The results of the abrasion and adhesion testing before and after artificial weathering are presented in this section. Solids tests were compared for each of the limewash samples. In addition, changes in appearance of the limewashes were recorded by colorimetry and photographic documentation. Test results were represented as an unweighted average of the results from the individual samples for each wash.
For each test except the artificial weathering, three replicates were prepared of each wash and the results averaged. Due to space limitations in the QUV, the artificial weathering was performed in duplicate. A ranking system was devised to evaluate the results of each test, and each limewash was ranked from best to worst for relative change in appearance, adhesion, and abrasion for samples both before and after weathering. Depending upon the number of limewashes for the sample, the rankings varied from 1 (worst) to 10 to 13 (best) (Table 2). The ranking was based on the unweighted averages of the result from each test. The ratings where two washes are grouped together are representative of washes with the same overall rating.
|Table 1. Limewash Recipes Showing Ingredients Used|
|Handmade Brick||Modern Brick||Weathered and
Rough-sawn New Wood
|Best 13||Wash A||Best 9||Wash B||Best 12||Wash E||Best 3||Wash E|
|12||Wash K||8||Wash D||11||Wash G||2||Wash D|
|11||Wash M||7||Wash A & K||10||Wash D||Worst 1||Wash G|
|10||Wash D||6||9||Wash A & I|
|9||Wash C||5||Wash E||8|
|8||Wash B||4||Wash F & M||7||Wash B & H|
|6||Wash L||2||Wash G||5||Wash F|
|5||Wash I||Worst 1||Wash C & I||4||Wash C|
|4||Wash H||3||Wash L|
|3||Wash F||2||Wash M|
|2||Wash E||Worst 1||Wash N|
The results of abrasion testing on all limewashed samples of handmade brick are presented in Figure 3. These results compare abrasion testing before and after artificial weathering. In all cases the limewash performed better before artificial weathering except for wash M, which performed better after artificial weathering. Washes that include a salt additive required the most volume of sand to abrade through to the substrate before artificial weathering. After artificial weathering, all limewashes performed markedly worse, with the exception of wash M, which performed more than twice as well. Washes A, B, L, and M required similar volumes of sand abrasion after artificial weathering. Limewashed modern-brick samples performed similarly to handmade brick.
The results of abrasion testing on limewashed wood samples, including both weathered and rough-sawn new wood, are presented in Figure 4. It should be noted that limewash was flaking off samples of washes C, D, and E before testing began. All limewashes were poor performers on wood substrates both before and after artificial weathering. None of the washes withstood more than 5 liters of sand abrasion. Several samples from washes L, M, and N retained insufficient limewash after artificial weathering to perform abrasion testing. The samples that were tested from washes L, M, and N took less than 250 milliliters to abrade to the wood substrate. Epoxy samples had results similar to the wood in the abrasion tests.
Figure 5 presents adhesion results on historic handmade brick. All limewashes performed similarly before and after artificial weathering except wash M, which performed better after artificial weathering. Before artificial weathering washes D, F, and H were well rated. After artificial weathering, wash M was the best rated, followed closely by washes C, E, G, and K. Washes D through G had powdering surfaces that made it more difficult to perform adhesion tests. In many of these adhesion tests there was little consistency between replicates, leading to a large standard deviation.
On the wood samples all washes performed similarly in the adhesion testing before and after artificial weathering. On most samples the limewash was beginning to flake off before testing, and there was not a solid, cohesive coat to remove with the tape. The best performers, washes E, F, and G, had a rating average in the middle of the scale and experienced between 1⁄16 inch and 1⁄8 inch of loss along the incision. Before artificial weathering wash A received an average rating and ranked with the best performers. After artificial weathering wash A was rated significantly lower, near the bottom of the group. The rest of the washes averaged a rating between 0A and 1A both before and after artificial weathering. The migration of salt through the brick samples during artificial weathering is one likely cause for the poor performance of wash A in tests after artificial weathering.
Handmade-brick and modern-brick samples performed exceptionally well during artificial weathering. All recipes on brick samples were rated 4A or 5A, the top rankings, and had an excellent appearance after artificial weathering. However, a marked difference could be seen in the performance of all washes on the adhesion and abrasion tests before and after artificial weathering.
There was a noticeable failure of the limewash on numerous weathered-wood and rough-sawn new wood samples during artificial weathering (Fig. 6). Washes D, E, and I were the only recipes that had a rating average above 4A. Washes F and G had the next highest average ratings but a large standard deviation. For all recipes applied to wood samples, artificial weathering generally removed limewash from the peaks of the grain on the weathered samples, leaving limewash remaining in the valleys. This may be a result of the valleys in the grain being created by the less dense spring growth that erodes faster than the harder, denser summer growth. 14 The lower density wood in the valleys in the grain or the valley itself may have provided assistance in the adhesion of the limewash.
On all materials washes A, B, and C showed the highest solids deposit, which may be a result of the salt additive (Fig. 7). Washes from the same recipe tended to have similar solids deposit regardless of the lime used. Washes G, H, and I, with the acrylic-emulsion additive, showed the lowest solids deposit. Washes D, E, and F, with the casein binder, had the second-highest solids deposit. Washes L, M, and N, which did not have additives, had solids deposits similar in amount to washes G, H, and I. The total color difference was calculated based on data from the Minolta colorimeter before and after artificial weathering. 15 Results were similar for all samples where limewash remained after testing. Samples where limewash remained after artificial weathering often showed lighter results than before weathering. One sample from wash G on the wood substrate had a drastic color change that was the result of a tan spot that developed during artificial weathering. It is unclear whether the tan spot was the result of tannins in the wood migrating through the limewash or a reaction of one of the additives.
During the visual inspection and documentation of samples before testing, crystallization was observed on washes A, B, and C. These washes had a salt additive. Samples were examined under a Leica MZ 8 stereomicroscope to confirm the crystallization (Fig. 8). In the stereomicroscope photographs, the crystals are readily apparent as raised, discolored material differing in composition from the limewash itself. These crystals could be either salt or sugar (from the molasses additive), since those were the only constituents capable of such crystallization. Their presence suggests that the additive did not become a cohesive part of the limewash matrix upon drying. Furthermore, since both salt and sugar are highly soluble, the crystals would be lost upon exposure to water, disrupting the matrix significantly and weakening the limewash.
During the artificial weathering, a white residue began to appear on the unexposed back of the modern brick samples from washes A, B, and C (Fig. 9). Using a Keymaster TRACer III Portable XRF for X-ray fluorescence analysis, the residue was studied. The results showed that the residue contained chlorine. The limewash on the surface, which was tested after artificial weathering, had almost no trace of chlorine. Thus, the chlorine on the backside of the samples suggests that the salt migrated from the limewash through the modern brick samples.
Originally published in APT BULLETIN: JOURNAL OF PRESERVATION TECHNOLOGY / 38:2-3, 2007
1. Laura Soulliere Gates, email to author, Aug. 17, 2006.
2. National Park Service Technical Information Center, ‘Class C’ Cost Estimating Guide: Historic Preservation and Stabilization (Denver: Denver Service Center, 1993), 18.
3. Colin Mitchell Rose, Traditional Paints, available from http://www.buildingconservation.com/articles
4. Abbott Lowell Cummings and Richard M. Candee, “Colonial and Federal America: Accounts of Early Painting Practices” in Paint in America: The Colors of Historic Buildings 14 (New York: Wiley, 1994), 14.
5. Scottish Lime Centre, Technical Advice Note 15: External Lime Coatings on Traditional Buildings (Edinburgh: Historic Scotland, 2001).
7. John Ashurst and Nicola Ashurst, Mortars, Plasters, and Renders, vol. 3 of English Heritage Technical Handbook (Great Britain: Gower, 1995), 47.
8. Roger W. Moss, “Nineteenth-Century Paints: A Documentary Approach” in Paint in America: The Colors of Historic Buildings (New York: Wiley, 1994), 55.
9. ASTM Subcommittee D01.24, Standard Test Methods for Viscosity by Ford Viscosity Cup, ASTM D 1200-94 (West Conshohocken, Pa.: ASTM, 1996).
10. Marcy Frantom, email to author, Sept. 12, 2005.
11. ASTM Subcommittee D01.23, Standard Test Methods for Abrasion Resistance of Organic Coatings by Falling Abrasive, ASTM D 968-93 (West Conshohocken, Pa.: ASTM, 1996).
12. ASTM Subcommittee D01.23, Standard Test Methods for Measuring Adhesion by Tape Test, ASTM D 3359-95 (West Conshohocken, Pa.: ASTM, 1996).
13. ASTM Subcommittee D01.27, Standard Practice for Conducting Tests on Paint and Related Coatings and Materials Using a Fluorescent UV-Condensation Light- and Water- Exposure Apparatus, ASTM D 4587-91 (West Conshohocken, Pa.: ASTM, 1996).
14. Pete Sotos, conversation with author, Nov. 15, 2006.
15. Ruth Johnston-Feller, Color Science in the Examination of Museum Objects: Nondestructive Procedures (Los Angeles: Getty Conservation Institute, 2001), 35.
16. L. Franke and I. Schumann, “Causes and Mechanisms of Decay of Historic Brick Buildings in Northern Germany,” in Conservation of Historic Brick Structures, ed. N. S. Baer, S. Fitz, and R. A. Livingston (Shaftsbury: Donhead, 1998), 26-34.
SARAH MARIE JACKSON joined NCPTT in 2005 as a graduate intern to continue the testing for the limewash study. In 2006 she accepted a permanent position with the Architecture and Engineering Program at NCPTT. She received a master’s degree in historic preservation from the Savannah College of Art and Design.
TYE BOTTING is a research staff member at the Institute for Defense Analyses. He served as the NCPTT/NSU joint faculty researcher for three years. He holds a PhD in nuclear chemistry from Texas A&M University, where he did post-doctoral work in nuclear engineering.
MARY STRIEGEL is responsible for NCPTT’s Materials Research Program, focusing on evaluation of preservation treatments for preventing damage to cultural resources. She also directs investigation of preservation treatments geared towards cemeteries and develops seminars and workshops nationwide. She holds a PhD in inorganic chemistry from Washington University in St. Louis.