Improving the Solubility of Aripiprazole by Multicomponent Crystallization

Abstract

Aripiprazole (ARI) is a third-generation antipsychotic with few side effects but a poor solubility. Salt formation, as one common form of multicomponent crystals, is an effective strategy to improve pharmacokinetic profiles. In this work, a new ARI salt with adipic acid (ADI) and its acetone hemisolvate were obtained successfully, along with a known ARI salt with salicylic acid (SAL). Their comprehensive characterizations were conducted using X-ray diffraction and differential scanning calorimetry. The crystal structures of the ARI-ADI salt acetone hemisolvate and ARI-SAL salt were elucidated by single-crystal X-ray diffraction for the first time, demonstrating the proton transfer from a carboxyl group of acid to ARI piperazine. Theoretical calculations were also performed on weak interactions. Moreover, comparative studies on pharmaceutical properties, including powder hygroscopicity, stability, solubility, and the intrinsic dissolution rate, were carried out. The results indicated that the solubility and intrinsic dissolution rate of the ARI-ADI salt and its acetone hemisolvate significantly improved, clearly outperforming that of the ARI-SAL salt and the untreated ARI. The study presented one potential alternative salt of aripiprazole and provided a potential strategy to increase the solubility of poorly water-soluble drugs.

1. Introduction

In the past few decades, a number of drugs have shown poor physicochemical properties, especially aqueous solubility and stability, often affecting their absorption in the gastrointestinal tract [1]. Multicomponent crystal formation is an effective strategy to improve pharmacokinetic profiles without altering the main chemical structures and inherent biological activity [2,3]. Co-crystals and salts are two common forms of multicomponent crystals which might have higher solubility and faster dissolution behavior compared to untreated drugs [4,5]. Compared to cocrystals, salt formation is the simplest and most cost-effective strategy and has significant advantages in addressing poor aqueous solubility because of ionizable drugs [6].

Aripiprazole (ARI), a third-generation antipsychotic, is a dopamine D2 receptor partial agonist and D1 receptor agonist which can ameliorate hyperprolactinemia induced by other antipsychotic drugs and cause fewer side effects, such as weight gain, diabetes, and dyslipidemia [7,8]. However, ARI belongs to the Biopharmaceutics Classification System (BCS) class II and its clinical use is limited by its poor aqueous solubility [9,10]. As a weakly basic drug, salt formation is one of the most popular and effective approaches to improve physicochemical properties, especially solubility [6,11,12]. Many salts of ARI have been reported. Freire et al. reported successively eight crystal structures of ARI salts with nitrate, perchlorate, oxalate, phthalate, homophthalate, thiosalicilate, and two different dihidrogenphosphates [13,14,15,16]. Nanubolu et al. reported five ARI salts with benzoic acid, 2,4-dihydroxy benzoic acid, 2,5-dihydroxy benzoic acid, salicylic acid, and hydrochloric acid [17]. However, these studies mainly focus on the structure illustration, and key pharmaceutical properties were not given. While Zhao et al. synthesized six ARI salts with gallic acid, 4-aminosalicylic acid, acetylsalicylic acid, maleic acid, fumaric acid, and malic acid and evaluated their solubility and dissolution profile [18], new ARI salts with ideal pharmaceutical properties are still in demand.

Continuing to explore the excellent salts of ARI, we successfully obtained a new ARI salt with adipic acid (ADI) and its acetone hemisolvate, along with a known ARI salt with salicylic acid (SAL). The molecular structures of the ARI and cocrystal formers (CCF) are displayed in Scheme 1. Their comprehensive characterizations were conducted using powder X-ray diffraction (PXRD) and differential scanning calorimetry (DSC). The crystal structures of the ARI-ADI salt acetone hemisolvate and ARI-SAL salt were elucidated by single-crystal X-ray diffraction (SXRD) for the first time. Furthermore, computational studies, including molecular electrostatic potential surface (MEPS) and Hirshfeld surface analysis (HSA), were applied to explore molecular interactions between the active pharmaceutical ingredient (API) and CCF. Above all, their pharmaceutical properties, such as powder hygroscopicity, stability, solubility, and the intrinsic dissolution rate (IDR), were evaluated. The results showed that the solubility and IDR of the ARI-ADI salt and its acetone hemisolvate significantly improved. This study provides a potential strategy to increase the solubility of poorly water-soluble drugs and gain a comprehensive understanding of the structure–property relationship.

2. Materials and Methods

2.1. Materials

The aripiprazole was purchased from Sun Chemical & Technology (Shanghai, China) Co., Ltd., and the CCF were purchased from J&K Scientific Ltd. (Beijing, China). All chemicals were used without further purification. All reagents used for this study were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Purified water was prepared by Millipore UP Water Purification System (Merck, Kenilworth, NJ, USA).

2.2. Preparation of Multicomponent Crystals

ARI-SAL salt. The equimolar amounts of ARI (224.2 mg, 0.5 mmol) and SAL (69.1 mg, 0.5 mmol) were dissolved in 30 mL of acetonitrile-water mixed solvent (v/v, 1:1) at 60 °C. The resulting solution was filtered and then kept for eight days at 35 °C. White block crystals were obtained. The bulk samples were prepared by slurry method. A mixture of ARI (448 mg, 1 mmol) and SAL (138 mg, 1 mmol) was added to 6 mL of acetonitrile-water mixed solvent (v/v, 1:1) and stirred for 12 h at a speed of 350 rpm. The solid obtained by filtering the solution was dried to a constant weight in a vacuum oven at 30 °C.

ARI-ADI salt acetone hemisolvate. The mixture of ARI (224.2 mg, 0.5 mmol) and ADI (36.5 mg, 0.25 mmol) was dissolved in 20 mL of acetone at 60 °C. The resulting solution was filtered and then kept for four days at 35 °C. White block crystals were obtained. The bulk samples were prepared by slurry method. A mixture of ARI (448 mg, 1 mmol) and ADI (73 mg, 0.5 mmol) was added into 6 mL of acetone solution and stirred for 12 h at a speed of 350 rpm. The solid obtained by filtering the solution was dried to a constant weight in a vacuum oven at 30 °C.

Additionally, ARI-ADI salt was obtained by drying ARI-ADI salt acetone hemisolvate in a vacuum oven at 70 °C for 2 h.

2.3. Characterization

2.3.1. Single Crystal X-ray Diffraction (SXRD)

The crystal structures were determined with a Bruker APEX-II CCD diffractometer using graphite monochromatic Mo-Kα radiation (λ = 0.71073 Å) at 296K. Frame integration was performed using SAINT (version 7.68A) [19]. The resulting raw data were scaled and the absorption was corrected using a multi-scan averaging of symmetry-equivalent data by SADABS [20]. The structure was solved by the direct method with the olex2 software and then refined via full-matrix least-squares procedures using SHELXL-2014 on F2 with anisotropic displacement parameters (ADPs) for non-hydrogen atoms [21,22]. The H atoms were located in idealized difference Fourier maps and refined as riding models with isotropic thermal parameters (Uiso(H) = 1.2 Ueq(C), Uiso(H) = 1.2 Ueq(N), and Uiso(H) = 1.5 Ueq(O)).

Diamond [23] (version 4.6.3, Crystal Impact GbR, Bonn, Germany) was used for preparing crystal packing diagrams.

2.3.2. Powder X-ray Diffraction (PXRD)

Powder X-ray diffraction (PXRD) patterns were recorded on a Bruker D8 ADVANCE X-ray powder diffractometer (Bruker, Germany) with Cu-Kα radiation at 40 KV and 40 mA. After sieving through 100 mesh, about 50 mg samples were measured in the 2-theta range of 5–50° at a scan rate of 8°/min.

2.3.3. Differential Scanning Calorimetry (DSC)

DSC analyses were taken on a Mettler Toledo DSC1 instrument (Mettler, Zurich, Switzerland). An amount of 3~5 mg samples were put into aluminum pans with pinhole lids and heated in the 30~300 °C temperature range at a constant heating rate of 10 °C/min under a nitrogen flux of 50 cm³/min.

2.4. Computational Studies

2.4.1. Acid Dissociation Constant (pKa)

The formation of the resultant supramolecules can be predicted according to the pKa difference (ΔpKa = pKa[base] − pKa[acid]) of corresponding acid/base pairs [24,25]. It is generally accepted that a salt will be formed if the ΔpKa value is greater than 3, and a ΔpKa value less than 0 will lead to the formation of cocrystals. However, if the ΔpKa value is in the range of 0~3, it will be not accurate to predict the resulting formation [25]. The pKa values of ARI and CCF were calculated by MarvinSketch 15.6.29 (ChemAxon, Budapest, Hungary) [26].

2.4.2. Molecular Electrostatic Potential Surface (MEPS)

The molecular structures of ARI and CCF were extracted from their crystal structures. Full geometry optimization and wave functions were performed by density function theory (DFT) using the B3LYP hybrid functional with 6-311G(d) basis set in the Gaussian 09 software. MEPS was mapped on the 0.001 a.u. electron density isosurface and analyzed by the Multiwfn program [27] and VWD program [28].

2.4.3. Hirshfeld Surface Analysis (HSA)

HSA and fingerprint plots were performed by CrystalExplorer 17.5 software [29], providing information about the nature of intermolecular interactions and their quantitative contribution to the Hirshfeld surface [30].

2.5. Powder Hygroscopicity

Dry glass weighing bottles (outside diameter 50 mm, height 25 mm) were placed in a dryer with ammonium chloride saturated solution at 25 °C ± 1 °C and weighed (m1) after 24 h of storage. Then, about 100 mg samples were put into the bottles and weighed accurately (m2), respectively. After storage at the aforementioned conditions with the caps opened for 24 h, each sample was weighed again (m3). The hygroscopicity could be calculated by the following equation.

Mass Change (%) = (m3 − m1) / (m2 − m1) × 100%.

2.6. Stability Test

The stabilities of ARI-SAL salt and ARI-ADI salt acetone hemisolvate were studied under high-temperature (60 ± 1 °C) and high-humidity (95 ± 5%) tests. Materials were randomly selected and exposed to high temperature and high humidity conditions. The storage times were 10 days and PXRD was applied to measure the final samples.

2.7. Solubility Experiment

2.7.1. Solubility Test

The solubility measurements were performed under water (pH 7.0), hydrochloric acid buffer (pH 1.2), acetate acid-sodium acetate buffered solution (pH 4.5), and phosphate buffer solution (pH 6.8), respectively. An excess number of samples were added to a test tube containing 10 mL of solvent and the tube was shaken in an orbital shaker (37 ± 0.5 °C) until reaching the equilibrium condition (48 h). The solution was filtered through a 0.45 µm mixed cellulose ester membrane and then analyzed by HPLC (LC-20, Shimadzu, Japan) equipped with a SP-20 ultraviolet detector at 254 nm wavelength. Quantitative tests were performed on a YMC C8 column (250 × 4.6 mm, 5 µm) with an external standard method. The mobile phase consisted of acetonitrile and water (containing 0.1% trimethylamine) (65:35, v/v), with a flow rate of 1.0 mL/min. Each test was performed in triplicate.

2.7.2. Intrinsic Dissolution Rate (IDR) Test

IDR is a key physicochemical parameter commonly used to assess in vivo dissolution and reflect bioavailability of drugs [31]. In this work, IDR was measured by the rotary basket method, which was applied to distinguish their dissolution properties. Prior to the IDR test, round discs of the samples should be compressed with a hydraulic press (Jintan Ruiding Machinery Co., Ltd. Changzhou, China). Specifically, 300 mg samples were compressed at a pressure of 115.2 MPa for 10 s to form smooth discs 8 mm in diameter. The acquired discs were coated with beeswax on three sides. The intrinsic dissolution study was performed at 100 rpm in 500 mL of hydrochloric acid buffer (pH 1.2) as a dissolution medium at 37 °C. The ARI concentration in solution was measured at the predetermined time interval by the same analysis method as the solubility test. The sink conditions were maintained during the entire dissolution experiment, and each test was performed in triplicate.

3. Results and Discussion

3.1. Characterization

3.1.1. SXRD Analysis

The crystal structures revealed that both ARI-SAL salt and ARI-ADI salt acetone hemisolvate were salts and the protonation occurred at the N₂ atom, forming a strong charge assisted hydrogen bond formed by the strongest acceptor from the carboxylate anion interacting with strongest N⁺-H donor of piperazinium (see Appendix A). The crystallographic data are listed in

Table 1. Crystallographic data of two multicomponent crystals of aripiprazole.

Compound Reference	ARI-SAL Salt	ARI-ADI Salts Acetone Hemisolvate
Chemical formula	C23H28Cl2N3O2·C7H5O3	C23H28Cl2N3O2·0.5(C6H8O4)·0.5(C3H6O)
Formula mass	586.49	550.48
Crystal system	monoclinic	triclinic
Space group	P 21/c	P-1
a/Å	15.082 (2)	7.6388(12)
b/Å	9.6912 (13)	10.7268(17)
c/Å	21.220 (3)	18.390(3)
α/°	90	97.229(3)
β/°	106.743 (2)	93.641(3)
γ/°	90	105.529(3)
Unit cell volume/Å3	2970.1 (7)	1432.9(4)
Temperature/k	296(2)	296(2)
No. of formula units per unit cell, Z	4	2
Crystal density (g/cm3)	1.312	1.276
No. of reflections measured	9612	7725
No. of independent reflections	6483	4047
Rint	0.0289	0.0321
Final R1 values (I > 2σ(I))	0.0806	0.0689
Final wR (F2) values (I > 2σ(I))	0.2526	0.2135
Final R1 values (all data)	0.1127	0.1308
Final wR (F2) values (all data)	0.2848	0.2725
F(000)	1232	582
Goodness of fit on F2	1.084	0.963
CCDC Number	1991810	2023732

and the hydrogen bonding parameters are given in

Table 2. Main hydrogen bonding parameters of two multicomponent crystals of ARI.

	D-H⋯A	D-H/Å	H⋯A/Å	D⋯A/Å	D-H⋯A/°
ARI-SAL Salt	C16-H16A⋯Cl1	0.97	2.675	3.248	118
	O5-H5⋯O4	0.82	1.789	2.497	143
	N2-H2⋯O3	0.98	1.691	2.670	177
	N2-H2⋯O4	0.98	2.503	3.122	121
	N1-H1⋯O1i	0.86	2.062	2.908	168
	C17-H17A⋯O5ii	0.97	2.542	3.467	160
	C27-H27⋯O1iii	0.93	2.360	3.287	174
Symmetry transformation: i: −x+2, −y, −z+1; ii: −x+1, y−1/2, −z+3/2; iii: −x+2, y+1/2, −z+3/2.
ARI-ADI salt acetone hemisolvate	C16-H16A⋯Cl1	0.97	2.595	3.214	121
	N1-H1⋯O1i	0.86	2.023	2.869	167
	N2-H2⋯O3	0.98	1.729	2.685	164
	C15-H15A⋯Cl1ii	0.97	2.957	3.448	112
	C11-H11A⋯π(Cg)iii	0.97	2.96	3.843	151
Symmetry transformation i: −x+1, −y−1, −z; ii: x+1, y, z; ii: x, −1+y, z. Cg is the centroid of C18/C19/C20/C21/C22/C23 atoms

. The C₁₆-H_16A⋯Cl₁ hydrogen bonds in Table 2 correspond to the typical intramolecular interaction characteristic of the dichlorophenyl-1-piperazinyl group in all reported ARI variants [16,17,18].

ARI-SAL salt. The ARI-SAL salt crystallized in the monoclinic P2₁/c space group with an aripiprazole cation, balanced by a salicylic counter-ion, in the asymmetric unit (Figure 1a). The O₅ atom of salicylic anion presented positional disorder over two sites with a 0.6: 0.4 site-occupancy. ARI-SAL salt, like other lactam compounds, formed a centrosymmetric N₁-H₁⋯O₁ dimer $R_2^2$(8) motif with its inversion-related molecule [32]. Interestingly, the dimer formed a 1D ribbon through the short Cl₁⋯O₁ interaction, which formed another new dimeric substructure (Figure 2a). Furthermore, the dimer units were arranged helically into the three-dimensional network (Figure 2b) by the bifurcated N-H⋯O hydrogen bonds (N₂⁺-H₂⋯O₃- and N₂-H₂⋯O₄ hydrogen bond) and two weaker C-H⋯O interactions (C₂₇-H₂₇⋯O₁ and C₁₇-H_17A⋯O₅ hydrogen bond), which generated two types of helices propagated along the b axis (Figure 2c,d).

ARI-ADI salt acetone hemisolvate. The ARI-ADI salt acetone hemisolvate crystallized in the triclinic P-1 space group. The ORTEP diagram contained an aripiprazole cation, half of an adipic acid counter-ion (bisected by an inversion center), and half of an acetone molecule (Figure 1b). The solvent acetone molecule was disordered with a position occupancy of 0.5 due to the location on the symmetric center, and acted as a filler in spatial network. The centrosymmetric N₁-H₁⋯O₁ dimer $R_2^2$(8) motif with its inversion-related molecule also can be found and formed a 1D ribbon by the strongest charged N₂⁺-H₂⋯O₃ hydrogen bond (Figure 3a). The 1D ribbon interacted with its adjacent ribbon and then formed a 2D sheet through a C₁₅-H_15A⋯Cl₁ hydrogen bond (Figure 3b). Finally, the 2D sheets were arranged into the three-dimensional network by a weak C₁₁-H_11A⋯π hydrogen bond (Figure 3c).

3.1.2. PXRD Analysis

The ARI multicomponent crystals were synthesized and characterized by PXRD. As seen in Figure 4, their experimental patterns show significant differences compared with those of ARI, which are confirmed as new crystalline forms due to the absence of the characteristic peak of ARI at 20.483° and 22.199°. The characteristic PXRD peaks of the ARI-SAL salt were 12.948°, 18.401°, and 19.849°, while those of the ARI-ADI salt were 25.079°, 14.976°, and 18.256°, which were different from those of its acetone hemisolvte with characteristic PXRD peaks at 18.197° and 24.897°. Importantly, the experimental patterns of ARI-ADI salt acetone hemisolvate and ARI-SAL salt coincided well with their own calculated PXRD patterns from single crystal structures, indicating that the bulk pure samples were prepared successfully.

3.1.3. DSC Analysis

As reported by Zhao Yanxiao [18], the DSC thermogram (Figure 5) of ARI exhibited two endothermic peaks and a small exothermic peak. The endothermic peak at 138.94 °C corresponded to the melting point of form III, while the second endothermic peak at 147.79 °C represented the melting point of form I, demonstrating that form III might be transformed to form I during the heating process. As expected, the DSC thermograms of ARI multicomponent crystals were distinct from those of ARI and CCF. A single melting endothermic peak was observed at 182.38 °C for the ARI-SAL salt, indicating the formation of pure solid phase. For the ARI-ADI salt acetone hemisolvate, a weak endothermic event of desolvation and a sharp melting endothermic peak were observed at 57.43 °C and 121.18 °C, respectively, which implied that it had a poor thermodynamic stability and may transform into a desolvated salt. As we intended, the ARI-ADI salt showed a single melting endothermic peak at 120.36 °C.

3.2. Theoretical Calculation

3.2.1. Acid Dissociation Constant (pKa)

ARI is a weak base with a pKa of 7.46, and both salicylic acid and adipic acid are weak acid excipients with pKa values of 2.79 and 3.92, respectively. According to the ΔpKa rule, if the difference of pKa between active pharmaceutical ingredients and CCF was greater than 3, the binary mixture tended to form a salt type [25]. Therefore, we predicted that both of the synthesized multicomponent crystals of ARI were salts, which could be confirmed by especially single-crystal X-ray diffraction.

3.2.2. Molecular Electrostatic Potential Surface

MEPs can provide some clear information in terms of the electrophilic and nucleophilic attack region of the molecule. The values of MEP, related to the strength of intermolecular interactions, can predict the formation possibility of multicomponent crystals. The MEPs mapped on van der waals (VDW) surface of ARI, SAL, and ADI are given in Figure 6, where the red color represents a positive electrical potential and the blue color represents a negative electrical potential. The pKa values were calculated by MarvinSketch and are presented in the black color.

As reported by Zhao Yanxiao [18], the global maxima and minima values of the electrostatic potential on the surface in ARI were +36.98 and −41.23 kcal/mol, which corresponded to carbonyl and amino group of the lactam ring, respectively, and formed the N-H⋯O homodimers as evidenced in this study. The N₂ and N₃ atoms of the piperazinyl group had little difference in electrical potential, but the N₃ atom had the strong p-π conjugative effect with adjacent benzene ring. As a result, the N₂ atom was vulnerable to electrophilic attack, and no ARI variant containing an exclusive protonation of the N₃ atom was observed [16]. The global maxima value of the electrostatic potential on the surface in SAL was +53.86 kcal/mol, which corresponded to carboxylate group with a low pKa value (pKa = 2.79). As result, the deprotonated N⁺-H⋯O hydrogen bond was formed in ARI-SAL salt. For ADI, the global and secondary maxima values were +49.71 and +49.66 kcal/mol, which corresponded to both of the carboxylate groups with identical nucleophilic reactivity, respectively. As expected, the deprotonated N⁺-H⋯O hydrogen-bonding contacts occurred in real ARI-ADI salt, where the stoichiometric ratio of ARI to ADI was 1:0.5.

3.2.3. Hirshfeld Surface Analysis

The 3D Hirshfeld surface and 2D fingerprint plots of two multicomponent crystals of ARI are given in Figure 7. In their 3D Hirshfeld surface, the large and deep red spots correspond to the close-contact N-H⋯O interactions. The H⋯H contacts appear as scattered points in the middle region of the 2D fingerprint plots, which were the predominate type of interactions. The relative percentage contributions of H⋯H contacts in theARI-SAL salt were markedly lower than that of the ARI-ADI salt acetone hemisolvate and compensated by the C⋯H/H⋯C contacts shown as a pair of “wings”. However, these interactions displayed higher interatomic distance. Interactions with less interatomic distance are presented in fingerprint plots as pointy regions. The reciprocal O⋯H contacts, corresponding to O-H⋯O, N-H⋯O, and C-H⋯O interactions, are presented as long, sharp, symmetrical spikes. Additionally, the Cl⋯H contacts, corresponding to the Cl⋯H-C, Cl⋯H-O and Cl⋯H-N interactions, appear as a pair of “wings” and have equivalent relative percentage contributions to their own Hirshfeld surfaces.

3.3. Powder Hygroscopicity

The moisture absorption is a crucial parameter that should be considered during the process of drug development, production, and storage. Therefore, the powder hygroscopicity was measured under 25 °C/80% RH condition. Ten days later, the hygroscopic gain of ARI was 0.32 ± 1.30%. Meanwhile, ARI-SAL salt and ARI-ADI salt and its acetone hemisolvate showed greater hygroscopicity, with hygroscopic gains of 2.60 ± 0.78%, 1.36 ± 0.32%, and 1.13 ± 0.7%, respectively. This might be presumably due to the presence of a polar group (i.e., –OH and –COOH) introduced by synthesizing carboxylate salts, leading to the formation of hydrogen bonds with water molecules [33].

3.4. Stability Test

Moisture and high temperatures may cause the transformation of drug crystal forms and degradation, especially for solvates. Thus, close attention was paid to their stability under high humidity and high temperatures in this work. Figure 8 illustrates that the ARI-SAL salt was stable, while the ARI-ADI salt acetone hemisolvate was unstable under a high temperature. There were significant differences at the two-theta range of 19° to 27° compared with the starting PXRD pattern. In conjunction with DSC patterns of ARI-ADI salt acetone hemisolvate, we speculated that desolvation occurred under high temperatures. Unexpectedly, the PXRD pattern of the ARI-ADI salt was inconsistent with that of its acetone hemisolvate under high temperatures, demonstrating that ARI-ADI salt might be unstable under high temperatures.

3.5. Solubility

3.5.1. Equilibrium Solubility Test

The results of the equilibrium solubility experiments are summarized in Figure 9. The equilibrium solubility of the ARI-ADI salt in water, hydrochloric acid buffer, and acetate acid-sodium acetate buffered solution were 0.13, 0.63, and 0.15 mg/mL, respectively, which was much better than that of the ARI and ARI-SAL salt but slightly smaller than that of the ARI-ADI salt acetone hemisolvate. However, the equilibrium solubility of the ARI-SAL salt was even worse than that of the ARI in both the hydrochloric acid buffer solution and acetate acid-sodium acetate buffered solution, though slightly better in water. In addition, the equilibrium solubility of ARI and its salts in the phosphate buffer solution were too low to be acquired accurately.

3.5.2. IDR Test

Given its kinetic nature, IDR assumes a better correlation with in vivo drug dissolution dynamics than solubility [34]. Thus, IDR studies were carried out at pH 1.2 in this work, and the results are summarized in Figure 10. The ARI held an IDR of 0.5051 mg·cm⁻²·min⁻¹, while the IDR of the ARI-ADI salt, slightly lower than that of its acetone hemisolvate, significantly increased to 0.9263 mg·cm⁻²·min⁻¹ and was about twice as much as that of the ARI. Not unexpectedly, the IDR of the ARI-SAL salt was only 0.3745 mg·cm⁻²·min⁻¹, significantly lower than that of the ARI.

These findings indicated that ARI-ADI salt and its acetone hemisolvate had predictable advantages in vivo absorption over ARI-SAL salt and the untreated ARI. This may be due to many causes. First, the relatively higher hygroscopicity and lower melting point may lead to easier dissolution in an aqueous solvent, because of it demanding less lattice energy to break. Molecular constituents were also one of the important causes. The CCF of ARI-ADI salt was more soluble than that of the ARI-SAL salt, and their solubility behaviors follow the rule of thumb that the greater solubility of the CCF, the more soluble the salt will be [18]. Finally, their spatial structures may also be associated with their differences in solubility. The ADI molecules linked with the solvent molecules to form a hydrophilic layer. As a result, water molecules permeate more easily into the layer stacking spatial structure and cause it to disintegrate in aqueous solvent.

4. Conclusions

In this study, a new ARI-ADI salt and its acetone hemisolvate were successfully synthesized and fully characterized by various methods, along with a reported ARI-SAL salt. Their crystal structures revealed that the proton transferring occurred from a carboxylic group of CCF to the N₂ atoms of the piperazine region from the ARI. Structural analysis showed that the ARI-SAL salt was arranged helically by two helices propagated along the b axis, while the ARI-ADI salt acetone hemisolvate presented a typical layer stacking with the help of weak C-H⋯Cl and C-H⋯π hydrogen bonds. The reactivity of ARI and CCF was predicted by MEPS and matched well with their hydrogen bond schemes. Hirshfeld surface analysis was also used to clarify the intermolecular interactions. Furthermore, the ARI-ADI salt had a significant advantage over the ARI and ARI-SAL salt in terms of equilibrium solubility and IDR. This study provided a valuable insight into the formation of multicomponent pharmaceutical salts and presented potential alternative formulations of ARI.