WO2012166053A1

WO2012166053A1 - A filtering membrane

Info

Publication number: WO2012166053A1
Application number: PCT/SG2012/000191
Authority: WO
Inventors: Shashi Ranjan; Yong Zhang
Original assignee: National University Of Singapore
Priority date: 2011-05-31
Filing date: 2012-05-31
Publication date: 2012-12-06

Abstract

There is provided a filtering membrane and method of forming thereof. The filtering membrane comprises a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least one side of the substrate, the arresting formations being configured to arrest the ingress of particulate entities of a selected particle oversize from entering the holes. There is also provided a mold and method of forming thereof as well as a microfluidic device incorporating the filtering membrane.

Description

A Filtering Membrane

Technical Field

The present invention generally relates to a filtering membrane and a method of forming the same. The present invention also relates to a mold and a method of forming the same. The present invention also relates to a microfluidic device.

Background

Recent technological developments have provided a variety of micro-scale devices for handling biologically relevant micron-sized entities, like cells or micro- beads, to satisfy ever growing biological and medical needs. Many of such biological and medical needs require either sorting or separation or patterning/trapping of these micron-sized entities and have attracted attention for developing new micro-devices.

A few examples of biological and medical separations include separation ^: of circulating tumor cells or epithelial cells from blood, separation of white blood cells from whole blood, separation of blood-cell subtypes, and isolation of stem cells from amniotic fluids. An example of cell-patterning for basic studies on cell biology include individual cell studies. In addition, separation and patterning of micro-beads conjugated with functional bio-molecules has provided a new tool for immunoassays .

Micro-particle separation has been achieved by various types of micro-devices based on either intrinsic biophysical differences (such as size, shape, polarisability or charge) or extrinsic differences (such as magnetic labeling of different cells) . Porous membrane -teehn-o-logy—h s—-been ex.t.e.aaive1y emp1oye_d___fo_r_ _separation of particles. Track-etched membranes with different pore sizes are commercially available. However, track-etched membranes have randomly placed pores with variations in pore shape and size and have relatively low pore density. Fusion of two or more pores can also be found in track- etched membranes which reduces the efficiency of filtration. These membranes cannot be fabricated with desired structures around the pore. Also, fabrication of such membranes requires expensive equipment or harsh chemicals for etching and hence is not suitable for high- end applications.

Other methods can be employed for fabrication of uniform-sized pores, such as etching of polymers or silicon-based substrate by photo-lithographically defined masks. However, these methods are also unsuitable for fabrication of three-dimensional structures and they require clean-room facilities for fabrication of such membranes .

Overall, fabrication of such membranes may be expensive, complicated and inefficient.

Additionally, porous membrane technology has a major problem of clogging or fouling during a filtration process. The clogging of pores results in increased pressure build-up and eventually failure of membrane function.

Currently, soft-lithography has been identified to be the most popular method for fabrication of microfluidic devices as it is inexpensive, fast and efficient. Once a mold having desired structures is fabricated . using clean-room facilities, many devices can be fabricated from this mold simply by replica-molding in any lab environment. However, while structures on the mold can easily be replicated by normal soft-lithography, -p-Q-r-o-us—membranes ma.de by this meJ:hod require some modifications after replication. An example of a modification that is required is that control of the thickness of the substrate layer^' is needed in order to make through-holes in the porous membrane produced.

In the past, attempts have been made to modify soft- lithography to fabricate porous membranes. However, there are no known methods of fabricating porous membranes with three-dimensional structures. Typically, an array of pillars or posts are used to create through-holes as pores in a thin film of polymer using a gas blowing method. However, the gas-blowing method is an empirical approach and is thus difficult to standardize and optimize this method. Hence, the efficiency of reproducing a large number of pores may not high.

Patterning of micro-particles has been demonstrated either by creating micro-traps using properties such as electrical, mechanical or optical, or by using chemically modified micro-patterned surfaces. While some of these approaches (like electrical or mechanical traps) may be used for achieving simultaneous functionalities, the efficiency of such approaches is low and sensitive to the process parameters. Further, mechanical traps using pillars, wire or ^VC shaped structures suffer from low throughput, clogging and incomplete separation.

In view of the above, different approaches have been developed for targeting the individual problems of patterning and separation. However, there are no such systems that demonstrate an ability to target separation and patterning at high efficiency at the same time.

There is therefore a need to provide a membrane that overcomes, or at least ameliorates, one or more of the disadvantages described above.

There is also a need to provide a simple and e.f-f.i-ci-en±_me_t . cLof fabricating such membranes . Summary

According to a first aspect, there is provided a filtering membrane comprising a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least one side of the substrate, the arresting formations being configured to arrest the ingress of particulate entities of a selected particle oversize from entering the holes.

Advantageously, the filtering membrane may not suffer from problems such as pore-clogging or filter fouling.

Advantageously, the filtering membrane may be used to filter a plurality of particulate entities or particles having different particle sizes so as to separate or pattern the particulate entities or particles. The particulate entities or particles that have a smaller particle size than the holes can pass through the holes while those having a larger particle size are trapped by the arresting formations. Although the larger particles are trapped by the arresting formations, due to the interstitial gaps in between adjacent arresting formations, smaller particulate entities or particles and/or fluid can still pass through the holes. Hence, the filtering membrane may not experience high pressure drops as compared to another membrane that does not have the arresting formations thereon.

Advantageously, the arresting formations may confer multi-functionalities to the filtering membrane such as enhanced^■ filtration, high efficiency patterning of particles and further functionality due to non-blockage e-f—-heie-s--due—-©--th-e-^ the trapped. particles may be separated from the un-trapped particles (that is, those particles which pass through the holes) while not restricting the flow of smaller sized particles and fluid. ■

Advantageously, the ability to trap particles while allowing fluid to flow around the trapped particles to the holes may allow for enhanced interaction (due to a graeter contact area) between the trapped particles and fluid. The trapped particles may be functionalized with, for example, protein, antibody or DNA, while the fluid may contain target analytes that can bind to the trapped particles while flowing to the holes.

According to a second aspect, there is provided a method for forming a filtering membrane comprising a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least one side of the substrate, the method comprising the steps of:

a. subjecting a mold comprised of a layer of a cross-linkable material to a cross-linking treatment to cross-link the cross-linkable material and to thereby form imprint formation structures thereon, wherein the degree of cross-linking throughout the mold is unequal throughout; and

b. applying a layer of the substrate to the mold under conditions to form imprints on the substrate that are complementary to the imprint formation structures

According to a third aspect, there is provided a mold having imprint formation structures thereon, the imprint formation structures having a first region . that has a substantially uniform cross-sectional area along the length of the first region and a second region that has a varying cross-sectional area along the length of ir^^~s^corrd^~¾^~g±on^": According to a fourth aspect, there is provided a method for forming a mold having imprint formation structures thereon, the method comprising the step of subjecting a mold comprised of a layer of a cross- linkable material to a cross-linking treatment to crosslink the cross-linkable material and to thereby form imprint formation structures thereon, wherein the degree of cross-linking throughout the mold is unequal throughout.

According to a fifth aspect, there is provided a microfluidic device comprising a channel for flow of a plurality of particulate entities having ^' different particle sizes therein; and a membrane in fluid communication with the flow channel, the membrane having a plurality of holes that extend through the membrane, and arresting formations provided around the openings of the holes on at least one side of the membrane.

According to a sixth aspect, there is provided a substrate having three-dimensional structures thereon, wherein the three-dimensional structures have a varying cross-sectional area along the length of the structures.

According to a seventh aspect, there is provided a method for forming a substrate having three-dimensional structures thereon, the method comprising the step of subjecting the substrate comprised of a layer of a cross- linkable material to a cross-linking treatment to crosslink the cross-linkable material and to thereby form three-dimensional structures thereon, wherein the degree of cross-linking throughout the substrate is unequal. Definitions

The following words and terms used herein shall have the meaning indicated:

The terms "particulate entity" or "particle" are to be interpreted broadly to include a variety of materials or substances that can be characterized in terms of its dimensions or size. The particulate entity or particle may include, but is not limited to, cells, both eukaryotic (e.g., leukocytes, erythrocytes or fungi) and prokaryotic (e.g., bacteria, protozoa or mycoplasma), viruses, cell components, macromolecules , microparticles , beads (including microbeads), etc.

The terms "cross-link" or "cross-linked" refer to an interconnection (usually chemical bonding) between monomers or co-monomers making up a polymer or between two types of discrete polymers.

The term "organic polymer" as used herein, refers to a polymeric material which has repeating units of a backbone composed mainly of carbon atoms or a ring containing carbon atoms, and which also contains hydrogen. The polymeric material may also contain other elements such as, for . example, sulphur, oxygen or nitrogen .

The terms "micron-sized" or "micron-range" refer to a dimension that is more than about 1 pm to about 1000 pm.

The word "substantially" does not exclude "completely" e.g. a composition which is "substantially free" from Y may be completely free from Ύ. Where necessary, the .word "substantially" may be omitted from the definition of the invention.

Unless specified otherwise, the terms "comprising" and "comprise", and grammatical variants thereof, are intended to represent "open" or "inclusive" language such that they include recited elements but also permit inclusion of additional, unrecited elements.

As used herein, the term "about", in the context of concentrations of components of the formulations, typically means +/- 5% of the stated value, more typically +/- 4% of the stated value, more typically +/- 3% of the stated value, more typically, +/- 2% of the stated value, even more typically +/- 1% of the stated value, and even more typically +/- 0.5% of the stated value.

Throughout this disclosure, certain embodiments may be disclosed in a range format. It should be understood that the description in range format is merely for convenience and brevity and should not be construed as an inflexible limitation on the scope of the disclosed ranges. Accordingly, the description of a range should be considered to have specifically disclosed all the , possible sub-ranges as well as individual numerical values within that range. For example, description of a range such as from 1 to 6 should be considered to have specifically disclosed sub-ranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6 etc., as well as individual numbers within that range, for example, 1, 2, 3, 4, 5, and 6. This applies regardless of the breadth of the range.

Certain embodiments may also be described broadly and generically herein. Each of the narrower species and subgeneric groupings falling within the generic disclosure also form part of the disclosure. This includes the generic description of the embodiments with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein. Detailed Disclosure of Embodiments

Exemplary, non-limiting embodiments of a filtering membrane will now be disclosed. The filtering membrane comprises a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least one side of the substrate, the arresting formations being configured to arrest the ingress of particulate entities of a selected particle oversize from entering the holes.

The holes may allow the ingress of particulalte entities or particles of a selected particle undersize to enter the holes.

The arresting formations may be integrally formed with, and extend from, the substrate. The arresting formations may be formed at the same time as the holes.

The arresting formations may be three-dimensional (that is, 3-D) structures that extend from at least one surface of the substrate. A number of arresting formations may be chosen as appropriate to surround a hole while trapping a particulate entity (or particle) above the hole. The number of arresting formations that surround a hole may be at least three, at least four, at least five or at least six.

The shape of the arresting formations is not particularly limited as long as they define a region, typically above the hole, for trapping a particulate entity or particle, while allowing fluid (and optionally smaller sized particulate entities) to enter the holes from the sides of the arresting formations. The arresting -formations may prevent the^' escape of the trapped particles when subjected to a wash. Due to the ability of the filtering membrane to allow fluid through the hole, the filtering membrane may not experience high pressure-dr_o_, which would otherwise be experienced by another filtering membrane without the arresting formations. The decrease in the pressure drop (as compared to another filtering membrane without the arresting formations) may be due to the arresting formations which trap the larger sized particles such that they do not enter the holes and get trapped there.

The arresting formations may generally taper from a base portion adjacent the holes to a smaller end portion. As such, the distance between adjacent arresting formations surrounding a hole may not be equal when extending from the base portion to the end portion. The shape of the arresting formations may be a pyramidal shape with an inclined surface being presented to the ^■ particulate entity. Other shapes may include frustum- shape, dome-shape, doll-shape, etc.

The arresting formation may be of the same material as the substrate. Hence, the arresting formations and substrate may be comprised of an organic polymer. The organic polymer may comprise monomers selected from the group consisting of methylesiloxanes , ethylenes, propylenes, methyls, pentenes, amides, sulphones, ethersulphones, esters, carbonates, acrylonitriles , butadienes, styrenes, imides, amic acids, phenols, styrene sulphonic acids, acrylic acids, methacrylic acids, saccharides, thiophenols and combinations thereof.

The organic polymer may be selected from the group consisting of polydimethylsiloxanes ; polyolefins such as polyethylene, polypropylene, polymethylpentene; polyamides and polyimides, including aryl polyamides and aryl polyimides; polystyrenes or substituted polystyrenes; polysulphones such as polyethersulphones; polyesters such as polyethylene terephthalates, polybutylene terephthalates; polyacrylates ; polycarbonates ; pgJLyamic__acids ; polyphenols; polystyrene sulphonic acids; polyacrylic acids; polymethacrylic acids; polythiophenols; polysaccharides such as agarose and alginate as well as co-polymers such as polyacrylic acid/butadiene copolymer or polystyrene/butadiene copolymer.

The arresting formations may be dimensioned in the micron-range. Accordingly, the height, length of the base portion and length of the end portion may be dimensioned in the micron-range. It is to be noted that the dimensions of the arresting formations are not particularly limited and is adjustable according to the size of the target particulate entity that is to be trapped by the arresting formations. The size and shape of the arresting formations may be dependent on the size and shape of the imprint formation structures of the mold material when forming the filtering membrane (as will be discussed further below) .

The peak-to-peak distance between adjacent arresting formations may also be in the micron-range.

The holes may have a substantially consistent cross- sectional shape when extending through the substrate. The cross-sectional dimension of the holes may be in the micron-range. The holes may be arranged throughout the substrate in a random distribution or in a non-random distribution .

Exemplary, non-limiting embodiments of a method for forming a filtering membrane comprising a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least, one side of the substrate will now be disclosed. The method comprises the steps of (a) subjecting a mold comprised of a layer of a cross-linkable material to an anisotropic cross- -1-inJci-ng txe3i.m_e.nt_ to cro_ss^link the cross-linkable^' material, wherein the degree of cross-linking throughout the mold is unequal, to thereby form imprint formation structures thereon; and (b) _. applying a layer of the substrate to the mold under conditions to form imprints on the substrate that are complementary to the imprint formation structures.

The anisotropic cross-linking treatment may comprise the step of exposing the mold to one of ultraviolet (UV) light or ■ ionizing radiation. The extent of desirable cross-linking may be dependent on the exposure energy of the cross-linking treatment, which in turn may depend on the time of the exposure based on the power of the light- energy source. In an embodiment where UV light is used, the time of exposure to UV light may be selected from the range of about 1 second to about 2 minutes, about 10 seconds to about 2 minutes, about 20 seconds to about 2 minutes, about 30 seconds to about 2 minutes, about 40 seconds to about 2 minutes, about 50 seconds to about 2 minutes, about 60 seconds to about 2 minutes, about 70 seconds to about 2 minutes, about 80 seconds to about 2 minutes, about 90 seconds to about 2 minutes, about 100 seconds to about 2 minutes, about 110 seconds to about 2 minutes, about 1 second to about 10 seconds, about 1 second to about 20 seconds, about 1 second to about 30 seconds, about 1 second to about 40 seconds, about 1 second to about 50 seconds, about 1 second to about 60 seconds, about 1 second to about 70 seconds, about 1 second to about 80 seconds, about 1 second to about 90 seconds, about 1 second to about 100 seconds, about 1 second to about 110 seconds, about 1 second to about 5 seconds, about 20 seconds to about 40 seconds. The mold may be exposed to the UV light for about 30 seconds.

The power used during the UV light treatment may be s-e-Le.ct_e_d,__f_rom the range of about 1 mW/cm² to about 10 mW/cm², about 1 mW/cm² to about 2 mW/cm², about 1_. mW/cm² to about 3 mW/cm², about 1 mW/cm² to about 4 mW/cm², about 1 mW/cm² to about 5 mW/cm², about 1 mW/cm² to about 6 mW/cm², about 1 mW/cm² to about 7 mW/cm², about 1 mW/cm² to about 8 mW/cm², about 1 mW/cm² to about 9 mW/cm², about 2 mW/cm² to about 10 mW/cm², about 3 mW/cm² to about 10 mW/cm², about 4 mW/cm² to about 10 mW/cm², about 5 mW/cm² to about 10 mW/cm², about 6 mW/cm² to about 10 mW/cm², about 7 mW/cm² to about 10 mW/cm², about 8 mW/cm² to about 10 mW/cm² and about 9 mW/cm² to about 10 mW/cm².

The cross-linkable material may be exposed to the UV light through a photo-mask, which is patterned as desired.

The cross-linking treatment may comprise the step of adding a second layer of a cross-linkable material to the first layer to thereby allow diffusion of cross-linkable material from the first layer to the second layer^'. The cross-linkable material in the two layers may be of the same material. The first layer may be treated with ultraviolet (UV) light or ionizing radiation as described above. The second layer may not require any cross-linking treatment as the cross-linking species from the first layer is able to diffuse to the second layer, that is, the cross-linking species is able to diffuse from the exposed region to the unexposed region. Alternatively, in order to increase the degree of anisotropic crosslinking; a cross-linking treatment may be applied to the second layer .

The cross-linkable material may be a photosensitive material. The cross-linkable material may be a negative cross-linking photoresist. The negative cross-linking photoresist may be SU-8, AZ negative photoresist., poly (vinyl cinnamate) , Novolaks or poly ( t-Boc styrene) . The thickness of the cross-linkable material is not especially limited and depends on the desired height of the arresting formations.

The method may comprise the step of providing the layer of the cross-linkable material onto a support material. The support material may be used to form channels for the subsequent micro-fluidic device. The support material may be silicon wafer or may be the same material as the cross-linkable material.

The method may comprise the step of providing a subsequent layer of cross-linkable material before or after the subjecting step (a). This layer may be treated and developed to form pillars that,^' when applied to the substrate, form complementary holes in the substrate. Accordingly, the mold may comprise imprint formation structures that comprise two regions, the first region containing the pillars and the second region containingthe cross-linked 3D structures. When this mold is applied to the substrate, the pillars of the mold form corresponding holes in the substrate while the cross- linked structures of the mold contribute to the formation of the arresting formations on the substrate.

The cross-linkable material may be deposited onto the support material or first cross-linkable layer by spin-coating. After each layer of cross-linkable material is deposited, the cross-linkable material may be subjected to a soft-baking step (or pre-exposure baking step) . This soft-baking step- may aid in the evaporation of solvents which are used to dissolve the material. The solvent allows spreading of the material on a surface (like silicon wafer) during spin-coating, but has to be removed after spin-coating as the solvent may interfere with the cross-linking step. The soft-baking step may be &a-r-r-i-ed- out—f-o-r—-a— eriod—-of—time—s-eLected—.from—t.he__ aag_e of about 1 minute to about 20 minutes, about 1 minute to about 5 minutes, about 1 minute to about 10 minutes, about 1 minute to about 15 minutes, about 5 minutes to about 20 minutes, about 10 minutes to about 20 minutes and about 15 minutes to about 20 minutes. The soft-baking step may be conducted at a temperature selected from the range of about 60°C to about 100°C, about 65°C to about 100°C, about 70°C to about 100°C, about 75°C to about 100°C, about 80°C to about 100°C, about 85°C to about 100°C, about 90°C to about 100°C, about 95°C to about 100°C, about 60°C to about.65°C, about 60°C to about 70°C, about 60°C to about 75°C, about 60°C to about 80°C, about 60°C to about 85°C, about 60°C to about 90°C and about 60°C to about 95°C. The soft-baking step may be carried out over two time periods and temperature. As such, the soft- baking step may be carried out for about 1 minute to about 3 minutes at a temperature of 65°C and then rampedto 15 minutes at a temperature of 95°C.

After the cross-linking treatment, the method may comprise the step of baking the exposed layer (or postexposure baking step) . Dependent on the time and temperature of the post-exposure baking step, this may either control the degree of cross-linking in the exposed region or complete the cross-linking process. In order to control the cross-linking process, the post-exposure bake can be carried out for a shorter period of time (such as from about 30 seconds to about 2 minutes) and a lower temperature (such as from about 60°C to about 95°C) as compared to the post-exposure bake for completing the cross-linking (which . can take place at about 15 minutes at 96°C) . Typically, a final post-exposure bake is applied to complete the cross-linking process which may involve anisotropic cross-linking of partially cross- li-nJ ed— spaci-as Hence, dependent on the sequence of the mold fabrication process, the post-exposure baking step may be carried out for a period of time from about 30 seconds to about 2 minutes, about 40 seconds to about 2 minutes, about 50 seconds to about 2 minutes, about 60 seconds to about 2 minutes, about 70 seconds to about 2 minutes, about 80 seconds to about 2 minutes, about 90 seconds to about 2 minutes, about 100 seconds to about 2 minutes, about 110 seconds to about 2 minutes, about 30 seconds to about 40 seconds, about 30 seconds to about 50 seconds, about 30 seconds to about 60 seconds, about 30 seconds to about 70 seconds, about 30 seconds to about 80 seconds, about 30 seconds to about 90 seconds, about 30 seconds to about 100 seconds and about 30 seconds to about 110 seconds. The temperature during this period may be from about 60°C to about 95°C, about 65°C to about 95°C, ^' about 70°C to about 95°C, about 75°C to about 95°C, about 80°C to about 95°C, about 85°C to about 95°C, about 90°C to about 95°C, about 60°C to about 65°C, about 60°C to about 70°C, about 60°C to about 75°C, about 60°C to about 80°C and about 60°C to about 85°C. The post-exposure bake may be carried out at two temperatures for two time periods, such as for one minute at 65°C and then for 1 minute at 95°C.

The final post-exposure baking step may be carried out for about 10 minutes to about 20 minutes, about 12 minutes to about 20 minutes, about 14 minutes to about 20 minutes, about 16 minutes to about 20 minutes, about 18 minutes to about 20 minutes, about 10 minutes to about 12 minutes, about 10 minutes to about 14 minutes, about 10 minutes to about 16 minutes and about 10 minutes to about 18 minutes. The temperature may be selected from about 95°C to about 100°C, about 95°C ^' to about 96°C, about 95°C to about S7°C,—about 95°C to about 98^;;C.,.._. about __.95^r:C -„c about 99°C, about 96°C to about 100°C, about 97°C to about 100°C, about 98°C to about 100°C and about 99°C to about 100°C. The final post-exposure, bake may include baking for about 3 minutes at 65°C, ramping to 96°C within 2 minutes and then 15 minutes at this temperature.

After the post-exposure bake, the mold may be subjected to a cooling step. The cooling step may be carried out for about 25 to about 35 minutes, or 30 minutes.

The cooled mold may be developed in order to obtain the imprint formation structures. The cooled mold may be developed in a developer solution which dissolves the material which is not cross-linked such that only the desired cross-linked material is left as the mold.

The applying step (b) may comprise the step of spin- coating the substrate layer onto ^'the mold.

The substrate layer may be subjected to a baking step in order to speed up the cross-linking process and solidify the substrate layer. The baking step may also allow the imprint formation structures to indent into the substrate and form the corresponding holes and arresting formations. The substrate layer may be baked for a certain period of time (such as from about 5 minutes to about 30 minutes) at a certain temperature (such as from 50°C to about 70°C) .

The method may further comprise the step of removing the substrate layer from the mold. The substrate layer may be removed from the mold through the use of solvents such as Iso-propanol or subjecting to heat at, for example, 70°C for about .30min.

The removed substrate layer may be subjected to a plasma treatment. The plasma treatment may result in the generation of charged species on the surface of the removed—s-ubs-txa±-e Layjer so s___to_ allow _^bonding of two layers of the substrate. The plasma treatment may be an oxygen-plasma treatment for a period of about 20 seconds to about 1 minute at a power of about 150 W to about 250 W. In one embodiment, the oxygen plasma treatment is carried out for 30 seconds at 200 W.

Exemplary, non-limiting embodiments of a mold having imprint formation structures thereon will now be disclosed. The mold comprises imprint formation structures which have a first region that has a substantially uniform cross-sectional area along the length of the first region and a second region that has a varying cross-sectional area along the length of the second region.

The mold may be comprised of a cross-linkable material. The cross-linkable material may be a photosensitive material. The cross-linkable material may be a negative cross-linking photoresist. The negative cross-linking photoresist may be SU-8, AZ negative photoresist, poly (vinyl cinnamate) , Novolaks or poly(t- Boc styrene) .

The first region of the imprint formation structures may be pillars which when applied to a substrate, form corresponding holes in the substrate. The pillars■ may be randomly or non-randomly distributed throughout the mold. The cross-sectional shape of the pillars is not particularly limited and may be circular, squarish or any other shape. The cross-sectional dimension (length or diameter as appropriate) of the pillar may be in the micro-scale.

.The second region results in the formation of the arresting formations on the substrate. The second region may be 3-D structures that are formed of cross-linked species. The mold may be subjected to a cross-linking r-ea-t-me-n-t—a-s—defined above in order to form the 3-D structures. The 3-D structures may have a varying cross- sectional area along the length of the second region such that the 3-D structures taper from one end towards the other end. The base portion may have a greater dimension as compared to the end portion, which is connected to the first region. The 3-D structures converge at the end portion. The dimensions of the 3-D structures may be in the micron-range.

Exemplary, non-limiting embodiments of a method for forming a mold having imprint formation structures thereon will now be disclosed. The method comprises the step of subjecting a mold comprised of a layer of a cross-linkable material to an anisotropic cross-linking treatment to cross-link the cross-linkable material, wherein the degree of cross-linking throughout the mold is unequal, to thereby form imprint formation structures thereon.

The- anisotropic cross-linking treatment may comprise the step of exposing the mold to one of ultraviolet (UV) light or ionizing radiation. The conditions for the UV light treatment are as described above.

The cross-linkable material may be exposed to the UV light through a photo-mask, which is patterned as desired .

The cross-linking treatment may comprise the step of adding a second layer of a cross-linkable material to the first layer to thereby allow diffusion of cross-linkable material from the first layer to the second layer. The cross-linkable material in the two layers may be of the same material. The first layer may be treated with ultraviolet (UV) light or ionizing radiation as described above. The second layer may not require any cross-linking treatment as the cross-linking species from the first la-yer is_^_able__to diffuse to the second layer, that is, the cross-linking species is able to diffuse from the exposed region to the unexposed region. Alternatively, in order to increase the degree of anisotropic crosslinking, a cross-linking treatment may be applied to the second layer.

Exemplary, non-limiting embodiments of a microfluidic device will now be disclosed. The microfluidic device comprises a channel for flow of a plurality of particulate entities having different particle sizes therein; and a membrane in fluid communication with the flow channel, the membrane having a plurality of holes that extend through the membrane, and arresting formations provided around the openings of the holes on at least one side of the membrane.

The membrane may be as described above with regard to the filtering membrane.

The microfluidic device may be used to separate ■ particles that are larger than the holes from those that are smaller than the holes. The smaller particles that pass through ^' the holes may be collected in a collector. The larger particles may be retained by the arresting formations while allowing fluid (and optionally smaller particles)^' to pass through the holes. By entrapping the larger particles while being exposed to the fluid, interactions between the particles and the fluid can take place. For example, the particle may be tagged with a ligand that reacts with a target analyte such that if the target analyte is present in the fluid sample, the target analyte will bind to the ligand and this interaction between the ligand and its target analyte can be dectected.

The microfluidic device may be used for the patterning of microparticles or microbeads . The particulate entities are not particularly limited and may include microbeads, cells, bacteria, fungi, yeasts, virus.

There is also provided a substrate having three- dimensional structures thereon, wherein the three- dimensional structures have a varying cross-sectional area along the length of the structures. This substrate may be of the same material as the mold described above. The three-dimensional structures of this substrate are similar to the imprint formation structures of the mold and may be formed as described above.

There is also provided a method for forming a substrate having three-dimensional structures thereon, the method comprising the step of subjecting the substrate comprised of a layer of a cross-linkable material to a cross-linking treatment to cross-link the cross-linkable material and to thereby form three- dimensional structures thereon, wherein the degree of cross-linking throughout the substrate is unequal. The cross-linking treatment is as described above.

Brief Description Of Drawings

The accompanying drawings illustrate a disclosed embodiment and serves to explain the - principles of the disclosed embodiment. It is to be understood, however, that the drawings are designed for purposes of illustration only, and not as a definition of the limits of the invention.

Fig. 1(a) shows a diagram of a photo-mask with transparent circles. Fig. 1(b) shows a diagram of a photo-mask with opaque circles.

Fig. 2 (a) (i) is a schematic diagram showing the arresting formations trapping particulate entities of different particle^' sizes while allowing. smaller particulate entities to pass through the holes. Fig. 2 (a) (ii) is a schematic diagram of a single trapped particulate entity surrounded by a number of arresting formations .

Fig. 2(b) (i) is a schematic diagram of one process to make the master mold by way of UV light exposure. Fig. 2 (b) (ii) is a schematic diagram of another process to make the master mold by way of diffusion. Fig. 2(c) is a schematic diagram of the process to form the substrate.

Fig. 3(a) (i) (left side) shows a top view schematic representation of a microfluidic device incorporating the filtering membrane while the right side shows the bottom view of the microfluidic device. Fig. 3(a) (ii) is an expanded view of the filtering membrane. Fig. 3(b) is a schematic diagram of the fluid flow in the microfuidic device.

Fig. 4(a) shows a scanning electron microscopic (SEM) micrograph image of the master-mold obtained from Example 2 at Ι,ΟΟΟχ magnification. Fig. 4(b) shows an SEM image of the porous PDMS film being peeled off from the master-mold from Example 3 at 55x magnification.. Figs. 4(c) to (f) show the SEM images of the top, bottom, side and the 45°-angled view of the final porous membrane film obtained from Example 3 at l,100x, l,200x, l,100x and Ι,ΙΟΟχ magnification respectively.

Fig. 5 shows an SEM^■ image and the associated measurements of a trap of the porous membrane obtained from Example 3.

Fig. 6(a) shows an SEM image at 850x magnification of the un-patterned clumps of beads on the membrane film integrated into the microfluidic device of Example 4. Fig. 6(b) shows an SEM image at 80 Ox magnificatTohT^"""of^~"" larger beads of size ΙΟμπι and 12 m being patterned on the film in Example 4. Fig. 6(c) shows a graph of the patterning efficiency of the different sizes of beads in Example 4. Fig. 6(d) shows the separation and patterning efficiencies of bead ratio groups for 5μιη beads to ΙΟμπι beads of 1:3, 1:2, 1:1, 2:1 and 3:1 in Example 4. Fig. 6(e) shows an SEM image at 750x magnification of patterned yeast and cancer cells on the membrane film integrated into the microfluidic device of Example 5. Fig. 6(f) shows the separation and patterning efficiencies of the cells in Example 5.

Fig. 7 shows the patterning efficiency of different bead concentrations in Example .

Fig. 8(a) shows a schematic diagram of the model used for studying the fluid flow velocity through the pores in Example 6. Fig. 8(b) shows the velocity profile in m/s obtained by the COMSOL Multiphysics.

Figs. 9(a) to (c) show the SEM images at l,800x, l,700x and 2,500x magnifications respectively of trapped beads with a size of ΙΟμιη, 12μτη and 20μτη respectively in the membrane filter obtained from Example 3 and used in Example 6. Figs. 9(d) to (f) show the SEM images at 500x, Ι,ΟΟΟχ and 500x magnifications respectively of the normal filter used in Example 6. Fig. 9(g) shows a plot of the percentage of beads that passed through the pores of the normal filter and the structured filter against the flow speed in Example 6.

Figs. 10(a) and (b) show the ,SEM images of the flow of Ιμπι beads . through the micro-structured filter at l,900x magnification and the normal filter l,800x magnification respectively in Example 6. Figs. 10(c) and

(d) show the SEM images of the flow of 3μηα beads through the rnicro-st IIctured ^~fiTEer^" at Ι7^~7ΌΌχ magiii^'fTeation ancf the normal filter at l,200x magnification respectively in Example 6. Figs. 10(e) and (f) show the SEM images at 2,000x and Ι,βΟΟχ magnifications respectively of the flow of Ιμκι and .3μπι beads through patterned 12μιτι beads on the micro-structured filter in Example 6. Fig. 10(g) shows the percentage of Ιμιτι and 3μηα beads that passed through the interstitial gaps between the trapped ΙΟμπι beads and the micro-structures of the normal filter and the structured filter used in Example 6. Fig. 10(h) shows the percentage of Ιμιη and 3μπι beads that passed through the interstitial gaps between the trapped 12μπι beads and the micro-structures of the normal filter and the structured filter used in Example 6. Fig. 10 (i) shows the efficiency in percentage of Ιμιη, 3μπι and 5μιη beads that passed through the interstitial gaps between trapped MDA-MB-231 ^' cells and the normal filter and the structured filter- used in Example 6.

Figs. 11(a) and (b) show the bright-field images of the target sample mold and the control sample mold respectively obtained in Example 7 to study the anisotropic cross-linking by the low-dose exposure method.. Figs. 11(c) and (d) show the SEM images of the target sample mold at l,400x magnification and the control sample mold respectively obtained in Example 7 to study the anisotropic cross-linking by the diffusion method .

Figs. 12(a) to (f) show the SEM images of the different shapes of structures, such as hour-glass shaped, popsicle-shaped and doll-shaped, obtained from Example 8. The SEM images in Figs. 12(a) to (e) are at 350x magnification and the SEM image in Fig. 12(f) are at 330x magnification. Figs. 12(g) to (i) show the graphs of the normalized values of different defining features against a variation of exposure energy at a PEB ramp-up temperature of 85°C to 90°C, 90°C to 96°C and 96°C to 100°C respectively in Example 8.

Figs. 13(a) and (b) show the SE images at 350x magnification of the film obtained at different exposure energies in Example 9. Fig. 13(c) show the graph of the normalized values of different defining features against the variation of exposure energy in Example 9.

Figs. 14(a) and (b) show the SEM images at 350x magnification of the film obtained from Example 10. Figs. 14(c) and (d) show the SEM images at 350x magnification of the film obtained in Example 11. Figs. 14(e) and (f) show the graph of the normalized values of different defining features against the different exposure energies obtained from Examples 10 and 11 respectively.

Figs. 15(a) and (b) show the SEM images at 350x magnification of the film obtained from Example 12. Fig. 15(c) shows the graph of the normalized values of different defining features against the different exposure energies obtained from Example^' 12.

Detailed Description of Drawings

Referring to Fig. 1(a) and Fig. 1(b), there are shown respective photo-masks 10. Fig. 1(a) shows a photo- mask 10 with transparent circles 12 while Fig. 1(b) shows a photo-mask 10 with opaque circles (14). Exemplary dimensions of the circles and inter-circle distances are also shown. The photo-masks are used during the production of- the master mold ^'during UV treatment.

Fig. 2(a) (i) is a schematic diagram showing the arresting formations 16 trapping particulate entities of different particle sizes such as particles 20a and 20b while allowing smaller particulate entity such as particle 22a to pass through the holes 18. As can be seen in this figure, the larger .sized particulate entities (particles 20a and 20b) are trapped by the inclined surface of the arresting formations 16 and do not completely block the holes 18. Hence, it is still possible for _. fluid (and optionally smaller-sized particulate entities) to flow around the larger sized particulate entities (particles 20a and 20b) and enter the holes 18.

Fig. 2(a) (ii) is a schematic diagram of a single trapped particulate entity 24 surrounded by a number of arresting formations 16. As can be seen in Fig. 2(a) (ii) , the fluid (with flow direction depicted by arrow 26) is able to enter the hole 18 through the gaps provided between the arresting formations 16.

Fig. 2(b) (i) is a schematic diagram of one process 100 to make the master mold 30 by way of UV light exposure. In the first step, a layer of photosensitive material 34 is coated onto a substrate 32. The photosensitive material may be baked for a period of time. The photosensitive material 34 is exposed through a photo-mask- 38 (for channels) to UV light 35 and is baked for a short time. In the second step, a second layer 37 of the same^' photosensitive material is coated onto the exposed layer 36 and soft-baked. The second layer 37 is exposed to a low dose of UV light 40 (without any photomask) and subjected to a post-exposure bake to generate partially cross-linked species 44. In the third step, a third layer of photosensitive material 39 is coated onto the second layer and soft-baked to allow interaction with partially cross-linked species 44. The third layer 39 is then exposed ^· to UV light 35 via another photo-mask 42 (with patterns to form the pillars) and then subjected to a post-exposure bake. In the fourth step, the layered photosensitive material is developed to form the master mold 30 having the imprint formation structures thereon 46.

Fig. 2(b) (ii) is a schematic diagram of another process 110 to make the master mold by way of diffusion. The first step is identical to that in process 100 of Fig. 2 (b) (i) above. In the second step, the difference between process 110 and process 100 is that the second layer 37 is exposed to UV light 35 via a photo-mask 42. The photo-mask of Fig. 1(a) is used as the photo-mask 42 here. The post-exposure bake after this UV exposure is also carried out for a shorter period of time as compared to that in the second step of process 100. The shorter post-exposure bake prevents complete cross-linking and allows diffusion to occur in the subsequent steps. The short post-exposure bake after the second exposure controls the cross-linking of the exposed layer 36 and does not allow complete cross-linking of the exposed layer 36. In the third step, the coating of the third layer 39 allows for a non-uniform distribution of cross- linking species 44 from the exposed layer 36 to the unexposed third layer 39. The remaining step is identical to the fourth step of process 100.

Fig. 2(c) is a schematic diagram of the process 120 to form the substrate 50. A layer of the substrate 48 is deposited onto the master mold 30 to cover the imprint formations structures 46 and baked. The thickness of the substrate 48^' is controlled to allow formation of through- holes by pillars in the mater mold. It is then removed from the master mold using 2-propanol to form a porous substrate 50 having the arresting formations 52 and holes 54 therethrough.

Fig. 3(a) (i) shows a top view schematic representation of a microfluidic device 80 incorporating the filtering membrane 50' on the left and a bottom view of the microfluidic device 80 on the right. Fig. 3(a) (ii) is an expanded view of the filtering membrane 50' showing the arresting formations 52' and the holes 54'.

Fig. 3(b) is a schematic diagram of the fluid flow in the microfuidic device 80. The fluid may contain a plurality of particulate entities and is introduced into the microfluidic device 80 via inlet 84. The fluid then flows in the channel 82 with outlet 88 open and outlet 80 closed. As the fluid flows through the channel 82, the particulate entities encounter the filtering membrane 50' where larger sized particulate entities are captured by the arresting formations (not shown) while smaller sized particulate entities pass through the holes 54' to a collector 90.

Examples

Non-limiting examples of the invention and a comparative example will be further described in greater detail by reference to specific Examples, which should not be construed as in any way limiting the scope of the invention.

Example 1

Low dose exposure energy method

In this example, a master-mold was fabricated using the low dose exposure energy method, in accordance with the schematic process diagram in Fig. 2(b) (i).

A first layer of epoxy-based negative photoresist, SU-8, was spin-coated (using Spin-coater, Cee 100, brewer science, Missouri, United States of America (USA) ) at 4500 revolutions per minute (rpm) on an oxygen-plasma treated silicon wafer. The oxygen plasma treatment was done using a PX-250 plasma chamber from March Instruments Incorporated, Massachusetts, USA. Thereafter, the silicon wafer was soft-baked twice, the first for 2 min at 65^°C and the second for 10 min at 95^°C on a hot-plate to remove solvents (Sawatec AG, Germany) .

It was then exposed to ultraviolet (UV)^■ light for 40 seconds (s) at a power of 7 mW/cm² through a photo-mask by a mask-aligner (Karl-SUSS MicroTec, Vermont, USA) . The patterned photo-mask schematically shown in Fig. 1(a) has a rectangular, array of circular open windows of 5μπι diameter each separated by 15μπι. The patterned photo-mask acts as a design for creating imprint formation structures, also termed in the examples as "micro- structures", as the patterned mask blocks UV exposure on the SU-8 under the mask. The photo-activated UV-exposed regions are thus able to form micro-structures.

A second layer of SU-8 was coated thereon at 6000 rpm. It was soft-baked twice, the first for 3 min at 65^°C and the second for 15 min at 95^°C.

It was then exposed to a low dose of UV light for 1.2s at about 7 mJ/cm² without any photo-mask to thereby create partially cross-linked SU-8 chains between the interface of the first and second layers. A post-exposure bake (PEB) was then performed twice, the first for 1 min at 65^°C and the second for 1 min at 95^°C.

A third layer of SU-8 was coated thereon at 5000 rpm. It was soft-baked twice, the first for 1 min at. 65^°C and the second for 15 min at 92^°C.

It was exposed again through the same photo-mask aligned to the same position for 50s to create the micro-

PEB was performed for 3 min at 65^°C and then a drop of SU-8 (soft-baked) was put over the region containing micro-structures. The temperature was then ramped to 96^°C within 2 min and was baked at this temperature for 15 min. It was cooled down before ^; developing it for a further 8 min. The master-mold was therefore obtained.

Example 2

Diffusion method

In this example, a master-mold was fabricated using the diffusion method, in accordance with the schematic _. process diagram in Fig. 2(b) (ii).

A first layer of epoxy-based negative photoresist, SU-8, was spin-coated at 4500 rpm on an oxygen-plasma treated silicon wafer and was soft-baked twice, the first for 2 min at 65°C and the second for 10 min at 95°C. It was then exposed to UV light for 40s at a power of 7 mW/cm² through a photo-mask, with a pattern designed to make micro-channels, by a mask-aligner to thereby create a first level of base micro-structures.

A second layer of SU-8 was coated thereon at 5000 rpm. It was soft-baked twice, the first for 3 min at 65°C and the second for 15 min at 95°C.

It was then exposed to UV light for 30s through the photo-mask schematically shown in Fig. 1(a) and aligned with the first photo-mask exposure.

PEB was then performed for 30s at temperatures ramping from 78 °C to 85 °C. This short PEB permitted the partial cross-linking at the UV■ exposed region of the first and second SU-8 layers.

A third SU-8^' layer was then coated at 5000 rpm, which allowed mixing of the exposed and unexposed regions, thereby permitting some amount^~~of^~dTffus^"ion^~~fTom ^' the exposed (photo-activated) layer to the unexposed layer. Due to such diffusion, a non-uniform distribution of cross-linked species was. observed. Thereafter, the third layer was soft-baked twice, the first for 1 min at 65^cC and the second for 15 min at 92°C.

It was then exposed to UV light through the same second photo-mask aligned to the same position for 50s. A final PEB was performed for 3 min at 65 °C and the temperature was ramped up to 96 °C within 2 min, to thereby aid in re-arranging the partially cross-linked chains in order to create double-level, structures having three-dimensional pyramidal micro-structures on the base micro-structures.

The substrate was then baked at this temperature for 15 min. Thereafter, the substrate was cooled down for 30 min before developing it for a further 8 min. The master- mold was therefore obtained.

A scanning electron microscopic (SEM) image of the master-mold obtained is shown in Fig. 4(a) at Ι,ΟΟΟχ magnification. The micro-structures of the master-mold is seen as bright circles in Fig. 4(a).

Example 3

Porous membrane and subsequent microfluidic device fabrication .

The master-mold obtained from Example 2 was used to fabricate a porous membrane in accordance with an embodiment of the present invention. This example was performed in accordance with the schematic process diagram of Fig. 2(c).

A polydimethylsiloxane (PDMS) base solution was thoroughly mixed with a curing ^' agent in a ratio of 10:1 to obtain a PDMS solution. The PDMS solution was degassed in vacuum desiccators for 30 min.

A small amount of PDMS solution was first dropped onto the arrayed structures of the master-mold to cap the structures and was baked for 15 min at 70^°C. The PDMS was then spun-coated at 1400 rpm onto the master-mold to cover the channels and baked for 10 min at 70^°C.

The cap was then removed and another layer of PDMS was coated at 2500 rpm. A weight was applied to remove excess PDMS and the coated array was baked for 5 min at 50°C. The weight was then removed and the arrayed pillars of the master-mold were wiped with a small piece of silicon wafer.

It was allowed to stand for 15 min before it was baked again for 60 min at 70^°C. To create the collector in the microfluidic device for collecting fluid or smaller particles passing through the arresting formations of the membrane, a PDMS-containing collector was aligned to the arrayed structures and the collector was bonded to the array using oxygen-plasma treatment.

Thereafter, it was submerged in 2-propanol and was baked for 30 min at 70^°C to release the master-mold. The porous membrane along with the PDMS collector was peeled off from the master-mold and the SEM image of the porous PDMS film is shown in Fig. 4 (b) at 55x magnification. The porous membrane was then dried in air. Holes were punched to fabricate inlets and outlets in the microfluidic device to allow fluidic connections, such as tubings, for the purpose of sample delivery or recovery to be connected to the device and the device was sealed from the top by a PDMS slab with punched holes by oxygen- plasma treatment for 30s at a power of 200W to obtain the microfluidic device. SEM images of the top, bottom, side and the 45°- angled view of the final porous membrane film obtained are shown in Figs. 4(c), (d) , (e) and (f) respectively. In particular, the 45°-angled view of the porous membrane shown in Fig. 4(f) shows that 3-D structures on top of 2- D pillars are created on one side of the film, evidenced by the pyramidal structures that surround funnel-like through-holes. However, the reverse side of the film does not have any micro-structures, as seen in Fig. 4(d).

Example 4

Bead separation and patterning

The porous membrane obtained from Example 3 was integrated into a microfluidic device as shown in the schematic diagram of Fig. 3(a) (i) .

In this example, the microfluidic device was used for separation of beads based on size and patterning of beads on the porous membrane .

The size, of each feature in a single through-hole of the porous membrane, termed "trap" herein, was measured and the size .measurements are shown in Fig. 5. Referring to Fig. 5, the solid lines indicate the distance measured, while the dotted lines represent the line of feature between which the distance is measured. The dimensions indicate that each trap is capable of holding any particle in the size-range of 6pm to 20μιη. Consequently, it is hypothesized that particles having a size less than 6 m would be able to flow out of the microfluidic device.

This hypothesis was experimentally verified in this example using different sizes of beads ranging from 5 m to 19pm to pass through the microfluidic device. The b.ea_ds__were counted using a hemocytometer . The beads were mixed in solution and diluted using deionized (DI) water to obtain a solution of beads. 100 ΐι of bead solution was pumped into the microfluidic device at a pre-optimized flow speed of 50 μΐι/min using a syringe pump (Kd Scientific, Massachusetts, USA) .

The beads that passed through the pores of the device were collected from the sink and were counted. Single beads that were either not positioned in a single trap or not separated were washed away from the device and were collected back. These beads were then pumped again into the device. The above process was repeated for about 2 to 3 times.

Thereafter, the beads that did not pattern on the porous membrane in the device or that were not singly patterned on a single trap were counted in order to determine the number and the percentage of singly- patterned beads. Separation and patterning efficiencies were calculated as per the formulae below.

_ _ _„,_.„_. . If**, nf JWHwi -w"jwawxffr«5i*-MW¾d! x^" 1 00

Sez'Oratior,- fri dsjwy =—— : ; ——: ; :

' ctsti tso.vf smaiie tne ss nseti, item&f simgiivw eed bis- beads X ICQ

Fatts-ming erricierty—— : ^". , , ■—~—; :

^'i otcit no. f big csac mtm i suF-e

Based on the separation efficiency formula above, more than about 95% of 5μιη sized beads were recovered from the device, confirming the hypothesis that particles having a size less than 6μιη would be able to flow out of the microfluidic device.

Further, based on the patterning efficiency formula, about 2-3% of beads formed un-patterned clumps and did not pass through the pores of the film. An SEM image at 850x magnification of the un-patterned clumps is shown in _Fi_.g. 6(a). It can be seen from Fig. 6(a) that the clumps did not pass through the pores. About 3-4% of beads were lost in the sink and the channels, while beads with a size of 7μηι or more did not pass through the pores and were patterned on the device. Fig. 6(b) shows the SEM image at 800x magnification of larger beads of size ΙΟμιή and 12μπι being patterned on top of the porous membrane.

The patterning efficiency of the different sizes of beads was charted in Fig. 6(c). From Fig. 6(c), it can be seen that beads of size ΙΟμιη and 12μιη were patterned at the highest efficiency at more than 90%, which is expected for the given dimensions of traps in this example. The 7μιη beads were patterned with slightly lower efficiency of about 85% since a single trap can capture more than one such bead and these beads may get stuck between the micro-structures of the pores during washing. The efficiency for patterning of 19μιη beads is low at about 30%. This is because the trapping of bigger particles may block the passageway of the neighboring pores which in turn hinder other big particles to be trapped in the vicinity. Further experiments were done to test the patterning efficiency of 19pm beads by varying the concentration of bead solution between 10⁴ beads/mL and 10⁵ beads/mL. The results of the patterning efficiency of the different bead concentrations are shown in Fig. 7. From Fig. 7,. it can be seen that as bead concentration of the 19μπι beads was reduced from 10⁵ beads/mL to 10⁴ beads/mL, patterning efficiency increased to about 80%, thereby confirming that larger beads, such as 19μπι. beads, are patterned with lower efficiency as compared to ΙΟμιη or 12μπι beads. These results also - explain that the patterning efficiency decreases if the number of beads in the . sample is more- than the number of pores. These experimental results suggest that the pyramidal-shaped micro-structures have helped in the patterning of different sizes of beads with reasonably high efficiency.

Additionally, the simultaneous separation and patterning of beads were also studied in this example by using a mixture of 5pm and ΙΟμπι beads in varying ratios. The concentration of the ΙΟμπι beads was fixed to 10⁵ beads/mL and the concentration of 5μπι beads was varied according to the desired ratio.

■ Separation and patterning efficiencies were calculated as per the formulae given above and the efficiencies of ratio groups for 5μπι beads to ΙΟμιη beads of 1:3, 1:2, 1:1, 2:1 and 3:1 are shown in Fig. 6(d). The one-way analysis of variance (ANOVA) to determine a single factor between each ratio group suggests that there is no significant change between separation or patterning efficiencies when the ratios were changed, as seen from Fig. 6(d) .

It can thus be concluded that high efficiency of separation and patterning was achieved and maintained with- no or little sensitivity to different bead mixtures of varying size ratios. The high efficiencies can be attributed to the pyramidal-shaped micro-structures which help to retain a trapped particle while washing thereby allowing reproducibility of the process while maintaining high separation and patterning efficiencies.

Example 5

Cell separation and patterning

In this example, the same microfluidic device of Example 4 was used for size-based separation and pa tre-r-ntng—o-f—-ϋν-i-ng—selis _: Cancer cells from cell-line MDA-MB-231 were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented by 10% fetal bovine serum (FBS) in a cell-culture flask. Cells were trypsinized and were counted using a hemocytometer . The cell solution was diluted in complete cell-culture medium to a concentration of 10⁵ cells/mL.

Yeast cells were taken from an agar plate and mixed in the cell-culture medium. Yeast cells represent smaller sized cells, while cancer cells represent bigger sized cells. The concentration was adjusted by dilution. Accordingly, the mixture of cancer cells and yeast cells was obtained in different ratios keeping the concentration of cancerous cells constant at 10⁵ cells/mL in the mixture .

100pL of the cell mixture were passed through the microfluidic device at a flow rate of 50 μΐ,/mln. The process- was repeated in the same manner as in Example 4.

Trapped cells were fixed in the device by using 4% paraformaldehyde in phosphate-buffered saline (PBS) for one hour.. The cells were then dehydrated by using different concentrations of ethanol solution varying from 10 vol% to 100 vol% of ethanol ^· with an increment of 10 vol% for 5 min incubation at each point. Thereafter, the cells were dried on the device by using different ratios of ethanol to hexamethyldisilazane (HMDS) with increasing concentration of HMDS (3:1, 2:1, 1:1, 1:2 and 1:3) and left to dry overnight.

The sample was sputter-coated using Jeol Sputter (JEOL Ltd, Japan) for about 30s to 40s at 30 mA before obtaining an .SEM image at 750x magnification of patterned cells on the membrane film integrated into the microfluidic device. The SEM image is shown in Fig. 6 (e.) . The. trapped cells in this SEM image looked smaller than the r^"_-¾^"rmaI^—size ^'~of^~a¹bout^~ 4^'~~to Z0^"pm ^_ b aTy^~¾ue to shrinkage of cells during sample preparation, i.e. fixation and dehydration, for SEM. Also, it can be seen from Fig. 6(e) that the cells were successfully patterned on the film, with an efficiency of about 85%.

Further, the separation and patterning efficiencies of the cancer and yeast cells were calculated according to the formulae above and the results are shown in Fig. 6(f) . As evidenced in Fig.. 6(f), human cancer cells were separated from yeast cells and were singly-patterned at above 80% efficiency.

Example 6

Fluid flow through interstitial-gaps

In this example, the possibility of fluid flow through interstitial gaps between a trapped particulate entity and the pore is studied. Fluid flow for filtration processes is essential to ensure that the porous membrane of a^" microfluidic device does not get clogged or fouled easily.

SEM. images at high magnification of trapped beads of different sizes are shown in Figs. 9(a) to (c) . The interstitial gaps between trapped beads and the pore are shown by arrows in Figs. 9(a) to (c). Specifically, Fig. 9(a) shows a lOum bead, Fig. 9(b) shows, a 12μπι bead and Fig. 9(c) shows a 20μιη bead.

A model for studying the fluid flow in the through- pores was developed and solved using the modeling and simulation software, COMSOL. A schematic diagram of the model is shown in Fig. 8(a). As seen in Fig. 8(a), the model is an array of through-holes surrounded by 3-D structures with two columns of through-holes filled with solid spheres which represents cells. The model was solvecT^~By^~~'tRe^~COMSOE^—so^~ff7ware^~al¾d^™ h^^~ve^~lbjl;y prbfTle ^"in^" m/s obtained is shown in Fig. 8(b). In Fig. 8(b), the holes with trapped cells (indicated by arrows) maintained a fluid velocity of about 0.007 to 0.009. Fig. 8(b) demonstrates that the flow rate was not drastically reduced when cells were trapped in the pores. Flow was maintained in the device. Hence, the above model shows that it is theoretically possible for interstitial gaps between trapped cells and pores to allow fluid to flow through.

To prove the theory that the interstitial gaps between trapped cells and pores are not blocked and allow fluid to flow through, a study was done using the porous membrane filter of Example 3 and a normal filter without any micro-structures. The normal filter was obtained by reversing the membrane of Example 3 as the other side of the membrane did not have any micro-structures.

ΙΟμπι sized beads were used in this example. The beads were patterned in the microfluidic device and fluid flow at flow-rates from 100 μΐι/min to 500 pL/min was maintained to study the relative difference- in the pressure drop due to blocking or partial blocking of pores .

Hypothetically, depending on the type of filter used, the pores of the filter can be fully or partially blocked when beads are patterned on the device. Hence, the liquid passing through the device would cause a pressure drop, depending on the extent of pore blockage. The pressure drop increases with increasing fluid flow- rate. Further, fully blocked pores would have .a higher pressure drop as compared to partially blocked pores. If the pressure drop is too high, the film may be damaged.

However, as seen from SEM images of the normal fritre—r—F±g-s-^—9-(-d-)--t-e—(-£->-,—the—-fi-im-was—no±__Qb.s.e.rjv_e_l_to be damaged even at high flow-rates. Fig. 9(d) shows beads pushed into the pores of the normal filter. Fig. 9(e) shows complete blocking of pores oy beads. The increased pressure drop caused some of beads to pass through pores in spite of their bigger size, as seen in Fig. 9(f) . This is because PDMS is a soft and flexible material, thus beads which are bigger than the pore size may also be pushed through the pore at a higher pressure drop.

The difference between the percentage of beads that passed through the two different types of filters used in this example indicates the relative difference in the pressure drop to thereby conclude on the amount of blockage of pores. A plot of the percentage of beads that passed through the pores against the flow speed was obtained for both types of filters and is shown in Fig. 9(g). It can be seen from Fig. 9(g) that the percentage of beads that passed through the micro-structured filter ^' was much lower as compared to the normal filter. This indicates a lower pressure drop in the micro-structured filter compared to normal filter. The lower pressure drop indicates that the fluid-flow is maintained around the pore according to Bernoulli's principle and the continuity principle, and can be attributed to the absence or at least a lower level of pore blockage by beads in the micro-structured filter. Though all pores may not be blocked or partially blocked and the pressure drop in the device may be a complex function of different parameters, but the relative difference in the pressure drop for the two type of filters is instructive.

Another^' experiment was performed to prove the same fact that the interstitial gaps between trapped cells and pores of the porous membrane filter of Example 3 are not blocked and allow fluid to flow through. Further, the experiment was performed to prove that particles smaller than the gap are able to pass through the gap.

Hypothetically, particles smaller than the interstitial gap should be able to pass through together with the fluid while particles bigger than the interstitial gap should be blocked or trapped around the beads that are patterned on the pores. In both of these situations, fluid flow through the gap can be confirmed.

To study this hypothesis, a study was conducted using both types of filters as before. ΙΟμιτι beads were patterned on the micro-structured filter and the normal filter. Beads of either Ιμιη or 3μπι in size were then diluted to the concentration of 3<10⁵ beads/mL and 100 μΐ. of bead solution was passed through the device at a flow- rate of 50 μΐ/min.

The percentage of lpm and 3μιη beads that passed through the pores of both types of filter devices is shown in Fig. 10(g). It can be seen in Fig. 10(g) that there is a significant difference in the percentage of l m beads passing through the micro-structured filter as compared to the normal filter. Specifically, the percentage of Ιμιη beads passing through the . micro- structured filter is about twice that of the normal filter. However, the percentage of 3μιη beads passing through the micro-structured filter is about half that of the normal filter.

This result is further^■ substantiated by SEM images of the micro-structured filter and the normal > filter. Figs. 10(a) and (b) show the SEM images of the flow of Ιμηι beads through the micro-structured filter and the normal filter respectively. It can be seen from the arrow in Fig. 10(a) that the Ιμιτι beads that were unable to pass throug —the—rntex-s-ttt-i-a-l—g-a-p-s—w-e-re—t-r-appe-d—-arou d—the— patterned ΙΟμηα beads. In contrast, Fig. 10(b) shows that the Ιμπι beads were scattered over the whole surface of the normal filter. Also from . Fig.^" 10(b), many beads can be seen trapped inside the pores of the normal filter. This indicates that the pressure drop is higher for the normal filter. Further, about 35% of 10pm beads passed through the normal filter, whereas no ΙΟμπι beads were observed to be pushed out of the micro-structured filter (data not shown) . Although the pores of the normal filter were observed to be completely blocked, about 30% of Ιμιτι beads were found to pass through the pores. This can happen due to unoccupied pores or due to the trapped beads passing through the pores.

Accordingly, as other parameters of the method were kept same, the significant differences between the two types of filters that were observed may be due to the presence of micro-structures around the funnel-like pores, thereby concluding that Ιμιη beads were able to pass through the interstitial gaps.

This conclusion was further confirmed from · the analysis of . results obtained from flowing 3μπι beads through both types of filters. Figs. 10(c) and (d) show the SEM images of the flow of 3μιη beads through the micro-structured filter and the normal filter respectively. It can be seen from Fig. 10(c) that the 3μιη beads were concentrated around the patterned ΙΟμπι beads in the micro-structured filter, indicating that fluid flow was maintained through the gap.

In the case of the normal filter in Fig. 10(d), the 3μιη beads were seen sitting over the patterned ΙΟμπι beads in the pores, which results in a higher pressure drop due to pore blockage. The higher percentage of 3μιη beads —paaaing through_ the normal filter as compared to the micro-structured filter (shown in Fig. 10(g) and described above) can be explained by the increased pressure drop in the normal filter which pushed some of the trapped 3μτη beads through the pores.

Another point to note from Fig. 10(g) is that the percentage of 3μιη beads passing through the micro- structured filter was significantly reduced as compared to Ιμκι beads. For the normal filter, the difference was not significant. Accordingly, the significant reduction of percentage of Ιμπι or 3μπι beads passing through the micro-structured filter can be used as a measure for studying the percentage of different sized beads passing through the interstitial gaps, as will . be discussed below.

The results were further confirmed by patterning

12μιη beads on the micro-structured filter, instead of lOum beads, to prove that an increase in bead size would decrease ^' the steric hindrance by increasing the interstitial gap between trapped particles and pores.

Similar to before, Ιμιτι and 3μιτι beads were separately passed through the micro-structured filter. The SEM images of the flow of Ιμπι and 3μιη beads through patterned 12μιη beads- on the micro-structured filter are shown in Figs . 10(e) and .(f) .

The percentage of beads passing through were calculated as shown in Fig. 10(h) and compared with the corresponding percentage for the ΙΟμιτι beads. The result shown in Fig. 10(h) shows a significant increase in percentage , ^" of smaller beads passing through the interstitial gap created by 12μιη beads as compared to ΙΟμιη beads. The relative percentage values of the ΙΟμιη and 12μιη beads is useful to prove that the fluid as well as smaller beads can pass through the interstitial gaps. The above results were further confirmed by patterning cells on both types of filters. Cancer cells from cell-line MDA-MB-231 were patterned on the micro- structured filter and normal filter and Ιμιη, 3 m and 5μπι beads were passed through the respective filter membranes thereafter. The efficiency in percentage of Ιμπι, 3μιη and 5pm beads that passed through the interstitial gaps between the trapped cancer cells and the normal filter and micro-structured filter respectively were charted and the graph is shown in Fig. 10 (i). From Fig. 10 (i), it can be seen that the percentage of lpm and 3μπι beads that passed through the gaps of the micro-structured filter is much higher than that of the normal filter, thereby confirming that the fluid as well as smaller beads are able to pass through the gaps.

Example 7

Anisotropic cross-linking of partially cross-linked material

In this example, the parameters of the method of

Example 1 were varied to study the anisotropic cross- linking of partially cross-linked polymeric chains between the SU-8 layers. The hypothesis being tested in this example is that exposing a layer of SU-8 with low energy would partially activate the layer to thereby induce partial cross-linking. The behavior of partially cross-linked chains can thus be studied by exposing the SU-8 again at certain regions only.

A first ^' layer of SU-8 was coated at a final rotational speed of 4000 rpm and was baked twice, the first for 2 min at 65°C and the second for 10 min at 96°C to remove solvents. The first exposure was performed for 2s (about 14 mJ/cm²) without any photo-mask. After coating the second layer of US-8, the second exposure was done for 40s (about 280 mJ/cm²) through the photo-mask. Again, the photo-mask shown in Fig. 1(a) was used here. The PEB was performed twice, the first for 1 min at 65°C and the second for 5 min at 96°C.

A control was prepared in which the first and second exposures were both performed for 40s (about 280 mJ/cm²) .

Bright-field images of the target sample mold and the control sample mold are shown in Figs. 11(a) and (b) respectively. From Fig. 11(a), pillars are shown as the bright circles which were developed due to exposure through the photo-mask. Unexpected connecting structures, as indicated by the arrow, were additionally observed between adjacent pillars. However, no such connecting structures were observed in the control sample, as evidenced in Fig. 11(b). This proves that the partially cross-linked chains were anisotropically cross-linked towards the cross-linked pillars.

Hypothetically, this phenomenon should also be observed for a partially cross-linked layer from the diffusion method.

To prove this hypothesis, the parameters of the method of Example 2 were varied to study the anisotropic cross-linking of partially cross-linked polymeric chains between the SU-8 layers. A first layer of SU-8 of the target sample was exposed to UV light without any photomask, thereby activating the layer. However, PEB was not performed ^' for this layer.

In the experimental control sample, PEB was performed for the exposed first SU-8 layer at 95°C. PEB helped to ^' complete the cross-linking of the photo- activated layer. A second layer was then coated for both samples. This allowed diffusion of the activated species from the first layer of the target sample to the second layer to create a partially cross-linked interface. However, in the control sample, inter-diffusion was restricted between the two layers.

The second layer of both samples was then exposed through the photo-mask and PEB was performed normally. SEM images of the target sample mold and the control sample mold are shown in Figs. 11(c) and (d) respectively. In Fig. 11(c), connecting structures, were also observed at the bottom of each pillar in the target sample mold. However, no such connecting structures were observed in Fig. . 11(d) for the control sample mold. Accordingly, these results confirm the hypothesis of the anisotropic cross-linking of partially cross-linked chains evidenced by the partial cross-linking occurring at the interface between first and second layer and the connecting structures forming at the bottom of each pillar.

Exposure, through the photo-mask containing an array of transparent circles should produce an array of distinctly separated pillars. However, connecting structures at regions beneath the photo-mask can be seen clearly in Figs. 11(a) and (c) . Therefore, the connecting structures can only be attributed to the first exposure since the first exposure was done uniformly without a photo-mask. Hence, the presence of connecting structures which are only in a particular orientation proves ^* the anisotropic cross-linking of partially^■ cross-linked SU-8 chains . In the following examples, different shapes and sizes of three-dimensional micro-structures were fabricated using different method parameters.

Example 8

In this example, the ramp-up temperatures of the PEB and the exposure energy were varied and studied.

One layer of SU-8 was coated at a final rotational speed of 2000 rpm for 40s on a clean silicon wafer using the spin-coater. It was then baked twice, the first for 5 min at 65°C and the second for 20 min at 96°C to remove solvents. Thereafter, it was cooled down for 10 min and was exposed to UV light at a required dose of exposure energy by using a UV light source through the photo-mask by a mask aligner-. The schematic diagram of the photomask used in this example is shown in Fig. 1(b). PEB was then performed at ramp-up temperatures of 85°C to 90°C, 90°C to 96°C and 96°C to 100°C.

At different ramp-up temperatures and exposure energies, different types of double-level structures, i.e. structures with two different shapes, were formed from the single layer of SU-8. The different shapes of structures, such as hour-glass shaped, popsicle-shaped and doll-shaped, are shown in the SEM images of Figs. 12 (a) to (f ) .

The SEM images of the film obtained from a ramp-up temperature of 85°C to 90°C at different exposure energies are shown in Figs. 12(a) to (c). Specifically, Fig. 12(a) was obtained ^' at an exposure energy of 700 mJ/cm², Fig. 12 (b) was obtained at an exposure energy of 750 mJ/cm² and Fig. 12 (c) was obtained at an exposure energy of 800 mJ/cm². It can be seen from Figs. 12(a) to (c) that hourglass shaped micro-structures were formed. Assuming these hour-glass shaped micro-structures to be conical, the plot showing the normalized values of different defining features against a variation of exposure energy at this ramp-up temperature is shown in Fig. 12(g) . The normalized values represent the variation of the different features of the micro-structures. The points marked with (*) indicate that a double-level structure was formed at exposure energies from about 700 mJ/cm² to about 800 mJ/cm², while a conical structure was formed at an exposure energy of 850 mJ/cm². As seen in Fig. 12(g), the normalized values did not vary very significantly and thus, a ramp-up temperature of 85°C to 90°C is empirically classified as a less sensitive zone.'

The SE images of the film obtained from a ramp-up temperature of 90°C to 96°C at different exposure energies are shown in Figs. 12(d) and (e) . Specifically, Fig. 12(d) was obtained at an exposure energy of 700 mJ/cm² and Fig. 12(e) was obtained at an exposure energy of 750 mJ/cm². It can be seen from Figs. 12(d) and (e) that doll- shaped micro-structures were formed. The plot of the normalized values of different defining features against the variation of exposure energy at this ramp-up temperature is shown in Fig. 12(h) . The points marked with (*) indicate that a double-level structure was formed at exposure energies from about 700 mJ/cm² to about 800 mJ/cm², while a conical structure was formed at an exposure energy of 850 mJ/cm². As seen in Fig. 12(h) ., there ^■ was relatively more variation in the normalized values and thus, a ramp-up temperature of 90°C to 96°C is empirically classified as a moderately sensitive zone.

The SEM image of the film obtained from a ramp-up temperature of 96°C to 100°C at different exposure energies is shown in Fig. 12(f). Specifically, Fig. 12(f) was obtained at an exposure energy of about 650 mJ/cm² to produce doll-shaped micro-structures. The plot of the normalized values of different defining features against the variation of exposure energy at this ramp-up temperature is shown in Fig. 12 (i) . The double-level structure was formed at an exposure energy of about 650 mJ/cm², while a conical structure was formed at other exposure energies. As seen in Fig. 12 (i), there were large variations in the normalized values, even yielding no micro-structures at an exposure energy of 800 mJ/cm². Thus, a ramp-up temperature of 96°C to 100°C is empirically classified as a highly sensitive zone.

Example 9

Single-layer Double-exposure method

In this example, a single layer of SU-8 was exposed, without any photo-mask, to UV light at a low value of exposure energy of 14 mJ/cm². It was then exposed through a photo-mask to an exposure energy ranging from 400 to 800 mJ/cm². PEB was performed at a temperature of 80°C.

The SEM images of the film obtained at different exposure energies are shown in Figs. 13(a) and (b) . Specifically,^■ Fig. 13(a) was obtained at an exposure energy of 400 mJ/cm² to produce an hour-glass shaped micro-structure. Fig. 13(b) was obtained at an exposure energy, of 500 mJ/cm² to produce a doll-shaped micro- structure.

The plot of the normalized values of different defining features against the variation of exposure energy is shown in Fig. 13(c). In Fig. 13(c), it can be seen that a double-level structure was formed at an exposure energy 15^"f^~" 0Γ1τΰΐ7cm²1and^~~5^~0^"0^~~m^~J;/cm²-.- Example 10

Double-layer method

In this example, it is shown that the double-layer method can be useful for fabrication of dome-like structures .

The first layer of SU-8 was coated at a final rotational speed of 4000 rpm and was baked twice, the first for 2 min at 65°C and the second for 10 min at 96°C to remove solvents. It was then cooled down for 8 min and exposed to UV light for 40s at 280 mJ/cm².

The second layer of SU-8 was immediately coated without performing PEB of the first layer. Thereafter, it was soft-baked and exposed through the photo-mask at different exposure energies. Finally, PEB was performed at 80°C.

SE images of the film obtained are shown in Figs. 1 (a) and (b) evidencing that dome-like structures were formed. The plot of the normalized values of different defining features against the different exposure energies is also shown in Fig. 14 (e) .

Example 11

Combination method

A combination of the double-layer method and single- layer double-exposure method ^' was used in this example to fabricate dome-shaped micro-structures.

The double-layer method was performed on a sample by coating a second layer of SU-8 without performing PEB of the first SU-8 layer. Thereafter, the sample was exposed, without a photo-mask, to UV light at a low dose of exposure energy of 7 mJ/cm². The sample was then exposed through the photo-mask schematically represented in Fig. 1(b) at different exposure energies. Finally, the sample was baked at 80°C.

Uniform dome-shaped structures were obtained and the

SEM images are shown in Figs. 14(c) and (d) .

The plot of the normalized values of different defining features against the different exposure _. energies is also shown in Fig. 14(f).

Example 12

Fabrication of top-modified 3-D structures

In this example, a method .was developed to have better control over micro-structure dimensions, tested by modifying the top or upper part of the micro-structures without losing dimensions at other parts.

A first layer of SU-8 was exposed to UV light at a low dose of exposure energy of 14 mJ/cm² and then another SU-8 layer was coated. Thereafter, the second exposure was done at varying exposure energies of about 400 to 800 mJ/cm². It was then baked at a PEB temperature of 80 °C.

3-D micro-structures with unique shapes were obtained on the film and the SEM images are shown in Figs . 15(a) and (b) .

The plot of the normalized values of different defining features against the different exposure energies ^''is shown in Fig. 15(c). It is shown in Fig. 15(c) that-, the change in base width did not vary as much as the change in top width at increasing exposure energies. The change in angle was also uniform. Accordingly, it is evidenced in this example that process conditions can be controlled to modulate the shape and size of such micro-structures. Applications

The disclosed filtering membrane provides an improved membrane for biological and medical applications.

The disclosed filtering membrane may advantageously provide simultaneous functionalities of patterning and separation.

Further, the disclosed filtering membrane may advantageously possess a lower probability of fouling when used in filtration processes. As a result, the maintenance period of the disclosed membrane may advantageously be lengthened.

The disclosed method for forming ^' a filtering membrane may advantageously possess higher efficiency and reproducibility than current methods. Advantageously, the disclosed method does not need to be performed in special environments such as in clean-room facilities.

The disclosed microfluidic device may be useful for different bead and cell based applications. Some of these applications are listed in the following.

A bead-pattern can be generated by the disclosed device at high efficiency to provide a functional micro- array. Each bead, functionalized with desired proteins, antibodies or DNA, can act ^' as a functional micro-array used for detection and medical diagnosis. The bead- pattern can .also be used for studying interactions between biomolecules . Further, the bead-pattern may advantageously possess high efficiency of detection or interaction as . the sample fluid flows around each p^t eTne ^~bea^"d^". Th¾ fur.cti or.alizea beaas may be par.terr.ed on the disclosed device for use in bacterial diagnosis, such as multiplex bacterial detection from a pathological sample.

The disclosed microfluidic device may be used in the isolation of rare cells, such as circulating tumor cells (CTCs) or epithelial cells, from blood for diagnostic or prognostic applications.

The disclosed microfluidic device may also be used in leukapheresis , which is the separation of cancerous blood cells from blood in order to reduce white blood cell count. This is possible because blood cancer cells are stiffer than normal cells and thus can be filtered out from the blood using the disclosed device. The disclosed device may be modified to be used in the continuous separation of cells for the management of leukemia.

The disclosed device may be used for single-cell micro-arrays in drug testing. Drug discovery may thus be done based on an individual cell response.

Cells trapped using the disclosed device may be used as cell-based biosensors for toxin detection or point-of- care diagnosis.

Cell encapsulation is important for cell-based therapy and single-cell encapsulation is reguired for better mass-transfer. The disclosed device can be used for encapsulation of single-cells. As arrayed cells do not block the holes, each cell can be accessed from the encapsulating solution. The encapsulated cells can either be removed from the device or can be cultured on the device itself. Singly encapsulated■ cells can be used for cell-therapy in the treatment of diseases. The encapsulated cells on the device can also be used as a cell-based platform for treatment of diseases. The singly encapsulated cells on the device can further be used for 3-D single-cell based drug testing.

Further, a cell-based . artificial pancreas for diabetes treatment may be possible by using singly- encapsulated cells on the device. This can be developed as a wearable device containing cells that can produce insulin (β-cells). Blood can pass through the device by interfacing the device with the body. The encapsulated cells can then sense and respond to glucose levels in the blood and control glucose levels by producing insulin accordingly. This same principle can be used for treating patients with liver failure by detoxification of blood.

The disclosed method for forming a mold having imprint formation structures thereon can be used for fabrication of optical components (such as micro-lens, mirror arrays or as a part of optical- microelectromechanical systems (MEMS) devices. These .devices can also be used as a part of photo-electronic devices and possibly for solar-cell applications for concentrating light.

The. disclosed structures may also be developed as part of MEMS devices. The double-level structures may be used for generating or sensing mechanical movements in MEMS devices. These structures can possibly be used for developing sensitive pressure-sensors, micro-valves, micro-nozzles or micro-needles.

The disclosed structures may be used as filters for particle separation due to the targeted trapping of bigger particles.

The disclosed structures may also be used as an efficient trap for cells. These structures may anchor the trapped cells to facilitate single-cell studies. Further, the patterned structures may be modified for growth of -s-t-em^ceLLs—tha e^rL-fQrJja er stem-cell studies. It will be apparent that various other modifications and adaptations of the invention will be apparent to the person skilled in the art after reading the foregoing disclosure without departing from the spirit and scope of the invention and it is intended that all such modifications and adaptations come within the scope of the appended claims.

Claims

1. A filtering membrane comprising a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least one side of the substrate, the arresting formations being configured to arrest the ingress of particulate entities of a selected particle oversize from entering the holes.

2. The filtering membrane as claimed in claim 1, wherein the arresting formations are integrally formed with, and extend from, the substrate.

3. The filtering membrane as claimed in claim 2, wherein the substrate is comprised of a material selected from an organic polymer.

4. The filtering membrane as claimed in claim 3, wherein the organic polymer comprises monomers selected from the group consisting of dimethylesiloxanes , ethylenes, propylenes, methyls, pentenes, amides, sulphones, ethersulphones , carbonates, acrylonitriles , butadienes, styrenes, imides, amic acids, phenols, styrene sulphonic acids, acrylic acids, methacrylic acids, saccharides, thiophenols and combinations thereof.

5. The filtering membrane as claimed in any one of the preceding claims, wherein the holes allow the ingress of particulate entities of a selected particle undersize to enter the holes.

6. The filtering membrane as claimed in any one of the preceding claims, wherein the arresting formations comprise at least three three-dimensional structures.

7. The filtering membrane as claimed in any one of the preceding claims, wherein the arresting formations generally taper from a base portion adjacent the holes to a smaller end portion.

8. The filtering membrane as claimed in any one of the preceding claims, wherein the distance between adjacent arresting formations surrounding a hole is not equal when extending from the base portion to the end portion.

9. · The filtering membrane as claimed in any one of the preceding claims, wherein the arresting formations are dimensioned to trap oversize particulate entities while allowing fluid to enter the holes from the sides of the arresting formations.

10. The filtering membrane as claimed in any one of the preceding claims, wherein the holes have a cross- sectional diameter in the micron-range.

11. The filtering membrane as claimed in any one of the preceding claims, wherein the dimensions of the arresting formations are in the micron-range.

12. A method for forming a filtering membrane comprising a substrate having a plurality of holes that extend through the substrate, and arresting formations provided around the openings of the holes on at least one side of the substrate, the method comprising the steps of:

a. subjecting a mold comprised of a layer of a cross-linkable material to an anisotropic cross-linking treatment to cross-link the cross-linkable material, wherein the degree of cross-linking throughout the mold is unequal, to thereby form imprint formation structures thereon; and

b. applying a layer of the substrate, to the mold under conditions to form imprints on the substrate that are complementary to the imprint formation structures.

13. The method as claimed in claim 12, comprising, after the subjecting step (a), the step of baking the mold to cause complete cross-linking of the cross- linkable material.

14. - The method as claimed in claim 12 or claim 13, wherein the cross-linking treatment comprises exposing the mold to UV light.

15. The method as claimed in any one of claims 12 to 14, wherein the cross-linking treatment comprises adding a second layer of the cross-linkable material to the first layer to thereby allow diffusion of cross- linkable material from the first layer to the second layer .

16. The method as claimed in any one of claims 12 to 15, wherein applying step (c) comprises the step of spin-coating the substrate layer onto the mold.

17. The method as claimed in claim 16, further comprising the step of removing the substrate layer from the mold.

18. A mold having imprint formation structures thereon, the imprint formation structures having a first region that has a substantially uniform cross-sectional area along the length of the first region and a second region that has a varying cross-sectional area along the length of the second region.

19. The mold as claimed in claim 18, wherein the second region of the imprint formation structures converge at a point.

20. The mold as claimed in claim 18 or claim 19, wherein . the imprint formation structures are dimensioned in the micron-range.

21. The mold as claimed in any one of claims 18 to

20, wherein the mold is comprised of a cross-linkable, photosensitive material.

22. The mold as claimed in claim 21, wherein the mold is comprised of a negative cross-linking photoresist.

23. A method for forming a mold having imprint formation structures thereon, the method comprising the step of subjecting a mold comprised of a layer of a cross-linkable material to an anisotropic cross-linking treatment to cross-link the cross-linkable material, wherein the degree of cross-linking throughout the mold is unequal, to thereby form imprint formation structures thereon.

24. The method as claimed in claim 23, comprising, after the subjecting step, the step of baking the mold to cause complete cross-linking of the cross-linkable material .

25. The method as claimed in claim 23 or claim 24, wherein the cross-linking treatment comprises exposing the mold to UV light.

26. The method as claimed in any one of claims 23 to 25, wherein the cross-linking treatment comprises adding a second layer of the cross-linkable material to the first ^' layer to thereby allow diffusion of cross- linkable material from the first layer to the secondlayer.

27. A microfluidic device comprising

a channel for flow of a plurality of particulate entities having different particle sizes therein; and

a membrane in fluid communication with the flow channel, the membrane having a plurality of holes that extend through the membrane, and arresting formations provided around the openings of the holes on at least one side of the membrane.

28. The microfluidic device as claimed in claim 27, further comprising a collector for collecting particulate entities that passes through the holes due to the smaller particle size of the particulate entities.

29. A substrate having three-dimensional structures thereon, wherein the three-dimensional structures have a varying cross-sectional area along the length of the structures .

30. A method for forming a substrate having three- dimensional structures thereon, the method comprising the step of subjecting the substrate comprised of a layer of a cross-linkable material to a cross-linking treatment to cross-link the cross-linkable material .and to thereby form three-dimensional structures thereon, wherein the degree of cross-linking throughout the substrate is unequal.