Study of detonation initiation in hydrogen/air flow

While extensive studies have been conducted concerning the formation of detonation waves in various combustible gaseous mixtures under static conditions since the 1950s, there is very little experimental work on simple flowing systems. In this study, experiments on the deflagration to detonation transition (DDT) of a hydrogen–air flow system were carried out to see the effects of tube diameter, equivalence ratio, and flow types in a premixed and non-premixed flow. Tube diameters used were 25, 50, and 100 mm. The premixed experiments show that the larger tube diameter provides a wider range in run-up distance, reduction of L _DDT/D (ratio of the run-up distance, L _DDT to tube diameter), and expansion of the detonable concentration limit by spreading the cell width. The result of the non-premixed experiments show that similar values of the run-up distance to the premixed experiments are obtained at an equivalence ratio of about 1.0, however, fluctuations of DDT occur near the DDT concentration limit. Under laminar flow conditions at a Reynolds number of less than 2,300, the difference between the two systems could not be observed. However, when the Reynolds number increases towards turbulent conditions, the DDT run-up distance decreases compared to that of static flow conditions. This paper is based on work that was presented at the 21st International Colloquium on the Dynamics of Explosions and Reactive Systems, Poitiers, France, July 23–27, 2007.     

DDT in a smooth tube filled with a hydrogen–oxygen mixture

Results of experimental study on DDT in a smooth tube filled with sensitive mixtures having detonation cell size from 1 to 3 orders of magnitude smaller than the tube diameter are presented. Stoichiometric hydrogen–oxygen mixtures were used in the tests with initial pressure ranging from 0.2 to 8 bar. A dependence of the run-up distance to DDT on the initial pressure is studied. This dependence is found to be close to the inverse proportionality. It is suggested that the flow ahead of the flame results in formation of the turbulent boundary layer. This boundary layer controls the scale of turbulent motions in the flow. A simple model to estimate the maximum scale of the turbulent pulsations (boundary layer thickness) at flame positions along the tube is presented. The largest scale of the turbulent motions at the location of the onset of detonation is shown to be 1 order of magnitude greater than the detonation cell widths, λ, in all the tests. It is suggested that the onset of detonation is triggered during flame acceleration as soon as the maximum scale of the turbulent pulsations increases up to about 10 λ. The model to estimate the maximum size of turbulent motions, δ, and the correlation δ≈ 10λ, give a basis for estimations of the run-up distances to DDT in tubes with internal diameter D > 20λ. PACS 47.40.-x; 47.27.Nz This paper was based on work that was presented at the 19th Inter-national Colloquium on the Dynamics of Explosions and Reactive Systems, Hakone, Japan, July 27 - August 1, 2003     

Influence of mixture sensitivity and pore size on detonation velocities in porous media

Experiments have been carried out to determine the dependence of the detonation velocity in porous media, on mixture sensitivity and pore size. A detonation is established at the top end of a vertical tube and allowed to propagate to the bottom section housing the porous bed, comprised of alumina spheres of equal diameter (1–32 mm). Several of the common detonable fuels were tested at atmospheric initial pressure. Results indicate the existence of a continuous range of velocities with change in Φ, spanning the lean and the rich propagation limits. For all fuels in a given porous bed, the velocity decreases from a maximum value at the most sensitive mixture near Φ≈1 (minimum induction length), toV/V _CJ≈0.3 at the limits. A decrease in pore size brings about a reduction inV/V _CJ and a narrowing of the detonability range for each fuel. For porous media comprised of spherical particles, it was possible to correlate the velocity data corresponding to a variety of different mixtures and for a broad range of particle sizes, using the following empirical expression:V/V _CJ=[1–0.35 log(d _c /d _p)]±0.1. The critical tube diameterd _c is used as a measure of mixture sensitivity andd _p denotes the pore diameter. An examination of the phenomenon at the composition limits, suggests that wave failure is controlled by a turbulent quenching mechanism.     

Detonability of THDCPD-exo–air mixtures

In the frame of industrial risk and propulsive application, the detonability study of JP10–air mixtures was performed. The simulation and measurements of detonation parameters were performed for THDCPD-exo/air mixtures at various initial pressure (1 bar < P ₀ < 3 bar) and equivalence ratio (0.8 < Φ < 1.6) in a heated tube (T ₀ ~ 375 K). Numerical simulations of the detonation were performed with the STANJAN code and a detailed kinetic scheme of the combustion of THDCPD. The experimental study deals with the measurements of detonation velocity and cell size λ. The measured velocity is in a good agreement with the calculated theoretical values. The cell size measurements show a minimum value for Φ ~ 1.2 at every level of initial pressure studied and the calculated induction length L _i corresponds to cell size value with a coefficient k = λ/L _i = 24 at P ₀ = 1 bar. Based on the comparison between the results obtained during this study and those available in the literature on the critical initiation energy E _c, critical tube diameter d _c and deflagration to detonation transition length L _DDT, we can conclude that the detonability of THDCPD–air mixtures corresponds to that of hydrocarbon–air mixtures.

---

This paper is based on the work presented at the 33rd International Pyrotechnics Seminar, IPS 2006, Fort Collins, July 16–21, 2006.     

DDT in methane-air mixtures

Experimental results from a study on the critical condition for deflagration-to-detonation transition (DDT) in methane-air mixtures are presented. Experiments were carried out at 293 K and 1 atm using methane-air mixtures with methane concentrations ranging from 5.5 to 17% vol. The tests were performed in detonation tubes with inner-diameters of 174 mm and 520 mm. Detonation cell widths l\lambda were determined in the tests for a range of methane concentrations. The results of DDT tests indicate that for a tube cross-sectional area blockage ratio (BR) of 0.3 the critical condition for DDT can be characterized by the d/l = 1d/\lambda = 1 criterion. However, for a BR = 0.6 the critical value d/ld/\lambda was significantly higher. The data also show that the critical condition for DDT can be described by L/l = 7L/\lambda = 7, where L is the characteristic length-scale of the channel volume between orifice plates. This length-scale is defined by a grouping of the orifice plate dimensions (inner and outer-diameter) and plate spacing.     

DDT in fuel–air mixtures at elevated temperatures and pressures

An experimental study was carried out to investigate flame acceleration and deflagration-to-detonation transition (DDT) in fuel–air mixtures at initial temperatures up to 573 K and pressures up to 2 atm. The fuels investigated include hydrogen, ethylene, acetylene and JP-10 aviation fuel. The experiments were performed in a 3.1-m long, 10-cm inner-diameter heated detonation tube equipped with equally spaced orifice plates. Ionization probes were used to measure the flame time-of-arrival from which the average flame velocity versus propagation distance could be obtained. The DDT composition limits and the distance required for the flame to transition to detonation were obtained from this flame velocity data. The correlation developed by Veser et al. (run-up distance to supersonic flames in obstacle-laden tubes. In the proceedings of the 4th International Symposium on Hazards, Prevention and Mitigation of Industrial Explosions, France (2002)) for the flame choking distance proved to work very well for correlating the detonation run-up distance measured in the present study. The only exception was for the hydrogen–air data at elevated initial temperatures which tended to fall outside the scatter of the hydrocarbon mixture data. The DDT limits obtained at room temperature were found to follow the classical d/λ = 1 correlation, where d is the orifice plate diameter and λ is the detonation cell size. Deviations found for the high-temperature data could be attributed to the one-dimensional ZND detonation structure model used to predict the detonation cell size for the DDT limit mixtures. This simple model was used in place of actual experimental data not currently available. PACS 47.40.-x; 47.70.Fw This paper was based on work that was presented at the 19th Interna-tional Colloquium on the Dynamics of Explosions and Reactive Sys-tems, Hakone, Japan, July 27 - August 1, 2003     

Interaction of detonation with inert gas zone

The results of experimental study on detonation interaction with the regions of low reactivity, generated by the injection of an inert gas into reactive mixture, are reported. A square cross-section 60×60 mm, 3.6 m long detonation channel was used. The experiments were done for stoichiometric H₂−O₂ mixture at 0.3 bar and 0.5 bar initial pressure and room temperature. The inert gas (Ar, He, N₂ or CO₂) was injected from 0.523 dm³ container into the main channel 1 s before ignition. The size of the inert zone was controlled by inert initial pressure. The behavior of detonation was studied using direct streak photography and pressure transducers. The study has shown that at low pressure of Ar, N₂ and CO₂ injection only a slight decrease of detonation velocity occurs. At higher injection pressures complete damping of detonation and flame extinguishment occur, followed by flame reigniton and DDT outside the inert zone. For low He injection pressures an increase in detonation velocity was recorded. For higher injection pressures, detonation damping occurred, followed by DDT process. The results have shown that CO₂ has the strongest effect on damping 2H₂+O₂ detonation, with N₂ and Ar in the next places, and He very far behind. The effectiveness of inert gas in detonation damping was found proportional to its molecular weight and reciprocal to its specific heat ratio. The numerical simulations of detonation propagation through inert gas zone were also performed using the one- dimensional Detonation Lagrangean code with simple energy release model. The results of simulations are in good qualitative agreement with experimental tendencies. In particular, the model has shown that the re-initiation of detonation is enhanced by smooth concentration gradients at inert/reactive interface. An abridged version of this paper was presented at the 15th Int. Colloquium on the Dynamics of Explosions and Reactive Systems at Boulder, Colorado, from July 30 to August 4, 1995     

An investigation of electromagnetic wave propagation in plasma by shock tube

This paper presents the electromagnetic wave propagation characteristics in plasma and the attenuation coefficients of the microwave in terms of the parameters he, v, w, L, wb. The φ800 mm high temperature shock tube has been used to produce a uniform plasma. In order to get the attenuation of the electromagnetic wave through the plasma behind a shock wave, the microwave transmission has been used to measure the relative change of the wave power. The working frequency is f = (2-35)GHz (ω=2πf, wave length A =15cm-8mm). The electron density in the plasma is ne = (3&#215;10^10-1&#215;10^14) cm^-3. The collision frequency v = (1&#215;10^8-6&#215;10^10) Hz. The thickness of the plasma layer L = (2-80)cm. The electron circular frequency ωb=eBo/me, magnetic flux density B0 = (0-0.84)T. The experimental results show that when the plasma layer is thick (such as L/λ≥10), the correlation between the attenuation coefficients of the electromagnetic waves and the parameters ne,v,ω, L determined from the measurements are in good agreement with the theoretical predictions of electromagnetic wave propagations in the uniform infinite plasma. When the plasma layer is thin (such as when both L and A are of the same order), the theoretical results are only in a qualitative agreement with the experimental observations in the present parameter range, but the formula of the electromagnetic wave propagation theory in an uniform infinite plasma can not be used for quantitative computations of the correlation between the attenuation coefficients and the parameters ne,v,ω, L. In fact, if ω&lt;ωp, v^2&lt;&lt;ω^2, the power attenuations K of the electromagnetic waves obtained from the measurements in the thin-layer plasma are much smaller than those of the theoretical predictions. On the other hand, if ω&gt;ωp, v^2&lt;&lt;ω^2 (just v≈f), the measurements are much larger than the theoretical results. Also, we have measured the electromagnetic wave power attenuation value under the magnetic field and without a magnetic field. The result indicates that the value measured under the magnetic field shows a distinct improvement.     

Mixed Convection Heat Transfer in the Annulus Between Two Concentric Vertical Cylinders Using Porous Layers

A numerical investigation of the steady-state, laminar, axi-symmetric, mixed convection heat transfer in the annulus between two concentric vertical cylinders using porous inserts is carried out. The inner cylinder is subjected to constant heat flux and the outer cylinder is insulated. A finite volume code is used to numerically solve the sets of governing equations. The Darcy–Brinkman–Forchheimer model along with Boussinesq approximation is used to solve the flow in the porous region. The Navier–Stokes equation is used to describe the flow in the clear flow region. The dependence of the average Nusselt number on several flow and geometric parameters is investigated. These include: convective parameter, λ, Darcy number, Da, thermal conductivity ratio, K _r, and porous-insert thickness to gap ratio (H/D). It is found that, in general, the heat transfer enhances by the presence of porous layers of high thermal conductivity ratios. It is also found that there is a critical thermal conductivity ratio on which if the values of Kr are higher than the critical value the average Nusselt number starts to decrease. Also, it found that at low thermal conductivity ratio (K _r ≈ 1) and for all values of λ the porous material acts as thermal insulation.     

Thermal convection in a porous layer: Effects of anisotropy and surface boundary conditions

The theory describing the onset of convection in a homogeneous porous layer bounded above and below by isothermal surfaces is extended to consider an upper boundary which is partly permeable. The general boundary condition p + λ ∂p/∂n = constant is applied at the top surface and the flow is investigated for various λ in the range 0 ⩽ λ < ∞. Estimates of the magnitude and horizontal distribution of the vertical mass and heat fluxes at the surface, the horizontally-averaged heat flux (Nusselt number) and the fraction of the fluid which recirculates within the layer are found for slightly supercritical conditions. Comparisons are made with the two limiting cases λ → ∞, where the surface is completely impermeable, and λ = 0, where the surface is at constant pressure. Also studied are the effects of anisotropy in permeability, ξ = K _H /K _V , and anisotropy is thermal conductivity, η = k _H /k _V , both parameters being ratios of horizontal to vertical quantities. Quantitative results are given for a wide variety of the parameters λ, ξ and η. In the limit ξ/η → 0 there is no recirculation, all fluid being converted out of the top surface, while in the limit ξ/η → ∞ there is full recirculation.     

Parameter Dependence of Stable Manifolds for Delay Equations with Polynomial Dichotomies

We establish the existence of Lipschitz stable invariant manifolds for semiflows generated by a delay equation x′ = L(t)x _t + f (t, x _t , λ), assuming that the linear equation x′ = L(t)x _t admits a polynomial dichotomy and that f is a sufficiently small Lipschitz perturbation. Moreover, we show that the stable invariant manifolds are Lipschitz in the parameter λ. We also consider the general case of nonuniform polynomial dichotomies.     

Non-linear rheology and flow-induced transition of a lamellar-to-vesicle phase in ternary systems of alkyldimethyl oxide/alcohol/water

 The time-dependent transformation of an ionically charged lamellar phase (L _α-phase) into a vesicle phase under the influence of shear is investigated using rheological and conductivity measurements. The L _α-phase consists of the zwitterionic surfactant tetradecyldimethylaminoxide (C₁₄DMAO), hexanol, oxalic acid and water. The experiments were carried out on the L _α-phase in a well defined state. It was prepared by a special route from the neighbouring L ₃-phase that consists of 100 mM C₁₄DMAO, 250 mM hexanol and 5 mM oxalicdiethylester (OEE). The OEE hydrolyses in the L ₃ -phase to oxalic acid and ethanol. The result is a virgin L _α-phase which consists of stacked bilayers and which has not been exposed to shear. When this low-viscous phase is subjected to shear it is transformed into a highly viscous vesicle phase. The transformation of the L _α-phase into vesicles under constant shear was monitored by recording the viscosity and conductivity with time. It is observed that at least three different time constants can be distinguished in the transformation process. The conductivity passes through a minimum (τ₁) in the direction of shear. The viscosity first passes through a minimum (τ₂) and then over a maximum (τ₃). It is concluded that τ₁ belongs to the complete alignment of the bilayer parallel to the wall, τ₂ to the beginning of the break-up of the bilayers to the vesicles and τ₃ to the complete transformation of the L _α- to the vesicle phase. When the shear rate was varied, it was noted that the product of the time constants and shear is constant. Received: 30 June 1999/Accepted: 30 August 1999     

A proof of the Benjamin-Feir instability

The existence and linear stability problem for the Stokes periodic wavetrain on fluids of finite depth is formulated in terms of the spatial and temporal Hamiltonian structure of the water-wave problem. A proof, within the Hamiltonian framework, of instability of the Stokes periodic wavetrain is presented. A Hamiltonian center-manifold analysis reduces the linear stability problem to an ordinary differential eigenvalue problem on ℝ⁴. A projection of the reduced stability problem onto the tangent space of the 2-manifold of periodic Stokes waves is used to prove the existence of a dispersion relation Λ(λ,σ, I ₁, I ₂)=0 where λ ε ℂ is the stability exponent for the Stokes wave with amplitude I ₁ and mass flux I ₂ and σ is the “sideband’ or spatial exponent. A rigorous analysis of the dispersion relation proves the result, first discovered in the 1960's, that the Stokes gravity wavetrain of sufficiently small amplitude is unstable for F ε (0,F₀) where F ₀ ≈ 0.8 and F is the Froude number.     

Thermorheological properties near the glass transition of oligomeric poly(methyl methacrylate) blended with acrylic polyhedral oligomeric silsesquioxane nanocages

Two distinct oligomeric species of similar mass and chemical functionality (M _w≈2,000 g/mol), one a linear methyl methacrylate oligomer (radius of gyration R _g≈1.1 nm) and the other a hybrid organic–inorganic polyhedral silsesquioxane nanocage (methacryl-POSS, r≈1.0 nm), were subjected to thermal and rheological tests to compare the behaviors of these geometrically dissimilar molecules over the entire composition range. The glass transition temperatures of the blends varied monotonically between the glass transition temperatures of the pure oligomer (T _g=−47.3°C) and the pure POSS (T _g=−61.0°C). Blends containing high POSS contents (with volume fraction φ _POSS≥0.90) exhibited enhanced enthalpy relaxation in differential scanning calorimetry (DSC) measurements, and the degree of enthalpy relaxation was used to calculate the kinetic fragility indices m of the oligomeric MMA (m=59) and the POSS (m=74). The temperature dependences of the viscosities were fitted by the free-volume based Williams–Landel–Ferry (WLF) and Vogel–Fulcher–Tammann (VFT) framework and a dynamic scaling relation. The calculated values of the fragility from the WLF–VFT fits were similar for the POSS (m=82) and for the oligomer (m=76), and the dynamic scaling exponent was similar for the oligomeric MMA and the POSS. Within the range of known fragilities for glass-forming liquids, the temperature dependence of the viscosity was found to be similarly fragile for the two species. The difference in shape of the nanocages and oligomer chains is unimportant in controlling the glass-forming properties of the blends at low volume fractions (φ _POSS<0.20). However, at higher volume fractions, adjacent POSS cages begin to crowd each other, leading to an increase in the fractional free volume at the glass transition temperature and the observed enhanced enthalpy relaxation in DSC.     

Detonation diffraction through different geometries

We performed the study of the diffraction of a self-sustained detonation from a cylindrical tube (of inner diameter d) through different geometric configurations in order to characterise the transmission processes and to quantify the transmission criteria to the reception chamber. For the diffraction from a tube to the open space the transmission criteria is expressed by d _c  = k _c ·λ (with λ the detonation cell size and k _c depending on the mixture and on the operture configuration, classically 13 for alkane mixtures with oxygen). The studied geometries are: (a) a sharp increase of diameter (D/d > 1) with and without a central obstacle in the diffracting section, (b) a conical divergent with a central obstacle in the diffracting section and (c) an inversed intermediate one end closed tube insuring a double reflection before a final diffraction between the initiator tube and the reception chamber. The results for case A show that the reinitiation process depends on the ratio d/λ. For ratios below k _c the re-ignition takes place at the receptor tube wall and at a fixed distance from the step, i.e. closely after the diffracted shock reflection shows a Mach stem configuration. For ratios below a limit ratio k _lim (which depends on D/d) the re-ignition distance increases with the decrease of d/λ. For both case A and B the introduction of a central obstacle (of blockage ratio BR = 0.5) at the exit of the initiator tube decreases the critical transmission ratio k _c by 50%. The results in configuration C show that the re-ignition process depends both on d/λ and the geometric conditions. Optimal configuration is found that provides the transmission through the two successive reflections (from d = 26 mm to D _ch = 200 mm) at as small d/λ as 2.2 whatever the intermediate diameter D is. This configuration provides a significant improvement in the detonation transmission conditions.

---

This paper was based on work that was partly presented at the International Conference on Combustion and Detonation, Zel’dovich Memorial II, Moscow, Russia, 30 August–3 September 2004, and at the 20th International Colloquium on the Dynamics of Explosions and Reactive systems, Montreal, Canada, 31 July–5 August 2005.     

The criterion of the existence or inexistence of transverse shock wave at wedge supported oblique detonation wave

A simplified theoretic method and numerical simulations were carried out to investigate the characterization of propagation of transverse shock wave at wedge supported oblique detonation wave.After solution validation,a criterion which is associated with the ratio Φ (u 2 /u CJ) of existence or inexistence of the transverse shock wave at the region of the primary triple was deduced systematically by 38 cases.It is observed that for abrupt oblique shock wave (OSW)/oblique detonation wave (ODW) transition,a transverse shock wave is generated at the region of the primary triple when Φ < 1,however,such a transverse shock wave does not take place for the smooth OSW/ODW transition when Φ > 1.The parameter Φ can be expressed as the Mach number behind the ODW front for stable CJ detonation.When 0.9 < Φ < 1.0,the reflected shock wave can pass across the contact discontinuity and interact with transverse waves which are originating from the ODW front.When 0.8 < Φ < 0.9,the reflected shock wave can not pass across the contact discontinuity and only reflects at the contact discontinuity.The condition (0.8 < Φ < 0.9) agrees well with the ratio (D ave /D CJ) in the critical detonation.     

Equivalence between Rank-One Convexity and Polyconvexity for Some Classes of Elastic Materials

Let W(F) = φ(λ ₁ ^s + λ ₂ ^s + λ ₃ ^s ) + ψ(λ ₁ ^r λ ₂ ^r + λ ₁ ^r λ ₃ ^r + λ ₂ ^r λ ₃ ^r ) + f(λ ₁ λ ₂ λ ₃) be a stored energy function. We prove that, for this function, rank-one convexity is equivalent to polyconvexity.under suitable assumptions on φ, ψ and f.     

Breakdown in a boundary layer exposed to free-stream turbulence

The natural secondary breakdown modes of a flat plate boundary layer exposed to a free-stream turbulence level of 6.7% are characterised by experimental visualisations. The used experimental set-up is a water channel and no external triggering of the instabilities is used. The visualisations show the presence of two secondary instability modes, a sinuous (antisymmetric) and a varicose (symmetric) mode. The amplitudes of both instabilities increase, according to a non-linear profile, in upstream direction. The amplitudes also experience temporal growth. When a critical amplitude is reached, roll-up structures appear. These structures develop and finally merge, resulting in a turbulent spot. The development of the amplitude is similar in both modes. However, the amplitude of the varicose instability varies between 0.38 and 0.63 [2 π A/ λ] and the sinuous amplitude between 0.11 and 0.23 [2 π A/ λ]. The propagation velocities of the sinuous and varicose instabilities are 0.91 [u/u _blas] and 0.87 [u/u _blas] respectively. This shows that both modes are located in a low speed streak.     

Investigation of detonation initiation in aluminium suspensions

Detonation initiation is investigated in aluminium/oxygen and aluminium/air mixtures. Critical conditions for initiation of spherical detonations are examined in analogy with the criteria defined for gaseous mixtures, which correlate critical parameters of detonation initiation to the characteristic size of the cellular structure. However, experimental data on the detonation cell size in these two-phase mixtures are very scarce, on account of the difficulty to perform large-scale experiments. Therefore, 2D numerical simulations of the detonation cellular structure have been undertaken, with the same combustion model for Al/air and Al/O₂ mixtures. The cell size is found to be λ = 37.5 cm for a rich (r = 1.61) aluminium–air mixture, and λ = 7.5 cm for a stoichiometric aluminium-oxygen mixture, which is in reasonable agreement with available experimental data. Calculations performed in large-scale configurations (up to 25 m in length and 1.5 m in lateral direction) suggest that the critical initiation energy and predetonation radius for direct initiation of the unconfined detonation in the aluminium–air mixture are, respectively, 10 kg of TNT and 8 m. Moreover, numerical simulations reveal that the structure of the detonation wave behind the leading front is even more complicated than in pure gaseous mixtures, due to two-phase flow effects. This paper is based on work that was presented at the 21st International Colloquium on the Dynamics of Explosions and Reactive Systems, Poitiers, France, July 23–27, 2007.     

Behaviour of the Reynolds stress on rough walls

Quadrant analysis is used to study the contributions, associated to the four quadrants of the (u,v) plane, to the production of the turbulent shear stress on rough walls. The measurements are described for a fully developed turbulent flow between two rough plates with varying the parameterλ _z(span/height ratio of roughness elements). The application of this technique indicates that the (Q ₂) events (ejections) and (Q ₄) events (sweeps) cause an intense production of the Reynolds stress —ρ as compared with the (Q ₁) and (Q ₃) quadrants. The (Q ₂) contribution to the Reynolds stress depends on the geometry factorλ _z. Variation of the parameterλ _zaffects the distributions of and forH⩽3, whereH is a particular threshold. Comparison with boundary layer flow shows that the region 0.2⩽y/h⩽0.7 is characterized by a Reynolds stress production, independent of the flow nature. The third moment of the longitudinal velocity fluctuations is found to be sensitive to the surface roughness.