European Journal of Pharmaceutical Sciences 37 (2009) 370–377

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences

journa l homepage: www.e lsev ier .com/ locate /e jps

Engineering novel topical foams using hydrofluroalkane emulsions stabilisedwith pluronic surfactants

Yanjun Zhaoa, Marc B. Brownb,c, Stuart A. Jonesa,∗

a Pharmaceutical Science Division, King’s College London, 150 Stamford Street, London, SE1 9NH, UKb MedPharm Ltd., Unit 3/Chancellor Court, 50 Occam Road, Surrey Research Park, Guildford, GU2 7YN, UKc School of Pharmacy, University of Hertfordshire, College Lane Campus, Hatfield, Hertfortshire, AL10 9AB, UK

a r t i c l e i n f o

Article history:Received 17 December 2008Accepted 12 March 2009Available online 25 March 2009

Keywords:FoamHydrofluoroalkanePluronic

a b s t r a c t

Aesthetics are very important for topical products and as a consequence elegant vehicles such as spraysand foams are often preferred by patients. Pressurised systems are ideal to dose foams, however, as solittle is known about the influence of formulation characteristics on foam properties, the rational design ofthese systems difficult. This study aimed to assess the capability of pluronic surfactants to stabilise topicalpressurised hydrofluoroalkane (HFA) emulsions and attempted to define the formulation characteristicsthat had an impact upon foam properties. In situ phase diagrams and conductivity measurements wereused to characterise the HFA emulsions. Cryo-scanning electron microscopy images, collapse time (Ct)and wetting time (Wt) were used to assess the foams post dosing, i.e. after removal of the HFA. The resultsindicated that foam stability was a direct function of HFA emulsion type; HFA-in-water (HIW) emulsions

Emulsion

Topical generated stable foams, they had 30–100 �m bubble diameter with c.a. 40 bubbles in a 0.45 mm × 0.40 mmarea; water-in-HFA (WIH) emulsions created quick-breaking foams they contained 20–200 �m sized bub-bles and had 20 bubbles in an area of 0.45 mm × 0.40 mm. Therefore, the rational design of pressurised

ved i

1

cpbcFpiustdDsiawd

ae

ftbp2tatasaft

0d

topical foams can be achie

. Introduction

Efficient drug delivery into the skin is difficult as the Stratumorneum (SC), the outmost layer, forms a formidable barrier to theenetration of therapeutic agents. This problem is compoundedy the inefficient release of drugs from the majority of commer-ial ointment and cream formulations (Surber and Smith, 2005).or example, the topical application of levothyroxine cream hasreviously been reported to deliver only 0.8% of the applied drug

nto the skin (Padula et al., 2008). The inadequacy of topical prod-cts has driven research into the development of novel vehiclesuch as sprays and foams. The manipulation of the administra-ion vehicle is a simple, low cost and efficient method to improverug delivery to the skin (Ricciatti-Sibbald and Sibbald, 1989).iclofenac, heparin, fluticasone propionate, lidocaine and several

ex hormones have all previously been incorporated within top-

cal sprays (Morgan et al., 1998; Hegarty et al., 2002; Kaygusuznd Susaman, 2003; Brunner et al., 2005; Gorski et al., 2005),hereas clobetasol propionate, betamethasone valerate, minoxi-il, and pyrethrins have all been incorporated in foams (Amerio et

∗ Corresponding author. Tel.: +44 20 7848 4843; fax: +44 20 7848 4800.E-mail address: stuart.jones@kcl.ac.uk (S.A. Jones).

utdlf

uf

928-0987/$ – see front matter © 2009 Elsevier B.V. All rights reserved.oi:10.1016/j.ejps.2009.03.007

f the formulation is analysed in situ.© 2009 Elsevier B.V. All rights reserved.

l., 2003; Tanojo et al., 2004; Reid and Kimball, 2005; Rundegrent al., 2005).

Foams have many distinct advantages over other topical dosageorms including ease of application, lower density and the abilityo alter skin moisturisation (Purdon et al., 2003). As a result of theirenefits, patient compliance is often improved with foams com-ared to more conventional dosage forms (McCarty and Feldman,004). In addition, the potential of contaminating the unused por-ion of the medication is minimised as the foam is often dosed fromsealed airtight container. Furthermore, foams have been reported

o enhance topical drug delivery efficiency. For example, Franz etl. (2000) showed that a clobetasol propionate foam produced aignificantly greater percutaneous drug absorption compared tosolution; the total drug adsorption after 12 h was 2.6% for the

oam and 1.2% using the solution. However, despite their advan-ages the development of topical foams for commercial use remainsnattractive as their development is often time consuming andhus costly. The behaviour of drug-loaded foams is often unpre-ictable and rational development is impossible as there are very

ittle published data that links formulation characteristics with theoam properties.

Pressurised foam systems usually contain a highly volatile liq-id propellant to enable ejection of the dose. In addition, theoam can also contain either aqueous or non-aqueous co-solvents

armaceutical Sciences 37 (2009) 370–377 371

tia1seattp

piattfaflpuwoBB

srses2eowget

2

2

o(LBaGopfF

2

2

wakwSf

Fig. 1. The in-house designed conductivity cell for the analysis of HFA emulsions(5b

p(w8v

2

aocLta(tpwLiecimU1iatdtcd

Y. Zhao et al. / European Journal of Ph

o enhance excipient compatibility. Non-aqueous systems usuallyncorporate non-polar solvents such as ethanol, acetone, hexadecyllcohol, glycol ethers and polyglycols (Ricciatti-Sibbald and Sibbald,989), but due to the skin irritancy problems arising from theseolvents, aqueous foams are preferred (Trumbore et al., 2007). How-ver, many topical therapeutic agents, for example steroids andntibiotics demonstrate poor solubility in aqueous vehicles andherefore innovative formulation strategies are required to facili-ate the development of homogeneous products containing theseharmaceutical actives.

Most of the commercially available foams use hydrocarbonropellants, which are explosive, flammable and causes great

nconvenience during production, consumption and disposal. Inddition, as a result of their hazardous nature no in situ analysis ofhese systems can be performed and therefore it is difficult to assesshe interactions between the propellants and other excipients in theormulations. Hydrofluoroalkane (HFA) propellants are attractivelternatives to hydrocarbons as they are non-explosive and non-ammable compressed gases which are usually liquefied underressure for storage. HFAs have been approved for pharmaceuticalse, but as most drugs and surfactants display low solubility in HFA,ithout the help of an organic co-solvent such as ethanol, no previ-

us studies have generated aqueous based HFA foams (Vervaet andyron, 1999; Dickinson et al., 2000; McDonald and Martin, 2000;utz et al., 2002; Gupta et al., 2003).

Blondino and Byron (1998) showed that some hydrophilicurfactants demonstrate appreciable solubility in HFA and moreecently, Ridder et al. (2005) demonstrated that several pluronicurfactants, exhibited good solubility in HFA propellants. The gen-ration of aqueous HFA foams may be possible if HFA solubleurfactants are used to emulsify the water in the HFA (Zhao et al.,008). However, it is very difficult using current scientific knowl-dge to predict what properties the HFA emulsions should exhibit inrder to generate elegant foams. Therefore, the aims of this studyere to determine the feasibility of using pluronic surfactants to

enerate HFA aqueous foam formulations and to investigate theffect of the formulation characteristics on the physical stability ofhe foams after dose application.

. Materials and methods

.1. Materials

The HFA propellants tetrafluoroethane (HFA 134a) and heptaflu-ropropane (HFA 227) were kindly donated by Solvay Fluor GmbHFrankfurt, Germany). Pluronic 10R5, 17R2, 17R4, 25R4, 31R1, L61,81, L101, L121, L31, L35, L43, L44NF, L62D, and L92 were provided byASF (New Jersey, USA). Poloxamer 188 (pluronic F68) and polox-mer 407 (pluronic F127) were acquired from BASF (Ludwigshafen,ermany). Methocel E4M (hydroxypropyl methylcellulose) wasbtained from Colorcon Ltd. (Dartford, UK). Sodium chloride wasrovided by Sigma–Aldrich Ltd. (Gillingham, UK). HPLC (high per-ormance liquid chromatography) grade water was sourced fromisher (Leicestershire, UK).

.2. Methods

.2.1. Pluronic hydrofluoroalkane solubilityThe determination of surfactant solubility in the liquefied HFAs

as conducted visually at ambient temperature (23 ± 2 ◦C) using

commonly described method (Blondino and Byron, 1998). A

nown amount of surfactant together with a small magnetic fleaas placed in a 10 ml plastic coated glass canister (Schott UK Ltd.,

tafford, UK) which was sealed with a CV20 continuous valve (Giftrom Rexam Beauty & Pharma, Suresnes, France). As liquefied pro-

2

iaw

1: connection cable, 2: conductivity probe, 3: continuous valve, 4: display screen,: temperature probe, 6: test plate, 7: glass window, 8: sealing screw, 9: plastic cellulk body, 10: conductivity meter).

ellants with vapour pressure of 5.72 bar (HFA 134a) and 3.90 barHFA 227) at 20 ◦C, respectively, HFAs were added gradually byeight using a pressurised filler (Pamasol Willi Mäder AG, CH-808 Pfäffikon SZ, Switzerland) until the surfactant dissolution wasisibly apparent.

.2.2. Hydrofluoroalkane emulsion preparationHPLC grade water and 0.1% (w/v) methocel E4M was added to

10 ml canister. The appropriate surfactant, pluronic F127, L62Dr L31, selected based on their solubility in HFA, was added. Theanister was sealed with a 100 �l metered spray valve (Valois UKtd., Bletchley, UK) and HFA was filled into the canister. To ensurehe homogeneity, the whole mixture was stirred over twelve hourst 1000 rpm using a motorless electronic magnetic stirrer plateVariomag® Telesystem HP15, Florida Scientific Services, Inc., Day-ona Beach, USA). The HFA mixtures containing pluronic F127 wererepared in order to represent HFA-in-water (HIW) emulsions, i.e.ater was the continuous phase. The mixtures containing pluronic

62D or L31 were formulated as water-in-HFA (WIH) emulsions,.e. HFA was the continuous phase. To confirm the type of the HFAmulsion system the conductivity was assessed using a pressurisedell that was designed in-house (Fig. 1) containing a conductiv-ty probe (Jenway epoxy bodied, K = 1) connected to a conductivity

eter (Jenway 470 conductivity meter IP65) (VWR, Leicestershire,K). The meter was calibrated with standard solutions (84 and2,880 �S, Jenway) (VWR, Leicestershire, UK) and the conductiv-ty of 0.1% (w/v) methocel containing 1% (w/v) sodium chloridend HFA 134a/227 was tested by simply filling the solvent intohe pressure cell. Selected foam systems (Table 1) were preparedirectly in the sealed pressure cell with a continuous valve usinghe previously described method (Blondino and Byron, 1998). Theonductivity of the solvent or the emulsion in the pressure cell wasetermined in triplicate.

.2.3. Hydrofluoroalkane emulsion stabilityThe emulsion stability was evaluated by recording the time that

t took for phase separation (creaming/sendimentation) to occur atmbient temperature. A stable emulsion was defined as the one forhich the phase separation time exceeded 60 min, otherwise the

372 Y. Zhao et al. / European Journal of Pharma

Table 1The foam formulations designed for foam physical stability testing (HFA 134a:tetrafluoroethane, HFA 227: heptafluoropropane, methocel solution: 0.1% (w/v)methocel solution containing 1% (w/v) sodium chloride, HIW: HFA-in-water emul-sion, WIH: water-in-HFA emulsion).

Formulations Compositions (w/w)

Dispersed phase Continuous phase Emulsifier

HIWa 20% HFA 134a 78% Methocelsolution

2% Pluronic F127

HIWb 20% HFA 227 78% Methocel 2% Pluronic F127

WW

eb(aUe

2

ttrdefiTlbudir

2

i5ustwswHce

2

eca

2

tAlm(t

eos

3

3

ies1iawppdapai

3

oocpo215% and 40% (w/w) of dispersed phase (HFA 134a), but this rangenarrowed to 15–30% (w/w) for the HFA 227-in-water emulsion(Fig. 2). The inclusion of methocel in the aqueous phase as a viscos-ity modifier had no effect on emulsion physical stability (data notshown). Both types of HFA-in-water emulsions, i.e. those contain-

Table 2The apparent solubility of pluronic and reverse pluronic surfactants in HFA 134a(tetrafluoroethane) and HFA 227 (heptafluoropropane) (w/w), all data were obtainedat ambient temperature (23 ± 2 ◦C) (n = 3). For reverse pluronics, the number fol-lowing the “R” multiplied by 10 indicates the approximate weight percent EO in themolecule. For pluronics, the first letter describes their physical form, for example, “L”for liquid and “F” for solid. The last digit multiplied by 10 indicates the approximateEO content in weight percent. MW: molecular weight, HLB: hydrophile–lipophilebalance. The MW and HLB were from manufacturer; the average numbers of EO andPO repeat units were calculated using the average MW.

Pluronicsurfactant

MW Repeat unitsEO/PO

HLB Solubility inHFA 134a

Solubility inHFA 227

F68 8400 76/29 18–23 <0.03% <0.03%F127 12600 100/65 29 <0.03% <0.03%10R5 1950 22/8 12–18 Up to 50% Up to 50%17R2 2150 10/15 1–7 Up to 50% Up to 50%17R4 2650 24/14 7–12 Up to 50% Up to 50%25R4 3600 33/19 7–12 1.18% Up to 50%31R1 3250 7/25 1–7 1.00% Up to 50%L31 1100 1/17 1–7 Up to 50% Up to 50%L61 2000 2/31 1–7 1.94% Up to 50%L81 2750 3/42 1–7 0.99% Up to 50%L101 3800 4/59 1–7 0.24% Up to 50%L121 4400 5/68 1–7 0.15% Up to 50%

solutionIHa 15% Methocel solution 83% HFA 227 2% Pluronic L62DIHb 30% Methocel solution 68% HFA 227 2% Pluronic L62D

mulsion was considered unstable. The compositions of both sta-le and unstable HFA emulsions were plotted in a ternary diagramthe dispersed phase, the continuous phase and surfactant as threepexes) using Originpro software (Silverdale Scientific Ltd., Bucks,K) with the objective of investigating the relationship betweenmulsion stability and excipient ratio.

.2.4. Foam collapse time and wetting time determinationAll the foams listed in Table 1 were tested. The properties of

he foams were evaluated semi-quantitatively by foam collapseime (Ct) and foam wetting time (Wt) according to a previouslyeported method (Sanders, 1979). Wetting as a consequence ofrainage was judged by simply discharging one actuation of foamxpanded from a 100 �l metered valve onto a Whatman® celluloselter paper (Diameter 70 mm, Grade 1, Fisher, Leicestershire, UK).he wetting time was defined as the time elapsed when the cel-ulose paper was first wetted visibly. Foam collapse was assessedy releasing all the foam (8 g) from a single canister via a contin-ous valve into a 200 ml glass receptacle. The collapse time wasefined as the time when the foam volume decreased to 50% of its

nitial volume (measured using calibrated graduation on the glasseceptacle).

.2.5. Foam cryo-SEM analysisFoam samples were sprayed onto a slotted stub and then frozen

mmediately by plunging into ‘slushed’ nitrogen (−210 ◦C). Aftermin, the stub containing the sample was transferred under vac-um into the cryopreparation chamber (Gatan Alto 2500 cryo-SEMystem attached to an FEI Quanta 200F field emission scanning elec-ron microscope) (FEI UK Ltd., Cambridge, UK). The specimen stageas fixed at −140 ◦C. The foam sample was fractured using preci-

ion rotary knife (part of the Gatan Alto system) and sputter-coatedith chromium using 20 mA for 120 s (XE200 Xenosput, Edwardsigh Vacuum, Crawley, UK). The samples were transferred to theryo-stage (−140 ◦C) and SEM images were recorded with an accel-rating voltage of 3 kV.

.2.6. Foam light microscopy analysisSelected foams from both HFA-in-water and water-in-HFA

mulsions (4 g) (Table 1) were applied from a container with aontinuous valve to a glass slide and a digital image of the foamppearance was quickly taken.

.2.7. Statistic analysisAll data were presented as mean ± standard deviation and sta-

istical analysis of data was performed using SPSS version 16.0.

statistically significant difference was determined at a minimal

evel of significance of 0.05. All data were checked in terms of nor-ality (Kolmogorov–Smirnov test) and homogeneity of variances

Levene’s test). The analysis of pluronic solubility and foam wet-ing time data was performed using Student’s t-test, whereas the

LLLLL

ceutical Sciences 37 (2009) 370–377

mulsion conductivity and foam collapse data were analysed usingne-way analysis of variance (ANOVA) with a Turkey’ HSD (honestlyignificant difference) post hoc test.

. Results

.1. Pluronic hydrofluoroalkane solubility

The apparent solubility of two solid and 15 liquid pluronicsn two HFA solvents were determined with a relative intra-xperimental standard deviation (RSD) of less than 5%. The twoolid pluronic surfactants showed very low solubility in both HFA34a and 227 (<0.03%, w/w), but the miscibility of the liquid pluron-cs in HFA 227 in all cases was >50% (w/w) (Table 2). Only 10R5, 17R2,nd 17R4 of the reverse pulronics tested showed high miscibilityith HFA 227 and HFA 134a. The molecular weight (MW) of theluronics appeared to influence their solubility in HFA. For exam-le, pluronic L61 and L121, which were structurally identical, butiffered in average MW from 2000 (L61) to 4400 (L121) displayedlmost a one order of magnitude difference in solubility (Table 1,< 0.01). No attempt to draw a correlation between surfactant HLBnd HFA solubility was made as the HLB scale is not an appropriatendex to use for such polydisperse polymers (Table 2).

.2. Hydrofluoroalkane emulsion stability

Ternary diagrams were plotted in order to investigate the effectsf the dispersed phase, continuous phase and pluronic surfactantsn the physical stability of the HFA emulsions inside the pressurisedanisters. The emulsions containing more than 60% (w/w) of dis-ersed phase were not investigated in this study as phase inversionccurred. The HFA 134a-in-water emulsion and the water-in-HFA27 emulsions remained stable when constructed using between

35 1900 11/16 18–23 2.19% Up to 50%43 1850 6/22 7–12 2.47% Up to 50%44NF 2200 10/23 12–18 2.09% Up to 50%62D 2360 5/33 1–7 1.51% Up to 50%92 3650 8/50 1–7 0.34% Up to 50%

Y. Zhao et al. / European Journal of Pharmaceutical Sciences 37 (2009) 370–377 373

Fig. 2. Ternary diagram of HFA-in-water emulsion physical stability. One hour was defined as the boundary between stable (filled circle) and unstable (empty square)emulsions (phase separation due to creaming/sedimentation). (A) Diagram of HFA 134a-in-water emulsions with pluronic F127 as the stabiliser and (B) Diagram of HFA227-in-water emulsions with pluronic F127 as the stabiliser, HFA 134a: tetrafluoroethane; HFA 227: heptafluoropropane.

F fined( 34a ee eptafl

ipTpee

fottwaoT(ttonpwc

3

The wetting time (Wt) of the two foams generated from thewater-in-HFA 227 emulsions were very short and significantlydifferent from each other at 9 ± 2 s (WIHa) and 15 ± 3 s (WIHb)(p < 0.01) (Table 3). In contrast, the Wt of the two foams gener-ated from the HFA-in-water emulsions were almost six times longer

Table 3The conductivity and physical stability of different foam systems. The stability wasdefined by the indices of foam wetting time (Wt) and collapse time (Ct). All the sam-ples were tested at ambient temperature (n > 10), HIW: HFA-in-water emulsion withmethocel solution as continuous phase; WIH: water-in-HFA emulsion with HFA 227as continuous phase; methocel solution: 0.1% (w/v) methocel aqueous solution con-taining 1% sodium chloride (w/v); HFA 227: heptafluoropropane. Foam compositionscan be found in Table 1.

Formulations Conductivity Foam wettingtime, Wt (s)

Foam collapsetime, Ct (s)

HIWa 15.7 ± 0.2 mS 61 ± 4 178 ± 25

ig. 3. Ternary diagram of water-in-HFA emulsion physical stability. One hour was dephase separation due to creaming/sedimentation). (A) Diagram of water-in-HFA 1mulsions with pluronic L62D as stabiliser. HFA 134a: tetrafluoroethane; HFA 227: h

ng both 134a and 227, were unstable unless the concentration ofluronic F127 surfactant that was included was <5% (w/w) (Fig. 2).he water-in-HFA emulsion could be stablised with up to 10% (w/w)luronic L62D surfactant (Fig. 3B), but a stable water-in-HFA 134amulsions could not be attained whatever the ratio of excipientsmployed (Fig. 3A).

Based on the emulsion physical stability results, four very dif-erent formulations were selected in order to assess the influencef formulation upon foam physical stability (Table 1). The conduc-ivity values of these HFA emulsions were determined in ordero ensure that no phase inversion had occurred. The HFA-in-ater emulsions displayed a conductivity of 15.7 ± 0.2 mS (HIWa)

nd 15.3 ± 0.4 mS (HIWb) compared to the 0.1% methocel aque-us solution containing 1% NaCl (w/v) at 16.6 ± 0.1 mS (Table 3).he conductivity of water-in-HFA 227 emulsions were 1.1 ± 0.0 mSWIHa) and 2.1 ± 0.2 mS (WIHb) compared to the HFA 227 conduc-ivity of −0.1 ± 0.0 �S (Table 3). Statistical analysis demonstratedhat there was no significant difference between the conductivities

f the HIWa and HIWb emulsions (p = 0.16), but that there was a sig-ificant difference between the two emulsions and the continuoushase (p < 0.01). The conductivities of WIHa and WIHb emulsionsere significantly different from each other and different to the

ontinuous phase (p < 0.01).

HMWWH

as the boundary between stable (filled circle) and unstable (empty square) emulsionsmulsions with pluronic L31 as the stabiliser and (B) diagram of water-in-HFA 227uoropropane.

.3. Foam stability upon hydrofluoroalkane evaporation

IWb 15.3 ± 0.4 mS 62 ± 5 509 ± 16ethocel solution 16.6 ± 0.1 mS N/A N/AIHa 1.1 ± 0.0 �S 9 ± 2 7 ± 1IHb 2.1 ± 0.2 �S 15 ± 3 11 ± 2

FA 227 −0.1 ± 0.0 �S N/A N/A

374 Y. Zhao et al. / European Journal of Pharmaceutical Sciences 37 (2009) 370–377

F on sca( on; HIi

cTCfo

Fafd

w

ig. 4. SEM (gatan Alto 2500 cryo-SEM system with an FEI Quanta 200F field emissiB) SEM image of the ‘HIWa’ foam. WIH: foam produced from water-in-HFA emulsin Table 1.

ompared to the foams from the water-in-HFA emulsions (p < 0.01).

he collapse time (Ct) of the foams correlated well with the Wt. Thet of the water-in-HFA 227 emulsions WIHb (11 ± 2 s) and WIHaoams (7 ± 1 s) were rapid but significantly different from eachther (p < 0.01). The Ct of the foams generated from the HFA-in-

ig. 5. The foam appearances from four types of HFA emulsion pre and postpplication (A) HIWa, (B) HIWb, (C) WIHa, and (D) WIHb; HIW: foam producedrom HFA-in-water emulsion; WIH: foam produced from water-in-HFA emulsion;etailed compositions were summarised in Table 1.

e

ebobbt2Tegs

4

(oplocma2h

nning electron microscope) image of foams, (A) SEM image of the ‘WIHa’ foam andW: foam produced from HFA-in-water emulsion. Foam compositions can be found

ater emulsions were significantly longer than the water-in-HFAmulsions at 509 ± 16 s, (HIWb) and 178 ± 25 s (HIWa) (p < 0.01).

Cryo-SEM images of HIWa and WIHa were generated in order toxamine the foam morphology and to qualitatively assess the bub-le density (Fig. 4). In a field of 0.45 mm × 0.40 mm, the numberf foam bubbles generated from the HIWa emulsion was 40 with aubble size range of 30–100 �m (counted from Fig. 4B). The averageubble number in the foam generated from WIHa emulsion was lesshan the HIWa emulsion at 20 and the bubble size ranged from 20 to00 �m which suggested a less stable foam (counted from Fig. 4A).he light microscopy images also indicated that the foams gen-rated from HFA-in-water emulsions were more stable than thoseenerated from water-in-HFA emulsions. The HIWa foam displayeduperior foam height and retained its white appearance (Fig. 5).

. Discussion

Non-ionic pluronic (PEO–PPO–PEO) and reverse pluronicPPO–PEO–PPO) surfactants are synthetic block copolymersf hydrophilic poly(oxyethylene) (PEO) and hydrophobicoly(oxypropylene) (PPO). The manipulation of block chain

ength within the structure of the molecule creates various gradesf pluronics differing in physicochemical characteristics. These

opolymers are commonly used as stabilisers of dispersed phar-aceutical formulations, e.g. emulsions and foams, due to their

mphiphilic properties (Sedev et al., 1999; Kawaguchi and Kubota,004). The behaviour of pluronic surfactants in aqueous solutionsas been well characterised, but in comparison investigation into

armac

tiectf

ecaiFphwwlTlr1

uPi(m(awcRtc

abt1efctptmttwaHwwaia2esdtd

eo

bttleh(tasnietsridbtuo

OttCdtvcFowpusb

yraoi(wt

tpaitismis

s

Y. Zhao et al. / European Journal of Ph

he interaction between these copolymer surfactants and non-onic vehicles such as HFA propellants remains limited (Svenssont al., 1999). As a consequence, in this study the behaviour of theseopolymers in HFA was investigated in an attempt to determineheir suitability for the formulation of non-polar pharmaceuticaloams.

Unlike the low molecular weight liquid pluronics, the appar-nt solubility of the high MW solid pluronics in HFA was poor andould be attributed to the strong pluronic intermolecular inter-ction, e.g. hydrogen bonding which reduces the extent of theirnteractions with HFA. Guo et al. (1999) employed FT-Raman andTIR spectroscopy to investigate the conformational structure ofluronic triblock copolymers. They found that solid pluronics with aigh PEO/PPO ratio exhibited extensive intermolecular interactionshich resulted in the formation of a secondary helical structure,hilst the molten liquid pluronics with a low PEO/PPO ratio had

ess intermolecular interactions and thus a disordered structure.he intramolecular interactions also affected the solubility of theiquid and reverse pulronics. For example, at the same PEO/PPOatio, the solubility of liquid pluronics and reverse pluronics in HFA34a decreased with increasing copolymer MW.

Liquid pluronics and reverse pluronics demonstrated higher sol-bility in HFA 227 (C3HF7) compared to in HFA 134a (C2H2F4).revious studies have shown the presence of both hydrogen bond-ng and dipole–dipole interactions between solute and HFA solventsByron et al., 1994; Vervaet and Byron, 1999). HFA 134a has a dipole

oment value of 2.06 debye, higher than 0.93 debye of HFA 227Solvay Fluor, 2008). If the dipole–dipole forces dominate the inter-ction between HFA and pluronics, their solubility in HFA 134aould be greater than in HFA 227 and as this was not the case in the

urrent study, it indicates that the hydrogen bonds were dominant.idder et al. (2005) showed stronger hydrogen bonding interac-ions between pluronics and HFA 227 compared to HFA 134a, whichoncurs with the hypothesis derived from the current work.

Due to the large interfacial free energy, a coarse emulsion isthermodynamically unstable dispersion. Thus, an emulsion sta-iliser such as a surfactant is usually used to decrease the interfaceension between the dispersed and continuous phases (Takeo,999). The phase behaviour of HFA-in-water and water-in-HFAmulsions is thought to depend on the affinity of the surfactantor both the HFA and water phases (Selvam et al., 2006). In theurrent work, a relatively hydrophilic pluronic, F127 was usedo produce HFA-in-water emulsions and two more hydrophobicluronics, L31 and L62D to generate water-in-HFA emulsions. Ashe HFA emulsions were held under pressure, the characterisation

ethods for conventional emulsions such as droplet size and set-ling rate assessment were not practically feasible. Therefore, withhe aid of glass canisters, the physical stability of HFA emulsionsas determined visually using the time to achieve phase separation

s a qualitative measure. According to this visual assessment theFA 134a-in-water emulsion with less than 15% dispersed phaseas not stable, probably because of the low emulsion viscosityhich favoured creaming (according to Stokes’ law) (Sepulveda et

l., 2003). When the dispersed phase increased above 40% phasenversion started to occur, which resulted in emulsion instability;n effect reported previously (Allouche et al., 2004; Thakur et al.,007). Similar physical stability trends were observed for other HFAmulsions, e.g. HFA 227-in-water emulsions containing the F127tabiliser were physically stable when formulated using 15–30% ofispersed phase (Fig. 2B) and water-in-HFA 227 emulsions using

he L62D stabiliser were only stable when containing 10–40% ofispersed phase (Fig. 3B).

The physical stability of both HFA-in-water and water-in-HFAmulsions were shown to be dependant on the concentrationf pluronic stabilisers employed. The adsorption of pluronic sta-

fetea

eutical Sciences 37 (2009) 370–377 375

iliser at the HFA–water interface in both types of emulsions wouldheoretically form monomolecular films to reduce the interfacialension, the surface free energy, and hence the tendency for coa-escence. Although the ability of pluronics to stabilise aqueousmulsions has been studied in depth, very little previous workas investigated the pluronic steric stabilisation in HFA emulsionsPeguin et al., 2007; Wu et al., 2008). Steric repulsion depends onhe nature, thickness, and completeness of the pluronic stabiliser-dsorbed layers on the emulsion droplets. At a low concentration oftabiliser, it was hypothesised that the pluronic copolymer wouldot fully cover the whole surface of emulsion droplets, resulting

n an unstable emulsion. As the pluronic concentration increased,nough stabiliser would be available to cover the entire surface ofhe emulsion droplets, giving rising to a physically stable emul-ion. A further increase of pluronic stabiliser concentration wouldesult in physical instability as the excess pluronic molecules wouldnteract with the pluronic stabiliser layer adsorbed on the emulsionroplets and destabilise the emulsion. This theory was supportedy the study of Barnes and Prestidge (2000) who found a similarrend when stabilising a dispersed phase in water. However, thenique properties of HFA mean that the exact interfacial behaviourf pluronics still needs further investigation.

According to the classic theory of Derjaguin–Landau–Verwey–verbeek (DLVO), charge on the surface of emulsion droplets helps

o stabilise the emulsion through repulsive electrostatic interac-ions (Derjaguin and Landau, 1941; Verwey and Overbeek, 1948).harge is ubiquitous in aqueous systems as the water has a highielectric constant (≈80), but in non-polar solvents such as HFA,he electrostatic barrier to charge is 40 times larger than in polarehicles and much greater than thermal energy kBT; therefore theharge effect is thought to be inconsequential (Hsu et al., 2005).or HFA emulsions, the surface charges are present at the interfacef water and HFA, but the contribution of electrostatic interactionsas assumed not to be sufficient to stabilise the emulsion com-ared to steric effects. This theory can be supported by the workndertaken by Lee et al. (1999) who found that the electrostatictabilisation in water-in-carbon dioxide emulsions was negligibleecause of the very short Debye length in non-polar systems.

The HFA emulsion type was confirmed using conductivity anal-sis by comparison with the continuous phase and found to beepresentative of the initial ratios of the phases employed. The devi-tion of both emulsion conductivity away from the conductivityf the continuous phase was probably a result of ion partition-ng caused by the osmotic effect (Bart et al., 1995). Leunissen et al.2007) has shown that ion partitioning occurs at many interfaceshere there was a difference in the dielectric constant between the

wo phases.The foams expanded from the HFA emulsions were assessed in

erms of Wt and Ct; two indices employed for foam evaluation inrevious work (Sanders, 1979). Although the means by which foamsre produced can be quite different, the mechanism of their collapses often similar. Gas diffusion drives bubble collapse kinetics andhis process is typically accompanied by drainage of the interven-ng fluid (Brady and Ross, 1944). Foam collapse can be retarded bytabilising the foam bubbles and slowing the rupture process. Oneethod of achieving this is to use surfactants at the bubble foam

nterface which can stabilise the bubbles via electrostatic and/orteric mechanisms (Tan et al., 2005).

HFA-in-water emulsions (HIWa and HIWb) generated physicallytable foams with a high bubble density, whereas quick-breaking

oams with a low bubble density were produced from water-in-HFAmulsions (WIHa and WIHb). It was hypothesised that the loca-ion of the HFA phase in the emulsion determined the formation ofither stable or quick-breaking foam. For example, when the WIHand WIHb emulsions (the HFA was located in the external phase)

3 arma

wteatAbttCs

iptohswHafittoc

5

dasetgwefbTd

A

P

R

A

A

B

B

B

B

B

B

B

DD

E

F

G

G

G

H

H

K

K

L

L

L

M

M

M

P

P

P

R

R

R

R

S

S

S

S

S

S

76 Y. Zhao et al. / European Journal of Ph

ere actuated from the canister, the HFA evaporated quickly andhe foam collapsed rapidly. In contrast, when the HIWa and HIWbmulsions (the HFA was located in the disperse phase) were actu-ted the HFA evaporation was slower as it was retarded by diffusionhrough the aqueous phase and this resulted in more stable bubbles.s a consequence the images of the foams showed a greater num-er of bubbles were generated for HFA-in-water systems comparedo water-in-HFA emulsions. A similar effect of propellant localisa-ion upon foam stability has been reported previously (Sciarra andutie, 1996) and concurs with the results obtained in the currenttudy.

The type of HFA propellants used in the foam formulations alsonfluenced the foam behaviour. For example, HIWa (HFA 134a asropellant) was less stable than HIWb (HFA 227 as propellant) inerms of foam collapse time. The effect of propellant propertiesn foam stability has previously been investigated and showed aigher miscibility of the two phases results in the generation of lesstable foams (Langevin, 2008). As HFA 134a showed much higherater solubility (193 ppm, 20 ◦C) than HFA 227 (58 ppm, 20 ◦C),FA 134a transport across water films was thought much fasternd this provides one explanation of why it generated less stableoams. In addition, due to the smaller size of HFA 134a (MW = 102),ts diffusion coefficient in water would be greater at the sameemperature when compared to HFA 227 (MW = 170) (accordingo Stokes–Einstein relation) (Einstein, 1956). The faster diffusionf HFA 134a could also play a role in the lower stability of HIWaompared with HIWb.

. Conclusions

The solubility of pluronic surfactants in HFA was found toepend on the physicochemical properties of both the surfactantnd HFA. The solid pluronics with high PEO/PPO ratio showed poorolubility in HFA compared to liquid pluronics. Both HFA-in-watermulsions containing the F127 and water-in-HFA emulsions con-aining the L62D were primarily stabilised by the steric hindranceenerated by the pluronic copolymers. It was found that the foametting time and collapse time correlated with the type of HFA

mulsion from which the foam was derived. A physically stableoam could be obtained from a HFA-in-water emulsion and a quick-reaking foam could be generated from a water-in-HFA emulsion.hese novel foam vehicles offer potential application in the topicalelivery of both hydrophilic and lipophilic therapeutic agents.

cknowledgements

The authors would like to thank the financial support from Med-harm Ltd. (UK) and the EPSRC.

eferences

llouche, J., Tyrode, E., Sadtler, V., Choplin, L., Salager, J.L., 2004. Simultaneous con-ductivity and viscosity measurements as a technique to track emulsion inversionby the phase-inversion-temperature method. Langmuir 20, 2134–2140.

merio, P.L., Capizzi, R., Milani, M., 2003. Efficacy and tolerability of natural syner-gised pyrethrins in a new thermo labile foam formulation in topical treatment ofscabies: a prospective, randomised, investigator-blinded, comparative trial vs.permethrin cream. Eur. J. Dermatol. 13, 69–71.

arnes, T.J., Prestidge, C.A., 2000. PEO-PPO-PEO block copolymers at the emulsiondroplet-water interface. Langmuir 16, 4116–4121.

art, H.J., Jungling, H., Ramaseder, N., Marr, R., 1995. Water and solute solubilizationand transport in emulsion liquid membranes. J. Membr. Sci. 102, 103–112.

londino, F.E., Byron, P.R., 1998. Surfactant dissolution and water solubilization in

chlorine-free liquified gas propellants. Drug Dev. Ind. Pharm. 24, 935–945.

rady, A.P., Ross, S., 1944. The measurement of foam stability. J. Am. Chem. Soc. 66,1348–1356.

runner, M., Dehghanyar, P., Seigfried, B., Martin, W., Menke, G., Muller, M., 2005.Favourable dermal penetration of diclofenac after administration to the skinusing a novel spray gel formulation. Br. J. Clin. Pharmacol. 60, 573–577.

S

S

ceutical Sciences 37 (2009) 370–377

utz, N., Porte, C., Courrier, H., Krafft, M.P., Vandamme, T., 2002. Reverse water-in-fluorocarbon emulsions for use in pressurized metered-dose inhalers containinghydrofluoroalkane propellants. Int. J. Pharm. 238, 257–269.

yron, P.R., Miller, N.C., Blondino, F.E., Visich, J.E., Ward, G.H., 1994. Some aspects ofalternative propellant solvency. Respir. Drug Deliv. IV 1, 231–242.

erjaguin, B.V., Landau, L., 1941. Acta Physicochim. URSS 14, 633.ickinson, P.A., Seville, P.C., McHale, H., Perkins, N.C., Taylor, G., 2000. An investiga-

tion of the solubility of various compounds in the hydrofluoroalkane propellantsand possible model liquid propellants. J. Aerosol. Med. 13, 179–186.

instein, A., 1956. Investigation on the Theory of Brownian Movement. Dover, NewYork, pp. 1–86.

ranz, T.J., Parsell, D.A., Myers, J.A., Hannigan, J.F., 2000. Clobetasol propionate foam0.05%: a novel vehicle with enhanced delivery. Int. J. Dermatol. 39, 535–538.

orski, G., Szopinski, P., Michalak, J., Marianowska, A., Borkowski, M., Geremek, M.,Trochimczuk, M., Brotanek, J., Sarnik, S., Semenka, J., Wilkowski, D., Noszczyk,W., 2005. Liposomal heparin spray: a new formula in adjunctive treatment ofsuperficial venous thrombosis. Angiology 56, 9–17.

uo, C., Liu, H.Z., Wang, J., Chen, J.Y., 1999. Conformational structure of triblockcopolymers by FT-Raman and FTIR spectroscopy. J. Colloid Interface Sci. 209,368–373.

upta, A., Stein, S.W., Myrdal, P.B., 2003. Balancing ethanol cosolvent concentrationwith product performance in 134a-based pressurized metered dose inhalers. J.Aerosol. Med. 16, 167–174.

egarty, A.M., Hodgson, T.A., Lewsey, J.D., Porter, S.R., 2002. Fluticasone propi-onate spray and betamethasone sodium phosphate mouthrinse: a randomizedcrossover study for the treatment of symptomatic oral lichen planus. J. Am. Acad.Dematol. 47, 271–279.

su, M.F., Dufresne, E.R., Weitz, D.A., 2005. Charge stabilization in nonpolar solvents.Langmuir 21, 4881–4887.

awaguchi, M., Kubota, K., 2004. Rheo-optical properties of silicone oil emulsionsin the presence of polymer emulsifiers. Langmuir 20, 1126–1129.

aygusuz, I., Susaman, N., 2003. The effects of dexamethasone, bupivacaine and top-ical lidocaine spray on pain after tonsillectomy. Int. J. Pediatr. Otorhinolaryngol.67, 737–742.

angevin, D., 2008. Aqueous foams: a field of investigation at the frontier betweenchemistry and physics. Chemphyschemistry 9, 510–522.

ee, C.T., Psathas, P.A., Johnston, K.P., deGrazia, J., Randolph, T.W., 1999. Water-in-carbon dioxide emulsions: formation and stability. Langmuir 15, 6781–6791.

eunissen, M.E., Zwanikken, J., van, R.R., Chaikin, P.M., van, B.A., 2007. Ion partitioningat the oil–water interface as a source of tunable electrostatic effects in emulsionswith colloids. Phys. Chem. Chem. Phys. 9, 6405–6414.

cCarty, A., Feldman, S., 2004. Study of patient preferences for ointment, cream andfoam formulations. J. Am. Acad. Dermatol. 50, 150.

cDonald, K.J., Martin, G.P., 2000. Transition to CFC-free metered dose inhalers–intothe new millennium. Int. J. Pharm. 201, 89–107.

organ, T.M., Reed, B.L., Finnin, B.C., 1998. Enhanced skin permeation of sex hor-mones with novel topical spray vehicles. J. Pharm. Sci. 87, 1213–1218.

adula, C., Pappani, A., Santi, P., 2008. In vitro permeation of levothyroxine acrossthe skin. Int. J. Pharm. 349, 161–165.

eguin, R.P.S., Wu, L.B., Da Rocha, S.R.P., 2007. The ester group: howhydrofluoroalkane-philic is it? Langmuir 23, 8291–8294.

urdon, C.H., Haigh, J.M., Surber, C., Smith, E.W., 2003. Foam drug delivery in der-matology: beyond the scalp. Am. J. Drug. Dev. 1, 71–75.

eid, D.C., Kimball, A.B., 2005. Clobetasol propionate foam in the treatment of pso-riasis. Expert. Opin. Pharmacother. 6, 1735–1740.

icciatti-Sibbald, D., Sibbald, R.G., 1989. Dermatologic vehicles. Clin. Dermatol. 7,11–24.

idder, K.B., vies-Cutting, C.J., Kellaway, I.W., 2005. Surfactant solubility and aggre-gate orientation in hydrofluoroalkanes. Int. J. Pharm. 295, 57–65.

undegren, J., Westin, A., Kohut, B., 2005. Hair growth efficacy assessment of anew topical minoxidil foam formulation in the stumptail macaque. J. Invest.Dermatol. 124, A98.

anders, P.A., 1979. General properties of foams. In: Sanders, P.A. (Ed.), Hand-book of Aerosol Technology. Van Nostrand Reinhold Company, New York,pp. 305–309.

ciarra, J.J., Cutie, A.J., 1996. Aerosol suspensions and emulsions. In: Lieberman,H.A., Rieger, M.M., Banker, G.S. (Eds.), Pharmaceutical Dosage Forms: DisperseSystems. Dekker, New York, pp. 417–424.

edev, R., Jachimska, B., Khristov, K., Malysa, K., Exerowa, D., 1999. Foamability ofPEO-PPO-PEO triblock copolymers (P85 and F108) and role of the foam films. J.Dispersion Sci. Technol. 20, 1759–1776.

elvam, P., Peguin, R.P.S., Chokshi, U., daRocha, S.R.P., 2006. Surfactant design forthe 1,1,1,2-tetrafluoroethane-water interface: ab initio calculations and in situhigh-pressure tensiometry. Langmuir 22, 8675–8683.

epulveda, E., Kildsig, D.O., Ghaly, E.S., 2003. Relationship between internal phasevolume and emulsion stability: the cetyl alcohol/stearyl alcohol system. Pharm.Dev. Technol. 8, 263–275.

olvay Fluor, 2008. Solkane® 227 pharma and 134a pharma–properties.http://www.solvay-fluor.com/product/properties/0,0,- EN-1000700,00.html

(accessed 07/01/2008).

urber, C., Smith, E.W., 2005. The mystical effects of dermatological vehicles. Der-matology 210, 157–168.

vensson, M., Alexandridis, P., Linse, P., 1999. Phase behavior and microstructurein binary block copolymer/selective solvent systems: experiments and theory.Macromolecules 32, 637–645.

armac

T

T

T

T

T

V

V

Y. Zhao et al. / European Journal of Ph

akeo, M., 1999. Emulsions and microemulsions. In: Takeo, M. (Ed.), Disperse Sys-tems. Wiley-VCH, Weinheim, pp. 244–256.

an, S.N., Fornasiero, D., Sedev, R., Ralston, J., 2005. The role of surfactant struc-ture on foam behaviour. Colloids Surf. A. Physicochem. Eng. Aspects 263,233–238.

anojo, H., Lenn, J., Huang, X., 2004. Compatibility of betamethasone 17-valerate andketoconazole foam formulations in combination topical therapy. J. Am. Acad.Dematol. 50, 173.

hakur, R.K., Villette, C., Aubry, J.M., Delaplace, G., 2007. Formulation-compositionmap of a lecithin-based emulsion. Colloids Surf. A. Physicochem. Eng. Aspects310, 55–61.

W

Z

eutical Sciences 37 (2009) 370–377 377

rumbore, M.W., Gurge, R.M., Hirsh, J.C., 2007. Topical pharmaceutical foam compo-sition. US Patent 20070154402.

ervaet, C., Byron, P.R., 1999. Drug-surfactant-propellant interactions in HFA-formulations. Int. J. Pharm. 186, 13–30.

erwey, E.J.W., Overbeek, J.T.G., 1948. Theory of the Stability of Lyophobic Colloids:

The Interactions of Sol Particles Having an Electric Double Layer. Elsevier, Ams-terdam, pp. 1–182.

u, L., Al-Haydari, M., da Rocha, S.R.P., 2008. Novel propellant-driven inhala-tion formulations: engineering polar drug particles with surface-trappedhydrofluoroalkane-philes. Eur. J. Pharm. Sci. 33, 146–158.

hao, Y., Jones, S.A., Brown, M.B., 2008. Topical formulations. UK Patent GB0715646.6.