REINFORCEMENT MATERIAL INCLUDING A POROUS LAYER MADE OF A PARTIALLY CROSS-LINKED THERMOPLASTIC POLYMER AND ASSOCIATED METHODS

20230373131 · 2023-11-23

Assignee

Inventors

Cpc classification

International classification

Abstract

The present invention relates to a reinforcement material including at least one fiber reinforcement associated on at least one of its faces with a porous layer, the porous layer(s) representing no more than 10% of the total weight of the reinforcement material, preferably from 0.5 to 10% of the total weight of the reinforcement material, and most preferably from 2 to 6% of the total weight of the reinforcement material, characterized in that the porous layer contains a partially cross-linked thermoplastic polymer. Another object of the invention is a precursor material of such a reinforcement material, as well as their preparation method and the methods for manufacturing a preform or a composite part from such materials.

Claims

1. A method for preparing a reinforcement material suitable for liquid composite molding, comprising the following successive steps: (a) providing a fibrous reinforcement; (b) providing at least one polymeric porous layer at least partially composed of a partially-cross-linked thermoplastic polymer; (c) associating the fiber reinforcement with the at least one polymeric porous layer; wherein said associating step (c) is accompanied or followed by heating the partially cross-linked thermoplastic polymeric porous layer resulting in its softening or melting, followed by cooling.

2. The method of claim 1, further comprising the following successive step: (d) cross-linking at least part of the cross-linkable functions present within the thermoplastic polymer under heat, UV, gamma, or beta radiation.

3. The method of claim 1, wherein said at least one polymeric porous layer is a porous film, a grid, a powder deposit, a fabric or a nonwoven or web.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0097] The following examples, with reference to the attached Figures, illustrate but in no way limit the invention.

[0098] FIG. 1 shows, very schematically, the partial cross-linking of the porous layer of a precursor reinforcement material presented in cross-section and including a fiber reinforcement associated on only one of its faces with a cross-linkable thermoplastic porous layer, leading to a reinforcement material according to the invention, then including a fiber reinforcement associated with a partially cross-linked thermoplastic porous layer.

[0099] FIG. 2 shows, in diagram form, the method used in the examples for the manufacture of reinforcement materials, referred to as “webbed UDs.”

[0100] FIG. 3 shows the level of cross-linking obtained for an HX2632 web and a webbed UD material made with such a web, based on the applied dose of beta irradiation.

[0101] FIGS. 4 to 6 show images obtained under optical microscopy, when various webs and the resin are placed between two glass slides and undergo (or do not undergo) heating.

[0102] FIGS. 7 and 8 show the DMA curves obtained on various resin/web samples according to EN Standard 6032 (1 Hz, 1° C./min, Amplitude 15 μm).

[0103] FIG. 9 shows the melting behavior curve of the HX2632 web following irradiation under 100 kGy of beta radiation by DSC, according to ISO Standard 11357-3.

[0104] FIG. 10 illustrates, in diagram form, a method for placing a reinforcement material.

[0105] FIG. 11 illustrates, in diagram form, the peel test used in the examples.

[0106] FIG. 12 illustrates, in diagram form, a preforming method for making a non-flat preform.

[0107] FIG. 13 shows where the diameter measurement mentioned in Table 5 is taken.

[0108] FIG. 14 shows the hygrothermal cycle used in studying the microcrack behavior reported in the examples.

[0109] FIG. 15 shows the cutting plane used when preparing the samples in the examples in order to study any microcracks that may be present.

[0110] FIG. 16 shows the densities of microcracks obtained in laminates made with the materials according to the invention or the prior art.

[0111] FIG. 17 is an image obtained under optical microscopy of a laminate obtained with a material of the prior art.

[0112] FIG. 18 shows the image obtained under electron microscopy of the Hylink powder and of the RTM6 resin placed between two glass slides, after heating to 180° C.

[0113] FIG. 19 shows the DMA curves obtained on various resin/web or resin/powder samples according to EN Standard 6032 (1 Hz, 1° C./min, Amplitude 15 μm).

DETAILED DESCRIPTION OF THE INVENTION

Materials/Products Used

[0114] The porous layers according to the invention were made with: [0115] 1) Either a web of fibers made of Platamid® HX2632 polymer sold by the Arkema company (copolyamide with terminal unsaturations enabling a three-dimensional network under UV, gamma, or beta treatment), which has a melting point of 117° C.—this web (referred to hereinafter as web HX2632) is obtained by melt blowing and has a mass per unit area of 100 μm prior to lamination onto the fiber reinforcement. The diameter of the fibers that compose it is 15 μm. The opening factor of such a layer, determined according to the method described in patent application WO 2011/086266, is 50+/−10%. [0116] 2) Or by depositing a thermoplastic copolyamide powder that is cross-linkable in temperature due to the presence of epoxy and isocyanate functions, enabling the creation of a T-shaped three-dimensional network, [missing text: presumably “such as the product”] Vestamelt Hylink (X1333) sold by Evonik, which has a melting point of 123° C. The cross-linking can be initiated at a temperature of 150° C.

[0117] The porous layers used for comparison purposes were made with: [0118] 1) Either a 1R8D04 thermoplastic web sold by the Protechnic company (66, rue des Fabriques, 68702—CERNAY Cedex—France), which has a melting point of 160° C.—this web (hereinafter referred to as web 1R8D04) is obtained by melt blowing and has a mass per unit area of 4 g/m.sup.2 and a thickness of 100 μm prior to lamination onto the fiber reinforcement. The diameter of the fibers that compose it is 15 μm. The opening factor of such a layer, determined according to the method described in patent application WO 2011/086266, is 50+/−10%. [0119] 2) or a web of fibers made of a thermoplastic polymer, PA11 LMNO, sold by the Arkema company, which has a melting point of 188° C.—this web (referred to hereinafter as web PA11 LMNO) is obtained by melt blowing and has a mass per unit area of 4 g/m.sup.2 and a thickness of 100 μm prior to lamination onto the fiber reinforcement. The diameter of the fibers that compose it is 15 μm. The opening factor of such a layer, determined according to the method described in patent application WO 2011/086266, is 50+/−10%. [0120] 3) or by depositing a layer of epoxy powder used in the fabric Hexcel Primetex 43098 S 1020 S E01 1F. The average diameter of the powder is 51 μm (D50, median value), and its glass transition temperature ranges from 54 to 65° C.

[0121] The fiber reinforcements used in all cases are carbon fiber unidirectionals sold by the applicant under the reference IMA 12K. The properties of these 12K fibers are summarized in Table 1 below.

[0122] The thermoset resins that can be used for making composite parts are the RTM6 and RTM230ST epoxy resins sold by the company Hexcel Composites, Dagneux France.

TABLE-US-00001 TABLE 1 Hexcel IMA 12K Tensile strength (Mpa) 6.067 Tensile modulus (GPa) 297 Final elongation at break (%) 1.8 Density (g/cm.sup.3) 1.79 Weight/length (g/m) 0.445 Diameter of filaments (μm) 5.1

Measurements Performed

[0123] DSC: Differential Scanning Analysis. The analyses were performed on a Q2000 apparatus by TA Instruments, Guyancourt, France.

[0124] DMA: Dynamic Mechanical Analysis. The analyses were performed on a Q800 apparatus by TA Instruments, Guyancourt, France.

[0125] Hot microscopy analysis: The analyses were performed on an Imager Axio M2m Microscope by Zeiss, Marly-le-Roi, France, equipped with a heating device by Linkam Scientific Instruments, Tadworth, UK.

[0126] Rheology: Viscosity analyses were performed on a HAAKE Mars rheometer by Thermofisher Scientific, Courtaboeuf, France.

Lamination of Webs—Production of a “Webbed UD” Reinforcement Material

[0127] The web is laminated directly on each side of the carbon fiber unidirectional laps by means of a machine (FIG. 2) specifically used for this purpose, immediately after the lap at the desired grammage has been formed. The carbon threads 1 are unrolled from carbon spools 3 attached to a creel 4, passing through a comb 5, are led into the shaft of the machine by means of a guide roller 6 and a comb 7, of a guide bar 8a. The carbon threads are preheated using a heating bar 9 and then are spread by the spreader bar 8b and the heating bar 10 to the desired carbon mass per unit area of the unidirectional lap 17. The web rolls 13a and 13b are unrolled without tension and transported using conveyor belts 15a and 15b attached between the freely rotatable rollers 14a, 14b, 14c, 14d and the heated bars 12a, 12b. The webs 2a and 2b are preheated in the zones 11a and 11b before being in contact with the carbon threads 1 and laminated on either side of two heated bars 12a and 12b, whose air gap is controlled. A calender 16, which may be cooled, then applies pressure onto the unidirectional lap with a web on each side 17. A return roller 18 redirects the product 17 toward the traction system including three rollers for drawing it 19 then rolling it 20, driven by an engine to form a roll composed of the formed material 17.

[0128] The test conditions for manufacturing carbon unidirectional laps combined with a web on each side (referred to as “webbed UD”) are listed in Table 2 below.

TABLE-US-00002 TABLE 2 Method parameters for implementing unidirectional laps associated with a web on each side Measured mass Web per unit area of Line preheating Bar temp. the unidirectional speed Bar temp. Bar temp. temp. (° C.) (° C.) Web (g/m.sup.2) (/min) (° C.) (9) (° C.) (10) (11a &11b) (12a & 12b) Web 210 2.4 60 65 85 100 HX2632 Web 210 2.4 200 200 160 180 1R8D04 Web 210 2.4 200 200 160 180 PA11 LMNO
A. Tests Performed when Webs Used

I. Influence of the Applied Radiation on the Produced Cross-Linked Portion

[0129] Web HX2632 is treated with various beta radiations (equipment by COMET, Flamatt, Switzerland, acceleration voltage of 150 kiloelectronVolt (kV) and irradiation doses from 50 to 100 kiloGrays (kGy).

[0130] The irradiation is performed before the web is associated with the unidirectional. Since thermoplastic copolyamides are soluble in formic acid, the cross-linked portion is determined as follows: the webs or webbed UD obtained following these irradiations are immersed in formic acid for 3 days at ambient temperature (23° C.), then the obtained residue is filtered and dried at 50° C. for 4 hrs. FIG. 3 shows the change in the percentage by weight represented by this residue (therefore corresponding to the cross-linked portion), relative to the total weight of the web obtained after cross-linking. The obtained results are the same regardless of whether the irradiation is performed on the web alone or on a web/unidirectional combination.

[0131] It appears that the cross-linked portion varies from 30 to 60% and varies based on the irradiation conditions used.

II. Influence of Cross-Linking on Solubility in the RTM6 Resin

[0132] Web HX2632, before irradiation, and RTM6 epoxy resin applied onto the web are placed between two glass slides, and the slides are placed under an optical microscope. The assembly then undergoes a temperature rise of 2° C./min up to a temperature of 180° C., corresponding to the final temperature upon infusion or injection of the resin when a composite part is being made. Therefore, this is the critical cycle for the web's temperature resistance since no step for precross-linking the resin is used.

[0133] FIG. 4 shows the image obtained at 23° C. (left) and at 180° C., therefore after cross-linking of the resin (right). It appears that the web dissolves in the resin when it is not partially reticulated.

[0134] FIG. 5 shows the image obtained at 180° C. when the web used is the web HX2632 that has undergo irradiation of 100 kGy of beta rays, using two thermoset resins: RTM6 (left) and RTM230ST (right). It appears that the obtained cross-linking makes the web insoluble in these two resins.

[0135] FIG. 6 shows the change in this insolubility based on the cross-linking level: treatment with 50 kGy of beta rays (35+/−5 by weight of cross-linked portion) (left), treatment with 100 kGy of beta rays (57+/−5 by weight of cross-linked portion) (right). The cross-linking levels are obtained by taking the arithmetic mean of six measurements and the standard deviation is defined as being quadratic mean of the deviations at the mean

[00001] ( 1 n .Math. i ( x i - x _ ) 2 ) .

The increase of the cross-linked portion therefore makes it possible to increase and control the insoluble portion of the web.

[0136] The photos presented show that the presence of the cross-linked portion of the porous layer makes it possible to retain the web's integrity in the liquid resin and appears to reduce the molecular mobility of the thermoplastic portion.

III. Influence of Cross-Linking Studied by DMA

[0137] These results can be correlated with the DMA curves (obtained according to NE Standard 6032) on the RTM6/web samples. These samples were prepared by impregnating the web which was kept vertical inside a metal mold. Once the impregnation with the RTM6 resin was complete, the samples were prepolymerized at 120° C. for 45 minutes, followed by a post-curing for 2 hours at 180° C. As can be seen in FIG. 7 (DMA Analysis, 1 Hz, 1° C./min—Amplitude 15 μm), the dose of irradiation applied to the web greatly influences the thermomechanical properties symbolized by the DMA results: by increasing the cross-linking level, it is possible to maintain the thermomechanical performance of the material until the glass transition of the epoxy-amine network of the RTM6 resin, and the thermoplastic contribution is negligible (Dose of 100 kGy in Beta electron beam). On the other hand, without cross-linking, the thermoplastic transition is clearly visible at about 80° C. and leads to a decrease in thermomechanical properties. This confirms the results obtained in optical microscopy and shows that by irradiating the web, it is possible to control the interactions between the web and the epoxy resin.

[0138] This is the point of the invention, since it appears that once it is treated with 100 kGy of Beta rays, the web HX2632 does not impact the thermomechanical properties of the thermoset resin, despite its low melting point.

[0139] Additionally, FIG. 8, obtained with the web HX2632 having undergone irradiation by an electron beam (Beta rays) of 100 kGy, shows that these observations are confirmed regardless of the resin used: RTM6 or RTM230ST.

IV. Study Using DSC

[0140] The final step for fully understanding the behavior of the web following irradiation was to verify that it was still able to melt after irradiation. For the most cross-linked webs (cross-linked portion representing 57% by weight, obtained after irradiation under 100 kGy electron beam and 35% by weight, obtained after irradiation under 50 kGy electron beam), it was verified that the cross-linked portion present in the web did not prevent the subsequent melting of the web.

[0141] Indeed, the fact that the web retains a hot-melt character is necessary for its later connection to the unidirectional, and during lay-up, in particular for making the preform.

[0142] The web's melting behavior after irradiation was observed using DSC, according to ISO Standard 11357-3. The curves obtained and presented in FIG. 9 show that irradiation by beta electron beam leads to a slight difference in terms of the web's melting point, but this occurs at around 100° C. in any case. This low melting point is very advantageous in terms of time and expense, and makes it possible to lower the temperature to be used during the later lay-up step for making the preform, which in the prior art is generally performed at temperatures above 150° C.

V. Studies Performed on the Reinforcement Materials According to the Invention

1) Cross-Linking on a Webbed UD

[0143] Measurements of the cross-linking level were taken on the webbed UD with an HX2632 web laminated onto each side of the unidirectional, under the conditions listed in Table 2. The voltage applied for electron beam irradiation was 150 kV and the dose applied was 100 kGy. Since carbon fiber does not react with irradiation but may act as a barrier to rays due to its density, the influence of the treatment on only one or on both sides of the material was evaluated. The webbed UD material irradiated by electron beam (beta rays) was immersed in formic acid for 3 days, then filtered and dried in order to evaluate the cross-linking level, as described above.

[0144] The results are summarized in Table 3 below and compared to the level of cross-linking obtained by irradiating only the web. It appears that the results obtained are similar on both materials, confirming that the treatment can be performed at two different steps of the method: on only the web, upstream of its association with the unidirectional, or on the webbed UD. For the latter, however, the treatment must be performed on both sides of the material due to the density of the reinforcement fibers, based on the cross-linking mode used (irradiation under electron beam). In the case of gamma irradiation, a treatment on only one side is sufficient.

Table 3 lists the % of cross-linked portion by weight obtained on the total weight of the porous layer under consideration.

TABLE-US-00003 Webbed UD Web 100 kGy electron beam (single 22 +/− 6 59 +/− side) 100 KGy electron beam (both 67 +/− 5 sides)

2) Lay-Up and Preforming

2.1) Lay-Up

[0145] The depositing of the webbed UD is performed continuously with application of pressure perpendicular to the depositing surface in order to apply it to this surface. Such methods, known as AFP (Automated Fiber Placement) or ATL (Automated Tape Lay-up) are described, e.g., in documents WO2014/076433 A1 or WO 2014/191667 and illustrated in FIG. 10. Various strips of intermediate material are placed one atop the other along parallel placement trajectories, so as to form plies 200.sub.1, 200.sub.2, etc. The device 300 activates the thermoplastic material (web, powder, etc.); that is it uses the hot-melt character of the material and is integrated into the placement unit 400. The latter is moved in order to place the various strips of material that are cut out at the end of the path. When a ply is fully placed, the orientation is modified in order to place the following ply along a placement path that is different from the previous ply. Each strip is placed parallel to the previous strip, with no space between the strips and with adhesive over the entire surface.

[0146] Such a method has been successfully used with the materials according to the invention. In particular, the following conditions were used:

Webbed UD reinforcement materials used: unidirectional laminated on both of its faces with an HX2632 web, then subjected to 50 kGy Beta radiation or to 100 kGy Beta radiation, on each of its faces. Placement power and speed ensured by a FANUC M16iB machine, sold by the FANUC company (Japan).

TABLE-US-00004 TABLE 4 Placement power Placement speed (Watts) (mm/sec.) HX2632 Web  830 430 1R8D04 Web 1000 370
Orientation and number of plies: [(45/0/90/135/0]).sub.2.

[0147] In the case of an activation of an epoxy-powder-type thermoset material used in the prior art, the primary advantage is the material's activation temperature, which is around 100° C. Conversely, this type of material generally leads to soiling of the placement unit 400. This is why thermoplastic materials, e.g., in web form (such as those cited in WO 2010/046609), are generally preferred, but those used in the prior art activate at temperatures above 150° C.

[0148] The webbed UD reinforcement material of the invention with an HX2632 web proposes to address the issue of the material's activation at temperatures below 150° C., even if it has a partially cross-linked thermoplastic material. In particular the webbed UD material with a cross-linked HX2632 web on each of the faces of the carbon lap can be placed at temperatures ranging from 80 to 130° C., more specifically from 100 to 120° C., which lowers by 20 to 40% the power necessary for the placement of the material in comparison with a traditional thermoplastic web and accelerates the placement speed, in particular during the many acceleration phases of the placement means (energy savings). The level of cross-linking of the thermoplastic web does not influence these results, since the residual thermoplastic portion ensures the thermal-bonding character of the material, regardless of the quantity. More specifically, the same placement parameters can be used with a thermoplastic percentage ranging from 30 to 70% by weight after cross-linking of the HX2632 web. These results confirm that the present invention enables especially advantageous placement operations on a large scale.

[0149] The quality of the material's placement was evaluated by means of a non-standard peel test used for comparative purposes. The peeling assembly is indicated in FIG. 11. To do this, two strips of webbed UD material are associated parallel to each other using the previously-specified placement power and speed. The duo thereby formed is peeled in traction at an arbitrary speed of 50 mm/min. over a distance of 200 mm (preforming distance mentioned in FIG. 11) in order to measure the generated peel force. The quality of the placement is thus evaluated by measuring the average of the peel force over 200 mm. The results presented in Table 5 compare the placement quality between a webbed UD reinforcement material with an IR8D04 thermoplastic web that is laminated on both faces and an HX2632 web treated with 50 or 100 kGy of Beta radiation, also laminated on both faces. The three webs have a grammage of 4 g/m.sup.2. It is clear that, despite the irradiation treatment undergone by the HX2632 web, this in no way affects its adhesive character and it is therefore possible to produce placement qualities that are equivalent to those obtained with a pure thermoplastic web.

Table 5 lists the peel force generated over 200 mm between two strips of webbed UD intermediate material according to the web used.

TABLE-US-00005 HX2632 50 KGy HX2632 1R8D04 Beta 100 KGy Beta Average peel force 0.13 +/− 0.03 0.12 +/− 0.02 0.14 +/− 0.04 over 200 mm (N)

2.2) Preforming

[0150] The multiaxial flat preform obtained in Paragraph 2.1 can subsequently undergo preforming by again using the thermobonding character of the thermoplastic material present on the unidirectional reinforcement material. To do this, the flat preform is positioned on the preforming tool at ambient temperature and a silicone vacuum bag covers the assembly, which is then heated in an oven at a temperature that activates the thermobonding character of the thermoplastic or thermoset material. A reduced pressure is then applied to enable the flat preform to be preformed according to the desired three-dimensional shape, then the assembly is cooled under reduced pressure, before the preform is recovered. The entire method is described in FIG. 12. At (i), we see a two-dimensional preform positioned on the tool, with application of the vacuum bag, at (ii), heating leads to activation of thermobonding, at (iii) there is application of the vacuum, the preforming step, followed by cooling with active vacuum at (iv) leading, after (v) and (vi), to the ready preform.

[0151] Whereas reinforcement materials such as those described in patent application WO 2010/046609 must be preformed at temperatures above 150° C., the materials of the invention made with the HX2632 web can be preformed at temperatures below 130° C., preferably below 120° C. This once again demonstrates its true utility in large-scale use for implementing the entire method at temperatures below 130° C.

[0152] To evaluate the quality of the obtained preform, a diameter measurement is performed on it. As was done for evaluating placement quality, three webbed materials are compared: lamination with 1RD04 webs, HX2632 webs that have undergone 50 kGy of Beta radiation, and HX2632 webs that have undergone 100 kGy of Beta radiation. The preforms such as those presented in FIG. 12 were first obtained by stacking 10 unidirectional plies [(45/0/90/135/0)].sub.2 with a mass per unit area of 210 g/m.sup.2. The preforming cycles, as well as the radius measurements taken on the obtained preforms, are presented in Table 6. FIG. 13 describes the site where the diameter is measured on the preform.

TABLE-US-00006 TABLE 6 HX2632 HX2632 Beta 1R8D04 Beta 50 KGy 100 kGy Preforming step 30 min. 30 min. 30 min. 170° C. 120° C. 120° C. Measured diameter (mm) 20.0 21.3 22.0 Theoretical diameter (mm) 20

[0153] The results clearly show the utility of the invention, since it makes it possible to significantly lower preforming temperatures without altering the quality of the obtained preform. In the example presented, a diameter of 22 mm for a theoretical diameter of 20 mm is considered perfectly acceptable.

3) Treatment of Panels

[0154] A preform measuring 340 mm×340 mm composed of the stack sequence adapted to the carbon grammage is placed inside a press injection mold. A frame of known thickness surrounding the preform yields the desired fiber content (FC).

[0155] Four reinforcement materials are compared, two previously-described ones according to the invention and two others used in the prior art (Table 7).

TABLE-US-00007 TABLE 7 Com- Com- parative parative Material 3 Material 4 material material according to according to 1 2 the invention the invention Reinforcement Hexcel IMA 12K fiber Porous layer Epoxy IR8D04 HX2632 web HX2632 web Powder web 50 kGy Beta 100 kGy Beta used in irradiated irradiated the after after Hexcel lamination lamination Primetex (lap (lap 43098 S irradiated) irradiated) 1020 S E011F Mass per unit 210 area of the reinforcement fibers of the webbed UD (g/m.sup.2)

[0156] The epoxy resin sold by Hexcel Composites under the reference HexFlow RTM6 is injected at 80° C. under 2 bars through the preform, which is kept at 140° C. inside the press. The pressure applied by the press is 5.5 bars. When the preform is full and resin is coming out of the mold, the outlet pipe is closed and the polymerization cycle begins (3° C./min up to 180° C. followed by 2 hrs. of post-curing at 180° C. and cooling at 5° C./min).

[0157] Test pieces are then cut out in dimensions suitable for performing compression after impact (CAI) tests, in-plane shear (IPS) tests, open-hole compression (OHC) tests, as well as crack initiation and propagation tests (GIc and GIIc), summarized in Table 8.

TABLE-US-00008 TABLE 8 IPS CAI GIc/GIIc OHC Orientation of plies [45/135]2 s [45/0/135/90]3 s [0]16 [45/0/135/90]3 s in the preform Test machine Instron 5582 Zwick Z300 Instron 2519 Zwick Z300 EN Standard 6031 6038 6033/6034 6036

[0158] The results obtained for all of these tests are listed in Tables 9 to 11. In the case of the GIc and GIIc tests, obtaining a value above 700 J/m.sup.2 is considered to be highly satisfactory and is obtained regardless of the material.

TABLE-US-00009 TABLE 9 IPS Material 3 Material 4 Comparative Comparative according to the according to the IPS (dry, 23° C.) material 1 material 2 invention invention Stress (MPa) 66 102 101 100 Modulus (MPa) 4.4 4.4 4.5 4.6 Material 4 according to IPS (dry, 70° C.) Comparative material 2 the invention Stress (MPa) 71 71 Modulus (MPa) 3.1 3.8 Material 4 according to IPS (dry, 120° C.) Comparative material 2 the invention Stress 54 52 (MPa) Modulus (MPa) 2.8 3.5

TABLE-US-00010 TABLE 10 CAI Standardized Material 3 Material 4 CAI Com- according according at 60% FC parative Comparative to the to the (dry, 23° C.) material 1 material 2 invention invention 30 J (MPa) 126 259 262 255 70 J (MPa) 192 217 211

TABLE-US-00011 TABLE 11 OHC Material 3 Material 4 Com- according according parative Comparative to the to the Compression material 1 material 2 invention invention OHC (MPa) 257 285 295 295

[0159] The mechanical results presented show that in addition to the method issues described previously, to which the materials of the invention respond, these materials also make it possible to obtain composite parts with optimal properties, particularly in terms of impact resistance (CAI), mechanical properties showing hole sensitivity such as the open-hole test (OHC), in-plane shear (IPS), or crack resistance (crack initiation and propagation, GIc, GIIc).

[0160] Specifically, it is possible to obtain a post-impact compressive strength above 250 MPa under an impact of 30 J.

[0161] We therefore note that, on the one hand, while the epoxy powder solves the issue of performing all of the steps of the dry preform embodiment at temperatures ranging from 80 to 130° C., it does not yield composite parts with optimal mechanical properties. Additionally, the traditional polyamide web does yield optimal mechanical properties but does not solve the issue of the low-temperature method.

[0162] Therefore, the present invention combines a method for making the dry preform at temperatures below 130° C. with optimal mechanical properties on composite parts.

4) Microcracks

[0163] Microcrack behavior is studied on a composite material whose dry preform is obtained by stacking 16 unidirectional plies [45°/0°/135°/90°]2s with a mass per unit area of 210 g/m.sup.2. The epoxy resin sold by Hexcel under the reference HexFlow RTM6 is injected at 80° C. under 2 bars through the preform, which is kept at 140° C. inside the press. The pressure applied by the press is 5.5 bars. When the preform is full and resin is coming out of the mold, the outlet pipe is closed and the polymerization cycle begins (3° C./min up to 180° C. followed by post-curing for 2 hrs. at 180° C. and cooling at 5° C./min).

[0164] Three reinforcement materials are compared, two according to the invention: materials 3 and 4 according to the invention, described previously, and one used in the prior art: comparative material 5, which is a reinforcement material similar to previously-described comparative material 2, but in which the porous layer 1R8D04 is replaced by a porous layer PA11 LMNO on each of the faces of the fiber reinforcement.

[0165] The obtained composite materials then undergo one or more hygrothermal cycles, in order to simulate the heat cycles and periods of humidity that an aeronautical part may have to withstand.

[0166] Samples measuring 50 mm×60 mm×4 mm are cut out in order to undergo the hygrothermal cycle defined below. Next, each sample is cut out again, then polished in order to count the number of cracks that appeared during the cycle.

Hyprothermal Cycle Presented in FIG. 14:

[0167] The cycle includes several repetitions of two phases:

[0168] A stationary phase for increasing humidity at 50° C. and under 95% humidity, followed by one hour of heat cycles. These hea cycles consist of a plateau lasting 15 minutes at −55° C., followed by a temperature increase lasting 15 minutes to reach 72° C., followed by a plateau lasting 15 minutes at 72° C. This plateau is followed by a new temperature change phase, returning to −55° C. This negative temperature is selected because it corresponds to what an aircraft may undergo during a subsonic flight. The positive temperature accelerates humidity desorption.

[0169] The load, due to its humidification period, causes a water concentration gradient within the sample. This concentration profile is different on the edges of the sample because the diffusion coefficients are greater in the direction of the fibers. Since the orientation of the fibers is different in each unidirectional ply, the diffusion coefficients are different as well, which generates a very complex water concentration profile on the edges of the sample. This phenomenon has been taken into account in the dimensioning of the sample and in the definition of the zone to be studied.

[0170] The hygrothermal cycles are performed inside a CTS (Climatic Testing System), model CS-70/280-15 from the Climatique et Thermique Service company (ZAC du Pujol, 13390 Auriol, France), including a system for cooling via a two-stage frigorigenic liquid release. A refrigeration unit, Type 30 RA-040-B 0327-PEE, from the Carrier company (CARRIER S.A.S. Route du Thil 01122 Montluel Cedex) circulates a considerable volume of recycled glycolized water at 10° C. inside the cooling system of the first stage of the climatic enclosure in order to ensure its operation. This type of device guarantees a cooling speed of 10° C./min even for temperatures below −50° C., which is close to the cold end of the temperature range for using the enclosure, set between 180° C. et −70° C.

[0171] The humidity inside the usable space of the enclosure is controlled and adjusted using a dew bath. A dryer was added to this device, specifically a ZANDER Type K-MTI dryer by the ZANDER company (45219 ESSEN, Germany) using dry air injection. Once the humidity level of the dryer is set at 0%, the space is considered to be fully dry.

[0172] In order to count the cracks by microscopic observation after the hygrothermal cycle, a sample preparation protocol was prepared. The only direct method for observing internal microstructures, such as microcracks, in a material is to cut out a portion and to polish the cut plane of the portion. This is a widely-used method. It consists of using diamond cloths and suspensions in order to abrade the surface to be polished with increasing fineness, in order to obtain the flatness needed for proper analysis.

[0173] We opted to use samples measuring 5×6 cm.sup.2. Therefore, there are two equivalent observation planes. In each sample that has undergone the hygrothermal cycle, cutouts, in accordance with FIG. 15, are made. The central sample 1 is observed on the observation planes presented in FIG. 15 after polishing.

[0174] The cutting planes P are perpendicular to the plane of the unidirectional plies. The method for polishing the sample to mirror status, which is helpful for observation, was simplified with regard to one metal, in the step using large-grain sandpaper due to its greater ductility. But a finer level of finishing, interspersed with ultrasonic bath cleaning, is necessary during the final polishing phase involving a diamond suspension.

[0175] The cutting of the samples leading to the final sample 100 is performed using a chainsaw with a silicon carbide circular blade. The cutting is carried out via gradual abrasion, with advance calibration of the blade's speed. Next, the following polishing protocol, which yields a mirror polish favorable to microscopic observation, is implemented at the cutting plane.

Polishing Protocol

[0176] The samples are coated in Resin 605 by LamPlan (acrylic resin polymerized with methyl methacrylate) and polished using an independent-pressure automatic polisher (Mecapol P320 by Presi).

[0177] To do this, the cut samples are placed on the bottom of a cylindrical mold. The surface to be polished is oriented toward the bottom of the mold. The mold is then filled with a cold-coating bicomponent resin (LamPlan, 605), which polymerizes spontaneously in about 15 min. The samples are then unmolded and polished according to the described protocol.

The various polishing steps are listed in Table 12 below:

TABLE-US-00012 TABLE 12 Step no. Paper grain Pressure per sample Minimum time 1 P240 250 g/cm.sup.2 40 s 2 P600 250 g/cm.sup.2 optional 3 P1000 250 g/cm.sup.2 50 s 3 min in an ultrasonic bath Step no. Diamond suspension Pressure per sample time 4 3 μm 100 g/cm.sup.2 7 min 3 min in an ultrasonic bath 5 1 μm 100 g/cm.sup.2 5 min

[0178] In steps 1, 2, and 3, we use, for the automatic polisher, a rotation speed of 150 rpm in counter-rotation for the plate and the maximum speed for the head (100 rpm). Impurities are rinsed away.

[0179] In steps 4 and 5, we use a rotation speed of 300 rpm in counter-rotation for the plate and the maximum speed for the head (100 rpm per minute). Impurities are removed using a lubricant applied dropwise.

[0180] Counting the cracks is then performed using analysis micrographic images obtained by a 5-megapixel digital camera (model U-TVO.5XC-2-4F04335 by OLYMPUS) mounted onto a microscope (model GX 51 F-T2 SN 4 G 0 9299 by OLYMPUS), with an ×5 lens (magnification ×50). The image analysis software used is “Analysis Pro Five,” sold by Olympus France SAS, Parc d'affaire Silic, 74 rue d'Arcueil BP 90165, 94533 Rungis cedex, France. For an observation of Ni fissures in the unidirectional ply i over a sample of length L with a lay-up that allows the cracks in p unidirectional plies to be clearly seen, a criterion d is defined according to the equation:

[00002] d = .Math. i Ni L × p

[0181] The p factor corresponds to the total number of unidirectional plies of the laminate minus the number of unidirectional plies whose carbon fibers are parallel to the observation plane, taking into account the fact that the cracks remain invisible within these unidirectional plies.

[0182] The d factor is a linear crack density, expressed in cm.sup.−1, which taking into account the choice of L, can be considered an intrinsic feature of the material under the relevant load.

[0183] The graphic in FIG. 16 shows the values of the d factors (referred to as crack density) obtained on various samples (given that the measurement method is destructive), after a determined number of hygrothermal cycles. It is clearly shown that the laminates in accordance with the invention have a much lower crack density, even zero density.

[0184] An image of a microcrack obtained after 400 hygrothermal cycles with the laminate of Comparative Example 23 is presented in FIG. 17.

[0185] This type of result shows another contribution of the invention: the addition of a partially cross-linked thermoplastic polymer porous layer minimizes and even eliminates the occurrence of microcracks.

B. Tests Performed Using a Powder Deposit

[0186] First, the Hylink binder was cross-linked for 30 minutes at 180° C. in an oven. Measurement of the weight of the cross-linked portion, as well as hot optical microscopy, were performed as in Parts A-I and II, in order to verify the behavior of the polymer. The content levels of cross-linked portion were measured by immersion in formic acid for 3 days at ambient temperature, then filtered and dried at 50° C. for 4 hrs. The results obtained were as follows: [0187] initial cross-linked portion: 7%+/−3%, relative to the total weight of polymer [0188] cross-linked portion after 30 minutes at 180° C.: 60%+/−9%, relative to the total weight of polymer.

[0189] FIG. 18 shows the image obtained with the optical microscope at 180° C., therefore following cross-linking of the present RTM6 resin. The photo presented shows the polymer's ability to cross-link partially under the increase of temperature, which allows it to remain insoluble in the RTM6 resin, in the same way as for the HX2632 web that underwent irradiation.

[0190] Here again, these results can be correlated with the DMA curves on the RTM6/Hylink binder samples (following EN Standard 6032). The samples were pre-polymerized at 120° C. for 45 minutes, followed by 2 hours of post-curing at 180° C. As the results presented in FIG. 19 show, the cross-linking treatment at the temperature applied to the Hylink binder has considerable influence on the DMA results: by increasing the level of cross-linking, it is possible to maintain the thermomechanical performance of the material in the same way as for the HX2632 web. Consequently, the thermoplastic contribution is negligible, which confirms the microscopy observations.