The effect of microbacterial digestion on Red Oak, White Pine, and WEST SYSTEM(R) 105/206 cured epoxy.

Bruce Niederer and William T. (Chip) Reynolds


Utilizing microbial digestion of petroleum based products for groundwater remediation and the cleaning of contaminated tanks on ships is well known. The microbial breakdown of petroleum based products into water and carbon dioxide has no perceivable negative effect on soils and steel tanks. Less understood is the effect of microbial breakdown on epoxy and wood. The study described below tested the effects of microbial digestion on both hardwood and softwood samples, and cured epoxy. Results indicate that the use of microbial products for cleaning petroleum based products from the bilge of boats constructed with wood/epoxy matrices will not degrade the structural integrity of the wood or the epoxy.

Introduction and Background

Discussion of several examples of oil contamination in bilges of wood and wood/epoxy built boats prompted Rob Munroe, Bruce Niederer, Joe Parker, and Jim Watson, of Gougeon Brothers, Inc., and Chip Reynolds, Captain of the replica vessel Half Moon, to examine several methods of cleaning and decontaminating oily bilges. Promising among the methods considered was the use of oil digesting microbes, a technique used in recent years for groundwater decontamination and for cleaning tanks and ballast water in ships. It is likely that as the use of microbial products expands they will become more widely available to the general public for applications such as the decontamination of bilges. Suppliers of the microbial products tested indicated that the microbes should have no effect on either the wood or epoxy. A quick survey of basic microbial physiology indicated that the microbes in question would use neither the cellulose nor epoxy as food. However, no documentation could be found of tests measuring the impact of microbial digestion on wood or epoxy substrates.

With this background, Parker, Niederer, and Reynolds devised a six week study to test empirically the compatibility of WEST SYSTEM 105/206 cured epoxy with two commercially available oil digesting microbacterial systems. The study also included Red Oak and White Pine samples to test the microbes impact on hard and soft woods that may be exposed in boat bilges. The compression modulus on cylinders of each material were measured before, and at weekly intervals after, exposure to microbial digestion, serving as the method for monitoring and assessing the effects of each product on the test materials.

This report describes the products, the methodology used, a discussion of the statistical analysis , and the conclusions and recommendations derived from the test, A summary of the data, in chart and table format, is included as an appendix.

Product Description


Chemical Specialties International
P.O. Box 738
Cameron Park, CA 95682
contact - Robbie Robinson

This product consists of four parts, two liquids and two salts, which must be weighed and measured, and then mixed with water.

Remediate Liquid

Microbes, Inc.
7876 Stage Hills Blvd.
Suite 107
Bartlett, TN 38133
contact - Jerry Flagg

This product is a one part liquid, packaged in a measured dispensing bottle, that is diluted in water.

Experimental Design and Methodology

The compression modulus of all samples was measured using the MTS equipment in the materials testing lab at Gougeon Brothers. A total of 30 epoxy compression cylinder samples were prepared using 105/206 @ (5 : 1) by weight, with average dimensions of 0.625" dia. x 1.5" long. Wood samples, 24 of each species, were prepared using 1" dia. dowels with a 0.5" dia. centered bore hole x 4.0" in length.

To optimize the environmental conditions most suited for the bacterial colonies to survive and multiply, an oven was maintained at a temperature of ~95oF and 6 Tupperware tubs with ~25 oz. of water were placed inside and allowed to come to temperature. Dissolved oxygen was supplied to each tub via fish tank aerator equipment. Two tubs of each test product were prepared and allowed to sit for 24 hrs, after which time ~ 6 - 10 drops of 10w30 motor oil were added to each, and the solutions were allowed to sit for another 24 hrs. This technique was suggested by the suppliers to allow the bacteria colonies to multiply to effective levels. Two tubs labeled (A) contained the Superbugs, two tubs labeled (B) contained the Remediate Liquid, and the remaining two tubs contained water only to serve as a control.

Following this 48 hr. preparation, 6 samples of each material were placed in each type of bacteria solution and the control water solution. The compression modulus of dry samples were measured at this point. Then, at one week intervals, for a period of six weeks, a sample of each material was pulled from the solutions and allowed to dry at room temperature for at least 4 hours. The samples were then carefully weighed and measured before testing the compression modulus. Additionally, one epoxy cylinder was suspended half way into each bacteria solution and then carefully weighed and measured to monitor changes in dimension and mass. Shore D hardness readings were recorded for each of these cylinders comparing the dry and emmersed ends.

An additional tub of water was kept at experimental temperature in the oven and used to replenish evaporated water in the test solution tubs to avoid thermal shock of the bacteria. Also, ~6 - 8 drops of oil were added each week to provide an adequate food source for the bacteria.

Statistical Analysis

When comparing two populations, it is sometimes appropriate to compare their variances or standard deviations. Because the means of each test group are so close (see tables 1-3) an analysis of means is less appropriate. This experiment measured the affect of one process, the exposure to microbial digestion , applied to the three materials as a function of time, and compared the results to those of control samples of the same materials exposed only to water, a different process. Engineers routinely apply the Hypothesis Test of Two Variances to assess process variation, which looks at the sampling distribution of a ratio between the ( of each sample population, forming a continuous distribution known as the "F" distribution. For these reasons this method was chosen.

For this analysis, a one -sided test was chosen because it more accurately reflects the objectives of the testing. The test hypothesis (Ho) is stated as follows: The variation2 of the test population is less than or equal to the variation2 of the control population. The alternative hypothesis (Ha) is then: The variation2 of the test population is greater than the variation2 of the control population. In making a one-sided test, the numerator of the test statistic (ratio) contains the sample variance corresponding to the larger population variance under the alternative hypothesis. Therefore, the test statistic always places the ( of the test population over that of the control population, regardless of either numerical value.

i.e. : [Ho : s12 < s22 ] and [Ha : s12 > s22], where F = s12/s22 if s12 = test pop. and s22 = control pop.

If the resulting calculated F value (Fc) is less than the theoretical F value (Ft) , found in literature and based on the degrees of freedom [n1 - 1, n2- 1] in each sample population and desired accuracy, then it can be said that there is no significant statistical difference between the compared results.

Statistical Results

Epoxy Ft = 5.05 @ 95% confidence level
Fc(1,A) = 1.09 , Fc(2,A) = 0.94
Fc(1,B) = 1.70 , Fc(2,B) = 1.46

Red Oak Ft = 5.05 @ 95% confidence level
Fc(A) = 1.40 , Fc(B) = 1.07

Wht. Pine F~ = 19.3 @ 95% confidence level*
Fc(A) = 0.91 , Fc(B) = 0.77

*Large Ft value due to small control sample population.

Conclusions and Recommendations

At the end of the six week study, the collected data (see figs. 1 - 3) was summarized, and the statistical analysis described above applied. All the calculated Fc values are less than the theoretical Ft values, therefore, no statistical difference exists between the compression moduli from samples in either bacterial solution and the water control solution. This observation is consistent with information provided by the manufacturers, which indicates that the bacteria would be able to attack the extended carbon backbone structure of a cured epoxy polymer, and would not use the wood as a food source. Based on our tests, we can recommend the use of these products in the bilges of boats coated with WEST SYSTEM(R) epoxy products, in both wood and/or FRP composite construction.

Appendix A

Epoxy Samples (105/206)

Analysis of epoxy samples subjected to microbacterial systems.

Oak Samples

Analysis of oak dowel samples subjected to microbacterial systems.

Pine Samples

Analysis of pine dowel samples subjected to microbacterial systems.

Return to the Preservation Conference Schedule page.


Copyright © 1997, Maritime Park Association
All Rights Reserved
Legal Notices and Privacy Policy
Version 1.01, 7 July 1997