Content uploaded by ANA PAULA Barreto Gomes
Author content
All content in this area was uploaded by ANA PAULA Barreto Gomes on Oct 26, 2015
Content may be subject to copyright.
Compatibility study between hydroquinone and the excipients
used in semi-solid pharmaceutical forms by thermal and non-
thermal techniques
I
´gor Prado de Barros Lima •Naiana Gondim P. B. Lima •Denise M. C. Barros •
Thays S. Oliveira •Ca
ˆndida M. S. Mendonc¸a •Euze
´bio G. Barbosa •
Fernanda N. Raffin •Tu
´lio F. A. de Lima e Moura •Ana Paula B. Gomes •
Ma
´rcio Ferrari •Cı
´cero F. S. Araga
˜o
Received: 3 November 2013 / Accepted: 3 August 2014 / Published online: 29 August 2014
ÓAkade
´miai Kiado
´, Budapest, Hungary 2014
Abstract Thermal techniques, such as differential scan-
ning calorimetry (DSC), thermogravimetry (TG), derivate
of TG curve, differential thermal analysis, and non-thermal
techniques such as fourier transform infrared (FTIR)
spectroscopy and X-ray diffractometry (XRD) were used to
evaluate the possible interactions between hydroquinone
(HQ) and excipients commonly used in semi-solid phar-
maceutical forms. The DSC curve of HQ showed a sharp
endothermic event between 173 and 179 °C indicating
melting point. No evidence of interaction was observed
between HQ and cetyl alcohol (CA), cetostearyl alcohol
(CTA), disodium ethylenediaminetetraacetate , and decyl
oleate. However, based on the thermoanalytical trials, a
physical interaction was suspected between HQ and di-
propylene glycol (DPG), glycerin (GLY), hydroxypropyl
methylcellulose (HPMC), imidazolidinyl urea (IMD),
methylparaben (MTP), and propylparaben (PPP). The
FTIR results show that for DPG, GLY, HPMC, MTP, and
PPP, there were no chemical interactions with HQ at room
temperature, but the heating promotes interaction between
HQ and HPMC. The FTIR spectra of HQ/IMD show the
chemical interaction at room temperature, which was also
observed with heating. The XRD results of mixtures
between HQ and DPG, HPMC, IMD, MTP, and PPP
indicate no interaction between these substances at room
temperature, but the heating modifies the HQ crystallinity
in these mixtures. All of these methods showed incom-
patibility between HQ and the excipient IMD.
Keywords Hydroquinone Compatibility Thermal
analysis FTIR XRD
Introduction
Hydroquinone is the most popular depigmenting agent, a
phenolic compound chemically known as 1,4-dihydroxy-
benzene that inhibits the conversion of DOPA to melanin
by tyrosinase inhibition. It covalently binds to histidine or
interacts with copper at the active site of tyrosinase. It has
been the gold standard for treatment of hyperpigmentation
for many decades. It is also supposed to inhibit DNA and
RNA synthesis and induce degradation of melanosomes
and destruction of melanocytes [1,2].
Hydroquinone is very difficult to formulate in a stable
preparation. It is a highly reactive oxidant that rapidly
combines with oxygen. Typically, HQ skin-lightening
creams have a creamy color that changes to a darker yellow
or brown as oxidation occurs. As the discoloration pro-
gresses, the HQ activity decreases. Products with any off-
color change should be immediately discarded [2].
This active pharmaceutical ingredient (API), what is
HQ, is mixed with selected excipients to prepare semi-solid
formulations for the treatment of hyperpigmentation. The
successful formulation of a stable and effective semi-solid
dosage form depends on the careful selection of the ex-
cipients used to make administration easier or more suit-
able, to improve patient compliance, to promote release
and bioavailability of the drug, and protect it from degra-
dation [3,4].
I
´. P. de Barros Lima N. G. P. B. Lima
D. M. C. Barros T. S. Oliveira C. M. S. Mendonc¸a
E. G. Barbosa F. N. Raffin T. F. A. de Lima e Moura
A. P. B. Gomes M. Ferrari C. F. S. Araga
˜o(&)
Department of Pharmacy, Federal University of Rio Grande do
Norte, Rua Gen. 9 Cordeiro de Faria s/n, Rua Ju
´lio Gomes
Moreira 1113, 303 A, Bairro: Barro Vermelho, Natal,
Rio Grande do Norte CEP 59.022-110, Brazil
e-mail: ciceroaragao@ufrnet.br; cicero.aragao@yahoo.com.br
123
J Therm Anal Calorim (2015) 120:719–732
DOI 10.1007/s10973-014-4076-9
Excipients are known to facilitate administration and
release of an active component, as well as protecting it
from the environment. Furthermore, excipients are con-
sidered pharmaceutically inert, but physical and chemical
interactions with active components are possible [4,5].
The study of drug–excipients compatibility represents an
important phase in the pre-formulation stage for the devel-
opment of all dosage forms. In fact, potential physical and
chemical interactions between drug and excipients can affect
the chemical nature, stability and drug bioavailability and,
consequently, their therapeutic efficacy and safety [3].
The thermoanalytical methods are useful at the pre-
formulation stage to obtain information on the physico-
chemical properties and thermal behavior of the active
substances, because they are related to its decomposition.
Furthermore, data acquired at this stage are extremely
important in critical decisions relating to subsequent phases
of development [6].
Brazil Thermoanalytical techniques, especially differ-
ential scanning calorimetry (DSC), thermogravimetry
(TG)/derivate of TG curve (DTG), and differential thermal
analysis (DTA), have been used a long time ago by phar-
macists for characterization of materials before their use
and/or at any other stages of the pre-formulation [7–9].
The DTA analytical technique was used in this study,
along with DSC, in order to recover the DTA technique for
use in compatibility studies of pharmaceutical samples.
The DTA, similar to the DSC, is used to measure the
melting point and heat of fusion, and it has been used for
more than five decades to evaluate interactions between
substances [10,11]. Several studies have confirmed that
these techniques can be considered as important tools in the
first stage of developing a formulation in the excipients
selection [12–14].
There are several differences between these two tech-
niques; DTA detects the temperature differences while the
DSC detects changes in enthalpy (heat flow difference).
The DTA is a robust technique and older than the DSC. On
the other hand, the DSC is derived from DTA, being more
sensitive than DTA. The DTA uses sample amount about
four times greater than the DSC and has as the lV as unit
(microvolts), since the DSC is expressed in mW (milli-
watt). The open alumina crucible is used on the DTA,
while for DSC is used closed aluminum crucible. By the
combination of DTA and DSC techniques many overlap-
ping stages can be observed and better interpreted [15].
Thermoanalytical trials have been proposed as a rapid
method for the evaluation of physicochemical interactions
between components of the formulation and, therefore, for
selection of excipients. However, interpretation of the
thermal data is not always simple, and it is necessary to
avoid misinterpretation and unreliable conclusions thor-
ough the evaluation [3]. Therefore, other analytic
techniques often have to be used to adequately interpret TA
findings. Besides TA techniques, fourier transform infrared
(FTIR) and X-ray diffractometry (XRD) have been used as
a complementary method for the evaluation of possible
interactions between the components [4].
Several reports are described in the literature to evaluate
the interactions between drugs and excipients using ther-
mal (DSC and DTA) and non-thermal (FTIR and XRD)
techniques [4,16–18].
The aim of the present work was to evaluate the possible
interactions between HQ and excipients commonly used in
semi-solid pharmaceutical forms using thermal (DSC, TG/
DTG and DTA) and non-thermal (FTIR and XRD)
techniques.
Materials and methods
Materials and samples
Besides the API, the names, classification, and provider of
the used excipients are listed in Table 1. The excipients
selected are part of the composition of an anionic cream
and a hydroxypropyl methylcellulose gel contained in the
national formulary of Brazilian pharmacopeia [19]. Binary
mixtures (BM) of HQ with each selected excipient were
Table 1 Raw materials used in the compatibility study
Sample Classification Provider
API
Hydroquinone (HQ) Active ingredient Viafarma
Excipients
Cetyl alcohol (CA) Emulsifying agent;
stiffening agent
Pharma
special
Cetostearyl alcohol
(CTA)
Emulsifying agent;
stiffening agent
Henrifarma
Dipropylene glycol
(DPG)
Humectant; cosolvent Galena
Disodium EDTA
(EDTA)
Chelating agent Viafarma
Glycerin (GLY) Humectant; emollient Galena
Hydroxypropyl
methylcellulose
(HPMC)
Gelling agent; viscosity-
increasing agent
Henrifarma
Imidazolidinyl urea
(IMD)
Antimicrobial
preservative
Fagron
Methylparaben (MTP) Antimicrobial
preservative
Pharma
special
Decyl oleate (DCO) Emolient Galena
Propylparaben (PPP) Antimicrobial
preservative
Pharma
special
720 I
´. P. de Barros Lima et al.
123
prepared in the 1:1 (m/m) ratio by simple physical mixture
of the components in agate mortar with pestle for 5 min.
The 1:1 (m/m) ratio was chosen to maximize the proba-
bility of observing any interaction.
Thermal techniques
DSC curves were obtained in a Shimadzu DSC-60 cell,
using closed aluminum pans with about 2 mg of samples,
under dynamic atmosphere of N
2
(flow rate of
50 mL min
-1
) and heating rate of 10 °C min
-1
in the
temperature range from 25 to 450 °C. Tests were carried
out individually with API and excipients, then with
recently prepared physical mixtures.
Highly pure Zn and In were used to calibrate the DSC
equipment, where experiments were run at 200 and 500 °C,
respectively. Through their melting points (156.65 and
419.50 °C for In and Zn, respectively) the areas under the
peaks were determined. Once the correction of the cali-
bration temperature was performed, the heat calibration
was corrected in which the enthalpy value for In and Zn
was 28.5 and 100.5 J g
-1
, respectively. Further, new
experiments were performed to assure that the melting
temperature varied in the range of ±0.5 °C and the values
of melting enthalpy (DH) in the range of ±1.0 J g
-1
. Once
these parameters were reached, the calibration was
accomplished.
TG and DTA curves were obtained on a SHIMADZU
thermobalance model TGA 60 (simultaneous TG/DTA),
using alumina pan (about 8 mg samples), heating rate of
10 °C min
-1
in the 25–900 °C temperature range, under
dynamic atmosphere of N
2
at 50 mL min
-1
. Tests were
carried out individually with API and excipients, then with
recently prepared physical mixtures.
The TGA 60 equipment was calibrated using In which
was heated up to 200 °C followed by correction of the
calibration temperature. Next, another experiment was run
with the purpose of checking whether the melting tem-
perature varied in ±0.5 °C.
Thermal curves were analyzed with the aid of the
SHIMADZU software TASYS to identify thermal events
presented as well as the temperatures (T
onset
, and T
peak
) and
energies (J g
-1
) involved in these events.
Non-thermal techniques
Fourier transform infrared spectroscopy in transmittance
mode was used. KBr pellets with 1 % mass of the pow-
dered material were produced. The spectra were obtained
using a PerkinElmer FTIR spectrometer, model Spec-
trum
TM
65, in the spectral area 400–4,000 cm
-1
, with a
resolution 0.5 cm
-1
.
In order to detect interactions between the active and
excipients ingredients, a correlational IR analysis was
performed. Such analysis was carried out by importing the
IR spectral data in Spectrum 10 software. The spectral
region from 400 to 4,000 cm
-1
was considered in this
approach. A theoretical IR spectrum of the active and ex-
cipients ingredients was built to establish a comparison
with the linear combination of the BM’s experimental
spectra. Subsequently, the Pearson’s correlation
(r) between the theoretical and experimental drug–excipi-
ent IR spectra was calculated. The deviation from ideality
(r=1) was interpreted as indication of problems for a
particular drug–excipient mixture analyzed. The compari-
son of the IR spectra of pure drug and BM drug/excipient is
widely used in the pharmaceutical literature.
The XRD of the samples was obtained in a Shimadzu
X-ray diffractometer model Maxima_X XRD-7000, with
X-ray tube sealed type using a Cu karadiation. The
scanned range was 5–80°(2h). The equipment was oper-
ated on 40.0 kV, 30.0 mA. Data were plotted by means of
the software Origin version 8.0.
Analyses were performed in the FTIR spectra and XRD
with pharmaceutical raw materials at room temperature and
heat treated at 190 °C. For heating, samples were placed on
a Kline plate with 12 excavation and then heated on a
Quimis drying oven model Q317 M in the heating rate of
5°C min
-1
to the temperature of 190 °C.
Results and discussion
Thermal behavior of HQ
In Fig. 1a, the DSC curve of HQ is presented revealing
important information. An analysis of the DSC curve of
HQ confirmed the presence of two endothermic events
characteristic of the thermal behavior of this material. The
DSC curve of HQ showed a sharp endothermic event
(event 1) between 173 and 179 °C indicating the melting
(T
onset
173 °C, T
peak
176 °C, DH258 J g
-1
). This peak
corresponds to the HQ melting, and it is consistent with the
melting range values described in the literature (between
172 and 174 °C) [20]. The second event observed (event 2)
in the DSC curve was also an endothermic one between
180 and 246 °C(T
onset
180 °C, T
peak
244 °C).
The TG/DTG curve (Fig. 1a) showed that HQ was
thermally stable up to 135 °C and then one mass loss stage
could be observed. The mass loss (98.2 % decrease)
occurred in the temperature range from 135 to 226 °C, with
T
onset
TG 149 °C. The DTG presented one peak (T
onset
DTG 172 °C, T
peak
DTG 217 °C) corresponding to the
mass loss observed in the TG.
Thermal and non-thermal techniques 721
123
The DTA curve (Fig. 1a) showed the first endothermic
event (event 3) between 173 and 184 °C indicating the
melting (T
onset
173 °C, T
peak
176 °C, DH1,270 J g
-1
) and
the second one (event 4—also endothermic) between 185
and 244 °C(T
onset
185 °C, T
peak
219 °C,
DH1,940 J g
-1
).
30
20
10
0
–10
Heat flow/mW
300
200
100
0
–100
Temperature difference/μV
50 100 150 200 250 300
Temperature/°C
100
80
60
40
20
–0
Mass loss/%
4
2
0
–2
–4
First derivate/mg min–1
60
40
20
40
20
0
Transmitance/% Transmitance/%
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
7000
7000
0
0
Intensity/a.u. Intensity/a.u.
10 20 30 40 50 60 70 80
10 20 30 40 50 60 70 80
2
θ
/°
HQ at room temperature
HQ heated at 190 °C
HQ at room temperature
HQ heated at 190 °C
DSC
DTG
DTA
2
1
4
3
Endo
HO OH
ab
c
TG
Fig. 1 a DSC, TG/DTG, and
DTA curves of HQ; bX-ray
diffractogram of HQ at room
temperature and heated at
190 °C; cFTIR spectra of HQ at
room temperature and heated at
190 °C
20
0
–20
–40
Endo
Endo
Endo
Heat flow/mW
Heat flow/mW
Heat flow/mW
20
0
–20
–40
Endo Heat flow/mW
50 100 150 200 250 300
Temperature/°C
50 100 150 200 250 300
Temperature/°C
50 100 150 200 250 300
Temperature/°C
50 100 150 200 250 300
Temperature/°C
10
0
–10
–20
–30
10
0
–10
–20
–30
HQ
HQ/CA
HQ/CTA
HQ/DPG
HQ/EDTA
HQ/GLY
HQ
HQ/HPMC
HQ/IMD
HQ/MTP
HQ/DCO
HQ/PPP
HQ
HPMC
IMD
MTP
DCO
PPP
HQ
CA
CTA
DPG
EDTA
GLY
ab
d
c
Fig. 2 DSC curves of all substances used in compatibility study: aand bHQ and excipients, cand dHQ and its 1:1 (m/m) mixtures
722 I
´. P. de Barros Lima et al.
123
Compatibility study with excipients
The selection of adequate excipients for formulation should
be based on the drug characteristic and its compatibility and
stability with other components [21]. Figures 2a, b and 3a, b
show DSC and DTA curves of all substances isolated (HQ
and excipients) used in the compatibility study. The DSC
and DTA curves of HQ-excipients mixtures are shown in
Figs. 2c, d and 3c, d.
The DSC curve of HQ/CA BM (Fig. 2c) shows a first
characteristic peak of the CA excipient (occurring between
50 and 55 °C) and two other events characteristic of HQ.
The melting peak of HQ is observed in this BM at tem-
peratures (T
onset
, and T
peak
—as can be seen in Table 2)
very similar to those temperatures of HQ alone, and the
DHinvolved in this event has its value reduced from 258 to
157 J g
-1
(Table 2). Analyzing the DTA curve of HQ/CA
BM (Fig. 3c), a similar behavior to that found in the DSC
0
0
–100
–200
–300
–400
–500
Temperature differance/μV
–100
–200
–300
–400
Temperature differance/μV
0
–100
–200
–300
–400
Temperature differance/μV
0
–100
–200
–300
–400
Temperature differance/μV
50 100 150 200 250 300
Temperature/°C
50 100 150 200 250 300
Temperature/°C
50 100 150 200 250 300
Temperature/°C
50 100 150 200 250 300
Temperature/°C
Endo
Endo
Endo Endo
HQ
CA
CTA
DPG
EDTA
GLY
HQ
HPMC
IMD
MTP
DCO
PPP
HQ
HQ/HPMC
HQ/IMD
HQ/MTP
HQ/DCO
HQ/PPP
HQ
HQ/CA
HQ/CTA
HQ/DPG
HQ/EDTA
HQ/GLY
ab
dc
Fig. 3 DTA curves of all substances used in compatibility study: aand bHQ and excipients, cand dHQ and its 1:1 (m/m) mixtures
Table 2 Temperatures and DH values of melting peak for physical
mixtures in DSC curve
Excipient in 1:1 mixture T
onset
/°CT
pico
/°CDH/J g
-1
HQ/CA 172 175 157
HQ/CTA 173 175 58
HQ/DPG 165 167 14
HQ/EDTA 175 177 166
HQ/GLY 174 184 60
HQ/HPMC* – – –
HQ/IMD* – – –
HQ/MTP* – – –
HQ/DCO 175 177 67
HQ/PPP* – – –
HQ 173 176 258
* Physical mixtures in which the melting peak of HQ was not
observed
Table 3 Temperatures and DH values of melting peak for physical
mixtures in DTA curve
Excipient in 1:1 mixture T
onset
/°CT
pico
/°CDH/J g
-1
HQ/CA 168 172 316
HQ/CTA 167 173 353
HQ/DPG* 170 214 3,110
HQ/EDTA 171 175 564
HQ/GLY* 191 228 4,120
HQ/HPMC* 175 213 1,440
HQ/IMD 152 164 845
HQ/MTP* 194 224 2,640
HQ/DCO 171 175 524
HQ/PPP 141 158 207
HQ 173 176 1,270
* Presence of a single event with the junction of two characteristics
events
Thermal and non-thermal techniques 723
123
curve of the same BM is observed, where a first charac-
teristic peak of the excipient and two other events of HQ.
In the DTA curve, the melting peak of HQ is also observed
in this BM at very similar temperatures to those of HQ
alone, but the DHinvolved in this event has its value
greatly reduced (Table 3).
In the DSC curve of HQ/CTA BM (Fig. 2c), the melting
peak of HQ occurred in this BM at temperatures (T
onset
,
and T
peak
—Table 2) very similar to those temperatures of
HQ alone, but the DHinvolved in this event has its value
strongly reduced from 258 to 58 J g
-1
(Table 2). In the
DTA curve of HQ/CTA BM (Fig. 3c), the melting peak of
HQ occurred also in this BM at very similar temperatures
to those of HQ alone, but the DHinvolved in this event has
its value strongly reduced from 1,270 to 353 J g
-1
(Table 3).
The similarity in the thermal behavior of CA and CTA
in this compatibility study can be explained by the simi-
larity in the chemical structures of these pharmaceutical
excipients, which also have the same function (emollient;
emulsifying) in semi-solid pharmaceutical forms.
Evaluating the DSC and DTA curves of HQ/EDTA BM
(Figs. 2c, 3c), the melting peak of HQ happened in this BM
at temperatures (T
onset
, and T
peak
—Tables 2,3) very similar
to those of HQ alone.
In the 1:1 physical mixtures, where there is no any
interaction between substances, the stages of heat flow
should remain virtually unchanged, similarly to when the
drug is alone. Through the comparison of the DSC and
DTA curves of HQ, CA, CTA, EDTA, and the BM of these
compounds, it was concluded for a little change in the
thermal stability of HQ in those mixtures (slight changes in
T
onset
, and T
peak
), suggesting no interaction between these
substances, even with some reductions in the DHinvolved
in the characteristic events of HQ.
In the DSC curve, the DHfor the HQ corresponds to
258 J g
-1
, the value expected in a 1:1 mixture is close to
129 J g
-1
and in the DTA curve, the DHcorresponds to
1,270 J g
-1
, the value expected in a 1:1 mixture is close to
635 J g
-1
; however, reductions isolated in the DHwithout
significant changes in temperatures (T
onset
, and T
peak
) not
necessarily represent drug–excipient interaction. The small
variations in the enthalpy’s values for BM can be attributed
to some heterogeneity in the small samples used for the
DSC experiments (3–4 mg).
Data from the DSC and DTA curves of these substances
were very similar, showing substantial complementarity of
these thermoanalytical techniques.
However, when evaluating the DSC curve of HQ/DPG
BM (Fig. 2c), the melting peak of HQ on this BM has its
temperatures (T
onset
and T
peak
—Table 2) anticipated to
lower values. The DHof the melting of HQ in the BM has
its value reduced to an appreciable extent from 258 to
14 J g
-1
(Table 2). When evaluating the DTA curve of
HQ/DPG BM (Fig. 3c), there is a junction (overlapping) of
two events characteristic of the drug, showing a single
event. This event presents T
onset
virtually unchanged, but
the T
peak
is shifted to higher temperature, undergoing the
T
peak
from 176 to 214 °C (Table 3). With respect to the
DHinvolved, there is increase in their values from 1,270 to
3,110 J g
-1
(Table 3), which can be explained only by the
junction of events in this BM.
With the objective of better explain the processes
occurring in this HQ/DPG BM, Fig. 4a shows the TG/DTG
curves of this BM. The TG curve of HQ/DPG (Fig. 4a)
shows one mass loss stage has its T
onset
anticipated to lower
values (from 182 to 167 °C), and in the DTG curve of this
BM (Fig. 4a), there is anticipation of T
onset
and T
peak
to
lower values.
Thus, the anticipation to lower values of melting peak of
HQ in the DSC curve of the HQ/DPG BM, allied to the
combination of events in the DTA curve, and together with
anticipation of T
onset
and T
peak
to lower values in the TG/
DTG curves, is indicative of interaction of HQ with DPG.
These data obtained from the thermal curves of HQ/
DPG mixture indicate a possible interaction with this
excipient, which can be attributed to the almost complete
drug dissolution in liquid pharmaceutical excipient. The
same behavior was observed in the compatibility study of
Salvio Neto et al. [17], where due to the complete drug
dissolution in the melt of the excipient, no melting of
prednicarbate was observed in the DSC curve of the
prednicarbate/stearyl alcohol mixture.
In regards to the DSC curve of HQ/GLY BM (Fig. 2c),
the melting peak of HQ on this BM has its temperatures
(Table 2) shifted to higher values, mainly T
peak
, but the
DHinvolved in this event has its value strongly reduced
from 258 to 60 J g
-1
(Table 2). When observing the DTA
curve of HQ/GLY BM (Fig. 3c), it is possible to see a
junction (overlap) of two characteristic events of the HQ,
showing a single event. This event presents the tempera-
tures shifted to higher temperatures, undergoing the T
onset
from 173 to 191 °C and T
peak
from 176 to 228 °C. With
respect to the energies involved, there is increased values
from 1,270 to 4,120 J g
-1
(Table 3), which can be
explained only by the junction of events in this BM.
In the same way as for the HQ/DPG, with the objective
of explain the processes occurring in HQ/GLY BM, Fig. 4b
shows the TG/DTG curves of this BM. The TG curve of
HQ/GLY (Fig. 4b) shows one mass loss stage has its T
onset
slightly shifted to higher temperature, from 182 to 184 °C,
and in the DTG curve (Fig. 4b), there is also displacement
to higher temperature, T
onset
DTG from 172 to 182 °C and
T
peak
DTG from 217 to 226 °C.
Thus, DSC and TG/DTG curves of HQ/GLY BM do not
show shifts in these curves of HQ for lower temperatures,
724 I
´. P. de Barros Lima et al.
123
indicating no interaction between these substances through
these techniques, but the junction of events in the DTA
curve of the HQ/GLY BM indicates that the other analyt-
ical method must be used for this mixture aiming at
investigating whether there is interaction between HQ and
GLY.
In the DSC curve of HQ/HPMC BM (Fig. 2d), no
characteristic events of HQ were observed, including the
melting peak of HQ which was absent. When analyzing the
DTA curve of HQ/HPMC BM (Fig. 3d), it is possible to
observe a junction (overlap) of two events characteristic of
the HQ, showing a single event. This event presents the
temperatures shifted to higher temperatures, undergoing
the T
onset
from 173 to 175 °C and T
peak
from 176 to 213 °C,
and the DHinvolved in this event has its value increased
from 1,270 to 1440 J g
-1
(Table 3).
The TG curve of HQ/HPMC (Fig. 4c) shows two mass
loss stages, being the first stage corresponding to the
decomposition of HQ, and has its T
onset
slightly anticipated
to lower values, and in the DTG curve of this BM (Fig. 4c),
there is two peaks, the first corresponding to the drug
decomposition, and in this peak, slight anticipation of T
onset
and T
peak
to lower values, T
onset
DTG from 172 to 162 °C,
and T
peak
DTG from 217 to 214 °C happens.
The absence of characteristic events of HQ in the DSC
curve of the HQ/HPMC BM, allied to the junction of
events in the DTA curve, is indicative of interaction of HQ
with HPMC.
In the DSC curve of HQ/IMD BM (Fig. 2d), no char-
acteristic events of HQ were observed, including the
melting peak of HQ, which was absent. For the DTA curve
of HQ/IMD BM (Fig. 3d), the melting peak of HQ is
markedly anticipated to lower temperatures, mainly T
onset
which was reduced from 173 to 152 °C (Table 3), corre-
sponding to a reduction of more than 10 % of T
onset
of HQ
alone. The DHinvolved in this event has its lower value
change from 1,270 to 845 J g
-1
(Table 3).
The TG curve of HQ/IMD (Fig. 4d) shows three mass
loss stages, being the first stage corresponding to the
decomposition of HQ, and has its T
onset
slightly anticipated
to lower values, and in the DTG curve of this BM (Fig. 4d),
there is two peaks, the first corresponding to the drug
decomposition and the second to the IMD. In the HQ/IMD,
decomposition peak happens strong anticipation of T
peak
to
lower value, T
peak
DTG from 217 to 165 °C.
Thus, the absence of characteristic events of HQ in the
DSC curve of the HQ/IMD BM, combined with extensive
anticipation of the melting peak of HQ in DTA curve,
allied to the anticipation of T
onset
to lower values in the TG
curve, and disappearance of the HQ peak in the DTG curve
are indicative of strong interaction of HQ with IMD.
The evaluation of the DSC and DTA curves of HQ/DCO
BM (Figs. 2d, 3d), shows the melting peak of HQ hap-
pened in this BM at very similar temperatures to those of
HQ alone, but the DHinvolved in this event has its value
reduced (Tables 2,3). Through the comparison of the DSC
and DTA curves of these compounds, it was concluded that
the thermal stability of HQ in this mixture changed a little,
suggesting no interaction between these substances.
The DSC curve of HQ/MTP BM (Fig. 2d) assumes
practically the same behavior of the DSC curve of MTP,
and no characteristic events of HQ were observed,
including the melting peak of HQ, which was absent.
However, there is the presence of an event between 110
100
50
0
Mass loss/%
100 200 300 400 500 600 700 800 900
Temperature/°C
100 200 300 400 500 600 700 800 900
Temperature/°C
100 200 300 400 500 600 700 800 900
Temperature/°C
100 200 300 400 500 600 700 800 900
Temperature/°C
100 200 300 400 500 600 700 800 900
Temperature/°C
100 200 300 400 500 600 700 800 900
Temperature/°C
100
50
0
Mass loss/%
100
50
0
Mass loss/%
100
50
0
Mass loss/%
100
50
0
Mass loss/%
100
50
0
Mass loss/%
0
–5
–10
First derivate/mg min–1
0
–5
–10
First derivate/mg min–1
0
–5
–10
First derivate/mg min–1 First derivate/m
g
min–1
First derivate/mg min–1
0
–5
–10
First derivate/mg min–1
0
–1
–2
–3
–4
0
–2
–4
–6
–8
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
HQ
IMD
IMD
MTP
MTP
PPP
PPP
HQ/PPP
HQ/PPP
HQ/HPMC
HQ/HPMC
HPMC
HPMC
GLY
GLY
HQ/GLY
HQ/GLY
HQ/MTP
HQ/MTP
HQ/DPG
HQ/DPG
HQ/IMD
HQ/IMD
DPG
DPG
abc
def
Fig. 4 TG/DTG curves of possible interactions of HQ with the excipients: aHQ, DPG, and HQ/DPG; bHQ, GLY, and HQ/GLY; cHQ, HPMC,
and HQ/HPMC; dHQ, IMD, and HQ/IMD; eHQ, MTP, and HQ/MTP; fHQ, PPP, and HQ/PPP
Thermal and non-thermal techniques 725
123
and 120 °C, which corresponds to a characteristic peak of
MTP. By analyzing the DTA curve of HQ/MTP BM
(Fig. 3d), it is possible to observe an event happening
between 109 and 123 °C, which corresponds to a charac-
teristic peak of MTP. After this, it is possible to observe a
junction (overlap) of two events characteristic of the HQ,
showing a single event. This event presents the tempera-
tures shifted to higher temperatures, undergoing the T
onset
from 173 to 194 °C and T
peak
from 176 to 224 °C, and the
DHinvolved in this event has its value increased from
1,270 to 2,640 J g
-1
(Table 3).
The TG curve of HQ/MTP (Fig. 4e) shows one mass
loss stage, it has its T
onset
slightly shifted to higher tem-
perature from 182 to 183 °C, and in the DTG curve
(Fig. 4e,) there is also displacement to higher temperature,
T
onset
DTG from 172 to 175 °C and T
peak
DTG from 217 to
223 °C.
From these considerations, the absence of characteristic
events of HQ in the DSC curve of the HQ/MTP BM,
combined with the junction of events in the DTA curve, is
indicative of interaction of HQ with MTP.
In the study of Lira et al. [22], the same behavior was
observed with MTP; there was an appreciable downward
shift of the drug peak temperature in the thermal curve of
the BM of lapachol/MTP, which can be indicative of some
drug–excipient solid interaction. DSC curves showed that
the peak at around 139 and 124 °C, which was observed for
lapachol and MTP, respectively, disappeared in the eutectic
mixture.
The DSC curve of HQ/PPP BM (Fig. 2d) assumes
practically the same thermal behavior of the DSC curve of
MTP, and no characteristic events of HQ were observed,
including the melting peak of HQ. By comparing the DTA
curves (Fig. 3d) of pure HQ and PPP with their 1:1 phys-
ical mixture, a first peak that is characteristic of PPP
(between 89 and 103 °C) and two other characteristic
events of HQ are observed. For the DTA curve of HQ/PPP
BM (Fig. 3d), the melting peak of HQ is markedly antic-
ipated to lower temperatures, T
onset
which was reduced
from 173 to 141 °C and T
peak
reduced from 176 to 158 °C
(Table 3), corresponding to a reduction of more than 15 %
of T
onset
of HQ alone. The DHinvolved in this event has its
value considerably reduced from 1,270 to 207 J g
-1
(Table 3).
The TG curve of HQ/PPP (Fig. 4f) shows one mass loss
stage, and has its T
onset
slightly shifted to higher tempera-
ture, from 182 to 183 °C, and in the DTG curve (Fig. 4f),
there is also displacement to higher temperature, T
onset
DTG from 172 to 175 °C and T
peak
DTG from 217 to
224 °C.
From these considerations, the absence of characteristic
events of HQ in the DSC curve of the HQ/PPP BM,
combined with extensive anticipation of the melting peak
of HQ in DTA curve, is indicative of interaction of HQ
with PPP.
In another study of compatibility [22], there was no
interaction between the PPP and the API (lapachol), whose
the physical mixture of lapachol and PPP in the DSC curve
can be considered to be the superposition of the DSC
curves of the two individual components. These results
show that physical interactions of components did not
occur within the mixture. Photovisual DSC was used to
confirm the results obtained by conventional DSC. These
data are not supported by the results found in our com-
patibility study with HQ/PPP BM.
Another analytical technique used in this study was the
FTIR spectroscopy that is a simple technique for the
detection of changes within drug–excipient mixtures.
The FTIR spectra of HQ at room temperature (Fig. 1c)
showed the characteristic absorption bands of 1,4-disub-
stituted mononuclear aromatic ring and symmetrical sec-
ondary phenol group. The broad and medium intensity
band at 3,269 cm
-1
is due to the hydrogen bonded OH
stretching vibration. The aromatic overtones and combi-
nation bands appearing from 2,010 to 1,740 cm
-1
region
confirm the presence of aromatic system. The intense band
at 2,856 cm
-1
represents C–H stretching vibration. The
bands at 1,635 and 1,516 cm
-1
are consistent with the
skeletal vibrations of the aromatic system. The bands at
1,388 and 1,353 cm
-1
are due to the O–H in-plane bending
vibrations. The absorptions at 1,095 cm
-1
are due to C–O
stretching vibrations. The substitution pattern is established
by strong C–H out-of-plane bending absorptions at 827 and
759 cm
-1
.
These data of FTIR spectrum are consistent with those
reported in the literature [23], confirming that the substance
is HQ and not indicating the presence of any other species.
The FTIR spectra at room temperature of HQ, excipients
and its 1:1 (m/m) physical mixtures with DPG, GLY,
HPMC, IMD, MTP, and PPP are shown in Fig. 5. In this
study, the analyses were performed in the FTIR spectra
with the excipients mentioned above because in the ther-
moanalytical trials, we observed possible interactions in the
BM of HQ with these excipients, thus the FTIR spectra acts
as a complementary method for the evaluation of possible
interactions between components.
The evaluation of the FTIR spectra of 1:1 (m/m) phys-
ical mixtures between HQ/DPG, HQ/GLY, HQ/HPMC,
HQ/MTP, and HQ/PPP (Fig. 5a, b, c, e, and f, respec-
tively), clearly shows that the spectra of these BM are the
result of the sum of the characteristic bands of individual
components. However, note for the five mixtures (HQ/
DPG, HQ/GLY, HQ/HPMC, HQ/MTP, and HQ/PPP) in
the spectra cited above that the absorption bands of the
excipients may be overlapping the skeletal vibrations of the
aromatic system of HQ at 1.635 cm
-1
without any
726 I
´. P. de Barros Lima et al.
123
suggestion of interaction. Therefore, no chemical interac-
tions occur between these substances at room temperature.
In another study of compatibility [16], there was inter-
action between the propylene glycol excipient and the
active ingredient (lipoic acid), whose active and excipient
bands in the DSC scan of lipoic acid/propylene glycol
mixture were missing. DSC results points toward some
incompatibility between lipoic acid/propylene glycol mix-
ture. Furthermore, IR spectra of the lipoic acid/propylene
glycol blend did not show characteristic bands of lipoic
acid at 3,030 and 945 cm
-1
, but presented a reduction of
intensity of the band at 1,250 cm
-1
. Thus, the results
obtained in the compatibility study of HQ with DPG are
not in agreement with the data found in the aforementioned
compatibility study.
According to the results of the thermal studies, some
changes in the FTIR spectra of mixtures of HQ, and some
of the excipient suggested a possible interaction between
the mixtures components, in agreement with the thermal
analysis findings. For instance, FTIR spectra of HQ/IMD at
room temperature (Fig. 5d) does not show the bands at
3,269 cm
-1
(relating to the hydrogen bonded OH stretch-
ing vibration), at 1,740 cm
-1
(relating to the aromatic
system), and at 1,635 cm
-1
(consistent with the skeletal
vibrations of the aromatic system). Besides the disappear-
ance of these bands mentioned above, there is the
appearance of new bands at 1,725 and 1,672 cm
-1
. Then,
there is a marked change in the FTIR spectra of HQ in the
mixture with the IMDto get a clear evidence for chemical
interaction between the HQ and IMD at room temperature.
These results of FTIR are in agreement with reports in
the literature indicating that the disappearance of an
absorption peak or reduction of the peak intensity com-
bined with the appearance of new peaks give a clear evi-
dence for interactions between the excipient and the drug
investigated [24].
In the compatibility study of Moyano et al. [16], there
was no interaction between the pharmaceutical pre-
servative (IMD) and the API(lipoic acid), where the
melting of lipoic acid was well retained in the DSC scan of
lipoic acid/IMD mixture, the band corresponding to lipoic
acid was observed without any new bands. Furthermore, IR
spectra of lipoic acid and its blend with the above men-
tioned excipient IMD showed the presence of characteristic
bands corresponding to lipoic acid. Thus, the results
obtained in the compatibility study of HQ with IMD are not
in agreement with the data found in the compatibility study
cited above.
Studies are constantly being conducted on the elabora-
tion of efficient methods to confirm the compatibility of
API and excipients, and scholars have continued to find
new more effective methods in the identification of
incompatibilities. Therefore, the analysis of the FTIR
spectra of drug/excipient BM was also conducted by a
quantitative method of Pearson’s correlation (r), the sta-
tistical methodology, generated by a computer that math-
ematically calculated the difference between the theoretical
and experimental FTIR spectra. We propose that the more
the computed correlations deviate from the unit (1.00), the
more drug–excipient interactions are present.
60
30
40
20
20
0
4000 3500 3000 2500 2000 1500 1000 500
Transmittance/%
60
30
40
20
10
20
0
0
Transmittance/%
60
30
40
60
20
30
10
Transmittance/%
Transmittance/%
60
30
40
20
20
0
0
Transmittance/%
60
30
30
30
0
0
Transmittance/%
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
60
30
30
30
HQ at room temperature HQ at room temperature HQ at room temperature
HQ at room temperature
HQ at room temperature
HQ at room temperature
DPG at room temperature
HQ/DPG at room temperature
GLY at room temperature
HQ/GLY at room temperature
HPMC at room temperature
HQ/HPMC at room temperature
PPP at room temperature
HQ/PPP at room temperature
MTP at room temperature
HQ/MTP at room temperature
IMD at room temperature
HQ/IMD at room temperature
ab c
de f
Fig. 5 FTIR spectra of HQ, excipients and its 1:1 (m/m) mixtures at room temperature: aHQ, DPG, and HQ/DPG; bHQ, GLY, and HQ/GLY;
cHQ, HPMC, and HQ/HPMC; dHQ, IMD, and HQ/IMD; eHQ, MTP, and HQ/MTP; fHQ, PPP, and HQ/PPP
Thermal and non-thermal techniques 727
123
Wesolowski et al. [25], conducted an evaluation of the
utility of the two chemometric methodologies, hierarchical
cluster analysis (HCA) and principal component analysis
(PCA), as supporting techniques for the identification of
potential incompatibilities that can occur in the pre-for-
mulation stage of a solid dosage drug form. The investi-
gation performed with the use of baclofen and selected
excipients has shown that with thermogravimetric analysis,
HCA, and PCA fulfill their role as supporting techniques in
the interpretation of the data obtained.
The Fig. 6shows the results of such analysis. It can be
observed that the HQ/IMD mixture (r
1
=0.7301) was the
sample that mostly deviated from the ideal correlation,
indicating moderate correlation between the FTIR spectra.
The literature [26,27] reported that the Pearson’s corre-
lation (r) close to unity (1.00) shows that the variables
being compared are similar. When the value of r is between
0.80 and 1.00, indicating high correlation, while between
0.5 and 0.80 shows moderate correlation, and r less than
0.50 indicates a low correlation between variables
compared.
This can be related to a chemical reaction of HQ and
IMD molecules resulting in a greater difference in the
FTIR spectra, showing that there is evidence that the IMD
released formaldehyde and thus oxidized the HQ.
The second most unstable mixture is the HQ/GLY
(r
2
=0.8399), followed by HQ/DPG (r
3
=0.9050), and
HQ/PPP (r
4
=0.9329). The FTIR spectra of HQ/HPMC
(r
5
=0.9877) and HQ/MTP (r
6
=0.9899) showed almost
perfect correlation with the theoretical FTIR spectra of HQ.
The Pearson’s correlation showed a small deviation of
FTIR spectrum of HQ, indicating a high correlation
between the theoretical and experimental FTIR spectra,
confirming no chemical interaction between the HQ and
the excipients mentioned above.
Thus, the Pearson’s correlation provided a more precise
mathematical analysis of the FTIR spectra of the samples
analyzed in this compatibility study. The advantage of the
FTIR spectra correlation analysis in comparison to the TA
is the lack of heat-induced alterations in the mixture.
The HQ was heated at 190 °C to be analyzed by FTIR
aiming at evaluating the chemical stability of the drug
depending on the heating, and this temperature (after the
melting point of HQ) was selected based on thermal
characterization performed by DSC and DTA curves of the
drug alone. The FTIR spectra of HQ heated at 190 °C
showed (Fig. 1c) virtually the same profile FTIR spectra of
HQ at room temperature, except for the absorption bands at
1,635 and 1,618 cm
-1
that became overlapping with the
heating, thus showing only one band in 1,637 cm
-1
.
In the study of Fulias et al. [28], a heating of pharma-
ceutical substances (phenazone, and phenylbutazone) was
conducted followed by FTIR analysis aiming at charac-
terizing such samples at room temperature and treated with
heating.
In compatibility study of HQ were conducted also
heating at 190 °C of excipients and BM of HQ with ex-
cipients to perform FTIR analysis, with the aim of evalu-
ating the influence of heating on chemical interactions
between the drug and excipient. The BM selected for the
heating at 190 °C were those that showed incompatibilities
in thermal and FITR analysis at room temperature. The
FTIR spectra of HQ, excipients and its 1:1 (m/m) physical
mixtures with excipients heated at 190 °C are shown in
Fig. 7.
In the study of Lima et al. [29], heating of BM of the
trioxsalen drug with various pharmaceutical excipients was
performed, and then the FTIR analyses were performed
with the aim of investigating the influence of heating in the
appearance of chemical interactions in the drug–excipient
mixtures.
The evaluation of the FTIR spectra of physical mixtures
heated at 190 °C between HQ/DPG, HQ/GLY, HQ/MTP,
and HQ/PPP (Fig. 7a, b, e, and f, respectively), clearly
shows that the spectra of these BM are the result of the sum
of the characteristic bands of individual components, being
in the same manner as observed for the same mixtures at
room temperature. However, noted for the two mixtures
(HQ/MTP and HQ/PPP) the disappearance of the absorp-
tion band (at 1,637 cm
-1
) of the skeletal vibrations of the
aromatic system of HQ heated at 190 °C, without any
suggestion of interaction. Thus, indicating that no chemical
interactions occur between these substances heated at
190 °C.
The FTIR spectra of HQ/HPMC BM heated at 190 °C
(Fig. 7c) show a change in behavior with the heating,
1.00
0.95
0.90
0.85
0.80
0.75
0.70
0.65
0.60
0.55
0.50
Pearson′s correlation
HQ/IMD
HQ/GLY
HQ/DPG
HQ/PPP
HQ/HPMC
HQ/MTP
Fig. 6 Plot of the Pearson’s correlation (r) between the theoretical
and experimental HQ/excipient FTIR spectra
728 I
´. P. de Barros Lima et al.
123
because there is mainly the disappearance of the absorption
bands at 1,858 cm
-1
(which indicates the presence of the
aromatic system), 1,354 cm
-1
(O–H in-plane bending
vibrations) and 1,096 cm
-1
(C–O stretching vibrations).
Thus, it can be said that the heating to 190 °C provides the
appearance of a chemical interaction between HQ and
HPMC, which was not observed at room temperature.
The FTIR spectra of HQ/IMD heated at 190 °C
(Fig. 7d) besides not show bands at 3,269, 1,740, and
1,637 cm
-1
as also was observed at room temperature, the
heating promoted the disappearance of the bands at 2,856
and 2,010 cm
-1
confirming the chemical interaction with
the heating, which was also observed at room temperature.
Another analytical technique used in this study was the
XRD, a direct measure of the crystal form of a material
being a plot of intensity versus the diffraction angle (2h). A
crystalline material exhibits a unique set of diffraction
peaks and the lack of crystalline API peaks, when a dosage
form is analyzed could indicate that the material is amor-
phous or that the loading is too low to detect using the
parameters chosen. XRD analysis is of immense help in
case of incompatibilities which can occur during processes
for selection of suitable excipients [30].
The XRD patterns of HQ at room temperature and HQ
heated at 190 °C have been represented in Fig. 1b. The
XRD for HQ at room temperature, exhibited sharp char-
acteristic peaks, revealing its high crystallinity, which can
be used as a fingerprint. The diffraction peak corresponding
to the highest intensity was observed at 2h°value at 20.25°.
Also, other important diffraction peaks were observed at
9.42°, 16.07°, 16.27°, 16.84°, 21.36°, and 30.60°.
As said before, the HQ was heated at 190 °Ctobe
analyzed by XRD aiming to evaluate the crystallinity of the
drug as a function of heating. Thus, the Fig. 1b shows that
in the HQ heated at 190 °C occurs a slight change at 2h°
value (in the pattern of crystallinity) compared to the HQ at
room temperature, because with the heating, the main
diffraction peaks can be observed in the sample heated, but
with a reduction in the intensity of some peaks (mainly at
9.42°), however, do not show loss of crystallinity in the HQ
heated.
The XRD pattern of HQ, excipients and its 1:1 (m/m)
physical mixtures with excipients at room temperature are
shown in Fig. 8. The BM selected for the XRD were those
that shown incompatibilities in the thermoanalytical trials
and FITR analysis.
The evaluation of the XRD of mixtures between HQ/
DPG at room temperature (Fig. 8a) clearly shows that the
important diffraction peaks (as in 2h°values of 9.42°,
16.07°, 16.27°, 16.84°, 20.25°, and 21.36°), which are
characteristic of crystallinity of HQ, remain practically
unchanged in this BM, except for the peak at 30.6°which
practically disappears in the mixture. However, when the
XRD of HQ/DPG heated at 190 °C is analyzed (Fig. 9a),
the disappearance of almost all important diffraction peaks
of HQ is observed, and there is the formation of a new peak
at 21.36°(high intensity). Thus, the XRD of HQ/DPG
indicates no interaction between these substances at room
temperature, but the heating modifies the HQ crystallinity.
The XRD of HQ/GLY at room temperature (Fig. 8b)
clearly shows that the important diffraction peaks, which
are characteristic of HQ crystallinity, remain practically
40
20
40
20
40
20
0
0
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1 4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
4000 3500 3000 2500 2000 1500 1000 500
Wavenumber/cm–1
Transmittance/%
40
20
40
20
40
20
0
0
Transmittance/%
40
20
40
20
20
0
0
Transmittance/%
40
20
40
20
20
0
0
0
Transmittance/%
40
20
20
20
0
0
Transmittance/%
40
20
40
60
20
40
20
0
0
0
Transmittance/%
HQ at 190 °C HQ at 190 °C HQ at 190 °C
HQ at 190 °C
HQ at 190 °C
HQ at 190 °C
IMD at 190 °C
HQ/IMD at 190 °C
MTP at 190 °C
HQ/MTP at 190 °C
PPP at 190 °C
HQ/PPP at 190 °C
HPMC at 190 °C
HQ/HPMC at 190 °C
GLY at 190 °C
HQ/GLY at 190 °C
DPG at 190 °C
HQ/DPG at 190 °C
abc
fed
Fig. 7 FTIR spectra of HQ, excipients and its 1:1 (m/m) mixtures heated at 190 °C: aHQ, DPG, HQ/DPG; bHQ, GLY, HQ/GLY; cHQ,
HPMC, HQ/HPMC; dHQ, IMD, HQ/IMD; eHQ, MTP, HQ/MTP; fHQ, PPP, HQ/PPP
Thermal and non-thermal techniques 729
123
unchanged in this BM, except for the peak at 28.91°that
increases its intensity in the mixture. The XRD of HQ/GLY
heated at 190 °C (Fig. 9b) can be considered as the overlap
of individual components without absence, shift or broad-
ening substantial of the peaks of HQ heated. Therefore, the
XRD of HQ/GLY indicates no interaction between these
substances at room temperature, and heated.
The XRD of HQ/HPMC and HQ/IMD at room tem-
perature (Fig. 8c, d, respectively), clearly show that the
important diffraction peaks, remain practically unchanged
in these BM, showing the overlap of individual compo-
nents without absence, shift or broadening substantial of
the peaks of HQ. In the XRD of HQ/HPMC and HQ/IMD
heated at 190 °C (Fig. 9c, d, respectively), the
6000
0
3000
3000
0
0
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
Intensity/a.u.
6000
0
2000
4000
0
0
Intensity/a.u.
6000
0
3000
3000
0
0
Intensity/a.u.
6000
0
1000
5000
0
0
Intensity/a.u.
6000
0
6000
5000
0
0
Intensity/a.u.
6500
0
6500
11000
0
0
Intensity/a.u.
HQ at room temperature
DPG at room temperature
HQ/DPG at room temperature
HQ at room temperature
GLY at room temperature
HQ/GLY at room temperature
HQ at room temperature
HPMC at room temperature
HQ/HPMC at room temperature
HQ at room temperature
IMD at room temperature
HQ/IMD at room temperature
HQ at room temperature
MTP at room temperature
HQ/MTP at room temperature
HQ at room temperature
PPP at room temperature
HQ/PPP at room temperature
abc
def
Fig. 8 X-ray diffractogram of HQ, excipients and its 1:1 (m/m) mixtures at room temperature: aHQ, DPG, HQ/DPG; bHQ, GLY, HQ/GLY;
cHQ, HPMC, HQ/HPMC; dHQ, IMD, HQ/IMD; eHQ, MTP, HQ/MTP; fHQ, PPP, HQ/PPP
6000
0
3000
100000
0
0
Intensity/a.u.
a
6000
0
2000
3000
0
0
Intensity/a.u.
d
6000
0
3000
6000
0
0
Intensity/a.u.
b
6000
0
2000
6000
0
0
Intensity/a.u.
c
6000
0
60000
6000
0
0
Intensity/a.u.
6000
0
22000
6000
0
0
Intensity/a.u.
f
e
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
10 20 30 40 50 60 70 80
2
θ
/°
HQ at 190 °C
IMD at 190 °C
HQ/IMD at 190 °C
HQ at 190 °C
MTP at 190 °C
HQ/MTP at 190 °C
HQ at 190 °C
PPP at 190 °C
HQ/PPP at 190 °C
HQ at 190 °C
HPMC at 190 °C
HQ/HPMC at 190 °C
HQ at 190 °C
GLY at 190 °C
HQ/GLY at 190 °C
HQ at 190 °C
DPG at 190 °C
HQ/DPG at 190 °C
Fig. 9 X-ray diffractogram of HQ, excipients and its 1:1 (m/m) mixtures heated at 190 °C: aHQ, DPG, HQ/DPG; bHQ, GLY, HQ/GLY; cHQ,
HPMC, HQ/HPMC; dHQ, IMD, HQ/IMD; eHQ, MTP, HQ/MTP; fHQ, PPP, HQ/PPP
730 I
´. P. de Barros Lima et al.
123
disappearance of all important diffraction peaks of HQ
heated is observed. Thus, the XRD of HQ/HPMC and HQ/
IMD indicates no interaction between these substances at
room temperature, but the heating modifies the crystallinity
of the HQ, indicating interaction with heating between HQ/
HPMC and HQ/IMD.
The XRD of HQ/MTP and HQ/PPP at room temperature
(Fig. 8e, f, respectively) clearly show that the important
diffraction peaks, remain practically unchanged in these
BM, showing the overlap of individual components with-
out absence, shift or broadening substantial of the peaks of
HQ. However, in the XRD of HQ/MTP and HQ/PPP
heated at 190 °C (Fig. 9e, f, respectively), the disappear-
ance of important diffraction peaks of HQ at 16.07°,
16.27°, and 30.6°is observed. In the XRD of HQ/MTP
heated (Fig. 9e), there is the formation of a new peak at
23.07°(high intensity). Therefore, the XRD of HQ/MTP
and HQ/PPP indicate no interaction between these sub-
stances at room temperature, but the heating modifies the
HQ crystallinity, indicating interaction with heating
between HQ/MTP and HQ/PPP.
In the present study, the interactions suggested were
mainly observed in the results obtained by thermal tech-
niques (DSC, DTA, and TG/DTG), but in the mixture of
HQ with IMD, both by thermal as the non-thermal (FTIR,
and XRD) techniques, there was interaction between these
components, showing a strong correlation between results
obtained by these analytical techniques.
The literature shows that IMD is one of the most widely
used preservative system in the world; it is a broad-spec-
trum antimicrobial preservative used in cosmetics and
topical pharmaceutical formulations; typical concentrations
used are 0.03–0.5 % m/m. It is effective between pH 3–9
and is reported to have synergistic effects when used with
parabens (MTP and PPP) [31].
It is known that pharmaceutical excipients commonly
used in oral solid dosage forms might also be sources of
formaldehyde. The results found in the study of Fujita et al.
[32] showed that the formaldehyde is generated by the
excipients lactose, D-mannitol, microcrystalline cellulose,
low-substituted hydroxypropyl-cellulose, magnesium stea-
rate, and light anhydrous silicic acid. The quality and safety
of pharmaceutical products can be significantly affected by
the presence of formaldehyde.
The literature [31] emphasizes that the IMD excipient is
produced from the condensation of allantoin with formal-
dehyde. This study suggests that the IMD—when associ-
ated with HQ—releases formaldehyde, thus promoting
these marked changes in the physicochemical characteris-
tics of HQ, which in turn affect the quality and safety of
semi-solid pharmaceutical formulations containing this
depigmenting agent. Therefore, it is recommended that the
IMD excipient be replaced by another preservative system
in formulations containing HQ.
Conclusions
The results of the present study confirmed the utility and
reliability of thermoanalytical analysis at the earliest stage
of pre-formulation studies as a valuable tool for a rapid
screening of a wide range of candidate excipients,
allowing a rapid evaluation of possible drug–excipient
interactions.
No evidence of interaction was observed between HQ
and excipients CA, CTA, EDTA, and DCO. However,
based on the thermoanalytical results alone, a physical
interaction was suspected between HQ and excipients
DPG, GLY, HPMC, IMD, MTP, and PPP.
The results of the FTIR studies of the HQ and its mix-
ture with excipient IMD, at room temperature and heated at
190 °C, were consistent with TA experiments, where a
chemical interaction between HQ and IMD was observed.
For excipients DPG, GLY, HPMC, MTP, and PPP, there
were no chemical interactions with HQ at room tempera-
ture; however, it can be said that the heating at 190 °C
provides the appearance of a chemical interaction between
HQ and HPMC.
The chemometric methodology, Pearson’s correlation,
based on the quantitative analysis of FTIR spectra showed
that the IMD was the excipient that mostly deviated from
the ideal correlation (r=1.00).
The results of the XRD of mixtures between HQ and
excipients DPG, GLY, HPMC, IMD, MTP, and PPP at
room temperature show that the important diffraction peaks
of HQ alone, remain practically unchanged in these BM.
However, the heating at 190 °C changes the crystallinity of
the HQ for the mixtures with excipients DPG, HPMC,
IMD, MTP, and PPP, thus indicates that no interaction
occurs between these substances at room temperature, but
the heating modifies the crystallinity of the HQ.
In the present study, the interactions suggested were
mainly observed in the results obtained by thermal tech-
niques, but in the HQ/IMD mixture, both by thermal as the
non-thermal techniques, it was observed interaction
between these components. Therefore, it is recommended
to replace the excipient IMD by another preservative sys-
tem in semi-solid pharmaceutical formulations containing
HQ.
Acknowledgements The authors thank Conselho Nacional de De-
senvolvimento Cientı
´fico e Tecnolo
´gico (CNPq), Coordenac¸a
˜ode
Aperfeic¸oamento de Pessoal de Nı
´vel Superior (CAPES), and Fun-
dac¸a
˜o de Apoio a
`Pesquisa do Estado do Rio Grande do Norte
(FAPERN).
Thermal and non-thermal techniques 731
123
References
1. Rendon M, Berneburg M, Arellano I, Picardo M. Treatment of
melasma. J Am Acad Dermatol. 2006;54:272–81.
2. Draelos ZD. Skin lightening preparations and the hydroquinone
controversy. Dermatol Ther. 2007;20:308–13.
3. Tita B, Fulias A, Bandur G, Marian E, Tita D. Compatibility
study between ketoprofen and pharmaceutical excipients used in
solid dosage forms. J Pharm Biomed Anal. 2011;56:221–7.
4. Roumeli E, Tsiapranta A, Pavlidou E, Vourlias G, Kachrimanis
K, Bikiaris D, Chrissafis K. Compatibility study between
trandolapril and natural excipients used in solid dosage forms.
J Therm Anal Calorim. 2013;111:2109–15.
5. Tomassetti M, Catalani A, Rossi V, Vecchio S. Thermal analysis
study of the interactions between acetaminophen and excipients
in solid dosage forms and in some binary mixtures. J Pharm
Biomed Anal. 2005;37:949–55.
6. Bharate SS, Bharate SB, Bajaj AN. Interactions and incompati-
bilities of pharmaceutical excipients with active pharmaceutical
ingredients: a comprehensive review. J Excip Food Chem.
2010;1:3–26.
7. Hardy MJ. Drug–excipient compatibility prediction by DSC.
Anal Proc. 1982;19:556–7.
8. Smith A. Use of thermal analysis in predicting drug–excipient
interactions. Anal Proc. 1982;19:559–61.
9. Giron D. Applications of thermal analysis in the pharmaceutical
industry. J Pharm Biomed Anal. 1986;4:755–70.
10. Ke B. Differential thermal analysis of high polymers. II. Effects
of diluents on melting behavior of polyethylenes. J Polym Sci.
1961;50:79–86.
11. Soares-Sobrinho JL, Soares MFLR, Lopes PQ, Correia LP, Souza
FS, Mace
ˆdo RO, Rolim-Neto PJ. A preformulation study of a new
medicine for Chagas disease treatment: physicochemical char-
acterization, thermal stability, and compatibility of benznidazole.
AAPS Pharm Sci Technol. 2010;11(3):1391–6.
12. Costa SPM, Silva KER, Medeiros GCR, Rolim LA, Oliveira JF,
Lima MCA, Galdino SL, Pitta IR, Neto PJR. Thermal behavior
and compatibility analysis of the new chemical entity LPSF/FZ4.
Thermoch Acta. 2013;562:29–34.
13. Soares MFLR, Soares-Sobrinho JL, da Silva KER, Alves LDS,
Lopes PQ, Correia LP, Souza FS, Mace
ˆdo RO, Rolim-Neto PJ.
Thermal characterization of antimicrobial drug ornidazole and its
compatibility in a solid pharmaceutical product. J Therm Anal
Calorim. 2011;104:307–13.
14. Lavor EP, Freire FD, Araga
˜o CFS, Raffin FN, Moura TFAL.
Application of thermal analysis to the study of anti-tuberculosis
drug compatibility. Part 1. J Therm Anal Calorim.
2012;108:207–12.
15. Haines PJ, Reading M, Wilburn FW. Differential thermal ana-
lysis and differential scanning calorimetry. In: Brown ME, editor.
Handbook of thermal analysis and calorimetry, Chap. 5, Vol. 1.
Principles and practice. Amsterdam: Elsevier Science; 1998.
p. 279–361.
16. Moyano MA, Broussalis AM, Segall AI. Thermal analysis of
lipoic acid and evaluation of the compatibility with excipientes.
J Therm Anal Calorim. 2010;99:631–7.
17. Salvio-Neto H, Nova
´k CS, Matos JR. Thermal analysis and
compatibility studies of prednicarbate with excipients used in
semi solid pharmaceutical form. J Therm Anal Calorim.
2009;97:367–74.
18. Barboza F, Vecchia DD, Tagliari MP, Silva MAS, Stulzer HK.
Differential scanning calorimetry as a screening technique in
compatibility studies of acyclovir extended release formulations.
Pharm Chem J. 2009;43:363–8.
19. Age
ˆncia Nacional de Vigila
ˆncia Sanita
´ria. Formula
´rio Nacional
da Farmacope
´ia Brasileira. 1 ed. Brası
´lia: Ministe
´rio da Sau
´de;
2005.
20. U.S. Pharmacopeia—National Formulary, United States Phar-
macopeial Convention. USP 36—NF 31; 2013.
21. Bertol CD, Cruz AP, Stulzer HK, Murakami FS, Silva MAS.
Thermal decomposition kinetics and compatibility studies of
primaquine under isothermal and non-isothermal conditions.
J Therm Anal Calorim. 2010;102:187–92.
22. Lira AM, Arau
´jo AAS, Bası
´lio IDJ, Santos BLL, Santana DP,
Mace
ˆdo RO. Compatibility studies of lapachol with pharmaceu-
tical excipients for the development of topical formulations.
Thermochim Acta. 2007;457:1–6.
23. Sherine B, Nasser AJA, Rajendran S. Inhibitive action of
hydroquinone—Zn
2?
system in controlling the corrosion of car-
bon steel in well water. Int J Eng Sci Technol. 2010;2:341–57.
24. Liltorp K, Larsen TG, Willumsen B, Holm R. Solid state com-
patibility studies with tablet excipients using non thermal meth-
ods. J Pharm Biomed Anal. 2011;55:424–8.
25. Wesolowski M, Rojek B, Piotrowska J. Application of chemo-
metrically processed thermogravimetric data for identification of
baclofen–excipient interactions. J AOAC Int. 2012;95:691–8.
26. Devore JL. Probability and statistics for engineering and the
sciences. 4th ed. Belmont: Duxbury Press; 1995.
27. Hupp AM, Marshall LJ, Campbell DI, Smith RW, McGuffin VL.
Chemometric analysis of diesel fuel for forensic and environ-
mental applications. Anal Chim Acta. 2008;606:159–71.
28. Fulias A, Ledeti I, Vlase G, Vlase T. Physico-chemical solid-state
characterization of pharmaceutical pyrazolones: an unexpected
thermal behavior. J Pharm Biomed Anal. 2013;81–82:44–9.
29. Lima NGPB, Lima IPB, Barros DMC, Oliveira TS, Raffin FN,
Moura TFAL, Medeiros ACD, Gomes APB, Araga
˜o CFS.
Compatibility studies of trioxsalen with excipients by DSC, DTA,
and FTIR. J Therm Anal Calorim. 2014;115:2311–8.
30. Chadha R, Bhandari S. Drug–excipient compatibility screening—
role of thermoanalytical and spectroscopic techniques. J Pharm
Biomed Anal. 2014;87:82–97.
31. Rowe RC, Sheskey PJ, Quinn ME. Handbook of pharmaceutical
excipients. 6th ed. London: Pharmaceutical Press; 2009.
32. Fujita M, Ueda T, Handa T. Generation of formaldehyde by
pharmaceutical excipients and its absorption by meglumine.
Chem Pharm Bull. 2009;57:1096–9.
732 I
´. P. de Barros Lima et al.
123
