WO2012143733A1

WO2012143733A1 - Nanoparticles

Info

Publication number: WO2012143733A1
Application number: PCT/GB2012/050884
Authority: WO
Inventors: Marco Norman GIARDIELLO; Thomas Oliver MCDONALD; Andrew Owen; Steven Paul Rannard
Original assignee: The University Of Liverpool
Priority date: 2011-04-21
Filing date: 2012-04-20
Publication date: 2012-10-26
Also published as: GB201106825D0

Abstract

The present invention relates to a porous composite material comprising a hydrophilic carrier carrying a magnetic material. The porous composite material can include a hydrophobic polymer, hydrophobic material, organic compound, and/or pharmaceutically active ingredient. The porous composite material is prepared by a modified emulsion templating and freeze-drying procedure. Magnetic nanoparticles are released from the porous composite material on application of water.

Description

NANOPARTICIES

The present invention relates to magnetic nanoparticies, and in particular to magnetic nanoparticies prepared by a modified emulsion templating and freeze-drying procedure.

Emulsion templating Is a procedure whereby the structure of a solid material is derived from the structure of a liquid emulsion. It has been used widely to prepare porous materials for a large variety of applications,

In an emulsion, a mixture of immiscible liquids takes the form of a continuous (non-dropiet) phase around an internal (droplet) phase, In emulsion templating, this structure is at least partially conserved and solidified, for example by polymerization or curing/crosslinking of reactive monomers in one or both of the phases or materia!(s) within the phases and/or by removal of the phase solvents. Thus, porous insoluble polymeric materials may be prepared. The polymeric materials are generally rendered insoluble through the use a crosslinking agent and the subsequent formation of a polymeric interpenetrating network that prevents dissolution.

Insoluble porous polymeric materials prepared by emulsion templating have numerous uses, for example as supports, carriers, immobilizers or absorbers. In one of numerous examples, emulsion templated supports can be used to facilitate clathrate formation and dissociation due to the porous and interconnected nature of the material, as disclosed in WO 2009/068912, As well as being useful in their own right, emulsion-tempiated materials can also be used as intermediates en-route to other useful materials.

WO 2004/01 1537 discloses porous water-soluble polymers and the preparation of these using emulsion tempiating methods. This document discloses several examples of oil-in-water emulsions [comprising a continuous (non-drop!et) aqueous phase and an internal (droplet) oil phase]. The continuous aqueous phase contains water and a hydrophi!ic polymeric material, for example poly (vinyl alcohol) (PVA), poly (ethylene glycol) (PEG) or poly (vinyl pyrrolidone) (PVP). The internal (droplet) phase contains an organic solvent, for example dichloromethane or cyclohexane. The oil-in-water emulsions of WO 2004/01 1537 are typically frozen to provide frozen droplets which are then freeze-dried to remove the water and organic solvent. This results in dry, solid beads comprising water-soluble polymeric material. No polymerisation, crosslinking or other reactive chemistry is used as the continuous phase comprises a solution containing a pre-formed polymeric material. The absence of crosslinking or reaction during the process renders the materia! capable of dissolving when water is added. In some examples of WO 2004/011537, other components are also present. The hydrophilic porous beads dissolve rapidly, thereby rapidly releasing additives entrapped within the beads. The additives disclosed in WO 2004/011537 include a water-soluble dye added to the aqueous phase, an oil-soluble dye added to the organic phase, and a zinc nitrate aqueous solution. This document also discloses the preparation of gold nanoparticfe / PEG composite beads by dissolving gold nanoparticles in water, adding PEG, and then freezing and freeze-drying the composite. As disclosed in for example WO 2005/073296 and Nature Nanotechnology Vol 3. August 2008, pp 508-511 , emulsion-templated water-soluble porous materials can contain water-insoluble materials, so that the addition of water may not only dissolve the porous material but also release such water-insoluble materials to produce aqueous dispersions. WO

2005/073298 discloses the formation of a "solution" of the hydrophobic compound Triclosan™ using such a method. WO 2005/075547 discloses other organic hydrophobic materials which may be dispersed in aqueous media using emulsion-templated water-soluble porous bodies.

WO 2005/073300 discloses an emulsion-templated porous material wherein a substantial part of the structure of the bodies is provided by a surfactant. WO 2005/075546 discloses emulsion-templated porous bodies which are soluble or dispersib!e in non-aqueous media. The lattice comprises a polymer which is soluble in water-immiscible non-aqueous media. We have now discovered that, surprisingly, magnetic materials may be incorporated into emulsion-templated porous composite materials.

The present invention provides a porous composite material comprising a hydrophilic carrier carrying a hydrophobic polymer and a magnetic material.

The present invention also provides a method for the preparation of a composite material comprising:

- providing an oil phase comprising an organic solvent which is immiscible with water, a hydrophobic polymer, and a magnetic material;

- providing an aqueous phase comprising water and a skeleton- forming material comprising one or more water-soluble polymer and/or surfactant;

- emulsifying said oil phase with said aqueous phase to form an oil- in-water emulsion;

- freezing said emulsion; and

- removing said water and said organic solvent by freeze-drying. The composite materiai of the present invention readily dissolves in water to provide highly-dispersed magnetic nanoparticles. Thus the solubilizing properties of the hydrophilic carrier, or skeleton, allow hydrophobic and magnetic materials to be dispersed well within an aqueous environment.

The magnetic nanoparticles are held within the composite material and released when the hydrophilic framework is so!ubilized by water. The nanoparticles comprise the hydrophobic polymer, magnetic material, and other components which were present in the oil phase, other than the solvent which has previously been removed by freeze-drying.

Much research is currently focusing on magnetic nanoparticles and their potential biomedical uses, ACS Appl, Mater. Interfaces 2011 , 3, 842-856 and Biomaterials 2010. 31 , 3694-3706 disclose magnetic nanoparticles for targeted therapy. However, the magnetic nanoparticles disclosed in the prior art have different structures and compositions to those of the present invention, and are made by different processes. In particular, in the present invention the magnetic material is surprisingly first dissolved or dispersed in the oil phase of an oil-in-water emulsion and the resultant nanoparticles are highly dispersible in water.

The magnetic material may be a metal oxide, for example magnetite, Optionally, the oil phase further comprises a hydrophobic compound or an organic compound. In particular the oil phase may comprise an active pharmaceutical ingredient (API). Thus the resultant magnetic

nanoparlicles may contain an API as well as have magnetic properties to allow effective targeted therapy.

Magnetic nanoparticles reported in the literature form dispersions which have limited storage times and stabilization problems, In contrast the present invention provides a solid format which can stay as a solid composite material prior to use; it is used to form dispersions when required. Furthermore, in comparison with the prior art, a higher percentage of the organic matter is associated with magnetic

nanoparticles. For example, in the present invention it is possible for over 60%, or over 70%, or over 75%, or even over 80% of the organic matter (e.g. hydrophobic material, organic compound, or active pharmaceutical ingredient) to have associated magnetic matter.

The hydrophobic polymer is soluble or dispersible in the organic solvent of the oil phase. In the resultant nanoparticle, the hydrophobic polymer provides inert solid bulk. However, the hydrophobic polymer may be absent or present at low concentrations, in which case the magnetic nanoparticles contain the magnetic material and other materials which were present in the oil phase. For example, the magnetic nanoparticles may contain the magnetic material and a water-insoluble organic material, e.g. an active pharmaceutical ingredient, with the hydrophobic polymer merely being an optional component.

The hydrophobic polymer may be any water-insoluble polymer that will dissolve into the oil phase such as, but not limited to, polyaikyl

methacrylates (eg poiymethyl methacrylate), polyaikyl acrylates (eg polybutyl acrylate), polyvinyl esters (eg polyvinyl acetate), polyvinyl ethers, polyesters (eg polycaprolactone), polycarbonates, and polystyrenics preferably polystyrene. When using the invention for pharmaceutical applications, the polymer is most preferably chosen from water-insoluble polymers contained within the US Food and Drug Administration Center for Drug Evaluation and Research lists of inactive ingredients

The organic solvent used in the oil phase is immiscible with water so as to form a discontinuous phase on emulsification.

The organic solvent may be chosen from, but not limited to haloforms (preferably dichloromethane, chloroform), esters (preferably ethyl acetate); alkanes (preferably heptanes, n-hexane, isooctane, dodecane, decane; - cyclic hydrocarbons, preferably toluene, xylene, cyclohexane); esters (preferably ethyl acetate); ketones (preferably 2-butanone); ethers

(preferably diethyl ether), most preferably toluene. When using the invention for pharmaceutical applications, the organic solvent is most preferably chosen from within the International Conference on

Harmonisation classification lists and selected from Ciass 2 or Class 3

The hydrophilic carrier of the composite material comprises one or more hydrophilic polymers and/or surfactants. The material used to make this water-soluble "skeleton" may comprise known pharmaceutically- acceptable polymers, surfactants and water soluble materials. The hydrophilic carrier may comprise a hydrophilic polymer which is a water soluble polymer as defined and listed in WO 2005/073298. In particular, preferred water soluble polymers include natural polymers (for example naturally occurring gums such as guar gum, alginate, locust bean gum or a polysaccharide such as dextran); modified natural polymers (for example cellulose derivatives cellulose acetate, methylcellulose, methyl- efhylcellulose, hydroxy-efhyiceilu!ose, hydroxy-ethylmethyl-cellulose, hydroxy-propylcellulose, hydroxy- propylmethylcellulose, hydroxy- propylbutylcellulose, ethylhydroxy-ethylcellulose, carboxy-methylcellu ose and its salts (eg the sodium salt - SC!VfC), or carboxy- methylhydroxyethylcellulose and its salts (for example the sodium salt)); homopolymers of or copolymers prepared from two or more monomers selected from: vinyl alcohol, acrylic acid, methacrylic acid, acrylamide, methacrylamide, acrylamide methylpropane sulphonates,

aminoalkylacrylates, aminoalkyl-methacrylates, hydroxyethylacrylate, hydroxyethylmethylacrylate, vinyl pyrrolidone, vinyl imidazole, vinyl amines, vinyl pyridine, ethyleneglycol and other alkylene glycols, ethylene oxide and other alkylene oxides, ethyleneimine, styrenesulphonates, ethyleneglycolacrylat.es and efhyleneglycoi methacrylate. Preferably the water soluble polymers is chosen from water-soluble polymers contained within the US Food and Drug Administration Center for Drug Evaluation and Research lists of inactive ingredients, most preferably polyvinyl alcohol (PVA), polyethylene glycol (PEG) or PVA-PEG copolymers, for example Kollicoat® Protect (BASF),

The hydrophilic carrier may comprise an anionic, non-ionic, amphoteric, zwitterionic or cationic surfactant which is as defined and listed in WO 2005/073300. Examples of suitable non-ionic surfactants include ethoxylated triglycerides; fatty alcohol ethoxylates; alkylphenol

ethoxylates; fatty acid ethoxylates; fatty amide ethoxylates; fatty amine ethoxylates; sorbitan alkanoates; ethyiated sorbitan alkanoates; alkyl ethoxylates; Pluronics(T ); alkyl polyglucosides; stearol ethoxylates; alkyl polyglycosides. Examples of suitable anionic surfactants include alkylether sulfates; alkylether carboxylates; alkyibenzene sulfonates; alkylether phosphates; dialkyl sulfosuccinates; sarcosinates; alkyl sulfonates; soaps; alkyl sulfates; alkyl carboxylates; alkyl phosphates; paraffin sulfonates; secondary n-alkane sulfonates; alpha-olefin sulfonates; isethionate sulfonates. Examples of suitable cationic surfactants include fatty amine salts; fatty diamine salts; quaternary ammonium compounds;

phosphonium surfactants; sulfonium surfactants; sulfonxonium

surfactants. Examples of suitable zwitterionic surfactants include N-alkyl derivatives of amino acids (such as glycine, betaine, aminopropionic acid); imidazoline surfactants; amine oxides; amidobetaines. Mixtures of surfactants may be used. In such mixtures there may be individual components which are liquid, provided that the carrier material overall, is a solid. A!koxylated nonionic's (especially the PEG/PPG Pluronic(TM) materials), phenol- ethoxylates (especially TR TON(TM) materials), alkyf su!phonates (especially SDS), ester surfactants (preferably sorbitan esters of the Span(Tlvl) and Tween(T ) types) and cationics (especially cetyltrimethylammonium bromide - CTAB) are particularly preferred as surfactant carrier materials. When using the invention for pharmaceutical applications, the surfactant is most preferably chosen from surfactants contained within the US Food and Drug Administration Center for Drug Evaluation and Research lists of inactive ingredients In particular, preferred surfactants include docusate sodium, sodium deoxycholate. Hyamine, sorbitan esters, polysorbates (eg Tween 20), PEG containing surfactants such as Vitamin-E PEG succinate, BRIJ and Pluronic surfactants, most preferably Solutol HS® 15",

The present invention will now be more particularly described, by way of example only, with reference to the accompanying figures.

The present invention is particularly useful in relation to water-insoluble organic compounds employed in aqueous media for pharmaceutical use. To overcome high water-insolubility the present invention provides nano- dispersions which have properties similar to those of molecular aqueous solutions. The present invention provides a generic method for producing hydrophobic nanoparticle dispersions via a combination ot modified emulsion templating and freeze-drying, The method can be used with a variety of hydrophobic active pharmaceutical ingredients (APIs) e,g, HAART therapeutics for HIV, or cytotoxic cancer therapeutics.

We have prepared nanoparticles bearing three hydrophobic ingredients, namely magnetite (FeaCU), polystyrene and oil red, to demonstrate that the technique can be used to generate composite nanoparticles bearing organic and inorganic components. There is great interest in the development of magnetic nanoparticles as they have shown promise in many biological applications, such as targeted drug delivery, ceil labelling and cell purification, tissue repair, contrast agents for magnetic resonance imaging (MR!) and hyperthermia for cancer therapy.

Figure 1 shows, by way of example only, a schematic representation of the emulsion templating/ freeze-drying technique. At first the immiscible wafer and oil phases are emulsified, for example via ultrasonication forming an oil in water (O/W) emulsion. This is followed by cryogenic freezing and solvent removal via freeze-drying producing a porous composite material. Figure 1. (a). The hydrophobic compounds are present in the volatile organic oil phase of the O/W emulsion white the aqueous phase contains a mixture of stabilizers (i.e. water-soluble polymers or surfactants or both), (b), Local concentration is increased as water is converted to ice crystals following cryogenic freezing, which leads to particle formation, (c). Both the organic and water solvents are removed via freeze-drying to produce a stable, porous composite material bearing both the water insoluble compounds as well as the water-soluble polymers and surfactants. The expanded part of (c) shows, schematically, a pore where the oil phase droplet used to be, and, in this pore, two nanoparticles. (d). Addition of water to the porous composites releases the hydrophobic compounds forming nanoparticie dispersions which bear similar properties to those of transparent molecular solutions. The expanded part of (d) shows, schematically, a nanoparticie, hydrophilic polymer, and surfactant.

Figure 2 shows photographs showing that a magnetic field directs the magnetic nanoparticles. The colouring is due to the oil red component. A lined background is used to help visualize the dispersion and

concentration of the nanoparticles.

Magnetic nanoparticles were prepared bearing three hydrophobic components, both organic and inorganic; Polystyrene, Oil Red;

Magnetite (FesC^). Nanoparticie dispersions were prepared using Toluene as the oil phase in a 1 :4 o/w ratio (0.5 ml total). Kollicoat Protect" (8mg) and Soiutoi HS^R (3mg) were employed as the polymer and surfactant stabilisers. Total hydrophobic mass - 1 mg, comprising 89% polystyrene, 1 % oil red and 10% Fea0₄ (added as 10 nm nanopar tides).

We have carried out various experiments to study the effect of the magnetic field on the nanoparticles and the dispersions. Over time the magnetic nanoparticles can be directed/ drawn by the magnetic field. Decrease in Z-average, Pdl, Number Average and Volume Average is indicative of the initial removal of larger magnetic nanoparticles creating a dispersion bearing smaller nanoparticles of more uniform size.

Our results show that the method can be used to prepare composite nanoparticles bearing three hydrophobic components; both organic and inorganic:

• Reduction in z-average over time in presence of magnetic field

· Loss of red colour and increase in transmittance over time

• Shown aggregation of nanoparticles when flowed past magnets through capillary -— differing aggregation depending on applied magnetic field

• SEM and TEM show nanoparticle structures both within the dispersion z- -average range and bearing Fe₃0₄. Figures 3 to 8 show SEM images. Figures 7 to 8 show TEM images.

The use of magnetic nanoparticle based therapy for cancer treatment Is born out of the knowledge that cell death begins to occur when ceils are heated above 41 °C. Furthermore, the rate of cell death as a result of increased temperature is more rapid in tumour cells due to their higher rates of metabolism. Upon target site accumulation alternating magnetic fields would be applied, raising the temperature of the nanoparticles, leading to cell necrosis and ultimately tumour death, with minimal damage to local healthy tissue cells in comparison to other therapeutic techniques. Introducing an external magnetic field, magnetic nanoparticles carrying anti-cancer drugs could be targeted to tumour sites. Targeting the anticancer drug containing magnetic nanoparticles to the site of a tumour will enable concentration of the drug at the site of action, meaning that potentially smaller doses of drug may be used which provides a better toxicity profile.

Metal oxides, such as maghemite {γ-Ρβ₂0₃) and magnetite (Ρβ₃0₄), loaded nanosystems are of particular interest, ahead of their pure metal analogues, due to their relatively low toxicity and sensitivity to oxidation. In order to improve biocompatibility and reduce aggregation of such magnetic metal oxide nanoparticles, surface modification is advantageous using, for example, polymer chains such as polyethylene glycol (PEG) or polyvinyl alcohol (PVA). Although their surface modification is a somewhat facile synthetic process, it still nevertheless requires chemical extra chemical procedures in the nanoparticle development. In contrast the present invention provides a generic method for producing organic nanoparticle dispersions via a combination of modified emulsion

templating and freeze-drying to generate composite nanoparticles bearing three hydrophobic ingredients, namely magnetite, polystyrene and oil red. The technique avoids the necessity to chemically modify the metal oxide surface while encapsulating the metal oxide inside a stable hydrophobic structure, which, when dispersed in water, shows similar properties to transparent aqueous solutions. Furthermore, the addition of oil red to the model magnetic nanoparticle composites give rise to coloured dispersions. Magnetic response is visually observed following colour changes upon the addition of a magnetic field, drawing out the red coloured magnetic particles from the dispersion.

10 nm average size magnetite (Fe₃04) magnetic nanoparticles, 5 mgm!~1 in toluene, were purchased from Sigma-Aldrich. Oil red, polystyrene and toluene were also purchased from Sigma Aldrich. Kollicoat Protect^® and Solutol HS^® 15 were purchased from BASF^®. 0.25 inch, cubic neodymium, boron iron oxide rare earth magnets were purchased from gyroscope.com. Dynamic Light Scattering (DLS)

Nanoparticle dispersions of 10 mgm!^"1 in 1 m! of water were prepared. The hydrodynamic diameter (z-average) was recorded using a DLS with cuvets, Further experiments in the presence and absence of a magnetic field were recorded where a neodymium, boron iron oxide magnet was coated with parafilm and suspended inside the cuvet, Just touching the top of the liquid, DLS was run as normal.

Electronic absorbsion

Nanoparticle dispersions of 10 mgrnr¹ in 1 ml of water were prepared. Electron absorbsion spectra were recorded using a lambda and cuvet. Magnetic field experiments were carried out similarly to the DLS experiments. %T was recorded between the range 350 to 800 nm both in the presence and absence of a neodymium, boron magnet iron oxide magnet was suspended inside the cuvet, coated with parafilm and just touching the top of the liquid inside the cuvet. Scans were recorded every 30 minutes until a plateau in %T was reached (50 hours).

Capillary flow experiments

10 mgrnr¹ dispersions of 3 ml volume in water were prepared. 0,5 ml was removed and an initial %T was recorded. This sample cuvet was kept and recorded again at the end of the experiment. The remaining 2.5 ml dispersion was divided into two and used to perform 2 experiments so resulting values could be compared with differing magnet arrangements and differing flow rates {f mlrr¹ and Sml^"1), Samples were passes through a mm capillary alongside which magnets were arranged. The experiment was carried out underneath an optical microscope. Magnet experiments were run in pairs and arrangements were as follows; 1 single magnet; past 2 magnets facing each other with poles aligned on either side of the capillary; past 2, 4 and 10 magnets in a row with poles aligned, facing the capillary kept approx 3 cm apart. The experiment was monitored with an optical microscope, and images were taken at time intervals depending on the flow rate (every 1 minute at 5m!h^"\ every 2 minutes at 1 m!h^"1),

Experiments were continued until 1 ml dispersant was collected in a cuvet. The %T was then recorded for the resulting dispersion.

Nanoparticle preparation

Stock solutions of polystyrene and oil red were prepared at 40 mgm!^"1 and 10 mgmi^"1 respectively in toluene. 15-20 nm Fe₃0₄ was purchased as 5 mgm!-1 in toluene. Two 1 ml stock solutions containing a total mass combination of 10 mgml^"1 were prepared; blank at 99% polystyrene, 1 % oil red, i.e., 247.5 micro-! (9.9 mg) polystyrene stock, 10 micro-l (0.1 mg) oil red stock and 742.5 micro-l toluene; magnetic at 89% polystyrene, 1 % oil red and 10% Fe₃0₄, i.e.. 222.5 micro-l (8.9 mg) polystyrene stock, 10 micro-l (0.1 mg) oil red stock, 200 micro-! (1 mg) Fesdj stock, and 587.5 micro-l toluene. The hydrophobic "active" stock solutions were then mixed for 2 hours to ensure homogeneity in the solution. An oil in water (O W) emulsion was prepared in which the hydrophobic compounds were

Π present in the volatile organic oil phase while the aqueous phase contains a mixture of stabilizers, the polymer Kollicoat Protect^® and the surfactant Solutol HS^® 15. Two aqueous solutions of the polymer and the surfactant were prepared, each at 22.5 mgm!^"1. The three stock solutions were added thus!y: 100 micro-l "active", 287.5 micro-l polymer, 132.5 micro-l surfactant, overall equating to 10 mg solid mass with a ratio of; 10% active, 60% polymer and 30% surfactant in a 1 :4 oil to water (OW) mix. The mixtures were then emulsified via sonication for 12 seconds flowing by immediate cryogenic freezing. Both the organic and water solvents are removed via freeze drying to produce a stable, porous composite material bearing the water insoluble hydrophobic compounds as well as the water- soluble polymers and surfactants. Upon addition of water to the porous composites the hydrophobic compounds are released as composite nanoparticles containing all the added hydrophobic material as a combination particle, forming dispersions which bear similar properties to those of transparent molecular solutions. 9 samples were made per toluene stock solution.

Z-ayerage vs. time in the presence of a magnetic f ield .

The porous composites that were produced following the emulsion templating and freeze-drying procedure were pink in colour, due to the presence of the added oil red, which formed pink dispersions upon addition of water. The hydrophobic dye was deliberately added in order to visually observe any colour changes to the dispersions upon introduction of a magnetic field. Uniform dispersions of sampies of both the blank composite, he, where the hydrophobic component equals 99% polystyrene and 1% oil red, and the magnetic composite, i.e. where the hydrophobic component equals 89% polystyrene, 1% oil red and 10 % Fe₃04, were dispersed in 1 ml of water and the released nanoparticles^s average hydrophobic diameter. Z-average, were measured via dynamic light scattering, DLS. Z-averages were measured as 812.9 nm and 924.77 nm for the blank and magnetic nanoparticles respectively, each showing reasonably low polydispersity with polydispersity indexes (Pdi) recorded as 0.481 and 0.382 respectively. The increase of approx 100 nm upon the replacement of 10% of the polystyrene with what equates to 0.1 mg of 15- 20 nm Fe₃0₄ is not surprising since small organic molecules are being replaced with larger, metal oxide nanoparticles, which suggests that incorporation of Fe₃0₄ into the nanoparticle forming a composite bearing al three hydrophobic components has occurs, rather than the dispersion containing individual nanopaticles of oil red, polystyrene and Fe₃Q₄ alone. To further prove this hypothesis, a magnetic fieid was applied to the Fe₃0 containing dispersions, Figure 11.

Figure 11 : Magnetic nanoparticle dispersions where in each picture the laft cuvet shows the dispersion with no exposure to a magnetic field and right shows dispersion with applied magnetic field after 24 hours, a and d show the magnet to the side of cuvet, b and e show the cuvet turned 90° clockwise after magnetic field is removed to show magnetic nanoparticle aggregation, c and f show magnet suspended above the dispersion.

The images in Figure 1 clearly show the removal of red coloured, magnetic nanoparticles from the dispersion upon the introduction of a magnetic field. The turbid aqueous dispersion becomes more transparent as colour is removed, as demonstrated by placing the cuvets in front of a lined background. No black lines are visible through the non-magnetised dispersion where as they are clearly visible once the magnetised nanoparticles have been pulled clear by the induced magnetic field.

Furthermore, since the magnets were placed either above or to the left of the dispersion it is obvious that the resulting increase in transparency must result from the application of the magnetic field and not from particle sedimentation, Images b and e show the resulting dispersion upon removal of the magnetic field showing nanoparticle aggregation occurs. Iron oxide nanoparticles tend to aglomerate and form clusters due to their inherent high surface area to volume ratio, dipole-dipole interactions and ferromagnetic behaviour. Increased aggregation occurs as these clusters are magnetised resulting in stronger interaction between the magnetic nanopartcles. Hence, the aggregated nanopaticles do not re-disperse upon removal of the magnetic field.

The removal of large particles over time in the presence of a magnetic field was observed by dynamic light scattering, DLS. The cuvet in images c and f , where the magnet is suspended just touching the top of the aqueous dispersion, was immediately placed in a DLS spectrometer after the magnet was introduced and the z-average was recorded every 10 seconds for a total of 80 minutes, The magnet was deliberately suspended above the solution to show that any change in z-average than was observed must result from the presence of the magnetic field and not from particle sedimentation. Furthermore, a control dispersion where no magnet was present was also run on the DLS spectrometer. Figure 12 shows the results,

Figure 12: Change in z-average over time in the presence (blue - lower - line) and absence (green - upper - line) of a magnetic field, 10 mgmf¹ dispersion, 1 ml water, 25°C. Figure 12 shows an overall decrease in z-average over time in the presence of a magnetic field, which does not occur in the absence.

Average particle size decreases from around 800 nm to 300 nm, while the control dispersion remains around 750 nm. The average size decreases as the larger magnetised nanoparticies present are removed from the dispersion, leaving the smaller particles behind. One could speculate that such smaller particles perhaps contain little or no Fe₃C and hence are not as greatly attracted to the magnet, which itself is now coated in the narnoparticuiate aggregates, reducing its overall magnetic strength, This explains why the solution remains slightly pink after the 80 minutes exposed to the magnet, Nevertheless, it is obvious that large magnetic particles are being removed from the dispersion by the imposed magnetic field, therefore must contain at least both polystyrene and FeaC^. The colour change further suggests that oil red is present in the nanoparticles produced; therefore there is strong evidence to suggest that composite nanoparticles bearing 3 hydrophobic ingredients have indeed been prepared following the emulsion templating and freeze-drying technique discussed here, !t is interesting to note that the percentage error for the magnetised sample is smaller than for the non-magnetised sample at 5% and 12% respectively, calculated from the standard deviation of the average of every 20 scans; each point in the plot shows an average of 20 scans, and is therefore representative of an average over 200 seconds. The difference in percentage error suggests that an overall loss in particle number occurs in the presence of the magnet, hence lowering the particle size range present in the dispersion and therefore lowering the overall error. This is further demonstrated in Figure 13 a-d, where the number average, volume average, derived count rate and polydispersity indexes are ali reduced over time as a result in the loss of larger particles from the dispersion. The remaining aqueous mixture would therefore contain smaller, more uniformly sized particles.

Figure 13: Magnetic nanoparticles in the presence of a magnetic field. Charts show reduction in (a) number average, (b) volume average, (c) derived count rate and (d) Pdl over time. 10 mgmr dispersion, 1 ml water, 25°C.

Dispersion transparency vs. time in the presence of a magnetic field.

From Figure 13 it is clear that the turbid dispersions become more transparent in the presence of the magnetic field. In order to quantify such observations electronic absorbance spectrometry was employed to monitor the percentage transmittance (%T) of light in the red region (-600 nm to 750 n n) of the electromagnetic spectrum through the sample (Figure 14 a-b). The cuvet was prepared in the same manner as for the DLS experiments, i.e., 10 mgmr¹ dispersion in 1 ml water with the magnet suspended above, just touching the surface of the liquid. An initial reading was taken at 0 minutes followed by a reading every 30 minutes until the observed percentage transmittance reached a plateau. Once again, a control dispersion where no magnetic field was applied was measured at the same intervals and for the same length of time.

Figure 14: %T vs, Time in the presence of magnetic field, a shows %T trace increase with time, b shows %T at 700 nm in the presence (filled circles) and absence (open circles) of the magnetic field. 10 mgml^"1 dispersion, 1 ml water, 25°C.

Percentage transmittance increased from around 0.2 % to 45 % after 50 hours in the presence of the magnetic field. Figure 14a clearly shows the increase in time in the red region of the electromagnetic spectrum. As the magnetic field pulls the red coloured particles away to aggregate on the face of the magnet more red light is allowed to pass through the remaining liquid, hence an increase in the %T between 600 and 750 nm, This further suggests that the particles themselves are red in colour, and hence further strengthens the hypothesis that the composite particles contain all three hydrophobic components of polystyrene. Fe₃C>4 and oil red. Figure 14b shows the overall increase from ~ 0.2 % to -45 % after 50 hours.

Furthermore, only a slight increase in %T from 0.25% to 1 ,0% was observed for the samples to which no magnet was introduced. This suggests that a slight amount of particle sedimentation occurred over the 50 hour time period, however the mere 0,8% increase is not high enough to suggest that the overwhelming increase of 44.8 %T in the presence of the magnetic is due to particle sedimentation, and must instead arise from the presence of the magnetic field pulling the coloured magnetic nanoparticles out of the dispersion. Indeed, an increase of 1 %T was seen after the first hour in the presence of the magnetic field.

It is noted that the experiment time was rather long at 50 hours before a plateau was reached, while the equivalent experiment run on the DLS reached a plateau after 80 minutes. Smaller, less Fe₃0₄ rich particles would not attracted to the magnet as rapidly as bigger particles.

Furthermore, the overall magnetic field strength would be reduced as the face of the magnet is coated with the magnetic particle aggregates. Figure 11 c and 11f clearly shows how such aggregates cover not only the face of the magnet which touches the liquid, but also collects on the sides of the magnet as well. Dynamic light scattering measures a particle size distribution that is more greatly affected by the presence of larger particles, Hence, rapid removal of such large particles would lead to a more dramatic decrease in the observed average particle size,

Furthermore, the decrease in number average, volume average, derived count rate and polydispersify indexes would also be expected (Figure 3a~ d) as fewer, smaller particles remain creating a more uniform particle sized dispersion. Electronic absorbsion spectrometry records simply the light transmitting through a sample, and hence is not affected by particle size. Although the smaller particles travel slower towards the magnetic field, they are still nevertheless red in colour and hence affect the observed %T. Therefore figure 4a-b is representative of complete magnetic particle removal. The fact that %T did not reach 100% shows that during nanoparticle preparation there will be some polystyrene, oil red and polystyrene/oil red composite particles produced, which are not attracted by the magnetic field and cause the remaining dispersant to be turbid, albeit far less turbid than it was prior to magnetic field exposure.

Magnetic particle aggregation following flow past a magnet.

Thus far, magnetic nanoparticle aggregation had been observed with a static magnet suspended above dispersions, so as to just touch the surface of the liquid. Experiments were therefore continued to observe whether magnetic nanopartic!es were removed as the dispersion was flowed past the magnetic field through a capillary. In the first experiment, 10 mgml^"1 dispersions were passed by 1 magnet, whose poles were aligned to face the capillary, i.e. so either the north or south pole was against the capiilary wall. The dispersion was passed a two flow rates; 1 mlh^" and 5 mlh^"1 until 1 ml was collected, therefore running for 60 minutes and 12 minutes respectively, The collected dispersion was taken and %T recorded. %T was also measured for the dispersion prior to interaction with the magnet and following the experiment with no magnetic interaction. The experiment was then repeated where the dispersion passed by 2, 4 and 10 magnets arranged in a row approx 3cm apart, and finally past 2 magnets sat either side of the capillary, with their poles attracting each other. Each experiment was carried out with a new dispersion, the %T of which was recorded prior to interaction with the magnet so as to normalise the results. Figure 9 shows the percentage increase in %T for each experiment at the two different flow rates, i.e. the percentage increase of the final %T of the dispersion which passed by the magnetic fields from the %T of the initial dispersion with no magnetic interaction. Figure 9: % increase in %T following interaction with magnetic fields at 1 mlh^"1 (blue - left - bars) and 5 mlh^"1 (red - right - bars). 10 mgml^"1 dispersion, 1 ml water, 25°C. Figure 10: photograph of f!ow through a capillary past a magnet, showing accumulation of oil red (OR) at the magnet.

Figure 9 shows that a greater increase in %T was observed at slower flow rates and with increased numbers of magnets. This is not surprising as under these conditions one would assume the increased exposure time to a magnetic field would result in the increased aggregation of nanoparticles at the magnet surface. An interesting observation was there being little difference in nanoparticle aggregation at both flow rates when the two magnets were placed either side of the capillary. Likewise, the lower increase in %T compared to when two magnets are placed in a row suggest less aggregation occurred as the dispersion flowed through a magnetic field rather than alongside it. Figures 15, 16a-f and 17a-f shows the images recorded for each of these experiments. It is clearly seen that a gradual accumulation of nanoparticle aggregates is observed at the edges of each magnet as the dispersion passes by. There is a spread across the entire face of the magnet when only one is present (Figure 15), This is not the case, however, when magnets are placed either side of the capillary, as aggregation occurs at the centre of the magnetic field (Figure 17a-f). One can clearly see the orientation of the magnetic field as particles aggregate in a cone like fashion between the two magnets. St became clear during the experiments that as the dispersion flowed from left to right, gradual increase in accumulation occurred as the dispersion flowed past the single magnets in a row. This was also initially seen when the magnets were placed facing one another, however once aggregation appeared to be so great the continually flowing dispersion was seen to remove aggragated magneted particles from the right hand side of the cone while depositing aggregates on the left. Therefore, at a certain point aggregation appeared to reach a limit with no more net increase in nanoparticles being deposited, and hence there being no great difference in %T increase between the two flow rates.

Figure 15: Images taken at the end of the experiment (3 images put together) following dispersion flow past the single agnet. Flow is from left to right, 10 mgm!^'1 dispersion, 25°C, flow rate 5 mlh^"1.

Figure 18: Images taken every 2 minutes (a to f) when dispersion flows past magnet. Flow is from left to right. The image is taken from the corner of the first magnet the dispersion passes. 10 mgml^"1 disperion, 25°C, flow rate 5 mlh^"1.

Figure 17: Images taken every 10 minutes (a to f) when dispersion flows between two magnets. Flow is from left to right. The image is taken from the centre of the two magnet the dispersion passes. 10 mgmf¹ disperion, 25°C, flow rate 1 mlh^"1.

Claims

1 , A porous composite material comprising a hydrophillc carrier

carrying a magnetic material.

2. A porous composite material comprising a hydrophiisc carrier

carrying a magnetic material and a hydrophobic polymer,

3, A porous composite material comprising a hydrophiisc carrier

carrying a magnetic materia! and a hydrophobic material or organic compound.

4. A porous composite material comprising a hydrophillc carrier

carrying a magnetic material and a pharmaceutically active ingredient.

5, A porous composite material comprising a hydrophillc carrier

carrying a magnetic material, a hydrophobic polymer and a hydrophobic material or organic compound.

6. A porous composite material comprising a hydrophillc carrier

carrying a magnetic nanoparticle.

7, A porous composite material as claimed in any preceding claim wherein the carried material is released on application of water to form a magnetic nanodispersion.

8. A porous composite material as claimed in any preceding claim wherein the carried materia! comprises a drug,

9. A porous composite material as claimed In any preceding claim wherein the drug is an anti-cancer drug.

10. A magnetic nanopartscle or a magnetic nanodispersson released by the application of water or an aqueous material to a porous composite material as claimed in any preceding claim,

11.A method for the preparation of a composite material comprising: a. providing an oil phase comprising an organic solvent which is

immiscible with water and a magnetic material;

b. providing an aqueous phase comprising water and a skeleton- forming materia! comprising one or more water-soluble polymer and/or surfactant;

c. emulsifying said oil phase with said aqueous phase to form an oii- in-water emulsion;

d. freezing said emulsion; and

e. removing said water and said organic solvent by freeze-drying,

12. A method as claimed in claim 11 wherein the oil phase of step a also contains a hydrophobic polymer.

13. A method as claimed In claim 11 or claim 12 wherein the oil phase of step a also contains a hydrophobic material or organic

compound.

14. A method as claimed in any of claims 11 to 13 wherein the oil

phase of step a also contains an active pharmaceutical ingredient,

15. A product obtainable by the method of any of claims 11 to 13.

18. A product as claimed in any of claims 1 to 10 or 15, or a method as claimed in any of claims 11 to 13, wherein the hydrophilic carrier or skeleton-forming water-soluble material is or comprises a water- soluble polymer.

17. A product as claimed in any of claims 1 to 10, 15 or 16, or a method as claimed in any of claims 11 to 13 or 18, wherein the hydrophilic carrier or skeleton-forming water-soluble material is or comprises a surfactant.

18. A product as claimed in any of claims 1 to 10 or 15 to 17, or a

method as claimed in any of claims 11 to 13, 16 or 17, wherein the magnetic material is or comprises a metal oxide.

19. A product or method as claimed in claim 18, wherein the magnetic material is or comprises magnetite.

20. A product as claimed in any of claims 1 to 10 or 15 to 19, or a method as claimed in any of claims 11 to 13 or 18 to 19, wherein the hydrophobic polymer is polystyrene.

21.A product as claimed in any of claims 1 to 10 or 15 to 20, or a method as claimed in any of claims 11 to 13 or 18 to 20, wherein the organic solvent is toluene.

22. A product as claimed in any of claims 1 to 10 or 15 to 21. or a method as claimed in any of claims 11 to 13 or 18 to 21 , wherein the water-soluble polymer is a PVA-PEG copolymer.

23. A product as claimed in any of claims 1 to 10 or 15 to 22 for use in therapy.

24. A product as claimed in any of claims 1 to 10 or 15 to 22 for use in cancer therapy.