Effect of the Chain Length and Temperature on the Adhesive

Oct 1, 2017 - Materials Science and Technology of Polymers, MESA+ Institute for Nanotechnology, University of Twente, NL-7500 AE Enschede, The Netherl...
1 downloads 15 Views 1MB Size
Subscriber access provided by UNIV OF ESSEX

Article

Effect of chain length and temperature on the adhesive properties of alkanethiol self-assembled monolayers Hubert Gojzewski, Michael Kappl, and Arkadiusz Ptak Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b01921 • Publication Date (Web): 01 Oct 2017 Downloaded from http://pubs.acs.org on October 2, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 1. The interaction potential (energy profile) of an adhesion bond loaded by a harmonic potential (solid curve) and the load-free potential (dashed line) 127x132mm (150 x 150 DPI)

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Adhesion force versus loading rate for nanoadhesion between a silicon nitride tip and a SAM of: a) 1-tetradecanethiol, and b) 1-hexadecanethiol, for different temperatures. 1 – region of weak dependence, 2 – region of strong dependence. The error bars were calculated as a standard deviation. 82x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 2. Adhesion force versus loading rate for nanoadhesion between a silicon nitride tip and a SAM of: a) 1-tetradecanethiol, and b) 1-hexadecanethiol, for different temperatures. 1 – region of weak dependence, 2 – region of strong dependence. The error bars were calculated as a standard deviation. 82x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Adhesion force versus loading rate for nanoadhesion between a silicon nitride tip and a 1tetradecanethiol SAM on Au(111) at T = 35°C. The error bars were calculated as a standard deviation. The red solid line is a result of the fitting with the DHS model. 82x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 4 of 30

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 4. Adhesion force per single molecule versus loading rate for the nanoadhesion between a silicon nitride tip and a 1-tetradecanethiol SAM on Au(111) at T = 35°C. The red solid line is a result of the fitting with the DHS model. 82x57mm (300 x 300 DPI)

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. The free-energy of activation (∆Gβ*) versus temperature for the adhesion between a silicon nitride tip and a single molecule of: 1-decanethiol (black ■),8 1-tetradecanethiol (red ●), and 1-hexadecanethiol (green▲). The error bars were calculated as a standard error within the DHS model’s fitting (Figure 4). 82x59mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 6 of 30

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 6. The distance between the bound state and an activation barrier on the free-energy interaction potential in the absence of external forces (xβ*) as a function of temperature for the adhesion between a silicon nitride tip and a single molecule of: 1-decanethiol (black ■),8 1-tetradecanethiol (red ●), and 1hexadecanethiol (green▲). The error bars were calculated as a standard error within the DHS model’s fitting (Figure 4). 82x59mm (300 x 300 DPI)

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Effect of chain length and temperature on the adhesive properties of alkanethiol self-assembled monolayers Hubert Gojzewski†‡, Michael Kappl§, and Arkadiusz Ptak†* †

Institute of Physics, Poznan University of Technology, Piotrowo 3, PL-60965 Poznan,

Poland ‡

Materials Science and Technology of Polymers, MESA+ Institute for Nanotechnology,

University of Twente, NL-7500 AE Enschede, the Netherlands §

Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, Germany

*

Corresponding author. E-mail: [email protected]

Abstract Stable and hydrophobic self-assembled monolayers of alkanethiols are promising materials for use as lubricants in micro- and nanodevices. We applied high-rate dynamic force spectroscopy measurements to study in detail the influence of chain length and temperature on the adhesion between methyl-terminated thiol monolayers and a silicon nitride tip. We used the Johnson-Kendall-Roberts model to calculate the number of molecules in adhesive contact and then the Dudko-Hummer-Szabo model to extract the information about the position and the height of the activation barrier per single molecule. Both parameters were determined and analyzed in the temperature range from 25°C to 65°C for three thiols: 1-decanethiol (measured previously), 1-tetradecanethiol and 1-hexadecanethiol. The increase of the activation barrier parameters versus chain length we associate with lower stiffness of longer molecules and higher effectiveness of adhesive bonds formation. The thermal changes of the parameters, however, we relate rather to rearrangements of molecules than direct influence of temperature on the adhesive bonds.

1 ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Introduction Alkanethiols are known to form self-assembled monolayers (SAMs) on gold (111) surfaces. If they are sufficiently hydrophobic and stable, they can serve as lubricants in microdevices and nanodevices, since the adhesion between a hydrophobic SAM and a nanoobject remains low as compared to liquid lubricants.1-2 Termination of alkanethiols with a methyl group guarantees hydrophobicity of the layer. The research for an optimal chain length of alkanethiols, however, requires taking into account not only structural stability of SAMs but also the independence of their adhesive properties of temperature changes in a typical operating range. Temperature can influence the SAM’s nanoadhesion either directly affecting the adhesive interactions or indirectly, through reorganization of the SAM’s structure. In the paper we perform systematic experimental studies to characterize the adhesive interaction potential as a function of temperature and we discuss both of the kinds of influence. Since standard experimental studies of temperature-dependent adhesion or friction, which are limited to the measurement and analysis of force (normal or lateral) versus temperature,3-4 cannot give a deep insight into the kinetics and energetics of adhesive processes, we extended the studies by using the so-called dynamic force spectroscopy. It is a measurement of adhesion force as a function of separation rate performed usually with an optical tweezers or an atomic force microscope (AFM).5 In our experiments we have used a modified AFM to expand the accessible range of separation rates to over 106 nm s−1. Applying theoretical models of mechanical-adhesive contact (the Johnson-Kendall-Roberts (JKR) model6) and thermally activated unbinding (the Dudko-Hummer-Szabo (DHS) model7), we could determine parameters of the interaction potential – like the energy and the position of activation barrier – per single thiol molecule. Finally, we discuss the dependence of the parameters on temperature, suggesting possible reasons for their non-monotonic character, as

2 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

well as the length of the thiol molecules. To exclude the influence of the odd-even effect in our research, we have chosen alkanethiols with an even number of carbon atoms only, namely: 1-decanethiol (measured by us earlier8), 1-tetradecanethiol, and 1-hexadecanethiol. We have limited the temperature range from 25 to 65°C to avoid phase transitions and in particular the melting of the molecular layers.9-11 Details on the structure-property relationships of the alkanethiol SAMs on Au(111) at elevated temperatures can be found in the Supporting Information.

Theory There are several theoretical models interpreting dynamic force spectroscopy data. One of the most commonly used is the so-called Bell-Evans (BE) model, which predicts a logarithmic dependence of the mean adhesion force (Fad) on the loading rate (rF):12-13 Fad =

 k BT  xβ ln rF  . 0  k Tk  xβ  B off 

(1)

Here kB is the Boltzmann constant, T is the absolute temperature, xβ is the distance between the bound state and an activation barrier on the free-energy potential in the absence of external forces, and k0off is the kinetic off-rate constant. The BE model fitted to experimental data enables researchers to extract the values of xβ and k0off from the slope and intercept of an Fad-ln(rF) line. Therefore, it has been widely applied to study specific interactions between or within biomolecules.14-17 Since the model does not introduce any restrictions to narrow its applicability exclusively to specific interactions, it has been applied also to study nanoadhesion between an AFM tip and the SAMs of silanes18 and thiols.19-20 However, the applicability of the BE model is generally limited, since it lumps all information about the potential shape into one parameter – xβ. In addition, it does not take into account rebinding processes (Figure 1).

3 ACS Paragon Plus Environment

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Friddle et al.21 proposed a model which considers the contribution of rebinding and can be applied to a multi-bond. It is based on the BE model but an additional parameter is defined – the equilibrium unbinding force – which together with xβ and k0off describes an adhesive bond. In addition, Friddle et al. showed that there exists a critical loading rate rc, given by a simple relation: rc =

k BT ν0 , xβ

(2)

above which the behavior of real adhesive bonds deviates from the BE model predictions sufficiently to apply a more precise model. Here, ν0 is the attempt frequency, which is the reciprocal of the characteristic time, tD = (lclts γ)/kBT, where lc is the length representing the thermal spread in the bound state, lts is the length representing the thermal spread in the transition state, and γ is the molecular damping coefficient.13

Figure 1. The interaction potential (energy profile) of an adhesion bond loaded by a harmonic potential (solid curve) and the load-free potential (dashed line)

4 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

One of such more detailed models is the DHS model formulated by assuming a stochastic character of the escape process from a potential well22 and specifying the free-energy potential shape. The prediction of this model for the mean rupture force versus the loading rate is:

ν    ∆Gβ   0   k BT koff exp + γ    k T ∆Gβ   k BT B     F ≅ ln 1 −   , νxβ  xβ rF ∆Gβ          

(3)

where: ∆Gβ is the free energy of activation in the absence of external forces, γ = 0.577 is the Euler-Mascheroni constant, and the parameter ν corresponds with the shape of the free-energy potential. As suggested by the authors the universal value for ν is 2/3. Then the DHS model can precisely characterize unbinding processes in the case of real interatomic or intermolecular interactions (described by van der Waals or Morse potentials for instance). Additionally, the DHS model enables us to extract not only k0off and xβ – like the BE model – but also ∆Gβ. Eq. 3 reduces to Eq. 1 of the BE model for ν = 1 and ∆Gβ >> kBT, however, these conditions are normally not fulfilled in force spectroscopy experiments. There are also alternate theoretical approaches, for instance, by Li and Ji.23 They combined a model developed by them for ultralow loading rates, with the DHS model for the loading rates higher than a threshold value, which is below the lower limit of the experimental data qualified by us to quantitative analysis (Figure 1, region 2).

Experimental Section Preparation of self-assembled monolayers and their characterization Two types of SAMs samples were prepared by chemisorption of 1-tetradecanethiol (Sigma-Aldrich) and 1-hexadecanethiol (Alfa-Aesar) on Au(111) substrates. Au(111) 5 ACS Paragon Plus Environment

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

substrates were immersed in 1 mM thiol solution in ethanol. After 24 h self-assembly samples were rinsed with ethanol, dried under nitrogen, and immediately used. The quality of the Au(111) substrates was controlled with X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and scanning tunnelling microscopy (STM) measurements. The SAM samples were characterized at nanoscale (STM imaging) and macroscale (contact angle measurements, XPS). For details see the Supporting Information.

AFM modification, calibration and measurements A commercial AFM (Bruker, EnviroScope with the NanoScope IIIa controller), modified and thereafter calibrated to perform dynamic force spectroscopy measurements at high surface separation rate, and high temperature was used. Details on the setup used can be found elsewhere.8 Briefly, an additional low-mass and high-frequency piezo-actuator (PI Ceramic, model PL055 PICMA) was mounted on a heating plate using thermal glue. The investigated samples adhered to the piezo-actuator were located in a hermetic chamber with slow nitrogen flow. The piezo-actuator was driven by a function generator (Agilent, model 33220A) combined with a custom-made voltage amplifier. This additional piezo-actuator allowed to expand the range of fast loading rates by about 3 orders of magnitude. Force-distance curves were collected by a fast data acquisition system (Alazartech, model ATS9462).20, 24 A heating system – consisting of two precisely regulated thermocouples (Lakeshore, model 331) – was employed to monitor temperatures of SAM samples and AFM tips. Both temperatures were kept equal to avoid heat transfer between the contacting surfaces, and thus influencing the stability and organization of thiol molecules in SAM and the cantilever deflection. Low spring constant (~0.5 N·m-1) and high resonance frequency (~120 kHz) silicon nitride cantilevers (Bruker, model MLCT-F, metal-layer coated) were used. The cantilevers were cleaned by Ar plasma cleaner (Harrick Scientific Corporation, model PDC-001) for 20 s before use. The cantilever spring constant was measured 10 times at 25°C using the thermal 6 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

noise method25 and the average value was taken for further calculations (minor influence of temperature on the spring constant was found in the range of 25-65°C). The radii of AFM tips were determined by using a low-voltage scanning electron microscopy (SEM) observation (Carl Zeiss, model LEO 1530 Gemini) and fitting a circle to the shape of the tip’s apex (ImageJ 1.45s). The data presented and analysed in this paper were measured according to the procedure “one tip for one sample” in the temperature range of 25-65ºC. 10-250 individual force curves were recorded for each loading rate (10 for the lowest and 250 for the highest loading rate). The loading rate was varied in a random order to avoid systematic errors related to the tip wear and SAM disorder under load.

Results and Discussion The adhesion force between a silicon nitride tip and the layer of methyl-terminated thiols increases monotonically with an increase in loading rate for both SAMs: 1-tetradecanethiol and 1-hexadecanethiol, and for different temperatures from 25°C to 65°C (Figure 2). For the 1-hexadecanethiol SAM the highest temperature was limited to 55°C; above this temperature changes occur in the ordering of the molecular layer, which affects the results leading to high measurement uncertainty. We will discuss the changes in response to temperature later in this section. The determined relationship between the adhesion force and the loading rate resemble that measured for 1-decanethiol SAM.8 It shows two regions, of weak and strong dependence, with the threshold loading rate about 105 nN/s for the 1-tetradecanethiol SAM (Figure 2a), and 4·105 nN/s for the 1-hexadecanethiol SAM (Figure 2b). Since both the thiol SAMs are hydrophobic and the experiments were performed in N2 atmosphere at low relative humidity, the influence of capillary forces as well as fluid viscosity can be neglected.

7 ACS Paragon Plus Environment

Page 14 of 30

Page 15 of 30

50

a) Adhesion force [nN]

o

65 C o 55 C o 45 C o 35 C o 25 C

40

30

2 20

1 10 1 10

2

3

10

4

10

5

10

6

10

10

Loading rate [nN/s]

b) 30

Adhesion force [nN]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

20

o

55 C o 45 C o 35 C o 25 C

10

2 1

0 1 10

2

10

3

10

4

10

5

10

6

10

Loading rate [nN/s]

Figure 2. Adhesion force versus loading rate for nanoadhesion between a silicon nitride tip and a SAM of: a) 1-tetradecanethiol, and b) 1-hexadecanethiol, for different temperatures. 1 – region of weak dependence, 2 – region of strong dependence. The error bars were calculated as a standard deviation. In the standard interpretation for the occurrence of two distinct regions with different slopes in DFS data, the existence of two activation barriers within the interaction potential of nanoadhesion is needed: (1) the outer barrier, which determines the dependence at low

8 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

loading rates; and (2) the inner one, which determines the dependence at larger loading rates when the outer barrier is already strongly lowered by external force at the moment of forceinduced unbinding.13 This interpretation cannot be applied to the results of our experiments as the studied non-specific interactions between a silicon nitride tip and methyl-terminated thiols, which come mainly from the van der Waals forces, are characterized by a barrier-free, Lennard-Jones type of potential. Instead, we assume an effective rebinding of single adhesive bonds (van der Waals interactions per single molecules terminated with a methyl group) as a reason for the weak dependence at low loading rates.8, 21, 26 A rupture of a single adhesive bond does not lead to the rupture of the whole adhesion contact between the AFM tip and a SAM, because each broken bond can easily rebind as long as the adjacent bonds still exist. Only a rupture of a significant fracture of adhesive bonds can lead to full detachment. Therefore, the adhesion force at lower loading rates (region 1) will be strongly influenced by rebinding and does not weaken with decreasing loading rate as steeply as it does at higher loading rates (region 2), where rebinding becomes more and more unlikely. Since the models of thermally activated escape from potential well, which describe unbinding (e.g. the DHS model), do not take into account the processes of rebinding, only region 2 should be considered for further analysis. In region 2 single adhesive bonds behave independently during rupture and hence the force distributed per single bond can be given as:

Fs =

Fad k BT  xβ Fad + ln N xβ  k BT

 ,  

(4)

where Fad is the resultant adhesion force (measured with the AFM), N is the number of single independent bonds in parallel, kB is the Boltzmann constant, T is the absolute temperature, and xβ is the distance between the bound state and the activation barrier on the effective freeenergy potential.27-28 Since Fad is measured and T is controlled in our experiments, two

9 ACS Paragon Plus Environment

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

parameters: xβ and N, have to be determined to calculate a precise value of the unbinding force per molecule. xβ was obtained with the DHS model fitted to region 2. We decided to apply this model since we exceeded the threshold value of the loading rate (Eq. 2) given by Friddle et al.21 Substituting xβ from our results (Table S1 in SI) and ν0 = 109 s-1 provided by Friddle et al., a value of rc ≈ 106 nN/s was obtained. Although this is around 2 orders of magnitude beyond the present standard experimental range, we exceeded this threshold using our modified AFM system with an additional Z-piezo-actuator and performing the experiments in air instead of liquid. An example result of the fitting for the adhesion between a silicon nitride tip and a 1tetradecanethiol SAM on Au(111) measured at 35°C is shown in Figure 3 (red solid curve). Fittings for a narrower range of loading rates within regime 2, i.e. for 7, 6, or 5 highest loading rates, give similar results, which indicates independence from the rebinding within regime 2. The values of parameter xβ, which were used in further calculations, determined for all measured temperatures for the SAMs of 1-tetradecanethiol and 1-hexadecanethiol, are given in Table S1 (SI). Since they were obtained for the nanoadhesion treated as a single effective bond, we needed to recalculate them per individual adhesive bonds to discuss their physical meaning.

10 ACS Paragon Plus Environment

Langmuir

50

Adhesion force [nN]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

40

2

30

20

10 1 10

1

2

10

3

10

4

10

5

10

6

10

Loading rate [nN/s]

Figure 3. Adhesion force versus loading rate for nanoadhesion between a silicon nitride tip and a 1-tetradecanethiol SAM on Au(111) at T = 35°C. The error bars were calculated as a standard deviation. The red solid line is a result of the fitting with the DHS model. The number of bonds in the adhesion contact (N) was estimated using the JKR model.6 The JKR model considers the contribution of adhesion within a mechanical contact between a homogeneous, elastic planar surface and a sphere. The radius of the contact area between an AFM tip and an elastic monolayer at the moment of rupture can be given as:  RF ac =  ad  Es

1

 3  , 

(5)

where, R is the radius of the tip curvature, and Es is the elastic modulus of the monolayer, assuming that the elastic moduli of the tip material and the gold substrate are much higher. Although different values of Es can be found in the literature depending on measurement method and configuration of the sample,29-32 we decided to use the value calculated from AFM measurements by DelRio et al.33 for a thiol SAM alone (without influence of the substrate), i.e. (0.86 ± 0.14) GPa. Most literature values were obtained from experiments or simulations with compression of the monolayer under load and therefore the contributions of 11 ACS Paragon Plus Environment

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

the underlying substrate as well as the steric repulsion in squeezed molecules was significant. In our experiment, however, the rupture of adhesion contact occurs at a small negative (stretching) load, when the molecules straighten up to a vertical position34 and are slightly stretched rather than compressed. Therefore, the effects of substrate and steric repulsion do not occur. CH3-terminated alkanethiols are found to adopt √3√330° ordering on Au(111) substrate, which leads to an area per molecule of 0.216 nm2.8, 11 The radii of the AFM tips were determined from SEM images as (59 ± 6) nm for the experiments with the 1tetradecanethiol SAM which results in 1600–2150 contacting molecules (depending on the temperature), and (50 ± 5) nm for the experiments with the 1-hexadecanethiol SAM, which results in 940–1350 contacting molecules. Detailed values of the contact area at rupture and the number of contacting molecules are presented in Table S2 (SI). Inserting xβ and N into Eq. 4, the adhesion force per single molecule could be calculated. In this manner, the total adhesion force dependences (an example in Figure 3, regime 2) were recalculated into the adhesion force per single molecule versus loading rate (an example in Figure 4). After this, the DHS model was applied to regime 2 of these new plots to extract parameters describing the shape of the interaction potential between an individual methyl group and the silicon nitride tip, namely xβ* and ∆Gβ* (Table S3). In the case of van der Waals potential, which does not exhibit any activation barrier until it is subjected to an external force (Figure 4), xβ* can be defined as the distance between the bound state and an apparent activation barrier, and ∆Gβ* – as the height of the apparent activation barrier. These quantities are defined by the first and third derivative of the interaction potential, U0(x), at the inflection point, xinf, respectively:7 1

xβ0* =

2[U 0′ ( xinf )]2

(6)

1

[− U 0′′′( xinf )]2 , 12 ACS Paragon Plus Environment

Langmuir

3 2 [2U 0′ ( xinf )]2 ∆Gβ0* = 3 1 [− U 0′′′( xinf )]2 .

(7)

The parameters describe the shape of a load-free interaction potential. When the adhesive bond is subjected to an external force (for instance by AFM cantilever in our experiments) an energy barrier and a second minimum are created as a result of added harmonic potential (Figure 5).21, 26

40

Adhesion force [pN]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

35

30

25 5

6

10

10

Loading rate [nN/s]

Figure 4. Adhesion force per single molecule versus loading rate for the nanoadhesion between a silicon nitride tip and a 1-tetradecanethiol SAM on Au(111) at T = 35°C. The red solid line is a result of the fitting with the DHS model. The dependences of ∆Gβ* versus temperature, and xβ* versus temperature for three thiols: 1-decanethiol,8 1-tetradecanethiol, and 1-hexadecanethiol are plotted in Figure 5 and 6, respectively. The general observation is that longer molecules have higher values of ∆Gβ* and xβ*. This can be explained by lower stiffness of longer molecules and hence longer stretching distance before the unbinding occurs, resulting in lower probability for transition over the

13 ACS Paragon Plus Environment

Page 21 of 30

energy barrier (Figure 1). Another explanation is that longer molecules can create more effectively adhesive bonds with a rough surface of the AFM tip, which was not assumed in our methodology. Therefore, the average values of the interaction potential parameters per molecule, in particular ∆Gβ*, are overestimated for longer molecules in comparison to shorter ones. The value of ∆Gβ* appears stable and independent of temperature for 1-decanethiol and 1-tetradecanethiol SAMs. For 1-hexadecanethiol an increase is observed whilst the temperature rises from 25°C to 35°C (Figure 5). A similar increase between 25°C and 35°C occurs in the case of xβ* for all three thiols: 1-decanethiol, 1-tetradecanethiol, and 1hexadecanethiol (Figure 6). Above 35°C no abrupt changes are observed for both ∆Gβ* and xβ*.

60

40

*

∆Gβ [kBT]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

20

0 20

30

40

50

60

70

o

T [ C]

Figure 5. The free-energy of activation (∆Gβ*) versus temperature for the adhesion between a silicon nitride tip and a single molecule of: 1-decanethiol (black ■),8 1-tetradecanethiol (red

●), and 1-hexadecanethiol (green▲). The error bars were calculated as a standard error within the DHS model’s fitting (Figure 4).

14 ACS Paragon Plus Environment

Langmuir

6

4

*

xβ [nm]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

2

0 20

30

40

50

60

70

o

T [ C]

Figure 6. The distance between the bound state and an activation barrier on the free-energy interaction potential in the absence of external forces (xβ*) as a function of temperature for the adhesion between a silicon nitride tip and a single molecule of: 1-decanethiol (black ■),8 1tetradecanethiol (red ●), and 1-hexadecanethiol (green▲). The error bars were calculated as a standard error within the DHS model’s fitting (Figure 4). We assume that the relative changes in temperature are too low to directly influence the van der Waals forces which, in the case of our experiments, constitute the main contribution to adhesive interactions. However, temperature can influence the interaction potential indirectly, through processes not considered by the model of thermal activation, for example, some reorganization of the molecules in the SAM during the unbinding. The energy of the activation barrier can be separated into enthalpic and entropic contributions:35

∆Gβ* = ∆Hβ* – T∆S

(8)

The enthalpic component to the bond strength (∆Hβ*) is temperature independent but the entropic component (T∆S) depends on temperature and can be negative or positive. Its sign and value depends on changes in the arrangement of molecular groups in the adhesion contact 15 ACS Paragon Plus Environment

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

zone during the unbinding process. Therefore, some changes of ∆Gβ* can occur even in a narrow range of temperature change. Another possible mechanism affecting the parameters, particularly xβ*, can be related to the thermal change of the angle between the longitudinal axis of the thiol molecules and the normal to the substrate.36 The tilt angle decreases with temperature, which leads to a reduction in the tip-sample distance necessary to rupture the adhesive contact. As a result, a slight decrease of xβ* in the interaction potential could be observed. Also an interplay between unbinding and rebinding (Figure 1) can influence the interaction potential parameters. The rates of unbinding and rebinding processes are determined basically by the heights of corresponding energy barriers in relation to kBT. If the probability of rebinding increases faster with temperature up to 35°C than that of unbinding, then an increase in ∆Gβ* can be observed. A similar effect has been observed for friction between a silicon AFM tip and some substrates, e.g., a silicon wafer.37 However, in the case of our results, extraction of the interaction potential parameters was performed for the region of high loading rates, where rebinding processes are negligible. Thus, the thermal rearrangements (but without decomposition or desorption) of the molecules rather than rebinding processes are responsible for observed changes of ∆Gβ* and xβ* versus temperature. Over 65°C, the two-dimensional melting of the SAMs occurs38-40 (Ramin and Jabbarzadeh report about 60°C for 1-dodecanthiol36), which leads to significant changes in the adhesion force; hence, we limited our analysis to 65°C.

Summary and Conclusions We applied high-rate dynamic force spectroscopy measurements to study in detail the influence of chain length and temperature on the adhesion between methyl-terminated thiol monolayers and a silicon nitride tip. We used the JKR model to calculate the number of molecules in adhesive contact and then applied the DHS model to extract the information 16 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

about the interaction potential per single molecule. We have decided to apply the DHS model for two reasons: i) it better describes the behavior of adhesive bonds at high loading-rates than other models of thermally activated escape from the potential well, and ii) it delivers information about the height of the activation barrier (∆Gβ*) additionally to its position (xβ*). Finally, we were able to determine the dependence of ∆Gβ* and xβ* versus temperature for two thiols: 1-tetradecanethiol and 1-hexadecanethiol. We compared the results with previously obtained data for 1-decanethiol.8 The increase of ∆Gβ* and xβ* versus chain length we associate with lower stiffness of longer molecules which facilitates adhesive bonds formation. The thermal changes of the parameters, however, we relate rather to rearrangements of molecules than direct influence of temperature on the adhesive bonds. Since the higher limit of adjusted temperatures is below the melting point, the rearrangements can be due to the thermal changes of the tilt angle of molecules in the layer. The demonstrated measurements and calculations enabled us to deepen the knowledge about adhesive bonds and their temperature dependence at the single molecule level.

Associated content Supporting Information: Details on the structure-property relationships of 1-tetradecanethiol and 1-hexadecanethiol SAMs on Au(111) at various temperatures; description of the sample preparation; characterization of Au(111) substrates with XRD (Fig. S1a), STM (Fig. S1b) and XPS (Fig. S3 and S4); characterization of the thiol SAMs with STM (Fig. S2) and XPS (Fig. S3 and S4); intermediate (Tables S1 and S2) and complete, final (Table S3) results. This information is available free of charge on the ACS Publications website at DOI:

Acknowledgements 17 ACS Paragon Plus Environment

Page 24 of 30

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Authors acknowledge the Ministry of Science and Higher Education in Poland for financial support within Project No. 06/62/DSPB/2173. Hubert Gojzewski acknowledges the Ministry of Science and Higher Education in Poland for the stipend Stypendium dla wybitnych mlodych naukowcow. We are thankful to Prof. Hans-Jürgen Butt from the Max Planck Institute for Polymer Research, who provided expertise that greatly assisted the research. We are also thankful to MSc. William Hannemann from the Univeristy of Twente for language proofreading of the manuscript and valuable remarks.

References (1)

Bhushan, B. Nanotribology and nanomechanics in nano/biotechnology. Phil. Trans. R.

Soc. A 2008, 366, 1499-1537. (2)

Srisombat, L.; Jamison, A. C.; Lee, T. R. Stability: A key issue for self-assembled

monolayers on gold as thin-film coatings and nanoparticle protectants. Colloids Surf., A 2011, 390, 1-19. (3)

Liu, H.; Bhushan, B. Investigation of nanotribological properties of self-assembled

monolayers with alkyl and biphenyl spacer chains (Invited). Ultramicroscopy 2002, 91, 185202. (4)

Yang, X.; Perry, S. S. Friction and molecular order of alkanethiol self-assembled

monolayers on Au(111) at elevated temperatures measured by atomic force microscopy. Langmuir 2003, 19, 6135-6139. (5)

Binnig, G.; Quate, C. F.; Gerber, C. Atomic force microscope. Phys. Rev. Lett. 1986,

56, 930-933. (6)

Johnson, K. L.; Kendall, K.; Roberts, A. D. Surface energy and the contact of elastic

solids. Proc. R. Soc. Lond. Ser. A-Math. Phys. Eng. Sci. 1971, 324, 301-313.

18 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(7)

Dudko, O. K.; Hummer, G.; Szabo, A. Intrinsic rates and activation free energies from

single-molecule pulling experiments. Phys. Rev. Lett. 2006, 96, 108101. (8)

Gojzewski, H.; Kappl, M.; Butt, H.-J.; Ptak, A. Influence of Temperature on the

Nanoadhesion of a Methyl-Terminated Thiol Monolayer: A New Insight with High-Rate Dynamic Force Spectroscopy. Langmuir 2016, 32, 4500–4508. (9)

Ishida, T.; Fukushima, H.; Mizutani, W.; Miyashita, S.; Ogiso, H.; Ozaki, K.;

Tokumoto, H. Annealing effect of self-assembled monolayers generated from terphenyl derivatized thiols on Au(111). Langmuir 2002, 18, 83-92. (10)

McCarley, R. L.; Dunaway, D. J.; Willicut, R. J. Mobility of the alkanethiol-gold(111)

interface studied by scanning probe microscopy. Langmuir 1993, 9, 2775-2777. (11)

Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-

assembled monolayers of thiolates on metals as a form of nanotechnology. Chem. Rev. 2005, 105, 1103-1169. (12)

Bell, G. I. Models for specific adhesion of cells to cells. Science 1978, 200, 618-627.

(13)

Evans, E.; Ritchie, K. Dynamic strength of molecular adhesion bonds. Biophys. J.

1997, 72, 1541-1555. (14)

Merkel, R.; Nassoy, P.; Leung, A.; Ritchie, K.; Evans, E. Energy landscapes of

receptor-ligand bonds explored with dynamic force spectroscopy. Nature 1999, 397, 50-53. (15)

Strunz, T.; Oroszlan, K.; Schafer, R.; Guntherodt, H. J. Dynamic force spectroscopy of

single DNA molecules. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 11277-11282. (16)

Taninaka, A.; Takeuchi, O.; Shigekawa, H. Reconsideration of dynamic force

spectroscopy analysis of streptavidin-biotin interactions. Int. J. Mol. Sci. 2010, 11, 2135-2151. (17)

Neuert, G.; Albrecht, C.; Pamir, E.; Gaub, H. E. Dynamic force spectroscopy of the

digoxigenin-antibody complex. FEBS Lett. 2006, 580, 505-509.

19 ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(18)

Ptak, A.; Makowski, M.; Cichomski, M. Characterization of nanoscale adhesion

between a fluoroalkyl silane monolayer and a silicon AFM tip. Complex character of the interaction potential. Chem. Phys. Lett. 2010, 489, 54-58. (19)

Ptak, A.; Kappl, M.; Moreno-Flores, S.; Gojzewski, H.; Butt, H. J. Quantitative

Characterization of Nanoadhesion by Dynamic Force Spectroscopy. Langmuir 2009, 25, 256261. (20)

Gojzewski, H.; Kappl, M.; Ptak, A.; Butt, H. J. Effect of Humidity on Nanoscale

Adhesion on Self-Assembled Thiol Monolayers Studied by Dynamic Force Spectroscopy. Langmuir 2010, 26, 1837-1847. (21)

Friddle, R. W.; Noy, A.; De Yoreo, J. J. Interpreting the widespread nonlinear force

spectra of intermolecular bonds. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 13573-13578. (22)

Kramers, H. A. Brownian motion in a field of force and the diffusion model of

chemical reactions. Physica 1940, 7, 284-304. (23)

Li, D.; Ji, B. Crucial roles of bond rebinding in rupture behaviors of single molecular

bond at ultralow loading rates. Int. J. Appl. Mech. 2015, 7, 1550015. (24)

Ptak, A.; Kappl, M.; Butt, H. J. Modified atomic force microscope for high-rate

dynamic force spectroscopy. Appl. Phys. Lett. 2006, 88, 263109. (25)

Hutter, J. L.; Bechhoefer, J. Calibration of atomic-force microscope tips. Rev. Sci.

Instrum. 1993, 64, 1868-1873. (26)

Li, D.; Ji, B. Predicted rupture force of a single molecular bond becomes rate

independent at ultralow loading rates. Phys. Rev. Lett. 2014, 112, 078302. (27)

Evans, E. Probing the relation between force - Lifetime - and chemistry in single

molecular bonds. Annu. Rev. Biophys. Biomolec. Struct. 2001, 30, 105-128. (28)

Williams, P. M. Analytical descriptions of dynamic force spectroscopy: behaviour of

multiple connections. Anal. Chim. Acta 2003, 479, 107-115.

20 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(29)

Kiridena, W.; Jain, V.; Kuo, P. K.; Liu, G. y. Nanometer‐scale Elasticity

Measurements on Organic Monolayers Using Scanning Force Microscopy. Surf. Interface Anal. 1997, 25, 383-389. (30)

Burns, A. R.; Houston, J. E.; Carpick, R. W.; Michalske, T. A. Friction and Molecular

Deformation in the Tensile Regime. Phys. Rev. Lett. 1999, 82, 1181-1184. (31)

Leng, Y.; Jiang, S. Atomic indentation and friction of self-assembled monolayers by

hybrid molecular simulations. J. Chem. Phys. 2000, 113, 8800. (32)

Ramin, L.; Jabbarzadeh, A. Effect of compression on self-assembled monolayers: A

molecular dynamics study. Model. Simul. Mater. Sci. Eng. 2012, 20. (33)

Delrio, F. W.; Jaye, C.; Fischer, D. A.; Cook, R. F. Elastic and adhesive properties of

alkanethiol self-assembled monolayers on gold. Appl. Phys. Lett. 2009, 94, 131909 (34)

Ramin, L.; Jabbarzadeh, A. Effect of load on structural and frictional properties of

alkanethiol self-assembled monolayers on gold: Some odd-even effects. Langmuir 2012, 28, 4102-4112. (35)

Zepeda, S.; Yeh, Y.; Noy, A. Determination of energy barriers for intermolecular

interactions by variable temperature dynamic force spectroscopy. Langmuir 2003, 19, 14571461. (36)

Ramin, L.; Jabbarzadeh, A. Odd-even effects on the structure, stability, and phase

transition of alkanethiol self-assembled monolayers. Langmuir 2011, 27, 9748-9759. (37)

Barel, I.; Urbakh, M.; Jansen, L.; Schirmeisen, A. Multibond dynamics of nanoscale

friction: The role of temperature. Phys. Rev. Lett. 2010, 104, 066104. (38)

Yang, G.; Liu, G. Y. New insights for self-assembled monolayers of organothiols on

Au(111) revealed by scanning tunneling microscopy. J. Phys. Chem. B 2003, 107, 8746-8759. (39)

Xiao, X.; Wang, B.; Zhang, C.; Yang, Z.; Loy, M. M. T. Thermal annealing effect of

alkanethiol monolayers on Au(111) in air. Surf. Sci. 2001, 472, 41-50.

21 ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(40)

Qian, Y.; Yang, G.; Yu, J.; Jung, T. A.; Liu, G. Y. Structures of annealed decanethiol

self-assembled monolayers on Au(111): An ultrahigh vacuum scanning tunneling microscopy study. Langmuir 2003, 19, 6056-6065.

22 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of Contents Image

23 ACS Paragon Plus Environment

Page 30 of 30