Posts Tagged Historic mortar
Lime has been used for thousands of years for building construction as an ingredient in mortar and plasters and limewash. The conversion process by which the material claims its name is from the lime cycle. It is described in most books as a clock face with the corresponding chemical reactions and changes to limestone that take place as it goes through the various cycles.
At the noon position on the clock face we have limestone [calcium carbonate – CaCO3]. As we move around the clock-face toward the three o’clock position we introduce heat. The heat needed to convert limestone to quicklime [calcium oxide – CaO] is 1,650F. At that temperature the CO2 is driven off, water is vaporized into a gas at a much earlier stage in the firing process at 212F. From the three o’clock position moving toward the six o’clock position we introduce water to the quicklime.
This converts the quicklime into lime putty while giving off an exothermic reaction causing the water to boil in a process called lime slaking. The lime putty [calcium hydroxide – Ca(OH)2] settles down into a consistency of thick Philadelphia cream cheese before its ready for use. The quicklime naturally takes the amount of water it needs and drains off the rest, so in a sense you over soak the quicklime during slaking and the material finds its natural balance as it settles down into a putty under the water – a process that generally takes 60 to 90 days to complete if left undisturbed.
Moving now from the six o’clock position to the nine o’clock position we introduce sand, and mix the lime mortar or plaster into a cohesive mixture usually in a volumetric ratio of 1 part lime putty to 2.5 parts sand. We end up at the nine o’clock position with our mixture ready for installation.
From the nine o’clock position back to the noon position we introduce water to the walls [by spray misting] in a series of wetting and drying cycles to encourage carbonation. Carbonation is defined as the process by which lime cures – or converts, back to limestone from which it originated. We suggest a minimum of nine (9) wetting and drying cycles to initiate this process after installation. And that’s it! That is the process of the Lime Cycle. We take limestone apart using fire, mix it back up with water and sand and we have lime-stone mortar in the end – a very durable long-lasting material.
The new term on the streets used by industry consultants to describe the details of how a building takes on water and then (hopefully) sheds it is “water management.” The longevity of historic masonry walls relies heavily on how water is managed in and around them. I am personally not yet convinced we can control water. I can work to manage where it goes, and possibly how long it stays – by redirecting it, but in the end it goes where it wants, the easiest way. When you attempt to fight water it usually wins. The ways water impacts a building depends on how long it stays – which is directly correlated to its architectural design, geographic location, topography, soil, the water table, the type of brick, stone or mortar, and whether the building has recently been restored.
Sometimes the understanding of historic load-bearing masonry walls built with lime mortar materials is not established, or respected, prior to a restoration project being undertaken. While the joints may look like they are in need of repointing due to deterioration, it’s important to know why they deteriorated in the first place. The cause is most likely from water saturation – then freezing and thawing or extreme temperature variations. One of the challenges is understanding that a building can, and does, breathe though its mortar joints as well as its masonry units.
The shear thickness of most load-bearing masonry walls keep the water out. The original building materials made for quick evaporation of the water on the surface of the walls and kept the inside dry, but this breathability does takes its toll on old lime mortar joints and they need to be repointed in high moisture areas every 75 to 100 years or so. Problems start when an architect specifies a replacement mortar that is harder than the original (in an effort to make it last longer) than potentially traps moisture inside the wall system. The effort in the restoration repairs is totally focused on keeping water out from coming in through the exterior side of the wall. The problem is that old masonry walls contain a certain amount of moisture already and often do not perform very well with harder/stronger mortar joints surrounding them.
When the goal of the restoration project is to create a Watertight Envelope you’d better run the other way – fast. “Watertight Envelope” and “Historic Load-bearing Masonry” should not be used in the same sentence. Keeping water on the outside would seem to be an honorable goal for any restoration project, but observing the current condition of some masonry buildings restored in the past 10 years tells us a much different story, a troubling one. Basically, the buildings subjected to this watertight-envelope theory are not doing very well.
Where waterproofing and harder cement-based mortars are applied we find decay patterns that are surprising – in just a decade after application. Instead of the mortar surfaces wearing, there is a new pattern of brick and stone decay. Strong osmotic and hydrostatic pressures build up in brick and stone that are subjected to these hard, strong, and water resistant materials.
Tomorrow we will discuss how water enters a building.
As the market increasingly becomes aware of the use of building lime for historic masonry restoration there will always be challenges in making sure everyone understands the decisions they are making, why, and most importantly, the materials they are working with. Take lime for instance, everyone seems to believe that going back to the old mixes of yesteryear is a better choice than that found in Isle 14 at the local Home Depot when it comes to mortar selection for historic masonry structures. But just knowing about a subject and really understanding a subject are two entirely different things. The product of lime is pretty basic. You have lime putty, purchased wet (Philadelphia cream cheese consistency) in a bucket or barrel, and you have dry hydrate lime purchased in a 50lbs. (fifty pound) bags (fluffy and very light weight).
Common cement/lime mortar mix formulations in the restoration industry center around 1:1:6; 1:2:9; and 1:3:12 (Type N, O, and K respectively- ASTM C270-10, proportion specification). The second numeral reflects the amount of lime to be added to the formulation to create the desired mortar and thus the characteristics of that mortar. Generally, a mortar with more lime will tend to have better workability, higher flexural bond and more autogenious healing properties than a mortar with less. If its compressive strength your after than 1:1:6 is your answer, if you are looking for the flexibility to accommodate for future movement than you will likely be happy with a 1:2:9 or 1:3:12 formula. And then of course there is the historic straight 1:2.5 lime-sand mortar almost always made with lime putty and not dry hydrate lime, let me explain one of the reasons why.
Lime, like portland cement, is measured as a dry powder when mixing individual ingredients at the job site. Small batches of mortar are mixed from opened bags using a coffee can or some other used drinking cup (seven-eleven big gulp cup works good) up to a five gallon bucket depending of the project needs. But here’s the real scoop – Dry hydrate lime experiences a significant volumetric loss when converted to a wet paste during mixing. Let me say that again, Dry hydrate lime experiences a significant volumetric loss when converted to a wet paste during mixing. Volume changes that occur when dry hydrated lime is converted to a wet paste can cause sizable errors in proportioning mortar formulations; the most likely error is over-sanding.
A given amount of hydrated lime occupies far more volume as a dry powder than it does after mixing with water. Thus, when lime is measured as a dry powder, less is actually put into the mixture than is used if the lime is measured as putty. When wetted, dry hydrate lime will typically contract, on average, to 75% of the original dry volume. Using a nominal 1:2:9 mixture (Type O) cement/lime/sand, the variation caused by wet verses dry measure of the lime results in a 1:1.5:9 mixture. This ratio exceeds the allowable sand content in ASTM C270 of 2.5 to 3 times the binder, and is actually 3.6 times the cement plus lime; thus an unintended over-sanded mixture results. To avoid this problem an additional amount of dry hydrate lime (25%) must be added to all formulations during the proportioning stage, or just use lime putty. Note: Portland cement does not experience this volumetric loss when converted to a wet paste during mixing.
Phillips, Morgan, A Source of Confusion about Mortar Formulas, APT Bulletin 1993 http://www.jstor.org/pss/1504465
I often get asked this time of year, “How late in the construction season can we work with lime-sand mortars?” Well the quick and fast answer is 45 days before the first hard frost. Which means you should have your projects wrapping up by the end of October at the latest just to be safe. The reason for this safe period (as suggested by lime mortar manufacturers) is because of the way lime-sand mortars initially cure, by carbonation – absorbing CO2 back into the material through wetting and drying cycles. Most specifications call for a minimum of nine (9) wetting and drying cycles of misting the walls down with generous amounts of water after installation and allowing them to dry out naturally – drawing in the CO2 from the evaporation of the water. Obviously, this wet-curing process can become problematic during freezing temperatures.
The Brick Industry Association, March 1992, Technical Note 1 states, “Mortar which freezes is not as weather-resistant or as watertight as a mortar that has not been frozen. Furthermore, significant reductions in compressive and bond strength may occur. Mortar having a water content over 6 to 8 percent of the total volume will experience disruptive expansive forces if frozen due to the increase in volume of water when it is converted to ice. Thus, the bond between the unit and the mortar may be damaged or destroyed.”
But what if your schedule backs you up against old man winter and you have no choice but to work into November? Don’t lose heart. We carried out some testing 8 years ago on a project in Chicago [Lime Mortars, Two Recent Case Studies, Ed A. Gerns and Joshua Freedland] to find out how late in the season you could repoint a building using lime-sand mortars. Trials were conducted at various times in the fall and early winter at approximately six week intervals. The last installation occurred 48 hours before the first frost on November 23, 2003. Observations during installation, following initial curing, and periodically through the winter and following year were noted. The high-calcium lime putty and sand mortar showed no signs of shrinkage cracks, the bond between the mortar and brick units was well-adhered, and no erosion or cyclic freeze-thaw damage was observed. We were working with butter joint brickwork of 1/8 inch mortar joints.
To supplement the insitu testing, limited concurrent laboratory testing was conducted to evaluate the depth of carbonation and the impact of freezing temperatures had on the depth and rate of carbonation of the lime. Two inch mortar cubes were made from the same mortar formulation (1:2.5) – no additional water was added. The mortar cubes (eight sets) each went under freezing temperatures once for a four-hour duration at various times after initial mixing. The exposure to freezing temperatures (10F) was established at 24 hrs; 48hrs; 72hrs; 96hrs; 1 week and 2 weeks. The mortar cubes were then broken at various times and the depth of carbonation was measured using a phenolthalein solution as an indicator.
After 2 weeks, the depth of carbonation suggested that mortar cubes that were exposed to a freezing temperature for a limited duration during the first week had less depth of carbonation than the cubes that did not experience any freezing temperatures. After three weeks, however, this difference was no longer observable. All mortar cubes seemed to equalize after three weeks. Interesting. The success of this trial may be a result of the forgiving nature of lime-sand mortars, the low water content of the repointing mortar, and the narrow joint width of this particular project.
This paper was presented at the 2005 International Building Lime Symposium in Orlando, Florida. Proceedings are available on CD (ISBN 0-9767621-0-2). The CD includes 39 papers by authors from 10 countries. Also, included are several important historical documents related to building lime–some as old as 1920.
The conference proceedings are available for $25 at: http://www.lime.org/documents/publications/free_downloads/summary-ibls-2005.pdf
Best way I know of to match a historic mortar is to first identify whether you are dealing with a cement-based binder or a straight lime binder. You can do this by dissolving a sample of original historic mortar in a solution of hydrochloric acid and some water. Watch the reaction of the material as the solution makes its first contact. It’s best to place the material into the solution rather than the solution into the material for best results. If it is a lime binder the material will break down quickly and form foam at the top of the material, bubbling and hissing as the calcium breaks down. If the solution just sits there with no reaction, only a few bubbles – but know foam or hissing action you have a portland cement mortar.
Allow the materials to soak in this solution until all the binder materials are gone. You can check and see this visually by looking at what is left – the aggregate of the mortar. It should be clean and free of any particles of binder still attached. Lime binder mortar can dissolve quickly, sometimes in a matter of hours. Cement binders can take up to several days to dissolve down. Now comes the fun part – identifying the aggregate or sand in the mortar.
Drain off the solution through filter paper to collect the fines. Dry the material in an oven at around 200F on a hot tray. Weigh the material into an even gram amount then run the sand through the series of ASTM E11 sieves specified through ASTM C144 and calculate the percentage of grain particles on each sieve. Then create a sand gradation chart which depicts to sand particle size, shape and color to make for an easy way to match the material.
Chances are if you match the sand and binder materials in a historic mortar you should not need to assistance that oxide pigments can provide. However, these materials are used in the industry to assist masons in matching mortars regularly. We typically stick with a 1 part binder to 2.5 parts sand in our formulas as this was the standard in the industry for the exception of butterjoint brickwork which is often a mixture of one part binder to one part sand by volume.
The performance characteristics of a historic mortar; bond strength, flexibility, breathability, vapor permeability, and compressive strength (in that order of importance) will typically fall into place if the necessary time and testing of the original material is carried out. Most historic lime mortars are very durable and well carbonated and worth replicating. It is not best practice to trump a historic mortar formulation and go to the next higher mortar type, i.e., Type L to a Type O for example. It is best to fix the problem of why the historic mortar deteriorated in the first place – more than likely a water related issue.
Testing masonry materials for durability and performance has been going on for some time. It is important to study and test materials prior to construction of an actual wall to prevent wasted material, time and labor. Some of the earliest recorded tests performed on masonry mortar ingredients were carried out by the U.S. Army Corps of Engineers in the construction of fortifications during the early part of the 19th century.
The durability of masonry buildings relied heavily upon the past performance of the actual structures and the master mason’s experience with the individual materials. Much of the heritage knowledge of making good mortar was passed down from generation to generation through the trades. Testing mortar ingredients historically involved the masons working with the architects in a team approach for common understanding. However, signs of change began to appear as early as the 1890s.
Uriah Cummings writes in his book, “American Cements,” which was first published in 1898. “With their former teaching and experience on the one hand, and the testing machine on the other, the question was not long in doubt. The machine was victorious, and henceforth all judgment founded on experience was laid aside and they became blind believers in the tensile strain tests. What matter though they were continually befogged by the frequent, unreasonable, and capricious pranks of the machine, they had found a god, and were determined to worship it. And so it came to be established as a fixed belief among engineers and architects that the best cement was the one which tested the highest, and the manufacturer had no alternative but to strive to make his product test as high as possible.”
Seems from the tone of Mr. Cummings writing that he knew the industry was going in the wrong direction toward high compression. Is it high compression that destroys historic masonry? Well indirectly it does. Most mortars that are very high in compressive strength are very low in vapor permeability. The ability a mortar has to capture and release water easily through evaporation. What tends to happen in a historic masonry wall is moisture infiltrates by various ways; rising damp, poor roof/parapet/flashing details; driving rain; capillary action through cracks among other ways. The water needs to escape from inside the walls through the mortar joints ideally keeping everything dry from the inside out. Hard, high strength mortar prevents water from escaping thus trapping it inside the wall potentially causing damage to the masonry units of brick and stone as well as terra cotta over the course of time. It’s always better to insist on a lower compressive strength lime mortar that readily breathes with the masonry allowing quick evaporation of water, and in addition, provides the natural flexibility needed for traditional load-bearing masonry walls to perform at their best.
Ever wonder why portland cement gets so hard? Why it is so high in compressive strength? This blog post is a continuation from an earlier one from last week titled “Hot Rocks” which most people seemed to enjoy. Early lime kilns could operate only up to around 1600F — just hot enough to activate the clay, turn the limestone (calcium carbonate) to quicklime (calcium oxide) and combine them both to form belite (dicalcium oxide) to make hydraulic lime.
Joseph Aspdin (1824) had achieved a slightly higher temperature of around 2200F forming some liquid phase and combining the belite with the remaining quicklime to create alite (tricalcium silicate), the base compound for portland cement. The vertical shaft kilns could burn slightly hotter, but temperatures were mainly kept below the maximum heat achievable to limit the liquid phase and prevent the danger of a kiln blockage, a situation created when the molten mass of materials cooled within the kiln. Kiln blockage was something kiln operators historically and understandably feared; crawling down into a shaft kiln and chopping away with a hammer and chisel at a molten mass of material was to be avoided at all costs.
The rotary kiln, however, not only helped to overcome the kiln blockage challenge, but it also allowed the higher temperatures necessary to produce portland cement with a much higher compressive strength. Before the rotary kiln was invented (1889) most kilns were constructed vertically, loaded with alternating layers of limestone and fuel (wood and/or coal or both) until the materials reached the top. A fire was started at the base of the kiln and allowed to burn for a day or two depending on how large the kiln was. The entire kiln was cooled down another few days until the materials were discharged from the bottom. The process would start over again for the next batch, a very labor intensive process. The rotary kiln also allowed for continuous feeding of the kiln without the usual starting and stopping and could run 365 days per year without interruption in production.
So just how hard is portland cement? Well, in a paper presented at the American Lime Conference in Lynchburg, Virginia in March 2003, Paul Livesey of Castle Cement presented the following information about the answer to this question. When matching a “portland cement” mortar from the period between 1871 and 1920 then, we should not be confused by the terminology. Portland cement of 1871 was a different material from that of 1920 which, in turn, is totally different from portland cement today.
Portland cement products produced at the turn of the last century were fired at temperatures from 2300F to 2600F. In comparison manufacturers making portland cement in this century fire at temperatures ranging from 2800F to 3000F. The differences in mineralogy and the effect that burning at higher temperatures has on strength development are well known. For example, in 1871, portland cement tested in the 1800 psi range; today’s portland cement are in the range of 8,000 psi, an increase of 344 percent in compressive strength. Indeed, the portland cement of 1871 bears more relation to modern, higher-strength hydraulic lime (NHL 5, 1000 to 1500 psi) than it does to modern portland cement.
And when making decisions on the appropriate mortar match for historic buildings remember that the National Park Service as well as leading American experts agree that it is always better to err on the side of a lower strength mortar replacement in order to protect the historic masonry materials. Even adding a small amount of modern portland cement can have a significant affect of increasing strength when maybe you did not intend to in the first place.
It was David O. Saylor back in 1871 that produced the first portland cement in the United States in Lehigh Valley, Pennsylvania. Between 1871 and 1920, American portland cement production skyrocketed, due in part to the increasing demand for automobiles and the attendant need for roadways during this period. The years 1871 to 1920 also saw a major change from “traditional” manufacture of cement to a more technically aware, science-based industry.
As a consequence, it was inevitable that cement products that reflected both traditional and modern scientific production methods were on the market at the same time. Thus, anyone looking to match a historic portland cement mortar from the time period between 1871 and 1920 will benefit from considering the evolution of cement production technology during this period.
As popularity of portland cement grew over the period, so did its compressive strength, from 1800 psi in 1871 to 3000 psi by 1920, an astounding 110 percent increase. An examination of the limestone and clay used to produce portland cement, however, shows that they changed very little after 1871. What, then, had changed? Mainly it was the production process and the ability to fire the raw materials to consistently higher temperature.
The main technological breakthrough came with the invention of the rotary kiln, which was first used commercially in Lehigh Valley, Pennsylvania in 1889. Initial trials by Ransome and Stoke in England had lacked the necessary financial backing to succeed. In the end, it was left to Seaman and Hurry in the United States to make the final technical refinements that could produce sufficient temperatures and efficiency to unleash the massive portland cement industry we know today.
The traditional volume mix design of 1 part lime putty to 3 parts sand may be insidious to follow straight up without more details. First, mix designs historically used quicklime as the 1 part of lime mixed to the 3 parts of sand by volume. Quicklime when it is slaked with water will increase volumetrically 70-100 percent – or basically double its size. This fact would reduce the sand content closer to that of a 1 part lime to 1.5 parts sand – a much sticker richer mix design.
Secondly, sand must be measured in a damp loose condition according to ASTM C270 when mixing mortar. Dried sand will bulk up to 30 percent and grow volumetrically by the addition of a small amount of water. This can send your mortar mix designs at the construction site off the specified requirements.
I recommend everyone take a moment to read Gerard Lynch’s article on the subject it is well done. http://www.buildingconservation.com/articles/mythmix/mythmix.htm
Ever wonder how Type N mortar came to be? or Type M for that matter? Well the story goes something like this…In 1931 a group of non-mortar producers and representatives from the lime and cement industries got together and formed a committee to discuss the issue.
The issue was that mortar “types” needed to be established to distinguish high compressive strength mortars from soft flexible ones, so in 1944 the designations using A-1 (2,500 psi); A-2 (1,800 psi); B (750 psi); C (350 psi); & D (75 psi) were adopted, with minimum compressive strength requirements specified.
In the United States, “A-1” had become synonymous with “the best” or “top quality” and some committee members felt that the designation for the higher compressive strength cement mortar was misleading. The possibility did exist that an architect desired a flexible lime mortar type for a particular project, but he might mistakenly specify the A-1 type, thinking it was the best. In an effort to avoid confusion on the subject, the committee adopted a new mortar type designation in 1954.
The new designation letters were taken from the two words, MASON WORK utilizing every other letter. The compressive strength minimums for each mortar type are still recognized in the current ASTM mortar specification C270.
(2,500 psi) Type M replaced A-1
(1,800 psi) Type S replaced A-2
(750 psi) Type N replaced B
(350 psi) Type O replaced C
(75 psi) Type K replaced D
Most historic load-bearing masonry buildings have original mortars with low compressive strength, but yet are very durable (well carbonated lime mortar). We have plenty of architectural inventory around the world that supports this statement. High compressive strength in historic masonry mortar (Type O or higher) is not a direct reflection of durability and maximum life-cycle performance.
In fact, to give you some perspective, a certain material scientist/university professor studied historic mortar for his entire career. Traveling the world he collected samples from some of the oldest historic masonry structures. Very seldom did he ever run across a historic mortar with compressive strength of over 300 psi.
As you climb the scale from Type K upward, you are adding more and more portland cement by volume. As a result, the mortar becomes less permeable, less breathable, and more inflexible in exchange for the increased compression. Historic masonry on the other hand needs mortars to accommodate building movement (flexibility), exchange moisture readily from the face of the wall (breathability), and most of all have excellent bond strength-all natural properties of lime mortar (“Type L” introduced in 1998, ref. NPS Preservation Brief 2).
Preservation Brief 2, “Repointing Mortar Joints in Historic Masonry Buildings” http://www.cr.nps.gov/hps/tps/briefs/brief02.htm
The History of Masonry Mortar in America 1720-1995 http://www.lulu.com/product/paperback/the-history-of-masonry-mortar-in-america-1720-1995/11271764