COMMENTARY

A Commentary on Scale-Up of Pan Coating Process Using

Microenvironmental Control
PREETANSHU PANDEY, DILBIR S. BINDRA

Drug Product Science and Technology, Bristol-Myers Squibb, New Brunswick, New Jersey 08901

Received 30 June 2014; revised 3 September 2014; accepted 9 September 2014

Published online 25 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.24191

ABSTRACT: Although significant progress had been made in developing general scale-up rules for an aqueous pan-coating process, there
are often scenarios where small-scale experiments are not found to be truly reflective of what may be observed at the large scale. This
article reviews some of the methods traditionally used for scale-up, identifies the gaps associated with the traditional scale-up rules, and
provides a perspective on a new real-time process monitoring tool that is capable of providing thermodynamic changes taking place in
the microenvironment of the substrate being coated. This tool has been used to ensure increased success during scale-up by maintaining
environmentally equivalent conditions between the processes, especially for systems that are sensitive to small thermodynamic changes.
C 2014 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci 103:34123415, 2014

Keywords: coating; scale-up; microenvironment; EEF factor; unit operations; quality by design (QbD); process analytical technology
(PAT); processing; solid dosage form

INTRODUCTION regulated. Additionally, the exhaust air %RH is not controlled

or monitored. Clearly, there is incomplete characterization of
A pan coater consists of a rotating perforated pan equipped
the process thermodynamics, even at the macroscopic level. In-
with spray gun(s) (Fig. 1). The spray gun system is generally
adequate control of the inlet air %RH has been shown to have
a two-fluid nozzle system where the coating suspension is at-
a negative effect on the quality of the coated product, with logo
omized into fine droplets via an air stream (atomization air).
bridging tablet defect observed at a high inlet air %RH.5
These atomized droplets travel through a preheated air stream
An approach for scaling up the coating process that is most
(inlet air) and impinge on the cascading layer of tablets. These
routinely used is to utilize established scale-up rules on in-
droplets coalesce, spread on the tablet surface, dry and form a
dividual process parameters.69 The process parameters that
thin film on the surface of the tablet.
play an important role in defining the coating process ther-
The thermodynamics of the coating process is governed
modynamics include inlet and exhaust air conditions, spray
largely by the factors such as inlet air conditions [temperature,
rate, airflow rate, gun-to-bed distance, atomization, and pat-
relative humidity (RH)], inlet airflow rate, gun-to-bed distance,
tern air pressure. Other factors such as pan speed (mixing-
heat losses, and spray conditions such as spray rate, droplet vis-
related) and batch size are important in defining the overall
cosity, droplet size, and so on. The thermodynamic conditions
coating operation but do not have a direct role in dictating
inside a coating pan can affect the quality of the film formed
the process thermodynamics. As discussed before, the exhaust
on the tablets. For example, a relatively dry environment can
air temperature is assumed to be reflective of the tablet-bed
result in tablet defects such as surface roughness, whereas a
temperature and is therefore maintained constant during scale
wet environment can cause tablets to stick or pick. Appearance
up. The inlet air temperature is adjusted in order to reach
and residual water content of coated tablets have been shown
the desired exhaust temperature. Spray rate plays a signifi-
to be a function of coating conditions.14
cant role in defining the thermodynamics of the overall coating
Traditionally, the coating process is controlled by maintain-
process. Spray rate is especially important if the drug prod-
ing the exhaust air temperature at a predetermined set point
uct is sensitive or unstable under high moisture conditions, as
by adjusting the inlet air temperature once other independent
residual water content in coated tablets can be directly related
variables such as spray rate and inlet air volume are fixed. The
to spray rate.1 At fixed formulation factors, the spray rate and
tablet-bed temperature (T) is assumed to be 2 C3 C lower than
the gun-to-bed distance affect the tablet-bed microenvironment
the exhaust temperature, as was measured in certain studies
significantly, which can impact the final coated drug product
by pointing an IR gun on the surface of the tablet bed (just
quality.10 During scale-up, it is desired to keep the spray rate
outside of the spray zone). The inlet air RH (%RH) or dew
as high as possible at the larger scale so as to reduce the pro-
point is often maintained below a certain set-point in a cGMP
cess cycle times. In order to compensate for the increased spray
(current Good Manufacturing Practice) manufacturing setting.
rate, the inlet airflow is increased, such that the ratio of in-
However, in most laboratory and pilot-scale settings, the inlet
let airflow to spray rate (drying capacity) is maintained con-
air %RH is not controlled or monitored. The inlet air is gen-
stant across scales. Maintaining drying capacity similar across
erally obtained from the process room, which may not be well
scales can be effective in keeping the macroscopic environment
across scales similar; however, this may not be the case at the
Correspondence to: Preetanshu Pandey (Telephone: +7322275918;
Fax: +7322275150; E-mail: preetanshu@gmail.com) tablet-level (microscopic).2 Alternatively, the inlet air flow is
Journal of Pharmaceutical Sciences, Vol. 103, 34123415 (2014) determined first based on recommendations by the manufac-

C 2014 Wiley Periodicals, Inc. and the American Pharmacists Association turer and equipment limitations.9 Once inlet airflow is fixed,

3412 Pandey and Bindra, JOURNAL OF PHARMACEUTICAL SCIENCES 103:34123415, 2014


Figure 1. Schematic representation of a pan coater. Also shown in
this figure are PyroButtonR data loggers that are fixed at different
locations inside the coater and also allowed to freely move with the
tablets to enable measurement of the microenvironment experienced
by the tablets during the coating process (reproduced with permission
from John Wiley and Sons).5
the spray rate is determined by keeping the ratio of inlet air-
flow and spray rate constant across scales. There are changes
in nozzle sizes and sometimes even spray gun manufacturer
during scale up; however, the aim should be to match the aver-
age droplet size coming out of the spray nozzle by making spray
rate and atomization air pressure adjustments. The atomiza-
tion and pattern air pressures or flow rates are scaled based on
maintaining droplet size distribution similar across scales.11,12
In absence of droplet size data, as a general rule of thumb, the
ratio of spray rate to atomization air flow rate, and atomization
air to pattern air ratio can be kept constant across scales.
It is worthwhile to note that all the scale-up factors dis-
cussed here are macroscopic factors and do not quantify the
changes experienced by the tablets at the microscopic level as
they get coated. The fundamental assumption is that changes
to the macroscopic parameters at different scales have sim-
ilar, if not the same, effect on the microenvironment experi-
enced by the tablets. For example, the exhaust air tempera-
ture is maintained constant across scales during scale-up with
the underlying assumption that by doing so the tablet-bed
temperature is maintained constant across scales. Although
this may turn out to be a reasonable assumption at certain
process conditions, this cannot be generalized. Recent studies
have clearly shown that the difference between the exhaust
temperature and tablet-bed temperature is a function of the
processing conditions, with larger deviation observed between
the two for wetter process conditions. The difference between
the exhaust and the tablet-bed temperature can, in certain
cases, be larger than 10 C depending on the process conditions
and also on the exhaust temperature measurement location in
the coater.2,13 The difference between the exhaust RH and the
tablet-bed RH (or the RH that the tablets experience) can be
higher than 20%.10,14 Such differences between the exhaust air
conditions and the conditions experienced by the tablets high-
light the importance of characterizing the conditions at the mi-
DOI 10.1002/jps.24191
COMMENTARY 3413
croscopic level (tablet microenvironment). This understanding
would be critical for drug products that are sensitive to heat and
moisture.
Mathematical models using heat and mass transfer equa-
tions have been developed to describe the coating process.7,1518
Ebey15 introduced the concept of using a single value, envi-
ronmental equivalency factor (EEF), to describe the thermo-
dynamic state of the coating process, with a high EEF value
indicating a dry process. EEF number is often used for under-
standing the effect of changing process parameters at a partic-
ular scale and also for scale-up purposes such that it is kept
constant during scale up. EEF value accounts for several pro-
cess parameters including inlet air temperature and RH con-
ditions and also provides the predicted exhaust air T and RH
values. However, this approach considers the coating process as
a black-box and the factors such as gun-to-bed distance, which
are well-known to have a significant role in the process,10 do not
appear in EEF calculations. Additionally, the heat losses dur-
ing the coating process are generally not accounted for in EFF
calculations, and therefore, the models tend to deviate from ex-
perimental results when used across scales, where heat losses
may be different. Some of the newer thermodynamic models
have improved upon Ebeys EEF model and accounted for heat
losses in their calculations.16,17,19 However, the approach is still
largely macroscopic in nature and such models do not directly
address the localized conditions experienced by the tablets. Al-
though the United States Food and Drug Administrations em-
phasis on process analytical technologies and quality by design
stress the importance of a thorough understanding of processes,
such a detailed knowledge of the coating process is still incom-
plete, especially in relation to the thermodynamic conditions in
the coating bed.
Recent studies by the authors and coworkers have es-
tablished the utility of a new real-time monitoring tool
(PyroButtons ; Opulus Ltd., Philadelphia, Pennsylvania) that
R
can be used to quantify the microenvironment experienced by
the tablets during coating. PyroButtons are data logging de-
R
vices that record T and RH data in real-time. These tablet-sized
devices (16 mm diameter, 6 mm height) were secured at spe-
cific locations in the equipment (e.g., inlet, exhaust, spray gun
handle, baffles), and also placed in the tablet bed and allowed
to tumble freely along with the tablets (Fig. 1). PyroButtons R
that move with the tablets provide information on the ther-
modynamic conditions (microenvironment) experienced by the
tablets during the coating process. There have been a few case
studies published recently that show the utility of the informa-
tion provided by the moving PyroButtons , where correlations
R
could be established between the measured microenvironment
and the coated drug products critical quality attributes (CQAs)
such as appearance, physical, and chemical stability. These case
studies can be classified into different categories of drug prod-
uct CQAs:
1. Appearance (logo bridging tablet defect): As part of un-
derstanding this CQA, it was shown via a case study
that the macroscopic events such as logo bridging coat-
ing defect correlated best to tablet-bed RH when com-
pared with the conventionally-measured coating process
parameters.14 It was shown that a process design space
can be established based on the tablet-bed microenviron-
ment (RH) where no logo bridging was observed (Fig. 2). A
Pandey and Bindra, JOURNAL OF PHARMACEUTICAL SCIENCES 103:34123415, 2014
3414 COMMENTARY
Figure 2. Case study demonstrating a correlation between tablet-
bed microenvironmental RH to logo bridging coating defect (reproduced
with permission from John Wiley and Sons).14
critical tablet-bed RH (30%) was established below which
no logo bridging was observed. It should be noted that this
critical bed RH would be specific to the coating formula-
tion, logo design, and tablet core used in that study but
should be independent of coating scale. The difference
between the tablet-bed conditions and the exhaust air
conditions was found to be a function of the process con-
ditions, with more deviation observed at wetter condi-
tions. Additionally, the tablet-bed microenvironment was
shown to be a cumulative effect of various process pa-
rameters, and therefore can be used as a single param-
eter for control and understanding the coating process
thermodynamics.10
2. Physical stability (bilayer tablet delamination): In this
case study, it was shown that the delamination tendency
of a coated bilayer tablet when placed on accelerated sta-
bility was related to the microenvironmental conditions
experienced by the tablets during the coating process.20
A higher tablet-bed RH led to a higher tendency of the
bilayer tablet to show defects along the bilayer tablet in-
terface, as shown in Figure 3. Interestingly, this study
showed that the bilayer tablets exhibited a memory of
the coating process conditions even when they were dried
to a similar LOD (loss on drying) at the end of the coating
process and no physical differences (appearance) could be
observed at the end of the coating run.
3. Chemical stability (degradation profile of a moisture-
sensitive drug product): This case study showed a cor-
relation between the degradation (hydrolysis) rate of a
drug product on stability to the microenvironment expe-
rienced by the tablets during the coating process.21 The
water activity of the coated tablets at the end of the coat-
ing run was found to be linked to the tablet-bed microen-
vironmental conditions.
These studies demonstrated that monitoring the microen-
vironment during the coating process provides a deeper un-
derstanding and better control of the coating process and its
effects on the drug product CQAs. Some of these case studies
also included scale-up experiments. For example, in the bilayer
Pandey and Bindra, JOURNAL OF PHARMACEUTICAL SCIENCES 103:34123415, 2014
Figure 3. Case study showing a correlation between tablet-bed rela-
tive humidity and a bilayer tablet delamination defect rate when put
on accelerated stability (40 C/75% RH, 24 h).20
tablet delamination case study, it was observed that when the
process was scaled from 24 in. scale coater to 36 in. coater
using a constant drying capacity scale-up rule, the resultant
tablet-bed microenvironment was drier at the 36 in. scale.20
The tablet-bed RH was observed to be 37.2% at 24 in. scale and
28.9% at the 36 in. scale. Given that a drier microenvironment
was favorable in terms of delamination/hairline cracks, a mis-
match in microenvironment across scales was not considered
to be an issue. However, it does highlight the limitations of
the traditional scale-up rules. In another case study related to
the chemical stability of a moisture-sensitive drug product, it
was shown that the chemical stability was directly related to
the final moisture content of the coated tablet.21 It was shown
that the traditional scale-up rules, such as keeping constant
EEF value constant across scales, did not result in an equiv-
alent microenvironment and also resulted in a different drug
product chemical degradation profile due to differences in the
resultant tablet water content (Table 1).
Based on these case studies, it is believed that a good match
of microenvironment across scales is essential to enable suc-
cessful scale up, where the drug product CQAs can be repro-
duced across scales. Often the traditional scale-up rules may
work well-enough and provide good starting points, and a mi-
croenvironmental level of scrutiny may not be required for a
drug product that is robust to the thermodynamic conditions
experienced during the coating process. However, maintaining
the same thermodynamic environment (T and RH) that tablets
experience across scales would undoubtedly ensure a better
scale-up outcome in terms of drug product CQAs, and is critical
Table 1. Scale-Up of Process Using EEF Rule Versus Constant
Microenvironment Rule
Scale Coated Tablet Average
(Pan diameter) EEF Water Activity Tablet-Bed RH%
15 in. 2.78 0.28 30
24 in. 2.77 0.35 40
24 in. 3.22 0.28 30
Adapted from Kestur et al.21
DOI 10.1002/jps.24191

for sensitive drug products, and likely for functional coating
scenarios.
An assurance of successful scale-up and good reproducibility
of CQAs across scales would allow for more experimentation at
a small-scale in order to identify the critical process parameters
for a given drug product and establish a process design space,
which can be then scaled-up based on matching the tablet-bed
microenvironment.
CONCLUSIONS
PyroButton data loggers are an effective PAT tool which can
R
provide a meaningful thermodynamic signature of the coating
process that can be correlated to drug product CQAs. Addition-
ally, such measurements allow for establishing a new method-
ology of scaling up the coating process thermodynamics at a
higher level of scrutiny, where the microenvironment experi-
enced by the tablets is maintained across scales in order to
achieve good reproducibility of CQAs.
REFERENCES
1. Ruotsalainen M, Heinam aki
J, Guo H, Laitinen N, Yliruusi J. 2003.
A novel technique for imaging film coating defects in the film-core inter-
face and surface of coated tablets. Eur J Pharm Biopharm 56(3):381
388.
2. Bley O, Siepmann J, Bodmeier R. 2009. Protection of moisture-
sensitive drugs with aqueous polymer coatings: Importance of coating
and curing conditions. Int J Pharm 378(12):5965.
3. Faroongsarng D, Peck GE. 1991. The swelling of core tablets during
aqueous coating I: A simple model describing extent of swelling and
water penetration for insoluble tablets containing a superdisintegrant.
Drug Dev Ind Pharmacy 17(18):24392455.
4. Pourkavoos N, Peck GE. 1993. The effect of swelling characteristics of
superdisintegrants on the aqueous coating solution penetration into the
tablet matrix during the film coating process. Pharm Res 10(9):1363
1371.
5. Pandey P, Bindra DS. 2013. Real-time monitoring of thermodynamic
microenvironment in a pan coater. J Pharm Sci 102(2):336340.
6. Pandey P, Turton R, Joshi N, Hammerman E, Ergun J. 2006. Scale-
up of a pan-coating process. AAPS PharmSciTech 7(4):102.
7. Porter SC. 2011. Scale-up of film coating. In Pharmaceutical process
scale-up; Levin M, Ed. 3rd ed. New York: Informa Healthcare, pp 444
489.
DOI 10.1002/jps.24191
COMMENTARY 3415
8. Porter SC. 2012. Coating of pharmaceutical dosage forms. In Rem-
ington the science and practice of pharmacy; Allen LVea, Ed. 22nd ed.
London: Pharmaceutical Press, pp 977987.
9. Porter S, Sackett G, Liu L. 2009. Chapter 33Development, opti-
mization, and scale-up of process parameters: Pan coating. In Develop-
ing solid oral dosage forms; Qiu Y, Chen Y, Zhang GGZ, Liu L, Porter
WR, Eds. San Diego: Academic Press, pp 761805.
10. Pandey P, Bindra DS, Felton LA. 2014. Influence of process pa-
rameters on tablet bed microenvironmental factors during pan coating.
AAPS PharmSciTech 15(2):296305.
11. Chen W, Chang S-H, Kiang S, Early W, Paruchuri S, Desai D. 2008.
The measurement of spray quality for pan coating processes. J Pharm
Innov 3:314.
12. Wang J, Hemenway J, Chen W, Desai D, Varia S, et al. 2012.
An evaluation of process parameters to improve coating efficiency
of an active tablet film-coating process. Int J Pharm 427:163
169.
13. Wobker MS, Mehrotra A, Carter B. 2010. Use of commercial data
loggers to develop process understanding in pharmaceutical unit oper-
ations. J Pharm Innov 5(4):169180.
14. Pandey P, Ji J, Subramanian G, Gour S, Bindra DS. 2014.
Understanding the thermodynamic micro-environment inside a pan
coater using a data logging device. Drug Dev Ind Pharm 40(4):542
548.
15. Ebey GC. 1987. Thermodynamic model for aqueous film coating.
Pharm Technol 11(Apr):4050.
16. Ende MT, Berchielli A. 2005. A thermodynamic model for or-
ganic and aqueous tablet film coating. Pharm Dev Technol 10(Jan):47
58.
17. Page S, Baumann KH, Kleinebudde P. 2006. Mathematical model-
ing of an aqueous film coating process in a Bohle Lab-Coater, part 1:
Development of the model. AAPS PharmSciTech 7(2):E42.
18. Page S, Baumann KH, Kleinebudde P. 2006. Mathematical model-
ing of an aqueous film coating process in a Bohle Lab-Coater: Part 2:
Application of the model. AAPS PharmSciTech 7(2):E43.
19. Strong JC. 2009. Psychrometric analysis of the environmental
equivalency factor for aqueous tablet coating. AAPS PharmSciTech
10(1):303309.
20. Zacour BM, Pandey P, Subramanian G, Gao JZ, Nikfar F. 2014. Cor-
relating bilayer tablet delamination tendencies to micro-environmental
thermodynamic conditions during pan coating. Drug Dev Ind Pharm
40(6):829837.
21. Kestur U, Pandey P, Badawy S, Lin J, Desai D. 2014. Controlling
chemical stability of a moisture-sensitive drug product through mon-
itoring and identification of coating process microenvironment. Int J
Pharm: in press.
Pandey and Bindra, JOURNAL OF PHARMACEUTICAL SCIENCES 103:34123415, 2014