Masonry_Materials_Analsyis_Report.pdf

January 2020

=BUILD ING CONSERVAT ION ASSOC IATES INC .

Stafford Slave Settlement Cumberland Island National Seashore, Georgia

Masonry Materials Analysis and Testing

Stafford Slave Settlement Cumberland Island National Seashore, Georgia

Prepared For Joseph K. Oppermann – Architect, P.A.

Prepared By

Building Conservation Associates, Inc.

242 Cherry Street

Philadelphia, PA 19106

BCA Team Dorothy Krotzer

CONTENTS

1.0 INTRODUCTION

2.0 METHODOLOGY

3.0 SUMMARY OF FINDINGS

3.1 Tabby Brick

3.2 Red Clay Brick

3.3 Mortar

3.4 Visual Observations from Site Visit

4.0 CONCLUSIONS AND RECOMMENDATIONS

APPENDICES

Appendix A: “Report on Physical Properties of Tabby Brick” (Highbridge Materials Consulting, Inc., 11/18/19)

Appendix B: “Report on Tabby Brick Compositional Analysis” (Highbridge Materials Consulting, Inc., 12/23/19)

Appendix C: “Report on Mortar Compositional Analysis” (Highbridge Materials Consulting, Inc., 12/23/19)

Appendix D: “Report on Petrography of Red Clay Brick” (Highbridge Materials Consulting, Inc., 12/24/19)

Appendix E: “Report on Salt Analysis of Tabby Brick” (Highbridge Materials Consulting, Inc., 12/31/19)

Appendix F: “Report on Clay Analysis of Tabby Brick” (Highbridge Materials Consulting, Inc., 12/31/19)

Stafford Slave Settlement, Cumberland Island National Seashore Page 1

January 2020 BUILD ING CONSERVAT ION ASSOC IATES INC .

1.0 INTRODUCTION

At the request of Joseph K. Oppermann – Architect, P.A., Building Conservation Associates, Inc.

(BCA) performed testing and analysis of multiple historic masonry materials removed from the Stafford Slave Settlement on Cumberland Island National Seashore in Georgia. The materials tested were removed from the remaining chimneys that are present at the ruins of the Stafford Slave Settlement. The chimneys are constructed of four primary masonry materials: tabby brick, red clay brick, mortar, and stucco/parging. The testing and analysis performed by BCA was limited to the tabby brick, red clay brick and mortar.

The materials testing is being performed as part of a larger project involving the structural stabilization of select chimneys that remain at the site. The work performed by BCA and presented in this report involved both physical testing and evaluation in the laboratory, as well as in situ visual assessment. The goal of the testing and evaluation was to document the physical properties of the historic masonry to guide certain aspects of the chimney stabilization project.

The following report summarizes the findings of both the field observations and laboratory testing.

The report is organized by material (tabby brick, red clay brick and mortar), with sections on methodology as well as final conclusions and recommendations. All work required for the execution of this testing and evaluation was performed by Dorothy S. Krotzer, BCA Regional Director with assistance from Highbridge Materials Testing, Inc. (Highbridge). Highbridge submitted six separate reports as part of their laboratory analysis and testing, all of which are appended to this report. While this report summarizes Highbridge’s findings and puts them in the context of the site, review of their in-depth and comprehensive reports is strongly recommended for a full understanding of the Stafford Slave Settlement’s masonry materials.

Stafford Slave Settlement, Cumberland Island National Seashore Page 2

2.0 METHODOLOGY

On 19-20 August 2019, BCA performed a site visit to Cumberland Island to remove masonry materials from the Stafford Slave Settlement for laboratory analysis and testing. A total of 21 individual material samples were removed, including several full brick units. A list of the samples is included on the following page (Table 1) and representative photographs of sample locations are included in Images 1-16. While on site, in addition to removing samples, BCA also examined the extant chimneys for typical conditions and patterns of deterioration. Observations from the site visit are summarized in Section 3.1 of this report.

All of the material samples removed by BCA were shipped to Highbridge. Once received, BCA and Highbridge collaborated to decide the type and extent of testing that should be performed in order to provide the most useful information to the National Park Service (NPS) and the project team. The testing regimen developed by BCA and Highbridge for the tabby brick, red clay brick and mortar is summarized below and the results are summarized in Section 3.0 of this report. For a detailed description of the methodology used for each testing procedure, see Highbridge’s reports in Appendices A- F of this report.

It should be noted that the tabby brick does not fit any modern material category, so the test methods had to be adapted from several different standards.

Testing and Analysis of Tabby Brick

• Compositional analysis of tabby brick. This testing combined petrographic examination and chemical analysis to identify constituents, estimate proportions, and assess overall condition of the tabby brick. An acid digestion to extract an aggregate sample for description and gradation was also performed.

• Gradation analysis of tabby brick aggregate. Acid digestion was performed on three additional brick samples and the content and gradation of the acid-insoluble constituents compared.

The reason for this analysis was because it was anticipated that there may be some variation in the tabby brick mix proportions.

• Physical property testing of tabby brick. This testing included compressive strength, absorption, and saturation coefficient determination on a sample of five brick specimens.

• Qualitative identification of salts in tabby brick. Identification of water-soluble salts present within the brick through x-ray diffraction analysis. Three samples were chosen for the analysis, two identified by BCA as distressed and one identified by BCA as sound.

• Qualitative identification of clays in tabby brick. X-ray diffraction analysis was performed on clays extracted from one brick sample to determine whether swelling varieties are present.

This testing was performed in order to evaluate the possibility that the weathering of the tabby brick is the result of the swelling behavior of clay constituents during wetting/drying cycles.

• Gradation analysis of local silt/clay. Analysis of a sample of local sediments removed from the shore of the island was performed because it is suspected that this material may have been used in the tabby manufacture. The laboratory disaggregated this loosely consolidated

Stafford Slave Settlement, Cumberland Island National Seashore Page 3 material and graded it through standard sieves. The color and gradation of the sediment was compared to the acid-insoluble material recovered from the tabby brick to see if there was any similarity.

Analysis of Mortar

• Compositional analysis of masonry mortar. This testing combined petrographic examination and chemical analysis to identify constituents, estimate proportions, and assess overall condition of the tabby brick bedding mortar. An acid digestion to extract an aggregate sample for description and gradation was also included.

Analysis of Red Clay Brick

• Petrographic examination of red clay brick. One sample of red clay brick was examined petrographically. The purpose of this petrographic examination is to evaluate the constituents and microstructural features of the brick and to assess any potential causes for deterioration.

Stafford Slave Settlement, Cumberland Island National Seashore Page 4

Table 1. Masonry Samples Removed

Sample No. Chimney No.

Material Type

Description

Stafford.1 18 Tabby brick bedding mortar

Taken from interior side of chimney above lintel.

Stafford.2 11 Tabby brick bedding mortar

Taken from location of roof peak, area of mortar build-up, mortar continues into adjacent joint, (i.e. same material as mortar in joints).

Stafford.3 1 Tabby brick bedding mortar

Small sample from exterior mortar joint, west exterior face

Stafford.4 4 Parging Parging does not appear to be original (different color than mortar original, evidence of soiling on face of mortar under parging).

Stafford.5 N/A – Sample Discarded Stafford.6 4 Red brick Found on ground next to chimney, presumed to be from west surround of hearth.

Stafford.7 16 Tabby brick East side of hearth, “f” marked on face Stafford.8 1 Tabby brick West side of hearth, interior-facing side of chimney, removed from middle wythe Stafford.9 18 Tabby brick Interior-facing side of chimney (same location as

Stafford.1), original orientation of brick not clear.

Stafford.10 10 Tabby brick Found in rubble pile next to chimney Stafford.11 10 Tabby brick Found in rubble pile next to chimney Stafford.12 22 Tabby brick Found in rubble pile next to chimney Stafford.13 22 Tabby brick Found in rubble pile next to chimney Stafford.14 22 Tabby brick Found in rubble pile next to chimney Stafford.15 22 Tabby brick Found in rubble pile next to chimney Stafford.16 22 Tabby brick Found in rubble pile next to chimney Stafford.17 22 Red brick Removed from back wall of hearth Stafford.18 4 Tabby brick Removed from west side of chimney, significant erosion of brick face Stafford.19 3 Tabby brick Removed from back exterior face of chimney, significant erosion of brick face Stafford.20 16 Tabby brick Removed from northwest corner of chimney, face only, “B” marked on back/interior of brick) mild erosion of brick face

Stafford.21 N/A “clay rock” from shore

Stafford Slave Settlement, Cumberland Island National Seashore Page 5

Image 1. Typical intact chimney with replaced wood lintel. Note the use of red brick for the hearth and firebox and the use of tabby brick elsewhere.

Stafford Slave Settlement, Cumberland Island National Seashore Page 6

Image 2. General setting of the majority of the chimneys, located in a shaded tree grove in approximately the middle of the island. Some chimneys were previously stabilized in 2008.

Stafford Slave Settlement, Cumberland Island National Seashore Page 7

Image 3. Location of the sample Stafford.1.

Stafford Slave Settlement, Cumberland Island National Seashore Page 8

Images 4 and 5. Location of the sample Stafford.2.

Stafford Slave Settlement, Cumberland Island National Seashore Page 9

Image 6. Location of the sample Stafford.3.

Image 7. Location of the sample Stafford.4.

Stafford Slave Settlement, Cumberland Island National Seashore Page 10

Image 8. Location of the sample Stafford.6.

Image 9. Location of the sample Stafford.9.

Stafford Slave Settlement, Cumberland Island National Seashore Page 11

Image 10. Location of the sample Stafford.17.

Image 11. Location of the sample Stafford.19.

Stafford Slave Settlement, Cumberland Island National Seashore Page 12

Images 12 and 13. Location of the sample Stafford.18.

Stafford Slave Settlement, Cumberland Island National Seashore Page 13

Images 14 and 15. Location of the sample Stafford.20.

Stafford Slave Settlement, Cumberland Island National Seashore Page 14

Image 16. Location of the sample Stafford.21, the “clay rock” or deposited sediment found along the shore of the island.

Stafford Slave Settlement, Cumberland Island National Seashore Page 15

3.0 SUMMARY OF FINDINGS

The chimneys of the Stafford Slave Settlement are constructed of four primary masonry materials: tabby brick (a cast product); red clay brick (a fired product); mortar; and stucco/parging. The parging is only present on some chimneys and is only partial where it does exist. It is unclear whether the parging, which is a similar brown color as the tabby brick, is original or added. The parging was not examined in detail as part of this study.

The findings summarized below are primarily derived from the results of the extensive laboratory analysis and testing performed by Highbridge (see Appendices A-F), with Section 3.4 summarizing BCA’s observations from the site visit.

3.1 Tabby Brick

The pale brown-colored tabby bricks at the Stafford Slave Settlement are handmade, cast products composed solely of non-hydraulic lime and sand. Unlike the more well-known tabby concrete, it does not include broken shells as a coarse aggregate addition, although numerous shell fragments are present in the brick and are readily observed. (Images 7 and 11) However, these shell fragments are interpreted to have originated from the burning of oyster shell to produce the lime used to make the brick, and not as intentional aggregate additions. These unburned shell fragments and other lime conclusions are fairly plentiful, making up between 7- 10% of the brick sample examined. In general, the tabby brick is more like a coarse lime mortar than a true tabby.

The aggregate found in the tabby brick is a very fine and narrowly-graded natural quartz sand typical of Southern coastal lime mortars of the 19th century. There is a trace amount of silt, clay, and organic matter, which were probably part of the sand originally. It is this material, even in a small amount, that is responsible for the pale brown color of the brick. The sands for the four brick examined are all essentially the same. In addition, the color and gradation of the sand in the tabby brick matches that of the sands in both the sediment sample and the mortar sample analyzed as part of this study. This evidence indicates that the sands for both the tabby brick and mortar came from the same general geological deposit, most likely on Cumberland Island. (Chart

1) It is interesting to note that the original sediment material has a high percentage of dark-colored fines (approximately 12%), which must have been rinsed off prior to being used for the tabby brick and the mortar because the percentage of these fines in the brick and mortar are significantly lower.

The proportion of ingredients was also examined in four different tabby bricks and found to be incredibly consistent. The bricks are made of lime and sand in a proportion of 1 part lime: 1.5 parts sand, by volume, with the lime interpreted as a dry hydrate and not a putty. If interpreted as a lime putty, the proportions are 1 part lime : 1.8 parts sand. The consistency of proportions is unusual and striking. It is at a level usually found only in modern premixed mortars where batching is done in a controlled plant setting. This level of consistency could have been the result of the bricks being made from a single, large, well-mixed batch of lime and sand. Or, more likely, the brick manufacturing process involved careful measuring procedures. Either way, the proportion consistency indicates a fairly controlled and somewhat sophisticated means of manufacture for the tabby brick.

The petrographic examination of the tabby brick shows a material of good quality and no inherent material flaws. Its ingredients are well-mixed, adequately consolidated, and effectively

Stafford Slave Settlement, Cumberland Island National Seashore Page 16 cured. The resulting product is highly cohesive. The binder is fully carbonated, as expected for the material and age. Typical of lime mortars, the tabby brick is soft and highly permeable. No deleterious effects from being in service were identified petrographically. The clay portion of the brick was extracted and analyzed to determine if they were of the swelling variety, which could contribute to the long-term erosion of the tabby brick. The analysis determined that swelling clays are not present in any appreciable concentration in the brick.

The physical properties of the tabby brick were also evaluated using a combination of test techniques and standards. The compressive strength of the brick ranges from 400to 600 psi, which is in the range expected for a well-cured, non-hydraulic lime based mortar. For the compressive strength testing, the laboratory took cores in two directions from each of the five bricks to determine compressive strength. The reason for this was to determine if there was any significant strength anisotropy, since the tabby brick could potentially contain relatively flat shell fragments that would likely have been compacted in a direction perpendicular to the bed face during the casting process resulting in a plane of weakness. Anisotropy can impact how the bricks decay in service. The results of the testing indicate that there is no statistically significant difference between the two groups of cores tested (perpendicular and parallel), suggesting the brick ingredients were not strongly preferentially aligned during casting. Therefore, anisotropy is not a concern. One additional core sample was tested in each orientation under saturated conditions because it was also suspected that the tabby brick may be appreciably weaker when wet. The results indicate that the strengths while saturated are 60% to 70% that of the dry values.

Absorption values of the tabby brick are rather high, as expected for non-hydraulic lime-based mortars. Strength is more variable than absorption and there is no good correlation between the two. This suggests that the variations in strength are not directly controlled by porosity.

The tabby brick were also tested for the presence of soluble salts. While salts were observed on the brick in several locations during the site visit, the quantity and type of salts was unknown.

Knowing the type of salts present in masonry can help to identify the source(s), which can aid in eliminating the source when possible. Three tabby bricks were tested for salt content, two were in a deteriorated condition and one was in sound condition, provided as a control of sorts. The two bricks in deteriorated condition contain approximately 1% soluble salts by weight. The salts include halite (NaCl), niter (KNO3) and darapskite (Na3(SO4)(NO3)(H2O). These types of salt are likely from a combination of seawater (the chloride and sulfate salts) and the decay products of naturally-occurring matter in the soil (the nitrate salts). They are typical of those expected to be associated with low-lying marine environments. The brick in good condition was also tested and found to contain very little salt. Although the salts could have been introduced through the sands used to make the tabby brick, it is more likely that they are the result of rising damp, migrating upward through capillary action from the underlying soil. The very low concentration of salts in the sound brick sample indicates that the salts are not present in the same quantity for all bricks, which would be the case if the salts were in the original brick mix.

The variety of salts documented in the two damaged brick are certainly capable of causing progressive decay of the brick matrix through repeated wetting and drying cycles. The sulfates would tend to be the most destructive, but none of the salts are considered innocuous. This makes sense when one considers that some of the most pronounced damage observed on site was at the edges or sides of the chimney bases, where wetting and drying cycles would occur more frequently. Because there is no evidence for the presence of swelling clays in the brick matrix, nor are there other common environmental factors such as freeze-thaw cycling that

Stafford Slave Settlement, Cumberland Island National Seashore Page 17 would damage the tabby brick, it is possible that soluble salts are one of the more significant causes of erosion in the tabby brick.

Stafford Slave Settlement, Cumberland Island National Seashore Page 18

Chart 1. This chart compares the sand profiles in the tabby brick (Sample 07), the bedding mortar (Sample 01), and the local sediment sample (Sample 21). Note that the profiles are essentially identical.

Building Conservation Associates, Inc. Report #: SL1443-03 Stafford Slave Settlement, CINS, Chimney Stabilization Page 13 of 28

Chart 5.2: Aggregate Sieve Analysis This chart compares the sand profiles of the tabby brick (Sample 07) and the bedding mortar evaluated for Highbridge Report SL1443-02 (Sample 01). The local sediment in Sample 21 is also plotted here. One curve represents the complete material and the other curve represents the material with the fines passing a No. 325 sieve removed.

0.010.101.0010.00

Cu m ul at iv e

Pe rc en t P as si ng

Grain Size (mm)

Tabby - Sample 07

Mortar - Sample 01

Local Sediment Sample 21

Local sediment without fines Sample 21

Stafford Slave Settlement, Cumberland Island National Seashore Page 19

3.2 Red Clay Brick

The red clay bricks at Stafford Slave Settlement are hand-molded, fired bricks. The brick is uniform in texture with a purplish-red matrix and darker spots distributed evenly throughout the unit. The material is sandy-textured, highly permeable and well-fired. Microscopically, the brick is found to have approximately even volumes of sintered clay, fine temper, and micropores. The temper, which constitutes approximately 30% of the total brick volume, is a fine quartz sand that was a natural component of the brick clay. The ingredients are well-mixed and there is no streaking of the clay. The overall composition and microstructure of these red bricks are typical of many Southern coastal bricks of the nineteenth century.

There is only minor hairline cracking is evident in the brick and no significant secondary mineralizations. The red brick material is intact and sound.

Stafford Slave Settlement, Cumberland Island National Seashore Page 20

3.3 Mortar

The mortar analyzed for the current study was removed from the tabby brick masonry. More specifically, it was removed from the interior (protected) side of Chimney 18 above the lintel.

(Image 3) The mortar is off-white in color (Munsell code approximately 10YR 8/1) and contains visible lime particles and incompletely burned shell fragments. It is relatively uniform in appearance. Although it was not analyzed as part of this study, the parging on the interior of the chimneys appears to be the same white-colored mortar material.

Laboratory analysis determined that the mortar is a common lime mortar. The lime is a non-hydraulic fat (high calcium) lime prepared from oyster shell and apparently screened to remove coarser shell fragments. The aggregate is a very fine and narrowly graded natural quartz sand. It appears to be the same sand as documented in the tabby brick and the sediment deposit sample, but with the fines removed. The lack of these fines, which impart a brownish color to the sediment and the tabby brick, are what make the mortar white in color.

The proportions of ingredients in the mortar are estimated at 1 part lime : 1.4 parts sand by volume, with the lime calculated as a putty. If calculated as a dry hydrate, the proportions are 1 part lime : 1.1 parts sand, by volume. This is a fairly common proportioning for historical lime mortars, especially ones that contained narrowly graded sands such as the Stafford Slave Settlement tabby brick mortar. Such sands have high void contents and require a greater amount of binder to ensure thorough consolidation.

The materials were thoroughly mixed, adequately consolidated, and well-cured. There are no sand steaks or coarse binder inclusions. The resulting product is highly cohesive, though like the tabby brick, it maintains the softness and high permeability typical of lime mortars.

Discontinuous microscopic shrinkage cracks were observed, but this is typical of lime mortars (especially those with a relatively high sand content) and not considered a defect. The binder is fully carbonated as expected for the material and age. Aside from the normal shrinkage cracks, there is no physical distress of any kind identified petrographically.

Stafford Slave Settlement, Cumberland Island National Seashore Page 21

3.4 Visual Observations from Site Visit

The condition of the masonry materials varies widely from chimney to chimney, and also on each individual chimney. However, in general, the tabby brick and the mortar exhibit the worst conditions. The red brick is generally in moderate to good condition. Typical material conditions observed for the tabby brick include: erosion, often severe and concave; cracking, through both brick units and mortar joints; and loss of entire units. The mortar shows signs of cracking in most joints, as well as separation of the mortar from the brick units in many locations. The mortar is also severely eroded in places, typically in locations where the tabby brick are also eroded. Salts were also observed on all of the masonry materials.

Although the conditions tend to be localized and somewhat random, some patterns of deterioration were observed. The most severe damage seems to be concentrated at the outside corners of the chimney bases, at the backs of the hearths, and in any location where a dense Portland cement re-pointing mortar or patching has been installed. In light of the information provided in the materials testing reports, this pattern makes sense. If cyclical wetting and drying of soluble salts is considered one of the primary causes of the severe tabby brick deterioration (namely erosion), then the pattern of eroded bricks at the bottom of the chimneys seems logical. The bricks at the bases of the chimneys are closer to the ground and the soluble salts are entering the brick masonry from the salt-laden ground moisture through rising damp and then hydrating or crystalizing (both of which cause damage to masonry). Damage from salts associated with rising damp tends to manifest themselves several inches or feet from the ground, depending on the type of salt and other environmental conditions, because this is the location where the salts are drying. The corners of the chimneys, which tend to exhibit some of the most extreme erosion, would tend to go through more frequent wetting and drying cycles than the rest of the chimneys, since they are exposed to air on two sides. The corners may also be more subjected to rain run-off from the “shoulders” of the chimneys, although this was not directly observed during the site visit.

The presence of Portland cement patches and repointing, although not prevalent, is also causing damage to the tabby brick where it does exist. The dense, impervious Portland cement material redirects moisture through the adjacent highly porous tabby brick and historic mortar, exacerbating their deterioration. Portland cements can also contain salts such as sulfates, which could also be entering the adjacent tabby brick masonry and contributing to their deterioration.

The loss of chimney stacks, which is a prevalent condition, is seen as a structural condition and not a material condition. However, where chimney stacks remain, the tabby brick tend to be in relatively good condition. This observation is consistent with the theory that the worst damage is occurring closer to the ground due to proximity to salt-laden ground moisture.

Stafford Slave Settlement, Cumberland Island National Seashore Page 22

Image 17. The thin, brown-colored parging that is present on some of the chimneys (usually in partial form) is visible on this chimney.

Image 18. The exterior side of the hearth is typically in poor condition on most chimneys, as is visible in this photograph.

Stafford Slave Settlement, Cumberland Island National Seashore Page 23

Image 19. This chimney clearly exhibits the typical deterioration pattern of eroded corners of the chimney base.

Image 20. Portland cement repair materials have caused deterioration in adjacent tabby brick.

Stafford Slave Settlement, Cumberland Island National Seashore Page 24

4.0 CONCLUSIONS AND RECOMMENDATIONS

As a result of this study, the materials and properties of the primary masonry materials used to construct the chimneys of the Stafford Slave Settlement are much better understood. These materials include cast tabby bricks made of lime and sand, fired red clay bricks, and lime mortar.

As documented by the analysis, the majority of these materials (the lime and sand used to make the mortar and the tabby brick, as well as the red brick) are typical of Southern coastal building materials of the 19th century. In addition, the ingredients for these materials (with the exception of the red clay brick) were most likely obtained from the island and manufactured on it as well.

Of particular interest is the great consistency of ingredient proportions, or batching, exhibited by the tabby brick, which is unusual for a 19th century mortar material and more typical of contemporary industrial mixing. The proportion consistency indicates a fairly controlled and somewhat sophisticated means of manufacture for the tabby brick.

The results of the laboratory analysis, in tandem with the field assessment, indicate that all of the masonry materials used to construct the chimneys of the Stafford Slave Settlement are considered sound from a material perspective. When not exposed to external conditions such as salts, Portland cement repair materials or structural stresses, the tabby brick and mortar are in relatively good condition as individual building materials. Their composition and physical characteristics are not inherently flawed, and it is only when an external factor is introduced that they are damaged.

The fundamental good quality and sound composition of these materials affects the recommendations that are made for the repair and stabilization of the chimneys. Because they are not intrinsically flawed and do not contain any inherent deleterious ingredients or negative physical properties, replication of the materials can be considered.

The following recommendations for the masonry portions of the chimneys are made based on the findings of the field assessment and laboratory analysis:

1. Selective repointing, or as needed for structural stabilization and reconstruction.

Repointing of open or severely deteriorated mortar joints is recommended for the long-term preservation of the chimneys. Re-pointing and re-setting of tabby and red brick will also be required as part of any stabilization or reconstruction efforts associated with the current project. Because the mortar used for the chimneys is a fairly traditional high-calcium, non-hydraulic lime mortar that has performed adequately over its life, a replication mortar based on the original mortar is recommended with two adjustments.

First, the original sand used for the mortar is very fine and somewhat narrowly graded. It does not comply with modern masonry and gradation profiles as specified by ASTM C144. Therefore, the sand for the replication mortar should be fine-grained, similar in color, and graded to be similar to the original particle size distribution of the aggregate while meeting the requirements of ASTM C144.

Secondly, because a less fine sand and more well-graded sand will be used, a replication mix based on slightly more sand is recommended. While the original mortar ratio was 1 part lime putty to 1.4 parts sand, the recommended replication mortar mix is 1 part lime putty: 2 parts sand (by volume). This mix is also more in line with current industry standards.

Stafford Slave Settlement, Cumberland Island National Seashore Page 25

It should be noted that lime putty mortars require special precautions to ensure the proper installation and long-term durability. Because lime putty mortars take longer to cure and carbonate than lime-cement hybrid or hydraulic lime mortars (full carbonation of non-hydraulic lime can take years), it is important to protect the mortar during its initial cure. In order to ensure a proper cure, we recommend that the following precautions be taken when using lime putty mortars:

• Install at least 28 days prior to expected freezing temperatures to prevent the mortar from freezing. This should not be an issue on Cumberland Island, but is nonetheless important to note.

• Do not install during the hottest summer months when mortar may dry out prematurely.

• Do not install in areas of perpetual dampness.

• Protect from sun, wind and rain for at least 14 days.

• Moisten mortar and allow to dry in regular cycles during the initial cure to allow for carbonation.

• Best results will be obtained from a mason who is familiar with lime putty mortars, as their workability and working time differ from that of cement-lime hybrid mortars.

The lime putty manufacturer’s precautions and recommended installation procedures should be consulted before beginning work.

2. Selective replacement of significantly deteriorated bricks, or as needed for structural stabilization and reconstruction.

Replacement of significantly damaged or deteriorated bricks, both tabby and red clay, is a significant part of any repair or stabilization effort for the chimneys. Fortunately, the tabby brick is a material that can be replicated fairly easily by hand and does not need to be made by a manufacturer. The original tabby brick, which were cast in molds or possibly shaped by hand, were composed of only lime and sand. New tabby bricks can be made that essentially replicates this mix. However, as with the replication mortar discussed above, two modifications to the original mix are needed: the use of a sand that matches the original as closely as possible while still meeting ASTM C144, and the use of slightly more sand than the original tabby brick. The recommended mortar mix for the replication tabby brick is 1 part hydrated lime: 2 parts sand (by volume). It should be noted that a dry hydrate is recommended for the tabby brick as opposed to a lime putty, which is what was used originally. The reason for this is that it will be much easier to thoroughly mix the required large batches of mortar for the tabby brick production if the mortar is initially mixed dry. Shrinkage will also be less of an issue if the dry hydrate is used instead of a putty, although some shrinkage is to be expected when casting these bricks of mortar.

The same considerations for working with lime-based mortars that are outlined above are also true for these tabby brick.

As for the red clay brick, these cannot be made by hand as they are a fired product. Instead, salvaged or newly manufactured bricks that visually match the original should be sourced.

Fortunately, far fewer red clay replacement brick are needed than replacement tabby brick.

Stafford Slave Settlement, Cumberland Island National Seashore Page 26

3. Removal of Portland cement repair materials. Existing Portland cement repair materials (parging, pointing) should be carefully removed and replaced with the recommended lime mortar above.

The removal of the Portland cement materials will most likely damage the tabby and red clay brick masonry. However, the removal of this material is felt to be critical to the long-term preservation of the masonry and the long-term benefits of its removal are felt to outweigh any damage that might occur during the removal process.

4. Accept rising damp of salt-laden ground moisture. Although salts have been identified as one of the primary causes of the deterioration of the tabby brick and mortar, it is virtually impossible to eliminate this factor from affecting the masonry. All of the chimneys come into direct contact with the soil and elimination of the salts from the ground is not an option. While damp proof coursing to address the issue of rising damp exists (in both physical and chemical applications), it is not felt to be appropriate for this site. It is an invasive procedure with sometimes limited benefit. The success of damp-proof coursing, especially chemical, is not proven and the idea of introducing another material into these chimney assemblies does not seem prudent.

Appendix A

“Report on Physical Properties of Tabby Brick” (Highbridge Materials Consulting, Inc., 11/18/19)

404 Irvington Street, Pleasantville, NY 10570 | 914-502-0100 | www.highbridgematerials.com

Report on Physical Properties of Tabby Brick

Stafford Slave Settlement, Cumberland Island National Seashore, Chimney Stabilization Cumberland Island, Camden County, GA

Prepared for

Client ID

BUIL005

Report No.

SL1443-01

Report Date 11/18/19

Building Conservation Associates, Inc. Report #: SL1443-01 Stafford Slave Settlement, CINS, Chimney Stabilization Page 2 of 12

Confidentiality This report presents the results of laboratory testing requested by the client to satisfy specific project requirements. As such, the client has the right to use this report as necessary in any commercial matters related to the referenced project. Any reproduction of this report must be done in full. In offering a more thorough analysis, it may have been necessary for Highbridge to describe proprietary laboratory methods or present opinions, concepts, or original research that represent the intellectual property of Highbridge Materials Consulting and its successors. These intellectual property rights are not transferred in part or in full to any other party. Presentation of any or all of the data or interpretations for purposes other than those necessary to satisfy the goals of the investigation are not permitted without the express written consent of the author.

The findings may not be used for purposes outside those originally intended. Unauthorized uses include but are not limited to internet or electronic presentation for marketing purposes, presentation of findings at professional venues, or submission of scholarly articles.

Standard of Care Highbridge has performed its services in conformance with the care and skill ordinarily exercised by reputable members of the profession practicing under similar conditions at the same time. No other warranty of any kind, expressed or implied, in fact or by law, is made or intended. Interpretations and results are based strictly on samples provided and/or examined.

Cover Image Photograph of a chimney structure at the Stafford Slave Settlement on Cumberland Island, GA courtesy of Ms. Dorothy Krotzer of Building Conservation Associates, October 2, 2019.

Respectfully submitted, John J. Walsh President/ Senior Petrographer Highbridge Materials Consulting, Inc.

Stafford Slave Settlement, CINS, Chimney Stabilization Page 3 of 12

1. Executive Summary This report presents compressive strength, absorption, and saturation coefficient values for tabby brick sampled from the Stafford Slave Settlement on Cumberland Island, GA. Five individual units were sampled for this testing and strength was performed in two orientations relative to the assumed casting direction. Testing was also performed in wet condition for Sample 12. Table 1.1 presents the average test results.

Table 1.1: Summary of Findings

Property Average Standard deviation Dry compressive strength - perpendicular (psi) 480 110 Dry compressive strength - parallel (psi) 450 70 Wet compressive strength - perpendicular (psi) 310 n/a Wet compressive strength - parallel (psi) 400 n/a Cold water absorption (%) 18.0 1.1 Boiling water absorption (%) 26.3 0.8 Saturation coefficient 0.68 0.04

The following statements may be offered regarding the test data:

• Compressive strengths range from about 400 to 600 psi. This is in the range expected for a well-cured non-hydraulic lime-based mortar. It is worth noting that any spalling of the brick faces would occur in tension. A rule of thumb for brittle solids is that tensile strength should be about an order of magnitude less than compressive strength.

• A statistical analysis was performed on the strength testing performed perpendicular and parallel to the assumed casting direction (Appendix II). No statistically significant difference is identified between the two groups. This suggests that the constituents were not strongly preferentially aligned during casting.

• For Sample 12, companion cores were tested under saturated conditions. This results in strengths that are 60% to 76% of the dry values.

• Absorption values are rather high as expected for non-hydraulic lime-based mortars. The variance is small for both cold and boiling water absorption.

• A graphical plot of strength vs. absorption is presented in Section 6. Strength is more variable than absorption and there is no good correlation between the two. This suggests that the variations in strength are not directly controlled by porosity.

Stafford Slave Settlement, CINS, Chimney Stabilization Page 4 of 12

2. Introduction On August 27, 2019, Highbridge received a set of masonry samples from Ms. Dorothy Krotzer of Building Conservation Associates reported to have been sampled from chimney structures at the Stafford Slave Settlement on Cumberland Island, GA. A summary of the samples, their identifications, and locations is presented in Appendix III. It is understood from the client that the site represents the ruins of a 19th century slave housing settlement in which only the chimney structures survive. The chimneys are reported to have been constructed of tabby brick with red clay brick used for the hearth and firebox.

At Ms. Krotzer's request, laboratory testing is performed on the samples to provide information that may assist in an ongoing stabilization effort. The testing plan was developed in discussions between Ms. Krotzer and Mr. John Walsh of Highbridge.

The plan was finalized in a telephone call on September 20 and includes the following:

1. Qualitative identification of salts in tabby brick

Surface erosions on tabby brick may be the result of salt hydration or crystallization. The client has requested an identification of water-soluble salts present within the brick through x-ray diffraction analysis. Three samples are chosen for the analysis, two the client has identified as distressed and one the client has identified as sound.

2. Qualitative identification of clays in tabby brick Weathering of the tabby brick may also be the result of the swelling behavior of clay constituents during wetting/drying cycles. X-ray diffraction analysis will be performed on clays extracted from one brick sample to determine whether swelling varieties are present.

3. Compositional analysis of tabby brick This testing combines petrographic examination and chemical analysis to identify constituents, estimate proportions, and assess overall condition of the tabby brick. An acid digestion to extract an aggregate sample for description and gradation is also included.

One sample will be examined comprehensively.

4. Gradation analysis of tabby brick aggregate It is anticipated that there may be some variation in the tabby brick mix proportions. Rather than examine several samples comprehensively, a less costly acid digestion will be performed on an additional three brick samples. The content and gradation of the acid-insoluble constituents will be compared.

5. Gradation analysis of local silt/clay The client has provided a sample of local sediments from the shore of the island. It is suspected that this material may have been used in the tabby manufacture. The laboratory will disaggregate this loosely consolidated material and grade it through standard sieves.

The color and gradation of the sediment will be compared to the acid-insoluble material recovered from the tabby brick.

6. Physical property testing of tabby brick The testing includes compressive strength, absorption, and saturation coefficient determination on a sample of five brick specimens.

Since the tabby brick potentially contains relatively flat shell fragments and would likely have been compacted in a direction perpendicular to the bed face, it is suspected that there could be significant strength anisotropy. If so, this might have a bearing on the manner in which the brick decays in service. The laboratory will take cores in two directions from each of the five bricks to determine compressive strength. It is also suspected that the tabby brick may be appreciably weaker when wet. One additional core sample will be tested in each orientation under saturated conditions.

7. Compositional analysis of masonry mortar This testing combines petrographic examination and chemical analysis to identify constituents, estimate proportions, and assess overall condition of the tabby brick bedding mortar. An acid digestion to extract an aggregate sample for description and gradation is also included. One sample will be examined comprehensively.

8. Petrographic examination of red clay brick One sample of red clay brick will be examined petrographically. The purpose of a petrographic examination is to evaluate the constituents and microstructural features of the brick and to assess any potential causes for deterioration observed by the client on site.

This report presents results of the physical properties of the tabby brick masonry units. Results for the other tests will be presented under separate cover when complete.

Stafford Slave Settlement, CINS, Chimney Stabilization Page 5 of 12

3. Methods of Examination The tabby brick does not fit any modern standard material category. As such, test methods were adapted from several standards and optimized for the tabby.

Compressive strength testing was performed on five specimens chosen from rubble pile sample. Samples 10, 12, 13, 14, and 16 were selected for the testing. Since tabby brick is a cast material, there is the possibility of anisotropic mechanical properties due to the preferential alignment of elongate or plate-like particles during consolidation. In order to evaluate this, the laboratory chose to test compressive strength in two dimensions. A better statistical comparison could be made if the same population of five brick were used for comparison. Additionally, it was desired to test at least one specimen under saturated conditions in both orientations. This would allow an assessment of the worst possible conditions.

In order to achieve all of this, it was decided to take core samples from the bricks rather than cutting larger blocks. Cores were taken at 1.5" diameter on a drill press fitted with a diamond coring bit. For each brick, one core was taken perpendicular to the bed surface, and another parallel to the bed surface and perpendicular to the face. For Sample 12, an additional pair of cores were taken to test strength under saturated conditions. The perpendicular cores were labeled with an "E" suffix after the brick number. The parallel cores were labeled with an "A" suffix. A "W" was appended to the sample identifications for the cores intended to be tested under wet conditions.

The ends of each core were trimmed square to remove any irregularities. The cutting was done to keep the cores as close to a 2 : 1 length to diameter ratio without exceeding this ratio. Most of the preparation and conditioning from this point onward was done in partial accordance with ASTM C67, the standard for clay brick. All cores were dried to constant weight at 110°C for not less than 24 hours. A very thin application of shellac was applied to the cut surfaces. Just enough was applied to slow the absorption of mix water from the capping compound without achieving any penetration into the tabby. From here, the core ends were capped with a high-strength gypsum plaster and cured in a low temperature oven overnight. The cores for wet conditioning were further immersed in water for 48 hours prior to testing. All cores were tested to failure in a compression machine meeting the requirements of ASTM E4-16. The test results were rounded to the nearest 10 psi. No correction factors were applied to account for length to diameter ratios less than 1.75. This is not expected to result in any appreciable error.

The absorption and saturation coefficient testing was performed in general accordance with ASTM C67/C67M-19.

Rectangular prisms were sawn from the portions of the tabby brick remaining after cores were sampled for strength testing.

The faces of the prism were trimmed to remove any weathered or soiled material. The absorption method chosen includes the 24-hour cold water absorption followed by the 5-hour boiling water absorption. Calculation of the saturation coefficient is also included.

The following personnel contributed to the examination:

Technician: M. Pattie Supervisors: H. Hartshorn J. Walsh

Stafford Slave Settlement, CINS, Chimney Stabilization Page 6 of 12

4. Compressive Strength Testing

Table 4.1: Compressive Strength Results - Perpendicular Cores

Specimen ID 10E 12E 13E 14E 16E Capped length (in.) 2.55 2.45 2.17 2.31 2.10 Diameter (in.) 1.47 1.48 1.48 1.48 1.48 Length/diameter ratio 1.73 1.66 1.47 1.56 1.42 Area (sq. in.) 1.71 1.72 1.71 1.71 1.71 Load (lbf) 1033 914 912 632 635 Compressive strength (psi) 610 530 530 370 370 Average (psi) 480 Standard deviation (psi) 110

Table 4.2: Compressive Strength Results - Parallel Cores

Specimen ID 10A 12A 13A 14A 16A Capped length (in.) 2.76 2.94 2.95 2.94 2.96 Diameter (in.) 1.47 1.48 1.44 1.46 1.47 Length/diameter ratio 1.88 1.99 2.05 2.01 2.01 Area (sq. in.) 1.70 1.72 1.63 1.67 1.70 Load (lbf) 848 887 646 593 799 Compressive strength (psi) 500 520 400 350 470 Average (psi) 450 Standard deviation (psi) 70

Table 4.3: Compressive Strength Results - Wet-Conditioned Cores

Specimen ID 12EW 12AW Capped length (in.) 2.60 2.95 Diameter (in.) 1.47 1.48 Length/diameter ratio 1.77 2.00 Area (sq. in.) 1.70 1.72 Load (lbf) 533 696 Compressive strength (psi) 310 400

Stafford Slave Settlement, CINS, Chimney Stabilization Page 7 of 12

5. Absorption and Saturation Coefficient Determination

Table 5.1: Test Results

Specimen ID Cold water absorption Boiling water absorption Saturation coefficient

(%) (%) (no units) 10 18.0 25.0 0.72 12 19.6 27.2 0.72 13 16.6 26.7 0.62 14 17.6 26.6 0.66 16 18.0 26.1 0.69 Average 18.0 26.3 0.68 Standard deviation 1.1 0.8 0.04

Stafford Slave Settlement, CINS, Chimney Stabilization Page 8 of 12

6. Strength vs. Absorption Analysis

Chart 6.1: Relationship Between Absorption and Compressive Strength The following graph plots the cold and boiling water absorption values against the compressive strength for the same test specimens. Each specimen was loaded perpendicular and parallel to the assumed casting direction of the tabby. The plot suggests that variations in compressive strength values do not appear to be related to the absorption or porosity of the material.

0.0 5.0 10.0 15.0 20.0 25.0 30.0

Co m pr es si ve st re ng th

(p si

Absorption (%)

Cold absorption vs. perpendicular strength

Cold absorption vs. parallel strength

Boiling absorption vs. perpendicular strength

Boiling absorption vs.