Bhushan B. Nanotribology and Nanomechanics: An Introduction

Подождите немного. Документ загружается.

1244 Bharat Bhushan

Decrease in surface height (nm)

Si substrate

Wear tests

PFTS

Decrease in surface height (nm)

Si substrate

ODMS

Decrease in surface height (nm)

Normal load (μN)

10 20 30 40 50 60

Al substrate

ODDMS

ODP

ODDMSODMSPFTSSi Al OP ODP

Critical load (μN)

Fig. 22.27. Decrease of sur-

face height as a function of

normal load after one scan

cycle for various SAMs on Si

and Al substrates, and com-

parison of critical loads for

failure during wear tests for

various SAMs

22 Characterization of MEMS/NEMS and BioMEMS/BioNEMS 1245

normal loadfor variousSAMs and correspondingsubstrates [124,164].As shown in

the ﬁgure, the SAMs exhibit a critical normal load beyond which the surface height

drastically decreases. Unlike SAMs, the substrates show a monotonic decrease in

surface height with increasing normal load, with wear initiating fromthe very begin-

ning, i.e., even for low normal loads. The critical loads correspondingto the sudden

failure are shown in Fig. 22.27b. Amongst all the SAMs, ODDMS and ODP show

the best performance in the wear tests. Out of the two alkyl SAMs, ODDMS/Si and

ODP/Al showed better wear resistance than ODMS/Si and OD/Al due to the eﬀect

of chain length. Wear behavior of the SAMs is reported to be mostly determined by

the molecule–substrate bond strengths.

Bhushan et al. [122] and Lee et al. [125] studied various ﬂuoropolymer mul-

tilayers and ﬂuorosilane monolayers on Si and a selected ﬂuorosilane on PDMS

surfaces. For nanoscale devices such as in nanochannels, monolayers are preferred.

They reported that all ﬂuorosilane ﬁlms increased the contact angle. The ﬂuorosi-

lane monolayer 1H, 1H, 2H, 2H–perﬂuorodecyltriethoxysilane (PFDTES) resulted

in a contact angle of about 100

◦

Based on these studies, a perﬂuoro SAM with a compliant layer should have

optimized tribological performance for MEMS/NEMS and BioMEMS/BioNEMS

applications.

22.3.3 Hard Diamond-Like Carbon (DLC) Coatings

Hard amorphous carbon (a-C), commonlyknown as DLC (implying high hardness)

coatings are deposited by a variety of deposition techniques including ﬁltered ca-

thodic arc (FCA), ion beam, electron cyclotron resonance chemical vapor deposi-

tion (ECR-CVD), plasma-enhanced chemical vapor deposition (PECVD), and sput-

tering [136,154]. These coatings are used in a wide range on applications including

tribological, optical, electronic, and biomedical applications. Ultrathin coatings (3–

10nm thick) are employedto protectagainst wear and corrosionin magnetic storage

applications– thin-ﬁlm rigid disks, metal-evaporatedtapes, and thin-ﬁlmread/write

heads –, Gillette Mach 3 razor blades, glass windows, and sunglasses. The coatings

exhibit low friction, high hardness and wear resistance, chemical inertness to both

acids and alkalis, lack of magnetic response, and optical band gaps ranging from

zero to a few eV, depending upon the deposition technique and its conditions. Se-

lected data on DLC coatings relevant for MEMS/NEMS applications is presented

in the following section on adhesion measurements.

22.4 Tribological Studies of Biological Molecules

on Silicon-Based Surfaces and of Coated Polymer Surfaces

22.4.1 Adhesion, Friction, and Wear of Biomolecules on Si-Based Surfaces

Proteins on silicon-based surfaces are of extreme importance in various applica-

tions including silicon microimplants, various bioMEMS such as biosensors, and

1246 Bharat Bhushan

therapeutics. Silicon is a commonly used substrate in microimplants, but it can have

undesired interactions with the human immune system. Therefore, to mimic a bi-

ological surface, protein coatings are used on silicon-based surfaces as a passiva-

tion layer, so that these implants are compatible with the body and avoid rejection.

Whether this surface treatment is applied to a large implant or a bioMEMS, the

function of the protein passivation is obtained from the nanoscale 3D structural

conformation of the protein. Proteins are also used in bioMEMS because of their

function speciﬁcity. For biosensor applications, the extensive array of protein activ-

ities provides a rich supply of operations that may be performed at the nanoscale.

Many antibodies (proteins) have an aﬃnity to speciﬁc protein antigens. For exam-

ple, pathogens(disease causingagents, e.g., virus or bacteria)trigger the production

of antigens which can be detected when bound to a speciﬁc antibody on the biosen-

sor. The speciﬁc binding behavior of proteins that has been applied to laboratory

assays may also be redesigned for in vivo use as sensing elements of a bioMEMS.

The epitope-speciﬁc binding properties of proteins to various antigens are useful

in therapeutics. Adhesion between the protein and substrate aﬀects the reliability

of an application. Among other things, the morphology of the substrate aﬀects the

adhesion. Furthermore, for in vivo environments, the proteins on the biosensor sur-

face should exhibit high wear resistance during direct contact with the tissue and

circulatory blood ﬂow without washing oﬀ.

Bhushan et al. [72] studied the step-by-step morphological changes and the ad-

hesion of a model protein – streptavidin (STA) – on silicon-based surfaces. Fig-

ure 22.28a presents a ﬂowchart showing the sequential modiﬁcation of a silicon sur-

face. In addition to physical adsorption, they also used nanopatterning and chemical

linker methods to improve adhesion. Nanopatterned surfaces contain a large edge

surface area, leading to high surface energy, which results in high adhesion. In the

chemical linker method, sulfo-NHS-biotin was used as a cross linker because the

bonds between the STA and the biotin molecule are some of the strongest nonco-

valent bonds known (Fig. 22.28b). It was connected to the silica surface through

a silane linker, 3-aminopropyltriethoxysilane (3-APTES). In order to make a bond

between the silane linker and the silica surface, the silica surface was hydroxylated.

Bovine serum albumin (BSA) was used before STA in order to block nonspeciﬁc

binding sites of the STA protein with silica surface. Figure 22.29 shows the step-by-

step morphological changes in the silica surface during the deposition process using

the chemical linker method. There is an increase in roughness of the silica sur-

face boiled in de-ionized (DI) water compared to the bare silica surface. After the

silanization process, there are many free silane links on the surface which caused

higher roughness. Once biotin was coated on the silanized surface, the surface be-

came smoother. Finally, after the deposition of STA, surface shows large and small

clumps. Presumably, the large clumps represent BSA and the smaller ones repre-

sent STA. To measure adhesion between STA and the corresponding substrates, an

STA-coated tip (or functionalized tip) was used and all measurements were made

in phosphate buﬀered saline (PBS) solution, a medium commonly used in protein

analysis and to simulate body ﬂuid. Figure 22.30 shows the adhesion values of var-

22 Characterization of MEMS/NEMS and BioMEMS/BioNEMS 1247

Silicon (cleaned)

Silica (thermal oxidation)

Pre-cycle cleaned

STA coated (adsorption)

Boiled in DI water

Silanized

(3-APTES monolayer)

Sulfo-NHS-biotin coated

(bonded to silane)

BSA coated

STA coated

(bonded to biotin)

Method II

Method I A

STA coated (adsorption)

Method I B

Patterned

SiO

NHO

Si CH

NHO

Si CH

NHO

Streptavidin has four biotin-

binding pockets. Two or one

may be attached to the biotin on the surface, with the

remaining 2 or 3 available to bind the biotin analyte.

Fig. 22.28. (a)Flowchart

showing the samples used and

their preparation technique,

and (b) a chemical structure

showing streptavidin protein

binding to the silica sub-

strate by the chemical linker

method

ious surfaces. The adhesion value between biotin and STA was higher than that for

other samples, which is expected.Edges of patterned silica also exhibited high adhe-

sion values. It appears that both nanopatterned surfaces and chemical linker method

increase adhesion with STA.

1248 Bharat Bhushan

Silica boiled

in DI water

σ = 0.12 nm

P–V = 3.0 nm

1 μm

Silanized

(3-APTES

monolayers)

silica

σ = 1.1 nm

P–V = 17.0 nm

After coated

with sulpho-

NHS-biotin

(bonded to

silane)

σ = 0.96 nm

P–V = 15.0 nm

After coated

with BSA

σ = 0.62 nm

P–V = 14.0 nm

After coated

with streptavidin

(bonded to biotin)

at 10 μg/ml

σ = 0.78 nm

P–V = 15.0 nm

Fig. 22.29. Morphological

changes in silica surface dur-

ing functionalization of silica

surface by chemical linker

imaged in PBS. Streptavidin

is covalently bonded at a con-

centration of 10 µg/ml [72]

22 Characterization of MEMS/NEMS and BioMEMS/BioNEMS 1249

Adhesive force (nN)

Unpatterned

silica

Edge of

patterned

silica

Silica boiled

in DI water

Silica coated

with

sulfo-NHS-biotin

3.5

3.0

2.5

2.0

1.5

0.5

Streptavidin

Sample surface

Adhesion measurement in PBS with functionalized tip

AFM tip

Fig. 22.30. Adhesion meas-

urements of silica, patterned

silicon, silica boiled in DI

water, and sulfo-NHS-biotin

using functionalized (with

streptavidin) tips obtained

from force–distance curves,

captured in PBS

Tokach ichu et al. [120] studied friction and wear of STA deposited by physical

adsorption and the chemical linker method. Figure 22.31 shows the coeﬃcient of

friction between the Si

tip and various samples. The coeﬃcient of friction is

less for STA-coated silica samples compared to uncoated sample. The streptavidin

coating acts as a lubricant ﬁlm. The coeﬃcient of friction is found to be depen-

Coefficient of friction

Pre-cycle

cleaned

silica

1 10 100

STA by adsorption

at μg/ml

Silanized

silica

NHS-

biotin

coated

silica

STA by

chemical

linker

10 μg/ml

0.01

0.02

0.03

0.04

Fig. 22.31. Coeﬃcient of fric-

tion for various surfaces with

and without biomolecules

1250 Bharat Bhushan

dent upon the concentration of STA, and decreases with an increase in the concen-

tration. Bhushan et al. [72] have reported that the density and distribution of the

biomolecules vary with concentration. At higher concentration of the solution, the

coated layer is more uniform and the silica substrate surface is highly covered with

biomolecules than at lower concentration. This means that the surface forms a con-

tinuous lubricant ﬁlm at higher concentration.

In the case of samples prepared by the chemical linker method, the coeﬃcient

of friction increases with an increase in the biomolecular chain length due to in-

creased compliance. When normal load is applied on the surface, the surface be-

comes compressed, resulting in a larger contact area between the AFM tip and the

biomolecules. Besides that, the size of STA is much larger than that of APTES and

biotin. This results in a tightly packed surface with the biomolecules, which results

in very little lateral deﬂection of the linker in the case of STA-coated biotin. Due to

this high contact area and low lateral deﬂection the friction force increases for the

same applied normal load compared to directly adsorbed surface. These tests reveal

that surfaces coated with biomolecules reduce the friction, but if the biomolecular

coating of the surface is too thick or the surface has some cushioning eﬀect, as seen

in the chemical linker method, that increases the coeﬃcient of friction.

2.5

–2.5

2.5

–2.5

0 5 nm

(μm)

2.5

–2.5

(nm) (nm) (nm)

(μm)

75% of free amplitude 50% of free amplitude

Cross sectional profiles and heights

30% of free amplitude

Phase images

05°

(μm)

Fig. 22.32. Wear mark images and cross-sectional proﬁles of precycle cleaned silica coated

with streptavidin at 10 µg/ml by physical adsorption at three normal loads (increasing from

left to right)

22 Characterization of MEMS/NEMS and BioMEMS/BioNEMS 1251

Figure 22.32 shows the wear maps of STA deposited by physical adsorption

at three normal loads. The wear depth increases with increasing normal load. An

increase in normal load causes partial damage to the folding structure of the strepta-

vidin molecules. It is unlikely that the chemical (covalent)bonds within the strepta-

vidin molecule are broken; instead, the folding structureis damagedleading to wear

mark. When the load is high, i.e., 30% of the free amplitude(≈ 8nN), the molecules

may have been removedby the AFM tip due to the eﬀect of indentation/ Because of

this, there is a signiﬁcant increase in the wear depth from 50% of the free amplitude

(≈ 6nN) to 30% of the free amplitude (≈ 8nN). The data show that biomolecules

will be damaged during sliding.

22.4.2 Adhesion of Coated Polymer Surfaces

As mentioned in Sect. 22.A, PMMA, PDMS, and other polymers are used in the

construction of micro/nanoﬂuidic-basedbiodevices. Adhesion between the moving

partsneeds to be minimized.Furthermore,if the adhesionbetweenthe microchannel

surface and the bioﬂuid is high, the biomolecules will stick to the microchannel

surface and restrict ﬂow. In order to facilitate ﬂow, surfaces with low bioadhesion

are required.

Tambe and Bhushan [169,170] and Bhushan and Burton [171] have reported

adhesive force data for PMMA and PDMS against an AFM Si

tip and a silicon

ball. Tokachichu and Bhushan [172] measured contact angle and adhesion of bare

PMMA and PDMS and coated with a perﬂuoro SAM of perﬂuorodecyltriethoxysi-

lane(PFDTES). Oxygenplasma treatmentwas usedfor hydroxylationof the surface

to enhance chemical bonding of the SAM to the polymer surface. They made meas-

urements in ambient and in PBS and fetal bovine serum (FBS); the latter is a blood

component. Figs. 22.33 and 22.34 show the contact angle and adhesion data. SAM-

coated surfaces have a high contact angle Fig. 22.33, as expected. The adhesion

valueof PDMS in ambient is high because of electrostatic charge present on the sur-

face. The adhesion values of PDMS are higher thanPMMA because PDMS is softer

Contact angle (deg)

PMMA

Virgin

Oxygen plasma treated

PFDTES coated

120

150

PDMS

Fig. 22.33. Sessile drop

contact-angle measurements

of virgin, oxygen-plasma-

treated and PFDTES-coated

PMMA and PDMS surfaces.

The maximum error in the

data is ±2

◦

[120]

1252 Bharat Bhushan

Adhesive force (nN)

PMMA

Virgin

Oxygen plasma treated

PFDTES coated

PDMS

Adhesive force (nN)

PMMA

Virgin

Oxygen plasma treated

PFDTES coated

PDMS

Adhesive force (nN)

PMMA

Virgin

Oxygen plasma treated

PFDTES coated

PDMS

Fig. 22.34. Adhesion meas-

urement of virgin, oxygen-

plasma-treated and PFDTES-

coated PMMA and PDMS

surfaces with bare silicon ni-

tride AFM tip (a)inambient,

and (b) in PBS environ-

ment, and (c) dip-coated tip

with FBS in a PBS environ-

ment [120]

than PMMA (elastic modulus = 5GPa and hardness = 410MPa [121]) and results

in a higher contact area between the PDMS surface and the AFM tip, and PMMA

does not develop electrostatic charge. When SAM is coated on PMMA and PDMS

surfaces, the adhesion values are similar, which shows that electrostatic charge on

virgin PDMS plays no role when the surface is coated. In the PBS solution, there

22 Characterization of MEMS/NEMS and BioMEMS/BioNEMS 1253

is a decrease in adhesion values because of the lack of a meniscus contribution.

The adhesion values in the FBS-coated tip in PBS are generally lower than for the

uncoated tip in PBS. In summary, the adhesion values of SAM-coated surfaces are

lower than bare surfaces in various environments.

22.5 Nanopatterned Surfaces

22.5.1 Analytical Model and Roughness Optimization

One of the crucial surface properties for surfaces in wet environmentsis nonwetting

or hydrophobicity.It is usually desirable to reduce wetting in ﬂuid ﬂow applications

and some conventional applications, such as glass windows and automotive wind-

shields, in order for liquid to ﬂow away along their surfaces. Reduction of wetting

is also important in reducing meniscus formation, consequently reducing stiction,

friction, and wear. Wetting is characterized by the contact angle, which is the an-

gle between the solid and liquid surfaces. If the liquid wets the surface (referred

to as a wetting liquid or a hydrophilic surface), the value of the contact angle is

0 ≤ θ ≤ 90

◦

, whereas if the liquid does not wet the surface (referred to as a nonwet-

ting liquidor a hydrophobicsurface),the valueof the contactangleis 90

◦

<θ≤180

◦

A surface is considered superhydrophobic if θ is close to 180

◦

. Superhydrophobic

surfaces should also have very low water contact angle hysteresis. One of the ways

to increase the hydrophobic or hydrophilic properties of the surface is to increase

surface roughness. It has been demonstrated experimentally that roughness changes

contact angle. Some natural surfaces, includingleaves of water-repellentplants such

as lotus, are known to be superhydrophobic due to their high roughness and the

presence of a wax coating Fig. 22.35. This phenomenon is called in the literature

the lotus eﬀect [173].

If a droplet of liquid is placed on a smooth surface, the liquid and solid surfaces

come together under equilibrium at a characteristic angle called the static contact

angle θ

; see Fig. 22.36. The contact angle can be determined from the condition of

the total energy of the system being minimized. It can be shown that

cosθ

= dA

/ dA

, (22.1)

where θ

is the contact angle for smooth surface, and A

and A

are the solid–

liquid and liquid–air contact areas. Next, let us consider a rough solid surface with

a typical size of roughness details smaller than the size of the droplet (on the order

of a few hundred microns or larger), Fig. 22.36. For a rough surface, the roughness

aﬀects the contact angle due to the increased contact area A

. For a droplet in

contact with a rough surface without air pockets, referred to as a homogeneous

interface, based on the minimization of the total surface energy of the system, the

contact angle is given as [174]

cosθ = dA

/ dA





(

/ dA

)

= R

cosθ

, (22.2)