See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/51701306

Efficient throughput method for hygroscopicity

classification of active and inactive
pharmaceutical ingredients by water vapor
sorption analysis

Article in Pharmaceutical Development and Technology · October 2011

Impact Factor: 1.2 · DOI: 10.3109/10837450.2011.618947 · Source: PubMed

CITATIONS READS

10 157

3 authors, including:

Vaibhav Sihorkar
Dr. Reddy's Laboratories
18 PUBLICATIONS 501 CITATIONS

SEE PROFILE

Available from: Vaibhav Sihorkar

Retrieved on: 12 May 2016
 Pharmaceutical Development and Technology, 2013; 18(2): 348–358
© 2013 Informa Healthcare USA, Inc.
ISSN 1083-7450 print/ISSN 1097-9867 online
DOI: 10.3109/10837450.2011.618947

RESEARCH ARTICLE

Efficient throughput method for hygroscopicity classification

of active and inactive pharmaceutical ingredients by water
vapor sorption analysis
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13

Vasudha Murikipudi, Piyush Gupta, and Vaibhav Sihorkar

Pharmaceutical Development, Aurigene Discovery Technologies Limited, Bollaram Road, Miyapur, Hyderabad,
Andhra Pradesh, India

Abstract
The conventional method of hygroscopicity determination proposed by Callahan and co-workers utilizes more
sample and time, may not be precise in all the cases, and is a relatively broader classification system. The method of
indicating degree of hygroscopicity as per European Pharmacopoeia considers equilibration of sample for 24 hours
under single humidity condition and may not necessarily ensure equilibration in all the cases. Additionally, both
these methods do not provide information on solid state changes occurring within the sample during the course
of experiment. This research work envisages an efficient throughput method for hygroscopicity determination, and
For personal use only.

validates it with active and inactive pharmaceutical ingredients using sorption analysis. Further, this method has
been performed under optimal equilibration conditions, in a throughput manner (consuming less sample and time),
with additional information on solid state changes occurring within the experimental conditions. This throughput
method would be a valuable tool for hygroscopicity assessment of new chemical entities, during drug development
in particular, and across all pharmaceutical materials in general.
Keywords:  hygroscopicity, water vapor sorption, adsorption, desorption, pre-formulation

Introduction is often used for characterizing a material as such,[15] or

The water-solid interaction(s) among pharmaceuticals
is of great interest in drug development process.[1–3] A
pharmaceutically active or inactive ingredient encom-
passes varied levels of humidity during different stages
of product development, that is to say, during synthesis,
storage and analysis of that ingredient, or during various
unit operations of drug product development like, spray
or freeze drying, wet milling, granulation, etc. The ad-/
ab-sorption of water molecules by a pharmaceutical
solid can largely affect its performance with regard to
stability, flow, wetness, dissolution, compressibility or
compactability, etc.[4–13] Further, varied handling or stor-
age conditions can alter the properties of materials with
regard to their equilibrium moisture content (EMC).
Hygroscopicity is a terminology commonly used for
describing the interaction of a material with water.[14] It
for making a selection among various polymorphs of a
drug molecule,[16–20] or for selecting salt-formers during
initial salt screening attempts.[21–24] Based on its extent
of interaction with water, a material can be classified as
non-hygroscopic or hygroscopic.
Hygroscopicity of a sample can be determined by a
multitude of analytical methods.[25] The earliest reported
and widely practiced method for hygroscopicity deter-
mination or classification was the one proposed by
Callahan and co-workers.[26] This method is referred to as
the “conventional method” herein. It is based upon the
determination of EMC of samples equilibrated at partic-
ular relative humidity (RH) using saturated salt solutions
in the well of the desiccators (Table 1).
This method for hygroscopicity classification of phar-
maceuticals is widely practiced as such,[27] or with some
modifications.[28] However, this method suffers from

Address for Correspondence:  Vaibhav Sihorkar, Pharmaceutical Development, Aurigene Discovery Technologies Limited, Bollaram Road,
Miyapur, Hyderabad, Andhra Pradesh 500 049, India. Tel: +91 40 4465 7777. Fax: +91 40 4465 7438. E-mail: vaibhav_s@aurigene.com;
IP Clearance No. H-01/1025/2010.
(Received 12 April 2011; revised 17 August 2011; accepted 18 August 2011)

348
 Efficient throughput method for hygroscopicity  349

Table 1.  Hygroscopicity classification.

Conventional method European pharmacopoeia
Classification Criteria Classification Criteriaa
Non-hygroscopic Essentially no moisture increase below 90% RH; less than 20% (w/w) – –
increase in moisture content above 90% RH in 1 week
Slightly hygroscopic Essentially no moisture increase below 80% RH; less than 40% (w/w) Slightly hygroscopic Increase in mass
increase in moisture content above 80% RH in 1 week < 2% & ≥ 0.2%
Moderately hygroscopic Moisture content does not increase >5% (w/w) below 60% RH; less Hygroscopic Increase in mass
than 50% (w/w) increase in moisture content above 80% RH in 1 week < 15% & ≥ 2%
Very hygroscopic Moisture content will increase as low as 40–50% RH; greater than 20% Very hygroscopic Increase in
(w/w) increase in moisture content above 90% RH in 1 week mass ≥ 15%
a
Percent water uptake at 25°C/80% RH in 24 h.
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13

drawbacks such as time, sample and resource consum-
ing, tedious, and may offer chances of cross-contamina-
tion. Moreover, any solid state changes occurring within
the sample during the course of the experiment can only
be investigated by analyzing and comparing the starting
and end sample, using suitable analytical method(s).
European Pharmacopoeia (EP) also provides a method
for the determination of the tendency of a sample to take
up atmospheric water. It has been cited as the method
for indicating the degree of hygroscopicity of a sample,
rather than its determination.[29] This method recom-
mends determination of increase in mass of a sample
upon storage at 25°C and 80% RH for 24 hours (Table 1).
This however, subjects the sample to an environment
For personal use only.
analysis method[33] that could be obtained by using a
sorption analyzer or dynamic vapor sorption (DVS)
instrument is proposed and validated in this paper. This
method can serve as an efficient throughput method
for hygroscopicity determination or classification of
active and inactive pharmaceutical ingredients. It is a
precise, accurate and robust method and utilizes less
sample and time. It involves determination of weight
change for a sample purged with humid nitrogen under
isothermal conditions. These gravimetric observations
could be made during increasing, as well as decreasing
humidity cycles, so as to capture the adsorption, as well
as desorption isotherms of the sample, respectively. The
shape, as well as any hysteresis in sorption isotherm

with limited moisture for 24 hours. It may or may not
necessarily yield the desired equilibration.
A comparison of results obtained from the data of
both these methods may be difficult. This is because both
the classification systems are sufficiently different. The
results differ in the way the sample is exposed to mois-
ture and hence the classification. Additionally, these
criteria, which were originally directed toward inactive
pharmaceutical ingredients, may not apply universally to
different areas of drug development.
A pharmaceutically active and inactive ingredient can
exist in crystalline and amorphous form. Furthermore,
within crystalline forms, anhydrous and hydrate forms
can also exist.[30] The hygroscopic nature of the sample
would vary based upon the solid-state under analysis.
That is, in general amorphous and anhydrous crystal-
line forms of a compound tend to ad-/ab-sorb more
moisture when compared with their respective crystal-
line and hydrate forms.[31] Hence, it becomes important
to determine the hygroscopicity of the respective solid-
state sample during the hygroscopicity determination
experiment. Ensuring that the sample does not undergo
any solid-state changes during the experiment is very
important,[32] so that the hygroscopicity classification
pertains to that particular solid-state of the compound
only and not of a different solid-state emerged after
moisture sorption.
Hence, keeping in view the limitations of conven-
tional and EP method for hygroscopicity classifica-
tion, we adopted another emerging methodology for
hygroscopicity determination. A water vapor sorption
can provide valuable information on the water-solid
interaction of the sample.[34] The equilibration time for
water sorption in this method is much shorter than
that in desiccators, which are used in the conventional
methods. This is because of the small sample size and
the continuous purge of gas to maintain a defined
humidity level. Various research works cite the use of
sorption analysis for hygroscopicity determination of
samples.[35,36] However, this method has not yet been
validated using active and/or inactive pharmaceutical
ingredients. Nonetheless it generates enough confi-
dence to be used as a classification system.
A major advantage of this method could be its appli-
cability to new chemical entities (NCEs) during regula-
tory submission of a new drug application. At the same
time, it gives information about the solid-state changes
occurring within the sample. The various mechanisms
by which solids interact with water,[34] and the important
role played by the crystalline or amorphous form of the
solid could also get analyzed by using sorption method.
In the present study, the hygroscopicity of active and
inactive pharmaceutical ingredients was determined
using conventional, EP and sorption analysis method.
The sorption method was assessed against the results
obtained from conventional & EP method. A compara-
tive analysis was then drawn between the results from
these methods. The solid state changes occurring in the
sample during sorption analysis were also evaluated and
correlated. The envisaged work was aimed at validating
sorption analysis as a potential tool to be used for hygro-
scopicity classification of samples.

© 2013 Informa Healthcare USA, Inc.

 350  V. Murikipudi et al.
Materials and methods
Materials
The selection of active and inactive pharmaceutical
ingredients for the present study was made in order to
represent compounds varying in their hygroscopicity.
They vary from low to high, as per the information avail-
able in literature. Further, the inactive pharmaceutical
ingredients were selected to represent commonly used
excipients used for preparing solid dosage forms. For
example, diluents, disintegrants, binders, lubricants, gli-
dants, adsorbents, polymers, etc. Based upon these crite-
ria, 30 inactive and 10 active pharmaceutical ingredients
were selected.
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13
The 30 inactive pharmaceutical ingredients selected
were; colloidal silicon dioxide (Aerosil 200), dicalcium
phosphate anhydrous (A-Tab Granular), microcrystalline
cellulose (Avicel PH 102), sodium salt of carboxymethyl
cellulose (Blanose), carboxy vinyl polymer (Carbopol
914), cellulose acetate phthalate, dextrose (D-Glucose),
dicalcium phosphate dihydrate (Di-Tab Granular),
cellulose ethyl ether (Ethyl cellulose), methyl prop-2-
enoate (Eudragit RL PO), sodium carboxy methyl starch
(Glycolys), sodium starch glycolate, hydroxypropyl
cellulose (Klucel LF), lactose anhydrous, lactose mono-
hydrate, partially pregelatinized starch (Lycatab C),
magnesium aluminum silicate (Magnabrite), mag-
For personal use only.
nesium carbonate anhydrous (Magnesium carbon-
ate), magnesium oxide heavy, magnesium stearate,
D-mannitol (Pearlitol), polyethylene glycol-8000 (PEG-
8000), hydroxypropyl methylcellulose (Pharmacoat
606), polyvinyl pyrrolidone (Plasdone K-30), polyoxy-
ethylene-propylene block copolymer (Poloxamer-188),
cross-linked polyvinyl pyrrolidone (Polyplasdone
XL-10), sodium stearyl fumarate, corn starch, silicon
dioxide (Syloid) and lastly talc. The 10 active pharma-
ceutical ingredients selected were Albendazole, Aspirin,
Enalaprilat, Famotidine, Irbesartan, Metoprolol suc-
cinate, Naproxen sodium, Niacin, Nimesulide and
Raloxifene hydrochloride.
All inactive and active pharmaceutical ingredients
used under this study were procured from certified com-
mercial sources. Additionally, as per the information
provided in their certificate of analysis, it was ascertained
that their residual organic solvent content was minimal
(within ICH limits), and all volatile losses were due to
moisture present.
Hygroscopicity assessment by conventional
method
The hygroscopicity data was generated using the method
reported by Callahan and co-workers.[26] In brief, an
excess amount of saturated salt solution (with excess
crystals) was placed in the well of the desiccators, and
the humidity and temperature were checked using a cali-
brated thermo-hygrometer. The EMC was determined
by placing accurately weighed samples of each material
 Pharmaceutical Development and Technology
(100–300 mg) in open and tarred petri-plates. Then these
were placed inside labeled desiccators containing one
of the saturated salt solutions (lithium chloride, potas-
sium carbonate, sodium chloride, potassium bromide
and potassium nitrate for 11, 43, 75, 83 and 93%RH,
respectively). After 7 days of storage at controlled room
temperature (CRT, 25°C ± 2°C) and particular humidity,
the samples were removed from the desiccators and the
decrease or increase in moisture was determined for
each sample by obtaining the final equilibrium weight
with the aid of a calibrated analytical balance.
The initial moisture content (A) of the samples was
determined using thermo-gravimetric analysis (TGA),
Q5000 IR TGA (TA Instruments, USA). The samples
(5–15  mg) were analyzed under dry nitrogen purge
(25 ml/min) in platinum crucibles at a heating rate of
10 °C/min from 25°C to 300°C. The percentage of weight
loss corresponding to the first transition of the deriva-
tive weight (%/°C) curve was integrated using Universal
Analysis 2000 software and considered as the initial
moisture content of the sample.
Initial moisture content of the sample was used to cal-
culate P (% dry basis), and EMC values were calculated
from P with the aid of the equations 1 and 2. EMC values,
at each RH tested, were tabulated, and samples were
classified based upon the Callahan and co-workers’ clas-
sification system (Table 1).
[(W . A/100) – B] . 100
P= (1)
W - [W . A/100]
P
EMC = (2)
P + 100
Where, P % moisture dry basis
W initial sample weight in grams
A % moisture at start
B Weight change at equilibrium in grams
EMC Equilibrium moisture content
Hygroscopicity assessment by European
pharmacopoeia method
The hygroscopicity data was generated using the method
reported in European Pharmacopoeia.[29] In brief, an
excess amount of saturated salt solution (with excess
crystals) of ammonium chloride was placed in the well
of the desiccators. The humidity and temperature were
checked using a calibrated thermo-hygrometer. Samples
of each material (100–300 mg) were weighed into open
and tarred plastic petri-plates and then placed into a
desiccator, which was labeled and maintained at CRT
and 80 ± 2%RH. After 24 hours of storage, the samples
were removed from the desiccator, and the final mass of
each sample was determined with the aid of a calibrated
analytical balance.
The percentage of increase in mass of each sample
was calculated using equation 3, and the samples were

classified for their hygroscopic nature as per the EP clas-
sification system (Table 1).
m3 - m2 100
Percentage of increase in Mass =
m2 - m1
Where,
m1 mass of empty Petri plate 
m2 mass of Petri plate and sample (Initial)
m3 mass of Petri plate and sample (after 24 h)
Hygroscopicity assessment by sorption analysis
method
This involved subjecting the sample to water vapor
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13
Efficient throughput method for hygroscopicity  351
(Table 3) pharmaceutical ingredients. This was done
to ensure the reproducibility of classification (as given
by Callahan and co-workers) under experimental set-
up used. Most of the excipients could be classified as
(3)
reported in literature,[26] except for a few.
Classification of samples under each class by conven-
tional method relies on EMC data generated at two RH
values. Due to this, it is difficult at times to fit a particular
sample under both criteria, as it may satisfy more than
one criterion of classification. It may lead to an overlap
of hygroscopicity class for the same sample and potential
exists to add subjectivity to the hygroscopicity class of the
sample selected in some cases.

sorption analysis using Q5000 SA sorption analyzer (TA
Instruments, USA) at 25°C. Vapor from the samples were
subjected to varying conditions of humidity under iso-
thermal conditions, and the responses were measured
gravimetrically. To avoid the possibility of change in the
solid state of the sample, initial drying of the sample
within the sorption analyzer was not performed. All the
samples were then analyzed “as is” from 25°C and 40%
RH. Around 5–15 mg sample was exposed to increasing
environmental humidity from 40%RH to 90%RH. The step
equilibrium criterion was set as 0.1% w/w in 10 min with
a maximum step time of 60 min. The samples were then
subjected to desorption from 90% RH to 0% RH. The step
For personal use only.
For e.g. Metoprolol succinate and Niacin could be
classified as slightly hygroscopic (refer to the classifica-
tion mentioned in Table 1). Moisture increase below 80%
RH is minimal (since the term “minimal” is not defined,
and covers broadly up to 10%, values of 2.87% and 3.19%
EMC are herein considered as minimal at 75% RH for
Metoprolol succinate and Niacin, respectively). Also, the
moisture increase above 80% RH is less than 40% (5.77%
and 5.81% EMC at 93% RH for Metoprolol succinate and
Niacin, respectively). The same data, however, quali-
fies them for moderately hygroscopic as well, because
increase in moisture content below 60% RH is less than
5% (1.80% and 1.30% EMC at 43% RH for Metoprolol
succinate and Niacin, respectively) and above 80% RH,

equilibrium criterion for desorption was the same as that

of adsorption. The adsorption-desorption was repeated
for one more cycle from 0%RH to 90%RH to 0%RH. The
percentge of weight change at 25°C and 80% RH in the first
adsorption cycle (measurement criterion analogous to
that mentioned in EP) was calculated as per equation 4.
W2 + W1
% Weight Change = × 100 (4)
W1
 Where,
W1 weight of sample at the start of the experiment
W2 weight of sample in equilibrium at 25°C and 80% RH
The presence of hysteresis between the adsorption
and desorption profiles was used to interpret the sample
characteristics. This was in terms of solid state changes
occurring under the experimental conditions of hygro-
scopicity determination within the sorption analyzer. To
cross check the interpretations drawn from the sorption
isotherm, all the samples were analyzed by powder X- ray
diffractometry (PXRD) before and after performing sorp-
tion analysis. PXRD of the samples was performed using
a Rigaku 2200 diffractometer (Rigaku, Japan) equipped
with a Cu K∝1/50kV/34mA X-Ray source. The X-rays were
received by a scintillation detector. All diffraction patterns
were collected in continuous mode at a scan speed of
3°/min with a scan step of 0.02° over range of 3–45 °2θ.
Results and discussion
Conventional method of hygroscopicity determination
was used to classify 30 inactive (Table 2) and 10 active
it is less than 50% (5.77% and 5.81% EMC at 93% RH for
Metoprolol succinate and Niacin, respectively).
Similarly, Enalaprilat and Naproxen sodium could
be classified as moderately hygroscopic, because, above
80% RH, increase in moisture content is less than 50%
(9.53% and 24.48% EMC at 93% RH for Enalaprilat and
Naproxen sodium, respectively). Though, below 60% RH
the increase in moisture content is more than 5% (5.95%
and 12.30% EMC at 43% RH for Enalaprilat and Naproxen
sodium, respectively). The same data still qualifies them
as very hygroscopic, as well as increase in moisture con-
tent is low at 40–50% RH (5.95% and 12.30% EMC at 43%
RH for Enalaprilat and Naproxen sodium, respectively)
and more than 20% above 90% RH for Naproxen sodium
(24.48% EMC at 93% RH). The same is the case with some
of the inactive pharmaceutical ingredients, like Aerosil,
which can be classified as non-hygroscopic or slightly
hygroscopic. Avicel PH 102, cellulose acetate phthalate,
D-Glucose, Eudragit RL PO, Magnesium carbonate and
Magnesium oxide can be classified as slightly or moder-
ately hygroscopic. Similarly PEG 8000, Lycatab C, Starch
and Syloid come under very hygroscopic classification.
Similar observations can also be seen with the original
work of Callahan and co-workers.[26] These results indi-
cate that the conventional method of hygroscopicity clas-
sification is relatively broad and overlapping.
Unlike conventional method, no historical data could
be traced for hygroscopicity classification of selected
samples by EP method. The results of hygroscopicity
classification of 30 inactive and 10 active pharmaceuti-
cal ingredients determined by EP method are compiled

© 2013 Informa Healthcare USA, Inc.

 352  V. Murikipudi et al.

Table 2.  Hygroscopicity classification of inactive pharmaceutical ingredients studied by conventional method.
EMC (%) values at 25°C & different RH
S. No. Inactive pharmaceutical ingredient 11% 43% 75% 83% 93% Hygroscopicity classa
1 Aerosil –1.23 0.20 3.38 6.97 15.36 NHb
2 A-tab granular 0.42 1.09 1.28 1.34 3.27 NH
3 Di-tab granular –0.16 0.01 1.08 1.33 2.14 NH
4 Ethyl cellulose –5.49 –0.83 –0.82 0.24 4.24 NH
5 Lactose anhydrous –32.2 0.72 0.9 1.38 2.85 NH
6 Lactose monohydrate –3.03 –0.26 0.65 2.09 3.32 NH
7 Magnesium stearate –0.8 0.02 0.95 1.79 2.56 NH
8 Pearlitol –0.55 0.27 1.12 1.55 3.26 NH
9 Sodium stearyl fumarate –1.22 –0.32 1.11 1.69 3.21 NH
10 Talc –0.57 0.65 1.06 1.64 2.25 NH
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13

11 Avicel PH 102 2.46 7.35 9.66 10.46 14.90 SHc

12 Cellulose acetate phthalate –3.14 –2.28 3.08 3.54 9.11 SHc
13 D–glucose –0.38 2.45 3.62 9.84 55.65 SHc
14 Eudragit RL PO –0.09 2.01 4.73 7.34 11.04 SHc
15 Magnesium carbonate –4.84 1.11 2.36 3.72 6.28 SHc
16 Magnesium oxide 1.85 2.61 5.33 6.80 9.28 SHc
17 PEG 8000 –3.43 –0.05 1.52 2.74 51.07 SHd
18 Poloxamer 188 –0.42 0.13 1.25 2.66 42.3 SHc
19 Klucel LF –2.2 3.38 6.64 9.35 19.12 MH
20 Lycatab C 6.67 9.47 12.90 13.04 21.15 MHd
21 Magnabrite 2.07 5.96 13.46 16.37 20.84 MHd
22 Pharmacoat 606 2.04 2.95 9.03 12.75 21.29 MH
23 Starch 3.50 9.85 12.45 13.43 18.35 MHd
24 Syloid –1.27 9.02 12.86 14.3 16.14 MHd
For personal use only.

25 Blanose 11.13 15.71 26.73 29.05 49.25 VH

26 Carbopol 914 2.43 5.63 16.66 24.84 37.24 VH
27 Glycolys 5.36 9.21 21.22 30.91 47.25 VH
28 Plasdone K-30 9.52 14.24 25.93 27.41 41.25 VH
29 Polyplasdone XL-10 0.85 16.16 20.32 24.46 33.35 VH
30 Sodium starch glycolate 6.93 18.35 23.07 27.75 44.75 VH
a
NH− Non-hygroscopic, SH− Slightly hygroscopic, MH− Moderately hygroscopic, VH− Very hygroscopic.
Samples can also be classified as bslightly hygroscopic, cmoderately hygroscopic & dvery hygroscopic.

Table 3.  Hygroscopicity classification of active pharmaceutical ingredients studied by conventional method.
EMC (%) values at 25°C & different RH

S. No. Active pharmaceutical ingredient 11% 43% 75% 83% 93% Hygroscopicity classa
1 Albendazole –0.28 0.31 1.54 1.92 3.54 NH
2 Aspirin –0.73 0.29 1.05 1.77 3.08 NH
3 Famotidine –0.47 0.82 1.85 2.83 3.94 NH
4 Irbesartan –0.84 0.29 1.94 2.58 3.83 NH
5 Nimesulide –0.47 0.23 1.38 1.65 3.19 NH
6 Raloxifene hydrochloride –0.46 0.46 1.18 2.02 3.08 NH
7 Metoprolol succinate –1.67 1.80 2.87 4.90 5.77 SHb
8 Niacin –0.97 1.30 3.19 4.84 5.81 SHb
9 Enalaprilat –0.29 5.95 7.64 7.65 9.53 MHc
10 Naproxen sodium 5.04 12.30 14.14 18.64 24.48 MHc
a
NH− Non-hygroscopic, SH− Slightly hygroscopic, MH− Moderately hygroscopic, VH− Very hygroscopic.
Samples can also be classified as bmoderately hygroscopic & cvery hygroscopic.

in Tables 4 and 5, respectively. These results do not cor-
relate well with that of the conventional method. This
might be due to the short equilibration time of 24 hours
at single RH condition applied in the EP method, as
compared to that of 7 days at each RH condition in con-
ventional method. Moreover, EP method only considers
the percentage of increase in mass occurring on storage
at 25°C and 80% RH for 24 h, and neither has it included
the initial moisture content, nor the EMC of sample for
its classification, unlike that of the conventional method.
The results may therefore, rely heavily on the thermal his-
tory of the starting material, and there lies a potential for

 Pharmaceutical Development and Technology

 Efficient throughput method for hygroscopicity  353

Table 4.  Hygroscopicity classification of inactive pharmaceutical Table 5.  Hygroscopicity classification of active pharmaceutical
ingredients studied by EP method. ingredients studied by EP method.
Inactive pharmaceutical % Increase Hygroscopicity Active pharmaceutical % Increase Hygroscopicity
S. No. ingredient in mass classa S. No. ingredient in mass classa
1 Aerosil –1.64 Not defined in EP 1 Albendazole 0.10 Not defined in EP
2 A-tab granular –1.68 Not defined in EP 2 Aspirin –0.03 Not defined in EP
3 D-glucose –0.23 Not defined in EP 3 Enalaprilat –0.08 Not defined in EP
4 Di-tab granular –0.07 Not defined in EP 4 Famotidine –0.31 Not defined in EP
5 Ethyl cellulose 0.15 Not defined in EP 5 Raloxifene hydrochloride 0.15 Not defined in EP
6 Lactose anhydrous –0.51 Not defined in EP 6 Irbesartan 0.24 SH
7 Lactose monohydrate –0.21 Not defined in EP 7 Metoprolol succinate 0.59 SH
8 Magnesium carbonate –1.10 Not defined in EP 8 Naproxen sodium 0.59 SH
9 Pearlitol 0.15 Not defined in EP 9 Niacin 1.37 SH
10 Talc 0.09 Not defined in EP 10 Nimesulide 0.30 SH
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13

11 Cellulose acetate 1.50 SH a

SH − Slightly hygroscopic.
phthalate
12 Eudragit RL PO 0.66 SH Table 6.  Hygroscopicity classification criterion by sorption
13 Lycatab C 0.52 SH analysis.
14 Magnesium oxide 0.35 SH Hygroscopicity class Criteriaa
15 Magnesium stearate 0.49 SH Non-hygroscopic < 0.2% (w/w)
16 PEG− 8000 0.21 SH Slightly hygroscopic 0.2–2% (w/w)
17 Pharmacoat 606 1.77 SH Hygroscopic / Moderately hygroscopic 2–15% (w/w)
18 Poloxamer-188 0.17 SH Very hygroscopic >15% (w/w)
19 Sodium stearyl fumarate 0.29 SH a
Percent water uptake at 25°C/80% RH in first adsorption cycle of
20 Avicel PH 102 3.26 H sorption isotherm
21 Blanose 6.15 H
22 Carbopol 914 9.27 H and initiating sorption analysis from ambient conditions,
For personal use only.

23 Glycolys 11.10 H i.e. 25°C and 40% RH) after attaining equilibration in
24 Klucel LF 5.19 H weight over the step equilibrium. The reasons for such an
25 Magnabrite 3.65 H approach were manifold. Foremost being its synchroni-
26 Plasdone K-30 5.87 H zation with the criterion of EP and conventional method.
27 Polyplasdone XL-10 6.03 H Callahan and co-workers (referred herein as conventional
28 Sodium starch glycolate 11.96 H method) utilized the percentage of moisture increase in
29 Starch 2.18 H sample mostly on exposure to high humidity conditions,
30 Syloid 2.69 H such as 80 and 90% RH. However, samples may behave dif-
a
SH − Slightly hygroscopic, H − Hygroscopic. ferently at times under different RH conditions. Therefore,
some samples may get unambiguously classified under
a material being classified at different degrees of hygro- more than one classification, as described elsewhere in
scopicity under differ experimental backgrounds. this manuscript. Further, hygroscopicity classification
The proposed efficient throughput method was used by sorption analysis has added advantage of sample and
to classify hygroscopicity class of 30 inactive and 10 active time conservation, along with the complete view of sorp-
pharmaceutical ingredients, and results were compared tion profile of “as is” (from I adsorption-desorption cycles)
with the data of hygroscopicity generated using conven- and desorbed (from II adsorption-desorption cycles)
tional and EP methods. The basis for validating water sample. Furthermore, this also provides with an addi-
sorption analysis as a potential tool for hygroscopicity tional information on potential solid-state transformation
classification was to attain moisture equilibration of the due to moisture sorption or desorption (hysteresis and/
sample with the most efficient mechanism with less time or hydrate formation potential during adsorption and
and compound. Additionally, to make unequivocal and desorption cycles) for any further investigation. However,
unambiguous hygroscopicity classification of compounds it is up to the discretion of experimenter to modify the
based on equilibrium moisture reached at an optimally temperature or humidity conditions for sample analysis
selected RH condition. The premise of the strategy in the by sorption analyzer. The experimenter can then evalu-
envisaged work toward validating a novel throughput ate the obtained data for arriving at logical conclusion
method of hygroscopicity classification, relied upon equil- to characterize drug and support drug product develop-
ibration concept of conventional method and an optimal ment. The authors also feel it is worth to maintain the
RH concept (and classification theme) of EP method. classification of samples as “Non-Hygroscopic” as in
The criterion for hygroscopicity classification by sorp- conventional method. The hygroscopicity classification
tion analysis has been taken as the EMC of the sample at system followed under envisaged work is detailed in Table
25°C and 80% RH. It is defined as the percentage of weight 6, and is derived for 30 inactive and 10 active pharmaceu-
change of “as is” sample (without pre-drying of sample tical ingredients listed in Tables 7 and 8, respectively.
© 2013 Informa Healthcare USA, Inc.
 354  V. Murikipudi et al.

Table 7.  Hygroscopicity classification of inactive pharmaceutical

ingredients studied by sorption analysis.
% Weight
Inactive pharmaceutical change Hygroscopicity
S. No. ingredient (25°C/80%RH) classa
1 D-glucose 0.15 NH
2 Di-tab granular 0.04 NH
3 Lactose anhydrous 0.06 NH
4 Lactose monohydrate 0.03 NH
5 Magnesium stearate 0.18 NH
6 Pearlitol 0.12 NH
7 Sodium stearyl fumarate 0.01 NH
8 Talc 0.06 NH
9 A-tab granular 0.47 SH
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13

10 Ethyl cellulose 1.72 SH

11 Eudragit RL PO 1.92 SH
12 Magnesium carbonate 0.46 SH
13 Magnesium oxide 0.57 SH
14 PEG - 8000 0.68 SH
15 Aerosil 3.58 H/MH
16 Avicel PH 102 6.08 H/MH
17 Carbopol 914 9.14 H/MH
18 Cellulose acetate phthalate 5.33 H/MH
19 Klucel LF 8.91 H/MH
20 Lycatab C 5.49 H/MH
21 Magnabrite 7.62 H/MH
22 Pharmacoat 606 10.43 H/MH
For personal use only.

23 Poloxamer-188 10.95 H/MH

24 Starch 7.95 H/MH
25 Syloid 4.01 H/MH
26 Blanose 23.72 VH
27 Glycolys 15.07 VH
28 Plasdone K-30 24.47 VH
29 Polyplasdone XL-10 19.23 VH
30 Sodium starch glycolate 22.66 VH Figure 1. Comparison of hygroscopicity classification of (a)
a
NH − Non-hygroscopic, SH − Slightly hygroscopic, H − Hygroscopic, inactive and (b) active pharmaceutical ingredients by different
MH − Moderately hygroscopic, VH − Very hygroscopic. methods. Key: NH–Non-hygroscopic, SH–Slightly hygroscopic,
MH–Moderately hygroscopic, H–Hygroscopic, VH–Very
Table 8.  Hygroscopicity classification of active pharmaceutical hygroscopic. * Criteria for “Non-Hygroscopic” materials not
ingredients studied by sorption analysis. defined in EP method of hygroscopicity classification.
% Weight
Active pharmaceutical change Hygroscopicity with those of conventional method. Additionally, both
S. No. ingredient (25°C/80%RH) classa these methods provide sufficient equilibration of sam-
1 Albendazole 0.011 NH ples under a specific humidity condition, although the
2 Aspirin 0.002 NH equilibration time and mechanisms are different. On
3 Famotidine 0.004 NH the other hand, EP method could probably have under-
4 Metoprolol succinate 0.006 NH evaluated the samples being either just “Hygroscopic”
5 Niacin 0.016 NH or “Slightly Hygroscopic”. Even though they could have
6 Nimesulide 0.002 NH been classified one level up into “Very Hygroscopic” or
7 Raloxifene hydrochloride 0.018 NH “Hygroscopic” class by other two methods. This suggests
8 Irbesartan 0.341 SH toward a potentially incomplete bulk equilibration of
9 Enalaprilat 1.679 SH samples within the timeframe of EP method, i.e. 24 h. On
10 Naproxen sodium 3.111 H/MH the contrary, an equivalent equilibration to that seen in
a
NH − Non-hygroscopic, SH − Slightly hygroscopic, H − Hygroscopic, conventional method could be achieved with sorption
MH − Moderately hygroscopic.
analysis despite the use of lesser amount of sample and
A comparative analysis of hygroscopicity classification lesser but adequate equilibration time due to highly effi-
of inactive and active pharmaceutical ingredients inves- cient contact of sample with water vapor in a micro-envi-
tigated under present study by different methodologies ronment of interaction chamber. The evident differences
is presented in Figure 1. An overview of the data suggests of classifying 30 inactive and 10 active pharmaceutical
that results of sorption analysis are in better correlation ingredients (Figure 1) using sorption could be attributed
 Pharmaceutical Development and Technology

Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13
For personal use only.
Figure 2.  Results of (i) sorption analysis and (ii) PXRD of Naproxen sodium (i) before and (ii) after sorption analysis.
to a hybrid approach of envisaged method, which clas-
sifies the compounds on EP criterion despite ensuring
equilibration as per conventional method.
The percentage of weight change determined for clas-
sifying the hygroscopic nature of the sample under sorp-
tion conditions can be considered as EMC of the sample
at that particular temperature and humidity. Moreover,
hygroscopicity classification by sorption analysis is fur-
ther beneficial in terms of exposure of sample to increas-
ing RH with equilibration at each RH step. This however,
is not the case with conventional and EP method wherein
the sample is being subjected to a fixed RH at a time.
An important aspect of envisaging a through-
put method for hygroscopicity classification was to
simultaneously generate pivotal information that
could serve as a means to understand the behavior
of materials under moisture ingress. All the inactive
© 2013 Informa Healthcare USA, Inc.
Efficient throughput method for hygroscopicity  355
pharmaceutical ingredients showed none or minimal
hysteresis in sorption isotherms. This was with further
complimented with no solid-state transformation as
determined by PXRD analysis, before and after sorp-
tion analysis (data not shown). Majority of the active
pharmaceutical ingredients also showed minimal or no
hysteresis in sorption isotherms, except for Enalaprilat
and Naproxen sodium. These samples were further
investigated for solid-state changes and were found to
be transforming into alternate solid-state during sorp-
tion analysis (Figures 2 and 3). This has been a critical
evaluation that can remain unaddressed if conven-
tional, or EP methods are being used.
The conventional method of hygroscopicity classifica-
tion utilizes 1–2 gm of sample for all RH conditions and
it takes 7 days for completion. The EP method of hygro-
scopicity classification on the other hand utilizes lesser
 356  V. Murikipudi et al.
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13
For personal use only.
Figure 3.  Results of (i) sorption analysis and (ii) PXRD of Enalaprilat (i) before and (ii) after sorption analysis.
sample (100–300 mg) and lesser time (1 day), but it has its
own shortcomings, as elaborated earlier. Determination
of hygroscopicity by sorption analysis requires 10–20 mg
of sample, and complete information could be captured
within 1–2 days across all RH values. In brief, the sorption
analysis saves lot of time, consumes lesser samples and
provides more technical inputs from a single experiment,
which can lead to a logical conclusion for classification
of hygroscopicity of active and inactive pharmaceuti-
cal ingredients when compared to that of conventional
method.
Conclusion
Hygroscopic nature of 30 inactive and 10 active pharma-
ceutical ingredients was determined using conventional,
 Pharmaceutical Development and Technology
EP and sorption analysis method. Conventional method
of hygroscopicity classification utilizes more time, sam-
ple, and does not provide any insight on the solid state
changes occurring within the sample under the experi-
mental conditions of analysis. A sample can convert from
anhydrous to hydrate state at high RH, and vice versa
at low RH conditions, chances do exists in the conven-
tional method and the starting material may undergo
polymorph change during the experiment. In this sce-
nario, no information would be available whether the
classification provided is of the starting polymorph or
the new polymorph of the sample by conventional or EP
methods. Moreover, the classification provided by con-
ventional method is very broad and overlapping, so that
same sample can be classified under two categories if the
data is interpreted by two different personnel.

EP method of hygroscopicity classification is very nar-
row and precise, but it considers single condition of 25°C
and 80% RH for its classification that may not necessarily
ensure complete equilibration and may under-evaluate
the ability of a sample towards moisture uptake. It also
utilizes more samples and does not provide information
on the solid state changes occurring within the sample
under the experimental conditions studied.
Efficient throughput method of hygroscopicity classifi-
cation utilizing sorption analysis is investigated herein, is
more robust, requires very less compound and time, and
provides precise and reproducible data. It also utilizes
concise classification at an optimal equilibration and an
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13
optimal RH, and considers EMC value for hygroscopicity
classification. The sorption isotherm also provides infor-
mation about the solid state changes occurring within the
sample. Hence, it is a very valuable method to classify the
hygroscopic nature of the sample, which can be applied
universally for all the active and inactive pharmaceutical
ingredients. This also includes early and late drug develop-
ment stages, where time and compound availability could
be limiting. Apart from offering compound, time and
resource conservation, it also provides valuable scientific
data from a single experiment, and may provide flexibility
to interpret and design their experiments accordingly.
For personal use only.
Acknowledgments
Authors wish to thank Dr. Reddy’s Laboratories Ltd. and
Aurigene Discovery Technologies Ltd. for their support
in this research work. Scientific support provided by TA
Instruments Ltd. in interpreting sorption analysis data is
also greatly acknowledged.
Declaration of interest
The authors report no declarations of interest.
References
1. Zografi G. States of water associated with solids. Drug Dev Ind
Pharm 1988;14:1905–26
2. Kontny M, Zografi G. Sorption of water by solids. In: Brittain H,
editor. Physical characterization of pharmaceutical solids. New
York: Marcel Dekker; 1995.386–418.
3. Reutzel-Edens S, Newman A. The physical characterization of
hygroscopicity in 17 pharmaceutical solids. In: Hilfiker R, editor.
Polymorphism in the pharmaceutical industry. Weinheim: Wiley-
VCH; 2006. 235–58.
4. Fassihi A, Persicaner P. Solid state interaction of bromazepam with
poly (vinylpyrrolidone) in the presence of moisture. Int J Pharm
1987; 37:167–70.
5. Padmadisastra Y, Kennedy R, Stewart P. Solid bridge formation in
sulfonamide-28 29 Emdex interactive systems. Int J Pharm 1994;
112:55–63.
6. Schmitt E, Davis CW, Long ST. Moisture-dependent crystallization of
amorphous lamotrigine mesylate. J Pharm Sci 1996;85:1215–1219.
7. Koizumi N, Kouji M, Itai S. Effect of water content and temperature
on the stability of lornoxicam tablets. Drug Stability 1997;1:202–8.
8. Rohrs BR, Thamann TJ, Gao P, Stelzer DJ, Bergren MS, Chao RS. Tablet
dissolution affected by a moisture mediated solid-state interaction
between drug and disintegrant. Pharm Res 1999;16:1850–1856.
© 2013 Informa Healthcare USA, Inc.
Efficient throughput method for hygroscopicity  357
  9. Cleaver J, Karatzas G, Louis S, Hayati I. Moisture-induced caking of
boric acid powder. Powder Technol 2004; 146:93–101.
10. Listiohadi Y, Hourigan J, Sleigh R, Steele R. Properties of lactose
and its caking behavior. Aust J Dairy Technol. 2005; 60:33–52.
11. Nail S. Physical and chemical stability considerations in the
development and stress testing of freeze-dried pharmaceuticals.
Drugs Pharm Sci 2005; 153:261–91.
12. Faqih AM, Mehrotra A, Hammond SV, Muzzio FJ. Effect of moisture
and magnesium stearate concentration on flow properties of
cohesive granular materials. Int J Pharm 2007;336:338–345.
13. Sun CC. Mechanism of moisture induced variations in true density
and compaction properties of microcrystalline cellulose. Int J
Pharm 2008;346:93–101.
14. Newman A, Reutzel-Edens S, Zografi G. Characterization of the
“hygroscopic” properties of active pharmaceutical ingredients.
J Pharm Sci 2008; 97(3):1047–1059.
15. Huang LF, Tong WQ. Impact of solid state properties on
developability assessment of drug candidates. Adv Drug Deliv Rev
2004;56:321–334.
16. Narurkar A, Purkaystha AR, Sheen P-C, Augustine MA.
Hygroscopicity of Celiprolol hydrochloride polymorphs. Drug Dev
Ind Pharm 1988; 14:465–74.
17. Kobayashi Y, Ito S, Itai S, Yamamoto K. Physicochemical properties
and bioavailability of carbamazepine polymorphs and dihydrate.
Int J Pharm 2000;193:137–146.
18. Kojima T, Sugano K, Onoue S, Murase N, Sato M, Kawabata Y et al.
Solid form selection of zwitterionic 5-HT4 receptor agonist. Int J
Pharm 2008;350:35–42.
19. Aaltonen J, Allesø M, Mirza S, Koradia V, Gordon KC, Rantanen
J. Solid form screening–a review. Eur J Pharm Biopharm 2009;71:
23–37.
20. Katrincic LM, Sun YT, Carlton RA, Diederich AM, Mueller RL, Vogt
FG. Characterization, selection, and development of an orally
dosed drug polymorph from an enantiotropically related system.
Int J Pharm 2009;366:1–13.
21. Morris KR, Fakes MG, Thakur AB, Newman AW, Singh AK, Venit JJ,
et al. An integrated approach to the selection of optimal salt form
for a new drug candidate Int J. Pharm 1994; 105:209–17.
22. Bastin R, Bowker M, Slater B. Salt selection and optimization
procedures for pharmaceutical new chemical entities. Org Process
Res Dev 2000; 4:427–435.
23. Machatha SG, Sanghvi T, Yalkowsky SH. Structure determination
and characterization of carbendazim hydrochloride dihydrate.
AAPS PharmSciTech 2005;6:E115–E119.
24. Guerrieri P, Jarring K, Taylor L. Impact of counter ion on the
chemical stability of crystalline salts of procaine. J Pharm Sci 2010;
99:3719–30.
25. Kumar L, Amin A, Bansal AK. An overview of automated systems
relevant in pharmaceutical salt screening. Drug Discov Today
2007;12:1046–1053.
26. Callahan J, Cleary G, Elefant M, Kaplan G, Kensler T, Nash R.
Equilibrium moisture content of pharmaceutical excipients. Drug
Dev Ind Pharm 1982; 8:355–69.
27. Visalakshi NA, Mariappan TT, Bhutani H, Singh S. Behavior of
moisture gain and equilibrium moisture contents (EMC) of various
drug substances and correlation with compendial information
on hygroscopicity and loss on drying. Pharm Dev Technol
2005;10:489–497.
28. Roskar R, Kmetec V. Evaluation of the moisture sorption behaviour
of several excipients by BET, GAB and microcalorimetric
approaches. Chem Pharm Bull 2005;53:662–665.
29. Anonymous. Characters section in monographs. European
Pharmacopoeia 6, version 6.8, Section 5.11 ed2010.
30. Hilfiker R, Blatter F, Raumer M. Relevance of solid-state properties
for pharmaceutical products. In: Hilfiker R, ed. Polymorphism:
in the Pharmaceutical Industry. Weinheim: WILEY-VCH Verlag
GmbH & Co. KGaA; 2006. 1–19.
31. Petit S, Coquerel G. The amorphous state. In: Hilfiker R, editor.
Polymorphism in the Pharmaceutical Industry. Weinheim: Wiley-
VCH; 2006.259–285.
 358  V. Murikipudi et al.
32. Yoshinari T, Forbes RT, York P, Kawashima Y. Moisture induced
polymorphic transition of mannitol and its morphological
transformation. Int J Pharm 2002;247:69–77.
33. Balbach S, Korn C. Pharmaceutical evaluation of early development
candidates “the 100 mg-approach”. Int J Pharm 2004;275:1–12.
34. Umprayn K, Mendes R. Hygroscopicity and moisture adsorption
kinetics of pharmaceutical solids: A review. Drug Dev Ind Pharm
1987; 13:653–693.
Pharmaceutical Development and Technology Downloaded from informahealthcare.com by Novartis Pharma on 01/14/13
For personal use only.
 Pharmaceutical Development and Technology
35. Jain A, Adeyeye M. Hygroscopicity, phase solubility and dissolution
of various 6789 substituted sulfobutylether β-cyclodextrins (SBE) and
danazol-SBE inclusion complexes. Int J Pharm 2001; 212:177–186.
36. van Drooge DJ, Hinrichs WL, Visser MR, Frijlink HW.
Characterization of the molecular distribution of drugs in glassy
solid dispersions at the nano-meter scale, using differential
scanning calorimetry and gravimetric water vapour sorption
techniques. Int J Pharm 2006;310:220–229.