Journal of Colloid and Interface Science 440 (2015) 78–83
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Volatile fluorinated nanoemulsions: A chemical route to controlled delivery of inhalation Anesthesia Ibrahim E. Salama a, Claire L. Jenkins a, Alun Davies a, Jeffrey N. Clark a,b, Antony R. Wilkes b, Judith E. Hall b, Alison Paul a,⇑ a b
School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff, UK Department of Anesthetics, Intensive Care and Pain Medicine, School of Medicine, Cardiff University, Cardiff, UK
a r t i c l e
i n f o
Article history: Received 11 June 2014 Accepted 16 October 2014 Available online 29 October 2014 Keywords: Anesthesia Drug-delivery Nanoemulsions Fluorocarbon Nanotechnology
a b s t r a c t Novel dispersions of the volatile inhalation anesthetic sevoflurane have been formulated that can provide controlled, sustainable release of anesthetic over clinically useful timescales. The emulsions can be simply formed with manual shaking, reproducibly yielding droplets of the order of 250 nm diameter, i.e. within the nanoemulsion range. Using a custom flow-rig, release of anesthetic gas from the emulsion has been evaluated, and clinically useful levels achieved through appropriate stirring of the formulation. Stirring can also be used to temporarily increase or decrease the amount of anesthetic released. Once consideration of the unusual nature of the fluorinated systems (phase separation by sedimentation rather than creaming), and the highly perturbed environment of their evaluation (under stirring and flow of gas), the observed behavior regarding sevoflurane evaporation can be reasonably well explained by existing theoretical models. Links between anesthetic release and emulsion structure have been defined, providing the basis for future development. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Nonionic partially fluorinated surfactants have been used to stabilize emulsions and microemulsions of fluorinated oils in water, e.g. for formation of blood substitutes [1–4] or as templating media for the preparation of porous materials [5–9]. For volatile emulsions previous fundamental studies have shown that oil evaporation depends both on the type of stabilizer used and on the solubility of the oil in the continuous water phase [10–12]: evaporation rates of water soluble oils approach that of the bulk oil, but evaporation of very low solubility oils is significantly hindered, an effect magnified by the use of a polymeric (as opposed to low molecular weight surfactant) stabilizer [10–12]. Other groups have focused on relationships between phase structures and the evaporation pathway, primarily for fragrance emulsions [13–15]. Here we utilize a volatile anesthetic as the evaporating component of an oil-in-water (o/w) emulsion. This is used to evaluate the feasibility of using a colloid science based approach to inhalation delivery of an anesthetic (as opposed to intravenous delivery [16]).
⇑ Corresponding author at: School of Chemistry, Main Building, Park Place, Cardiff CF10 3AT, UK. E-mail address: [email protected] (A. Paul). http://dx.doi.org/10.1016/j.jcis.2014.10.021 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
Currently, safe and accurate anesthesia delivery is reliant on large, complex and expensive machines operated by very specialized, highly trained personnel. All of these factors limit their use outside of modern clinical settings, and a simpler, safer and cheaper delivery system could provide an alternative means of anesthesia delivery. This is particularly relevant for challenging situations including pre-hospital medicine, in-the-field anesthesia, warfare anesthesia and in the challenging conditions often encountered in developing countries. Ethically, the delivery of anesthesia and analgesia are mandatory for surgical procedures worldwide, but in developing countries access to safe anesthesia can be limited due to a shortage of trained staff, and compromised by failure of complex equipment, leading to reduced access to healthcare. The aim of the work is therefore to provide proof of concept for a dispersion based anesthetic delivery device. Anesthetic is released in vapor form from a formulation, and picked up by a carrier flowing through the device and to the patient (Fig. 1). Specific targets are to demonstrate (i) release of clinically safe and useful concentrations of anesthetic (0.5–4 vol.% in the carrier gas stream). This is evaluated over a 60 min timescale at fixed gas flows of 1 L min1 and 4 L min1, chosen to exemplify typical clinically used flow rates; (ii) to obtain higher release (up to 8 vol.%) for shorter periods to create a release profile suitable for both induction and maintenance of anesthesia, and (iii) to provide a means of adjusting anesthetic release in a controllable and responsive
79
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
manner, without changing the carrier-gas flow used. This controlled variability is essential for the potential clinical use of the device and formulation.
Table 1 Formulation composition and oil droplet sizes for different performance sevoflurane containing emulsions. Carrier gas flow rate/L min1
Release level/ vol%
Surfactanta concentration/ wt.% (aq)
vol% anesthetic in formulation
Average droplet diameter/ nm
1 1 1 1 1 4 4 4 4
4b 3 2 1 0.5 4 3 2 0.5
18 10 7 4 8 25 22 17 7
31 19 13 6 6 44 36 29 13
209 259 261 239 188 256 206 384 188
2. Materials and methods 2.1. Materials Sevoflurane (99.98%, Abbot, UK) and Zonyl FSN-100 (technical grade, ABCR, Germany) were both used as received. Industrial nitrogen gas (BOC, UK) was used as the carrier gas throughout. Water was reagent-grade produced by a RiOs 5 purification system (Millipore, USA).
(±2) (±0.6) (±4) (±5) (±4) (±5) (±2) (±5) (±4)
a
2.2. Methods
Zonyl FSN-100. 4 vol% equates to 2 maximum alveolar concentration (MAC). One MAC is the alveolar concentration of a volatile anesthetic that produces no movement in 50% of spontaneously-breathing patients during skin incision. b
2.2.1. Sample preparation All formulations were prepared on 150 mL scale in 500 mL DURAN laboratory bottle with a screw cap and equilibrated at 20 °C prior to testing. Formulation compositions are given in Table 1. Emulsions were prepared by vigorously shaking (by hand) a known volume of the anesthetic with a pre-prepared aqueous surfactant solution, for a fixed time of 60 s. Method validation experiments were performed using different shaking times, and independent preparation by different personnel to ensure robustness of the preparation procedure. This simple method of dispersion greatly simplifies the formulation stability requirements, as the emulsion itself is formed at point of use. 2.2.2. Characterization Ternary phase diagrams were constructed by visual inspection. The continuous phase of the emulsion was determined by a standard drop-test method (an added drop of the continuous phase will disperse freely into the bulk; a drop of the dispersed phase will separate at the top or bottom of the sample, depending on density difference). Emulsion droplets were characterized by light-microscope imaging using an Olympus BX50 system microscope (Olympus, UK) fitted with JVC TK-C1380 color video camera (JVC, Japan) and analyzed using Image J software (Fiji, USA). Additional measurements were obtained from dynamic light scattering analysis using a
Brookhaven ZetaPlus (Brookhaven Instruments Ltd., USA). For light scattering measurements the emulsions were diluted by a factor of 20–50 depending on the emulsion concentration. 2.2.3. Release testing The controlled release of anesthetic agents from the formulations was evaluated using the experimental flow-rig shown in Fig. 1. The upper module consists of a 250 mL cylindrical plastic container (which holds the formulation and a magnetic stirrer bar) attached to a custom built Teflon fitting which has an inlet port for entrance of the free carrier gas and an outlet port for the exit of the anesthetic-loaded carrier gas. The module was placed on a standard magnetic stirrer (Heidolph MR 3002, Germany) for controlled stirring. Nitrogen gas was passed through the flow module at a controlled flow rate, typically 1 L min1 and 4 L min1. In lab experiments nitrogen was used as the carrier gas for convenience, although control experiments using air and air/O2 mixtures showed that the performance was independent of gas composition. The concentration of the anesthetic agent in the outlet stream was measured with a standard anesthetic monitor (Capnomac Ultima, Datex Instrumentarium Inc., Helsinki, Finland). The formulation
carrier gas in anaesthec monitor
sampling port
to paent carrier gas and anaesthec gas out
(not to scale) Fig. 1. Schematic of test flow-rig. SA = 60 cm2. Typical stirring rates 100–800 rpm. Typical formulation volume 120 cm3.
80
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
was thermostatted using a double-walled glass water-jacket connected to a circulating water bath to maintain temperatures other than 20 °C. The temperature of the formulation was monitored with a thermocouple using a digital thermometer (KM 3013, Kane-May Ltd., UK). Identification of the volatile species in the emulsion was carried out by GC–MS using a Waters GCT Premier instrument. 2.2.4. Rheological analysis Rheological measurements on surfactant solutions and emulsion formulations were carried out on a Malvern CVO and a TA AR-100-N rheometer using both controlled stress and controlled rate modes using a 2° 60 mm cone and plate or 25 mm diameter cup and bob geometry. Typical setting were (i) a stress range of 0.015–4.743 Pa in logarithmic steps (up and down), with a 0.05 acceptance of steady state and maximum total step time of 100 s; (ii) stress range 0.015–1.022 Pa in logarithmic steps (up and down) with a 0.05 acceptance of steady state and maximum total step time of 20 s. Experiments were performed at a fixed temperature of 20 °C. 2.2.5. Target concentrations The amount of anesthetic delivered is described in terms of the minimum alveolar concentration (MAC). One MAC is the alveolar concentration of a volatile anesthetic that produces no movement in 50% of spontaneously-breathing patients during skin incision. The vapor phase concentration equating to one MAC is anesthetic dependent and is 2% v/v for sevoflurane. Between 1 and 4 MAC are required at different stages of anesthesia (induction, maintenance), and the precise MAC value required for equivalent anesthetic effect may vary for different patients/patient types (e.g. neonatal, geriatric, bariatric). Hence values of 0.5, 1, 2, 3 and 4% were set as target deliverable levels of anesthetic. 3. Results and discussion 3.1. Phase behavior of sevoflurane/FSN-100/water mixtures The ternary phase diagram for sevoflurane, water and the fluorinated surfactant FSN-100 (Fig. 2) shows a transparent microemulsion region is observed at up to 15 wt.% sevoflurane, with an
opaque o/w emulsion phase observed at higher anesthetic content. Emulsions containing up to 60 vol.% sevoflurane were obtained that remained stable for >24 h. Higher anesthetic content resulted in formation of a Winsor I system, with sevoflurane separating at the bottom of the samples. This classical phase progression is consistent with early phase studies with similar stabilizers and different fluorinated oils [4]. Characterization of the emulsion phases by microscopy and light scattering indicated the presence of discrete oil droplets, with droplet size distributions typical of nanoemulsion systems (Table 1 and Fig. S1). 3.2. Release profiles of sevoflurane nanoemulsions Preliminary release studies using the custom flow rig indicated that dispersion stability was important for stable anesthetic release (Fig. S2). During evaporation a viscous surface layer of concentrated polymeric surfactant forms and detected anesthetic levels drop rapidly to zero (i.e. below the detection limit of 0.1% v/v). Water evaporation from concentrated o/w emulsions is known to concentrate and compress the oil droplets [17], and drying from emulsions is widely used in the coatings industry for film forming applications. The presence of concentrated surface mesophases has also been shown to retard evaporation [18]. The drop in sevoflurane release observed is, therefore, understandable. Whilst further investigation of the viscous surface structure is of great interest, its formation can be prevented by continuous stirring of the emulsion, allowing clinically useful release rates to be obtained, and good control of anesthetic release achieved. 3.3. Effect of stirring rate on the released anesthetic concentration Fig. 3 shows the release profiles from stirred (fixed rpm) o/w emulsions, tailored to provide clinically useful anesthetic concentrations, with release maintained over timeframes appropriate for typical clinical procedures. Despite the extremely simple dispersion process used, reproducibility in performance is excellent (Fig. S3). Using increased amounts of formulation extends the duration of release but does not affect release level. In addition, an increase in carrier gas flow rate from 1 to 4 L min1 can render a given formulation suitable for use in sedation, rather than anesthesia (Fig. 3(C)), demonstrating the potential of these formulations for multiple purposes. Gradual stirring rate adjustment can be used to maintain constant anesthetic levels within the regulatory requirement (ISO 5358) of ±20% of the indicated value (Fig. 4(A)). Clinically, however, it is essential to vary and control the level of anesthetic delivered during a procedure to account for patient response to surgical stimuli. In the traditional anesthetic vaporizer this is achieved by changing gas flow. Here, changing the stirring rate, (Fig. 4(B)) allows the desirable property of changes in release independent of gas flow. Thus, the release profile targeted in aim (ii) can be obtained [19]. In fact, the range of concentrations accessible with changing stirring rate are such that a single formulation could be used to provide constant delivery of anesthetic at different levels from 0.5 to 4% v/v [17,19]. 3.4. Effect of temperature on the released anesthetic concentrations
Fig. 2. Partial phase diagram (wt.%) of Zonyl FSN-100/sevoflurane/water system at 20 °C. Labels denote o/w microemulsion (A), stable o/w emulsion (B) and unstable o/w emulsion (Winsor I) (C).
Here temperature has been fixed to a constant and thermostatted 20 °C. Ultimately, and particularly for in-the-field applications, it will be necessary to compensate for environmental temperature induced effects on the anesthetic release. As shown in Fig. 5, at constant stirring rate there is a difference in the resulting anesthetic concentration in the carrier gas at different temperatures. However, as Fig. 5 also shows, variation of the stirring rate can normalize the release to a chosen value.
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
81
10
Sevoflurane / vol.%
A
1
0.1 0
10
20
30
40
50
60
me / min 10
Sevoflurane / vol.%
B
1
0.1 0
10
20
30
40
50
60
me / min Fig. 4. Effect of stirring rate on sevoflurane release from formulation containing 15 mL sevoflurane and 10 L ml of an aqueous solution of 7 wt.% Zonyl FSN-100 under carrier gas flow rate of 1 L min1. Panel (A): output maintained is 2 ± 0.2%. Panel (B) stirring rate varied to produce a range of outputs from a single formulation.
10
Sevoflurane / vol.%
C
1
0.1 0
10
20
30
40
50
60
me / min Fig. 3. Sevoflurane release obtained from the formulations in Table 1. In all panels, sevoflurane concentration in carrier gas 4% (squares); 3% (triangles); 2% (circles); 1% (diamonds); 0.5% (inverted triangles). Panel (A) at 1 L min1 gas flow, panel (B) at 4 L min1 gas flow, panel (C) shows the effect of gas flow rate for the 2 % sample from (A) run at 1 L min1 (circles) ; 4 L min1 (inverted triangles).
modeling of evaporation from systems under flow presents significant challenges, due to unknown parameters of interest for a given system [17,20]. These difficulties notwithstanding, extensive work from groups at Hull University have provided various theoretical descriptions that can assist in understanding the observed behavior. As described by Fletcher, Binks and co-workers, the evaporation rate (E) of a liquid is related to the mass (M) of the evaporating molecule, its vapor diffusion rate (Dv), the surface area available for evaporation (A), the absolute temperature (T) and the height of the stagnant vapor phase (hv) above the liquid, which acts as a retarding barrier to evaporation (Eq. (1)) (P = equilibrium vapor pressure, z relates to the flow of the carrier gas in the stagnant vapor layer) [10,17]. The height of the stagnant layer depends in a complex manner on the flow of gas across the surface, but generally as flow (F) increases hv decreases and evaporation increases (Eq. (1)).
E¼ 3.5. Evaporation rate of sevoflurane nanoemulsions: theoretical consideration The current formulation shows proof of concept, and the design of optimal systems for controlled release of volatiles may be accelerated by understanding the current data in the context of a theoretical framework. It is acknowledged in the literature that
dm MADv Pz ¼ dt hv RT
ð1Þ
At higher flow rates, the contribution from the stagnant vapor phase becomes negligible in comparison to mass transport of the oil to the surface, and evaporation is more closely linked to dispersion structure (Eq. (2)). This relates to the solubility of the oil in the continuous water phase Sw, the diffusion of molecular oil through the continuous phase Dw, and the bulk water barrier
82
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
10 oC 20 oC 40 oC
Sevoflurane / vol.%
10
10
1 1
0.1
0.1
0.01
0.01
0
10
20
30
40
50
60
0
10
20
me / min
30
40
50
60
me / min
MADw Sw hw 1=3
s ¼ 2r=/
ð2Þ
sevoflurane release
2.0 1.8 1.6
count rate
280
In our studies, the influence of the design of the flow rig cannot be entirely decoupled from the effects of the formulation. Although measured anesthetic concentrations are flow rate dependent due to dilution effects, routing of the carrier gas into (rather than across the top of) the sample chamber is expected to remove the stagnant vapor layer above the formulations (i.e. hv becomes negligible). In addition, due the direction of flow towards the formulation surface effective mixing of the anesthetic vapor into the carrier gas stream is achieved (i.e. increased z in Eq. (1)). Therefore, according to the theoretical framework of Binks et al. [12,17], evaporation rate is expected to be dependent on the thickness of the water layer separating the droplets from the air/emulsion interface, hw, which also relates to the spatial arrangement of the droplets via s. In an o/w emulsion where creaming occurs, droplets cluster under gravity forming a dense layer at the top of the emulsion, decreasing s and hw, thereby causing an increase in evaporation due to faster transport to the surface. For o/w emulsions with fluorinated oils the droplets have a higher mass density than water and phase separation occurs via sedimentation, rather than creaming. Hence, the contribution from formation of an upper bulk water layer to increasing hw, outweighs the effect of the reduction in s caused by droplet clustering, and E decreases. It is therefore understandable that without stirring only stable emulsions were effective in providing any measureable release of anesthetic, and that at 20 °C anesthetic evaporation from the emulsions was up to 20 times less than the (clinically dangerous) release levels observed for pure sevoflurane. In the stirred systems however, settling is prevented. This increases s, but decreases hw by ensuring droplet transport to the air/emulsion interface. Evaporation therefore increases. As can be seen from Table 1, formulations designed for use at higher flow rates contain higher amounts of sevoflurane, and correspondingly higher concentrations of surfactant. As droplet sizes show no systematic change, (for the most part these are rather similar at 200–250 lm) it seems that the biggest change is in the number of droplets present (Eq. (3)) giving a decrease in s and therefore hw that is expected to increase E under otherwise
3.0
droplet size
ð3Þ droplet szie/ nm
EhighF ¼
2.2
2.5
260 2.0
240 1.5
220
count rate/ Million cps
layer thickness hw. In addition, hw is related to the spacing of the droplets s, which in turn is related to droplet size (in Eq. (3) r is the droplet radius) and the volume fraction of oil in the dispersion /oil.
Sevoflurane / vol.%
Fig. 5. Minimising the effect of temperature on sevoflurane release profile of a formulation containing 15 mL sevoflurane and 55 mL of an aqueous solution of 9 wt.% Zonyl FSN-100 stirred at different rates under nitrogen flow rate of 1 L min1 using a thermostatted glass flow cell (S.A. = 20 cm2). The formulations were stirred at 400, 350 and 200 rpm at 10, 20 and 40 °C, respectively.
1.0
200 0
20
40
60
0.5
me/ min Fig. 6. Change in droplet size during evaporation of sevoflurane from formulation containing 15 mL sevoflurane dispersed in 105 mL of a 7 wt.% solution of Zonyl FSN100 under 1 L min1 carrier gas flow rate, with constant stirring of 250 rpm.
equivalent conditions [10,17]. An increase in viscosity would decrease mass transport by increasing diffusion time. However, rheological studies indicate that despite the effective increase in polymer concentration, viscosity actually decreases as evaporation occurs (supplemental). Therefore, due to the relatively constant droplet sizes, the experimentally observed increase in anesthetic release (at constant stirring rate) with increasing sevoflurane content in the formulation is as predicted by theory. Studies following droplet size during evaporation at constant stirring rate have shown that both droplet size and measured gas phase sevoflurane concentrations remain steady for a given period of time, after which both are observed to decrease (Fig. 6). Count rate observations during PCS studies are consistent with shrinkage of droplets in the initial ‘steady release’ phase, followed by a reduction in the number of droplets (decreased count rate), upon which a concomitant drop in sevoflurane concentration is observed. It is at this point that stirring rate must be increased if constant gas phase sevoflurane concentrations are to be maintained. The increase in stirring rate required to maintain steady release over time is hence consistent with requiring more mixing to bring the equivalent amount of oil to the surface region. This in turn results from a rapid increase in droplet separation (s) as the number of
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
stable
HC creaming
FC sedimenng
srred
hw
hw
s
s
E
E
83
S depends on φoil as lile variaon In droplet size hw small and constant due to srring
increased srring compensates for decreasing s to maintain small hw
Fig. 7. Schematic illustration of changes in variables in Eqs. (2) and (3) for various emulsion systems.
droplets decreases and an accompanying decrease in droplet volume as / decreases (see Fig. 7). These data and theoretical considerations indicate that the use of stirring compensates for compositional changes in the formulation. Overall the evaporation rate is therefore dependent on Sw, as hw is maintained at a low value by stirring, and all other parameters in Eq. (2) are constant or negligible. This suggests that further control of the evaporation could be obtained by incorporation of an additive to increase solubility of sevoflurane in the continuous phase. However, increasing the complexity of the formulation in this instance has been resisted in an attempt to minimize potential regulatory hurdles for development. It is also acknowledged that the fluorinated surfactant used here is not approved for use in inhalation drug delivery. The formulation described herein serves to provide proof of concept for a dispersion based formulation suitable for delivery of inhalation anesthesia. Returning to the theoretical models, although the nanoemulsions formulated are designed for ultimate use in systems far from the conditions under which the theoretical framework was developed (perturbing gas flows and stirring rates that cause turbulence of the surface), as long as differences in the nature of the phase separation (high density of the fluorinated oil) are taken into account, behavior under lower flow and stirring rates can to some extent be predicted by the equations presented. This allows comparisons between different systems to be made that link emulsion composition and structure to the release characteristics obtained. Hence a framework has been evaluated within which new formulations can be developed and tailored for a specific release performance. Ultimately, the target is a formulation that provides constant release over the required timescale without the need for adjustments to the stirring rate. 4. Conclusions Novel anesthetic-in-water emulsions have been formulated that allow controlled release of a volatile anesthetic gas under clinically relevant gas flow conditions. This work represents a conceptually novel application of colloid science in a medical field which has seen relatively little chemistry-based innovation in recent years. Experimental studies [this paper,19,21] indicate a strong link between formulation microstructure and anesthetic release under flow and shear. Existing theoretical frameworks for evaporation from emulsions [10,12,17,20] provide a reasonable account of the observed behavior, but this system highlights the challenge of fully understanding evaporation of volatile components under air flow and stirred conditions [17]. Future development of the dispersion based anesthesia concept would be greatly facilitated by an enhanced theoretical understanding of behavior in these unusual systems.
Evaporation of the anesthetic from the formulation is highly sensitive to agitation, allowing stirring to be used to control and optimize the anesthetic release profile. This proof of concept study shows that simply produced emulsions, such as those described, could be used as the basis for safe, controllable delivery of anesthetic gas via a cheap, portable device that could be much simpler to use than conventional anesthesia systems. Acknowledgments This research was funded by Cardiff University via Early Stage Development funding, and via the Cardiff Partnership Fund. Cardiff University support is also acknowledged for funding for patent filing (WO 2013/041850 A2). Welsh Government research funding under the A4B scheme is also recognised. Rheological experiments were carried out by Formumetrics Ltd., UK (Bristol). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2014.10.021. References [1] S. Bertilla, P. Marie, M. Krafft, Semifluorinated Alkanes as Stabilizing Agents of Fluorocarbon Emulsions, in: K. Kobayashi, E. Tsuchida, H. Horinouchi (Eds.), Artificial Oxygen Carrier, vol. 12, Springer, Tokyo, 2005, pp. 237–251. [2] M. Krafft, Curr. Opin. Colloid Interface Sci. 8 (2003) 251–258. [3] P. LoNostro, S.-M. Choi, C.-Y. Ku, S.-H. Chen, J. Phys. Chem. B. 103 (1999) 5347– 5352. [4] G. Mathis, P. Leempoel, J.C. Ravey, C. Selve, J.J. Delpuech, J. Am. Chem. Soc. 106 (1984) 6162–6171. [5] R. Bleta, J.L. Blin, M.J. Stébé, J. Phys. Chem. B. 110 (2006) 23547–23556. [6] J.L. Blin, R. Bleta, M.J. Stebe, J. Colloid Interface Sci. 300 (2006) 765–773. [7] J.L. Blin, P. Lesieur, M.J. Stébé, Langmuir 20 (2004) 491–498. [8] J.L. Blin, M.J. Stébé, J. Phys. Chem. B. 108 (2004) 11399–11405. [9] F. Michaux, J.L. Blin, M.J. Stébé, J. Phys. Chem. B. 112 (2008) 11950–11959. [10] I. Aranberri, K.J. Beverley, B.P. Binks, J.H. Clint, P.D.I. Fletcher, Langmuir 18 (2002) 3471–3475. [11] I. Aranberri, B.P. Binks, J.H. Clint, Chem. Commun. (2003) 2538. [12] I. Aranberri, B.P. Binks, J.H. Clint, P.D.I. Fletcher, Langmuir 20 (2004) 2069– 2074. [13] A. Al-Bawab, F. Odeh, A. Bozeya, P.A. Aikens, S.E. Friberg, Flav. Fragr. J. 24 (2009) 155–159. [14] S.E. Friberg, J. Phys. Chem. B. 113 (2009) 3894–3900. [15] O. Santos, M.F. Camargo, K. Boock, M. Bergamaschi, P. Rocha Filho, J. Dispersion Sci. Technol. 30 (2009) 394–398. [16] J.P. Jee, M.C. Parlato, M.G. Perkins, S. Mecozzi, R.A. Pearce, Anesthesiology 116 (2012) 580–585. [17] B.P. Binks, P.D. Fletcher, B.L. Holt, P. Beaussoubre, K. Wong, Langmuir 26 (2010) 18024–18030. [18] K.J. Beverley, J.H. Clint, P.D.I. Fletcher, PCCP 2 (2000) 4173–4177. [19] J.E. Hall, A.R. Wilkes, A. Paul, 2013. WO2013041850; GB2012/052302. [20] K.J. Beverley, J.H. Clint, P.D.I. Fletcher, PCCP 1 (1999) 149–153. [21] A. Paul, A. Wilkes, I. Salama, N. Goodwin, J. Hall, Br. J. Anaesth. 112 (1) (2014) 189–190P.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Volatile fluorinated nanoemulsions: A chemical route to controlled delivery of inhalation Anesthesia Ibrahim E. Salama a, Claire L. Jenkins a, Alun Davies a, Jeffrey N. Clark a,b, Antony R. Wilkes b, Judith E. Hall b, Alison Paul a,⇑ a b
School of Chemistry, Cardiff University, Main Building, Park Place, Cardiff, UK Department of Anesthetics, Intensive Care and Pain Medicine, School of Medicine, Cardiff University, Cardiff, UK
a r t i c l e
i n f o
Article history: Received 11 June 2014 Accepted 16 October 2014 Available online 29 October 2014 Keywords: Anesthesia Drug-delivery Nanoemulsions Fluorocarbon Nanotechnology
a b s t r a c t Novel dispersions of the volatile inhalation anesthetic sevoflurane have been formulated that can provide controlled, sustainable release of anesthetic over clinically useful timescales. The emulsions can be simply formed with manual shaking, reproducibly yielding droplets of the order of 250 nm diameter, i.e. within the nanoemulsion range. Using a custom flow-rig, release of anesthetic gas from the emulsion has been evaluated, and clinically useful levels achieved through appropriate stirring of the formulation. Stirring can also be used to temporarily increase or decrease the amount of anesthetic released. Once consideration of the unusual nature of the fluorinated systems (phase separation by sedimentation rather than creaming), and the highly perturbed environment of their evaluation (under stirring and flow of gas), the observed behavior regarding sevoflurane evaporation can be reasonably well explained by existing theoretical models. Links between anesthetic release and emulsion structure have been defined, providing the basis for future development. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Nonionic partially fluorinated surfactants have been used to stabilize emulsions and microemulsions of fluorinated oils in water, e.g. for formation of blood substitutes [1–4] or as templating media for the preparation of porous materials [5–9]. For volatile emulsions previous fundamental studies have shown that oil evaporation depends both on the type of stabilizer used and on the solubility of the oil in the continuous water phase [10–12]: evaporation rates of water soluble oils approach that of the bulk oil, but evaporation of very low solubility oils is significantly hindered, an effect magnified by the use of a polymeric (as opposed to low molecular weight surfactant) stabilizer [10–12]. Other groups have focused on relationships between phase structures and the evaporation pathway, primarily for fragrance emulsions [13–15]. Here we utilize a volatile anesthetic as the evaporating component of an oil-in-water (o/w) emulsion. This is used to evaluate the feasibility of using a colloid science based approach to inhalation delivery of an anesthetic (as opposed to intravenous delivery [16]).
⇑ Corresponding author at: School of Chemistry, Main Building, Park Place, Cardiff CF10 3AT, UK. E-mail address: [email protected] (A. Paul). http://dx.doi.org/10.1016/j.jcis.2014.10.021 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
Currently, safe and accurate anesthesia delivery is reliant on large, complex and expensive machines operated by very specialized, highly trained personnel. All of these factors limit their use outside of modern clinical settings, and a simpler, safer and cheaper delivery system could provide an alternative means of anesthesia delivery. This is particularly relevant for challenging situations including pre-hospital medicine, in-the-field anesthesia, warfare anesthesia and in the challenging conditions often encountered in developing countries. Ethically, the delivery of anesthesia and analgesia are mandatory for surgical procedures worldwide, but in developing countries access to safe anesthesia can be limited due to a shortage of trained staff, and compromised by failure of complex equipment, leading to reduced access to healthcare. The aim of the work is therefore to provide proof of concept for a dispersion based anesthetic delivery device. Anesthetic is released in vapor form from a formulation, and picked up by a carrier flowing through the device and to the patient (Fig. 1). Specific targets are to demonstrate (i) release of clinically safe and useful concentrations of anesthetic (0.5–4 vol.% in the carrier gas stream). This is evaluated over a 60 min timescale at fixed gas flows of 1 L min1 and 4 L min1, chosen to exemplify typical clinically used flow rates; (ii) to obtain higher release (up to 8 vol.%) for shorter periods to create a release profile suitable for both induction and maintenance of anesthesia, and (iii) to provide a means of adjusting anesthetic release in a controllable and responsive
79
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
manner, without changing the carrier-gas flow used. This controlled variability is essential for the potential clinical use of the device and formulation.
Table 1 Formulation composition and oil droplet sizes for different performance sevoflurane containing emulsions. Carrier gas flow rate/L min1
Release level/ vol%
Surfactanta concentration/ wt.% (aq)
vol% anesthetic in formulation
Average droplet diameter/ nm
1 1 1 1 1 4 4 4 4
4b 3 2 1 0.5 4 3 2 0.5
18 10 7 4 8 25 22 17 7
31 19 13 6 6 44 36 29 13
209 259 261 239 188 256 206 384 188
2. Materials and methods 2.1. Materials Sevoflurane (99.98%, Abbot, UK) and Zonyl FSN-100 (technical grade, ABCR, Germany) were both used as received. Industrial nitrogen gas (BOC, UK) was used as the carrier gas throughout. Water was reagent-grade produced by a RiOs 5 purification system (Millipore, USA).
(±2) (±0.6) (±4) (±5) (±4) (±5) (±2) (±5) (±4)
a
2.2. Methods
Zonyl FSN-100. 4 vol% equates to 2 maximum alveolar concentration (MAC). One MAC is the alveolar concentration of a volatile anesthetic that produces no movement in 50% of spontaneously-breathing patients during skin incision. b
2.2.1. Sample preparation All formulations were prepared on 150 mL scale in 500 mL DURAN laboratory bottle with a screw cap and equilibrated at 20 °C prior to testing. Formulation compositions are given in Table 1. Emulsions were prepared by vigorously shaking (by hand) a known volume of the anesthetic with a pre-prepared aqueous surfactant solution, for a fixed time of 60 s. Method validation experiments were performed using different shaking times, and independent preparation by different personnel to ensure robustness of the preparation procedure. This simple method of dispersion greatly simplifies the formulation stability requirements, as the emulsion itself is formed at point of use. 2.2.2. Characterization Ternary phase diagrams were constructed by visual inspection. The continuous phase of the emulsion was determined by a standard drop-test method (an added drop of the continuous phase will disperse freely into the bulk; a drop of the dispersed phase will separate at the top or bottom of the sample, depending on density difference). Emulsion droplets were characterized by light-microscope imaging using an Olympus BX50 system microscope (Olympus, UK) fitted with JVC TK-C1380 color video camera (JVC, Japan) and analyzed using Image J software (Fiji, USA). Additional measurements were obtained from dynamic light scattering analysis using a
Brookhaven ZetaPlus (Brookhaven Instruments Ltd., USA). For light scattering measurements the emulsions were diluted by a factor of 20–50 depending on the emulsion concentration. 2.2.3. Release testing The controlled release of anesthetic agents from the formulations was evaluated using the experimental flow-rig shown in Fig. 1. The upper module consists of a 250 mL cylindrical plastic container (which holds the formulation and a magnetic stirrer bar) attached to a custom built Teflon fitting which has an inlet port for entrance of the free carrier gas and an outlet port for the exit of the anesthetic-loaded carrier gas. The module was placed on a standard magnetic stirrer (Heidolph MR 3002, Germany) for controlled stirring. Nitrogen gas was passed through the flow module at a controlled flow rate, typically 1 L min1 and 4 L min1. In lab experiments nitrogen was used as the carrier gas for convenience, although control experiments using air and air/O2 mixtures showed that the performance was independent of gas composition. The concentration of the anesthetic agent in the outlet stream was measured with a standard anesthetic monitor (Capnomac Ultima, Datex Instrumentarium Inc., Helsinki, Finland). The formulation
carrier gas in anaesthec monitor
sampling port
to paent carrier gas and anaesthec gas out
(not to scale) Fig. 1. Schematic of test flow-rig. SA = 60 cm2. Typical stirring rates 100–800 rpm. Typical formulation volume 120 cm3.
80
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
was thermostatted using a double-walled glass water-jacket connected to a circulating water bath to maintain temperatures other than 20 °C. The temperature of the formulation was monitored with a thermocouple using a digital thermometer (KM 3013, Kane-May Ltd., UK). Identification of the volatile species in the emulsion was carried out by GC–MS using a Waters GCT Premier instrument. 2.2.4. Rheological analysis Rheological measurements on surfactant solutions and emulsion formulations were carried out on a Malvern CVO and a TA AR-100-N rheometer using both controlled stress and controlled rate modes using a 2° 60 mm cone and plate or 25 mm diameter cup and bob geometry. Typical setting were (i) a stress range of 0.015–4.743 Pa in logarithmic steps (up and down), with a 0.05 acceptance of steady state and maximum total step time of 100 s; (ii) stress range 0.015–1.022 Pa in logarithmic steps (up and down) with a 0.05 acceptance of steady state and maximum total step time of 20 s. Experiments were performed at a fixed temperature of 20 °C. 2.2.5. Target concentrations The amount of anesthetic delivered is described in terms of the minimum alveolar concentration (MAC). One MAC is the alveolar concentration of a volatile anesthetic that produces no movement in 50% of spontaneously-breathing patients during skin incision. The vapor phase concentration equating to one MAC is anesthetic dependent and is 2% v/v for sevoflurane. Between 1 and 4 MAC are required at different stages of anesthesia (induction, maintenance), and the precise MAC value required for equivalent anesthetic effect may vary for different patients/patient types (e.g. neonatal, geriatric, bariatric). Hence values of 0.5, 1, 2, 3 and 4% were set as target deliverable levels of anesthetic. 3. Results and discussion 3.1. Phase behavior of sevoflurane/FSN-100/water mixtures The ternary phase diagram for sevoflurane, water and the fluorinated surfactant FSN-100 (Fig. 2) shows a transparent microemulsion region is observed at up to 15 wt.% sevoflurane, with an
opaque o/w emulsion phase observed at higher anesthetic content. Emulsions containing up to 60 vol.% sevoflurane were obtained that remained stable for >24 h. Higher anesthetic content resulted in formation of a Winsor I system, with sevoflurane separating at the bottom of the samples. This classical phase progression is consistent with early phase studies with similar stabilizers and different fluorinated oils [4]. Characterization of the emulsion phases by microscopy and light scattering indicated the presence of discrete oil droplets, with droplet size distributions typical of nanoemulsion systems (Table 1 and Fig. S1). 3.2. Release profiles of sevoflurane nanoemulsions Preliminary release studies using the custom flow rig indicated that dispersion stability was important for stable anesthetic release (Fig. S2). During evaporation a viscous surface layer of concentrated polymeric surfactant forms and detected anesthetic levels drop rapidly to zero (i.e. below the detection limit of 0.1% v/v). Water evaporation from concentrated o/w emulsions is known to concentrate and compress the oil droplets [17], and drying from emulsions is widely used in the coatings industry for film forming applications. The presence of concentrated surface mesophases has also been shown to retard evaporation [18]. The drop in sevoflurane release observed is, therefore, understandable. Whilst further investigation of the viscous surface structure is of great interest, its formation can be prevented by continuous stirring of the emulsion, allowing clinically useful release rates to be obtained, and good control of anesthetic release achieved. 3.3. Effect of stirring rate on the released anesthetic concentration Fig. 3 shows the release profiles from stirred (fixed rpm) o/w emulsions, tailored to provide clinically useful anesthetic concentrations, with release maintained over timeframes appropriate for typical clinical procedures. Despite the extremely simple dispersion process used, reproducibility in performance is excellent (Fig. S3). Using increased amounts of formulation extends the duration of release but does not affect release level. In addition, an increase in carrier gas flow rate from 1 to 4 L min1 can render a given formulation suitable for use in sedation, rather than anesthesia (Fig. 3(C)), demonstrating the potential of these formulations for multiple purposes. Gradual stirring rate adjustment can be used to maintain constant anesthetic levels within the regulatory requirement (ISO 5358) of ±20% of the indicated value (Fig. 4(A)). Clinically, however, it is essential to vary and control the level of anesthetic delivered during a procedure to account for patient response to surgical stimuli. In the traditional anesthetic vaporizer this is achieved by changing gas flow. Here, changing the stirring rate, (Fig. 4(B)) allows the desirable property of changes in release independent of gas flow. Thus, the release profile targeted in aim (ii) can be obtained [19]. In fact, the range of concentrations accessible with changing stirring rate are such that a single formulation could be used to provide constant delivery of anesthetic at different levels from 0.5 to 4% v/v [17,19]. 3.4. Effect of temperature on the released anesthetic concentrations
Fig. 2. Partial phase diagram (wt.%) of Zonyl FSN-100/sevoflurane/water system at 20 °C. Labels denote o/w microemulsion (A), stable o/w emulsion (B) and unstable o/w emulsion (Winsor I) (C).
Here temperature has been fixed to a constant and thermostatted 20 °C. Ultimately, and particularly for in-the-field applications, it will be necessary to compensate for environmental temperature induced effects on the anesthetic release. As shown in Fig. 5, at constant stirring rate there is a difference in the resulting anesthetic concentration in the carrier gas at different temperatures. However, as Fig. 5 also shows, variation of the stirring rate can normalize the release to a chosen value.
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
81
10
Sevoflurane / vol.%
A
1
0.1 0
10
20
30
40
50
60
me / min 10
Sevoflurane / vol.%
B
1
0.1 0
10
20
30
40
50
60
me / min Fig. 4. Effect of stirring rate on sevoflurane release from formulation containing 15 mL sevoflurane and 10 L ml of an aqueous solution of 7 wt.% Zonyl FSN-100 under carrier gas flow rate of 1 L min1. Panel (A): output maintained is 2 ± 0.2%. Panel (B) stirring rate varied to produce a range of outputs from a single formulation.
10
Sevoflurane / vol.%
C
1
0.1 0
10
20
30
40
50
60
me / min Fig. 3. Sevoflurane release obtained from the formulations in Table 1. In all panels, sevoflurane concentration in carrier gas 4% (squares); 3% (triangles); 2% (circles); 1% (diamonds); 0.5% (inverted triangles). Panel (A) at 1 L min1 gas flow, panel (B) at 4 L min1 gas flow, panel (C) shows the effect of gas flow rate for the 2 % sample from (A) run at 1 L min1 (circles) ; 4 L min1 (inverted triangles).
modeling of evaporation from systems under flow presents significant challenges, due to unknown parameters of interest for a given system [17,20]. These difficulties notwithstanding, extensive work from groups at Hull University have provided various theoretical descriptions that can assist in understanding the observed behavior. As described by Fletcher, Binks and co-workers, the evaporation rate (E) of a liquid is related to the mass (M) of the evaporating molecule, its vapor diffusion rate (Dv), the surface area available for evaporation (A), the absolute temperature (T) and the height of the stagnant vapor phase (hv) above the liquid, which acts as a retarding barrier to evaporation (Eq. (1)) (P = equilibrium vapor pressure, z relates to the flow of the carrier gas in the stagnant vapor layer) [10,17]. The height of the stagnant layer depends in a complex manner on the flow of gas across the surface, but generally as flow (F) increases hv decreases and evaporation increases (Eq. (1)).
E¼ 3.5. Evaporation rate of sevoflurane nanoemulsions: theoretical consideration The current formulation shows proof of concept, and the design of optimal systems for controlled release of volatiles may be accelerated by understanding the current data in the context of a theoretical framework. It is acknowledged in the literature that
dm MADv Pz ¼ dt hv RT
ð1Þ
At higher flow rates, the contribution from the stagnant vapor phase becomes negligible in comparison to mass transport of the oil to the surface, and evaporation is more closely linked to dispersion structure (Eq. (2)). This relates to the solubility of the oil in the continuous water phase Sw, the diffusion of molecular oil through the continuous phase Dw, and the bulk water barrier
82
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
10 oC 20 oC 40 oC
Sevoflurane / vol.%
10
10
1 1
0.1
0.1
0.01
0.01
0
10
20
30
40
50
60
0
10
20
me / min
30
40
50
60
me / min
MADw Sw hw 1=3
s ¼ 2r=/
ð2Þ
sevoflurane release
2.0 1.8 1.6
count rate
280
In our studies, the influence of the design of the flow rig cannot be entirely decoupled from the effects of the formulation. Although measured anesthetic concentrations are flow rate dependent due to dilution effects, routing of the carrier gas into (rather than across the top of) the sample chamber is expected to remove the stagnant vapor layer above the formulations (i.e. hv becomes negligible). In addition, due the direction of flow towards the formulation surface effective mixing of the anesthetic vapor into the carrier gas stream is achieved (i.e. increased z in Eq. (1)). Therefore, according to the theoretical framework of Binks et al. [12,17], evaporation rate is expected to be dependent on the thickness of the water layer separating the droplets from the air/emulsion interface, hw, which also relates to the spatial arrangement of the droplets via s. In an o/w emulsion where creaming occurs, droplets cluster under gravity forming a dense layer at the top of the emulsion, decreasing s and hw, thereby causing an increase in evaporation due to faster transport to the surface. For o/w emulsions with fluorinated oils the droplets have a higher mass density than water and phase separation occurs via sedimentation, rather than creaming. Hence, the contribution from formation of an upper bulk water layer to increasing hw, outweighs the effect of the reduction in s caused by droplet clustering, and E decreases. It is therefore understandable that without stirring only stable emulsions were effective in providing any measureable release of anesthetic, and that at 20 °C anesthetic evaporation from the emulsions was up to 20 times less than the (clinically dangerous) release levels observed for pure sevoflurane. In the stirred systems however, settling is prevented. This increases s, but decreases hw by ensuring droplet transport to the air/emulsion interface. Evaporation therefore increases. As can be seen from Table 1, formulations designed for use at higher flow rates contain higher amounts of sevoflurane, and correspondingly higher concentrations of surfactant. As droplet sizes show no systematic change, (for the most part these are rather similar at 200–250 lm) it seems that the biggest change is in the number of droplets present (Eq. (3)) giving a decrease in s and therefore hw that is expected to increase E under otherwise
3.0
droplet size
ð3Þ droplet szie/ nm
EhighF ¼
2.2
2.5
260 2.0
240 1.5
220
count rate/ Million cps
layer thickness hw. In addition, hw is related to the spacing of the droplets s, which in turn is related to droplet size (in Eq. (3) r is the droplet radius) and the volume fraction of oil in the dispersion /oil.
Sevoflurane / vol.%
Fig. 5. Minimising the effect of temperature on sevoflurane release profile of a formulation containing 15 mL sevoflurane and 55 mL of an aqueous solution of 9 wt.% Zonyl FSN-100 stirred at different rates under nitrogen flow rate of 1 L min1 using a thermostatted glass flow cell (S.A. = 20 cm2). The formulations were stirred at 400, 350 and 200 rpm at 10, 20 and 40 °C, respectively.
1.0
200 0
20
40
60
0.5
me/ min Fig. 6. Change in droplet size during evaporation of sevoflurane from formulation containing 15 mL sevoflurane dispersed in 105 mL of a 7 wt.% solution of Zonyl FSN100 under 1 L min1 carrier gas flow rate, with constant stirring of 250 rpm.
equivalent conditions [10,17]. An increase in viscosity would decrease mass transport by increasing diffusion time. However, rheological studies indicate that despite the effective increase in polymer concentration, viscosity actually decreases as evaporation occurs (supplemental). Therefore, due to the relatively constant droplet sizes, the experimentally observed increase in anesthetic release (at constant stirring rate) with increasing sevoflurane content in the formulation is as predicted by theory. Studies following droplet size during evaporation at constant stirring rate have shown that both droplet size and measured gas phase sevoflurane concentrations remain steady for a given period of time, after which both are observed to decrease (Fig. 6). Count rate observations during PCS studies are consistent with shrinkage of droplets in the initial ‘steady release’ phase, followed by a reduction in the number of droplets (decreased count rate), upon which a concomitant drop in sevoflurane concentration is observed. It is at this point that stirring rate must be increased if constant gas phase sevoflurane concentrations are to be maintained. The increase in stirring rate required to maintain steady release over time is hence consistent with requiring more mixing to bring the equivalent amount of oil to the surface region. This in turn results from a rapid increase in droplet separation (s) as the number of
I.E. Salama et al. / Journal of Colloid and Interface Science 440 (2015) 78–83
stable
HC creaming
FC sedimenng
srred
hw
hw
s
s
E
E
83
S depends on φoil as lile variaon In droplet size hw small and constant due to srring
increased srring compensates for decreasing s to maintain small hw
Fig. 7. Schematic illustration of changes in variables in Eqs. (2) and (3) for various emulsion systems.
droplets decreases and an accompanying decrease in droplet volume as / decreases (see Fig. 7). These data and theoretical considerations indicate that the use of stirring compensates for compositional changes in the formulation. Overall the evaporation rate is therefore dependent on Sw, as hw is maintained at a low value by stirring, and all other parameters in Eq. (2) are constant or negligible. This suggests that further control of the evaporation could be obtained by incorporation of an additive to increase solubility of sevoflurane in the continuous phase. However, increasing the complexity of the formulation in this instance has been resisted in an attempt to minimize potential regulatory hurdles for development. It is also acknowledged that the fluorinated surfactant used here is not approved for use in inhalation drug delivery. The formulation described herein serves to provide proof of concept for a dispersion based formulation suitable for delivery of inhalation anesthesia. Returning to the theoretical models, although the nanoemulsions formulated are designed for ultimate use in systems far from the conditions under which the theoretical framework was developed (perturbing gas flows and stirring rates that cause turbulence of the surface), as long as differences in the nature of the phase separation (high density of the fluorinated oil) are taken into account, behavior under lower flow and stirring rates can to some extent be predicted by the equations presented. This allows comparisons between different systems to be made that link emulsion composition and structure to the release characteristics obtained. Hence a framework has been evaluated within which new formulations can be developed and tailored for a specific release performance. Ultimately, the target is a formulation that provides constant release over the required timescale without the need for adjustments to the stirring rate. 4. Conclusions Novel anesthetic-in-water emulsions have been formulated that allow controlled release of a volatile anesthetic gas under clinically relevant gas flow conditions. This work represents a conceptually novel application of colloid science in a medical field which has seen relatively little chemistry-based innovation in recent years. Experimental studies [this paper,19,21] indicate a strong link between formulation microstructure and anesthetic release under flow and shear. Existing theoretical frameworks for evaporation from emulsions [10,12,17,20] provide a reasonable account of the observed behavior, but this system highlights the challenge of fully understanding evaporation of volatile components under air flow and stirred conditions [17]. Future development of the dispersion based anesthesia concept would be greatly facilitated by an enhanced theoretical understanding of behavior in these unusual systems.
Evaporation of the anesthetic from the formulation is highly sensitive to agitation, allowing stirring to be used to control and optimize the anesthetic release profile. This proof of concept study shows that simply produced emulsions, such as those described, could be used as the basis for safe, controllable delivery of anesthetic gas via a cheap, portable device that could be much simpler to use than conventional anesthesia systems. Acknowledgments This research was funded by Cardiff University via Early Stage Development funding, and via the Cardiff Partnership Fund. Cardiff University support is also acknowledged for funding for patent filing (WO 2013/041850 A2). Welsh Government research funding under the A4B scheme is also recognised. Rheological experiments were carried out by Formumetrics Ltd., UK (Bristol). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2014.10.021. References [1] S. Bertilla, P. Marie, M. Krafft, Semifluorinated Alkanes as Stabilizing Agents of Fluorocarbon Emulsions, in: K. Kobayashi, E. Tsuchida, H. Horinouchi (Eds.), Artificial Oxygen Carrier, vol. 12, Springer, Tokyo, 2005, pp. 237–251. [2] M. Krafft, Curr. Opin. Colloid Interface Sci. 8 (2003) 251–258. [3] P. LoNostro, S.-M. Choi, C.-Y. Ku, S.-H. Chen, J. Phys. Chem. B. 103 (1999) 5347– 5352. [4] G. Mathis, P. Leempoel, J.C. Ravey, C. Selve, J.J. Delpuech, J. Am. Chem. Soc. 106 (1984) 6162–6171. [5] R. Bleta, J.L. Blin, M.J. Stébé, J. Phys. Chem. B. 110 (2006) 23547–23556. [6] J.L. Blin, R. Bleta, M.J. Stebe, J. Colloid Interface Sci. 300 (2006) 765–773. [7] J.L. Blin, P. Lesieur, M.J. Stébé, Langmuir 20 (2004) 491–498. [8] J.L. Blin, M.J. Stébé, J. Phys. Chem. B. 108 (2004) 11399–11405. [9] F. Michaux, J.L. Blin, M.J. Stébé, J. Phys. Chem. B. 112 (2008) 11950–11959. [10] I. Aranberri, K.J. Beverley, B.P. Binks, J.H. Clint, P.D.I. Fletcher, Langmuir 18 (2002) 3471–3475. [11] I. Aranberri, B.P. Binks, J.H. Clint, Chem. Commun. (2003) 2538. [12] I. Aranberri, B.P. Binks, J.H. Clint, P.D.I. Fletcher, Langmuir 20 (2004) 2069– 2074. [13] A. Al-Bawab, F. Odeh, A. Bozeya, P.A. Aikens, S.E. Friberg, Flav. Fragr. J. 24 (2009) 155–159. [14] S.E. Friberg, J. Phys. Chem. B. 113 (2009) 3894–3900. [15] O. Santos, M.F. Camargo, K. Boock, M. Bergamaschi, P. Rocha Filho, J. Dispersion Sci. Technol. 30 (2009) 394–398. [16] J.P. Jee, M.C. Parlato, M.G. Perkins, S. Mecozzi, R.A. Pearce, Anesthesiology 116 (2012) 580–585. [17] B.P. Binks, P.D. Fletcher, B.L. Holt, P. Beaussoubre, K. Wong, Langmuir 26 (2010) 18024–18030. [18] K.J. Beverley, J.H. Clint, P.D.I. Fletcher, PCCP 2 (2000) 4173–4177. [19] J.E. Hall, A.R. Wilkes, A. Paul, 2013. WO2013041850; GB2012/052302. [20] K.J. Beverley, J.H. Clint, P.D.I. Fletcher, PCCP 1 (1999) 149–153. [21] A. Paul, A. Wilkes, I. Salama, N. Goodwin, J. Hall, Br. J. Anaesth. 112 (1) (2014) 189–190P.
