The biorelevant concentration of Tween 80 solution is a simple alternative medium to simulated fasted state intestinal fluid

Abstract

The aim of the present study is to investigate whether the use of the biorelevant concentration of conventional surfactants as an alternative medium to simulated fasted state intestinal fluid (FaSSIF) for drugs with different acid–base properties is feasible. First, the equilibrium solubility of seven drugs representing diverse acid–base properties was determined. The solubility of those drugs was studied in pH 6.5 blank FaSSIF, FaSSIF containing physiological levels of sodium taurocholate and lecithin, and blank FaSSIF containing a series of concentrations of Tween 80 or sodium dodecyl sulfate (SDS) at 37 °C using the shake-flask method. The results demonstrated that the solubility of the seven drugs in 0.07% and 0.10% Tween 80 solutions matched better with the biorelevant media FaSSIF than SDS solutions (0.05% to 0.10%). Then the dissolution behavior of the BCS class II drug lacidipine was determined in SDS solutions, Tween 80 solutions and FaSSIF. Compared with FaSSIF, 0.07% Tween 80 was proved to be the best biorelevant-mimetic medium for lacidipine oral solid formulations. Then the dissolution tests of micronized lacidipine, carbamazepine, glimepiride and carvedilol tablets were conducted as internal and external validations, respectively. The results all demonstrated that the biorelevant concentration of Tween 80 could be set as 0.07% and such a cost-effective medium will hold a great potential in drug development and quality control of oral solid dosage forms.

---

Introduction

Dissolution tests are of great importance in the development of oral drug products for two significant objectives, i.e., to act as a quality control (QC) specification for ensuring bath to bath consistency, and to establish in vitro/in vivo correlation (IVIVC) for forecasting drug in vivo performance.¹ However, the simple aqueous buffer dissolution media with or without certain amounts of surfactants commonly adopted by the pharmacopeia are principally used to determine manufacturing variables and stability changes during storage, but usually fail to discriminate different products with different release behavior and are then unable to guarantee the in vivo bioequivalence (BE) and efficacy.^2–4

Biorelevant media are adapted to simulate human small intestinal fluids, including fasted state simulated intestinal fluid (FaSSIF) and fed state simulated intestinal fluid (FeSSIF) in terms of pH, buffer capacity and osmolality, as well as bile salt and phospholipids’ concentration.^5,6 Biorelevant media have been proved valuable in establishing IVIVC for poorly soluble drugs.⁷ Though the biorelevant media play an important role in formulation development, their high labor requirements and great cost restrict their routine application in industrial and research fields, especially for the flow-through cell method which requires a large volume of dissolution media.^8,9 In addition, biorelevant media are surfactant-enriched solutions and the application of these media is also restricted in biphasic dissolution systems.¹⁰ As a consequence, it would be desirable if conventional surfactants could be used to replace these biorelevant media for evaluating the dissolution behavior of oral solid formulations.¹¹

Zoeller and Klein¹² tried to simplify biorelevant media with a single surfactant for FaSSIF and a surfactant mixture for FeSSIF. They found the type and concentration of surfactants had a remarkable influence on the surface tension of the media, while there was little impact by pH, osmolality, and buffer capacity. Lehto et al.¹³ evaluated such properties as solubilization, wettability, diffusivity and surface tension between two conventional surfactant solutions (anionic surfactant sodium dodecyl sulfate (SDS) and non-ionic surfactant Tween 80) and FaSSIF. However, only neutral drugs were determined in this report. Peter et al. found that acidic and basic drugs showed different solubility behavior compared with neutral drugs in FaSSIF.¹⁴ So it is of great importance to investigate whether the use of conventional surfactant solutions to substitute FaSSIF for drugs with different acid–base properties is feasible.

In the study, we measured the solubility of seven drugs representing different acid–base properties in two conventional surfactant solutions (anionic surfactant sodium dodecyl sulfate (SDS) and non-ionic surfactant Tween 80), and compared the results of FaSSIF to those of the two conventional surfactant solutions. Then the dissolution behavior of the BCS class II drug lacidipine was determined in these media to confirm the results. The dissolution tests of micronized lacidipine, carbamazepine, glimepiride and carvedilol tablets were conducted as internal and external validations, respectively.

Results and discussion

The solubility in FaSSIF and SDS solutions

Seven poorly soluble drugs were selected as the model compounds to analyze their apparent solubility in pH 6.5 blank FaSSIF, FaSSIF and blank FaSSIF containing a series of SDS concentrations. These drugs representing different acid–base properties include three neutrals, two acids and two bases (Table 1). We intended to determine the extent to which it is possible to increase the apparent solubility of poorly soluble compounds in FaSSIF and SDS solutions. Bearing this in mind, the drugs’ solubility in SDS and FaSSIF solutions are summarized in Table S1 (ESI†).

Table 1 Physicochemical parameters of the seven drugs

Compound	Acidic drugs		Neutral drugs			Basic drugs
	Indomethacin	Glimepiride	Carbamazepine	Spironolactone	Lacidipine	Carvedilol	Itraconazole
a Data provided by Drugbank database.b Taken from ref. 15.
log P	4.27^a	3.50^a	2.45^a	2.78^a	5.51^b	4.19^a	5.66^a
pKa	4.50^a	4.32^a	15.96^a	18.01^a	2.50^b	8.74^a	3.70^a

---

The S_SDS/S_FaSSIF ratio helps to analyze the relationship between SDS solutions and FaSSIF and can be expressed as follows:

where S_SDS is solubility of drugs obtained in SDS solutions and S_FaSSIF is solubility of drugs obtained in FaSSIF.

The equilibrium solubility of the three neutral drugs (lacidipine, spironolactone, and carbamazepine) in the SDS solution gradually increased along with the SDS concentration between 0.05–0.10%, but it sharply increased in 0.3% and 0.5% SDS solutions, due to the formation of SDS micelles.¹⁶ Additionally, the solubility of spironolactone and carbamazepine in 0.3% and 0.5% SDS solutions were 10-fold higher than lacidipine. The reason for this phenomenon was that spironolactone and carbamazepine had better wetting characteristics than lacidipine.¹³ Fig. 2a shows the S_SDS/S_FaSSIF change of the three neutral drugs with various SDS concentrations. With FaSSIF as the control, the S_SDS/S_FaSSIF for 0.05%, 0.07% and 0.10% SDS matched well with the biorelevant media FaSSIF. Even though the three neutral drugs have different wetting characteristics and lipophilicity, they all showed similar solubility in FaSSIF compared with SDS solutions (0.05–0.1%).

Fig. 1 Chemical structures of the seven selected drugs.

Fig. 2 S_SDS/S_FaSSIF of seven drugs with different acid–base properties in FaSSIF and SDS solutions. S_SDS/S_FaSSIF is the ratio between the solubility of drugs representing different acid–base properties obtained in SDS solutions as compared to FaSSIF ((a): neutral drugs, (b): acidic drugs, (c): basic drugs).

The glimepiride and indomethacin represented weakly acidic drugs. The solubility of glimepiride was still very low in SDS solutions. The large increase in solubility resulted in a shift from BCS class II to class I in the apparent classification of indomethacin in FaSSIF and SDS solutions.¹⁷ Although the two acidic drugs manifested different solubility-changing trends in SDS solutions, the relationship between FaSSIF and SDS solutions was further studied. As shown in Fig. 2b, the increase in S_SDS/S_FaSSIF of glimepiride in 0.05% SDS, 0.07% SDS and 0.10% SDS was 1.5-fold, 2-fold and 3-fold, respectively. Based on these data, the solubility of glimepiride in SDS solutions ranging from 0.05–0.10% gradually became higher than that in FaSSIF.

The solubility of carvedilol, a model basic drug, in the 0.05% and 0.10% SDS solutions, with concentration below CMC, showed a slight decrease. Since carvedilol was in its cationic state at pH 6.5 and SDS was predominantly in an anionic state, SDS then interacted with the cationic carvedilol to form an insoluble ionic complex leading to reduced solubility. When the concentration of SDS rose above CMC (10 mmol L⁻¹), the solubility of carvedilol was increased in 0.3% and 0.5% SDS solutions, due to micellar solubilization. In addition, SDS interacted with the cationic carvedilol, perhaps enhancing the partition of carvedilol into the cationic micelles, rather than the formation of the insoluble precipitates.¹⁶ As for itraconazole, its solubility in FaSSIF and SDS solutions was much like the neutral drugs, since it is a very weak drug (pK_a = 3.70) and in its molecular form at pH 6.5. As shown in Fig. 2c, the S_SDS/S_FaSSIF of carvedilol for 0.05–0.10% SDS was less than one, but that of itraconazole increased with SDS concentration, ranging from 10- to 100-fold. Based on the above results, SDS solutions ranging from 0.05–0.10% generally showed a higher solubilization capability than FaSSIF for drugs representing different acid–base properties.

The solubility in FaSSIF and Tween 80 solutions

The solubility of the seven drugs in Tween 80 and FaSSIF solutions are summarized in Table S2 (ESI†). The equilibrium solubility of the seven drugs in Tween 80 solutions gradually increased along with its concentration between 0.01–0.30%. It was observed that the increasing trends in Tween 80 solutions were less sharp than in SDS solutions. The lower solubility in Tween 80 may be largely related to the larger molecular weight of Tween 80 and the limited number of Tween 80 micelles.¹⁸ The solubility of carvedilol increases gradually at all concentrations (0.01–0.30%) of Tween 80, which is different from the situation in SDS solutions. Being non-ionic in nature, Tween 80 didn’t exhibit any kind of ionic interaction with basic drugs like carvedilol which would not precipitate in Tween 80 solutions.¹⁶ The solubility of indomethacin in FaSSIF and SDS solutions was greatly increased, and the phenomenon resulted in a shift from BCS class II to class I in the apparent classification. The same phenomenon was also found in Tween 80 solutions with indomethacin.¹⁷

The S_{Tween 80}/S_FaSSIF ratio helps to analyze the relationship between Tween 80 solutions and FaSSIF and can be expressed as follows:

where S_{Tween 80} is the solubility of drugs obtained in Tween 80 solutions and S_FaSSIF is the solubility of drugs obtained in FaSSIF.

As shown in Fig. 3, the S_{Tween 80}/S_FaSSIF of drugs increased with Tween 80 concentration, and these drugs (spironolactone, carbamazepine, glimepiride, indomethacin and carvedilol) all show a slightly increasing trend. With FaSSIF as the control, the solubility of the seven drugs in 0.07% and 0.10% Tween 80 solutions matched well with the biorelevant media FaSSIF, indicating that Tween 80 solutions and FaSSIF have nearly equivalent solubilization ability. To sum up, the solubilization behavior of the seven drugs representing different acid–base properties in Tween 80 solutions ranging from 0.07% to 0.10% was closer to the biorelevant media FaSSIF than SDS solutions (0.05–0.10%).

Fig. 3 S_{Tween 80}/S_FaSSIF of seven drugs with different acid–base properties in FaSSIF and Tween 80 solutions. S_{Tween 80}/S_FaSSIF is the ratio between the solubility of drugs obtained in Tween 80 solutions as compared to FaSSIF ((a): neutral drugs, (b): acidic drugs, (c): basic drugs).

Dissolution in FaSSIF and SDS solutions of the two lacidipine formulations

Solubility plays an important role in the dissolution of a drug substance from oral solid dosage forms. As we can see from the above results, the solubilizing effect of SDS solutions (ranging from 0.05% to 0.10%) and Tween 80 solutions (0.07% and 0.10%) for neutral drugs is in good agreement with that of FaSSIF. So we selected a neutral drug (lacidipine) as a model compound to investigate the dissolution behavior in SDS solutions, Tween 80 solutions and FaSSIF.

The solubility of lacidipine in 0.05% SDS, 0.07% SDS, 0.10% SDS and FaSSIF was 2.4, 4.2, 6.9 and 4.6 μg mL⁻¹, respectively. Therefore, the three SDS concentrations of 0.05%, 0.07% and 0.10% were selected for further investigation of their comparable solubilizing capacity to FaSSIF.

Fig. 4 shows the dissolution profiles of two formulations of lacidipine in FaSSIF and the above three SDS solutions with concentrations ranging from 0.05% to 0.1%. The dissolution of formulation A was complete in FaSSIF and SDS solutions within 30 min (Fig. 4A), but the dissolution rate was slower in FaSSIF than in SDS solutions, due to the higher translational diffusion coefficient of SDS than FaSSIF (12 × 10⁻⁶ versus 0.69 × 10⁻⁶ cm² s⁻¹).¹³ However, the dissolution of formulation B was incomplete even within 60 min and the ranked order of the dissolution media was as follows: 0.10% SDS > 0.07% SDS > FaSSIF > 0.05% SDS (Fig. 4B). The dissolution rate of formulation B was in parallel with the SDS concentration. With FaSSIF as the control, the calculated f₂ values for 0.05% SDS, 0.07% SDS and 0.10% SDS solutions were 46, 44 and 47 for formulation A and 70, 69 and 46 for formulation B, respectively. Apparently, the SDS solutions didn’t match well with the biorelevant media FaSSIF. It was reported that SDS at concentrations above the CMC value resulted in a surface tension of about 30 mN m⁻¹, which is much lower than that of FaSSIF (45 mN m⁻¹).^1,13 Thus the surfactant SDS appeared unsuitable as an alternative to FaSSIF. Meanwhile, the results indicated that the dissolution of formulation B was extremely sensitive to the concentration of surfactants, and as a result the selection of surfactant concentration should be judicious.

Fig. 4 Dissolution profiles of two lacidipine formulations in FaSSIF and SDS solutions ((A): formulation A; (B): formulation B, data are expressed as mean ± S.D., n = 3).

Dissolution in FaSSIF and Tween 80 solutions of the two lacidipine formulations

It has been demonstrated that Tween 80 has closer solubilization properties and surface tension to FaSSIF than SDS.¹³ So the dissolution behavior in Tween 80 solution was further investigated. As shown in Table S2 (ESI†), the solubility of lacidipine was very low in 0.01% Tween 80, which is below its CMC value (approximately 0.012%).¹⁹ Then the solubility increased linearly with Tween 80 concentration from 0.03% to 0.30%. The solubility in 0.07% and 0.10% Tween 80 was close to that in FaSSIF, and so these two levels of Tween 80 were chosen to conduct further investigations.

As shown in Fig. 5, the two formulations demonstrated similar dissolution profiles in 0.07% Tween 80 solution to FaSSIF, with f₂ values being 69 for formulation A and 83 for formulation B, respectively. In 0.10% Tween 80 solution, formulation A showed a faster dissolution rate and formulation B exhibited a higher dissolution, compared to those in FaSSIF. The comparable surface tension and lacidipine solubility could contribute to the similar dissolution performances of the two lacidipine formulations in FaSSIF and 0.07% Tween 80. This result was also in line with the previous report that danazol and spironolactone showed the same dissolution efficiency in FaSSIF and 0.07% Tween 80.¹³

Fig. 5 Dissolution profiles of two lacidipine formulations in FaSSIF and Tween 80 solutions (— for formulation A and --- for formulation B, data are mean ± S.D., n = 3).

The dissolution rate was faster in SDS than that in Tween 80, especially for formulation A. As we know, the dissolution process is the combined result of drug solubilization and drug diffusion through the diffusion layer into the dissolution medium. In this respect, the diffusivity of drug molecules and drug–micelle complexes plays an important role. The diffusivity of drug–micelle complexes is several-fold less than the drug molecule, and the net change in the dissolution rate was the sum of enhanced solubility and declined effective diffusivity.^18,20 Compared with SDS, the higher molecular weight of Tween 80 (1310 versus 288.4 g mol⁻¹) and the greater aggregation weight of its micelles (76 000 versus 15 900 g mol⁻¹)^21,22 resulted in lower diffusivity of drug–micelle complexes and hence a slower dissolution rate.¹⁸ Taken together, 0.07% Tween 80 was a better alternative to FaSSIF than 0.05–0.1% SDS solutions.

The verification of the physiologically relevant concentration of Tween 80

Internal validation. Another preparation of lacidipine, self-made micronized tablets, was utilized to determine whether 0.07% Tween 80 solution and FaSSIF could show similar dissolution behavior. The dissolution behavior of the micronized tablets in the two test media was quite similar with the f₂ value being up to 97 (Fig. 6). This result illustrated that the biologically relevant concentration of Tween 80 could be set as 0.07%.

Fig. 6 Dissolution profiles of micronized lacidipine tablets in FaSSIF and 0.07% Tween 80 solution (data are mean ± S.D., n = 3).

External validation. To further verify the biorelevant concentration of Tween 80, a neutral drug carbamazepine, a weakly acidic drug glimepiride and a weakly basic drug carvedilol were used for external validation.

The dissolution of carbamazepine tablets was comparable in FaSSIF and 0.07% Tween 80 (Fig. 7). The dissolutions of two glimepiride commercial tablets were conducted in blank FaSSIF, FaSSIF and 0.07% Tween 80 solution. As a weakly acidic compound, glimepiride tablets demonstrated poor dissolution (10–18%) at physiological pH (blank FaSSIF) and the dissolution could be improved to a certain extent in the biorelevant conditions (Fig. 8). In FaSSIF and 0.07% Tween 80 solution, the dissolutions of formulation A and B were both increased by two-fold. With FaSSIF as the control, the f₂ values were 81 for A and 83 for B in 0.07% Tween 80. As shown in Fig. 9, the dissolution of carvedilol tablets was comparable in FaSSIF and 0.07% Tween 80. The dissolution of commercial tablets in 0.07% Tween 80 solution and FaSSIF was ∼80% within 1 h. With FaSSIF as the control, the f₂ values were 68 for carvedilol tablets in 0.07% Tween 80.

Fig. 7 Dissolution profiles of carbamazepine tablets in FaSSIF and 0.07% Tween 80 solution (data are mean ± S.D., n = 3).

Fig. 8 Dissolution profiles of two glimepiride formulations in blank FaSSIF, FaSSIF and 0.07% Tween 80 solution (— for formulation A and --- for formulation B, data are expressed as mean ± S.D., n = 3).

Fig. 9 Dissolution profiles of carvedilol tablets in FaSSIF and 0.07% Tween 80 solution (data are mean ± S.D., n = 3).

The above external validation results of oral solid dosage forms of carbamazepine, glimepiride and carvedilol further confirmed that the 0.07% Tween 80 was a promising alternative to FaSSIF.

As for the fasted state, the condition of the small intestine is relatively complicated which is difficult to simulate with simple surfactants.^23,24 Nevertheless, mixed micelles may be more feasible. Zoeller and Klein¹² proposed the mixed micelles formed by natural bile components and the results showed the dissolution profiles of ketoconazole obtained from a blank FeSSIF consisting of 0.25% Tween 80 and 0.25% triethanolamine were superimposable to those obtained from FeSSIF. In further studies, we will study the combination of Tween 80 and other biosurfactants as an alternative to FeSSIF. To sum up, such biorelevant cost-effective 0.07% Tween 80 solutions would be valuable in drug development and preparation.

Experimental section

Materials and methods

Materials. Lacidipine was obtained from Kangya of NingXia Pharmaceuticals Co. Ltd. (Ningxia, China). The two commercial formulations of lacidipine were formulation A LACIPIL® 4 mg tablet (GlaxoSmithKline, England) and formulation B Sileping™ 4 mg tablet (Harbin Pharmaceutical Group Sanchine Corporation, Harbin, China). Glimepiride was purchased from Dahua Weiye Pharmaceutical Chemical Co. Ltd. (Wuhan, China). Two products of glimepiride were Amaryl® 2 mg tablet and Wansuping™ 2 mg tablet, which were purchased from Aventis Pharma S.P.A (Beijing, China) and Wanbang Biochemical Pharmaceutical Co. Ltd (Jiangsu, China), respectively. Carbamazepine was a gift from Suzhou Wanqing Pharmaceutical Co. Ltd (Suzhou, China). The commercial tablets of carbamazepine were produced by Fudan Fuhua Pharmaceutical Co. Ltd (Shanghai, China). The commercial tablets of carvedilol were purchased from Qilu Pharmaceutical Co. Ltd (Shandong, China). Indomethacin was purchased from Kangya of NingXia Pharmaceuticals Co. Ltd. (Ningxia, China). Spironolactone was obtained from Shandong Xinhua pharmaceutical Co. Ltd. (Shandong, China.). Carvedilol and itraconazole were purchased from Yinhe of Hubei Pharmaceuticals Co. Ltd. (Hubei, China). Sodium taurocholate and sodium taurodeoxycholate were purchased from Beijing Aoboxing Biotech Co. Ltd. (Beijing, China). Egg phosphatidylcholine was purchased from Lipoid GmbH (Germany). Sodium hydroxide pellets, hydrochloric acid, sodium chloride, potassium dihydrogen phosphate, ammonium dihydrogen phosphate, sodium dihydrogen phosphate monohydrate, sodium dodecyl sulfate (SDS) and Tween 80 were all of analytical grade and purchased from Tianjin Bodi Chemical Holding Co. Ltd. (Tianjin, China). Methanol, acetonitrile, acetic acid and diethanolamine were of chromatographic grade. Deionized-distilled water was used throughout the study. The structure of the model compounds is presented in Fig. 1 and their physicochemical parameters are summarized in Table 1.

Methods.
Media preparation. The preparation procedures of the biorelevant media were performed as described previously in the literature. The concentrations of SDS and Tween 80 added to the blank FaSSIF were 0.05–0.50% and 0.01–0.30% (w/v), respectively. The compositions of blank FaSSIF, FaSSIF and blank FaSSIF containing a series of SDS and Tween 80 concentrations are summarized in Table S3 (ESI†).
Determination of the equilibrium solubility. The equilibrium solubility of drugs (including lacidipine, spironolactone, carbamazepine, glimepiride, indomethacin, carvedilol and itraconazole) was studied in pH 6.5 blank FaSSIF, FaSSIF, and blank FaSSIF containing a series of SDS and Tween 80 concentrations by the shake-flask method. An excess amount of drug powder (n = 3) was added into conical flasks containing 30 mL of different media and shaken at 100 rpm for 24 h at 37 °C. The clear solution above the precipitate was collected and centrifuged at 3500 rpm for 10 min. The supernatant was filtered through a 0.22 μm filter. The withdrawn samples were filtered, diluted properly and assayed by HPLC.
Dissolution tests. Dissolution tests were carried out using the USP Apparatus 2 (Paddle) setup (ZRS-8G; TIANDA TIANFA Technology Co., Ltd, Tianjin, China). Dissolution media including blank FaSSIF, FaSSIF, 0.05% SDS, 0.07% SDS, 0.10% SDS, 0.07% Tween 80 and 0.10% Tween 80 were degassed before the dissolution tests proceeded and equilibrated at 37 °C. Each vessel was filled with 500 mL of media. Samples were withdrawn at 5, 10, 20, 30, 45, and 60 min. Dissolution in each medium was carried out in triplicate (n = 3). At predetermined time intervals, 5 mL of samples were withdrawn and immediately filtered through 0.45 μm filters, and 5 mL of fresh medium was replaced. Withdrawn samples were analyzed by the HPLC method.
Preparation of micronized tablets of lacidipine. Lacidipine was micronized with the air-flow crushing method by jet mill. Then the particle size distribution and the crystal form were investigated. The lacidipine micronized tablets were prepared by a wet granulation compression method.
HPLC method. All samples were analyzed by the HPLC method. The chromatographic conditions are specified in Table S4 (ESI†).

Conclusions

The solubility performance of the seven drugs in Tween 80 solutions at 0.07% and 0.10% matched better with the biorelevant media FaSSIF, compared to SDS solutions (0.05–0.10%). With FaSSIF as the selection criteria, the biorelevant concentration of Tween 80 was determined to provide more similar surface tension and dissolution performances. These biorelevant, simple and low-cost dissolution media will hold great potential in the quality control of oral solid dosage forms.

Acknowledgements

We thank financial support from the National Natural Science Foundation of China (No. 81173009) and Technology bureau in Shenyang (No. ZCJJ2013402). We also thank financial support from the Project for New Century Excellent Talents of Ministry of Education (No. NCET-12-1015).

Notes and references

1. K. Gowthamarajan and S. K. Singh, Dissolution Technol., 2010, 17(3), 24–32.
2. C. Xu, M. Zou, Y. Wang, Y. Liu, J. Yan, Y. Wu and G. Cheng, Drug Dev. Ind. Pharm., 2012, 38(6), 679–688.
3. M. Sun, J. Sun, S. He, Y. Wang, Y. Sun, X. Liu and Z. He, Drug Dev. Ind. Pharm., 2012, 38(9), 1099–1106.
4. Z. He, D. Zhong, X. Chen, X. Liu, X. Tang and L. Zhao, Eur. J. Pharm. Sci., 2004, 21(4), 487–491.
5. J. B. Dressman, G. L. Amidon, C. Reppas and V. P. Shah, Pharm. Res., 1998, 15(1), 11–22.
6. E. Galia, E. Nicolaides, D. Hörter, R. Löbenberg, C. Reppas and J. Dressman, Pharm. Res., 1998, 15(5), 698–705.
7. B. M. Lue, F. S. Nielsen, T. Magnussen, H. M. Schou, K. Kristensen, L. O. Jacobsen and A. Müllertz, Eur. J. Pharm. Biopharm., 2008, 69(2), 648–657.
8. W. Brown, Dissolution Technol., 2005, 30(5), 28.
9. M. Kakhi, Eur. J. Pharm. Sci., 2009, 37(5), 531–544.
10. D. J. Phillips, S. R. Pygall, V. B. Cooper and J. C. Mann, J. Pharm. Pharmacol., 2012, 64(11), 1549–1559.
11. T. S. Wiedmann, W. Liang and L. Kamel, Pharm. Res., 2002, 19(8), 1203–1208.
12. T. Zoeller and S. Klein, Dissolution Technol., 2007, 14(4), 8–13.
13. P. Lehto, H. Kortejärvi, A. Liimatainen, K. Ojala, H. Kangas, J. Hirvonen, V. P. Tanninen and L. Peltonen, Eur. J. Pharm. Biopharm., 2011, 78(3), 531–538.
14. T. N. Krisztina, S. Vera, V. Gergely and H. Peter, J. Pharm. Biomed. Anal., 2013, 83, 279–285.
15. C. N. Wu, L. F. Kou, P. Q. Ma, L. F. Gao and B. Li, RSC Adv., 2015, 5, 19844–19852.
16. S. Chakraborty, D. Shukla, A. Jain, B. Mishra and S. Singh, J. Colloid Interface Sci., 2009, 335, 242–249.
17. J. H. Fagerberg, O. Tsinman, N. Sun and K. Tsinman, Mol. Pharm., 2010, 7(5), 1419–1430.
18. A. Balakrishnan, B. D. Rege, G. L. Amidon and J. E. Polli, J. Pharm. Sci., 2004, 93(8), 2064–2075.
19. N. Fotaki, W. Brown, J. Kochling, H. Chokshi, H. Miao, K. Tang and V. Gray, Dissolution Technol., 2013, 20(3), 6–13.
20. J. R. Crison, V. P. Shah, J. P. Skelly and G. L. Amidon, J. Pharm. Sci., 1996, 85(9), 1005–1011.
21. M. M. Nerurkar, N. F. Ho, P. S. Burton, T. J. Vidmar and R. T. Borchardt, J. Pharm. Sci., 1997, 86(7), 813–821.
22. N. J. Turro and A. Yekta, J. Am. Chem. Soc., 1978, 100(18), 5951–5952.
23. B. Kloefer, P. V. Hoogevest, R. Moloney, M. Kuentz, M. L. S. Leigh and J. Dressman, Dissolution Technol., 2010, 17(3), 6–13.
24. S. M. Gómez, D. M. Cristancho and F. Martínez, J. Mol. Liq., 2013, 179, 110–117.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra17674c

---