betaFeedback

WorksPhysics and Astronomy

Control of Static Friction by Designing Grooves on Friction Surface

Abstract

This study numerically investigated the friction of viscoelastic objects with grooves. A 3D viscoelastic block with grooves on a rigid substrate is slowly pushed from the lateral side under uniform pressure on the top surface. The local friction force at the interface between the block and the substrate obeys Amontons' law. Numerical results obtained using the finite element method reveal that the static friction coefficient decreases with increasing groove width and depth. The propagation of the precursor slip is observed before bulk sliding. Furthermore, bulk sliding occurs when the area of slow precursor slip reaches a critical value, which decreases with increasing groove size. A theoretical analysis based on a simplified model reveals that the static friction coefficient is related to the critical area of the precursor, which is determined by the instability of the precursor. A scaling law for the critical area is theoretically predicted, and it indicates that the decrease in the effective viscosity due to the formation of the grooves leads to a decrease in the static friction coefficient. The validity of the theoretical prediction is numerically confirmed.

Graphical Abstract

Graphical abstract showing a schematic simulation system, a cross-section of a grooved friction surface, friction force versus displacement, and static friction coefficient versus groove depth

Figure 13.

Introduction

Friction forces occur in different situations, such as sliding parts of machines and contact surfaces between tires and the ground, and prevent the relative motion between two objects in contact. Friction forces are desirable in applications requiring low slippage, whereas they are undesirable in sliding parts of machines due to energy loss. Therefore, the control of friction forces is important in engineering [1, 2, 3, 4, 5, 6, 7]. One of the known methods to control friction forces is designing friction surfaces by forming grooves. Generally, grooves on the surfaces of tires and sliding parts of machines are formed to reduce undesirable lubrication in wet conditions, which leads to a decrease in the friction coefficient and results in accidental slippage [8, 9, 10, 11]. However, the dependence of friction force on grooves in dry conditions has not been established clearly.

Generally, for friction between solids in dry conditions, Amontons’ law is expected to hold [1, 2, 3, 4, 5, 6, 7]. According to Amontons’ law, the friction coefficient does not depend on the external pressure or size and shape of the object. However, the phenomenological explanation of Amontons’ law is based on the adhesion of microscopic asperities at the friction interface [1, 2, 3, 4, 5, 6, 7, 12, 13], and implicitly assumes the uniformity of the stress field. For macroscopic objects associated with the non-uniform stress field, Amontons’ law may not hold. Therefore, the friction coefficient may depend on the shape of the macroscopic objects in dry conditions.

In fact, recent studies on the friction of objects with flat friction surfaces have shown that Amontons’ law is not satisfied when the non-uniformity of the stress field is significant [14, 15, 16, 17, 18]. In Refs. 16, 18, numerical simulations and analysis of a simplified model have revealed the mechanism of breakdown of Amontons’ law in viscoelastic materials. The analysis clarified that the local precursor slip before bulk sliding due to the non-uniform stress field leads to the breakdown of Amontons’ law, and that the static friction coefficient exhibits characteristic load dependence. The relationship between the precursor slip and breakdown of Amontons’ law and the load dependence of the static friction coefficient have been verified in experiments on acrylic glass blocks [17]. Precursor slip relates to earthquake [19, 20, 21, 22, 23, 24] and fracture [25, 26, 27, 28, 29, 30, 31], and has been extensively studied in experiments [15, 17, 19, 20, 24, 25, 26, 27, 28, 29, 32, 33] and numerical simulations [18, 23, 30, 31, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45]. However, many of these studies have considered only flat friction surfaces, and the effects of grooves in the friction surface on frictional properties and precursor slip are yet to be discovered.

Recently, several studies have been conducted to reveal the dependence of the friction coefficient on the macroscopic shape of the friction surface. In Refs. 46–51, shapes of friction surfaces were represented by a spatial dependence of the local friction coefficient in 1D or 2D spring-block models. The frictional properties of the models vary with the spatial pattern of the local friction coefficient [46, 47, 48, 49, 50, 51], and these results have been applied to experiments of macroscopic objects [52, 53]. However, it is unclear to what extent the results of the spring-block models with a spatial pattern of local friction coefficient reflect the effect of the actual surface shape. Experiments with rubber and gel blocks have also revealed the dependence of the friction coefficient on the macroscopic shape of the friction surface [54, 55]. However, it is unclear whether the results for relatively soft objects such as rubber and gel can be applied to harder materials where Amontons’ law is locally satisfied.

In this study, using the finite element method (FEM), we numerically investigate the friction of a 3D viscoelastic material with grooves in a dry condition, where the friction force locally obeys Amontons’ law. The dependence of the static friction coefficient on the groove shape is investigated. We find that the static friction coefficient is a decreasing function of the groove width and depth. We also observe that local precursor slip occurs before bulk sliding of the viscoelastic material. The bulk sliding occurs when the area of the precursor slip reaches a critical value. The static friction coefficient is scaled by the normalized critical area of the precursor slip. The propagation of the precursor slip is analytically studied based on a simplified model. We derive the conditions for the onset of bulk sliding and the dependence of the static friction coefficient on the groove shape. The results show that the static friction coefficient decreases due to the decrease in effective viscosity as the groove width and depth increase. Both pillars in the friction surface and main body supporting them play an important role. Our results aid in the improvement of sliding interface design by making grooves for both wet and dry conditions.

Model and Methods

We consider grooved viscoelastic blocks on a rigid substrate under a uniform external pressure PextP_{\mathrm{ext}} with width WW, length LL, and height HH along the xx, yy, and zz axes, respectively, as shown in Figure 1 (PDF p. 4). The rigid substrate is at z=0z = 0. A rigid plate with a width of WW and a height of 0.5H0.5H pushes the side of the block at y=0y = 0 and 0.5H≤z≤H0.5H \le z \le H with a slow constant velocity VV along the yy direction. This study considers a longitudinal groove parallel to the yy direction, as shown in Figure 1 (PDF p. 4). The number of pillars in the friction surface of the block is denoted by nxn_x. The pillars are equally spaced with width lgl_g. The height and width of the pillar are denoted by dd and W/nx−lgW/n_x - l_g, respectively. The cross-section perpendicular to the yy direction is symmetrical, as shown in Figure 1 (PDF p. 4)b. The pillar is in contact with the rigid substrate, as shown in Figure 1 (PDF p. 4)c. The ratio ϕ\phi of the area of the non-contact surface to the area of the bottom surface is given by ϕ=lgnx/W\phi= l_g n_x/W, and the contact area of the friction surface is given by LW(1−ϕ)LW(1 - \phi). Here, ϕ=0\phi= 0 corresponds to a rectangular block without grooves.

Figure 1 (PDF p. 4). Schematic of the system. (a) Grooved viscoelastic block moving on a rigid substrate. (b) Cross-section perpendicular to the yy direction indicated as I in (a). (c) The bottom of the block indicated as II in (a). The blue region represents the contact area between the rigid substrate and the block.

The equation of motion for the viscoelastic object is given by

ρu¨i=∑j∂jσij(1)\rho\ddot{u}_i = \sum_j \partial_j \sigma_{ij} \tag*{(1)}

with density ρ\rho, displacement vector u\mathbf{u}, and stress tensor σ\boldsymbol{\sigma}; where σij\sigma_{ij} is the ijij component of σ\boldsymbol{\sigma}, uiu_i is the ii component of u\mathbf{u}, and u¨i\ddot{u}_i is its second-order time derivative. We adopt the Kelvin-Voigt model for σ\boldsymbol{\sigma}, where σ\boldsymbol{\sigma} is given by σ=σ(E)+σ(V)\boldsymbol{\sigma} = \boldsymbol{\sigma}^{(\mathrm{E})} + \boldsymbol{\sigma}^{(\mathrm{V})} with the elastic stress σ(E)\boldsymbol{\sigma}^{(\mathrm{E})} obeying Hooke’s law and the viscous stress σ(V)\boldsymbol{\sigma}^{(\mathrm{V})} proportional to strain rate, which reduces the elastic waves caused by the deformation of the block. We assume that the viscoelastic material of the block is isotropic. The ijij component

of the elastic stress tensor σij(E)\sigma_{ij}^{(\mathrm{E})} is given by

σij(E)=E1+νϵij+νE(1+ν)(1−2ν)∑kϵkkδij(2)\sigma_{ij}^{(\mathrm{E})} = \frac{E}{1+\nu}\epsilon_{ij} + \frac{\nu E}{(1+\nu)(1-2\nu)} \sum_k \epsilon_{kk}\delta_{ij} \tag*{(2)}

with Young’s modulus EE, Poisson’s ratio ν\nu, Kronecker’s delta δij\delta_{ij}, and strain tensor ϵij\epsilon_{ij}. The ijij component of the viscous stress tensor σij(V)\sigma_{ij}^{(\mathrm{V})} is given by

σij(V)=η1ϵ˙ij+η2∑kϵ˙kkδij(3)\sigma_{ij}^{(\mathrm{V})} = \eta_1\dot{\epsilon}_{ij} + \eta_2 \sum_k \dot{\epsilon}_{kk}\delta_{ij} \tag*{(3)}

with the two viscosity coefficients η1\eta_1 and η2\eta_2 and the strain rate tensor ϵ˙ij\dot{\epsilon}_{ij} [56].

The boundary conditions on the top surface of the block at z=Hz = H are given by σzz=−Pext\sigma_{zz} = -P_{\mathrm{ext}} and σxz=σyz=0\sigma_{xz} = \sigma_{yz} = 0. At surfaces except the top and bottom of the block, free boundary conditions (∑jσijnj=0\sum_j \sigma_{ij}n_j = 0) are applied, where njn_j is the jj component of the normal vector n\boldsymbol{n} to the surface. The boundary conditions at the contact surface with the rigid plate at y=0y = 0 are given by σxy=σzy=0\sigma_{xy} = \sigma_{zy} = 0 and u˙y=V\dot{u}_y = V, where u˙y\dot{u}_y is the velocity along the yy direction. We set VV sufficiently small to push the block quasi-statically.

The friction between the block bottom and substrate obeys Amontons’ law locally. Since the substrate is rigid, the zz-direction displacement uzu_z satisfies uz≥0u_z \ge0. At the bottom, the tangential stress vector t(x,y)=(σxz,σyz)\boldsymbol{t}(x,y) = (\sigma_{xz},\sigma_{yz}) at the position (x,y)(x,y) is given by

t=−vvσ(fric),(4)\boldsymbol{t} = -\frac{\boldsymbol{v}}{v}\sigma^{(\mathrm{fric})}, \tag*{(4)}
σ(fric)(x,y)=μ(v(x,y))p(x,y),(5)\sigma^{(\mathrm{fric})}(x,y) = \mu(v(x,y))p(x,y), \tag*{(5)}

where σ(fric)\sigma^{(\mathrm{fric})} is the frictional stress, v(x,y)=(u˙x,u˙y)\boldsymbol{v}(x,y) = (\dot{u}_x,\dot{u}_y) is the slip velocity vector with velocities along the ii direction u˙i\dot{u}_i, and v(x,y)=∣v∣v(x,y) = |\boldsymbol{v}| is the slip velocity [57]. The bottom pressure p(x,y)=−σzz(x,y,z=0)p(x,y) = -\sigma_{zz}(x,y,z = 0) is set to satisfy uz≥0u_z \ge0, where p=0p = 0 for uz>0u_z > 0. Here, μ(v)\mu(v) is the local friction coefficient depending on vv. In the static region with v(x,y)=0v(x,y) = 0, μ(v)\mu(v) is lower than μS\mu_S, and set to balance the local internal shear stress with the frictional stress. In the slip region with v(x,y)>0v(x,y) > 0, μ(v)\mu(v) is given by

μ(v)={μS−(μS−μK)v/vc,0<v<vcμK,v≥vc(6)\mu(v) = \begin{cases} \mu_S - (\mu_S-\mu_K)v/v_c, & 0 < v < v_c \\ \mu_K, & v \ge v_c \end{cases} \tag*{(6)}

where μS\mu_S and μK\mu_K are the local static and dynamic friction coefficients, respectively. Here, vcv_c is the characteristic velocity. The local Amontons’ law is expected to hold when a local region considered in the interface contains a sufficiently large number of real contact points, and has a negligibly small spatial variation in internal stress [12, 13, 58]. Note that the rate and state-dependent friction law [7] might be more appropriate to represent the local friction, but it coincides with the velocity-weakening friction law in Eq. (6) for a sufficiently large slip length, which is satisfied in the poly methyl methacrylate (PMMA) experiments [17, 26, 32, 59]. Hence, we have adopted Eq. (6). For blocks without grooves, the analysis using the velocity-weakening friction law [16] has been shown to reproduce the PMMA experimental results [17].

We numerically solve Eq. (1) using FEM. The viscoelastic block is divided into cubes with length Δx\Delta x, comprising six tetrahedrons. The displacements of its nodes are evolved based on Eq. (1), and the displacement and velocity within each element are approximated using linear interpolation. The local friction coefficient μ(v)\mu(v) is approximately given as

μ(v)={μSv/ve,0≤v≤veμS−(μS−μK)v/vc,ve<v<vcμK,v≥vc(7)\mu(v) = \begin{cases} \mu_{\mathrm{S}}v/v_{\mathrm{e}}, & 0 \le v \le v_{\mathrm{e}} \\ \mu_{\mathrm{S}} - (\mu_{\mathrm{S}}-\mu_{\mathrm{K}})v/v_{\mathrm{c}}, & v_{\mathrm{e}} < v < v_{\mathrm{c}} \\ \mu_{\mathrm{K}}, & v \ge v_{\mathrm{c}} \end{cases} \tag*{(7)}

with a sufficiently small velocity scale vev_{\mathrm{e}}. The state with 0≤v≤ve0 \le v \le v_{\mathrm{e}} corresponds to the static region, and the state with v>vev > v_{\mathrm{e}} corresponds to the slip region. We set ve/V=2.5×10−2v_{\mathrm{e}}/V = 2.5 \times10^{-2} to satisfy ve≪V,vcv_{\mathrm{e}} \ll V,v_{\mathrm{c}}, and use Δx/H=1/48\Delta x/H = 1/48, Δtvs/H≈10−6\Delta t v_{\mathrm{s}}/H \approx10^{-6}, and V/vs=2.83×10−5V/v_{\mathrm{s}} = 2.83 \times10^{-5} with vs=E/ρv_{\mathrm{s}} = \sqrt{E/\rho}. Here, vsv_{\mathrm{s}} represents the elastic wave velocity. We set the driving speed VV to satisfy the condition V≪vsV \ll v_{\mathrm{s}}, where the elastic waves are sufficiently dissipated. We have confirmed that the dependence of the results on the driving velocity VV is negligible under the condition of V≪vc≪vsV \ll v_{\mathrm{c}} \ll v_{\mathrm{s}}.

In our simulation, we first apply a uniform pressure PextP_{\mathrm{ext}} to the top surface and relax the system to an equilibrium state. From the time t=0t = 0 after the relaxation, the rigid plate pushes the side of the block with a constant velocity VV, and the calculation continues until a periodic stick-slip is observed.

We set the length and width of the block to L/H=4L/H = 4 and W/H=1W/H = 1, respectively. Qualitatively similar results are obtained for L/H=2L/H = 2, as shown in Appendix § 12. We adopt ν=0.34\nu= 0.34, η1vs/(HE)=2.83\eta_{1}v_{\mathrm{s}}/(HE) = 2.83, η2/η1=1\eta_{2}/\eta_{1} = 1, μS=0.38\mu_{\mathrm{S}} = 0.38, μK=0.1\mu_{\mathrm{K}} = 0.1, and vc/vs=4.81×10−4v_{\mathrm{c}}/v_{\mathrm{s}} = 4.81 \times10^{-4} following previous simulations [16, 18]. We select the number of pillars in the friction surface as nx=3n_x = 3, and confirm that the dependence of the numerical results on nxn_x is small, as shown in Appendix § 13. In this study, we investigate the dependence on the external pressure PextP_{\mathrm{ext}}, fraction of non-contact area ϕ\phi, and groove depth dd.

Results

Numerical Simulation

Figure 2 shows the friction force FTF_{\mathrm{T}} against the displacement of the rigid plate U=VtU = Vt at time tt for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and d/H=0.5d/H = 0.5. Here, FTF_{\mathrm{T}} is given by the force on the rigid plate in the yy direction. In Figure 2, FTF_{\mathrm{T}} is normalized by the normal load FN=PextLWF_{\mathrm{N}} = P_{\mathrm{ext}}LW applied to the top surface of the block. The thin and thick solid lines represent the results for ϕ=0\phi= 0 and ϕ=0.5\phi= 0.5, respectively. For each ϕ\phi, FT/FNF_{\mathrm{T}}/F_{\mathrm{N}} increases approximately linearly with UU, and rapidly decreases after reaching a peak value. When the rapid decrease occurs, the entire system slides, and the block returns to a static state after reaching a minimum value close to the local dynamic friction coefficient μK\mu_{\mathrm{K}}. The increase and decrease in FT/FNF_{\mathrm{T}}/F_{\mathrm{N}} repeat periodically, which corresponds to stick-slip motion. We define the maximum value of FT/FNF_{\mathrm{T}}/F_{\mathrm{N}} in the periodic stick-slip region as the macroscopic static friction coefficient μM\mu_{\mathrm{M}}, which is lower than the local static friction coefficient μS\mu_{\mathrm{S}}. Figure 2 shows that μM\mu_{\mathrm{M}} for the block with grooves is lower than that for the flat block.

Figure 2. Ratio of friction force FTF_{\mathrm{T}} to applied normal force FNF_{\mathrm{N}} against displacement of the rigid plate UU for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and d/H=0.5d/H = 0.5. The thin and thick solid lines represent the results for ϕ=0\phi= 0 and ϕ=0.5\phi= 0.5, respectively. The thin and thick horizontal solid lines represent macroscopic static friction coefficient μM\mu_{\mathrm{M}} for ϕ=0\phi= 0 and ϕ=0.5\phi= 0.5, respectively. The dotted and dashed lines represent μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

In Figure 3, we plot the macroscopic static friction coefficient μM\mu_{\mathrm{M}} against the groove depth dd for different values of ϕ\phi with Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and 0.0060.006. Note that the results for ϕ=0\phi= 0 are independent of dd. For each PextP_{\mathrm{ext}}, μM\mu_{\mathrm{M}} is a decreasing function of dd. As dd approaches 00, μM\mu_{\mathrm{M}} converges to that for ϕ=0\phi= 0. The macroscopic static friction coefficient μM\mu_{\mathrm{M}} is a decreasing function of ϕ\phi. These results indicate that the static friction force decreases as the size of the groove increases. Comparing Figure 3a and b, we find that μM\mu_{\mathrm{M}} is a decreasing function of PextP_{\mathrm{ext}}, which is consistent with the results of previous studies on rectangular blocks without grooves [16, 17, 18].

Figure 3. Macroscopic static friction coefficient μM\mu_{\mathrm{M}} against dd for different values of ϕ\phi with (a) Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and (b) Pext/E=0.006P_{\mathrm{ext}}/E = 0.006. The dotted and dashed lines represent μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

In Figure 4, we present the spatial distributions of the slip velocity vv in the friction surface at z=0z = 0 for the displacements U=U1U = U_1, U2U_2, U3U_3, and U4U_4 shown in Figure 2 with Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, ϕ=0.5\phi= 0.5, and d/H=0.5d/H = 0.5. Here, we select U1/L=4.6×10−3U_1/L = 4.6 \times10^{-3}, U2/L=5×10−3U_2/L = 5 \times10^{-3}, U3/L=5.4×10−3U_3/L = 5.4 \times10^{-3}, and U4/L=5.57×10−3U_4/L = 5.57 \times10^{-3} in the periodic stick–slip region. In Figure 4, the blue area represents the static region, and the yellow-green and yellow areas represent the sliding regions with v≤Vv \le V and v>Vv > V, respectively.

Figure 4. Spatial distributions of the slip velocity vv in the friction surface at z=0z = 0 for U=U1U = U_1, U2U_2, U3U_3, and U4U_4 shown in Figure 2 for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, ϕ=0.5\phi= 0.5, and d/H=0.5d/H = 0.5. The blue area represents the static region. The yellow-green and yellow areas represent the slip regions with v≤Vv \le V and v>Vv > V, respectively. The rigid plate pushes the block at y=0y = 0. The white area represents the groove region.

The quasi-static precursor slip with v≤Vv \le V begins to propagate from the region near the rigid plate at y=0y = 0 for U=U1U = U_1, and the area of precursor slip expands quasi-statically as UU increases to U2U_2 and U3U_3. After U3U_3, the area of precursor slip develops rapidly, and the entire system begins to slide, leading to bulk sliding at U4U_4 (see Supplementary Videos). During bulk sliding, the slip velocity vv exceeds VV. We confirm that the displacement due to these slips is approximately along the yy direction for all parameters.

Figure 5 shows the normalized slip area S~=S/[LW(1−ϕ)]\widetilde{S} = S/[LW(1 - \phi)] against U/LU/L for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and d/H=0.5d/H = 0.5. Here, the precursor slip area SS, defined by the sum of the yellow-green and yellow areas in Figure 4, is normalized by the contact area in friction surface LW(1−ϕ)LW(1 - \phi). When S~=0\widetilde{S} = 0, the entire friction surface is static, while S~=1\widetilde{S} = 1 indicates bulk sliding where the entire friction surface is sliding. The thin and thick solid lines represent the results for ϕ=0\phi= 0 and ϕ=0.5\phi= 0.5, respectively. The normalized precursor slip area S~\widetilde{S} increases gradually with UU for small S~\widetilde{S}, but the oscillation of S~\widetilde{S} appears as S~\widetilde{S} becomes large. The slip associated with the oscillation in S~\widetilde{S} is called bounded rapid precursor (BRP). In BRP, the slip front propagates close to the elastic wave speed, but the slip quickly slows down and stops. According to

Figure 5. Normalized precursor slip area S~\widetilde{S} against U/LU/L for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and d/H=0.5d/H = 0.5. The thin and thick lines represent the results for ϕ=0\phi= 0 and ϕ=0.5\phi= 0.5, respectively. The thin and thick horizontal lines represent the normalized critical area of precursor slip S~c\widetilde{S}_{\mathrm{c}} for ϕ=0\phi= 0 and ϕ=0.5\phi= 0.5, respectively. Crosses represent the peaks of the oscillation of S~\widetilde{S} in the region of the periodic stick-slip motion for ϕ=0\phi= 0.

our analysis in Sect. 3.2, the BRP is caused by oscillatory instability. The oscillation of S~\tilde{S} in Figure 5 and small drops of FT/FNF_{\mathrm{T}}/F_{\mathrm{N}} in Figure 2 before bulk sliding is caused by the sequence of the BRP [16]. Each BRP reduces the stress and S~\tilde{S}, but they both recover quickly due to a slight increase in the driving force. The BRP becomes significant depending on the values of the parameters. When S~\tilde{S} reaches a threshold value S~c\tilde{S}_{\mathrm{c}}, the propagation speed of S~\tilde{S} suddenly increases, and S~\tilde{S} reaches unity, which corresponds to the bulk sliding.

We evaluate the critical slip S~c\tilde{S}_{\mathrm{c}} for bulk sliding in the periodic stick-slip motion as the maximum value of S~\tilde{S} in the sequence of the BRP. For example, we have plotted the peaks of S~\tilde{S} due to BRP in the region of periodic stick-slip motion, U/L>3U/L > 3, as crosses for ϕ=0\phi= 0 in Figure 5. The last peaks before bulk sliding represent S~c\tilde{S}_{\mathrm{c}}, which are shown as horizontal lines in Figure 5. Figure 5 shows that S~c\tilde{S}_{\mathrm{c}} decreases with increasing ϕ\phi.

In Figure 6, we show the normalized critical area of the precursor slip S~c\tilde{S}_{\mathrm{c}} against the groove depth dd for different values of ϕ\phi with Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and 0.0060.006. We find that S~c\tilde{S}_{\mathrm{c}} is a decreasing function of dd. As dd approaches 00, S~c\tilde{S}_{\mathrm{c}} approaches that for ϕ=0\phi= 0. We also find that S~c\tilde{S}_{\mathrm{c}} decreases with increasing ϕ\phi. Comparing Figure 6a and b, we see that S~c\tilde{S}_{\mathrm{c}} is a decreasing function of PextP_{\mathrm{ext}}, which is consistent with the results of previous studies on blocks without grooves [16, 17, 18].

Figure 6. Normalized critical area of precursor slip S~c\tilde{S}_{\mathrm{c}} against dd for different values of ϕ\phi with (a) Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and (b) Pext/E=0.006P_{\mathrm{ext}}/E = 0.006.

The dependence of the macroscopic static friction coefficient μM\mu_{\mathrm{M}} on ϕ\phi and dd shown in Figure 3 is similar to that of S~c\tilde{S}_{\mathrm{c}} shown in Figure 6. This similarity indicates a close relation between μM\mu_{\mathrm{M}} and S~c\tilde{S}_{\mathrm{c}}. In fact, as shown in Figure 7, μM\mu_{\mathrm{M}} is an almost linear function of S~c\tilde{S}_{\mathrm{c}} for different values of ϕ\phi and dd with Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and 0.0060.006. Figure 7 also shows that μM\mu_{\mathrm{M}} lies between μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}. This scaling of μM\mu_{\mathrm{M}} using S~c\tilde{S}_{\mathrm{c}} is consistent with the results of previous studies on blocks without grooves [16, 17, 18].

Figure 7. Macroscopic static friction coefficient μM\mu_{\mathrm{M}} against S~c\widetilde{S}_{\mathrm{c}} for different values of ϕ\phi and dd. The filled and open symbols represent the results for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and Pext/E=0.006P_{\mathrm{ext}}/E = 0.006, respectively. The solid line represents the analytical results given by Eq. (20). The dotted and dashed lines represent μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

Figure 8a shows the spatial distribution of the ratio of the frictional stress σ(fric)\sigma^{(\mathrm{fric})} to the bottom pressure pp in the friction surface at U=U3U = U_{3} for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, ϕ=0.5\phi= 0.5 and d/H=0.5d/H = 0.5. Here, U=U3U = U_{3} indicates the state just before bulk sliding, as shown in Figure 2 and Figure 5. Note that the local static friction in the static state with v=0v = 0 takes any values for 0<σ(fric)/p<μS0 < \sigma^{(\mathrm{fric})}/p < \mu_{\mathrm{S}}. As shown in Supplementary Videos and previous studies [16, 18], the ratio σ(fric)/p\sigma^{(\mathrm{fric})}/p returns to the value near μK\mu_{\mathrm{K}} in the entire area just after the bulk sliding. As the block is pushed, σ(fric)/p\sigma^{(\mathrm{fric})}/p reaches the local static friction coefficient μS\mu_{\mathrm{S}} near the region pushed by the rigid plate. The area with σ(fric)/p≈μS\sigma^{(\mathrm{fric})}/p \approx\mu_{\mathrm{S}}

Figure 8. Spatial distribution of stress in the friction surface at U=U3U = U_3 for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, ϕ=0.5\phi= 0.5, and d/H=0.5d/H = 0.5. (a) Spatial distribution of ratio of frictional stress σ(fric)\sigma^{(\mathrm{fric})} to bottom pressure pp. (b) Spatial distribution of pp. The rigid plate pushes the block at y=0y = 0. The white area in (a) represents the region without contact. The white area in (b) represents the groove area.

gradually increases as UU increases. The region with σ(fric)/p≈μS\sigma^{(\mathrm{fric})}/p \approx\mu_{\mathrm{S}} corresponds to the slip region at U=U3U = U_3 in Figure 4, while σ(fric)/p\sigma^{(\mathrm{fric})}/p remains near μK\mu_{\mathrm{K}} in the static region.

fig:8b shows the spatial distribution of the bottom pressure pp at U=U3U = U_3 for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, ϕ=0.5\phi= 0.5 and d/H=0.5d/H = 0.5. Although a uniform pressure PextP_{\mathrm{ext}} is applied at the top surface, the spatial average of pp becomes Pext/(1−ϕ)P_{\mathrm{ext}}/(1-\phi), because the contact area in the friction surface, LW(1−ϕ)LW(1-\phi), is smaller than the area of the top surface, LWLW, due to the grooves. We confirm that the bottom pressure is p≈Pext/(1−ϕ)p \approx P_{\mathrm{ext}}/(1-\phi) in most areas except for the regions near y=0y = 0 and LL. The spatial distribution of pp is almost independent of the time tt, as shown in Supplementary Videos.

Theoretical Analysis

We theoretically analyze the effect of the longitudinal grooves shown in Sect. 3.1 based on a simplified model [16, 18]. The precursor slip is approximately uniform in the xx direction and propagates toward the yy direction, as shown in Figure 4. Therefore, we neglect displacements in the zz and xx directions and consider only the yy-dependent displacement along the yy direction. Since the bottom pressure pp at z=0z = 0 is approximately uniform, as shown in Figure 8b, we assume p=Pext/(1−ϕ)p = P_{\mathrm{ext}}/(1-\phi). Additionally, since the deformation is significant in the region near the bottom before bulk sliding in our 3D simulations, as shown in Appendix § 14, we focus on the slip and deformation in the region 0≤z/H≤α0 \le z/H \le\alpha with a constant α\alpha, as shown in the red shaded area in Figure 9.

Figure 9. (a) Schematic of the derivation of the simplified model for grooved viscoelastic block. (b) Cross-section perpendicular to the yy direction indicated as I in (a). The red shaded areas represent the region from the bottom to the height z=αHz=\alpha H. The dotted rectangle represents the element with infinitesimal width dy\mathrm{d}y.

We consider the equation of motion for a thin element at yy with small width dy\mathrm{d}y indicated by the dotted rectangle in Figure 9a. The mass of the element is given by ρA(ϕ,d)dy\rho A(\phi,d)\mathrm{d}y, where A(ϕ,d)A(\phi,d) is the cross-sectional area of the red region in Figure 9b excluding the groove. In Figure 9a, the forces acting on the left and right surfaces of that element are given by A(ϕ,d)σyy(y,t)A(\phi,d)\sigma_{yy}(y,t) and A(ϕ,d)σyy(y+dy,t)A(\phi,d)\sigma_{yy}(y+\mathrm{d}y,t), respectively. Here, the normal stress in the yy direction is denoted by σyy\sigma_{yy}. The friction force acting on the bottom is given by μPextWdy\mu P_{\mathrm{ext}}W\mathrm{d}y. The equation of motion for the displacement qy(y,t)q_y(y,t) of the thin element along the yy direction is given by

ρA(ϕ,d)dy q¨y(y,t)=A(ϕ,d)[σyy(y+dy,t)−σyy(y,t)]−μ(q˙y(y,t))PextWdy(8)\rho A(\phi,d)\mathrm{d}y\,\ddot{q}_y(y,t)=A(\phi,d)\left[\sigma_{yy}(y+\mathrm{d}y,t)-\sigma_{yy}(y,t)\right]-\mu(\dot{q}_y(y,t))P_{\mathrm{ext}}W\mathrm{d}y \tag*{(8)}

where q˙y(y,t)\dot{q}_y(y,t) and q¨y(y,t)\ddot{q}_y(y,t) are the first and second-order time derivatives of qy(y,t)q_y(y,t), respectively. We assume a plane stress state, where the normal stress σyy(y,t)\sigma_{yy}(y,t) is given by

σyy(y,t)=E1∂qy(y,t)∂y+ηt∂q˙y(y,t)∂y(9)\sigma_{yy}(y,t)=E_1\frac{\partial q_y(y,t)}{\partial y}+\eta_t\frac{\partial\dot{q}_y(y,t)}{\partial y} \tag*{(9)}

with the elastic modulus E1=E/[(1+ν)(1−ν)]E_1=E/[(1+\nu)(1-\nu)] and viscous modulus ηt=η1(η1+2η2)/(η1+η2)\eta_t=\eta_1(\eta_1+2\eta_2)/(\eta_1+\eta_2).

The cross-sectional area A(ϕ,d)A(\phi,d) is given by

A(ϕ,d)=A0[1−κ(ϕ,d)](10)A(\phi,d)=A_0[1-\kappa(\phi,d)] \tag*{(10)}

with the cross-sectional area A0=αHWA_0=\alpha HW for ϕ=0\phi=0. Here, κ(ϕ,d)\kappa(\phi,d) is the reduction rate of the cross-sectional area by the groove,

κ(ϕ,d)={ϕd/(αH),0≤d≤αHϕ,αH<d≤H(11)\kappa(\phi,d)= \begin{cases} \phi d/(\alpha H), & 0\le d\le\alpha H \\ \phi, & \alpha H<d\le H \end{cases} \tag*{(11)}

Substituting Eqs. (9) and (10) into Eq. (8) and taking the limit of dy→0\mathrm{d}y \to0, we obtain

ρ(1−κ)q¨y(y,t)=(1−κ)[E1∂2qy(y,t)∂y2+ηt∂2q˙y(y,t)∂y2]−μ(q˙y(y,t))PextαH.(12)\rho(1-\kappa)\ddot{q}_y(y,t)=(1-\kappa)\left[E_1\frac{\partial^2q_y(y,t)}{\partial y^2}+\eta_t\frac{\partial^2\dot{q}_y(y,t)}{\partial y^2}\right]-\frac{\mu(\dot{q}_y(y,t))P_{\mathrm{ext}}}{\alpha H}. \tag*{(12)}

The boundary conditions are given by ∂qy(L,t)/∂y=0\partial q_y(L,t)/\partial y=0 for the free boundary at y=Ly=L and qy(0,t)=Uq_y(0,t)=U for the fixed boundary at y=0y=0.

We set t=0t=0 just after the bulk sliding, where the friction coefficient is given by μ=μK\mu=\mu_{\mathrm{K}}. When a precursor slip occurs with the normalized slip area S~\widetilde{S} for U>0U>0, the friction coefficient is given by μ=μS\mu=\mu_{\mathrm{S}} in the region 0≤y/L≤S~0\leq y/L\leq\widetilde{S}, because the slip distances of the precursors are significantly smaller than that in bulk sliding. In the other regions, μ\mu remains μK\mu_{\mathrm{K}} due to the frictional stress drop after the bulk sliding. This is confirmed by direct numerical calculations of Eq. (12) and qualitatively consistent with the results in Sect. 3.1. For sufficiently slow driving with q¨y≈0\ddot{q}_y\approx0 and q˙y≈0\dot{q}_y\approx0, the quasi-static solution of qyq_y in Eq. (12) is analytically derived as described in Appendix § 15. In this quasi-static solution qa(y)q_a(y), S~\widetilde{S} is given as an increasing function of UU.

We conduct a stability analysis based on Eq. (12) following the procedure in the previous studies [16, 18]. Substituting qy(y,t)=qa(y)+δq(y,t)q_y(y,t)=q_a(y)+\delta q(y,t) into Eq. (12) with the perturbation δq(y,t)\delta q(y,t), we obtain the equation for δq(y,t)\delta q(y,t) as

ρ(1−κ)δq¨(y,t)=(1−κ)[E1∂2δq(y,t)∂y2+ηt∂2δq˙(y,t)∂y2]−(μS−μK)PextvcαHδq˙(y,t).(13)\rho(1-\kappa)\delta\ddot{q}(y,t)=(1-\kappa)\left[E_1\frac{\partial^2\delta q(y,t)}{\partial y^2}+\eta_t\frac{\partial^2\delta\dot{q}(y,t)}{\partial y^2}\right]-\frac{(\mu_{\mathrm{S}}-\mu_{\mathrm{K}})P_{\mathrm{ext}}}{v_{\mathrm{c}}\alpha H}\delta\dot{q}(y,t). \tag*{(13)}

Note that δq(y,t)\delta q(y,t) has a non-zero value in the region 0<y/L<S~0<y/L<\widetilde{S}, and δq(y,t)\delta q(y,t) remains zero in the other region due to static friction. Since the perturbation δq(y,t)\delta q(y,t) is zero for y=0y=0 and S~<y/L<1\widetilde{S}<y/L<1, δq(y,t)\delta q(y,t) is expressed as

δq(y,t)=∑mqmeλmtsin⁡kmξ,(14)\delta q(y,t)=\sum_m q_m e^{\lambda_m t}\sin k_m\xi, \tag*{(14)}

where mm is a positive integer, qmq_m is a constant, λm\lambda_m is the eigenvalue of the time evolution operator with km=mπk_m=m\pi and ξ=y/(S~L)\xi=y/(\widetilde{S}L). Substituting Eq. (14) into Eq. (13), multiplying by 2sin⁡knξ2\sin k_n\xi with positive integer nn, and integrating in 0<y<S~L0<y<\widetilde{S}L, we obtain

(1−κ)ρL2λm2+(1−κ)E1km2S~2+(1−κ)ηtkm2S~2λm−(μS−μK)PextL2vcαHλm=0.(15)(1-\kappa)\rho L^2\lambda_m^2+(1-\kappa)E_1\frac{k_m^2}{\widetilde{S}^2}+(1-\kappa)\eta_t\frac{k_m^2}{\widetilde{S}^2}\lambda_m-\frac{(\mu_{\mathrm{S}}-\mu_{\mathrm{K}})P_{\mathrm{ext}}L^2}{v_{\mathrm{c}}\alpha H}\lambda_m=0. \tag*{(15)}

The perturbation δq(y,t)\delta q(y,t) is unstable in the case of Re⁡λ1>0\operatorname{Re}\lambda_1>0 and Im⁡λ1≠0\operatorname{Im}\lambda_1\ne0. The latter condition, Im⁡λ1≠0\operatorname{Im}\lambda_1\ne0, induces the oscillatory motion. However, the backward motion of the oscillation reduces the frictional stress, and the local slip stops when it becomes smaller than the local maximum static frictional stress, which causes the reduction of the slip area S~\widetilde{S}. This oscillatory instability corresponds to the BRP. The BRP continues in a certain region of S~\widetilde{S} because the frictional stress increases again by the drive, and intermittent slip events are observed until the perturbation develops and causes bulk sliding in the case of Re⁡λ1>0\operatorname{Re}\lambda_1 > 0 and Im⁡λ1=0\operatorname{Im}\lambda_1 = 0. In Eq. (15), we find that the stability conditions of the system are generally determined by the competition between the viscosity represented by the third term and velocity-weakening friction represented by the fourth term on the left-hand side. These terms are considered as the stabilizing and destabilizing factors, respectively. The stabilizing factor decreases due to S~−2\tilde{S}^{-2} in the third term as the precursor slip area S~\tilde{S} increases. When S~\tilde{S} reaches the critical area S~c\tilde{S}_{c}, the destabilizing factor overwhelms the stabilizing factor, and the perturbation δq(y,t)\delta q(y,t) becomes unstable. Therefore, S~\tilde{S} increases rapidly just after reaching S~c\tilde{S}_{c}, as shown in Figure 5, and bulk sliding occurs. The viscous term is proportional to 1−κ1-\kappa. The velocity-weakening friction term is proportional to the load on the top of the block but independent of κ\kappa. Thus, if κ(ϕ,d)\kappa(\phi,d) increases by increasing ϕ\phi and dd, the viscosity becomes effectively smaller, which leads to the decrease of S~c\tilde{S}_{c}.

As S~\tilde{S} increases, the mode with m=1m=1 in Eq. (14) becomes unstable first, which determines S~c\tilde{S}_{c}. By definition, the maximum value of S~c\tilde{S}_{c} does not exceed 1, and for S~c<1\tilde{S}_{c}<1, S~c\tilde{S}_{c} satisfies

π2(1−κ)ηtS~c−2+2π(1−κ)LρE1S~c−1=(μS−μK)PextL2vcαH(16)\pi^{2}(1-\kappa)\eta_{\mathrm{t}}\tilde{S}_{c}^{-2}+2\pi(1-\kappa)L\sqrt{\rho E_{1}}\tilde{S}_{c}^{-1}=\frac{(\mu_{\mathrm{S}}-\mu_{\mathrm{K}})P_{\mathrm{ext}}L^{2}}{v_{\mathrm{c}}\alpha H} \tag*{(16)}

which is derived from Eq. (15). Therefore, S~c\tilde{S}_{c} is given by

S~c=min⁡(S~c∗,1)(17)\tilde{S}_{c}=\min(\tilde{S}_{c}^{*},1) \tag*{(17)}

where min⁡(a,b)\min(a,b) is a function that takes the smaller value between aa and bb, and S~c∗\tilde{S}_{c}^{*} is the solution of Eq. (16) given by

S~c∗=πηtL(−ρE1+ρE1+μS−μK1−κPextηtvcαH)(18)\tilde{S}_{c}^{*}=\frac{\pi\eta_{\mathrm{t}}}{L\left(-\sqrt{\rho E_{1}}+\sqrt{\rho E_{1}+\frac{\mu_{\mathrm{S}}-\mu_{\mathrm{K}}}{1-\kappa}\frac{P_{\mathrm{ext}}\eta_{\mathrm{t}}}{v_{\mathrm{c}}\alpha H}}\right)} \tag*{(18)}

For S~c≪1\tilde{S}_{c}\ll1, S~c\tilde{S}_{c} is approximately given by

S~c≃π{[1−κ(ϕ,d)]αμS−μKηtvcPextH}12HL(19)\tilde{S}_{c}\simeq\pi\left\{\frac{[1-\kappa(\phi,d)]\alpha}{\mu_{\mathrm{S}}-\mu_{\mathrm{K}}}\frac{\eta_{\mathrm{t}}v_{\mathrm{c}}}{P_{\mathrm{ext}}H}\right\}^{\frac{1}{2}}\frac{H}{L} \tag*{(19)}

This result indicates that the normalized critical area of the precursor slip S~c\tilde{S}_{c} is a decreasing function of PextP_{\mathrm{ext}} and the size of grooves because κ(ϕ,d)\kappa(\phi,d) in Eq. (19) increases with ϕ\phi and dd, as described in Eq. (11). These analytical results are qualitatively consistent with those of the FEM simulations shown in Figure 6.

The macroscopic static friction coefficient μM\mu_{\mathrm{M}} can be analytically derived in our simplified model [16, 18]. Since the ratio of the local frictional stress to bottom pressure is μS\mu_{\mathrm{S}} in the slip region and μK\mu_{\mathrm{K}} in the static region, μM\mu_{\mathrm{M}} just before the bulk sliding is given by

μM=μK+(μS−μK)S~c(20)\mu_{\mathrm{M}}=\mu_{\mathrm{K}}+(\mu_{\mathrm{S}}-\mu_{\mathrm{K}})\tilde{S}_{c} \tag*{(20)}

This result is qualitatively consistent with the FEM simulations shown in Figure 7, where Eq. (20) is represented by the solid line. Substituting Eq. (20) into Eq. (19), we obtain

μM−μK≃π((μS−μK)[1−κ(ϕ,d)]αηtvcPextH)12HL.(21)\mu_{\mathrm{M}}-\mu_{\mathrm{K}} \simeq\pi\left((\mu_{\mathrm{S}}-\mu_{\mathrm{K}})[1-\kappa(\phi,d)]^{\alpha}\frac{\eta_{\mathrm{t}}v_{\mathrm{c}}}{P_{\mathrm{ext}}H}\right)^{\frac{1}{2}}\frac{H}{L}. \tag*{(21)}

This equation, together with Eq. (11), implies that the macroscopic static friction coefficient μM\mu_{\mathrm{M}} is a decreasing function of PextP_{\mathrm{ext}}, ϕ\phi and dd. The analytical results are qualitatively consistent with the FEM simulations shown in Figure 3.

Equations (19) and (21) indicate that S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} for different ϕ\phi and dd are scaled by the reduction rate of cross-sectional area κ(ϕ,d)\kappa(\phi,d). Figure 10 and Figure 11 respectively show S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} obtained from the FEM simulations against κ(ϕ,d)\kappa(\phi,d). Both S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} are scaled by κ(ϕ,d)\kappa(\phi,d) and decrease with increasing κ(ϕ,d)\kappa(\phi,d). The solid line in Figure 10 represents the analytical result given by Eq. (17). The solid line in Figure 11 represents the result given by Eqs. (17) and (20). Here, we set α=0.2\alpha= 0.2 to semi-quantitatively reproduce the results of the FEM simulations in Sect. 3.1. The deformation before bulk sliding is significant only in the region z/H<0.5z/H < 0.5, as shown in Appendix § 14.

Figure 10. Normalized critical area of precursor slip S~c\widetilde{S}_{\mathrm{c}} against κ(ϕ,d)\kappa(\phi,d) for different values of ϕ\phi and dd with (a) Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and (b) Pext/E=0.006P_{\mathrm{ext}}/E = 0.006. The symbols represent the results of the FEM simulations. The solid lines represent the analytical results given by Eq. (17) with α=0.2\alpha= 0.2.

Figure 11. Macroscopic static friction coefficient μM\mu_{\mathrm{M}} against κ(ϕ,d)\kappa(\phi,d) for different values of ϕ\phi and dd with (a) Pext/E=0.003P_{\mathrm{ext}}/E = 0.003 and (b) Pext/E=0.006P_{\mathrm{ext}}/E = 0.006. The symbols represent the results of the FEM simulations. The solid lines represent the analytical results given by Eqs. (17) and (20) with α=0.2\alpha= 0.2. The dotted and dashed lines indicate μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

This result is consistent with the estimate of α=0.2\alpha= 0.2. The numerical results of FEM are semi-quantitatively consistent with the theoretical analysis. These results explain the decreases of μM\mu_M and S~c\tilde{S}_c with the increases of ϕ\phi and dd in Figure 3 and Figure 6. Here, ϕ\phi represents the decrease in the contact area of the pillars, and dd represents the decrease in the size of the main body. These values determine the reduction rate of the cross-sectional area κ(ϕ,d)\kappa(\phi,d). Thus, it can be concluded that μM\mu_M and S~c\tilde{S}_c decrease with increasing ϕ\phi or dd because of the decrease in effective viscoelasticity due to the reduction of the cross-sectional area, leading to the decline of the stability and bulk sliding with a smaller size of the precursor slip.

Discussion

Generally, grooves on friction surfaces are designed to control lubrication properties in wet conditions. It is considered that the friction coefficient at the wet interface increases with the width and depth of the grooves, because they can eject more lubricant from the friction interface [8, 9, 10, 11]. However, this study reveals that the groove size also affects friction in dry conditions. The static friction coefficient in dry conditions decreases with increases in groove width and depth. This is opposite to the usual consideration for the friction in the wet case. Even in wet conditions, the friction force at the solid-solid interface determines the total friction force after the ejection of the lubricant. These results should aid in improving the design of sliding interfaces with grooves for both wet and dry conditions.

The influence of the groove shape on friction in dry conditions has recently been investigated based on different models, where the effect of grooves is represented by the spatial distribution of the local friction coefficient [46, 47, 48, 49, 50, 51, 52, 53]. These previous works report a decrease in the static friction coefficient by forming longitudinal grooves, which is consistent with our results. However, the effect of the depth of the longitudinal grooves is ignored in their models, while our study is based on a realistic 3D system and reveals its importance. Moreover, previous studies have investigated different patterns of grooves including transversal grooves. Their results have suggested that complex shapes in the friction surface used in various industrial products such as shoe soles and tires and adopted on the surface of living things such as snakes [48, 60, 61] affect the frictional properties. Therefore, the extension of our study to these complex shapes will lead to more efficient guiding principles for groove design.

In this study, the parameter values for virtual materials are adopted to reduce the computational load, which does not correspond to those for real materials. These are selected to compare the results with those in previous simulations [16, 18]. However, the mechanism of changes in the friction coefficient revealed by our theory in Sect. 3.2 is universal and independent of specific parameter values. The numerical results for flat friction surfaces in Ref. [16] have been reproduced in experiments on PMMA [17]. The occurrence of the quasi-static precursor (slow slip event, SSE) and the dependence of its destabilization on pressure are also confirmed in an experiment on PMMA [19]. Therefore, we expect our results will be experimentally verified in future work.

Conclusion

Friction surfaces of products such as shoe soles, tires, and sliding parts of machines have grooves. Several studies on grooves have focused on controlling lubrication properties via their design. However, it has been empirically known that grooves affect friction even in dry conditions, although no theoretical explanation exists. In this study, we have performed numerical simulations of viscoelastic objects using 3D FEM to clarify the effect of longitudinal grooves on static friction in a dry condition. We have revealed that the static friction coefficient is a decreasing function of the groove size, and that precursor slip occurs before bulk sliding. The static friction coefficient is scaled by the normalized critical area of the precursor slip. Based on the simplified model, we have theoretically derived the equation for the static friction coefficient depending on the groove size. The theoretical result indicates that the static friction coefficient decreases with the reduction rate of the cross-sectional area in the viscoelastic object. The decrease in the cross-sectional area reduces the effective viscosity, which enhances the instability of the precursor slip and decreases the static friction coefficient. Our results provide new guiding principles for groove design for static friction control beyond the empirical laws for both wet and dry conditions. Investigation of different types of grooves and their effect on dynamic friction will be the subject of future studies.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments This research used computational resources of Fugaku at the RIKEN Center for Computational Science (Project: hp220372), Yukawa-21 at YITP, Kyoto University, SQUID at the Cybermedia Center, Osaka University, ohtaka and kugui at ISSP, the University of Tokyo, the supercomputers at RCCS, Okazaki, Japan (Project: 23-IMS-C126), and JSS3 at JAXA. We would like to thank Editage (www.editage.com) for English language editing.

Funding This work was supported by the Japan Society for the Promotion of Science KAKENHI (Grant Numbers JP21H01006, JP22KJ2190, JP23K03248, and JP23K03252).

Author information

Contributions All authors contributed to the study conception and design. The FEM simulation and analysis based on a simplified model are performed by WI. The first draft of the manuscript was written by WI, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Ethics declarations

Competing interests The authors declare no competing interests.

Supplementary information

The online version contains supplementary material.

Results for L/H=2L/H = 2

In this appendix, we show the results for L/H=2L/H = 2. Figure 12a and b respectively show S~c\tilde{S}_{c} and μM\mu_{M} against dd for different values of ϕ\phi with Pext/E=0.003P_{\mathrm{ext}}/E = 0.003. We find that S~c\tilde{S}_{c} and μM\mu_{M} are decreasing functions of ϕ\phi and dd, which is consistent with the results for L/H=4L/H = 4, as shown in Figure 3 and Figure 6. Compared to Figure 3a and Figure 6a for the identical pressure, we find that S~c\tilde{S}_{c} and μM\mu_{M} increase with decreasing LL.

Image not extracted

Figure 12. (a) Normalized critical area of precursor slip S~c\tilde{S}_{c} and (b) macroscopic static friction coefficient μM\mu_{M} against dd for different values of ϕ\phi with L/H=2L/H = 2 and Pext/E=0.003P_{\mathrm{ext}}/E = 0.003. The dotted and dashed lines indicate μS\mu_{S} and μK\mu_{K}, respectively.

Image not extracted

Figure 13b. Normalized critical area of precursor slip S~c\widetilde{S}_{\mathrm{c}} and (b) macroscopic static friction coefficient μM\mu_{\mathrm{M}} against κ(ϕ,d)\kappa(\phi,d) for different values of ϕ\phi and dd with L/H=2L/H=2 and Pext/E=0.003P_{\mathrm{ext}}/E=0.003. The solid lines represent the analytical results of Eqs. (17) and (20) with α=0.2\alpha=0.2. The dotted and dashed lines indicate μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

Figure 13a and b respectively show S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} against κ(ϕ,d)\kappa(\phi,d) with Pext/E=0.003P_{\mathrm{ext}}/E=0.003. We find that S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} are decreasing functions of κ(ϕ,d)\kappa(\phi,d). The solid lines show the analytical results for α=0.2\alpha=0.2 based on Eqs. (17) and (20). It is shown that κ(ϕ,d)\kappa(\phi,d) for α=0.2\alpha=0.2 can scale S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} even for the system with L/H=2L/H=2. The decreases in S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} by increasing LL is consistent with Eqs. (19) and (21) for the theoretical analysis in Sect. 3.2.

Image not extracted

Figure 13a. Normalized critical area of precursor slip S~c\widetilde{S}_{\mathrm{c}} against κ(ϕ,d)\kappa(\phi,d) for different values of ϕ\phi and dd with L/H=2L/H=2 and Pext/E=0.003P_{\mathrm{ext}}/E=0.003. The solid lines represent the analytical results of Eqs. (17) and (20) with α=0.2\alpha=0.2. The dotted and dashed lines indicate μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

Dependence on Number of Pillars nxn_x in Friction Surface

In this appendix, we verify the dependence of the S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} on the number of pillars of the block nxn_x for L/H=4L/H=4, d/H=0.5d/H=0.5, and ϕ=0.5\phi=0.5. Figure 14a and b show S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} against nxn_x for Pext/E=0.003P_{\mathrm{ext}}/E=0.003 and 0.0060.006, respectively. In Fig. 14, we find that S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} are almost independent of nxn_x. According to the theoretical results given by Eqs. (19) and (21), S~c\widetilde{S}_{\mathrm{c}} and μM\mu_{\mathrm{M}} depend on the reduction rate of the cross-sectional area

Image not extracted

Figure 14a. Normalized critical area of precursor slip S~c\widetilde{S}_{\mathrm{c}} against nxn_x for Pext/E=0.003P_{\mathrm{ext}}/E=0.003 and 0.0060.006 with L/H=4L/H=4, d/H=0.5d/H=0.5, and ϕ=0.5\phi=0.5. The dotted and dashed lines represent μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

Image not extracted

Figure 14b. Normalized critical area of precursor slip S~c\widetilde{S}_{\mathrm{c}} and (b) macroscopic static friction coefficient μM\mu_{\mathrm{M}} against nxn_x for Pext/E=0.003P_{\mathrm{ext}}/E=0.003 and 0.0060.006 with L/H=4L/H=4, d/H=0.5d/H=0.5, and ϕ=0.5\phi=0.5. The dotted and dashed lines represent μS\mu_{\mathrm{S}} and μK\mu_{\mathrm{K}}, respectively.

κ(ϕ,d)\kappa(\phi,d), which is independent of nxn_x, as described in Eq. (11). This analytical result explains the behaviors of S~c\tilde{S}_c and μM\mu_M in Fig. 14.

Acceleration Distribution

Figure 15 demonstrates the spatial distribution of the yy-direction acceleration u¨y\ddot{u}_y for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, d/H=0.5d/H = 0.5, and ϕ=0.5\phi= 0.5 in the cross-section at x=0.5Wx = 0.5W with U/L=5.43×10−3U/L = 5.43 \times10^{-3}, just before bulk sliding. The critical length in the yy direction of the precursor slip is estimated to be S~cL\tilde{S}_cL, because the precursor slip propagates uniformly in the xx-direction, as shown in Figure 4. The acceleration around the region of y/H=0y/H = 0 and 0.5≤z/H≤10.5 \le z/H \le1 is small, because the displacement is fixed at the rigid plate position at the region. Moreover, the acceleration is negligible in the static region at the bottom (z=0z = 0) with σ(fric)/p≈μK\sigma^{(\mathrm{fric})}/p \approx\mu_K for S~cL≤y≤L\tilde{S}_cL \le y \le L, as shown in Figure 8a. Therefore, the region with significant acceleration before bulk sliding is concentrated in the region for 0≤y≤S~cL0 \le y \le\tilde{S}_cL near the bottom, as shown in Figure 15.

Figure 15. Spatial distribution of acceleration along the yy direction, u¨y\ddot{u}_y, on cross-section at x=0.5Wx = 0.5W for Pext/E=0.003P_{\mathrm{ext}}/E = 0.003, d/H=0.5d/H = 0.5, and ϕ=0.5\phi= 0.5 at U/L=5.43×10−3U/L = 5.43 \times10^{-3}, just before bulk sliding. The gray thick line for y/H=0y/H = 0 and 0.5≤z/H≤10.5 \le z/H \le1 indicates the position of the rigid plate.

Quasi-static Solution

Using the equation of motion (12), boundary conditions, and assumption of friction coefficient μ\mu in Sect. 3.2, we obtain the quasi-static solution q0(y)q_0(y) at U=0U = 0 as

q0(y)=μKPext2(1−κ)E1αH(y2−2Ly).(22)q_0(y) = \frac{\mu_K P_{\mathrm{ext}}}{2(1-\kappa)E_1\alpha H}\left(y^2-2Ly\right). \tag*{(22)}

The quasi-static solution qa(y)q_a(y) for U>0U > 0 is obtained as

qa(y)={q1(y),0≤y≤S~Lq0(y),S~L≤y≤L(23)q_a(y) = \begin{cases} q_1(y), & 0 \le y \le\tilde{S}L \\ q_0(y), & \tilde{S}L \le y \le L \end{cases} \tag*{(23)}

With the assumption of μ\mu and connectivity condition at y=S~Ly = \tilde{S}L, q1(S~L)=q0(S~L)q_1(\tilde{S}L) = q_0(\tilde{S}L) and dq1(S~L)/dy=dq0(S~L)/dydq_1(\tilde{S}L)/dy = dq_0(\tilde{S}L)/dy, we obtain q1q_1 as

q1(y)=q0(y)+(μS−μK)Pext2(1−κ)E1αH(y2−2S~Ly+S~2L2).(24)q_1(y) = q_0(y) + \frac{(\mu_S-\mu_K)P_{\mathrm{ext}}}{2(1-\kappa)E_1\alpha H}\left(y^2-2\tilde{S}Ly+\tilde{S}^2L^2\right). \tag*{(24)}

To satisfy the boundary condition at y/L=0y/L = 0, S~\tilde{S} is derived as

S~=[2(1−κ)E1αHU(μS−μK)PextL2]12.(25)\tilde{S} = \left[\frac{2(1-\kappa)E_{1}\alpha H U}{(\mu_{S}-\mu_{K})P_{\mathrm{ext}}L^{2}}\right]^{\frac{1}{2}}. \tag*{(25)}

References

  1. [1]Bowden, F.P., Tabor, D.: The Friction and Lubrication of Solids. Oxford University Press, New York (1950)
  2. [2]Persson, B.N.J.: Sliding Friction: Physical Principles and Applications, 2nd edn. Springer, Berlin (2000)
  3. [3]Popov, V.L.: Contact Mechanics and Friction: Physical Principles and Applications, 2nd edn. Springer, Berlin (2017)
  4. [4]Rabinowicz, E.: Friction and Wear of Materials, 2nd edn. John Wiley & Sons, New York (1995)
  5. [5]Dowson, D.: History of Tribology, 2nd edn. John Wiley & Sons, New York (1998)
  6. [6]Bhushan, B.: Principles and Applications of Tribology, 2nd edn. John Wiley & Sons, New York (2013)
  7. [7]Baumberger, T., Caroli, C.: Solid friction from stick-slip down to pinning and aging. Advances in Physics 55(3-4), 279–348 (2006). https://doi.org/10.1080/00018730600732186
  8. [8]Li, K.W., Chen, C.J.: The effect of shoe soling tread groove width on the coefficient of friction with different sole materials, floors, and contaminants. Applied Ergonomics 35(6), 499–507 (2004). https://doi.org/10.1016/j.apergo.2004.06.010
  9. [9]Li, K.W., Chin, J.C.: Effects of tread groove orientation and width of the footwear pads on measured friction coefficients. Safety Science 43(7), 391–405 (2005). https://doi.org/10.1016/j.ssci.2005.08.006
  10. [10]Li, K.W., Wu, H.H., Lin, Y.-C.: The effect of shoe sole tread groove depth on the friction coefficient with different tread groove widths, floors and contaminants. Applied Ergonomics 37(6), 743–748 (2006). https://doi.org/10.1016/j.apergo.2005.11.007
  11. [11]Yamaguchi, T., Umetsu, T., Ishizuka, Y., Kasuga, K., Ito, T., Ishizawa, S., Hokkirigawa, K.: Development of new footwear sole surface pattern for prevention of slip-related falls. Safety Science 50(4), 986–994 (2012). https://doi.org/10.1016/j.ssci.2011.12.017
  12. [12]Archard, J.F.: Elastic deformation and the laws of friction. Proceedings of the Royal Society of London A 243(1233), 190–205 (1957). https://doi.org/10.1098/rspa.1957.0214
  13. [13]Dieterich, J.H., Kilgore, B.D.: Imaging surface contacts: power law contact distributions and contact stresses in quartz, calcite, glass and acrylic plastic. Tectonophysics 256(1-4), 219–239 (1996). https://doi.org/10.1016/0040-1951(95)00165-4
  14. [14]Bouissou, S., Petit, J.P., Barquins, M.: Normal load, slip rate and roughness influence on the polymethylmethacrylate dynamics of sliding 1. Stable sliding to stick-slip transition. Wear 214(2), 156–164 (1998). https://doi.org/10.1016/S0043-1648(97)00242-1
  15. [15]Ben-David, O., Fineberg, J.: Static friction coefficient is not a material constant. Physical Review Letters 106(25), 254301 (2011). https://doi.org/10.1103/PhysRevLett.106.254301
  16. [16]Otsuki, M., Matsukawa, H.: Systematic breakdown of Amontons’ law of friction for an elastic object locally obeying Amontons’ law. Scientific Reports 3, 1586 (2013). https://doi.org/10.1038/srep01586
  17. [17]Katano, Y., Nakano, K., Otsuki, M., Matsukawa, H.: Novel friction law for the static friction force based on local precursor slipping. Scientific Reports 4, 6324 (2014). https://doi.org/10.1038/srep06324
  18. [18]Iwashita, W., Matsukawa, H., Otsuki, M.: Static friction coefficient depends on the external pressure and block shape due to precursor slip. Scientific Reports 13, 2511 (2023). https://doi.org/10.1038/s41598-023-29764-w
  19. [19]Selvadurai, P.A., Glaser, S.D., Parker, J.M.: On factors controlling precursor slip fronts in the laboratory and their relation to slow slip events in nature. Geophysical Research Letters 44(6), 2743–2754 (2017). https://doi.org/10.1002/2017GL072538
  20. [20]Yamaguchi, T., Morishita, M., Doi, M., Hori, T., Sakaguchi, H., Ampuero, J.-P.: Gutenberg-Richter’s law in sliding friction of gels. Journal of Geophysical Research: Solid Earth 116(B12), 12306 (2011). https://doi.org/10.1029/2011JB008415
  21. [21]Obara, K., Kato, A.: Connecting slow earthquakes to huge earthquakes. Science 353(6296), 253–257 (2016). https://doi.org/10.1126/science.aaf1512
  22. [22]Kato, A., Ben-Zion, Y.: The generation of large earthquakes. Nature Reviews Earth and Environment 2, 26–39 (2021). https://doi.org/10.1038/s43017-020-00108-w
  23. [23]Petrillo, G., Lippiello, E., Landes, F.P., Rosso, A.: The influence of the brittle-ductile transition zone on aftershock and foreshock occurrence. Nature Communications 11, 3010 (2020). https://doi.org/10.1038/s41467-020-16811-7
  24. [24]Xu, S., Fukuyama, E., Yamashita, F., Kawakata, H., Mizoguchi, K., Takizawa, S.: Fault strength and rupture process controlled by fault surface topography. Nature Geoscience 16, 94–100 (2023). https://doi.org/10.1038/s41561-022-01093-z
  25. [25]Svetlizky, I., Fineberg, J.: Classical shear cracks drive the onset of dry frictional motion. Nature 509, 205–208 (2014). https://doi.org/10.1038/nature13202
  26. [26]Bayart, E., Svetlizky, I., Fineberg, J.: Fracture mechanics determine the lengths of interface ruptures that mediate frictional motion. Nature Physics 12, 166–170 (2016). https://doi.org/10.1038/nphys3539
  27. [27]Svetlizky, I., Kammer, D.S., Bayart, E., Cohen, G., Fineberg, J.: Brittle fracture theory predicts the equation of motion of frictional rupture fronts. Physical Review Letters 118(12), 125501 (2017). https://doi.org/10.1103/PhysRevLett.118.125501
  28. [28]Berman, N., Cohen, G., Fineberg, J.: Dynamics and properties of the cohesive zone in rapid fracture and friction. Physical Review Letters 125(12), 125503 (2020). https://doi.org/10.1103/PhysRevLett.125.125503
  29. [29]Givrtzman, S., Fineberg, J.: Nucleation fronts ignite the interface rupture that initiates frictional motion. Nature Physics 17, 1037–1042 (2021). https://doi.org/10.1038/s41567-021-01299-9
  30. [30]Kammer, D.S., Radiguet, M., Ampuero, J.-P., Molinari, J.-F.: Linear elastic fracture mechanics predicts the propagation distance of frictional slip. Tribology Letters 57, 23 (2015). https://doi.org/10.1007/s11249-014-0451-8
  31. [31]Castellano, M., Lorez, F., Kammer, D.S.: Nucleation of frictional slip: A yielding or a fracture process? Journal of the Mechanics and Physics of Solids 173, 105193 (2023). https://doi.org/10.1016/j.jmps.2022.105193
  32. [32]Rubinstein, S.M., Cohen, G., Fineberg, J.: Dynamics of precursors to frictional sliding. Physical Review Letters 98(22), 226103 (2007). https://doi.org/10.1103/PhysRevLett.98.226103
  33. [33]Maegawa, S., Suzuki, A., Nakano, K.: Precursors of global slip in a longitudinal line contact under non-uniform normal loading. Tribology Letters 38, 313–323 (2010). https://doi.org/10.1007/s11249-010-9611-7
  34. [34]Burridge, R., Knopoff, L.: Model and theoretical seismicity. Bulletin of the Seismological Society of America 57(3), 341–371 (1967). https://doi.org/10.1785/BSSA0570030341
  35. [35]de Sousa Vieira, M.: Self-organized criticality in a deterministic mechanical model. Physical Review A 46(10), 6288 (1992). https://doi.org/10.1103/PhysRevA.46.6288
  36. [36]Braun, O.M., Barel, I., Urbahk, M.: Dynamics of transition from static to kinetic friction. Physical Review Letters 103(19), 194301 (2009). https://doi.org/10.1103/PhysRevLett.103.194301
  37. [37]Capozza, R., Urbahk, M.: Static friction and the dynamics of interfacial rupture. Physical Review B 86(8), 085430 (2012). https://doi.org/10.1103/PhysRevB.86.085430
  38. [38]Ozaki, S., Inanobe, C., Nakano, K.: Finite element analysis of precursors to macroscopic stick-slip motion in elastic materials: analysis of friction test as a boundary value problem. Tribology Letters 55, 151–163 (2014). https://doi.org/10.1007/s11249-014-0343-y
  39. [39]Scheibert, J., Dysthe, D.K.: Role of friction-induced torque in stick-slip motion. Europhysics Letters 92(5), 54001 (2010). https://doi.org/10.1209/0295-5075/92/54001
  40. [40]Amundsen, D.S., Scheibert, J., Thøgersen, K., Trømborg, J., Malthe-Sørenssen, A.: 1D model of precursors to frictional stick-slip motion allowing for robust comparison with experiments. Tribology Letters 45, 357–369 (2012). https://doi.org/10.1007/s11249-011-9894-3
  41. [41]Trømborg, J.K., Sveinsson, H.A., Scheibert, J., Thøgersen, K., Amundsen, D.S., Malthe-Sørenssen, A.: Slow slip and the transition from fast to slow fronts in the rupture of frictional interfaces. Proceedings of the National Academy of Sciences of the U.S.A. 111(24), 8764–8769 (2014). https://doi.org/10.1073/pnas.1321752111
  42. [42]Radiguet, M., Kammer, D.S., Gillet, P., Molinari, J.-F.: Survival of heterogeneous stress distributions created by precursory slip at frictional interfaces. Physical Review Letters 111(16), 164302 (2013). https://doi.org/10.1103/PhysRevLett.111.164302
  43. [43]Albertini, G., Karrer, S., Grigoriu, M.D., Kammer, D.S.: Stochastic properties of static friction. Journal of the Mechanics and Physics of Solids 147, 104242 (2021). https://doi.org/10.1016/j.jmps.2020.104242
  44. [44]Taloni, A., Benassi, A., Sandfeld, S., Zapperi, S.: Scalar model for frictional precursors dynamics. Scientific Reports 5, 8086 (2015). https://doi.org/10.1038/srep08086
  45. [45]de Geus, T.W.J., Popović, M., Ji, W., Rosso, A., Wyart, M.: How collective asperity detachments nucleate slip at frictional interfaces. Proceedings of the National Academy of Sciences of the U.S.A. 116(48), 23977–23983 (2019). https://doi.org/10.1073/pnas.1906551116
  46. [46]Capozza, R., Pugno, N.M.: Effect of surface grooves on the static friction of an elastic slider. Tribology Letters 58, 35 (2015). https://doi.org/10.1007/s11249-015-0510-9
  47. [47]Costagliola, G., Bosia, F., Pugno, N.M.: Static and dynamic friction of hierarchical surfaces. Physical Review E 94(6), 063003 (2016). https://doi.org/10.1103/PhysRevE.94.063003
  48. [48]Costagliola, G., Bosia, F., Pugno, N.M.: Hierarchical spring-block model for multiscale friction problems. ACS Biomaterials Science and Engineering 3(11), 2845–2852 (2017). https://doi.org/10.1021/acsbiomaterials.6b00709
  49. [49]Costagliola, G., Bosia, F., Pugno, N.M.: A 2-D model for friction of complex anisotropic surfaces. Journal of the Mechanics and Physics of Solids 112, 50–65 (2018). https://doi.org/10.1016/j.jmps.2017.11.015
  50. [50]Costagliola, G., Bosia, F., Pugno, N.M.: Correlation between slip precursors and topological length scales at the onset of frictional sliding. International Journal of Solids and Structures 243, 111525 (2022). https://doi.org/10.1016/j.ijsolstr.2022.111525
  51. [51]Maegawa, S., Itoigawa, F., Nakamura, T., Matsuoka, H., Fukui, S.: Effect of tangential loading history on static friction force of elastic slider with split contact surface: model calculation. Tribology Letters 65, 37 (2017). https://doi.org/10.1007/s11249-017-0811-2
  52. [52]Berardo, A., Costagliola, G., Ghio, S., Boscardin, M., Bosia, F., Pugno, N.M.: An experimental-numerical study of the adhesive static and dynamic friction of micro-patterned soft polymer surfaces. Materials and Design 181, 107930 (2019). https://doi.org/10.1016/j.matdes.2019.107930
  53. [53]Balestra, S., Costagliola, G., Pegoraro, A., Picollo, F., Molinari, J.-F., Pugno, N.M., Vittone, E., Bosia, F., Sin, A.: Experimental and numerical study of the effect of surface patterning on the frictional properties of polymer surfaces. Journal of Tribology 144(3), 031704 (2022). https://doi.org/10.1115/1.4052777
  54. [54]Maegawa, S., Itoigawa, F., Nakamura, T.: Effect of surface grooves on kinetic friction of a rubber slider. Tribology International 102, 326–332 (2016). https://doi.org/10.1016/j.triboint.2016.05.019
  55. [55]Gao, Y., Wang, Y.: Research on the dynamic behaviors in the sliding friction of silicone. Tribology Letters 71, 15 (2023). https://doi.org/10.1007/s11249-022-01675-3
  56. [56]Landau, L.D., Lifshitz, E.M., Kosevich, A.M., Pitaevskii, L.P.: Theory of Elasticity, 3rd edn. Butterworth-Heinemann, Oxford (1986)
  57. [57]Wriggers, P.: Computational Contact Mechanics, 2nd edn. Springer, Berlin (2006)
  58. [58]Dieterich, J.H., Kilgore, B.D.: Direct observation of frictional contacts: New insights for state-dependent properties. Pure and Applied Geophysics 143, 283–302 (1994). https://doi.org/10.1007/BF00874332
  59. [59]Baumberger, T., Berthoud, P., Caroli, C.: Physical analysis of the state- and rate-dependent friction law. II. Dynamic friction. Physical Review B 60(6), 3928–3939 (1999). https://doi.org/10.1103/PhysRevB.60.3928
  60. [60]Baum, M.J., Heepe, L., Gorb, S.N.: Friction behavior of a microstructured polymer surface inspired by snake skin. Beilstein Journal of Nanotechnology 5, 83–97 (2014). https://doi.org/10.3762/bjnano.5.8
  61. [61]Baum, M.J., Kovalev, A.E., Michels, J., Gorb, S.N.: Anisotropic friction of the ventral scales in the snake Lampropeltis getula californiae. Tribology Letters 54, 139–150 (2014). https://doi.org/10.1007/s11249-014-0319-y

Paper details

vlnc:W_h53qaha2
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents