Effect of temperature and trace water on surface forces studied by liquid AFM

EFFECT OF TEMPERATURE AND TRACE WATER ON SURFACE FORCES STUDIED BY LIQUID AFM LEONARD LIM TZE WEI (B.Sc (Hons.), National University of Singapore) A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF PHYSICS NATIONAL UNIVERSITY OF SINGAPORE 2008 Acknowledgements To those who have worked in the Biosensor Lab #05-02 and Interfaces Lab #02-02 (IMRE) Thank you for your friendship and help throughout my years in IMRE, especially Dr Cedric Troadec, Dr Aaron Lau, and Mr Tan Yee Yuan for their countless hours of discussions and Dr Roderick Lim for his invaluable lessons on the fine art of AFMing To the Cashewians, thank you for your friendship and advice, especially Boss, Kathy, Kenny and Setha I am eternally grateful to my parents and mother-in-law for their support and patience throughout the course of my studentship To Siddy, thank you for helping me out in the sticky situations all these years To Jessica Koh, thank you for all the support and help in getting me to keep my focus To the 1SAPians, thank you for your understanding and support, especially Glenn, CP, Michiel, Victor and Karen I would like to express my deepest gratitude to Professor Andrew Wee for his great support and advice throughout my candidature both experimentally and administratively Last but not least, to the “old man”, Dr Sean O’Shea for his stewardship, patience and generosity The experiences and lessons learnt have been invaluable The discussions about science and life in general have taught me much about what this world is all about i Table of Contents Acknowledgements i Table of Contents ii Summary v List of Figures vi List of Tables xii List of Publications xiii Chapter One: Introduction 1.1 Motivation 1.2 Thesis Outline Chapter Two: Literature Survey 2.1 Solvation Forces 2.2 Surface Force Apparatus Measurements 2.3 Atomic Force Microscopy Measurements 17 2.4 Computer Simulations 24 2.5 Concluding Remarks 26 Chapter Three: Experimental Methods 28 3.1 AFM Techniques 28 3.1.1 Cantilever Characterization 31 3.1.2 Tip Modification 36 3.1.3 Tip Characterization 40 3.1.4 Piezo Calibration 41 3.1.5 Force Measurements 42 ii 3.1.6 Conduction AFM 46 3.1.7 Temperature Control 47 3.2 STM Imaging 49 3.2.1 STM Tip Preparation 49 3.3 Materials 50 3.3.1 Liquids 50 3.3.2 Substrate 52 3.4 Miscellaneous Techniques 53 3.4.1 Trace Water Measurement 54 Chapter Four: Effects of Temperature and Tip Radius 56 4.1 Introduction 56 4.2 Experimental 58 4.2.1 General Force Curves 59 4.2.2 Important caveats regarding the force curve data 65 4.2.2.1 Inherent variability of data 66 4.2.2.2 Influence of Tip Size 69 4.2.2.3 The relevant forces to consider 76 4.2.2.4 General conclusions regarding the quality of data 81 4.3 Experimental Force Curves at Different Temperatures 82 4.4 Discussion: Physical Basis of Temperature Dependent Observations 97 4.4.1 Large Temperature Change is not due to Density 98 4.4.2 Monolayer melting 99 4.4.3 Squeeze Out of the Confined Liquid : The Preferred Mechanism 105 4.5 Analytical Model of Temperature Dependent Squeeze Out 111 4.5.1 Analytical Model 113 4.5.2 Model results 120 iii 4.6 Conclusion 126 Chapter Five: Effects of Water 130 5.1 Preparation of Liquids 130 5.2 Solvation Forces Measured in Liquids with Water 133 5.2.1 OMCTS with Water 133 5.2.2 Hexadecane and Hexadecane with Water 135 5.2.3 Dodecanol with Water 136 5.3 STM Imaging & NMR 138 5.4 Discussion I 141 5.5 Adhesion Force in Other Alcohols 145 5.6 True Tip-HOPG Contact 147 5.7 Variation of Adhesion with Tip Material 149 5.7.1 Si Cantilevers 149 5.7.2 Au Coated Cantilevers 151 5.7.3 Al Coated Cantilevers 152 5.8 Drying Experiments 154 5.8.1 Boiling 154 5.8.2 Molecular Sieve 156 5.8.3 Freeze-thaw Method 157 5.8.4 Chemical Method – Sodium Sulphate 158 5.9 Discussion of Repulsive Adhesion Force 158 5.10 Conclusion 166 Chapter Six: Conclusions and Outlook 169 Bibliography 172 Appendix A: Tables of temperature dependent data A-1 iv Summary Solvation forces in confined liquids have been studied using the Atomic Force Microscope (AFM), principally the effect of temperature, tip shape and trace amounts of water in the liquid The effect of temperature on solvation forces have been studied in the liquids OMCTS, nhexadecane, and n-dodecanol Discrete solvation layers can be observed for all three liquids at all the temperatures measured (298K to 348K) However, with increasing temperature there is a significant decrease in the magnitude of the measured solvation forces and a reduction in the number of solvation oscillations which can be observed The normalized solvation force data, F/Rtip, has also been found to differ between AFM tips of different radius of curvature (Rtip = 15nm to 100nm) with a clear trend of decreasing F/Rtip with increasing Rtip The effect of trace water, with the exception of the OMCTS-HOPG system where the data is inconclusive and no comment can be made, has been found to cause a decrease in the magnitude of the maximum force (Fmax) for each layer, a decrease in the number of observable jumps, and a decrease in the exponential decay length in the liquids nhexadecane and n-dodecanol v List of Figures Figure 1.1 Schematic of the structure of a simple liquid confined between two parallel walls Taken from Ref [1] Figure 2.1 Typical force interaction curves of DLVO theory Taken from Ref [1] Figure 2.2 Schematic of a SFA for use in liquids Taken from Ref [1] 10 Figure 2.3 Experimental results of force F as a function of separation D between two curved mica surfaces of radius R = cm separated by OMCTS at 22°C Taken from Ref [2] 12 Figure 2.4 Oscillatory solvation force superimposed on a monotonic force Taken from Ref [1] 15 Figure 2.5 Principle of STM Taken from Ref [3] 18 Figure 2.6 Principle of a basic AFM Taken from Ref [3] 19 Fig 3.1 Simple schematic of AFM operation 28 Figure 3.2 (a) Homebuilt vacuum chamber for measurement of the fundamental resonance frequency of a cantilever (b) Typical plot of the measurement taken, with the resonance frequency at 15.41 kHz 34 Figure 3.3 Experimental setup for attaching beads 37 Figure 3.4 Graphic description of the bead attachment process (a) A small quantity of adhesive is applied onto a silicon substrate and the tip brought up close to it This picture is taken at x100 magnification (b) Magnification is changed to x500 and the tip is brought closer to the adhesive (c) x500 The tip is partially submerged into the adhesive (d) The substrate with the adhesive is removed and replaced by one with silica spheres dispersed on them A single sphere is chosen visually and the tip is brought up close to the sphere (e) x500 The tip is brought in contact with the sphere (f) x500 The tip is then withdrawn from the Si substrate and the bead is now mounted on the tip (g) SEM image of the tip apex mounted with the bead Scale bar = 1µm 39 Figure 3.5 SEM image of an AFM tip (a) Overall image of tip and cantilever (b) Zoomed in image of the tip 40 vi Figure 3.6 Schematic of how IPSD vs Z (position sensitive detector current signal vs piezo position) are converted to force vs distance curves for (a) a simple “ideal” force curve (infinitely hard tip and sample without surface forces) (b) Infinitely hard materials with long-range repulsion (c) Deformable materials without surface forces (d) deformable materials with attraction and adhesion force Taken from Ref [4] 42 Figure 3.7 (a) Typical raw data curve and (b) its corresponding converted force-tip sample distance curve 44 Figure 3.8 Conduction AFM setup 47 Figure 3.9 Graphical plot of the setpoint temperature (Red ▲), substrate temperature (Purple *) and liquid temperature (Hexadecane – Blue ■, OMCTS – Yellow ♦, Dodecanol – Green ●) 48 Figure 3.10 Schematic of the molecular structure of the various liquids used 51 Figure 3.11 x nm STM image of HOPG 53 Figure 4.1 Typical force curves near the HOPG surface for (a) OMCTS with Rtip = 40nm, (b) n-dodecanol with Rtip = 25nm, and (c) n-dodecanol with Rtip = 100 nm The black curve ( ) is for tip approach and the red circles (○) for tip retract The tip-sample distance D=0 corresponds to the tip in mechanical contact with the HOPG The label n=1 shows the first solvation layer, n=2 the second layer, etc The force F1 is the peak-peak force (maximum force minus minimum force) in the n=1 layer 61 Figure 4.2 Two examples of SEM images of a Si3N4 tip after use 63 Figure 4.3 Comparison of AFM solvation data for different tip radii in hexadecane The graphs show the same data as in the associated Table The “ratio” in the Table shows Fn/Rtip for n=1 divided by Fn/Rtip for n=2 72 Figure 4.4 Comparison of AFM solvation data for different tip radii in Dodecanol The graphs show the same data as in the associated Table The “ratio” in the Table shows Fn/Rtip for n=1 divided by Fn/Rtip for n=2 73 Figure 4.5 Generalized schematic of a spherical tip in contact with a flat substrate (heavy solid lines) The pressure distribution comprises a repulsive, Hertzian contribution (P1) and an adhesive part (P2) The net pressure (dashed line) has tension components around the contact periphery and maximum at the tip apex 77 vii Figure 4.6 Contact radius as a function of force for the Hertz, DMT and JKR models of an elastic sphere-flat contact Fads is fixed at -1.5 nN To create this plot the value of (3Rtip/4K) is set to 10-18 m3N-1, which is the typical order for AFM experiments 78 Figure 4.7 Force curves (approach is black, retract is red) near a HOPG surface immersed in hexadecane taken at a) 25ºC, b) 45ºC and c) 75ºC with Rtip = 45 nm The tip-sample distance D=0Å corresponds to the tip in contact with the HOPG 82 Figure 4.8 Force curves (approach is black, retract is red) near a HOPG surface immersed in dodecanol taken at a) 25ºC, b) 45ºC and c) 75ºC with Rtip = 29 nm The tip-sample distance D=0Å corresponds to the tip in contact with the HOPG 84 Figure 4.9 Force curves (approach is black, retract is red) near a HOPG surface immersed in OMCTS taken at a) 25ºC, b) 45ºC and c) 55ºC with Rtip = 40 nm The tip-sample distance D=0Å corresponds to the tip in contact with the HOPG 85 Figure 4.10 Temperature dependence of the normalized peak-peak solvation force for OMCTS near a HOPG surface for the a) n=1 and b) n=2 layers Data is only presented for the smallest radius tips used (R = 20nm, R = 40nm) There is a small but clear decrease in Fn/R with increasing temperature The errors are large in a relative sense because of the difficulty in measuring the very small forces involved with OMCTS layers 87 Figure 4.11 Temperature dependence of the normalized maximum solvation force for hexadecane near a HOPG surface for the a) n=1 and b) n=2 layers There is a clear decrease in Fmax/R with increasing temperature Data is only presented for the smallest radius tips used (R = 15nm, R = 30nm) Data was also taken for the n=3 layer but the forces are very small and only a qualitative statement can be made that Fmax decreases with increasing temperature 88 Figure 4.12 Temperature dependence of the maximum solvation force Fmax in hexadecane near HOPG The Fmax/R data for different tip radius (Rtip = 15, 30, 36, 45, 50, 52, 55, 80 nm) is normalized to the corresponding value of Fmax/R at 25oC A clear decrease in Fmax/R with increasing temperature is observed for all tips The n=2 data falls slightly more rapidly than the n=1 data 89 Figure 4.13 a) Temperature dependence of the peak-peak solvation force for dodecanol near a HOPG surface for the n=1 layer Data is only presented for the smallest radius tips used (R = 20nm to 29nm) b) Two data sets from (a) replotted to give an indication of the errors involved The R=23nm data is offset by 1oC to for clarity These large errors are typical 91 Figure 4.14 Temperature dependence of the maximum solvation force and corresponding adhesion force for dodecanol near a HOPG surface for the n=1 layer The dashed line separates the two data sets This data is used to find the peak-peak values of Figure 4.13 Some of the data sets were offset by 1oC to make the data easier to visualize 92 viii Figure 4.15 a) Temperature dependence of the peak-peak solvation force for dodecanol near a HOPG surface for the n=2 layer Data is only presented for the smallest radius tips used (R = 20nm to 29nm) Some errors are shown to give an idea of the magnitudes involved b) Temperature dependence of the maximum solvation force and corresponding adhesion force The dashed line separates the two data sets This data is used to find the peak-peak values of (a) Some of the data sets were offset by 1oC to make the data easier to visualize 93 Figure 4.16 a) Temperature dependence of the peak-peak solvation force for dodecanol near a HOPG surface for the n=3 layer Data is only presented for the smallest radius tips used (R = 20nm to 29nm) Some errors are shown to give an idea of the magnitudes involved b) Temperature dependence of the maximum solvation force and corresponding adhesion force The dashed line separates the two data sets This data is used to find the peak-peak values of (a) Some of the data sets were offset by 1oC to make the data easier to visualize 94 Figure 4.17 a) Temperature dependence of the peak-peak solvation force for dodecanol near a HOPG surface for the n=1 layer Data taken in 2005 using larger radius tips (R = 40nm) b) Temperature dependence of the maximum solvation force and corresponding adhesion force The dashed line separates the two data sets This data is used to find the peak-peak values of (a) The R=42.5nm data set is offset by 1oC for clarity 96 Figure 4.18 a) STM image of the dodecanol monolayer (n=1) on HOPG Image size 15nm x 15nm STM image conditions 20pA, 0.5V b) AFM image the hexadecane monolayer on HOPG Image size 15nm x 15nm.[5] 100 Figure 4.19 Temperature dependence of the peak-peak solvation force Fn in a) OMCTS (n=1,2), and b) dodecanol (n=1,2,3) The Fn/R data for different tip radius is normalized to the corresponding value of Fn/R at 25oC (Rtip = 20 and 40nm for OMCTS; Rtip = 20, 20, 23, 25, 25, 29, 40, 42.5nm for dodecanol) The “n=1 yr 2005” is a data set taken with dodecanol from a different bottle The circled data is from single tip (Rtip=20nm) which shows atypical behavior 102 Figure 4.20 Simulation results at two temperatures of the average pressure between two blocks with Xenon confined in between[6] The two blocks move towards each other starting at distance d=0 where ~4 Xe monolayers are confined The discontinuities show sudden removal of a lubrication layer In this simulation the lubricant layers are not strongly pinned to the substrate, a situation similar to our experiments using a HOPG surface 107 Figure 4.21 Simulated results of the pressure required to squeeze out the last layer of confined butane and iso-butane as a function of temperature [207] The results at low temperature (