FLUORESCENT OXYGEN SENSING INK

BACKGROUND OF THE INVENTION

[0001] Chronic non-healing wounds (e.g., diabetic foot and bed sores) impact over 6.5 million Americans per year, costs in excess of $25 billion to treat on an annual basis, and are on the rise due to increasing levels of obesity and diabetes compounded by an aging population. Current treatments are expensive, labor intensive, and generic, relying on regular cleaning, debridement, oxygen therapy, and topical or systemic administration of antibiotics. Commercially-available dressings (e.g., alginate, hydrogels, hydro-colloids, foams, etc.) have not proven to be significantly effective in reducing the burden. An ideal dressing integrates sensors (pH, oxygen, and inflammatory mediators), drug/cell delivery (antibiotics, growth factors, stem cells, and oxygen), and electronic intelligence to drastically improve wound care by measuring individual responses and enabling appropriate adjustments to therapy.

[0002] Suboptimal oxygenation of the wound bed is a major healing inhibitor in chronic wounds. Unlike acute injuries that receive sufficient oxygen via a functional blood vessel network, chronic wounds often suffer from the lack of a proper vascular network; thus being incapable of providing sufficient oxygen for tissue growth. While the lack of oxygen may trigger vascular regeneration, the severity and depth of wounds can prevent adequate regeneration, causing wound ischemia. Modern medical treatment of hypoxic chronic wounds typically employs hyperbaric oxygen therapy, which requires bulky equipment and often exposes large areas of the body to unnecessarily elevated oxygen concentrations that can damage healthy tissue. A more practical approach is topical oxygen therapy (TOT) in which the dressing itself can generate the required oxygen.

SUMMARY OF THE INVENTION

[0003] One aspect of the present invention is an ink that can be utilized to fabricate

"smart" dressings for chronic wounds. A fluorescent oxygen sensing ink includes an organic solvent, a polymer binder such as ethyl cellulose, and a fluorescent dye that is dispersed or dissolved in the solution. The ink can be printed on a thin flexible substrate such as paper, and the ink forms a moisture resistant flexible film that can be utilized in an oxygen sensor. The smart dressing measures the amount of oxygen present in a wound and pumps more oxygen as necessary. The smart dressing integrates oxygen delivery and sensing onto a single low-cost, manufacturable, flexible dressing. The smart dressing may be fabricated on a biocompatible substrate (e.g. paper) that incorporates patterned catalytic oxygen generating regions and an array of oxygen sensors connected to an electronic readout module. The use of a paper substrate provides structural stability and flexibility while simultaneously offering printability, selective gaseous filtering, and physical/chemical protection. However, it will be understood that the smart dressing is not limited to paper substrates, and virtually any hydrophobic to partially hydrophilic substrate may be utilized.

[0004] These and other features, advantages, and objects of the present invention will be further understood and appreciated by those skilled in the art by reference to the following specification, claims, and appended drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

[0005] FIG. 1 is a schematic cross sectional view of a smart wound dressing with

integrated oxygen sensing and delivery according to one aspect of the present invention;

[0006] FIG. 2 is a schematic isometric view of a test setup for measuring oxygen diffusion into agarose gel;

[0007] FIG. 3 is a graph showing oxygen generation and diffusion into agarose gel;

[0008] FIG. 4 is a graph showing 3D distribution of oxygen diffusion rate inside agarose gel from a single oxygen generation spot;

[0009] FIG. 5A is a graph showing excitation spectrum (RFU) for PdTFPP dissolved in chloroform;

[0010] FIG. 5B is a graph showing excitation spectrum (RFU) for PdTFPP dye dissolved in chloroform (dried for 10 minutes);

[0011] FIG. 5C is a graph showing excitation spectrum (RFU) for PdTFPP dissolved in heptane;

[0012] FIG. 5D is a graph showing excitation spectrum (RFU) for PdTFPP dissolved in heptane (dried for 10 minutes);

[0013] FIG. 6 is a graph showing the excitation spectrum of (RFU) of Ru(dpp) ₃ Cl2

dissolved in chloroform; [0014] FIG. 7 is a graph showing the excitation spectrum (RFU) of PdTFPP with PS on filter paper;

[0015] FIG. 8 is a graph showing the excitation spectrum of PdTFPP with PDMS on filter paper;

[0016] FIG. 9A is a graph showing emission spectrum for PdTFPP dissolved in chloroform;

[0017] FIG. 9B is a graph showing PdTFPP dye dissolved in chloroform (dried for 10

minutes);

[0018] FIG. 9C is a graph showing PdTFPP dye dissolved in chloroform (dried for 30

minutes);

[0019] FIG. 10 is a graph showing the emission spectrum of Ru(dpp) ₃ Cl2 dissolved in

chloroform;

[0020] FIG. 11 is a graph showing emission spectrum of PdTFPP with PS on filter paper;

[0021] FIG. 12 is a graph showing the emission spectrum of PdTFPP with PDMS on filter paper;

[0022] FIG. 13A is a graph showing the emission spectrum of a RedEye ^® patch in water with 0% dissolved oxygen concentration: (a) 0%; (b) 20%;

[0023] FIG. 13B is a graph showing the emission spectrum of a RedEye ^® patch in water with 20% dissolved oxygen concentration;

[0024] FIG. 14A is a graph showing the emission spectrum of PDMS encapsulated

Ru(dpp) ₃ Cl2 in water with a dissolved oxygen concentration of 0%; 20%;

[0025] FIG. 14B is a graph showing the emission spectrum of PDMS encapsulated

Ru(dpp) ₃ Cl2 in water with a dissolved oxygen concentration of 20%;

[0026] FIG. 15 is a schematic diagram of an oxygen sensor and electronic interfacing circuitry;

[0027] FIG. 16 is a schematic view of inkjet-printed oxygen sensitive dye with 7.5 mm diameter circular spot size;

[0028] FIG. 17A is a schematic diagram showing the design of the microfluidic network in an oxygen generation patch or module, including: (a) a larger (85 mm x 65 mm) model;

[0029] FIG. 17B is a schematic diagram showing the design of the microfluidic network in an oxygen generation patch or module, including a smaller (52 mm x 45 mm) model;

[0030] FIG. 18 is a schematic showing the use of pressure rollers for improving bonding; [0031] FIG. 19 is a photographic image showing test patches bonded with the use of pressure rollers;

[0032] FIG. 20 is a schematic showing an experimental setup and 0 ₂ sensing;

[0033] FIG. 21 shows graphs of optical and electrochemical responses of (a) a

commercially available Redeye oxygen sensing patch, and (b) printing Ruthenium ink according to one aspect of the present invention;

[0034] FIG. 22 is an electrical schematic of an oxygen sensing circuit;

[0035] FIG. 23 shows cytotoxicity of smart dressing components for oxygen generation;

[0036] FIG. 24 is an image of oxygen sensitive dye printed on unrasted parchment paper;

[0037] FIG. 25 is an image showing inject printed oxygen sensitive dye with 7.5 mm

diameter circular spot size;

[0038] FIG. 26 is a schematic showing fabrication of an oxygen delivery patch;

[0039] FIG. 27 is a photographic image of an oxygen delivery patch;

[0040] FIG. 28 is a photographic image of fluid (oxygen) patch arrays of size 1x2 and 2x2;

[0041] FIG. 29 is a chart showing Cytotoxicity test results of smart dressing components, wherein cells were maintained in complete growth medium (Eagle's Minimum Essential Medium) ("EMEM") polydimethylsiloxane ("PDMS"), double-sided tape ("TT"), RU (Ruthenium dye printed on parchment paper as 1, 2 or 3 layers) ("IRU," "2RU," and "3RU," respectively), and negative control extract (NC) made from low density polyethylene tubing;

[0042] FIG. 30 is a graph showing positive cytotoxicity control for WST-1 assay;

[0043] FIG. 31 is a chart showing cytotoxicity of smart dressing components following sterilization by a Sterrad process or by dipping in 100% ethanol;

[0044] FIG. 32 is a graph showing positive cytotoxicity control for WST-1 assay;

[0045] FIG. 33 is a chart showing cytotoxicity of paper sterilized by a Sterrad process or

70% isopropanol; filter paper ("FP"); parchment paper ("PP"); laser-treated parchment paper ("LTPP"); parchment paper calendered by rollers 1 and 2 ("Call-2"); parchment paper calendared by rollers 2 and 3 ("Cal2-3"); positive cytotoxicity control (PC");

negative cytotoxicity control ("NK"); and

[0046] FIG. 34 is a graph showing positive and negative WST-1 cytotoxicity controls. DETAILED DESCRIPTION

[0047] With reference to FIG. 1, a smart dressing 1 according to one aspect of the

present invention includes a substrate layer 5 forming a back-bone onto which one or more oxygen generating modules 10 and oxygen sensing modules 20 are printed. A fluid conduit 6 is fluidly connected to a pump/reservoir unit 12 including a reservoir IB and pump 14. A network of low-profile and flexible microfluidic channels 15 are formed by PDMS layers 8 and 9 that are bonded to the substrate layer 5. Fluid channels 15 guide and delivers hydrogen peroxide from fluid conduit 6 to the catalyst-printed regions 10. A wound-facing side 2 of smart dressing 1 includes a collagen-glycosaminoglycan biodegradable matrix (INTEGRA ^® ) 3 which provides a scaffold for cellular invasion and capillary growth while permitting oxygen exchange between the sensors/generators 20, 10 and a wound bed 4. The 3 matrix is retained in the wound 4 after initial application. The sensors/generators 20, 10 together with layer 5 and PDMS 8, 9 form a module 25 that can be delaminated from the matrix 3 and replaced periodically.

[0048] As discussed in more detail below, other aspects of the present invention include reliable processes for inkjet printing the oxygen sensing/generating modules 10, 20 as well as suitable lamination and bonding techniques (e.g., plasma, adhesives) for integrating the various layers of the smart dressing 1. An electronic readout and control module 16 is connected to an edge 17 of the smart dressing 1 via an edge-mounted connector 18 and a reservoir/pump module 12. The smart dressing 1 is connected to the reservoir/pump module 12 via fluid conduit 6 to supply FhC^ through channels 15. The reservoir/pump module may be fabricated via soft micro-molding techniques or other suitable processes.

[0049] Substrate layer 5 may comprise laser-treated parchment paper. In particular, laser-treated parchment paper possesses high mechanical strength (> 70MPa) to withstand human motion, high elastic modulus when dry (>300kPa) for easy handling during fabrication, low elastic modulus (<50kPa) when wet for interfacing with similarly soft tissue, permeability to gas and not water at low pressures, and permeable to oxygen diffusion. When laser-rastered, the surface energy of the paper increases.

[0050] One aspect of the present invention is a fluorescent oxygen sensing ink and

process for printing the ink. The ink generally includes a solvent, a dye, and a polymer binder. The solvent may comprise aqueous buffers or an organic solvent such as ethanol, dimethyl sulfoxide (DMSO), dimethyl formamide, isopropyl alcohol, acetone, toluene, and mixtures thereof. The fluorescent dye may comprise complexes of ruthenium, osmium tetroxide, rhodium acetate, chromium, palladium or other dye that fluoresces when exposed to UV/visible light in the presence of oxygen. The ink may comprise any polymer based material that provides uniform dispersion or is completely soluble in the ink system for different additive print manufacturing processes such as screen, gravure, flexography, inkjet, aerosol jet. The particle size of the polymers dispersed in the ink solution are dependent on the nozzle size of the inkjet heads nozzles if inkjet printing processes are utilized. For example, if the nozzle diameter is 21 pm, then the particle size should be less than 0.2 pm to avoid agglomeration and clogging of print head nozzles. For printing the surface tension of the ink is preferably less than the surface energy of the substrate to adhere well. The surface energy of the substrate (e.g. paper) can be modified by employing UV, corona, plasma, sintering, or laser engraving processes to increase a surface energy of the substrate. The surface characteristics of the substrate can be modified as desired without adversely affecting or damaging the other

characteristics/properties of the substrate. The fluorescent ink according to the present invention can be printed on any hydrophobic to partially hydrophilic substrates.

However, the ink typically cannot be printed on a substrate that is completely

hydrophilic. An advantage of the ink according to the present invention is that it does not require any transparent or translucent substrate or any additional protective coating materials.

[0051] As noted above, an ink according to the present invention includes one or more polymer binders that include alkyl substituted cellulosic materials. These alkyl substituted cellulose materials may be represented by Formula (I):

In Formula (I), Ri, R ₂ , and R ₃ may each independently be hydrogen or an alkyl group having 1-8 carbons including, for example, methyl, ethyl, propyl, isopropyl, n-butyl, sec- butyl, pentyl, or combinations thereof. In some aspects, the polymer binder is ethyl cellulose. Ethyl cellulose has the following chemical structure where Ri is ethyl, R ₂ is ethyl, and R ₃ is hydrogen as represented by Formula (II):

[0052] Ethyl cellulose does not contain any sulphonic or phosphonic groups or

naphthylene groups. Cellulose containings repeating anhydroglucose rings having hydroxyl groups at the 2', 3', and 6' positions that can treated with an alkaline solution resulting in an alkali cellulose which in turn is reacted with ethyl chloride to yield ethyl cellulose. In this reaction some hydroxyl (-OH) groups are replaced by ethoxyl (C ₂ H ₅ ) groups. In some aspects, the degree of substitution of the 2', 3', and 6' hydroxyl groups may be from about 1.0 to about 3.0, from about 1.2 to about 2.6, from about 2.3 to about 3.0, or from about 1.8 to about 2.2. In other aspects, the degree of substitution may be greater than 10%, greater than 20%, greater than 30%, greater than 40%, greater than 50%, greater than 60%, greater than 70%, greater than 80%, greater than 90%, greater than 95%, greater than 98%, greater than 99% where the percentage is relative to the substitution of the 2', 3', and 6' cellulosic hydroxyl groups, or the hydroxyl groups may be quantitatively substituted with ethyl or other alkyl group. Not to be bound by theory but the increasing reactivity of the 2', 3', and 6' hydroxyl groups, respectively, will affect the substitution positioned as would be appreciated by one skilled in the art.

[0053] Ethyl cellulose is an excellent water barrier film and provides moisture resistance.

In contrast, other polymers such as nitrocellulose dissolve in water, and have poor moisture resistance. In other words, ethyl cellulose has hydrophobicity. Also, ethyl cellulose provides excellent film formation, adhesion, high mechanical flexibility, and also allows greater film coverage compared to other cellulose derivatives. For example, nitrocellulose requires additional materials including synthetic resins (such as alkylated resins, maleic resins, ketone resins, urea resins, polyurethane resins, polyacrylates, polyester and polyacrylate resins containing hydroxyl groups) and plasticizers (diisobutyl phthalate (DIBP), dicyclohexyl phthalate (DCHP), epoxidized soya oil (ESO), triphenyl phosphate) to the ink system in order to provide uniform film formation, adhesion, flexibility to printed layer of dye.

[0054] Although polydimethylsiloxane (PDMS) or polystyrene may be utilized in some ink formulations according to one aspect of the present invention, PDMS or polystyrene disperses in the ink solution and binds to the ruthenium ink. In contrast, ethyl cellulose completely dissolves in the ink solution, rather than be dispersed, and binds with the ruthenium dye to form a moisture resistant, flexible, continuous, and uniform film. In general, binders such as ethyl cellulose that dissolve completely or nearly completely in the ink solution provide better film formation, adhesion, and flexibility than binders that disperse in the ink system. An ink solution according to the present invention requires minimal materials and simple fabrication steps, and forms a continuous uniform film with superior adhesion and flexibility.

[0055] An ink formula according to the present invention may include 98 weight percent of an organic solvent, a 1 weight percentage of dye, and a 1 percentage of a polymer that is preferably completely dissolved in the ink solution. The ink compositions can be varied as per the requirements of the additive printing processes.

[0056] To evaluate the ability to increase the oxygen concentration in a wound bed, oxygen diffusion was investigated on a surrogate wound bed (FIG. 2) comprising a sample of 0.3% agarose gel. An acrylic chamber 32 with open top 34 was assembled to hold the agarose gel sample 30. The chamber 32 includes an array of 2 mm holes 36 through a side wall 32 to allow insertion of an oxygen probe 40. Prior to testing, 0.3 % agarose gel is prepared and stored in a hypoxic environment until ready for use. During testing, the agarose gel 30 is placed in the chamber 32. An oxygenation platform 45 was constructed by bonding laser-machined parchment paper 46 to PDMS 47 patterned with a chamber (3x3x2mm ³ ) and a channel (18xlx2mm ³ ). The laser- treated region within the chamber was a 3x3 mm ² catalyst spot (deposited as described above). The chamber 34 was filled with 30% H2O2 through the guide channel using a syringe pump to begin oxygen generation.

[0057] The oxygenation platform 45 was placed on top (in contact with) of the gel 30.

The test chamber 34 was then sealed with a Parafilm barrier to prevent significant oxygenation form the atmosphere. The same oxygen probe 40 is then inserted into a hole 36 of the test chamber 34, penetrating the gel 30 until the tip is positioned 3mm directly below the catalyst spot of the parchment paper 46. For this test the oxygen probe 40 was covered with a protector needle (not shown) to prevent mechanical damage to the probe 40 during insertion. The remaining holes 36 in the chamber 34 were are sealed with adhesive tape to prevent oxygen diffusion from the atmosphere. The oxygen concentration in the gel 30 was monitored over time.

[0058] In clinical applications, the oxygenation platform may have an interfacial material between the parchment paper 5 and the wound to create intimate contact with the wound bed. To simulate this, the above experiment was repeated with a commercial dermal regeneration matrix (Integra ^® , available from Integra Life Sciences Corp.) as the interface. The dermal regeneration matrix is 900 pm thick and is composed of cross- linked bovine tendon collagen and glycosaminoglycan that is indicated for the treatment of acute and chronic wounds, including diabetic skin ulcers. A 1 cm ^c 1 cm sample of the dermal regeneration matrix was cut with a razor blade and sandwiched between the oxygenation platform 45 and the agarose gel 30. The rest of the experiment proceeded as above. As a control experiment, this test was repeated with empty microfluidics (i.e., no H2O2).

[0059] To investigate the range spatial effect of an oxygenation spot on a gel substrate, the oxygenation experiments were repeated for multiple locations, and the rate of oxygenation was plotted as a function of both vertical and horizontal distance from the generation spot.

[0060] The results from the diffusion experiments into agarose gel 3 mm deep are

presented in FIG. 3. For the case without a dermal regeneration matrix, the solid bold line curve shows a monotonically-increasing oxygen level (from a partially hypoxic level of 15 % to 40 % 3 hours later) in the agarose gel 3 mm below an oxygenation spot. The curves show saturation in the oxygen level since for these experiments, a fixed amount of

H2O2 was used (rather than a continuous flow). Although the level shown is not 100 % saturation, the results do show that the platform is able to successfully raise the oxygen concentration 3mm within the gel BO to levels which are far from hypoxic. Therefore, if the gel 30 were a wound, it would be reasonable to expect improved healing as deep as 3mm (or more) as a result of the oxygenation platform.

[0061] The two remaining curves represent the tests with dermal regeneration matrix and show a different trend. In particular, the solid thin line curve, corresponding to the setup with dermal regeneration matrix and peroxide-filled microfluidics contains an initial shallow slope; this lag in the increase of oxygenation can be attributed to the extra time required for oxygen to diffuse through the dermal regeneration matrix (integra) layer. After 2.5 hours, however, the solid thin line curve exhibits its highest rate of change in oxygen concentration (slope of 18.9% per hour); this rate is similar to the largest rate of the sample without dermal regeneration matrix (17.1 % per hour), suggesting that although the dermal regeneration matrix causes an initial lag in oxygen diffusion, the eventual diffusion rate of oxygen approaches that of the oxygen generation platform 45. For comparison, the oxygen level does not increase during this time for the sample (dashed line curve) that does not contain peroxide in the microfluidics.

[0062] One feature of the curves that should be clarified is the initial drop in oxygen for the two dermal regeneration matrix samples. For both of these cases, the data shows an initially normoxic oxygen level. This corresponds to the reading of the oxygen probe 40 in atmosphere, prior to insertion into the gel 30 (at time 0). Following insertion, the oxygen concentration drops steadily. Although a quick drop in oxygen concentration (to hypoxic levels in the gel) might be expected, the curves show a 20-30 minute steady decay which may be attributed to atmospheric oxygen trapped in the probe protector needle (described in the experimental setup above) which needs time to diffuse into the gel 30. After 30 minutes, however, the curves reach their minimum values (the oxygen level in the hypoxic gel, <15 %0 ₂ ).

[0063] FIG. 4 shows the 3D spatial oxygen concentration by diffusing through a 0.9mm thick dermal regeneration matrix into the hypoxic gel. The maximum oxygen diffusion rate is 0.09 %/min (percentage per minute) at the surface of the gel just below the catalyst spot (0mm depth and 0mm horizontal distance); while the minimum oxygen diffusion rate is 0.004 %/min at the position of 2.2mm depth and 15mm horizontal distance inside the gel. The oxygen diffusion rate show a normal distribution in both the depth and horizontal direction. Within the 80% area under the normal curve, the critical oxygen diffusion rate is calculated by 0.09/e=0.03 %/min. The oxygen generated from a 3x3mm2 catalyst spot can therefore cover a range with the radius of 10mm following the surface and the depth of 2.2mm directly beneath it. The oxygen concentration distribution through a single oxygen generation source provides an experimental baseline for designing an oxygen generation platform with multiple sources to achieve an efficient (optimal) oxygen delivery rate for a large scale chronic wound.

[0064] Another aspect of the present invention is a mass-reproducible technique for creating PDMS micro channels and bonding them to parchment paper patterned with selective catalyst and oxygen sensitive dye deposited. A repeatable bonding procedure between PDMS and parchment paper provides for high-volume production. Several mass production technologies, such as screen printing, inkjet printing, lamination, etc., may be utilized to produce smart dressings according to the present invention.

[0065] PdTFPP (5,10,15,20-Tetrakis(pentafluorophenyl)-21H,23H-porphine palladium(ll)) and Ru(dpp) ₃ Cl2 (Tris (4,7-diphenyl-l,10-phenanthroline) ruthenium II dichloride) are suitable candidates for oxygen sensing materials due to their wide application as oxygen indicator. When the fluorescent dyes are exposed to UV/visible light (for example 455 nm blue light) in the presence of oxygen, the oxygen atoms srike the fluorescing complex and cause a change in energy which quenches its fluorescence. A higher oxygenated environment creates a higher possibility for such collisions to happen between oxygen atoms and the fluorescent material, resulting in a lower fluorescence level. The fluorescent properties of two materials being capsulated in both poly-styrene and PDMS (polydimethylsiloxane) were then measured to determine the suitability of the materials in an oxygen sensing system.

[0066] PdTFPP was purchased from Sigma-Aldrich and Ru(dpp) ₃ was purchased from

Cayman Chemical. A solution of the material was made by dissolving lmg of PdTFPP or

Ru(dpp)3CI2 powder into lmL of chloroform. For PdTFPP, lmg of powder was also dissolved into lmL of heptane for testing. Poly-styrene (PS) sensing patches were prepared by mixing PS and dissolved fluorescent solution at a ratio of 1:10 by weight. 20 pL of mixed solution was then cast onto filter paper with a diameter of 8mm. The patch was left in the nitrogen chamber for drying for 24-hours before testing. The PDMS encapsulated sensing patches were fabricated by firstly depositing fluorescent solution onto the filter paper. PDMS was then added to the filter paper after the removal of the solvent. The same amount of dye was used for all samples. The fluorescence spectrum was measured using a UV spectrophotometer. A zero oxygen solution was prepared by nitrogen bubbling a 0.15mol/L Na2S03 solution for 30 minutes.

[0067] Excitation spectrums of different oxygen sensing materials were measured first, as shown in FIG. 5. The x-axis is the wavelength of the excitation light, and the y- axis represents the RFU (relative fluorescence unit). An excitation peak at 407 nm can be detected. For comparison, PdTFPP was dissolved into two different solvents at same concentration. While comparing the chloroform solution, FIG. 5 (a-b), and heptane solution, FIG. l(c-d), a smoother plot was obtained with chloroform solution. With reference to FIG. 5 (b)(d), a higher RFU can also be observed after 10 minutes of evaporation of the solvent. These results demonstrate that, the existence of a solvent affects the photo-property of the sensing material. Thus, to increase the sensitive and stability of the sensing film, any solvent that may be present should be removed completely before further fabrication.

[0068] With further reference to FIG. 6, the excitation spectrum of Ru(dpp) ₃ Cl2 was also tested. Within the visible light range (390 nm-700 nm), an excitation peak at 460 nm was detected. As compared to PdTFPP, a blue light source can be used for Ru(dpp) ₃ Cl2 while UV light is required for PdTFPP. Ru(dpp) ₃ C is therefore preferred for oxygen sensors that are embedded (integrated) with a wound dressing system.

[0069] The excitation spectrum was also measured for PdTFPP deposited on filter paper with PS (poly-styrene) and PDMS, as shown in FIGS. 7 and 8, respectively. A higher peak RFU can be observed with PS. When the sensing material being mixed with PDMS, more photo noise was introduced and fluorescent intensity (RFU) at the peak decreased. As a result, PS can provide a better photo-stability protection to PdTFPP without

compromising the photo characteristic.

[0070] Emission characterization was then conducted. When dissolved with chloroform,

PdTFPP has three emission peaks. With reference to FIG. 9, a peak at 675 nm remains while solvent is being evaporated. Thus, the effect of solvent on the fluorescent material can be again confirmed from the emission profile. With reference to FIG. 10, the photo property of dissolved Ru(dpp) ₃ C in chloroform was not affected by the solvent. A clear peak is illustrated at 625 nm. FIGS. 11 and 12 compare the emission profile while PdTFPP being deposited with PS and PDMS. Same trend was observed as the sample with PDMS had three emission peaks which is not expected for oxygen sensing.

[0071] For the validation of the concept, emission intensity under different oxygen levels was also tested using the same experiment setup. For comparison, RedEye ^® was purchased which has been widely used as non-invasive oxygen indicator. PDMS encapsulated Ru(dpp) ₃ Cl2 and PS encapsulated PdTFPP were also tested. The results are shown in FIGS. 13-15. A fluorescent intensity difference of 3.38 times was obtained from the commercial patch. For Ru(dpp) ₃ Cl2 and PdTFPP, a difference of 1.4 times and 2.79 times, respectively, were measured under various oxygenated environments. The emission intensity decreases with an increase of the oxygen concentration in water from 0% to 20%.

[0072] In order to protect the functionality of the sensing materials during sterilization, the photo properties of different retrieved samples were measured after a sterilization process (H2O2 vapor treatment), as shown in Table 2. Ru(dpp) ₃ Cl2 was tested to be more vulnerable to the H2O2 vapor treatment and PS is needed for protection. PdTFPP is more stable as compared to Ru(dpp) ₃ Cl2; however, a decrease of the photo-reaction intensity can be observed after sterilization when no polymer was added.

T¾ί¾¾ί ii. ed? urnu d%i' dsS&diSd? [0073] Printing techniques for the manufacturing of electronics on flexible substrates have been developed. Printing methods that have been investigated for the direct printing of electronics include, but are not limited to, inkjet, flexographic, screen and rotogravure printing.

[0074] An oxygen detection sensor according to one aspect of the present invention may be fabricated using inkjet printing process. In inkjet printing, precise control of ink interactions at the substrate surface may depend on the ink formulation and substrate morphology. Typically, in inject printing, the viscosity of the ink should be below about 10 centipose (cP) to properly jet the ink from the nozzles of cartridge. In printed electronics, various applications may require substrates with different surface properties.

[0075] Four different substrate materials were tested for oxygen generation. These

materials included: unrasted parchment paper, laser rastered parchment paper, unrasted Tyvek ^® paper and laser rasted Tyvek ^® paper. For the preparation of ink, Tris(4,7- diphenyl-l,10-phenanthroline)ruthenium(ll) dichloride dye, ethanol (ACS

spectrophotometric grade) solvent and polymers (binders) such as polystyrene (38% emulsified in H ₂ 0) and polydimethylsiloxane (PDMS) was used.

[0076] Substrate characteristics such as roughness and thickness were measured using a

Bruker ContourGT-K interferometer. The average thickness of the unrastered

parchment, rastered parchment, unrastered Tyvek ^® and rastered Tyvek ^® substrates were measured to be 73.8 ± 2.2 pm, 73.3 ± 1.1 pm, 206.2 ± 4.2 pm, and 208.8 ± 6.3 pm, respectively. The root mean square (RMS) roughness of the unrastered parchment, rastered parchment, unrastered Tyvek ^® and rastered Tyvek ^® substrates were measured to be 7.0 ± 0.5 pm, 6.5 ± 0.4 pm, 5.1 ± 0.1 pm and 4.6 ± 0.4 pm, respectively. From the measured values, it is understood that thickness of the substrates were not impacted by laser rastering. The roughness of the rastered Tyvek ^® paper substrate was decreased by 10.9% when compared to the unrastered Tyvek ^® paper. Similarly, the roughness of the rastered parchment paper substrate was decreased by 7.8 % when compared to the unrastered parchment paper.

[0077] Two test ink solutions were prepared for testing, including:

1. Dye + Ethanol + Polystyrene (1:100:1, by mass) - 0.3 g of dye and 0.3 g of polystyrene were mixed with 30 g (40 ml) of ethanol solvent. 2. Dye + Ethanol + PDMS (1:100:1, by mass) - 0.3 g of dye and 0.3 g of PDMS were mixed with 30 g (40 ml) of ethanol solvent.

Both test ink solutions were mixed on a hotplate with magnetic stirrer at 525 rpm; overnight (12 hours) under a fume hood to obtain homogenous ink solutions.

[0078] The Z-number is a dimensionless constant, and a measure of density, surface

tension and viscosity. For proper jetting of ink, the Z-number should be in the range of about 2 to about 10. The formula for Z is:

Where:

d is the nozzle diameter (21.5 miti),

p is the liquid density,

Y is the surface tension and

h is the ink viscosity.

[0079] Inks with viscosity less than 10 cP are typically preferred for inkjet printing. For the dye + ethanol + polystyrene based test ink solution (solution #1), the surface tension was measured using the FTA200 and is 21.95 ± 0.1 dynes/cm. The measured density of test ink solution #1 was 0.766 g/ml.

[0080] To determine the viscous behavior of ink solution #1 under a broad range of

temperatures from 20° C to 60° C, the rheometer was used. The shear rate was maintained at 1000 (1/s) and the viscosity was decreased from 3.77 cP to 2.11 cP for the temperature range of 20° C to 60° C. After substituting the measured values, Z-numbers ranging from 5 to 9 were calculated as the temperature increased from 20° C to 60° C. The test results show that ink solution #1 is suitable for inkjet printing at room

temperature.

[0081] For the dye + ethanol + PDMS based ink solution (solution #2), the measured

surface tension was 21.81 ± 0.08 dynes/cm. The measured density of ink solution #2 was 0.7669 g/ml. For the shear rate of 1000 (1/s), viscosity decreased from 3.76 cP to 2.10 cP for the temperature range of 20° C to 60° C. The calculated Z-number for ink solution #2 increased from 5 to 9 as the temperature increased from 20° C to 60° C. Based on these test results, it is evident that ink solution #2 is also suitable for inkjet printing at room temperature.

[0082] The Z-number and other characteristics such as viscosity, density and surface tension for both ink solutions are similar. This may be because the effect of very small quantities of polymers, polystyrene and PDMS in the ink solutions is negligible.

[0083] The contact angle was measured for ink solution #1 on the four substrate

materials (unrastered parchment paper, laser rasted parchment paper, unrasted Tyvek ^® paper and laser rastered Tyvek ^® paper) using an FTA 200 instrument. The measured contact values of the ink drops on the unrastered parchment paper, rastered parchment paper, unrastered Tyvek ^® paper and rastered Tyvek ^® paper substrates were 20.7 ± 0.1 degrees, 35.3 ± 0.5, 13.2 ± 0.7 degrees and 12.4 ± 0.1 degrees, respectively. While measuring contact angles with the FTA200 instrument it was observed that the ink drops were spreading rapidly on both the unrastered and rastered Tyvek ^® substrates. Even though it is evident from the contact values that all the substrates possess good wetting characteristics, the tested Tyvek ^® substrates may not be suitable for printing due to rapid spreading of ink on the surface. Thus, surface modifications of Tyvek ^® substrates may be required. For example, plasma or UV treatment may be utilized to alter the surface properties (to increase contact angle) of Tyvek ^® materials so that spreading of ink on the surface can be controlled prior to curing of the ink.

[0084] An oxygen sensing and electronic interfacing system (bandage) according to the present invention can be used to measure and maintain a suitable amount of oxygen at the interface of the wound and the bandage. In short, the smart dressing 1 of the present invention measures the amount of oxygen present, and pumps more oxygen as necessary.

[0085] Smart Dressing 1 includes an optical oxygen sensor (module) 20 with a signal processing circuit to monitor the amount of oxygen present. A ruthenium-based dye may be utilized as the oxygen-dependent (sensing) compound. Similar to other dyes, when the ruthenium dye is excited by a blue LED, it produces an orange fluorescence. The fluorescence signal is dependent on the amount of oxygen present. In contrast to hypoxic conditions, it is known that when oxygen is present, fluorescence is less intense and decays more quickly. By characterizing the fluorescence, the amount of oxygen present can be quantified (measured). One method to quantify oxygen is to excite the dye until it reaches a steady state, then directly measure the peak intensity as well as the time it takes to decay. However, this method is sensitive to the precise positioning of the LED source, the amount of background light present and the inevitable photobleaching of the dye over time. To avoid these issues, a system according to one aspect of the present invention modulates the excitation blue LED at a frequency between 20 kHz and 75 kHz. The phase difference between blue excitation and the resulting orange fluorescence signal can be measured. This phase difference changes with the amount of oxygen present.

[0086] A DC bias to an excitation source (blue LED) may be provided to turn it on. An AC signal may then be superimposed to modulate the intensity of the blue LED. The resulting fluorescence signal has a phase shift that increases with the amount of oxygen present. The fluorescence signal may be amplified and processed so that its phase can be compared with the excitation signal. This phase shift is an indicator (measurement) of the amount of oxygen present.

[0087] Phase detection may be accomplished with digital logic by using a single exclusive or (XOR) gate. When the XOR gate receives two in-phase signals, its output is low (0 volts DC). When the XOR gate receives two completely-out-of-phase signals, its output is high (e.g. 4.5 V DC). When the signals are slightly out-of-phase, the output of the XOR gate is high for a short time, and then low for the rest of the cycle. The assertion time (pulse width) for one cycle of the XOR gate's output increases as the input signals move more out-of-phase. The output of the XOR gate may be low-pass-filtered to produce a DC output that corresponds to the phase of its inputs.

[0088] Signal generation, measurement and decision making may be mediated by a

microcontroller 50. The microcontroller 50 generates a square wave at a specified frequency, which is converted to a DC-biased sine wave by a filter 52. The DC-biased sine wave from filter 52 drives a blue LED 52, so that blue light from the LED 54 excites the ruthenium dye 56. The excited ruthenium dye 56 fluoresces, and the sinusoidal fluorescence (orange) is measured with a highly-sensitive photodiode 58. The

photodiode 58 includes a light filter so that it picks up only the orange fluorescence of the dye 56, and not the blue LED's emission. A processed fluorescence signal is sent from the photodiode 58 to a phase comparator (detector) 60. Because the phase detector 60 requires square wave inputs, the sinusoidal fluorescence signal is converted to a square wave by a comparator circuit. After the fluorescence signal is converted, its output is sent to the phase comparator 60.

[0089] Finally, the microcontroller 50 receives a DC input from the phase comparator 60 that represents the current oxygen present. The microcontroller 50 then decides whether or not to pump hydrogen peroxide to generate more oxygen.

[0090] As discussed above, untreated Tyvek ^® is generally not suitable for printing.

Therefore, the surface properties of the Tyvek ^® substrate may be altered by treating its surface with a fusion UV system. Typically, UV treatment raises the surface energy of a substrate through oxidation which in turn increases the polar energy, potentially providing improved wetting. For test purposes, unrastered and rastered Tyvek ^® substrates were UV treated 1 to 4 times. The contact angle of the ink drops on the substrates were then measured (Table 3).

[0091] The contact angle of the unrastered and rastered Tyvek ^® substrate before UV treatment was measured as 13.2 ± 0.7 degrees and 12.4 ± 0.1 degrees, respectively. The contact angles of the unrastered Tyvek ^® substrates that were UV treated for 1 time, 2 times, 3 times and 4 times were measured as 14.8 ± 0.1 degrees, 13.4 ± 1.1 degrees, 12.4 ± 1.1 degrees and 12.7 ± 1.6 degrees, respectively. Similarly, the contact angles of the rastered Tyvek ^® substrates that were UV treated for 1 time, 2 times, 3 times and 4 times were 10.5 ± 1.1 degrees, 13.8 ± 1.0 degrees, 14 ± 1.4 degrees and 12.1 ± 0.6 degrees, respectively.

Table 3: Contact angles of the UV treated Tyvek ^® substrates.

[0092] From the measured contact angles and through the live video option (spreading and absorbing behavior of the drops on the substrate can be seen) in the FTA 200 software, it was concluded that the impact of the UV treatment on the surface of the Tyvek ^® substrates is minimal. Therefore, further characterizations on the Tyvek ^® substrate were not performed.

[0093] It was observed from measured values that the roughness of parchment paper samples was not consistent. In order to obtain a similar smoothness over the surface of parchment paper, a calendering process was employed. A calendering machine was used to calender both sides of the parchment paper, with an applied pressure of 35 Psi (241 kPa). It was observed that, due to calendering, the roughness of the parchment paper was reduced from 8.7 ± 1.7 pm to 5.5 ± 0.4 pm (Figure 1.3 (a) and (b)). [0094] During testing, multi-layer samples (5 layer, 3 layer and 1 layer) of ruthenium dye based ink, with ethanol as solvent and PDMS as binder, was inkjet printed on to both unrastered and rastered parchment paper in an array of circular spots with a diameter of 5 mm with 20 pm drop spacing, using a DIMATIX inkjet printer (DMP 2831). The ruthenium ink with polysterene as binder could not be inkjet printed because of its comparatively large particle size (<500 nm). The ruthenium ink with PDMS was loaded into a DIMATIX DMC-11610 cartridge (10 pi) through a 25 mm disposable Whatman syringe filter, with a poly vinylidene difluoride filter (PVDF) filter membrane of 0.2 pm pore size, to filter any large particles that may have agglomerated in order to achieve smooth printing. Each layer of the printed ink was cured on the stage of the inkjet printer for 5 minutes at 60 °C. A 27 V actuation voltage was applied at 5 kHz firing frequency. The corresponding waveform and cleaning cycles employed for inkjet printing are shown in FIGS. 16 and 17, respectively.

[0095] Roughness and thickness measurements were performed to characterize the print quality of the printed samples (Table 4). The root mean square (RMS) roughness of the 5 layer, 3 layer and 1 layer printed sample was measured to be 6.0 ± 0.03 pm, 6.4 ± 0.48 pm and 6.7 ± 0.44 pm, respectively for the unrastered parchment paper. Similarly, a roughness of 6.0 ± 0.42 pm, 6.2 ± 0.31 pm and 6.8 ± 0.29 pm was measured for the 5 layer, 3 layer and 1 layer printed samples, respectively for the rastered parchment paper. Before printing, the roughness of the unrastered and rastered parchment paper substrates were measured to be 7.0 ± 0.5 pm and 6.5 ± 0.40 pm, respectively. From the measured roughness values, it is thus understood that the thickness of the substrates could not be measured because the printed ink did not cover the entire roughness of the substrates (Inkjet printing provides a layer that is about 0.5 pm thick).

Table 4 Roughness measurement of the inkjet printed ruthenium dye based ink.

[0096] After initial tests for oxygen sensing, the circular spot size was increased to 7.5 mm to increase the concentration of dye. With reference to FIG. 16, an array of circular spots was inkjet printed (10 layer, 7 layer and 5 layer) on the unrastered parchment paper with 10 pm drop spacing and cured at 45° C.

[0097] From the printed samples, it was observed that thermally cured ruthenium

particles were falling off from the parchment paper. This was due to poor adhesion between the ink and the parchment paper substrate. If the surface energy of the substrate is higher than the surface tension of the ink at least by 10 dynes/cm, then the ink binds/adheres well to the substrate. The surface energy of the calendered parchment paper was measured with the FTA 200 using Owens-Wendt method and was calculated as 21.99 dynes/cm. The surface tension of the ink is 21.81 ± 0.08 dynes/cm. The poor adhesion of the ink was caused, at least in part, by the small difference (less than 10 dynes/cm) between the surface energy of the substrate and surface tension of the ink. To improve adhesion, the surface energy of the substrate may be improved (raised) either by UV or corona treatment. During testing, the surface of parchment paper was UV treated 4 times using the Fusion UV System ISOOMB. The surface energy of the UV treated parchment paper was measured to be 22.09 dynes/cm. Thus, the impact of the UV treatment on the parchment paper is minimal. Similarly, no impact was observed with the corona treatment on the surface of parchment paper.

[0098] As discussed above, smart dressing 1 (FIG. 1) includes fluid channels 15. FIGS. 17A and 17B show microfluidic networks 65A and 65B, which represent examples of the channels 15 of FIG. 1. White regions or lines 67 comprise fluid channels, and black hexagonal areas 69 represent locations (cells) of the oxygen sensing dye. FIGS. 17A and 17B shows two designs, a smaller one 65A, and a larger one 65B, to allow the patch to be adapted for wound of various sizes. The designs 65A and 65B exhibit a honeycomb pattern due to the spatial and radial uniformity of the designs. The size of the hexagonal unit cells 69 was determined based on the test results discussed above. Specifically, testing showed that a 1 mm oxygen-generating spot could influence oxygenation within a 5-10 mm radius. Thus, the unit cells 69 have a radius of 7.5 mm. Testing of large-area bonding of PDMS to parchment paper via the use of partially-cured PDMS (as an interface between the two layers) was also performed. Testing revealed that this technique is not sufficiently strong for assembling larger devices because it often results in delamination and/or leakages.

[0099] To remedy the leakage and reliability issue, a bonding process utilizing lamination rollers was developed. Such rollers are typically found on hot lamination machines. During testing, a commercially available hot laminator machine 68 (Apache) was incorporated into the fabrication process as shown in FIG. 18. Specifically, paper/PDMS bilayers 72 were passed through rollers 70 before curing was complete, allowing the rollers 70 to apply pressure and squeeze out trapped gasses from the interface between the paper and PDMS. This technique allowed the successful creation of patches with stronger bonding and no leakages. Examples of these are shown in FIG. 19. The left- hand subfigure (image) shows a patch with Mn0 ₂ catalyst deposited on only the junctions of the microchannels (a few of which are identified with red circles), whereas the right-hand subfigure (image) shows a patch with channels that have been lined with MnC>2 (an alternative, pattern-less technique). The resulting patches exhibit functionality and robustness while remaining thin for conforming to human skin. [00100] As discussed above, smart dressing 1 senses (measures) oxygen levels, and provides controlled flow of oxygen to a wound based, at least in part, on measured oxygen levels. Oxygen delivery is initiated with injected H ₂ 0 ₂ over the printed Mn0 ₂ on a parchment paper. The volume and duration of oxygen delivered to the wound is precisely controlled. The volume duration of oxygen is also observed (measured) while the smart wound dressing is worn by a patient. The concentration of 0 ₂ is monitored (measured) in real time to control the 0 ₂ delivery based on a user's demand, or the condition of the patient.

[00101] A Ruthenium complex may be used to measure the concentration of oxygen at the wound. During testing, the performance of printed Ru(dpp) ₃ CI ₂ (Ruthenium dye) with different combinations of materials was characterized. First, a commercial optical oxygen sensor (Redeye ^® patch from Ocean Optics) was characterized and observed.

Then, printed Ruthenium dye was characterized with different compositions of PDMS, polystyrene, and ethanol/chloroform.

[00102] The main mechanism of this sensor is the attraction of oxygen atoms by the

Ruthenium-complex. Ruthenium dye excites when exposed to light have a wavelength of about 455 nm (Blue), and emits fluorescence of 610 nm (red). The Ruthenium dye fluorescence quenching is observed when oxygen atoms collide to fluorescence

Ruthenium-complex transferring its energy. Thus, the quenching fluorophore results in lower fluorescent intensity in an environment having a high concentration of oxygen environment. This printed oxygen sensor can work in range of 0 to 100 % of oxygen environment. It is used mostly for in situ and real-time monitoring of oxygen generation in water. For the accuracy in measurement and observations, both emitted wavelength of the fluorophore and dissolved oxygen concentration in water were characterized by optical and electrochemical sensors.

[00103] For testing purposes, Ruthenium dye 84 (FIG. 20) was printed (diameter of 7.5 mm) on a parchment paper, then attached to transparent double-sided tape 74. The tape 74 was then placed on a wall 76 of a water container 78 facing outside for optical measurement as shown in FIG. 20. Water 80 was deoxygenated by pumping N ₂ gas over

30 minutes. After deoxygenating, initial dissolved oxygen concentration of the water was measured with an electrochemical probe 82. The measured oxygen concentration was about 0.2 ppm at room temperature. It will be understood that "normal" (untreated) water contains oxygen concentration of 8 to 9 ppm (1 ppm = 1 mg/L) in air. An optical probe 86 was placed about 2 mm away from the wall 76 of the water container 78 and positioned perpendicular to the dye 84. The electrochemical probe 82 was dipped into the water 80 deep enough so that a thermal sensor embedded in the probe 82 was completely submerged under the water 80. Two probes were calibrated before the measurement using two-point calibration (0 % and 100 % saturated water). A stirring magnet 88 was placed under the water 80 and stirred at 150 rpm. This ensured that to make 0 ₂ concentration inside the container was at equilibrium at all time.

[00104] Characterization was first conducted with a Redeye ^® patch, and compared to a first printed Ruthenium sensor. To optimize the performance of the printed dye, multiple layers of sensors with different compositions were tested. These tests were conducted with one test sensor unit left in air up to 21 % of oxygen concentration (9 ppm) and another pumped with 0 ₂ gas close to 100 % of oxygen dissolved in the water.

[00105] Dissolved oxygen measurement and corresponding fluorescent lifetime of a

Redeye ^® patch (commercially available oxygen sensor, Ocean optics) was first

characterized by leaving deoxygenated water 80 in air above open top 77 of container 78, thereby slowly dissolving (absorbing) oxygen. An initial reading of the Redeye ^® patch in the deoxygenated water had maximum fluorescent lifetime of 20.342 psec at 0.62 ppm of dissolved oxygen. Data was collected every 15 minutes until the oxygen concentration reached equilibrium to air (9 ppm). The graph (a) of FIG. 21 shows that fluorescent lifetime of the Redeye ^® patch exponentially decreases over time until it is in equilibrium to air. After water reaches the equilibrium with air around 9 ppm, the fluorescent lifetime was saturated around 9 psec as shown from the plot. Resulting fluorescent lifetime at 9 ppm was 9.247 psec.

[00106] A first batch of printed Ruthenium test sensors according to one aspect of the present invention was characterized using the same setup (FIG. 20) as the Redeye ^® patch. The samples of Ruthenium dye were printed in multiples layers of 3, 5, and 10- layers with compositions of Ruthenium dye, polystyrene, and chloroform in 1:1:100 ratio.

Sensors with 3 multiple layers were printed on two different conditioned substrates

(parchment paper) that were laser treated and non-laser treated. The laser treatment was performed to increase the adhesion of the Ruthenium dye to the paper and for better absorbance of 0 ₂ . Each sample was submerged and placed on the wall 76 of the container 78 under deoxygenated water 80, then measured using both optical probe 86 and electrochemical probe 82, leaving water to dissolve (absorb) oxygen from air overtime. Compared to the Redeye ^® patch, the printed Ruthenium dye showed 0.2 times smaller fluorescence lifetime at the initial reading of deoxygenated water. The fluorescent decay of the Redeye patch was 10 times faster than the printed sensor as shown by the lower graph (b) of FIG. 21.

[00107] The lower graph (b) of FIG. 21 shows that, the emitted wavelength from the excited Ruthenium dye was successfully detected using the optical probe 86. The fluorescent lifetime of the printed sensors also had exponential decay over time like the Redeye ^® patch. However, multiple layered samples did not show a significant variation except 10-layered samples. Printed sensors with 10-layers had the lowest performance among the group due to the poor adhesion between the dye and the parchment paper. Some fragments of particles of the Ruthenium dye fell off from the parchment paper. Other test samples such as the B and 5 layer samples also were not uniformly. However, the sensing of these non-uniform 3 and 5 layer samples was not significantly different as was the case for the 10 layer samples. Laser treatment of parchment paper did not show a significant change in sensing as the plot shows between laser treated and non-treated samples.

[00108] During testing, a second batch of sensors were printed. These test samples

included single layer samples and, 2 and 3 layer samples. The samples were printed on unrastered parchment paper. Materials used to print this batch were Ruthenium dye, ethanol, ethyl cellulose mixed in 1:1:100 ratio by weight. The second batch samples showed better uniformity compared to the first printed sensors. These samples were primarily characterized in a deoxygenated water container 78 having an opening 77 exposed to air/oxygen such that the water continuously dissolved oxygen from the air until the oxygen concentration in the water reached equilibrium.

[00109] Another aspect of the present invention is a portable circuit which uses the

fluorescence quenching method to monitor oxygen concentration.

[00110] By exciting the ruthenium dye periodically with blue light, measuring the periodic fluorescence of the dye and calculating the delay between excitation and emission, it is possible to extrapolate oxygen concentration using the Stern-Volmer formula. [00111] As discussed above, in connection with FIG. 15, a microcontroller-generated square wave may be fed into a series of low-pass filters, which act as a square-to-sine converter. The sine wave is then used to drive a blue LED 54 and excite the ruthenium dye 56. A transimpedance amplifier is used to capture the sinusoidal fluorescence of the ruthenium dye 56, and this fluorescence signal is then compared to the original excitation signal so that the microcontroller 50 can calculate the amount of oxygen present and control the hydrogen peroxide pump accordingly.

[00112] Circuit operation, corresponding to the schematic:

1. Tl MSP430G2553 Microcontroller provides a square wave at a pre programmed frequency

2. A series of low-pass filters remove the high frequency content of the square wave, acting as a square-to-sine converter.

3. The sine wave is fed into a filtering operational amplifier, the output of which drives the excitation LED

4. The transimpedance amplifier (Rll = photodiode) picks up the fluorescence signal

5. The sinusoidal fluorescence is converted into a square wave for processing

6. An XOR gate compares the original square wave to the fluorescent square wave. As the phase between the two signals increases, the outputted pulse width increases.

7. The output pulse of the XOR gate is low-pass filtered to a DC voltage. Higher duty cycle of XOR gate corresponds with higher DC voltage.

[00113] The cytotoxicity of the materials used for the fabrication of the smart wound dressing 1 described above was investigated following standard ISO 10993-05

(Cytotoxicity) and ISO 10993-12 (Sample preparation and reference materials).

[00114] All samples were < 0.5 mm thick and prepared as 8 mm-diameter discs (surface area of 0.50 cm ² ). Samples were sterilized by the STERRAD ^® process (low temperature hydrogen peroxide gas plasma) and then extracted for 24 h/37°C in complete growth medium (Eagle's Minimum Essential Medium + 10% horse serum + 100 lU/ml penicillin +

100 pg/ml streptomycin) using an extraction ratio of 6 cm ² /ml. At the time of the extraction, L-929 mouse fibroblast cells (NCTC clone 929: CCL 1, American Type Culture Collection, Manassas, VA, USA) in passage S were lifted from the culture flask using trypsin/EDTA. An aliquot was counted using trypan blue, and then cells were re suspended in complete growth medium at a density of 1 x 105 cells/ml. Cells were dispensed into wells of 96-well culture plates (1 x 104 cells/well) and cultured at 37°C in a humidified atmosphere of 5% C0 ₂ /95% air. After 24 h, the culture medium was removed and replaced with 100 pi of extractant. Some wells received sodium dodecyl sulfate (0 to 400 mM in EMEM; positive controls), low-density polyethylene extract (1.25 cm ² LDPE/ml EMEM; negative control) or complete growth medium alone. Cells were then cultured for an additional 24 h. Images (mag. of 100 and 200x) of cell cultures were recorded by photo microscopy using a Olympus CK40 inverted microscope and Insight2 SPOT camera (Diagnostic Imaging) and the number of attached and dead cells were manually counted at a later time using ImageJ (NIH). In addition, images were graded for morphological evidence of cytotoxicity using the ISO 10993-5 standard, where the 0 to 4 scale represents no, slight, mild, moderate or severe cytotoxicity, respectively.

Subsequently, cells in culture plates were washed once with Hank's Balanced Salt Solution and metabolic activity was measured by incubating cells with 100 mI of WST-1 cell proliferation reagent (Roche Diagnostics) for up to 4 h at 37°C. To determine cytotoxicity, absorbance of the medium in wells was measured at 450 nm after 2 and 4 h using a microplate reader (PHERAstar) and was corrected using absorbance

measurements at 630 nm and using blanks. To check for mycoplasma contamination of the cultures, medium was saved and tested using the luminescent MycoAlert Plus mycoplasma detection kit (Lonza).

[00115] Absorbance is proportional to the amount of formazan product generated by the metabolic activity of cells. Thus, lower absorbance values correlate with increased cytotoxicity. Mean absorbance values for cells treated with extracts of palladium, palladium + polystyrene or palladium + PDMS on paper substrates (0.734, 0.816 or 0.811, resp.) are similar to values for cells incubated in EMEM alone (0.827) or the LDPE extract

(negative control; 0.753). However, cells treated for 24 h with the extracts of ruthenium or ruthenium + polystyrene show considerable cytotoxicity (corrected absorbance readings of 0.318 or 0.089, resp.), with only 38.4% or 10.7%, respectively, of the metabolic activity of cells that were cultured in EMEM alone. The extract created from ruthenium + PDMS on paper was borderline non-cytotoxic, having a corrected mean absorbance reading of 0.625 or 75.6% (readings below 70% would be considered cytotoxic according to ISO 10993-05).

[00116] The results are confirmed qualitatively via microscope images. Micrographs of cells treated with EMEM, palladium, palladium-polystyrene, palladium-PDMS and ruthenium-PDMS extracted media had cytotoxicity scores of 0-1, while micrographs of cells treated with ruthenium and ruthenium-polystyrene extracted media had scores of 2-3. According to the standards, scores > 2 are considered to be cytotoxic.

[00117] The results of the cytotoxicity assay on the various combinations of oxygen- generation materials are summarized in FIG. 23. Mean absorbance values for cells treated for 24 h in EMEM or extracts of Tyvek ^® were 0.768 or 0.822, respectively.

However, extracts made from Mn0 ₂ on Tyvek ^® or Mn0 ₂ on parchment paper were cytotoxic, producing mean absorbance values of 0.005 or 0.399, respectively. This represents metabolic activity of only 0.7% or 52%, respectively, of healthy cells.

Morphological grading confirmed the findings. Cells treated with EMEM or extracts of Tyvek ^® , Mn0 ₂ on Tyvek ^® and Mn0 ₂ on parchment paper produced scores of 0, 1, 4, and 3, respectively.

[00118] Cells in additional wells were cultured in 0-400 mM sodium dodecyl sulfate in

EMEM for 24 h to serve as positive cytotoxicity controls. The test results confirmed that increasing concentrations of SDS produced a graded and increasing cytotoxic response as expected. Mean absorption readings fell from about 0.7 to 0 between 0 and 300 pM SDS. Morphology scores ranged from 0 to 2 between 0 and 150 pM SDS. Concentrations of SDS above 200 pM were cytotoxic.

[00119] To test cell cultures for mycoplasma contamination, a MycoAlert Plus kit was used. Luminescence of the test solution is measured in the presence of reagent alone or reagent plus substrate and ratios are calculated and compared to positive and negative controls that are purchased with the test kit. Ratios < 0.9 are negative and >1.2 are positive for mycoplasma. Borderline values between 0.9 and 1.2 are retested after 24 h. The cell cultures used for the cytotoxicity study produced a ratio of 0.31 (negative), while positive and negative controls produced ratios of 22.38 and 0.32, respectively.

[00120] Using appropriate positive and negative controls, extracts of palladium,

palladium + polystyrene and palladium + PDMS were non-cytotoxic. Ruthenium + PDMS was marginally non-cytotoxic. Extracts of ruthenium alone on paper and ruthenium + polystyrene were cytotoxic. Extracts of Mn0 ₂ on parchment paper or on Tyvek ^® were cytotoxic, but Tyvek ^® alone was non-cytotoxic. Cultures tested negative for mycoplasma.

[00121] Based on the print quality of the ruthenium dye (Ru + Ethanol + PDMS) ink

discussed above, it was concluded that the film formation of Ru dye and its adhesion with parchment paper is poor. This may be due to the insolubility of PDMS with the ink system. To improve both the film formation and adhesion, among various binders such as ethyl cellulose (polymer) ethyl cellulose was chosen because of its solubility in ethanol and better film formation properties. Ruthenium dye (powder form) is mixed with ethanol and ethyl cellulose in a 1:100:1 weight ratio on a hotplate with magnetic stirrer at 700 rpm for 20 hours at room temperature.

[00122] As discussed above, for ink jet printing the Z-number should be in the range of about 2 to about 10. Also, inks having a viscosity that is less than about 10 cP are typically preferred for inkjet printing.

[00123] For the ruthenium dye + ethanol + ethyl cellulose based ink solution, the

measured surface tension is 21.48 ± 0.12 dynes/cm. The measured density of the ink solution is 0.78 g/ml. To determine the viscous behavior of the ink solution under a broad range of temperatures from 20° C to 60° C, a rheometer was used. The shear rate was maintained at 1000 (1/s) and the viscosity was decreased from 5.6 cP to 3.4 cP for the temperature range of 20° C to 60° C. After substituting the measured values, Z-numbers ranging from 3.4 to 5.5 were calculated as the temperature increased from 20° C to 60°

C. It is therefore evident that the ink is suitable for inkjet printing at room temperature.

[00124] During further testing, multi-layer samples (3 layer, 2 layer and 1 layer) of the ruthenium dye based ink, with ethanol as solvent and ethyl cellulose as binder, were inkjet printed on to unrastered parchment paper in an array of circular spots with a diameter of 7.5 mm, with 10 pm drop spacing and resolution of 2540 dpi, using a

DIMATIX inkjet printer (DMP 2831). The ruthenium ink solution was loaded into a DIMATIX DMC-11610 cartridge (10 pi) through a 25 mm disposable Whatman syringe filter, with a poly vinylidene difluoride filter (PVDF) filter membrane of 0.45 pm pore size, to filter any large particles that may have agglomerated in order to achieve smooth printing. Each layer of the printed ink was cured on the stage of the inkjet printer at 55 °C. A 40 V actuation voltage, applied at 5 kHz firing frequency, was employed for inkjet printing the ruthenium ink. The printed samples on the unrasted parchment paper are shown in FIG. 24.

[00125] From the printed samples, it was observed that the film formation and coverage of ruthenium dye with ethyl cellulose binder is good when compared to the ruthenium dye with PDMS binder. However, the adhesion between the parchment paper and multiple layers of ruthenium dye (with ethyl cellulose binder) was potentially insufficient.

[00126] As discussed above, the surface energy of calendered parchment paper was

measured with the FTA 200 using Owens-Wendt method and was calculated as 21.99 dynes/cm. The surface tension of the ruthenium dye + ethanol + ethyl cellulose based ink solution is 21.48 ± 0.12 dynes/cm. As also discussed above, the difference between the surface energy of the substrate and surface tension of the ink should be greater than 10 dynes/cm to achieve good adhesion between the substrate and ink. Various surface treatments such as UV (Fusion UV Systems 1300MB), corona (Electro-technic BD-20v corona treater) and sintering (Novacentrix pulseforge 1200) have been employed to improve/modify the surface energy of calendered unrastered parchment paper.

However, it is observed that these treatments have minimal or no impact on the surface of parchment paper.

[00127] However, testing revealed that laser surface treatments significantly alter the surface energy of parchment paper. Specifically, when the surface of calendered parchment paper is subjected to a laser ablation/rastering process using a PLM 6MW laser machine (available from Universal Laser Systems), the surface energy increased to 64 dynes/cm. The surface energy values shows that the laser rastering process has a strong impact on the surface of parchment paper. Also, during testing, the contact angle of ruthenium ink with ethyl cellulose binder with parchment paper was measured as 30.37 ± 1.35 degrees. This implies good wetting properties.

[00128] The ruthenium ink solution with ethyl cellulose binder was inkjet printed on to laser rasted parchment paper using the same settings discussed above. Photographs of the printed samples with multiple layers of ruthenium dye on the laser rastered parchment paper are shown in FIG. 25. It was observed that the adhesion between the ruthenium dye and the laser rastered paper is very good (confirmed by placing/sticking and removing a scotch tape on the printed dye). Also, digital microscope images (not shown) confirmed that the film formation and coverage of the ruthenium dye was good. [00129] However, some burnt fibers (black spots) were evident in the rasted area due to the application of high power intensity during the laser rasting process. In order to reduce or eliminate burning of fibers, a profile of power intensity and laser speed effects on the surface energies of the parchment paper may be utilized to identify a suitable laser rasting process that provides a surface energy value above 32 dynes/cm without burning of paper fibers.

[00130] A suitable binder (ethyl cellulose) was identified and used in the ruthenium ink system in place of PDMS and z-number has been calculated. The ethyl cellulose binder provided acceptable printed ruthenium film formation and coverage. Proper adhesion may be provided by laser rastered calendered parchment paper for inkjet printing.

[00131] As discussed above, an oxygen generation patch may be fabricated using partial cured PDMS to bond parchment paper with laser-rastered spots to PDMS with molded microchannels. This method is capable of creating a flexible and conformable wound dressing patch. However, this process is time consuming, which may interfere with large scale production. Thus, processes that are suitable for large scale (high speed) production have been developed. Testing showed that the improved processes improved the mechanical properties of the oxygen delivery system/platform and reduced fabrication cost.

[00132] With reference to FIG. 26, a method 90 may be utilized to fabricate an oxygen delivery patch. First, at step 91, double sided transparent tape 95 (e.g. 3M 300LSE) is bonded to a layer of PDMS 96 utilizing an oxygen plasma process. At step 92, the tape 95 is then laser-rastered (ablated) to form fluid channels 97 in a predefined honeycomb pattern 99 (see also FIG. 27). The tape 95 (with channels 97) and PDMS 96 form a first subassembly 98. The PDMS layer 96 may also be laser-rastered to a certain depth, provided the thickness of the rastered regions of the PDMS layer are not reduced to a level affecting the robustness of the patch 100. At step 93, a layer of parchment paper

102 is laser-rastered (ablated) at selected surface regions 103, and oxygen catalyst 104 is inkjet printed on to the rastered spots 102 to form a second subassembly 105. Oxygen catalyst 104 may be printed utilizing ruthenium dye/ink (Ru + Ethanol + PDMS) as discussed above. It will be understood that forming the first subassembly 98 (steps 91 and 92) and forming the second subassembly 105 (step 93) may occur at the same time or at different times. The oxygen catalyst 104 forms a honeycomb pattern that aligns with the channels 97 of the tape 95 and PDMS layer 96. At step 94, the first and second subassemblies 98 and 105, respectively, are bonded together with catalyst 104 forming a sidewall that closes off channels 97 to form fluid conduits 106 having a honeycomb pattern 99 (FIG. 27). During step 94, the parchment paper 102 is oxygen plasma bonded to the tape 95.

[00133] Peel strength testing of a patch 100 fabricated according to process 90 (FIG. 26) showed that the interface bond between PDMS layer 96 and parchment paper 102 is about 7N per 2cm width. This is about twice the peel strength obtained using partially cured PDMS as the bonding glue.

[00134] Bonding strength testing was also conducted on a patch fabricated according to process 90 (FIG. 26). This testing was conducted to determine if the patch 100 can withstand the pressure resulting from pumping hydrogen peroxide with a certain flow rate through the microchannels 106 during use of patch 100. In one test, the outlet was open and fluid was pumped at an escalated flow rate. In a second test, the outlet was sealed, and fluid was pumped with a fixed flow rate. Testing showed that a patch 100 can withstand up to 30 PSI with a flow rate up to 7ml/min in the open outlet case. Patch 100 can withstand up to about 3 PSI with a fixed flow rate at 30 ul/min in the closed outlet case.

[00135] The required flow rate for a wound dressing is about 10 mI/min. Thus, the test results show that a patch 100 fabricated according to process 90 (FIG. 26) meets the requirement of a sustained H ₂ 0 ₂ pumping with a flow rate of about 10 mI/min for several hours.

[00136] Robustness testing to determine the effect of bending/curving of patch 100 was also conducted. Patch 100 is designed to conform to a shape/curvature of a patient's skin around a wound. The curvature may vary for different patients and wounds. In general, the patch 100 must not leak during continuous pumping of H ₂ 0 ₂ . During the robustness test, the patch 100 was folded into six different configurations ranging from about 90 degrees to about 180 degrees (fully folded). Thus, a patch 100 was first tested at a bend/fold (curvature) of about 90 degrees, followed by testing at a greater bond/fold of about 108 degrees, followed by a bend/fold of about 126 degrees, etc. until the maximum bend/fold of 180 degrees (6 ^th curvature) was reached. H ₂ 0 ₂ was then continuously pumped through the fluid microchannels/conduits 106 at a constant flow rate of about 0.1 ml/min for 6 hours.

[00137] The fluid pressure inside the microchannels 106 was also measured continuously for all six curvatures. The test demonstrated that the patch 100 provided a constant pressure range from about 0.4 to about 0.5 PSI. This indicates that the patch 100 can sustain up to at least about 6 hours of continuous operation under a maximum 180 degree folding state (zero pressure would be detected if leakage had occurred).

[00138] The process 90 (FIG. 26) is scalable to provide increased production efficiency.

Specifically, with reference to FIG. 28, patch arrays (e.g. 1x2 and 2x2) may be fabricated using the procedure 90 (FIG. 26) utilized for a single patch. An array (e.g. 2x2 array) does not require additional fabrication time compared to fabrication of a single patch 100. Thus, the process 90 and patch 100 provide improved mechanical properties and also increase fabrication efficiency in a scalable production process.

[00139] Additional characterization (testing) of the Ruthenium oxygen sensors and

substrate (parchment paper) was conducted by measuring dissolved oxygen in deoxygenated water. This additional testing was conducted using substantially the same test set up described above in connection with FIG. 20. First, multiple layers of the Ru dye (ink) (Ruthenium based ink with ethyl cellulose binder) were tested to optimize its performance. This formula produced a more uniform printing of the Ru dye (ink) on the parchment paper. Also, laser treated parchment paper was tested to determine if laser treating increased the adhesion of the printed Ru dye.

[00140] A test oxygen sensor was fabricated by printing Ruthenium (Ru) dye on a piece of parchment paper (Diameter = 7.5 mm), and the parchment paper was bonded to double sided tape. Referring again to FIG. 20 the Ru printed side of the test sensor was taped to the wall 70 of the water container 78 facing outside for optical measurement in substantially the same manner discussed above in connection with FIG. 20.

Deoxygenated water was prepared before the experiment, and oxygen concentration was measured with both electrochemical and optical oxygen probes 82, 86 respectively. During the experiment, oxygen gas was injected into the water 80 through external tubing (not shown). The stirring magnet 88 was utilized to ensure uniform 0 ₂ concentration. The experiment was conducted with three different sensors, namely, sensors having single, double, and triple layers of Ru dye. [00141] The objective of this experiment was to test the fluorescence lifetime decay of the single and multi-layered Ru dye samples. Larger fluorescence lifetime decay from multi layered Ru dye was expected. From the previous experiment of printing Ru dye, highly concentrated Ru particles showed difficulties in printing due to the viscosity of the ink and mixing with solvent. Therefore, a method of multi-layer printing was selected to increase its range of quenching decay time of the fluorescence with more oxygen absorbance at the sensor. For this test, oxygen gas was injected to the deoxygenated water then measured with optical (psec) and electrochemical probes (1 mg/L = 1 ppm). Oxygen gas injection was stopped when the measurement was taken. The gas injection continued until the oxygen concentration reached 27 mg/L, which was the limit of electrochemical probe 82. Double and triple layered Ru dye samples were prepared, and the fluorescent lifetime decay performance was measured (with oxygen gas injected). Fluorescence lifetime decays exponentially over saturation of oxygen gas in the liquid. The fluorescence lifetime decay was measured up to about 25 to about 27 mg/L due to limit of the measuring device. Nevertheless, measurement was compared at 9 mg/L, since it is 21 % of oxygen concentration in room temperature. Lifetime decay were - 0.101 and -0.109 psec for double and triple layer samples, respectively. Triple layered samples showed higher changes in quenching decay time of fluorescence. However this different is not significant compared to the results for the double layer samples. Also, the single layer samples showed better quenching fluorescence at around 0 percent dissolved oxygen, resulting in larger changes of fluorescence lifetime at 9 mg/L. Oxygen absorbed from the parchment paper through the multi-layered Ru dye may not be effectively diffused through each layer. Also, the gradients of multi-layer printed Ru dye samples were more significant compared to the gradients of single layer samples. Thus, multiple layers of printed Ru dye do not appear to be effective with respect to increasing the performance of quenching fluorescence decay.

[00142] Additional testing was also conducted to compare the performance of printed oxygen sensors on rastered and unrastered parchment paper to determine if rastering provides increased adhesion. As discussed above, printed Ru dye on unrastered parchment paper tended to adhere poorly, and particles from the printed sensor fell off the unrastered parchment paper. [00143] During testing, parchment paper was rastered with a laser engraving machine. Test samples were fabricated by printing Ru dye in single, double, and triple layers on laser engraved (rastered) parchment paper. Three experiments were repeated for each group of test samples. Both unrastered and rastered parchment paper showed exponential fluorescence lifetime decay. At an atmospheric oxygen level of 21%, the single layer test samples resulted in a faster fluorescent lifetime drop compared to the multiple layer test samples.

[00144] The decay rate up to 9 mg/L was similar in both unrastered and rastered single layered test samples. As a result, the multi-layered Ru dye test samples had a smaller change of lifetime decay compared to the single layer test samples. Consistent with prior observations, the printed Ru dye on unrastered test samples tended to separate from the parchment paper, and edge portions of the printed Ru dye fell apart during most of the experiments. Test samples having a single layer of Ru dye printed on rastered parchment paper had significantly better adhesion. Based on these results, a suitable oxygen sensor can be fabricated by printing a single layer of Ru dye onto a rastered parchment paper surface.

Fluorescence lifetime decay up to 9 mg/L

[00145] The cytotoxicity of the materials used for the fabrication of the smart wound dressing 1 was investigated following standard ISO 10993-05 (Cytotoxicity) and ISO 10993-12 (Sample preparation and reference materials). This subsection describes the methods and presents the results of the cytotoxicity experiments.

[00146] Samples were sterilized by the STERRAD ^® process (low temperature hydrogen peroxide gas plasma) and then extracted for 24 h/37° C in complete growth medium (Eagle's Minimum Essential Medium + 10% horse serum + 100 lll/ml penicillin + 100 pg/ml streptomycin) using an extraction ratio of 6 cm ² /ml. In some experiments, additional samples were sterilized by dipping samples into 100% ethanol or 75% isopropanol for 5 minutes and allowing time to air dry before extraction. At the time of the extraction, L-929 mouse fibroblast cells (NCTC clone 929: CCL 1, American Type Culture Collection, Manassas, VA, USA) in passage 3-10 were lifted from the culture flask using trypsin/EDTA. An aliquot was counted using trypan blue, and then cells were re suspended in complete growth medium at a density of 1 x 10 ⁵ cells/ml. Cells were dispensed into wells of 96-well culture plates (1 x 10 ⁴ cells/well) and cultured at 37° C in a humidified atmosphere of 5% C0 ₂ /95% air. After 24 h, the culture medium was removed and replaced with 100 pi of extractant. Some wells received sodium dodecyl sulfate (SDS; 0 to 400 mM in EMEM; positive controls), low-density polyethylene extract (1.25 cm ² LDPE/ml EMEM; negative control) or complete growth medium alone. Cells were then cultured for an additional 24 h. Subsequently, cells in culture plates were washed once with HBSS and metabolic activity was measured by incubating cells with 100 mI of WST-1 cell proliferation reagent (Roche Diagnostics) for up to 4 h at 37° C.

[00147] To determine cytotoxicity, absorbance of the medium in wells was measured at

450 nm after 2 and/or 4 h using a microplate reader (PHERAstar) and was corrected using absorbance measurements at 630 nm and using blanks. Absorbance levels are proportional to the metabolic activity of cells and therefore inversely related to cytotoxicity. To check for mycoplasma contamination of the cultures, medium was saved and tested using the luminescent MycoAlert Plus mycoplasma detection kit (Lonza). Statistical significance was determined using analysis of variance and Tukey-Kramer post test.

[00148] The results of the cytotoxicity measurements of the various materials used in the smart dressing are shown in FIGS. 31 and 32. The low metabolic activity of cells treated with the extracts of parchment paper, PDMS, double-sided tape, and 3-, 2- or 1-layer

Ruthenium dye printed on parchment paper was significantly less than the activity of cells treated with the LDPE extract (negative control) or cells treated with growth medium and was comparable to cells treated with 300-400 mM SDS (positive controls)

(FIG. 30). We hypothesized that the apparent toxicity of the individual materials could be related to residual contaminants from the Sterrad process. To test this, samples of parchment paper, double-sided tape, PDMS and the three materials combined were sterilized by Sterrad. Duplicate samples were sterilized by dipping in 100% ethanol for 5 minutes and then air-drying before extraction with complete growth medium (37° C/24 h). FIG. SI shows that the cytotoxicity of parchment paper ("paper"), alone or combined with double sided tape ("tape") and PDMS ("3-Layer"), was independent of the sterilization method. However, the effect of the double-sided tape extract on metabolic activity was not significantly different than the negative control ("NC") or Eagle's

Minimum Essential Medium ("EMEM") treated samples. This was in contrast to experiment 1, where the extract of double-sided tape induced significant cytotoxicity. This may have been a result of extractingthe tape with the backing paper left on in experiment 1 and removing it in experiment 2. Cellulosics are known absorbers of H ₂ 0 ₂ and can be chemically modified by H ₂ 0 ₂ .

[00149] To further examine a possible interaction between paper and the Sterrad process, samples of filter paper and parchment paper where treated with the Sterrad process or sterilized by immersion in 70% isopropanol. Additional samples of parchment paper calendered between specific rollers where sterilized by Sterrad to determine if the devices could be the source of the toxic contaminants. FIG. 33 shows that extracts of filter paper (FP), parchment paper (PP), laser-treated parchment paper (LTPP) and calendered parchment paper (CALI-2 and CAL 2-3) sterilized by the Sterrad process were significantly cytotoxic and comparable to the cytotoxicity of 400 mM SDS (FIG. 34). By contrast, extracts of filter paper and parchment paper dipped in isopropanol were not cytotoxic and were comparable to cells maintained in EMEM or extracts of double-sided tape without backing. This confirms the previous findings of an interaction between paper and the Sterrad process, which renders the paper cytotoxic.

[00150] Sterilized samples of parchment paper appeared to be cytotoxic due to possible contaminants resulting from the Sterrad process. The cytotoxicity associated with the Sterrad process was reduced by washing parchment paper samples for 5 minutes in HBSS followed by equilibration for 5 minutes in complete growth medium.

LIST OF NON-LIMITING EMBODIMENTS

[00151] Embodiment A is a fluorescent oxygen sensing ink. The composition of

Embodiment A includes an organic solvent, polymer binder in the organic solvent, and fluorescent dye particles disposed in the organic solvent wherein the fluorescent dye particles bind to the alkyl cellulose particles after printing to form a moisture resistant flexible and comformable film.

[00152] The composition of Embodiment A wherein the polymer binder includes alkyl cellulose particles comprising methyl cellulose, ethyl cellulose, propyl cellulose, isopropyl cellulose, n-butyl cellulose, sec-butyl cellulose, pentyl cellulose, or combinations thereof; silicone based polymers such as polydimethylsiloxane (PDMS), Ecoflex; and polystyrene.

[00153] The composition of Embodiment A or Embodiment A with any of the intervening features wherein the alkyl cellulose polymer have a degree of substitution from about 1.0 to about 3.0.

[00154] The composition of Embodiment A or Embodiment A with any of the intervening features wherein the organic solvent includes at least one substance or a mixture of substances chosen from the group consisting of ethanol, dimethyl sulfoxide (DMSO), dimethyl-formamide, iso propyl alcohol, acetone, and toluene.

[00155] The composition of Embodiment A or Embodiment A with any of the intervening features wherein the fluorescent dye complexes comprise a material selected from the group consisting of ruthenium, osmium tetroxide, rhodium acetate, palladium and chromium.

[00156] The composition of Embodiment A or Embodiment A with any of the intervening features wherein the size of particles in the ink system should be less than 1/100 of the nozzle diameter to avoid agglomeration and clogging of print nozzles during inkjet printing. For example, if the nozzle diameter is 21 pm, then the particle size should be less than 0.2 pm to avoid agglomeration and clogging of print head nozzles.

[00157] The composition of Embodiment A or Embodiment A with any of the intervening features wherein the ink is capable of being printed on hydrophobic to partially hydrophilic substrates, but not completely hydrophilic substrates.

[00158] Embodiment B is a method of fabricating an oxygen sensor. The method

comprising: providing a liquid ink solution including a solvent, fluorescent ink particles dispersed in the solvent, and a polymer binder dissolved in the solution, wherein the polymer binder particles are bound to the fluorescent ink particles, providing a thin flexible substrate having a surface that is hydrophobic to partially hydrophilic, printing the liquid ink solution on the surface of the thin flexible substrate. [00159] The method of Embodiment B wherein the polymer binder comprises an alkyl cellulose, silicone based polymers such as PDMS, Ecoflex; and polystyrene.

[00160] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the alkyl cellulose comprises methyl cellulose, ethyl cellulose, propyl cellulose, isopropyl cellulose, n-butyl cellulose, sec-butyl cellulose, pentyl cellulose, or combinations thereof.

[00161] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the alkyl cellulose has a degree of substitution from about 1.0 to about 3.0.

[00162] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the size of particles in the ink system should be less than 1/100 of the nozzle diameter to avoid agglomeration and clogging of print nozzles during inkjet printing. For example, if the nozzle diameter is 21 pm, then the particle size should be less than 0.2 pm to avoid agglomeration and clogging of print head nozzles.

[00163] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the fluorescent dye complexes comprise a material selected from the group consisting of ruthenium, osmium tetroxide, rhodium acetate, palladium and chromium.

[00164] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the substrate comprises any paper/coated papers such as parchment, TYVEK™ wax coated, chromatography; any polyester films such as polyethylene terephthalate (PET), polyethylene-naphthalate (PEN); any polyimide films such as KAPTON™, UPILEX™; any polyurethane plastics/thermoplastic elastomers such as thermoplastic polyurethane; any silicon based organic polymers such as

polydimethylsiloxane (PDMS) and ECOFLEX™.

[00165] The method of Embodiment B or Embodiment B with any of the intervening

features wherein treating a surface of the substrate, to alter its surface energy, by utilizing a process selected from the group consisting of UV treatment, corona treatment, plasma treatment, sintering, and laser engraving.

[00166] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the organic solvent includes at least one substance chosen from the group consisting of ethanol, DMSO, dimethyl formamide, iso propyl alcohol, acetone, and toluene.

[00167] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the ink can be deposited on the substrate using additive print manufacturing processes such as screen, inkjet, flexography, aerosol jet or gravure.

[00168] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the organic solvent includes at least one substance or a mixture of substances chosen from the group consisting of ethanol, DMSO, dimethyl formamide, iso propyl alcohol, acetone, and toluene.

[00169] The method of Embodiment B or Embodiment B with any of the intervening

features wherein the liquid ink solution includes about 75% to about 99% solvent, from about .1% to about 5% fluorescent ink particles, and from about .1% to about 20% polymer binder particles.