Vol.3, No.11, 936-947 (2011) Natural Science
http://dx.doi.org/10.4236/ns.2011.311120
Fabrication and characterization copper/diamond
composites for heat sink application using powder
metallurgy
Zeinab Abdel Hamid1*, Sayed F. Moustafa1, Fatma A. Morsy2, Nevien Abdel Atty Khalifa2,
Fatma Abdel Mouez1
1Central Metallurgical R & D Institute, CMRDI, Cairo, Egypt; *Corresponding Author:
2Faculty of Science, Chemistry department, Helwan University, Helwan, Egypt.
Received 27 July 2011; revised 30 August 2011; accepted 10 September 2011.
ABSTRACT
Copper composites reinforced with diamond par-
ticles were fabricated by the powder metallur-
gical technique. Copper matrix and diamond
powders were mixed mechanically, cold com-
pacted at 100 bar then sintered at 900˚C. The
prepared powders and sintered copper/diamond
composites were investigated using X-ray dif-
fraction (XRD) and scanning electron micro-
scope equipped with an energy dispersive X-ray
analysis (SEM/EDS). The effect of diamond con-
tents in the Cu/diamond composite on the dif-
ferent properties of the composite was studied.
On fracture surfaces of the Cu/uncoated dia-
mond composites, it was found that there is a
very weak bonding between diamonds and pure
copper matrix. In order to improve the bonding
strength between copper and the reinforcement,
diamond particles were electroless coated with
NiWB alloy. The results show that coated dia-
mond particles distribute uniformly in copper
composite and the interface between diamond
particles and Cu matrix is clear and well bonded
due to the formation of a thin layer from WB2,
Ni3B, and BC2 between Cu and diamond inter-
faces. The properties of the composites materi-
als using coated powder, such as hardness,
transverse rupture strength, thermal conductiv-
ity, and coefficient of thermal expansion (CTE)
were exhibit greater values than that of the
composites using uncoated diamond powder.
Additionally, the results reveals that the maxi-
mum diamond incorporation was attained at 20
Vf%. Actually, Cu/20 Vf% coated diamond com-
posite yields a high thermal conductivity of 430
W/mK along with a low coefficient of thermal
expansion (CTE) 6 × 10–6/K.
Keywords: Powder Metallurgy; Ceramic-Matrix
Composites (CMCs); Ceramics; Coating;
Electroless
1. INTRODUCTION
In order to dissipate the heat generated in electronic
packages effectively, suitable materials must be selected
as heat spreaders and heat sink [1-6]. The ideal material
working as heat sink and heat spreader should have a
low coefficient of thermal expansion (CTE) and a high
thermal conductivity. Copper as it has the highest ther-
mal conductivities in metal, has been used for decades as
the material of choice as heat sink for semiconductor
electronic packages. Unfortunately, copper has become a
bottleneck in removing heat from semiconductor devices,
and its thermal conductivity becomes not enough to dis-
sipate the heat generated from the new generation of
semiconductor components. In addition to bulk thermal
conductivity which must be removed, thermal expansion
coefficients of semiconductors of silicon or gallium ar-
senide are low compared to other materials and much
smaller than that of copper which is usually used as a
heat sink for semiconductors. The differences in coeffi-
cient of thermal expansion properties leads to formation
of stresses, which can lead to the component becoming
distorted, detached or even fractured. Stresses may form
during production of the semiconductor chip, specifi-
cally during the cooling phase from the soldering tem-
perature to room temperature. However, temperature
fluctuations also occur when the package is operating,
ranging from –50˚C to 200˚C, which can lead to thermo-
mechanical stresses. As copper and other metals exhibit
thermal expansion rates that are an order of magnitude
greater than those of silicon and gallium arsenide, it is
problematic to attach these metals to semiconductor
chips. In fact, many packaging solutions sacrifice on

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947
937937
thermal conductivity, choosing less efficient heat con-
ductors such as ceramics in order to address this issue
[6].
There is a great demand for a new material to be used
as heat sink substrate. This new material should have
high thermal conductivity greater than that of copper (i.e.
as great as possible) in order to dissipate the high heat
generated from the semiconductor component during
operation, and in the same time its coefficient of thermal
expansion must match as closely as possible to that of
the semiconductor component. This new material could
be developed by means of composite materials such as
Cu/diamond composite. Since diamond is characterized
by high thermal conductivity about 6 times that of cop-
per but its disadvantage that its thermal expansion is
smaller than that of semiconductors. So by adding cop-
per to diamond we can have the required characteristics
by controlling the percentages of copper to diamond
[7-9]. The major problems encountered with “diamond
on copper” synthesis are the low nucleation density, film
cracking, and poor wetability. When diamond particles
are embedded in a copper matrix, the interface plays a
crucial role in determining the thermal conductivity, the
CTE and also the mechanical properties of the composite.
An ideal interface should provide good adhesion and
minimum thermal boundary resistance.
It is well known that alloying of copper with a strong
carbide diamond forming element promotes wetting and
bonding of diamond. Even in the case of solid phase
bonding (e.g. hot pressing), high bonding strength was
observed for copper alloys with minor additions of car-
bides from (Ti, Cr, B or Zr) [10,11]. Chromium or tung-
sten is believed to be a good promoter due to its abilities
to inhibit bulk copper formation, to improve copper ther-
mal stability and to increase copper dispersion. The main
objective of this investigation is to synthesize and char-
acterize Cu/diamond composite materials for electronic
application obtained by powder metallurgy technique.
Coating diamond particles using electroless technique
added to the copper matrix to fabricate Cu/diamond
composite has been investigated. Finally, the properties
of Cu/coated diamond composites were compared with
the same materials containing uncoated diamond.
2. EXPERIMENTAL PROCUDURES
2.1. General
In the present work, the fabrication of Cu/diamond
composites using powder metallurgy technique was in-
vestigated in detail. The investigated diamond powders
of micron grain sizes (20 - 40 µm) type RVD were sup-
plied by Polaris Diamond Powder Co., Ltd. The dia-
mond powder was electroless coated with NiWB. Cop-
per powder using in this study has been fabricated using
electroless technique.
2.2. Surface Treatment for Diamond and
Electroless NiWB Plating
As the diamond surface has higher interface energy,
the adhesion of surface with the coating layer is bad, and
it makes the diamond shatter easily. For solving this, the
diamond surface must be treated and activated to im-
prove the coating adhesion. The following steps were
used for surface treatment.
Washing by acetone (30 min.); etching by acidic solu-
tion using 50% HNO3; sensitization using stannous
chloride solution; activation using palladium chloride
solution; then immersed in the plating solution [12].
The diamond after sensitization and activation were
chemically coated with NiWB. The chemical composi-
tion and the operating conditions of the plating solution
were illustrated in Table 1. Additionally, the chemical
composition and the operating conditions for producing
the copper powder using electroless technique were il-
lustrated in Table 2. Analytical reagents and distilled
water were used to prepare the plating solutions.
Table 1. Chemical composition and operating conditions of
electroless NiWB plating solution.
Composition Concentration, gl–1
NiWB
Nickel sulphate 10
sodium citrate 20
Dimethyl amine borane (DMAB) 0.4
sodium Tungestate 20
Operating conditions
Temperature, ˚C 85
pH 6.5
Time, hrs 2
Table 2. The chemical composition and the operating condi-
tions of electroless copper solution.
Composition Concentration gl–1
Copper sulphate 35
Sodium potassium tartrate 170
Sodium hydroxxide 50
Formaline 200 mlL–1
Operating conditions
pH 11 - 13
Temperature, ˚C 25

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947 938
2.3. Composites Production
The amount of powder and coating thickness must be
controlled during the production of composite. If a too
large amount of diamond or a too thick coating for the
functional layer is used the thermal conductivity of the
copper matrix material will be reduced very severe. The
copper powders were mixed with the diamond particles
to prepare composites then milled for 20 min, compacted
under the pressure 100 bar then sintered at 900˚C for 1
hr using hydrogen atmosphere.
2.4. Analysis and Characterizations
Scanning electron microscope (SEM) with a link en-
ergy dispersive X-ray spectroscopy (EDS) detector at-
tachment, model JEOL, JSM-5410, were used to assess
the surface morphology, particle size, particle shape,
agglomeration of particles and the compositions of the
various Cu/diamond composites. For SEM, the dry
powder was dusted onto a carbon tape, which was stuck
onto the copper holder disk of the microscope, gently
blown with compressed air before introducing into the
microscope chamber.
The phase identification was determined using X-ray
diffraction PANalytical X’pert PRO (45 kv/40 mA) ad-
vanced with Cu target (λ = 1.54 A0). The boron content
of the NiWB thin film deposited was determined by in-
ductively coupled plasma-mass spectrometer (ICP-MS)
after dissolving the deposited layer in nitric acid solu-
tion.
2.5. The Densities Measurements
The densities of the sintered powders were measured
according to MPIF Standards 42, 1998, using Archimedes
rule. The density (δ) of the samples was calculated ac-
cording to the following formula [13,14]:

aaw
WWW   gm/cm3 (1)
where Wa and Ww are the weights in air and water re-
spectively.
2.6. The Electrical Resistivity
Measurements
The electrical resistivity of the sintered materials was
measured using the four-probe method by using Omega
CL 8400 micrometer device. The rectangular sintered
specimen was placed in a specially designed jig for mak-
ing the electrical connection. Measurements were taken
for the longitudinal and the transverse directions of the
sample. The resistivity (
 ) was calculated according to
the following equation by using High Precision Micro-
ohmmeter.


RA L   (2)
where, R is the resistance in micro ohm, L is the meas-
ured length in cm, A is the cross section area in cm2, and
ρ is the resistivity in µ··cm.
2.7. The Macrohardness Measurements
The macrohardness values of the investigated materi-
als were measured as the average of 5 readings over the
surface of the specimens using Vicker’s macrohardness
Test type HPV 30 A and the using load was 5 kg for 15 s.
2.8. Transverse Rupture Strength (TRS)
Measurements
The rupture test was performed using compression
testing machine and a test fixture, according to MPIF
Standard 41. All the samples of the powders were com-
pacted into rectangular compacts of dimension of (30 ×
10 × 6) mm3. In the transverse rupture fixture, the test
bar was placed centrally located and perpendicular to the
supporting rods with the top up. The fixture with the test
bar was placed between the plates of the compression
testing machine and the load was applied at constant rate
of 2.5 mm/min, until the test bar fractured. The trans-
verse rupture strength for sintered samples was calcu-
lated according to the following expression [13]:
2
3
2
PL
TRS tW  (3)
where: TRS = Transverse rupture strength in N/mm2
P = Fracturing (rupture) load (N)
L = the distance between the supporting rods (25.4
mm)
t = the thickness of the sample in (mm)
W = the width of the sample in (mm)
2.9. Coefficient of Thermal Expansion (CTE)
CTE was measured with a Netzsch dilatometer model
402 C push rod. The samples were tested in open air
between 25˚C and 160˚C at a heating rate of 3 K/min.
Further on, the reproducibility was checked by measur-
ing the same sample twice. The system was calibrated
with a copper standard prior to the sample test run. Co-
efficient of thermal expansion (CTE) value at a given
temperature was calculated from the thermal expansion
curves (change in length versus temperature) recorded
during heating up the specimen from room temperature
up to 160˚C, according to this relationship:
L
CTE TL





(4)
where L represents the original length of the specimen

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947
939939
and L stands for the length change after a thermal cy-
cling.
2.10. Thermal Conductivity
The thermal conductivity was calculated from electri-
cal resistivity measurement using Wiedemann and Franz
equation which derived a relation equation between the
thermal and electrical conductivities [15]. The Wiede-
mann-Franz equation is as follows:
22
82
2
π2.443 10JΩKs
3
B
kL
Te



 (5)
where,  is thermal conductivity (W/mK),  is electrical
conductivity μ·Ω·cm–1, T is absolute temperature in de-
gree Kelvin, KB is Boltzmann constant, and L is Lorentz
number.
3. RESULTS AND DISCUSSION
3.1. Surface Treatment of Diamond
Powders Surface Cleaning and Etching
The coating process of diamond powders is highly
dependent on the pre-treatment of the powders them-
selves. The first pretreatment step is to remove any for-
eign matter may be existed from the reinforcement
powders. Acetone was used to remove any organic mat-
ter, such as grease or oil, and finally used 50% HNO3 for
further cleaning the powder surfaces and also to etch
their surfaces. This pretreatment increases the ability of
diamond surface to be coated, increases the homogeneity
of the coating film and improves the adhesion of coating
layer. Although the coating bath is the most visible and
complex, any process problem is almost not due to the
chemistry of the bath itself, but is more likely due to the
pre-treatment of the surface of the substrate.
Activation (Metallization) of Diamond Surfaces
The main purpose of the metallization process for
diamond powders is to etch and provide active sites that
can react with the activators as well as to convert the
hydrophobic to hydrophilic surface. Adding sensitized
powders to the activation bath, the tin ions that are ad-
sorbed on the diamond surfaces, reduce the palladium
ions in the activation step according to the following
equation:
Sn+2 + Pd+2  Sn+4 + Pd0 (1)
This step results in imparting a uniform surface film
to the diamond powders, which ensures uniform adsorp-
tion of subsequent coating and therefore promotes better
coating [16].
3.2. NiWB Electroless Plating
The coatings are deposited on the surface of diamond
found in the solution once its pH value becomes 6.5 and
its temperature 85˚C. NiWB coating process used in this
investigation is a chemical reduction process by di-
methyl amine borane (DMAB) which served as a reduc-
ing agent to supply electrons for the reduction process.
The diamond powders coated with NiWB are investi-
gated by scanning electron microscope and the X-ray
diffraction. The scanning electron micrograph of the
diamond coated is shown in Figure 1. Continuous coat-
ing is seen on the surfaces of diamond. Figure 2 illus-
trated the chemical composition and the obtained aver-
age data of the coated layer using EDS. It was interest-
ing to note that the nickel content of the deposit is higher
than the tungsten content, suggesting an anomalous be-
havior that involves a preferential deposition of electro-
chemically less noble in the deposit. Saito et al. [17]
have also observed this anomaly and suggested that the
low overvoltage for nickel deposition is the reason for its
90 μm40 μm
× 750 × 1500
Figure 1. SEM image for the diamond coated with NiWB film.

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947 940
Figure 2. EDS analysis of the diamond coated with electroless NiWB.
preferential deposition in the electroless deposit. The B
content in the film was determined by inductively coupled
plasma-mass spectrometer (ICP-MS). The analysis proved
that the B content in the NiWB film was 5 wt%. View
Within Article
XRD pattern for uncoated and coated diamond before
and after heat treatment is illustrated in Figure 3. Figure
3(a) shows the peaks at 2θ of 440 and 750 which at-
tributed to the pure diamond powder, while XRD pattern
for diamond powder coated with NiWB as a deposit (Fig-
ure 3(b)) is seen the presence of sharp peak represent
the presence of W at 35˚ and peaks represent the pres-
ence of diamond at 75˚ and broad peak with low inten-
sity at 45˚ represent the deposition of Ni alloy in amor-
phous structure. When the coated powders after heat
treatment under reducing atmosphere at 900˚C for 1 hr
were examined by XRD in order to find out any reaction,
the pattern reveals that the amorphous coated converted
to crystalline one. Crystallization induced by heating
was confirmed by the appearance of diffraction lines
corresponding to WB2 at 2θ of 37˚, BC2 at 75.5˚, Ni3B at
45˚ and C at 41.5, 51, 54.5 and 60˚ as those indicated in
Figure 3(c).
The mechanism of electroless NiB has been studied by
many researchers [18,19]. Nevertheless, the mechanism
of formation of NiWB remains unclear. In traditional
NiB solution, the acid hydrolysis of DMAB occurs ac-
cording to the following equations [17]:
R2NHBH3  R2NH + BH3 + H2 + H +
 R2NH
+ + B + 5/2 H 2)
(CH2)3 NHBH3 + 3H2O + H+
 (CH2)3 NHBH2+ + H3BO3 + 3H (3)
Most authors believe that the major species supplying
electrons for metal-ion reduction is BH3OH–. The hy-
drolysis investigation of DMAB shows that hydrolysis is
pronounced at pH  5. So, a significant amount of
DMAB is wasted by the hydrolysis, and consequently
the electroless deposition in this range should be avoided
[20,21]. In the pH region above 5, the consumption of
DMAB by hydrolysis approaches a minimum. Mallory
proved that the rate of Ni deposition increases with an
increase in DMAB for all investigated pH values within
the range 6 to 11 [20]. However, it should be noted that
an increase in pH within this range leads to a decrease in
the rate of Ni deposition. This can be attributed to the
increase in the solution stability (probably because of the
tendency to hydrolyze at very high pH). Under these
conditions, the reduction reaction may start in the bulk
solution and the rate of deposition decreases. Conse-
quently, the deposition efficiency decreases. Additionally,
Mallory suggested that the preferred operating pH range
for Ni deposition with DMAB is 6 to 7 (near neutral)
[21]. Generally, DMAB has three active hydrogen atoms
bonded to the boron, and theoretically should reduce
three metal ions (such as Ni) for each ion of BH3OH–.
BH3OH– + 3Ni2+ + 2H2O  3Ni+ B(OH)3 +5H+ (4)
The boron reduction can be represented by the follow-
ing reaction:
(CH2)3 NH + 2BH3 + H+  (CH2)3 NH2+ + 2B + 3H2 (5)
The tungsten deposition mechanism itself is not clear
yet. There are some proposed models that explain the
fact that tungsten cannot be deposited either electro-
chemically or chemically by itself, but it can be deposited

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947
941941
Figure 3. X-ray diffraction for diamond uncoated and coated with NiWB where, (a) diamond
pure, (b) diamond coated before heat treated and (c) diamond coated after heat treatment at
900˚C.
with iron group metals [22]. This mechanism should
take into consideration the pH and that the metals are
complexes with the citrate. In the simplest case, this ion
should have the formula [(WO4) (cit)]x−5 but this is,
however, unstable. Several protonated forms of this
complex are known, such as [(WO4) (cit) (H)x]x−5, where
x can assume values of 1 - 3. The reduction of these ions
can be presented as:
WO2+ + 2 H2O + 6 e–  W + 4 OH– (6)
3.3. Composites Fabrication
Consolidation of metallic powders is mainly carried
out using compaction followed by sintering. Sintering is

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947 942
a thermal process which increases the strength of a
powder mass by bonding adjacent particles via diffusion
or related atomic level events. As a result of this opera-
tion, the material acquires the required physical and
mechanical properties. Most of the properties of a pow-
der compact are improved with sintering. All the pre-
pared composites were investigated by the SEM, and
exposed to different physical and mechanical tests.
In order to make clear the difference in wetability of
un-coated and coated diamond with Cu matrix, the sur-
face morphologies of the sintered materials with differ-
ent volume fraction diamond mixed with copper were
observed by SEM. Figures 4(a)-(f) illustrate the disper-
sion of uncoated or coated diamond particles in the cop-
per matrix. One can notice that the distribution of un-
coated diamond (Figures 4(a)-(c)) or coated diamond
(Figures 4(d)-(f)) increases with increasing diamond Vf%
in the composite. The Cu/coated diamond composite has
low porosity than Cu/un-coated diamond composite.
Additionally, the distribution of diamond in the compos-
ite made from Cu/coated diamond powders is more uni-
form with good wetability than those of the composite
made from Cu/uncoated diamond powders. Since the
uncoated diamond particles are easy to be stripped off
during mechanical polishing, small pits are left on the
surface of samples. While, the composite made from
Cu/coated diamond powders has low porosity content
due to less Cu-C contacts during sintering, and the me-
tallic binders (coated film) decrease all the existing cavi-
ties between Cu particles. The uniform distribution and a
maximum value of the diamond in the composite were
observed at 20% Vf.
3.4. Characterizations of the Fabricated
Composites
3.4.1. Densities Calculation
The density of the copper/diamond composite is one
of the most important properties due to its effect on the
other properties. It is very sensitive to composition and
porosity in the sample and is widely used as a quality
control test. The densities and relative densities of the
sintered copper/diamond composite materials with dif-
ferent diamond Vf% are reported in Table 3. The rela-
tive density of the sintered materials was calculated ac-
cording to the following formula:
Relative density

%%Rds t   (7)
where δs, and δt are the densities of sintered density
(actual density) and theoretical density respectively.
The data reveals that the density of composite materi-
als containing coated diamond with NiWB has higher
value than that composite containing uncoated diamond
Table 3. The densities and relative densities percentage of the
sintered materials with different Vf% of diamond.
Investigated composites  s  t Rd%
Cu/uncoated diamond with
different Vf%
10 5.78 8.33 69.46
15 5.86 8.045 72.25
20 6.30 7.76 81.18
30 6.66 7.19 60.00
Cu/coated diamond with NiWB
with different Vf%
10 6.27 8.33 88.6
15 6.08 8.04 82.29
20 5.50 7.76 99.7
30 4.50 7.19 92.6
particle. This is can be attributed to the excellent
wetability between Cu matrix and coated diamond than
in case of composite containing uncoated diamond due
to high solubility of Ni in Cu. Coated film is used as a
bonding matrix because its wetting or capillary action
during solid phase sintering allows the achievement of
high densities; this improvement could be attributed to
the ability of B and W to form carbide phases as shown
in XRD analysis (Figure 3). XRD pattern can’t prove
the formation of tungsten carbide phase owing to low
concentration of W compared to the concentration of Ni
in the coated layer or carbon of the substrate. This re-
sults agreement with Q. Sun [9] who proved that pure
liquid copper does not wet diamond, while well in pres-
ence of carbide which promotes wetting and bonding of
diamond.
3.4.2. Hardness Measurements
Hardness measurements were used to investigate the
strengthening effect of diamond particles as a rein-
forcement material and to distinguish the surface imper-
fection. Table 4 illustrates the hardness of the investi-
gated composite materials. As can be seen, hardness of
sintered materials made from coated diamond is higher
than those made of uncoated diamond powders. Also
hardness increases with increasing Vf% of diamond
powder in the matrix. It is evident that Cu-20 Vf%
coated diamond composites have the highest hardness
values of all investigated composites, not only because
of the strengthening effect of diamond but also due to
the high density and low porosity content. While, in the
case of the uncoated diamond, there is no any adhesion

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947
943943
Figure 4. Surface morphologies of the as-polished Cu/uncoated diamond and Cu/NiWB
coated diamond composites prepared by PM technique containing different Vf% diamond
where: (a)-(c) Cu/10, 20, 30 Vf% uncoated diamond respectively and (d)-(f) Cu-10, 20, 30
Vf% NiWB coated diamond.
or bond between the diamond and the copper matrix
which causes the formation of pores in the final compos-
ite and hence showed low hardness.
3.4.3. Electrical Conductivity Measurements
The electro friction materials that exposed to high
current densities or high voltages are produced from a
composite material that having high electrical and ther-
mal conductivities. Copper/diamond composite has a
good sliding, antifriction properties, and high electrical
and thermal conductivity [23]. The measured resistivity
was converted to electrical conductivity IACS% (Inter-
national Annealed Copper Standard) according to the
ASTM standard B 193-72.
The results of electrical resistivity measurements of
the fabricated composites are shown in Table 5. It can
be noticed that the electrical resistivity for all investi-
gated Cu/diamond composites decreases with increasing
Vf% of diamond up to 20 V% and then slightly increase.
At the same time, the electrical resistivity for the com-
posite fabricated from coated diamond with NiWB is
lower than that of composite fabricated from uncoated
diamond. This can be rationalized by the high porosity
of the Cu/uncoated diamond composite. This porosity
was resulted from the weak bond between diamond and
the Cu matrix as shown in Figure 4. This in turn, creates
voids, which decreases the mobility of the free electron
causing high resistivity [24,25]. The electrical conductiv-
ity is directly affected by porosity, the greater the void
content the lower is the electrical conductivity. While, in

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947 944
Table 4. The Vicker Macro-Hardness values of the investigated
composites.
Investigate Composites Hardness Values, HV
Pure copper 52.00
Cu/uncoated diamond
Cu/10 V% uncoated diamond 55.00
Cu/15 V% uncoated diamond 55.00
Cu/20 V% uncoated diamond 56.00
Cu/30 V% uncoated diamond 52.00
Cu/coated diamond with NiWB
Cu/10 V% coated diamond 66.35
Cu/15 V% coated diamond 77.04
Cu/20 V% coated diamond 87.45
Cu/30 V% coated diamond 74.24
Table 5. The electrical resistivity of the investigated compos-
ites.
electrical resistivity
Investigate Composites
Resistivity (µ·cm) IACS%
Pure copper chemically deposited 2.1 82.00
Cu/uncoated diamond
Cu/10 V% uncoated diamond 2.33 74.00
Cu/15 V% uncoated diamond 2.085 82.5
Cu/20 V% uncoated diamond 1.96 87.75
Cu/30 V% uncoated diamond 2.00 86.00
Cu/coated diamond with NiWB
Cu/10 V% coated diamond 1.78 97.00
Cu/15 V% coated diamond 1.74 99.00
Cu/20 V% coated diamond 1.73 99.40
Cu/30 V% coated diamond 1.75 98.30
case of Cu/coated diamond composites the pores is low
due to the presence of a metallic bond between the
coated layer on the diamond surface and the copper ma-
trix. The NiWB coated diamond composite has the
highest and more conductive sample due to the follow-
ing reasons:
 The highest density and lowest porosity of the materi-
als.
 The good adhesion force between the carbide coating
with diamond surface on one side and the copper ma-
trix on the other side.
 The presence of a carrier layer at the interface be-
tween the copper matrix and the diamond, which
transfer the high electrical conductivity properties of
the copper to the diamond.
 The high hardness of carbide layers which increase
the effect of compaction pressure on Cu-matrix on one
side and protect diamond from fragmentation on the
other side.
3.4.4. Transverse Rupture Strength (TRS)
Transverse rupture strength (TRS) is the most com-
mon method for determining the fracture strength of
composite materials; it is the maximum stress that is
encountered during a three-point bending test. In the
same time the TRS testing method is more suitable than
the tensile measurement for determining the strength of
materials processed by powder metallurgy. There are
several reasons for its popularity in practice. First of all,
TRS is very sensitive to porosity levels. When porosity
level is high, TRS values will be not only poor but also
very inconsistent. Therefore, it has historically being
used as an indicator of the quality of sintered composite
materials in manufacturing. Secondly, because of its
sensitivity to pores and other defects, TRS is often also
viewed as a measure of “toughness”.
The effect of diamond Vf% (10−30) in the Cu matrix
on the TRS was studied and the values were listed in
Table 6. According to the data obtained, the TRS in-
creases with increasing the diamond Vf% in the Cu ma-
trix and attains the optimum values at 20%.
One can notice that the TRS of the Cu/coated dia-
mond composites are much higher than those made from
Cu/uncoated diamond composites. This can be explained
by the fact that, the coated layer which consists of car-
bides and borides acts as crack propagation layer that
transfer the strength of diamond to the copper matrix and
Table 6. The TRS values for the investigated composites.
Investigate Composites TRS Values, N/mm2
Cu/uncoated diamond
Cu/10 V% uncoated diamond 210.00
Cu/15 V% uncoated diamond 238.00
Cu/20 V% uncoated diamond 250.00
Cu/30 V% uncoated diamond 270.00
Cu/coated diamond with NiWB
Cu/10 V% coated diamond 438.65
Cu/15 V% coated diamond 753.37
Cu/20 V% coated diamond 755.00
Cu/30 V% coated diamond 652.60

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947
945945
also permit a good adhesion between the diamond and
the copper matrix that gives more density and less po-
rosity, consequently more strengthened composite. But
in case of the uncoated diamond composite, there is no
real bonding between diamond and copper matrix, so the
adhesion force is very weak, which creates space bond-
ing that deteriorate the strength of the composite.
But if we view the fracture process during TRS testing
as consisting of crack initiation and propagation proc-
esses and assume the ideal case when the effects of po-
rosity is negligible, the crack initiation process will
dominate when the hardness is high and the fracture
toughness is very low. Therefore the higher the hardness
is, the higher the stress that is needed for crack initiation
and hence the higher the TRS. This means that, the TRS
of the investigated Cu/coated diamond is much greater
than the Cu/uncoated diamond. The explanation of the
result is as follows; in Cu/coated diamond the crack re-
sistance is greatest in the coating film, followed by the
Cu grains and the Cu/diamond interface. The coating
grains act to impede rapid crack growth and absorb more
energy than the uncoated diamond, thus increasing the
fracture toughness of the material.
The fracture micrographs of the investigated materials
are shown in Figure 5. Figures 5(a) and (b) indicate the
fracture surface of the Cu/uncoated diamond composite
at low and high magnification respectively. Figure 5(b)
illustrates poor bonding on (111) diamond surfaces with
Cu matrix as shown in XRD analysis. Marked A in Fig-
ure 5(b) indicates regions where the Cu matrix has ad-
hered to diamond surfaces. This adhesion between Cu
and diamond was consistently observed on (220) dia-
mond faces as described by P. W. Ruch et al. [6]. Fig-
ures 5(c) and (d) show the fracture surface of the Cu/
Figure 5. SEM micrograph of the fracture surface of Cu/uncoated and coated diamond com-
posite, where (a) Cu/uncoated diamond at low magnification, (b) at high magnification, (c)
Cu/NiWB coated diamond composite at low magnification, and (d) Cu/NiWB coated diamond
composite at high magnification.

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947 946
NiWB coated diamond composite at low and high mag-
nification respectively. Figure 5(d) illustrates the im-
proving in the adhesion force between coated diamond
surface and Cu matrix. NiWB coated diamond acts as a
filler continuous layer that decreases the pores at the
interface and so gives a degree of adhesion that increases
the strength of the final composite.
3.4.5. Thermal Conductivity and CTE of
Investigated Composites
Table 7 shows the variation of thermal conductivity
and CTE of pure Cu and different composites with op-
timum Vf% diamond (20% Vf). It is clear that thermal
conductivity increased from 378 for Cu/uncoated dia-
mond composite to 430 W/mk for Cu/coated diamond
composites. This means that coated diamond improved
the thermal conductivity of the Cu/diamond composite.
Adding uncoated diamond particles to Cu matrix caused
slight increase (378 W/mK) in the thermal conductivity
of pure Cu. The low thermal conductivity of this com-
posite was indicating a high thermal barrier resistance.
This result due to separation between copper and dia-
mond, i.e. diamond particles are surrounded by cavities,
which lead to low chemical affinity between copper and
diamond. Therefore, it is difficult to produce a bond of
low thermal resistance and high mechanical strength
between the matrix and the reinforcement. While adding
NiWB coated diamonds lead to increase the thermal
conductivity of the composites, which demonstrated the
effectiveness of the coated layer to obtain a good ther-
mal contact between the matrix and the diamond parti-
cles.
Thermal expansion coefficient of copper/diamond
composites, namely, Cu/20 Vf% uncoated diamond and
Cu/20 Vf% coated diamond was measured. It can be
seen that the CTE of Cu/coated diamond decreases to 6
× 10–6 K−1, which is about one-third of that of the pure
copper (17 × 10–6). The lower CTE of Cu/coated dia-
mond composite is related to the good bonding between
the metal matrix and the diamond particle. This indicates
that NiWB coated diamond are a promising reinforce-
ment to lower the CTE of heat sink materials.
4. CONCLUSIONS
1) Diamond powders can be coated with NiWB alloy
by electroless technique.
2) The sintered materials made from coated powders
exhibit better structure homogeneity, higher densifica-
tion, electrical resistivity, hardness and TRS properties
than those made from mixed powders.
3) This work demonstrates that high thermal conduc-
tivities can be achieved for Cu/coated diamond compos-
Table 7. Thermal conductivity and CTE of the investigated
composites.
Investigate Composites Thermal conductivity,
(W/mk) CTE, 10–6/k
Pure copper 352 17
Cu/20 V% uncoated diamond378 10
Cu/ 20 V% coated diamond430 6
ite. The formation of a carbide and boride layer is crucial
to enable the manufacturing of Cu/diamond heat sinks
with high thermal conductivities up to about 430 W/mK
combined with CTE of 6 × 10–6/K.
4) Further work address some critical characteristics
of composites, in order to promote a better coupling of
matrix and reinforcement, thus leading to improved
physical and mechanical characteristics.
REFERENCES
[1] Ellsworth, M.J. (2004) Chip power density and module
cool technology projections for the current decade. Pro-
ceedings of the 9th Intersociety Conference on Thermal and
Thermo Mechanical Phenomena in Electronic Systems,
Las Vegas, 2, 707-708.
[2] Zweben, C. (1998) Advances in composite materials for
thermal management in electronic packaging. Journal of
Management, 50, 47-51.
[3] Zweben, C. (2001) Thermal management and electronic
packaging applications. ASM Handbook, 21, 1078-1084.
[4] German, R.M., Hens, K.F. and Johnson, J.L. (1994)
Powder metallurgy processing of thermal management
materials for microelectronic applications. International
Journal of Powder Metallurgy, 30, 205-215.
[5] Evans, T.C. (1998) Thermal properties of electronic ma-
terials. Thermal Management Handbook for Electronic As-
semblies, McGraw-Hill, New York.
[6] Ruch, P.W., Beffort, O., Kleiner, S., Weber, L. and Ug-
gowitzer, P.J. (2006) Selective interfacial bonding in
Al(Si)-diamond composites and its effect on thermal
conductivity. Composites Science and Technology, 66,
2677-2685. doi:10.1016/j.compscitech.2006.03.016
[7] Clyne T.W. (2000) Thermal and electrical conduction in
MMCs. In: Kelly, A., Chou, T.W., Talreja, R., Clyne,
T.W., Warren, R., Carlsson, L., et al., Eds., Comprehensive
Composite Materials. Metal-Matrix Composites, Elsevier,
Amsterdam, 3.
[8] Weber, L. and Tavangar, R. (2007) On the Influence of
active element content on the thermal conductivity and
thermal expansion of Cu–X (X = Cr, B) diamond com-
posites. Journal of Materials Science, 57, 988-991.
[9] Schubert, T., Ciupinski, Ł., Zielinski ,W., Michalski, A.,
Weigärber, T. and Kieback, B. (2008) Interfacial charac-
terization of Cu/diamond composites prepared by powder
metallurgy for heat sink application. Scripta Materialia,
58, 263-266. doi:10.1016/j.scriptamat.2007.10.011
[10] Schubert, T.H., Ciupiński, Ł., Morgiel, J., Weidmüller, H.,
Weissgärber, T. and Kieback, B. (2007) Advanced com-

Z. A. Hamid et al. / Natural Science 3 (2011) 936-947
947947
posite materials for heat sink applications. Euro PM 2007—
PM Functional Materials, Toulouse, 15-17 October 2007.
[11] Lowenheim, F.A. (1974) Modern electroplating. John
Wiley and Sons, New York.
[12] Metal Powder Industries Federation, Princeton (1998).
[13] Fang, Z. and Eason, J.W., (1993) Nondestructive evalua-
tion of WC-Co composites using magnetic properties.
International Journal of Powder Metallurgy, 29 , 259-
265.
[14] Deborah, C.D.L. (2001) Applied materials science applica-
tion of engineering materials in structural, electronics,
thermal and other industries. Chapman and Hall, London.
[15] Moustafa, S.F., Abdel, H.Z., Osama, B.G. and Hussien, A.
Synthesis of WC hard materials using coated powders.
Advanced Powder Metall, 22, 596-601.
[16] Saito, T., Sato, E., Matsuoka, M. and Iwakura, C.J.,
(1998) Electroless deposition of Ni-B, Co-B and N-Co-B
alloys using dimethyl amineborane as a reducing agent.
Applied Electrochemistry, 28, 559-563.
doi:10.1023/A:1003233715362
[17] Osaka ,T. Takano, N., Kurokawa, T., Kaneko, T. and
Ueno, K. (2003) Characterization of chemically-depo-
sited NiB and NiWB thin films as a capping layer for
ULSI application. Surface and Coatings Technology,
169-170, 124-127. doi:10.1016/S0257-8972(03)00186-5
[18] Duhin, A., Sverdlov, Y., Feldman, Y. and Shacham-Dia-
mand, Y. (2009) Electroless deposition of NiWB alloy on
p-type Si(100) for NiSi contact metallization. Electro-
chimica Acta, 54, 6036-6041.
doi:10.1016/j.electacta.2009.01.062
[19] Mallory, G.O. (1971) The electroless nickel plating bath:
effect of variables on the process. Plating, 58, 319-322.
[20] Mallory, G.O. and Hadju, J.B. (1991) Electroless plating:
Fundamentals and applications. AESF, Orlando.
[21] Abdel, A.A., Barakat, H., Abdel, H.Z. (2008) Synthesis
and characterization of electroless deposited Co–W–P
thin films as diffusion barrier layer. Surface & Coatings
Technology, 202, 4591-4597.
doi:10.1016/j.surfcoat.2008.03.023
[22] Pearlstein, F. and Weightman, R.F. (1974) Electroless
cobalt deposition from acid baths. Journal of The Elec-
trochemical Society, 121, 1023-1028.
doi.org/10.1149/1.2401971
[23] Ellis, D.L. and McDanels, D.L. (1993) Thermal Conduc-
tivity and thermal expansion of graphite fiber-reinforced
copper matrix composites. Metallurgical and Materials
Transactions A, 24, 43-52. doi:10.1007/BF02669601
[24] Abdel, G. and El-Kad, O. (2005) Improvement of wet-
tability of carbon fiber by in-situ carbide coating. Ph.D.
Thesis, Cairo University, Cairo.
[25] Khattab, N.M.M. (2001) Ph.D. Thesis, Department of
Chemistry, Faculty of Girls, Ain Shams University,
Cairo.