Dr. Peter Raynor, at the University of Minnesota, conducted an independent study comparing the real life performance of electrostatic versus mechanical filter media. 

This study was conducted over a 19-week period in 2001 by Dr. Raynor and Soo Jae Chae from the University of Minnesota, School of Public Health, Division of Environmental and Occupational Health and was sponsored by Johns Manville.

Findings of this study show that, in this real-life application, filters made with mechanical fiberglass filter media had significantly better efficiency performance than filters made with electrostatically charged synthetic filter media.

In June of 2001, MERV 14 rigid-pleat filters using fiberglass filter media were installed in a 60,000‑cfm HVAC system in the Basic Sciences and Biomedical Engineering Building on the University of Minnesota campus. In the same building in an adjacent identical 60,000‑cfm HVAC system, were installed MERV 14 rigid-pleat filters using electrostatically-treated synthetic filter media. Each unit contained thirty 24” by 24” by 12" filters. The HVAC systems used 100% outside air from the same source location with no recirculation.

Each unit had thirty 24” x 24” x 2" prefilters. The prefilters were changed as required throughout the 19 weeks of the test using pressure drop as the determining factor.

Dr. Raynor and S. J. Chae developed a method of sampling the air upstream and down steam of the filters so that all of the measurements could be taken while the HVAC systems were in normal operation.  The particulate from the air stream was classified and counted in specific ranges from 0.117 micrometers to 3.05 micrometers.

The first set of data was collected immediately after the filters were installed and the units restarted. The last measurements were taken 134 days later. The HVAC systems were computer controlled responding to the air needs of the building. The pressure drop across the test filters was recorded every hour by the computer control system.

For all particle size ranges, the synthetic filters showed significant decreases in efficiency that continued for the first 12-14 weeks of the study. The mechanical filters maintained a consistent efficiency performance throughout the duration of the test. With 99% of all airborne particles being under 1-micron in size, decreased efficiency in this range should require critical assessment. The coarse fiber filters also dropped in efficiency in particle ranges above 1-micron in size. For example, at a particle size of 1.0 µm the efficiency of the synthetic filter at Day 0 was 93% efficient but after Day 134 the efficiency was 75% after dropping as low as 70%. In comparison the mechanical filters had an efficiency of 93% on Day 0 and remained at this performance level through the test to Day 134.

Before installation, four of each of the filters were weighed and then weighed again after removal 134 days later. The mechanical filters showed a weight gain of 7.1 grams per square foot of media while the synthetic filters removed only 4.8 grams per square foot. The pressure drop of the mechanical filters increased from 0.61 to 0.98 in W.G. while the pressure drop of the synthetic filters increased from 0.66 to 0.95 in W.G. during the test.