WO2012068259A1

WO2012068259A1 - Carbon fiber

Info

Publication number: WO2012068259A1
Application number: PCT/US2011/061008
Authority: WO
Inventors: Makoto Kibayashi; Satoshi Seike; Lawrence A. Pranger; Anand Valliyur Rau
Original assignee: Toray Carbon Fibers America Inc.
Priority date: 2010-11-16
Filing date: 2011-11-16
Publication date: 2012-05-24
Also published as: EP2640879A1; KR20130001301A; KR101408880B1; CN103069063A; JP2014500912A; EP2640879A4; US20120123053A1

Abstract

A carbon fiber is coated with a sizing at an amount X between 0.1 and 0.3 weight%. The sizing is formed of a heat resistant polymer or a precursor of the heat resistant polymer. The amount X of the sizing is expressed with a following formula: where W₀ is a weight of the carbon fiber with the sizing, and W₁ is a weight of the carbon fiber without the sizing.

Description

CARBON FIBER

CROSS-REFERENCE TO RELATED APPLICATIONS

This is a PCT application claiming priority of a prior

US patent application No. 12/947,160, filed on 11/16/2010, pending.

BACKGROUND OF THE INVENTION AND RELATED ART STATEMENT

The present invention relates to a carbon fiber with a sizing capable of achieving superior resistance against thermal decomposition.

Carbon fiber reinforced plastics (CFRP) have superior mechanical properties such as high specific strength and high specific modulus; therefore, they are used for a wide variety of applications, e.g., aerospace, sports equipment, industrial goods, and the like. In particular, CFRP with a matrix consisting of a thermoplastic resin has a great advantage such as quick molding characteristics and superior impact strength. In recent years, research and development efforts in this area have been flourishing.

In general, polymer type composite materials tend to show reduced strength and modulus under high temperature conditions. Thereby, heat resistant matrix resins are necessary in order to maintain desired mechanical properties under high temperature conditions. Such heat resistant matrix resins include a thermosetting polyimide resin, a urea-formaldehyde resin, a thermoplastic polyimide resin, a polyamideimide resin, a polyetherimide resin, a polysulfone resin, a polyethersulfone resin, a polyetheretherketone resin, a polyetherketoneketone resin, and a

polyphenylenesulfide resin.

CFRP with heat resistant matrix resins are molded under high temperature conditions, so a sizing must withstand thermal decomposition. If the sizing experiences thermal decomposition, voids and some other problems occur inside a composite, resulting in undesired composite mechanical properties. Accordingly, a heat resistant sizing is an essential part of CFRP for better handleability, superior interfacial adhesive capability, controlling fuzz

development, etc.

A conventional heat resistant sizing has been developed and tried in the past. For instance, US Patent No.

4,394,467 and US Patent No. 5,401,779 have disclosed a polyamic acid oligomer as an intermediate .agent generated from a reaction of an aromatic diamine, an aromatic

dianhydride, and an aromatic tetracarboxylic acid diester. When the intermediate agent is applied to a carbon fiber at an amount of 0.3 to 5 weight% (or more desirably 0.5 to 1.3 weight%), it is possible to produce a polyimide coating. However, the sizing amount of 0.3 to 5 weight% does not seem efficient in terms of drape ability and spreadability for resin impregnation. The composite mechanical properties tend to be lower than a desirable level.

In US Patent No. 5,155,206 and US Patent No. 5,239,046, a composition of the polyamideimide as the sizing has been disclosed. However, the sizing amount that is essential to obtain the optimal mechanical properties has not been disclosed.

In view of the problems described above, an object of the present invention is to provide a carbon fiber with high mechanical property in addition to superior resistance to thermal decomposition and capability for resin impregnation.

· Further objects and advantages of the invention will be apparent from the following description of the invention.

SUMMARY OF THE INVENTION In order to attain the objects described above, according to the present invention, a carbon fiber is coated with a sizing at an amount X between 0.05 and 0.3 weight% . The sizing is formed of a heat resistant polymer or a precursor of the heat resistant polymer. The amount X of the sizing is expressed with a following formula: χ ₌ Έ^ _{x 100} where W₀ is a weight of the carbon fiber with the sizing, and i is a weight of the carbon fiber without the sizing.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a graph showing a relationship between strand tensile strength and sizing amount (Kapton type polyimide,

T800SC-24K, KAPTON is a registered trademark of E. I. du

Pont de Nemours and Company) ;

Fig. 2 is a graph showing a relationship between drape value and sizing amount (Kapton type polyimide, T800SC-24K) Fig. 3 is a graph showing a relationship between rubbing fuzz and sizing amount (Kapton type polyimide,

T800SC-24K) ;

Fig. 4 is a graph showing a relationship between ILSS and sizing amount (Kapton type polyimide, T800SC-24K) ;

Fig. 5 is a graph showing a TGA measurement result of

T800S type fiber coated with Kapton type polyimide;

Fig. 6 is a graph showing a TGA measurement result of Kapton type polyimide;

Fig. 7 is a graph showing a relationship between strand tensile strength and sizing amount (ULTEM type

polyetherimide, T800SC-24K, ULTEM is a registered trademark of Saudi Basic Industries Corporation) ; Fig. 8 is a graph showing a relationship between drape value and sizing amount (ULTEM type polyetherimide, T800SC- 24K) ;

Fig. 9 is a graph showing a relationship between rubbing fuzz and sizing amount (ULTEM type polyetherimide, T800SC-24K) ;

Fig. 10 is a graph showing a relationship between ILSS and sizing amount (ULTEM type polyetherimide, T800SC-24K) ;

Fig. 11 is a graph showing a TGA measurement result of T800S type fiber coated with ULTEM type polyetherimide;

Fig. 12 is a graph showing a TGA measurement result of ULTEM type polyetherimide;

Fig. 13 is a graph showing a relationship between strand tensile strength and sizing amount (ULTEM type polyetherimide, T700SC-12K) ;

Fig. 14 is a graph showing a relationship between drape value and sizing amount (ULTEM type polyetherimide, T700SC- 12K) ;

Fig. 15 is a graph showing a relationship between rubbing fuzz and sizing amount (ULTEM type polyetherimide, T700SC-12K) ;

Fig. 16 is a graph showing a relationship between ILSS and sizing amount (ULTEM type polyetherimide, T700SC-12K) ;

Fig. 17 is a graph showing a relationship between strand tensile strength and sizing amount (Methylated melamine-formaldehyde, T700SC-12K) ;

Fig. 18 is a graph showing a relationship between drape value and sizing amount (Methylated melamine-formaldehyde, T700SC-12K) ;

Fig. 19 is a graph showing a relationship between rubbing fuzz and sizing amount (Methylated melamine- formaldehyde, T700SC-12K) ; Fig. 20 is a graph showing a relationship between ILSS and sizing amount (Methylated melamihe-formaldehyde, T700SC- 1-2K) ;

Fig. 21 is a graph showing a TGA measurement result of T700S type fiber coated with methylated melamine- formaldehyde ;

Fig. 22 is a graph showing a TGA measurement result of methylated melamine-formaldehyde;

Fig. 23 is a graph showing a relationship between strand tensile strength and sizing amount (Epoxy cresol novolac, T700SC-12K) ;

Fig. 24 is a graph showing a relationship between drape value and sizing amount (Epoxy cresol novolac, T700SC-12K) ;

Fig. 25 is a graph showing a relationship between rubbing fuzz and sizing amount (Epoxy cresol novolac,

T700SC-12K) ;

Fig. 26 is a graph showing a relationship between ILSS and sizing amount (Epoxy cresol novolac, T700SC-12K) ;

Fig. 27 is a graph showing a TGA measurement result of T700S type fiber coated with epoxy cresol novolac;

Fig. 28 is a graph showing a TGA measurement result of epoxy cresol novolac;

Fig. 29 is a graph showing adhesion strength between a T800S type fiber and polyetherimide resin;

Fig. 30 is a graph showing adhesion strength between a T700S type fiber and polyetherimide resin;

Fig. 31 is a schematic view showing a measurement procedure of drape value;

Fig. 32 is a schematic view showing a measurement instrument of rubbing fuzz;

Fig. 33 is geometry of dumbbell specimen for Single Fiber Fragmentation Test; Table 1 shows a relationship between strand tensile strength and sizing amount (Kapton type polyimide, T800SC- 24K) ;

Table 2 shows a relationship between drape value and sizing amount (Kapton type polyimide, T800SC-24K) ;

Table 3 shows a relationship between rubbing fuzz and sizing amount (Kapton type polyimide, T800SC-24K) ;

Table 4 shows a relationship between ILSS and sizing amount (Kapton type polyimide, T800SC-24K) ;

Table 5 shows a relationship between strand tensile strength and sizing amount (ULTEM type polyetherimide,

T800SC-24K) ;

Table 6 shows a relationship between drape value and sizing amount (ULTEM type polyetherimide, T800SC-24K) ;

Table 7 shows a relationship between rubbing fuzz and sizing amount (ULTEM type polyetherimide, T800SC-24K) ;

.Table 8 shows a relationship between ILSS and sizing amount (ULTEM type polyetherimide, T800SC-24K) ;

Table 9 shows a relationship between strand tensile strength and sizing amount (ULTEM type polyetherimide,

T700SC-12K) ;

Table 10 shows a relationship between drape value and sizing amount (ULTEM type polyetherimide, T700SC-12K) ;

Table 11 shows a relationship between rubbing fuzz and sizing amount (ULTEM type polyetherimide, T700SC-12K) ;

Table 12 shows a relationship between ILSS and sizing amount (ULTEM type polyetherimide, T700SC-12K) ;

Table 13 shows a relationship between strand tensile strength and sizing amount (Methylated melamine-formaldehyde, T700SC-12K) ;

Table 14 shows a relationship between drape value and sizing amount (Methylated melamine-formaldehyde, T700SC- 12K) ; Table 15 shows a relationship between rubbing fuzz and sizing amount (Methylated melamine-formaldehyde, T700SC- 12K) ;

Table 16 shows a relationship between ILSS and sizing amount (Methylated melamine-formaldehyde, T700SC-12K) ;

Table 17 shows a relationship between strand tensile strength and sizing amount (Epoxy cresol novolac, T700SC- 12K) ;

Table 18 shows a relationship between drape value and sizing amount (Epoxy cresol novolac, T700SC-12K) ;

Table 19 shows a relationship between rubbing fuzz and sizing amount (Epoxy cresol novolac, T700SC-12K) ;

Table 20 shows a relationship between ILSS and sizing amount (Epoxy cresol novolac, T700SC-12K) ;

Table 21 shows a comparison result of composite properties;

Table 22 shows adhesion strength between a T800S type fiber and polyetherimide resin; and

Table 23 shows adhesion strength between a T700S type fiber and polyetherimide resin.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

Embodiments of the present invention will be explained with reference to the accompanying drawings.

In the embodiment, a commercially available carbon fiber is used (including graphite fiber) . Specifically, a pitch type carbon fiber, a rayon type carbon fiber, or a PAN (polyacrylonitrile) type carbon fiber is used. Among these carbon fibers, the PAN type carbon fibers that have high tensile strength are the most desirable for the invention.

Among the carbon fibers, there are a twisted carbon fiber and a never twisted carbon fiber. The carbon fibers have preferably a yield of 0.06 - 4.0 g/m and a filament number of 1,000 to 48,000. In order to have high tensile strength and high tensile modulus in addition to preventing single filament breakage from happening during the carbon fiber production, the single filament diameter should be within 3 um to 8 um, more ideally, 4 μπι to 7 urn.

Strand strength is 4.5 GPa or above. 5.0 GPa or above is more desirable. 5.5 GPa or above is even more desirable. Tensile modulus is 200 GPa or above. 220 GPa or above is more desirable. 240 GPa or above is even more desirable. If the strand strength and modulus of the carbon fiber are below 4.5 GPa and 200 GPa, respectively, it is difficult to obtain the desirable mechanical property when the carbon fiber is made into composites materials.

The desirable sizing amount on carbon fiber is between 0.05 and 0.3 weight%. If the sizing amount is less than 0.05 weight%, when carbon fiber tow is spread with some tension, fuzz becomes an issue. If on the other hand, the sizing amount is above 0.3 weight%,. the carbon fiber is almost completely coated by the heat, resistant polymer and would develop voids, resulting in poor density (low), and poor spreadability . When this occurs, even low viscosity resins such as epoxy resins have experienced reduced impregnation; thereby leading to low mechanical properties. From an environmental standpoint, if the sizing amount is less than 0.3 weight%, little volatile could generate.

The desirable relation B/A is over 1.05, and more desirable relation B/A is over 1.1, where A is IFSS

(Interfacial Shear Strength) of unsized fiber and B is IFSS of sized fiber in the present invention whose surface treatment must be same as the unsized fiber. IFSS can be measured with a single fiber fragmentation test, and unsized fiber could be de-sized fiber. A single fiber fragmentation test procedure and a de-sizing method will be described later. The continuous process including carbonization, sizing application, drying and winding is preferred. If the process is not continuous, the possibility of fuzz generation and contamination becomes higher.

In order to obtain composites with high mechanical properties, it is desirable to use continuous fiber when molding, and chopped and/or long fiber reinforced

thermoplastic pellet may also be used. In terms of the types of carbon fibers, chopped fiber for mold injection, continuous fiber for filament winding or pultrusion, or weaving, braiding, or a mat form could be also used.

In order for the carbon fiber to have superior

spreadability and effective resin impregnation, a drape ability (measured by the procedures described below) can be defined as drape value having less than 15 cm, 12 cm or less is better, 10 cm or less is even more desirable, 8 cm or less is most desirable.

As to the matrix resin, either thermosetting or thermoplastic resins could be used. As for the

thermosetting resins, the invention is not limited to any particular resins, and a thermosetting polyimide resin, an epoxy resin, a polyester resin, a polyurethane resin, a urea resin, a phenol resin, a melamine resin, a cyanate ester resin, and a bismaleimide resin may be used. As for the thermoplastic resin, resins, mostly heat resistant resins, that contain oligomer could be used. The invention is not limited to any particular heat resistant thermoplastic resins, and a thermoplastic polyimide resin, a

polyamideimide resin, a polyetherimide resin, a polysulfone resin, a polyethersulfone resin, a polyetheretherketone resin, a polyetherketoneketone resin, and a

polyphenylenesulfide resin may be used.

A heat resistant polymer is a desirable sizing agent to be used for coating the carbon fiber. The sizing agents include a phenol resin, a urea resin, a melamine resin, a polysulfone resin, a polyethersulfone resin, a

polyetheretherketone resin, a polyetherketoneketone resin, a polyphenylenesulfide resin, a polyimide resin, a

polyamideimide resin, a polyetherimide resin, and others.

Generally, a polyimide is made by heat reaction or chemical reaction of polyamic acid. During the imidization process, water is generated as a condensation product;

therefore, it is important to complete imidization before composite fabrication. Otherwise, voids could become a problem due to water generation. A water generation ratio W of 0.05% or less is acceptable, and 0.03% or less is desirable. Ideally, 0.01% or less is optimal. The water generation ratio at the imidization process can be defined by the following equation:

W(%) = B/A x 100

where a weight A is measured after holding 2 hours at 110 degrees Celsius and a weight difference B is measured between 130 degrees Celsius and at 415 degrees Celsius under air atmosphere with TGA (holding 110 degrees Celsius for 2 hours, then heating up to 450 degrees Celsius at 10 degrees Celsius/min) .

An imidization ratio X of 80% or higher is acceptable, and 90% or better is desirable. Ideally, 95% or higher is optimal. The imidization ratio X is defined by the

following equation:

X(%) = (1 - D/C) x 100

where a weight loss ratio C of a polyamic acid without being imidized and a weight loss ratio D of a polyimide are measured between 130 degrees Celsius and 415 degrees Celsius under air atmosphere with TGA (holding 110 degrees Celsius for 2 hours, then heating up to 450 degrees Celsius at 10 degrees Celsius/min) . A degree of imidization is qualitatively measured using an infrared absorption spectrum of the polyimide with FTIR (Fourier transform infrared spectroscopy) which enables to measure the spectrum absorption level of an imide bond (C=0 stretching vibration) at approximately 1,780 cm^"1.

A weight loss ratio Ws based on the sizing amount can be defined by the following equation:

Ws (%) = E/F x 100

where a weight F is the amount of the sizing and a weight difference E is measured between 130 degrees Celsius and 415 degrees Celsius under air atmosphere with TGA (holding 110 degrees Celsius for 2 hours, then heating up to 450 degrees Celsius at 10 degrees Celsius/min) .

The weight loss ratio based on the sizing amount of 7% or less is acceptable, and 5% or less is desirable. Ideally, 3% or less is optimal.

The heat resistant polymer is preferably used in a form of an organic solvent solution, a water solution, a water dispersion or a water emulsion of the polymer itself or a polymer precursor. A polyamic acid which is the precursor to a polyimide is enabled to be water soluble by

neutralization with alkali. It is better for alkali to be water soluble. Chemicals such as ammonia, a monoalkyl amine, a dialkyl amine, a trialkyl amine, and tetraalkylammonium hydroxide could be used.

Organic solvents such as DMF (dimethylformamide) , DMAc (dimethylacetamide) , DMSO (dimethylsulfoxide ) , NMP (N- methylpyrrolidone) , THF (tetrahydrofuran) , etc. could be used. Naturally, low boiling point and safe solvents should be selected. It is desirable that the sizing agent is dried and sometimes reacted chemically in low oxygen concentration air or inert atmosphere such as nitrogen to avoid forming explosive mixed gas. After the heat resistant polymer or polymer precursor is applied to the carbon fiber, it is dried and sometimes reacted chemically in order to obtain heat resistant polymer coating.

The sizing has a glass transition temperature above 100 degrees Celsius. Above 150 degrees Celsius is better. Even more preferably the glass transition temperature shall be above 200 degrees Celsius.

A glass transition temperature is measured according to ASTM E1640 using a Differential Scanning Calorimetry (DSC) .

A thermal decomposition onset temperature of a sized fiber is preferably above 300 degrees Celsius. 370 degrees Celsius or higher is more desirable, 450 degrees Celsius or higher is most desirable. When a thermal decomposition onset temperature is measured, first, a sample with a weight of about 5 mg is dried in an oven at 110 degrees Celsius for 2 hours, and cooled off in a desiccator at room temperature for 1 hour. Then it is placed on a thermogravimetric analyzer (TGA) under air atmosphere. Then, the sample is analyzed under an air flow of 50 ml/minute at a heating ratio of 10 degrees Celsius/minute. A weight change is measured between less than 120 degrees Celsius and 650 degrees Celsius. The decomposition onset temperature of a sized fiber is defined as a temperature at which an onset of a major weight loss occurs. From the TGA experimental data, the sample weight, expressed as a percentage of the initial weight, is plotted as a function of the temperature

(abscissa) . By drawing tangents on a curve, the

decomposition onset temperature is defined as an

intersection point where tangent at a steepest weight loss crosses a tangent at minimum gradient weight loss adjacent to the steepest weight loss on a lower temperature side. If it is difficult to measure a decomposition onset temperature of a sized fiber, a sizing can be used in place of a sized fiber. <30% weight reduction temperature>

A 30% weight reduction temperature of a sizing is preferably higher than 350 degrees Celsius. 420 degrees Celsius or higher is more desirable. 500 degrees Celsius or higher is most desirable. When a 30% weight reduction temperature is measured, first, a sample with a weight of about 5 mg is dried in an oven at 110 degrees Celsius for 2 hours, and cooled off in a desiccator at room temperature for 1 hour. Then it is placed on a thermogravimetric analyzer (TGA) under air atmosphere. Then, the sample is analyzed under an air flow of 50 ml/minute at a heating ratio of 10 degrees Celsius/minute. A weight change is measured between less than 120 degrees Celsius and 650 degrees Celsius. From the TGA experimental data, the sample weight, expressed as a percentage of the initial weight, is plotted as a function of the temperature (abscissa) . The

30% weight reduction temperature of the sizing is defined as a temperature at which the weight of the sizing reduces by 30% with reference to the weight of the said sizing at 130 degrees Celsius.

A sizing agent application method includes a roller sizing method, a submerged roller sizing method and/or a spray sizing method. The submerged roller sizing method is desirable because it is possible to apply a sizing agent very evenly even to large filament count tow fibers.

Sufficiently spread carbon fibers are submerged in the sizing agent. In this process, a number of factors become important such as a sizing agent concentration, temperature, fiber tension, etc. for the carbon fiber to attain the optimal sizing amount for the ultimate objective to be realized. Often, ultrasonic agitation is applied to vibrate carbon fiber during the sizing process for better end results.

In order to achieve a sizing amount 0.05 to 0.3 weight% on the carbon fiber, the bath sizing concentration is preferably 0.05 to 2.0 weight! , more preferably 0.1 to 1.0 weight% .

After the sizing application process, the carbon fiber goes through the drying treatment process in which water and/or organic solvent will be dried, which are solvent or dispersion media. Normally an air dryer is used and the dryer is run for six seconds to fifteen minutes. The dry temperature should be set at 200 degrees Celsius to 450 degrees Celsius, 240 degrees Celsius to 410 degrees Celsius would be more ideal, 260 degrees Celsius to 370 degrees Celsius would be even more ideal, and 280 degrees Celsius to 330 degrees Celsius would be most desirable.

In case of thermoplastic dispersion, it is desirable that it should be dried at over the formed or softened temperature. This could also serve a purpose of reacting to the desired polymer characteristics. For this invention, the heat treatment will possibly be used with a higher temperature than the temperature used for the drying treatment. The atmosphere to be used for the drying treatment should be air; however, when an organic solvent is used in the process, an inert atmosphere involving elements such as nitrogen could be used.

<Winding process> The carbon fiber tow, then, is wound onto a bobbin.

The carbon fiber produced as described above is evenly sized. This helps make desired carbon fiber reinforced composites materials when mixed with the resin.

Examples

Examples of the carbon fiber will be explained next.

Following methods are used for evaluating properties of the carbon fiber.

The polyimide type sizing amount (weight%) is measured by the following method.

(1) About 5 g carbon fiber is taken.

(2) The sample is put in an oven at 110 degrees Celsius for 1 hour.

(3) It is put in a desiccator to be cooled off at the

ambient temperature (room temperature) .

(4) A weight o is weighed.

(5) For removing the sizing by alkaline degradation, it is put in 5% KOH solution at 80 degrees Celsius for 4 hours.

(6) The de-sized sample is rinsed with enough water and put in an oven for 1 hour at 110 degrees Celsius.

(7) It is put in a desiccator to be cooled off at the

ambient temperature (room temperature) .

(8) A weight Wi is weighed.

The sizing amount (weight%) is calculated by the following formula . Sizing amount (weight%) = (W₀ - Wi)/(W₀) x 100

The sizing amount (weight!) other than polyimide type sizing is measured by the following method. (1) About 2 g carbon fiber is taken.

(2) The sample is put in an oven at 110 degrees Celsius for 1 hour.

(3) It is put in a desiccator to be cooled off at the ambient temperature (room temperature) .

(4) A weight W₀ is weighed.

(5) For removing the sizing, it is put in a furnace at 450 degrees Celsius for 20 minutes.

(6) The de-sized sample is put in a nitrogen purged

container for 1 hour.

(7) A weight Wi is weighed.

The sizing amount (weight%) is calculated by the following formula .

Sizing amount (weight%) = (W₀ - Wi)/(W₀) x 100

Tensile strength and tensile modulus of the strand specimen made of polymer coated carbon fiber and epoxy resin matrix is measured by ASTM D4018.

The carbon fiber tow is cut about 50 cm long from the bobbin without applying any tension. One end of the specimen is glued on the desk and draped, and a weight is placed on the other end of the specimen. After a twist and/or bend of the specimen are removed, the specimen is placed for 30 minutes. The weight is 30 g for 12,000 filaments and 60 g for 24,000 filaments, so that 1 g tension is applied per 400 filaments. After the weight is released from the specimen, the specimen is placed on a rectangular table such that a portion of the specimen is extended by 25 cm from an edge of the table having 90 degrees angle as shown in Fig. 31. The specimen on the table is fixed with an adhesive tape without breaking so that the portion hangs down from the edge of the table. A distance D (Fig. 31) between a tip of the specimen and a side of the table is defined as the drape value. <Fuzz count>

As shown in Fig. 32, the carbon fiber tow is slid against four pins with a diameter of 10 mm (material:

chromium steel, surface roughness: 1 - 1.5 μπι RMS) at a speed of 3 meter/minute in order to generate fuzz. The carbon fiber initial tension is 500 g for the 12,000 filament strand and 650 g for 24,000 filament strand. The carbon fiber is slid against the pins by an angle of 120 degrees. The four pins are placed (horizontal distance) 25 mm, 50 mm and 25 mm apart (refer to Fig. 32) . After the carbon fiber passes through the pins, a fuzz blocks light incident on a photo electric tube from above, so that a fuzz counter counts the fuzz count.

ILSS of the composites consisting of the polymer coated carbon fiber and an epoxy resin matrix is measured by ASTM D2344.

Specimens are prepared with the following procedure.

(1) Two aluminum plates (length: 250 ^χ width: 250 ^χ

thickness: 6 (mm)), a Kapton film (thickness: 0.1 (mm)), a Kapton tape, a mold release agent, an ULTEM type

polyetherimide resin sheet (thickness 0.26 (mm)), which must be dried in a vacuum oven at 110 degrees Celsius for at least 1 day, and carbon fiber strand are prepared.

(2) The Kapton film (thickness: 0.1 (mm)) coated with a mold release agent is set on an aluminum plate. (3) The ULTEM type polyetherimide resin sheet (length: 90 ^χ width: 150 ^χ thickness: 0.26 (mm)), whose grease on the surface is removed with acetone, is set on the Kapton film.

(4) A single filament is picked up from the carbon fiber strand and set on the ULTEM type polyetherimide resin sheet.

(5) The filament is fixed at the both sides with a Kapton tape to be kept straight.

(6) The filament (filaments) is overlapped with another ULTEM type polyetherimide resin sheet (length: 90 * width: 150 x thickness: 0.26 (mm)), and Kapton film (thickness: 0.1 (mm)) coated with a mold release agent is overlapped on it.

(7) Spacers (thickness: 0.7 (mm)) are set between two aluminum plates.

(8) The aluminum plates including a sample are set on the pressing machine at 290 degrees Celsius.

(9) They are heated for. ten minutes contacting with the pressing machine at 0.1 MPa .

(10) They are pressed at 1 MPa and cooled at a speed of 15 degrees Celsius/minute being pressed at 1 MPa.

(11) They are taken out of the pressing machine when the temperature is under 180 degrees Celsius.

(12) Dumbbell specimen where a single filament is embedded in the center along the loading direction has the center length 20 mm, the center width 5 mm and the thickness 0.5 mm as shown in Fig. 33.

SFFT is performed at a strain rate of approximately 4 % /minute counting the fragmented fiber number in the center 20 mm of the specimen at every 0.64% strain with a polarized microscope until the saturation of fragmented fiber number. The preferable number of specimens is more than 2 and Interfacial Shear Strength (IFSS) is obtained from the average length of the fragmented fibers at the saturation point of fragmented fiber number. IFSS can be shown as below, where Of is the strand strength, d is the fiber diameter, L_c is the critical length (=4*L_b/3) and L_b is the average length of fragmented fibers.

0,-d

l^FSS=^T

<De-sizing process>

De-sized fiber may be used for SFFT in place of unsized fiber. De-sizing process is as follows.

(1) Sized fiber is put in a furnace of nitrogen atmosphere at 500 degrees Celsius, where the oxygen concentration is less than 7 weight%.

(2) The fiber is kept in the furnace for 20 minutes.

(3) The de-sized fiber is cooled at room temperature in nitrogen atmosphere for 1 hour.

Example 1, Comparative Example 1:

Unsized 24K high tensile strength, intermediate modulus carbon fiber "Torayca" T800SC (Registered trademark by Toray Industries; strand strength 5.9 GPa, strand modulus 294 GPa) was used. The carbon fiber was continuously submerged in the sizing bath containing polyamic acid ammonium salt of 0.1 to 1.0 weight% . The polyamic acid is formed from the monomers pyromellitic dianyhydride and 4 , 4¹ -oxydiphenylene . After the submerging process, it was dried at 300 degrees Celsius for one minute in order to have poly (4,4 '- oxydiphenylene-pyromellitimide) (Kapton type polyimide) coating .

The tensile strengths of both the sizing amount of 0.05 to 0.3 weight% (Example 1) and 0.31 to 0.5 weight%

(Comparative Example 1) were measured. The results are shown in both Table 1 and Fig. 1. The error bar in the figure indicates the standard deviation. Additionally mechanical properties of unsized fiber are also shown.

Example 2, Comparative Example 2:

The same as the above Example 1 and Comparative Example

1, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 2) and the other with 0.31 to

0.5 weight% (Comparative Example 2) to test the drape value. The result is indicated in both Table 2 and Fig. 2. The error bar in the figure indicates the standard deviation. As the sample of Example 2 has superior drapeability than that of Comparative Example 2, the sample of Example 2 demonstrates the superior spreadability and impregnation. Additionally drape value of unsized fiber are also shown.

Example 3, Comparative Example 3:

The same as the above Example 1 and Comparative Example

1, the samples were made, i.e. one with sizing amount of

0.05 to 0.3 weight% (Example 3), the other with 0.31 to 0.5 weight% (Comparative Example 3) and unsized fiber

(Comparative Example 3) to conduct a fuzz count test. The result is shown in Table 3 and Fig. 3. The error bar in the figure indicates the standard deviation. The fuzz count of unsized fiber is extremely high and the fiber with 0.05 to 0.3 weight% amount sizing showed almost equal fuzz count as the fiber with 0.31 to 0.5 weight% amount sizing, indicating that the low sizing amount (0.05 to 0.3 weight%) carbon fiber could be processed as easily. Example 4, Comparative Example 4:

The same as the above Example 1 and Comparative Example

1, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 4) and the other with 0.31 to 0.5 weight! (Comparative Example 4) to conduct an ILSS test. The result is indicated in both Table 4 and Fig. 4. The error bar in the figure indicates the standard deviation. The ILSS measurements of the both samples taken from the test are almost identical, verifying that the low sized (0.05 to 0.3 weight%) carbon fiber also has superb

interfacial adhesion. Additionally ILSS of unsized fiber are also shown.

Example 5:

Thermogravimetric analysis (TGA) was conducted under air atmosphere. The heat decomposition onset temperature of the same carbon fiber as the above Example 1 is 510 degrees Celsius as shown in Fig. 5. The heat decomposition onset temperature of the sizing is 585 degrees Celsius and the 30% weight reduction temperature is 620 degrees Celsius as shown in Fig. 6, confirming the heat resistance is in excess of 500 degrees Celsius.

Example 6, Comparative Example 5:

Unsized 24K high tensile strength, intermediate modulus carbon fiber "Torayca" T800SC (Registered trademark by Toray Industries; strand strength 5.9 GPa, strand modulus 294 GPa) was used. The carbon fiber was continuously submerged in the sizing bath containing polyamic acid

dimethylaminoethanol salt of 0.1 - 2.0 weight%. The polyamic acid is formed from the monomers 2, 2' -Bis (4- (3, 4- dicarboxyphenol ) phenyl) propane dianhydride and meta- phenylene diamine. After the submerging process, it was dried at 300 degrees Celsius for one minute in order to have 2, 2-Bis (4- (3, 4-dicarboxyphenol) phenyl) propane dianhydride-m- phenylene diamine copolymer (ULTEM type polyetherimide) coating. The imidization ratio was 98%.

The tensile strengths of both the sizing amount of 0.05 to 0.3 weight% (Example 6) and 0.31 to 0.7 weight% (Comparative Example 5) were measured. The results are shown in both Table 5 and Fig. 7. The error bar in the figure indicates the standard deviation. The test sample of

Example 6 had a higher tensile strength than that of

Comparative Example 5. Additionally mechanical properties of unsized fiber are also shown.

Example 7, Comparative Example 6:

The same as the above Example 6 and Comparative Example 5, the samples were made, i.e. one with sizing amount of

0.05 to 0.3 weight% (Example 7) and the other with 0.31 to

0.7 weight%- (Comparative Example 6) to test the drape value.

The result is indicated in both Table 6 and Fig. 8. The error bar in the figure indicates the standard deviation. The sample of Example 7 has superior drapeability than that of Comparative Example 6. Additionally drape value of unsized fiber are also shown.

Example 8, Comparative Example 7:

The same as the above Example 6 and Comparative Example

5, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 8) and the other with 0.31 to 0.7 weight% (Comparative Example 7) to conduct a fuzz count test. The result is shown in Table 7 and Fig. 9. The error bar in the figure indicates the standard deviation. The fuzz count of the both samples is almost equal. The unsized carbon fiber generated much fuzz, indicating the

effectiveness of sizing in preventing fuzz occurrence. Example 9, Comparative Example 8:

The same as the above Example 6 and Comparative Example 5, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 9) and the other with 0.31 to 0.7 weight% (Comparative Example 8) to conduct an ILSS test. The result is indicated in both Table 8 and Fig. 10. The error bar in the figure indicates the standard deviation. The ILSS measurements of the both samples taken from the test^" are almost identical. Additionally ILSS of unsized fiber are also shown.

Example 10:

Thermogravimetric analysis (TGA) was conducted under air atmosphere. The heat decomposition onset temperature of the same carbon fiber as the above Example 6 is over 550 degrees Celsius as shown in Fig. 11. The heat decomposition onset temperature of the sizing was 548 degrees Celsius and the 30% weight reduction temperature is 540 degrees Celsius as shown in Fig. 12, confirming the heat resistance is in excess of 500 degrees Celsius.

Example 11, Comparative Example 9:

Unsized 12K high tensile strength, standard modulus carbon fiber ""Torayca" T700SC (Registered trademark by Toray Industries - strand strength 4.9 GPa, strand modulus 230 GPa) was used. The carbon fiber was continuously submerged in the sizing bath containing polyamic acid dimethylaminoethanol salt of 0.1 - 2.0 weight%. The polyamic acid is formed from the monomers 2, 2' -Bis (4- (3, 4- dicarboxyphenol) phenyl) propane dianhydride (BPADA) and meta- phenylene diamine (m-PDA) . After the submerging process, it was dried at 300 degrees Celsius for one minute in order to have ULTEM type polyetherimide coating. The imidization ratio was 98%. The tensile strengths of both the sizing amount of 0.05 to 0.3 weight% (Example 11) and 0.31 to 0.7 weight% (Comparative Example 9) were measured. The results are shown in both Table 9 and Fig. 13. The error bar in the figure indicates the standard deviation. The test sample of Example 11 had a higher tensile strength than that of Comparative Example 9. Additionally mechanical properties of unsized fiber are also shown.

Example 12, Comparative Example 10:

The same as the above Example 11 and Comparative

Example 9, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 12) and the other with 0.31 to 0.7 weight% (Comparative Example 10) to test the drape value. The result is indicated in both Table 10 and Fig. 14. The error bar in the figure indicates the standard deviation. The sample of Example 12 has superior drapeability than that of Comparative Example 10.

Additionally drape value of unsized fiber are also shown. Example 13, Comparative Example 11:

The same as the above Example 11 and Comparative

Example 9, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 13), the other with 0.31 to 0.7 weight% (Comparative Example 11) and unsized fiber (Comparative Example 11) to conduct a fuzz count test. The result is shown in Table 11 and Fig. 15. The error bar in the figure indicates the standard deviation. The fuzz count of the both samples is almost equal. The carbon fiber without a sizing agent generated much fuzz indicating the effectiveness of sizing in preventing fuzz occurrence.

Example 14, Comparative Example 12:

The same as the above Example 11 and Comparative

Example 9, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 14) and the other with 0.31 to 0.7 weight% (Comparative Example 12) to conduct an ILSS test. The result is indicated in both Table 12 and Fig. 16. The error bar in the figure indicates the standard deviation. The ILSS measurements of the both samples taken from the test are almost identical, verifying that the low sized (0.05 to 0.3 weight%) carbon fiber also has superb interfacial adhesion. Additionally ILSS of unsized fiber are also shown.

Example 15, Comparative Example 13:

Unsized 12K high tensile strength, standard modulus carbon fiber "Torayca" T700SC (Registered trademark by Toray Industries - strand strength 4.9 GPa, strand modulus 230 GPa) was used. The carbon fiber was continuously submerged in the sizing bath containing 0.2 to 1.6 weight% of

methylated me1amine-formaldehyde resin. The tensile

strengths of both the sizing amount of 0.05 to 0.3 weight% (Example 15) and 0.31 to 0.7 weight% (Comparative Example 13) were measured. The results are shown in both Table 13 and Fig. 17. The error bar in the figure indicates the standard deviation. Additionally mechanical properties of unsized fiber are also shown. Example 16, Comparative Example 14:

The same as the above Example 15 and Comparative

Example 13, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 16) and the other with 0.31 to 0.7 weight% (Comparative Example 14) to test the drape value. The result is indicated in both Table 14 and Fig. 18. The error bar in the figure indicates the standard deviation. The sample of Example 16 has superior drapeability than that of Comparative Example 14.

Additionally drape value of unsized fiber are also shown.

Example 17, Comparative Example 15:

The same as the above Example 15 and Comparative

Example 13, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 17) and the other with 0.31 to 0.7 weight% (Comparative Example 15) to conduct a fuzz count test. The result is shown in Table 15 and Fig. 19. The error bar in the figure indicates the standard deviation.

Example 18, Comparative Example 16:

The same as the above Example 15 and Comparative

Example 13, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 18) and the other with 0.31 to 0.7 weight% (Comparative Example 16) to conduct an ILSS test. The result is indicated in both Table 16 and Fig. 20. The error bar in the figure indicates the standard deviation. Additionally ILSS of unsized fiber are also shown.

Example 19:

Thermogravimetric analysis (TGA) was conducted under air atmosphere. The heat decomposition onset temperature of the same carbon fiber as the above Example 15 is 390 degrees Celsius as shown in Fig. 21. The heat decomposition onset temperature of the sizing only is 375 degrees Celsius and the 30% weight reduction temperature is 380 degrees Celsius as shown in Fig. 22, confirming the heat resistance is in excess of 350 degrees Celsius.

Example 20, Comparative Example 17:

Unsized 12K high tensile strength, standard modulus carbon fiber "Torayca" T700SC (Registered trademark by Toray Industries - strand strength 4.9 GPa, strand modulus 230 GPa) was used. The carbon fiber was continuously submerged in the sizing bath containing 0.1 to 2.0 weight% of epoxy cresol novolac resin. The tensile strengths of both the sizing amount of 0.05 to 0.3 weight% (Example 20) and 0.31 to 0.8 weight% (Comparative Example 17) were measured. The results are shown in both Table 17 and Fig. 23. The error bar in the figure indicates the standard deviation.

Additionally mechanical properties of unsized fiber are also shown.

Example 21, Comparative Example 18:

The same as the above Example 20 and Comparative

Example 17, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 21) and the other with 0.31 to 0.8 weight% (Comparative Example 18) to test the drape value. The result is indicated in both Table 18 and Fig. 24. The error bar in the figure indicates the standard deviation. The sample of Example 21 has superior drapeability than that of Comparative Example 18.

Additionally drape value of unsized fiber are also shown.

Example 22, Comparative Example 19:

The same as the above Example 20 and Comparative

Example 17, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 22) and the other with 0.31 to 0.8 weight! (Comparative Example 19) to conduct a fuzz count test. The result is shown in Table 19 and Fig. 25. The error bar in the figure indicates the standard deviation.

Example 23, Comparative Example 20:

The same as the above Example 20 and Comparative

Example 17, the samples were made, i.e. one with sizing amount of 0.05 to 0.3 weight% (Example 23) and the other with 0.31 to 0.8 weight% (Comparative Example 20) to conduct an ILSS test. The result is indicated in both Table 20 and Fig. 26. The error bar in the figure indicates the standard deviation. The ILSS measurements of the both samples taken from the test are almost identical. Additionally ILSS of unsized fiber are also shown.

Example 24:

Thermogravimetric analysis (TGA) was conducted under air atmosphere. The heat decomposition onset temperature of the same carbon fiber as the above Example 20 is 423 degrees Celsius as shown in Fig. 27. The heat decomposition onset temperature is 335 degrees Celsius and the 30% weight

reduction temperature is 420 degrees Celsius as shown in Fig. 28, confirming the heat resistance is in excess of 300 degrees Celsius.

Example 25, Comparative Example 21, 22:

As indicated in Examples 11 the carbon fiber with about

0.2% heat resistant sizing (Examples 25), "Torayca" T700SC- 12K-60E and Unsized fiber T700SC-12K (Comparative Examples 21, 22) were used.

Unidirectional specimens were obtained by stacking thermoplastic tapes made of carbon fiber strand and PPS resin. In accordance with EN2850, the compression tests were conducted. As a result, as indicated in Table 21,

Example 25 is superior to the Comparative Examples 21 and 22. Example 26, 27, Comparative Example 23, 24:

As indicated in Examples 1 and 6 the carbon fiber with about 0.2% heat resistant sizing (Example 26, 27), "Torayca" T800SC-24K-10E and Unsized fiber T800SC-24K (Comparative Examples 23, 24) were used.

Fig. 29 and Table 22show the results of SFFT using

polyetherimide resin. From the results it can be shown the IFSS of Example 26 and 27 are 5% higher than that of

Comparative Example 23 and 24. Example 28, 29, 30, Comparative Example 25:

As indicated in Examples 11, 15 and 20 the carbon fiber with about 0.2% heat resistant sizing (Examples 28, 29, 30) and Unsized fiber T700SC-12K (Comparative Examples 25) were used.

Fig. 30 and Table 23 show the results of SFFT using polyetherimide resin. It can be shown the IFSS of Example 28 through 30 are 5% higher than that of Comparative Example 25. While the invention has been explained with reference to the specific embodiments of the invention, the

explanation is illustrative and the invention is limited only by the appended claims.

Claims

What is claimed is:

1. A carbon fiber coated with a sizing at an amount X between 0.05 and 0.3 weight%, said sizing being formed of a heat resistant polymer or a precursor of the heat resistant polymer, said amount X being expressed with a following formula :

where Wo is a weight of the carbon fiber with the sizing, and Wi is a weight of the carbon fiber without the sizing.

2.. The carbon fiber according to claim 1, wherein said heat resistant polymer has a thermal decomposition onset temperature higher than 300. degrees Celsius.

3. The carbon fiber according to claim 1, wherein said heat resistant polymer has a thermal decomposition onset temperature higher than 370 degrees Celsius.

4. The carbon fiber according to claim 1, wherein said heat resistant polymer has a thermal decomposition onset temperature higher than 450 degrees Celsius.

5. The carbon fiber according to claim 1, wherein said heat resistant polymer has a 30% weight reduction

temperature higher than 350 degrees Celsius.

6. The carbon fiber according to claim 1, wherein said heat resistant polymer has a 30% weight reduction

temperature higher than 420 degrees Celsius.

7. The carbon fiber according to claim 1, wherein said heat resistant polymer has a 30% weight reduction

temperature higher than 500 degrees Celsius.

8. The carbon fiber according to claim 1 having an interfacial shear strength A greater than an interfacial shear strength B of the carbon fiber without the sizing to satisfy a relation of A > B, said interfacial shear strength A and B being measured with a single fiber fragmentation

9. The carbon fiber according to claim 7 having the interfacial shear strength A satisfying a relation of A/B ≥ 1.05.

10. The carbon fiber according to claim 7 having the interfacial shear strength A satisfying a relation of A/B 1.10.

11. The carbon fiber according to claim 1, wherein said heat resistant polymer or said precursor is applied to the carbon fiber in a form of an organic solution, an aqueous solution, an aqueous dispersion, or an aqueous emulsion.

12. The carbon fiber according to claim 1 being produced through a fabrication process including a

carbonization process, a sizing application process, a drying process, and a continuous winding process.

13. The carbon fiber according to claim 1 being produced through a fabrication process including a drying process at a temperature higher 200 degrees Celsius for longer than 6 seconds.

14. The carbon fiber according to claim 1 being produced through a fabrication process including a drying process at a temperature higher 240 degrees Celsius for longer than 6 seconds.

15. The carbon fiber according to claim 1 being produced through a fabrication process including a drying process at a temperature higher 280 degrees Celsius for longer than 6 seconds.

16. The carbon fiber according to claim 1, wherein said heat resistant polymer includes at least one of a phenol resin, a melamine resin, a urea resin, a polyimide resin, a polyamideimide resin, a polyetherimide resin, a polysulfone resin, a polyethersulfone resin, a polyetheretherketone resin, a polyetherketoneketone resin, and a

polyphenylenesulfide resin.

17. The carbon fiber according to claim 1 having a tensile modulus between 200 and 600 GPa.

18. The carbon fiber according to claim 1 having a tensile strength between 4.5 and 7 GPa.

19. The carbon fiber according to claim 1 having a drape value less than 15 cm.

20. The carbon fiber according to claim 1 being formed of filaments having a number between 1,000 and 48,000.

21. A composite material comprising the carbon fiber according to claim 1 and a thermoplastic resin..

22. A composite material comprising the carbon fibe rding to claim 1 and a thermosetting resin.