by Sarah Jackson
Executive Summary
Results from limewash study of wood samples, August 30, 2005
Introduction
This report is an executive summary of testing undertaken by NCPTT in partnership with the Cane River Creole National Historical Park (CARI) and Quality Finish.
In December 2003 meetings were held to discuss applying limewash to the brick cabins at Magnolia Plantation in hopes of prolonging the life of the structures. We developed a study to determine the durability of traditional and modified limewash recipes within certain criteria. We wanted to identify a low cost limewash that would last for five years when applied in approximately three coats As the project progressed we decided to test limewash on wood, since CARI expressed interest in applying limewash to wood structures at Oakland plantation. Epoxy filler is being used at CARI to help stabilize historic wood structures, leading us to add epoxy fillers as a third material of study. In collaboration with its partners, NCPTT designed a program of testing for limewash on brick, wood, and epoxy.
Quality Finish researched possible limewash recipes that may have been used locally by interviewing community members. They were not able to identify recipes used in the community and thus turned to published limewash recipes, including a limewash identified through NPS contracting schedules. We worked together to prepare wood and brick test samples. Starting in November 2004, Quality Finish applied nine separate washes on weathered wood and rough-sawn new wood. On all wood samples, an Edison acrylic emulsion was used as a primer since the limewash alone would not bind to the wood. The same nine recipes plus an additional wash containing lime putty and water were applied to handmade and modern bricks. Table 1 contains information about the limewashes tested. Sarah Jackson, a student intern at NCPTT, manufactured epoxy samples for testing. The limewash recipes that perform the best on wood samples will be applied to epoxy samples and run through the same tests.
As of August 2005, NCPTT has completed scheduled testing on the wood samples. We expect results on limewash on brick samples by mid-October 2005. This report will focus on the results of durability testing of limewash recipes on wood.
Testing Methods
We tested limewashed wood samples using artificial weathering, adhesion, and abrasion testing
according to ASTM standard methods of testing. We photographed samples before and after each test
and monitored them for color change with a Minolta colorimeter. To perform artificial weathering, we
placed the samples in a QUV weatherometer and subjected them to four hours of UV light exposure and
four hours of condensation in the dark continuously alternating for 800 hours. This accelerated
weathering gives us an idea of how the limewash may age over time. For the adhesion test, we cut an
“X” through the limewash to the sample, then placed pressure sensitive tape over the cut. Upon removal
of the tape, we assessed how much limewash was removed. This test evaluates how firmly the
limewash bonds to the samples. In the abrasion test, we placed samples in a holder and poured
measured amounts of falling sand on to the surface to determine how much sand is needed to remove a 4
mm (.1575 inch) diameter circular area of limewash. The abrasion test allows us to rank how a
limewash will stand up to abrasion from wind- and rain-borne particles and from people touching the
buildings. With the help of Quality Finish, we also performed a solids test to determine how much
limewash is being applied to the samples.
Test results
We evaluated limewashed wood samples before and after artificial weathering using visual
appearance, abrasion testing and adhesion testing. We represent test results as an average of the results
from the individual samples for each wash. We prepared three replicates for each wash. Individual
results for each sample are recorded in the appendix that is included with the report. All limewash on
wood performed average at best and completely failed at worst.
For unweathered samples, washes E, H, and F performed the best in adhesion testing. This
means that washes E, H, and F bound most tightly to the wood prior to weathering. Upon abrasion
testing, unweathered samples of washes B, A, and G performed the best. This means that they formed a
harder coating that was more cohesive.
The performance of artificially weathered limewashes was similar to unweathered samples with
some slight variations. After artificial weathering, we visually evaluated each wash and found that
washes D, E, and I outperformed the other limewashes. Also, we found that washes E, G, and F were
better performers in the adhesion tests. When we performed abrasion tests on the weathered
limewashed samples, we found washes B, G, and D were slightly better than other washes.
We devised a ranking system in order to evaluate the results of each test. Based on their
performance, we ranked each limewash from best to worst for change in appearance, adhesion, and
abrasion for samples both before and after weathering. This was a relative ranking from best to worst.
Those with the best performance ranked a 9 and those with the worst performance ranked a 1. Then we
totaled the rankings for each wash.
Based on our ranking system, we found that overall limewash recipe E performed the best out of
the recipes we tested on the wood samples. Wash E is made from Graymont’s Niagara lime putty,
water, casein binder, molasses, and clove oil. Wash E had a better appearance upon artificial
weathering, and performed well in the adhesion tests both before and after weathering. Wash E only
performed in the middle of the group, not being either the best or the worst, in abrasion tests both before
and after weathering. We should note that Wash E had one sample that performed unusually better than
other wash E samples in the abrasion test and was omitted as an outlier.
We found the second best performer was wash G which is made from Graymont’s “Ivory”
hydrated lime, water, and acrylic binder. Wash G performed well in three of the tests -- in abrasion
testing both before and after weathering and in the adhesion test after weathering. However, it should
be noted that wash G produced a powdery surface which lacked cohesiveness. It performed poorly
when judged by appearance after artificial weathering and in adhesion testing before weathering.
We found the third best performer was Wash D which is made from Graymont’s “Ivory”
hydrated lime, water, casein binder, molasses, and clove oil. For the artificial weathering it performed
on par with Wash E and for the other tests it was in the middle of the group.
Recommendations based on results
Based on current analysis of the results, we can state that none of the limewashes tested were
highly durable. Most washes tested display average to marginal performance when applied to wood
based on our laboratory testing. Of the recipes tested, wash E was the best performer and is our
recommendation for use at the Cane River Creole National Historical Park. It performed well in most
tests and maintained good appearance upon artificial weathering.
Alternative limewashes include wash G and wash D. While wash G, which includes an acrylic
binder, was the second best performer, we feel that the recipe is not a true limewash but rather a weak
paint. The powdery nature of the surface observed upon weathering is likely the result of the acrylic
medium encapsulating the lime particles and hindering the ability of the lime to fully carbonate. Wash
D holds a good appearance but is an average performer in adhesion and abrasion testing.
We note that all samples were subjected to color analysis before and after artificial aging. All
limewashes display some color change over time and tend to darken and yellow. Since all washes
changed approximately the same, we did not include these tests in our recommendations.