WO2011116972A1

WO2011116972A1 - Nanoparticulate photoinitiators

Info

Publication number: WO2011116972A1
Application number: PCT/EP2011/001488
Authority: WO
Inventors: Sylvia Kuhn; Michael Schmitt; Rolf Hempelmann
Original assignee: Universitaet des Saarlandes
Current assignee: Universitaet des Saarlandes
Priority date: 2010-03-25
Filing date: 2011-03-24
Publication date: 2011-09-29
Anticipated expiration: 2012-09-25
Also published as: DE112011101038T5; EP2368919A1

Abstract

A nanoparticulate, isolatable photoinitiator comprises a nanoparticulate photosemiconductor absorbing visible or UV light and an initiation free radical precursor which is attached to the photosemiconductor and which on irradiation with a wavelength which is absorbed by the photosemiconductor fragments into a neutral stable molecule and a free radical having a life time of 5 χ 10^-3 s to 15 s. Further described are compositions comprising the nanoparticulate photoinitiator, a process for preparing polymerized or hardened articles or part of articles by the aid of the nanoparticulate photoinitiator, a product of the process and the use of the nanoparticulate photoinitiator in photopolymerization and/or photohardening.

Description

Nanoparticulate photoinitiators

Field of the invention The present invention relates to nanoparticulate photoinitiators for free-radical polymerization and also to a polymerization process utilizing them.

Background of the invention The well-known Kolbe reaction involves RCOO^" ions being electrochemically oxidized to an RCOO* free radical and then CO₂ is eliminated, leaving behind a free radical R^» (alkyl free radical in the case of alkanoic acids). Two free radicals R* (alkyl free radicals for example) then dimerize to R-R by forming a C-C bond. In the photo-Kolbe reaction, first described by B. Kraeutler et al. (B. Kraeutler and A.J. Bard, J. Am. Chem. Soc, 99, 7729 (1977)), a photoexcited semiconductor performs the oxidation of the carboxylate group. In this case, it is the electron hole in the semiconductor material (T1O2 for example) which oxidizes the carboxylic acid added to the dispersion to form a carboxyl free radical which fragments into the volatile product CO2 and a free radical which, if it is reactive (such as the methyl free radical), dimerizes with a further free radical. Stable, sterically hindered free radicals, such as the triphenylmethyl free radical, do, however, not dimerize, as expected (B. Kraeutler, CD. Jaeger and A.J. Bard, J. Am. Chem. Soc, 100, 4903-4905, 1978). B. Kraeutler et al. (B. Kraeutler, H. Reiche and A.J. Bard, J. Polym. Science Part C: Polym. Lett. Ed., 17, 535-538 (1979)) have also succeeded in initiating a free-radical chain reaction in a glacial acetic acid dispersion of TiO₂ and methyl methacrylate (MMA), presumably through the methyl free radical intermediate. If, however, neat, anhydrous vinyl acetate and a little glacial acetic acid are irradiated in the presence of dispersed TiO₂, no polymerization takes place (D. Yang, X.Y. Ni, W.K. Chen, Z. Weng, Journal of Photochemistry and Photobiology A-Chemistry 2008, 195, 323). It is an aim of the invention to enable a free-radical polymerization to be initiated similarly to the photo-Kolbe reaction under more diverse conditions than was hitherto the case with the photo-Kolbe reaction. A further problem addressed by the present invention will now be described.

Customary photoinitiators for offset printing inks, coatings and adhesives are reactive low molecular weight compounds having a molar mass M < 400.

Examples are PhC(OCH₃)₂C(0)Ph (Irgacure 651 ) PhC(0)C(CH₃)₂OH (Darocur 1 173) or isopropylthioxanthone. UV irradiation splits the C(0)-CR₂ bond to form mesomerism-stabilized free radicals which initiate the polymerization and/or crosslinking of monomers, prepolymers, oligomers and/or polymers in printing ink, coating or adhesive formulations. The art refers to this as the curing of the binders or as the drying of the UV printing inks. During the curing process, however, the conversion of the photoinitiator is only on the order of 10% (J. -P. Fouassier, Photoinitiation, Photopolymerization, and Photocuring: Fundamentals and

Applications, Hanser Gardner Pubns, 1995). The remaining proportion of the initiator molecules is able to diffuse or migrate within and out of the matrix of the cured system (such as printing ink, coating or adhesive).

The high chemical reactivity of the initiator molecules means that they are toxic to humans. Accordingly, these compounds do pose a health risk if they emanate from the printing inks, coatings or adhesives. There are relevant opinions from the German Federal Institute for Risk Assessment (BfR). One quote (English translation) by way of example: "Food packaging carries printed text in order to provide consumers with information. The printing inks contain chemical

substances that can migrate to food and be ingested. This also applies to photoinitiators. They are used to harden printing ink as rapidly as possible." (BfR Opinion No. 028/2008 "Replacement of isopropylthioxanthone (ITX) in printing inks by non-assessed substances is not appropriate" , http://www.bfr.bund.de/, 2008). BfR Opinion No. 044/2005 "Constituents of printing inks in beverages from cartons" (English translation), http://www.bfr.bund.de/, 2005, shows that commercial photoinitiators are found in the interior of food packaging. This represents a serious problem for printing press builders and printing ink producers. The problem has not been solved to the present date. It was initially believed that non-fragmenting initiators, so-called 2nd generation initiators, would be an alternative to fragmenting photoinitiators (B. Miiller, U. P.oth, Lackformulierung und Lackrezeptur, Vol. 2, Vincentz, Coating Compendien, 2005). However, they have failed to become established in commercial practice. There has been only limited implementation to date of the solutions whereby the photoinitiators are attached to organic matrices (see for example EP 0 859 797 B1 and EP 1 449 645 B1 ). In effect, photoactive groups are attached to complex organic macromolecules. The disadvantage with this is that the latter frequently bear a high proportion of further functional groups, the health risk of which is unknown. They further serve as unreactive supports which restrict mobility but thereby also have an effect on reactivity, making it likely that an increased proportion of requisite initiator has to be used. Moreover, the synthesis of initiators of this type is costly and

inconvenient.

Another important field in which free radical photoinitiators are used is plastic dental compositions cured or hardened with UV light. Here it would also be desirable to have photoinitiators which do not diffuse or migrate out of the hardened polymer matrix into the surrounding tissue.

Summary of the invention In a first aspect, the present invention provides nanoparticulate, isolatable photoinitiators comprising a nanoparticulate photosemiconductor absorbing visible or UV light and an initiation free radical precursor which is attached to the photosemiconductor and which on irradiation with a wavelength which is absorbed by the photosemiconductor fragments into a neutral stable molecule and a free radical having a life time of > about 5 χ 10^"3 s to about 15 seconds. A second aspect of the present invention is a composition comprising a compound or composition capable of being polymerized or hardened by a free radical process and the particulate photoinitiator according to the invention. A further aspect of the present invention is a process for preparing a polymerized or hardened article or part of an article, wherein a composition according to the second aspect of the present invention is irradiated with visible or UV light.

In another aspect the present invention relates to a product prepared by the above process.

In yet another aspect the present invention relates to the use of a particulate photoinitiator according to any of claims 1 to 10 for photoinitiating a free radical polymerization and/or hardening of one or more monomers, prepolymers, oligomers and/or polymers capable of being polymerized or hardened by a free radical process.

Brief description of the drawings Figure 1 is a schematic showing the construction of a nanoparticulate photoinitiator according to the invention, where VB is valence band and LB is conduction band.

Figure 2 shows a schema of a free-radical polymerization (R^* = initiation free radical or free-radical initiator, = free polymer radical of chain length 1 ,

P_n ^* = free polymer radical of chain length n, M = monomer).

Figure 3 shows the infrared spectrum in multiple reflection of a nanoparticulate photoinitiator obtained as per Example 1 compared with the unmodified photo- semiconductor. Figure 4 shows the infrared spectrum in multiple reflection of a nanoparticulate photoinitiator obtained as per Example 2 compared with the unmodified photo- semiconductor. Figure 5 shows a polymerization kinetics diagram with the nanoparticulate photo- initiators of Examples 1 and 2, with the commercial initiator Darocur 1173 and without initiator.

Detailed Description

The particulars shown herein are by way of example and for purposes of illustrative discussion of the embodiments of the present invention only and are presented in the cause of providing what is believed to be the most useful and readily understood description of the principles and conceptual aspects of the present invention. In this regard, no attempt is made to show structural details of the present invention in more detail than is necessary for the fundamental understanding of the present invention, the description taken with the drawings making apparent to those skilled in the art how the several forms of the present invention may be embodied in practice.

The nanoparticulate photoinitiators according to the invention comprise a nanoparticulate photosemiconductor which absorbs visible or UV light and is surface modified with an initiation free radical precursor. "Nanoparticle" or "nanoparticulate" herein refers to particles having a size in the range from about 1 to about 100 nm, preferably in the range from about 1 to about 50 nm, more preferably in the range from about 1 to about 30 nm and even more preferably in the range from about 2 to about 15 nm, for example about 5 nm. The term "photoinitiator" is used in its customary meaning with which a person skilled in the art is sufficiently familiar; that is, a photoinitiator herein is a material which releases a free radical on irradiation with UV or visible light. The nanoparticulate photosemiconductors are preferably selected from metal oxides, metal sulphides and selenides, lll-V photosemiconductors and SiN.

Examples of photoconducting metal oxides are ZnO, T1O2, Sn02, NaTa0₃, KTa0₃, Fe₂0₃, Zr0₂, WO₃, ITO (indium-doped tin oxide) and ATO (antimony-doped tin oxide). Photoconducting sulphides and selenides can be selected for example from ZnS, CdS, ZnSe, CdSe, AI₂Se₃. Examples of lll-V photosemiconductors are GaN, InN and GaP. SiN is a further useful photosemiconductor. These photosemiconductors may all also be doped in order that a suitable band gap as hereinbelow elucidated may be created.

The band gaps of the photosemiconductors are such that the photosemiconductors are capable of absorbing, or being stimulated with light, in the UV or visible wavelength range. By doping the semiconductor, the UV or visible light absorption maximum can be shifted into a desired wavelength range.

The surface of the nanoparticulate photosemiconductors is modified or

functionalized with an initiation free radical precursor (also referred to as a mediator herein).

An initiation free radical (or else free-radical initiator) for the purposes of the invention is capable of initiating and maintaining a free-radical polymerization (see Fig. 2).

This free initiation radical is attached to the surface such that the assembly formed from photosemiconductor and attached photoinitiator precursor, i.e. the nanoparticulate photoinitiator according to the invention, is isolatable. The attaching bond can come about by adsorption (chemisorption) or by an ionic bond.

An initiation free radical precursor for the purposes of the present invention is a compound whence, under suitable conditions, as hereinbelow elucidated, an initiation free radical (or a free-radical initiator) for a free-radical polymerization can emerge.

When the photoinitiator according to the invention is irradiated with a wavelength which is absorbed by the photosemiconductor, the initiation free radical precursor according to the invention will fragment into a neutral molecule and a free radical having a life time (in an inert gas atmosphere) in the range from about 5 χ 10^"3 s to about 15 s, and preferably in the range from about 5 ^χ 10^"2 s to aboutI O s, for example in the range from about 10^"1 s to about 1 s.

The neutral stable molecules eliminated from the initiation free radical precursor are for example selected from CO2, a phosphine substituted with one or more substituents, ammonia, a primary, secondary or tertiary amine, sulphur, SO2 and S0₃, preferably from C0₂, triphenylphosphine and a secondary amine. The initiation precursor is preferably selected such that the neutral stable molecule emanating from it is inert under the reaction conditions.

The free radical which emanates from the initiation free radical precursor has a life time of at least about 5 ^χ 10^"3 s. Free radicals of this type are stabilized free radicals. They include primarily mesomerism-stabilized free radicals, but also to a certain extent sterically hindered free radicals, for example alkanoyl free radical, more particularly the acetyl free radical, the acetonyl free radical, benzoyl free radical, benzyl free radical, tert-butyl free radical and many further free radicals stabilized by acyl or carbonyl, aryl or heteroaryl groups and also free radicals stabilized by halogen, OH, nitrogen-containing, phosphorus-containing and sulphur-containing groups for example.

However, the free radical should also not be fully unreactive, i.e. have a life time of more than about 15 s. The triphenylmethyl free radical for example would be an unsuitable free radical because it is too unreactive for steric reasons. The initiation free radical precursor is usually attached to the nanoparticulate photosemiconductor via the group which forms the neutral stable molecule after exposure to light. Examples of suitable initiation free radical precursors are a-keto carboxylic acids, β-carbonylalkylphosphonium salts, 1 ,1-dimethylalkanethiols, arylmethanesulphonic acids, arylmethanesulphinic acids and amines such as triethanolamine

hydrochloride. Particular preference is given to pyruvic acid, benzoylformic acid and acetonylphosphonium chloride. Pyruvic acid on irradiation of the photoinitiator according to the invention with suitable light waves fragments into CO2 and a stabilized acetyl free radical, benzoylformic acid fragments into C0₂ and a stabilized benzoyl free radical and acetonyltriphosphonium chloride fragments into triphenylphosphine and a stabilized acetonyl free radical. The initiation free radical precursor can optionally also itself be a photoinitiator which absorbs radiation having a wavelength which is absorbed by the photo- semiconductor, and in the process produces a free radical which can serve as initiation free radical (as in the case of benzoylformic acid for example). However, such an initiation free radical precursor is a more efficient free-radical initiator when in a state of attachment to a photosemiconductor.

When the photosemiconductor surface is basic in character, as is the case with certain oxides (ZnO and Zr0₂ for example), an acidic free initiation radical precursor is generally chosen.

When the photosemiconductor surface is acidic in character, as is the case with certain other oxides (Ti0₂ and Sn0₂ for example), a basic or cationic free initiation radical precursor is generally chosen, for example a triphenylphosphonium salt. However, by raising the acidity of the carboxylic acids (-M or -I effect), it is also possible to bind them to a broadly acidic particle surface.

A person skilled in the art knows which photosemiconductors have an acidic surface and which have a basic surface (see e.g. Jolivet, J.-P.; Henry, M.; Livage, J.; Bescher, E. Metal Oxide Chemistry and Synthesis: From Solution to Solid State; Wiley, 2000.) It is similarly known how to realize a negative inductive effect (-1) or a negative mesomeric effect (-M) in the case of a carboxylic acid for example (see e.g. Bruckner, R. Reaktionsmechanismen; Spektrum Akademischer Verlag, 2004; Vol. 3.).

The fragmentation mechanism of the present nanoparticulate photoinitiators on exposure to light is believed - without wishing to be tied to this - to proceed similarly to the mechanism of the photo-Kolbe reaction, although with the latter the photoinitiator is not in a state of attachment to the surface of a photo- semiconductor.

This presumed mechanism is depicted in Figure 1 for a basic photosemiconductor. An electron of the photosemiconductor is excited into the conduction band and leaves a positive hole behind in the valence band. This positive hole strips the depicted carboxylate moiety of an electron and the carboxylate radical becomes a carboxyl free radical moiety as a result. This free radical moiety detaches C0₂. Unlike with the original photo-Kolbe reaction (see introduction), this free radical does not dimerize at once, since it is stabilized, but initiates a free-radical polymerization, as shown in Figure 2.

Since the free radical is stabilized according to the invention and hence is relatively long-lived, it can be used to initiate polymerizations in media which are unsuitable for the conventional photo-Kolbe reaction, i.e. in which no initiation of the polymerization takes place (as in solvent-free monomers for example).

The initiation free radical precursors, if suitably adapted to the monomer and/or the polymer which forms (in respect of polarity for example), can also ensure a colloidal stabilization of the nanoparticulate photosemiconductors in the

polymerization. Colloidal stabilization leads to an increased specific surface area and hence to higher activity on the part of the photoinitiators and, on account of the better distribution (ideally dispersal to primary particle size), to a more homogeneous polymerization. The particulate photoinitiators according to the invention, if dispersed down to primary particle size in transparent binder systems, can form dispersions therein, before and after the polymerization, that are transparent to the visible spectrum.

Below the absorption limit of the photosemiconductor, the UV range in the case of Ti0₂ for example, they are not transparent, as expected.

The particulate photoinitiators according to the invention are immobilized in the cured polymeric matrix of the composition, they do not migrate through the polymer matrix. This is firstly because of their size and secondly because the curing reaction, as confirmed by ESR experiments, proceeds from the particle surface, i.e. around the particles. The relative proportion of captured free radicals in this more immobile region of the matrix is elevated compared to curing with molecular photoinitiators.

Another very significant aspect is that there are no loose undecomposed initiation free radical precursors in the polymer matrix, since they are attached to the particle surface.

The synergistic effect between the photosemiconductor nanoparticle and the initiation free radical precursor (mediator molecule) was surprisingly found to increase photocatalytic activity to such an extent that the use as a nanoparticulate photoinitiator for radiatively curing materials, such as printing inks, coatings and adhesives, is possible in the same manner as with conventional molecular free radical photoinitiators (e.g. Irgacure 651 (2,2-dimethoxy-1 ,2-di(phenyl)ethanone) or Darocur 1173 (2-hydroxy-2-methyl-1-phenyl-1-propanone)). The reaction times of the photocatalytically functionalized photoreactive nanoparticles are comparable to those of the currently customary molecular photoinitiators. Compared with these commercially available photoinitiators, however, the photocatalytically

functionalized photoreactive nanoparticles according to the invention have the advantage that they are immobilized in the cured material, such as printing inks, coatings and adhesives, and hence cannot egress and contaminate the

environment.

One of ordinary skill in the art will recognize that the use of the nanoparticulate photoinitiators of the present invention is not limited to radical polymerization or hardening of the materials mentioned above.

The nanoparticulate photoinitiators can be used in connection with any

monomeric, prepolymeric, oligomeric and/or polymeric compound or composition, for example a binder, that can be polymerized or hardened by actinic radiation in combination with a currently customary molecular radical photoinitiator.

One example is plastic dental compositions which can be cured or hardened with UV light. Other examples are photohardenable plastic sheets or shaped articles.

The compounds or compositions that can be polymerized or hardened by actinic radiation in combination with a radical photoinitiator are well-known to one of ordinary skill in the art. Usually, they comprise at least one carbon-carbon double or triple bond. However, particularly in hardening compounds already polymeric, other bonds that react with a free radical moiety, such as carbon-hydrogen bonds from which a hydrogen atom can be abstracted leaving a reactive radical, may also participate in the curing process.

It is impossible to list all the compounds or compositions that can be polymerized or cured in this manner, and therefore it is referred to the many textbooks which comprehensively cover this field, e.g. J. -P. Fouassier, Photoinitiation,

Photopolymehzation, and Photocuring: Fundamentals and Applications, Hanser Gardner Pubns, 1995; N. S. Allen, M. Edge, I. R. Bellobono and E. Selli (editors), Current trends in polymer photochemistry. Prentice Hall Europe, Hemel

Hempstead, 1995; and Radation Curing of Polymeric Materials, ACS Symposium Series, Vol. 417, American Chemical Society 1990. The amount of the particulate photoinitiator in the compositions to be polymerized or cured may be in the range of from > 0 to about 20% by weight, e.g. about 0.5 % to about 18% or about 16% or about 14% by weight, preferably from about 1 % to about 12 % or about 10 % or about 2% to about 8% by weight, e.g. about 3% to about 6% or about 7 % by weight, such as about 4% by weight.

EXAMPLES

Example 1

1.1. Preparing a suspension of nanocrystalline zinc oxide

A quantity of 20.5 g (0.15 mol) of anhydrous zinc chloride is dissolved in 180 ml of methanol. Similarly, 12 g (0.3 mol) of sodium hydroxide are dissolved in 120 ml of methanol. After clearing and cooling down, the sodium hydroxide solution is rapidly combined with the vigorously stirred ZnC solution. Cloudiness appears very quickly to indicate formation of the zinc oxide by hydrolysis. Stirring is continued at room temperature for about 12 hours. The reaction mixture is centrifuged at 3500 revolutions per minute. The supernatant is entirely discarded and the solid phase is separated. The remaining sodium chloride is removed by washing twice with methanol or preferably ethanol and phase separation by centrifuging. To avoid agglomeration, care must be taken to ensure that the zinc oxide is not dried to completion, but remains pasty. The zinc oxide has the cubic zincblende structure and consists of spherical crystallites about 8 nm in diameter. These particulars are confirmed by transmission electron microscopy, light scattering and x-ray diffraction.

1.2. Surface functionalization with benzoylformic acid

The entire zinc oxide prepared by the method of Example 1 , about 12.2 g, is redispersed in 100 ml of ethanol before heating to 70°C. A maximum of 31 mmol of benzoylformic acid (Ph-C(O)-COOH, Ph = phenyl) are dissolved in 10 ml of solvent (ethanol or methanol) and combined with the dispersion. After one hour, the modified zinc oxide is separated off by centrifugation. The precipitate is purified by rinsing with 100 ml of solvent each time and renewed centrifugation. This step is repeated 2 to 3 times. Infrared spectroscopy in multiple reflection can be used to corroborate successful functionalization, see Figure 3. The bands at 1600 cm^"1 and 1400 cm^"1 are specific to a carboxylate and confirm the attachment to the ZnO surface.

An ethanolic dispersion of the product (white paste) can be stored in a tightly closed container away from light.

Example 2

Surface functionalization with pyruvic acid

12.2 g of zinc oxide prepared by the method of Example 1.1. are redispersed in 100 ml of methanol followed by heating to 70°C. A maximum of 31 mmol of pyruvic acid (CH₃-C(0)-COOH) are dissolved in 10 ml of solvent (ethanol or methanol) and combined with the dispersion. After one hour, the modified zinc oxide is separated off by centrifugation. The precipitate is purified by rinsing with 100 ml of ethanol each time and renewed centrifugation. This step is repeated 2 to 3 times. Infrared spectroscopy in multiple reflection can be used to corroborate successful functionalization, see Figure 4. The bands at 1600 cm^"1 and 1400 cm^"1 are specific to a carboxylate and confirm the attachment to the ZnO surface.

An ethanolic dispersion of the product (white paste) can be stored in a tightly closed container away from light. Example 3

3.1. Incorporation into a binder system or a printing ink, coating or adhesive formulation

A photoinitiator paste prepared as in Example 1 or Example 2 is dispersed in ethanol. The nanoscale functionalized metal oxide can be incorporated into a binder in a conventional manner, for example a pan grinder, an Ultraturrax or a three-roll stand, as a dispersion. Since the weight content of the alcoholic dispersion is easy to determine (e.g. by weight loss upon evaporation of solvent), the content in the binder is easy to compute. The content of material according to the invention (semiconductor with mediator on its surface) can be up to 20 percent by weight. 3.2. Conventional testing of UV curing

Curing is conventionally tested via tacticity determinations using a UV

polymerization range for example. A suitable range is the Minicure laboratory range, type M-25- -URS-TR-SLC, from 1ST METZ of Nurtingen, Germany, having a maximum lamp power output of 200 watt cm^'1 and a maximum passage speed of 100 m min^'1. The binder system (e.g. 75 wt.% epoxy acrylate oligomer and 25 wt% dimethylolpropanetetraacrylate) incorporating the material according to the invention is applied to a test strip of PVC, for example, using a test printing press. The layer thickness amounts to a few μιτι, but is quantified by weighing. An add-on of 1.5 g/m² is typical. The laboratory dryer is set to a radiative power of

90 watt cm^"1 and a passage speed of 70 m s^"1 for example. Every passage is followed by a thumbprint test to see whether the system is already fully cured. The number of passages until curing is complete is a measure of the reaction rate of the polymerization. The rule for a given binder system is: the fewer the passages needed for curing, the higher the reactivity of the photoinitiator. This system of measurement can be used to determine that the cure of an acrylate (e.g. 75 wt.% epoxy acrylate oligomer and 25 wt% dimethylolpropanetetraacrylate) incorporating the material according to the invention is more than twice as fast as the cure of this acrylate without initiator. 3.3. Kinetics of polymerization reaction via real time Raman spectroscopy

Cure kinetics can also be determined via in-situ FT Raman spectroscopy, for example using a Bruker Optics Multiram spectrometer. To this end, the binder system (e.g. 75 wt.% epoxy acrylate oligomer and 25 wt%

dimethylolpropanetetraacrylate) incorporating the material according to the invention is introduced into a cell having a path length of 0.05 mm. This cell serves as cuvette (sample holder) in the Raman spectrometer. The front end of the cuvette in the Raman spectrometer is irradiated by the Raman excitation laser and the optics of the Raman spectrometer collect the scattered Raman light reflected off the front end. Recording a Raman spectrum only takes 25 seconds and is repeated under control of the UV lamp.

When no Raman measurement is carried out, the cuvette is irradiated on the back with UV flash light (66 flashes per s) via a lightguide. This UV light triggers the cure in the binder system, and so the kinetics of the polymerization reaction can now be tracked in situ using real time Raman spectroscopy. To this end, the intensity of a double bond vibration band of the monomer at about 1635 cm^"1 is studied. The polymerization consumes the double bonds, which leads to an intensity decrease over time. Such a measurement is depicted in Figure 5.

It is clearly apparent that the modified zinc oxides effectuate a decrease in the intensity of the double bond band and hence a cure of the coating film. Although optimization of the system is not yet concluded, the samples initiated by the nanoparticulate photoinitiators cure approximately at the same rate as the classic initiator Darocur 1 173 used. The fact that at large times the proportion achieved in the intensity of the double bond vibration bands is higher testifies to the low mobility, i.e. the additional encapsulation, of the initiators. Example 4

4.1. Production of nanocrystalline titanium dioxide in the anatase form

To a mixture of 192 g (0.68 mol) of titanium tetraisopropoxide and 235 ml of 1 - pentanol are added 1 1.2 ml of fuming HCI solution (37% strength) followed by stirring at room temperature for 10 minutes. Thereafter, 14.2 ml of twice-distilled water are slowly added dropwise under vigorous agitation to form a slightly cloudy reaction mixture, which is subsequently stirred at 130°C for 24 h. During this time, titanium dioxide is formed by hydrolysis and polycondensation, as is apparent from the precipitation of a solid from the mixture. A rotary evaporator is then used to remove the volatile components at 65°C and gradual reduction in pressure down to 2 mbar. The titanium dioxide thus produced is present in the tetragonal anatase form and consists of spherical crystallites 5 nm in diameter. These particulars were confirmed by transmission electron microscopy, dynamic light scattering and x-ray diffraction.

4.2. Surface functionalization with acetonyltriphenylphosphonium chloride in diluent aery late

The dry surface-modified nanoscale T1O2 powder produced according to the present invention is used to prepare a 10 per cent by weight dispersion in toluene. The powder is only dispersible in apolar organic solvents at room temperature by stirring. A stable, clear and optically transparent dispersion is formed. A diluent acrylate (such as dimethylolpropanetetraacrylate) and

acetonyltriphenylphosphonium chloride ((Ph₃PCH-C(O)-CH₃)⁺Cr) are additionally added to this dispersion. In the process, a 45:55% by weight ratio of T1O2 to acetonyltriphenylphosphonium chloride is set. Finally, toluene is removed from the mixture by evaporation.

This provides a transparent, nanoscale formulation into which the further components needed for preparing printing inks can be stirred directly.

The T1O2 functionalized with acetonyltriphenylphosphonium chloride can alternatively be prepared in only toluene as solvent and subsequently isolated by removing the solvent and washing as described for functionalized ZnO in

Examples 1 and 2.

4.3. Incorporation into a printing ink formulation

The components typically used for preparing printing ink formulations are stirred into the preparation described in section 4.2. (surface-modified nanoscale T1O2 particles in diluent acrylate). These components include acrylates (e.g. 75 wt.% epoxy acrylate oligomer and 25 wt% dimethylolpropanetetraacrylate), rheology modifiers (such as polyethylenewaxes), film-forming agents (e.g. polyacrylates) and standard pigments. The total amount of surface-modified nanoscale T1O2 particles without diluent acrylate comprises not more than 4% by weight of the overall formulation in order that curing may be achieved. The typical mixing time to homogenize the components is only 1 - 3 min without the use of special dispersing equipment, since the homogeneous distribution of the particles has already taken place in the diluent acrylate.

4.4. Application

Application can take place on absorbent substrates (paper strips for example) or non-absorbent substrates (PVC strips for example) using a manual, semiautomatic or fully automatic roll printing press. Typically, 1.5 g of overall formulation are applied per 1 m² of coating area.

4.5. Curing

Curing is done using an 1ST UV Minicure laboratory range (M-25-1-Tr-SS). Not only the irradiated intensity (100 mW/cm² to 200 mW/cm²) but also the belt speed (from 10 m/min to 100 m/min) can be varied. Typically, unpigmented systems are cured at 200 mW cm^"2 within one passage at a passage speed of 50 m/min. In the case of pigmented systems, the rate of cure varies with the colour pigment used, between 1-2 (magenta) and 2-3 (black) passages. Without addition of the material according to the invention, the unpigmented material takes 4 passages to cure and the pigmented material between 9 (magenta) and 11 (black).

Example 5

5.1. Preparing a suspension of nanocrystalline titanium dioxide in the rutile form

A quantity of 26.8 g (0.14 mol) of TiCI₄ is passed into a solution of 375 ml of water and 15 ml of HCI (37% strength) under acetone/ice cooling. The reaction mixture is stirred for 1 hour and then heated to 60°C for 3 hours, the previously cloudy solution clarifies first. After about 1 hour, the Ti0₂ gradually precipitates. Stirring is continued at room temperature for about 12 hours. The reaction mixture is centrifuged at 4500 revolutions per minute. The supernatant is discarded and the titanium dioxide is then first washed with water and then twice with ethanol. The titanium dioxide should remain pasty and not be dried completely. Rutile is formed therefrom for thermodynamic reasons.

5.2. Surface functionalization with benzoylformic acid

The entire titanium dioxide prepared by the method is redispersed in 100 ml of ethanol before heating to 70°C. A maximum of 31 mmol of benzoylformic acid (Ph- C(0)-COOH, Ph = phenyl) are dissolved in 20 ml of ethanol and combined with the dispersion. After one hour, the modified titanium dioxide is cooled and separated off by centrifugation. The titanium dioxide is purified by rinsing with 00 ml of ethanol each time and renewed centrifugation. This step is repeated 2 to 3 times.

An ethanolic dispersion of the product (white paste) can be stored in a tightly closed container away from light. 5.3. Incorporation into the printing formulation

Incorporation is done similarly to that described in point 3.1.

The disclosure of all patents, patent applications, books and journal articles cited herein is herewith incorporated in its entirety by reference.

It is noted that the foregoing examples have been provided merely for the purpose of explanation and are in no way to be construed as limiting of the present invention. While the present invention has been described with reference to an exemplary embodiment, it is understood that the words which have been used herein are words of description and illustration, rather than words of limitation. Changes may be made, within the purview of the appended claims, as presently stated and as amended, without departing from the scope and spirit of the present invention in its aspects. Although the present invention has been described herein with reference to particular means, materials and embodiments, the present invention is not intended to be limited to the particulars disclosed herein; rather, the present invention extends to all functionally equivalent structures, methods and uses, such as are within the scope of the appended claims.

Claims

1. A nanoparticulate, isolatable photoinitiator comprising a nanoparticulate photosemiconductor absorbing visible or UV light and an initiation free radical precursor which is attached to the photosemiconductor and which on irradiation with a wavelength which is absorbed by the photosemiconductor fragments into a neutral stable molecule and a free radical having a life time of about 5 ^χ 10^"3 s to about 15 s.

2. The nanoparticulate photoinitiator according to Claim 1 , characterized in that the photosemiconductor is selected from metal oxides, metal sulphides and selenides, lll-V photosemiconductors and SiN which may be doped in order to create a suitable band gap.

3. The nanoparticulate photoinitiator according to Claim 2, characterized in that it is selected from ZnO, Zr0₂, Sn0₂ and T1O₂.

4. The nanoparticulate photoinitiator according to any one of Claims 1 to 3, characterized in that the part of the initiation free radical precursor which forms the neutral stable molecule after fragmentation is attached to the photosemiconductor.

5. The nanoparticulate photoinitiator according to any one of Claims 1 to 4, characterized in that the bond attaching the initiation free radical precursor to the photosemiconductor is an ionic bond or comes about by chemisorption.

6. The nanoparticulate photoinitiator according to any one of Claims 1 to 5, characterized in that the neutral stable molecule is selected from C0₂, a

phosphine substituted by one or more substituents, ammonia, a primary, secondary or tertiary amine, sulphur, S0₂ and S0₃.

7. The nanoparticulate photoinitiator according to Claim 6, characterized in that the neutral stable molecule is selected from CO₂, triphenylphosphine and a secondary amine.

8. The nanoparticulate photoinitiator according to any one of Claims 1 to 7, characterized in that the free radical is selected from alkanoyl free radical, more particularly acetyl free radical, acetonyl free radical and benzoyl free radical.

9. The nanoparticulate photoinitiator according to any one of Claims 1 to 7, characterized in that the initiation free radical precursor is itself a photoinitiator which absorbs visible or UV light and in the process gives rise to a free radical which can serve as free initiation radical.

10. The nanoparticulate photoinitiator according to any one of Claims 1 to 9, characterized in that the life time of the free radical is in the range from 10^"1 s to 10 s.

1 1 . A composition comprising a compound or composition capable of being polymerized or hardened by a free radical process and the particulate

photoinitiator according to any of claims 1 to 10.

12. The composition according to claim 1 1 wherein the concentration of the photoinitiator is in the range of > 0 wt.% to about 20 wt.%.

13. The composition according to claim 1 1 or 12, wherein the compound or composition capable of being polymerized or hardened by a free radical process comprises one or more monomers, prepolymers, oligomers and/or polymers.

14. The composition according to any of claims 1 1 to 13, wherein the

composition capable of being polymerized or hardened by a free radical process is a printing, coating or adhesive composition.

15. The composition according to any of claims 1 1 to 13, wherein the

composition capable of being polymerized or hardened by a free radical process is a dental composition.

16. A process for preparing a polymerized or hardened article or part of an article, wherein a composition according to any of claims 11 to 15 is irradiated with visible or UV light.

17. The process for preparing a polymerized or hardened article or part of an article according to claim 16, wherein the article or part of an article is a printing, coating or adhesive layer, a dental filling or an artificial dentition.

18. A product prepared by the process of claim 16 or 17.

19. Use of a particulate photoinitiator according to any of claims 1 to 10 for photoinitiating a free radical polymerization and/or hardening of one or more monomers, prepolymers, oligomers and/or polymers capable of being polymerized or hardened by a free radical process.