简介概要

Effect of electrolyte temperature on the nano-carbonitride layer fabricated by surface nanocrystallization and plasma treatment on a γ-TiAl alloy

来源期刊：Rare Metals2009年第5期

论文作者：M. Aliofkhazraei A. Sabour Rouhaghdam H. Hassannejad

文章页码：454 - 459

摘要：The distribution of nano-carbonitrides produced by the treatments of surface nanocrystallization and plasma electrolytic carbonitriding on a γ-TiAl was investigated by means of figure analysis. The skewness and kurtosis of Gaussian shape distribution curves were studied and the effect of electrolyte temperature was determined. The usage of lower temperatures of the electrolyte is more suitable for achieving lower sizes of complex nano-carbonitrides. The surface roughness of treated samples was measured and it was observed that there is an optimum level of electrolyte temperature for surface roughness increase （difference between two measured data）.

Rare Metals 2009,28(05),454-459

Effect of electrolyte temperature on the nano-carbonitride layer fabricated by surface nanocrystallization and plasma treatment on a γ-TiAl alloy

M. Aliofkhazraei A. Sabour Rouhaghdam H. Hassannejad

Department of Materials Engineering, Faculty of Engineering, Tarbiat Modares University

作者简介：A. Sabour Rouhaghdam E-mail: sabour01@gmail.com, sabour01@modares.ac.ir;

收稿日期：16 January 2009

基金：funded by the National Elite Foundation of Iran and Iranian Nanotechnology Initia-tive is appreciated;

Effect of electrolyte temperature on the nano-carbonitride layer fabricated by surface nanocrystallization and plasma treatment on a γ-TiAl alloy

Abstract：

The distribution of nano-carbonitrides produced by the treatments of surface nanocrystallization and plasma electrolytic carbonitriding on a γ-TiAl was investigated by means of figure analysis. The skewness and kurtosis of Gaussian shape distribution curves were studied and the effect of electrolyte temperature was determined. The usage of lower temperatures of the electrolyte is more suitable for achieving lower sizes of complex nano-carbonitrides. The surface roughness of treated samples was measured and it was observed that there is an optimum level of electrolyte temperature for surface roughness increase (difference between two measured data).

Keyword：

carbonitride; electrolyte temperature; nanostructures; plasma electrolytic deposition; roughness; γ-TiAl;

Received： 16 January 2009

1. Introduction

It is well known that diffusion processes can exert significant influences on the tribological properties of different alloys.Titanium based alloys are used for fabricating components of aerospace engines.The innovation of appropriate new coatings with improved performance for titanium is still an interesting and important research area.One way to improve the surface hardness and passive layer stability of different kinds of titanium based alloys is by using the plasma electrolysis technique [ 1] .The pulsed plasma electrolytic carbonitriding PPEC/N technique is a relatively new group of saturation treatments which have been used on different metals [ 2] .Electrolytes of this electrolytic method are ecologically friendly and non-hazardous [ 3] .

Despite the considerable effort that has been invested into understanding the different properties of treated samples,it seems that little attention has been given to the role of PPEC/N in wear and corrosion procedures.Deep understanding of the subject can not be achieved without a clear view of the role of morphology of the nanostructured layers.It is therefore the aim of this paper to further clarify the important role of the nanocrystalline carbonitrided layer and its role on protectingγ-Ti Al.

Nanocrystalline carbonitride layers can be synthesized successfully through the pulsed plasma electrolytic carbonitriding method,which is caused by the bombardment of the sample in an organic based electrolyte [ 4] .It was clearly found that there is a direct relationship between the nanostructure of the compound layer and its different properties Considering the Gaussian shape of normal distribution for achieved nano-carbonitrides,it is better to have a norma distribution with its specific kurtosis and skewness levels Thus,in this paper,the surface roughness of pulsed plasma electrolytic carbonitridedγ-Ti Al was studied with respect to different deviations from normal status.The fourth momen distributions of nano-carbonitrides were studied and they revealed direct relations with the roughness of differen treated samples.

2. Experimental

2.1. Materials and treatments

A set of samples was machined from aφ20 mm rod of Ti-48Al-2Cr-2Nb(Al 33.5 wt.%,Cr 2.55 wt.%,Nb 2.67wt.%,Ti balance)(referred to asγ-Ti Al).The sample was a cylinder of 5 mm in length and 20 mm in diameter with a small hole(φ4 mm)at its surrounding area for holding purposes.Specimens were ground with SiC emery paper up to2500 grit,cleaned with distilled water,and degreased with acetone in an ultrasonic bath for 10 min before the treatment.

2.2. Duplex treatments

For the surface nanocrystallization(SN)process,samples were subjected in the cylindrical steel container of a high energy mill with enough WC/Co balls(φ6 mm)for 3 h under vacuum conditions [ 5, 6] .The average size of nanocrystals was obtained as about 74 nm after this treatment.Pulsed plasma electrolytic carbonitriding(PPEC/N)treatment was conducted with a 30 k W power supply.It was performed with an electrical DC source and a pulser box.A cooling system was used for controlling the electrolyte temperature.Our experimental set-up can be found in our previous articles such as Ref. [ 7] .After loading the sample,the carbonitriding reactor was circulated for the purpose of adjusting the electrolyte temperature by a pump and the sample was located in the middle of a cylinder shaped stainless steel anode.The sample was then treated at an average current density of 0.7 A/cm²,a frequency of 5 kHz,and a cathodic peak voltage of 600 V at different electrolyte temperatures.The current density was directly related to the electrical conductivity of the electrolyte,which was adjusted by the addition of different concentrations of sodium carbonate to the triethanolamine based electrolyte.The resulting sparks then bombarded the sample with high-energy radicals and produced the carbonitrided surface.The distance between the cathode and anode was 50 mm.The samples were allowed to cool down to room temperature from the carbonitriding temperature by taking them out during sparking from the reactor.

2.3. Analysis

The surface roughness of treated samples was measured via a Taylor-Hobson Surtronic 25 roughness checker.The obtained data was analyzed by Talyprofile(Gold)software.To measure the average size of nanocrystallites,5 SEM nanostructures with the same magnification were analyzed through commercial software for figure analysis,called a4i Docu,for each treated sample.Scanning electron microscopy(SEM)was performed using a Philips XL-30.Different measurements were interpolated to obtain average results.At least 40 measurements were done in each nanostructure to minimize systematical errors.The nanostructure of the coating was also evaluated by atomic force microscopy(AFM).The AFM part was a Nano Scope II from Digital Instruments,USA.Non-scraping Si₃N₄-tips were used throughout.The distribution of nanocrystals was plotted and they showed a Gaussian-shape.Several different distributions(for several temperatures of electrolyte)have been used to give a more accurate fit to the average size of nanocrystallites than is possible using a Gaussian distribution.In order to obtain realistic profiles,the first four moments of the distribution functions(as can be seen in Eq.(1)-(4))must be taken into account.The first moment R_p,known as the range,refers to the peak of the distribution.The range is the average size of nanocrystallites at the maximum value of distribution and does not affect the shape of the profile.The second momentΔR,or straggle,gives the width of the distribution.The normalized third momentγ(Eq.(5))indicates the skewness for a given distribution profile.It will have a value of zero for a pure Gaussian distribution.If the skewness is negative,the distribution falls toward the surface more smoothly and decreases sharply on the other side of the distribution peak.The steepness of the slopes is opposite for positive skewness.The normalized fourth momentβ(Eq.(6)),called kurtosis,describes the extent of the tail for any distribution.Kurtosis also indicates whether a distribution is sharply peaked or more flat topped A distribution with a high value of kurtosis will be sharply peaked with a long exponential tail.For a pure Gaussian distribution the kurtosis has a value of 3.Schematic figures of kurtosis and skewness and their changes can be seen in Fig.1.

Fig.1.Schematic changes of kurtosis and skewness of distri-bution curves:(a)1?positive skewness,2?negative skewness,3?skewness=0,for normal distribution;(b)1?kurtosis<32?kurtosis>3,3?kurtosis=3,for normal distribution.

3. Results

For each reported data,3 events were calculated to achieve sufficient accuracy.Fig.2 is an example of achieved nanocrystallites during the carbonitriding process.Four parameters that were extracted from the best fit of the distribution functions are shown in Fig.3(a)-3(d).For the range in Fig.3(a),a linear relationship was obtained with the temperature of the electrolyte.The results for the straggle show the same curvature as a function of electrolyte temperature in Fig.3(b).

4. Discussion

4.1. Plasma electrolytic carbonitriding

4.1.1. Effect of electrolyte temperature on the distribution of nanocrystallites

Changes of the different moments of distribution are shown in Figs.3(a)-3(d)during surface hardening.It is revealed that different moments of distribution change linearly with an increase in electrolyte temperature during the coating process.The electrolyte temperature in this process is one of the most important factors in identifying the surface temperature around the sample.By increasing the electrolyte temperature,local quenching and formation of nanocrystallites will occur at higher temperatures,and the average size of nanocrystallites will increase.A normal distribution of nanocrystallites will be obtained when the electrolyte temperature nears 50°C,and this can be seen from Figs.3(c)and 3(d)(for skewness=0 and kurtosis=3).

Fig.2.(a)Nano-carbonitrides fabricated at an electrolyte temperature of 68°C;(b)AFM nanostructure of minimum ob-tained roughness.

Fig.3.Distribution curves of different parameters versus electrolyte temperature for PPEC/N treatment:(a)range;(b)straggle;(c)skewness;(d)kurtosis.

4.1.2. Effect of electrolyte temperature on roughness in-creasing

It is common knowledge that the best properties of nanocrystallites can be achieved at their normal narrow distribution,so the changes of one important related property of coatings to the distribution of nanocrystallites(surface roughness)are plotted in Fig.4.It can be seen in Fig.4 that the roughness of the coating increases slightly by increasing the electrolyte temperature,but the amount of increase(difference between each two measured data)will decrease towards 50°C and increase after that,which means that from this point of view,the electrolyte temperature of 50°C is an optimum level.In fact,increasing the electrolyte temperature will cause two inverse effects(increasing the average size of nanocrystallites and decreasing the effect of micro and nano explosions on the surface of the cathode(sparks))on the roughness of the treated sample.This optimum leve is the point at which the distribution of nanocrystallites is in its normal mode.By increasing the skewness and kurtosis of distribution toward its normal mode,however,the effect of sparks on the total profile of the surface is more than the shape of nanocrystallites,but the amount of increasing roughness will decrease.Increasing the skewness and kurtosis of distribution after its normal mode will increase roughness,which is usually not appropriate in industrial usages.

4.2. Surface nanocrystallization+plasma electrolytic carbonitriding

4.2.1. Effect of electrolyte temperature on the distribution of nanocrystallites

Figs.5(a)-5(d)illustrates the changes for differen moments of distribution during duplex treatment.All of the changes are similar to the PPEC/N treatment without surface nanocrystallization,and somehow the average size of nanocrystallites has been changed.The regressed equations for the electrolyte temperature can again be expressed linearly.It appears that the effect of surface nanocrystallization and residual stresses just show themselves on the growth mechanism of nanocrystallites and their average size,and do not affect the relation of the electrolyte temperature with the duplex process.By increasing the electrolyte temperature,cathodic reactions will more and more affect the coating and the relation between the electrolyte temperature and the distribution of nanocrystallites retains its linear status.A normal distribution of nanocrystallites will be obtained at an electrolyte temperature near 40°C,and this can be seen from Figs.3(c)and 3(d)(for skewness=0 and kurtosis=3).It is also obvious that the effect of surface nanocrystallization is that of reducing the electrolyte temperature to achieve normal distribution.

Fig.4.Roughness increasing versus electrolyte temperature for PPEC/N treatment.

Fig.5.Distribution curves of different parameters versus electrolyte temperature for duplex treatments:(a)range(b)straggle(c)skewness(d)kurtosis.

4.2.2. Effect of electrolyte temperature on roughness in-creasing

The changes of surface roughness of coatings have been plotted in Fig.6.The roughness will increase by itself due to duplex treatment,but the amount of its increase will show a similar trend as a single PPEC/N treatment.The amount of roughness increase(difference among each two measured data)will decrease towards 40°C and increase after that,which means that,again,the electrolyte temperature of 40°C is an optimum level.Decreasing the effect of micro and nano sparks have a higher effect on roughness than an increase in the average size of nanocrystallites,which leads to a higher surface roughness for duplex treatment than single PPEC/N treatment,but the roughness increase is smaller in amount for the duplex treatment.

Fig.6.Roughness increasing versus electrolyte temperature for duplex treatments.

4.3. Effect of electrolyte temperature on growth kinetics for obtained layers

It has been revealed that the thicknesses of the diffusive layer or compound layer for cathodic plasma electrolytic saturation treatments change linearly with the square root of the treatment time [ 7] .The thickness of each layer was measured on the cross section of the treated samples.The thicknesses of layers fabricated in constant time increase with an increase in electrolyte temperature during carbonitriding treatment(however the roughness will also increase)Also,the surface nanocrystallization(SN)process causes layers to rapidly form;thus the layer thickness in equal carbonitriding treatment time and applied voltage with pre SN treatment is higher than that without pre SN treatment.The increase in the thicknesses of layers proceeds in accordance with a linear relation,which becomes non-linear for duplex treatment.Due to different mechanisms of carbonitriding processes that work by accelerating the carbon radicals from those that work by thermal activation and diffusion at long periods,the equations on growth kinetics of the method presented here by formulas resulted from interpolating different obtained data for effective factors with R_fitting≥0.98which shows excellent fittings.Eq.(7)shows the relationship between the thickness of the influenced layer(tota thicknesses of the layer with increased hardness in the raw sample plus the compound layer)and the electrolyte temperature.

where T_h and T are the thickness of the influenced layer and the electrolyte temperature during carbonitriding treatment respectively.αandβare constants and are related to the treatment time forαand the modified transmission coefficient of treated media forβ.

By applying the pre SN process,the above equation is modified in this way:

As it can be seen from Eq.(8),T_h′increases more rapidly than T_h,compared with Eq.(7).In fact,the SN process affects the formation mechanism of the nanocrystalline carbonitride layer more than the shape or properties of carbonitride nanocrystallites.The nanocrystalline surface of substrates will lead to faster formation of the hard layer at lower electrolyte temperatures than a microcrystalline surface.

5. Conclusions

Based on experimental results,the following conclusions can be drawn from the present investigation on the duplex treatments of surface nanocrystallization and pulsed plasma carbonitridedγ-Ti Al.

(1)Normal distribution of nanocrystallites will be obtained when the electrolyte temperature is near 50°C.

(2)Increasing the electrolyte temperature will cause two inverse effects(increasing the average size of nanocrystallites and decreasing the destructive effect of sparks)on the roughness of the treated sample.

(3)Increasing the skewness and kurtosis of distribution after its normal mode will increase the roughness,which usually is not appropriate in industrial usages.

(4)Surface nanocrystallization will affect the growth mechanism and roughness of the hard layer while its effec on thickness growth is more than the roughness.

(5)Under similar conditions,the nanocrystalline surface of the substrate will lead to faster formation of the hard layer at lower electrolyte temperatures than the microcrystalline surface.

Acknowledgements

The authors wish to express their thanks to Dr.J.Curran(Cambridge University)for his useful guidance during the investigation of different aspects of plasma electrolysis.Partial work of this project which was funded by the Nationa Elite Foundation of Iran and Iranian Nanotechnology Initiative is appreciated.