Mechanical Contact Spectroscopy: Characterizing Nanoscale

Apr 9, 2019 - ... Characterizing Nanoscale Adhesive Contacts via Thermal Forces ... Investigation of the Hydrophobic Nature of Metal Oxide Surfaces Cr...
0 downloads 0 Views 2MB Size
Download PDF
Article Cite This: Langmuir XXXX, XXX, XXX−XXX

pubs.acs.org/Langmuir

Mechanical Contact Spectroscopy: Characterizing Nanoscale Adhesive Contacts via Thermal Forces Alexandr Jonaś ,̌ *,† Martin Kochanczyk,‡,⊥ Alexandro D. Ramirez,‡,∥ Michael Speidel,§ and Ernst-Ludwig Florin*,‡ †

The Czech Academy of Sciences, Institute of Scientific Instruments, Královopolská 147, 61264 Brno, Czech Republic Center for Nonlinear Dynamics and Department of Physics, University of Texas at Austin, Austin, Texas 78712, United States § European Molecular Biology Laboratory, Meyerhofstrasse 1, 69117 Heidelberg, Germany Downloaded via UNIV OF SOUTH AUSTRALIA on April 20, 2019 at 14:31:09 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: The adhesion of micro- and nanoparticles to solid substrates immersed in liquids is a problem of great scientific and technological importance. However, the quantitative characterization of such nanoscale adhesive contacts without rupturing them still presents a major experimental challenge. In this article, we introduce mechanical contact spectroscopy (MCS), an experimental technique for the nondestructive probing of particle adhesion in liquid environments. With MCS, the strength of adhesive contacts is inferred from residual position fluctuations of adherent particles excited by thermal forces. In particular, the strength of adhesion is correlated with the standard deviation of the particle lateral position x, with smaller position standard deviations ξ = Δx 2 indicating higher adhesive strength. For a given combination of particles, substrate, and immersion medium, the adhesion is characterized by the mechanical contact spectrum, which is a histogram of ξ values obtained from tracking an ensemble of adherent particles. Because the energy of thermal excitation at room temperature is very small in comparison to the typical total energy of adhesive contacts, the studied contacts remain in equilibrium during the measurement. Using MCS, we study the adhesion of micrometer-sized particles to planar solid substrates under a wide range of environmental conditions, including liquid immersion media of varying ionic strength and adhesion substrates with different chemical functionality of their surfaces. These experiments provide evidence that MCS is capable of reproducibly detecting minute changes in the particle−substrate work of adhesion while at the same time covering the range of adhesive contact strength relevant in the context of surface chemistry, biology, and microfabrication.

■

INTRODUCTION The phenomenon of small particle adhesion is frequently encountered in diverse areas of science and technology.1−3 The formation of adhesive mechanical contacts between micrometer- and submicrometer-sized particles and various types of substrates is critical to the stability and durability of paints and prints.4,5 Contamination of silicon wafers due to the unwanted adhesion of dust particles poses a serious complication in semiconductor manufacturing and processing.6,7 Precise microscale control of adhesion forces determines success or failure in the fabrication of complex integrated microelectromechanical systems.8,9 In life sciences, bacterial adhesion represents the first step necessary for the formation of biofilms,10 and selective adhesion mediated by specialized molecules is a prerequisite for targeted intracellular vesicular transport,11 biological tissue assembly,12 and the regulation and development of the organism’s immune system.13 Adhesion is generally a result of a complex interplay of various types of attractive and repulsive intermolecular © XXXX American Chemical Society

interactions (electrostatic, van der Waals, acid−base, covalent and hydrogen bonding, hydrophobic, and Born repulsion) between material objects.14,15 The overall effect of these interactions can be summarized by the change in free energy upon separating to infinity two material surfaces that are initially in contact with each other. This free-energy change is called the work of adhesion and is defined as W = γ1 + γ2 − γ12, where γ1 and γ2 are the free energies per unit surface area of materials 1 and 2 with respect to the medium surrounding them and γ12 is the free energy per unit area of the interface between materials 1 and 2.14 The work of adhesion serves as a quantitative characteristics of the strength of adhesive contacts. In particular, adhesive contacts that display higher strength (i.e., higher resistance to externally applied forces) are associated with higher values of W. Received: December 7, 2018 Revised: April 3, 2019 Published: April 9, 2019 A

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

experimental technique for the nondestructive characterization of adhesive contacts formed between micrometer- and submicrometer-sized colloidal particles and substrates in liquid environments. With MCS, the strength of adhesion is inferred from the residual lateral position fluctuations of the adherent particles excited by thermal forces. In particular, under the given experimental conditions, higher adhesive strength is indicated by smaller standard deviations of the particle positions. In other words, the adhesive contacts of particles displaying smaller position fluctuations are more resistant to rupture by externally applied forces and torques. The specific studied combination of particles, substrate, and immersion medium is then described by its mechanical contact spectrum, which is a histogram of positional standard deviations obtained from tracking an ensemble of adherent particles. Thermal excitation represents a relevant energy scale for probing colloidal interactions that lead to particle adhesion.39,40 At the same time, the energy of thermal excitation at room temperature is very small in comparison to the typical total energy of the adhesive contacts. Consequently, thermal forces do not appreciably change the size of the particle−substrate contact area. The studied adhesive contact remains in equilibrium at all times, and thus the MCS measurement reflects the equilibrium work of adhesion. We show that MCS is a noninvasive technique for probing nanoscale adhesive contacts that is capable of detecting minute changes in the particle−substrate work of adhesion while also covering a range of adhesive contact strength relevant in the context of surface chemistry, biology, and microfabrication.

Traditionally, on the macroscopic level, the strength of adhesion between two bodies is tested by “pull-off” or “peeloff” techniques.16−18 However, the results of these macroscopic measurements cannot be extrapolated directly to the adhesion of microscopic particles because the strength of adhesion on the microscale depends strongly on the local contact geometry and roughness of surfaces in contact.19,20 Moreover, the contribution of individual types of forces to the overall work of adhesion changes with changing dimensions of adhering objects.21 An investigation of adhesive contacts on the micrometer and submicrometer scales typically involves pull-off types of measurement.22,23 In this approach, the force required to separate two adhering bodies is determined using various techniques such as atomic force microscopy (AFM),24,25 the colloidal particle interaction apparatus,26,27 the surface force apparatus,14 centrifugation,28 and hydrodynamic detachment.29,30 All of these techniques require the contact to be ruptured to determine the pull-off force. Therefore, they are nonequilibrium in nature, and the measured work of separation depends on the loading rate.31,32 Because of the energy dissipated during the adhesive contact fracture, the measured work of separation is in general higher than the true work of adhesion.33,34 Alternatively, the strength of particle−substrate adhesion can be determined from the measurement of the size of the contact area using optical35 or electron microscopy36 and an appropriate model of the adhesive contact. However, because the resolution of optical microscopy is limited by light diffraction, this technique is restricted to studying contacts of diameter >500 nm, well above the characteristic dimensions of contact areas of submicrometer-sized particles. Electron microscopy achieves nanometer resolution but operates under conditions not compatible with in situ measurements under a wide range of relevant environmental conditions (air at atmospheric pressure and liquids). A nondestructive method for characterizing the adhesion of micrometer-sized particles to solid substrates via acoustic excitation of the adhesion substrate was proposed in ref 37. In this approach, the studied substrate with adherent particles is excited with acoustic pulses, and the axial displacement of the particles is recorded using reflected-light laser interferometry. The work of adhesion can be subsequently determined from the measured natural frequency of the particle’s oscillatory motion using a simplified model of adhesive contact of elastic solids.38 This technique was used to study the adhesion of 21.4 μm polystyrene particles to various substrates in air.37,38 However, its application to studying particle adhesion in liquid environments is complicated by the complex hydrodynamics of the liquid in the vicinity of the particle that is excited by the oscillating substrate. Furthermore, the use of reflected-light laser interferometry for the particle position tracking makes the tracking of particles with sizes comparable to the wavelength of used light challenging. In addition, this technique tracks only one particle at a time; thus, recording a representative data set for the given particle−substrate combination is rather timeconsuming. Despite recent advances fueled by the scientific and technological importance of particle adhesion, the quantitative characterization of adhesion of immersed micro- and nanoparticles carried out without rupturing the adhesive contact still presents a major experimental challenge. In this article, we introduce mechanical contact spectroscopy (MCS), a novel

■

EXPERIMENTAL METHODS

Mechanics of Adhesive Contacts. When two elastic bodies are in contact with adhesive forces acting between them, deformation occurs in the contact zone and a finite size contact area forms even in the absence of external compressive forces. This process eventually reaches equilibrium when the elastic forces resulting from the body deformation just balance the adhesion forces. Within the framework of continuum mechanics, the equilibrium size of the contact area can be related to the elastic properties of the materials, the size and shape of the adhering bodies, and the work of adhesion.41 In the following section, we are interested in the special case of an elastic sphere adhering to a rigid planar substrate. There exist two well-established models that allow the determination of the equilibrium contact area radius. The Johnson−Kendall−Roberts (JKR) model of adhesive contacts42 applies to the case of a relatively large, compliant sphere adhering to a substrate with strong, short-range adhesive forces that are assumed to act exclusively within the contact area. On the other hand, the Derjaguin−Muller−Toporov (DMT) model43 is applicable to small, stiff spheres and weak, long-range adhesive forces also acting outside of the contact area. The JKR and DMT models that have been frequently adopted in the literature to evaluate the results of pull-off force experiments carried out using various force probe techniques24,27 are the opposite limiting cases of a general theory of adhesive contacts covering all ranges of action of the adhesive forces.44 Thus, they provide boundaries for the expected scaling of the contact area size with the work of adhesion. When no additional external compressive load is applied to the adherent spherical particle, both models predict that the radius a of the contact area scales as a = (KR2W /E*)1/3

(1)

where K = 9π/2 [3π/2] for the JKR [DMT] model, R is the radius of the sphere, and E* is the effective elastic modulus of the system defined as 1/E* = (1 − ν12)/E1 + (1 − ν22)/E2, with E1 [E2] and ν1 [ν2] being the Young’s modulus and the Poisson’s ratio of the sphere [substrate] material, respectively.41,44 In aqueous environments, the B

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir work of adhesion W is typically in the range of ∼1−100 mJ·m−2.45,46 The corresponding radius of the contact area for a polystyrene sphere (E1 = 2.3 GPa, ν1 = 0.35) with a radius of R = 500 nm in contact with a flat glass substrate (E2 = 62 GPa, ν2 = 0.22) is then in the range of ∼11−52 nm for the JKR model and in the range of ∼8−36 nm for the DMT model. Young’s modulus of glass is much larger than that of polystyrene; thus, E* = 2.5 GPa ≈ E1, and we can assume that the deformation in the contact zone takes place exclusively in the sphere. Adherent Colloidal Particle as a Thermally Driven Harmonic Mechanical Oscillator. The position of an unbound colloidal particle immersed in a fluid stochastically changes due to random collisions between the particle and molecules of the surrounding fluid. These collisions give rise to a fluctuating net force acting on the particle center of mass, with magnitude depending on the particle size, fluid viscosity, and temperature. When such a particle is confined in a harmonic potential U(x) = kx2/2, its position fluctuates around the location of the potential minimum, with a mean displacement whose magnitude can be calculated from the equipartition theorem.47 Conversely, the stiffness k of the confining potential can be calculated from the variance ⟨Δx2⟩ of the particle displacement x as k = kBT /⟨Δx 2⟩

(2)

with kB being the Boltzmann constant and T being the fluid temperature. Thus, the confining forces acting on the particle can be directly determined by a series of position measurements.48 The formation of an adhesive contact between an immersed elastic particle and a rigid substrate is analogous to particle confinement in a three-dimensional potential well. Figure 1a illustrates the particle− substrate contact along with the relevant forces acting on the particle. The adhesive force Fadhesion pulls the particle toward the substrate until the restoring elastic force Felastic originating from deformations of the particle balances it. Thus, the particle is confined along the surface normal direction (the z axis) by the interplay between these two forces. In the lateral direction (the x axis), the particle can potentially roll or slide across the surface, with the relative motion being opposed by the friction force Ffriction. The rolling of an elastically deformed particle over the substrate involves the simultaneous compression of the particle material at the leading edge and relaxation at the trailing edge of the contact area. Resistance to rolling (i.e., the rolling friction) then stems in part from the dissipation of elastic energy in the particle bulk during this process: the energy recovered by relaxation of the deformed particle material is lower than the energy originally required for deformation.30,34 For nanometer-scale contact areas, rolling friction is further enhanced and frequently dominated by the dissipation of surface energy due to adhesion hysteresis: the energy needed to open the crack at the trailing edge of the contact is higher than the energy released during the crack closure at the leading edge.31−34 The sliding motion of the particle is opposed by the sliding friction force that depends on the external load applied normally to the contact area and on the adhesive force Fadhesion. Because of the effects of adhesion, the sliding friction can be nonzero even in the absence of external compressive loads.14,49 Experimental evidence shows that an adherent microscopic particle can withstand substantial lateral forces, and upon pushing, its center of mass can be elastically displaced by tens of nanometers before the onset of rolling.50,51 Forces required to induce particle sliding are then significantly higher than those needed to initiate particle rolling.49−51 Hence, adhesive and friction forces can effectively immobilize the particle, resisting the tendency of the particle to roll or slide across the surface. Indeed, microrheologic studies by Sharma et al. demonstrated that micrometer-sized colloidal particles that are immersed in an aqueous solution and interact with a rigid glass substrate can elastically bond to the glass surface, with the bond stiffness controlled by the experimental conditions.52 Let us now assume that the adherent particle is subject to a randomly fluctuating thermal force Fthermal that tries to displace it. If the effective confinement forces Fadhesion and Ffriction are large enough to prevent the particle from sliding or rolling, then the action of Fthermal on the particle causes bending and shearing of the particle material in the contact region (Figure 1b). These minute

Figure 1. Schematics of adhesive contact between a particle and a flat substrate. (a) Forces relevant to the confinement of the adherent particle: Fadhesion − adhesive force, Felastic − total elastic force given by the integral of the sum of compressive and tensile elastic stresses over the contact area, and Ffriction − static friction force. (b) Thermal force Fthermal causes bending and shearing deformations of the particle contact region that result in excess forces ΔFexcess with a nonzero net torque. These excess forces are proportional to the deformations caused by thermal forces and tend to restore the particle equilibrium position, thus representing an effective confining potential well for the adherent particle. Small compressive/stretching deformations Δz of the contact region are translated into much larger lateral displacements Δx of the particle center: Δx = ΔzR/a, where R is the particle radius and a is the contact area radius. Note that the figure is not drawn to scale. In reality, a ≪ R. deformations slightly disrupt the circular symmetry of the contact and induce additional stresses within the contact area whose integral leads to excess forces, ΔFexcess, acting on the particle. These excess forces are associated with a nonzero net torque that tends to restore the particle’s equilibrium position and determines the effective stiffness of the contact. A model of this adhesion-induced torque that leads to rolling friction predicts linear scaling of the adhesion torque with the work of adhesion W, the radius of the particle R, and the lateral displacement Δx of the particle center from its equilibrium configuration with no external torque applied.34 It should be emphasized that the additional deformations of the contact zone due to the thermal force are small in comparison to the primary deformations caused by adhesion. The total adhesive energy of the contact Ua that includes contributions from the surface energy needed to separate the particle from the substrate and the elastic energy stored in the deformed adherent bodies can be expressed as14 Ua = 0.6πa2W =

6π ijj K 2/3R4/3W 5/3 yzz jj zz z 10 jk E*2/3 {

(3)

where the contact radius a was substituted from eq 1. For the case of a 1-μm-diameter polystyrene particle adherent to glass (E* = 2.5 GPa) with a rather low work of adhesion of 1 mJ·m−2,46 Ua is ∼58kBT for C

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir the JKR model and ∼28kBT for the DMT model (i.e., much higher than the thermal energy kBT that drives the stochastic position fluctuations of the particle). Thermal forcing adopted in our adhesion experiments differentiates our approach from the studies reported in ref 53 and 54, where adherent spheres rolled under the influence of a constant external bias force and controlled mechanical noise. In these experiments, the contact area between the sphere and the supporting substrate moved, and the dynamics of the system was strongly affected by nonlinear, hysteretic forces with the character of dry friction.55 In such a scenario, the bonds between the particle and the substrate repeatedly break and form. In our experiments, in contrast, we exploit solely the weak thermal excitation and, within the formalism introduced in ref 55 and also adopted in ref 53, operate in the regime where the driving force is too weak to overcome the friction. Therefore, the particle remains completely stuck on the surface, and the location and size of the contact area remain approximately constant. According to the theoretical model of microscale rolling friction proposed by Krijt et al.,34 the adhesive torque that prevents the particles from rolling across the surface is directly proportional to the lateral displacement of the particle center with respect to the relaxed state of the contact. Therefore, particle confinement through adhesion can be described by a harmonic potential with a single characteristic value of stiffness k. Because the adhesive torque scales linearly with W and R, the effective stiffness of the contact k is expected to have the same dependence on these quantities. Intuitively, k increases with increasing W and R because the amount of particle material that has to be deformed in the contact region becomes larger. From the equipartition theorem represented by eq 2, it then follows that the observed variance ⟨Δx2⟩ of the adherent particle’s position decreases with increasing W. Thus, the particle position variance can be used as a sensitive measure of the strength of adhesion for a given combination of particle/substrate/immersion medium. Physicochemical properties of the particles and the substrate surface inevitably display certain variability. Therefore, to obtain a robust indicator of the adhesion strength for the studied system, it is necessary to analyze the position fluctuation patterns of multiple particles recorded under identical experimental conditions. Such a set of values of ⟨Δx2⟩ then forms the basis for generating mechanical contact spectra. In principle, the strength of particle adhesion can be assessed from analyzing either the axial (normal to the surface) or lateral (parallel to the surface) position fluctuations of the particle. However, in practice, it is much easier and more precise to determine the lateral position of a submicrometer-sized object. This can be accomplished, for example, by using video microscopy that also allows the simultaneous tracking of multiple particles.56 Moreover, the lateral position fluctuations of an adherent particle are significantly larger than the particle’s axial fluctuations due to the geometric amplification effect. (See Figure 1b for an explanation.) For the above reasons, we based our experimental approach on monitoring the lateral position fluctuations of adherent particles. Sample Preparation. Using the concept introduced in the previous section, we studied experimentally the strength of adhesion of micrometer-sized particles to substrates of varied chemical functionality immersed in buffers of controlled ionic strength. A careful adjustment of the parameters of the immersion medium and the adhesion substrate allowed us to systematically investigate the sensitivity and limits of operation of mechanical contact spectroscopy. We used carboxylate-modified green fluorescent polystyrene beads of 1 μm diameter (product no. F-8823; Invitrogen/Thermo Fisher Scientific) as model adherent particles. For the adhesion experiments, the beads were diluted in MES [2-(N-morpholino)ethanesulfonic acid] or PBS [phosphate-buffered saline] buffers with controlled pH. We adjusted the concentration of sodium chloride (NaCl) in the buffer to change the ionic strength and, consequently, the level of electrostatic shielding of the Coulombic interaction between the particles and the substrate. We employed cleaned glass coverslips as adhesion substrates; the coverslips were either plain or functionalized with organosilane layers in order to adjust the adhesive properties of their surfaces. In the beginning of an experiment, a bead suspension

was injected into a sample chamber formed from a stainless steel holder with the adhesion substrate attached to the bottom and another cleaned glass coverslip attached to the top (inset in Figure 2).

Figure 2. Experimental setup for mechanical contact spectroscopy. The position of adherent fluorescent particles is tracked using a standard epifluorescence microscopy arrangement with mercury lamp illumination. The inset shows the detail of the sample chamber that is formed between the adhesion substrate and a cleaned coverslip attached to the opposite sides of a stainless steel holder. Because the density of polystyrene (ρp = 1050 kg·m−3) is slightly larger than the density of water (ρw = 1000 kg·m−3), the beads gradually settled on the surface of the adhesion substrate. After loading the beads, the sample chamber was sealed and mounted on a fluorescence microscope to carry out the tracking of the adherent particle positions. Detailed sample preparation steps including the description of the used materials, substrate cleaning procedures, and surface functionalization protocols can be found in the Supporting Information. Particle Tracking. Our experimental technique for characterizing the strength of particle−substrate adhesion exploits the analysis of residual lateral position fluctuations of the adherent particles. To follow the particle motion, we used high-resolution video tracking based on epifluorescence microscopy. (See Figure 2 for an illustration of the particle tracking setup.) The samples were observed through an oil-immersion objective lens (Plan Neofluar 100×/N.A. 1.3; Zeiss) with a high-sensitivity cooled digital CCD camera (PCO SensiCam QE; Cooke Corp) mounted on an inverted microscope (Axiovert 10; Zeiss). We recorded a stack of in-focus images of adherent beads that was subsequently processed off-line to extract the lateral positions of the beads in each frame of the stack using a custom-written code developed in IgorPro (Wavemetrics). In particular, the positions of the beads in each frame of the stack were found by fitting a 2D Gaussian function to the bead image and identifying the bead position with the location of the center of the fitted 2D Gaussian.57 With this technique, bead positions could be obtained with subnanometer resolution on the millisecond time scale.58 (See the Supporting Information for the results of resolution tests of our particle tracking system.) Additional benefits of working with fluorescent beads lie in the relative simplicity of the experimental setup, straightforward image processing, and the possibility of reliable discrimination of the target particles against the background. Typically, the bead images were acquired with a camera exposure time of 1 ms, and the recorded trajectories of the beads consisted of 500 positions. Video-based particle tracking provides simultaneous information on the position of multiple particles located in the camera’s field of view, which greatly enhances the speed of acquiring sufficient data for generating a mechanical contact spectrum. Using the full size of the CCD chip with the 100× magnification objective lens, we could track as many as D

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir ∼60 beads per field of view. When selecting the beads for the analysis, we excluded the beads that were closer than about three bead diameters from each other because their proximity could lead to a bias in determining their positions with our tracking algorithm. In addition, there can be a real physical correlation between the position fluctuations of neighboring beads due to hydrodynamic coupling.59 These requirements set the practical limit on the useful number of beads per field of view that can be tracked simultaneously. Because the adhesive properties of the studied substrate can vary slightly across the surface, we acquired position tracks of adherent beads at multiple randomly selected locations within the sample and included all of them in the final analysis in order to obtain representative characteristics of the given adhesion system. When the random residual position fluctuations of an adherent particle are sampled using a detector with a finite bandwidth, the true particle position track is averaged over the detector integration time. This low-pass filtering effect leads to an underestimation of the particle’s position variance ⟨Δx2⟩ and, according to eq 2, an overestimation of the effective stiffness k of the particle confinement. In the context of mechanical contact spectroscopy, this then translates into overestimating the work of adhesion. For a particle confined in a harmonic potential well, it is in principle possible to recover the real variance of the particle position fluctuations; in order to do that, the characteristic relaxation time τ of the particle in the potential well must be known.60,61 However, for immersed adherent particles located in the vicinity of a solid surface, the precise determination of τ is rather challenging because of complex hydrodynamic coupling effects62 and additional contributions to the overall damping force acting on the particle.30 Alternatively, when only a relative comparison of different particle adhesion systems is of interest, it is possible to use the low-pass-filtered particle position data directly to characterize the adhesion. In this article, we adopt the latter approach, bearing in mind its limitations for the determination of the absolute value of the work of adhesion.

Figure 3. Residual thermal position fluctuations of an adherent particle immersed in a liquid. (a) Typical time trace of the lateral position x of an adherent particle determined by the particle tracking algorithm. (b) Histogram P(x) of the particle position trace shown in (a). Experimental data (bars) are well fitted with a Gaussian (black curve), thus justifying the harmonic potential model of the adhesive contact. The width of the Gaussian is given by the particle position variance ⟨Δx2⟩; the square root of the variance ξ = Δx 2 then characterizes the strength of the particle−substrate adhesion. Experimental parameters: 1 μm carboxylate-modified polystyrene particle, glass substrate, MES buffer (pH 6) with 200 mM NaCl, camera integration time 1 ms.

■

RESULTS AND DISCUSSION Particle adhesion experiments reported in this article were performed on planar substrates with varied physicochemical properties immersed in buffers of different ionic strength. Adjusting the properties of the substrate surface and the buffer allowed us to tune the strength of adhesion in a controlled way and examine the response in the observed position fluctuations of the adherent particles. Generation of Mechanical Contact Spectra. Figure 3a shows a typical track of the lateral position x of an adherent particle recorded over a period of ∼35 s; the particle position fluctuations are clearly confined and display a finite variance ⟨Δx2⟩ that does not change with time. The fact that ⟨Δx2⟩ is time-independent is consistent with the notion of the suppressed long-range rolling or sliding motion of the particles immobilized on the surface by adhesion. The histogram P(x) of the particle lateral positions can be related to the potential energy U(x) of the particle−surface interaction via the Boltzmann distribution: P(x) = P0 exp(−U(x)/kBT).48 As illustrated in Figure 3b, the histogram of lateral positions of the adherent particle has a Gaussian shape. The Gaussian form of P(x) that can be expressed as P(x) = P0 exp(−x2/[2⟨Δx2⟩]) then implies that the particle lateral confinement through adhesion can be described by a harmonic potential U(x) = kx2/2 with the effective stiffness k = kBT/⟨Δx2⟩. The formula for k is just another statement of the equipartition theorem of eq 2. On the basis of the reasoning presented in the Experimental Methods section, we can link the observed values of ⟨Δx2⟩ to the strength of the adhesive contact. For each tracked particle, the variance ⟨Δx2⟩ of its lateral position is calculated. The square root of the variance,

ξ = ⟨Δx 2⟩ , then represents the mean fluctuation of the particle about its equilibrium position. Values of ξ obtained for all adherent particles tracked under particular experimental conditions (i.e., the given adhesion substrate and buffer) form the basis for generating a mechanical contact spectrum. The mechanical contact spectrum is a histogram of ξ of all considered particles; it characterizes the adhesive properties of the given combination of particles, substrate, and immersion buffer. A typical example of the mechanical contact spectrum is shown in Figure 4. The averaged mean fluctuation ξ̅ of the whole analyzed particle population reports on the characteristic strength of the particle−substrate adhesion, with smaller values of ξ̅ indicating higher adhesive strength. The overall shape and width of the spectral peak then report on the heterogeneity of the adhesive properties of the substrate and individual particles; peak broadening corresponds to a greater spread of the adhesive strength within the studied particle population and is also affected by the particle size distribution. Using experimental data acquired under varied adhesion conditions, we demonstrate below that mechanical contact spectra indeed represent a sensitive indicator of the adhesive contact strength. Furthermore, repeated adhesion experiments carried out under constant experimental conditions illustrate the robustness of mechanical contact spectra that reproduce very well with the adhesion substrates prepared in parallel, following the same cleaning and surface functionalization E

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

particles and the substrate possess surface charges, then the most straightforward way of altering the strength of adhesion between them is by screening the electrostatic component of their interaction. This can be achieved by changing the ionic strength I of the immersion buffer. Higher ionic strength leads to stronger shielding of the electrostatic forces between the two bodies and thus a shorter range of their action quantified by the Debye length κ −1 = (εkBT )/(2NAe 2I ) .14 In the expression for κ−1, ε is the absolute permittivity of the immersion medium, NA is Avogadro’s number, e is the elementary charge, and a monovalent 1:1 salt solution is assumed, with the ionic strength equal to the salt concentration. In the case of the attractive electrostatic force, the strength of adhesion decreases with decreasing κ−1, and for the repulsive electrostatic force, the strength of adhesion increases with decreasing κ−1. Thus, by manipulating the value of κ−1, the particle adhesion strength can be gradually adjusted. In our experiments, we studied the adhesion of carboxylatemodified polystyrene particles to bare glass surfaces in MES buffer at pH 6 with the ionic strength adjusted by the addition of a controlled amount of NaCl. Figure 5 shows the mechanical contact spectra obtained for the concentration of NaCl in the solution increasing gradually from 100 to 200 mM (panels a−d). With higher salt concentration, the averaged particle displacement ξ̅ decreases and the spectral peak becomes narrower, which indicates the increasing strength of particle adhesion and its higher uniformity across the surface. This result can be explained if we consider the charging state of both surfaces under the actual experimental conditions. At pH 6, carboxylate groups at the particle surface are largely dissociated (carboxylate pK ≈ 4.9 63), and the particles are strongly negatively charged. (According to the manufacturer’s

Figure 4. Mechanical contact spectrum of adherent particles immersed in a liquid. The spectrum is formed by a histogram of the mean position fluctuations ξ obtained for a population of particles adherent to a particular substrate immersed in a given buffer. The histogram counts are normalized with respect to the total number of particles N included in the spectrum so that the sum over all histogram bins is equal to 100. The black vertical bar on the left side of the spectrum indicates the resolution limit of our particle tracking technique (0.7 nm under the given experimental conditions). The averaged mean fluctuation ξ̅ of the adherent particle population then reports on the typical strength of particle adhesion. Experimental parameters: 1 μm carboxylate-modified polystyrene particles, glass substrate, MES buffer (pH 6) with 175 mM NaCl, and camera integration time 1 ms.

protocols. (For details, see Figures S3 and S4 and the related discussion in the Supporting Information.) Adhesion to Plain Glass Surfaces. The simplest situation for studying particle adhesion in a liquid environment is represented by a bare solid substrate immersed in a buffer of well-defined pH and ionic strength. If both the adherent

Figure 5. Mechanical contact spectra recorded at varied ionic strength of the immersion buffer. Presented spectra characterize the adhesion of 1μm-diameter carboxylate-modified polystyrene beads to cleaned bare glass surfaces in MES buffer (pH 6) with increasing NaCl concentration: (a) 100 mM NaCl, (b) 150 mM NaCl, (c) 175 mM NaCl, and (d) 200 mM NaCl. F

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 6. Functionalization of glass surfaces with organosilane molecules. (a) 3-[2-(2-Aminoethylamino)ethylamino]propyl-trimethoxysilane (DETA), a positively charged molecule with a molecular length of ∼2.7 nm.69 (b) (3-Mercaptopropyl)trimethoxysilane (MPTMS), a neutral molecule with a molecular length of ∼0.9 nm.70 Note that the figures are not drawn to scale.

Figure 7. Mechanical contact spectra of silanized glass surfaces recorded at a constant ionic strength of immersion buffer. The presented spectra characterize the adhesion of 1-μm-diameter carboxylate-modified polystyrene beads (negatively charged) to (a) a DETA-coated glass surface and (b) an MPTMS-coated glass surface in PBS buffer (pH 7.6) with 13.7 mM NaCl. Silanized substrates are positively charged (DETA) and neutral (MPTMS) at the working buffer pH. The black vertical line shown in part b identifies the value of ξ̅ observed in part a.

specification, the surface charge density is −32.5 μC·cm−2). At neutral and slightly acidic pH, a clean glass surface carries a small negative charge that arises from the dissociation of terminal SiOH (silanol) groups.63 Hence, the electrostatic force between the particles and the substrate is repulsive and destabilizes the adhesion, which is then maintained by other attractive interactions, mostly van der Waals and Lewis acid− base.64 In fact, when the experiment is performed in pure MES buffer whose ionic strength is negligible,65 the long-range repulsion is so strong that it completely prevents the particles from adhering. Thus, some minimal amount of NaCl has to be added to the solution to enable the particles to overcome the repulsive potential barrier and start adhering. With increasing concentration of NaCl in the buffer, electrostatic repulsion is attenuated, and the particles start adhering more strongly. While the change in NaCl concentration in the solution allows the continuous adjustment of the adhesion strength, it can be used only to control the adhesion within a limited range. Too much salt leads to particle clustering, as the electrostatic repulsive force, which stabilizes the negatively charged colloidal particles in the suspension, is shielded along with shielding the particle−substrate repulsion.14 Partial particle clustering was already observed with the highest salt concentration of 200 mM used to obtain the data of Figure 5d. To test the mechanical contact spectroscopy at a higher strength of adhesion that characterizes, for example, relevant interactions between hydrophobic surfaces in aqueous environ-

ments45,46 or acid−base interactions,66 a different strategy has to be used. Adhesion to Silanized Glass Surfaces. The strength of particle adhesion is largely determined by the physicochemical properties of the surface of the adhesion substrate. Thus, the deposition of a carefully chosen molecular layer over the substrate can make its surface strongly adhesive or, conversely, repellent for the particles. In our experiments, we modified the adhesiveness of the bare glass surfaces by coating them with organosilane layers. Organosilanes that can be covalently immobilized on substrates with terminal silanol groups, such as glass or silica, are frequently employed as adhesion promoters. The popularity of silane-based coatings is derived from the wide choice of available surface chemistries and good control over the homogeneity of the coated surface.67 A suitable choice of the silane terminal group allows the adjustment of the surface charge independently of the surface charge of the bare substrate. Furthermore, the surface charge can be reversibly tuned by changing the pH of the immersion buffer.68 Two different silane molecules were used in our adhesion experiments that were carried out in PBS buffer at pH 7.6. To prepare a positively charged surface, we coated glass coverslips with 3-[2-(2-aminoethylamino)ethylamino]propyl-trimethoxysilane (DETA; see Figure 6a). This molecule contains three protonatable amine groups, and at a neutral pH in aqueous solutions, it is a trivalent cation.71 The proton-accepting capacity of the amine groups implies the basic character of the G

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 8. Mechanical contact spectra of silanized glass surfaces recorded at varied ionic strength of immersion buffer. The presented spectra characterize the adhesion of 1-μm-diameter carboxylate-modified polystyrene beads to an MPTMS-coated glass surface in PBS buffer (pH 7.6) with (a) 13.7 mM NaCl and (b) 137 mM NaCl. The inset in part b shows both mechanical contact spectra in a common system of axes. Note the different scale of the horizontal axes between parts a and b.

DETA-coated surface.68 Furthermore, the presence of a terminal N atom in the silane molecule facilitates the formation of hydrogen bonds with hydrogen atoms bound to highelectronegativity atoms (O, N) on the opposite adhering surface. The second silane used in our adhesion experiments was (3mercaptopropyl)trimethoxysilane (MPTMS; see Figure 6b). In contrast to aminosilane, mercaptosilane with an SH− terminal group is neutral at pH 7.6 and does not possess the character of a chemical base. In addition, the SH− group of MPTMS does not participate in hydrogen bonding. This difference between the chemical nature of aminosilane and mercaptosilane coatings is reflected in contact angle measurements suggesting that MPTMS renders the coated glass surface more hydrophobic (the contact angle of water has been reported to be ∼17° on DETA films72 and >60° on MPTMS films73). In both studied cases, the presence of the silane layer leads to a slight attenuation of the van der Waals interaction between the particle and the substrate due to the decrease in the substrate Hamaker constant.74 (See eq S4 in the Supporting Information for the van der Waals interaction formula.) Additional details on the surface silanization protocol can be found in the Supporting Information. Figure 7 shows the mechanical contact spectra of carboxylate-modified polystyrene particles adherent to DETA-coated glass (a) and MPTMS-coated glass (b) in PBS buffer (pH 7.6) containing 13.7 mM NaCl. This salt concentration is approximately 8 times lower than that used in the adhesion experiments with plain glass substrates shown in Figure 5a, and it is not sufficient to induce the adhesion of carboxylate-modified beads to the plain glass surface. The spectra presented in Figure 7 illustrate the remarkable difference in the adhesiveness of the two studied silanized surfaces for carboxylate-modified particles. In the case of the DETA-coated substrate, we observe, in accord with the expectation, strong adhesion due to the attractive electrostatic interaction between the negatively charged bead and the positively charged substrate that is further augmented by acid− base bonding and hydrogen bonding between the carboxylate groups on the bead surface and the amine groups on the substrate.68 This is indicated by the very small value of ξ̅ of the adherent particle population. In contrast, shorter, uncharged MPTMS molecules in the low-ionic-strength solution do not fully shield the negative charge of the glass substrate arising

from the hydrolyzed silanol groups.63 Consequently, the residual electrostatic repulsion between the bead and the substrate and the absence of acid−base and hydrogen bonding lead to weaker adhesion, resulting in significantly larger particle position fluctuations (more than 10-fold in comparison to the DETA adhesion case). Nevertheless, the beads adhere to the surface and do not detach within the experiment duration of 30 to 45 min. Similar to the adhesion experiments with plain glass surfaces, the adhesion strength of particles to silanized substrates can be tuned by adjusting the ionic strength of the immersion buffer. This is illustrated in Figure 8, which shows the mechanical contact spectra recorded for carboxylate-modified polystyrene particles adherent to MPTMS-coated glass in PBS buffer with 13.7 mM (a) and 137 mM (b) concentrations of NaCl at the value of the solution pH fixed at 7.6. Here, the low salt concentration of 13.7 mM corresponds to the conditions used in the measurements reported in Figure 7, and the high salt concentration of 137 mM corresponds to the plain glass experiments reported in Figure 5. A 10-fold increase in the solution’s ionic strength causes strong shielding of the residual electrostatic repulsion, therefore leading to a significant increase in the strength of particle adhesion to the MPTMS surface. This is clearly indicated in the mechanical contact spectra by the large shift of the average particle position fluctuations from ξ̅ = 14.8 nm recorded in 13.7 mM buffer to ξ̅ = 1.0 nm recorded in 137 mM buffer. It should be noted that in 137 mM NaCl the strength of carboxylate-modified particle adhesion to the MPTMS-coated surface noticeably exceeds the strength of adhesion of the same particles to a plain-glass surface at a similar ionic strength of the immersion buffer (cf. Figure 5). Sensitivity and Operating Range of Mechanical Contact Spectroscopy. Both the sensitivity and limits of operation of mechanical contact spectroscopy are wellillustrated by adhesion experiments conducted with plain and silanized glass surfaces that cover a wide range of conditions encountered in practical situations involving particle adhesion in liquid environments. The results obtained with 1 μm carboxylate-modified polystyrene particles adherent to plain glass surfaces in aqueous buffers of varying ionic strength demonstrate that mechanical contact spectra are sensitive to small changes in the work of adhesion between the particles and substrates immersed in a liquid. With this simple, wellH

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

ξ̅amine, and ξ̅mercaptan can be estimated from the literature data obtained using chemical force microscopy. This technique measures the pull-off force Fad required to retract a chemically modified AFM cantilever with a known tip radius Rtip from a substrate, the surface of which is also functionalized with a molecule of interest.45,66 Fad can then be converted to the work of adhesion W assuming the JKR model of the particle adhesive contact.14,42 On the basis of the detailed discussion presented in the Supporting Information, we estimated the values of the work of adhesion as Wamine = 19.8 mJ·m−2 for the adhesion of carboxylate-modified polystyrene particles to amine-coated glass substrates and Wmercaptan = 0.9 mJ·m−2 for the adhesion of carboxylate-modified polystyrene particles to mercaptan-coated glass substrates. Hence, mechanical contact spectroscopy can cover the range of the work of adhesion spanning at least an order of magnitude. The measured ξ̅ decreases with the growing work of adhesion; therefore, the current range of measurable W is limited from above by the resolution limit of our particle tracking algorithm which is ∼0.7 nm for 1 μm greenfluorescent particles (Supporting Information). This limitation, however, is not fundamental because the particle tracking resolution can be improved by using a different video imaging strategy, for example, employing a pulsed laser to illuminate the fluorescent particles that provides a better signal-to-noise ratio for the particle position determination and reduces the particle photobleaching. In all studied cases, measured values of ξ̅ lie well above the resolution limit, and thus mechanical contact spectroscopy in its current implementation furnishes reliable information on the particle−substrate adhesion under given experimental conditions. Despite the resolution limit of the particle position tracking, ξ̅ in the adhesion experiments with silanized glass surfaces could be determined with angstrom-level reproducibility (Figure S4). This follows from the fact that the values of ξ̅ represent the average position fluctuations of particle populations including typically tens of particles. Because the standard error of ξ̅ that quantifies the precision of determining ξ̅ is inversely proportional to the square root of the number of particles N included in the spectra, MCS peaks can be located with high precision by analyzing sufficiently large numbers of particles for the given sample. Effect of Surface Heterogeneity and Roughness on the Shape of Mechanical Contact Spectra. In addition to the value of ξ̅, mechanical contact spectra can be characterized by the width of the spectral peak (cf. Figures 7 and 8). This quantity directly reflects the distribution of the particle− substrate work of adhesion that was also observed in the particle detachment experiments30,64 and chemical force microscopy measurements.45,66 Such a distribution of the work of adhesion is a consequence of morphological and chemical heterogeneities (surface asperities and contamination) present on both adhering surfaces.19,20 Theoretical treatment of the adhesion of colloidal particles and substrates with surfaces bearing nanometer-scale roughness,76,77 and direct measurement of the interaction forces between such surfaces40,78 revealed a strong dependence of the work of adhesion on the size and number of asperities protruding from the surfaces. Thus, an ensemble of particles with random surface asperities adherent to a substrate with chemical properties that are not completely homogeneous is expected to yield a spectral peak of a finite width. Another effect that influences the observed particle position fluctuations is the

defined, and controllable model adhesion system, we were able to induce a reproducible shift of the averaged particle position fluctuations ξ̅ toward noticeably smaller values by increasing the ionic strength I of the medium from 100 to 175 mM while keeping the fixed solution pH at 6. (See Figure S3 in the Supporting Information.) The corresponding change in the particle−substrate work of adhesion W can be estimated from linking the value of W to the overall interaction energy ΔGΣ between the two adhering surfaces. Within the framework of extended DLVO theory, ΔGΣ can be expressed as the sum of the electrostatic, retarded van der Waals, Lewis acid−base, and Born repulsion components.15,64 (See the Supporting Information for full expressions of the individual interaction energy components and a list of relevant interaction parameters.) The work of adhesion is then equal to the depth of the primary minimum ΔGΣmin of the overall interaction energy divided by the area of the contact.14,42 Change of the solution’s ionic strength solely affects the electrostatic component ΔGEL of the total interaction energy ΔGΣ. In particular, an increase in I leads to a decrease in the Debye length κ−1 and a decrease in the absolute values of the electrostatic surface potentials ψP and ψS of the particle and the substrate due to the presence of counterions.14,75 This in turn leads to a weaker electrostatic repulsion between like-charged particle and substrate and, consequently, stronger particle adhesion. Detailed calculations presented in the Supporting Information indicate that the work of adhesion increases from W100 mM = 4.7 mJ·m−2 at I = 100 mM to W175 mM = 5.2 mJ·m−2 at I = 175 mM. This corresponds to the relative change in the work of adhesion δW = (W175 mM − W100 mM)/W100 mM ≈ 0.11. Thus, the acquired mechanical contact spectra reflect a change in the work of adhesion on the order of 10%. In the experiments with carboxylate-modified particles adherent to plain glass surfaces, this 10% change in the work of adhesion is accompanied by a shift in the average particle position fluctuations from ξ̅ = 4.7 nm recorded in 100 mM NaCl to ξ̅ = 3.3 nm recorded in 175 mM NaCl. In principle, with a sufficient number of adherent particles included in the mechanical contact spectrum, we are able to reliably detect even smaller shifts dξ̅ in the value of ξ̅. Assuming a general power-law relationship between W and ξ̅ of the form W = Aξ̅n, the relative change in W is linked to the relative change in ξ̅ as dW/W = n dξ̅/ξ̅. Thus, for a fixed dξ̅, the relative resolution dW/W of determining the value of W improves with increasing ξ̅ (i.e., for a particle/substrate combination that displays weaker adhesion). On the basis of the above analysis of the adhesion of carboxylate-modified polystyrene particles to plain glass surfaces, the estimated typical resolution of mechanical contact spectroscopy with respect to detecting the changes in the work of adhesion is in the range of a few percent. Adhesion experiments with silanized glass surfaces provide evidence that, apart from being highly sensitive, mechanical contact spectroscopy can also cover a relevant range of the particle−substrate work of adhesion in aqueous solutions. In the experiments carried out with negatively charged carboxylate-modified particles under low-ionic-strength conditions (13.7 mM NaCl), we observed reproducible values of ξ̅ that spanned an order of magnitude. In particular, for the case of strong adhesion to positively charged amine-coated substrates, we recorded ξ̅amine = 1.1 nm, whereas for the case of weak adhesion to neutral mercaptan-coated substrates, we recorded ξ̅mercaptan = 14.8 nm (Figures 7 and S4). The values of the work of adhesion Wamine, Wmercaptan corresponding to the observed I

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir finite time required for forming contact to reach equilibrium; this transition phenomenon was observed with polystyrene particles manipulated by optical tweezers.52 As the particles included in the mechanical contact spectra form their contacts with the substrate at random times, they are generally in different phases of the contact formation process at the time of observation and thus can display different position fluctuations. For weakly adherent particles, thermal forces alone can even be sufficient to induce the detachment of the particle from the substrate.79 However, even though the surface heterogeneities and finite time window of the adhesive contact formation cause the broadening of spectral peaks, the experimental results are well-reproduced as long as the experiments are performed under strictly controlled, identical environmental conditions and the spectra contain information from a sufficient number of adherent particles. A general trend observed in the mechanical contact spectra is the narrowing of the spectral peak with increasing strength of adhesion (cf. Figures 7 and 8). This phenomenon can be explained by taking the compliance of asperities on the particle surface into account. With higher adhesive forces, the asperities protruding from the particle surface deform as the adhering bodies are pressed into contact by the attractive interaction potential. Hence, the particle surface resumes a more regular shape, and the contact area becomes better defined. Moreover, stronger adhesion shortens the time scale of the contact relaxation,52 and thus a higher fraction of adherent particles reach a fully equilibrated state by the time of data acquisition. As a result, the differences between individual particles are diminished and, consequently, the spectral distribution becomes narrower.

immersion media that can be exchanged reversibly in situ and the monitoring of the dynamics of the particle response to changing environmental conditions. Mechanical contact spectra of adhesive contacts are generated from high-resolution particle position tracks. Because such tracks can be obtained with subnanometer precision even for particles with sizes of a few hundreds of nanometers,58 MCS in principle allows the efficient investigation of nanoparticle adhesion in liquids. In addition, MCS is not limited to the study of transparent adhesion substrates. A slight modification of the optical layout of the experimental apparatus would permit the investigation of opaque substrates (e.g., silicon wafers, metal sheets, or ceramic materials) with a wide range of potential technological applications. Besides fluorescence, other optical contrasting techniques for particle observation could be adopted, such as differential interference contrast (DIC), phase contrast, or dark field. This would further widen the range of particle−substrate combinations and experimental conditions accessible to MCS. In an independent series of experiments, we have studied the selective adhesion of fluorescent particles with diameters ranging between 40 nm and 1 μm to patterned silicon wafers. In addition, we have also carried out pilot DIC-based MCS experiments with 1-μm-diameter particles adherent to glass substrates and verified that subnanometer resolution can be achieved in DIC microscopy with nanoparticles of the diameters as small as 200 nm (unpublished results). These studies indicate the feasibility of expanding the use of MCS into such experimental scenarios. Experimental data presented in this article describe changes in the adhesion strength due to the modification of nonspecific components of the total interaction potential (mostly electrostatic and acid−base/hydrogen bonding) that can be easily achieved by adjusting the substrate and immersion buffer properties (surface charge and hydrophobicity, ionic strength of the solution). MCS, however, is not restricted to characterizing nonspecific particle adhesion: it is applicable to monitoring the changes in the strength of any type of adhesive interaction for particle sizes ranging from nanometers to micrometers and various types of substrates. Thus, the technique can be readily used for the screening of surface contamination, for the assessment of the reproducibility of surface preparation protocols, for the quantification of the efficiency of preventing unwanted particle adhesion in singlemolecule experiments, and so forth. In this context, the study of specific ligand−receptor-type adhesion mediated, for example, by the interaction between antigen and antibody is of particular interest because it triggers a number of physiologically relevant processes (e.g., targeted intracellular transport, cell immune response, tissue formation, etc.). Further information contained in the MCS data can be extracted by creating a theoretical model of thermally excited adhesive contacts that would provide a quantitative link between the observed position fluctuations of adherent particles and the work of adhesion. Such a model, which could be verified against finite element method simulations, is currently under development.

■

SUMMARY AND CONCLUSIONS In this article, we have introduced mechanical contact spectroscopy (MCS), a novel experimental technique for characterizing adhesive contacts of immersed micrometer-sized particles to solid substrates under a wide range of environmental conditions. In MCS, the strength of adhesion is quantified via the analysis of residual thermal position fluctuations of adherent particles. The standard deviation, ξ, of the particle position track is correlated with the work of adhesion, W, with smaller ξ indicating higher W. Adhesive properties of a given combination of particles, substrates, and immersion media are summarized in mechanical contact spectra that are readily obtained from the parallel tracking of an ensemble of adherent particles. In MCS, the adhesion is probed solely by the thermal forces imparted by the immersion medium surrounding the particle, and no additional external force has to be applied. Hence, the experimental arrangement for performing MCS experiments is very simple and can be easily implemented using a sufficiently stable fluorescence microscope. Because no direct mechanical access to the studied particles is required, it is possible to investigate particle adhesion in environments with complex topography (e.g., porous materials or irregularly shaped substrates) or even inside a closed boundary, provided that at least a part of it is transparent for the light used for the particle position tracking. The minute magnitude of thermal forces relative to the forces of adhesion assures that the studied adhesive contacts remain in equilibrium at all times; virtually no bonds between the particles and the substrate are ruptured. The nondestructive character of MCS allows, in principle, the study of the same population of adherent particles in different

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.8b04074. J

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

■

(14) Israelachvili, J. N. Intermolecular and Surface Forces, 3rd ed.; Elsevier, Academic Press: Amsterdam, 2011. (15) Van Oss, C. J. Interfacial Forces in Aqueous Media, 2nd ed.; Taylor & Francis: Boca Raton, FL, 2006. (16) Gent, A. N.; Kaang, S. Pull-off Forces for Adhesive Tapes. J. Appl. Polym. Sci. 1986, 32 (4), 4689−4700. (17) Chalker, P. R.; Bull, S. J.; Rickerby, D. S. A Review of the Methods for the Evaluation of Coating-Substrate Adhesion. Mater. Sci. Eng., A 1991, 140, 583−592. (18) Sickfeld, J. Pull-Off Test, an Internationally Standardized Method for Adhesion Testing-Assessment of the Relevance of Test Results. In Adhesion Aspects of Polymeric Coatings.; Mittal, K. L., Ed.; Springer: Boston, 1983; pp 543−567.. (19) Walz, J. Y. The Effect of Surface Heterogeneities on Colloidal Forces. Adv. Colloid Interface Sci. 1998, 74 (1−3), 119−168. (20) Duffadar, R.; Kalasin, S.; Davis, J. M.; Santore, M. M. The Impact of Nanoscale Chemical Features on Micron-Scale Adhesion: Crossover from Heterogeneity-Dominated to Mean-Field Behavior. J. Colloid Interface Sci. 2009, 337 (2), 396−407. (21) Rimai, D.; Quesnel, D. J. Fundamentals of Particle Adhesion; Polymer Surfaces and Interfaces Series; Global Press: Moorhead, MN, 2001. (22) Moore, N. W.; Houston, J. E. The Pull-Off Force and the Work of Adhesion: New Challenges at the Nanoscale. J. Adhes. Sci. Technol. 2010, 24 (15−16), 2531−2544. (23) Jacobs, T. D. B.; Martini, A. Measuring and Understanding Contact Area at the Nanoscale: A Review. Appl. Mech. Rev. 2017, 69 (6), 060802. (24) Butt, H.-J.; Cappella, B.; Kappl, M. Force Measurements with the Atomic Force Microscope: Technique, Interpretation and Applications. Surf. Sci. Rep. 2005, 59 (1−6), 1−152. (25) Leite, F. L.; Bueno, C. C.; Da Róz, A. L.; Ziemath, E. C.; Oliveira, O. N. Theoretical Models for Surface Forces and Adhesion and Their Measurement Using Atomic Force Microscopy. Int. J. Mol. Sci. 2012, 13 (12), 12773−12856. (26) Erath, J.; Schmidt, S.; Fery, A. Characterization of Adhesion Phenomena and Contact of Surfaces by Soft Colloidal Probe AFM. Soft Matter 2010, 6 (7), 1432. (27) Kappl, M.; Butt, H.-J. The Colloidal Probe Technique and Its Application to Adhesion Force Measurements. Part. Part. Syst. Charact. 2002, 19 (3), 129. (28) Kulvanich, P.; Stewart, P. Fundamental Considerations in the Measurement of Adhesional Forces between Particles Using the Centrifuge Method. Int. J. Pharm. 1987, 35 (1−2), 111−120. (29) Busnaina, A.; Taylor, J.; Kashkoush, I. Measurement of the Adhesion and Removal Forces of Submicrometer Particles on Silicon Substrates. J. Adhes. Sci. Technol. 1993, 7 (5), 441−455. (30) Bergendahl, J.; Grasso, D. Prediction of Colloid Detachment in a Model Porous Media: Hydrodynamics. Chem. Eng. Sci. 2000, 55 (9), 1523−1532. (31) Ghatak, A.; Vorvolakos, K.; She, H.; Malotky, D. L.; Chaudhury, M. K. Interfacial Rate Processes in Adhesion and Friction. J. Phys. Chem. B 2000, 104 (17), 4018−4030. (32) Attard, P. Interaction and Deformation of Elastic Bodies: Origin of Adhesion Hysteresis. J. Phys. Chem. B 2000, 104 (45), 10635−10641. (33) She, H.; Malotky, D.; Chaudhury, M. K. Estimation of Adhesion Hysteresis at Polymer/Oxide Interfaces Using Rolling Contact Mechanics. Langmuir 1998, 14 (11), 3090−3100. (34) Krijt, S.; Dominik, C.; Tielens, A. G. G. M. Rolling Friction of Adhesive Microspheres. J. Phys. D: Appl. Phys. 2014, 47 (17), 175302. (35) Moy, V. T.; Jiao, Y.; Hillmann, T.; Lehmann, H.; Sano, T. Adhesion Energy of Receptor-Mediated Interaction Measured by Elastic Deformation. Biophys. J. 1999, 76 (3), 1632−1638. (36) Rimai, D. S.; DeMejo, L. P.; Vreeland, W. B.; Bowen, R. C. Adhesion Induced Deformations of a Highly Compliant Elastomeric Substrate in Contact with Rigid Particles. Langmuir 1994, 10 (11), 4361−4366.

Description of materials and sample preparation procedures; discussion of resolution and precision of particle position tracking; discussion of the reproducibility of mechanical contact spectra; and development of the theoretical framework for the quantitative characterization of adhesive interactions between the particles and the substrate (PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (A.J.). *E-mail: fl[email protected] (E.-L.F.). ORCID

Alexandr Jonás:̌ 0000-0002-3555-6901 Present Addresses ⊥

Luminex Corporation, 12212 Technology Boulevard, Suite 130, Austin, Texas 78727, United States. ∥ Institute for Computational Biomedicine and the Department of Physiology and Biophysics, Weill Cornell Medical College, New York, New York 10065, United States. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS We thank Samo Fišinger for assistance with the initial stages of the project. REFERENCES

(1) Nel, A. E.; Mädler, L.; Velegol, D.; Xia, T.; Hoek, E. M. V.; Somasundaran, P.; Klaessig, F.; Castranova, V.; Thompson, M. Understanding Biophysicochemical Interactions at the Nano-Bio Interface. Nat. Mater. 2009, 8 (7), 543−557. (2) Mani, M.; Gopinath, A.; Mahadevan, L. How Things Get Stuck: Kinetics, Elastohydrodynamics, and Soft Adhesion. Phys. Rev. Lett. 2012, 108 (22), 226104. (3) Visser, J. Particle Adhesion And Removal: A Review. Part. Sci. Technol. 1995, 13 (3−4), 169−196. (4) Mazumder, M. K.; Wankum, D. L.; Sims, R. A.; Mountain, J. R.; Chen, H.; Pettit, P.; Chaser, T. Influence of Powder Properties on the Performance of Electrostatic Coating Process. J. Electrost. 1997, 40− 41, 369−374. (5) Steward, P. A.; Hearn, J.; Wilkinson, M. C. An Overview of Polymer Latex Film Formation and Properties. Adv. Colloid Interface Sci. 2000, 86 (3), 195−267. (6) Bowling, R. A. An Analysis of Particle Adhesion on Semiconductor Surfaces. J. Electrochem. Soc. 1985, 132 (9), 2208. (7) Particle Control for Semiconductor Manufacturing; Donovan, R. P., Ed.; Marcel Dekker: New York, 1990. (8) Dejeu, J.; Bechelany, M.; Rougeot, P.; Philippe, L.; Gauthier, M. Adhesion Control for Micro- and Nanomanipulation. ACS Nano 2011, 5 (6), 4648−4657. (9) Maboudian, R.; Howe, R. T. Critical Review: Adhesion in Surface Micromechanical Structures. J. Vac. Sci. Technol., B: Microelectron. Process. Phenom. 1997, 15 (1), 1−20. (10) Perni, S.; Preedy, E. C.; Prokopovich, P. Success and Failure of Colloidal Approaches in Adhesion of Microorganisms to Surfaces. Adv. Colloid Interface Sci. 2014, 206, 265−274. (11) Tareste, D.; Shen, J.; Melia, T. J.; Rothman, J. E. SNAREpin/ Munc18 Promotes Adhesion and Fusion of Large Vesicles to Giant Membranes. Proc. Natl. Acad. Sci. U. S. A. 2008, 105 (7), 2380−2385. (12) Balda, M. Epithelial Cell Adhesion and the Regulation of Gene Expression. Trends Cell Biol. 2003, 13 (6), 310−318. (13) Yamada, S.; Nelson, W. J. Synapses: Sites of Cell Recognition, Adhesion, and Functional Specification. Annu. Rev. Biochem. 2007, 76 (1), 267−294. K

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Measurements on an Attached Tracer Particle. In Methods in Cell Biology.; Elsevier, 2010; Vol. 95, pp 601−615.. (62) Happel, J.; Brenner, H. Low Reynolds Number Hydrodynamics: With Special Applications to Particulate Media. In Mechanics of Fluids and Transport Processes; Nijhoff, M., Ed.; Kluwer: Boston, 1983. (63) Behrens, S. H.; Grier, D. G. The Charge of Glass and Silica Surfaces. J. Chem. Phys. 2001, 115 (14), 6716−6721. (64) Bergendahl, J.; Grasso, D. Prediction of Colloid Detachment in a Model Porous Media: Thermodynamics. AIChE J. 1999, 45 (3), 475−484. (65) Good, N. E.; Winget, G. D.; Winter, W.; Connolly, T. N.; Izawa, S.; Singh, R. M. M. Hydrogen Ion Buffers for Biological Research. Biochemistry 1966, 5 (2), 467−477. (66) Noy, A.; Frisbie, C. D.; Rozsnyai, L. F.; Wrighton, M. S.; Lieber, C. M. Chemical Force Microscopy: Exploiting ChemicallyModified Tips To Quantify Adhesion, Friction, and Functional Group Distributions in Molecular Assemblies. J. Am. Chem. Soc. 1995, 117 (30), 7943−7951. (67) Plueddemann, E. P. Silane Coupling Agents; Springer: Boston, 1991.. (68) Giesbers, M.; Kleijn, J. M.; Cohen Stuart, M. A. Interactions between Acid- and Base-Functionalized Surfaces. J. Colloid Interface Sci. 2002, 252 (1), 138−148. (69) Metwalli, E.; Haines, D.; Becker, O.; Conzone, S.; Pantano, C. G. Surface Characterizations of Mono-, Di-, and Tri-Aminosilane Treated Glass Substrates. J. Colloid Interface Sci. 2006, 298 (2), 825− 831. (70) Kurth, D. G.; Bein, T. Surface Reactions on Thin Layers of Silane Coupling Agents. Langmuir 1993, 9 (11), 2965−2973. (71) Fang, Y.; Hoh, J. H. Cationic Silanes Stabilize Intermediates in DNA Condensation. FEBS Lett. 1999, 459 (2), 173−176. (72) Chrisey, L. Covalent Attachment of Synthetic DNA to SelfAssembled Monolayer Films. Nucleic Acids Res. 1996, 24 (15), 3031− 3039. (73) Cras, J.; Rowe-Taitt, C.; Nivens, D.; Ligler, F. Comparison of Chemical Cleaning Methods of Glass in Preparation for Silanization. Biosens. Bioelectron. 1999, 14 (8−9), 683−688. (74) Gotoh, K.; Tao, J.; Tagawa, M. Adhesion Interaction in Water/ n-Alcohol Mixtures between Silanized Silica and Polymer Particles. J. Adhes. Sci. Technol. 1999, 13 (11), 1307−1320. (75) Particle Deposition and Aggregation: Measurement, Modelling and Simulation; Elimelech, M., Ed.; Colloid and Surface Engineering Series; Butterworth-Heinemann: Oxford, 1998. (76) Suresh, L.; Walz, J. Y. Effect of Surface Roughness on the Interaction Energy between a Colloidal Sphere and a Flat Plate. J. Colloid Interface Sci. 1996, 183 (1), 199−213. (77) Bhattacharjee, S.; Ko, C.-H.; Elimelech, M. DLVO Interaction between Rough Surfaces. Langmuir 1998, 14 (12), 3365−3375. (78) Suresh, L.; Walz, J. Y. Direct Measurement of the Effect of Surface Roughness on the Colloidal Forces between a Particle and Flat Plate. J. Colloid Interface Sci. 1997, 196 (2), 177−190. (79) Wang, Z.; Jin, Y.; Shen, C.; Li, T.; Huang, Y.; Li, B. Spontaneous Detachment of Colloids from Primary Energy Minima by Brownian Diffusion. PLoS One 2016, 11 (1), No. e0147368.

(37) Murthy Peri, M. D.; Cetinkaya, C. Non-Contact MicrosphereSurface Adhesion Measurement via Acoustic Base Excitations. J. Colloid Interface Sci. 2005, 288 (2), 432−443. (38) Saeedi Vahdat, A.; Azizi, S.; Cetinkaya, C. Nonlinear Dynamics of Adhesive Micro-Spherical Particles on Vibrating Substrates. J. Adhes. Sci. Technol. 2013, 27 (15), 1712−1726. (39) Liebert, R. B.; Prieve, D. C. Species-Specific Long Range Interactions between Receptor/Ligand Pairs. Biophys. J. 1995, 69 (1), 66−73. (40) Bevan, M. A.; Prieve, D. C. Direct Measurement of Retarded van Der Waals Attraction. Langmuir 1999, 15 (23), 7925−7936. (41) Shull, K. R. Contact Mechanics and the Adhesion of Soft Solids. Mater. Sci. Eng., R 2002, 36 (1), 1−45. (42) Johnson, K. L.; Kendall, K.; Roberts, A. D. Surface Energy and the Contact of Elastic Solids. Proc. R. Soc. London, Ser. A 1971, 324 (1558), 301−313. (43) Derjaguin, B. V.; Muller, V. M.; Toporov, Y. P. Effect of Contact Deformations on the Adhesion of Particles. J. Colloid Interface Sci. 1975, 53 (2), 314−326. (44) Maugis, D. Adhesion of Spheres: The JKR-DMT Transition Using a Dugdale Model. J. Colloid Interface Sci. 1992, 150 (1), 243− 269. (45) Clear, S. C.; Nealey, P. F. Chemical Force Microscopy Study of Adhesion and Friction between Surfaces Functionalized with SelfAssembled Monolayers and Immersed in Solvents. J. Colloid Interface Sci. 1999, 213 (1), 238−250. (46) Clint, J. H.; Wicks, A. C. Adhesion under Water: Surface Energy Considerations. Int. J. Adhes. Adhes. 2001, 21 (4), 267−273. (47) Reif, F. Fundamentals of Statistical and Thermal Physics.; Waveland Press: Long Grove, IL, 2009. (48) Florin, E.-L.; Pralle, A.; Stelzer, E. H. K.; Hörber, J. K. H. Photonic Force Microscope Calibration by Thermal Noise Analysis. Appl. Phys. A: Mater. Sci. Process. 1998, 66 (7), S75−S78. (49) Dominik, C.; Tielens, A. G. G. M. Resistance to Sliding on Atomic Scales in the Adhesive Contact of Two Elastic Spheres. Philos. Mag. A 1996, 73 (5), 1279−1302. (50) Ding, W.; Howard, A. J.; Peri, M. D. M.; Cetinkaya, C. Rolling Resistance Moment of Microspheres on Surfaces: Contact Measurements. Philos. Mag. 2007, 87 (36), 5685−5696. (51) Sümer, B.; Sitti, M. Rolling and Spinning Friction Characterization of Fine Particles Using Lateral Force Microscopy Based Contact Pushing. J. Adhes. Sci. Technol. 2008, 22 (5−6), 481−506. (52) Sharma, P.; Ghosh, S.; Bhattacharya, S. Microrheology of a Sticking Transition. Nat. Phys. 2008, 4 (12), 960−966. (53) Goohpattader, P. S.; Mettu, S.; Chaudhury, M. K. Stochastic Rolling of a Rigid Sphere in Weak Adhesive Contact with a Soft Substrate. Eur. Phys. J. E: Soft Matter Biol. Phys. 2011, 34 (11), 120. (54) Chaudhury, M. K.; Goohpattader, P. S. Noise-Activated Dissociation of Soft Elastic Contacts. Eur. Phys. J. E: Soft Matter Biol. Phys. 2012, 35 (12), 131. (55) Gennes, P.-G. de. Brownian Motion with Dry Friction. J. Stat. Phys. 2005, 119 (5−6), 953−962. (56) Crocker, J. C.; Grier, D. G. Methods of Digital Video Microscopy for Colloidal Studies. J. Colloid Interface Sci. 1996, 179 (1), 298−310. (57) Thompson, R. E.; Larson, D. R.; Webb, W. W. Precise Nanometer Localization Analysis for Individual Fluorescent Probes. Biophys. J. 2002, 82 (5), 2775−2783. (58) Speidel, M.; Jonás,̌ A.; Florin, E.-L. Three-Dimensional Tracking of Fluorescent Nanoparticles with Subnanometer Precision by Use of off-Focus Imaging. Opt. Lett. 2003, 28 (2), 69. (59) Dufresne, E. R.; Squires, T. M.; Brenner, M. P.; Grier, D. G. Hydrodynamic Coupling of Two Brownian Spheres to a Planar Surface. Phys. Rev. Lett. 2000, 85 (15), 3317−3320. (60) Wong, W. P.; Halvorsen, K. The Effect of Integration Time on Fluctuation Measurements: Calibrating an Optical Trap in the Presence of Motion Blur. Opt. Express 2006, 14 (25), 12517. (61) Taute, K. M.; Pampaloni, F.; Florin, E.-L. Extracting the Mechanical Properties of Microtubules from Thermal Fluctuation L

DOI: 10.1021/acs.langmuir.8b04074 Langmuir XXXX, XXX, XXX−XXX