Journal of Thermal Analysis and Calorimetry (2025) 150:15117–15127

[Link]

Characterization and study of drug–excipient compatibility

of Lisinopril by TGA, FTIR and HPLC
Renata C. Silva1 · Marcello G. Trevisan1 · Jerusa S. Garcia1

Received: 10 October 2024 / Accepted: 25 August 2025 / Published online: 23 September 2025
© Akadémiai Kiadó Zrt 2025

Abstract
Given that drug–excipient interactions in formulations can alter chemical properties, stability and bioavailability, thus com-
promising efficacy, this study focuses on assessing the compatibility between Lisinopril (LIS) and selected pharmaceutical
excipients commonly employed in its formulations. LIS was characterized by powder X-ray diffraction (PXRD), Fourier-
transform infrared spectroscopy (FTIR), thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC).
To investigate potential interactions, LIS, eight excipients, and their respective binary mixtures (1:1, w/w) were subjected
to analysis by TGA, FTIR, and high-performance liquid chromatography (HPLC). A comparison between experimental and
simulated TGA/DTG curves was made using Pearson correlation coefficient to determine possible chemical incompatibilities
due thermal degradation. TGA results showed a probable interaction of the LIS with lactose monohydrate and mannitol.
FTIR results reveal an interaction of LIS with mannitol due to fact that hydroxyl is present in it. Results from HPLC analyses
have evidenced interactions found in the TG analyses, in addition to a loss of drug content corresponding to 6 and 19% in
binary mixtures with lactose monohydrate and mannitol, respectively.

Keywords Lisinopril dihydrate · Compatibility study · TGA​· FTIR · HPLC

Introduction Interactions between the drug and excipients used in its

formulation represent a critical aspect in pharmaceutical
Lisinopril dihydrate (LIS) (Fig. 1) is a drug whose chem- research, possibly leading to changes in chemical, physical
ical formula is [N2-[(S)-1-carboxy-3-phenylpropyl]-L- and therapeutic properties of its pharmaceutical formulation.
lysyl]-L-proline dihydrate, having molecular mass of Such interactions are called incompatibilities and, conse-
441.5 g ­m ol−1, solubility of 97 mg m­ L −1 and empirical quently, the therapeutic efficacy and safety of the drug might
formula of ­C 21H31N3O5 0.2 ­H2O [1]. LIS is a pharmaco- be compromised [3].
logically active compound that functions as a long-acting Research focused on the development of new drugs
inhibitor of the angiotensin-converting enzyme (ACE). often encompasses drug excipient compatibility studies.
This drug is commonly prescribed for the management Furthermore, there are studies documenting drug excipient
of various cardiovascular and renal disorders, including incompatibilities in drugs having already been marketed [4].
systemic arterial hypertension, congestive heart failure, In the search for an effective product formulation, distinct
myocardial infarction, and complications related to dia- analytical methods have been used in drug excipient compat-
betic nephropathy. Its most stable crystalline form is ibility studies, such as: thermal, spectroscopic and chroma-
Form 1, and the one typically found in commercial drugs tographic methods [5].
is Lisinopril dihydrate [2]. There is no widely accepted standard protocol for evalu-
ating drug-excipient compatibility, but it is recommended
to employ a combination of several techniques [6]. In this
* Jerusa S. Garcia context, the present study aimed to assess the compatibility
[Link]@[Link]
of LIS with selected excipients typically used in the prepara-
1
Laboratory of Analysis and Characterization tion of its solid dosage forms.
of Pharmaceuticals – LACFar, Institute This is the very first compatibility study on LIS. Typically,
of Chemistry, Federal University of Alfenas, Alfenas, the initial technique used in drug-excipient compatibility studies
Minas Gerais 37130‑001, Brazil

Vol.:(0123456789)
15118 R. C. Silva et al.

NH2 Methods
LIS characterization
×2 H2O
O OH
Thermogravimetric analysis (TGA)
N
N TGA of LIS were performed using a Q600 instrument (TA
H Instruments, USA). Approximately 6 mg of the sample
O was accurately weighed and placed in a 100 µL alumina
OH
O
crucible. The analysis was conducted under a nitrogen
atmosphere with a constant flow rate of 50 mL min⁻1,
Fig. 1  Chemical structure of lisinopril dihydrate applying a heating rate of 10 K min⁻1 from 303 to 873 K.
The calibration of the TGA was performed in accordance
is DSC. However, it was impossible through this study owing with the manufacturer's recommended protocol, using an
to the proximity of melting and decomposition temperatures of indium (In) standard (purity > 99.9%), and baseline/mass
the drug, which could compromise the interpretation of results. calibration was conducted with a traceable standard mass.
Therefore, it was opted to initiate the investigation of LIS com-
patibility with selected excipients using the TGA technique. To
aid in assessing TGA data, a simulated curve was devised and Differential scanning calorimetry (DSC)
then compared to experimental curves using Pearson correla-
tion coefficient to assess their similarity; this approach has been DSC measurement was carried out on a Q20 instrument (TA
successfully employed in similar studies [7]. LIS, excipient and Instruments, USA). Around 3 mg of LIS was placed in an
binary mixture samples have also been analyzed by HPLC and open aluminum crucible with a volume of 40 µL. The analy-
FTIR aiming to further and clarifying results found in TGA sis was conducted under a dry nitrogen atmosphere at a flow
analyses. The characterization of the pure drug was carried out rate of 50 mL min⁻1, employing a heating rate of 10 K min⁻1
using various analytical techniques, including thermal analysis over the temperature range of 313–483 K. The instrument
(DSC and TGA), powder X-ray diffraction (PXRD), and Fou- was calibrated for both temperature and heat flow (differen-
rier-transform infrared spectroscopy (FTIR). tial scanning calorimetry) using an indium (In) standard with
a well-defined melting point and enthalpy.
Experimental
X‑ray powder diffraction
Materials
The crystal structure of LIS was analyzed by powder X-ray
For the compatibility evaluation, binary mixtures were pre- diffraction (PXRD) using a Rigaku Ultima IV diffractometer.
pared using LIS dihydrate (lot #M1Q8795) with eight phar- The measurement were carried out over a 2θ range of 3–40°,
maceutical-grade excipients (purity greater than 99%). Excipi- under operating conditions of 40 kV and 30 mA. Data acqui-
ents included colloidal silicon dioxide (CSD), croscarmellose sition was performed in Step Scan mode with a step size of
sodium (CCS), dicalcium phosphate dihydrate (DCPD), lac- 0.02° and a counting time of 0.5 s per step. Cu Kα radiation
tose monohydrate (LAC), mannitol (MAN), microcrystalline (λ = 1.54178 Å) was employed as the X-ray source, with a
cellulose (MC), magnesium stearate (MS), and starch (ST). nickel filter used to suppress the Kβ radiation. The sample,
High-performance liquid chromatography (HPLC) analyses consisting of a thin, solvent-free powder layer, was prepared
were performed using acetonitrile and methanol (HPLC grade) on glass slides prior to analysis.
as solvents, and phosphoric acid was employed to adjust the
pH of the mobile phase.
Characterization and study of drug–excipient compatibility of Lisinopril by TGA, FTIR and HPLC 15119

Fourier transform infrared spectroscopy between the simulated and experimental curves of the mix-
tures was quantitatively evaluated using the Pearson correla-
This analysis was performed using an attenuated total reflec- tion coefficient [7]. This procedure is performed to facilitate
tance Fourier transform infrared (ATR-FTIR) (Nicolet iS50, an identification of incompatibilities. If there are incompat-
Thermo Scientific, USA). Spectra were acquired at 298 K ibilities and formation of volatile products, there will be a
in the range of 4000–400 ­cm−1 (4 ­cm−1of resolution and 64 profile different from simulated ones.
coadded scans per sample). Before all analysis, a background
spectrum was properly collected using the same conditions. High performance liquid chromatography (HPLC)

LIS compatibility study HPLC was employed to verify whether mass variations
observed in TGA results at high temperatures are related to
Sample preparation drug content loss. After heating in a laboratory oven, sam-
ples were transferred to 10 mL volumetric flasks and diluted
Binary mixtures were prepared in a 1:1 mass ratio (w/w), with a methanol/water/acetonitrile mixture in the volumet-
accurately weighed using an analytical microbalance, and ric ratio of 80:17.5:2.5 (v/v/v). The resulting solutions were
placed into 1.5 mL polyethylene microtubes containing sonicated in an ultrasonic bath for 40 min and subsequently
stainless steel spheres. Each sample, with approximately analyzed by HPLC. All procedures were performed in trip-
6 mg, was homogenized for 3 min using a vortex mixer. licate. Calibration curves were established using stand-
To ensure analytical consistency, the entire content of each ard solutions at concentrations of 20, 40, 60, 80, 100, and
individually prepared mixture was used, thereby minimiz- 120 mg L⁻1, based on peak area measurements.
ing potential shortcoming related to sample homogeneity. HPLC analyses were carried out using an Ultimate
After preparation, the samples were immediately analyzed 3000 LC system (Thermo Scientific, USA). The chroma-
by TGA. tographic separation was performed on a Roc C18 col-
To the isothermal stress testing, another set of binary umn (4.6 mm × 150 mm, 5 µm particle size), following the
mixtures were also prepared in the same previous prepara- method proposed by Arayne et al. [8]. The mobile phase
tion procedure. Samples were subjected to heat treatment was composed of methanol, deionized water, and acetoni-
in a laboratory oven to simulate the TGA conditions and trile in the volumetric proportion of 80:17.5:2.5 (v/v/v),
allowing a better interpretation of the results. It is important with the pH adjusted to 3.0 using concentrated phosphoric
to mentioned that freshly prepared binary mixture at room acid. The flow rate was maintained at 0.7 mL min⁻1, with
temperature is generally insufficient to cause immediate the column temperature at 303 K. Detection was conducted
incompatibility. Therefore, the thermal treatment consisted at a wavelength of 225 nm, and the injection volume was
of weighing 6 mg of each sample (including LIS, excipients, 2 µL. All solvents used—methanol (ChromAR) and ace-
and homogenized drug–excipient mixtures), which were tonitrile (CROMASOLV®, ≥ 99.9%)—were of HPLC grade,
then placed in a laboratory drying oven at 443 K for 5 min. and deionized water was obtained using a laboratory-grade
This temperature was selected based on preliminary experi- purification system. The method was validated according to
ments, ensuring that it remained 5 K below the thermal deg- ICH (Q2) R1 [9].
radation temperature of LIS, thus avoiding drug decomposi-
tion. Heating time and temperature were established based Fourier transform infrared spectroscopy
on preliminary assessments [9] confirming no degradation
of the drug under these conditions. The heated samples were Prior to FTIR analysis, samples of LIS, individual excipi-
applied to HPLC and FTIR analyses. ents, and binary mixtures were heated in a laboratory oven
at 443 K for 5 min. Subsequent spectroscopic measurements
Thermogravimetric analysis (TGA) were conducted under the same conditions described in the
characterization section.
TGA were carried out for pure LIS, each individual excipi-
ent, and their respective binary physical mixtures (1:1 w/w)
using a TG Q20 instrument (TA Instruments, USA). All Results and discussion
measurements followed the experimental conditions estab-
lished in the LIS thermal characterization section. LIS characterization
To support interpretation of the thermal behavior, simu-
lated TGA curves were generated for each binary system The TG/DTG and DSC thermal profiles are depicted in
by mathematically combining the individual curves of LIS Fig. 2a. The DSC curve presents two overlapping endother-
and the corresponding excipient. The degree of similarity mic events ranging from 313.0 to 416.7 K. According to
15120 R. C. Silva et al.

Fig. 2  a LIS thermal analysis 0

results: DSC (313–483 K; 100 1.0
10 K ­min−1; ­N2 atmosphere,
50 [Link]−1; closed alu-
minum crucibles), TGA and 80

Heat flow/mW
Derivate mass/% K–1
Mass loss/%
DTG curves (313–873 K;
10 K ­min−1; ­N2 atmosphere, –1
50 mL ­min−1; aluminum 60
0.5
crucibles); b FTIR spectra of
LIS at 298 K in the range of
40
4000–400 ­cm−1 (64 scans at
4 ­cm−1 resolution); c X-ray
diffractogram of LIS: 5–55°, 20 –2

Endo
40 kV; 30 mA
0.0
0
400 500 600 700 800

Temperature/K
(a)

100

90

3099 1205
Transmittance/%

80

70
1608
1390
60 1453
1653

1569
50 1542

40
4000 3500 3000 2500 2000 1500 1000 500

Wavenumber/cm –1

(b)
Intensity/a.u

10 20 30 40 50
2 /°
(c)
Characterization and study of drug–excipient compatibility of Lisinopril by TGA, FTIR and HPLC 15121

TGA results, there is a mass loss of 8.3% attributed to dehy- components, it is found that the API is possibly compatible
dration at temperatures ranging from 313.0 to 378.5 K. Such with selected excipients. However, if results from the TGA
mass loss corresponds to two water molecules turning from of the binary mixture reveals a different profile compared
dihydrate to their anhydrous form, once the molecular mass to individual components, it possibly indicates that there
of LIS is 441.5 [Link]−1. Afterwards, an additional endother- might have been an interaction [13]. It is evident that such
mic event was noted in the DSC curve ranging from 436.4 modifications in thermal profiles are caused by the forma-
to 479.46 K with ­Tonset at 447 K and enthalpy of 174.57 J. tion of volatile compounds causing mass loss. Therefore, it
g−1. Literature data indicate that when LIS is analyzed by is worth conducting TGA compatibility studies associated
DSC using an open aluminum crucible, there is a third endo- with other techniques, such as FTIR and HPLC.
thermic event corresponding to its melting point at T ­ onset of After obtaining results of thermal analyses from pure
460 K [10]. components and binary mixtures, a simulated curve of the
At the same melting temperature range of the drug binary mixture was plotted. For such a purpose, data corre-
(436.4–479.46 K), melting occurs through decomposi- sponding to the TGA curve of LIS and excipients at tempera-
tion, and mass loss is observed in the TGA curve. The TG tures ranging from 313 to 823 K were used. The data matrix
curve reveals that the drug presented mass loss from 448 K has 6,022 lines, both for LIS and excipients, which have
onwards, leaving 9.5% of residue. Literature reports indicate been compared to plot the simulated curve point by point.
that LIS melting occurs at temperatures ranging between 440 Each point of the simulated curve was calculated through
and 460 K [11, 12]. the following equation:
Melting and decomposition events occur at very close [( ) ( )]
temperatures, which hampers the determination of the final Simulated curve MLIS x0.5 + Mexcipient x0.5
melting point and initial decomposition temperature of the
where MLIS corresponds to the TGA curve related to LIS,
drug. Concerning thermal behavior, the thermal event is
and ­Mexcipient to the TGA curve of the excipient.
poorly reproducible when melting is preceded by molecule
Thus, the TGA curves of LIS and those found for binary
decomposition owing to the difficulty in integrating the ther-
mixtures obtained experimentally and through simulations
mal event. Thus, this behavior rendered a drug-excipient
are shown in Fig. 3. Most binary mixtures, except for lac-
compatibility study of LIS by DSC impossible. Therefore,
tose and mannitol, presented no significant changes in their
TGA were performed instead.
thermal profiles compared to individual components, thus
The FTIR spectrum of LIS (Fig. 2b) displays character-
revealing no interaction between the API and the evaluated
istic absorption bands consistent with its functional groups.
excipients.
The band at 3099 cm⁻1 corresponds to aromatic C–H stretch-
A visual analysis of simulated and experimental curves
ing vibrations. The absorption at 1608 cm⁻1 is attributed to
of binary mixtures assists in verifying possible differences.
C=C stretching within the aromatic ring, while the band at
However, correlation methods are effective, since they can
1653 cm⁻1 is associated with C=O stretching vibrations of
assess the similarity between two profiles more accurately.
the amide group. Bands observed at 1569 and 1542 cm⁻1 are
In this study, Pearson correlation coefficient (­ r2) was used to
ascribed to asymmetric stretching of the carboxylate group.
compare experimental and theoretical profiles; this approach
Additionally, the absorption at 1453 cm⁻1 corresponds to
has been successfully employed in similar studies [7]. As
–CH₂ bending vibrations. The band at 1390 cm⁻1 also arises
shown in Fig. 4, the Pearson correlation coefficient reveals a
from asymmetric stretching of the carboxylate group. The
deviation from the ideal value (­ r2 = 1.00), particularly for the
presence of a tertiary amine is confirmed by the C–N stretch-
binary mixtures of LIS with the excipients LAC and MAN,
ing band at 1205 cm⁻1 [10–12].
which exhibited the lowest correlation values.
The X-ray diffraction pattern obtained for LIS is showed
Experimental TGA and simulated TG curves were com-
in Fig. 2c. The observed diffractogram corresponds to the
pared, and results for LIS with LAC and MAN binary mix-
dihydrate phase of lisinopril, known as Form 1, which rep-
tures are shown in Fig. 4. Figure 4a shows LAC monohydrate
resents its most stable crystalline form and was employed in
dehydration up to 423 K, with mass loss of approximately
the present study. This result is consistent with data previ-
4.4%. From 493 K, this excipient initiates a decomposition
ously reported in the literature [12].
process with 81% initial mass loss. Comparing the experi-
mental binary mixture with the simulated one, their thermal
LIS compatibility
profiles can be divided into 3 regions. In region I (313 to
453 K), the simulated curve resembles the experimental one.
TG compatibility study
In region II (453–675 K), the simulated curve is above the
experimental one, indicating that there was greater mass loss
When the thermal profile of the drug-excipient binary
experimentally. This region encompasses LIS decomposition
mixture resembles the thermal patterns of individual
15122 R. C. Silva et al.

100 100
LIS + CCS 100 LIS + CSD
100 90

Simulated mass%
Simulated mass%
90
80 80

Mass loss/%
80
Mass loss/%

70
60 80
60
60
40
40 50 70
y = 0.998x – 0.960 LIS
LIS y = 1.0275x – 1.8639
CSD
CCS 40 20
20 R2 = 0.9899 LIS + CSD R2 = 0.9887
LIS + CCS 60
30 Simulated
Simulated
0
0
100 90 80 70 60 50 40 30 400 500 600 700 800 100 95 90 85 80 75 70 65 60
400 500 600 700 800
Temperature/K Experimental mass% Temperature/K Experimental mass%

(a) (b) (c) (d)

100 100
LIS + MC
100 LIS + DCPD 100

Simulated mass%
90 80
80 Simulated mass% 80
Mass loss/%

Mass loss/%
80 60
60 60

70
40 40 40
LIS LIS
DCPD y = 0.925x + 7.893 y = 1.1185x – 11.6492
60 MC
20 LIS + DCPD 20
R2 = 0.9941 LIS + MC 20 R2 = 0.9912
Simulated Simulated
0 50 0
400 500 600 700 800 100 90 80 70 60 50 400 500 600 700 800 100 90 80 70 60 50 40 30 20
Temperature/K Experimental mass% Temperature/K Experimental mass%
(e) (f) (g) (h)
100 100
100 LIS + MS 100 LIS + ST

Simulated mass%
Simulated mass%

80 80
80 80
Mass loss/%

Mass loss/%
60 60 60 60

40 40 40
LIS 40 LIS y = 1.0197x – 1.8309
MS y = 1.130x – 7.843 ST
20 20 LIS + ST R2 = 0.9894
LIS + MS 20
20 R2 = 0.9797 Simulated
Simulated
0 0
400 500 600 700 800 100 90 80 70 60 50 40 30 20 400 500 600 700 800 100 90 80 70 60 50 40 30 20
Temperature/K Experimental mass% Temperature/K Experimental mass%
(i) (j) (k) (l)

Fig. 3  TGA curves of LIS, excipient and experimental and simulated binary mixtures: a LIS+CCS; c LIS+CSD; e LIS+DCPD; g LIS+MC; i
LIS+MS; k LIS+ST Correlation charts of binary mixtures: b LIS+CCS; d LIS+CSD; f LIS+DCPD; h LIS+MC; j LIS+MS; l LIS+ST

and, in the case of its binary mixture with LAC, there might Figure 5 shows the DTG curves of LIS, the individual
have been the formation of different volatile thermal deg- excipients, and their respective binary mixtures, including
radation products in comparison with heating the excipi- both experimental and simulated profiles. Some binary mix-
ent or drug separately. The opposite occurred at region III tures, such as those of LIS with excipients CSD, DCPD,
(675–823 K), i.e. the experimental curve is above the simu- MS and ST, showed no significant changes in the thermal
lated curve, thus revealing smaller mass loss than expected profiles compared to individual components.
and indicating greater thermal stability of the binary mix- Regarding the binary mixtures of the drug with excipi-
ture. This behavior suggests a protective thermal effect of the ents CCS and MC, the experimental DTG curve indicated
excipient in this temperature range, despite both compounds greater thermal stability compared to that found for the
having already been thermally degraded and converted to a simulated DTG curve. Figure 5b shows that the experi-
condensed mass of carbon. mental curve of LIS with CCS reached its peak between
Regarding the MAN excipient (Fig. 4c), thermal decom- 534 and 667 K, while the simulated curve for the same
position is observed at temperatures ranging from 534 to mixture exhibits its corresponding peak at lower tempera-
673 K, leaving 1.9% residue. When comparing the experi- tures ranging between 524 and 599 K. The same behav-
mental thermogravimetric curve of the LIS–MAN mixture ior was observed for the drug mixed with MC, where its
with the corresponding simulated curve, the thermal profile experimental peak was found at temperatures ranging
can be divided into three regions. In region I (313–373 K), between 540 and 667 K, while the simulated curve shows
the simulated curve resembles the experimental one. In its corresponding peak between 557 and 637 K.
region II (373–532 K), the experimental curve is slightly The binary mixture of LIS with excipients LAC and
above the simulated curve, suggesting the occurrence of a MAN presented a different experimental DTG curve pro-
slight mass loss, as expected. In region III (532–823 K), the file from that found for the simulated curve. The experi-
simulated curve is above the experimental one, indicating mental curve indicated smaller thermal stability, since
that there was greater mass loss experimentally. peaks were observed at lower temperatures compared to
those found for the simulated DTG curve. Concerning the
Characterization and study of drug–excipient compatibility of Lisinopril by TGA, FTIR and HPLC 15123

I II III
100 100

80 80

Simulated mass/%
Mass loss/%

60 60

40 40
y = 1.117x – 5.158
LIS
20 LAC 20 R2 = 0.9471
LIS + LAC
Simulated
0 0
400 500 600 700 800 100 90 80 70 60 50 40 30
Temperature/K Experimental mass/%

(a) (b)
I II
III
100 100

80 80

Simulated mass/%
Mass loss/%

60 60

40 40

LIS y = 0.8548x – 16.445

20 MAN 20
R2 = 0.912
LIS + MAN
Simulated
0 0
400 500 600 700 800 100 80 60 40 20 0
Temperature/K Experimental mass/%

(c) (d)

Fig. 4  TGA curves of LIS, LAC, experimental (LIS+LAC) and simulated binary mixtures; b Pearson correlation coefficient of LIS + LAC; c
TGA curves of LIS, MAN, experimental (LIS+MAN) and simulated binary mixtures; d Pearson correlation coefficient of (LIS+MAN)

binary mixture of LIS with LAC, the experimental curve HPLC compatibility study
exhibited peaks at lower temperatures from 435 K onward
compared to those found for the simulated curve. Regard- The use of HPLC for LIS quantification enables the assess-
ing the binary mixture of LIS with MAN, all peaks for the ment of potential chemical interactions with excipients, as
experimental curve were observed at lower temperatures such interactions may lead to increase of degradation prod-
than those for the simulated curve. For instance, a peak ucts and consequent reduction of API content. Compatibility
is observed between 482 and 625 K for the experimen- of LIS with excipients was evaluated using the HPLC–UV
tal curve, while the corresponding peak of the simulated method, which demonstrated a linear analytical response
curve was found at temperature ranging between 567 and (y = 0.099x–0.205; ­R2 = 0.999). LIS analysis results after
662 K. heating at 443 K for 5 min are shown in Fig. 6a. No peaks
The Pearson correlation coefficient was also applied related to the excipients under analysis were observed
to estimate the similarity between the DTG curves of the through the method. A single LIS peak was observed at
binary mixtures, following the same approach used for the 1.85 min, and none of the studied excipients presented a
TGA data. The mixtures of LIS with LAC and MAN exhib- peak in this region or in any other, therefore evidencing the
ited the lowest correlation coefficients, with r­ 2 values of used method selectivity. No significant thermal degradation
0.1751 and 0.1148, respectively. of LIS was caused by the heating process, since this peak
corresponded to approximately 98% of the quantity of LIS
in the solution before heating.
15124 R. C. Silva et al.

1.6
1.0 LIS LIS LIS
1.0
CSD 1.4 CCS DCPD
LIS + CSD LIS + CCS LIS + DCPD
Simulated 1.2 Simulated Simulated
0.8 0.8

Derivate mass/% K–1

Derivate mass/% K–1

Derivate mass/% K–1

1.0
0.6 0.6
0.8

0.6
0.4 0.4
0.4
0.2 0.2
0.2

0.0
0.0 0.0
– 0.2
400 500 600 700 400 500 600 700 800 400 500 600 700 800
Temperature/K Temperature/K Temperature/K
(a) (b) (c)
1.4 3.0
LIS LIS 3.0 LIS
1.2 LAC MAN MC
LIS + LAC 2.5 LIS + MAN LIS + MC
Simulated Simulated 2.5 Simulated
1.0

Derivate mass/% K–1

Derivate mass/% K–1

Derivate mass/% K–1

2.0
0.8 2.0

1.5
0.6 1.5

0.4 1.0
1.0

0.2 0.5 0.5

0.0
0.0 0.0
– 0.2
400 500 600 700 800 400 500 600 700 800 400 500 600 700 800
Temperature/K Temperature/K Temperature/K
(d) (e) (f)
2.0
LIS LIS
2.5
MS ST
LIS + MS LIS + ST
Simulated 1.5 Simulated
2.0
Derivate mass/% K–1
Derivate mass/% K–1

1.5 1.0

1.0
0.5
0.5

0.0
0.0

400 500 600 700 800 400 500 600 700 800
Temperature/K Temperature/K
(g) (h)

Fig. 5  DTG curves of LIS, excipient and experimental and simulated mixtures: a LIS+CCS; b LIS+CSD; c LIS+DCPD; d LIS+LAC; e
LIS+MAN; f LIS+MC; g LIS+MS; h LIS+ST

The LIS–excipient binary mixtures were prepared at found for mixtures with lactose and mannitol can modify
the same concentration used for the LIS standard solu- the stability of LIS, thus potentially affecting final product
tion (400 mg L⁻1), following the experimental conditions quality.
described in the Methods section. Table 1 shows the peak Drugs containing an amine group, such as LIS, com-
areas and corresponding relative LIS contents for all binary promise product stability due to the occurrence of Mail-
systems. lard reaction when formulated with reducing disaccharides
The chromatograms of binary mixtures of LIS with or other pharmaceutical ingredients containing carbonyl
most excipients proposed herein, except for LAC and groups, such as lactose [14].
MAN, exhibited no changes in peak areas (chromatograms Mannitol, nonetheless, can interact with LIS through its
not shown), thus confirming Pearson correlation coeffi- hydroxyls. There are reports in literature on the incompat-
cient data. The prevalence of LIS relative content in binary ibility of bisoprolol fumarate with mannitol based on HPLC
mixtures indicates that these excipients are compatible and results showing 77% degradation due to the drug interaction
can be used for solid formulations. Otherwise, mixtures with mannitol hydroxyl groups [15].
of LIS with LAC and MAN (Fig. 6b) showed reduction of
LIS peak areas (approximately 6 and 19%, respectively),
thus indicating chemical interactions between the API
and these excipients. Chemical interactions such as those
Characterization and study of drug–excipient compatibility of Lisinopril by TGA, FTIR and HPLC 15125

Fig. 6  a Chromatograms of 70

excipients and LIS b Chromato-
grams of binary mixtures show- 56
ing LIS content loss. Analysis
conditions: C18 column, 42

mAU
4.6 mm × 150 mm, particle
size of 5 µm; methanol/water/ 28
acetonitrile phase in the ratio
80:17.5:2.5 (v/v/v) at pH 3.0 14

0
LIS
CSD
CCS
DCPD
LAC
MAN
MC
MS
ST
0 1 2 3 4 5 6
Time/min
(a)

64

48

mAU
32

16

LIS

LIS+MAN

LIS+LAC

0 1 2 3 4 5 6
Time/min
(b)

FTIR compatibility study mixtures of LIS and excipients, simple milling at room
temperature caused no chemical interaction observed by
Infrared spectroscopy was employed as a complementary FTIR. All FTIR spectra of binary mixtures represent an
technique to investigate potential chemical interactions overlapping of FTIR spectra of pure components. These
between LIS and the selected excipients. In the analysis of results evidence that LIS is compatible with all the
binary mixtures, evidence of such interactions may include selected excipients at room temperature (Fig. 7a).
the appearance of new absorption bands, broadening of However, regarding the LIS-MAN mixture, the band at
existing ones, or variations in band intensities [16]. 3565 ­cm−1 disappeared, in addition to greater intensity in ranges
To compared temperature effect in the binary mix- between 3400–2266 ­cm−1 and 1680–400 ­cm−1. The spectrum
tures, the spectra of samples were analyzed before and region corresponding to the range between 3700–3200 ­cm−1 is
after heating to 443 K using FTIR (Fig. 7). No spectral typical of O–H stretching vibrations, thus confirming that the
change was observed comparing pure LIS before and band at 3565 cm⁻1 is attributed to hydroxyl groups. The analysis
after heat treatment. Regarding freshly prepared binary results of the LIS-MAN binary mixture indicate that the drug
15126 R. C. Silva et al.

Table 1  Peak area and relative content of LIS in the samples Conclusions
Peak area/mAU*min Relative content/%
Thermal analysis techniques, particularly thermogravi-
LIS 7.62 ± 0.07 98.0 ± 1.0 (a) metric analysis (TGA), offer a rapid and effective path for
LIS+CCS 7.58 ± 0.08 97.4 ± 1.1 (a) preliminary evaluations in pharmaceutical formulation
LIS+CSD 7.69 ± 0.15 97.4 ± 1.0 (a) studies. Nevertheless, for a comprehensive assessment of
LIS+DCPD 7.64 ± 0.13 98.3 ± 1.6 (a) drug–excipient compatibility, the use of multiple comple-
LIS+LAC 7.13 ± 0.02 91.8 ± 0.3 (b) mentary analytical methods is recommended to obtain more
LIS+MAN 6.10 ± 0.21 79.0 ± 2.6 (b) reliable and conclusive results. In this context, HPLC and
LIS+MC 7.70 ± 0.29 99.1 ± 0.4 (a) FTIR are especially valuable. The combined analyses of
LIS+MS 7.62 ± 0.19 98.0 ± 2.4 (a) binary mixtures using TGA, HPLC, and FTIR consistently
LIS+ST 7.52 ± 0.16 96.8 ± 2.1 (a) indicated incompatibilities between Lisinopril and lactose
Values in the same column followed by the same letter do not show monohydrate and mannitol. Thus, the integrated application
statistically significant differences (p > 0.05) according to Student’s of these techniques proved to be an effective strategy for
t-test evaluating the compatibility of Lisinopril with the selected
excipients.

Acknowledgements The authors gratefully acknowledge the financial

LIS support provided by FAPEMIG, CNPq, and CAPES.

LIS+CSD Funding Fundação de Amparo à Pesquisa do Estado de Minas Gerais

Transmittance/%

LIS+CCD
LIS+DCPD
LIS+LAC
LIS+MAN References
LIS+MC
1. Helal F, Lane ME. Transdermal delivery of Angiotensin Con-
LIS+MS
verting Enzyme inhibitors. Eur J Pharm Biopharm. 2014;88:1–7.
LIS+ST [Link]
2. Grosu IG, Filip X, Miclăus MO, Filip C. Review-hydrogen-medi-
3500 3000 2500 2000 1500 1000 500 ated noncovalent interactions in solids: what can NMR crystal-
Wavenumber/cm–1 lography tell about? Molecules. 2020;25:1–24. [Link]
3390/​molec​ules2​51637​57.
(a) 3. Abrantes CG, Duarte D, Reis CP. An overview of pharmaceutical
excipients: safe or not safe? J Pharm Sci. 2016. [Link]
LIS 1016/j.​xphs.​2016.​03.​019.
LIS+CSD
4. Santos WM, Nóbrega FP, Andrade JC, Almeida LF, Conceição
MM, Medeiros ACD, et al. Pharmaceutical compatibility of dexa-
LIS+CCD methasone with excipients commonly used in solid oral dosage
LIS+DCPD forms. J Therm Anal Calorim. 2021;145:361–78. [Link]
Transmittance/%

LIS+LAC 10.​1007/​s10973-​020-​09753-1.
5. Gorain B, Choudhury H, Pandey M, Madheswaran T, Kesharwani
LIS+MAN P, Tekade RK. Drug–excipient interaction and incompatibilities.
LIS+MC Dos Form Des Parameters. 2018; [Link]
0-​12-​814421-​3.​00011-7.
LIS+MS
6. Chadha R, Bhandari S. Drug-excipient compatibility screening
LIS+ST role of thermoanalytical and spectroscopic techniques. J Pharm
Biomed Anal. 2014;87:82–97. [Link]
3000 2500 2000 1500 1000 500 2013.​06.​016.
Wavenumber/cm–1 7. Daniel JSP, Cruz JC, Catelani TA, Garcia JS, Trevisan MG.
Erythromycin-excipients compatibility studies using the thermal
(b) analysis and dynamic thermal infrared spectroscopy coupled with
chemometrics. J Therm Anal Calorim. 2021;143:3127–35. [Link]
Fig. 7  FTIR spectra of LIS samples and its mixtures: a before and b doi.​org/​10.​1007/​s10973-​020-​09691-y.
after heating for 5 min at 443 K 8. Arayne MS, Sultana N, Naveed S. RP-HPLC Method for the deter-
mination of lisinopril in active pharmaceutical ingredients, dos-
age forms and human serum. Open Access Sci Rep. 2013;3:1–5.
probably interacts with hydroxyl groups present in mannitol, [Link]
9. Silva, RC. Drug-excipient compatibility studies of Brono-
thus suggesting their interaction. pride, Lisinopril and Pantoprazole by thermal analysis, infrared
Characterization and study of drug–excipient compatibility of Lisinopril by TGA, FTIR and HPLC 15127

spectroscopy and liquid chromatography. Doctorate Thesis. Fed- 14. Ghaderi G, Monajjemzadeh F. Review of the physicochemical
eral University of Alfenas–Unifal-MG. 2025. methods applied in the investigation of the maillard reaction
10. Sorrenti M, Catenacci L, Cruickshank DL, Caira MR. Lisino- in pharmaceutical preparations. J Drug Delivery Sci Technol.
pril dihydrate: single-crystal X-ray structure and physico- 2020;55:1–9. [Link]
chemical characterization of derived solid forms. J Pharm Sci. 15. Marothu VK, Yerramothu P, Gorrepati M, Majeti S, Mamidala
2013;102:3596–603. [Link] SK, Nellutla A. Application of HPLC to assess the compatibil-
11. Danafar H, Rostamizadeh K, Davaran S, HAmidi M. Drug-con- ity of bisoprolol fumarate with selected excipients in mixtures
jugated PLA–PEG–PLA copolymers: a novel approach for con- by isothermal stress testing Ann Pharm Fr. Camille Desmoulins:
trolled delivery of hydrophilic drugs by micelle formation. Pharm Elsevier; 2015. p. 442–51.
Dev Technol. 2017;22:947–57. [Link] oi.o​ rg/1​ 0.3​ 109/1​ 08374​ 50.​ 16. Rojek B, Wesolowski M. FTIR and TG analyses coupled with
2015.​11259​20. factor analysis in a compatibility study of acetazolamide with
12. Sbârcea L, Udrescu L, Lede I, Szabadai Z, Fulias A, Sbârcea excipients. Spectrochim Acta Part A Mol Biomol Spectrosc.
C. β-Cyclodextrin inclusion complexes of lisinopril and zofeno- 2019;208:285–93. [Link]
pril physicochemical characterization and compatibility study of
lisinopril-β-cyclodextrin with lactose. J Therm Anal Calorim. Publisher's Note Springer Nature remains neutral with regard to
2016;123:2377–90. [Link] jurisdictional claims in published maps and institutional affiliations.
13. Ledeti I, Bolintineanu S, Vlase G, Circioban D, Ledeti A, Vlase
T, et al. Compatibility study between antiparkinsonian drug Lev- Springer Nature or its licensor (e.g. a society or other partner) holds
odopa and excipients by FTIR spectroscopy, X-ray diffraction exclusive rights to this article under a publishing agreement with the
and thermal analysis. J Therm Anal Calorim. 2017;130:433–41. author(s) or other rightsholder(s); author self-archiving of the accepted
[Link] manuscript version of this article is solely governed by the terms of
such publishing agreement and applicable law.