www.elsevier.com/locate/apsusc

Applied Surface Science 252 (2006) 7361–7372

Corrosion resistance and lubricated sliding wear

behaviour of novel Ni–P graded alloys as an

alternative to hard Cr deposits

Liping Wang a,b, Yan Gao a, Tao Xu a,*, Qunji Xue a,*

a State Key Laboratory of Solid Lubrication, Lanzhou Institute of Chemical Physics,

Chinese Academy of Sciences, Lanzhou 730000, PR Chinab Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China

Received 8 April 2005; received in revised form 23 August 2005; accepted 23 August 2005

Available online 23 September 2005

Abstract

Alternative process to hexavalent chromium, substitute materials and new designs are urgently needed owing to the

requirement of ‘‘clean’’ manufacture. This comparative study was conducted to systematically investigate the corrosion

resistance and lubricated sliding wear behavior of graded Ni–P alloy deposits produced from a single plating bath by

electrodeposition and hard Cr deposits, using potentiodynamic polarization and reciprocating ball-on-disc tribometer. Results

showed that Ni–P deposits heat-treated at 400 8C with maximum hardness exhibited more than two orders of magnitude higher

corrosion resistance than hard Cr deposits in 10 wt.% HCl solution. The Stribeck curves for the heat-treated Ni–P gradient

deposits and hard Cr under lubrication conditions were obtained with accurate control of normal load and sliding speed during

the wear process, three main different regimes corresponding to different lubrication mechanism were identified. Heat-treated

Ni–P gradient deposits showed relatively poor wear resistance than hard Cr deposits under the lubrication conditions, which may

be attributed to superior oil-retaining surface structure and the unique ‘‘nodular’’ effect of hard Cr in wear process.

# 2005 Elsevier B.V. All rights reserved.

Keywords: Corrosion resistance; Graded Ni–P deposits; Electrodeposition; Lubrication; Hard chromium

* Corresponding authors. Tel.: +86 931 496 8169;

fax: +86 931 496 8169.

E-mail addresses: pingpingwlp@hotmail.com (T. Xu),

lpwang@lsl.ac.cn (Q. Xue).

0169-4332/$ – see front matter # 2005 Elsevier B.V. All rights reserved

doi:10.1016/j.apsusc.2005.08.040

1. Introduction

Electrodeposition as an industrial activity has been

practiced for over 150 years. Nowadays, the electro-

deposition industries are undergoing fundamental

changes due to the environmental problems. Huge

environmental pressures increasingly require that

.

L. Wang et al. / Applied Surface Science 252 (2006) 7361–73727362

certain established plating processes be substituted by

more environmental friendly technologies. The devel-

opment of ‘‘clean’’ technologies in the electroplating

industry is today an essential task required and initiated

by environmental laws of countries around the world

[1,2]. Undoubtedly, from an environmental point of

view, chromium electrodeposition, which has a wide

range of industrial applications in the automotive,

aerospace, mining and petrochemical fields [3,4], is one

of the most critical electrodeposition processes. In all

environmental regulations, chromic acid (CrO3), which

is mainly used in hard Cr plating have been recognized

as both highly toxic and carcinogenic chemicals, and

was identified by the U.S. Environmental Protection

Agency (EPA) as one of 17 ‘‘high priority’’ toxic

chemicals. Consequently, the use of hexavalent

chromates will require special waste disposal methods

and expensive breathing apparatus, and exhaust

systems must be employed to deal with emissions

during processing [5]. For these reasons, alternative

process, substitute materials and new designs have been

under study for many years. Alternatives such as

composite coatings and trivalent chromium deposits

have been investigated in recent years. Alloy electro-

deposits including Ni–W, Ni–P, Co–W and ternary or

quaternary alloys have been considered to replace the

conventional hard chromium deposits [6,7]. Unfortu-

nately, extremely limited deposits could completely

replace the conventional hard chromium owing to the

comprehensive properties Cr deposit possesses, such as

high hardness, low friction coefficient, excellent wear

and corrosion resistance.

A possible approach for the preparation of such

kind of Ni-based alloy coatings as an alternative to

hard chromium is to introduce the new concept of

functionally graded deposits (FGDs), which originally

evolved from the application of functionally graded

materials (FGMs), since the property gradient in the

FGMs is caused by a position-dependent chemical

composition, microstructure or atomic order [8–10], It

has been found in our previous research work that

functionally graded Ni–P deposits (Ni–P FGDs)

exhibited much better adhesive strength, smaller

thermal stress induced by heat treatment and high

wear resistance when compared to ungraded Ni–P

deposits. The hardness of graded Ni–P alloys after

heat treatment at 400 8C are close to or even higher

than that of conventional hard chromium and heat-

treated Ni–P gradient deposits exhibited better wear

resistance than hard Cr deposit both at dry sliding wear

and high temperature wear conditions [11]. Conse-

quently, gradient design of alloy composition inside

the deposits could solve the classic hard chromium

problems to adapt the properties of the coatings to

special demands. In some cases hard Cr deposits was

often used in oil-lubricated wear conditions and even

provide excellent protection against corrosion in

petrochemical fields. Therefore, further investigations

on the corrosion resistance of newly developed Ni–P

gradient deposits and the tribological behaviour under

the oil-lubricated conditions are needed.

The aim of present work is to systemically

investigate the corrosion resistance and oil-lubricated

wear behavior of the heat-treated Ni–P gradient

deposits and to compare their behaviour to that of

conventional hard chromium deposits.

2. Experimental

Ni–P deposits were deposited on AISI-1045 steel

substrates by direct current electrodeposition process.

The anode was a pure Ni plate. The basic compositions

of the electrolyte are as follows: 240 g/l nickel sulfate,

30 g/l nickel chloride, 30 g/l boric acid and 20 g/l

phosphorous acid. The temperature of the plating bath

was kept at 70 8C. The pH of the plating bath was 1.5

adjusted by ammonia water or dilute sulfuric-acid.

Prior to the deposition, the substrates were mechani-

cally polished to a 0.10–0.12 mm surface finish, then a

sequence of cleanings were performed to remove

contamination on the substrate surface, the steel

substrates were activated for 20 s in a mixed acidic

bath, then rinsed with distilled water. The Ni–P

gradient deposits (Ni–P FGDs) with six layers were

electrodeposited by gradually changing the current

density from 5 to 30 A/dm2. The detailed controlling

parameters are seen in Ref. [11].

Hard Cr deposits with microhardness in the range

of 980–1050 HV and approximately 40 mm in thick-

ness was also deposited on steel substrates from a

conventional plating bath mainly containing chro-

mium trioxide and sulfuric acid, similar to what is

widely used in industry.

Microstructure investigation of cross-sectioned

deposits was performed using a JSM-5600Lv scanning

L. Wang et al. / Applied Surface Science 252 (2006) 7361–7372 7363

Fig. 1. The cross-sectional SEM morphology of the electrodepos-

ited Ni–P FGDs.

electron microscopy (SEM). The P content in the

direction of deposits thickness was measured using a

Kevex sigmaTM energy dispersive X-ray spectro-

scopy (EDS) analysis tool coupled to the SEM. The

phases in the coatings were determined by means of

X-ray diffraction (XRD) techniques. Microhardness

of the deposits was determined using a Vicker’s

microhardness indenter. The final value quoted for the

hardness of a coating was the average of 10

measurements.

To evaluate the corrosion resistance and possible

passivation behavior of the graded Ni–P and hard Cr

deposits, potentiodynamic anodic polarization curves

were acquired and the corrosion potential (Ecorr) and

corrosion current density (icorr) were determined using

the Tafel extrapolation method. Measurements were

respectively performed in 10 wt.% HCl and 10 wt.%

NaOH solutions at a temperature of 20 8C, using a

CHI660A Potentiostat/galvanostat system. A conven-

tional three-compartment plastic cell was used for the

electro-chemical investigations. The samples with

defined area of 0.24 cm2 were exposed to the

electrolyte solution. A saturated calomel electrode

(SCE) was used as the reference electrode whereas a

platinum electrode served as the counter electrode.

The specimen was first immersed in the corrosion

solution until a stable open-circuit potential (Eocp) was

reached before dynamic scanning at 10 mV/s. After

getting the stable Eocp, the upper and lower potential

limits of linear sweep voltammetry (LSV) were set at

30 mV more positive and negative than Eocp.

The wear tests under oil-lubricated sliding condi-

tions at room temperature were performed on a

reciprocating ball-on-disc UMT-2MT tribometer

(Center for Tribology, Inc., California, USA) in air.

The lubrication oil used in this study was CF-4 diesel

oil, which is commercially available from Great Wall

Lubricant Corporation of China. Si3N4 ceramic balls

of 3 mm diameter were used as the counter body. The

normal load in the wear tests was in the range of 2-

100N, whereas the sliding speed was between 2.2 and

33 cm/s. The friction coefficient and sliding time were

recorded automatically during the test. The wear

volume was measured using a surface profilometer,

the wear rates of all the deposits were calculated using

the equation of K = V/SF, where V is the wear volume

in mm3, S the total sliding distance in m and F is the

normal load in N. For each set of experimental

conditions, three tests were repeated and the results

given below refer to average values.

3. Results and discussion

3.1. Structure and composition

The cross-sectional micrograph of the Ni–P FGDs

which had a total thickness of approximately 36 mm is

shown in Fig. 1. It can be clearly observed that the Ni–P

FGDs exhibited a dense structure and strong bonding

between the deposit and steel substrate was achieved.

Moreover, Ni–P FGDs exhibited perfect compatibility

between the six sublayers and no obvious interface

between sub-layers can be seen. The distribution of P

contents in as-deposited Ni–P FGD and after heat

treatment at 400 8C in the direction of thickness is

shown in Fig. 2 [11]. It is evident that the P content

decreases gradually from the coating-substrate inter-

face to the top surface, which is in accordance with the

experimental design of the Ni–P gradient deposits. In

addition, the graded composition of Ni–P gradient

deposits was not changed after heat-treated at 400 8C.

Previous study has shown that as-deposited Ni–P

FGDs became increasingly amorphous with increas-

ing the distance form the surface. After annealing at

400 8C for 1 h, diffraction peaks corresponding to the

Ni3P and nickel phase in the XRD pattern were

observed simultaneously [11], indicating the precipi-

tation of dispersed hard Ni3P intermetallic compounds

L. Wang et al. / Applied Surface Science 252 (2006) 7361–73727364

Fig. 2. The distribution P content in the Ni–P FGDs before and after

heat treatment at 400 8C.

Fig. 4. LSV curves at the vicinity of the open-circuit potential for

Ni–P FGDs, measured in a 10 wt.% HCl solution.

in a nickel matrix. High hardness of Ni–P FGDs in the

range of 900–1100HV can be obtained as a result of

precipitation hardening by nickel phosphide (Ni3P)

precipitates at high temperature [12].

3.2. Corrosion behavior

The polarization curves measured in 10 wt.% HCl

solution for Ni–P FGDs with an annealing temperature

in air at 200, 400 8C for 1 h are shown in Fig. 3 as

curves A–C, respectively. The electrochemical beha-

vior of hard Cr deposit measured in the same solution

is also shown as curve D for a comparison purpose.

The corrosion resistance, Rcorr, was determined from

Fig. 3. Potentiodynamic polarization curves obtained for Ni–P

FGDs and hard Cr deposits, measured in 10 wt.% HCl solution.

the slopes of the potential-current plots measured by

LSV in the range of �30 mV about the open-circuit

potential (Eocp). Typical LSV curves of as-deposited

Ni–P FGDs and the deposits after heat-treated at 200,

400 8C measured in 10 wt.% HCl solution is shown in

Fig. 4, respectively. The corrosion resistance is

calculated on the basis of the following equation

[13,14]:

Rcorr ¼dE

di

����E¼Eocp

� DE

Di(1)

The corrosion potential (Ecorr) and corrosion

current density (icorr) calculated using Tafel extra-

polation method and the corrosion resistance for Ni–P

FGDs and hard Cr deposits calculated on the basis of

Eq. (1) are summarized in Table 1. By combining

Fig. 3 and Table 1, among the Ni–P FGDs, the

corrosion potential of graded Ni–P deposits is

positively shifted from �290 to �166 mV with

increasing the annealing temperature from room

temperature to 400 8C. Moreover, the corrosion

current density (icorr) and corrosion resistance of

heat-treated Ni–P FGDs at 200 8C is the lowest, slight

following by as-deposited Ni–P FGDs. In is clearly

shown that the heat-treated Ni–P FGDs at 200 8Cshow nobler Ecorr, lowest icorr and thus potentially

better corrosion resistance in the active region. The

heat-treated Ni–P FGD at 400 8C exhibited much

higher corrosion current density than as-deposited

FGDs, but showed very close corrosion resistance to

L. Wang et al. / Applied Surface Science 252 (2006) 7361–7372 7365

Table 1

Corrosion potential (Ecorr), corrosion current (icorr), and corrosion resistance (Rcorr) of Ni–P FGDs and hard Cr deposits in 10 wt.% HCl solution

System studied Ecorr (mV vs. SCE) icorr (A cm�2) Rcorr (V cm�2)

Graded Ni–P as-deposited �290 3.738E�6 1.03E5

Graded Ni–P heat treated at 200 8C �232 3.329E�6 1.81E5

Graded Ni–P heat treated at 400 8C �166 6.080E�6 1.04E5

Hard chromium �778 1.337E�3 4.31E2

that of as-deposited Ni–P FGDs, which is consistent

with the report of [15,16]. Many studies concerning

the corrosion resistance of Ni–P deposits have shown

that, the corrosion behavior of Ni–P deposits depend

significantly on three principal factors, namely, the

degree of amorphous state, extent of internal stress and

the percentage of phosphorus content in deposits [16].

As a consequence, the better corrosion resistance of

as-deposits Ni–P FGDs in amorphous state is due to

their homogeneous structure and the absence of grain

boundaries, dislocations, kink sites and other surface

defects [17]. The best corrosion resistance for the Ni–

P FGDs heat-treated at 200 8C, were both due to the

retained amorphous structure and the stress relaxation

by plastic flow and the onset of intrinsic stress at this

temperature. For the Ni–P FGDs heat-treated at

400 8C, very close corrosion resistance to as-deposited

Ni–P FGDs can be understood on the basis of a

competition between the following two antagonistic

mechanisms. On one hand, the phase transformation

from amorphous to a crystalline structure at annealing

temperature of 400 8C made available more number of

grain boundaries which are prone to corrosion attack.

On the other hand, the formation of high thermo-

Fig. 5. SEM micrographs of hard Cr deposit before (a) and after (b) polariz

and poor resistance to Cl� attack.

dynamic stable intermatallic compounds Ni3P which

is easy for passivation and the formation of Ni oxide

film on the surface of deposits could increase the anti-

corrosion ability of heat-treated Ni–P FGDs at 400 8C[18]. Accordingly, the highest hardness of heat-treated

Ni-P deposits at 400 8C was obtained without by the

sacrifice of corrosion resistance.

Note from Table 1 that the corrosion potential of hard

Cr deposits are�778 mV, which is much more negative

than that of heat-treated Ni–P FGD at 400 8C, and also

exhibited more than two orders of magnitude higher

corrosion current density than that of Ni–P FGDs. This

revealed that heat-treated Ni–P FGDs exhibited super-

ior corrosion resistance than hard Cr deposits in HCl

solution. Above results further confirmed that heat-

treated Ni–P FGDs at 400 8C with better wear

resistance could be used for corrosion protection

application in acidic and Cl�-containing solutions.

The poor corrosion resistance of hard Cr deposits in

HCl solution can be further confirmed by the SEM

morphologies before and after corrosion test as shown

in Fig. 5a and b. The typical surface morphology of

conventional hard chromium is shown in Fig. 5a. The

nodular surface and the network of cracks on deposits

ation test in 10 wt.% HCl solution, showing severe crevices corrosion

L. Wang et al. / Applied Surface Science 252 (2006) 7361–73727366

Fig. 6. SEM micrographs of heat-treated Ni–P FGDs at 400 8C before (a) and after (b) polarization test in 10 wt.% HCl solution, showing

localized corrosion in the interface of Ni matrix and Ni3P precipitates.

Fig. 7. Polarization curves obtained for Ni–P FGDs after heat-

treated at 400 8C and hard Cr deposits, in 10 wt.% NaOH solution.

can be clearly observed. During chromium electro-

deposition, intensive hydrogen evolution reaction

occurred. Then, unstable chromium hydride (such

as hexagonal CrH) was formed; the hexagonal CrH is

thought to decompose to body-centered cubic (BCC)

chromium with a 15% volume contraction. Hence,

cracks are then formed as a consequence of the

decomposition of chromium hydrides and shrinkage

of the crystallographic structure [19]. After corrosion

tests, the nodular structure was entirely replaced by

crevices which existed along the walls of the original

cells, and cracks became wider and more than as-

deposited hard Cr deposits as shown in Fig. 5b. This

indicated that severe dissolution of Cr species due to

the Cl� attack in acids environment, and thus the

exposed area of specimens in HCl solution became

greenblack quickly. Because natural crevice morphol-

ogy of the crack act as active paths for the penetration

of Cl� inside of deposits, and then the crevices

propagated and coalesced gradually. Hence, much

more crevices were formed on the surface and even

some upper layer was corroded away completely. As a

consequence, hard Cr deposits exhibited a rather poor

resistance to Cl� attack. The above SEM morphology

of hard Cr after polarization tests, consistent with

those data evaluated by polarization curves, further

support the conclusion that hard Cr deposits exhibited

much poor corrosion resistance in acidic and Cl�-

containing solutions. As for the heat-treated Ni–P

FGDs at 400 8C, quite similar morphologies before

and after polarization tests were observed shown in

Fig. 6a and b. Uniform distribution of hardened Ni3P

precipitates in the nickel matrix and relatively smooth

surface without the cracks can be clearly seen, which

is in agreement with the report of [20]. The

morphology after corrosion test shows localized

corrosion in the interface of Ni matrix and Ni3P

precipitates, but still retain smooth, bright finish and

continuous deposits as shown in Fig. 6b. From all the

above results and discussion, the heat-treated Ni–P

FGDs at 400 8C with the maximum hardness exhibited

superior anticorrosion properties against the Cl�

attack in acids when compared with hard Cr deposits.

The polarization curves measured in 10 wt.%

NaOH solution for heat-treated Ni–P FGDs at

400 8C and hard Cr deposit are comparatively shown

in Fig. 7. Data of Ecorr, icorr and Rcorr of above two

deposits measured in 10 wt.% NaOH solution are

comparatively summarized in Table 2. It is clearly

shown that the heat-treated Ni–P FGDs at 400 8Cshow positive corrosion potential but a little higher

corrosion current than hard Cr deposits in the active

L. Wang et al. / Applied Surface Science 252 (2006) 7361–7372 7367

Table 2

Corrosion potential, corrosion current and corrosion resistance of graded Ni–P and hard Cr deposits in 10 wt.% NaOH solution

System studied Ecorr (mV vs. SCE) icorr (A cm�2) Rcorr (V cm�2)

Ni–P FGDs heat-treated at 400 8C �626 3.44E�6 3.83E5

Hard Cr deposits �638 1.62E�6 6.85E5

region. As a consequence, the heat-treated Ni–P FGDs

at 400 8C with maximum hardness exhibited a little

poor corrosion resistance than hard Cr deposits in

alkaline environment.

3.3. Lubricated sliding friction and wear behavior

In industry most contacts are lubricated in order to

control friction and wear. In real applications, such as

cylinder liners, piston ring, rolls and machine tools,

contacts operate in a specific lubrication regime. In

order to optimize the contacts with regard to friction

on the one hand and lifetime on the other hand, it is

necessary to be able to predict the lubrication regime

in which such contacts operate [21]. In the lubricated

sliding wear study, the friction coefficient during

lubrication is potentially influenced by sliding speed

(v), mean contact pressure (Pa) or normal load, and

dynamic viscosity (h). In lubrication theory, these

three quantities often appear in a single quantity called

the Sommerfeld number (S), this number is defined as

[22,23]:

S ¼ hv

RaP(2)

where h is the dynamic viscosity of oil in Pa s, v the

sliding speed in m/s and Ra is the combined CLA

Fig. 8. Stribeck curves for friction coefficient under lubrication conditions

at 400 8C and (b) hard Cr deposits.

surface roughness in m, defined by

Ra ¼ ½R2a1 þ R2

a2�1=2

(3)

with Ra1 and Ra2 the CLA surface roughness of upper

surface 1 and lower surface 2, respectively, in which P

the mean contact pressure at the contact zone in Pa is

defined as

P ¼ 0:3870

�NE2

R2

�1=3

(4)

where N is the normal load applied in wear test, R the

radius of upper sliding ball and E is the effective

elastic modulus, which can be calculated:

1

E¼ 1� n2

1

E1

þ 1� n22

E2

(5)

with E1 and E2 the elastic modulus of the upper surface

1 and lower surface 2, respectively. And n is the

Poisson’s ratio of materials.

Experiments on lubrication of material surfaces as

a function of Sommerfeld number often reduce to the

Stribeck curve. Plotting the Stribeck curve is still a

convenient method for examining the effect of the

important variables of sliding speed and normal load

to indicate lubrication mechanisms and predict the

lubrication regime. The Stribeck curves for the heat-

as a function of Sommerfeld number for (a) heat-treated Ni–P FGDs

L. Wang et al. / Applied Surface Science 252 (2006) 7361–73727368

Fig. 9. SEM micrographs of the worn surface of (a) heat-treated Ni–P FGDs and hard Cr deposits after sliding at 33 cm/s under a 5 N normal load

for 9000 cycles under EHD lubrication conditions.

Fig. 10. Effect of load on the wear rate of heat-treated Ni–P FGDs

and hard Cr deposits with a constant sliding speed of 22 cm/s.

treated Ni–P FGDs at 400 8C and hard Cr deposits

under lubrication conditions are shown in Fig. 8a and

b, respectively, which provides an insight into the

lubrication mechanisms. From Fig. 8, three main

regimes can be identified, each one corresponding to a

different lubrication mode [24,25]. In regime I at high

Sommerfeld number, the surfaces are fully lubricated

by Elastohydrodynamic lubrication (EHD mode) with

the friction coefficient rising as Sommerdeld number

increases further. And the ball and the deposits were

completely separated by oil film and there is no

contact between the sliding surfaces. Using surface

profiler and SEM observation on wear tracks of above

two deposits, no visible wear scar and wear volume

can be measured. However, very slight microcracks

and traces are observed on the sliding surface as shown

in Fig. 9a and b, which may be caused by ineffective

lubrication during the first few cycles, before a true

EHD lubrication mode is established in this system.

This can be confirmed by the higher friction

coefficient of deposits at first sliding cycles owing

to the absence of a full oil film.

As Sommerfeld number reduces as a result of either

an increase in load or a decrease in sliding speed in

regime II, which corresponds to mixed lubrication

(ML) with the friction coefficient increasing to a high

value as Sommerfeld number decreases. In which the

surfaces get closer and metal-to-ceramic contact takes

place locally. The load is carried by both the oil

lubricant film and the deposit-ball contact. SEM

observation and wear volume loss measurement on

wear track of above two deposits exhibited that cracks,

small craters and small detachment of debris are

existed on the worn track. This is the typical of the

mixed lubrication mode.

Further increase in normal load or decrease in

sliding speed makes the Stribeck curve develop a

transition to boundary lubrication regime (BL),

identified as regime III as shown in Fig. 8. Under

boundary lubrication, intensive deposit-to-ceramic

contact of the sliding surface takes place and the

load is completely carried by the contacts. Thus, the

friction coefficient is higher than other lubrication

mode but much lower than dry sliding friction

conditions (dry sliding friction coefficient are in the

range of 0.5 � 0.08). SEM and wear volume

measurement show that many brittle cracks, craters

and severe detachment of deposits was observed under

the boundary lubrication conditions.

During the mixed lubrication, the effects of normal

load and sliding speed on the wear rate of heat-treated

Ni–P FGDs and hard Cr deposits are comparatively

shown in Figs. 10 and 11, respectively. It is observed

L. Wang et al. / Applied Surface Science 252 (2006) 7361–7372 7369

Fig. 11. Effect of sliding speed on the wear rate of heat-treated Ni–P

FGDs and hard Cr deposits with a constant load of 60 N.

Fig. 13. Effect of sliding speed on the friction coefficient of Ni–P

FGDs and hard Cr deposits with a constant load of 60 N.

that either an increase in load or decrease in sliding

speed while keeping other parameters constant, results

in an increase in wear rate. From Fig. 10, the wear rate

increased sharply with an increase of normal load

whereas the wear rate of two deposits decreased

gradually with the increase in sliding speed. In

addition, the heat-treated Ni–P FGDs exhibited

relatively higher wear rate than hard Cr deposits

under the oil-lubricated wear conditions. The reason

for this will be provided in the later discussion.

Effects of normal load and sliding speed on the

friction coefficient of heat-treated Ni–P FGDs and

hard Cr deposits are comparatively shown in Figs. 12

and 13, respectively. It seems that both deposits

Fig. 12. Effect of load on the friction coefficient of Ni–P FGDs and

hard Cr deposits with a constant sliding speed of 22 cm/s.

exhibited very similar variations as function of normal

load and sliding speed. Heat-treated Ni–P FGDs show

a little higher friction coefficient than hard Cr

deposits. The friction coefficient increases from

0.045 to 0.062 when the load decreases from 30 to

100 N and the friction coefficient increases from 0.087

sharply to 0.030 with the increase of sliding speed

from 5.5 to 33 cm/s. The above results indicated that

effect of normal load on the friction coefficient is

minor, whereas a strong dependence of the friction

coefficient on the sliding speed in mixed lubrication

mode. The above effects of normal load and sliding

speed on wear rate and friction coefficient of

electrodeposits are consistent with previous reports

on lubricated tribological behavior of multiplayer Ni–

W–P deposits and bulk materials in mixed lubrication

mode [22,24,26]. Panagopoulos et al. [22] thought that

an increase in load would significantly deteriorate the

contact situation and increase the contact pressure,

thus result in much higher wear rate. Whereas a

decrease in sliding speed will not greatly increase the

contact pressure, only more extensive metallic

contact. This made the wear loss will stay at relatively

low levels. However, in the case of friction coefficient

variations, a decrease in sliding speed would lead to a

significant decrease of the minimum lubrication film

thickness, thus more extensive metallic contact

happened, and increase the friction coefficient.

However, an increase in load will not produce such

a significant decrease in lubrication film thickness.

Hence, variations of friction coefficient for above two

L. Wang et al. / Applied Surface Science 252 (2006) 7361–73727370

deposits were more affected by sliding speed rather

than the normal load.

The relatively poor wear resistance of heat-treated

Ni–P FGDs compared with hard Cr deposits under oil

lubricated conditions, can be further explained by

SEM observation on wear track as shown in Fig. 14. It

is clearly that evolution of wear process for above two

deposits under lubrication, undergoes three different

wear stages:

(1) S

Fig.

evolu

tage I as shown in Fig. 14a and d (under a load of

60 N and a reciprocating frequency of 20 Hz for

1200 cycles), the wear is quite mild without debris

or clearly distinguished wear track for both two

deposits, microcracks parallel to the sliding

14. SEM morphologies of the worn tracks of heat-treated Ni–P FGDs (a–c)

tion of wear process under the ML or BL lubrication conditions.

direction on surface were observed both for two

deposits, however, hard Cr deposits exhibited

much more microcracks on sliding surface than

heat-treated Ni–P FGDs. This can be attributed to

the pre-existing cracking structure of hard Cr

deposits, since these pre-existing cracks on as-

deposited Cr deposits are susceptible to cracking

under the combined stress of load and shear.

(2) S

tage II as shown in Fig. 14b and e (for 12 000

cycles). The above initial microcracks propagated

driven by the stress field of the sliding counter-

faces to produce voids and free chips of deposits

as debris particles. This stress imposed on the

deposits during reciprocating wear process is

cyclic, namely compressive immediately in front

and hard Cr deposits (d–f) at different wear stages, showing the

L. Wang et al. / Applied Surface Science 252 (2006) 7361–7372 7371

of the contact area and tensile immediately behind

it [27]. Comparative observations reveals that

much more cracks are also present on the wear

surfaces of hard Cr deposits when compared to

heat-treated Ni–P FGDs, however, relatively more

chipping failure and brittle detachment of deposits

are observed on the worn track of heat-treated Ni–

P FGDs. On one hand, close observation of

Fig. 14e shows that these cracks was stopped at

the nodule boundaries which is the typical

structure of hard Cr deposits as shown in

Fig. 5a, thus preventing chipping failure and

detachment of deposit as debris (we termed this

nodular effect). One the other hand, under oil

lubrication conditions nodular topography of hard

Cr deposit provide superior wettability for oil than

Ni–P FGDs, and the cracks between the nodules

provide a capillary action which increases the

spreading of the lubricating oil on the worn

surface [27,28], which enhances its durability and

a litter lower friction coefficient than heat-treated

Ni–P FGDs under lubricated wear conditions.

This is also the reason why more cracking worn

surface was observed on hard Cr deposits, but

shows less wear loss than heat-treated Ni–P

FGDs.

(3) S

tage III as shown in Fig. 14c and f (for 36 000

cycles). Owing to the relatively poor fracture

resistance of heat-treated Ni–P FGDs, subsequent

action of the sliding counterface on the existing

chips or brittle craters in stage II rapidly leads to

the quick propagation of wide cracks in the

direction of thickness and even toward outside of

wear track. This results in the observed cata-

strophic breakdown of deposits as shown in

Fig. 14c, which always leads to the much wider

worn track than hard Cr deposit under the same

wear conditions. For hard Cr deposits in this stage,

resulted cracks in stage II eventually intersect and

join up to form discrete chips, which become

detached from the sliding surface. In addition,

fatigue wear was observed owing to the abrasive

action of hardened debris trapped between the

sliding surfaces.

Above results and discussions indicated that, for

heat-treated Ni–P FGDs, the wear procedure under oil

lubrication conditions is: formation of micro-

cracks! chipping failure and brittle craters forma-

tion! catastrophic breakdown of deposits. Whereas

the wear process of hard Cr deposits is: formation of

more microcracks based on pre-existing cracking

structure! crack propagation was inhibited due to

nodular and superior wettability for oil owing to

nodular topography! discrete chips detached from

the sliding surface. Zhang et al. [29] found for hard

coatings that the cracks propagation is the key stage;

the delay of this stage will effectively prolong the

lifetime of coatings. As a consequence, the heat-

treated Ni–P FGDs exhibited relatively poor wear

resistance than hard Cr deposits with superior oil-

retaining surface structure under the lubrication

conditions. However, in view of previous comparison

on wear resistance of above two electrodeposits [11],

the heat-treated Ni–P FGDs exhibited better wear

resistance when compared with electrodeposited hard

Cr deposits under both dry sliding wear and high

temperature wear conditions, Thus, heat-treated Ni–P

FGDs may be an potential alternative to hard Cr

deposits in terms of both corrosion resistance and wear

resistance under less progressive environment.

Frankly speaking, complete replacement of electro-

deposited hard Cr deposits will be a long and

complicated process and will be limited in the near

future to certain applications. Further research will be

done to further improve the wear resistance of graded

Ni–P deposits under oil-lubricated conditions.

4. Conclusions

Ni–P gradient deposits with a graded change of P

content within the deposits were successfully produced

by an electredeposition process in a single plating bath.

The systemic investigation on the properties of these

Ni–P FGDs with a comparison to electrodeposited hard

Cr deposits led to the following conclusions.

Heat treatment of graded Ni–P deposits in air show

positive effect on the anticorrosion properties of

deposits due to the formation of Ni3P and Ni oxide.

The heat-treated Ni–P FGDs at 400 8C with the

maximum hardness exhibited more than two orders of

magnitude higher corrosion resistance against the Cl�

attack in acids and showed a little poor corrosion

resistance in alkaline environment when compared

with hard Cr deposits.

L. Wang et al. / Applied Surface Science 252 (2006) 7361–73727372

The Stribeck curves for the heat-treated Ni–P

FGDs and hard Cr deposits under lubrication

conditions are obtained with accurate control of

normal load and sliding speed during the wear process,

three main different regimes corresponding to three

different lubrication mechanism were identified. In

addition, effects of normal load and sliding speed on

wear rate and friction coefficient of heat-treated Ni–P

FGDs and hard Cr deposits are similar. However, the

heat-treated Ni–P FGDs exhibited relatively poor wear

resistance than hard Cr deposits under the lubrication

conditions, which may be attributed to superior oil-

retaining surface structure and the unique ‘‘nodular’’

effect of hard Cr in wear process.

Acknowledgements

The authors gratefully acknowledge the National

Natural Science Foundation of China (Grant Nos.

50271080 and 50323007), the 863 Program of China

(No. 2003AA305670) and the Innovative Group

Foundation from NSFC (Grant No. 50421502) for

financial support of this research work. The authors

are also thankful to Dr. Jun Liang for the profitable

advice and discussions.

References

[1] D. Landolt, J. Electrochem. Soc. 149 (2002) 9.

[2] B. Navinsek, P. Panjan, I. Milosev, Surf. Coat. Technol. 116–

119 (1999) 476.

[3] M. Heydrzadeh Sohi, A.A. Kashi, S.M.M. Hadavi, J. Mater.

Process. Technol. 138 (2003) 219.

[4] S. Han, J.H. Lin, S.H. Tsai, S.C. Chung, D.Y. Wang, F.H. Lu,

H.C. Shih, Surf. Coat. Technol. 133/134 (2000) 460.

[5] S. Eskin, O. Berkh, G. Rogalsky, J. Zahavi, Plating Surf.

Finish. 85 (1998) 79.

[6] M. Donten, H. Cesiulis, Z. Stojek, Electrochimica Acta 45

(2000) 3389.

[7] H. Capel, P.H. Shipway, S.J. Harris, Wear 255 (2003) 917.

[8] B. Kieback, A. Neubrand, H. Riedel, Mater. Sci. Eng. A 362

(2003) 81.

[9] H.Z. Wang, S.W. Yao, S. Matsumura, Surf. Coat. Technol. 157

(2002) 166.

[10] L.P. Wang, Y. Gao, T. Xu, Q.J. Xue, J. Phys. D: Appl. Phys. 38

(2005) 1318.

[11] L.P. Wang, Y. Gao, T. Xu, Q.J. Xue, Surf. Coat. Technol., in

press.

[12] D.H. Jeong, U. Erb, K.T. Aust, G. Palumbo, Scripta Materialia

48 (2003) 1067.

[13] I. Paseka, Electrochimica Acta 40 (1995) 1641.

[14] A. Bai, P.-Y. Chuang, C.-C. Hu, Mater. Chem. Phys. 82 (2003)

93.

[15] H. Ashassi-Sorkhabi, S.H. Rafizadeh, Surf. Coat. Technol. 176

(2004) 318.

[16] Y.S. Huang, X.T. Zeng, X.F. Hu, F.M. Liu, Electrochimica

Acta 49 (2004) 4313.

[17] R. Raicheff, V. Zaprianova, J. Mater. Sci. Lett. 19 (2000) 3.

[18] C.F. Malfatti, J. Zoppas Ferreira, C.B. Santos, B.V. Souza, E.P.

Fallavena, S. Vaillant, J.-P. Bonino, Corros. Sci. 47 (2005) 567.

[19] D.R. Gabe, J. Appl. Electrochem. 27 (1997) 908.

[20] H.X. Li, H.Y. Chen, S.Z. Dong, J.S. Yang, J.F. Deng, Appl.

Surf. Sci. 125 (1998) 115.

[21] E.R.M. Gelinck, D.J. Schipper, Tribol. Int. 33 (2000) 175.

[22] C.N. Panagopoulos, V.D. Papachristos, L.W. Christoffersen,

Thin Solid Films 366 (2000) 155.

[23] D.J. Schipper, P.H. Vroegop, A.W.J. de Gee, Lubricat. Sci. 3

(1991) 191.

[24] J. Sheng, D.T. Gawne, J. Mater. Sci. 32 (1997) 913.

[25] H. Czichos, Tribology, Elsevier, Amsterdam, 1978, p. 132.

[26] E. Bozyazı, M. Urgen, Ali Fuat Cakır, Wear 256 (2004) 832.

[27] F.G. Arieta, D.T. Gawne, Surf. Coat. Technol. 73 (1995) 105.

[28] F.G. Arieta, D.T. Gawne, Surf. Coat. Technol. 70 (1995) 243.

[29] R.J. Zhang, Z.W. Lin, Z.P. Cui, Q. Song, Wear 147 (1991) 227.