Field of the Invention

The present invention relates to methods for post-polymerisation and post-curing annealing treatment of thermoset polymers and products obtaina- ble by such methods.

Background of the Invention

Many chemical reactions involve the use of harsh organic solvents such as trichloromethane (chloroform). These reactions need to be carried out in apparatus that is resistant to the solvents such that the apparatus will not be degraded during the progress of the reaction.

In such instances, it is common practice to use apparatus made from glass. Glass is beneficial as it exhibits a high resistance to harsh solvents. Further, it is optically transparent and therefore allows the progress of the re- action to be monitored visually. Glass can withstand high temperatures and in toughened form is resistant to breakages. Mass-produced glassware is generally relatively inexpensive and therefore a good choice for apparatus produced on a large scale.

However, glass is not cost-effective to process on a small scale. In instances where a one-off reaction vessel is required, or a small number of reaction vessels are required, it is expensive and time consuming to have these fabricated from glass.

This is particularly relevant in the field of microfluidic devices. Microflu- idic devices are very small reaction vessels that are used to investigate the physical and chemical properties of liquids and gases on a microscale. They typically take the form of small chips having a series of channels of the order of several hundred micrometres to a few micrometres in size. Microfluidic devices enable the use of much smaller quantities of samples and reagents, which is both cost effective and can significantly reduce the processing time for a reaction. Further, so-called“lab-on-a-chip” systems allow multiple reac- tion stages to take place on a single, micro-device.

Microfluidic devices can be made from glass and such devices benefit from the properties of glass, including the resistance to solvents and the opti- cal transparency. Further, the surface chemistry of glass is well-known and can therefore be taken into account. However, it is expensive and cumber- some to fabricate microfluidic devices from glass.

To address the downsides of manufacturing microfluidic devices from glass, polymers have been used to fabricate microfluidic devices, typically thermoplastic polymers.

It will be understood that a“thermoplastic polymer” is a polymer that does not have a cross-linked structure. In thermoplastic polymers, interac- tions between the polymer chains are restricted to interactions such as hy- drogen bonding, which can be broken when the polymer is heated. As a re- sult, thermoplastic polymers can be melted and reshaped upon the applica- tion of heat.

Other Thermoplastic polymers commonly used for microfluidic devices include fluoropolymers such as fluorinated ethylene propylene and polyimides such as Kapton®. However, using these materials to make devices is often laborious and expensive and thus they do not provide a cost effective alterna- tive to glass.

Thermoset polymers can also be used to manufacture microfluidic de- vices. It will be understood that a“thermoset polymer” is a polymer that is cross-linked during curing such that there are chemical bonds between the polymer chains. The cross-linked structure of thermoset polymers means that upon application of heat after the curing process, they do not melt.

One commonly used example of a thermoset polymer used to fabricate microfluidic devices is polydimethylsiloxane. Polydimethylsiloxane is relative- ly inexpensive and is transparent to visible light and thus allows micro- channels and their contents to be observed. However, polydimethylsiloxane ages with use and thus the performance of a device fabricated from this ma- terial deteriorates over time. Further, it has poor chemical compatibility with many organic solvents which limits the suitability of devices made from this material to predominantly aqueous applications.

Various attempts have been made to address the issue of solvent re- sistance of polymers. For example, W02012/140194 discloses a process of producing a thermoplastic polymer using thiol-ene addition polymerisation. It indicates that, after polymerisation, the sulfur atoms in the thioether groups can be oxidised to their sulfoxide and sulfone functions and doing so can re- sult in a change of properties, including changes in solubility or lack thereof in particular solvents. However, even after the oxidation process, the polymer is soluble in chloroform and is therefore not suitable for use with chloroform.

US2016/0145392 discloses an ene-thiol based curable composition. The aim is to provide a composition which has good solvent resistance, par- ticularly in relation to solvents such as water, acetic acid, isopropyl alcohol and methyl ethyl ketone. The document discloses that the desired solvent resistance characteristics can be achieved through the selection of the thiol and ene components in order to tailor the properties of the curable composi- tion. Thus, it indicates that desired solvent resistance can be achieved through choice of monomers forming the polymer. However, the product is not suitable for use with harsh solvents such as trichloromethane.

EP0067976 relates to cross-linkable polysulfones and polethersul- fones. It provides a general indication that a“medium” extent of crosslinking is desirable to give a thermoplastic-like material but with solvent resistance similar to that of a thermoset. There is no indication that the product would be suitable for use with harsh solvents such as trichloromethane.

It is an object of the present invention to address one or more draw- backs of the conventional means. In particular, it is an object of the present invention to provide a method of treating a thermoset polymer which results in a polymer with improved solvent resistance, particularly against harsh sol- vents such as chloroform.

Summary of the Invention

Thus, according to a first aspect of the present invention, there is provided a method for treatment of a thermoset polymer with fully consumed functional groups and crosslinked monomers after polymerization, the method comprising the steps of:

i) providing a fully polymerised and cured thermoset polymer contain- ing a plurality of sulfur atoms; and ii) subjecting the cured thermoset polymer to one of the following post-polymerization annealing processes:

a) heating the cured thermoset polymer to a temperature of from 50°C to 250°C, and maintaining the cured thermoset polymer at a tem- perature of from 50°C to 250°C for a period of from 1 minute to 72 hours in an oxygen-containing environment; and/or

b) exposing the cured thermoset polymer to UV radiation of wavelength from 10nm to 400nm at an intensity of from 50 mW/cm ² to 150 mW/cm ² for a period of from 1 minute to 24 hours; or

c) contacting the cured thermoset polymer with an oxidising agent for a period of from 10 minutes to 24 hours.

By post-polymerisation and post curing annealing process it is meant a process that is carried out on a fully polymerised and cured starting polymer material, i.e, with fully consumed functional groups and crosslinked mono- mers after polymerization and curing.

It has surprisingly been found that the methods result in treated thermoset polymers that exhibit higher resistance to solvents such as chloro- form than the untreated form of the thermoset polymer. The increase in sol- vent resistance is sufficiently large for the thermoset polymers treated in this way to be used in applications where they come into contact with harsh sol- vents such as chloroform without dissolving and, thus, without the polymer- based product changing shape. The pre-existing crosslinking and depletion of functional groups allows the post-polymerization annealing step to be car- ried out without disintegrating or losing the shape of the polymer, i.e., non- fully polymerised and/or crosslinked thermoplasts would melt at annealing temperatures. Thus, the method of the present invention concerns the treat- ment to render the polymer properties after a previously completed polymeri- sation and crosslinking processes.

Further, the resulting polymers are typically transparent or translu- cent and relatively cheap and easy to fabricate and manipulate, making the polymers treated according to these methods an attractive, favourable alter- native to glass and conventional polymer solutions.

The observed effects were unexpected since there was previously no indication that cured thermoset polymers containing a plurality of sulfur atoms could be treated in these ways to improve solvent resistance. On the contra- ry, it was conventionally expected that application of the relatively harsh con- ditions would degrade the polymer rather than bring about a desired change in properties.

The mechanism by which the change in properties of the thermoset polymer is achieved is not at present fully understood. However, without wishing to be bound by any theory, it is believed that the post-polymerization and post-curing annealing treatments result in an increase in the cross-linking density of the polymer thereby resulting in the change in properties. The change in cross-linking density may also be accompanied by other changes at a molecular level. The role of the sulfur atoms in the process is not fully understood.

Whilst the improvement in solvent resistance is particularly relevant for microfluidic devices, it will be understood that the applications are much broader and can encompass many other areas including reactor technology, electrical and electronics components, medical equipment and even more general applications where chemical resistance is required.

Further, the post-polymerization and post-curing annealing treatments have also been found to improve mechanical resistance, such as scratch re- sistance, and therefore said treatments can also be used in instances where these properties are desirable, such as in the treatment of surfaces prone to scratching, for example flooring or spectacle lenses.

In step a), the polymer sample may be heated in an oven, may be heated by radiation with an infrared lamp and/or may be heated by contact with a heated species such as superheated steam.

In step b), the polymer sample may be placed within or under a UV lamp.

In step c), the polymer sample may be submerged or immersed in a liquid form of the oxidising agent.

The specific conditions used will depend on the nature of the polymer and the type of post-polymerization and post-curing annealing treatment be- ing conducted. Minor variations in the end points of the ranges outlined in the claims are possible as would be understood by the skilled person.

The method may include only step a), only step b) or only step c). Alternatively, the method may include step a), step b) or step c) in combina- tion with other steps. The method may include a combination of the claimed steps, for example steps a) and b), steps a) and c) or steps b) and c), option- ally with other steps. Alternatively, all three steps may be combined.

Further advantageous embodiments of the method are outlined in Claims 2 to 13.

In a), the polymer is held at an elevated temperature for a period of from 1 minute to 72 hours. The polymer may be maintained at a temperature of from 50°C to 250°C for a period of from 5 minutes to 48 hours, or from 5 minutes to 24 hours, or from 5 minutes to 18 hours or from 5 minutes to 12 hours or from 5 minutes to 6 hours, or from 30 minutes to 48 hours, or from 30 minutes to 24 hours, or from 30 minutes to 18 hours or from 30 minutes to 12 hours, or from 30 minutes to 6 hours, or from 1 hour to 48 hours, or from 1 hour to 24 hours, or from 1 hour to 18 hours or from 1 hour to 12 hours or from 1 hour to 6 hours.

In a), any temperature in the range of from 50°C to 250°C may be used. For example, the polymer may be heated to and maintained at a tem- perature of from 120°C to 220°C.

It has been found that post-polymerization and post-curing annealing heat treatment provides an effective way to adjust the polymer’s properties such that its solvent resistance increases. In particular, the heat treatment is quick and simple to perform in practice, whilst providing significant improve- ment in the properties of the polymer.

The most suitable time period and/or temperature may depend on, inter alia, the nature of the polymer.

In some embodiments, the oxygen containing environment contains at least 5% (v/v) oxygen, preferably at least 10% (v/v) oxygen, optionally from 15% (v/v) oxygen to 25% (v/v) oxygen. Heating the cured thermoset polymer in air has been found to be effective. Heating in air provides a convenient method of providing sufficient oxygen for the desired effects. The role of oxy- gen in the process is unconfirmed at present, but it has been found that it is desirable to carry out the heating step in an oxygen containing environment in order for the heating process to provide a desired increase in solvent re- sistance.

In b), the wavelength of radiation is from 10nm to 400nm. The wave- length of the radiation may be from 350nm to 400nm, preferably 365nm. It has been found that short wavelength UV is particularly effective at modifying the properties of the polymer, whilst being convenient to apply in practice.

In b), the radiation intensity is from 50mW/cm ² to 150mW/cm ² . The UV radiation may have an intensity of 75mW/cm ² to 100mW/cm ² , preferably 90W/cm ² . These intensities provide sufficient radiation to effect modification of properties, whilst allowing the modification to occur in a relatively short pe- riod of time.

In b), the cured thermoset polymer may be exposed to UV radiation for a period of from 1 minute to 24 hours. For example, the cured thermoset polymer may be exposed to the UV radiation for a period of from 2 minutes to 12 hours, or from 2 minutes to 6 hours, or from 2 minutes to 3 hours, or from 2 minutes to 2 hours, or from 30 minutes to 6 hours, or from 30 minutes to 4 hours, or from 30 minutes to 3 hours, or from 1 hour to 6 hours, or from 1 hour to 4 hours, or from 1 hour to 3 hours.

It will be understood that the time period used may depend on the nature of the polymer and/or the nature of the UV radiation.

In many applications, including industrial applications, polymers are routinely cured using UV radiation. Thus, using UV radiation to provide a post-curing treatment is particularly attractive since in such cases no addi- tional apparatus would be required. Polymer samples could be cured in the usual way and then the same apparatus used to provide the post-curing treatment, thereby providing a streamlined and efficient process.

In c), any oxidising agent may be used which is capable of oxidising sulfur atoms of thiol or thiolether groups. For example, the oxidising agent may be hydrogen peroxide, peracids such as 3-Chlorobenzene-1- carboperoxoic acid (MCPBA), ethaneperoxoic acid or 2,2,2- trifluoroethaneperoxoic acid, potassium peroxymonosulfate (oxone), potassi- um persulfate or ozone.

In some embodiments, the oxidising agent may be an aqueous so- lution of hydrogen peroxide of from 0.1 % (v/v) hydrogen peroxide to 40% (v/v) hydrogen peroxide. These concentrations of hydrogen peroxide have been found to enable acceptable modification of the polymer’s properties. Further, these concentrations of hydrogen peroxide are relatively easy to handle, whilst acting upon the polymer within a relatively short time scale to give the desired results.

In c), the cured thermoset polymer is contacted with the oxidising agent for a period of from 10 minutes to 24 hours. For example, the cured thermoset polymer may be contacted with the oxidising agent for a period of from 30 minutes to 18 hours, or from 30 minutes to 10 hours, or from 30 minutes to 4 hours, or from 30 minutes to 2 hours, or from 1 hour to 20 hours, or from 2 hours to 20 hours, or from 5 hours to 20 hours. It will be understood that the time period used may depend on the nature of the polymer and/or the nature of the oxidising agent. In some instances, a contact time in the region of 16 hours may be particularly advantageous, for example from 14 hours to 18 hours.

Oxidising agents, and in particular those in liquid form, can provide a particularly convenient method of treating the polymer since it can simply be submerged in the oxidising agent to effect the treatment.

The cured thermoset polymer contains a plurality of sulfur atoms. At least some of the sulfur atoms may be present in the form of thiol and/or thio lether groups.

The cured thermoset polymer may be a thiol-ene polymer. A thiol-ene polymer is a polymer formed via a thiol-ene reaction i.e. reaction between a thiol and an alkene. The reaction is typically conducted in the presence of a radical initiator, or catalyst. The reaction is a so-called“click” reaction and typically uses mild reaction conditions, results in high yields and without any by products. The reaction is also insensitive to ambient oxygen and water. Use of a thiol-ene polymer is particularly advantageous as it provides a straightforward route to a polymer which is highly responsive to the post- polymerization and post-curing treatments of the present invention. Further, such polymers can typically undergo simple modification with surface altering molecules, generally have good mechanical stiffness and have high optical transparency, even after treatment according to the present invention.

The cured thermoset polymer may contain a plurality of heterocyclic structures. Without wishing to be bound by any theory, it is believed that these structures can contribute to the chemical stability of the treated poly- mer. It is also believed that they provide rigidity. However, the mechanism is not at present fully understood.

The heterocyclic structures may be single rings or fused rings and may be mono-substituted or may have multiple substitutions.

Triazine structures are particularly favourable. It will be understood that the term triazine structures encompasses groups containing any isomeric form of triazine i.e. 1 ,2,3-triazine, 1 ,2,4-triazine or 1 ,3,5-triazine. Combina- tions thereof may be present, and these may be present as fused rings.

Additionally, or alternatively, pyridine and derivatives, diazines, tetraazines or other heterocyles may be present.

The cured thermoset polymer may contain a photoinitiator. The pho- toinitiator can exaggerate the effects of the post-polymerization and post- curing treatments of the present invention. The photoinitiator used may de- pend on the nature of the polymer and/or the treatment method. In some in- stances, the photinitiator is 2,4,6-trimethylbenzoylphenyl phosphinate (T- POL). The photoinitiator may be added to the monomers forming the polymer i.e. before polymerisation.

The step of providing a cured thermoset polymer may involve providing an off-the-shelf product in which the polymer is purchased in a form that has already been polymerised and cured. According to a second aspect of the present invention, there is provid- ed a treated thermoset polymer obtainable by the method outlined above.

According to a third aspect of the present invention, there is provided an article manufactured from the treated thermoset polymer outlined above. The article may have many different forms. In some embodiments, the article is a microfluidic device. Such a device is relatively easy to fabricate by com- parison with conventional methods, whilst exhibiting desirable properties such as solvent resistance.

According to a fourth aspect of the present invention, there is provid- ed a method of forming an article coated with a treated thermoset polymer, the method comprising the steps of:

applying a cured thermoset polymer to one or more surfaces of an article to give a cured thermoset polymer coating;

and subjecting the cured thermoset polymer coating to a process according to step ii) of the first aspect of the present invention.

According to a fifth aspect of the present invention, there is provided a coated article obtainable by the method of the fourth aspect. The article may take many different forms. In some embodiments, the article is a coated re- actor, a coated medical device or a coated flooring.

It will be understood that reference to a“cured thermoset polymer” is intended to mean a polymer that has been treated such that it has some de- gree of crosslinking. This could be any extent of crosslinking density, from low through to high crosslinking density. Polymers which have been cured, but which have the potential to undergo further crosslinking are not excluded.

It will be understood that“solvent resistance” refers to the ability of the polymer to resist change when in contact with solvent. It may be that the treated polymer can be in contact with the solvent for a greater period of time before undergoing any change, or that the extent of change of the polymer at a given time period is less when the solvent resistance is higher.

One way to measure solvent resistance is to create an article from the polymer having one or more channels. A flow of solvent through the channels can then be established, and the channel width measured on a pe- riodic basis or after a given length of time. The higher the solvent resistance of the polymer, the greater extent that the channel width will be maintained. Low solvent resistance is signified by significant swelling of the polymer, thereby leading to a decrease in channel width. High solvent resistance is signified by low levels of swelling, thereby leading to maintenance of the channel width. Such a method is outlined in the examples which follow.

Channel width decrease is given by the following formula:

% width decrease = (( initial width - final width)/ initial width)x100

Brief Description of the Drawings

The present invention will now be described, by way of non-limiting ex- ample, with reference to the following drawings in which:

Figure 1 is a graph showing the impact of duration of exposure to ele- vated temperature on channel width decrease of thiol-ene chips when ex- posed to chloroform;

Figure 2 is a graph showing the impact of duration of exposure to UV radiation on channel width decrease of thiol-ene chips when exposed to chlo- roform;

Figure 3 is a graph showing the impact of hydrogen peroxide concen- tration on channel width decrease of thiol-ene chips when exposed to chloro- form;

Figure 4a is similar to Figure 1 , except that it provides a comparison of different elevated temperatures and different durations;

Figure 4b is analogous to Figure 4a, except that photoinitiator was added to the bulk material when the data was generated;

Figure 5a is a graph showing the impact duration of exposure to chlo- roform has on channel width decrease of thiol-ene chips that have been treated at 200°C for 60 hours;

Figure 5b is analogous to Figure 5a, except that photoinitiator was added to the bulk material when the data was generated;

Figure 6a is similar to Figure 3, except that it provides a comparison of other hydrogen peroxide concentrations and different durations;

Figure 6b is analogous to Figure 6a, except that photoinitiator was added to the bulk material when the data was generated;

Figure 7 a shows combined XPS spectra of a control sample (60 hrs at room temperature), a heat treated sample (60h at 200°C), and a FI2O2 treated sample (16h in 30% FI2O2) and in particular shows the S2p binding region for sulfur detection;

Figure 7b is the individual spectrum from Figure 7a of the control sam- ple, showing presence of unmodified thiols and single oxygen modified thiols, namely sulfoxides;

Figure 7c is the individual spectrum from Figure 7a of the heat treated sample (60h at 200°C) showing no oxygen modified thiols were present;

Figure 7d is the individual spectrum from Figure 7a of the sample treated with hydrogen peroxide (16h in 30% FI2O2) showing all three sulfur species present namely, sulfones, sulfoxides and unmodified thiols;

Figure 8 shows XPS spectra of carbon 1 S/2 for the control sample (60 hrs at room temperature) (top) and XPS spectra of heat treated sample (60h at 200°C) (bottom)- figures show full width at half maximum values for each peak fit;

Figure 9a shows FT-IR spectra of thiol-enes (without photoinitiator) heat treated for 70 hours at 200 ^° C compared with a non-treated control, with changes visible at 1676cnrr ¹ which corresponds to carbonyl groups, possibly from the ester-amide;

Figure 9b is analogous to Figure 9a, but shows FT-IR spectra for thiol- enes heat treated for 10 days at 200 ^° C compared with a non-treated control;

Figure 10 shows Raman spectra of thiol-enes (without photoinitiator) heat treated at 70°C and 130°C for 72 hours and compared with a non- treated control, with change of the alkene being clearly visible (peak around 929 and 670 cm-1 );

Figure 11 a shows mechanical analysis of thiol-enes (without photoiniti- ator), for a non-treated control; Figure 11 b shows mechanical analysis of thiol-enes (without photoiniti- ator), for a sample treated at 200°C for 70 hours; and

Figure 12 is a graph showing the impact of the presence of air or argon during the elevated heat post-polymerization and post-curing treatment on channel width decrease of thiol-ene chips when exposed to chloroform;

Figure 13 shows the results of 1 -hour chloroform exposure on various thiol-ene formulations, before and after 40-hour 200 °C heat treatment.

In all cases, samples were fabricated in accordance with Example 1 below unless indicated otherwise.

Detailed Description

Example 1 - Provision of a cured thermoset polymer by formation of thiol-ene microfluidic chips

Stoichiometric thiol-ene chips were fabricated from thiol monomer (pentaerythritol tetrakis(3-mercaptopropionate)) and tri-allyl monomer (1 ,3,5- triallyl- 1 ,3,5-triazine-2,4,6(1 H,3H,5H)-trione) using the method outlined in Sikanen, T. M., Lafleur, J. P., Moilanen, M. E., Zhuang, G., Jensen, T. G., & Kutter, J. P. (2013) - Fabrication and bonding of thiol-ene-based microfluidic devices; Journal of Micromechanics and Microengineering, 23(3), 037002, the contents of which are herein incorporated by reference.

In particular, polydimethylsiloxane (PDMS, Sylgard® 184) moulds were created. Each chip was moulded in two parts, namely a base part and a top part, such that when the base and the top were assembled they would form a simple chip having a single channel with a width of 500 pm and depth of 250 pm.

Thiol monomer (pentaerythritol tetrakis(3-mercaptopropionate)) and tri- allyl monomer (1 ,3,5-triallyl- 1 ,3,5-triazine-2,4,6(1 H,3H,5H)-trione) were used in a stoichiometric ratio. The components were mixed to form a thiol-ene monomer mixture and degassed. Degassed thiol-ene monomer mixture was then poured into the moulds and the mixture cured for 12 seconds each side under a mercury UV lamp such that it was exposed to radiation of wavelength of 365nm at an intensity of 90mW/cm ² . The chips were assembled by manu- ally aligning and pressing together the top and bottom pieces. Assembled chips were cured for 5 minutes each side at wavelength of 365nm and inten- sity of 90mW/cm ² to give a cured thermoset polymer sample.

Example 2 - Heat Treatment

Cured thermoset polymer was provided by using a chip formed in ac- cordance with Example 1. The chip was then placed in an oven at a tempera- ture of 200°C for 16 hours.

To assess the impact of the treatment, after removing the chip from the oven, chloroform was pumped through the channel of the chip at a flow rate of 10 pL/minute for a period of 1 hour. The channel width was then measured after the exposure to chloroform by taking microscopic images of the chan- nels and channel widths determined from the pixel widths using ImageJ.

It was found that the channel width had decreased by 0.11 ± 0.46%.

A further test was conducted in which cured thermoset polymer was also provided by using a chip formed in accordance with Example 1. The chip was then placed in an oven at a temperature of 150°C for two hours.

To assess the impact of the treatment, after removing the chip from the oven chloroform was pumped through the channel of the chip at a flow rate of 10 pL/minute for a period of 1 hour. The channel width was then measured after the exposure to chloroform and it was found that the channel width had decreased by 2.71 ± 0.48%.

Example 3 - UV exposure

Cured thermoset polymer was provided by using a chip formed in ac- cordance with Example 1. The chip was placed under a mercury UV lamp such that it was exposed to radiation of wavelength of 365nm at an intensity of 90mW/cm ² for a period of 2 hours.

To assess the impact of the treatment, after removing the chip from the lamp, chloroform was pumped through the channel of the chip at a flow rate of 10 pL/minute for a period of 1 hour. The channel width was then measured after the exposure to chloroform and it was found that the channel width had decreased by 3.91 ± 0.93%.

It was observed that the UV exposure also resulted in an increase in temperature of the sample.

Example 4 - Contact with Oxidising Agent

Cured thermoset polymer was provided by using a chip formed in ac- cordance with Example 1. The chip was submerged in 10% (v/v) hydrogen peroxide for a period of 16 hours.

To assess the impact of the treatment, after removing the chip from the hydrogen peroxide, chloroform was pumped through the channel of the chip at a flow rate of 10 pL/minute for a period of 1 hour. The channel width was then measured after the exposure to chloroform and it was found that the channel width had decreased by 5.84 ± 0.29%.

Comparative Example 1

By way of reference, when a chip was fabricated in accordance with Example 1 , but no further (post-polymerization or post-curing) treatment ap- plied and chloroform was pumped through the channel of the chip at a flow rate of 10 pL/minute for a period of 1 hour, the channel width deacrease was found to be: 25.16 ± 0.51 %.

Example 5 - Inclusion of Photoinitiator

Further chips were formulated in accordance with Example 1 , except that 0.5% (v/v) photoinitiator, namely TPO-L, was added to the monomer mix- ture. TPO-L is 2,4,6-trimethylbenzoylphenyl phosphinate and is available un- der the brand name Omnirad®. The process also differed from Example 1 in that the mixture was only cured for 2 seconds on the non-bonding sides and under an intensity of 17 mW/cm ² . Otherwise, the process of chip formulation was the same as for Example 1.

The photoinitiator-containing chips were then exposed to the conditions of similar to that of Examples 1 - 4. The heat treatment gave the following reductions in channel width: 16 hours at 200 °C: 0.01 ± 0.17%

1 hour at 200 °C: 0.79 ± 0.23%

The UV exposure of Example 3 gave the following reduction in channel width: 0.73 ± 0.09%

The oxidizing agent process of Example 4 gave the following reduction in channel width: 0.89 ± 0.20%.

Comparative Example 2

By way of reference, when a chip was fabricated in accordance with Example 1 , but including photoinitiator as outlined in Example 5, and no fur- ther (post-polymerization or post-curing) treatment applied, when chloroform was pumped through the channel of the chip at a flow rate of 10 pL/minute for a period of 1 hour, the channel width decrease was found to be: 6.07 ± 0.46%.

Example 6 - Impact of duration of heat treatment on solvent resistance

Five chips were fabricated according to the method outlined in Exam- pie 1.

The first of these chips was then set aside as a control. The four re- maining chips were heated to a temperature of 180°C using an oven. One of the chips was held at this temperature for 10 minutes, another of the chips was held at this temperature for 20 minutes, a third chip was held at this tem- perature for 40 minutes and the final chip was held at this temperature for 2 hours.

In order to measure the chloroform resistance, after removal from the oven, chloroform was pumped through the channels of each chip at the flow rate of 10 pL/minute for 1 hour or until delamination or clogging of the chip occurred.

All samples were conducted in triplicates and percent (%) difference in channel widths were plotted in GraphPad Prism. The decrease in channel width for each of the chips is shown in Figure 1. Error bars represent stand- ard deviation.

Example 7 - Impact of duration of UV exposure on solvent resistance

Four chips were fabricated according to the method outlined in Exam- pie 1.

The first chip was set aside without further treatment. The second chip was exposed to the UV radiation for a period of 20 minutes in total i.e. 10 minutes longer than the control. The third chip was exposed to the UV radia- tion for 40 minutes i.e. 30 minutes longer than the control. The final chip was exposed to the UV radiation for 2 hours i.e. 1 hour and 50 minutes longer than the control. UV radiation was provided in the same way as for Example 3.

In order to measure the chloroform resistance, after removal from the UV lamp, chloroform was pumped through the channels of the chip at the flow rate of 10 pL/minute for 1 hour or until delamination or clogging of the chip occurred. All samples were conducted in triplicates and percent (%) differ- ence in channel widths were plotted in GraphPad Prism. The decrease in channel width for each of the chips is shown in Figure 2. Error bars represent standard deviation.

Example 8 - Impact of concentration of oxidising agent on solvent resistance

Three chips were fabricated according to the method outlined in Ex- ample 1.

The first of these chips was then set aside as a control. The second and third chips were submerged in an aqueous solution of hydrogen peroxide for a period of one hour, at room temperature, in the dark. One of the chips was submerged in 10% (v/v) hydrogen peroxide and the other was sub- merged in 20% (v/v) hydrogen peroxide.

In order to measure the chloroform resistance, after removal from the hydrogen peroxide, chloroform was pumped through the channels of the chip at the flow rate of 10 pL/minute for 1 hour or until delamination or clogging of the chip occurred. All samples were conducted in triplicates and percent (%) difference in channel widths were plotted in GraphPad Prism. The decrease in channel width for each of the chips is shown in Figure 3. Error bars repre- sent standard deviation.

Example 9 - Impact of temperature of heat treatment on solvent resistance

Further experiments similar to those of Example 6 were conducted to assess the impact of temperature on the effects of the heat treatment. Three sample groups were established, each consisting of three chips fabricated as outlined in Example 1. All of the chips of the first group were heated to 100°C. The first chip was held at this temperature for 1 hour, the second was held at this temperature for 2 hours and the third was held at this temperature for 16 hours.

This was then repeated for the second group, in which all of the chips were heated to 150°C, with one of the chips being held at this temperature for 1 hour, another being held at this temperature for 2 hours and the remaining chip being held at this temperature for 16 hours.

A similar process was then repeated for the third group, in which all of the chips were heated to 200°C, with one of the chips being held at this tem- perature for 1 hour, another being held at this temperature for 2 hours and the remaining chip being held at this temperature for 16 hours.

A control was also prepared, with the chip being fabricated as outlined in Example 1 but with no further (post-polymerization or post-curing) treat- ment.

In order to measure the chloroform resistance of all of the chips, chlo- roform was pumped through the channels of the chip at the flow rate of 10 pL/minute for 1 hour or until delamination or clogging of the chip occurred.

The results are shown in Figure 4a.

To assess the impact of photoinitiator, the experiments above were all repeated on chips fabricated in accordance with Example 5. The results are shown in Figure 4b.

Example 10 - Extended chloroform exposure to heat treated sample

In order to assess the duration of solvent resistance of a heat treated sample, a chip was prepared according to Example 2, and a further chip was prepared in a similar way except that it was treated for 60 hours at 200°C ra- ther than 16 hours. Chloroform was pumped through the channels of the chips at the flow rate of 10 pL/minute for a period of 48 hours and periodic measurements of the channel width were taken. The results are shown in Figure 5a, with an untreated control for comparison.

Figure 5b shows the results of an analogous process except where the samples contained photoinitiator.

Example 11 - Comparison of impact of concentration of oxidising agent on solvent resistance over time

Further experiments similar to Example 8 were conducted, using vary- ing concentrations of hydrogen peroxide. The results are shown in Figure 6a.

Figure 6b shows the results of an analogous process except where the samples contained photoinitiator.

Discussion

Figures 7a - 7d were generated to try and better understand the mechanisms involved. They suggest that oxidation of sulfur may be a factor, but that the extent may vary depending on the treatment method.

Figure 10 shows a change in the alkene peaks after heat treatment.

With regard to Figure 11 , it can be seen that the glass transition tem- perature of the untreated sample is 51 ^° C while the heat treatment increased it to 118 ^° C. It is understood from this that the heat treatment is increasing the crosslinking density of the polymer. The storage modulus (comparable to Young’s modulus = elasticity) at room temperature significantly increases from 16 to 50 GPa.

Figure 12 shows the results when 0.5% TPO-L chips were heated at 200°C for 60 hours in the presence of air or under argon. Controls are chips at room temperature for 60 hours. Chips were exposed to chloroform for 16 hours and channel swelling measured. n=3. The results indicate that heating in an oxygen containing environment is desirable.

As shown in Fig. 13, the following were tested: a control, triallyloxy- triazine with PTMP, NOA-81 commercial glue, and Ostemer 322 commercial thiol-ene. All monomers are stoichiometric with regards to the allyl and thiol functional groups. All mixed monomers contain 0.5% TPO-L photoinitiator. All chips were cured under 90 mW/cm ² for 10 minutes after assembly. Both Ostemer 322 conditions were heated for 1 hour at 110°C as suggested by the manufacturer. All samples run in triplicates, error bars represent standard de- viation.

Assistance was provided in generating the graphs of Figs. 7a-d, 8, 10 and 11. In particular, Figs 7a-d, and 8 were generated with assistance from Nicolas Emile Bovet, Department of Chemistry, Copenhagen University. Fig. 10 was generated with assistance from Magnus Edinger, Department of Pharmacy, Copenhagen University. Fig. 11 was generated with assistance from Eric Ofosu Kissi, Department of Pharmacy, Copenhagen University.