- Email: [email protected]
Decoupling the contribution of dispersive and acid-base components of surface energy on the cohesion of pharmaceutical powders
International Journal of Pharmaceutics 475 (2014) 592–596
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Decoupling the contribution of dispersive and acid-base components of surface energy on the cohesion of pharmaceutical powders Umang V. Shah a , Dolapo Olusanmi b , Ajit S. Narang b , Munir A. Hussain b , Michael J. Tobyn c, Jerry Y.Y. Heng a, * a Surfaces and Particle Engineering Laboratory (SPEL), Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK b Bristol-Myers Squibb Pharmaceuticals, 1 Squibb Drive, New Brunswick,NJ 08903, USA c Bristol-Myers Squibb Pharmaceuticals, Reeds Lane, Moreton, Wirral CH46 1QW, UK
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 July 2014 Received in revised form 9 September 2014 Accepted 11 September 2014 Available online 16 September 2014
This study reports an experimental approach to determine the contribution from two different components of surface energy on cohesion. A method to tailor the surface chemistry of mefenamic acid via silanization is established and the role of surface energy on cohesion is investigated. Silanization was used as a method to functionalize mefenamic acid surfaces with four different functional end groups resulting in an ascending order of the dispersive component of surface energy. Furthermore, four haloalkane functional end groups were grafted on to the surface of mefenamic acid, resulting in varying levels of acid-base component of surface energy, while maintaining constant dispersive component of surface energy. A proportional increase in cohesion was observed with increases in both dispersive as well as acid-base components of surface energy. Contributions from dispersive and acid-base surface energy on cohesion were determined using an iterative approach. Due to the contribution from acid-base surface energy, cohesion was found to increase 11.7 compared to the contribution from dispersive surface energy. Here, we provide an approach to deconvolute the contribution from two different components of surface energy on cohesion, which has the potential of predicting powder flow behavior and ultimately controlling powder cohesion. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Dispersive surface energy Acid-base surface energy Silanization De-coupling Cohesion
1. Introduction Inter-particle interaction is argued to be governed by the material surface properties. Mechanisms for inter-particle interaction can be classified as two broad categories, physical and chemical interactions. Chemical interactions involves mainly covalent, ionic, metallic or electrostatic bonds, whereas physical interactions are a result of intermolecular forces, for example van der Waals and hydrogen bonding (Kendall, 1994). In addition to chemical and physical interactions, mechanical interlocking and diffusion are other two mechanisms widely discussed in the literature (Maeda et al., 2002). In industrial particle processing, instantaneous formation of menisci in capillaries between adhered particles is unavoidable and in such scenarios capillary forces of adhesion and inter-particle contact area becomes increasingly
* Corresponding author. Tel.: +44 207 594 0784; fax: +44 207 594 5700. E-mail addresses: [email protected], http://www.imperial.ac.uk/spel (J.Y.Y. Heng). http://dx.doi.org/10.1016/j.ijpharm.2014.09.018 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
important (Rabinovich et al., 2002). For the purpose of this study, the discussion is focused on different intermolecular forces based on inter-particle interaction mechanisms. Furthermore, the analysis is limited to the surface energetic heterogeneity/ homogeneity not taking into consideration role of any structural or compositional heterogeneity. In the current literature, focusing on the cohesion of pharmaceutical materials, a number of reports have considered the role of surface energy on cohesion and powder flow properties (Barra et al., 1996, 1998; Bhandari and Howes, 2005; Chen et al., 2010; Deng and Davé, 2013; Han et al., 2013; Jallo et al., 2011; Kilbury et al., 2012; Moreno-Atanasio et al., 2005; Spillmann et al., 2008; Traini et al., 2005; Young et al., 2003, 2004). Barra et al. investigated the effect of the surface energy and cohesion parameters proposed by Wu (Wu, 1973) and Rowe (Rowe, 1989a,b) to predict the maximum value of interaction parameters or strength of interaction between particles of binary mixture. Furthermore they also studied the influence of polar and dispersive fractions of two interacting materials on prediction (Barra et al., 1996; Barra et al., 1998). Moreno-Atanasio et al. used distinct
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
element method (DEM) to simulate the effect of surface energy on unconfined yield stress (UYS), revealing that an increase in surface energy by an order of magnitude produced similar increase in simulated UYS (Moreno-Atanasio et al., 2005). Traini et al. used atomic force microscopy as a tool to investigate adhesion–cohesion balance in pressurized metered dose inhalers, demonstrating a linear correlation between theoretical work of cohesion/adhesion calculated from contact angle, inverse phase gas chromatography and atomic force microscopy measurements (Traini et al., 2005). Chen et al. and Jallo et al. used surface modification, either using silanization of aluminum particles or using dry-coating method to coat surface using silica particles, to reduce cohesion. Reduction in cohesion was attributed to the reduction in surface energy; silanization of aluminium was found to result in a reduction of the surface energy, and subsequently measured cohesion values of silanized aluminum were observed to be lower, compared to unsilanized aluminum (Chen et al., 2010; Jallo et al., 2011). On the basis of the findings of Chen et al., Han et al. investigated effect of dry coating on passivating the high energy sites of micronized ibuprofen for improving flowability recently. Surface energy heterogeneity was observed to reduce as a result of dry-coating and the surface energy follows a descending trend with increasing coating resulting in reduction in cohesion (Han et al., 2013). It is apparent from the current literature that surface energy has a major role to play in controlling cohesion. However, whilst recent literature reports have suggested that a higher surface energy may result in higher cohesion and suggested routes to passivate higher surface energy sites, no fundamental understanding on the contribution from surface energy on cohesion compared to other surface attributes have been reported. Recently methodology for de-coupling roles of different surface properties, particularly, particle shape, surface area and surface energy has been established (Shah et al., 2014a,b). Considering that different components of surface energy can contribute towards cohesion on the basis of contribution from intermolecular forces, this study focuses on developing an approach for de-coupling the contribution from dispersive and acid-base component of surface energy on cohesion. 2. Materials Mefenamic acid (2-(2,3-dimethylphenyl) amino benzoic acid) (99.0%), n-heptane (99.0%), n-octane (99.0%), n-nonane (99.0%), n-decane (99.0%), dichlorodimethylsilane (>99.5%,), dodecyl triethoxysilane (technical grade), vinyltrimethoxysilane (>97.0%), triethoxyphenylsilane (>98.0%), (3-iodopropyl)
Milled MA - Un-Silanised Milled MA - Silanised – (-F) group Milled MA-Silanised - (-Cl) group Milled MA - Silanised – (-Br) group Milled MA - Silanised – (-I) group
60 55
45 40 35 30 25 0.01
0.02
0.03 0.04 0.05 0.06 0.07 Fractional Surface Coverage (n/nm) (-)
Fig. 1. g d profiles for milled mefenamic acid silanized with functional end groups.
0.08
F,
Cl,
0.09
Br, and
3. Methods 3.1. Silanization of milled mefenamic acid Milled mefenamic acid powders were silanized using a protocol reported in the literature (Al-Chalabi et al., 1990). In a typical process, 500 mg of mefenamic acid powder was added to a 50 mL 5% (v/v) solution of appropriate silane in cyclohexane. The mixture was refluxed at 80 C for 24 h. Then, the reaction mixture is allowed to cool down to room temperature and filtered using generalpurpose laboratory filter paper (Whatman, UK) followed by drying in a vacuum oven at 80 C for 4 h. Post silanization, the silanized mefenamic acid powders were stored in a glass vial at ambient conditions. 3.2. Surface energy analysis Surface energy analyzer (SEA, Surface Measurement Systems Ltd., London, UK) was used for surface energy heterogeneity characterization. Approximately 300 mg of mefenamic acid was packed in pre-silanized iGC columns (Surface Measurement Systems Ltd., London, UK) and conditioned for 2 h at 30 C followed by pulse injection measurements. Methane was used to determine the column dead time. Helium at a flow rate of 10 sccm was used as a carrier gas for all injections for the columns packed with un-silanized mefenamic acid, whereas 3 sccm helium flow rate was used for columns packed with silanized mefenamic acid. A series of dispersive n-alkane probes (hexane, heptane, octane, nonane and decane) at a range of concentrations were injected in order to achieve target surface coverages (n/nm) ranging from 0.7% to 10%. Net retention volumes were calculated using the commonly applied Schultz method (Schultz et al., 1987). Mono-polar probes (dichloromethane and ethyl acetate) were injected at the same concentrations to determine non-dispersive interactions. The surface energy due to the non-dispersive interactions was calculated using the vOCG method reported in the literature (Das et al., 2010; Van Oss et al., 1988). Principles of the techniques
Milled MA - Un-Silanised Milled MA - Silanised – (-F) group Milled MA - Silanised – (-Cl) group Milled MA - Silanised – (-Br) group Milled MA - Silanised – (-I) group
10
50
0
trimethoxysilane (95.0%), (3-bromopropyl) trimethoxysilane (97.0%) and trimethoxy(3,3,3-trifluoropropyl) silane (97.0%) were purchased from Sigma Aldrich, Dorset, UK. Methanol (>99.5%), ethyl acetate (>99.5%), dichloromethane (>99.0%), n-hexane (>99.0%), and cyclohexane (>99.0%) were received from VWR BDH Prolabo, Lutterworth, UK and (3-chloropropyl) trichlorosilane (>97.0%) was received from Alfa Aesar, Heysham, UK. All chemicals were used as received.
Acid-Base Surface Energy (γAB) (mJ/m2)
Dispersive Surface Energy (γd) (mJ/m2)
65
0.1
I
593
9 8 7 6 5 4 3 2 1 0 0
0.01
0.02
0.03 0.04 0.05 0.06 0.07 Fractional Surface Coverage (n/nm) (-)
Fig. 2. g AB profiles for milled mefenamic acid silanized with functional end groups.
0.08 F,
Cl,
0.09 Br, and
0.1 I
594
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
and a review of currently literature including theory, can be found elsewhere (Ho and Heng, 2013).
0
5
Dispersive Surface Energy (γd)(n/nm=0.02) (mJ/m2) 10 15 20 25 30 35
40
45
35 Total cohesion (due to acid-base and dispersive surface energy) Net Cohesion (due to acid-base surface energy) Net Cohesion (due to dispersive surface energy)
3.3. Uniaxial compression test 30 25 Cohesion (kPa)
A uniaxial compression test was used for powder cohesion measurements. Cylindrical compacts of 5 mm diameter were prepared using an evacuable IR die (Specac Ltd., Slough, UK) at a minimum of three different consolidation loads (10 N, 20 N and 40 N). Post consolidation, confinements were removed and yield load was measured using SMS texture analyzer TA.XT2i (Stable Micro Systems Ltd., Godalming, UK) equipped with a 5 kg load cell in a displacement compression mode, with compression speed of 0.02 mm s 1. Consolidation and yield load values were divided by the contact area to convert into consolidation and yield stress, respectively. Yield stress obtained was plotted as a function of consolidation stress. A linear regression line can be plotted for yield stress as a function of consolidation stress. Linear regression line was extrapolated to find intercept with y-axis showing yield load at zero consolidation load, which is cohesion. Theoretical principles of this test are detailed elsewhere (Head, 1994; Wang, 2013).
15
5
y = 0.23x
0 0
1
2
4.1.1. Contribution of acid-base component of surface energy on cohesion Dispersive (g d) and acid-base (g AB) surface energy heterogeneity profiles for mefenamic acid silanized with four different haloalkane functional end groups, chloropropyl, trifluoropropyl, bromopropyl and iodopropyl are presented in Figs. 1 and 2, respectively. Post silanization, the surface energies (both g d and g AB) of mefenamic acid powders remained constant with increasing fractional surface coverage, suggesting energetic homogeneity. For surface energy measurements to be representative of the entire material surface properties, typical fractional coverages used for analysis ranges from n/nm = 0.02 to 0.05 (Gamble et al., 2012, 2013; Shah et al., 2014a,b). The analysis of energetically homogeneous surfaces remains similar for different fractional surface coverages. Considering the range of fractional surface coverages typically used to provide material representative surface energy, n/nm = 0.02 was selected for analysis of both silanized and unsilanized materials.
Unconfined Yield Stress (kPa)
5
6
7
8
9
haloalkane functional groups were found to have very similar surface energy values at different fractional surface coverages (40.0 0.3 mJ/m2). All haloalkanes selected for this study have a propyl chain attached to the halogen atoms as a spacer and can provide very similar dispersive interactions. The acid-base component of surface energy was found to decrease in the order of Cl > F > Br > I functional groups. For fractional surface coverage n/nm = 0.02, the acid-base components of surface energy for chloropropyl, trifluoropropyl, bromopropyl and iodopropyl are 7.7 mJ/m2, 5.2 mJ/m2, 3.9 mJ/m2, and 2.2 mJ/m2, respectively. The order of decrease in the acid-base surface energy observed in this study can be explained by the functional end group properties, due to the electronegativity of haloalkanes. Unconfined yield stress was measured for powders silanized with haloalkane functional end groups and the data is shown in Fig. 3. Cohesion values were calculated from unconfined yield stress measurements following the method reported by Head, and plotted as a function of acid-base surface energy (Fig. 4) (Head, 1994). Cohesion was found to increase linearly with increasing acid-base surface energy. The dispersive component of the surface energy for the powders silanized with haloalkanes is very similar (40.0 0.3 mJ/m2), hence this increase in total surface
Dispersive Surface Energy (γd) (mJ/m2)
Milled MA - Un-Silanised Milled MA - Silanised – (-F) group Milled MA - Silanised – (-Cl) group Milled MA - Silanised – (-Br) group Milled MA - Silanised – (-I) group
140
4
Fig. 4. Cohesion as a function of dispersive and acid-base surface energy.
65
160
3
Acid-Base Surface Energy (γAB )(n/nm=0.02) (mJ/m2)
g d profiles for mefenamic acid silanized with different
4.1. Isolating the effect of different components of surface energy on cohesion
180
y = 2.64x R² = 0.75
20
10
4. Results and discussion
200
y = 2.64x + 9.03 R² = 0.75
120 100 80 60 40
Milled MA - Un-Silanised Milled MA - Silanised - Methyl group Milled MA - Silanised - Vinyl group Milled MA - Silanised - Phenyl group Milled MA - Silanised - Dodecyl group
60 55 50
45 40 35 30
20 25
0 0
500
1000 1500 Consolidation Stress (kPa)
2000
2500
Fig. 3. Unconfined yield stress as a function of consolidation stress for mefenamic acid silanized F, Cl, Br, and I functional end groups.
0
0.01
0.02
0.03 0.04 0.05 0.06 0.07 Fractional Surface Coverage (n/nm) (-)
0.08
0.09
0.1
Fig. 5. g d profile for milled mefenamic acid silanized with methyl, vinyl, phenyl, and dodecyl functional end groups.
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
energy is solely attributed to the increase in the acid-base component of the surface energy. Considering the linear relationship between cohesion and acid-base of surface energy, the intercept of the best fit line for cohesion as a function of acid-base surface energy (when g AB = 0 mJ/m2), will be attributed to the dispersive surface energy. The cohesion (9.0 kPa) at the intercept of best fit line for cohesion as a function of acid-base surface energy is net cohesion due to the dispersive surface energy at 40.0 mJ/m2 (Fig. 4). By subtracting the cohesion due to the dispersive component (9.0 kPa) from the total cohesion, the contribution of g AB on cohesion can be determined as shown in Fig. 4. 4.1.2. Contribution of dispersive component of surface energy on cohesion g d heterogeneity profiles, before and after silanization of mefenamic acid, are presented in Fig. 5. gd remained constant with increasing surface coverages for mefenamic acid silanized with different functional end groups, whereas the g d for unsilanized milled mefenamic acid was observed to decrease. Therefore, it can be suggested that silanization results in energetically homogenous surfaces. g d for silanized mefenamic acid was observed in the ascending order from methyl, dodecyl, phenyl and vinyl functional end groups. Acid base surface energy for surfaces silanized with vinyl and phenyl functional groups were found to be higher compared to that of surfaces functionalized with methyl and dodecyl functional groups. Variations here could be due to distribution of charge density and dipole moments. Surface energy for mefenamic acid silanized with methyl functional groups was found to vary minimally within the error bars from 32.7 mJ/m2 to 31.6 mJ/m2 with increasing fractional surface coverage from 0.7% to 10%. Dichlorodimethylsilane is the silane used for grafting methyl functional end group on to the surface. This molecule has no spacer and the methyl moiety is directly attached to the terminal end group, such that it provides no molecular flexibility to the functional end group. Therefore the grafted (or deposited) methyl functional end group, which is known to be unreactive, is very stable. Dodecyltriethoxysilane ((OC2H5)3 Si CH2(CH2)10CH3) was used for grafting. Mefenamic acid grafted with dodecyl end group has CH2(CH2)10CH3 functional group attached to Si without any spacer. Dodecyl is a long chain functional end group and results in a relatively higher dispersive surface energy compared to a methyl functional group, i.e. surface energy heterogeneity profile varies from 36.6 mJ/m2 to 34.9 mJ/m2 with increasing fractional surface coverage from 0.7% to 10%. Mefenamic acid silanized with phenyl and vinyl functional
groups resulted in surface energy heterogeneity profiles ranging from 40.2 mJ/m2 to 40.5 mJ/m2, and 42.9 mJ/m2 to 42.8 mJ/m2, respectively for fractional surface coverage ranging from 0.7% to 10%, thus demonstrating homogeneity. Considering the isostere at 2% fractional surface coverage, dispersive component of surface energy was found to be 32.7 mJ/m2 for methyl, 36.3 mJ/m2 for dodecyl, 40.7 mJ/m2 for phenyl and 42.3 mJ/m2 for vinyl silanized surfaces, and 46.4 mJ/m2 for un-silanized surfaces. The acid-base component of surface energy at an isostere of 2% fractional surface coverage was calculated to be 0.4 mJ/m2 for methyl, 0.8 mJ/m2 for dodecyl, 3.0 mJ/m2 for phenyl and 3.0 mJ/m2 for vinyl silanized surfaces (Fig. 6). Uniaxial compression test was used for measurements of unconfined yield stress at three different consolidation stresses for the silanized mefenamic acid and results are presented in Fig. 7. With a decrease in the dispersive component of the surface energy, a decrease in unconfined yield stress was observed for 2040 kPa and 1020 kPa consolidation stress. For 510 kPa consolidation stress unconfined yield stress for surfaces silanized with phenyl and vinyl are within experimental errors, and decrease in unconfined yield stress was observed in the order of the surfaces silanized with phenyl vinyl > dodecyl > methyl. Cohesion values calculated for methyl silanized surface was 10.3 kPa, dodecyl functionalized surface was 12.1 kPa, vinyl functionalized surface was 16.0 kPa and phenyl functionalized surface was 15.4 kPa. A proportional increase in cohesion as a function of dispersive component of the surface energy was observed. To de-couple the contribution of the dispersive surface energy on cohesion from the acid-base component, the correlation developed in Section 4.1.1, was used to calculate net cohesion due to the acid-base component of surface energy. However, the correlation developed between acid-base surface energy and cohesion is only specific to mefenamic acid. The approach reported here can be applied to establish correlations for other systems. The total cohesion calculated is attributed to the total surface energy, which has two components – the acid-base component and the dispersive component. To calculate the cohesion due to dispersive surface energy, the cohesion due to acid-base surface energy calculated previously was subtracted from the total cohesion. A linear regression line was fitted to the net cohesion (due to dispersive surface energy calculated) as a function of dispersive surface energy. As this regression line represents net cohesion due to dispersive surface energy, it was fitted with a zero intercept, suggesting at zero dispersive surface energy, calculated cohesion is also zero. An iterative approach was adopted to converge
160
Milled MA - Un-Silanised Milled MA - Silanised – Methyl group Milled MA - Silanised – Vinyl group Milled MA - Silanised – Phenyl group Milled MA - Silanised – Dodecyl group
4.5 4.0 3.5
Milled MA - Un-Silanised Milled MA - Silanised - Methyl group Milled MA - Silanised - Vinyl group Milled MA - Silanised - Phenyl group Milled MA - Silanised - Dodecyl group
140 Unconfined Yield Stress (kPa)
Acid-Base Surface Energy (γAB) (mJ/m2)
5.0
3.0 2.5 2.0 1.5 1.0
595
120 100 80 60 40 20
0.5 0.0 0
0.01
0.02
0.03 0.04 0.05 0.06 0.07 0.08 Fractional Surface Coverage (n/nm) (-)
0.09
0.1
Fig. 6. g AB profiles for milled mefenamic acid silanized with methyl, vinyl, phenyl, and dodecyl functional end groups.
0 0
500
1000 1500 Consolidation Stress (kPa)
2000
2500
Fig. 7. Unconfined yield stress as a function of consolidation stress for mefenamic acid silanized with methyl, vinyl, phenyl, and dodecyl functional end groups.
596
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
regression lines of net cohesion as a function of acid-base and dispersive surface energy. Using the linear regression for the net cohesion (due to dispersive surface energy) as a function of dispersive energy, cohesion due to the dispersive energy at 40.0 mJ/m2 was calculated and the obtained cohesion was used to set the intercept of linear fit for total cohesion as a function of acid base surface energy. Such iterations were continued until the cohesion value due to dispersive surface energy at 40.0 mJ/m2 calculated using both linear regressions (net cohesion as a function of acid-base and dispersive surface energy), converged (9.03 kPa). Fig. 4 shows the linear regression fits obtained as a result of iterative approach, showing the correlation between net cohesion and dispersive as well as acid-base surface energy. Net cohesion calculated due to the dispersive component of surface energy was found to be 9.2 kPa for surfaces silanized with methyl, 10.0 kPa for surfaces silanized with dodecyl, 7.8 kPa for surface silanized with phenyl and 8.3 kPa for surface silanized with vinyl functional end groups. Considering the approach adopted here to calculate net cohesion due to the dispersive component only, a linear correlation, intersecting at the origin, for cohesion as a function of dispersive surface energy was established and represented by an equation y = 0.226x (i.e., when dispersive surface energy is zero, cohesion is zero as net cohesion is only due to dispersive surface energy). The correlation between dispersive surface energy and cohesion is only specific to mefenamic acid. For the model system investigated, the contribution from g AB was found to result in 11.7 higher cohesion compared to contributions from g d. That is, when contributions from g d was eliminated as a factor in cohesion for material with same surface area, the remaining cohesion can be estimated from g AB. Furthermore, relationships between dispersive as well as acid-base surface energy and net cohesion due to contribution from dispersive as well as acid-base surface energy were established. The role of different components of surface energy on cohesion is system specific and also depends on intrinsic properties of the material. In addition, the contribution from the dispersive and acidbase components of surface energy on cohesion can be different for different materials and also depend on the experimental conditions. The approach presented in this study shows the potential for developing a fundamental understanding of contributions from different surface energy components on cohesion, which will permit controlling cohesion by engineering particle surface properties either via appropriate processing methods or crystal engineering. 5. Conclusion Here, an approach for de-coupling the different components of surface energy has been demonstrated. Silanization was used as a tool to tailor surface energies of mefenamic acid. Methyl, dodecyl, phenyl and vinyl functional groups were grafted on the mefenamic acid surface to investigate role of g d, whereas a series of haloalkane functional groups were grafted to study role of g AB on cohesion. Powder cohesion was found to increase in linear correlation with surface energy. A linear correlation between g AB and total cohesion was developed and used for determining contribution from g d on cohesion. An iterative approach was employed to converge the relationship between net cohesion (due to g d) and dispersive surface energy, and total cohesion and acid-base surface energy. For the model system investigated, contributions from g d and g AB on cohesion were decoupled and a correlation between net cohesion and surface energy (due to the g d and g AB) was established. Increase in cohesion was found to be 11.7 higher due to contribution from g AB compared to that of g d. Findings of this study not only provided fundamental understanding on the
effect of surface energy on cohesion but also can be used for quantification of contributions from different components of surface energy. References Al-Chalabi, S.A.M., Jones, A.R., Luckham, P.F., 1990. A simple method for improving the dispersability of micron-sized solid spheres. J. Aerosol Sci. 21, 821–826. Barra, J., Lescure, F., Doelker, E., 1996. Influence of surface free energies and cohesion parameters on pharmaceutical material interaction parameters–theoretical simulations. Pharm. Res. 13, 1746–1751. Barra, J., Lescure, F., Falson-Rieg, F., Doelker, E., 1998. Can the organization of a binary mix be predicted from the surface energy, cohesion parameter and particle size of its components? Pharm. Res. 15, 1727–1736. Bhandari, B., Howes, T., 2005. Relating the stickiness property of foods undergoing drying and dried products to their surface energetics. Drying Technol. 23, 781–797. Chen, Y., Jallo, L., Quintanilla, M.A.S., Dave, R., 2010. Characterization of particle and bulk level cohesion reduction of surface modified fine aluminum powders. Colloids Surf. A: Asp. 361, 66–80. Das, S.C., Larson, I., Morton, D.A.V., Stewart, P.J., 2010. Determination of the polar and total surface energy distributions of particulates by inverse gas chromatography. Langmuir 27, 521–523. Deng, X., Davé, R., 2013. Dynamic simulation of particle packing influenced by size, aspect ratio and surface energy. Granul. Matter 15, 401–415. Gamble, J.F., Davé, R.N., Kiang, S., Leane, M.M., Tobyn, M., Wang, S.S.Y., 2013. Investigating the applicability of inverse gas chromatography to binary powdered systems: an application of surface heterogeneity profiles to understanding preferential probe-surface interactions. Int. J. Pharm. 445, 39–46. Gamble, J.F., Leane, M., Olusanmi, D., Tobyn, M., Šupuk, E., Khoo, J., Naderi, M., 2012. Surface energy analysis as a tool to probe the surface energy characteristics of micronized materials—a comparison with inverse gas chromatography. Int. J. Pharm. 422, 238–244. Han, X., Jallo, L., To, D., Ghoroi, C., Davé, R., 2013. Passivation of high-surface-energy sites of milled ibuprofen crystals via dry coating for reduced cohesion and improved flowability. J. Pharm. Sci. 102, 2282–2296. Head, K.H., 1994. Manual of Soil Laboratory Testing, second ed. Pantech Press, New York. Ho, R., Heng, J.Y.Y., 2013. A review of inverse gas chromatography and its development as a tool to characterize anisotropic surface properties of pharmaceutical solids. Kona Powder Part. J. 30, 164–180. Jallo, L.J., Chen, Y., Bowen, J., Etzler, F., Dave, R., 2011. Prediction of inter-particle adhesion force from surface energy and surface roughness. J. Adhes. Sci. Technol. 25, 367–384. Kendall, K., 1994. Adhesion: molecules and mechanics. Science 263, 1720–1725. Kilbury, O.J., Barrett, K.S., Fu, X., Yin, J., Dinair, D.S., Gump, C.J., Weimer, A.W., King, D. M., 2012. Atomic layer deposition of solid lubricating coatings on particles. Powder Technol. 221, 26–35. Maeda, N., Chen, N., Tirrell, M., Israelachvili, J.N., 2002. Adhesion and friction mechanisms of polymer-on-polymer surfaces. Science 297, 379–382. Moreno-Atanasio, R., Antony, S.J., Ghadiri, M., 2005. Analysis of flowability of cohesive powders using distinct element method. Powder Technol. 158, 51–57. Rabinovich, Y.I., Adler, J.J., Esayanur, M.S., Ata, A., Singh, R.K., Moudgil, B.M., 2002. Capillary forces between surfaces with nanoscale roughness. Adv. Colloid Interface Sci. 96, 213–230. Rowe, R.C., 1989a. Binder-substrate interactions in granulation: a theoretical approach based on surface free energy and polarity. Int. J. Pharm. 52, 149–154. Rowe, R.C., 1989b. Polar/non-polar interactions in the granulation of organic substrates with polymer binding agents. Int. J. Pharm. 56, 117–124. Schultz, J., Lavielle, L., Martin, C., 1987. The role of the interface in carbon fibre–epoxy composites. J. Adhes. 23, 45–60. Shah, U.V., Olusanmi, D., Narang, A., Hussain, M., Tobyn, M., Hinder, S., Heng, J.Y.Y., 2014a. Decoupling the contribution of surface energy and surface area on the cohesion of pharmaceutical powders. Pharm. Res doi:http://dx.doi.org/10.1007/ s11095-014-1459-3 (in press). Shah, U.V., Olusanmi, D., Narang, A.S., Hussain, M.A., Gamble, J.F., Tobyn, M.J., Heng, J. Y.Y., 2014b. Effect of crystal habits on the surface energy and cohesion of crystalline powders. Int. J. Pharm. 472, 140–147. Spillmann, A., Sonnenfeld, A., Rudolf von Rohr, P., 2008. Effect of surface free energy on the flowability of lactose powder treated by PECVD. Plasma Process. Polym. 5, 753–758. Traini, D., Rogueda, P., Young, P., Price, R., 2005. Surface energy and interparticle force correlation in model pMDI formulations. Pharm. Res. 22, 816–825. Van Oss, C.J., Chaudhury, M.K., Good, R.J., 1988. Interfacial Lifshitz-van der Waals and polar interactions in macroscopic systems. Chem. Rev. 88, 927–941. Wang, D., 2013. Advanced Physical Characterisation of Milled Pharmaceutical Solids. Imperial College London, London PhD Thesis. Wu, S., 1973. Polar and nonpolar interactions in adhesion. J. Adhes. 5, 39–55. Young, P.M., Price, R., Tobyn, M.J., Buttrum, M., Dey, F., 2003. Investigation into the effect of humidity on drug–drug interactions using the atomic force microscope. J. Pharm. Sci. 92, 815–822. Young, P.M., Price, R., Tobyn, M.J., Buttrum, M., Dey, F., 2004. The influence of relative humidity on the cohesion properties of micronized drugs used in inhalation therapy. J. Pharm. Sci. 93, 753–761.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Decoupling the contribution of dispersive and acid-base components of surface energy on the cohesion of pharmaceutical powders Umang V. Shah a , Dolapo Olusanmi b , Ajit S. Narang b , Munir A. Hussain b , Michael J. Tobyn c, Jerry Y.Y. Heng a, * a Surfaces and Particle Engineering Laboratory (SPEL), Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK b Bristol-Myers Squibb Pharmaceuticals, 1 Squibb Drive, New Brunswick,NJ 08903, USA c Bristol-Myers Squibb Pharmaceuticals, Reeds Lane, Moreton, Wirral CH46 1QW, UK
A R T I C L E I N F O
A B S T R A C T
Article history: Received 26 July 2014 Received in revised form 9 September 2014 Accepted 11 September 2014 Available online 16 September 2014
This study reports an experimental approach to determine the contribution from two different components of surface energy on cohesion. A method to tailor the surface chemistry of mefenamic acid via silanization is established and the role of surface energy on cohesion is investigated. Silanization was used as a method to functionalize mefenamic acid surfaces with four different functional end groups resulting in an ascending order of the dispersive component of surface energy. Furthermore, four haloalkane functional end groups were grafted on to the surface of mefenamic acid, resulting in varying levels of acid-base component of surface energy, while maintaining constant dispersive component of surface energy. A proportional increase in cohesion was observed with increases in both dispersive as well as acid-base components of surface energy. Contributions from dispersive and acid-base surface energy on cohesion were determined using an iterative approach. Due to the contribution from acid-base surface energy, cohesion was found to increase 11.7 compared to the contribution from dispersive surface energy. Here, we provide an approach to deconvolute the contribution from two different components of surface energy on cohesion, which has the potential of predicting powder flow behavior and ultimately controlling powder cohesion. ã 2014 Elsevier B.V. All rights reserved.
Keywords: Dispersive surface energy Acid-base surface energy Silanization De-coupling Cohesion
1. Introduction Inter-particle interaction is argued to be governed by the material surface properties. Mechanisms for inter-particle interaction can be classified as two broad categories, physical and chemical interactions. Chemical interactions involves mainly covalent, ionic, metallic or electrostatic bonds, whereas physical interactions are a result of intermolecular forces, for example van der Waals and hydrogen bonding (Kendall, 1994). In addition to chemical and physical interactions, mechanical interlocking and diffusion are other two mechanisms widely discussed in the literature (Maeda et al., 2002). In industrial particle processing, instantaneous formation of menisci in capillaries between adhered particles is unavoidable and in such scenarios capillary forces of adhesion and inter-particle contact area becomes increasingly
* Corresponding author. Tel.: +44 207 594 0784; fax: +44 207 594 5700. E-mail addresses: [email protected], http://www.imperial.ac.uk/spel (J.Y.Y. Heng). http://dx.doi.org/10.1016/j.ijpharm.2014.09.018 0378-5173/ ã 2014 Elsevier B.V. All rights reserved.
important (Rabinovich et al., 2002). For the purpose of this study, the discussion is focused on different intermolecular forces based on inter-particle interaction mechanisms. Furthermore, the analysis is limited to the surface energetic heterogeneity/ homogeneity not taking into consideration role of any structural or compositional heterogeneity. In the current literature, focusing on the cohesion of pharmaceutical materials, a number of reports have considered the role of surface energy on cohesion and powder flow properties (Barra et al., 1996, 1998; Bhandari and Howes, 2005; Chen et al., 2010; Deng and Davé, 2013; Han et al., 2013; Jallo et al., 2011; Kilbury et al., 2012; Moreno-Atanasio et al., 2005; Spillmann et al., 2008; Traini et al., 2005; Young et al., 2003, 2004). Barra et al. investigated the effect of the surface energy and cohesion parameters proposed by Wu (Wu, 1973) and Rowe (Rowe, 1989a,b) to predict the maximum value of interaction parameters or strength of interaction between particles of binary mixture. Furthermore they also studied the influence of polar and dispersive fractions of two interacting materials on prediction (Barra et al., 1996; Barra et al., 1998). Moreno-Atanasio et al. used distinct
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
element method (DEM) to simulate the effect of surface energy on unconfined yield stress (UYS), revealing that an increase in surface energy by an order of magnitude produced similar increase in simulated UYS (Moreno-Atanasio et al., 2005). Traini et al. used atomic force microscopy as a tool to investigate adhesion–cohesion balance in pressurized metered dose inhalers, demonstrating a linear correlation between theoretical work of cohesion/adhesion calculated from contact angle, inverse phase gas chromatography and atomic force microscopy measurements (Traini et al., 2005). Chen et al. and Jallo et al. used surface modification, either using silanization of aluminum particles or using dry-coating method to coat surface using silica particles, to reduce cohesion. Reduction in cohesion was attributed to the reduction in surface energy; silanization of aluminium was found to result in a reduction of the surface energy, and subsequently measured cohesion values of silanized aluminum were observed to be lower, compared to unsilanized aluminum (Chen et al., 2010; Jallo et al., 2011). On the basis of the findings of Chen et al., Han et al. investigated effect of dry coating on passivating the high energy sites of micronized ibuprofen for improving flowability recently. Surface energy heterogeneity was observed to reduce as a result of dry-coating and the surface energy follows a descending trend with increasing coating resulting in reduction in cohesion (Han et al., 2013). It is apparent from the current literature that surface energy has a major role to play in controlling cohesion. However, whilst recent literature reports have suggested that a higher surface energy may result in higher cohesion and suggested routes to passivate higher surface energy sites, no fundamental understanding on the contribution from surface energy on cohesion compared to other surface attributes have been reported. Recently methodology for de-coupling roles of different surface properties, particularly, particle shape, surface area and surface energy has been established (Shah et al., 2014a,b). Considering that different components of surface energy can contribute towards cohesion on the basis of contribution from intermolecular forces, this study focuses on developing an approach for de-coupling the contribution from dispersive and acid-base component of surface energy on cohesion. 2. Materials Mefenamic acid (2-(2,3-dimethylphenyl) amino benzoic acid) (99.0%), n-heptane (99.0%), n-octane (99.0%), n-nonane (99.0%), n-decane (99.0%), dichlorodimethylsilane (>99.5%,), dodecyl triethoxysilane (technical grade), vinyltrimethoxysilane (>97.0%), triethoxyphenylsilane (>98.0%), (3-iodopropyl)
Milled MA - Un-Silanised Milled MA - Silanised – (-F) group Milled MA-Silanised - (-Cl) group Milled MA - Silanised – (-Br) group Milled MA - Silanised – (-I) group
60 55
45 40 35 30 25 0.01
0.02
0.03 0.04 0.05 0.06 0.07 Fractional Surface Coverage (n/nm) (-)
Fig. 1. g d profiles for milled mefenamic acid silanized with functional end groups.
0.08
F,
Cl,
0.09
Br, and
3. Methods 3.1. Silanization of milled mefenamic acid Milled mefenamic acid powders were silanized using a protocol reported in the literature (Al-Chalabi et al., 1990). In a typical process, 500 mg of mefenamic acid powder was added to a 50 mL 5% (v/v) solution of appropriate silane in cyclohexane. The mixture was refluxed at 80 C for 24 h. Then, the reaction mixture is allowed to cool down to room temperature and filtered using generalpurpose laboratory filter paper (Whatman, UK) followed by drying in a vacuum oven at 80 C for 4 h. Post silanization, the silanized mefenamic acid powders were stored in a glass vial at ambient conditions. 3.2. Surface energy analysis Surface energy analyzer (SEA, Surface Measurement Systems Ltd., London, UK) was used for surface energy heterogeneity characterization. Approximately 300 mg of mefenamic acid was packed in pre-silanized iGC columns (Surface Measurement Systems Ltd., London, UK) and conditioned for 2 h at 30 C followed by pulse injection measurements. Methane was used to determine the column dead time. Helium at a flow rate of 10 sccm was used as a carrier gas for all injections for the columns packed with un-silanized mefenamic acid, whereas 3 sccm helium flow rate was used for columns packed with silanized mefenamic acid. A series of dispersive n-alkane probes (hexane, heptane, octane, nonane and decane) at a range of concentrations were injected in order to achieve target surface coverages (n/nm) ranging from 0.7% to 10%. Net retention volumes were calculated using the commonly applied Schultz method (Schultz et al., 1987). Mono-polar probes (dichloromethane and ethyl acetate) were injected at the same concentrations to determine non-dispersive interactions. The surface energy due to the non-dispersive interactions was calculated using the vOCG method reported in the literature (Das et al., 2010; Van Oss et al., 1988). Principles of the techniques
Milled MA - Un-Silanised Milled MA - Silanised – (-F) group Milled MA - Silanised – (-Cl) group Milled MA - Silanised – (-Br) group Milled MA - Silanised – (-I) group
10
50
0
trimethoxysilane (95.0%), (3-bromopropyl) trimethoxysilane (97.0%) and trimethoxy(3,3,3-trifluoropropyl) silane (97.0%) were purchased from Sigma Aldrich, Dorset, UK. Methanol (>99.5%), ethyl acetate (>99.5%), dichloromethane (>99.0%), n-hexane (>99.0%), and cyclohexane (>99.0%) were received from VWR BDH Prolabo, Lutterworth, UK and (3-chloropropyl) trichlorosilane (>97.0%) was received from Alfa Aesar, Heysham, UK. All chemicals were used as received.
Acid-Base Surface Energy (γAB) (mJ/m2)
Dispersive Surface Energy (γd) (mJ/m2)
65
0.1
I
593
9 8 7 6 5 4 3 2 1 0 0
0.01
0.02
0.03 0.04 0.05 0.06 0.07 Fractional Surface Coverage (n/nm) (-)
Fig. 2. g AB profiles for milled mefenamic acid silanized with functional end groups.
0.08 F,
Cl,
0.09 Br, and
0.1 I
594
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
and a review of currently literature including theory, can be found elsewhere (Ho and Heng, 2013).
0
5
Dispersive Surface Energy (γd)(n/nm=0.02) (mJ/m2) 10 15 20 25 30 35
40
45
35 Total cohesion (due to acid-base and dispersive surface energy) Net Cohesion (due to acid-base surface energy) Net Cohesion (due to dispersive surface energy)
3.3. Uniaxial compression test 30 25 Cohesion (kPa)
A uniaxial compression test was used for powder cohesion measurements. Cylindrical compacts of 5 mm diameter were prepared using an evacuable IR die (Specac Ltd., Slough, UK) at a minimum of three different consolidation loads (10 N, 20 N and 40 N). Post consolidation, confinements were removed and yield load was measured using SMS texture analyzer TA.XT2i (Stable Micro Systems Ltd., Godalming, UK) equipped with a 5 kg load cell in a displacement compression mode, with compression speed of 0.02 mm s 1. Consolidation and yield load values were divided by the contact area to convert into consolidation and yield stress, respectively. Yield stress obtained was plotted as a function of consolidation stress. A linear regression line can be plotted for yield stress as a function of consolidation stress. Linear regression line was extrapolated to find intercept with y-axis showing yield load at zero consolidation load, which is cohesion. Theoretical principles of this test are detailed elsewhere (Head, 1994; Wang, 2013).
15
5
y = 0.23x
0 0
1
2
4.1.1. Contribution of acid-base component of surface energy on cohesion Dispersive (g d) and acid-base (g AB) surface energy heterogeneity profiles for mefenamic acid silanized with four different haloalkane functional end groups, chloropropyl, trifluoropropyl, bromopropyl and iodopropyl are presented in Figs. 1 and 2, respectively. Post silanization, the surface energies (both g d and g AB) of mefenamic acid powders remained constant with increasing fractional surface coverage, suggesting energetic homogeneity. For surface energy measurements to be representative of the entire material surface properties, typical fractional coverages used for analysis ranges from n/nm = 0.02 to 0.05 (Gamble et al., 2012, 2013; Shah et al., 2014a,b). The analysis of energetically homogeneous surfaces remains similar for different fractional surface coverages. Considering the range of fractional surface coverages typically used to provide material representative surface energy, n/nm = 0.02 was selected for analysis of both silanized and unsilanized materials.
Unconfined Yield Stress (kPa)
5
6
7
8
9
haloalkane functional groups were found to have very similar surface energy values at different fractional surface coverages (40.0 0.3 mJ/m2). All haloalkanes selected for this study have a propyl chain attached to the halogen atoms as a spacer and can provide very similar dispersive interactions. The acid-base component of surface energy was found to decrease in the order of Cl > F > Br > I functional groups. For fractional surface coverage n/nm = 0.02, the acid-base components of surface energy for chloropropyl, trifluoropropyl, bromopropyl and iodopropyl are 7.7 mJ/m2, 5.2 mJ/m2, 3.9 mJ/m2, and 2.2 mJ/m2, respectively. The order of decrease in the acid-base surface energy observed in this study can be explained by the functional end group properties, due to the electronegativity of haloalkanes. Unconfined yield stress was measured for powders silanized with haloalkane functional end groups and the data is shown in Fig. 3. Cohesion values were calculated from unconfined yield stress measurements following the method reported by Head, and plotted as a function of acid-base surface energy (Fig. 4) (Head, 1994). Cohesion was found to increase linearly with increasing acid-base surface energy. The dispersive component of the surface energy for the powders silanized with haloalkanes is very similar (40.0 0.3 mJ/m2), hence this increase in total surface
Dispersive Surface Energy (γd) (mJ/m2)
Milled MA - Un-Silanised Milled MA - Silanised – (-F) group Milled MA - Silanised – (-Cl) group Milled MA - Silanised – (-Br) group Milled MA - Silanised – (-I) group
140
4
Fig. 4. Cohesion as a function of dispersive and acid-base surface energy.
65
160
3
Acid-Base Surface Energy (γAB )(n/nm=0.02) (mJ/m2)
g d profiles for mefenamic acid silanized with different
4.1. Isolating the effect of different components of surface energy on cohesion
180
y = 2.64x R² = 0.75
20
10
4. Results and discussion
200
y = 2.64x + 9.03 R² = 0.75
120 100 80 60 40
Milled MA - Un-Silanised Milled MA - Silanised - Methyl group Milled MA - Silanised - Vinyl group Milled MA - Silanised - Phenyl group Milled MA - Silanised - Dodecyl group
60 55 50
45 40 35 30
20 25
0 0
500
1000 1500 Consolidation Stress (kPa)
2000
2500
Fig. 3. Unconfined yield stress as a function of consolidation stress for mefenamic acid silanized F, Cl, Br, and I functional end groups.
0
0.01
0.02
0.03 0.04 0.05 0.06 0.07 Fractional Surface Coverage (n/nm) (-)
0.08
0.09
0.1
Fig. 5. g d profile for milled mefenamic acid silanized with methyl, vinyl, phenyl, and dodecyl functional end groups.
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
energy is solely attributed to the increase in the acid-base component of the surface energy. Considering the linear relationship between cohesion and acid-base of surface energy, the intercept of the best fit line for cohesion as a function of acid-base surface energy (when g AB = 0 mJ/m2), will be attributed to the dispersive surface energy. The cohesion (9.0 kPa) at the intercept of best fit line for cohesion as a function of acid-base surface energy is net cohesion due to the dispersive surface energy at 40.0 mJ/m2 (Fig. 4). By subtracting the cohesion due to the dispersive component (9.0 kPa) from the total cohesion, the contribution of g AB on cohesion can be determined as shown in Fig. 4. 4.1.2. Contribution of dispersive component of surface energy on cohesion g d heterogeneity profiles, before and after silanization of mefenamic acid, are presented in Fig. 5. gd remained constant with increasing surface coverages for mefenamic acid silanized with different functional end groups, whereas the g d for unsilanized milled mefenamic acid was observed to decrease. Therefore, it can be suggested that silanization results in energetically homogenous surfaces. g d for silanized mefenamic acid was observed in the ascending order from methyl, dodecyl, phenyl and vinyl functional end groups. Acid base surface energy for surfaces silanized with vinyl and phenyl functional groups were found to be higher compared to that of surfaces functionalized with methyl and dodecyl functional groups. Variations here could be due to distribution of charge density and dipole moments. Surface energy for mefenamic acid silanized with methyl functional groups was found to vary minimally within the error bars from 32.7 mJ/m2 to 31.6 mJ/m2 with increasing fractional surface coverage from 0.7% to 10%. Dichlorodimethylsilane is the silane used for grafting methyl functional end group on to the surface. This molecule has no spacer and the methyl moiety is directly attached to the terminal end group, such that it provides no molecular flexibility to the functional end group. Therefore the grafted (or deposited) methyl functional end group, which is known to be unreactive, is very stable. Dodecyltriethoxysilane ((OC2H5)3 Si CH2(CH2)10CH3) was used for grafting. Mefenamic acid grafted with dodecyl end group has CH2(CH2)10CH3 functional group attached to Si without any spacer. Dodecyl is a long chain functional end group and results in a relatively higher dispersive surface energy compared to a methyl functional group, i.e. surface energy heterogeneity profile varies from 36.6 mJ/m2 to 34.9 mJ/m2 with increasing fractional surface coverage from 0.7% to 10%. Mefenamic acid silanized with phenyl and vinyl functional
groups resulted in surface energy heterogeneity profiles ranging from 40.2 mJ/m2 to 40.5 mJ/m2, and 42.9 mJ/m2 to 42.8 mJ/m2, respectively for fractional surface coverage ranging from 0.7% to 10%, thus demonstrating homogeneity. Considering the isostere at 2% fractional surface coverage, dispersive component of surface energy was found to be 32.7 mJ/m2 for methyl, 36.3 mJ/m2 for dodecyl, 40.7 mJ/m2 for phenyl and 42.3 mJ/m2 for vinyl silanized surfaces, and 46.4 mJ/m2 for un-silanized surfaces. The acid-base component of surface energy at an isostere of 2% fractional surface coverage was calculated to be 0.4 mJ/m2 for methyl, 0.8 mJ/m2 for dodecyl, 3.0 mJ/m2 for phenyl and 3.0 mJ/m2 for vinyl silanized surfaces (Fig. 6). Uniaxial compression test was used for measurements of unconfined yield stress at three different consolidation stresses for the silanized mefenamic acid and results are presented in Fig. 7. With a decrease in the dispersive component of the surface energy, a decrease in unconfined yield stress was observed for 2040 kPa and 1020 kPa consolidation stress. For 510 kPa consolidation stress unconfined yield stress for surfaces silanized with phenyl and vinyl are within experimental errors, and decrease in unconfined yield stress was observed in the order of the surfaces silanized with phenyl vinyl > dodecyl > methyl. Cohesion values calculated for methyl silanized surface was 10.3 kPa, dodecyl functionalized surface was 12.1 kPa, vinyl functionalized surface was 16.0 kPa and phenyl functionalized surface was 15.4 kPa. A proportional increase in cohesion as a function of dispersive component of the surface energy was observed. To de-couple the contribution of the dispersive surface energy on cohesion from the acid-base component, the correlation developed in Section 4.1.1, was used to calculate net cohesion due to the acid-base component of surface energy. However, the correlation developed between acid-base surface energy and cohesion is only specific to mefenamic acid. The approach reported here can be applied to establish correlations for other systems. The total cohesion calculated is attributed to the total surface energy, which has two components – the acid-base component and the dispersive component. To calculate the cohesion due to dispersive surface energy, the cohesion due to acid-base surface energy calculated previously was subtracted from the total cohesion. A linear regression line was fitted to the net cohesion (due to dispersive surface energy calculated) as a function of dispersive surface energy. As this regression line represents net cohesion due to dispersive surface energy, it was fitted with a zero intercept, suggesting at zero dispersive surface energy, calculated cohesion is also zero. An iterative approach was adopted to converge
160
Milled MA - Un-Silanised Milled MA - Silanised – Methyl group Milled MA - Silanised – Vinyl group Milled MA - Silanised – Phenyl group Milled MA - Silanised – Dodecyl group
4.5 4.0 3.5
Milled MA - Un-Silanised Milled MA - Silanised - Methyl group Milled MA - Silanised - Vinyl group Milled MA - Silanised - Phenyl group Milled MA - Silanised - Dodecyl group
140 Unconfined Yield Stress (kPa)
Acid-Base Surface Energy (γAB) (mJ/m2)
5.0
3.0 2.5 2.0 1.5 1.0
595
120 100 80 60 40 20
0.5 0.0 0
0.01
0.02
0.03 0.04 0.05 0.06 0.07 0.08 Fractional Surface Coverage (n/nm) (-)
0.09
0.1
Fig. 6. g AB profiles for milled mefenamic acid silanized with methyl, vinyl, phenyl, and dodecyl functional end groups.
0 0
500
1000 1500 Consolidation Stress (kPa)
2000
2500
Fig. 7. Unconfined yield stress as a function of consolidation stress for mefenamic acid silanized with methyl, vinyl, phenyl, and dodecyl functional end groups.
596
U.V. Shah et al. / International Journal of Pharmaceutics 475 (2014) 592–596
regression lines of net cohesion as a function of acid-base and dispersive surface energy. Using the linear regression for the net cohesion (due to dispersive surface energy) as a function of dispersive energy, cohesion due to the dispersive energy at 40.0 mJ/m2 was calculated and the obtained cohesion was used to set the intercept of linear fit for total cohesion as a function of acid base surface energy. Such iterations were continued until the cohesion value due to dispersive surface energy at 40.0 mJ/m2 calculated using both linear regressions (net cohesion as a function of acid-base and dispersive surface energy), converged (9.03 kPa). Fig. 4 shows the linear regression fits obtained as a result of iterative approach, showing the correlation between net cohesion and dispersive as well as acid-base surface energy. Net cohesion calculated due to the dispersive component of surface energy was found to be 9.2 kPa for surfaces silanized with methyl, 10.0 kPa for surfaces silanized with dodecyl, 7.8 kPa for surface silanized with phenyl and 8.3 kPa for surface silanized with vinyl functional end groups. Considering the approach adopted here to calculate net cohesion due to the dispersive component only, a linear correlation, intersecting at the origin, for cohesion as a function of dispersive surface energy was established and represented by an equation y = 0.226x (i.e., when dispersive surface energy is zero, cohesion is zero as net cohesion is only due to dispersive surface energy). The correlation between dispersive surface energy and cohesion is only specific to mefenamic acid. For the model system investigated, the contribution from g AB was found to result in 11.7 higher cohesion compared to contributions from g d. That is, when contributions from g d was eliminated as a factor in cohesion for material with same surface area, the remaining cohesion can be estimated from g AB. Furthermore, relationships between dispersive as well as acid-base surface energy and net cohesion due to contribution from dispersive as well as acid-base surface energy were established. The role of different components of surface energy on cohesion is system specific and also depends on intrinsic properties of the material. In addition, the contribution from the dispersive and acidbase components of surface energy on cohesion can be different for different materials and also depend on the experimental conditions. The approach presented in this study shows the potential for developing a fundamental understanding of contributions from different surface energy components on cohesion, which will permit controlling cohesion by engineering particle surface properties either via appropriate processing methods or crystal engineering. 5. Conclusion Here, an approach for de-coupling the different components of surface energy has been demonstrated. Silanization was used as a tool to tailor surface energies of mefenamic acid. Methyl, dodecyl, phenyl and vinyl functional groups were grafted on the mefenamic acid surface to investigate role of g d, whereas a series of haloalkane functional groups were grafted to study role of g AB on cohesion. Powder cohesion was found to increase in linear correlation with surface energy. A linear correlation between g AB and total cohesion was developed and used for determining contribution from g d on cohesion. An iterative approach was employed to converge the relationship between net cohesion (due to g d) and dispersive surface energy, and total cohesion and acid-base surface energy. For the model system investigated, contributions from g d and g AB on cohesion were decoupled and a correlation between net cohesion and surface energy (due to the g d and g AB) was established. Increase in cohesion was found to be 11.7 higher due to contribution from g AB compared to that of g d. Findings of this study not only provided fundamental understanding on the
effect of surface energy on cohesion but also can be used for quantification of contributions from different components of surface energy. References Al-Chalabi, S.A.M., Jones, A.R., Luckham, P.F., 1990. A simple method for improving the dispersability of micron-sized solid spheres. J. Aerosol Sci. 21, 821–826. Barra, J., Lescure, F., Doelker, E., 1996. Influence of surface free energies and cohesion parameters on pharmaceutical material interaction parameters–theoretical simulations. Pharm. Res. 13, 1746–1751. Barra, J., Lescure, F., Falson-Rieg, F., Doelker, E., 1998. Can the organization of a binary mix be predicted from the surface energy, cohesion parameter and particle size of its components? Pharm. Res. 15, 1727–1736. Bhandari, B., Howes, T., 2005. Relating the stickiness property of foods undergoing drying and dried products to their surface energetics. Drying Technol. 23, 781–797. Chen, Y., Jallo, L., Quintanilla, M.A.S., Dave, R., 2010. Characterization of particle and bulk level cohesion reduction of surface modified fine aluminum powders. Colloids Surf. A: Asp. 361, 66–80. Das, S.C., Larson, I., Morton, D.A.V., Stewart, P.J., 2010. Determination of the polar and total surface energy distributions of particulates by inverse gas chromatography. Langmuir 27, 521–523. Deng, X., Davé, R., 2013. Dynamic simulation of particle packing influenced by size, aspect ratio and surface energy. Granul. Matter 15, 401–415. Gamble, J.F., Davé, R.N., Kiang, S., Leane, M.M., Tobyn, M., Wang, S.S.Y., 2013. Investigating the applicability of inverse gas chromatography to binary powdered systems: an application of surface heterogeneity profiles to understanding preferential probe-surface interactions. Int. J. Pharm. 445, 39–46. Gamble, J.F., Leane, M., Olusanmi, D., Tobyn, M., Šupuk, E., Khoo, J., Naderi, M., 2012. Surface energy analysis as a tool to probe the surface energy characteristics of micronized materials—a comparison with inverse gas chromatography. Int. J. Pharm. 422, 238–244. Han, X., Jallo, L., To, D., Ghoroi, C., Davé, R., 2013. Passivation of high-surface-energy sites of milled ibuprofen crystals via dry coating for reduced cohesion and improved flowability. J. Pharm. Sci. 102, 2282–2296. Head, K.H., 1994. Manual of Soil Laboratory Testing, second ed. Pantech Press, New York. Ho, R., Heng, J.Y.Y., 2013. A review of inverse gas chromatography and its development as a tool to characterize anisotropic surface properties of pharmaceutical solids. Kona Powder Part. J. 30, 164–180. Jallo, L.J., Chen, Y., Bowen, J., Etzler, F., Dave, R., 2011. Prediction of inter-particle adhesion force from surface energy and surface roughness. J. Adhes. Sci. Technol. 25, 367–384. Kendall, K., 1994. Adhesion: molecules and mechanics. Science 263, 1720–1725. Kilbury, O.J., Barrett, K.S., Fu, X., Yin, J., Dinair, D.S., Gump, C.J., Weimer, A.W., King, D. M., 2012. Atomic layer deposition of solid lubricating coatings on particles. Powder Technol. 221, 26–35. Maeda, N., Chen, N., Tirrell, M., Israelachvili, J.N., 2002. Adhesion and friction mechanisms of polymer-on-polymer surfaces. Science 297, 379–382. Moreno-Atanasio, R., Antony, S.J., Ghadiri, M., 2005. Analysis of flowability of cohesive powders using distinct element method. Powder Technol. 158, 51–57. Rabinovich, Y.I., Adler, J.J., Esayanur, M.S., Ata, A., Singh, R.K., Moudgil, B.M., 2002. Capillary forces between surfaces with nanoscale roughness. Adv. Colloid Interface Sci. 96, 213–230. Rowe, R.C., 1989a. Binder-substrate interactions in granulation: a theoretical approach based on surface free energy and polarity. Int. J. Pharm. 52, 149–154. Rowe, R.C., 1989b. Polar/non-polar interactions in the granulation of organic substrates with polymer binding agents. Int. J. Pharm. 56, 117–124. Schultz, J., Lavielle, L., Martin, C., 1987. The role of the interface in carbon fibre–epoxy composites. J. Adhes. 23, 45–60. Shah, U.V., Olusanmi, D., Narang, A., Hussain, M., Tobyn, M., Hinder, S., Heng, J.Y.Y., 2014a. Decoupling the contribution of surface energy and surface area on the cohesion of pharmaceutical powders. Pharm. Res doi:http://dx.doi.org/10.1007/ s11095-014-1459-3 (in press). Shah, U.V., Olusanmi, D., Narang, A.S., Hussain, M.A., Gamble, J.F., Tobyn, M.J., Heng, J. Y.Y., 2014b. Effect of crystal habits on the surface energy and cohesion of crystalline powders. Int. J. Pharm. 472, 140–147. Spillmann, A., Sonnenfeld, A., Rudolf von Rohr, P., 2008. Effect of surface free energy on the flowability of lactose powder treated by PECVD. Plasma Process. Polym. 5, 753–758. Traini, D., Rogueda, P., Young, P., Price, R., 2005. Surface energy and interparticle force correlation in model pMDI formulations. Pharm. Res. 22, 816–825. Van Oss, C.J., Chaudhury, M.K., Good, R.J., 1988. Interfacial Lifshitz-van der Waals and polar interactions in macroscopic systems. Chem. Rev. 88, 927–941. Wang, D., 2013. Advanced Physical Characterisation of Milled Pharmaceutical Solids. Imperial College London, London PhD Thesis. Wu, S., 1973. Polar and nonpolar interactions in adhesion. J. Adhes. 5, 39–55. Young, P.M., Price, R., Tobyn, M.J., Buttrum, M., Dey, F., 2003. Investigation into the effect of humidity on drug–drug interactions using the atomic force microscope. J. Pharm. Sci. 92, 815–822. Young, P.M., Price, R., Tobyn, M.J., Buttrum, M., Dey, F., 2004. The influence of relative humidity on the cohesion properties of micronized drugs used in inhalation therapy. J. Pharm. Sci. 93, 753–761.