Bhushan B. Nanotribology and Nanomechanics: An Introduction

Подождите немного. Документ загружается.

930 Bharat Bhushan

Wear and Scratch Resistance

Wear resistance was studied on an area of 1µm×1 µm. The variation of wear depth

with normal load is presented in Fig. 17.15 [39]. HDT exhibits the best wear re-

sistance. A critical normal load (marked by arrows in Fig. 17.15) appears in the

curves for all of the SAMs tested. When the normal load is smaller than the critical

normal load, the monolayer only changes in height very slightly in the scan area.

When the normal load is higher than the critical value, the change in height of the

SAM increases dramatically. Relocation and accumulation of BPT molecules is ob-

served during the ﬁrst few scans, which leads to the formation of a larger terrace.

Wear studies of a single BPT terrace indicate that the wear life of BPT increases

exponentially with terrace size [40,41].

The scratch resistances of Au(111) and SAMs were studied using a continuous

AFM microscratch technique. Figure 17.16a shows coeﬃcient of friction proﬁles

as a function of increasing normal load, and corresponding tapping mode AFM

surface height images of the scratches captured on Au(111) and SAMs [41]. Fig-

ure 17.16a indicates that there is an abrupt increase in the coeﬃcient of friction for

all of the tested samples. The normal load associated with this event is termed the

critical load (indicated by the arrows labeled “A”). Initially during scratching, all of

the samples exhibit a low coeﬃcient of friction, indicating that the friction force is

dominated by the shear component. This is in agreement with AFM image analysis,

which shows negligibledamage on the surfaces prior to the critical load. At the criti-

cal load, a clear grooveis formed, which is accompanied by material piling up at the

sides of the scratch. This suggests that the initial damage that occurs at the critical

load is due to plowing associated with plastic deformation,and this causes the sharp

rise in the coeﬃcient of friction. Beyond the critical load, debris can be seen as well

as material pile-up at the sides of the scratch. Figure 17.16b summarizes the critical

loads for the various samples obtained in this study. It clearly indicates that all of

the SAMs increase the critical load of the corresponding substrate.

Decrease of surface height (nm)

–1

01234567

Normal load (μN)

Au(111)

HDT

BPT

BPTC



Fig. 17.15. Wear depth as

a function of normal load

after one scan cycle

17 Self-Assembled Monolayers (SAMs) for Controlling Adhesion 931

Coefficient of friction

0.6

0.4

0.2

Normal load (

μN)

0 20 40 60 80 100 120

Coefficient of friction

0.6

0.4

0.2

Normal load (

μN)

0 20 40 60 80 100 120

Coefficient of friction

0.6

0.4

0.2

Normal load (

μN)

0 20 40 60 80 100 120

Au(111) HDT BPT

Critical load (μN)

Materials

Au(111)

3 μm

HDT

BPT

HDT

3 μm

BPT

3 μm

0 40 nm

Fig. 17.16. (a)Coeﬃcient

of friction proﬁles during

scratch as a function of nor-

mal load, and correspond-

ing AFM surface height

images as well as (b) criti-

cal loads estimated from

coeﬃcient of friction pro-

ﬁles and AFM images for

Au(111), HDT/Au(111)and

BPT/Au(111) [41]

The mechanisms responsible for the sudden drop in surface height with increas-

ing load during the wear and scratch tests need to be understood.Barrena et al. [88]

observed that the heights of self-assembled alkylsilane layers decrease in discrete

amounts with normal load. This step-like behavior is due to discrete molecular

932 Bharat Bhushan

tilts, which are dictated by the geometrical requirements of the close packing of

molecules. Only certain angles are allowed due to the zigzag arrangement of the

carbon atoms. The relative height of the monolayer under pressure can be calcu-

lated by the following equation













−1/2

, (17.6)

where L is the total length of the molecule, h is the height of the SAMs in the

tilt conﬁguration (monolayer thickness), a is the distance between alternate carbon

atoms in the molecule, d is the separation of the molecules, and n is the step number.

The values of a (0.25nm) and d (0.47nm) are used in the calculation of HDT. The

calculated and measuredrelative heights of HDT are listed in Table 17.11. When the

normal loads are smaller than the critical values in Fig. 17.15,the measured relative

height values of HDT are very close to the calculated values. This means that HDT

underwent step tilting below the critical normal load.

The residual SAM thickness after wear under critical normal load was measured

by proﬁling the worn ﬁlm using AFM. The results are listed in Table 17.12. For an

alkanethiol monolayer, the relationship between the monolayer thickness h and the

intercept length L

can be expressed as (Fig. 17.17)

h = bcos(α)n+ L

, (17.7)

where b is the length of the projection of the C

−

C bond onto the main chain axis

(b = 0.127nm for alkanethiol), n is the chain length deﬁned by CH

(CH

)

SH, and

α is the tilt angle [80]. The L

values have also been calculated for BPT and BPTC,

based on the same principle, and using the bond lengths reported in reference [81],

Table 17.7. This indicates that the measured residual thickness values of SAMs

Table 17.11. Calculated





−

and measured





relative heights of an HDT self-

assembled monolayer [39]

Steps Calculated









−

Measured





(n)

1 0.883

2 0.685 0.674

3 0.531 0.532

4 0.425 0.416

5 0.352 0.354

6 0.299

Calculations are based on the assumption that the molecules tilt in discrete

steps (n) upon compression with a diamond AFM tip [88]

b,c,d,e

These measured values correspond to normal loads of 0.50 µN, 1.57 µN,

2.53 µNand4.03 µN, respectively

17 Self-Assembled Monolayers (SAMs) for Controlling Adhesion 933

Table 17.12. Calculated L

values and measured residual ﬁlm thicknesses of SAMs under

critical load

Residual thickness

(nm) (nm)

HDT 0.24 0.25

BPT 0.39 0.42

BPTC 0.33 0.38

Calculated using the equation h = bcos(α)n+ L

[80]

Measured by AFM using a diamond tip under critical normal

load. All of the data are the mean values of three tests

Fig. 17.17. Illustration of the relationship between the compo-

nents of the equation h = bcos(α)n+ L

[39]

under critical load are very close to the calculated intercept length (L

)values.This

means that the Si

tip approaches the interface and SAMs are severely worn

away from the substrate under the critical normal load. This is because chemical

adsorption bond strength at the interface (S

−

Au) is generally smaller than other

chemical bond strengths in SAM spacer chains (Table 17.8).

The change in the wear mechanism of a SAM with increasing normal load is

therefore illustrated in Fig. 17.18. Below the critical normal load SAMs undergo

Critical load

Decrease of surface height (arb. units)

Normal load (arb. units)

–1

123456

Fig. 17.18. Illustration of

the wear mechanisms of

SAMs with increasing normal

load [41]

934 Bharat Bhushan

step orientation; at the critical load SAMs wear away from the substrate due to the

weak interface bond strengths; while above the critical normal load severe wear

takes place on the substrate. To improve the wear resistance, the interface bond

must be enhanced; a rigid spacer chain and a hard substrate are also preferable in

this regard.

17.4.3 Alkylsilane and Perﬂuoroalkylsilane SAMs on Si(100)

and Alkylphosphonate SAMS on Al

The nanotribological parameters of perﬂuorodecyltricholorosilane (PFTS),

−

(CF

)

−

(CH

)

−

SiCl

, n-octyldimethyl (dimethylamino)silane (ODMS),

−

(CH

)

−

Si(CH

)

−

N(CH

)

(n= 7),and n-octadecylmethyl(dimethylamino)-

silane (n = 17) (ODDMS) vapordeposited on Si substrate were investigated,as were

those of octylphosphonate(OP),

−(CH

)

−

−OH

(n = 7) and octadecylphosponate (ODP) (n = 17) on Al substrate. The perﬂuo-

roalkylsilane SAM was selected because ﬂuorinated ﬁlms are known to have low

surface energy. Alkylsilanes with two diﬀerent chain lengths (with 8 and 18 carbon

atoms) were selected in order to compare their nanotribological performance with

that of PFTS as well as to study the inﬂuence of chain length. The alkylphosphonate

SAMs (with 8 and 18 carbon atoms) on Al were selected due to their industrial use

in applications such as digital projection displays.

Surface Free Energy and Contact Angle Measurement

As stated earlier, adhesive force arises from the presenceof a thin liquidﬁlm such as

a mobile lubricant or an adsorbed water layer that causes meniscus bridges to build

up around asperities due to surface energy eﬀects. The intrinsic attractive force aris-

ing from meniscus contributions depends on the surface tension of the ﬁlm and the

contact angle and may result in high friction and wear. Surface energies and contact

angles were measured in order to evaluate the hydrophobicity. Figure 17.19 shows

a Zisman plot for the SAMs deposited on Si and their surface energies obtained

using various alkane liquids [43]. Zisman analysis data was not available for the

Si substrate because the alkane liquids used for the measurement spreads instantly

across such surfaces. A signiﬁcantly lower critical surface tension or surface energy

was observed for PFTS (12.9mN/m for PFTS/Si) than for ODMS (24.7mN/m

for ODMS/Si) or ODDMS (23.9mN/m for ODDMS/Si). The surface energies for

ODMS and ODDMS were comparable. This suggests that the surface was covered

by these SAMs to a comparable degree without bare substrate appearing.

Contact angles were measured for various SAMs on Si and Al substrates so

that they could be compared. The measured values for the samples are shown in

17 Self-Assembled Monolayers (SAMs) for Controlling Adhesion 935

01.0

0.9

0.8

0.7

0.6

0.5

0.4

252015

cos (contact angle)

Surface tension of alkane (mN/m)

Contact angle (deg)

ODDMS/Si

ODMS/Si

PFTS/Si

23.9 mN/m

24.7 mN/m

12.9 mN/m

Fig. 17.19. Zisman plots

for PFTS/Si, ODMS/Si

and ODDMS/Si used to

calculate the critical sur-

face tension, a measure of

surface energy, which is

givenbythex-intercept

[cos(contact angle) = 1] of

the line ﬁtted to the data

Table 17.13. A summary of RMS roughness, contact angle and ﬁlm thicknesses of various

SAMs

SAM/substrate Acronym RMS roughness Contact angle Film thickness

(nm) (deg.) (nm)

Silicon(111) Si 0.07 48 –

Perﬂuorodecyltricholoro- PFTS/Si 0.09 112 1.8

silane/Si

Octyldimethyl(dimethyl- ODMS/Si 0.10 99 1.9

amino)silane/Si

Octadecyldimethyl- ODDMS/Si 0.10 92 2.1

(dimethylamino)silane/Si

Aluminium Al 1.73 74 –

Octylphosphonate/Al OP/Al – 108 ≈ 1.9

Octadecylphosphonate/Al ODP/Al – 115 ≈ 2.1

Kasai et al. [44]

Kulik (personal communication)

Fig. 17.20a [43,47]. A summary of RMS roughness measured using an AFM, con-

tact angles and ﬁlm thickness measured using an ellipsmeter are summarized in

Table 17.13.Signiﬁcantly better water repellency was observed for PFTS compared

to bare Si with natural oxide. The contact angle for PFTS/Si was ≈ 110

◦

whereas

it was ≈ 50

◦

for Si. The contact angle generally increases with decreasing surface

energy [89], which is consistent with the data obtained. The water contact angles

of ODMS and ODDMS were also large (≈ 100

◦

), implying high surface hydropho-

bicity. These contact angles can be inﬂuenced by the packing density as well as the

sample roughness [90], which probably accounts for the slightly higher contact an-

gles for the SAMs deposited on Al substrates compared to those on Si substrate.

936 Bharat Bhushan

Contact angle (deg)

120

100

Adhesive force (nN)

Friction force (nN)

Coefficient of friction

00.0

3.0

2.5

2.0

1.5

1.0

0.5

0.07

0.06

0.05

0.04

0.03

0.02

0.01

Si PFTS ODMS ODDMS Al OP ODP

Si substrate Al substrate

At 5 nN normal load

Fig. 17.20. (a) The static

contact angle, adhesive force,

friction force and coeﬃcient

of friction measured using an

AFM for various SAMs on Si

and Al substrates

The

−

groups in ODMS and ODDMS are nonpolarand are known to contribute

to the water repellency; however, OP and ODP (which have diﬀerent head groups to

ODMS and ODDMS) exhibited greater surface hydrophobicity. Among the SAMs,

PFTS and ODP exhibited the highest contact angle.

17 Self-Assembled Monolayers (SAMs) for Controlling Adhesion 937

Friction force (nN)

25 50075

100

Normal load (μN)

Si substrate

Si (0.06)

PFTS (0.022)

Si (0.06)

ODMS (0.017)

ODDMS (0.018)

Al (0.02)

OP (0.008)

ODP (0.005)

Si substrate

Al substrate

Fig. 17.20. (continued)

(b) friction force vs. normal

load plots for various SAMs

on Si and Al substrates

AFM Adhesion and Friction Measurements under Ambient Conditions

Figure 17.20a gives the adhesive force, friction force and the coeﬃcient of friction

measured under ambient conditionsusing AFM for various SAMs deposited onto Si

and Al substrates, while Fig. 17.20 shows friction force–normalload plots for these

systems [43,47]. Figure 17.21 shows surface height and friction force maps for Si

and PFTS, ODMS and ODDMS on Si [43].

The bare substrates gave much higher adhesive force than the SAM coatings.

ODMS and ODDMS show adhesive forces comparable to OP and ODP despite

their lower water contact angles. These SAMs have the same tail groups, and during

AFM measurements the AFM tip only interacts with the tail groups, whereas the

contact angles can also be inﬂuenced by the head groups in these SAMs. This is

probably why the adhesive forces for these SAMs are comparable. PFTS, which has

one of the highest contact angles, showed the lowest adhesion.

938 Bharat Bhushan

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

1.00

0.75

0.50

0.25

0 0.25 0.50 0.75 1.00

2 nm0

μm μm μm μm

HighLow

Silicon PFTS on Si

ODMS on Si ODDMS ON Si

Surface height Friction force Surface height Friction force

Fig. 17.21. Surface height (left) and friction force (right) maps for (a)Si,PFTS/Si, ODMS/Si

and ODDMS/Si [43]

Friction force images of SAMs are more uniform than those of Si. The coeﬃ-

cient of friction was higher for the bare substrates than for the corresponding SAMs

deposited on them. The SAMs deposited on the Si substrate showed higher coeﬃ-

cients of friction than those deposited on the Al substrates. The primary reason for

this is believed to be the greater roughness of the Al substrates and possibly higher

packing densities of SAMs on Al as mentioned earlier. For the SAMs deposited on

Si substrates, the ones with ﬂuorocarbonbackbones were foundto have highercoef-

ﬁcients of friction than those with hydrocarbon backbones. This might be due to the

higher stiﬀness of the ﬂuorocarbon backbone [44]. It is harder to rotate a ﬂuorocar-

bon backbone because the F atom is larger than the H atom [91]. The C

−

C bonds of

hydrocarbon chains, on the other hand, have more freedom to rotate. We presented

a molecular spring or brush model earlier that explained why less compliant SAMs

exhibit more friction. SAMs with higher spring constants or stiﬀer backbones may

need more energy to be elastically deformed during sliding, so these SAMS show

more friction. In terms of the inﬂuence of the chain length, it has been reported that

the friction coeﬃcients for SAM surfaces decrease with the carbon backbone chain

length (n) up to 12 carbon atoms (n ≈12) [70]. However, this eﬀect of chain length

on the coeﬃcient of friction was not apparent in these data.

Eﬀect of Relative Humidity, Temperature and Sliding Velocity on AFM Adhesion

and Friction

The inﬂuence of relative humidity was studied for various SAMs. Its eﬀects on ad-

hesive force, friction force at a normal load of 5nN, the coeﬃcient of friction and

17 Self-Assembled Monolayers (SAMs) for Controlling Adhesion 939

Adhesive force (nN)

40.0

35.0

30.0

25.0

20.0

15.0

Relative humidity (%)

100

Friction force (nN)

Coefficient of friction

Critical normal load (μN)

Adhesive force (nN)

10.0

0.00

00.0

8.00

6.00

4.00

2.00

0.10

0.08

0.06

0.04

0.02

80.0

60.0

40.0

20.0

120

100

10 20 30 40 50 60 70 80 90

ODDMS/Si

ODMS/Si

PFTS/Si

Normal load = 5 nN

ODP

Al substrate

Fig. 17.22. Relative eﬀect

of humidity on the adhesive

force, the friction force, the

coeﬃcient of friction and the

microwear for various SAMs

on (a) Si substrates [44], and

(b) Al substrates