Monday, October 20, 2014

Mortar and Brick Testing

Pictures by T. Westfall, Sam Osborne, and Kathryn Young


This weekend, we focused on preparing the brick door opening to begin repair next weekend.  We cleaned out chunks of old mortar and broken bricks lying between us and more healthily bonded bricks.


We also scrubbed the dust out with brushes and sponges.  


Graduate Student T. Westfall scrubs some dusty bricks
Graduate Student Kathryn Young scrubbing bricks
We collected samples of each of the four types of mortar we observed on the wall, crushed them into powder and dropped them into a muriatic acid solution, to see what the binding agent was and what the different types of aggregates were.  Below, some aggregate on my fingers. 


Discussing the mortar experiments with Archaeologist Darlene Applegate and Archaeology Research Student Katie O'Grady

We learned that the mortar we are most confident was applied by the owner had small pebbles of unslaked limestone in it, whereas the best restoration match, for color and texture we could find on the wall did not have these unslaked limestone pieces in it. 


Note the small, white flecks, especially in the upper left corner of the picture.  I originally thought they might be Mussel shells, but Dr. Applegate confirmed they were unslaked lime/limestone.
This best match sample also did not have as much total lime as the original.  We know this, because the chemical reaction (fizzing) when the mortar dust hit the acid was much stronger in the “original” mortar.  We also learned that some of the repairs made to the wall, including ones that placed chunks of broken brick out of sequence with the Flemish Bond pattern, and used gobs of mortar to fill broken brick holes in the wall, were made by an owner, using the old formula.  To the right of my finger, you'll see a gob of limestone-flecked mortar.


A second restoration mortar sample was tested, which matched the texture of the first, but not the color.  It also seemed to be heavier on sand and lighter on binder/plasticizer.  It too had less total lime than the “original” mortar.  The fourth total mortar looked to be at least partially comprised of Portland Cement.  



This sample produced the “cloudy” effect on the water characteristic of cement based mortars, but it also bubbled significantly, at a more steady rate, easily as much or more than the more aesthetically pleasing restoration mortars.  We originally thought this was 20th century mortar applied by an occupant, with Portland Cement.  We now think that this was at least related to the mortar recommended to us by a local specialty brickyard, and by a preservation masonry text (Published by the National Preservation Trust), where standard modern mortar ratios are modified with additional sand, and double the lime content.  We tried 1 part Portland Cement, 3 parts Hydrated Lime, 8-10 parts sand. 

Testing some mortar bonds on cheap bricks.  Its a little sandy.
Not only did this experiment teach us that we don’t want to use the brickyard recommended recipe, but it also taught us that some of the most visually problematic mortars on the house were applied in good faith, with an effort to soften the mortar and preserve the building, regardless of who applied them or when. 

We also prepared our materials: we practiced cutting different kinds of bricks to the correct sizes and shapes we’ll need.  We experimented with different types of mason’s chisels, different cutting angles and rhythms, on different types of bricks.  

T. chops up a brick into various pieces.
The shirt helps cushion the impact and prevent spider-cracking.


We practiced mixing mortar in different consistencies and ratios.  We acquired a second type of sand, in addition to our builder’s sand.  It was dug from a riverbed near Bonneyville and was given to us by Biological Preserve Facility Manager, Curtis.  We think it will be superior to our builder’s sand because it will have less air pockets between grains, as well as softer edges, which will make a softer, more workable mortar.  Additionally, as an aggregate, the sand’s color is closer to the original color. 

When we realized the mixture we were preparing to use definitely had the wrong color and might be too strong, we stopped and cleaned up.  It was time to do some research.


Here’s what we found:

According to Brick Mason/Material Culture Scholar Gerard Lynch, in the UK, much of the historically re-worked recipes for lime-based mortar (with or without Portland Cement as a binder) recommend some form of a 1:3 ratio of lime to sand.  This is consistent with what we found in available literature.  He states that this information was culled from journals, letters, ledgers and other primary source documents without an eye to the actual technological process that was used to create hydrated lime (which is what we intended to use).  He states that prior to WWI hydrated lime was delivered to building sites as baked quicklime and “dry slaked” or “hot slaked” on site to avoid having to carry heavy, expanded, wet hydrated lime.  This meant that the 1:3 ratio reflects lime pre-dry-slaking. (Check out Dr. Lynch's videos on Youtube.) During dry-slaking, lime expands 60%-100%.  So, the final product was significantly larger in volume than what has been the general practice in Europe and the US for re-creating historic lime mortars.  This argument is strongly supported by our muriatic acid analysis, where the oldest lime had significantly more lime content as well as incompletely processed limestone pebbles.  

He also recommends using more mature “fat lime putty” as a plasticizer, which we may experiment with in the coming weeks.  

Thanks for reading.  Let me know if you have questions, concerns or want to come help/watch on Saturday.  

(NS)

1 comment:

  1. Your blog provided us with valuable information to work with. Each & every tips of your post are awesome. Thanks a lot for sharing. Keep blogging.. 6f5 birmingham

    ReplyDelete

Community Open House

Thanks for stopping by the Gardner Historic House blog! If you'd like to visit the actual house, now's your chance.

The Gardner House will be open to the public THIS SATURDAY, September 27, 2014 from 9:00am to 3:00pm. SEE YOU THERE! For more information, go to
or

We look forward to seeing you there!