SOW_Attachment_2_-_Terracotta_Investigation.pdf

MARCH 20, 2016

FINAL REPORT

FINDLEY TERRACOTTA STUDY

Task Order No. GSP0515SC7031

Bailey Edward Project No. 1105-14-08

PAUL FINDLEY FEDERAL BUILDING, SPRINGFIELD, ILLINOIS

philipjhermanek Text Box Attachment #2: The terracotta Study documents shall be for reference only.

TABLE OF CONTENTS

Introduction ……………………………………………… 1

Executive Summary ……………………………………………… 2

Technical Report

Masonry Investigation ……………………………………………… 4

Existing Conditions ……………………………………………… 7

Terracotta Coating ……………………………………………… 8

Mortar ……………………………………………… 10

Conditions Causing the Deterioration ……………………………………………… 11

Recommendations to Remediate ……………………………………………… 12

Proposed Temporary Repairs ……………………………………………… 13

Appendices ………………………………………………

Appendix A: Thornton Tomasetti Report 7 Pages

Appendix B: Universal Construction Testing Report, Product Data for Plasthal, Synthalat

9 Pages

Appendix C: Drawings and Sketches 1 Page

Appendix D: Cost Estimate Extended Scope Reduced Scope

17 Pages 17 Pages

Appendix E: Minutes October 8, 2015 January 14, 2016 Teleconference Comments responses

1 Pages 3 Pages 2 Pages

Appendix F:

Contractor Notes on Terracotta Removals August 2014

4 Pages

Appendix G: Photos from Site October 8, 2015 And April 15, 2014

6 Pages philipjhermanek Typewritten Text Not included philipjhermanek Typewritten Text Not included

PAUL FINDLEY FEDERAL BUILDING,

SPRINGFIELD, ILLINOIS

Introduction

The Paul Findley Federal Building and U.S. Courthouse at 600 East Monroe Street in Springfield (IL0173ZZ) was constructed in 1930. The classically designed federal building consists of three stories and a basement, and is surfaced with limestone veneer with punched windows and a decorative terra cotta cresting at the cornice of the building. The building houses District Court functions and other federal government functions. The building is located downtown in the central business district, surrounded by sidewalk on the north and west elevations, parking on the east elevation and parking/loading dock/public lane on the south elevation. The Findley building has been listed on the National Register of Historic Places since 1978.

Bailey Edward (BE) was hired to prepare a study on the terracotta portion of the building, providing recommendations for the repair of the exterior terracotta along the parapet, identifying the conditions that caused the deterioration, providing recommendations to remediate the conditions that lead to the deterioration, as well as supplying suggestions for appropriate temporary interventions to safeguard the building. The study was informed by executing selective demolition of investigative openings. The final conceptual design includes plan and elevations and detail documents, photo documentation of the investigative openings, and cost estimates for proposed solutions. All solutions meet the Secretary of the Interiors Standards for Preservation.

The terracotta in question is part of a cornice band continuous around the top of the main facades of the building. As part of another project to review the masonry on the south facade, Bailey Edward observed terracotta spalling during a site visit on April 14, 2014. Further to BE’s report of this dangerous situation, the GSA hired a contractor to sound the terracotta, and anything that was found to be loose was removed at that time. Small pieces, too small for reuse, were removed and disposed. The contractor documented their observations on photos which were submitted to the GSA - see appendix F. During the October 8, 2015 start-up meeting for this project, there was little if any additional damage observed by Bailey Edward or Thornton Tomasetti on a brief walk-through, but the consultant team did not sound any terracotta at that time.

At the request of the GSA, a peer review engineer from Thornton Tomasetti (TT) was retained to provide an additional point of view. The letter report of TT is appended to this report, and the points raised in their report were discussed with TT and have been incorporated into the body of this report. BE also retained a masonry contractor, J. J. Braker, to perform the initial removals of selected terracotta units in two locations. Once the units were removed and the conditions assessed and photographed, the units were replaced and the openings were repaired, returning the area to its previous condition.

Executive Summary

On October 7, 2015, Bailey Edward lead an on-site investigation of the terracotta cornice of the Paul Findley Federal Building in Springfield Illinois to study potential causes of the observed terracotta spalling. The plan was to walk the parapet, observe the general conditions, select a number of masonry units to carefully remove and reinstall, photograph observations and conditions, and write up the observations, analysis and proposed solutions. In addition to Susan Turner of Bailey Edward, Bill Bast of Thornton Tomasetti reviewed the masonry openings performed by from J J Braker. Based on the locations of spalling, one masonry unit on the east cornice and one on the north were removed. The west cornice is relatively stable. (See Appendix C for roof plan and Appendix G for photographs.) Samples of coated terracotta were removed for testing (See Appendix B for test results). Based on the observations on site, it can be stated that:

1) The terracotta has been coated, and that coating improperly covers the masonry joints. The coating is trapping in moisture, which in turn freezes and thaws with the weather, and deteriorates the mortar. The moisture in the mortar is causing anchorage corrosion and section loss, and there is little, if any, structural utility left. (Fig 1B)

2) The coating is failing through cracking, thinning, and peeling. (see figures 9 and 10) Based on the testing, the coating appears to be an acrylic resin of low porosity. Its presence on the terracotta, the joints and the edges is deleterious to the drying function inherent in the masonry that it coats. (see appendix B)

3) There is selective spalling of the original glaze, sometimes down into the terracotta bisque substrate face, and there are pieces of detail spalling off the top of units. It is believed that this is due to differential thermal movement. (see figures 9 and 10)

4) The cornice pieces have been subjected to multiple campaigns of repointing as evidenced by the multiple mortar types observed in the investigative openings. Mortar joints are failing due to cracking, failure of the cementitious bond of the mortar, and saturation.

5) There was thick Portland mortar bedding on the masonry units observed in the masonry openings performed (see figure 1). This strong mortar constrains the masonry from moving with thermal expansion and contraction which can lead to fracturing of the unit, or displacement of a run of units. (see figure 8). The assembly is generally stable, but current patterns of deterioration will continue to occur exponentially over time.

The recommendations of the consultant team are as follows:

1) In the short term, retain a consultant to provide semi-annual inspection of the terracotta, to review and sound the terracotta, and report back on any deterioration which will occur between the time of writing this report and the time when the repairs will be undertaken. It would also be possible to laser scan the cornice, and use the scan as a base line to monitor incremental displacement of the cornice over time.

2) Carefully remove and reinstall the terracotta units to address the hard bedding mortar, the corroded anchors, and provide through-wall flashing. The investigative openings revealed corroded anchors, and without reconstruction of the top terracotta course, there is no mechanical anchorage inherent in the assembly. While this is an aggressive approach, there is no other way to address these problem areas.

3) Remove the coating, which traps water that appears to be entering the assembly through fine fissures.

4) Selectively repair the terracotta (spalls or missing details which have spalled).

5) After assessment of the appearance of the terracotta once the coating has been removed, either repair missing glaze where lost, or recoat all the terracotta with a breathable, very porous coating, but not over the mortar joint.

6) Repoint the terracotta.

7) Install sealant or lead cames in skyward facing joints.

The anticipated cost of the full recommended repairs is approximately $900,000 in 2016 dollars.

If the GSA decides to take a less aggressive approach, a lot of issues can be addressed, but neither the constraint caused by the bedding mortar, nor the deteriorated anchors, nor the through wall flashing could be addressed. The less aggressive, scaled down approach scope of work will include

1) In the short term, retain a consultant to provide semi-annual inspection of the terracotta, to review and sound the terracotta, and report back on any deterioration which will occur between the time of writing this report and the time when the repairs will be undertaken. It would also be possible to laser scan the cornice, and use the scan as a base line to monitor incremental displacement of the cornice over time.

2) Cut out all mortar joints as deeply as possible.

3) Remove the coating, which traps water that appears to be entering the assembly through fine fissures.

4) Selectively repair the terracotta (spalls or missing details which have spalled).

5) After assessment of the appearance of the terracotta once the coating has been removed, either repair missing glaze where lost, or recoat all the terracotta with a breathable, very porous coating, but not over the mortar joint.

6) Repoint the terracotta, in lifts suitable to the depth of joint cut.

7) Install sealant or lead cames in skyward facing joints.

The anticipated cost of the scaled down repairs is approximately $ 325,000 in 2016 dollars. These amounts are for construction only and not including design or contingencies. It is anticipated that the work would take two to three months over a summer.

Masonry Investigation

Bailey Edward (BE) visited the site in advance of the start-up meeting. Based on the deterioration patterns, two locations, one on the east parapet and one on the north parapet, were selected. There was little deterioration on the west parapet to warrant an investigative opening.

EAST MASONRY OPENING

The masons removed the indicated unit on the east with little effort. It had been fully mortar bedded, with shoved joints on the sides, and a backer rod and sealant skyward facing joint. There was some breakage of the interior flange of the unit in the process. Once the unit was removed, the opening revealed at least three mortars extant: white/grey Portland mortar, pinky-beige masonry mortar, and red Portland-based mortar. Close review of the opening revealed that there was no evidence of moisture in the cavity and the cavity was dry. There had been no rain in the previous 7 days. However, when the mortar was removed at the NE inside corner, an existing metal anchor was observed to be corroded. They have little if any structural utility left. (see Figure 1B)

The existing copper counter flashing present under the unit is believed to be original, and extends to the midpoint of the horizontal coping unit, with an upturned edge that was engaged in the mortar. The outer lip of the counter flashing overhangs a flashing the covers the roofing.

FIG. 1A East Masonry Opening FIG. 1B East Masonry Opening Inside Corner Farther along the east parapet to the north, a skyward facing sealant joint was observed to be badly failing. It was pulled out by the contractor to demonstrate the extent to which it had failed. At that location, about ten feet away from the previous opening, the pointing mortar was red Portland, with substrate bedding mortar of grey Portland which was saturated and had the consistency of sand.

Corroded anchorage with extreme section loss

FIG. 2 Joint where failed sealant removed

FIG. 3 Open joint

NORTH MASONRY OPENING

The north opening was more problematic to remove. The selected unit was considerably smaller, and previously cracked.

The joints were cut out on all sides, as was performed with the east unit. In the process of prying out the unit, the unit fractured along the crack line that was evident in the selection of the unit.

The opening revealed one mortar: a white/grey Portland cement mortar, in a very thick bed that engaged with the open bottom of the terracotta unit. The cavity was damp. There was evidence of a long term fracture at the face where it was discolored due to infiltrating dirt. There was moisture in the cavity, along with staining along the edge of the cavity. The moisture may have caused freeze/thaw movement which fractured the unit.

Red Portland pointing mortar Foam rod wet, undersized for joint

Too much depth of sealant

Removed brittle sealant

Edges of sealant failed

Coating over sealant failed

Opened joint shows evidence of cutting damage from previous efforts

Coating thinned, failed

Mortar below sealant and last pointing mortar deteriorated to sand, was damp

Coating failed

The existing copper counter flashing present under the unit is believed to be original. It was not fully exposed to the inner edge as it was buried in the mortar. The outer lip of the counter flashing overhangs a flashing that covers the roofing.

FIG. 4 North masonry opening FIG. 5 Interior of north opening, note dampness, water stains

FIG. 6 Repaired north masonry unit back in place (East masonry unit replaced but not photographed)

Existing Conditions

FIG. 7 Historic Drawings Detail FIG.8 Existing Conditions Sketch

The existing condition is slightly different from the depiction of the assembly in the original drawings. Due to thickness and size differences, some of the relationships among the pieces have changed from the drawing, as illustrated. The right hand coping piece has a drip edge which overhangs the flashing, and there is no counter flashing. The “floor” of the gutter is raised from the original drawing’s level. It is speculated that additional insulation was added in a previous roofing campaign. The EPDM roofing runs up and under the ribbed copper roofing panels about four feet up the slope.

Generally, the configuration of the various pieces permits the efficient shedding of rainwater off the terracotta and off the building.

Generally, the parapet is not exhibiting any extreme signs of distress that would indicate a requirement for any new control joints, although sections appear to be list towards the interior. Its general stability is likely due to the changes in direction in plan, occurring multiple times on each façade along its length. The parapet is low in height, resulting in a squat configuration. Its interlocked units of varying sizes and orientations, with some reinforcing present, leads the design team to the opinion that a seismic retrofit is not necessary.

COPPER FLASHING,

NO COUNTERFLASHING

TERRACOTTA

DRIP EDGE

POSSIBLE

INSULATION

Terracotta Coating Terracotta is generally a very permeable material, made from fired clay. The fired clay interior, called bisque, has the ability to absorb and evaporate moisture. A fired unit gets its water resistance suitable for a building’s wall system from the glaze that is applied during its manufacture. These glazed units are historically assembled in the wall lime-based mortar that was softer than the unit. The softness of the mortar permits the mortar some movement with the expansion and contraction caused by thermal movement. When the mortar is wetted by weather, or cracked due to larger thermal movement, any moisture entering the assembly was dried out through the porosity of the mortar inherent in historic mortars. This has been effectively prevented by the application of a coating over the joint.

Over time, Portland cement was introduced into the mortars, and they have become harder and more brittle. Thermal contraction of the terracotta led to cracking of the mortar. Once the crack permitted moisture to enter the mortar, it did not dry out since Portland mortars are not as permeable, and the retained water would freeze, expand, and cause cracking of the softer terracotta. (For the purposes of this work, it is assumed that the existing terracotta strength is at the mid-range of the typical terracotta strengths. In the interests of retaining as much historic material as possible, it is not recommended to sacrifice a terracotta unit to determine its empirical strength.)

In reviewing the history of the repairs on the building, it is believed that the terracotta was coated in 1996, the same year as the EPDM rubber roofing was installed. The coating was applied to both the terracotta unit and the mortar joint, which removed the unit’s ability to dry out between wettings. Further, the coating failed, in some places peeling off, permitting the ingress of moisture. The coating not only traps moisture, but prevents the mortar (the system’s drying component) from drying out.

This coating was sent for analysis to the laboratories of Universal Construction Testing, using terracotta samples removed from the building by Bailey Edward on their review in 2014. The full materials testing report can be found in Appendix B.

There appears to be two components to the coating, a polyester bonding agent, and an acrylic resin coating over it. The polyester bonding agent is believed to be Plasthall P-650. The technical data for this product is appended to the report in Appendix B. The acrylic resin coating is believed to be Synthalat-ACRYL 150. The technical data for this product is appended to the report in Appendix B.

An enquiry has been made to the GSA to locate the actual product used on the project. Manufacturers of each component have been contacted to determine what the porosity is for their product, and whether the porosity of their component would change if added to the other component. The responses to these enquiries have not yet been received at time of submittal.

Removal of a coating will require design phase testing, to ensure that what is proposed is suitable to the substrate, is effective, and does not negatively impact the substrate. It is not anticipated that the cleaning will cause any discoloration, although it is possible there was discoloration due to weathering that the coating was intended to cover in the first place.

There are a number of removal methods such as paint removers, a peel and strip, and other such methods which do not require abrasive methods. Once the testing is performed, recommendations will be made as to the proposed methodology, and whether or not to re-coat the terracotta within the project. Removal of the coating could be done in-situ, but may be easier to rinse and dry outside of the wall assembly.

Once the coating is removed, the terracotta glaze can be assessed. It is likely that there were some appearance issues with the historic finish which triggered the decision to coat it. If the finished surface that will be exposed after coating removal is deemed to be unsightly, it may be that the decision will be made to re-coat the terracotta, but this time using a modern breathable (porous) coating.

If it is deemed that the existing historic finish is presentable, but there are selected areas with missing glaze, the exposed bisque can be spot recoated. Where there is loss of the bisque, a terracotta patching material can build up the profile back to the original, and then spot recoated to match the adjacent finishes for a matching final appearance. If the design approach of a new overall coating is selected, then the patching work can occur without the application of spot recoating.

Figure 9 Typical deterioration patterns (north parapet shown)

Figure 10 Typical deterioration patterns – street face (north parapet shown)

Mortar

Terracotta detail spalled off, removed

Incipient crack in terracotta

Sealant joint failed

Coating thinned, peeled, failed

Sealant joint failed

Terracotta horizontal joint failure, removed

Coating thinned, peeled, failed

Sealant joints failed

The existing mortar was tested elsewhere on the building in a previous project, resulting in the recommendation to use a type N hydrated lime mortar. Mortar mock-ups need to be specified in the specifications of the future repair project, for the general contractor to establish a good match.

The existing terracotta has experienced multiple campaigns of repointing and resetting already. When the terracotta units were selectively removed, it was observed that there were multiple colors of hard Portland mortar in the bed joint (not just as repointing mortars). Due to its brittle nature, Portland mortars can crack and permit liquid moisture in, but its non-porous nature would prevent that moisture from dissipating. Resetting the units in less brittle, porous, hydrated lime mortar could be beneficial to the masonry’s long-term durability, but it would be at the risk of damage to the historic fabric substrate during its dismantling.

Rebuilding the top two courses of the parapet is an option that will permit the installation of a flashing to channel water away from the building, and to permit a good flashing detail with the roof membrane.

Additionally, some of the lengths of the runs between corners list towards the interior, which could be corrected in the resetting. There are rusted straps and pins that were encountered during the investigative work which can be replaced with stainless steel while the pieces are being reset.

The terracotta can be repaired in-situ, without removal, but that scope will not address the hard Portland bedding mortar, will not permit the installation of new stainless steel strap or pins, and will not permit the installation of through-wall flashing.

The decision to repair the terracotta in situ versus rebuilding the terracotta rests with the General Services Agency. Long lasting repairs can be affected with good detailing utilizing either approach.

Parapet listing towards interior

Conditions Causing the Deterioration

There are multiple causes of the current deterioration processes at work on the coping, which is causing the decorative elements to spall.

1) A coating has been applied to the terracotta and to the joints. Terracotta is generally a very permeable material, which gets its water resistance from the glaze that is applied in its manufacture. The coating was applied to both the terracotta unit and the mortar joint, which removed the unit’s ability to dry out between wetting. Further, the coating has fissured, and in some places crazed and fallen off, permitting the ingress of moisture. However, the coating over the mortar (the system’s drying component) prevents any moisture ingress from drying out.

2) Differential movement caused by thermal forces and day/night temperature fluctuations. Overnight, the temperature drops, and the entire terracotta unit has its temperature lowered consistent with the outdoor air temperature. At sunrise, the east façade receives direct sun, and when the temperature rises rapidly in the sun, the back side (which may also have frost on it) does not. The differential temperature causes internal stress in the unit. This is exacerbated by moisture trapped by the coating, and these water molecules freeze and expand, which has resulted in units’ spalling. Spalling occurs to the greatest extent on the east elevations, consistent with this observation but does occur on other facades as well, more related to the moisture.

3) Masonry joints between terracotta units are deteriorated, both through thermal movement causing cracking, or via weathering which has led to the absence of the mortar. The mortar between these units provides stability to the assembly, enabling the individual units to act as one. Missing or cracked mortar permits movement among the units, and also permits moisture to enter the assembly, which can lead to freeze/thaw deterioration.

4) Observed anchors were corroded, due to moisture cycling on ferrous metal. These anchors have corroded to the extent that they have lost section thickness. They have little if any structural utility left.

5) The mortar type varies dramatically among the units reviewed. The appearance ranges from a very reddish Portland-based mortar, to a grey Portland mortar, to more of an historic mortar in soft tan colors with a variegated aggregate. Given the distinctly differing appearance and assumed composition, that relative to strength and expansion/contraction, it is likely that the mortars are behaving differentially contributing to movement and cracking.

6) Copper flashings are present under the coping pieces. These extend only half way below the coping pieces.

They appear to be in good condition, but would be brittle if they were to be disturbed. They have been bent up and away for previous roofing campaigns and may not survive another (future) roofing replacement.

7) Oxidation of the terracotta anchors (cramps, fasteners) is apparent in the openings. This is caused by the wetting and drying cycling, and has resulted in the deterioration of the iron content to the extent that they have become brittle and will not provide their original strength.

8) While walking the roof to review the terracotta units, the overall condition of the roof was incidentally observed.

The copper roof and flashings are in good condition. The rubberized EPDM roof membrane was observed to have roof membrane protection pads that have become loose which are being blown off of their original position, sometimes landing over drains and blocking drainage. Further, the overall condition of the rubber roofing membrane is worn, exhibiting patches, telescoping of the substrate seams though it, discoloration due to exposure, weak field seams, minor tenting at the perimeter and some minor wrinkles and bulges. While it is serviceable in the short term, it is towards the end of its useful life. Additional foot traffic during repair of the terracotta will likely result in damage to the thin membrane.

Recommendations to Remediate

There are many different courses of action that could be undertaken, depending on budget, schedule, and desired life cycle costing. Following the Secretary of the Interior’s Standards for preservation, it is preferable to repair in lieu of replace. To that end, BE and TT recommends the following:

1) Disassemble and rebuild the parapet to remove the multiple setting mortar campaigns, replace failed corroded anchors, and install through-wall flashing.

a. Over the years, a number of repointing and resetting campaigns have occurred on the terracotta, as evidenced by the multiple mortars encountered in the investigative openings. These Portland mortars are brittle, and can crack with thermal movement. These cracks then permit the ingress of moisture.

Due to the impermeable nature of Portland cement, the water can’t evaporate out, which leads to freeze-thaw damage. Rebuilding of the parapet would give the opportunity to remove these hard Portland cement mortars, and consistently install a softer hydrated lime mortar to permit better movement and evaporation.

b. Rebuilding the parapet would permit the resetting of the parapet in-plane and true. Currently the parapet has areas which list slightly to the roof side. The listing is not serious, but is of concern.

c. Rebuilding the top terracotta courses will provide the ability to provide more perfect detailing for flashings and bedding mortar type.

d. The copper flashings could be replaced in kind, while the new roofing membrane could be run up and under it.

Repairs can be affected without total disassembly and rebuilding of the parapet, but any solution less than this will not address the terracotta displacement, the failed anchorage, the through-wall flashing.

2) As a minimum, repoint the entire parapet.

a. Given the existing condition of the parapet (as in, it is not showing signs of distress other than the localized spalling) it is not anticipated that aggressive control joints or seismic interventions are required.

b. Repointing mortar will be designed to be porous to assist in the drying out of any wind-driven rain which enters the assembly.

c. Mortar will be designed to be weaker than the low-end range of terracotta strength, so that in instances of movement, the mortar will act sacrificially.

d. Provide new sealant joints on all skyward facing joints.

3) Whether rebuilding or repointing, remove the existing coating from the individual units. The failing coating is permitting ingress of moisture, and also causing entrapment of moisture within the units, which causes deterioration when combined with freeze/thaw cycling.

a. There are multiple possible removal methods for the coating. Generally, it is believed that a commercial paint stripper or a peel and strip product will be effective to remove it. Design phase testing will need to be performed to determine the most effective method to be specified.

b. It is not anticipated that coating removal will cause any discoloration, but it is anticipated that the underlying terracotta was likely an inconsistent appearance (either staining, or crazing of the glaze) was the reason a coating was applied.

4) Once the coating removal is accomplished, review the condition of individual units of terracotta and the overall appearance, and identify required repairs of areas of spalled glaze or spalls in the terracotta substrate.

a. It is not recommended to replace the terracotta units, since they are in good condition. Based on the pattern of failure, it is not believed that there is an inherent issue with the original fabrication of the units.

b. Units would be only selectively replaced if they are badly damaged in the process of repointing/rebuilding.

c. Assess whether the appearance after coating removal is acceptable, or whether to reinstate coating. If it is decided to replace the coating once the terracotta repairs and the repointing are completed, reinstate the coating with a new, breathable coating, avoiding application over mortar joints. This will permit the drying process through the mortar joint, and prevent damage to the coating during future repointing campaigns.

5) Replace the EPDM roofing membrane. Masonry work on the parapet will damage the membrane, which is at the end of its useful life.

Proposed Temporary Repairs

To address any potential dynamic failure of the terracotta, it was requested that temporary repairs be placed, such as wire mesh, or similar temporary measures. The cost to install metal mesh would be high, once the material, labor and access costs are considered for the mesh, the anchorage, and the scaffolding / cranes. It would also place the building fabric at risk of additional damage to the terracotta, since the anchorage would need to be robust enough to secure the mesh under wind loading and would need to be placed in the existing narrow mortar joints. These larger anchors could damage the edges of the terracotta adjacent to the mortar joint into which the anchors would be fastened.

Instead of this temporary, unattractive and possibly damaging measure, it is recommended that an architect or engineer familiar with terracotta be hired to sound the masonry semi-annually. Based on the amount and configuration of the existing terracotta, this could be performed in one day at little expense, and with no risk to the building. Any material found to be loose or damaged could be removed at that time. The assistance of a mason could be used if larger pieces require removal, but that is not anticipated. BE believes this would be a more effective and economical solution for the

GSA.

To the extent of the observation performed prior to the meeting, the terracotta is apparently stable at this time, fourteen

(14) months after the contractor intervention in August of 2014, and no work has been completed since that time.

Appendix A: Thornton Thomasetti Report

Terra Cotta Parapet Condition Assessment

Paul Findley Federal Building, Springfield, IL January 22, 2016 | TT Project # C15188.00

1.0 INTRODUCTION

Bailey Edward engaged Thornton Tomasetti to perform a condition assessment of the deteriorated terra cotta parapet copings and decorative finials along portions of the roof of the Paul Findley Federal Building and U.S. Courthouse located at 600 East Monroe Street in Springfield, Illinois. According to the U.S. General Services Administration (GSA), the three-story building was constructed in 1930 and is clad in gray limestone with decorative terra-cotta crests, combining Federalist, Neoclassical, and Art Deco styles of architecture.

In general, the building is C-shaped with a pitched roof clad in standing seam copper panels.

The perimeter of the roof consists of a walkway serving as a type of gutter formed by a low (approximately 12-inch tall) exterior parapet. The standing-seam metal panels transition to a roof membrane, a few feet above the walkway, that wraps up the interior side of the parapet and under a copper flashing. The parapet is capped with limestone coping stones along the interior-facing, southern portions of the building and terra cotta copings with regularly spaced finial decorations along the exterior-facing, north, east and west, portions. All of the terra cotta has a beige-white coating that we understand was applied circa 1996 and is not original to the building.

William Bast, P.E., S.E. of Thornton Tomasetti visited the site on October 8, 2015. At that time, he met with Ms. Susan Turner of Bailey Edward and representatives from the building.

Thornton Tomasetti’s assessment was limited to visual observations of accessible areas and components. We documented various conditions throughout the roof using digital photographs taken from various vantage points throughout the exterior walkway. In addition, we observed the materials and conditions within two probe openings made in the roof parapet and copings.

2.0 OBSERVATIONS

2.1 GENERAL OBSERVATIONS AND SICUSSION

Based on our observations made during the site visit, we noted the following:

Cracked, spalled, and/or missing coating on the horizontal and vertical surfaces of the terra cotta copings and finial decorations. Portions of the terra cotta forming the finial decorations along the eastern parapet elevation have spalled away exposing their unprotected bisque. While the majority of the terra cotta deterioration appeared to be concentrated along the eastern portion of the building, we noted evidence of coating and sealant peeling, spalling and cracking throughout the entire parapet perimeter. In addition, we observed a finial decoration along the west parapet elevation exhibiting large cracks indicating an incipient spall. The building staff was unable to remove this area by hand.

Indications of improper mortar bed setting at the base of some of the finial decorations. The mortar bed does not reach the exterior of the decoration, creating a recess where water may collect.

De-bonded sealant in the joints connecting limestone copings along the west parapet elevation. Missing sealant in the ends of limestone copings where they meet the smokestack on the interior west elevation.

We did not observe expansion joints in the parapet at any point throughout the building’s perimeter. This is not unusual for buildings of this vintage.

The limestone copings and standing seam copper roof generally do not exhibit evidence of deterioration requiring remediation or repair. However, the roof membrane exhibits evidence it may be nearing the end of its service life. We did not conduct a detailed evaluation of the roofing materials as it was beyond the scope of this assignment.

Evidence of previous terra cotta repairs that do not exhibit deterioration.

2.2 PROBE OPENINGS

Bailey Edward identified two locations at which to create investigative openings, based on the observed deterioration patterns. One location was chosen on the east parapet elevation and one on the north.

Within the eastern elevation opening, we noted three distinct types of mortar delineated by their color; two Portland cement-type mortars and one masonry cement-type mortar. We did not observe visible moisture nor evidence of water infiltration throughout the materials within the opening interior. We noted two copper flashings that appear to be original to the building extending from below the coping and over the roof membrane that is lapped onto the parapet interior.

Within the northern elevation opening, we only noted a single type of Portland cement mortar. We observed damp materials and staining consistent with water infiltration. In addition, the exposed terra cotta unit exhibited a significant crack that may be caused by freeze-thaw cycles related to the apparent water infiltration.

3.0 MATERIALS TESTING

Thornton Tomasetti engaged Universal Construction Testing (UCT) to perform laboratory analysis on the terra cotta coating to determine its chemical composition and general resistance to moisture infiltration. Bailey Edward provided UCT with two samples (S-1 and S-2) of partially-coated, spalled terra cotta removed from the building.

UCT tested both samples for general moisture infiltration by applying water droplets to the surface of the coated, uncoated, and cracked surfaces of the terra cotta and noting whether the moisture was eventually absorbed or not. The results of these tests indicate the coating is generally impermeable to water. However, the water applied to the uncoated and cracked coating was readily and eventually absorbed, respectively. Therefore, an intact coating will prevent the transmission of water either into or out of the terra cotta, while a cracked coating or bare terra cotta provide a pathway for moisture infiltration.

UCT conducted infrared spectroscopy analysis on only one of the received samples (S-2) to determine the chemical composition of the coating. UCT’s laboratory results indicate the beige-colored surface coating consists of an approximately 0.08-0.10 mm thick, high-quality acrylic resin of polyethyl-acrylate polystyrene copolymer similar to products available under the trade name Synthalat-ACRYL 150. In addition, UCT identified an apparent bonding agent beneath the surface coating that consists of a 0.15 mm thick, white-colored, polycondensation of polyester adipate (carboxylate), possibly containing hexandiol, similar to products available under the tradename Plasthall P-650. Based on these chemical compositions removal may require the use of a commercial paint stripper or a product containing xylene and acetone.

4.0 CONCLUSIONS

The cracks and general deterioration of the coping system have allowed moisture and water vapor to penetrate the terra cotta substrate. The applied coating is not original to the building and the laboratory results indicate it is impervious to moisture transmission. In addition, the roofing membrane and flashing system may block alternative paths for moisture migration through the interior parapet face. Therefore, in the presence of uneven heating during sunrise/sunset and freeze thaw cycles this trapped moisture has caused the observed terra cotta spalling, particularly along the east elevation.

5.0 RECOMMENDATIONS

Based on our observations, materials testing results, conclusions and discussion with Bailey Edward, we recommend the following:

Remove the existing coating from all terra cotta materials, using products as noted in the UCT report above.

Repair deteriorated sealant and mortar joints.

While the parapet does not require repair or reconstruction, it may be beneficial to remove the miscellaneous and incompatible mortar types while facilitating the replacement of the roof membrane and original copper flashings. However, based on the configuration or the parapet, its condition, and mortar materials typically used during the original construction, it is not necessary to install expansion joints. In addition, we understand seismic retrofits to the parapet are not likely to be required.

Since the majority of the terra cotta units are in serviceable condition, we only recommend selective replacement of significantly damaged units; full scale replacement is not required.

6.0 LIMITATIONS AND SIGNATURES

Thornton Tomasetti’s professional services have been performed in accordance with the standards of skill and care generally exercised by other professional consultants acting under similar circumstances and conditions at the time the services were performed.

Thornton Tomasetti’s findings, conclusions and opinions are based on Thornton Tomasetti’s visual observations, professional experience, interviews with those knowledgeable with the conditions pertinent to the subject investigation, evaluation of documentation and sound investigation practices.

While Thornton Tomasetti’s findings are summarized as of the date of issuance, should new information or additional documentation become available, Thornton Tomasetti may amend or revise its opinions and recommendations accordingly.

This report shall not be construed to warrant or guarantee the building and/or any of its components under any circumstances. Thornton Tomasetti shall not be responsible for latent or hidden defects that may exist, nor shall it be inferred that all defects have been either observed or recorded.

We appreciate the opportunity to work with you on this project. Please do not hesitate to contact us at 312.596.2000 should you have any questions about our work and conclusions.

Sincerely, THORNTON TOMASETTI, INC.

Erik Wetzler, P.E., S.E. William D. Bast, P.E., S.E., SECB Associate Principal

APPENDIX A – REPRESENTATIVE PHOTOGRAPHS

T h o rn to n

T o m a se tt i

Spalled finial decorations along east elevation o rn to n

T o m a se tt

East investigation opening. View to the north ho rn to n T om as e tt

East investigation opening. View to northeast.

o rn to n

T o m a se tt

East investigation opening. View to south ho rn to n T om as e tt

Terra cotta “attic stock” o rn to n

T o m a se tt

Terra cotta “attic stock”

Appendix B: Universal Construction Testing Report

CHICAGO

61 Garlisch Dr.

Elk Grove Village, IL60007 P 847-459-9090 F 847-459-9015

DALLAS / FT WORTH

SAN ANTONIO / SO. TEXAS

AUSTIN / WACO

HOUSTON

MIAMI

972.432.6666 210.775.1637 512.551.0336 281.446.7363 954.676.4147

PROJECT NUMBER:

PROJECT NAME:

DATE:

15-213 600 E. Monroe Street – Infrared Spectroscopy & Porosity Testing 10.28.2015

PAGE | 1

Ms.Susan D. Turner, AIA, LEED AP, PMP Specialist Bailey Edward 35 E Wacker Dr., Suite 2800 Chicago, IL 60601

Re: Infrared Spectroscopy and porosity testing of a Surface Coating on a terracotta sample 600 E Monroe Street Springfield, Illinois

Dear Ms. Turner;

Universal Construction Testing (UCT) is pleased to present the results of the Infrared Spectroscopy and porosity testing of a surface coating on two (2) terracotta samples that were reportedly removed from the referenced location. The samples were retrieved by one of our staff members from your office on October 16, 2015.

The two samples (identified by us as Samples S-1 & S-2) appear to be fractured segments from a larger piece of terracotta. Both samples were tested for porosity, however, infrared spectroscopy was conducted only on Sample S-2. The base material for both samples is a rust colored terracotta and the exposed surfaces are coated with a beige-colored surface coating.

Photos of both samples are shown below in Figures 1 through 4.

Figure 1 – Back of Sample S-1 Figure 2 – Front of Sample S-1

Elk Grove Village, IL60007 P 847-459-9090 F 847-459-9015

DALLAS / FT WORTH

SAN ANTONIO / SO. TEXAS

AUSTIN / WACO

HOUSTON

MIAMI

972.432.6666 210.775.1637 512.551.0336

DATE:

15-213

PAGE | 2

Figure 3 – Front of Sample S-2 Figure 4 – Back of Sample S-2

Methods of Analysis

Porosity – Water droplets were applied to multiple surfaces of both samples. The water was visually observed to determine if it was being absorbed or did not absorb into the surface.

Droplets were applied to both coated and uncoated terracotta as well as cracked or chipped areas on the samples.

Extraction – The surface coating on Sample S-2 was chemically extracted with Xylene, Chloroform, Hexane and Tetrahydrofuran. The liquid extract was then analyzed by the following method:

Infrared Spectroscopy – The liquid extract was dried on a universal diamond ATR (Attenuated Total Reflectance Accessory) for Infrared Spectroscopy, then analyzed using a Perkin Elmer “Spectrum Model One” Fourier Transform Infrared (FTIR) Spectrometer, with a configuration of:

Resolution of 4 cm-1 and eight accumulations (number of scans) per sample.

Results

Water droplets applied to the surface of the coated areas on both samples beaded and did not absorb. Water droplets applied to the uncoated and chipped terracotta absorbed readily into the exposed terracotta surfaces of both samples. For the cracked areas, the water initially beaded, however it slowly absorbed into the sample and eventually completely absorbed into the samples.

The beige-colored surface coating on the terracotta is approximately 0.08 to 0.10 mm thick, and appears to be a high-quality Acrylic Resin, similar to a polyethyl-acrylate polystyrene copolymer.

Elk Grove Village, IL60007 P 847-459-9090 F 847-459-9015

DALLAS / FT WORTH

SAN ANTONIO / SO. TEXAS

AUSTIN / WACO

HOUSTON

MIAMI

972.432.6666 210.775.1637 512.551.0336

DATE:

15-213

PAGE | 3

Beneath the surface coating on the terracotta is a 0.15 mm thick layer of white-colored bonding agent which appears to be a polycondensation product of polyester adipate (carboxylate) with possibly some hexandiol (hexamethylene-glycol).

The surface coating on the terracotta is difficult to remove. To remove this coating from the in-place terracotta, either a commercial paint stripper, or a mixture of Xylene and Acetone could be used; these chemicals are typically available at hardware stores.

The Spectral analysis graphs are shown below in Analysis 1 and 2 and were derived by using Bio Rad “KnowitAll” software where known chemical spectral signatures are overalyed on top of the actual laboratory results.

Analysis 1

Overlay of Top Hit with IR Spectrum of Sample Coating, showing:

1) Test Sample: “S-2” (Black)

2) Acrylic Resin: Tradename “Synthalat” (Green)

3) Polyester Bond Agent: Tradename “Plasthall” (Red)

Elk Grove Village, IL60007 P 847-459-9090 F 847-459-9015

DALLAS / FT WORTH

SAN ANTONIO / SO. TEXAS

AUSTIN / WACO

HOUSTON

MIAMI

972.432.6666 210.775.1637 512.551.0336

DATE:

15-213

PAGE | 4

Analysis 2

Overlay of Top Hit with IR Spectrum of Sample “S-2”:

Same as Analysis 1 above, including a possible 3rd ingredient: “Hexamethylene.Glycol” from bond agent.

Mostly: Acrylic Resin, Poly(Ethyl-Acrylate-Styrene-copolymer)

Elk Grove Village, IL60007 P 847-459-9090 F 847-459-9015

DALLAS / FT WORTH

SAN ANTONIO / SO. TEXAS

AUSTIN / WACO

HOUSTON

MIAMI

972.432.6666 210.775.1637 512.551.0336

DATE:

15-213

PAGE | 5

We appreciated the opportunity to be of service to you on this project. Should you have any questions or require additional information, please feel free to contact us at your convenience.

Respectfully Submitted, Michael F. Pistilli, FACI Chemist, Petrographer

Mark E. Hughes, P.E.

General Manager

S Y N T H A L A T A 1 5 0

Issued: 12/2009-3 Page 1 / 2

Alter Postweg 35 D-21614 Buxtehude www.synthopol.com

Tel.: +49 (0) 41 61 / 70 71-0 Fax: +49 (0) 41 61 / 80 130 info@synthopol.com

Persönlich haftende Gesellschafterin Synthopol Chemie Dr. Koch Verwaltungs GmbH Registergericht Buxtehude B 2576

Geschäftsführer Dr. Henning Ziemer

All data is given to the best of our knowledge, for the purpose of information and advice. No claims of any kind - including in respect of patent rights - can be inferred herefrom.

Charakteristics: Polyisocyanate-curing acrylic resin

Supplied as: A = 60% in xylene/shellsol A/ butyl acetate (2:1:1) B = 60% in butyl acetat

Use: 2-pack coatings which are air- and oven-drying in combination with aliphatic polyisocyanates, for industrial coatings affording high mechanical strength, good adhesion and resistance to weathering.

Properties: A B hydroxyl value 130-150 130-150 (in-house method AV-F-H003) hydroxyl content ca. 4.5% ca. 4.5% (relative to nvc) flow time in sec.

(50% in Xylol) A:(in-house method AV-F-V003) 140-180 ----------- B:(in-house method AV-F-V009 ----------- 150-180

Gardner colour value A:(50% in Xylol) < 2 < 2 B:(as supplied) (in-house method AV-F-F007) non-volatile content 60 +/- 1% 60 +/- 1% (Lieferform) (in-house method AV-F-F003)

Flammpunkt in °C ca. 30 ca. 25 (as supplied) (in-house method AV-F-F006) density in g/ml 0.99 1.01 (as supplied)

(in-house method AV-F-D001)

Properties and fields of use: The main field of application for SYNTHALAT-A 150 is in coatings for metals and plastics which are air- and oven-drying in combination with aliphatic polyisocyanates. This type of 2-pack coating is preferred where the size or temperature sensitivity of objects to be coated stands in the way of achieving the required high-grade coating qualities. Examples are car refinishing, original finishing of buses and trams, plastic items, etc.

Film properties: In combination with aliphatic polyisocyanates such as "Desmodur N" they afford non-yellowing finishes with excellent weather resistance and gloss retention. Such films have good resistance to scratching and abrasion, combined with good solvent resistance. Their good resistance to water and aggressive chemicals should also be emphasised. SYNTHALAT-A 150 differs from A 045, A 055, A 065 and A 085 in its higher cross-linking density, which affords films having significantly greater abrasion and solvent resistance. These values are exceeded only by A 190 with its higher cross-linking density.

Pigmentation: SYNTHALAT-A 150 has high pigment absorption and very good pigment wetting properties.

Any neutral pigments or fillers are suitable for pigmentation. Basic pigments and pigments containing soluble metal compounds may have a catalytic effect on curing and shorten the pot life of coating batches once mixed.

S Y N T H A L A T A 1 5 0

Issued: 12/2009-3 Page 2 / 2

Alter Postweg 35 D-21614 Buxtehude www.synthopol.com

Tel.: +49 (0) 41 61 / 70 71-0 Fax: +49 (0) 41 61 / 80 130 info@synthopol.com

Persönlich haftende Gesellschafterin Synthopol Chemie Dr. Koch Verwaltungs GmbH Registergericht Buxtehude B 2576

Geschäftsführer Dr.