• <tr id="yyy80"></tr>
  • <sup id="yyy80"></sup>
  • <tfoot id="yyy80"><noscript id="yyy80"></noscript></tfoot>
  • 99热精品在线国产_美女午夜性视频免费_国产精品国产高清国产av_av欧美777_自拍偷自拍亚洲精品老妇_亚洲熟女精品中文字幕_www日本黄色视频网_国产精品野战在线观看 ?

    Influence of quenching medium on the dendrite morphology,hardness,and tribological behaviour of cast Cu-Ni-Sn spinodal alloy for defence application

    2023-12-27 04:09:52BipinSankarKarthikShankarVamuSunilHemanthKashyapNikhilNairAdarshNairAbhinav
    Defence Technology 2023年12期

    Bipin Sankar,Karthik V Shankar,Vamu Sunil,Hemanth Kashyap S,Nikhil Nair,Adarsh A.Nair,Abhinav P M

    Department of Mechanical Engineering, Amrita Vishwa Vidyapeetham, Amritapuri, India

    Keywords:Quenching Water Brine SAE 40 oil Grain size Hardness Wear rate Spinodal decomposition CuSn Spinodal alloy CuNiSn

    ABSTRACT Cu-Ni-Sn spinodal alloys (Spinodal bronze) are potential materials with robust applications in components associated with defence applications like bearings,propellers,bushes,and shafts of heavily loaded aircraft,off-road vehicles,and warships.This paper presents a comparative study using water,Brine solution,and SAE 40 oil as the quenching media in regular bronze (Cu-6Sn) and spinodal bronze(Cu-9Ni-6Sn) alloys.Morphological analysis was conducted by optical microscopy,transmission electron microscopy (TEM),and X-ray diffraction technique (XRD) on bronze and spinodal bronze samples immersed in the three different quenching media to understand the grain size and hardness values better.Tribological analysis was performed to analyze the effect of quenching media on the wear aspects of bronze and spinodal bronze samples.The hardness value of the brine-aged spinodal bronze samples was as high as 320 Hv,and the grain size was very low in the range of 60 μm.A quantitative comparison between brine-aged regular bronze and brine-aged spinodal bronze showed that the hardness(Hv) was almost 80%higher for brine-aged spinodal bronze.Further,the grain size was approximately 30%finer for spinodal bronze when compared with regular bronze.When the load was increased in spinodal bronze samples,there was an initial dip in wear rate followed by a marginal increase.There was a steady increase in friction coefficient with a rise in load for brine-aged regular bronze and spinodal bronze samples.These results indicate that brine solution is the most effective quenching medium for cast Cu-Ni-Sn spinodal alloys.

    1.Introduction

    Copper and its alloys have a perfect blend of mechanical,electrical,and tribological properties like strength,conductivity,machinability,and corrosion resistance,making it an apt and automatic choice for industrial applications.Amongst the copper alloys,bronze,a cost-effective alloy of copper and tin(Cu-Sn),has high electrical and thermal conductivity,better mechanical strength values like tensile strength and yield strength,high dimensional precision,and is resistant to corrosion as well.This makes bronze suitable for manufacturing components like bushings,bearings,and propellers related to automobile,aircraft,marine,and defence applications.Studies have also shown that Cu-Sn has been employed as anti-wear and friction-reducing material.However,bronze is susceptible to hazards like bronze disease,a green film formation on the surface when exposed to moist air,which can adversely affect the above-mentioned properties.Copper beryllium (Cu-Be) alloys are another category with prospective industrial applications,but these alloys also possess health hazards.

    Cu-Ni-Sn spinodal alloys developed by the Bell Laboratory have evolved as a potential and prospective alternative to Cu-Be alloys.The mechanical properties such as tensile strength,yield strength,ductility,and hardness that Cu-Ni-Sn alloys,strengthened by spinodal decomposition,are equivalent to Cu-Be alloys[1-4].There are toxic and hazardous challenges associated with Cu-Be alloys,unlike in Cu-Ni-Sn alloys [3].Cu-Ni-Sn alloys are highly tolerant and resistant to corrosion with marginal friction and wear characteristics and high toughness values.Bearings used in defence appliances demand lighter weight,high robustness,superior strength,more resilience,and resistance to damage,and the spinodal bronze alloys satisfy these criteria positively [3-5].Numerous significant studies have already been done to analyze and assess Cu-Ni-Sn spinodal alloys' morphological and mechanical characteristics [6-8].A conventionally wrought spinodal hardened Cu-Ni-Sn alloy has a superior yield strength of 758 MPa compared to the commonly used aluminium bronze with a yield strength of 621 MPa,indicating that weight can be lowered for components used in the aerospace and automobile industry by employing Cu-Ni-Sn alloys [9].Commercialising wrought Cu-Ni-Sn spinodal alloys had drawbacks due to the formation of hot cracks during rolling and forging operations.Casting and powder metallurgy then evolved as a potential alternative for manufacturing Cu-Ni-Sn spinodal as it eliminated hot crack formation to an extent enabling the fabrication of more significant section components.Microstructural analysis of the copper-based spinodal alloys is more invasive and definite using Transmission Electron Microscopy (TEM) in comparison to optical microscopy(OM) and Scanning Electron Microscopy (SEM) as the modulated spinodal structure in the Nanoscale making visibility complex.

    Studies analysing the tribological characteristics of Cu-Ni-Sn spinodal alloys have been limited [10-12].Ilangovan et al.[12]conducted a tribological study on sand-cast Cu-7Ni-4Sn spinodal bronze.It was confirmed that specific wear rate(SWR)was mainly influenced by ageing time,sliding velocity,and load applied.Singh et al.[13,14] has investigated the dry sliding wear behaviour and microstructural analysis of Cu-Ni-Sn spinodal alloy against SS 440C steel.When these 2 specimens slide against each other,a transfer layer possessing Nano crystalline grains is formed at the interface.The authors have found that this transfer layer is the primary source of wear debris and contains oxides.A Cu-15Ni-8Sn spinodal alloy was considered for the investigation.When a load of 980 N was applied at a sliding velocity of 0.25 m/s in the Argon atmosphere,the coefficient of friction (COF) was 0.34,and when the sliding velocity decreased to 0.10 m/s,c.o.f reduced to 0.294.These values indicate a reduction in wear rate with respect to the decrease in sliding velocity.Further,the authors confirmed that the wear's severity would be higher whenever the load is high under dry conditions [13,14].

    Defence components often demand high strength and high conductivity.The cu-9Ni-6Sn alloy was chosen for this work because for these reasons.Cu-9Ni-6Sn spinodal possesses strength as high as 1200 MPa and conductivity of 8.4% IACS.Appropriate thermo mechanical processing enables the tensile strength of Cu-9Ni-6Sn to be adjusted as well.Cu-9Ni-6Sn alloy also possesses a very high Rockwell hardness value of 30 HRC[15].These characteristics make Cu-9Ni-6Sn alloy a suitable choice in defence applications.

    The typical heat treatment procedure broadly involves homogenisation,solutionising,and ageing.Quenching is another critical intermediate process in which the alloy specimen is rapidly cooled to maintain and enhance the mechanical properties and the phase distribution.Some commonly employed quenching media are water,brine solution,SAE40,and paraffin [16,17].The quenching media should be selected to minimise residual stresses formed during phase transformation [18].Using water and oil as quenchants has its merits and demerits.A martensitic structure is formed when water is employed as the quenching medium.Still,the additional tempering operation must be performed to get the desired properties,which incurs more cost.In contrast,oil as a quenchant leads to adverse mechanical properties due to insufficient cooling rates[18].These limitations have led to trials of brine solution,polymeric solutions,and SAE oils as prospective quenching media.

    The role of quenching media in the tribological behaviour of cast bronze and spinodal bronze (Cu-Ni-Sn) alloys is still to be explored and investigated.This investigation aims to conduct a tribological analysis of cast Cu-6Sn and Cu-9Ni-6Sn alloy subjected to quenching in 3 different media.The three quenching media are water,brine solution,and SAE 40 oil.Subsequently,the effect of load on wear rate (WR) and coefficient of friction (COF),sliding velocity on WR and COF,and the effect of quenching media on WR and COF are studied.The hardness value corresponding to the three quenching mediums was also determined.

    2.Formulation of work

    2.1.Material selection procedure

    The toxic properties and high cost associated with Cu-Be alloys lead to the present work's choice of Cu-Sn and Cu-Ni-Sn spinodal alloys.Cu-Sn alloys(Bronze)are abundant in availability and nontoxic in nature.These bronze alloys possess high thermal conductivity,better corrosion resistance,and excellent elasticity.However,the Cu-Sn alloys fabricated through continuous casting feature defects like diminished surface quality,porosity,and inverse segregation [19,20].The Cu-Ni-Sn spinodal alloy has superior strength,toughness,conductivity values,and less cost than Cu-Be alloys.The selection of copper ternary alloy composition was based on the spinodal decomposition perspective and the weight percentage of tin and nickel(Sn from 2 to 8 wt%,Ni from 2 to 20 wt%)[19,20].The alloy sample composition of Cu-6Sn and Cu-9Ni-6Sn was preferred for this work.

    2.2.Preparation of Cu-Sn and Cu-Ni-Sn alloy by casting

    Permanent mould casting (PMC) was employed for fabricating Cu-6Sn and Cu-9Ni-6Sn alloy.The alloy composition was melted in a clay graphite crucible with an industrial electric furnace.PMC is identical to sand casting but utilises permanent mould with high thermal conductivity,enabling repeated use and thermal cycling.The higher the thermal conductivity,the higher the heat transfer rate and the lesser the time for solidification.As shown in Fig.1(a),the industrial electric furnace was maintained at 1250°C following the Cu-Ni-Sn ternary phase diagram[12].Fig.1(b)and 1(c)depicts the casting sequence,and Fig.1(c) shows the cast rods for experimental investigation.Cast Cu-Sn and Cu-Ni-Sn rods of the desired dimensionsφ36 mm×156 mm in length were obtained by pouring the molten metal into the permanent mould coated with LUBRIKOTE die coat material.The cast and machined Cu-6Sn,and Cu-9Ni-6Sn rods are shown in Fig.2.After the metal casting process,the developed alloys were subjected to spectrochemical analysis using optical spectroscopy (SPAS-02 spectrometer).Four samples were tested from each alloy,and an average value was calculated.The spectroscopy results of the developed Cu-6Sn and Cu-9Ni-6Sn alloys are reported in Table 1.The authors observed from Table 1 that the composition of the developed alloys was found to be within the variation of ±0.2 wt%-0.3 wt% from the original composition.

    Table 1 Average spectroscopy results of developed alloys.

    Fig.1.Image showing: (a) Industrial melting furnace;(b),(c) Sequence of casting;(d) Cast rods.

    Fig.2.Image showing: (a),(b) Cast and machined rods;(c) Heat treatment process.

    2.3.Heat treatment process

    The cast Cu-6Sn and Cu-9Ni-6Sn rods were subjected to various stages of heat treatment commencing with homogenisation,solutionising,quenching,and finally ageing.The solutionising process was done at 825°C for 10 h[21],followed by a quenching or rapid cooling to maintain the α phase in the supersaturated condition.The quenching media were water,SAE 40 oil,and brine solution.The water quenching was done after the specimens were heated to 825°C and immersed in a water bath of 30°C for 10 min.The heated specimens were immersed in a brine solution of 20°C for 10 min for brine quenching.The heat-treated specimens were oil quenched at room temperature for 10 min.The quenched rods were then aged at 350°C for 3 h.Fig.2(c) depicts a schematic diagram of the heat treatment process followed in this present investigation.The heat treatment process,as well as the cast rod fabrication,took place under inert argon conditions.Fig.3 depicts the experimental methodology followed in this present investigation.Advanced instrumental characterisation (XRD and TEM)techniques were used to characterise the developed Cu-6Sn and Cu-9Ni-6Sn alloys.Table 2 shows the heat treatment conditions of the produced alloys.

    Table 2 Various conditions of samples used.

    Fig.3.Experimental methodology.

    2.4.Metallurgical analysis

    The morphological analysis was conducted using Carl Zeiss optical microscope.The permanent mould-cast alloy specimens of Cu-6Sn and Cu-9Ni-6Sn were ground using a bench grinder.Subsequently,burrs were removed from the samples using a polishing machine.Emery sheets of various grades (coarse to fine grades A46,A54,A70,A80,A90,and A100) were utilised to finish the alloy specimens further.The final alloy specimens equipped for microstructural analysis were done in alumina using a disc polisher that smoothened the specimen surface set for microstructure analysis.X-ray diffraction(XRD)was conducted using a Philips PW 1710 diffractometer with a graphite-diffracted beam monochromator and a copper anode tube.Side bands were observed using step scan mode.The XRD analysis was conducted at 40 kV voltage with a current of 40 mA.The scan rate was set as 1°·min-1.The phase identification was carried out using XRD with Cu Kβ radiation.The TEM analysis was performed on the cast Cu-6Sn regular bronze and Cu-9Ni-6Sn spinodal alloy.Specimen discs of the spinodal bronze alloy diameter 3 mm were prepared for transmission electron microscopy through standard jet polishing in a solution containing 30% nitric acid and 70% methanol at approximately -30°C.A beam of electrons was then passed through the disc specimen casing,an interaction between electrons and discs.Images formed from these interactions were magnified by an objective lens that appeared on the fluorescent screen.A Philips 400 T analytical microscope operating at 125 V was employed for the TEM analysis,including microstructural imaging and selected area diffraction patterns (SADPs).

    2.5.Conduct of hardness test

    A Vickers microhardness tester (Mitutoyo Model: HM-210A)was employed to determine the microhardness of the cast alloy specimens.The hardness tester provided a diamond indenter to create micro-level indentations on the alloy specimens.The indentation depth was analysed using a combination of LED illumination and an objective lens.The hardness tester was loaded with 50 gf for a dwell time of approximately 15 s.The alloy specimen surfaces were polished by emery sheets of 2000-5000 grit size.As and when the diamond indenter touches the specimen surface,measurement was taken of the indentation using a microscope.Four test trials were conducted across five locations to measure the hardness.The average of these readings was taken as the hardness value of the specimens in terms of HV.

    2.6.Conduct of tribological test

    A pin-on-disc apparatus (DUCOM,TR-20NEO) was used for the tribological test,as shown in Fig.4.The test was conducted as per the ASTM G-99 standard,and the alloy specimens were machined toφ12 mm×35 mm length each.An electronic weighing machine of up to 0.1 mg was used to weigh the specimens.The steel disc of the apparatus (EN31 hardened steel) was maintained at constant RPM using a controller.The alloy specimen was kept against the disc jammed between the jaws,and the inclination effect was negotiated through uniform screwing.The weights were gradually added to the machine to ensure constant loading to the arm.During the test,lubricants were not used to maintain a dry condition.Consistent contact between the alloy specimens and the disc was ensured through the applied load when wear occurred.The procedure was tested with various wear testing parameters,as shown in Table 3.The wear test for all the conditions was performed thrice,and the average value was calculated.

    The worn-out surface was characterised using field emission scanning electron microscopy(FESEM)and energy-dispersive X-ray spectroscopy(EDS).FESEM and EDS analysis were also employed to determine the elemental composition of the tribo layer.

    3.Results and discussions

    3.1.Metallurgical analysis of Cu-6Sn and Cu-9Ni-6Sn spinodal bronze

    The microstructure of alloys primarily depends on three factors.These include temperature,time of ageing,and deformation before or during heat treatment [22].Optical microscopy (OM) was employed to study the microstructure of regular bronze Cu-6Sn and spinodal bronze Cu-9Ni-6Sn.The as-cast structures shown in Fig.5(a) and (b) indicate the presence of dendrites in traditional bronze and spinodal bronze,respectively.Besides Cu-6Sn,previous research on Cu-10Sn cast alloys also had a dendritic structure in the as-cast form [20,22].Further,an investigation conducted by Pal et al.[23]indicates the presence of sharp grain boundaries and subgrains with wider grain boundaries on the optical microstructure of cast Cu-15Ni-2.7 Sn alloy and Cu-15Ni-9Sn alloy,respectively.

    Fig.5.Optical image of dendritic microstructure in as-cast condition for (a) Regular Cu-6Sn alloy;(b) Cu-9Ni-6Sn alloy.

    Upon solutionising,these dendritic structures disappeared from Cu-6Sn and Cu-9Ni-6Sn,as shown in Fig.6(a)and 6(b),to form a single-phase solid solution.It was reported that Cu-10Sn alloy,upon homogenisation and solutionising,possessed a microstructure without dendrites [23].Further,Chieu et al.[22],based on investigating a homogenised and solutionized Cu-9Ni-6Sn alloy,confirmed that the optical microstructure is a single-phase solid solution.

    Fig.6.Optical image of solutionised microstructure of (a) Regular Cu-6Sn alloy;(b) Cu-9Ni-6Sn alloy.

    An as-quenched microstructure is significant for an alloy because it influences properties like fatigue resistance and toughness in the as-tempered stage.In the case of copper alloys,when the quenching medium used is water,Cu alloys become softer.Water-quenched copper alloys may possess a fine-grain structure.Park et al.[24] observed that bronze alloys with 15 wt%to 24 wt% tin content is water quenched at 700°C;the optical microstructure possesses β and β martensite phases having a needle-like structure.The subsequent reduction in tin content to 15 wt% further diminished the presence of the martensite phase.Based on the combined inference from this pattern and optical microstructure in Fig.7(a),it may be concluded that as-quenched(water) Cu-6Sn alloys will possess a negligible martensite phase.Fig.7(b) also indicates uniformity in microstructure for waterquenched Cu-9Ni-6Sn alloy.

    Fig.7.Optical microstructure of (a) Cu-6Sn quenched in water;(b) Cu-9Ni-6Sn quenched in water;(c) Cu-6Sn quenched in oil;(d) Cu-9Ni-6Sn quenched in oil;(e) Cu-6Sn quenched in brine;(f) Cu-9Ni-6Sn quenched in brine.

    Oil quenching does not cause many distortions to the specimen as water quenching.But the rate of cooling is relatively slower.Oilquenched samples may exhibit peak hardness during the initial stages of aging[25].Fig.7(c)and 7(d)indicate a minimal uniformity in the microstructure with a more refined grain structure for SAE 40 oil-quenched Cu-6Sn and Cu-9Ni-6Sn alloys.Unlike water quenching,Brine quenching has minimal distortion effect on the specimen.Fig.7(e)and 7(f)represent a fine-grained microstructure for both brine-quenched Cu-6Sn and Cu-9Ni-6Sn alloys.

    After homogenisation and quenching treatment,the Cu-6Sn and Cu-9Ni-6Sn samples were subjected to the ageing process.Ageing is a heat treatment process to enhance strength and hardness.Fig.8(a) indicates that dendrites may resurface on waterquenched bronze upon ageing.This resurfacing of dendrites may be attributed to the temperature gradient formed during solidification.It also shows that some precipitates are formed along the grain boundary.Coarse grains can be seen on oil-quenched and brinequenched bronze upon ageing,as shown in Fig.8(c) and 8(e).After the homogenisation and solutionising process,the microstructure of a Cu-Ni-Sn spinodal alloy will possess an α phase with a supersaturated solid solution.Upon ageing,the cellular structure gradually develops in the microstructure of Cu-Ni-Sn spinodal alloys with equilibrium α and γ phases at the grain boundaries.Spinodal decomposition forms a modulated structure with metastable γ phase having DO22structure formed at the tin-rich region[10,11].The more the ageing time,the more the cellular network in the microstructure.However,such a cellular structure can hardly be seen in the optical microstructures of Cu-9Ni-6Sn aged alloy shown in Fig.8(b),8(d)and 8(f).This leads to doing a TEM analysis to interpret the microstructure better.

    Fig.8.Optical microstructure of the aged sample of (a) water-quenched Cu-6Sn alloy;(b) water-quenched Cu-9Ni-6Sn alloy;(c) Oil quenched Cu-6Sn alloy;(d) Oil quenched Cu-9Ni-6Sn alloy;(e) Brine Cu-6Sn quenched alloy;(f) Brine Cu-9Ni-6Sn quenched alloy.

    3.2.Grain size measurement

    Table 4 and Fig.9 show the grain size value of cast Cu-6Sn and Cu-9Ni-6Sn spinodal alloy.It was observed that the brine quenched and brine aged spinodal bronze had the smallest grain size in the range of 75 μm and 60 μm,respectively.The cooling rate is a crucial factor that influences the grain size of the alloy.The faster the cooling rate,the smaller the grain size and the number of precipitated phases in the matrix.Among the three quenching media chosen,brine had the fastest cooling rate,which was the principal reason for the smaller grain size.The higher cooling rate of the brine solution may be attributed to the high concentration of NaCl (salt) which possess a wide operating temperature range.

    Table 4 Grain size values of developed alloys.

    Fig.9.Comparison of grain size for different samples.

    3.3.TEM analysis of Cu-9Ni-6Sn spinodal alloy in quenched and aged condition

    3.3.1.Brine quenched Cu-9Ni-6Sn alloy

    Transmission electron microscopy(TEM)was conducted on the Cu-9Ni-6Sn alloy brine quenched specimen.TEM images in Fig.10 indicate bright,dark,and even grey regions.The bright regions possessed the solute-depleted β phase,whereas the grey-colored regions contained the super saturated β phase,with a facecentred cubic (FCC) crystal structure.The TEM images in Fig.10 did not indicate discontinuous precipitation (DP),as DP colonies could hardly be seen.Fig.10(a) and 10(b) show the presence of Shockley partial dislocations capable of gliding divided by faulted regions which indicate that the stacking fault energy of Cu-9Ni-6Sn alloy may be lower than pure copper.This dip in stacking fault energy may be attributed to adding tin.The scratch line marks shown in Fig.10(c) point toward the twinning dislocations predominant in copper alloys.Twinning is an orientation-dependent deformation that could be in the size of nanometers,observable primarily through TEM only.It can also be inferred that complete decomposition of the solid solution may have taken place in the as-quenched stage since no structural features can be seen in the TEM micrographs shown in Fig.10.The Selected area electron diffraction(SAED)pattern of brine quenched spinodal bronze alloy is shown in Fig.11(a) and 11(b).Fig.11(a)depicts the pattern of Cu,and Fig.11(b)shows the pattern for β-Cu,which infers a relatively high concentration of tin.It was observed that the β-Cu phase decreased significantly upon ageing.

    Fig.10.TEM micrograph of brine quenched Cu-9Ni-6Sn spinodal alloy.

    Fig.11.SAED pattern for (a) Cu;(b) β -Cu;in brine quenched Cu-9Ni-6Sn spinodal alloy.

    3.3.2.TEM analysis of brine-aged Cu-9Ni-6Sn alloy

    Fig.12(a)-12(d) shows the TEM image of brine-aged Cu-9Ni-6Sn spinodal alloy.The image clearly indicates that the significant characteristic of the brine-aged Cu-9Ni-6Sn alloy is the modulated microstructure.It can be further inferred that there is a homogeneous distribution in the matrix,which indicates that nucleation will happen neither at the grain boundaries nor at the dislocations The combined factors of modulated microstructure,homogeneity and absence of nucleation barrier confirm the occurrence of spinodal decomposition.The bright regions shown in Fig.12(a) and 12(b) correspond to the modulated tin-rich region,whereas the dark regions shown in Fig.12(c)correspond to the tinlean regions.The metastable discontinuous γ phase cannot be seen in any TEM micrographs.Further,the bright spots in Fig.12(a)and 12(b) corresponds to the DO22structure and the grey spots corresponds to overlapped L12structure.The presence of both DO22and L12structures points to spinodal decomposition.Besides,the dark regions in Fig.12(c) indicate the presence of an intermetallic compound,Ni3Sn.The SAED patterns in Fig.13(a)and 13(b)confirm the perfect copper and Ni3Sn intermetallic phase for the aged spinodal bronze alloy.Spinodal decomposition is further authenticated by spinodal satellites on the TEM film by short time exposure(for almost 15 s),as indicated in Fig.13(b).The bright dots in Fig.13(b) represent the reflections of the DO22ordered structure.The TEM analysis conducted in the present investigation was in accordance with the studies of Zhao and Notis (1998).

    Fig.12.TEM images of aged Cu-9Ni-6Sn alloy quenched in brine.

    Fig.13.SAED pattern for (a) Cu;(b) Ni3Sn in aged Cu-9Ni-6Sn spinodal alloy quenched in brine.

    The SAED pattern in Fig.13(b)confirms the intermetallic phase Ni3Sn.Ni3Sn phase develops as a plane layer with fragile column grain,with its grain size only half the grain size of Ni3Sn2.The interfacial properties between Ni3Sn and Ni are significant in optimising the crack to avoid crack formation.An ionic bond in the Ni3Sn phase possesses a cubic structure.Studies have shown that Ni/Ni3Sn interface has very high bonding strength with strong resistance to crack formation [26].

    3.3.3.XRD analysis

    XRD pattern observed in regular bronze(Cu-6Sn)and spinodal bronze(Cu-9Ni-6Sn) is illustrated in Fig.14(a) and 14(b),respectively.Spinodal bronze is distinct from regular bronze because of the Ni3Sn intermediate phase,which might be in the form of elongated precipitates.The formation of the Ni3Sn phase may be attributed to the higher composition of Nickel in the alloy compared to Sn.When the concentration of Ni is high,the thickness of the Ni3Sn layer can also increase [27].There is a positive correlation between the growth of the Ni3Sn phase and the diffusion of Ni.Further,the addition of Nickel can increase the crystallisation temperature of Cu-Ni-Sn alloys,which act as a catalyst for forming the intermediate Ni3Sn phase.Besides,Nickel has a more substantial influence on the tin when compared with copper,fueling the diffusion of intermetallic phases like Ni3Sn in Cu-Ni-Sn alloys.Therefore,the principal reason behind the formation of the Ni3Sn phase is the diffusion process of Ni in spinodal bronze.

    Fig.14.XRD analysis of (a) regular bronze (b) spinodal bronze.

    3.4.Hardness test

    The hardness value of regular bronze (Cu-6Sn) and spinodal bronze (Cu-9Ni-6Sn) will be on the higher side due to solid solution(SS)strengthening and spinodal decomposition,respectively.Spinodal decomposition impacts properties such as the hardness and yield strength of the copper alloys due to the formation of DO22and L12-ordered structures.The modulated microstructure formed from spinodal decomposition is one of the principal reasons behind the high hardness of spinodal bronze.The hardness values of regular bronze and spinodal bronze are shown in Fig.15 and Table 5,and it was observed that brine-quenched and brine-aged Cu-9Ni-6Sn specimens had the highest hardness.This is mainly attributed to the fact that grain size is the smallest for brinequenched specimens due to its rapid cooling rate.Since the grain size is smaller,the grain boundaries formed are more,which increases hardness.Further,the intermetallic compound Ni3Sn formed could also be a reason for improved hardness.Since the grain size of the Ni3Sn phase is finer,and there is a dominant grain boundary diffusion,the hardness of the brine quenched and aged Cu-9Ni-6Sn increases.

    Table 5 Hardness values of developed alloys.

    Fig.15.Hardness values of various specimens.

    3.5.Tribological analysis

    The tribological study is significant in the context of energy management,reliability,and life of a component,making it a pivotal analysis in military applications.It focuses on three significant aspects: wear,friction,and lubrication.Static and kinetic friction leads to the deterioration of material surfaces,known as wear,which could be adhesive or abrasive wear.Sliding velocity and load influence friction between components,thereby,the wear of material.When sliding velocity increases,there will be a subsequent increase in the temperature of the interacting surfaces,irrespective of the ambient temperature.In most cases,there is a proportional relationship between applied load and wear rate.The contact conditions experienced and intense applications of military components signify the importance of studying the effect of load and sliding velocity through tribology tests.

    3.5.1.Effect of load on wear rate and coefficient of friction

    Scanning electron microscopy (SEM) and Energy dispersive XRay analysis (EDX) on the wear track surface and wear debris provide an insight into the wear mechanism involved in Cu-6Sn and Cu-9Ni-6Sn alloys.Zhang et al.[10]have confirmed an inverse relationship between the wear rate and hardness of the alloys by Archard's adhesion theory of wear.In defence applications,very high localised pressure and localised temperature are common and frequent,leading to microscopic projections which result in adhesive wear.The wear rate of both alloys was determined from mass loss using Eq.(2),where WR=wear rate in mg/Nm,

    Load in Newton,and sliding distance in meter

    Military equipment and components need to withstand wear and corrosion.Wear rate happens because of the heat dissipation in components,causing delamination.Table 6 depicts the wear rate and coefficient of friction values of regular bronze alloy (Cu-6Sn)and spinodal alloy (Cu-9Ni-6Sn) with standard deviation for varying loads at different quenching mediums.Fig.16(a) indicates that the wear rate increases with an increase in load for bronze alloy(Cu-6Sn)irrespective of water,SAE 40 oil,and brine solution as quenching media.Fig.16(b) indicates an initial dip in wear rate with an increase in load for spinodal bronze (Cu-9Ni-6Sn alloy)irrespective of water,SAE 40 oil,and brine solution as quenching media.When the load was increased beyond 20 N,the wear rate increased substantially.Adhesive wear was the principal wear mechanism associated with Cu-9Ni-6Sn alloy.The initial dip in wear rate for spinodal bronze could be attributed to the formation of the tribe-oxide layer or mechanically mixed layer (MML).The EDS analysis of the MML/oxide layer is shown in Fig.17.

    Table 6 Variation of wear rate and COF with varying load for quenching medium.

    Fig.16.Effect of load on wear rate and coefficient of friction.

    Fig.17.EDS analysis of MML layer obtained at 20 N.

    This oxide layer,a mechanical mix,forms due to the hightemperature friction on the interacting surfaces of the components.High-strength tribe-oxide layers generated can offer better wear resistance during initial loading.But as the load increases,the tribe-oxide layer breaks,and direct contact occurs between the pin and disc,causing the wear rate to increase in spinodal bronze alloy Cu-9Ni-6Sn.On the contrary,for Cu-6Sn bronze,the continuous increase in wear rate with the load,as shown in Fig.17(a),may be due to a weak tribe-oxide layer that is incapable of withstanding the increasing load.Fig.17(b) indicates that the wear rate is the least when employing brine solution among the three quenching media.

    Fig.16(c) and 16(d) indicate an increase in friction coefficient with an increase in load for both Cu-6Sn and Cu-9Ni-6Sn alloys irrespective of water,SAE 40 oil,or brine solution employed as the quenching media.For minimal loads in the scale of micro to nano newton range,the friction coefficient was observed to be very low[28]The increase in friction coefficient with the rise in load for both Cu-6Sn and Cu-9Ni-6Sn alloys could be attributed to the adhesion strength.When the applied load was less,there was no proper metallic contact as the tribo oxide layer formed (weak in Cu-6Sn and strong in Cu-9Ni-6Sn)was adequate to separate the surfaces.But as the load increased,the trio-oxide layer broke,causing intermetallic contact and increasing the friction coefficient.However,Fig.16(c)and 16(d)point out that the friction coefficient was lowest for Cu-6Sn and Cu-9Ni-6Sn when the brine solution was employed as the quenching media.For brine-quenched spinodal bronze,the highest friction coefficient value was 0.44-0.45,as shown in Fig.16(d),whereas,for brine-quenched bronze,shown in Fig.16(c),the highest friction coefficient value was found to be 0.48.

    3.5.2.Effect of sliding velocity on wear rate and coefficient of friction

    Table 7 shows the wear rate and coefficient of friction values of regular bronze and spinodal alloy for varying sliding velocity at different quenching mediums.Fig.18(a)and 18(b)indicate that the wear rate decreased for both Cu-6Sn and Cu-9Ni-6Sn alloy with increased sliding velocity irrespective of water,SAE 40 oil,or brine solution as the quenching medium.The wear rate was the least when the brine solution was employed as the quenching medium.When the sliding velocity was kept increasing up to 3 m/s,it was observed that a tribo layer/MML was formed on the surface of both Cu-6Sn and Cu-9Ni-6Sn alloy,thereby reducing the material removal rate.The EDS analysis and elemental mapping of the mechanically mixed layer observed in the Cu-9Ni-6Sn spinodal alloy are depicted in Fig.19.

    Table 7 Variation of wear rate and COF with varying sliding velocity for different quenching medium.

    Fig.18.Effect of sliding velocity on wear rate and coefficient of friction.

    For a brine quenched Cu-9Ni-6Sn alloy shown in Fig.18(b),the wear rate became as low as (2 × 10-4) mg/Nm when the sliding velocity was 3 m/s in comparison to brine quenched Cu-6Sn where the wear rate was approximately (4.25 × 10-4) mg/Nm shown in Fig.18(a).This low wear rate with minimal rough surfaces enhances the fact that Cu-9Ni-6Sn alloy undergoes adhesion wear.

    Fig.18(c)and 18(d)indicated a steady increase in the coefficient of friction with an increase in sliding velocity in both Cu-6Sn and Cu-9Ni-6Sn alloys,irrespective of the quenching media.This increase in friction coefficient may be attributed to the adhesive interaction between the pin and disc.However,Fig.18(d)points to the fact that the coefficient of friction is lowest for a brinequenched spinodal bronze when sliding velocity is high.Besides that,Fig.18(c) and 18(d) show that when the sliding velocity was 3 m/s,the brine quenched Cu-6Sn had a friction coefficient as high as 0.52,whereas for the brine quenched Cu-9Ni-6Sn alloy at the same sliding velocity of 3 m/s,the friction coefficient was found to be lower at approximately 0.46~0.48.

    3.5.3.Worn-out surface analysis

    Fig.20 depicts the FESEM images of bronze quenched and aged in brine solution under varying loads of 10 N (Fig.20(a))and 30 N(Fig.20(b)),respectively.On further analysis of both images,deep cracks and wear debris indicate the significant impact of load on the sliding surfaces.In Fig.20(a),deep cracks,ridges,and surface peeloff can be observed.This can be attributed to the abrasive wear mechanism between the alloy and its sliding counterpart.Further,mild scratches can also be observed from the image,which also gives an inference on the increased friction between the surfaces due to load.In Fig.20(b),wear debris increased considerably,and the grooves intensified more than in Fig.20(a).Hence it indicates that at a 30 N load,severe abrasion might have occurred while sliding.As the debris increased over repeated sliding,it might have trapped between the disc and alloy surface resulting in three body wear.This further created ploughing and shallow grooves.

    Fig.20.FSEM image showing the effect of load on wear rate (Bronze -Brine condition,load -(a) 10 N;(b) 30 N,constant Sliding velocity -3 m/s,sliding distance -1000 m).

    So,in the case of bronze,wear was less at load 10 N,and as the load increased,the wear rate also increased substantially.

    Fig.21(a)-21(c) shows the FESEM images of worn-out surfaces of spinodal bronze subjected to varying loads.The worn-out surfaces are categorized by varying loads of 10 N,20 N,and 30 N while sliding distance and sliding velocity are kept constant.

    Fig.21.Effect of load on wear rate(spinodal bronze-Brine condition,load -(a)10 N;(b) 20 N;(c) 30 N constant Sliding velocity -3 m/s,sliding distance -1000 m).

    A detailed analysis based on Fig.21(a)-21(c) shows that the surface has undergone abrasion and delamination.The propagation of subsurface cracks leading to plastic deformation on the surface is eminent in Fig.21(a).Besides this,the presence of grooves and wear debris points to abrasive wear.When the load was increased to 20 N,the worn-out surface,as seen in Fig.19(b),showed a different shift.As shown in Fig.19(b),the worn-out surface of spinodal bronze displayed comparatively lesser wear.The increased load has caused specific welding between the asperities and their sliding counterpart.This has resulted in plastic deformation of the surface,as shown in Fig.21(b).There are also darkcolored regions that show the presence of surface oxidation.The wear rate decrease can be attributed to the formation of a tribo layer through oxidation,indicated in Fig.21(b).This tribolayer reduces the surface contact between the disc and the sliding surface,thereby causing a significant reduction in the coefficient of friction.

    Upon further analysis of Fig.21(c),severe cracks and delamination can be observed.The increased load has increased the friction between two surfaces resulting in the extinction of the tribolayer.This has further resulted in extensive wear on the surface of the specimen.More and more wear particles and delamination are pre-dominant in the worn-out surface,as seen in Fig.21(c).

    Thus it can be summed up that,in the case of spinodal bronze alloy,the wear rate initially followed a dip due to the formation of a tribo oxide layer before it started increasing with increased load.Adhesion was the wear mechanism that was dominant in spinodal bronze.

    4.Conclusions

    The effect of quenching media(water,SAE40 oil,brine solution)on the morphological,mechanical,and tribological characteristics of regular cast bronze (Cu-6Sn) and spinodal bronze(Cu-9Ni-6Sn) was conducted in this study,and the following key observations were made.

    (1) Brine-aged regular bronze and spinodal bronze possess a fine-grained microstructure compared to water-aged and oil-aged specimens.The grain size of brine-aged regular bronze and spinodal bronze was 30-40% finer when compared with water-aged and oil-aged specimens.

    (2) Brine-aged spinodal bronze possesses the maximum hardness compared to water-aged and oil-aged spinodal bronze.The hardness (Hv) of brine-aged spinodal bronze was 20-30% higher when compared with the respective wateraged and oil-aged specimens.

    (3) Wear rate is least when brine is employed as the quenching medium for spinodal bronze and regular bronze.Compared with oil as the quenching medium for spinodal bronze,the wear rate dipped almost 100% for the brine quenched spinodal bronze.Also,in comparison with water as the quenching medium,the wear rate decreased by 40-50% for brine-quenched spinodal bronze.

    (4) Coefficient of friction is lowest for regular bronze and spinodal bronze when brine is used as the quenching medium.Brine-quenched regular bronze,and spinodal bronze possess a low friction coefficient of 0.40-0.4,6 whereas oil quenched and water-quenched regular bronze and spinodal bronze samples have a friction coefficient greater than 0.5.

    These observations suggest that brine solution is the most effective quenching medium for characteristics enhancement of regular cast bronze and spinodal bronze.

    Declaration of competing interest

    The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

    视频区图区小说| 国内揄拍国产精品人妻在线| 99久久精品热视频| 最近中文字幕高清免费大全6| 午夜爱爱视频在线播放| 久久久久久久国产电影| 欧美激情国产日韩精品一区| 欧美人与善性xxx| 成人免费观看视频高清| 日日摸夜夜添夜夜添av毛片| 在线看a的网站| 国产高清不卡午夜福利| 欧美日韩一区二区视频在线观看视频在线 | 一级毛片电影观看| 欧美bdsm另类| 亚洲精品,欧美精品| 免费看a级黄色片| eeuss影院久久| 亚洲av成人精品一二三区| 亚洲成人中文字幕在线播放| 久久人人爽人人片av| 国产亚洲av嫩草精品影院| 亚洲国产精品成人久久小说| 插阴视频在线观看视频| 高清欧美精品videossex| 97精品久久久久久久久久精品| 香蕉精品网在线| 如何舔出高潮| 少妇人妻精品综合一区二区| 欧美zozozo另类| 国产黄a三级三级三级人| 综合色丁香网| 成年免费大片在线观看| 亚洲av成人精品一区久久| 免费看光身美女| 亚洲成人av在线免费| 两个人的视频大全免费| www.av在线官网国产| 内射极品少妇av片p| 国产一区有黄有色的免费视频| 在线免费观看不下载黄p国产| 久久久久精品性色| 成人鲁丝片一二三区免费| 97在线人人人人妻| 午夜精品一区二区三区免费看| 国产精品国产av在线观看| 国产毛片a区久久久久| 夫妻性生交免费视频一级片| 日本黄大片高清| 成人毛片60女人毛片免费| 国产极品天堂在线| av播播在线观看一区| 国产精品福利在线免费观看| 人妻系列 视频| 涩涩av久久男人的天堂| 精品国产露脸久久av麻豆| 日本wwww免费看| 日日摸夜夜添夜夜爱| 狂野欧美白嫩少妇大欣赏| 午夜免费男女啪啪视频观看| 亚洲欧美精品专区久久| 丝袜美腿在线中文| 在线亚洲精品国产二区图片欧美 | 日韩大片免费观看网站| 禁无遮挡网站| 人妻 亚洲 视频| 制服丝袜香蕉在线| 欧美另类一区| 欧美3d第一页| 人人妻人人看人人澡| 亚洲精品日韩在线中文字幕| 久久久精品欧美日韩精品| 狂野欧美白嫩少妇大欣赏| 国产成人午夜福利电影在线观看| 日本与韩国留学比较| 亚洲一级一片aⅴ在线观看| 亚洲精品久久午夜乱码| 九九爱精品视频在线观看| 精品酒店卫生间| 成人国产麻豆网| 天天躁日日操中文字幕| 亚洲国产成人一精品久久久| 亚洲av在线观看美女高潮| 日本三级黄在线观看| 亚洲国产成人一精品久久久| 99热这里只有精品一区| 午夜精品一区二区三区免费看| 精品一区二区三卡| 欧美区成人在线视频| 少妇丰满av| 欧美zozozo另类| 别揉我奶头 嗯啊视频| 人妻制服诱惑在线中文字幕| 久久精品国产自在天天线| 日韩成人av中文字幕在线观看| 久热久热在线精品观看| 三级男女做爰猛烈吃奶摸视频| 一区二区av电影网| 一本一本综合久久| 看黄色毛片网站| 婷婷色麻豆天堂久久| 18禁在线播放成人免费| 亚洲精品中文字幕在线视频 | 丝袜脚勾引网站| 80岁老熟妇乱子伦牲交| av国产免费在线观看| 卡戴珊不雅视频在线播放| 久久精品国产亚洲av天美| 国产av码专区亚洲av| 国产成年人精品一区二区| 亚洲国产色片| 日韩,欧美,国产一区二区三区| 久久国产乱子免费精品| 亚洲国产最新在线播放| 中文字幕久久专区| 久久国产乱子免费精品| 亚洲高清免费不卡视频| 亚洲人与动物交配视频| 国产探花极品一区二区| 亚洲av.av天堂| 国产淫片久久久久久久久| 国产亚洲av片在线观看秒播厂| 精品久久久久久久人妻蜜臀av| 97超碰精品成人国产| av福利片在线观看| 亚洲一区二区三区欧美精品 | 热re99久久精品国产66热6| 日本黄色片子视频| 亚洲丝袜综合中文字幕| 麻豆成人av视频| 神马国产精品三级电影在线观看| 国产高清有码在线观看视频| 欧美极品一区二区三区四区| 三级国产精品欧美在线观看| 亚洲av电影在线观看一区二区三区 | 99精国产麻豆久久婷婷| 国产男人的电影天堂91| 在线免费观看不下载黄p国产| 十八禁网站网址无遮挡 | 免费高清在线观看视频在线观看| 91精品伊人久久大香线蕉| 久久精品国产鲁丝片午夜精品| av.在线天堂| 精品久久久久久久久av| 99久久精品国产国产毛片| 国产精品伦人一区二区| 成人免费观看视频高清| 26uuu在线亚洲综合色| 麻豆国产97在线/欧美| 中文乱码字字幕精品一区二区三区| 国产久久久一区二区三区| 国产老妇女一区| 国产成人91sexporn| 亚洲四区av| 亚洲丝袜综合中文字幕| 精品久久久久久久久亚洲| 大片免费播放器 马上看| 97在线人人人人妻| 日日啪夜夜撸| 国产一区二区亚洲精品在线观看| 欧美zozozo另类| 麻豆成人午夜福利视频| 激情 狠狠 欧美| 国产成人免费无遮挡视频| 亚洲欧美清纯卡通| 亚洲成人一二三区av| 最近最新中文字幕大全电影3| 高清视频免费观看一区二区| 麻豆久久精品国产亚洲av| 99热网站在线观看| 深爱激情五月婷婷| 午夜爱爱视频在线播放| 97在线人人人人妻| 成人高潮视频无遮挡免费网站| 神马国产精品三级电影在线观看| 成人亚洲精品一区在线观看 | 精品亚洲乱码少妇综合久久| 黄色配什么色好看| 成人黄色视频免费在线看| 夫妻午夜视频| 国产69精品久久久久777片| 少妇高潮的动态图| 成人国产av品久久久| 亚洲国产成人一精品久久久| 久久久久久九九精品二区国产| 国产黄片视频在线免费观看| 美女被艹到高潮喷水动态| 国产精品无大码| 丰满少妇做爰视频| 亚洲,欧美,日韩| 日日撸夜夜添| 一级av片app| 在线天堂最新版资源| 狂野欧美激情性xxxx在线观看| 国产爽快片一区二区三区| 两个人的视频大全免费| 精品久久久久久久末码| 色5月婷婷丁香| 亚洲国产精品专区欧美| 国产精品久久久久久精品古装| 欧美日韩综合久久久久久| av一本久久久久| 成人亚洲精品av一区二区| 欧美三级亚洲精品| 午夜福利视频精品| 日日摸夜夜添夜夜添av毛片| 国产国拍精品亚洲av在线观看| 婷婷色av中文字幕| 亚洲av日韩在线播放| 最近中文字幕高清免费大全6| 国产免费又黄又爽又色| 欧美精品一区二区大全| 精品国产三级普通话版| 一级毛片久久久久久久久女| 欧美成人精品欧美一级黄| 少妇高潮的动态图| 搡女人真爽免费视频火全软件| 亚洲色图综合在线观看| 18禁动态无遮挡网站| 国产精品一区二区三区四区免费观看| av播播在线观看一区| 久久99热这里只有精品18| 久久综合国产亚洲精品| 免费观看的影片在线观看| 干丝袜人妻中文字幕| 日韩成人伦理影院| 2021天堂中文幕一二区在线观| 18禁在线播放成人免费| a级毛色黄片| 久久久精品免费免费高清| 日韩av在线免费看完整版不卡| 亚洲av福利一区| 最后的刺客免费高清国语| 18禁动态无遮挡网站| 国产精品一区二区在线观看99| 少妇丰满av| av卡一久久| 国产乱人偷精品视频| av福利片在线观看| 少妇裸体淫交视频免费看高清| 精品国产三级普通话版| 欧美国产精品一级二级三级 | 国产精品无大码| 啦啦啦中文免费视频观看日本| 久久久久性生活片| 80岁老熟妇乱子伦牲交| 国产精品国产三级专区第一集| 国产精品精品国产色婷婷| 热re99久久精品国产66热6| 看免费成人av毛片| 99久久精品热视频| 久久久久九九精品影院| videos熟女内射| 人妻夜夜爽99麻豆av| 少妇的逼好多水| 亚洲av免费高清在线观看| 午夜精品一区二区三区免费看| av国产免费在线观看| 国产成人aa在线观看| 一个人看的www免费观看视频| 国产黄a三级三级三级人| 日本三级黄在线观看| 2021天堂中文幕一二区在线观| 国产精品嫩草影院av在线观看| 亚洲欧美日韩东京热| 亚洲自偷自拍三级| 久久精品夜色国产| 午夜精品一区二区三区免费看| 国产熟女欧美一区二区| 97超视频在线观看视频| 天天躁日日操中文字幕| 晚上一个人看的免费电影| 亚洲国产精品999| 欧美日韩在线观看h| 舔av片在线| 一级av片app| 美女脱内裤让男人舔精品视频| 亚洲av男天堂| 国产精品熟女久久久久浪| 久久久精品免费免费高清| 日韩不卡一区二区三区视频在线| 日本欧美国产在线视频| 午夜视频国产福利| 亚洲精品乱久久久久久| 国产大屁股一区二区在线视频| 国产精品福利在线免费观看| 在线免费观看不下载黄p国产| 亚洲欧美一区二区三区黑人 | 一级片'在线观看视频| 久久久久国产精品人妻一区二区| 夫妻性生交免费视频一级片| 日本av手机在线免费观看| 国产午夜精品一二区理论片| 国国产精品蜜臀av免费| 成人一区二区视频在线观看| 国产欧美另类精品又又久久亚洲欧美| 最近最新中文字幕大全电影3| 免费播放大片免费观看视频在线观看| 女的被弄到高潮叫床怎么办| 久久久亚洲精品成人影院| 免费观看的影片在线观看| 亚洲av二区三区四区| 美女xxoo啪啪120秒动态图| 亚洲精品一区蜜桃| 少妇人妻久久综合中文| 国产精品伦人一区二区| 伊人久久精品亚洲午夜| 久久精品综合一区二区三区| 男女无遮挡免费网站观看| 99久久中文字幕三级久久日本| 亚洲精品国产av蜜桃| 国产免费一级a男人的天堂| 舔av片在线| 久热这里只有精品99| 免费电影在线观看免费观看| 国产大屁股一区二区在线视频| 亚洲av成人精品一二三区| 久热久热在线精品观看| 国产永久视频网站| 天天躁日日操中文字幕| 亚洲欧美日韩东京热| 国国产精品蜜臀av免费| 亚洲图色成人| 成人黄色视频免费在线看| 麻豆精品久久久久久蜜桃| 久久热精品热| 少妇人妻精品综合一区二区| 最近中文字幕2019免费版| 精华霜和精华液先用哪个| 国产精品麻豆人妻色哟哟久久| 国产探花在线观看一区二区| freevideosex欧美| 久久人人爽av亚洲精品天堂 | 禁无遮挡网站| 色5月婷婷丁香| 久久99热这里只有精品18| 日韩 亚洲 欧美在线| 国产探花极品一区二区| 欧美精品国产亚洲| 欧美丝袜亚洲另类| 国产av码专区亚洲av| 欧美日韩视频高清一区二区三区二| 成年女人在线观看亚洲视频 | 国产精品国产av在线观看| 色哟哟·www| 一级毛片aaaaaa免费看小| 3wmmmm亚洲av在线观看| 成年av动漫网址| 久久久久精品久久久久真实原创| 欧美成人午夜免费资源| 男女下面进入的视频免费午夜| 国产精品一二三区在线看| 国内精品美女久久久久久| 亚洲av免费高清在线观看| 欧美三级亚洲精品| 麻豆成人午夜福利视频| 亚洲激情五月婷婷啪啪| 最近的中文字幕免费完整| 中文字幕av成人在线电影| 熟女人妻精品中文字幕| 午夜福利在线在线| 三级国产精品片| 久久久午夜欧美精品| 亚洲精品成人av观看孕妇| 天天躁夜夜躁狠狠久久av| 国产黄色视频一区二区在线观看| 尤物成人国产欧美一区二区三区| 久久久久久久久大av| 日本av手机在线免费观看| av在线app专区| av在线观看视频网站免费| 婷婷色综合www| 黄色配什么色好看| 在线观看国产h片| 日日啪夜夜爽| 亚洲av中文av极速乱| 亚洲av免费高清在线观看| 内地一区二区视频在线| 精品国产露脸久久av麻豆| 亚洲av.av天堂| 久久久色成人| 国产精品人妻久久久久久| 精品国产三级普通话版| 一级黄片播放器| 久久久久久久久久人人人人人人| av又黄又爽大尺度在线免费看| 蜜桃亚洲精品一区二区三区| 三级国产精品欧美在线观看| 精品人妻熟女av久视频| 国产成人精品久久久久久| 好男人视频免费观看在线| 午夜免费观看性视频| 少妇人妻 视频| 少妇人妻精品综合一区二区| 99热这里只有精品一区| 日韩一本色道免费dvd| 日韩成人av中文字幕在线观看| 丝袜喷水一区| 欧美bdsm另类| 少妇人妻精品综合一区二区| 九九在线视频观看精品| a级毛片免费高清观看在线播放| 亚洲久久久久久中文字幕| 天天躁夜夜躁狠狠久久av| 男插女下体视频免费在线播放| 大香蕉久久网| 精品久久国产蜜桃| 直男gayav资源| 男人舔奶头视频| 国语对白做爰xxxⅹ性视频网站| 熟女av电影| 精品一区二区免费观看| 少妇人妻精品综合一区二区| 你懂的网址亚洲精品在线观看| 欧美性感艳星| 国产一区有黄有色的免费视频| 日韩中字成人| 免费黄色在线免费观看| 日本-黄色视频高清免费观看| 老司机影院成人| 夜夜看夜夜爽夜夜摸| 国产在视频线精品| 日本熟妇午夜| 免费观看性生交大片5| 亚洲欧美日韩东京热| 天天躁日日操中文字幕| 国产免费福利视频在线观看| 一级黄片播放器| 国产 一区精品| 中文字幕制服av| 亚洲国产精品专区欧美| 男女无遮挡免费网站观看| 精品一区在线观看国产| 亚洲欧美日韩东京热| 大片免费播放器 马上看| 成人亚洲精品一区在线观看 | 精品久久久久久久久av| av在线亚洲专区| 一二三四中文在线观看免费高清| 一级毛片黄色毛片免费观看视频| 男人爽女人下面视频在线观看| 国内少妇人妻偷人精品xxx网站| 人妻系列 视频| 日韩中字成人| 国产高清三级在线| 久久ye,这里只有精品| 不卡视频在线观看欧美| 美女内射精品一级片tv| 久久鲁丝午夜福利片| 日本爱情动作片www.在线观看| 亚洲av二区三区四区| 亚洲最大成人手机在线| 91午夜精品亚洲一区二区三区| 又爽又黄无遮挡网站| 草草在线视频免费看| 一本久久精品| 国产精品久久久久久久电影| 国产美女午夜福利| 亚洲欧美日韩无卡精品| 亚洲天堂国产精品一区在线| 欧美另类一区| 国产毛片a区久久久久| 日韩免费高清中文字幕av| 丝袜脚勾引网站| 日本与韩国留学比较| 蜜桃久久精品国产亚洲av| 97超视频在线观看视频| 亚洲欧美成人精品一区二区| 国产精品女同一区二区软件| 免费av观看视频| 欧美性感艳星| 午夜激情福利司机影院| 又黄又爽又刺激的免费视频.| 成人黄色视频免费在线看| 国产中年淑女户外野战色| 国产永久视频网站| 久久久久国产精品人妻一区二区| 亚洲精品久久久久久婷婷小说| 久久久久性生活片| 亚洲一区二区三区欧美精品 | 大香蕉久久网| 69av精品久久久久久| 免费看不卡的av| 男女边吃奶边做爰视频| 亚洲丝袜综合中文字幕| 成人国产av品久久久| 91久久精品国产一区二区成人| 国产精品国产av在线观看| 久久久久久国产a免费观看| 欧美一级a爱片免费观看看| 国产国拍精品亚洲av在线观看| 欧美变态另类bdsm刘玥| 国产亚洲91精品色在线| 日韩欧美一区视频在线观看 | 制服丝袜香蕉在线| 乱系列少妇在线播放| 国产永久视频网站| 国产精品一区www在线观看| 纵有疾风起免费观看全集完整版| 男人和女人高潮做爰伦理| 插逼视频在线观看| 国产精品久久久久久精品电影小说 | 亚洲性久久影院| 亚洲,一卡二卡三卡| 欧美区成人在线视频| 在现免费观看毛片| 亚洲精品视频女| 一级av片app| 少妇丰满av| 久久久久久久国产电影| 哪个播放器可以免费观看大片| 欧美3d第一页| 2021天堂中文幕一二区在线观| 久久久久性生活片| 蜜臀久久99精品久久宅男| 免费在线观看成人毛片| 欧美三级亚洲精品| 亚洲综合精品二区| 亚洲经典国产精华液单| 一级毛片电影观看| 精品少妇黑人巨大在线播放| 卡戴珊不雅视频在线播放| 亚洲成人av在线免费| 日韩免费高清中文字幕av| 麻豆久久精品国产亚洲av| 水蜜桃什么品种好| 久久久午夜欧美精品| 99热这里只有是精品在线观看| 国产一区二区三区av在线| 18禁动态无遮挡网站| 免费av毛片视频| 国产成人精品一,二区| 久久久久久国产a免费观看| 国产伦理片在线播放av一区| 免费大片18禁| 人妻一区二区av| 久久人人爽av亚洲精品天堂 | 日本免费在线观看一区| 麻豆成人午夜福利视频| 国产极品天堂在线| 蜜桃亚洲精品一区二区三区| 搡女人真爽免费视频火全软件| 久久精品国产鲁丝片午夜精品| av免费在线看不卡| 老师上课跳d突然被开到最大视频| 久久久久久久久大av| 国产免费又黄又爽又色| 有码 亚洲区| 国产毛片在线视频| av国产精品久久久久影院| 亚洲色图av天堂| 亚洲精品国产av成人精品| 久久久久久九九精品二区国产| 久久久久性生活片| 3wmmmm亚洲av在线观看| av在线app专区| 高清av免费在线| 日韩成人伦理影院| 欧美zozozo另类| 日本免费在线观看一区| 精品人妻熟女av久视频| 精品人妻视频免费看| 涩涩av久久男人的天堂| 国国产精品蜜臀av免费| 人妻制服诱惑在线中文字幕| 禁无遮挡网站| 久久99精品国语久久久| 国产精品蜜桃在线观看| 在线观看一区二区三区| 免费少妇av软件| 在线观看国产h片| 成年版毛片免费区| 亚洲精品国产av成人精品| 中文字幕av成人在线电影| 久久精品久久久久久久性| 亚洲精品久久久久久婷婷小说| 欧美高清性xxxxhd video| 国产国拍精品亚洲av在线观看| 搞女人的毛片| 国产成人a∨麻豆精品| 九九在线视频观看精品| 麻豆成人午夜福利视频| 国产一区二区三区av在线| 国产亚洲av片在线观看秒播厂| 国产视频首页在线观看| 黄色日韩在线| 亚洲国产精品成人久久小说| 黄片无遮挡物在线观看| 国产老妇伦熟女老妇高清| 亚洲精品亚洲一区二区| 国产精品蜜桃在线观看| xxx大片免费视频| 国产精品99久久久久久久久| 国模一区二区三区四区视频| 纵有疾风起免费观看全集完整版| 黄色视频在线播放观看不卡| 亚洲精品成人久久久久久| 成人美女网站在线观看视频| 亚洲欧洲日产国产| 国产毛片a区久久久久| 在线免费十八禁| 亚洲精品成人久久久久久| 国产日韩欧美在线精品| 亚洲精品国产成人久久av| 日韩国内少妇激情av| 欧美日韩视频高清一区二区三区二| 国产白丝娇喘喷水9色精品| 99久国产av精品国产电影| 亚洲精品久久久久久婷婷小说| 久久久久久久大尺度免费视频| 国产精品秋霞免费鲁丝片| 国产淫语在线视频| 观看美女的网站| 熟女电影av网| 在线免费观看不下载黄p国产|