BACKGROUND OF THE INVENTION

1. Field of the Invention This invention relates to nonwoven fabrics containing wood pulp and synthetic fibers made by hydroentangling a fibrous web on a flattened woven support member. 2. Description of the Related Art The hydroentangling (or hydraulic needling) process for producing spunlaced nonwoven materials have been used for many years. In the hydroentangling process, a loose aggregation of fibers is positioned on a screen or some type of apertured support and subjected to a series of high-pressure water jets to bind or entangle the fibers to form a fabric. Also, relatively low pressure water jets (consolidator or condensing jets) can be used in advance of the first series of main hydroentangling jets to provide some initial structural integrity to the loose fibers. Screens or apertured support members are also used in this consolidation process. Conventional hydraulic needling processes are described in United States Patent No. 3,485,706, to Evans. The support member can be porous, such as a perforated plate, or a metal or plastic belt or screen that is woven from round or other shaped strands, monofilaments, or yarns. Woven screens are generally formed from metal wires and plastic filaments that have smooth surfaces. Alternately, as described in Smith et al. International Publication Number WO 2003/031711 , the web-supporting member can have a surface that includes rough-surfaced yarns that inhibit movement of the nonwoven fiber web relative to the web-supporting surface. United States Patent No. 4,868,958 to Suzuki et al describes hydroentangling a fibrous web on a smooth-surfaced plate that includes a plurality of drainage holes. United States Patent No. 4,805,275 to Suzuki et al describes water jet treatment of a fibrous web that includes conducting a prelϊffiinaTyelhtangling treatment on a smooth-surfaced water-impermeable belt followed by entangling treatment on a plurality of smooth-surfaced water impermeable rolls. The water impermeable rolls are multistagedly and parallely arranged in order to provide effective draining treatment. International Publication WO 2004/061183 describes a support belt made from flattened filaments used in hydroentangling processes. International Publication WO 2004/061183 describes a support belt that is subjected to a calendering process to deform flatten at least a portion of the constituent filaments. Both are assigned to Albany International Corporation (Albany, NY). The abrasion resistance and tensile properties of hydraulically entangled nonwoven fabrics can be improved by increasing hydroentangling pressures or using more powerful jets. However, these approaches can result in a reduction in barrier properties, which is undesirable for certain applications, such as medical fabrics. When woven belts are used as the support member during hydroentanglement, the fibers can become entangled or "needled" into the woven support over time. This may require physical removal of the unwanted fibers or, alternately, running scrap scrim through the machine in an attempt to remove the errant fibers, both of which involve delay or stopping the production line. When the fibers cannot be removed from the woven support member, the support member usually must be replaced, typically at high cost. There remains a need for hydraulically entangled nonwoven fabrics having improved physical properties such as improved abrasion resistance and strength. In addition, there is a need for nonwoven medical fabrics having improved barrier properties.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 schematically depicts intersecting strands in a conventional support member. Figure 2 depicts a ground fiber from a flattened support member. Fϊg"ure"3 is'a' phb'tomϊcrograph of the top surface of a fabric made using a conventional support member. Figure 4 is a photomicrograph of the top surface of a fabric made using a flattened support member. Figure 5 is a photomicrograph of the bottom surface of a fabric made using a conventional support member. Figure 6 is a photomicrograph of the bottom surface of a fabric made using a flattened support member.

DETAILED DESCRIPTION OF THE INVENTION Woven support substrates, such as screens, used in the hydroentangling process have "knuckles", which are raised areas that are formed where the strands extending in one direction weave over those extending in another direction, as depicted in Fig.1. Strand 2 is depicted as overlying strand 1 forming knuckle 3, representative of a conventional warp and weft woven pattern. The terms "flattened belt" and "flattened screen" as used herein refer to a woven support substrate, such as a belt or screen that has been woven from filaments or strands, wherein the knuckle areas have been flattened after weaving. It has been found that significant advantages are achieved when nonwoven fabrics containing about 5 -50 % wood pulp and about 95 - 50% synthetic fibers are made by using flattened belts as described herein in a hydroentangling process. The advantages include improvement in properties, such as, but not limited to tensile strength, abrasion resistance and hydrostatic head (hydrohead). Further, it has been found that "needling" is significantly reduced with the use of flattened belts. Also, a wood pulp/poly(ethylene terephthalate) (WP/PET) hydroentangled fabric made with flattened belts have thinner rows of wood pulp on the top surface than fabrics that were hydroentangled on round belts. Likewise, the bottom surface of such fabrics had significantly more wood pulp distributed substantially evenly across the bottom surface compared to the relatively thin lines of wood pulp aligned primarily in the machine direction for fabrics made on round belts. This is demonstrated in Figs. 3-6, wherein the top surface is substantially the wood pulp side and the'Mttom sTϊrfaό'S 'is Substantially the PET or polyester side. It is noted that the fabrics represented in the photomicrographs in Figs 3-6 were made using plastic belts in both the round form and in the flat form. The use of a flattened belt for hydroentangling results in a fabric having improved barrier properties compared to fabrics obtained using a round belt, making the fabrics of the present invention especially suitable for use as medical fabrics and other end uses where liquid barrier properties are desirable. The observation of improved hydrostatic head using flattened belts was unexpected in view of the thinner rows of wood pulp observed on the WP side as observed in photomicrographs of similar samples. Conventional wisdom is that in order to achieve good barrier properties in wood pulp-based medical fabrics, an even layer of unbroken WP (except for jet tracks) is desired to act as a substrate for the fluorochemical repellent. Surprisingly, despite the thinner rows of WP on the WP side of the hydroentangled fabrics of the invention, the fabrics of the invention exhibited improved hydrostatic head compared to fabrics made using round belts. Without wishing to be bound by theory, it is possible that the secondary layer of wood pulp formed on the fabric backside offers an additional receptor layer for the fluorochemical. It is also possible that the use of flattened belts/screens reduces the amount of wood pulp that is washed out of the web during the hydroentangling step compared to round belts/screens. When the strands are metal, the knuckle areas are flattened by removing part of the top surface of the strands at the knuckle areas, such as by grinding. When the top surfaces of the metal strands are partially ground away, flattened surfaces (lands) are formed. The amount of the metal strands ground away from the knuckles on the top surface of woven metal screens/belts was calculated from measurements made using optical photomicrographs taken of the flattened surface of the belt at a magnification of 5Ox. The knuckle areas that were ground away appeared as ovals on the flattened surface of the screen with the major dimension of the oval aligned with the length of the metal strands. When the strands are plastic, the knuckles and surrounding areas are flattened by applying pressure, (with or without heat) to the knuckle 'afe&s-tatiδtrigj thtfisilrafttteiD deform and flatten into an elliptical cross- sectional shape in the knuckle areas, in which case the flattened knuckle areas are also referred to as lands. For example, a woven plastic substrate can be calendered to form a flattened substrate as described in WO 2004/061183. To determine the extent of flattening of the plastic strands on the top surface of flattened plastic screens/belts, SEM photomicrographs were taken at a magnification of 20Ox for cross-sections through the center portion of a knuckle area of the screen/belt. The aspect ratio of a plastic filament at the flattened knuckle areas was calculated as provided in the Test Methods section below. A precursor (prior to flattening) belt/screen that is woven from round strands is referred to herein as a "round belt" or "round screen". The terms "flattened belt" and "flattened screen" are not intended to include substrates that have been woven from flat (e.g. rectangular or square) strands that are not further flattened after weaving. It is understood that the belt or screen can be in the form of a planar substrate such as a belt, or can form the surface of a drum, or can be of some other arrangement known in the art. The term "synthetic fiber" refers to various synthetic materials that can be used with wood pulp in nonwoven fabrics, such as polyolefin, polyamide and polyester. "Polyester" is more typically used and is intended to embrace polymers wherein at least 85% of the recurring units are condensation products of dicarboxylic acids and dihydroxy alcohols with linkages created by formation of ester units. Examples of polyesters include poly(ethylene terephthalate) (PET), poly(1 ,3-propylene terephthalate), poly(1 ,4-butylene terephthalate), and blends and copolymers thereof. The term "copolymer" as used herein includes random, block, alternating, and graft copolymers prepared by polymerizing two or more comonomers and thus includes dipolymers, terpolymers, etc. The term "wood pulp" includes cellulosic material in the form of paper webs as well as in particulate form, such as fluff and the like. The term "machine direction" (MD) is used herein to refer to the direction in which a nonwoven web is produced (e.g. the direction of travel which the fibers are laid down during formation of the nonwoven web). The term "cross direction" (CD) refers to the direction generally perpendicular to the machine direction in the plane of the web.

TEST METHODS In the description above and in the examples that follow, the following test methods were employed to determine various reported characteristics and properties, unless noted otherwise. INDA refers to the Association of the Nonwovens Fabric Industry. ASTM refers to the American Society of Testing Materials. Basis Weight (BW) of a sample was measured according to INDA Standard Test IST 130.1 (01). Hydrostatic head (HH) is a measure of the resistance of a sheet to penetration by liquid water under a static pressure and was measured according to INDA IST 80.6 (01). In the examples below, hydrostatic head was measured for repellent-finished fabrics by placing the reservoir of water on the WP side of the fabric and observing the PET side of the fabric, and is reported in units of mbar. Measurements were made on the number of samples specified in the Examples and averaged to obtain the mean hydrostatic head. Strength of nonwoven fabrics was measured as sheet grab tensile strength, measured according to INDA IST 110.1 (ASTM D5034-95). Measurements were made on 4 - 6 samples and an average strength was calculated for both the MD and CD. The average MD and CD strengths were averaged to obtain the average strength for a fabric. Martindale Abrasion is a measure of the abrasion resistance of nonwoven fabrics and was measured on the polyester side of the fabrics according to INDA IST 20.5 (ASTM D4966-98). Martindale Abrasion was measured on 6 samples for each example and the individual results for each Example were averaged. Martindale Abrasion is measured on a scale of 1 - 6, with lower Martindale numbers corresponding to better abrasion resistance (best = 1). Conditions used were150 cycles on dry fabric. THe 'af peCtϊattό' bf a" plastic filament was calculated as the ratio of the minimum and maximum cross-sectional dimensions, measured at the knuckle area through the center point of a strand. Measurements were made on two strands located on the top surface of each plastic screen and the two measurements were averaged to obtain the aspect ratios reported in the examples below. The percent diameter of metal strands ground away was calculated using the formulas found in Lang's Handbook of Chemistry, Fourth Edition (1941), Appendix, page 12. The value 100h/D is reported in the examples as the percent diameter ground away where h = rise and D = diameter of the strand prior to grinding away part of the surface.) The rise is the vertical distance "h" shown in Figure 2, measured at the center of and perpendicular to land "I" and extending to the original outer surface of the strand that has been ground away. Measurements of the minor dimension at the center of each land (corresponding to the length of land "I" in Figure 2 were made on three lands per sample and 100h/D calculated for each and averaged to determine the % ground away for belt CF used in Example 27 below. Air permeability of woven screens/belts was measured according to INDA 1ST 70.1 (01) (ASTM D737 - 96) and is reported in units of ft3/min/ft2.

EXAMPLES In the following examples, hydroentangled fabrics were prepared on a laboratory-scale table washer, on which consolidation and entangling steps are performed as batch processes. The table washer included a continuous belt with speed control that runs underneath one or more entangling jets. A secondary belt (either a flattened belt or a control belt (precursor to flattened belt)) was placed on the top surface of the continuous table washer belt and the polymeric fiber (e.g. PET) and wood pulp (WP) layers were placed on top of the secondary belt. In each example, a consolidation step was performed wherein the polyester web was impacted with relatively low pressure water jets prior to adding the paper layer, followed by running the layered WP/PET web underneath the primary hϊgti'preS'SCire'WaMr jet for multiple passes using the specified jet profile to generate the desired degree of entanglement. The layered WP/PET webs were entangled on the paper side. Some examples were hydroentangled on the same belt that was used for consolidation, while other examples used a hydroentanglement belt that was different from the consolidation belt. The following codes are used to identify the screens used as support members in the Examples that follow: BeIt A - Triotex 114TLM, woven plastic belt purchased from Albany International Corporation (Albany, NY), aspect ratio of plastic strands = 1.0, air permeability = 426 ft3/min/ft2. BeIt AIF - Formed by flattening Belt A, obtained from Albany International Corporation (Albany, NY), aspect ratio = 0.75, air permeability = 318 ft3/min/ft2. Belt A2F - Formed by flattening Belt A, obtained from Albany International Corporation (Albany, NY), aspect ratio = 0.61 , air permeability = 285 ft3/min/ft2. Belt B - Formtec 55LD, woven plastic belt with aspect ratio of plastic strands = 1.0, purchased from Albany International Corporation (Albany, NY), air permeability = 1140 ft3/min/ft2. Belt BF - Formed by flattening Belt B, obtained from Albany International Corporation (Albany, NY), aspect ratio = 0.67, air permeability = 705 ft3/min/ft2. BeIt C - 75 mesh metal screen woven from round metal strands obtained from Albany International Corporation (Albany, NY). Belt CF - Formed by grinding the high knuckle areas of Belt C to form a flattened metal screen. The percent diameter ground away was calculated to be about 20%.

Example 1 and Control Example A In this example, hydraulically entangled fabrics were made using a line speed of 75 yd/min on a table washer using flattened plastic belt A1F for Example 1 and precursor round plastic belt A for Control Example A. The nonwoven fabric was formed from layered wood pulp ("WP") and poiy^etnyiene terepntπaiatfe) fiber ("PET") webs. The wood pulp (WP) paper layer was Barrier Green (H. C. Kraft) paper obtained from Domtar, Inc. (Montreal, Quebec). The PET web was a 100% PET Rando web, product code T612 SDW obtained from DAK Americas (Charlotte, NC), having a basis weight of about 1 oz/yd2 and denier per filament of about 1.35. The PET fiber web was pre-consolidated on belt A1 F for Example 1 and on belt A for Control Example A using a first 254 psi consolidation jet followed by a second 368 psi consolidation jet. Following consolidation, the WP layer was placed on top of the pre-consolidated PET web and the combined layers hydroentangled on the same belt. The jet strip had 40 holes/inch, each hole having a diameter of 5 mils. The combined WP/PET layers were then hydroentangled with the WP layer facing the jets in 9 passes, adjusting the jet pressure to simulate a series of different jets as would be experienced in a commercial scale line. The jet profile was 114, 170, 283, 453, 906, 941 , 941 , 1019, and 1019 psi. The hydroentangled fabrics were air-dried and had an average basis weight of about 2.0 oz/yd2. Martindale abrasion and strength properties for the hydroentangled samples are reported in Table 1. An increase in average strength of 14.7% was observed for the sample prepared on the flattened belt compared to the sample prepared on the precursor belt. In addition, the abrasion resistance was about two times higher when the flattened belt was used compared to the sample prepared on the precursor belt.

Table 1. Strength and Abrasion Properties for Example 1 and Comparative Example A Examples 2 - 6 and Comparative Example B These examples demonstrate combinations of round and flattened belts used in the pre-consolidation and hydraulic entanglement steps to form WP/PET hydroentangled fabrics and the improvement achieved using flattened belts. The PET fiber layer and WP layer used in these examples were the same as those described above for Example 1. The line speed was 75 yd/min. The PET fiber web was pre-consolidated on either round Belt B (Examples 3-4, and Comparative Example B) or flattened Belt BF (Examples 2, 5-6) using the consolidation jet pressures specified above for Example 1. After consolidation, the web was removed from the consolidation belt and placed on the entanglement belt. After layering with the WP layer, the combined layers were hydroentangled on round Belt A (Example 2 and Comparative Example B), flattened Belt A1 F (Examples 3 and 5), or flattened Belt A2F (Examples 4 and 6) using the jet profile described above for Example 1. Martindale abrasion and strength properties for the hydroentangled samples are reported in Table 2. The largest increase in average strength (about 10-12%) was observed for Examples 5 and 6, which were both consolidated and hydroentangled on flattened belts. The examples which used a combination of a round belt and a flattened belt (Examples 2 - 4) showed some improvement in strength compared to all round belts (Comparative Example B), but less than the improvement achieved using all flattened belts. All of the Examples 2-6 exhibited better abrasion resistance than Comparative Example B. Table 2. Strength and Abrasion Properties for Examples 2 - 6 and Comparative Example B

1average basis weight was calculated and reported for samples used to measure MD strength and samples used to measure CD strength These examples demonstrate use of combinations of round and flattened belts in the consolidation and hydraulic entanglement of WP/PET nonwoven fabrics using a carded PET web instead of a Rando web, and the improvement in strength achieved using flattened belts. In these examples, the WP paper was the same paper that was used in Example 1. The PET web was obtained from Hamby Textile Research (Garner, NC) and was a carded web of 1.5 dpf T54 SD W PET fiber from DAK Americas (Charlotte, NC). The PET web was consolidated on the consolidation belt specified below in Table 3 with a 46.3 psi jet angled at an angle θ from the perpendicular of 30 degrees, as described in Oathout et al. U.S. Patent Application Publication No. US2002/0116801 , followed by three additional consolidation passes (θ = 0 degrees) at 93, 510, and 256 psi. After layering the WP layer on the consolidated PET layer, the combined layers were hydroentangled on the same belt in three additional passes with jets at 256, 194, and 324 psi, respectively, followed by transferring the combined webs to the hydroentanglement belt. Round Belt A was used as the hydroentanglement belt in Example 7 and Comparative Example C; Belt A1 F was used in Examples 8 and 10; and Belt A2F was used in Examples 9 and 11. The following jet profile was used for hydroentangling on the hydroentanglement belt: 1144, 1181 , 1279, 1149, 1149, 926, and 509 psi. The line speed during consolidation and hydroentanglement was 50 yd/min. The results are shown in Table 3. All samples of the invention exhibited higher strength than the control example (strength increased between about 13% - 37%). Table 3. Strength Properties for Examples 7 - 11 and Comparative Example C

1average basis weight was calculated and reported for samples used to measure MD strength and samples used to measure CD strength Examples 'Ϊ2' - 13J and' Comparative Example D

These examples demonstrate combinations of round and flattened

belts, using the same WP layer that was used in Examples 7-11 , and a

double layer of the carded PET web described in Examples 7 - 11. Two

layers of the PET web (instead of one layer as used in Examples 7 - 11)

were superimposed and consolidated and pre-entangled as described in

Examples 7 - 11 prior to combining with the WP paper layer. Belt B was

used as the consolidation belt for Comparative Example D and Belt BF

was used for Examples 12 - 13. The combined WP and PET layers were

then hydroentangled using the same jet profile described in Examples 7 -

11. The hydroentanglement belt used in these examples was Belt A for

Comparative Example D, Belt A1 F for Example 12, and Belt A2F for

Example 13. All of the examples of the invention exhibited increased

strength and improved abrasion resistance compared to Comparative

Example D.

Table 4. Strength and Abrasion Properties for Examples 12 - 13 and Comparative Example D

1 average basis weight was calculated and reported for samples used to

measure MD strength and samples used to measure CD strength Examples 14 - 16 and Comparative Example E These examples demonstrate the effect of combinations of round and flattened belts on barrier properties of repellent-finished hydroentangled WP/PET fabrics. The WP and PET layers used were those described above in Examples 2 - 6. The line speed and jet profile was also the same as used in Examples 2 - 6. A repellent finish was applied to the hydroentangled fabrics using a padder and squeeze roll setup. The padder was a Type KLFH/K padder manufactured by Ernst Benz Ag (Sweden) and operated at 7 yd/min and a nip pressure setting of "7" as determined by the equipment set point. The repellent finish used was 2.5% Freepel® 1225, manufactured by Freedom Chemical (Charlotte, NC) and 1.5% Zonyl 8315 manufactured by E.I. du Pont de Nemours and Company (Wilmington, DE)1 dissolved in de-ionized water. After applying the finish, the samples were dried for two minutes at a temperature of 1650C in a Mathis Lab dryer, suspended in air with pins. Hydrostatic head measurements were made on four samples for each example and averaged to obtain a mean hydrostatic head. The results are shown in Table 5. The results show that the second belt, on which the majority of hydroentangling occurs, has the greatest impact on the hydrostatic head.

Table 5. Hydrostatic Head measurements for Examples 14 - 16 and Comparative Example E

1 mbar =1.02 cm of water at 40C

Examples 17 - 21 and Comparative Examples F - J These examples demonstrate the impact of the degree of entanglement on hydrostatic head for fabrics prepared on combinations of rόύ'ntf'atid'Ttattenfed1 bifls.-The degree of entanglement was varied by varying the line speed of the table washer (slower speeds = higher degree of entanglement). The WP used was the same as that described above in Examples 2 - 6. The PET layer was a carded web of 1.5 denier PET fibers, Type 237 SDW from DAK Americas (Charlotte, NC). The PET web was consolidated on the consolidation belt, removed from the consolidation belt and combined with the WP layer on the hydroentanglement belt, followed by hydroentangling. The consolidation jet pressures and hydroentangling jet profile were the same as those described in Example 1. The hydroentangled fabrics were treated with a repellent finish as described above for examples 14 - 16. For each example, hydrostatic head measurements were made on six samples, each sample measuring 6 inches x 6 inches, and averaged to obtain a mean hydrostatic head. Sample weights were measured for each 6 in x 6 in sample and averaged to obtain a mean sample weight that was used to calculate a normalized hydrostatic head. The results are shown in Table 6 and show about a 3.5 mbar higher hydrostatic head for samples prepared using flattened belts compared to comparative examples made using round belts over the entire range of line speed. Table 6. Hydrostatic Head measurements for Examples 17-21 and Comparative Examples F-J

Hydrostatic head normalized to a sample weight of 1.653 g In the'tabϊe above" a"hd elsewhere in the specification, I x E is the energy impact product delivered by water jets impinging on a fabric web and is calculated as described in US Patent No. 5,459.912 to Oathout.

Examples 22 - 26 and Comparative Examples K - O The samples in these examples were prepared as described for Examples 17-21 , except that I x E was varied by changing the jet profile instead of the line speed, which was held constant in these examples at 50 yd/min. The jet profiles were calculated to give the same I x E factors as Examples 17-21 and Comparative Examples F - J. The specific jet pressures and belts used, as well as hydrostatic head values are reported in Table 7. Similar to Examples 17-21 and Comparative Examples F - J, the fabrics of the present invention had about a 3 mbar improvement in hydrostatic head.

Hydrostatic head normalized to a sample weight of 1.655 g Exanf1blfer27 and CoMpsirative Example P This example demonstrates use of a flattened metal screen to make nonwoven fabrics of the present invention. The paper and PET staple fiber layers were the same as those used in Example 1 and were consolidated and hydroentangled as described in Example 1 except that flattened metal screen CF was used for Example 27 and round metal screen C was used for Comparative Example P (for both the consolidation and entanglement steps). The fabrics were treated with a repellent finish as described in Examples 14 - 16. Examination of photomicrographs of the top (WP) and bottom (PET) layers showed that there was more WP on the PET side for Example 28 compared to Comparative Example P, however the difference was not as significant as observed for similar samples that were entangled on plastic belts. The results are summarized in Table 8. Hydrostatic head values were measured for four samples for each example and averaged to obtain the HH value shown in the table. All properties were somewhat improved for the fabric of the invention compared to the comparative example.

Table 8. Strength, Abrasion, and HH Properties for Example 27 and Comparative Example P

1average basis weight was calculated and reported for samples used to measure MD strength and samples used to measure CD strength