Bovine Serum Albumin adhesion force measurements using an Atomic

Force Microscopy

Chun-Chih Lai B. Sc. (Chemical Engineering),

National Chung Hsing University, Taiwan

Faculty of Built Environment and Engineering

Queensland University of Technology

2 George Street, QLD 4000, Brisbane, Australia

A thesis submitted as a requirement for the degree of:

Master of Engineering

At the Queensland University of Technology

2006

i

Abstract

In this thesis, a direct method of Atomic Force Microscopy (AFM) technique has been

developed to measure the adhesion forces between BSA and two different surfaces:

mica (a hydrophilic surface); and polystyrene (a hydrophobic surface); in PBS

solution. We have shown possible to measure interactions between proteins and

substrate surface directly without any modification to the substrate and the AFM tip;

this means protein molecules can keep the natural elastic property within the force

measurements. The average measured value of adhesion forces between BSA and

mica is 0.036 ± 0.002 nN, and between BSA and polystyrene is 0.066 ± 0.003 nN. The

polystyrene surface is more adhesive to BSA than the mica surface. This is consistent

with previous research, which assessed that hydrophobic surfaces enhance protein

adhesion but hydrophilic surfaces do not.

ii

Contents

Abstract ............................................................................ i

List of figures............................................................................ iv

List of tables ...........................................................................vi

Abbreviations ..........................................................................vii

Statement of Original Authorship .......................................... ix

Acknowledgments ..................................................................... x

1. Introduction........................................................................... 1

2. Literature Review ................................................................. 4

2.1. Introduction.....................................................................................................4 2.2. Atomic Force Microscopy (AFM) ..................................................................4

2.2.1. Surface techniques and advantages of using AFM ..............................4 2.2.2. Principles of AFM................................................................................6 2.2.3. Sample requirements............................................................................6

2.3. Adhesion Force Measurements of Proteins with Atomic Force Microscopy .7 2.3.1. Globular protein adsorption on solid surfaces .....................................7 2.3.2. Adhesion forces models of proteins...................................................13 2.3.3. AFM methods for protein-surface adhesion force measurements .....16

2.4. The future potential of atomic force microscopy..........................................22 2.5. Conclusion ....................................................................................................23

3. Experimental Technique .................................................... 25

3.1. AFM ..............................................................................................................25 3.1.1 AFM tip...............................................................................................25 3.1.2. Piezoelectric scanner..........................................................................27 3.1.3. Detection mechanism.........................................................................27 3.1.4. Modes of AFM...................................................................................29

3.2. Force measurement procedure facility..........................................................32

iii

3.2.1. Spring constant calibration ................................................................33 3.2.2. The force measurement......................................................................36 3.2.3. Force curves .......................................................................................38

4. Experimental Method......................................................... 42

4.1. Atomic force microscopy..............................................................................42 4.2. BSA adsorption onto the surfaces .................................................................43

5. Results and Discussion ....................................................... 44

5.1. Experimental results......................................................................................44 5.1.1. Bare mica surface in PBS ..................................................................45 5.1.2. BSA adsorbed on mica surface in PBS ..............................................46 5.1.3. Bare polystyrene in PBS ....................................................................48 5.1.4. BSA adsorbed on polystyrene surface in PBS ...................................49

5.2. Discussions ...................................................................................................50

6. Conclusion ...............................................................................

................................................................................ 59

7. References................................................................................

................................................................................ 61

Appendix A – The AFM force curves of BSA-mica in PBS . 66

Appendix B – The AFM force curves of BSA-polystyrene in

PBS......................................................................... 71

Appendix C – The bare mica force curves in PBS and the

bare polystyrene force curves in PBS.................. 78

Appendix D – Data, SEM images and details of the

calibration procedure for the two cantilevers..... 80

iv

List of figures

Figure 1: Ordering of polypeptide chains into an �–helix (left) and a parallel �–sheet (right) (Norde, 2003). ..................................................9

Figure 2: Schematic representation of the Gouy-Stern model of an electrical double layer, indicating the distribution of the counterions ( ) and co-ions ( ), compensating the negative surface charge and the resulting

potential decay (Norde, 2003)..............................................................11 Figure 3: Schematic representation of a covalently attached BSA molecule

on an AFM tip surface (Wang et al., 2004)..........................................17 Figure 4: Theoretical force-distance curve. ................................................18 Figure 5: Schematic of force-distance curves (Green et al., 2002).............19 Figure 6: Schematic representation of the tip of an AFM (Morris et al., 1999).

..............................................................................................................26 Figure 7: Schematic representation of the atomic force microscopy (Morris

et al., 1999). .........................................................................................28 Figure 8: Force-distance curve in AFM......................................................29 Figure 9: The NT-MDT SMENA liquid head .............................................33 Figure 10: The resonance curve..................................................................34 Figure 11: Resonance curve fitted by Gaussian equation for an AFM

cantilever..............................................................................................35 Figure 12: The photo show how to adjust the laser beam focus on the AFM

cantilever in the liquid. ........................................................................37 Figure 13: The diagram shows the where the force spectroscopy mode is in

the SPM window menu........................................................................37 Figure 14: The indication of how the signal-displacement curves are

converted to cantilever deflection-displacement curves ......................38 Figure 15: Force-distance curves ................................................................39 Figure K1: Force-distance curve. First stage of the approach to the surface

..............................................................................................................40 Figure K2: Force-distance curve: the repulsive region...............................40 Figure H1: Force-distance curve: pulling the cantilever in the repulsive

region ...................................................................................................41 Figure H2: Force-distance curve: the retracting cantilever stretches the BSA

molecules .............................................................................................41 Figure 16: Photo of AFM head and sample assembly during force

measurements.......................................................................................43 Figure 17: Bare mica force curves in PBS..................................................46

v

Figure 18: BSA-mica force-displacement curves in PBS...........................47 Figure 19: Bare polystyrene force curves in PBS.......................................48 Figure 20: BSA-polystyrene force curves in PBS ......................................50 Figure 21: The comparison of adhesion forces between BSA-mica and

BSA-polystyrene..................................................................................52 Figure 22: The comparison of adhesion works between BSA-mica and

BSA-polystyrene..................................................................................57 Figure A1: The adhesion force curves of BSA and mica............................66 Figure A2: The adhesion force curves of BSA and mica............................66 Figure A3: The adhesion force curves of BSA and mica............................67 Figure A4: The adhesion force curves of BSA and mica............................67 Figure A5: The adhesion force curves of BSA and mica............................68 Figure A6: The adhesion force curves of BSA and mica............................68 Figure A7: The adhesion force curves of BSA and mica............................69 Figure A8: The adhesion force curves of BSA and mica............................69 Figure A9: The adhesion force curves of BSA and mica............................70 Figure B1: The adhesion force curves of BSA and polystyrene .................71 Figure B2: The adhesion force curves of BSA and polystyrene .................71 Figure B5: The adhesion force curves of BSA and polystyrene.................73 Figure B6: The adhesion force curves of BSA and polystyrene .................73 Figure B7: The adhesion force curves of BSA and polystyrene .................74 Figure B8: The adhesion force curves of BSA and polystyrene .................74 Figure B9: The adhesion force curves of BSA and polystyrene .................75 Figure B10: The adhesion force curves of BSA and polystyrene ...............75 Figure B11: The adhesion force curves of BSA and polystyrene ...............76 Figure B12: The adhesion force curves of BSA and polystyrene ...............76 Figure B13: The adhesion force curves of BSA and polystyrene ...............77 Figure C1: no interactions were observed from the bare mica surface. .....78 Figure C2: no interactions were observed from the bare mica surface. .....78 Figure C3: no interactions were observed from the bare PS surface..........79 Figure C4: no interactions were observed from the bare PS surface..........79 Figure D1: One of resonance curves of cantilever A is fitted by Gaussian

equation................................................................................................80 Figure D2: The SEM image of cantilever A. ..............................................82 Figure D3: The SEM image of cantilever B. ..............................................83 Figure D4: One of resonance curves of cantilever A is fitted by Gaussian

equation................................................................................................84

vi

List of tables

Table 1: The information of Cantilever A and Cantilever B (See Appendix D for

experimental details)………………………………………………....44

Table 2: The table shows the maximum of adhesion force values of BSA-mica

and BSA-polystyrene (Errors of the average force values are the

standard deviations)……………………………………………….…51

Table 3: The student t test of adhesion forces of BSA-mica (Group A) and

BSA-polystyrene (Group B) (the student t test was calculated by

KaleidaGraph, Synergy software)…………………………………....52

Table 4: The adhesion forces between BSA and hydrophilic surfaces………..54

Table 5: The adhesion forces between BSA and hydrophobic surfaces………54

Table 6: The table shows the overall work for pulling BSA off the mica surface

and the overall work for pulling BSA off the polystyrene…………...55

Table D1: The resonance frequency values and the Q factor values were obtained

and calculated from the 10 resonance frequency curves of cantilever A

...….......................................................................................................81

Table D2: The resonance frequency values and the Q factor values were obtained

and calculated from the 10 resonance frequency curves of cantilever B

….…………………………………………………………………….85

vii

Abbreviations

Φ I the electrostatic potential

ρ∞ the bulk concentration of electrolytes

Φe exterior electrostatic potential

ρi the charge distribution

-Lg -lactoglobulin ( -Lg)

�fi (fc) the hydrodynamic function of the cantilever

εi the dielectric permittivity of the protein interior

AFM atomic force microscopy

Al32 Hamaker constant

b the width of the AFM cantilever

BSA bovine serum albumin

d the displacement AFM cantilever

DLVO Derjaguin-Landau-Verweey-Overbeek

dtip the deflection of the AFM tip

E the electronic charge

ESCA electron spectroscopy for chemical analysis

F force

f the bandwidth of the resonance frequency curve

f1 the lower cutoff frequency

f2 the upper cutoff frequecny

fc the resonance frequency

viii

I the value of current signal

k Boltzmann constant

kz the spring constant of the AFM cantilever

L the length of the AFM cantilever

PBS phosphate-buffered saline

Qf the quality factor of the resonance frequency fc

R the radius of the sphere

SAMs self-assembled monolayers

SEM scanning electron microscopy

SFA surface force apparatus

T the temperature

XPS X-Ray photoelectron spectroscopy

Wmica adhesion work between BSA and mica Wpolystyrene adhesion work between BSA and polystyrene

Zi the ion valency

Z the nearest distance between the sphere and the plane

� the density of the fluid

ix

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To the best

of my knowledge and belief, the thesis contains no material previously published or

written by another person except where due reference is made.

Chun-Chih Lai

x

Acknowledgments

I thankfully acknowledge my supervisor Professor John M. Bell for the guidance,

support, encouragement within my entire Master’s study.

I also would like to appreciate Professor Nunzio Motta, Dr. Thor Bostrom and Dr.

Edeline Wentrup-Byrne for their great helps and instructions regarding the AFM

technique. Thanks also to Dr. Cameron Wilson for his advice regarding protein

properties.

I would like to thank Mr. Loc Duong for his support with SEM. Thanks to Miss

Christina Theodoropoulos for her help with SEM, protein solution preparation.

Finally, I would like to appreciate my parents and my wife for their constant support

and encouragement.

1

1. Introduction

Many fundamental studies in recent few years have dealt with the adsorption

behaviour of proteins onto different surfaces using various experimental methods and

instruments. Adhesion properties between protein and surfaces in solutions have been

a significant area of research in biology and biotechnology providing researchers with

the key information to attached cells and living tissues onto surface. This is of critical

importance in medical applications like bone implants, prosthesis, dentures, etc.

Protein adsorption plays a vital role in determining the nature of the tissue-implant

interface, since the adsorbed proteins can significantly affect biomaterial surface

properties such as blood coagulation and cell adhesion. Bovine Serum Albumin (BSA)

is a globular protein and the most abundant protein in plasma. Although the structure

and the adsorption of BSA have been investigated with various instruments such as

X-Ray photoelectron spectroscopy (XPS) (Fitzpatrick, Luckham, Eriksen, &

Hammond, 1992), the adhesion interactions between BSA and the different surface

(mica-hydrophilic and polystyrene-hydrophobic) measured directly by AFM in PBS

are still not well understood.

Developed in 1986 Atomic force microscopy (AFM) has become quickly a versatile

and unique technique to probe interactions between proteins and surfaces in a real

physiological environment (Morris et al., 1999). Since then, more and more

researchers are using AFM to explore further the protein adsorption in solution at the

nanometer (10-9m) scale. The forces involved in the adhesion between protein and the

surface are the order of 10-9 N (or 1 nN) and AFM probes are perfectly suited to

measure forces in this range and in real time.

2

In this work, a direct method of identifying the interaction between protein and

surface properties by AFM has been established. We used AFM to measure the

adhesion forces between BSA and two different surfaces (mica and polystyrene)

without any modification of AFM tip or the substrate surface; this means protein

molecules can keep the natural elastic property during the force measurements. The

force-distance curves allow us to compare the interactions between BSA and different

surface properties from the quantified adhesion forces. In future, this AFM technique

can be exploited to study protein adhesion to new synthetic biomaterials promoting a

nanotechnology approach to tissue engineering research.

The remainder of this thesis consists of five chapters. Those five chapters are: � Chapter 2: Literature review – The main findings in the area of protein

adhesion are reviewed, with reference to the most important publication in the field. The main work that has guided our research is the fundamental book of Israelachvili, “Intermolecular and surface forces” (Israelachvili, 1992).

� Chapter 3: Experimental technique – Atomic Force Microscopy is explained

here, starting from the basic concepts, and developing the practical realization of a Force-distance experiment, including the spring constant calibration of the cantilever.

� Chapter 4: Experimental methodology – The experimental realization, sample

preparation, and measurement technique have been performed and described in this chapter.

� Chapter 5: Results and discussion – The results of the Force-distance

measurements are presented and discussed, explaining in details the physical interaction between AFM tip, BSA protein and surface.

3

� Chapter 6: Conclusion - The AFM technique has applied successfully to identify the protein adhesions with the hydrophobic and hydrophilic surfaces.

� References

� Appendix A – All adhesion force curves of BSA and mica in PBS

� Appendix B – All adhesion force curves of BSA and polystyrene in PBS

� Appendix C – Two examples of adhesion force curves of bare mica in PBS and

two examples of adhesion force curves of bare polystyrene in

PBS.

� Appendix D – Data, SEM images and details of the calibration procedure for the

two cantilevers used in this study.

4

2. Literature Review

2.1. Introduction

Scientists and engineers are now utilizing atomic force microscopy (AFM) to develop

new techniques to understand the adsorption mechanisms of proteins and the adhesion

interactions between proteins and surfaces. The goal of this chapter is to discuss

atomic force microscopy-based techniques for evaluating protein adhesion on

biomaterial surfaces.

This discussion includes:

� Advantages and principles of using atomic force microscopy;

� Basic theories of protein adsorption on solid surfaces;

� Atomic force microscopy technique to measure protein adhesion forces;

� Previous research and the issues uncovered in this research;

� Future potential of atomic force microscopy.

2.2. Atomic Force Microscopy (AFM)

2.2.1. Surface techniques and advantages of using AFM

Surface analytical techniques and methods in biomaterial surfaces provide

information about the surface characteristics related to biological responses (Dee,

2002). For example, contact angle analysis can measure the angle of contact (θ)

between a liquid and a surface, and give information on the interaction about a

particular liquid with a surface. If the liquid is water, a smaller contact angle is a

5

hydrophilic surface, on which water spreads to a greater extent; a larger contact

angle indicates a hydrophobic surface, on which water beads (Dee, 2002). Scanning

electron microscopy (SEM) can give images of surface topography. X-Ray

photoelectron spectroscopy (XPS, also called electron spectroscopy for chemical

analysis, ESCA) is based on the process of photoemission and measures the

distribution of electrons for each energetic level in atoms, molecules and solids. By

using x-rays to eject electrons from inner-shell orbital, this technique is widely used

for accurate chemical analysis of the surfaces. The surface force apparatus (SFA) is

another surface analytical instrument that can measure the surface forces between two

mica sheets. One of the surfaces is held fixed and the other is mounted on a spring to

measure the force between the surfaces. However, the main drawbacks of SFA are

lack of spatial resolution, the restriction to mica surfaces, and the jump-to-contact

instability in the force measurement due to large interacting surface areas (Patil et al.,

2005).

More recently, the introduction of atomic force microscopy (Binning & Quate, 1986)

has transformed surface science. This versatile microscopy, that scans the surface with

a tiny tip attached to a cantilever, has crucial advantages that other surface techniques

do not have, namely, to image the surface and to interact with it on a nanometer scale.

This allows researchers to measure the interactions between the surface and molecules

by modifying the surface or the tip of AFM and to analyse electrochemical

phenomenon. In addition, AFM can be used in vacuum, air and liquid environments.

Particularly, biologists can make use of AFM to observe samples with the fluid cell

under physiological conditions (Zhang, Bremmell, Kumar & Smart, 2004).

Furthermore, the observation and analysis of the interactions between proteins and

samples can be achieved and more detailed dynamic information can be obtained

6

using AFM. That is why AFM has been become a uniquely valuable device for

biomaterial researchers. Consequently, this literature review will focus on the

principles, functions and sample requirements of AFM, protein adsorption in the

aqueous environment, the theories of adhesion behaviours, adhesion forces

measurements between proteins and several substrates with AFM, and the future

potential of using AFM in the biomaterial research area.

2.2.2. Principles of AFM

Atomic force microscopy is a technique designed to investigate both conductors and

insulators on an atomic scale. Using the variation of forces between the surface and

the AFM tip can, in principle, measure any type of force, not only the interatomic

forces arising from atomic potentials (attractive and repulsive), but electromagnetic

forces arising from charged surfaces or particles as well (Binning & Quate, 1986).

2.2.3. Sample requirements

Contemporary AFMs allow any sample size to be used (Jandt, 2001), but the

restrictions of the scanners limit the typical imaging area from 1 1 nm2 to 250

250 m2. The z-direction range of AFM scanners is usually limited to around 8 m,

hence, rougher or bent samples cannot be imaged. Samples should be smooth and

clean.

For high-resolution imaging of molecules, flat substrates are needed. For example, if

the diameter of a protein is about a few nanometers (Scheuring, 2001), but the

substrate also has a roughness of a few nanometers, we may not be able to resolve the

7

protein in AFM height image, because it is hidden by the surface topography. There

are a number of experiments involving imaging single molecules on different flat

substrates, for instance, silicon (Ortega-Vinuesa, Tengvall & Lundstrom, 1998), glass

(Fritz et al., 1995), HOPG (Ta, Sykes & McDermott, 1998) and mica (Quist, Bjorck,

Reimann, Oscarsson & Sundqvist, 1995), that show good sub nanometer resolution.

2.3. Adhesion Force Measurements of Proteins with Atomic Force Microscopy

AFM has been used by many investigators to carry out adhesion force measurements

of proteins and material surfaces. The AFM cantilever can be functionalized at the end

with particles of specific characteristic to measure surface forces between the

“particle” and the surface. Duker et al. (1991) introduced this method by attaching a

silica colloidal sphere (3.5 m) on the end of AFM tip, and they also performed the

force measurements between silica surfaces with the “colloid” probe as a function of

surface separation, salt concentration, and pH (Duker, Senden & Pashley, 1992). This

technique can be used in the measurement of other forces between different colloid

particles and surfaces. This section discusses the theories of globular protein

adsorption, the methods of achieving protein adhesion force measurements, previous

studies and current issues.

2.3.1. Globular protein adsorption on solid surfaces

The adsorption behaviour of proteins plays a crucial role in determining the nature of

the tissue-implant interface, since the adsorbed proteins can significantly have an

effect on biomaterial surface properties. Furthermore, the focus of protein adsorption

research has often been on the rearrangement of protein molecules. The reason is not

8

only its relevance to the biological functioning of the molecules, but also the essential

importance in the mechanism of the adsorption process. Additionally, although

globular proteins, for example bovine serum albumin (BSA), represent only a

minority of the available protein mass, they form the greatest proportion of protein

species. Many applications and implications are relevant to the adsorption of the

globular proteins, for instance, biomedical engineering, biosensors and immunological

test systems (Norde, 2003). As a result, the following focuses on the adsorption of

globular proteins.

2.3.1.1. The stability of globular proteins in aqueous solution

The adsorption process depends on intramolecular forces within the protein molecules

that might lead to an alteration of protein conformation (Valle-Delgado et al., 2004).

Several factors determine the protein conformation: conformational entropy of the

protein; hydrophobic interactions; Lifshitz-van der Waals interactions; and hydrogen

bonding. The conformational entropy of the protein and hydrophobic interactions are

the major factors, because their opposing effects, the comparable magnitude in the

Gibbs energy, are much larger than other factors (Norde, 2003). These factors are

mentioned briefly below (Hlady & Buijs, 1996; Norde, 2003).

Conformational entropy

Most globular proteins contain a large amount of �-helices and �-sheets (Fig. 1).

These structures are stabilized by hydrogen bonds between the peptide units in the

polypeptide backbone. These intramolecular hydrogen bonds decrease the rotational

mobility of the bonds in the polypeptide chain and, therefore, the conformational

entropy. Thus, the stable structure is possible only if interactions within the protein

9

molecule and interactions between the protein molecules and its environment are

sufficiently favourable to compensate for the low conformational entropy (Norde,

2003).

Figure 1: Ordering of polypeptide chains into an �–helix (left) and a parallel �–sheet (right) (Norde, 2003).

Hydrophobic interactions

Protein molecules contain both polar and nonpolar sections. Dehydration of nonpolar

components under aqueous conditions causes a rise of the entropy of the water

molecules released and, as a result, in the lowering of the Gibbs free energy. This

hydrophobic dehydration results in polar parts of the polypeptide to link in water. The

significance of hydrophobic dehydration for protein folding was first introduced in a

10

previous study (Kauzmann, 1959), and it is now regarded as the primary driving force

for the folding process.

Lifshitz-van der Waals interactions

Lifshitz-van der Waals interactions come from interactions between fixed and/or

induced dipoles, and are very sensitive to the distance of separation L between the

dipoles. Upon folding the polypeptide chain into a compact structure, the entire effect

of dipolar interactions for promoting protein stability is not unambiguous, although it

is generally assumed that due to the relatively high packing density, dipolar

interactions tend to enhance a compact structure (Nir, 1977).

Hydrogen bonds

Most of the hydrogen bonds in globular proteins are between the amide and carbonyl

groups of the polypeptide backbone (Fig. 1). In �-helices and �-sheets such intrachain

hydrogen bonds reinforce each other since they are aligned more or less parallel to

one another. Although the net effect on the stabilization of the protein structure is still

unclear, the hydrogen bonds may be a part of the stabilizing protein structure if the

polypeptide chain and side chain that are able to form hydrogen bonds are forced into

the nonpolar interior of a compact protein (Norde, 2003).

2.3.1.2. Globular protein adsorption from aqueous solution onto solid surfaces

Although globular protein adsorption on a solid surface is an important research field

not only for a number of biological systems but also in the development of diverse

biomedical applications, some aspects of protein adsorption are still not well

understood, such as the extent and the rate of protein conformational change (Hlady &

Buijs, 1996). In this section we discuss some of the forces which come into play in the

11

globular protein adhesion process on solid surfaces.

Electrostatic protein-surface interaction

In an aqueous environment the protein molecule and the surface are electrically

charged, and there is an effect of electrostatic protein-surface interactions on the

behaviour of protein adsorption. When the surface and the protein move toward each

other, their electrical double layers overlap (Fig. 2) and give rise to electrostatic

interactions, as illustrated in Figure 2.

Figure 2: Schematic representation of the Gouy-Stern model of an electrical double layer, indicating the distribution of the counterions ( ) and co-ions ( ), compensating the negative

surface charge and the resulting potential decay (Norde, 2003).

There are a few papers that emphasize the effect of the heterogeneous charge

distribution in the protein molecule (Roth & Lenhoff, 1993, 1995; Yoon & Lenhoff,

1992). Itoh et al. (1995) showed that the adsorption behaviour of -lactoglobulin (

-Lg) on a stainless steel surface at pH 6.85 at 75 was altered by chemical

modification of the carboxyl and amino groups in -Lg molecules. When -Lg

was charged negatively by the chemical modification of the carboxyl groups, the

amount of negatively charged -Lg adsorbed was larger than that of native -Lg.

12

On the other hand, when -Lg was charged positviely by the chemical modification

of the amino groups, -Lg was not adsorbed on the same surface. The discoveries

suggest that electrostatic interactions do affect the adsorption of -Lg onto a surface

with a negative charge.

Hydration alterations

As mentioned above, protein molecules contain both polar and nonpolar moieties. In

an aqueous environment the globular proteins tend to be buried in the inner part of the

molecule in which they are shielded from contact with water. Early works (Gorbunov,

Lukyanov, Pasechnik & Vakhrushev, 1986; Regnier, 1987) have demonstrated that the

hydrophobicity of the protein influences its adsorption. On the other hand, protein

adsorption behaviour may be not only influenced by the hydrophobicity of the protein

surface, but the overall hydrophobicity of the protein may also be relevant.

Effects of ionic concentration on protein adsorption behaviour

The effect of ionic solution concentration and pH value has also been examined (Chen

et al., 1998; Frank & Belfort, 2000; Vengasandra, Sethumadhavan, Yan & Wang,

2003). Vengasandra et al. (2003) utilized AFM to measure the alteration of protein

binding rate by electrolyte. They discovered that when an electrolyte is present in the

medium, both long-range interactions (protein-surface) and short-range interactions

(protein-protein) require a longer time for bonding formation (Vengasandra et al.,

2003).

13

2.3.2. Adhesion forces models of proteins

2.3.2.1. Derjaguin-Landau-Verweey-Overbeek model

Derjaguin-Landau-Verweey-Overbeek forces

The Derjaguin-Landau-Verweey-Overbeek (DLVO) theory has been used to define

the stability of dispersions for decades. In the DLVO theory, the net balance of

attractive and repulsive potential energies at any point in the separation between

simple model particles (or surfaces) is calculated, based upon knowledge of

parameters including the particle surface potential, the dielectric constant and ionic

strength of the medium, and the effective “Lifhsitz-Hamaker constant” of the particle

in that medium (Rowe, 2001). The Lifhsitz-Hamaker constant defines the strength of

the van der Waals forces, and the particle attraction.

At the solid-liquid interface, electrostatic double-layer forces are always present

between charged surfaces in electrolyte solutions. As mentioned above, the protein

molecule and the surface are electrically charged in an aqueous environment, and

there is an effect of electrostatic protein-surface interactions on the behaviour of

protein adsorption. Also, DLVO theory assumes that the stability of colloidal

suspensions in aqueous electrolyte solutions is controlled by long-range forces

between the particles (or surfaces), particularly in the balance between attractive van

der Waals and repulsive electrostatic forces (Leckband & Sivasankar, 1999). The van

der Waals and the electrostatic forces have been shown to control protein aggregation,

crystallization, and adsorption (Marra & Israelachvili, 1984). The expression for the

van der Waals interaction energy between a sphere and an infinite plane is:

14

���

���

��

�

�

++

++−=∆

ZRZ

ZRR

ZRA

F planesphere 2ln

)2(6132

/

(1)

where R is the radius of the sphere, Z is the nearest distance between the sphere and

the plane, and Al32 is the Hamaker constant, essentially a material property (Roth &

Lenhoff, 1993). The electrostatic potential ΦΦΦΦ i within the protein is described by the

Poisson equation:

i

ii

ερ−=Φ∇ 2 (2)

where ρρρρi represents the charge distribution and εεεεi the dielectric permittivity of the

protein interior, with its dielectric constant generally taken to be in the range from

2 to 4 (Roth & Lenhoff, 1993). For an exterior electrostatic potential ΦΦΦΦ e is described

by the linearized Poisson-Boltzmann equation:

ee k Φ=Φ∇ 22

(3)

where κκκκ2 is kT

Zieeε

ρ ∞222, e is the electronic charge, Zi the ion valency, k the

Boltzmann constant, T the temperature, and ρρρρ∞∞∞∞ the bulk concentration of electrolytes

before dissociation.

Non-DLVO forces

Some studies confirmed that the DLVO theory describes the long-range interactions

between proteins, as mentioned above. On the other hand, direct force measurements

also quantified the non-DLVO forces (e.g. hydrodynamic and solvation forces) to the

15

entire force fields that mediate protein interactions (Helm, Knoll & Israelachvili, 1991;

Leckband, Schmitt, Israelachvili & Knoll, 1994).

The hydrodynamic force occurs in the drainage of the thin film of fluid between the

small particles, and the 2-10 nm dimensions of most proteins are substantially smaller

than colloidal particles (diameters 10 nm). Thus, hydrodynamic force must also be

considered relevant to the magnitudes of the double-layer and van der Waals forces

between them. It should be noted that it is not the same as the friction experienced by

the protein as it translates through the medium (Leckband & Sivasankar, 1999).

Solvation forces or hydration forces occur when molecules must overcome the

obstacle caused by the water layer adsorbed on the surface of the macromolecule. The

energy for displacing this water regulates the self-assembly of protein aggregates and

other biomolecules. The raised intermolecular repulsion because of solvation layers is

a well-accepted phenomenon. Further, the exclusion of water from protein interfaces

stabilizes the interaction. This solvent-mediated attraction is associated with the

disruption of hydrogen bonds between water molecules by the non-polar solute

(Israelachvili & Weenerstrom, 1996; Leikin, Parsegian, Rau & Rand, 1993).

To sum up, although the DLVO theory can be used in biological colloids, protein

interactions are complicated because short-range forces have a significant impact on

the rates and strengths of protein associations when involve, for example, solvation

and hydrodynamic forces. Therefore, in order to describe the solution behavior of

proteins, DLVO theory and other forces’ fields (e.g. non-DLVO forces) must be

considered (Leckband & Sivasankar, 1999).

16

2.3.3. AFM methods for protein-surface adhesion force measurements

Adhesion force measurements between protein and material surfaces can be generally

divided into two parts which are the preparation of AFM tips and the adhesion force

measurement itself. To prepare the AFM tips the researchers immobilized the proteins

onto the AFM cantilever tips (Chen et al., 1998; Sethuraman, Han, Kane & Belfort,

2004; Wang, Palmer, Schwartz & Razatos, 2004). Then, adhesion force measurements

are conducted by using the protein-functionalized AFM tip to approach and retract

from the surface. Adhesion force values can be obtained from AFM force-distance

curves.

2.3.3.1. Modifications of the AFM tip

There are methods of controlling the surface topography and chemistry using

self-assembled monolayers (SAMs) as substrates. SAMs not only imitate the chemical

properties of a surface, but also have the ability to immobilize adsorbed molecules,

which is advantageous for AFM imaging of single molecules. Therefore, a few

approaches using SAMs have enabled the measurement of the interactions between

proteins and substrates (Sethuraman et al., 2004; Wang et al., 2004).

Wang et al. (2004) evaluated protein adhesion to biomaterials with the AFM. First of

all, the AFM cantilevers (Si3N4) were coated with gold. Secondly, SAMs were

attached on the gold-coated surface of the AFM tips. Finally, bovine serum albumin

(BSA) was immobilized on the SAM covering the AFM tip surface (Fig. 3).

17

Figure 3: Schematic representation of a covalently attached BSA molecule on an AFM tip surface (Wang et al., 2004).

Chen et al. (1998) used another way to coat protein onto AFM tip. Silicon-nitride

AFM cantilevers were cleaned and oxidized with oxygen plasma before being

immersed into a toluene solution of 3-aminopropyldimethylethoxysilane for two

hours. Then, the cantilevers were rinsed in methanol followed by potassium

phosphate buffer before being immersed into aqueous solutions of glutaraldehyde for

half an hour and then moved into BSA solutions for an hour.

2.3.3.2. Adhesion force measurements

AFM is often used to obtain both topography and chemical and mechanical properties

of a surface. In addition, AFM has the capability to perform force measurements. The

tip on the cantilever of the AFM is a force sensor with a sensitivity of pN (Ruger &

Hansma, 1990). Consequently, AFM can be utilized to perform the measurement of

force-distance curves, providing quantitative information on forces between the tip

and the sample as a function of tip-sample distance.

The total tip-sample force is the sum of two components: a long-range attractive force

and a short-range repulsive force. As the atoms of the tip and sample are brought

18

together they initially weakly attract each other. This attractive force increases until

the interacting atoms are so close that their electron clouds begin to repel one another

electrostatically. This electrostatic repulsion continues to weaken the attractive force

as the separation distance decreases. The attractive force goes to zero (i.e., approaches

the limit of zero) when the distance comes within the length of a chemical bond (a

few angstroms). Once the total van der Waals forces (repulsive forces) are positive,

the atoms are in “contact”.

Figure 4: Theoretical force-distance curve.

A practical example of an acquisition of a force-distance curve is shown in Figure 5.

During the approach, when the sample-tip distance is large the cantilever is

undeflected (I). At a certain point the tip jumps into contact with the sample, because

the surface exerts an attractive force on the tip (II). As the tip is moved closer to the

surface, a repulsive force starts to overcome the attraction (contact region) and the

force-distance curve has a linear behaviour (II-III). When the tip is removed from the

sample surface, the repulsive force on the tip decreases (III-IV). The adhesive force

between the tip and the sample keeps the tip still in contact with the sample and

breaks off when the maximum adhesion force is overcome (IV

19

Figure 5: Schematic of force-distance curves (Green et al., 2002).

Cantilever calibration

Hooke’s law (Eq. 4) can be applied to calculate the force (F) of interaction between

the tip and the sample. Although the spring constant is defined by the manufacturer,

the actual spring constant may show variations from this value by an order of

magnitude (Green et al., 2002). Therefore, it is necessary to calibrate the spring

constant of each cantilever. Several methods of calibrating the spring constant have

developed by measuring the resonant frequency of the cantilever with a little mass

attached to the tip (Cleveland, Manne, Bocek & Hansma, 1993), measuring its free

oscillation (Sader, Larson, Mulvaney & Lee R, 1995) or measuring the thermal noise

(Hutter & Bechhoefer, 1993). The details of the procedure that has been used in this

research (free oscillation) will be explained in the chapter regarding experimental

techniques.

20

Previous research

Since AFM has been invented in 1986, several AFM techniques have been developed

and applied to measure the interactions between proteins and surface, but those

techniques still have drawbacks need to be considered. Three previous researches

described below can represent general AFM techniques about the adhesion force

measurement between proteins and the substrate surface. The advantages and the

disadvantages of these techniques have also been addressed.

Adhesion forces between individual ligand-receptor pairs

Adhesive force measurements of ligand-receptor interactions have been performed

(Florin, Moy & Gaub, 1994) by using a avidin (receptor)-functionalized tip to probe a

biotinylated (ligand) agarose bead. Under the condition of positioning the AFM tip on

the top of the agarose bead by an eplilight microscope, the area between the AFM tip

and the bead allowed only a limited number of molecular pairs to interact, and the

unbinding forces of individual molecular pairs have been obtained.

The advantage of this method would gain an unbinding force of a molecular pairs due

to the limited interaction area between the AFM tip and the substrate surface, but the

disadvantage would be because the AFM tip (silicon nitride) is sharp and much harder

than the protein molecules, the AFM tip may destroy and penetrate the protein

molecules until the tip is hit the surface during the force measurement. Hence, the

interaction which has been measured could be only the interaction between the AFM

tip and the agarose bead.

21

Interactions between proteins and different substrates by atomic force

microscopy

Sagvolden et al. (1998) showed a method capable of measuring the adhesion forces of

proteins to solid surfaces. Proteins are first attached to glass micro spheres that are

subsequently adsorbed to various substrates. After a relaxation time, the proteins

attached micro spheres are dislodged using the AFM cantilever and the force values

are obtained by the AFM cantilever deflection. The experimental results have shown

proteins adsorb better on a hydrophobic polystyrene substrate than hydrophilic glass

surfaces. Additionally, those researchers found when proteins were cross-linked with

glutaraldehyde on the surface, the adhesion forces of cross-linked proteins were lower

than the adhesion forces of natural proteins, which suggested using glutaraldehyde to

immobilize proteins on the surface may change the nature of the interaction between

proteins and surface. The main disadvantages of this technique are described below:

1. Because the mass of a micro sphere is larger than protein molecules, it would cause

damage on the flexibility of the protein molecules attached on the micro spheres when

the micro spheres have arrived on the substrate. Hence, due to the damaged flexibility

of the protein molecules, the measured adhesion interactions would be not the

“nature” interactions between proteins and the substrate.

2. Because the protein molecules may not attach to the micro sphere surface evenly,

the adhesion interactions which have been obtained may not the interactions between

proteins and the substrate surface.

22

Interactions between proteins adsorbed on different substrates using atomic

force microscopy with a colloid probe

Valle-Delgado et al. (2004) used a colloid AFM probe technique which was attached a

micro sphere with a diameter of 5 µm to the end of a AFM cantilever to measure the

interaction forces between bovine serum albumin (BSA) layers adsorbed on different

substrates (silica and polystyrene) at a function of pH and slat concentration. The

experimental results were fit to theoretical predictions in DLVO theory satisfactorily

and obtained an opportunity to calculate the hydration forces between proteins.

The colloid probe technique is to use a micro sphere attached to the end of the AFM

cantilever instead of the sharp AFM tip, and it has advantages that the sharp tip may

not have. For example, the proteins functionalized colloid probe can measure the

interactions between the proteins and the surface without penetrating the protein

molecules during the action of approaching to the surface. However, after attaching

the micro sphere to the end of AFM cantilever, the original spring constant of the

AFM cantilever can be changed, which may obtain inaccurate force measurements.

Hence, the spring constant of the colloid probe needs to be calibrated after the micro

sphere has been glued to the end of AFM cantilever.

2.4. The future potential of atomic force microscopy

Since the invention of AFM in 1986, researchers have been trying to utilize AFM to

understand surface interactions in a wide variety of situations. Carbon nanotubes that

have diameters as small as 5 nm can improve the lateral resolution, and they can be

employed in fine organic or biological samples because they have a low force

constant that makes them strong but limiting the maximum force (Green et al., 2002).

23

Some authors have used the AFM tip functionalized with nanotubes to obtain the

high-resolution surface images (Woolley, Cheung, Hafner & Lieber, 2000). Cho and

Sigmund. (2002) suggested that using a multiwalled carbon nanotube (MWNT) as a

nanosized probe can overcome the limitation of an AFM tip for studying nanoparticles,

due to the small diameter, low surface roughness and minimization of subsidiary

effects.

2.5. Conclusion

For years the adsorption of globular proteins plays a significant role and has been

investigated in many research areas, for example, biomedical engineering. Globular

protein adsorption onto a solid surface is influenced mainly by conformation entropy

and hydrophobic interactions, and electrostatic and hydrophobic interactions play the

major roles in determining the behaviour of globular protein adsorption on solid

surfaces. Because Bovine serum albumin (BSA) is globular proteins and has been

often studied for the adsorption of proteins, BSA was chosen to be used in this thesis.

Since 1986 atomic force microscopy has been utilized to develop techniques that can

give researchers an opportunity to explore and understand the adhesion interactions of

the protein molecules with different surfaces at the atomic level. Although the AFM

techniques have several advantages that other surface instruments can not achieve (e.g.

the spatial images, various ranges of substrates and real time measurements in liquid

condition), they still have few disadvantages, for example, the protein attached AFM

tip may not measure the “real” interaction between proteins and proteins, because the

“hard and sharp” AFM tip can penetrate the “soft” protein molecules during the force

measurement. Hence, in order to eliminate these disadvantages, we have established

24

an AFM technique which not only can keep the nature flexibility of proteins during

the force measurement, but also can measure directly and simply the “real” adhesion

interactions between proteins and the substrate surface. The following chapter will

describe and explain how this novel AFM technique works.

25

3. Experimental Technique

This chapter introduces the principles of atomic force microscopy (AFM) and the

methods of carrying out force measurements.

3.1. AFM

Atomic force microscopy is a technique designed to investigate both conductors and

insulators on an atomic scale by exploiting the forces between the surface and the

AFM tip. By keeping constant the force while the tip sweeps over the surface, it is

possible to obtain microscopic images of the surface with nearly atomic resolution.

Generally the force considered in AFM is the repulsive or attractive force generated

by the interaction of the two atomic potentials of the tip and of the surface. However

it is possible in principle to use any type of force, not only the interatomic forces, but

electromagnetic forces as well (Binning & Quate, 1986). In force spectroscopy mode

the AFM can measure the values of these forces as a function of the tip-sample

distance. The following section discusses the main features of AFM and explains how

it can achieve such high and fine resolution.

3.1.1 AFM tip

The most important part of AFM is the tip, because AFM images samples by sensing,

rather than by looking as other microscopes do (in the extended sense also SEM and

TEM “look” at the sample, since electrons behave like waves as photons do), and

the tip is the sensing tool (Morris, Gunning & Kirby, 1999). An AFM tip is shown

below in the electron micrograph (Fig. 6).

26

Figure 6: Schematic representation of the tip of an AFM (Morris et al., 1999).

The AFM tip is a micro-fabricated, particularly sharp point made of a hard material

(usually Si, SiO2 or SiC) installed on the end of a cantilever, a little bar of Si or

similar material, that must have good elastic properties in the used range. The

cantilever allows the tip to move up and down in z-direction, and it has a force

constant suited to respond to the force between the tip and the sample. The resolution

of an AFM depends strongly on the shape of the tip (Wiesendanger, 1995). The

smaller and sharper the tip is, the smaller is the surface area sampled by the tip (Jandt,

2001). The deflection of the cantilever as a spring can be converted to force,

according to Hooke’s law:

F = - kzd (4)

Where kz is the force constant of the cantilever and d is its displacement. The minus

sign indicates that the force is opposite to the displacement of the tip.

27

3.1.2. Piezoelectric scanner

The second crucial component of the AFM is the piezoelectric scanner. Generally,

AFM uses one of two fundamental scanning mechanisms, namely, scanning the

sample or scanning the tip. These two scanning mechanisms both rely on a

piezoelectric scanner. The scanner is usually a tube made of zirconate titanate

ceramics (PZT) that converts electric signals produced by the control electronics into

the mechanical scanning action along x-y and z directions. The distance in the

z-direction between the probe and the sample is usually kept constant by a feedback

mechanism in the AFM controller (constant force mode). A detailed mathematical

study (Taylor, 1993) of the dynamics of tube scanners (with respect to a piezoelectric

tripod) has addressed both their advantages and disadvantages. One of the few

disadvantages is the motion of the end of the tube which drives the tip or sample in

the x-y directions, tracing out an arc rather than a straight line, causing an effect

known as “eyeballing” when it is scanning on a large area.

3.1.3. Detection mechanism

The final feature of AFM is the detection mechanism. When scanning across the

sample, the deflection of the tip should be monitored. There are several different ways

of detecting the motion of the tip. The most common method, the optical lever system,

is illustrated below (Figure 7):

28

Figure 7: Schematic representation of the atomic force microscopy (Morris et al., 1999).

A laser beam is concentrated on the end of the tip, and reflected from the back of the

tip to a photodiode detector that is divided into four segments. While the tip is

travelling on the surface, the angle of laser reflection from the tip is changing as well,

and the laser reflection that is falling onto the photodiode detector is producing

changes in intensity in each of its quadrants. This chain of action affects the

mechanical amplifier that is sensitive enough to detect the tip movement at atomic

scale during scanning. The difference in laser intensity between the top two segments

and the bottom two segments generate the electrical signal that quantifies the z

deflection (up and down) of the tip, while the difference in the laser intensity between

the right and left segments can be used to quantify the lateral deflection of the tip.

29

3.1.4. Modes of AFM

Generally, AFM has three operating modes for imaging: the contact mode,

semi-contact mode and non-contact mode. In all three modes the cantilever is

scanning across the sample surface, and the topography of a sample is exhibited by

interpreting the signal of the cantilever deflection. By approaching the surface (Fig. 8),

the tip feels an attractive force (non-contact region) due to the long-range part of the

atomic potential up to an equilibrium point where the attractive force (negative)

reaches its maximum (minimum in the force-distance curve). Over this point the

repulsive component of the atomic potential (due to the repulsion of the atomic clouds

connected to the Fermi principle) starts to be larger than the attractive part. This is the

contact region. The region around the minimum is used in the intermittent contact

mode (tapping mode).

Figure 8: Force-distance curve in AFM

30

Contact mode

Using the tip in contact mode may be appropriate when the samples are sufficiently

hard (e.g. titanium) because the tip is in close contact with the sample. For softer

samples this mode may destroy the surface: for example, when studying the weakly

adsorbed biomolecules on surfaces. An advantage of the scanning of contact mode is

that it can be executed in air, vacuum and fluid environments.

Care must be taken to keep the tip always in close contact: this is obtained by using an

appropriate feedback to maintain the deflection signal at the set point value. The set

point value should not be too far from the zero deflection level (i.e. the tip should not

be pressed too much on the surface) because this might result in the destruction of the

tip and of the surface. On the other hand, a too small difference between the value of

set point and the zero deflection level could cause instabilities in the tracking system

(Morris et al., 1999).

The contact mode can also be a basis for the operation of other contact techniques,

because the contact mode uses the tip deflection, which can be converted according to

Hooke’s law (Equation 1) into a net force (Hartmann, 1998). The net force is

composed of a long-range attractive interaction between the probe and the sample and

a short-range repulsive interaction between the probe and the sample (Hartmann,

1998). With an appropriate detection system (four quadrant photodetector) lateral

forces can be measured. Since the tip is in stable contact with the surface, the

scanning motion results in some torsion of the tip in the lateral x-y plane that also

reflects the frictional properties of a surface (Jandt, 2001).

31

Non-contact mode

The non-contact mode of AFM is used to study the long-range attractive forces and

the tip never actually touches the sample surface during the scan. However, the forces

applied to the tip are extremely low, resulting in low deformation and shear. As a

result, this is the most suitable method for soft samples (e.g. polymers) since the risk

of possible destruction of the sample surface is greatly reduced. Choosing a flexible

tip with a low resonant frequency is important since van der Waals forces would have

little deflective effect on a stiff cantilever (Morris et al., 1999). The non-contact mode

can be used to test various tip-sample interactions such as magnetic and electrostatic

interactions. For this purpose, a special coating of the tip is necessary. For example,

magnetic force imaging can provide information about the surface magnetization of

the sample, hence, the tip has to be coated with a thin magnetic film.

Semi-contact mode

The semi-contact mode is also known as the “tapping mode”. In the semi-contact

mode the tip is oscillating (frequencies in air 50-500 kHz, in fluids approximately 10

kHz) with amplitude of several tens of nm. This oscillation is driven by the cantilever

piezodriver, so that the tip approaches the surface several thousands of times per

second, slightly touching or “tapping” the surface (Jandt, 2001). The contact time

between the tip and the sample is greatly reduced by using the semi-contact mode,

and the friction force is minimized while the tip is travelling over the surface. This

should prevent the tip from attracting or sweeping the molecules on the surface

(Putman, Van der Werf, De Grooth, Van Hulst & Greve, 1994). For this reason

biological samples are mostly scanned by the semi-contact mode. In the semi-contact

mode, phase imaging can be achieved, because the various material properties lead to

a phase shift of the tip oscillation, relative to the signal sent to the cantilever

32

piezodriver. Therefore, using phase imaging we can also acquire some characteristic

properties of the sample, for example, surface adhesion and viscoelasticity.

Force spectroscopy mode

Additionally, AFM not only has the function for the surface imaging, but also has the

function for studying different kinds of surface forces. The function for AFM to

measure the surface forces is called force spectroscopy mode. The AFM force

spectroscopy mode is based on the changes of the cantilever deflection as the AFM tip

approaches to the surface or retracts from it, and the AFM cantilever deflection can be

converted according to Hooke’s law (Equation 1) into a force value(Hartmann, 1998).

Heniz, et al. (1999) has explained the AFM force spectroscopy mode can be used to

study the adhesion interactions between proteins and the substrate surface. Hence, in

this work the AFM force spectroscopy mode has been utilized to establish the

technique of estimating the interactions between proteins and different surface

properties from the force-distance curves.

3.2. Force measurement procedure facility

The atomic force microscopy is a NT-MDT SMENA model equipped with a

50µ scanner. The SMENA is a stand-alone STM/AFM head that allows performing

measurements virtually on any surface. It is equipped also of an optional liquid head

which permits the dipping of the AFM cantilever in a liquid solution. This head has

been used extensively in this work.

33

Figure 9: The NT-MDT SMENA liquid head

Sample preparation

The sample is prepared in an appropriate size (1 cm2) and glued to a substrate which

is mounted on a metallic sample holder. The sample holder is then immersed into a

liquid cell (a small Petri dish filled with the appropriate solution) leaning on a

magnetic stage, which has the role of keeping the sample holder still.

AFM can detect forces in the range of 10-9 Newton (nano-Newton). There are three

main steps to obtain a reliable force measurement. The first is the calibration of the

AFM cantilever spring constant, the second is the force measurement, and the final

step is the conversion from photodiode current to force.

3.2.1. Spring constant calibration

An AFM cantilever can be considered as a spring Hooke’s law (Eq. 4) can therefore

be applied to calculate the forces applied to the AFM tip. Since the shape of all the

AFM cantilevers in this study is a rectangular, the method developed by Sader et al.

(1999) is particularly suitable. The equation used is:

34

kz = 0.1906 � b2 L Qf fc2 �

fi (�f) (5)

Where � is the density of the fluid, b and L are, respectively, width and length of the

AFM cantilever, Qf is the quality factor of the resonance frequency fc, and �fi (fc) is

the hydrodynamic function of the cantilever. The Qf is defined as the resonance

frequency (center frequency) fc divided by the bandwidth �f:

Qf = fc / (f2-f1) = fc /�f (6)

f2 is the upper and f1 the lower cutoff frequency. On a graph (Fig. 10) of response

versus frequency, the bandwidth is defined as the 3 dB change in level (voltage,

current, or power) on either side of the centre frequency. The bandwidth is not the

same as the "full width at half maximum" or FWHM. It is necessary to measure the

resonance frequency of the thermal oscillations, without applying any exciting

voltage. This is performed by leaving the cantilever far enough from the surface and

setting the amplification of the lock-in detector to 106. A typical frequency curve is

shown in Figure 10 below:

Figure 10: The resonance curve

35

Error calculations

In order to evaluate the accuracy of our measurements, we have to compute the error

in the cantilever calibration. The Q factor (fc/∆fc) and the standard deviation of the

resonance frequency (fc) of an AFM cantilever can be obtained by fitting a Gaussian

equation to the resonance frequency curve. An example is shown below (Fig. 11):

Figure 11: Resonance curve fitted by Gaussian equation for an AFM cantilever

The total relative error on the spring constant can be obtained by error propagation.

The errors for Q factor and resonance frequency are obtained from the Gaussian fit.

The length and width of the AFM cantilever can be measured on the SEM image

based on the scale bar, and the error (0.79 ) is obtained from some SEM calibrations.

Therefore, the errors of the measured length (∆∆∆∆L) and width (∆∆∆∆b) of an AFM

cantilever can be calculated from the equations

0

0.5

1

1.5

2

5 10 15 20 25 Frequency (kHz)

y = Ao+A1*exp(-(x-fc)2/σσσσ2)

Error Value 0.0043 0.1144 Ao 0.019 1.331 A1 0.013 13.231 fc 0.019 1.119 σσσσ

0.984 Chisq

0.981 R

• Amplitude

−−−− Gaussian fitted

13.231

36

∆∆∆∆L = 0.0079 L (7a)

∆∆∆∆b = 0.0079 b (7b)

The relative error of the AFM cantilever spring constant (∆∆∆∆kz), by assuming no error

on the fluid density � is:

∆∆∆∆kz / kz = 2 (∆∆∆∆b / b) + ∆∆∆∆L / L + ∆∆∆∆ Qf / Qf + 2 (∆∆∆∆ fc / fc) (8)

3.2.1.1. Experimental works

The AFM cantilever A was chosen to do the bare mica adhesion force measurements

and the BSA-mica adhesion force measurements. The AFM cantilever B was chosen

to do the bare polystyrene adhesion force measurements and the BSA-polystyrene

adhesion force measurements. We have obtained 10 resonance frequency curves of

cantilever A and 10 resonance frequency curves of cantilever B, respectively. Also, we

fitted the Gaussian equation to those resonance frequency curves of cantilever A and

cantilever B. Appendix D contains all data, SEM images and details of the calibration

procedure for the two cantilevers.

3.2.2. The force measurement

Before the force measurement, the AFM has to be set in the contact mode. The laser

beam which is reflected from the back of the cantilever to the photodiode should be

previously adjusted in order to reach the four quadrant sensor. After the immersion in

the liquid (usually in a Petri dish) the laser beam needs to be readjusted. This can be

achieved by placing the Petri dish on a stand so that the laser can be seen across the

glass on a white paper placed on the table (Fig. 12).

37

Figure 12: The photo show how to adjust the laser beam focus on the AFM cantilever in the liquid.

The force measurements are conducted by the AFM force spectroscopy mode. After

the AFM cantilever is landed in the contact mode, the force spectroscopy mode can be

opened by pressing the “operation” button in the SPM window menu (Fig. 13).

Figure 13: The diagram shows the where the force spectroscopy mode is in the SPM window menu.

The laser is seen across the glass on the white paper.

38

3.2.3. Force curves

During the force measurement the force spectroscopy mode allows the AFM

cantilever to approach the surface and then retracts from it at a fixed speed. In this

way the interaction between the tip and the surface is detected. The raw signal is

current (nA) versus displacement (nm). Fig. 14 indicates how the deflection

signal-displacement curves can be transferred to the cantilever

deflection-displacement curves:

Figure 14: The indication of how the signal-displacement curves are converted to cantilever deflection-displacement curves

By using the slope of the approach curve m we can write:

dtip = I / m (9)

Where dtip (nm) is the deflection of the AFM tip, I is the value of current signal

(nA). Once the spring constant kz and the cantilever deflection are obtained via

equation 9, Hooke’s law (equation 4) can be applied to calculate the force between the AFM tip and the sample.

-4

-3

-2

-1

0

1

2

3

-200 -100 0 100 200 300 400

Mica current-displacement curves 0428Tip approach (nA) Tip retract (nA)

Displacement (nm)

m = ∆∆∆∆I/∆∆∆∆x ∆∆∆∆I

∆∆∆∆X

39

Figure 15: Force-distance curves

The current-displacement curves have been converted to the force curves (Fig 15).

The following will describe how the force curves describe the AFM tip behaviours of

picking up BSA molecules.

-20

-15

-10

-5

0

5

10

15

20

-100 0 100 200 300 400

Mica force 0428

Tip approach (nN) Tip retract (nN)

Displacement (nm)

40

AFM tip approaching

When the AFM tip starts to approach the surface, at a distance of about 150 nm, the

cantilever should begin to bend downward before the actual surface is attained at

about 5-20 nm. This is caused by an attraction between the surface and the tip.

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-20 0 20 40 60 80 100 120

Tip approach (nN)

Distance (nm)

Figure K1: Force-distance curve. First stage of the approach to the surface

However, in water the electrostatic forces (like Van der Waals) are greatly reduced and

it is quite difficult to see the bending. The final stage of the approach is a repulsive

behaviour, with the force increasing as the tip is pushed towards the surface; this is

typical of a rigid substrate, showing that the tip reaches the actual mica surface.

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-20 0 20 40 60 80 100 120

Tip approach (nN)

Distance (nm)

Figure K2: Force-distance curve: the repulsive region

41

AFM tip retracting

When the AFM tip is raised, the force distance curve reflects initially the elastic

behaviour of the cantilever leaving the mica surface.

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-20 0 20 40 60 80 100 120

Tip retract (nN)

Distance (nm) Figure H1: Force-distance curve: pulling the cantilever in the repulsive region

Then the BSA molecules attach to the tip start again to pull downward, being

stretched by the cantilever (Fig H2).

-0.15

-0.1

-0.05

0

0.05

0.1

0.15

0.2

-20 0 20 40 60 80 100 120

Tip retract (nN)

Distance (nm)

Figure H2: Force-distance curve: the retracting cantilever stretches the BSA molecules

In conclusions, we have been able to calibrate the instrument and to measure the

forces acting between the tip and the mica surface in liquids. We will show the

complete set of results in next chapter.

42

4. Experimental Method

In this work we have developed a technique to directly identify the interactions

between protein molecules and the substrate surface by using AFM in liquid. In

literature hydrophobic interactions play a significant role in the protein adhesion

which means the adhesion forces between proteins and a hydrophobic surface are

larger than the adhesion forces between proteins and a hydrophilic surface. Therefore,

we chose Bovine serum albumin (BSA) as the proteins and mica and Polystyrene as

substrates. Because mica is hydrophilic (Leonenko, Z. V., Carnini, A., Cramb D.

T., 2000), and polystyrene is hydrophobic, the adhesion force values between BSA

and two different surfaces would be obvious comparable.

4.1. Atomic force microscopy

Our AFM (NT-MDT Solver P47, Moscow, Russia) was equipped with a CSG-11

350µm length AFM cantilever (Silicon). All AFM force measurements were

conducted in phosphate-buffered saline (PBS) buffer solution at 22 using the AFM

SMENA liquid head. All accessories were cleaned by rinsing in ethanol and deionised

water before using the liquid head. The spring constants of two AFM cantilevers were

calibrated by using the method of Sader et al. (1999) as discussed in the previous

chapter. The sample was fixed to a magnetic sample holder by adhesive tape, and the

ensemble was carefully placed into a glass dish with sufficient amount of PBS buffer

solution to completely cover the sample surface (Fig. 16). All experiments were

carried out in the Analytical Electron Microscopy Facility (AEMF) where a constant

temperature of 22° is maintained.

43

Figure 16: Photo of AFM head and sample assembly during force measurements.

4.2. BSA adsorption onto the surfaces

Bovine serum albumin (BSA, A2153-10G, Sigma-Aldrich, Australia) is a kind of

globular proteins and a single polypeptide chain consisting of about 583 amino acid

residues and no carbohydrates. We used the same species of BSA for our experiments

in this work.

A cleaved mica (muscovite) surface was dipped into the solution of 10 gL-1 BSA

(A2153-10G, Sigma-Aldrich, Australia) with 10 ml 1M phosphate-buffered saline

(PBS) buffer solution (pH 7.4) for 3 hours. The solution of 10 gL-1 BSA (A2153-10G,

Sigma-Aldrich, Australia) with 10 ml 1M phosphate-buffered saline (PBS) buffer

solution (pH 7.4) was in a polystyrene 55mm petri dish (S5514UV10, Techno-Plas

PTY LTD, Australia) for 3 hours and covered the whole area of the polystyrene dish

bottom. The mica surface and the polystyrene surface were rinsed with appropriate

amount of deionised water to clear loosely bound BSA before the AFM experiment.

SMENA head

Glass dish

44

5. Results and Discussion

5.1. Experimental results

Four sets of AFM adhesion force measurements have been performed:

1. bare mica (expand PBS in full) (PBS)

2. BSA adsorbed mica surface in PBS

3. bare polystyrene in PBS

4. BSA adsorbed polystyrene surface in PBS

A CSG-11 AFM cantilever (Cantilever A) was chosen for the experiment on mica,

while a different cantilever (Cantilever B) was chosen to carry out the force

measurements on polystyrene. While belonging to the same batch, and having the

same dimensions, the two cantilevers have slightly different properties. The measured

values of length, width, Q factor, resonance frequency and spring constant value for

the two cantilevers are shown in Table 1.

CSG-11 Length

(µm)

Width

(µm) Q factor

Resonant

Frequency

(kHz)

The spring

constant (N/m)

Cantilever A

(mica) 343±3 40.0±0.3 12.0±2.5 13.25±0.03 0.026±0.006

Cantilever B

(polystyrene) 347±3 35.3±0.2 13.2±2.9 12.90±0.08 0.025±0.006

Table 1: The information of Cantilever A and Cantilever B (See Appendix D for experimental details)

45

The results of the force-distance measurements are obtained as a current (the

deflection of the AFM cantilever), which is converted into a force using the method

described in Chapter 3. These have been displayed as force vs displacement curves..

The X-axis (displacement) represents that the displacement between the tip and the

substrate surface, and the Y-axis represents that the force values measured when the

tip was approaching or retracting from the substrate surface. The red curve describes

the force actions of when the AFM tip was approaching to the substrate surface, and

the blue curve describes the force actions of when the AFM tip was retracting from

the substrate surface. In this work we are interested in the protein adhesions to

different surfaces, so we focus on the results on the pull-off force curve of each force

measurement.

5.1.1. Bare mica surface in PBS

Figure 17 contains one of the many Force curves (more than 100) we have acquired

on bare mica surface in PBS solution. All measurements appear very similar, like the

one reported in Fig 17. Upon approach, no force is detected until a short-range

repulsive force is observed starting at about 10 nm; upon retract no hysteresis is

observed and the retracting line is perfectly superposed to the approach line. No

attractive interactions were observed.

46

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm)

Figure 17: Bare mica force curves in PBS

5.1.2. BSA adsorbed on mica surface in PBS

A total of 120 force-displacement curves were measured at four different places on the

same mica surface in PBS. Owing to the small fraction of the surface covered, only 9

adhesion force measurements between the BSA molecules and the mica surface were

recorded (See Appendix A). One of those ten adhesion force measurements is shown

in Fig. 18 as an example. The approach curve (red) shows that when AFM tip was

approaching the surface a repulsive force raised up smoothly compared to the

approach curve in Figure 17, which suggests the AFM tip hits an elastic layer of BSA

molecules before hitting the mica surface (Lin, S., Chen, J. L., Huang, L. S., & Lin H.

W., 2005). The retraction curve shows an adhesion of about 0.03 nN, followed by

47

several jumps of about 0.01nN each. We have observed that when the AFM tip starts

retracting from the mica surface, the repulsive force decreases dramatically. Then, a

short-range attractive force is generated by the BSA molecules attached to the tip and

this causes jumps of about 0.01 nN, possibly corresponding to the unfolding of the

proteins (Florin, Moy & Gaub, 1994; Heinz, W. F., Hoh, J. H. 1999). Due to the

elastic properties of the BSA molecules we think that the AFM tip stretches one or

more BSA molecules, and they detach one at a time until the tip or the BSA detaches

completely. At this point the tip returns to its original position entirely. The average

adhesion force value of the interactions between the BSA molecules and the mica

surface measured by our experiment is 0.036 nN, and the standard error is 0.002nN.

-0.1

-0.05

0

0.05

0.1

0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure 18: BSA-mica force-displacement curves in PBS

48

5.1.3. Bare polystyrene in PBS

The force curves (Fig. 19) show the interactions between the AFM tip and the bare

polystyrene surface in PBS solution. In the approach part (red), a short-range

repulsive force was detected at a distance of ~ 25 nm, and the repulsive force had

began to decrease dramatically since the AFM tip started retracting from the

polystyrene surface. No interactions were observed.

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

TIp approach (nN)Tip retract (nN)

Displacement (nm) Figure 19: Bare polystyrene force curves in PBS

49

5.1.4. BSA adsorbed on polystyrene surface in PBS

A total of 100 force measurements were measured at four different places on the same

polystyrene surface in PBS, and only 13 adhesion events were recorded (See

Appendix B). An example of the force curves between the BSA molecules and the

polystyrene surface is below (Fig. 20). Once the AFM tip is close to the surface a

repulsive force raises up smoothly compared to the approach curve in Figure 19,

which suggests to his the elastic BSA molecules (Lin, S., Chen, J. L., Huang, L. S., &

Lin H. W., 2005). Then, during the retraction the repulsive force decreased

dramatically, similarly to what occurs to the bare polystyrene surface, but suddenly

jumps are detected. We have fitted the elastic properties of the BSA molecules by

fitting the small curves appearing like jumps which are possibly the unbinding forces

of BSA molecules (Florin, Moy & Gaub, 1994; Heinz, W. F., Hoh, J. H. 1999).

We believe that the AFM tip is unfolding the BSA molecules until the AFM tip had

picked the BSA molecules completely and returned to its original position. The mean

adhesion force value of the interactions between the BSA molecules and the

polystyrene surface is 0.066 nN, and the standard error is 0.003 nN.

50

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100

Tip approach (nN)Tip retract (nN)

Displacement (nm)

Figure 20: BSA-polystyrene force curves in PBS

5.2. Discussions

Table 2 has listed the maximum of measured adhesion force of BSA-mica (Group A)

and BSA-polystyrene (Group B). Two groups were analyzed statistically by an

unpaired Student t test. The statistical analysis of the results shows clearly that the

adhesion force between BSA and polystyrene is twice the adhesion force between

BSA and mica, and the mean difference of the mean adhesion forces between the

BSA-mica and the BSA-polystyrene is about 0.03 nN (Table 3). The t Probability is

smaller than 0.0001 which means the mean difference between Group A and Group B

at 99.9999 % level is highly significant, and this shows our quantitative experimental

results are statistically significant. Figure 21 shows the comparison of adhesion forces

between BSA-mica and BSA-polystyrene and a clear separation between the group of

BSA-polystyrene data and the group of BSA-mica data. Because during the adhesion

51

force measurements we used the same source of the AFM cantilevers and BSA, the

difference of the measured adhesion forces between BSA-mica and BSA-polystyrene

could be due to the interactions between BSA and different surfaces, not the

interactions between the AFM tip and BSA. Furthermore, hydrophobic interaction is a

strong attractive force between non-polar species interacting in an aqueous solution

(Tilton, R. D., Robertson, C. R., & Gast, A. P., 1991; Castner, D. G. & Ratner, B. D.,

2002), and in the literature review it has been discussed that the hydrophobic

interaction is the main driving force of the protein molecules adsorption onto the

surface and the stabilization of a compact protein structure due to the decreasing of

free energy (Norde, W., 1986).

Item

Maximum of adhesion force values between BSA and mica (nN) in each force measurement (Group A)

Maximum of adhesion force values between BSA and polystyrene (nN) in each force measurement (Group B)

1 0.028 0.057

2 0.029 0.083 3 0.030 0.068 4 0.032 0.082 5 0.037 0.058 6 0.042 0.065 7 0.046 0.086 8 0.043 0.063 9 0.036 0.061

10 0.053 11 0.056 12 0.060 13 0.060

Table 2: The table shows the maximum of adhesion force values of BSA-mica and

BSA-polystyrene.

52

Group A (BSA-mica) Group B (BSA-polystyrene) Count 9 13 Mean 0.036 0.066

Variance 0.00004 0.0001 Standard Deviation 0.007 0.01

Mean Difference -0.03 Degrees of Freedom 19

t Value -7.85 t Probability 0.0001

Table 3: The student t test of adhesion forces of BSA-mica (Group A) and

BSA-polystyrene (Group B) (the student t test was calculated by KaleidaGraph,

Synergy Sofeware).

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0 2 4 6 8 10 12 14

BSA-mica adhesion forceMean BSA-mica adhesion forceMean - standard deviation (BSA-mica) Mean + standard deviation (BSA-mica)BSA-PS adhesion forceMean BSA-PS adhesion forceMean - standard deviation (BSA-PS)Mean + standard deviation (BSA-PS)

Counts

Figure 21: The comparison of adhesion forces between BSA-mica and BSA-polystyrene.

53

Ying et al. (2002) modified silicon surfaces as hydrophilic and hydrophobic surfaces.

They found the surface concentration of adsorbed BSA on the hydrophobic surface

was twice the surface concentration of adsorbed BSA on the hydrophilic surface. In

other words, the adhesion force between the BSA molecules and the hydrophobic

surface should be larger than the adhesion force between the BSA molecules and the

hydrophilic surface. In table 4 and table 5 Chen et al. (1998) used BSA-functionalized

AFM tip to measure the adhesion forces with –NH2 SAM surface (a hydrophilic

surface) and -CH3 SAM surface (a hydrophobic surface). They also found that the

adhesion force between BSA and -CH3 SAM surface are four times larger than the

adhesion force between BSA and –NH2 SAM surface due to the hydrophobic

interaction between BSA and –CH3 SAM surface. The AFM cantilevers they used

were stiffer than the AFM cantilevers we used due to the different spring constants,

the magnitudes of their average adhesion forces are larger than the magnitudes of our

average adhesion forces.

Sagvolden et al. (1998) developed a method capable of measuring the adhesion forces

of proteins to solid surfaces. Proteins were first attached to glass microspheres. After a

certain time, the BSA-coated microspheres were spread on glass surface and

polystyrene surface in buffer solution. The BSA-coated glass microspheres were

dislodged using the AFM cantilever, and the force values were obtained by the AFM

cantilever deflections. The results had larger adhesion forces between BSA-coated

microsphere and polystyrene surface than the adhesion forces between BSA-coated

microsphere and glass surface as well because of the hydrophobic interactions (Table

4 and Table 5). However, the technique they developed still has disadvantages. For

example, the AFM cantilever could need to overcome the total mass of the

BSA-coated glass microspheres, the friction forces between the microspheres and the

54

surface and the interactions between BSA and the surface to dislodge the BSA-coated

glass microspheres, so the measured adhesion forces may not be the direct

interactions between BSA and the surface. When BSA molecules are at a hydrophobic

polystyrene surface in aqueous solution, due to the hydrophobic interactions

BSA-coated microspheres may become sticky to adhere to each other and spread not

evenly on the surface. Thus, the AFM cantilever may not dislodge only one

microsphere at once within each force measurement, and the measured adhesion

forces could be greatly larger (Table 5).

On AFM

tip On substrate

Average forces (nN)

AFM cantilever spring constants

(N/m) Reference

None BSA adsorbed mica surface 0.036 ± 0.002 0.026 Lai (2006)

BSA -NH2 terminated SAM 4.6 0.32 Chen et al.

(1998)

None BSA adsorbed glass

microspheres-glass surface 0.07 0.3

Sagvolden

(1998)

Table 4: The adhesion forces between BSA and hydrophilic surfaces.

On AFM

tip On substrate

Average forces (nN)

AFM cantilever spring constants

(N/m) Reference

None BSA adsorbed polystyrene

surface 0.066 ± 0.003 0.025 Lai (2006)

BSA -CH3 terminated SAM 19 ± 9 0.32 Chen et al.

(1998)

None

BSA adsorbed glass

microspheres-polystyrene

surface

99 0.3 Sagvolden

(1998)

Table 5: The adhesion forces between BSA and hydrophobic surfaces.

55

From the table 2 the adhesion forces between BSA and mica surface or polystyrene

can be obtained. However, the values of adhesion forces only can identify the protein

affinity to the surface, but they can not characterize the measured adhesion

interactions which are protein-surface interactions or the combinations of the

protein-surface interactions and the unfolding of protein molecules. Thus, in order to

clarify the adhesion interactions in the force curves we have integrated the adhesion

works from each detected adhesion force curves.

Item The overall work of pulling off BSA from

mica surface (aJ)

The overall adhesion work of pulling off BSA from

polystyrene surface (aJ)

1 0.25 ± 0.02 1.08 ± 0.24

2 0.29 ± 0.04 2.19 ± 0.35

3 0.60 ± 0.01 2.43 ± 0.32

4 0.30 ± 0.01 1.80 ± 0.05

5 0.20 ± 0.05 1.51 ± 0.04

6 0.34 ± 0.04 0.84 ± 0.04

7 0.31 ± 0.06 4.06 ± 0.08

8 0.48 ± 0.02 0.68 ± 0.02

9 0.34 ± 0.02 2.67 ± 0.07

10 1.23 ± 0.05

11 1.20 ± 0.02

12 1.09 ± 0.01

13 0.71 ± 0.02

Mean 0.35 1.65

Standard deviation 0.12 0.97

Table 6: The table shows the overall work for pulling BSA off the mica surface and the overall work for pulling BSA off the polystyrene.

56

In table 6 the overall (Wmica and Wpolystyrene) work required to pull BSA molecules off

the surface has been integrated from each force curves and obtained the mean values

by KaleidaGraph, Synergy Sofeware. Also, for calculating the relative errors of each

adhesion work value we calculated the difference between the integral for the original

data and the integral for the data which was actually the moving average of the

original data over 5 points. The relative errors of each BSA-mica adhesion work are

in the range of 3 % ~ 25 % of each original BSA-mica adhesion work data, and the

relative errors of each BSA-polystyrene adhesion work are in the range of 1 % ~ 22 %

of each original BSA-polystyrene adhesion work data.

In figure 22 the mean and standard deviation of Wmica are 0.35 ± 0.12 aJ, and the third

Wmica and the eighth Wmica are larger than the mean Wmica. This is because the third

Wmica and the eighth Wmica may not only include the energy of pulling BSA off the

mica surface, but also including the energy of detaching BSA from the AFM tip. The

mean and standard deviation of Wpolystyrene are 1.65 ± 0.97 aJ. Because the AFM tip

could be unfolding the BSA molecules, the seventh Wpolystyrene in figure 22 is twice

the mean Wpolystyrene and probably includes the conformational energy of BSA

molecules. The ninth Wpolystyrene is also larger than the mean Wpolystyrene, which could

be also due to the combination energy of BSA-polystyrene interactions and

BSA-AFM tip. Figure 22 shows the mean Wpolystyrene is obviously larger than the

mean Wmica, which also demonstrates that on polystyrene surface the AFM tip needs

larger energy to pull off BSA molecules due to the hydrophobic interactions. This

result is also consistent with the results of the adhesion forces of BSA-mica and

BSA-polystyrene.

57

Figure 22: The comparison of adhesion works between BSA-mica and BSA-polystyrene.

Generally, our AFM technique provides us to compare the protein adhesions with two

different surfaces in a simple and direct way. The results have also proved that by

using our AFM technique we are able to investigate protein adhesion force

measurements without damaging the natural elastic property of protein molecules, and

it is an advantage over other AFM techniques described in the section of previous

research. Nevertheless, our AFM technique still has the disadvantages which need to

improve. For example, because the AFM cantilevers we chose have the low spring

constants, during the protein adhesion force measurements the softer AFM cantilever

may be too sensitive in aqueous solution and detect the needless interactions. Also, if

0

1

2

3

4

0 2 4 6 8 10 12 14

BSA-mica adhesion work

Relative error Mean BSA-mica adhesion work Mean-standard deviation (BSA-mica) Mean+standard deviation (BSA-mica)

BSA-PS adhesion work Relative error Mean BSA-PS adhesion work Mean-standard deviation (BSA-PS) Mean+standard deviation (BSA-PS)

Counts

58

we can blunt our sharp AFM tip, it may be able to measure the protein-surface

interactions more easily because of the stronger protein attachment onto a blunt AFM

tip than onto a sharp AFM tip within the adhesion force measurements.

59

6. Conclusion

Protein adsorption plays a significant role in determining the nature of the

tissue-implant interface, because the adsorbed proteins can importantly affect

biomaterial surface properties. For decades many researchers have dedicated to

discover the protein adsorption action with different surface, but no surface analysis

techniques such as SFA can observe the real-time protein interactions with the

different desired substrate surfaces in liquid environment at nano scale until the AFM

was invented in 1986. The AFM not only can provide the atomic resolution surface

image, but also offer the ability that can measure the interaction forces between

particles. Hence, we have used the AFM outstanding advantage to develop the AFM

technique that can measure the interactions between proteins and the surface.

The AFM technique we have established allows obtaining the direct measurements of

the interactions between protein and surfaces and consists of three main parts. At first,

in order to obtain more precise results of the AFM force measurements we have used

the method of Sader et al. (1999) to calibrate the AFM cantilever spring constant.

Secondly, in the force spectroscopy mode of NT-MDT SMENA head the AFM allows

us to obtain the relationships between the protein molecules and the surface properties

by using the AFM tip to pick up proteins from the surface. Thirdly, we convert the

AFM cantilever deflection signal (amplitude) curves to the force (nN) curves by using

the equation 9.

In this work, we selected the same proteins (BSA) adsorbed onto mica (hydrophilic)

and polystyrene (hydrophobic). The results show that BSA molecules prefer to adsorb

onto polystyrene surface than mica surface, and they are also consistent to the

60

previous observations (Castner, D. G. & Ratner, B. D., 2002) that hydrophobic

surfaces promote protein adhesion but hydrophilic surfaces do not. Consequently, the

AFM technique we have developed allow to measure and compare directly the

adhesions between proteins and different surfaces without any modification to the

substrate or to the AFM tip; this means we can keep the nature elastic property of

protein molecules during force measurements.

61

7. References Binning, G., & Quate, G. (1986). Atomic force microscopy. Physical review letters, 56,

930-933. Castner, D. G. & Ratner, B. D. (2002). Biomedical surface science: foundations to

frontiers. Surface Science, 500, 28-60. Chen, X., Patel, N., Davies, M. C., Roberts, C. J., Tendler, S. J. B., Williams, P. M., et

al. (1998). Application of protein-coated scanning force microscopy probes in measurements of surface affinity to protein adsorption. Applied Physics A:

Materials Science & Processing, 66, s631-s634. Cho, J.-M., & Sigmund, W. M. (2002). Direct surface force measurement in water

using a nanosize colloidal probe technique. Journal of Colloid and Interface

Science, 245(2), 405-407. Cleveland, J. P., Manne, S., Bocek, D., & Hansma, P. K. (1993). A nondestructive

method for determining the spring constant of cantilevers for scanning force microscopy. Review of Scientific Instruments, 64, 403-405.

Fitzpatrick, H., Luckham, P. F., Eriksen, S., & Hammond, K. (1992). Bovine serum albumin adsorption to mica surfaces. Colloids and Surfaces, 65, 43-49.

Duker, W. A., Senden, T. J., & Pashley, R. M. (1992). Measurement of forces in liquids using a force microscope. Langmuir, 8, 1831-1836.

Florin, E., Moy, V. T., & Gaub, H. E. (1994). Adhesion forces between individual ligand-receptor pairs. Science, 264, 415-417.

Frank, B. P., & Belfort, G. (2000). Atomic force microscopy for low-adhesion surfaces: thermodynamic criteria, critical surface tension, and intermolecular forces. Langmuir, 17, 1905-1912.

Fritz, M., Radmacher, M., Allersma, M. W., Cleveland, J. P., Stewart, R. J., Hansma, P. K., et al. (1995). Imaging microtubules in buffer solution using tapping mode atomic force microscopy. Proceedings of SPIE - The International Society for

Optical Engineering, 2384, 150-157. Gorbunov, A. A., Lukyanov, A. Y., Pasechnik, V. A., & Vakhrushev, A. V. (1986).

Computer simulation of protein adsorption and chromatography. Journal of

Chromatography A, 365, 205-212. Green, N. H., Allen, S., Davies, M. C., Roberts, C. J., Tendler, S. J. B., & Williams, P.

M. (2002). Force sensing and mapping by atomic force microscopy. TrAC

Trends in Analytical Chemistry, 21(1), 65-74. Hartmann, U. (1998). Atomic force microscopy. In J. A. Helsen & H. J. Breme (Eds.),

Metals as biomaterials (pp. 376). New York: Wiley.

62

Helm, C., Knoll, W., & Israelachvili, J. (1991). Measurement of ligand-receptor interactions. PNAS, 88(18), 8169-8173.

Heinz, W. F., Hoh, J. H. (1999). Spatially resolved force spectroscopy on

biological surfaces. Trends in Biotechnology, 4, 143-150. Hlady, V., & Buijs, J. (1996). Protein adsorption on solid surfaces. Current Opinion in

Biotechnology, 7(1), 72-77. Hutter, J. L., & Bechhoefer, J. (1993). Calibration of atomic-force microscope tips.

Review of Scientific Instruments, 64(7), 1868-1873. Israelachvili, J., & Weenerstrom, H. (1996). Role of hydration and water structure in

biological and colloidal interactions. Nature, 379, 219-225. Israelachvili, J. (1992). Intermolecular and surface forces (SECOND EDITION). San

Diego: Academic Press Itoh, H., Nagata, A., Toyomasu, T., Sakiyama, T., Nagai, T., Saeki, T., Nakanishi,

K. (1995). Adsorption of ß-lactoglobulin onto the surface of stainless steel particles. Journal, Bioscience, Biotechnology and Biochemistry, 59,

1648-1651. Jandt, K. D. (2001). Atomic force microscopy of biomaterials surfaces and interfaces.

Surface Science, 491(3), 303-332. Kauzmann, W. (1959). Some factors in the interpretation of protein denaturation.

Advances in protein chemistry, 14, 1-63. Kay C Dee, D. A. P., Rena Bizios. (2002). An Introduction to tissue-biomaterial

interactions. New Jersey: John Wiley and Sons, Inc. Lea, A. S., Pungor, A., Hlady, V., Andrade, J. J. D., Herron, J. N., & Voss, E. W.

(1992). Manipulation of proteins on mica by atomic force microscopy. Langmuir, 8, 68-73.

Leckband, D., & Sivasankar, S. (1999). Forces controlling protein interactions: theory and experiment. Colloids and Surfaces B: Biointerfaces, 14(1-4), 83-97.

Leckband, D. E., Schmitt, F. J., Israelachvili, J. N., & Knoll, W. (1994). Direct Force Measurements of specific and nonspecific protein interactions. Biochemistry,

33(15), 4611-4624. Leikin, S., Parsegian, V. A., Rau, D. C., & Rand, R. P. (1993). Hydration forces.

Annual Review of Physical Chemistry, 44(1), 369-395. Leonenko, Z. V., Carnini, A., Cramb D. T. (2000). Supported planar bilayer

formation by vesicle fusion: the interaction of phospholipid vesicles with surfaces and the effect of gramicidin on bilayer properties using atomic force microscopy. Biochimica et Biophysica Acta, 1509, 131-147

63

Lin, S., Chen, J. L., Huang, L. S., & Lin H. W. (2005). Measurements of the forces in protein interactions with atomic force microscopy. Current Proteomics,2, 55-81.

Marra, J., & Israelachvili, J. (1984). Direct measurements of forces between phosphatidylcholine and phosphatidylethanolamine bilayers in aqueous electrolyte solutions. Biochemistry, 24, 4608-4618.

Mori, O., & Imae, T. (1997). AFM investigation of the adsorption process of bovine serum albumin on mica. Colloids and Surfaces B: Biointerfaces, 9, 31-36

Morita, S., Wiesendanger, R., & Meyer, E. (2002). Noncontact atomic force

microscopy. New York: Springer-Verlag Berlin Heidelberg. Morris, V. J., Gunning, A. P., & Kirby, A. R. (1999). Atomic froce microscopy for

biologists. Lodon: Imperial College Press. Mueller, H., Butt, H. J., & Bamberg, E. (1999). Force measurements on myelin basic

protein adsorbed to mica and lipid bilayer surfaces done with the atomic force microscope. Biophysical Journal, 76, 1072-1079.

Nir, S. (1977). van der Waals interactions between surfaces of biological interest. Progress in Surface Science, 8, 1-58.

Norde, W. (1986). Adsorption of proteins from solution at the solid-liquid interfaces, Advances in Colloid and Interface Science, 25, 267-340.

Norde, W. (2003). Driving forces for protein adsorption at solid surfaces. In M. Malmsten (Ed.), Biopolymers at interfaces (2nd ed., pp. 21-43). New York: Marcel Dekker.

Ortega-Vinuesa, J. L., Tengvall, P., & Lundstrom, I. (1998). Aggregation of HSA, IgG, and fibrinogen on methylated silicon surfaces. Journal of Colloid and

Interface Science, 207(2), 228-239. Patil, S., Matei, G., Dong, H., Hoffmann, P. M., Karakse, M., & Oral, A. (2005). A

highly sensitive atomic force microscope for linear measurements of molecular forces in liquids. Review of Scientific Instruments, 76, 103705(1)-103705(7).

Putman, C. A. J., Van der Werf, K. O., De Grooth, B. G., Van Hulst, N. F., & Greve, J. (1994). Tapping mode atomic force microscopy in liquid. Applied Physics

Letters, 64(18), 2454-2456. Quist, A. P., Bjorck, L. P., Reimann, C. T., Oscarsson, S. O., & Sundqvist, B. U. R.

(1995). A scanning force microscopy study of human serum albumin and porcine pancreas trypsin adsorption on mica surfaces. Surface Science,

325(1-2), L406-L412. Regnier, F. E. (1987). The role of protein structure in chromatographic behavior.

Science, 238, 319-323.

64

Rezwam, K., Meier, L. P., & Gauckler, L. J. (2005). Lysozyme and bovine serum albumin adsorption on uncoated silica and AlOOH-coated silica particles: the influence of positively and negatively charged oxide surface coatings. Biomaterials, 26, 4351-4357.

Roth, C. M., & Lenhoff, A. M. (1993). Electrostatic and van der Waals contributions to protein adsorption: computation of equilibrium constants. Langmuir, 9, 962-972.

Roth, C. M., & Lenhoff, A. M. (1995). Electrostatic and van der Waals contributions to protein adsorption: comparison of theory and experiment. Langmuir, 11, 3500-3509.

Rowe, A. J. (2001). Probing hydration and the stability of protein solutions -- a colloid science approach. Biophysical Chemistry, 93(2-3), 93-101.

Ruger, D., & Hansma, P. K. (1990). Atomic force microscopy. Physics today, 43(10), 23-30.

Sader, J. E., Chon, J. W. M. Mulvaney, P. (1999). Calibration of rectangular

atomic force microscope cantilevers. Review of Scientific Instruments, 70, 3967-3969.

Sader, J. E., Larson, I., Mulvaney, P., & Lee R, W. (1995). Method for the calibration of atomic force microscopy cantilevers. Review of Scientific Instruments,

66(7), 3789-3798. Sagvolden, G., Giaever, I., & Feder, J. (1998). Characteristic protein adhesion forces

on glass and polystyrene substrates by atomic force microscopy. Langmuir, 14, 5984-5987.

Sethuraman, A., Han, M., Kane, R. S., & Belfort, G. (2004). Effect of surface wettability on the adhesion of proteins. Langmuir, 20, 7779-7788.

Scheuring, S., Fotiadis, D., Möller, C., Müller, S. A., Engel, A., Müller, D. J. (2001). Single proteins observed by Atomic Force Microscopy. Single Molecules, 2(2), 59-67.

Ta, T. C., Sykes, M. T., & McDermott, M. T. (1998). Real-time observation of plasma protein film formation on well-defined surfaces with scanning force microscopy. Langmuir, 14(9), 2435-2443.

Taylor, M. E. (1993). Dynamics of piezoelectric tube scanners for scanning probe microscopy. Review of Scientific Instruments, 64, 154-158.

Tilton, R. D., Robertson, C. R., & Gast, A. P., (1991). Manipulation of hydrophobic interactions in protein adsorption. Langmuir, 7, 2710-2718.

65

Valle-Delgado, J. J., Molina-Bolívarb, J. A., Galisteo-Gonzáleza, F., Gálvez-Ruiza, M. J., Feilerc, A., & Rutlandc, M. (2004). Interactions between bovine serum albumin layers adsorbed on different substrates measured with an atomic force microscope. Physical Chemistry Chemical Physics, 7, 1482-1486.

Vengasandra, S., Sethumadhavan, G., Yan, F., & Wang, R. (2003). Studies on the protein-receptor interaction by atomic force microscopy. Langmuir, 19, 10940-10946.

Wang, S. M., Palmer, L. B., Schwartz, J. D., & Razatos, A. (2004). Evaluating protein attraction and adhesion to biomaterials with atomic force microscope. Langmuir, 20, 7753-7759.

Wiesendanger, R. (1995). Scanning probe microscopy and spectroscopy: methods and

applications. Cambridge: Cambridge University Press. Weisenhorn, A. L., Maivald, P., Butt, H. J., & Hansma, P. K. (1992). Measuring

adhesion, attractive, and repulsion between surfaces in liquids with an atomic-force microscope. Physical Review B, 45, 11226-11232.

Woolley, A. T., Cheung, C. L., Hafner, J. H., & Lieber, C. M. (2000). Structural biology with carbon nanotube AFM probes. Chemistry & Biology, 7(11), R193-R204.

Ying, P., Yu, Y., Jin, G., & Tao, Z. (2003). Competitive protein adsorption studied with atomic force microscopy. Colloids and Surface B: Biointerfaces, 32, 1-10.

Yoon, B. J., & Lenhoff, A. M. (1992). Computation of the electrostatic interaction energy between a protein a charged surface. The Journal of Physical

Chemistry, 96, 3130-3134.

Zhang, H., Bremmell, K., Kumar, S., & Smart, R. S. C. (2004). Vitronectin adsorption

on surfaces visualized by tapping mode atomic force microscopy. Journal of

Biomedical Materials Research Part A, 68A(3), 479-488.

66

Appendix A – The AFM force curves of BSA-mica in PBS

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A1: The adhesion force curves of BSA and mica

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A2: The adhesion force curves of BSA and mica

67

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A3: The adhesion force curves of BSA and mica

-0.1

-0.05

0

0.05

0.1

0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A4: The adhesion force curves of BSA and mica

68

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A5: The adhesion force curves of BSA and mica

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A6: The adhesion force curves of BSA and mica

69

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A7: The adhesion force curves of BSA and mica

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A8: The adhesion force curves of BSA and mica

70

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure A9: The adhesion force curves of BSA and mica

71

Appendix B– The AFM force curves of BSA-polystyrene in PBS

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B1: The adhesion force curves of BSA and polystyrene

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B2: The adhesion force curves of BSA and polystyrene

72

-0.1

-0.05

0

0.05

0.1

0 40 80 120 160

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B3: The adhesion force curves of BSA and polystyrene

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B4: The adhesion force curves of BSA and polystyrene

73

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B5: The adhesion force curves of BSA and polystyrene

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B6: The adhesion force curves of BSA and polystyrene

74

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100 120 140

Tip approach (nN)Tip retract (nN)

Displacement (nm)

Figure B7: The adhesion force curves of BSA and polystyrene

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100 120

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B8: The adhesion force curves of BSA and polystyrene

75

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80 100 120 140

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B9: The adhesion force curves of BSA and polystyrene

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B10: The adhesion force curves of BSA and polystyrene

76

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B11: The adhesion force curves of BSA and polystyrene

-0.1

-0.05

0

0.05

0.1

-10 0 10 20 30 40 50 60

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B12: The adhesion force curves of BSA and polystyrene

77

-0.1

-0.05

0

0.05

0.1

-20 -10 0 10 20 30 40 50 60

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure B13: The adhesion force curves of BSA and polystyrene

78

Appendix C – The bare mica force curves in PBS and the bare polystyrene force curves in PBS

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure C1: no interactions were observed from the bare mica surface.

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure C2: no interactions were observed from the bare mica surface.

79

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

TIp approach (nN)Tip retract (nN)

Displacement (nm) Figure C3: no interactions were observed from the bare PS surface.

-0.1

-0.05

0

0.05

0.1

-20 0 20 40 60 80

Tip approach (nN)Tip retract (nN)

Displacement (nm) Figure C4: no interactions were observed from the bare PS surface.

80

Appendix D – Data, SEM images and details of the calibration procedure for the two cantilevers.

Cantilever A

10 resonance curves of cantilever A were obtained and showed in Appendix A. Figure

D1 is one of the resonance curves of the cantilever A and has been fitted by the

Gaussian equation. After fitting the Gaussian equation to those 10 resonance

frequency curves, the average value of the resonance frequency (fc) is 13.246 which

obtained from the 10 values of the resonance frequency, and the error (standard

deviation) can be also calculated (Table D1 shows those values).

Figure D1: One of resonance curves of cantilever A is fitted by Gaussian equation

0

0.5

1

1.5

2

10 15 20 Frequency (kHz)

y = A0+A1*exp(-(x-f0)2/σσσσ2

Error Value 0.0079 0.3281 A0 0.036 0.842 A1 0.025 13.261 f0 0.037 0.723 σσσσ

3.342 Chisq 0.859 R

• Amplitude

−−−− Gaussian fitted

13.261

81

The table D1 shows the values of the resonance frequency and the Q factor from the

10 resonance frequency curves of cantilever A. The Q factor values can be calculated

by the equation 6 from those 10 resonance frequency curves. The average value of Q

factor (Qf) is 12.048 and the error (standard deviation) is 2.531.

Resonance Frequency

(kHz) Q Factor

1 13.259 9.545

2 13.234 15.868

3 13.215 15.864

4 13.257 9.945

5 13.246 14.017

6 13.235 10.639

7 13.307 10.324

8 13.261 10.660

9 13.209 9.289

10 13.240 11.917

Average 13.246±0.027 12.048±2.531

Table D1: The resonance frequency values and the Q factor values were obtained and calculated from the 10 resonance frequency curves of cantilever A.

In order to measure the length and width of the cantilever A the SEM image of

cantilever A was taken (Fig. D2). The measured length (L) and width (b) of the

cantilever A are 342.8µm and 40.0µm, respectively. The errors (∆∆∆∆L and ∆∆∆∆b) can be

calculated by equation 7a and equation 7b.

82

Figure D2: The SEM image of cantilever A.

Since

b = 40.0µm

L = 342.8µm

Qf = 12.048

fc = 13.246 kHz

We have applied those four values to Sader’s method (Equation 5) and obtained the

spring constant of the cantilever A (kz = 0.0257 N/m). The error of the spring constant

of the cantilever A (∆∆∆∆kz) is 0.0061 can be calculated by the equation 8. Where

∆∆∆∆b = 0.3µm

∆∆∆∆L = 2.8µm

∆∆∆∆Qf = 2.531

∆∆∆∆fc = 0.027 kHz

AFM

cantilever A

83

Cantilever B

In order to measure the length and width of the cantilever B the SEM image of

cantilever A was taken (Fig. D3). The measured length (L) and width (b) of the

cantilever A are 347.1µm and 35.3µm, respectively. The errors (∆∆∆∆L and ∆∆∆∆b) can be

calculated by equation 7a and equation 7b.

Figure D3: The SEM image of cantilever B.

10 resonance curves of cantilever B were also obtained and showed in Appendix A.

Figure D4 is one of the resonance curves of the cantilever B and has been fitted by the

Gaussian equation. After fitting the Gaussian equation to those 10 resonance

frequency curves, the average value of the resonance frequency (fc) is 12.896 which

obtained from the 10 values of the resonance frequency, and the error (standard

deviation) can be also calculated (Table D2 shows those values).

AFM

cantilever B

84

Figure D4: One of resonance curves of cantilever A is fitted by Gaussian equation

The table D2 shows the values of the resonance frequency and the Q factor from the

10 resonance frequency curves of cantilever A. The Q factor values can be calculated

by the equation 6 from those 10 resonance frequency curves. The average value of Q

factor (Qf) is 13.22 and the error (standard deviation) is 2.873.

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

6 8 10 12 14 16 18 20

Frequency (kHz)

y = A0+A1*exp(-(x-f0)2/σσσσ2222

Error Value 0.0077 0.3879 A0 0.039 0.694 A1 0.028 12.878 f0 0.041 0.619 σσσσ

3.294 Chisq 0.796 R

• Amplitude

−−−− Gaussian fitted

12.878

85

Resonance Frequency

(kHz) Q Factor

1 12.885 9.739

2 13.046 14.227

3 12.841 13.912

4 12.757 12.757

5 12.892 10.916

6 12.878 12.371

7 12.901 14.287

8 12.929 8.251

9 12.853 18.520

10 12.971 14.380

Average 12.896±0.077 13.22±2.873

Table D2: The resonance frequency values and the Q factor values were obtained and calculated from the 10 resonance frequency curves of cantilever B.

Since

b = 35.3µm

L = 347.1µm

Qf = 13.22

fc = 12.896 kHz

We have applied those four values to Sader’s method (Equation 5) and obtained the

spring constant of the cantilever A (kz = 0.0257 N/m).

86

The error of the spring constant of the cantilever A (∆∆∆∆kz) is 0.0062 can be calculated

by the equation 8. Where

∆∆∆∆b = 0.2µm

∆∆∆∆L = 2.7µm

∆∆∆∆Qf = 2.873

∆∆∆∆fc = 0.077 kHz