THE EFFECT OF MOISTURE ON POWDER PROPERTIES
John Jozefowski, Pemco Corporation, USA
Steve Kilczewski, Pemco Corporation, USA
Richard Kmoch, Pemco Corporation, USA
Scott Levy, Pemco Corporation, USA

Abstract
Electrostatic powder performance in a customer's facility is dependent on many factors. The influence of relative humidity will be investigated in two ways.
A powder will be intentionally saturated with moisture to simulate moist circulating air and several parameters will be examined.
Powder room conditions will also be altered to simulate customer environments and properties will again be tested.
Conditions for optimum powder performance will be discussed.

Introduction
The performance of an electrostatic powder in an enameling facility is critical in achieving the high quality and acceptance rates that are sought for the various types of processing lines today.
There are many factors that can detrimentally influence the application of electrostatic powders, primarily upon recirculation. These factors can be difficult at times to isolate and institute proper corrective actions to resolve process line problems and variations.
One of the negative conditions that can cause severe operating problems is high relative humidity in the powder application room.
A series of experiments were designed principally to quantify the effects of intentionally saturating an electrostatic powder under very tightly controlled powder room conditions, and secondly, to develop powder systems that would be minimally affected by varying atmospheric conditions.
Subsequently, specific physical powder properties were measured as a function of varying degrees of powder saturation.
Electrostatic powder users are always recommended to control the atmosphere (relative humidity and temperature) in their powder rooms.
Hence, powder room conditions at most US manufacturing facilities are environmentally controlled as well as the condition of the incoming compressed air used in the powder booths and associated application equipment.

Experiment
Powder performance was measured in a variety of ways. First, ten-minute powder adherence was measured during each experiment according to DP test procedures (1). This test was performed twice on each sample and the results were averaged.
Ten-second powder transfer efficiency was measured using the "Roper method", which calculates the application weight (g/ft2) per unit time(2).
The average of three trials was reported. Average powder fluidity was also calculated using a SAMES Fluidimeter AS100 (3).
Finally, the resistivity of the experimental powder was measured over a thirtyminute period with an Industrial Development Bangor Powder Resistivity Meter Model 465 (4).
Samples of the powder were taken after each experiment to measure the actual moisture saturation of the powder. Free moisture and crystalline moisture were measured using a Leco RMC-100 Rapid Moisture Analyzer.
The ground coat powder used in these experiments was a commercially available dual-purpose black powder for the range industry. A dual-purpose powder has the ability to function as both a base coat and an appearance cover coat. This powder was virgin, as milled material, with an average particle size of 28 microns.
A fifty-pound homogeneous sample was obtained from the manufacturing process line and used for all testing. Baseline powder performance parameters were determined before the experimental testing began.

In order to demonstrate the reaction of electrostatic powder to increased moisture, the laboratory experiments were divided between two aspects.
First, the powder control room's relative humidity was altered while keeping the room temperature constant.
Normal laboratory room conditions are 68 ºF ± 2 ºF (20 ºC ± 1 ºC) and 42% ± 4% relative humidity. While the temperature set point remained constant at 68 ºF (20 ºC), the relative humidity was lowered to 40% and a sample of powder was left exposed in the laboratory testing area for sixteen hours.
This powder was then loaded into the powder spray booth and tested as stated above. The humidity of the room was also raised to 53% while keeping the temperature constant. Another fresh sample of as-milled powder was left exposed for 16 hours and powder properties were again measured.
The second part of the laboratory experiment involved subjecting the powder to moist fluidizing air. This experiment was designed to simulate the effects of humidified compressed air in a customer's plant. The powder room conditions were adjusted to nominal operating conditions. For this experiment, the incoming fluidizing air in the laboratory powder booth hopper was bubbled through a laboratory bomb filled with water. The outgoing fluidizing air was held at approximately 70% relative humidity and routed into the powder hopper. The powder properties mentioned previously were again measured.
After analyzing the data from the beginning of the experiment, it was decided to expand the initial test parameters. A sample of this basecoat powder was taken from a production facility. This sample represented recirculated powder with an average particle size of 22 microns. This powder had been yielding excellent quality and acceptance rates at the customer's facility. Baseline tests were run on this powder before being fluidized in the laboratory powder booth with humidified air. Powder adherence, transfer efficiency, fluidity and resistivity were measured in the same manner as the original powder. These samples were also submitted for moisture analysis.

Results
Leaving the virgin powder exposed to different humidity levels in the powder room did not seem to alter any of the electrostatic powder properties. As can be seen in Table 1, the powder adherence, transfer efficiency, fluidity and resistivity remained relatively constant during these experiments. The results of the free and crystalline moisture analyses did not vary over the course of these experiments.
When the virgin, as manufactured powder was subjected to the humidified fluidizing air, the results were not as expected. As shown in Table 2, the powder prperties were not significantly altered even with exposures of up to 48 hours.
Powder adherence, transfer efficiency, fluidity and resistivity remained relatively stable. The free and crystalline moistures also remained stable, with only a slight increase in moisture content of the powder. figures 1-4 summarize the trends in the powder properties of the virgin powder segment.
The recirculated powder, however, exhibited a much different behavior. These results are shown in Table 3. In particular, powder adherence decreased to zero in only 24 hours. A decrease in resistivity was also measured. Transfer efficiency dropped slightly over the 24-hour test. Fluidity remained relatively unaltered again. figures 5-8 depict the changes in powder properties over the saturation time. As can be seen in Table 3, free moisture values of the powder increased over the 24-hour test period as well.

Discussion
Exposing the virgin powder to varying room humidity did not change the powder properties significantly. Since the room conditions could not be varied that greatly, no real differences in powder performance were seen. The atmospheric control equipment in the powder lab can only vary the humidity in the room over a small range and will hold the humidity level to ± 4%. Furthermore, the size of the room and its relative lack of significant insulation prohibited the required humidity levels from being maintained over the experiments time span.
Even so, the adsorption of moisture into the powder over these relatively small humidity ranges would be slow and no major shift in powder properties would be expected. Small fluctuations in powder room conditions are therefore not expected to significantly alter powder properties. This fact was demonstrated during this experiment.
When the virgin, as manufactured powder was fluidized using humidified air, the results did not replicate the production experience with this specificelectrostatic powder. The as-milled powder absorbed only a slight amount of moisture over the 48-hour period. As a result, the powder properties did not vary significantly.
Problems with powder properties encountered at customer facilities have been documented and attributed to moisture but the results up to this point were not supported by any laboratory experiments. However, these problems usually occurred after a powder had been on-line at a plant for a period of time.
To simulate these conditions a recirculated powder was obtained from a customer's operating powder booth and was included as part of this experimet. The recirculated powder testing did show the results that were reported periodically from operating electrostatic powder processing lines. After fluidizing in humid air for as little as three hours, the recirculated powder showed a considerable loss of powder adherence and resistivity indicating a saturation of the powder with water. The free moisture measurements support these results.
An explanation of the experimental differences between the virgin and recirculated powders can be attributed to the overall shift in the particle size distribution upon recirculation. The original ground coat powder had an average particle size of 28 microns, but upon recirculation, this dropped to approximately 22 microns as shown in Table 4. This shift in particle size represents a change in the specific surface area of the powder from 0.58 square meters per gram for the virgin material to 0.73 square meters per gram for the recirculated powder. This change signifies an increase is total surface area by over 25%. This surface area increase provides additional susceptibility to moisture adsorption, which in turn, decreases the glass system's resistivity and powder adherence.
As shown in figure 9, the loss of powder adherence is rapid and significant.
Powder adherence loss can be attributed to the fact that abrasion and collisions between glass particles may alter the encapsulation of the electrostatic powder. New fractured glass surfaces may be a result of the recirculation process. Since only 30% of the initial material is deposited on the part to be enameled, approximately 70% of the electrostatic powder continues to recirculate during production processing. This material will be subjected to erosion of the encapsulant system.
The resistivity decrease can be attributed to the adsorption of moisture that increases the conductivity of the bulk, recirculated powder. Therefore, the resistivity value decreases with increased exposure time, as water is a very good
electrical conductor.

Conclusion
These experiments confirm that powder properties can be significantly affected upon recirculation. Environmental control in the powder application room is critical for good powder performance. Conditions will vary from one manufacturing facility to another, but there are certain operating procedures that should always be followed to ensure consistent powder properties. The following steps should be used as guidelines to minimize deteriorating performance of electrostatic powders upon recirculation:

Electrostatic powder manufacturers are striving to develop powder systems that will be relatively stable to changes in the customer's operating environment.
If these recommendations are not followed, part yields at customer facilities may potentially decline.

Acknowledgement
The authors would like to thank PEMCO International for the use of their facility in completing this effort.

References
1. Daily procedure Test Manual. DP - 76. "Electrostatic Adherence - Electrostatic Dry Powder".
2. Daily Procedure Test Manual. DP - 84 . " Transfer Efficiency - Electrostatic Dry Powder".
3. Daily Procedure Test Manual. DP - 79. " Fluidity - Electrostatic Dry Powder".
4. Daily Procedure Test Manual. DP - 77. " Volume Resistivity - Electrostatic Dry Powder.

The effect of environments with various relative humidity on the properties of an as manufactured powder. Very little change occurred during these tests

The powder properties of an as manufactured ground coat after being subjected to humid fluidizing air. The virgin powder showed no significant change in properties and only a slight increase in moisture content

The effects of fluidizing with humid air on the properties of a recirculated powder. A significant drop in powder adherence occurred during this test. A slight drop in transfer efficiency and a loss of resistivity also occurred

Particle size and specific surface area as obtained from a Malvern particle size analyzer. Both virgin, as manufactured powder and a recirculated sample were tested

figure 1 - The 10-minute powder adherence of a virgin, as manufactured powder after fluidizing with humid air for up to 48 hours. Powder adherence stayed above the 90% level throughout the test

figure 2 - The transfer efficiency of a virgin, as manufactured powder after fluidizing with humid air for up to 48 hours. The transfer efficiency grew slightly over the 24-hour period

figure 3 - The fluidity of a virgin, as manufactured powder after fluidizing with humid air for up to 48 hours. The fluidity of the virgin powder did not shift significantly during the experiment

figure 4 - The resistivity of a virgin, as manufactured powder after fluidizing with humid air for up to 48 hours. The resistivity remains relatively constant over the 24- hour period. The slight increase in resistivity is within the accuracy of the instrument

figure 5 - The powder adherence of a recirculated powder after fluidizing with humid air for up to 24 hours. Powder adherence declined rapidly when fluidized with humid air

figure 6 - The transfer efficiency of a recirculated powder after fluidizing with humid air for up to 24 hours

figure 7 - The fluidity of a recirculated powder after fluidizing with humid air for up to 24 hours

figure 8 - The resistivity of a recirculated powder after fluidizing with humid air for up to 24 hours

figure 9 - Direct comparison of the powder adherence of virgin and recirculated powders. The powders were both fluidized using humid air for at least 24 hours