(600ae) Application of Differential Scanning Calorimetry and Mathematical Modeling for the Development of ESTER Phosphate Coatings

For many years, chromate-based coating processes have been used by industry to generate protective coatings on metallic surfaces to ensure the success of subsequent applications.  For example, the automobile industry uses coatings to pre-treat car bodies to allow for the adhesion of paint.  Coatings can also be used to prevent oxidation on the surface of aluminum home siding.  However, the chromates used in these processes generate significant amounts of wastewater and are cause for environmental concern; subsequently, disposal of the process waste is very expensive.  These processes have been subject to strict regulations under the Clean Water Act (Reye et al, 2004).  In this paper,  a practical alternative is developed  using tert-butylated triphenyl phosphate (trade name Durad) with an intermetallic additives (Fe [II] acetate) to generate thin films on an aluminum substrate via Chemical Vapor Deposition (CVD).  Durad was chosen due to its molecular similarity to tri-cresyl phosphate, which was one of the most commonly used industrial lubricants.  However, an isomer of the substance has demonstrated carcinogenic effects in humans, and its use has receded.  It has been replaced by other phosphate esters that demonstrate the ability to promote thin films, which Durad has (Morales et al., 2000). 

Prior research has shown that the films are produced in the kinetically controlled region of the reaction mechanism for the surface reaction between the vapor and the substrate (Nagarajan et al., 2006).  Micro-reaction environments leading to these films are studied experimentally with the use of a Differential Scanning Calorimeter (DSC).  The DSC can be used to collect heat flow and temperature data for a given temperature range and at a specified rate for the heating of a sample.  The heat flow data is then used to correlate thermal data to the conversion of the limiting reactant.  Two methods of kinetic analysis can be used to extract the kinetic parameters (namely, the activation energy, pre-exponential factor and reaction order) of the overall process leading to phosphate-based coatings.  The first method assumes a power law model for the reaction rate.  This model can be combined with the design equation for a non-isothermal semi-batch reactor to correlate the dynamics of the conversion of the limiting reactant with the kinetic parameters.  However, the need to assume a model can be problematic.  It has been demonstrated that some kinetic data can be adequately described by more than one model (Vyazovkin 2006).  Further, this method requires a considerable amount of data filtering.  The uncertainty regarding the usefulness of the points that get discarded can be cause for concern.  An alternative approach is a recent method developed for the quantification of thermal analysis data. This method, frequently referred to as a model-free technique, uses isoconversional analysis and compares the reactant conversion vs. temperature results from different heating rates side-by-side.  This can be used to derive the activation energy as a function of the conversion.  This is unlike the first method, which treats the activation energy as a constant; however, a constant value for the activation energy is likely only valid for single-step gas phase reactions (Vyazovkin 2006).  The pre-exponential factor can also be expressed as a function of the conversion.

Some preliminary results from the first method of analysis has yielded activation energy values ranging between 300-350 kJ/mol and first-order power-law kinetics. Preliminary results from the model-free method of analysis suggest much lower values for the activation energy.  However, the results seem to be affected by large uncertainties, combined with other apparent inconsistencies, indicate that the application of this method of analysis needs to be refined.  One of the problems associated with the experiments described in this paper are associated with  the semi-batch nature of the reaction environment.

Extensions to the model-free method and its adaptation to the coating process described in this paper are presented.  Comparison between the results obtained by both methods is also discussed in detail.  Additionally, previous work was completed by using experimental protocols with linear heating rates. In modulated mode, however, the user can specify a non-linear heating rate to the substance (stair step, saw tooth and sinusoidal incline temperature profiles, for instance).  This approach is believed to present advantages over the standard mode of operation because it allows the operator to account for heat flow driving the reaction independently of that due to phase change or other physical transformation.  This significantly increases the reliability of the results.  With the kinetic parameters derived from DSC experiments, a finite-element based mathematical model is used to simulate proof-of-concept scale-up experiments in a deposition furnace.  Preliminary fluid flow, thermal fields, and mass transport results predicted through computational fluid dynamic simulations will be discussed.

 

 

REFERENCES

Morales, W. and Handschh R.F., “A Preliminary Study on the Vapor/Mist Phase Lubrication of a Spur Gearbox”, Lubrication Engineering, Sept. 2000, vol. 56, no. 9, pp. 14-19.

Nagarajan, A., Garrido, C., Gatica, J.E., Morales, W., “Phosphate Reactions as Mechanisms of High-Temperature Lubrication”, NASA TM,May 2006-214060.

Reye, J.T., McFadden, L.S., Gatica, J.E., Morales, W., “Conversion Coatings for Aluminum Alloys by Chemical Vapor Deposition Mechanisms”, NASA TM, January 2004-212905.

Vyazovkin, S., “Model-Free Kinetics: Staying Free of Multiplying Entities without Necessity”,

            Journal of Thermal Analysis and Calorimetry, 2006, vol. 83, no. 1, pp. 45-51.