PORTLAND CEMENT, CONCRETE AND MORTARS
Portland cement mortars are the most universally and effectively used materials for ruins stabilization in the western United States. The use of lime mortars is more prevalent in the East. Portland has a long history of use in stabilization. It was employed by Cosmos Mindeleff at Casa Grande as early as 1891, in mortar to set brickwork foundations and patches. It was used there again during the winters of 1906 through 1908 by J. W. Fewkes, in plaster coating to prevent erosion at the bases of walls. At about this time the first heavy concrete capping was placed on the Sun Temple in 1915.
Concrete has great resistance to weathering when properly made and placed in thick sections, or in thinner sections upon an unyielding material.
Up-to-date mixes of portland cement give compression strengths of 4 to 6 thousand pounds per square inch, and mortar strengths (concrete without the coarse aggregate) of 3 to 4 thousand pounds per square inch. These strengths are far in excess of any compressive strains that will be encountered in ruins stabilization.
Concrete can be colored almost any desired tint, either by the addition of dry mortar colors or by stains. It is also possible to achieve a desired tint by selecting sand or an aggregate with certain color characteristics, or by using naturally colored cements.
While various textures, ranging from glassy smooth to rough, can be obtained with concrete, it is practically impossible to match a soil mortar texture because of the size of the sand grains.
Concrete shrinks somewhat upon setting, and is increased appreciably by a high water-to-cement ratio, and by too rapid curing.
The tensile strength of concrete is very low. It is a very brittle material, an attribute that is generally not realized, or often ignored. Specifications for tensile strengths of mortars (portland cement and sand for use in setting stone) call for strengths of from 275 to 350 pounds per square inch. As Bauer states: "The question is frequently raised as to the value of a tension test when the concrete rarely ever is considered as having any tensile strength." Lack of such strength is the cause of most concrete failures, particularly the cracking of very thin capping.
Resistance to Moisture
Ordinary concrete or mortar is not particularly water-resistant. Even good concrete will absorb up to 20 percent of its weight in water. This absorption accounts for the fact that ordinary concrete used in the bases of walls, or as foundations, will not prevent the capillary rise of moisture into the wall above. In ordinary commercial work, waterproofing of concrete is accomplished in various ways:
1. With the use of a special waterproofing agent, usually tannin, added to the cement at the time of manufacture
2. Addition, at the time of mixing, of hydrated lime or waterproofing compounds such as emulsified asphalts.
3. The use of mixes richer than those not intended to retain water, plus smaller amounts of mixing water. Restricting the mixing water will result in a much denser concrete; the longer concrete is moist-cured, the more dense and water-resistant it will be.
In most commercial work, however, concrete or mortar that must be waterproofed is treated after laying. Basement walls are protected by porous fill and drains, and then mopped with hot asphalt or painted with various materials intended to seal the exterior surface. The tops of masonry walls (stone, brick, tile, cinder blocks, etc.) laid in concrete mortars are either coated with asphalt or covered with asphalt-impregnated cloth, fiber glass, or paper, or with a combination of these materials. With these restrictions, then, the waterproofing of concrete and mortar for stabilization is practically confined to the addition of such compounds as emulsified asphalts at the time of mixing, or by the use of special waterproof cements.
Field Control of Concrete and Mortar
The design of concrete mixes and the drawing of specifications for cement and aggregates are functions of the branch of engineering of the particular agency engaged in the work, and that branch should be consulted for specifications. The Portland Cement Association, the American Concrete Institute, and the Building Research Institute also cooperate in disseminating building information. However, the data to follow on proportions and mixes are supplied for the field man who may not have technical assistance readily available for minor jobs.
The strength of concrete or cement mortar is affected by a number of factors. Among these are the ratio of aggregate-to-cement, ratio of water-to-cement, length of mixing, gradation of the aggregate, and length of moist-curing. Of all these factors, the ratio of water-to-cement is most important. Few supervisors and workmen seem to realize this. Strength curves for various water-cement proportions should be examined (fig. 6) as they show the rapid loss of strength as the ratio of water to cement increases.
In most specifications, the water-to-cement ratio is a controlling factor in design. Table 1, taken from a handbook, The Design and Control of Concrete Mixtures, issued by the Portland Cement Association, is intended as a reference for the construction of concrete piles, thin walls, light structural members, and exterior beams and columns.
TABLE 1. Optimum water-to-cement ratios for various climates and conditions.
Trial mixes may be made in the field to determine proportions of sand and cement for use with a given water-to-cement ratio chosen for required strength. A weighed or known quantity of aggregate is used, and when a suitable mix has been obtained, the unused material is weighed or measured and the quantity required for the mix is determined. To conduct a trial mix, a set of scales and a measure or box of 1/10th cubic foot capacity are needed.
For example, let us suppose that the concrete or mortar employed in stabilization will be laid in thin sections and will often be exposed to extreme weather, but will not be in constant contact with water. A ratio of 6 gallons of water to the sack is chosen. Following is an outline for the preparation of a trial mix:
Standard sized sacks of cement weigh 94 pounds and have a volume of 1 cubic foot. One gallon of water weighs 8.33 pounds. Fill the 1/10th-cubic-foot measure with cement. The cement should weigh 9.4 pounds.
Measure out, say 4 volumes of sand and weigh them. In this example, call the weight 32 pounds.
Six gallons of water to the sack would be 6 x 8.33 or 49.98 pounds of water for a whole sack. Water required for 1/10th sack is 4.998 or roughly 5 pounds.
Mix the 5 pounds of water with the 1 volume of cement (1/10th cubic foot or 9.4 pounds). The water-to-cement ratio for the mix is now established and no more water should be added.
To this mix add some of the 32 pounds of sand previously measured out. Continue to add sand until a workable mix is obtained.
At this point weigh and measure the sand that remains of the unused 4 volumes or 32 pounds.
If the sand remaining measures 1-1/4th volumes, the mix used at 6 gallons of water to the sack is, by volume, cement part, sand 2-3/4ths parts.
If the weight of the remaining sand is, say 10 pounds, the mix is 1 part cement to 2-2/3rds parts sand by weight. With this ratio established, larger mixes may be made, always based on a portion of a sack of cement.
For 1/3rd of a sack of cement mix (a convenient size to use in the smaller mixers): cement, 1/3rd sack; water, 2 gallons; sand, 9/10ths cubic foot (at 1 part cement to 2-3/4ths parts sand by volume).
The field worker should be aware that an increasing number of gray portland cements have appeared on the market in recent years, ranging from Types I to III and from Types IA through IIIA. These have been created by demands of industry and technology. Type I is a general purpose cement suitable for most purposes. Type II cement is intended for use in mass concrete where control of hydration heat is necessary. It is recommended for use in structures of considerable size, where cement of moderate hydration heat tends to minimize rises in temperature. Such structures include dams, large piers, heavy abutments and retaining walls. Type III is a high, early strength cement designed for concrete which must be put into service rapidly, and is used in industrial and highway maintenance. Types IA-IIIA include a regular Type I portland cement into which an air entraining agent is integrally ground. This provides resistance to freezing and thawing, and prevents scaling action of de-icing salts and chemicals used in snow and ice removal. These cements are recommended for pavements, highways, etc. In addition to gray portland cements of six or eight types, several kinds of white cement have been developed. Finally, there is a waterproof gray cement known in the Southwest as plastic cement.
The important point to remember is that Type I portland gray cement should be used for most purposes under discussion here. Exceptions include masonry cements discussed below which, although forming a class of their own, are simply derivatives of portland.
Both Trinity white cement and a new, natural tan colored product (known by the trade name Warmtone, manufactured by the Trinity Portland Cement Company) have definite uses in stabilization, particularly in stuccos and plasters. It should be kept in mind, however, that in the restoration of a mortar with a high lime content, a lime mortar should be used. Differing coefficients of expansion and contraction between portland and lime mortars make the combination inadvisable for pointing or patching.
Masonry cement is a combination of portland and hydrated lime, usually in proportions of one-to-one, packaged in 1-cubic-foot bags weighing from 65 to 80 pounds.
The use of masonry cement is more widespread than straight portland cement mortars in commercial setting of brick, tile, masonry blocks, etc. The major reason for this is that mortars containing lime are much more workable than portland cement mortars. They are easier to handle and they trowel much better. Furthermore, masonry cement is cheaper. However, masonry cements are not as strong as portland cement mortars in tension or compression, although they should not be ruled out for pointing. Masonry cements are much less resistant to abrasion and should not be employed in capping or sections that may receive some visitor traffic. They may be colored with dry mortar colors in the same manner as portland cement.
The principal objection of masonry cements for use in capping is their lack of resistance to moisture penetration. Since manufacturers of waterproofing materials do not recommend their products for inclusion in lime-mortars, there seems to be no way to overcome this limitation. Therefore lime-mortar cements are not recommended for capping. However, recent experiments show that the water-repellent capacity of masonry cement can be improved by adding a waterproofed gray cement (Plastic, Medusa or equivalent). The ratio may be one part in five of the total mix, or equal parts of masonry cement and waterproofed gray cement, with three parts of sand (1-1-3). It should be mixed dry and the aggregate should contain just enough water for troweling. Because of their advantageous characteristics and slightly lower cost, these mortars should prove of value in the setting of large patches where very slight shrinkage and good bond with stone are needed.
Mixes of lime-cement mortars are usually made by volume and not by a water-to-cement ratio. Most common mixes call for one part masonry cement, and two or three parts fine sand, with sufficient water to make a workable mortar.
Soil-cement is a mixture of portland with a suitable local soil brought to a workable consistency by the addition of water. The amount of cement varies from 5 to 20 percent by volume of the finished project. The primary reasons for using soil-cement in lieu of other mortars in stabilization work are 1) cost, 2) the extreme difficulty in obtaining concrete sand in some locations, 3) appearance in some special applications, and 4) suitability for use in monolithic soil structures.
A comparison of the qualities of soil-cement intended for ruins stabilization with the same material in its various commercial applications is difficult. Soil-cement used for roads, airport runways and hard standings, ditch linings, etc., is always laid fairly dry (at optimum moisture) and is heavily compacted. The compactive effort in such applications is equivalent to the force exerted by a 5-1/2 pound tamper dropped 75 times a distance of 12 inches upon 1/30th of a cubic foot of the material. Obviously, such requirements of moisture and compaction cannot be obtained when laying mortar. All material published on soil-cement is directed toward these commercial applications, and there is no known research on its use as a mortar.
The earliest use of soil-cement in stabilization appears to be its employment by J. F. Motz in 1934. He used it as a mortar between stone at Wupatki National Monument during a Civil Works Administration project. It was also used in the 1935 repairs at Tuzigoot in another WPA project supervised by Edward H. Spicer and Louis C. Caywood. While no records of this work are available, and it is not possible to determine precisely the type of mix used, the soil-cement samples at both areas show a rich mixture of cement (not less than 20 percent by volume) and an admixture of red mortar color. Some of this material is still in place at Tuzigoot after 37 years, attesting to its durability if properly made.
The first controlled tests of soil-cement made expressly for stabilization purposes were those conducted by A. E. Buchenberg between December 1941 and May 1946 (Buchenberg, 1951). These were carried on at Wupatki National Monument, and were simply outdoor exposure tests made on 6 x 10 x 3/4th-inch briquettes of the mortar. It should be noted that they were made by mortar samples only and did not include samples of masonry laid with this material, and that the briquettes were made in a semi-compacted state, a condition that is not possible to obtain when laying mortar.
At the conclusion of the 4-year period of testing, Buchenberg found the material satisfactory. "A mixture of soil and cement stands up very well under exposure to climatic conditions, with an approximate admixture of from 15% to 20% cement..." (Buchenberg, 1946) A series of tests was carried on later at Chaco Canyon to determine the durability of soil-cement mixtures under conditions of wetting and drying, and freezing and thawing, and the strength of the bond between mortar and stone under the same conditions.
Only two general types of soils in the tests were available for use in stabilization in the area. One variety, classed as Soil A in the notes, is a heavy, dark-colored and poorly drained clay, 64 percent combined silt and clay. It appears to fall into USBPR Classes A-6 and A-7. The other, classed as Soil B, is a silt-loam, rather fine, poorly graded, 18 percent combined silt and clay, probably USBPR Class A-5. Both soils may be considered as fine-grained under the Unified Soil Classification System. (See: Earth Manual: a Guide to the Use of Soils as Foundations and as Construction Materials for Hydraulic Structures, U.S. Department of the Interior, Bureau of Outdoor Recreation, 1963, Washington D.C.) Soils containing more than 50 percent visible particles are classed as coarse-grained, while those containing less are fine-grained. For classification purposes, the No. 200 sieve (0.074 mm., or .003-inch) will separate the fine- and coarse-grained types. Particles of .003-inch size are about the smallest that can be seen by the unaided eye.
Mixes of varying cement content were made above optimum moisture in these tests, and at a consistency which could be troweled. The wet-dry tests were run on molded cylinders of soil-cement approximately 1/30th cubic foot in volume. Test cylinders were moist cured for seven days after molding. Each soil-cement cylinder was placed under water at room temperature for five hours. They were then dried for 42 hours, which completed one cycle. Although 12 cycles are considered a complete test, the cylinders in question were run through 22 cycles. Standard wet-dry tests for soil-cement specified by the American Society for Testing Materials (ASTM Designation D-559-44) require that each cylinder be wire brushed at the end of each cycle, the cylinder weighed, and the soil-cement loss computed. Lack of suitable balances prevented the completion of this part of the test.
The same kind of cylinders, cured the same length of time, were used in the freeze-thaw test (fig. 7). After curing, the soil-cement cylinders were placed on absorptive pads which extended into a pain of water. Thus, the soil-cement had a constant supply of moisture and there was no limit to the amount available for absorption. Soil-cement and portland will absorb large quantities of moisture (up to 20 percent) and it is useless to try a freeze-thaw test without making moisture available.
Standard procedures (ASTM Designation D-560-44) require that the cylinders be frozen for 22 hours at -10° F. This requirement could not be followed in the field. No freezing cabinet was available, so the cylinders were frozen outdoors at temperatures ranging downward to -24° F. After freezing, the cylinders were thawed for the required 22 hours, the completion of one cycle.
No tests were known to determine the strength of bond between rock and mortar, so the following procedure was carried out in the field: After the wet-dry and freeze-thaw cylinders were made, two average sized rocks were dampened, the mortar spread roughly on one, and the second rock tapped lightly into place upon the first. Thickness of the mortar between the rocks averaged 3/8ths inch. These were cured for seven days. The mortar specimens were tested for resistance to wetting and drying for six cycles along with and in the same manner as the cylinders. It was found that this soil-cement was far more resistant to wetting and drying than to freezing and thawing. At this point, the wet-dry test on mortar was stopped and the rock samples with mortar between were transferred to the freeze-thaw test. Thus, all the mortar samples passed six cycles of wetting and drying before undergoing the freeze-thaw test. A summary of the results of 10 exemplary tests is presented in table 2 and in the following observations:
1. Mixes of straight Soil A (adobe) were very hard to work and tended to crack. They were disregarded and do not appear in the tabulation of results.
2. As little as 10 percent cement by volume gave satisfactory resistance to wetting-drying cycles. All cylinder samples at or above 10 per cent cement passed 22 cycles without failure, and six cycles without destroying the bond between rock and mortar.
3. Cement in the amount of at least 14 percent for cylinder samples was required to resist the action of repeated freezing and thawing. Cylinders at cement content less than 14 percent disintegrated at the third cycle. Observation of the cylinders undergoing tests indicates that no amount of cement will prevent the soil-cement from absorbing moisture either under freeze-thaw or wet-dry conditions. Cement strength must be high enough to resist the force of freezing after the moisture has entered the specimen. Moisture rise in a cylinder 4 inches high was 2 to 3 inches. When cylinders failed under freezing, there was no gradual wearing away of the exterior, but rather a breaking up of the entire area that had absorbed moisture
4. Freeze-thaw tests on bond were far more severe than either freeze-thaw or wet-dry tests on cylinders. Thus, samples which pass a freeze-thaw test may not retain a strong bond with stone when subjected to weathering. This failure is probably attributable to the fact that some of the moisture entering the soil-cement is absorbed by clay particles which causes the entire mass to expand and break the bond.
Note should be taken of samples 54, 55, and 56 (table 2). Soil B, which contains only 18 percent combined silt and clay was mixed with equal proportions of concrete sand. These samples exhibited the greatest resistance to the freeze-thaw test on mortar bond. They still absorbed moisture, but it caused no appreciable weakening of the bond. The clay content was reduced to 9 percent which evidently is low enough to prevent much change in the volume of the mass upon repeated freezing and thawing.
5. Mixes containing large amounts of clay were less stable than those with less clay and more sand. Twenty percent cement with clay produced no more strength than 14 percent with a lower clay content.
6. In all cases where bond failed, the edges of the mortar remained hard and sharp. Failure of the bond does not necessarily mean that the soil-cement has disintegrated.
7. The addition of a bitumen to samples 23 through 31 demonstrates that the additives do not increase the moisture resistance of soil-cements, but merely compound the problem of tinting.
TABLE 2. Results of 10 exemplary soil-cement tests.
Compressive and Tensile Strength
Soil-cement is usually tested by its reaction to the wet-dry and freeze-thaw tests, and seldom by compression. No tests for tension have been found. Recorded compression tests indicate 28-day strengths of from 200 to 1,078 pounds per square inch. This is considerably less compressive strength than is developed by concrete mortar, but under all normal circumstances it is sufficient.
A. E. Buchenberg made a suitable mix without the addition of mortar color at Wupatki National Monument by carefully selected, dark red soils. However, in most areas the addition of standard mortar or cement colors will be required to obtain proper tint.
The texture of soil-cement is good. It more nearly matches the texture of prehistoric mortar than does either cement or lime-cement mortars.
Soil-cement is no more moisture resistant than concrete or cement mortars. In the cylinders tested, there was a moisture rise of from 2 to 3 inches in cylinders 4 inches high. The strength of the cement in the mix prevents disintegration with such high moisture absorption.
Soil-cement is not recommended as a mortar for setting stone unless its use is unavoidable in remote areas or necessary to simulate original material. In these instances the cement content must be unusually high, 15 to 20 percent, to insure the best possible bond. The more sandy soils should be selected. Because of its good texture, soil-cement is particularly adapted to applications where large amounts of mortar must be left exposed; these include heavy grouting of exposed longitudinal sections where facing is not replaced (figs. 8 and 9).
Soil-cement is particularly suited to large repair sections in monolithic soil structures, and is also valuable in stabilizing the sides or ends of cuts in excavations which must be left open for interpretive purposes (fig. 10).
Procedures for making optimum moisture and density tests for determination of cement requirements are not reproduced here inasmuch as they require equipment and data not generally available in field areas. Detailed instructions may be found in two publications of the Portland Cement Association: Soil Cement Mixtures; Laboratory Handbook, and Testing and Construction Criteria for Soil-Cement for Highway Ditch Linings, Levee Faces and Similar Structures; and Manual on Sampling and Testing for Construction Control, Airport Pavings, Roads and Streets, available from the U.S. Corps of Engineers.
Field Control of Soil-Cement
In general practice, the cement content for soil-cement must be based on the percent of cement by volume to the volume of the completed product. It is also standard practice to make trial cylinders of the mix at varying cement content, and with several available soils, subject these test cylinders to standard ASTM wet-dry and freeze-thaw tests before establishing a mix for any particular job. The following information on cement requirements is suggested as a basis for conducting field tests of soil-cement mixtures: (Tests should be made well in advance of actual work.)
Sandy soil, well graded: 8, 10, and 12 percent cement will harden 79 percent of these soils. A few will require cement volumes of 19 percent or more.
Coarse, little binder: 12 and 14 percent cement will harden 73 percent of these soils. A few may require up to 18 percent cement.
Silt soil: 12, 14, and 16 percent cement will harden 69 percent of these soils, while 28 percent will require 18 percent or over.
Clay soils: 14, 16, and 18 percent cement is required to harden most of these soils, while a few will need cement content as high as 21 per cent.
A definite trend is observable for the cement requirements to increase as the silt and clay content of the soil rises. Heavier clay soils not only require higher cement content, but are much harder to handle and place than are the lighter and more sandy types; these heavy soils should be avoided if at all possible.
A simple test to determine classification and gradation of soil may be made as follows:
1. Fill a straight-sided jar about one-third full of earth which has been screened by a No. 4 sieve (about 1/4th inch).
2. Add water to fill jar about two-thirds full.
3. Cover jar and shake vigorously until all of the earth is in suspension.
4. Allow earth to settle until the various particle-size divisions are visible (about 30 minutes).
The coarser material, No. 4 (about 1/4th inch) to No. 10 (about 3/32nd inch) will settle to the bottom (figs. 11, 12). The medium-sized material, No. 10 (about 3/32nd inch) to No. 40 (about 1/64th inch) will then settle. The fine material and colloids, No. 40 (about 1/64th inch) to No. 200 (about 3/1000th inch) and smaller, will rise to the top.
A good distribution of all particle sizes from large to small indicates a well graded soil. The soil is poorly graded if all particles are uniform in size, or if there is an absence of intermediate sizes. If the visible particles (up to 200 sieve) make up more than 50 percent of the sample, the soil is coarse-grained. Soils containing less than 50 percent visible particles are fine-grained. If the sample being tested is a fine-grained soil, coarse sand (No. 4 to No. 10) must be added to make up 50 percent of the visible particles to provide a suitable soil for making soil-cement. Soil samples should then be tested with varying percentages of cement to determine optimum mix. If tinting is required to match proposed work, cement color should be admixed during these tests.
Soil-cement may be mixed by hand for small jobs or in a standard plaster mixer with rotating blades. It is difficult to handle in a drum type mixer, and cannot be worked to advantage unless in a plastic condition. It should be resilient from the beginning to keep the workmen from adding uncontrolled amounts of water. It need not be so thin as to be "buttered on" when used as a mortar, but should be plastic enough so that when the stone is set and tapped in place, the mortar will flow to meet the inequalities on the surface of the stone. The weakest part of such construction is the bond. The stone should be well dampened to aid the mortar in adhering, and to prevent the stone from absorbing excess moisture from the mortar. It takes at least seven days of damp curing for soil-cement to setthe same as other cement mixes. The same precautions should be taken in curing soil-cement as are taken with portland or masonry cements. Keep it damp under wet earth, sacks, or other covering.
Soil-bitumen, or asphalt stabilized soil, is a mixture of emulsified asphalt and soil with sufficient mixing water for the purpose intended. Upon evaporation of the mixing water, the soil particles are left covered with a thin film of bitumen. There is no chemical action involved, nor does the bitumen act as a binder, glue, or other cementing agent. In all soil-bitumen mixes, the strength of the resulting product depends solely upon the cohesive action or quality of the clay particles in the soil.
In the commercial field, the use of soil-bitumen has been limited to the manufacture of bricks for building construction. One point that is brought out in specifications for buildings of this type is that manufacturers of emulsified bitumens do not recommend the use of their product on mortar for setting stabilized bricks or adobes. They recommend only the use of a sand-concrete mortar without the addition of lime. On the other hand, experiments conducted by the U.S. Office of Indian Affairs with stabilized adobes indicate that a soil-bitumen mortar, used in the same proportions as that for adobes, may be successfully employed (Hubbel, 1943, p. 91).
The first extensive use of soil-bitumen in ruins stabilization was made by the C.C.C. Mobile Unit at Chaco Canyon National Monument in 1937. Its use was dictated by funds quite limited in proportion to the number of main-days available for labor. Since the termination of that program, the use of soil-bitumen has been limited to small areas of plating, walkways and similar applications.
Properly made and laid soil-bitumen projects are durable as demonstrated by jobs at Pueblo Bonito, Wijiji, Kin Klizhin and Aztec Ruins, completed from 1937 to 1940. With a few exceptions which can be traced to poor technique, this work is standing up well. Its useful life without repairs is estimated at 25 to 30 years. Where unsatisfactory results have occurred, often in applications not connected with stabilization such as plaster, roof platings, thin walks, etc., failure can be attributed to three primary causes: A durable soil-bitumen mix cannot be made where the soil contains more than 2 percent of soluble salts. Lack of proper tests for salts resulted in several failures. Improper amounts of bitumen, in an excess over that necessary to waterproof the clay particles in the soil, results in excessive shrinkage and a total lack of bond. Improper applications result in failure, as in thin plating and roof coverings where it was mistakenly presumed that the bitumen would provide more strength than that inherent in the soil.
Failures of soil-bitumen in ruins stabilization are seen in these instances: 1) where it was used in very thin "membrane capping" (failures were the result of cracking and lack of bond, not loss of waterproofing); 2) very thin vertical patches at the sides of doorways where there was lack of sufficient bond with the remaining wall (some failures were also due to excessive use of bitumen in the mix); and 3) vertical sections and patches at corners in doorways, etc., which were subject to heavy visitor traffic.
The most satisfactory employment of soil-bitumen is seen in: 1) heavy capping where three to five courses of masonry were relaid in the mortar (most effective in thick core-type walls where there is an excess of mortar over stone); 2) in areas of large patches; and 3) in repairs made to correct basal erosion in damp areas. Soil-bitumen mortars are far more moisture resistant than other types.
Present day uses of soil-bitumen are rather limited and no extensive data are presented here. Details on testing, strengths, water permeability, selection of soils, etc., can be found in several publications, including Specifications, Bitadobe Brick and Building Conforming to Minimum Requirements Approved by Pacific Coast Building Officials Conference (supplied by American Bitumuls Co., 200 Bush Street, San Francisco, Calif.); Technical Paper No. 53, Hydropel Emulsified Asphalt, issued by the American Bitumuls Company; Bulletin BMS 78, Structural, Heat Transfer and Water Permeability Properties of Five Earth Wall Constructions, National Bureau of Standards, Washington, D.C.; and in an unpublished manuscript, "Interim Report on Experimental Ruin Stabilization at Wupatki National Monument (Buchenberg, Ms., 1946) on file, Arizona Archeological Center, National Park Service, Tucson, Ariz.)
The first requisite of soil or soil-bitumen mortars is that it contain not more than 2 percent soluble salts. If there is doubt, the soil should be tested by a competent laboratory because the deleterious effects of the salts will not become apparent until some time after the mortar has been laid. If it is satisfactory in this respect, the next requirement is that it contain from 30 to 45 percent clay, or combined silt and clay, which will pass a 200-mesh screen. Should the only soils obtainable contain too large a percentage of clay, they may be tempered by the addition of sand.
Trial bricks can be made using 4 to 8 percent bitumen by weight to roughly determine the best proportions of bitumen to be added. After a drying period of two weeks, the bricks can be examined for shrinkage and cracking. Those that are satisfactory in this respect can be further tested by soaking in water, or subjected to a stream of water to determine their resistance to erosion. In general, the most satisfactory mixes will be those which are water-resistant but do not contain an excess of bitumen.
During hot, dry weather, soil-bitumen capping and plating will require some protection to prevent too rapid drying and resultant cracking. Probably the best and easiest covering to apply is damp soil, preferably a light sandy loam that dampens easily. It can be applied after the surface has become firm and it need not be removed.
Emulsified asphalts are added to concrete and cement mortars at the time of mixing to reduce moisture absorption. They are intended for use in foundations and mortars; manufacturers claim a reduction in moisture absorption of from 75 to 85 percent. Claims are also made for increased workability and consequent reduction in the water-cement ratio.
Experiments have been made with a heavy asphalt called "Floor Mastic Binder" intended for use in floors and foundations. It was incorporated in concrete used as a foundation for damp kiva walls. Two drawbacks were encountered: First, it was extremely hard to handle; it gave the concrete a molasses-like consistency that could not be poured, shoveled or troweled. Secondly, it was coal-black, which required that it be well disguised with soil pointing. It would appear to be unsatisfactory as a mortar.
The emulsified asphalt known by the trade name of Hydropel and intended specifically for use in mortars was satisfactorily employed in damp walls in Chettro Kettle during 1947. It was also used successfully at the large West Ruin, Aztec, for the construction of curtain walls along the foundations of rooms where capillary moisture and erosion of basal stones were creating severe problems and threatened the collapse of upper wall sections. Its color is somewhat dark, and it appears to be unaffected by mortar colors incorporated in the mix. This is a minor inconvenience and, if necessary, can be overcome by pointing with soil or other native mortars.
Concrete mortars should be tinted with dry earth colors to properly blend with the remainder of the structure. Correct shades of mortar or cement colors are often difficult to obtain from local sources. Color charts may be obtained from manufacturers and the materials ordered well in advance of the intended starting date. Note should be made that the colors are certified lime proof, non-fading, and that 85 percent will pass a 200-mesh screen.
Samples should be made up in advance of actual work to test the reaction of the color to the particular cement in use; some brands of cement are harder to color than others. Color and texture of the sand aggregate affect the final product very noticeably. If possible, sand that matches the ruin should be selected. The color of the dry mix is a fair guide to the appearance of the completed mix, but in any event do not use more than 10 percent color since it may affect the strength of the concrete.
Add the color to the dry cement first. It is most convenient to mix a quantity of the cement and color in a large can or half barrel. It does not matter too much if all the various batches are not exactly the same shade; in fact, it may be necessary, where much mortar is exposed, to change tints often to duplicate changes in the original material.
The problem of bonding new mortar or concrete to old surfaces so that it will remain permanently in place has been the subject of considerable research and development, particularly in the decade 1958-1968. Only two companies and their products will be mentioned, since excellent results have been obtained with their products under field conditions. However, this discussion is not intended to denigrate other firms which may be manufacturing similar products of equal merit.
In the commercial field, it was found that a straight portland cement grout would provide an adequate bond between the new and old concrete, but only if the new section were thick and heavy, if edges were square cut, and if the entire patch area were roughened and cleaned. This requires considerable time and effort. The grout, too, must be mixed and applied very carefully. Too much moisture in the grout causes shrinkage and subsequent cracking; too little results in bond failure. The topping has to be applied when the grout hardens to the proper point of setting. This laborious, complicated, and time consuming method was necessary to secure an integral bond between new and old concrete. Even with these precautions, failures were all too common. Under many conditions, it became evident that portland cement grout by itself could not provide an adequate bond.
In an attempt to solve this problem, the first bonding agents were developed, i.e., rubber latex, polyvinyl acetate emulsions, and compounds with an asphalt base. But natural rubber latexes are affected by alkalies present in cement, and therefore tend to break down. They also oxidize in the presence of air. The use of bituminous materials resulted in a resilient surface but had scarcely any advantage as bonding agents.
Chemists approached the problem from the most promising of these directions. It was well known that polyvinyl acetates have excellent adhesive properties and, in emulsion form, are compatible with all mortars and concretes. They are relatively inexpensive. Unfortunately, these materials are also normally water soluble. They tend to re-emulsify and weaken when continuously exposed to moisture, as in outdoor applications or on cellar walls. Chemists then discovered a method whereby high polymer resins can be internally plasticized to withstand exposure to moisture and vapors without re-emulsifying, yet retain their superior adhesive qualities. Daraweld is such a commercial product. When properly used, it will not soften or disintegrate because of moisture, and can be used in damp areas, both indoors and outdoors. It will also stop most leaks in below-grade masonry. The coupling action of colloidal resin particles is so strong in Daraweld grouts and mortars that they will bond not only to concrete and masonry, but to tile, wood, steel and glass. Containers and tools used in handling Daraweld must be cleaned with water immediately after use. Mixed in proportions ranging from 1:1 to 1:4 with water, in the usual grouting or pointing mortar, it has been used very effectively in numerous ruins repair jobs. Also, where thin patches, overlays or tinted plasters are employed, it can be mixed with water and soil to form a very durable soil mortar.
Daraweld mixed 1:1 with water, sprayed or brushed on an adobe or soil mortar surface such as the eroding front portions of Montezuma Castle and followed by patching or plastering with soil mortar which matches the original, will provide an excellent bond for the new work. In 1960 this same mixture was applied as a spray to the floor of an early pithouse site, an interpretive archeological exhibit, near Montezuma Well. This treatment has been a major factor in prolonging the preservation of a fragile exhibit.
Daraweld is manufactured by the Dewey and Almy Chemical Division of the W. R. Grace and Company, with offices in Cambridge, Mass.; Chicago, Ill.; San Leandro, Calif.; and Montreal, Canada. It is sold in 1-, 5-, 29-, and 54-gallon drums, available through prominent construction supply firms in most of the larger cities.
The Dow Chemical Company of Midland, Mich., manufactures a Dow Latex 560, which, added to cement mortar, provides increased adhesive or bonding strength, better compressive, tensile, and flexural strength, and improved alkali and dilute acid resistance. This material provided an excellent bond and the necessary strength for a stone masonry retaining wall anchored to bedrock on a 60 percent slope in Mummy Cave, Canyon de Chelly.
Most modern treatments for the preservation of wood are based upon poisons such as penta-chlorphenol which prevents the growth of fungi. They may be used in stabilization for preservation of replaced timbers, or for the treatment of wood members already in place. Some of the newer preservatives, sold in gallon lots or more, contain both pentachlorphenol and silicone, thus providing combined protection against fungi and moisture. Some preparations are sold in concentrated form and are designed to be mixed with petroleum oils. Kerosene and very light fuel oils or napthas have been found satisfactory. It is suggested that prior to actual use, tests be made with the wood involved. Some preparations will produce "blooming," the appearance of iridescent colors on the surface. Since this is apparently due to the petroleum dilutent, various types can be tried and those with the best appearance employed. A word of caution: when mixing and handling these preparations, extreme care must be taken to protect the eyes and face. Limited contact with the hands is not serious, but prolonged contact must be avoided.
New material or beams which have been removed and are to be replaced can best be treated by soaking. Timbers in place are harder to handle. Exposed parts can be painted in several applications, but the portions of any timbers which need treatment most are those embedded in walls or other parts of the structure. The most satisfactory solution though not perfect is to drill a 1/2 or 3/4ths-inch hole on an angle into the embedded end and keep this full of solution until it has soaked through to the surface, plugging the hole after the treatment.
Silica aerogel, an excellent non-toxic powder, has been used effectively in recent years to treat ceiling and roof timbers in situ which were heavily infested with, powder post beetles and other insects. A special rubber syringe with adaptable nozzles, filled with powder, is used to dust the borings and larvae runs. The powder causes death by suffocation.
The long history of waterproofing agents applied to prehistoric structures, particularly those of soil construction at Casa Grande National Monument, is reviewed elsewhere in this volume. The history was one of failure. Failure in almost every instance was due to the fact that the waterproofing compounds always formed an impervious skin or surface. Moisture absorption from the soil below the wall, differential expansion of the pervious surface and the wall interior, and other factors forced a separation of the treated surface from the remainder of the wall. The damage caused by loss of the treated surface was often greater than that from natural weathering. While these waterproofings were suitable for such hard surfaces as brick or limestone, they were not applicable in the treatment of soil walls. Additional difficulty was encountered where treatments discolored the aboriginal surface or gave it a sheen or gloss.
Silicone waterproofings developed for above-grade masonry in the 1960's (late spinoffs from American technical developments in World War II) provide a water repellent surface and avoid many of the difficulties encountered with the older type. Their effectiveness is based on their peculiar surface structure. They do not form an impervious skin. Rather, when applied to masonry and other semiporous material, their surface becomes a network of small open cells or pores which repel water. It also permits moisture, entrapped within the wall, to evaporate through this honeycomb surface. As a result, there has been no separation of treated surface from the remainder of the wall in structures which were treated as long as 10 years ago. Furthermore, the silicone preparations do not form a glossy or noticeable surface. The treated surface will appear somewhat darker for some time after application, but disappears within a few weeks.
Silicone water repellents do not harden surfaces to which they are applied. If the surface is friable before treatment, it will remain so following treatment. Should the repellents be applied to areas that are brushed by heavy visitor traffic or otherwise abraded, regular inspection must be made to insure that the treatment is not lost. Inspections made during a rain will instantly disclose areas on which the repellent is no longer effective.
Silicone Types and Content
The basic silicones for water repellents are made by a few large manufacturers. They are then processed by other manufacturers into a bewildering array of masonry and other water repellents. Of the types which have been tested for use on monolithic soil or masonry structures, only those which have a hydrocarbon base or vehicle are approved. Other types, with a water base and caustic action, may prove satisfactory but have not as yet undergone sufficient testing to be recommended.
When testing of silicone products began, a number of producers were contacted regarding the guaranteed silicone content of their products. One manufacturer assured a silicone content of not less than 5 percent and not more than 8 percent. This was the highest percentage guaranteed, although one manufacturer offered to produce a water repellent at any desired silicone content. As a result of these inquiries, all testing and application of silicone water repellents in the Southwest has been made with material having a silicone content of not less than 5 percent. It should be made clear, however, that a 5 percent content will be too high for very low porosity substances such as granite, and that the silicone content should be adjusted to the porosity of the material on which it is used.
To date, applications of silicone water repellents have been made on the following structures: south wall of the "Big House" at Casa Grande and numerous walls in Compound A (figs. 13, 14) which had been previously covered with a soil-cement coating, and heavy soil-cement sections in Compound B. At Tumacacori the entire granary was treated and an application was made on the patched upper surface of the barrel-vault roof on the sacristy.
The applications on the above structures were made because of the extreme porosity of their surfaces and their susceptibility to weathering. Silicone water repellents should greatly extend the life of cement mortar capping and similar repairs. However, on typical porous Southwestern structures it is recommended that no application be made at a rate greater than 1 gallon to 50 square feet. Application is best made by a low pressure spray adjusted so that a very coarse stream is produced. Do not use a fine spray. Application should be heavy enough so that there is a rundown of from 6 to 8 inches below the strip being sprayed. The hydrocarbon vehicle is toxic if breathed in a confined space, and it is important that workmen wear an approved type painter's mask or respirator.
The use of plastics in ruins stabilization is in the experimental stage. They appear to have some uses in the special applications noted below.
There is an almost unlimited number of plastic compounds of the polyethylene and polystyrene types. They are specifically compounded for a variety of uses. Some, which are produced in pellet or granular form, can be dissolved in various solvents, the type depending upon the characteristics of the plastics. Some plastic coatings can be dissolved in water. The thought behind recent experiments is that plastics may find a use in hardening floors of such exhibits as pit houses, floors with special features, and possibly the soil walls of pit houses and similar substructures.
One type of polystyrene in pellet form will dissolve in benzene. Penetration of this solution into soil is excellent. The result is an impregnated soil which is waterproof and resistant to most acids or other substances which would be encountered on a substructure. However, it is not durable enough to be walked upon. This may be the result of improper technique in application, since polystyrene is a thermosetting plastic and may require additional heat in order to reach its maximum strength. On the other hand, compounds of greater impact strength, now being tested, may provide sufficiently durable without the addition of heat. Some emulsified types of polyethylene in liquid form, used for strengthening a variety of products from asbestos shingles to paperboards, also may be suitable for these applications.
Epoxy resins such as Bakelite's resin No. E R L 2795 are receiving increasing commercial use for applications ranging from cementing precast curbs and gutters on concrete roadways to the cementing of loose section of coal mine roofs (The Coal Age, vol. 63, Jan. 1957).
The most promising new product on the market is a concrescent epoxy adhesive of the Adhesive Engineering Company, San Carlos, Calif., a firm which has perfected a structural concrete bonding process. Introduced within the past 10 years, this process may have application in structural repair of ruins, particularly those with cracked masonry walls.
By this method, all crack surfaces are temporarily sealed except for occasional open ports. A gun is placed against one of the ports and an extremely strong, fast-setting epoxy injected into the crack at pressures of up to 300 p.s.i. When all ports have been filled, the temporary surface seal is removed. The epoxy is forced under pressure throughout the entire crack system, totally filling lateral and hairline cracks. Crack depth is no limitation.
The equipment used is compactly contained on a small, easily maneuvered two-wheeled hand truck. It consists of two containers holding the components of the fast curing, two-part epoxy adhesive (the resin and the curing agents) used for injection, and power driven pumping equipment that provides pressure for the operation. Two flexible hoses extend from the truck to the hand-held injection gun. The two-part epoxy is fed through these and mixed in a special in-line mixing chamber at the gun.
Several conditions must be met if this technique is to be successful in ruins stabilization: 1) as is the case for whatever method is used to repair cracks in walls, the cause of the cracks must be eliminated; 2) optimum pressure for injection of the epoxy is to be determined, preventing any possible hydraulic pressure which might damage the remainder of the wall; and 3) provisions must be made for injecting epoxy at depths beneath the face of the wall, leaving sufficient space to perform surface repairs which will match surrounding masonry and thus obscure the epoxy-filled cracks.
While it is doubtful that the day is fast approaching when an entire site can be indefinitely preserved in either plastics or resins, some of the new materials now under test may provide solutions to specific preservation problems.
A preservative spray which has enjoyed considerable success in the Eastern United States, particularly on stone and brick masonry, is Hydrozo Clear Coating. It was first produced in the early 1900's by E. E. Blackman, State Archeologist for the Nebraska State Historical Society in Lincoln, Nebr. According to the manufacturer's literature, Blackman "discovered and unearthed ancient pottery shards in a very wet soil that normally would have caused their disintegration, and further study at the Museum showed a resinous substance was responsible for their unique preservation. Several years of research and experimentation were necessary before this resin was reproduced by Mr. Blackman as Hydozite; and he then perfected a liquid formula, Hydrozo Clear Coating, for masonry and other porous surfaces." Since that time, the Hydrozo Products Company has produced other water repellent coatings for above and below grade stone and brick masonry, as well as for wood and concrete.
Only the Hydrozo Clear Coating will be considered here. The National Park Service has used this coating to preserve the restoration of some historical buildings, including Congress Hall in Philadelphia; the Mission of San Carlos, St. Augustine, Fla.; and the Old Meeting House in Alexandria, Va. Resembling paint oil or turpentine, Hydrozo Clear Coating consists of waterproof synthetic gum in formula with a volatile hydrocarbon thinner. It penetrates pores by capillary action, sealing the surface.
Specifications: Areas to be treated should be given two coats of Hydrozo. All surfaces must be dry and the temperature should be 55° F. or higher. If there is water or heavy dampness apparent on the surface, it will repel the coating. At temperatures lower than 55° F., Hydrozo may congeal on the surface. Surfaces must be clean, free from cracks, and all repointing which is needed should be done before application. If alkali is apparent, it should be removed with a brick cleaning compound or a solution of 15 percent muriatic acid. An acid, of course, should never be used on such substances as lime mortars, limestone, or marble. Hydrozo Clear should be applied with a wetting action and not by flooding. There should be very little if any run-down of materials as opposed to the application of silicones mentioned above. Brush, roller or airless spray may be used.
Hydrozo passes the ASTM submersion test with a repellency rating of 98 percent. The coating is resistant to acids, alkali, moisture, salt brine and sunlight. Mineral gum solids will penetrate 1/4th inch or more below surface. The coating has been tested by the National Bureau of Standards (Q.M. 095-S-Std. test report). It has a minimum vapor transmission rate of 5.7 grams per 100 square inches for a 24-hour period which assures breathing. While Hydrozo may be applied at temperatures lower than 55° F., it must be worked into the surface so that no material remains to cause clouding. If clouding does occur, it will disappear with warm weather or a light application of mineral spirits.
Hydrozo has limited use in the Southwest; it should have more use, particularly on surfaces of stone masonry, soil-cement veneer, and capping. Obviously, this gum and mineral based resin will not preserve friable adobe, but it will slow deterioration of adobe walls that are firm at the time of treatment.
The use of nonselective herbicides in keeping areas free of weeds, particularly room interiors, has had a brief history in the Southwest. First trials of a wettable powder, "Telvar-W," were made at Chaco Canyon in 1954. This small successful test was followed the next year by application at Aztec Ruins, Tuzigoot, and the Tumacacori National Monuments, and later at Grain Quivira. We are particularly indebted to then Superintendent John Stratton and Archeologist Peck for the careful photographic and written records of results at Tuzigoot. After three years use, they report excellent results in keeping rooms and other areas clear of local weeds including Russian Thistle, Puncturevine, Arizona Poppy and Trailing Four O'clock. The only vegetation which persisted were small clumps of an unidentified bunch grass. Less complete records from other areas indicate similar success in eradicating local weeds.
The only herbicide which has been well tested by the Stabilization Unit is this wettable powder, "Telvar-W." Numerous other herbicides are on the market, many of which may prove equally effective. Some types containing large percentages of borax, and which require heavy concentrations, were not tested. It was feared that the borax might produce undesirable leaching in masonry walls. Beginning in 1968, the staff at Chaco Canyon turned to the soil sterilant weed killer termed NS-610, a product of the National Chemsearch Corporation. Fast acting, non-selective, and low in toxicity, this product effectively kills all grass and weeds for a full season or longer, and is used successfully in the large open sites at Chaco Canyon. One to two gallons of NS-610 diluted with 10 gallons of water will sterilize 1,000 square feet applied in a coarse, wet spray.
In general, nonselective herbicides act upon the root systems of plants and sterilize the soil for varying periods of time, depending upon the rate of application, porosity of the soil and amount of rainfall. Application after the growing season begins and plants have attained some size is not as effective as applications made at the start of the growing season when plants are in the seedling stage. In any case the manufacturer's directions should be followed for maximum effectiveness.
Herbicides, as described above, are authorized for use in keeping rooms and small areas within sites free of weeds. Their use is far more economical than hand cutting or hoeing. In larger areas where a large expanse of barren ground might be objectionable, consideration should be given to low types of ground cover or sod where such can be established. In any event, herbicides must be strictly controlled to prevent washing into areas of valuable plantings.
In some archeological locations extensive damage can be caused by rodents, particularly rats and ground squirrels burrowing through or under masonry or adobe walls. Rodents have been particularly destructive in the cliff dwellings at Tonto National Monument where the structures are located in a dry cave, and at the schoolhouse at Tumacacori where the adobe is enclosed within a modern protective structure. Rodents at the schoolhouse had found a protected runway between the original walls and those of the enclosing structure
In areas of moderate infestation and where the affected structure is somewhat isolated, as at the Tumacacori schoolhouse, poisons are effective. Needless to say, extreme care must be exercised in their use. It is recommended that only those poisons employing a warfarin base be permitted. (Warfarin based rodenticides are stocked by GSA.) The warfarin base has a cumulative effect and must be consumed repeatedly over a period of time to be fatal. Such poisons are not as dangerous as those containing arsenic, strychnine and similar quick-acting ingredients. In any event, the poison should be placed in an approved bait station with a locked cover.
In areas of heavy infestation and in areas where the population is extremely dense in the immediate vicinity, poisons have limited usefulness unless a determined effort is made to reduce the rodent population over a wide area. In such instances the use of poisons should be combined with repellents which act to repel rodents from selected areas. Various areas in the Southwest participated in a recent test of rodent repellents on a small scale; the limited tests indicate that such materials, either in liquid or pellet form, might prove successful. Further tests in confined archeological sites are being continued.
A general list and evaluation of reference material on repellents used to protect pine seedlings is available from the U.S. Department of Agriculture, Pacific Northwest Forest and Range Experiment Station, entitled Comparison of 2 Rodent Repellents in Broadcast Seeding Douglas-fir, May 1957.
A note of caution is in order. Pesticides have assumed both national and international significance. Their use and misuse have seriously affected the environment. Before toxic chemicals are employed in an eradication program, including those briefly listed here, the user should consult with State (Agricultural Extension Service) and Federal (Interior) authorities for up-to-date information regarding restrictions, prohibitions, and alternative materials or methods for conducting a pest control program.
Last Updated: 16-Apr-2007