BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to a material comprising filaments, each filament having a diameter less than 2 μm, such that the use of this material as a filler provides composites with very high effectiveness for electromagnetic interference (EMI) shielding and very low surface impedance. This invention also relates to a process for making this material.

2. Description of the Prior Art

Electrically conducting polymer-matrix composites are used for electromagnetic interference (EMI) shielding, the need of which is increasingly important due to the increasing sensitivity and abundance of electronics. Polymer-matrix composites with low surface inpedance are used for lightweight microwave waveguides, which are commonly associated with radars in military aircrafts and satellites. Polymer-matrix components, as opposed to monolithic metals, are desirable due to their moldability. They are desirable compared to metal-coated monlithic polymers due to the fact that the coatings are prone to damage by scratching, abrasion or wear. These composites are conducting due to the presence of an electrically conducting filler, which can be discontinuous (such as particles and short fibers) or continuous (such as continuous fibers). Discontinuous fillers are in general less effective than continuous fillers in decreasing the electrical resistivity ofthe composite. However, unlike DC (direct current) electrical conduction, EMI shielding and conduction at high frequencies (such as 1 GHz) do not require continuity or percolation ofthe conducting phase, though continuity or percolation helps. Moreover, discontinuous fillers are suitable for composite fabrication by injection molding and, if the discontinuous filler is fine enough in size, even by ink jet printing or screen printing. Due to the lower cost and greater versatility of composite fabrication for discontinuous fillers compared to continuous fillers, discontinuous fillers are widely used for making electrically conducting composites, especially those for EMI shielding.

Discontinuous fillers that have been used in polymer matrices for EMI shielding and for providing low surface impedance include metal particles (e.g., Ni particles), metal flakes (e.g., Al flakes), carbon particles, (e.g., graphite and carbon black), carbon fibers, metal fibers (e.g., stainless steel fibers, nickel fibers), metal coated carbon particles, and metal coated carbon fibers (e.g., Ni coated carbon fibers). For any filler, the EMI shielding effectiveness increases and the surface

impedance decreases with increasing filler volume fraction in the composite, but the maximum filler volume fraction is limited by the poor composite mechanical properties at high filler volume fractions resulting from the poor filler-matrix bonding. For materials and process cost saving and good mechanical properties, the attainment of a high shielding effectiveness and a low surface impedance at a low filler volume fraction is desirable. Because electromagnetic waves at high frequencies (such as microwaves) interact with a conductor only near its surface, a smaller unit size ofthe conducting filler enables a higher shielding effectiveness and a lower surface impedance to be attained at the same filler volume fraction. The only submicron diameter filament filler that has been previously used for EMI shielding is carbon filaments (catalytically grown from carbonaceous gases); a shielding effectiveness at 1 GHz of 45 dB was attained at a filler content of 60 wt.% (about 45 vol.%), see

Makoto Katsumata, Hidenori Yamanashi, Hitoshi Ushijima and Morinobu Endo, SPIE Proc, Vol. 1916 (Smart Materials), edited by Vijay K. Varadan, 1993, p. 140- 148. These carbon filaments have not been previously used for materials with low surface impedance. They are in contrast to conventional carbon fibers (made from pitch or polymer precursors), which typically have diameter in the range 7-11 μm. The coating of conventional carbon fibers with nickel has been previously shown to greatly increase the shielding effectiveness, so that a shielding effectiveness at 1 GHz of 76 dB was attained, see Mai Murthy, 4th Int. SAMPE Electronics Conference, Vol. 4 (Electronic Materials - Our Future), edited by Ron W. Allied, Robert J. Martinez and Ken B. Wischmann, 1990, p. 806-818. The nickel-coated carbon fibers (not filaments) in the prior art contain a very minor volume fraction of nickel, as the nickel coating is very thin compared to the diameter of the fiber, see C.P. Beetz, Jr. and R.J. Schmitt, in SAMPE Journal, Vol. 19, no. 3, p. 6-9, 1983. For example, R.E. Evans, D.E. Hall and B.A. Luxon, in SAMPE Quarterly, Vol. 17, no.4, p. 18-26 (1986), teach the use of a nickel coating that is 0.35 μm thick, compared to the 7.0 μm diameter for the bare carbon fiber and the 7.8 μm diameter for the nickel-coated carbon fiber.

Fibers that are all nickel are made in the prior art by either deformation processing (such as wire drawing) or melt processing, both of which are expensive processes compared to coating by metal plating. Moreover, the difficulty of deformation processing and melt processing increases with decreasing fiber diameter; fibers of diameter less than 2 μm either cannot be produced by these processes or can be produced only at high costs.

The processes of plating a metal on to carbons include electroplating and electroless plating in the prior art. Electroless plating requires no electrical contacts to be made on the carbons, whereas

electroplating does. Electrical contacts can be conveniently made on continuous forms of carbons, but not on discontinuous forms. Therefore, electroless plating is commonly used for discontinuous forms of carbon, such as graphite powder and short carbon fibers, whereas electroplating is commonly used for continuous forms of carbon, such as continuous carbon fibers. The carbon filaments related to this invention are discontinuous, so they cannot be electroplated in the same way as continuous carbon fibers, which, in the electroplating process, are fed by rollers into an electrolyte bath such that the rollers also serve as electrical contacts. In the prior art, particles (discontinuous) are electroplated by allowing them to bombard the cathode multiple times, see Eiki Takeshima, Kiyoshi Takatsu, Youichi Kojima and Takahiro Fujii, in U.S. Patent 4,908,106 (1990), and Conrad M. Yu and John D. Illige, in U.S. Patent 4,316,786 (1982). During bombardment, electrical contact is made between the particle and the cathode, thereby allowing electroplating on the particle. This bombardment method is not suitable for the carbon filaments, which are submicron in diameter (hence easy to break upon bombardment) and are intertwined and fluffy as in cotton wool (hence difficult to project for bombardment).

3. Objectives ofthe Invention

An objective of this invention is to provide a material comprising filaments, each filament having a diameter less than 2 μm, for use as a filler in composites for EMI shielding and for giving a low surface impedance, such that the filament's fabrication is inexpensive and the EMI shielding effectiveness is superior to that of fillers in the prior art.

Another objective is to provide a filler in composites for EMI shielding, such that a high shielding effectiveness is attained even at a low filler volume fraction.

Yet another objective is to provide a process for making the said filament or filler. Still another objective is to provide a process for electroplating filaments that are in a discontinuous form.

SUMMARY OF THE INVENTION There is provided a material in the form of a metal or metal-coated filament, having diameter less than 2 μm, such that the incorporation of this material in a matrix that is incapable of electromagnetic interference shielding results in a composite that is capable of shielding electromagnetic interference and that has a low surface impedance.

There is also provided a process for making the said filament. This process comprises (i) removing polyaromatic hydrocarbons from the surface of carbon filaments that are made from

carbonaceous gases, and (ii) plating the resulting carbon filaments with a metal. There is moroever provided a process for said plating by electroplating.

There is furthermore provided a composite comprising said filaments and a matrix, which is a polymer, a ceramic or a polymer-ceramic combination.

BRIEF DESCRIPTION OF THE DRAWINGS For a better understanding of the invention, reference should be had to the accompanying drawings wherein:

FIG. 1 is a schematic showing the experimental set-up for electroplating carbon filaments. FIG. 2-5 are scanning electron microscope (SEM) photographs. FIG. 2 and FIG. 3 show carbon filaments at two magnifications; FIG. 4 and FIG. 5 show nickel filaments at two magnifications.

FIG.6 is a graph showing the variation ofthe volume electrical resistivity of composites with filler volume fraction, (a) Carbon filaments, (b) Nickel filaments, (c) Nickel fibers of diameter 2 μm. (d) Nickel fibers of diameter 20 μm.

FIG. 7 AND FIG. 8 are SEM photographs of Ni fibers of diameter 2 μm at two magnifications.

FIG. 9 AND FIG. 10 are SEM photographs of Ni fibers of diameter 20 μm at two magnifications. FIG. 11 is a schematic showing the experimental set-up for measuring the EMI shielding effectiveness.

FIG. 12 is a graph ofthe EMI shielding effectiveness vs. frequency for composites containing Ni filaments at various volume fractions.

FIG. 13 is a graph ofthe EMI shielding effectiveness vs. frequency for composites containing 22 μm diameter Ni fibers at various volume fractions.

FIG. 14 is a graph ofthe EMI shielding effectiveness vs. frequency for composites containing 20 μm diameter Ni fibers at various volume fractions.

FIG. 15 is a graph ofthe EMI shielding effectiveness vs. frequency for composites contauiing carbon filaments at various volume fractions. FIG. 16 is a schematic showing the method for measuring surface impedance and EMI shielding effectiveness.

5

DESCRIPTION OF THE EMBODIMENTS The following description of the embodiments is for illustrative purposes only and is not intended to limit the scope of the invention in any way.

Example I This example pertains to a nickel-coated carbon filament containing 94 vol.% Ni and 6 vol.%

C, such that the coated filament had a diameter of 0.4 μm, compared to a diameter of 0.1 μm for the carbon filament which forms its core. Due to the large proportion of nickel, this filament is referred to as a nickel filament.

This example also pertains to a method of making said filaments. Furthermore, this example pertains to a composite material comprising said filaments and a polymer.

The nickel filament was 0.404 ± 0.022 μm in diameter, such that it contained a carbon core of diameter 0.096 ± 0.018 μm. Each nickel filament contained 94.4 vol.% Ni and 5.6 vol.% C. The carbon core was present because the nickel filaments were fabricated by the electroplating of nickel on to carbon filaments.

The carbon filaments were provided as Grade ADNH by Applied Sciences, Inc. (Cedarville,

Ohio). Prior to electroplating, the surface ofthe carbon filaments was treated by washing in acetone, which removed surface contaminants in the form of mainly polyaromatic hydrocarbons, as identified by mass spectrometry conducted in conjunction with gas chromatography. The hydrocarbons removed include pentadecane, tetradecane, aliphatic hydrocarbon, phenanthrene, fluoranthene, naphthalene and fluorenone. The surface contaminants were due to the process of making the carbon filaments by carbonaceous gases, as shown by the observation (also by mass spectrometry in conjunction with gas chromatography) of similar polyaromatic hydrocarbons in the tarry residue left on the wall of the carbon filament growth chamber after carbon filament growth. Without the treatment, the nickel coating was not uniform and resulted in composites orders of magnitude higher in electrical resistivity compared to the case with the treatment, see Xiaoping Shui and D.D.L. Chung, in Proc. Int. SAMPE Electronic Materials and Processes Conference, 7, Electronic Materials and

Processes Revolution for a Peace Time World, 1994, p. 39-42, for the case without the treatment.

This means that the surface contaminants affected the quality of the nickel coating, such that the electrical conduction path through the nickel coating became less continuous.

The electroplating was conducted by using a nickel anode 10, as shown in FIG. 1 and a nickel sulfate electrolyte solution 12, such that a mixture 14 of carbon filaments and air bubbles was allowed

to float in the electrolyte 12 with a cathode 16 (a copper mesh) above the mixture in the electrolyte 12. The incorporation of air bubbles was achieved by prior mixing in a rotary blade blender. The bubble incorporation was necessary in order for the filaments to float. Due to the contact between the top of the filament-bubble mixture and the cathode and the contact between adjacent filaments in the mixture, all the filaments in the mixture were electroplated with nickel.

The nickel in the nickel filaments was crystalline, as shown by x-ray diffraction using CuKα radiation. Because the carbon filaments were not straight, the resulting nickel filaments were also not straight, as shown in FIG. 5 for both the acetone-washed carbon filaments and the resulting nickel filaments. Due to the bent morphology and large aspect ratio, determination of the exact length of the carbon or nickel filaments was not possible. Nevertheless, a lower limit of the length was found by scanning electron microscopy (SEM) to be 100 μm. The submicron nature of either nickel or carbon filaments did not allow measurement ofthe electrical resistivity of single filaments.

The polymer used for all composites was polyether sulfone (PES), a thermoplastic provided as Victrex PES 4100P by ICI. Its properties are shown in Table I. The nickel fibers used for comparing with the nickel filaments were (i) 2 μm in diameter and 2000 μm in length, as provided by

Ribtec (Gahanna, Ohio, U.S. A.), and (ii) 20 μm in diameter and 1000 μm in length, as provided as Fibrex by National-Standard Co. (Corbin, Kentucky, U.S.A.). Due to the large length ofthe nickel fibers of diameter 2 μm (which resemble cotton wool), the dispersion of these fibers were most difficult. The carbon filaments used for comparing with the nickel filaments were 0.1 μm in diameter and > 100 μm in length, as provided as Applied Sciences, Inc. (Cedarville, Ohio, U.S.A.).

The composites were fabricated by forming a dry mixture of the polymer powder and the filler and subsequent hot pressing in a steel mold at 310°C (processing temperature for PES, as recommended by ICI) and 13.4 MPa for ~ 30 min. For the nickel filaments and the 20 μm diameter Ni fibers, the mixing was carried out dry in a ball mill. For the carbon filaments, mixing was carried out wet — with water in a blender, and then the wet mix was dried at 120°C. For the 2 μm diameter

Ni fibers, mixing was performed by hand, as neither the abovementioned dry mixing nor wet mixing was possible.

The volume electrical resistivity ofthe composites was measured by the four-probe method, which is a standard method of electrical resistivity measurement. This method used four probes, i.e., four electrical contacts. The four probes were around the perimeter of the composite sample (a rectangular bar) such that the probes were in four planes perpendicular to the direction of resistivity measurement (i.e., direction ofthe applied electrical current). The outer two probes were for passing

the current; the inner two probes were for voltage measurement. This voltage divided by this current gave the electrical resistance R of the part of the sample between the inner probes. The electrical resistivity p is given by the equation

_f . _Λ ά , (1)

where A is the cross-sectional area perpendicular to the direction of resistivity measurement and t is the length between the inner probes along the direction of resistivity measurement. Silver paint was used as electrical contacts. FIG. 6 and Table II give the resistivity vs. filler volume fraction for various PES-matrix composites. The Ni filaments were slightly less effective than the Ni fibers of diameter 2 μm but much more effective than the Ni fibers of diameter 20 μm in providing low resistivity. T at the Ni filaments of diameter 0.4 μm were slightly less effective than the Ni fibers of diameter 2 μm is because the cross-sectional area ofthe Ni filaments (FIG.5) is not as uniform as that of the Ni fibers (FIG. 7) and probably also because the resistivity ofthe Ni filaments is higher than that ofthe Ni fibers. That the Ni filaments were much more effective than the Ni fibers of diameter 20 μm (FIG. 5 and FIG. 10) is because of their large aspect ratio (250 for Ni filaments and 50 for Ni fibers). The difference in effectiveness for providing composites of low resistivity between the Ni filaments and the Ni fibers of diameter 20 μm decreased with increasing filler volume fraction. At all filler volume fractions, the carbon filaments, though smaller in diameter and more uniform in cross- sectional area (FIG. 3), were less effective than the Ni filaments because of their high resistivity. The EMI shielding effectiveness was measured using the coaxial cable method. The sample 11, was shown in FIG. 11 , was in the form of an annular ring of outer diameter 3.8 in (97 mm) and inner diameter 1.25 in (32 mm). The sample thickness was 2.85 mm for all the composites, 3.1 mm for solid copper and 4.0 mm for solid stainless steel. The sample was placed in Elgal SET 19A shielding effectiveness tester 13, which was connected using coaxial cables 15 to Hewlett-Packard 8510A network analyzer 17. The frequency was scanned from 1 to 2 GHz such that 101 shielding effectiveness data points were taken within this frequency range. The average of these 101 data for each sample is shown in Table III, together with the standard deviation of these 101 data in parentheses. FIG. 12-15 give all the data for each composite sample. The error of each data point was better than ± 1 dB at < 10 dB, and ±5 dB at > 70 dB; the error increased with increasing attenuation (dB).

Table III shows that, at the same filler volume fraction, the shielding effectiveness was highest for the Ni filaments. At 7 vol.%, the advantage ofthe Ni filaments compared to the other filaments was most significant. The shielding effectiveness attained by the Ni filaments at 7-19 vol.% was comparable to those of copper and stainless steel, which were thicker than the composites. Even at 43 vol.%, the 20 μm diameter Ni fibers gave lower shielding effectiveness than the Ni filaments at 7 vol.%. Even at 19 vol.%, the 2 μm diameter Ni fibers and the carbon filaments gave lower shielding effectiveness than the Ni filaments at 7 vol.%. The high shielding effectiveness associated with the Ni filaments is attributed to the combination of small diameter (smaller than the Ni fibers) and low resistivity (lower than all except the 2 μm diameter Ni fibers) ofthe Ni filaments. The shielding effectiveness ofthe composite with 13 vol .% carbon filaments is higher than that of the composite with ~ 45 vol.% carbon filaments reported by Katsumata et al. (1993). This difference is attributed to the difference in composite processing and to the difference in the carbon filaments used.

The surface impedance and the EMI shielding effectiveness at 1 -2 GHz were measured in the same way, except that the former was obtained by reflection of the radiation off the sample, whereas the latter was obtained by transmission ofthe radiation through the sample, as illustrated in FIG. 16, where

S _n - ^b (2) ^β ι

is the reflection coefficient (related to the sample impedance Z.), and

(3)

is the transmission coefficient (related to the shielding effectiveness). The machine characteristic impedance Z ₀ is 50 Ω. Since

S - ^z ~ ^~z ° - ^ZtjlZ " ^" * (4)

^{11 "} Z _L +Z _C ^' ZJZ. + l '

Thus,

Z _ l +g„ _ l * |S _n |g» (5) z. ^" ι-s„ ^" 1-|*„|**

a l + |J ₁₁ | ² +2|-? _u | coβφ (6)

An HP8510A network analyzer and an HP computer (with HP Basic Operation System) were used to obtain both | S _n | and φ at 1-2 GHz. An HP APC-7 calibration kit was used to calibrate the system. Then, | Z^ | was obtained from Eq. (6). The frequency was scanned from 1 to 2 GHz such that 101 data points were taken within this frequency range. Table IV shows the EMI shielding effectiveness and surface impedance, both averaged over the 101 data points in the frequency range 1 -2 GHz, for the PES-matrix composites with various fillers (Ni filaments, carbon filaments, 2 μm diameter Ni fibers and 20 μm Ni fibers) at various volume fractions and for solid copper, nickel and stainless steel. Shown in parentheses are the standard deviation ofthe 101 data. The shielding effectiveness results are the same as those in Table III. The Ni filaments gave composites of very high shielding effectiveness and very low surface impedance even at a low filler volume fraction . For example, at 7 vol.%, the Ni filaments gave a shielding effectiveness of 87 dB and an impedance of 1.2 Q; both properties are close to those of solid copper. By increasing the volume fraction to 13%, the impedance was even less than that of solid copper. For the carbon filaments, even at 19 vol.%, the impedance was above that of solid copper. For the 2 μm diameter Ni fibers, the volume fraction needed to be 19% in order for the impedance to be below that of solid copper. For the 20 μm diameter Ni fibers, even at 43 vol.%, the impedance was above that of solid copper. Hence, both the impedance and shielding effectiveness results are consistent in indicating the exceptional ability of the Ni filaments.

Due to aerospace applications related to shielding or waveguides, the specific EMI shielding effectiveness (EMI shielding effectiveness divided by the density) and the specific surface conductance

(surface conductance divided by the density, where conductance is the reciprocal of impedance) are relevant quantities, which are listed in Table V for the composite at the filler volume fraction that gives the highest such quantities in each ofthe four categories in Table IV. Both specific quantities are highest for both the composite with 19 vol.% carbon filaments and that with 7 vol.% nickel filaments. Compared to all composites of Table V, solid copper gave the lowest values for both specific quantities.

Although carbon filaments provide composites with the same specific EMI shielding effectiveness and the same specific surface conductance as the nickel filaments, they provide composites with lower shielding effectiveness and higher surface impedance than nickel filaments at the same filler volume fraction. While nickel is the preferred metal in Example I, any other suitable metal could be used.

Examples of other suitable metals are copper, cobalt, silver, gold, tin or zinc, or alloys thereof.

While Example I has considered the use of carbon filaments as the core of metal-coated filaments, other filaments such as ceramic filaments can be substituted therefor.

While Example I has considered forming the metal coating on the filaments by electroplating, other processes such as electroless plating, chemical vapor deposition and physical vapor deposition can be substituted therefor.

While Example I has considered the use of acetone as a solvent to remove the contaminant in the form of mainly polyaromatic hydrocarbons from the surface of carbon filaments, other solvents such as methylene chloride, toluene and carbon disulfide may be substituted therefor. While Example I has considered washing with a solvent for removing the contaminant in the form of mainly polyaromatic hydrocarbons from the surface of carbon filaments, other hydrocarbon removal methods, such as heating, may be substituted therefor.

While Example I has considered composite materials with polyether sulfone as the polymer matrix, other polymers (such as silicone, polyimide, polyvinylidene fluoride and epoxy), ceramics (such as Portland cement, silicates, phosphates, silicon carbide, silicon nitride, aluminum nitride and titanium diboride) or polymer-ceramic combinations (such as latex-cement combinations) may be substituted for polyether sulfone.

11

TABLE I Properties of polyether sulfone polymer

T _g 220 - 222° C

Density 1.37 g/cm ³ Particle size 100 - 150 μm

Tensile strength 45.93 ± 1.12 MPa

Tensile modulus 2.64 ± 0.19 GPa

Elongation at break (3.1 ± 0.3)%

Electrical resistivity > 10 ¹⁰ Q.cm Coefficient of thermal expansion 55 x lO^ K

SUBSTITUTE SHEET (fcULE 26)

TABLE II Electrical resistivity of composites with different fillers

Resistivity of composites (Q.cm)

Filler vol.% Ni filaments 2 μm Ni fibers 20 μm Ni fibers C filaments

3 5.00 x 10 ^" ' 2.40 x 10 ^"2 2.20 x 10 ⁷ 1.16 x lO ²

7 2.20 x 10 ^'2 2.50 x 10 ^"3 2.20 x 10 ³ 3.78 x 10°

13 3.50 x 10 ^"3 6.90 x 10 ^"4 2.20 x 10° 3.68 x 10 ^" '

19 1.50 x lO ^*3 2.90 x 10^ 5.30 x 10 ^'2 1.16 x 10 ^" '

25 9.58 x 10 ^"4 2.10 x 10" 5.00 x 10 ^"3 6.22 x 10 ^'2

37 4.16 x 10^ 1.35 x 10^ 6.60 x 10 ^"4 3.57 x 10 ^"2

43 2.81 x 10"* 1.06 x Iff ⁴ 2.80 x 10 ^"4 /

TABLE III

EMI shielding effectiveness (dB), averaged in the range 1-2 GHz.

The standard deviations are shown in parentheses.

3 vol.% 7 vol.% 13 vol.% 19 vol.% 25 vol.% 37 vol.% 43 vol.%

Ni filaments 42.2(2.4) 86.6(5.1) 83.7(5.3) 91.7(6.6)

C filaments 20.6(1.3) 31.8(1.7) 53.6(3.5) 73.9(5.1)

2 μm Ni fibers 45.2(2.5) 58.1(4.2) 60.3(3.2) 71.7(4.6)

20 μm Ni fibers 4.9(1.9) 10.5(2.3) 38.4(1.9) 73.7(4.4)

Copper 90.2(5.0) Stainless steel 88.9(4.0)

Table IV EMI shielding effectiveness and surface impedance at 1-2 GHz. Standard deviations are shown in parentheses.

Material Filler EMI shielding Surface vol.% effectiveness (dB) impedance (Q)

Ni filaments PES composites 3 42.2(2.4) 2.65(0.21)

7 86.6(5.1) 1.22(0.13)

13 83.7(5.3) 1.09(0.15)

19 91.7(6.6) 1.10(0.15)

Carbon filaments PES composites 3 20.6(1.3) 4.11(0.78)

7 31.8(1.7) 2.78(0.22)

13 53.6(3.5) 1.96(0.18)

19 73.9(5.1) 1.48(0.15)

2 μm Ni fibers/PES composites 3 45.2(2.5) 1.76(0.20)

7 58.1(4.2) 1.36(0.18)

13 60.3(3.2) 1.29(0.17) I

19 71.7(4.6) 1.14(0.15) 1

20 μm Ni fibers/PES composites 19 4.9(1.9) 17.0(5.2)

25 10.5(2.3) 7.93(2.41)

37 38.4(1.9) 1.98(0.39)

43 73.7(4.4) 1.40(0.25)

Copper / 90.2(5.0) 1.20(0.13)

I Nickel / 74.4(5.5) 1.21(0.22)

1 Stainless steel / 88.9(4.0) 1.24(0.22)

Table V. Properties of the composite with the best overall (electrical and mechanical) performance for each filler type.

EMI shielding Surface Strength Ductility effectiveness impedance (MPa) (%)

(dB) (Q)

7 vol.% nickel filaments 86.6 1.22 51.5 1.01

19 vol.% carbon filaments 73.9 1.48 27.2 0.67

19 vol.% 2 μm nickel fibers 71.7 1.14 66.1 0.58

43 vol.% 20 μm nickel fibers 73.7 1.40 34.6 0.3