New Trends in Corrosion Analysis of Al-Sn Alloy Duplex System

The corrosion characterization of binary Al-Sn alloy systems has been statistically analyzed in the light of developed model equations. It was observed that the modeled corrosion penetration rate values generated using the developed model equations are in tandem with the experimental values.


Introduction
It has since been a worldwide problem for man to combat the menace of material degradation.Many structural failures and general loss of valuable engineering materials have been traced to be caused basically by analysis and not synthesis [1] as over 313 failure cases studied [2] have shown that well over 56.90% are due to corrosion.The causes of materials degradation with the associated environmental variables have been well explained in many literatures [3][4][5][6][7][8][9][10][11][12][13][14].
A critical look at all forms of corrosion show that they are merely a statistical phenomenon hence, the yardstick behind the adoption of statistics techniques in this present study.However, even with the somewhat semi-empirical nature of corrosion (as it show the relations between available data and measurements that do not necessarily reveal any relation between cause and effect), statistical models attempt to determine the fundamental relationship between sets of input data (predictors) and targets (predictands) [15,16].
The adoption of statistical analysis in corrosion analysis in metals has been predicted for years but, its usage has been hampered primarily by the specialty skill needed in applying this principle which, its impediment lies basically on the lack of reference frame for its application [16].
In this present study, we will apply the non-linear regression analysis technique to obtain model equations that will be used to determine the corrosion parameter of interest and other statistical parameters adequate for discussing and understanding the phenomenon of corrosion (in this case, Al-Sn alloys of compositions 2.5% and 4.5% by weight respectively of Sn) in selected media environments: HCl and NaCl of concentrations 0.5 M and 1.0 M respectively, using the statistical software SPSS ™ .
The data of Idenyi et al. [17] has been used for this analysis.

Results
The results of the corrosion penetration rate values for the experimental and modeled values are as shown in Tables 1 to 8, while Table 9 is the model equation of the corrosion parameters of the various Al-Sn alloy systems in the various concentrations of hydrochloric acid and brine environments.

Discussion of Results
A cursory look at Tables 1-8 clearly reveals that the modeled corrosion penetration values of the various binary alloy samples subjected to the varying concentrations (0.5 M and 1.0 M) of hydrochloric acid and brine environments respectively, are in tandem with the observed experimental values.This is further confirmed by the nearly perfect coefficient of correlations of all the composites which is in the range 0.94 1.00 R ≤ ≤ (see Table 9).The implication of this high positive coefficient of correlation is that the modeled values are in good agreement with the experimental data.Thus, our model can be of good usage in studying effect of the studied environments on Al-Sn binary alloy system on expanded time scale.In order to further confirm the dependence of the corrosion penetration rate mainly on the exposure  penetration rate in the whole environments is accounted for by the corresponding variation in the exposure time.The remaining 7.36% may be due to alloy composition and other factors not incorporated in the model equations.This is overwhelmingly significant and it further confirms that the developed model equations will be a good predictor of the corrosion trend in the various duplex Al-Sn alloy systems being investigated.

Conclusion
The statistical analysis of the corrosion behaviour of Al-Sn duplex alloy system has been investigated.It can be observed that the modeled values of CPR obtained from our model equations correlates well with the experimental data.This is attributed to the nearly perfect positive coefficient of correlation that obtained in the analysis which is consistent for all the alloy compositions.