WORKING WITH HECKEL PLOTS: METHOD FOR CHECKING THE PLAUSIBILITY OF FORCE DISPLACEMENT DATAa b b a aK. Duchatsch , J. Falconet , G. Tardy , R.F. Lammens , B. Fretter

a Department of Pharmaceutical Technology, University of Bonn, Gerhard-Domagk-Str. 3, 53121 Bonn, Germany b Medelpharm, 43, Porte du grand Lyon, 01700 Neyron, France

I R D INT O UCT ON:

M E ALS & E HO :AT RI M T DS

M t ria sa e l

M t dse ho

Fo cha c e isi g wde mpres beha i ur a d de o m tion e n sm f at r l f n er ra t r n po r co sion v o n f r a m cha i s o m e ia s o te th He ke l t is use I scr b s the o u e duction f a wder d d ing pre o by pl ttingc l P o d. t de i e v l m re o po be ur com ssi n o l (1/e) v rs s pre su e. He b , e cti ns o he ke l t des ibi g t lo di g ph se presn e u s r re y th se o f t Hec l P o cr n he a n a re ent the def rm tio be a ur o t w r. h lue o th o very n itive t sm l e ro s n the in-o a n h vio f he po de T e va f e sl pe is se s o a l r r i di densi y ter i at o hi r ta ting are a du to la k f a u a o ee t de m n i n w ch du ing ble , m inly e c o cc r cy f th

1[ ]di c m n nsdu e the re e n e positi ns a d pu ting . To inim se he er o aspla e e t tra c rs, f re c o n nch til etc m i t se r rs, n a curat xter l ig m a r m syste s esta ishe a t pa o sim tor t lc mc e e na he ht- e su e ent m wa bl d t he com cti n ula S y a 200R r racte sing po de un r o di ns o o pr du tion e Stylc m 200R s h se. Fo cha ri w r de c n tio cl se t o c , th a wa c o n be use f i bil to m l te a eti g spe li the on s pre ail o ri us produ tio bleca o ts a ity si u a t bl n eds ke e v ing n va o c n ta t pre es. ur e m r it w s a so q p d w th i f o v p T ea th dv nta ess F th r o e, a l e ui pe i a pa r o c nca e unches. his l ds to e a a g tha capp ng a lam ting mig be de ted earl i dev l pm n B ca ck l ts t o lyt i nd ina ht tec y n e o e t. e use He el P o s r ng depe n e m a m n a cura y, a pl u ibil t k o cquire da s v loped.nding o th e sure e t c c a s i y chec f the a d ta wa de e

! o pa tio im to Sty a 200R (Mede ph r , Fra ) si u a i f re ar sse aC m c n s ula r lc m l a m nce m l ting d f e nt rot y pre s t va o s spe ds, i str m nted w th d el nd rna n p a em nt nsduce i .ri u e n u e i loa c ls a inte l i ductive dis l c e tra rs (F g 1+ ta a ition a a alysi ft r n l (M lpha m a c2), da cquis nd n ng so wa e: A a is ede r , Fr n e)! pu s: 9 m c nca e R m f a , 28 m fla l EuBnche m o v 15, 10 m l t 11. m t, a l ! exte n l di c m n m a m n s m (F g c e di a m n tr n du er mp ifie ,r a spla e e t e sure e t yste i . 3): 4 indu tiv spl ce e t a s c s, a l r

[ ]2ll h ei r, rma y), v ra l a curacy 10 µ da acquisi n sof are: D Q4 (Hucke o twa , (a Sc r be Ge n o e l c m , ta tio tw A S f ree m ny), th r a t m c n be c mp d n - nch TSMG r a e exte n l sys e a la e to a y B type pu (Eu or )

! re e n e ta ts: m - pe m ta t bl o f a unches h g l n f ed Av ce t bl l f re c ble do e sha d e l a et f r l t p ; i h y de si i i l a et with equara u s e c n a e pu s secup di s a th o c v nche to be u d

De r i ng ref r n e po i ote m ni e e c s ti ns:Of op d e of the die is ta n s e r e nce po n altho gh the abs lute po tio o upp r a d ten, the t e g ke a th ef re i t u o si n f e nlo r pu i re v nt or th n di ble hei ht e sure e t. here o e, on y t sta c tw e we nch s ir le a f e i - e ta t g m a m n T f r l he di n e be e n

the punches is determined by applying small forces (ca. 100 N) on a reference tablet, three times before and after a set of experiments.

Measuring reference tablet height:Before and after establishing the reference positions, the height of the reference tablet (metal as well as Avicel) is measured with a linear gage having an accuracy of 3 µm.

Considering the elastic deformation:To calculate the exact tablet height under pressure the elastic deformation of punches and machine parts must be taken in account. For the external displacement measurement system, the deformation between punch tip and mounting clamp of the transducers is calculated from Hooke's law and punch

[3]geometry. For the internal displacement measurement system, a calculation is not possible and therefore, the deformation has to be determined experimentally.

Measuring the Avicel reference tablet height turn out to be within an accuracy of 5 µm which is little larger than the accuracy of 3 µm of the linear gage. The reasons therefore are the difficulty of positioning a concave tablet exactly vertically under the contact tip of the linear gage without tilting as well as the roughness of the tablet surface. The reproducibility of repeatedly determining the reference positions for the in-die powder bed height measurement without removing or changing the position of the reference tablet turn out to be within 5 µm. Establishing reference positions before and after a series of experiments is reproducible within 10 µm.

RESULTS:

Reproducibility of reference data

Fig. 4: inconsistent Heckel Plots of Avicel Ph102; 10mm flat punches

Fig.1: The compaction simulator Stylcam 200R

upper punch

lower punch

dietable

displacementtransducer

plunger

clamp

Fig.3: measurement setup of the external displacement transducers

Method for Checking the Plausibility

When working with Heckel Plots, small errors can have a large influence on the results. An easy and fast method for checking the plausibility of displacement data is examining the curve progression of Heckel Plots at different maximum compaction pressures. Those sections of the Heckel Plot describing the loading phase should coincide because at the same powder density the pressure should prevail as long as the influence of tableting speed is moderate. Only near the reversal point of the punch movement, the curves start to deviate due to relaxation because of low densification speeds. Fig. 4 shows an example of non coinciding curves due to a change of the mounting positions of the external displacement transducer clamps which results in different reference positions before and after the experiment. This error would not be recognised when examining just one Heckel Plot.In Fig. 5 a consistent set of Heckel Plots is shown with congruent sections describing the loading phase. These data are obtained with the concave punches under conditions close to production concerning the velocity of the simulated tablet press (effective dwell time of 7.5 ms).But consistent sets of Heckel Plots are only indicating correct measurement data. Also the shape of the Heckel curves must be considered.

Fig. 6: consistent but false measurement data of Technische Tablettiermischung, 11,28mm flat punches

Fi 6 g v an xampl of c si e t u alsg. i es e e on st n , b t f e ot e e n r s co i pl s of r lativ de sity ve su mpact onr ssu e T e r e u d l t e p e r . h blue cu v s wo l ead o th

c l ion at n t t ower p anonc us th de si y ge s l u on cr f or h c se se, an t thin ease o f ce, w i h is non n d hat e e sity m m i e at pr x. p d n axi um s obtain d ap o 30 er

c n h e r axi u , t ie t of t e pr ssu e m m m. Usually h s n m is e w t h f s ede sity maximu locat d i hin t e ir t 5 p r

[4] ter h r ssu e maxim (c . ) Th cent af t e p e r um f Fig. 5 . e

r o or e e i g t f i u at neas n f r c iv n he alse values s calc l i g i h e f rm . t r g tw t th wrong de o ation De e minin he fo o e x l is l c m nde rmati n for th e terna d p a e e t

m asu e e t st m s eas th f e t r ale r m n sy e i ier an or th in e n on n Fi . th e ar o H e P ote. I g 7 e r levant p t f two eck l l s

i d im eou y h e an h acqu re s ultan sl with t e int rnal d t ee terna i la mea u m yx l d sp cement s re ent s stem sh w or an e. i tino s good acc d c Th s is a good star g

n sin t e n e n syst m, ich s poi t for u g h i t r al e wh isi t e tu tt on e f e out de h ac al table ing z e and th re or

e r e. c oser inv on t easy to ope at A l estigation h i n t o di pl cementd ffere ces be ween the tw s a

u e e t d a v at on h meas r m n systems an alid i of t ei r al on ar ar r d o ente n e p ticul ly with egar t th

c w ll e h e f tu eac uracy i b made in t e n ar u r .

p e o t n exac H e P w tIt is ossibl to b ai t eck l lots i h co a e p n es u d c d s c tnc v u ch n er on ition lose o

od c e f iv l i e o mpr u tion ( f ect e dwel t m f 7.5 s).

C S SONCLU ION :

Determining the reference positions for the in-die height measurement with a highly densified tablet seems to be a good method although the precision can be optimised slightly for example by using reference tablets with a smooth surface. To check the plausibility of Heckel data, a set of experiments made at different maximum compaction pressures up to 400 to 600 MPa is strongly recommended. Moreover, the shape of each Heckel curve must be scrutinised. As obvious from the data, Heckel Plots are an error-prone method describing deformation behaviour of powders, due to the difficulties in measuring the in-die tablet height sufficiently accurately. Therefore, it might be difficult to evaluate literature data of Heckel Plots properly.First results show, that internal displacement transducers of the Stylcam mounted outside the tabletting zone can measure the in-die tablet height in good accordance to the external ones. Closer investigations on the comparability of the two systems are on-going. The external measurement configuration described in this paper has the advantage of being able to calculate the deformation from the punch geometry. It can be adapted for most single-punch tablet presses. The internal measurement setup of the Stylcam has the advantage of location and is therfore easy to use.

[1] Sonnergaard, J.M.; International Journal of Pharmaceutics 193(1), 63-71 (1999)[2] Fretter, B. et al; Tablet Tech, Brussels (2007)[3] Fretter, B. et al; 6th World Meeting on

Barcelona (2008)[4]

Schierstedt D., Müller F.; Pharm.Ind. 44 Nr.9 (1982), S.932-937

REFERENCES:

Pharmaceutics, Biopharmaceutics and Pharmaceutical Technology,

Fig. 7: comparing Heckel Plots of internal and external displacement measurement system, Technische Tablettiermischung, 11,28mm flat punches

FIP Pharmaceutical Sciences 2010 World Congress in Association with the AAPS Annual Meeting and Exposition; November 2010, New Orleans, USA

Fig.2: measurement setup of the internal displacement transducers

upper punch

lower punch

dietable

punchholder

punchholder

displacementtransducer

machinepart

Fig. 5: plausible Heckel Plots of Ca-Phosphate; 9mm concave punches