC
and ibuprofen as a poorly water-soluble drug
The plots of percent increase in matrix swelling
against t for matrices containing ibuprofen and
Table 5
The swel l i ng behavi our of HP MC compact s at pH 7.2
Vi scosi t y gr ade a S ( 10 - l) ( %/ ~/ s)
K4 2.34 + 0. 04
K15 2.30 _+ 0. 30
K30 3.00 _+ 0.02
K50 2.30 + 0.07
/3s ( 10- 3) ( %/ s )
3.68 + 0.08
3.42 + 0.06
3.35 + 0.04
5.49 _+ 0.16
y- i nt er cept c 3 63o min
0.32 16.67 + 0.48
0.21 15.97 + 1.25
0.03 18.77 _+/).72
0.48 20.06 _+ 0.19
164 L.S. Cheong Wan et al. / Int ernat i onal Journal of Pharmaceutics 116 (1995) 159-168
di f f er ent amount of HPMC of vari ous viscosity
gr ades ar e pr es ent ed in Fig. 2. The swelling of
mat r i ces i ncreases wi t h t he concent r at i on of poly-
mer. Mat r i ces cont ai ni ng 5% HPMC di d not have
suffi ci ent pol ymer f o r significant measur abl e
swelling of mat r i ces to occur, t he amount of
swollen l ayer f or med at 30 mi n (630 min ) bei ng less
t han 1% of t he original t hi ckness ( Tabl e 6). A
mi ni mum amount of 20% HPMC was r equi r ed
f or t he mat r i ces to swell t o 10% of t hei r initial
thickness.
The swelling r at es at t r i but ed t o t he Case I
mechani sm i ncr eased sharpl y when t he HPMC
concent r at i on was vari ed f r om 5 t o 50% of t he
t ot al wei ght cont ent (Fig. 3). The ef f ect of HPMC
concent r at i on on swelling rat es was less mar ked
at hi gher pol ymer cont ent ( > 50% HPMC of high
viscosity grade). A sat ur at i on st at e was at t ai ned
beyond 40% HPMC cont ent of t hese mat ri ces.
Thi s is because as t he HPMC becomes hydr at ed
and forms a swollen gel, di ssol ut i on and surface
er osi on of this wat er l ogged gel occur si mul t ane-
ously. The wat er upt ake r at e also slows down in
very high pol ymer concent r at i on as t he swollen
pol ymer can in t ur n affect t he wat er upt ake rat e.
The cont r i but i on of t he Case II mechani sm to
mat ri x swelling was relatively less t han t hat of t he
Case I mechani sm ( Tabl e 6). As obser ved in t he
case of Case I swelling rat es, t he rat es of swelling
of t he Case II mechani sm general l y i ncr eased
with HPMC concent r at i on. It is i nt erest i ng t o
not e t hat negat i ve rat es wer e obt ai ned f or mat ri -
ces cont ai ni ng 5% HPMC and 10% HPMC K4.
Thi s coul d be due t o t he part i al col l apse of mat ri -
ces duri ng swelling meas ur ement which will be
discussed shortly.
3.4. Swelling behaviour of matrices containing
HPMC and propranolol as a water-soluble drug
The swelling behavi our of mat ri ces cont ai ni ng
HPMC of vari ous viscosity gr ade and pr opr anol ol
Table 6
Effect of varying t he amount of HPMC of various viscosity grade on the swelling of HPMC- i bupr of en matrices at pH 7.2
Viscosity grade Percent HPMC as a ( X 10 - 2) ( %/ x/ s) /3 s b ( X 10 - 4) ( %/ S) y-i nt ercept c 3 t]30 min
K4 5 0.66 + 0.05 - 0.84 + 0.09 0.09 0.20 + 0.08
i 0 4.36 + 0.22 0.46 + 0.38 0.37 2.17 + 0.23
20 16.38 + 0.56 6.68 + 0.98 0.67 8.96 + 0.52
30 20.07 + 0.42 7.93 _+ 0.72 0.77 10.82 _+ 1.34
40 29.48 _+ 0.24 7.12 _+ 0.41 0.25 14.13 _+ 1.36
50 36.49 _+ 0.39 9.10 +_ 0.68 0.19 17.36 _+ 1.32
K15 5 0.14 _+ 0.01 - 0. 17 _+ 0.02 0.01 0.04 _+ 0.02
10 2.82 _+ 0.17 1.68 _+ 0.30 0.31 1.76 +_ 0.15
20 19.97 _+ 0.42 4.96 _+ 0.74 0.54 10.04 +_ 1.07
30 24.74 _+ 0.34 5.31 _+ 0.60 0.52 12.01 _+ 0.66
40 32.03 +_ 0.26 5.59 _+ 0.45 0.37 15.03 _+ 1.49
50 32.02 _+ 0.15 9.65 +_ 0.26 0.15 15.58 _+ 0.88
K30 5 1.26 _+ 0.08 - 0. 52 _+ 0.13 0.16 0.59 +_ 0.14
10 2.41 _+ 0.16 3.87 + 0.29 0.27 2.00 _+ 0.43
20 19.94 _+ 0.42 5.38 _+ 0.37 0.23 9.72 _+ 0.53
30 24.19 +_ 0.35 7.85 +_ 0.61 0.59 12.35 +_ 2.58
40 39.44 _+ 0.21 1.67 _+ 0.37 0.28 17.33 +_ 0.42
50 38.08 _+ 0.18 4.60 + 0.32 0.26 17.10 +_ 1.68
K50 5 4.47 + 0.16 - 3. 80 _+ 0.28 - 0 . 1 4 1.06 + 0.33
10 4.99 + 0.11 0.79 _+ 0.20 0.11 2.35 + 0.25
20 21.21 + 0.30 3.07 + 0.51 0.57 10.19 + 0.53
30 27.87 + 0.48 0.93 +_ 0.84 0.71 12.71 + 0.42
40 34.53 + 0.32 6.86 + 0.56 0.68 16.49 _+ 0.71
50 31.65 + 0.64 20.73 _+ 1.11 0.79 17.92 + 0.57
p < 0.0001.
h Not significant.
L.S. Cheong Wan et al. / International Journal of Pharmaceutics 116 (1995) 159-168 165
3 0
2 5
Mode l
- v - A c t u a l
- -&- C a s e I
I - C a s e I I
c o u p l e d C a s e I - C a s e I I
20
10
y ,
. A
w , L
~ t A

A . '
A. v
v
O I I I I
0 1 0 2 0 3 0 4 0 5 0 6 0
T i m e ( r a i n )
Fig. 1. A typical gr aph showing the goodness-of-fit of vari ous
model s compar ed to the actual swelling dat a of HPMC K4
compact at pH 7.2.
3 0
Per cent H P M C
50
0 4 O
2 5
~ 3 0
- ~ - 2 0
+ 1 0
2 0 - * 5
1 5
/
1 0 / ~ e ~ D ~ e
0 1 0 2 0 3 0 4 0 5 0 6 0
T i m e ( m i n )
Fig. 2. Effect of variation of amount s of HPMC K4 on the
swelling of HPMC- i bupr of en matrices at pH 7.2.
Table 7
Swelling profiles of HPMC- pr opr anol ol mat ri ces cont ai ni ng varying amount s of HPMC in distilled wat er
Viscosity grade Percent HPMC a s ( X 10- 1) ( %/ ~/ S) fls ( 10- 2) ( %/ s ) y-i nt ercept c27 ~30 mi~
K4 5 9.68 + 0.38 - 4 . 7 9 _+ 0.26 - 0. 01 0.00 + 0.00
10 1.40 + 0.17 - 0. 20 _+ 0.03 3.52 5.56 + 0.89
25 2.47 + 0.04 - 0. 14 _+ 0.01 - 1.06 10.14 + 2.28
50 4.59 + 0.10 - 0. 17 + 0.02 - 1.25 15.37 +_ 2.35
75 1.97 _+ 0.31 0.66 + 0.06 - 2.06 19.25 +_ 0.89
K15 5 2.38 + 0.42 - 1.36 + 0.34 - 0. 28 0.00 _+ 0.00
10 1.84 + 0.07 - 0. 32 _+ 0.02 - 0. 22 0.91 _+ 1.57
25 3.35 + 0.03 - 0. 23 + 0.00 - 0. 27 10.14 + 2.28
50 5.05 + 0.28 - 0. 11 _+ 0.05 - 2. 91 17.27 + 1.12
75 4.83 + 0.17 0.26 _+ 0.03 - 2. 29 23.68 _+ 1.00
K30 5 8.13 _+ 0.72 - 5. 64 _+ 0.59 0.10 1.87 _+ 1.96
10 3.10 _+ 0.22 - 0. 60 _+ 0.06 1.81 2.31 +_ 3.83
25 4.28 _+ 0.53 - 0. 36 + 0.09 0.52 11.95 +_ 1.78
50 6.56 _+ 0.06 - 0. 28 _+ 0.01 - 0. 99 21.94 + 0.33
75 4.53 _+ 0.15 0.26 _+ 0.03 - 1.25 27.07 _+ 2.29
K50 5 2.53 _+ 0.95 0.10 + 0.90 - 0 . 3 4 0.00 + 0.00
10 5.83 + 0.37 - 1.66 ___0.14 - 0. 09 2.31 + 2.73
25 4.22 +_ 0.07 - 0 . 3 0 + 0.01 1.38 13.67 + 0.12
50 6.21 _+ 0.11 - 0. 13 _+ 0.02 - 0. 63 23.51 +_ 0.46
75 1.44 _+ 0.30 0.85 _+ 0.05 - 1.76 20.99 _+ 2.62
K100 5 4.03 _+ 1.33 - 0. 49 +_ 1.27 - 0. 49 0.00 +_ 0.00
10 4.18 +_ 0.23 - 0 . 8 0 _+ 0.07 - 0. 16 3.84 _+ 2.34
25 4.50 +_ 0.11 - 0 . 3 4 +_ 0.02 2.04 14.57 +_ 0.12
50 4.80 _+ 0.26 0.09 _+ 0.04 - 2. 97 19.77 + 0.20
75 4.97 +_ 0.23 0.47 0.04 - 2.88 27.60 + 0.66
166 L. S. Cheong Wan et al. / International Journal of Pharmaceutics 116 (1995) 159-168
[ _ ~ 1 0 0 -
7 5 -
"~ s 0 -
" 2 5 -
1 ~ 1 0 0 .
i i l s l t y
( x l O 3 c p $ )
5 . 0 0
1 0 0 . 0 0
7 6 . 2 5 . - " " "
5 2 . 5 0 .
u
~ . 6 , 5 5
. . . . . . . . . .
- . . . . 3 0 . 7 7
2 8 . 7 5 , ~ - - - - - - - - - - ~ 2 4 . 9 9 " " "
. . . . 1 3 . 4 ~ 3 ~ ~ ' ' ' ' ~
. . . . ; . 65 . . . . . . . . . . . 4
1 ~ 8 7 -1
5 . 0 0
4 . 0 1 5 . 5 2 7 . 0 3 1 3 . 5 5 0 . 0
Viscosity ( x 1 0 3 c p s )
Fig. 3. Sur f ace and cont our pl ot s depi ct i ng t he effect of
concent r at i on and vi scosi t y of HP MC on t he a s coeffi ci ent of
HP MC i bupr of en mat r i ces at p H 7.2.
in distilled wat er was st udi ed. Swelling of mat ri -
ces was f ast er dur i ng t he first few mi nut es f or
mat r i ces cont ai ni ng a low pol ymer cont ent ( 5-
10%). Thi s was due to a f ast er hydr at i on of t hese
mat r i ces wi t h abundant soluble drug. When t he
swelling meas ur ement was al l owed t o ext end over
a l onger per i od, mat ri ces cont ai ni ng a low HPMC
level swelled t o a maxi mum thickness. Thi s was
fol l owed by a per i od of const ant swelling r at e
bef or e a r educt i on in t he 6 val ues was achi eved.
The swollen mass di mi ni shed in t hi ckness and
slowly col l apsed. Thi s observat i on was of t en seen
in mat r i ces with pol ymer cont ent < 10% and can
be at t r i but ed to di ssol ut i on of t he wat er-sol ubl e
drug. The t hi ckness of t he swollen l ayer ar ound
t he mat ri x cor e vari ed wi t h t he pol ymer cont ent
of t he mat ri x systems. The per cent swelling at 30
min, 830 min was used to compar e t he i nfl uence of
HPMC concent r at i on and viscosity gr ade on t he
swelling behavi our of HPMC- pr opr anol ol mat ri -
ces ( Tabl e 7). General l y, 830 min i ncr eased with
t he concent r at i on of pol ymer in t he formul at i ons.
The amount of gel at i nous l ayer f or med also var-
ied with t he viscosity gr ade of HPMC used, gen-
eral l y a t hi cker l ayer of swollen mass was ob-
t ai ned with mat ri ces cont ai ni ng similar HPMC
cont ent of hi gher viscosity grade.
The meas ur ement of mat ri x swelling involved
t he use of a gauge to r ecor d t he vert i cal i ncrease
in t he t hi ckness of t he swollen mass. In r esponse
t o t he swelling pr essur e of t he mat ri x with time,
t he compr essi on of t he tip of t he gauge woul d
resul t in a const ant back pr essur e (1.47 N) on t he
swelling matrix. I f t he upwar d swelling f or ce of
t he mat ri x wer e low, t he opposi ng back pr essur e
woul d resul t in a l ower readi ng. Thi s pr oduced
negat i ve val ues of Case II coeffi ci ent s due t o an
under est i mat i on of t he ext ent of mat ri x swelling.
In mat ri ces with low HPMC cont ent , l eachi ng of
dr ug weakened t he swollen mass. As t he mat ri x
swelled, t he dr ug- depl et ed swollen mass in-
cr eased in vol ume and t he swollen st r uct ur e
weakened. Hydr at i on of t he pol ymer caused a
viscous gel at i nous l ayer t o be f or med ar ound t he
matrix. Thi s coul d act as a shield t hat r egul at ed
liquid ent r y into t he matrix. The mat ri x cor e was
not wet t ed and r emai ned dry. The swollen pe-
r i pher al l ayer and t he i nner dry cor e wer e dis-
tinctly separ at e under light mi croscopi c exami na-
tion.
Two opposi ng forces occur in t he mat ri x when
pol ymer swells. The adhesi ve f or ce of HPMC
bi nds t he mat ri x t oget her . As t he part i cl e swells,
t he mat ri x exper i ences i nt r amat r i x swelling f or ce
pr omot i ng di si nt egrat i on. The amount of gel-for-
mi ng subst ance was small in mat ri ces with a low
pol ymer cont ent ( < 10%). On t he ot her hand,
t hese mat r i ces had a l arge amount of highly solu-
ble drug ( > 90%). A cont i nuous HPMC gel was
not f or med and pocket s of pol ymer mass wer e
f ound wi t hi n t he matrix. Di ssol ut i on of dr ug cre-
at ed void spaces wi t hi n t he swollen por t i on of t he
matrix. Thi s f ur t her weakened t he gel st ruct ure.
In t he meant i me, wat er cont i nued to ent er t he
matrix. When t he wat er-l ogged mat ri x coul d not
L.S. Cheong Wan et al. / International Journal of Pharmaceutics 116 (1995) 159-168 167
suppor t t he amount of fluid in t he matrix, t he
swollen mass col l apsed due t o t he small back
pr essur e exer t ed by t he gauge. Thi s obser vat i on
was of t en seen in mat r i ces cont ai ni ng t he sol ubl e
drug, pr opr anol ol or wi t h low HPMC cont ent
( 5- 10% HPMC in HPMC- i bupr of en mat ri ces).
In t he initial stage ( < 5 rain), HPMC- pr o-
pr anol ol mat r i ces wi t h > 25% pol ymer swelled
slower t han t hose with a l ower pol ymer cont ent .
A high amount of pr opr anol ol i mpr oved t he sur-
face wet t i ng of t he matrix. Thi s di f f er ence in t he
swelling profi l es can be cont r ast ed by compar i ng
with t he swelling profi l es of HPMC- i bupr of en
mat ri ces.
[ ~ 2 0 -
o ~ 1 5 -
o
-g
~ 0
~ ~ ; . . ~ -52
5 7 . 5 2 5 V i s c o s i t y ( x 1 0 3 c p s )
4 0 . 0
P e r c e n t H P M C 5 ~ 4
f
J
5 7 . 5 -
' 5 , g 3
4 0 . 0 -
2 2 . 5 -
. _ j .
. - ~ ~ ~ ~, 8 . 1 4 j
" " " / f 9 . 2 9 X __ ~ ~
5 0 , - 2 ' 0 s ' 2 r ' 6 1 0 0
V i s c o s i t y ( x l O 3 c p s )
Fig. 4. Surface and cont our plots depicting t he effect of
concent rat i on and viscosity of HPMC on t he a S coefficient of
HPMC- pr opr anol ol matrices.
The swelling charact eri st i cs of HPMC- pr o-
pr anol ol mat r i ces coul d be descr i bed by a pol yno-
mial equat i on (Eq. 3). The coeffi ci ent s of bot h
Case I and Case II mechani sms general l y in-
cr eased when t he concent r at i on a n d / o r viscosity
gr ade of HPMC wer e i ncr eased ( Tabl e 7). Unl i ke
HPMC- i bupr of en mat ri ces, t he coeffi ci ent s of t he
Case I mechani sm of HPMC- pr opr anol ol mat ri -
ces did not i ncrease pr opor t i onal l y t o t he HPMC
cont ent but exhi bi t ed a mi ni mum- maxi mum rela-
t i onshi p (Fig. 4). Thi s was due to t he di f f er ence
in dr ug solubilities of t he mat ri ces. The st r uct ur e
of t he swollen mass of HPMC- i bupr of en mat ri ces
was mor e rigid and f i r mer t han t hat of HPMC-
pr opr anol ol mat ri ces. Bei ng a highly soluble drug,
mor e pr opr anol ol l eached out of t he swollen mass
compar ed t o t he poorl y wat er-sol ubl e drug,
i buprofen. Leachi ng of dr ug left behi nd a highly
por ous and weak mat ri x st ruct ure. The back pres-
sure exer t ed by t he gauge on t he swollen mass
compr essed t he swollen mass. In t he cases of
mat ri ces with 5% pol ymer, t he amount of swollen
mass f or med was small. Swelling of HPMC in t he
mat ri ces was not summat i ve. Isol at ed HPMC gel
pocket s f or med as HPMC swelled wer e responsi -
bl e f or t he very high a S coeffi ci ent s obt ai ned (Fig.
4). A r at her l i near por t i on of t he gr aph bet ween
10 and 25% i ndi cat es t he effect of i ncreasi ng
pol ymer cont ent on t he Case I swelling coeffi -
cients. In mat r i ces with high pol ymer concent r a-
t i on ( > 25%), t he highly por ous swollen mass was
severel y compr essed and t he t hi ckness of t he
swollen l ayer was r educed. The t ort uosi t y of t he
mat ri x was i ncreased. The r at e of Case I di ffusi on
of liquid i nt o t he mat ri x was t her ef or e r educed.
4. Concl usi ons
The swelling charact eri st i cs of HPMC mat ri ces
wer e descr i bed by a pol ynomi al equat i on. The
coeffi ci ent s of bot h Case I and Case II mecha-
nisms general l y i ncr eased when t he concent r at i on
a n d / o r viscosity gr ade of HPMC wer e i ncreased.
For HPMC- i bupr of en mat ri ces, t he swelling rat es
at t r i but ed t o t he Case I mechani sm i ncr eased
sharpl y when t he HPMC concent r at i on was var-
ied f r om 5 t o 50% of t ot al wei ght cont ent . The
168 L.S. Cheong Wan et al. / International Journal of Pharmaceut ics 116 (1995) 159-168
effect of HPMC concent rat i on on swelling rates
was less marked at higher polymer cont ent ( >
50% HPMC of high viscosity grade). A saturation
state was attained beyond 40% HPMC cont ent of
these matrices. Unlike HPMC-i buprofen matri-
ces, the coefficients of t he Case I mechanism of
HPMC-propranol ol matrices did not increase in
proport i on to HPMC cont ent but exhibited a
minimum-maximum relationship. This was due
the difference in drug solubilities in t he matrices.
References
Faroongsarng, D. and Peck, G.E., The swelling of core tablets
duri ng aqueous coatings: I. A simple model describing
extent of swelling and wat er penet rat i on for insoluble
t abl et s containing a superdisintegrant. Drug Dev. Ind.
Pharm., 17 (1991) 2439-2455.
Faroongsarng, D. and Peck, G.E., The swelling of core tablets
during aqueous coating: II. An application of t he model
describing extent of swelling and wat er penet rat i on for
insoluble tablets. Drug Dec. Ind. Pharm., 18 (1992) 1527.
Mitchell, K., Ford, J.L., Roston, C., Armstrong, D.J., Elliot,
P.N.C. and Hogan, J.E., Swelling behavi our of cellulose
et her matrix tablets. J. Pharm. Pharmacol., 43 (1991) 76P.
Peppas, N.A. and Sahlin, J.J., A simple equat i on for t he
description of solution release: III. Coupling of diffusion
and relaxation. Int. J. Pharm., 57 (1989) 169-172.
Wan, L.S.C., Heng, P.W.S. and Wong, L.F., Effect of addi-
tives on liquid upt ake into HPMC matrices. Sci. Tech.
Pharm. Pharm. Sci. (France), (1993c) in press.
Wan, L.S.C., Heng, P.W.S. and Wong, L.F., Liquid penet ra-
tion into matrices. Sci. Tech. Pharrn. Pharrn. Sci. (France),
3 (1993b) 477-487.
Wan, L.S.C., Heng, P.W.S. and Wong, L.F., Relationship
bet ween swelling and drug release in a hydrophilic matrix.
Drug Dev. Ind. Pharm., 19 (1993a) 1201-1210.
West man, L. and Lindstrom, T., Swelling and mechanical
properties of cellulose hydrogels: I. Preparat i on, charac-
terisation and swelling behaviour. J. AppL Polym. Sci., 26
(1981a) 2519-2532.
Westman, L. and Lindstrom, T., Swelling and mechanical
properties of cellulose hydrogels: II. The relation bet ween
the degree of swelling and creep compliance. J. AppL
Polym. Sci., 26 (1981b) 2533-2544.
West man, L. and Lindstrom, T., Swelling and mechanical
properties of cellulose hydrogels: III. Temperat ure effects
on t he swelling and compliance levels studied by dilatome-
try and 1H-NMR spectroscopy. J. Appl. Polym. Sci., 26
(1981c) 2545-2560.
Westman, L. and Lindstrom, T., Swelling and mechanical
properties of cellulose hydrogels: IV. Kinetics of swelling
in liquid water. J. Appl. Polym. Sci., 26 (1981d) 2561-2572.

Vous aimerez peut-être aussi