B09_Att_25_Emergency_Dock_Inspection_Report_0002.pdf

34 William Way

Bellingham, MA 02019

(508) 966 9092 childseng.com

Emergency Dock Inspection at Fort Sumter

FOSU 211428

June 2022

Submitted to:

DHM Design

900 South Broadway

Denver, CO 80209

Submitted by:

DRAFT REPORT

Emergency Dock Inspection at Fort Sumter

Submitted by:

Childs Engineering Corporation

34 William Way

Bellingham, MA 02019

508-966-9092 www.childseng.com

Submitted to:

DHM Design

900 South Broadway

Denver, CO 80209

Childs Project: 2983-22.01

Inspection Dates: March 22 through March 25, 2022

TABLE OF CONTENTS

1.0 Introduction

1.1 Purpose and Scope

1.2 Method of Investigation

2.0 Description of Structure

3.0 Observations and Evaluations

3.1 Field Observations

3.2 Concrete Testing and Analysis

3.3 Structural Analysis

4.0 Recommendations

APPENDIX A – Photographs

APPENDIX B – Figures

APPENDIX C – Concrete Test Results

APPENDIX D – Structural Analysis

June 2022

1.0 Introduction

1.1 Purpose and Scope

The Fort Sumter Dock was recently rehabilitated due to impacts from ferry vessel docking. The initial engineering called for the replacement of numerous pre-cast concrete planks below the concrete topping slab on the dock damaged from repeated ferry impact upon existing wooden piles. Additionally, the repairs included adding new steel fender piles and adjacent dolphin piles independent of the existing dock to distribute loads and reduce further damage.

On December 16, 2021, a substantial completion inspection of the initial repair contract was conducted. During the previous inspection in October 2021, it was noted that newly developed cracks in the existing deck planks had formed. Upon further investigation during this inspection, at least ten of the existing twenty-three lower deck planks at the “T” section of the pier have cracks and spalling that has developed. In some cases, the individual plank area affected is nearly 50% of the plank surface area. This cracking is similar to what was observed during the original repair inspections of the now replaced deck planks. A previous analysis of the now replaced deck planks indicated a remaining service life of approximately 2 to 3 years.

Although the berthing loads have been reduced from the deck, damage may have already been done to these planks. Overloading and or reinforcing steel corrosion may have finally hit the point that the damage was able to manifest itself to the surface.

It is intended that the dock will be replaced in approximately 5 to 10 years under separate contract.

The tasks performed included: an underside investigation and collect necessary core samples at approximately three (3) locations on the underside of the dock within the location of the cracked underside planks, obtain information and lab testing from the core samples and other in the field investigation to determine recommended repairs to provide additional longevity until a replacement dock is constructed, and provide a short repair recommendation executive summary with anticipated repair costs.

1.2 Method of Investigation

On March 22 through March 25, 2022, three engineers from Childs Engineering

(Childs) conducted a topside inspection of the deck planks at the “T” portion of the Fort

Sumter ferry dock located at Fort Sumter in Charleston Harbor, Charleston, SC. The investigation included a Level I and Level II inspection of the accessible portions of the deck planks in accordance with ASCE Manual of Practice. During the inspection, the outside temperature was approximately 50 to 70 degrees Fahrenheit with weather ranging from mostly sunny to overcast with heavy rain. The inspection took place throughout the tide cycle with the underdeck portions investigated during low water with seas of 1 to 2 feet. The Level I inspection was performed to confirm as-built plans and detect any obvious major damage/deterioration due to over stress, impact damage, or severe corrosion. The Level II inspection is a complete detailed investigation of planks directed toward detecting and describing damaged or deteriorated areas and judging the structural integrity of the components. Photographs of typical conditions are located in

Appendix A. Field note drawings showing the deck construction and the location of defects are located in Appendix B. Additionally, full depth concrete cores were taken from three locations on the pier using an electric concrete core drill with stand and a 4-inch diameter coring bit. After the removal of cores, the holes were filled with a non-shrink grout and ground smooth with the deck surface. Also, twenty powdered samples of the concrete planks were taken from the underside of five planks at depths of 0 to 1 inch, 1 to 2 inches, 2 to 3 inches, and 3 to 4 inches utilizing an electric hammer drill with a carbide drill bit. The samples were visually analyzed and sent to a concrete testing laboratory for further analysis. The results of the testing are presented in Appendix C. Following review of the testing results and field information, a structural analysis of the elements was performed to assist in the overall evaluation of the planks. Calculations and results are presented in Appendix D.

2.0 Description of Structure

The Fort Sumter ferry dock is a concrete “T” shaped pier, constructed of prestressed concrete piles, poured in place concrete pile caps, prestressed deck planks, and a cast in place deck topping slab. In total, there are thirty planks arranged in six spans with five planks per span. The location of each plank is referenced to the bent and pile row that each plank spans. Bents are labeled at each pile cap, starting with bent A at the west and ending at bent K to the east. Pile rows are labeled from the outshore moving in, starting with pile row one near the docking face and ending with pile row three near the trestle or access portion of the pier. The planks spanning between bents A and B are at a raised portion of the pier and are approximately three feet higher in elevation than the remainder of the planks. The lower planks are located within the splash zone of the water column, with the lower portion of the planks submerged during storm surge events.

The planks are made of precast, prestressed concrete with three hollow voids near the third points of the plank length and quarter points of the plank width. The planks are fifteen inches in thickness and four feet wide by approximately twenty-one feet, three inches long. Overall, there are four upper prestressing stands located approximately four inches from the top of the plank and eight prestressing strands located approximately four inches from the bottom face of the planks.

3.0 Observations and Evaluations

3.1 Field Observations

The top decking slab showed no signs of distress. Sounding of the top surface indicated large areas of delamination between the deck topping and the planks over approximately 20 percent of the total surface area. Initial observations of the underside of the planks showed a range of deterioration from no visual deterioration to cracking of the concrete up to 1/16th of an inch in thickness, to delamination indicated by horizontal cracking and displacement of the concrete surface.

Cracking was found on the bottom surface of seventeen of the thirty planks. The cracks were typically hairline in width, but up to 1/16th inch. The crack lengths ranged from two feet to eleven feet and were often grouped on the planks, running in the longitudinal length of the plank. No cracks were noted on the planks spanning from bent

A to bent B.

Delamination of the concrete surface with visual indications were confirmed by sounding the concrete surface with a hammer. Based on the sounding of the concrete, the actual delaminated areas were found to be more numerous and much larger than the visual indications. In all, fifteen planks were found to have delaminations. The area of delamination ranged from approximately four-square feet to eighty square feet, the entire underside of the plank. No indications of delamination were found on the planks spanning from bent A to Bent B.

At bent B, the planks spanning from bent B to Bent C are exposed at their vertical face. Here, it was noted that the plank strands exhibited signs of corrosion with transverse cracking located typically at the low location of the strands.

Photos of typical conditions and deterioration are located in Appendix A. Drawings showing the location of deterioration are located in Appendix B.

Of the thirty planks that were investigated with this task order, twenty-five are located in an area where they have significant exposure to the water. The five planks located at the raised deck section are located high enough above the waterline, so that their exposure is less critical. This is observed in the visual and tactile deterioration that was lacking on these planks compared to the rest of the planks. Due to this, the focus of the concrete testing was determined to focus of the lower elevation planks where over 50 percent of the planks display deterioration over a substantial portion of their exposed surface.

3.2 Concrete Testing and Analysis

The concrete cores and powder samples were sent to a testing lab to evaluate the compressive strength of the concrete, perform a petrographic analysis, and to test for chloride intrusion.

The concrete was found to be hard with a compressive strength of 8,759 pounds per square inch, which is considered to be a good concrete strength.

The petrographic analysis found some anomalies in the concrete structure, but nothing that was alarming or indicated significant deterioration. Some of the findings included micro cracking, ettringite formation, and leaching of calcium hydroxide. None of these items appear to have detrimental effects on the performance of the planks.

The chloride testing showed exceedingly high chloride content in the samples with significant levels of chlorides well in excess of minimum levels necessary for the initiation of steel corrosion at the depth of the lower reinforcing steel. Fracture planes observed in the lower portion of all three of the cores are likely attributed to this corrosion of the embedded steel. Typically, steel corrosion is initiated when the chloride ion content of the concrete is approximately 0.03 percent based on the weight of the concrete. At the level of the lower reinforcing steel in our samples, the chloride ion content was found to be

0.050, 0.068, and 0.089 in the samples taken from planks with no observed deterioration, and 0.308, 0.366, and 0.395 in the samples taken from planks where deterioration was noted. This is more than ten times the level for corrosion initiation and more than three times the levels found in the previous investigation in 2017.

The full concrete laboratory analysis is located in Appendix C.

3.3 Structural Analysis

The structural analysis of the concrete was performed based on the reference information provided in as-built drawings from the original construction of the dock along with the results of the concrete strength testing. Based on the level of chloride ions within the concrete and the deterioration that is observed, it is known that the reinforcing steel has begun to corrode. The actual amount of steel strand cross sectional area loss due to corrosion is unknown but can be estimated based on historical data of similar structures and previous experience. When considering the original condition of the planks, it is estimated that the allowable uniform live load was approximately 290 pounds per square foot. When considering the damage and deterioration that has occurred due to the elevated levels of chloride ions, it can be assumed that the allowable uniform live load may require a reduction to as little as forty pounds per square foot.

Deterioration of the prestressing stands also results in a reduction of bond strength between the steel and the concrete, increased stresses in the anchorage zone, and reduced capacity of the steel strands. These factors reduce the elastic properties of the planks and may lead to a more brittle failure, which could occur with little to no warning.

The full structural analysis is located in Appendix D.

4.0 Recommendations

During the previous investigation of the pier in 2017, two deck planks on the “T” portion were found to have damage and deterioration. At that time, it was recommended that the planks be replaced to increase the overall service life expectancy of the pier to

5-10 years. This investigation occurred 5 years later and damage and deterioration is now found on approximately 70 percent of the deck planks located in the critical area. It is recommended that the remaining service life of the pier be considered as limited to 2 to

3 years based on the findings presented in this report. It is recommended that the deck planks of the pier be replaced within that time frame or loading restrictions be put in place.

A phased replacement plan may be considered, but there may be a need to restrict the use of portions of the pier prior to their replacement. It is highly suggested that a follow up investigation of the structure be performed with in 6 to 9 months of receipt of this report to determine the advancement of the deterioration and to reconsider the use and loading of the planks. Destructive analysis of a select plank will give great insight into the corrosion percentage of the steel, will allow for more accurate loading and service life analysis, and should be strongly considered for the next scheduled inspection.

Prestressed deck planks have few economic alternatives for repair once corrosion and spalling occur. While the piles and piles caps have a longer expected service life, the best option for the planks is replacement. It is estimated that the probable construction cost for labor and materials of the replacement of the deck planks of the “T” portion of the pier could be $2,750,000. This excludes engineering design costs, additional construction phase costs, and other costs associated with visitor access. Due to the cost and interruptions that occur during a rehabilitation of the deck structure, strong consideration should be given to the replacement of the pier.

Childs Engineering appreciates the opportunity to present our findings and recommendations from our recent investigation. If you have any questions or comments on this report, please do not hesitate to contact me.

Nicholas Sarata, P.E.

Vice President

310.776.1253 saratan@childseng.com

APPENDIX A

Photographs

Photo 1 – Typical under deck construction of pier.

Photo 2 – Deck plank in good visual condition.

Photo 3 – Deck plank with corrosion related cracking and sound concrete.

Photo 4 – Deck plank in visually good condition with areas of sounded voids.

Photo 5 – Deck planks with visual cracking and delaminations.

Photo 6 – Deck plank with larger delaminated area than indicated by visual cracks.

Photo 7 – Delaminated area of deck plank removed to sound concrete to a depth of 2.5 inches.

Photo 8 – Vertical face of deck planks at Bent B showing prestressing strand corrosion and delamination.

APPENDIX B

Figures

ELEVATED DECK SPAN

DECK PLANK REPLACED IN 2021

EMERGENCY DOCK INSPECTION AT FORT SUMTER

FIELD NOTES - MARCH 2022

PLAN OF "T" PORTION OF PIER

CAST IN PLACE DECK SPAN

NICHOLAS Sarata

EMERGENCY DOCK INSPECTION AT FORT SUMTER

FIELD NOTES - MARCH 2022

DECK PLANK SPAN BENT A TO B

VISUAL CRACK, LENGTH

DELAMINATION, SQUARE FOOTAGE

DECK PLANK SPAN BENT B TO C

VISUAL CRACK, LENGTH

DECK PLANK SPAN BENT C TO H

VISUAL CRACK, LENGTH

DELAMINATION, SQUARE FOOTAGE

DECK PLANK SPAN BENT H TO I

VISUAL CRACK, LENGTH

DELAMINATION, SQUARE FOOTAGE

DECK PLANK SPAN BENT I TO J

VISUAL CRACK, LENGTH

DECK PLANK SPAN BENT J TO K

VISUAL CRACK, LENGTH

EMERGENCY DOCK INSPECTION AT FORT SUMTER

SECTION OF "T" PORTION OF PIER

10" 20" 10"

4" 2" 8" 6" 8" 6" 8" 2" 4"

EMERGENCY DOCK INSPECTION AT FORT SUMTER

FIELD NOTES - MARCH 2022

PLANK DETAILS

21'-3"

APPENDIX C

Concrete Test Results

Concrete Research & Testing, LLC • 400 Frank Road, Columbus, Ohio 43207 • 614.443.0085

CRT Report No. P-1752

Petrographic Examination and Chloride Ion Testing of Specimens from The Ferry Terminal Pier at Fort

Sumter National Monument

May 17, 2022

Prepared for: Childs Engineering

Bellingham, Massachusetts

Childs Project No. 2983-22.01 www.ConcreteTesting.com

INTRODUCTION

Three concrete cores and twenty powdered concrete samples were received from Nicholas Sarata of

Childs Engineering Corporation on April 1, 2022. The samples were removed from the Ferry

Terminal Pier located at Fort Sumter in Charleston, South Carolina. The pier was constructed in 1991.

The pier deck is comprised of precast-prestressed hollow core planks with a welded-wire reinforced concrete topping. A diagram and site photos of the pier are presented in Figure 1 through Figure 3 of

Appendix A. The diagram and photographs were provided by Childs Engineering.

Currently, roughly 75 percent of the planks exhibit deterioration at the underside of the planks (see

Figure 3); the distress ranges from cracking to localized spalling. The current distress was observed in

December of 2021.

Concrete Research & Testing, LLC (CRT) was requested by Childs Engineering to perform a petrographic examination on the precast concrete plank portion of the cores to determine the cause of the cracking distress. It was also requested that chloride ion testing be performed on the powdered concrete samples and on samples removed from one of the cores.

SAMPLE DESCRIPTIONS

The received cores were labeled “Core #1,” “Core #2” and “Core #3.” Photographs of the cores in the as-received condition are presented in Figure 4 through Figure 6. Each of the cores is comprised of two distinct concretes: the precast/prestressed plank portion and a cast-in-place overlay portion. The cores have a diameter of 3.7 in. and were taken through the full depth of the overlay and substrate planks.

Core #1 has a length of 17.4 in. The overlay concrete comprises the upper 2.3 in. of the core and was received debonded from the substrate concrete. The bond failure occurred at the interface of the two materials and within weak material in the bottom of the overlay concrete. One end of the core represents the wearing surface of the overlay concrete. The wearing surface is worn, exposing aggregate particles across the entire surface. The other end of the core represents the bottom, formed surface of the precast plank. A fracture plane is present at a depth of 0.7 to 1.1 in. above the bottom surface of the core. The overlay portion of the core contains 4 mm diameter welded-wire steel reinforcement at a depth of 1.5 to 1.7 in. below the wearing surface. The substrate portion of the core contains three No. 3 rebars located at depths of 4.6, 5.7 and 14.0 in. below the core wearing surface.

Core #2 has a length of 18.0 in. The overlay concrete comprises the upper 2.7 in. of the core and is well bonded to the substrate concrete. One end of the core represents the wearing surface of the overlay concrete. The wearing surface is worn, exposing aggregate particles across the entire surface.

The other end of the core represents the bottom, formed surface of the precast plank. The core was received in two pieces with a horizontally-oriented fracture plane at a depth of 9.6 in. below the wearing surface of the core. The overlay portion of the core contains 4 mm diameter welded-wire steel reinforcement at depths of 1.6 and 1.8 in. below the wearing surface. The substrate portion of the core contains the imprint of a No. 3 rebar located at a depth of 6.2 in. below the core wearing surface.

Core #3 has a length of 18.2 in. The overlay concrete comprises the upper 3.1 in. of the core and is well bonded to the substrate concrete. One end of the core represents the wearing surface of the overlay concrete. The wearing surface is worn, exposing aggregate particles across the entire surface.

The other end of the core represents the bottom, formed surface of the precast plank. The core was received in three pieces with horizontally-oriented fracture planes at depths of 9.8 and 15.8 to 16.3 in.

below the wearing surface of the core. The overlay portion of the core contains 4 mm diameter welded-wire steel reinforcement at depths of 2.0 and 2.1 in. below the wearing surface.

The received powders were removed from five locations (four depths for each location). The powdered samples were labeled 1 A-D through 5 A-D.

EXAMINATION AND TEST METHODS

It was intended to use a portion of Core #1 for compressive strength testing, however, the full depth of the core was sectioned along with the other portions of the core for the petrographic work (see

Figure 7).

Preliminary stereomicroscopic examinations were performed on the cores. The cores were reassembled with epoxy, and epoxy was applied to the core wearing surfaces to protect and preserve features of the near-surface concrete during the specimen preparation. The cores were then saw-cut into thirds; saw-cuts were made parallel to the end surfaces. The upper, middle, and lower core portions were then saw-cut perpendicular to the end surfaces. Two parallel saw-cuts were made to obtain a 0.75-in.-thick section from the center of each portion. The depth of carbonation was measured on the concrete following the exposure of a freshly saw-cut surface to phenolphthalein solution.

The saw-cut sections were prepared for microscopic examination by lapping on a steel wheel with progressively-finer diamond and silicon carbide abrasives. The lapped sections were examined under a stereomicroscope having a magnification range of 3.5X to 220X. Visual and microscopic examinations were also performed on fresh fractured surfaces of the cores. The microscopic examinations were performed following the relevant guidelines outlined in ASTM C856, “Standard

Practice for Petrographic Examination of Hardened Concrete.”

The water-cement ratio (w/c) of the concrete was estimated based on observations and qualitative assessments of the cement paste, including hardness, textural features, color, abundance of unhydrated cement particles and rate of water absorption. The hardness is determined by probing the cement paste with a sharpened metal tool. The appearance of the probed areas is also used in the w/c assessment.

Acid-soluble chloride ion testing was performed on the powdered samples and on samples removed from one of the cores. The chloride ion testing was performed using Germann Instruments’ Rapid

Chloride Test.

TEST RESULTS

Chloride Ion Content – Chloride ion testing was performed to determine if the concrete contains elevated concentrations of chlorides. The results of the acid-soluble chloride testing of the powders are shown in the table below.

Powder

ID

Powder Depth

Chloride Ion

Content, %

1A 0 to 1 in. 0.162

1B 1 to 2 in. 0.155

1C 2 to 3 in. 0.137

1D 3 to 4 in. 0.089

2A 0 to 1 in. 0.587

2B 1 to 2 in. 0.516

2C 2 to 3 in. 0.454

2D 3 to 4 in. 0.308

3A 0 to 1 in. 0.562

3B 1 to 2 in. 0.454

3C 2 to 3 in. 0.399

3D 3 to 4 in. 0.366

4A 0 to 1 in. 0.226

4B 1 to 2 in. 0.147

4C 2 to 3 in. 0.104

4D 3 to 4 in. 0.068

5A 0 to 1 in. 0.430

5B 1 to 2 in. 0.347

5C 2 to 3 in. 0.347

5D 3 to 4 in. 0.395

Samples were removed from the bottom of Core #1 at depths of 1 in., 2 in., 3 in., and 4 in. from the bottom surface of the core. The chloride ion contents are shown below:

Core

ID

Sample Depth,

in. from bottom

Chloride Ion

Content, %

Core #1

1.0 0.147

2.0 0.092

3.0 0.077

4.0 0.050

For concretes with no intentionally added chlorides and which have not been subsequently exposed to chloride-bearing solutions, the baseline chloride ion content, as measured by the procedure used here, is typically less than 0.02 percent (based on dry concrete weight).

As can be seen from the results, the planks exhibit very high chloride ion contents. The minimum total chloride ion content of concrete that can lead to corrosion of embedded reinforcing steel is typically considered to be around 0.20% chloride based on weight of cement (ACI 222, “Protection of Metals in

Concrete Against Corrosion” §3.2, p. 222R-10). For a concrete having a cement content of 500 to

600 lb/yd3, this calculates to roughly 0.03% based on the weight of concrete.

EXAMINATION RESULTS

The examination was limited to the precast/prestressed planks. Photographs showing the lapped sections prepared from the cores are presented in Figure 7.

Aggregates – The coarse aggregate is a 0.75 in. maximum size crushed igneous rock consisting of granite. The granite is mottled in shades of gray, white, pink and salmon. The coarse aggregate particles are hard, dense, and exhibit very low water absorptions.

The fine aggregate is a natural sand is comprised primarily of quartz particles. The sand contains minor amounts of feldspar, granite, and magnetite particles and trace amounts of pyroxene/amphibole, garnet, coal, and mica particles.

The coarse and fine aggregates are judged to be good quality and are generally uniformly distributed throughout the examined concrete, although an area in the lower portion of Core #2 is deficient in coarse aggregate (see Figure 7).

Air Voids/Consolidation – The concrete represented by the cores is non-air-entrained.

Core #1 has an estimated total air void content of 1 to 2 percent. The concrete represented by the core is well consolidated with small amounts of entrapped air voids having typical sizes of 1 to 6 mm and a maximum size of 8 mm.

Core #2 has an estimated total air void content of 1 to 2 percent. The concrete represented by the core is well consolidated with small amounts of entrapped air voids having typical sizes of 1 to 7 mm and a maximum size of 11 mm.

Core #3 has an estimated total air void content of 2 to 3 percent. The concrete represented by the core exhibits fair consolidation with numerous entrapped air voids in the size range of 1 to 7 mm and occasional voids in the size range of 11 to 24 mm.

Cement Paste – The cement paste of the precast/prestressed components is comprised of well-hydrated portland cement. In all three cores, the cement paste is judged to be very good quality having estimated water-cement ratios (w/c) in the ranges of 0.40 to 0.45 in Core #1 and Core #2 and 0.42 to

0.47 in Core #3.

The cement paste at the top of the substrate concrete exhibits areas of elevated w/c. Small bleed voids are typically associated with these regions of high w/c cement paste. The elevated w/c cement paste is present across significant portions of the top 0.5 to 2.0 mm in Core #1 and 0.5 to 1.0 mm in Core #3.

Only minor areas in the top 0.5 to 1.0 mm of Core #2 exhibit the elevated w/c cement paste. This layer of weak cement paste can affect the bonding of the overlay concrete. Photomicrographs showing the top of the substrate concrete are provided in Figure 8 through Figure 10.

Reinforcing Steel – Core #1 contains three No. 3 rebars located at depths of 2.3, 3.4 and 11.7 in.

below the top of the substrate concrete; the bottom rebar has 3.2 in. of concrete cover. The upper two rebars do not exhibit corrosion. The lower rebar exhibits a small amount of corrosion, and two tight cracks extend up to 3 mm from the rebar. Photomicrographs of the lower rebar are shown in

Figure 11.

Core #2 contains the imprint of a No. 3 rebar located at a depth of 3.5 in. below the top of the substrate concrete. No corrosion product is present on the imprint surface.

Gaps Around Aggregate Particles – Core #2 exhibits gaps at the perimeter of aggregate particles which are filled with secondary deposits judged to be ettringite. The gaps have typical widths of

0.02 to 0.05 mm at the perimeter of coarse aggregate particles and 0.01 to 0.02 mm at the perimeter of fine aggregate particles. Photomicrographs of examples of the gaps at the perimeter of aggregate particles are presented in Figure 12 through Figure 14.

Fractures/Cracking Distress – Each of the cores exhibits complete fracture planes oriented roughly parallel to the end surfaces. The cores typically exhibit parallel cracking distress in the vicinity of the fracture planes. As the cores broke open at the fractures, the widths of the pre-existing cracks in situ are not known.

Core #1 contains a fracture located 0.7 to 1.1 in. above the bottom of the core (see Figure 4 and

Figure 7). The fracture typically extends around coarse aggregate particles. Parallel cracking is present between 2 and 10 mm above the fracture and 1 to 2 mm below the fracture; the cracks have widths of 0.05 to 0.08 mm. The core also exhibits tight cracks extending from rebar in the core. A crack extends 7 mm from the rebar located 3.5 in. below the top of the substrate concrete, and two cracks extend up to 3 mm from the lower rebar (see Figure 11).

Core #2 contains a fracture located at a depth of 7.0 to 7.3 in. below the top of the substrate (~8.4 in.

from the bottom of the core; see Figure 5 and Figure 7). The fracture extends around and through aggregate particles. Minor amounts of parallel cracking are present within about 1 mm of the fracture;

the cracks have a width of 0.02 mm. Significant horizontally-oriented cracking is present in the lower portion of the core, located about 3.6 to 4.6 in. above the bottom of the core. The cracks are discontinuous but are present across the full width of the core and have widths of 0.01 to 0.1 mm.

Small amounts of randomly-oriented microcracks are present throughout the core. The cracks have typical widths of 0.01 to 0.03 mm and are typically filled with ettringite; photos showing these cracks are shown in Figure 15 and Figure 16 (see also Figure 12 and Figure 13). A vertically-oriented crack is present at the bottom of the core; the crack extends to a depth of 13 mm and has a width of 0.1 mm

(see Figure 17).

Core #3 contains fractures located at depths of 6.7 in. below the top of the substrate and 12.7 to

13.2 in. below the top of the substrate (~1.9 to 2.4 in. from the bottom of the core; see Figure 6 and

Figure 7). The upper fracture typically extends around coarse aggregate particles but extends through a few coarse aggregate particles. Small amounts of cracking are present 0.5 to 2.0 mm above and below the upper fracture; the cracks have widths of 0.05 mm or less. The lower fracture typically extends around coarse aggregate particles but extends through a few coarse aggregate particles. A small amount of cracking is present 1 to 4 mm above the lower fracture, and a significant crack is present 0.1 to 0.9 in. below the fracture; the cracks have widths of ≤0.12 mm.

Secondary Deposits – White, crystalline secondary deposits judged to be ettringite are present in all three cores.

Core #1 contains ettringite in air voids through the full depth of the core. In the upper portion, air voids contain moderate amounts of ettringite, and small voids (≤0.2 mm) are often completely filled.

The amount of ettringite decreases with depth, and air voids in the lower portion contain only minor to small amounts of ettringite.

Core #2 contains abundant ettringite in air voids, in cracks and in gaps at the perimeter of aggregate particles (see Figure 12 through Figure 16). Voids up to 0.3 mm are often completely filled with ettringite.

Core #3 contains ettringite in air voids through the full depth of the core. In the upper portion, air voids contain small to moderate amounts of ettringite, and some smaller voids (≤0.2 mm) are completely filled. The amount of ettringite decreases with depth, and air voids in the lower portion contain only minor amounts of ettringite.

The bottom surface of each core is encrusted with a layer of white to tan secondary deposits (see

Figure 4 through Figure 6). The deposits have thicknesses of ≤0.2 mm in Core #1, 0.3 to 0.5 mm in

Core #2, and ≤0.2 mm mm in Core #3. Photomicrographs of the deposits are shown in Figure 17.

The deposits are judged to be calcite related to leaching of calcium hydroxide from the concrete. The calcium hydroxoide converts to calcium carbonate (calcite) upon interacting with carbon dioxide in the atmosphere. Calcite is also present within a crack at the bottom surface of Core #2 (see Figure 17).

Carbonation – The depth of carbonation was measured on the cores following the application of phenolphthalein solution.

In Core #1, the cement paste at the top of the substrate concrete is carbonated to depths of 1.0 to

1.5 mm, and the cement paste at the bottom of the core is not carbonated.

In Core #2, the cement paste at the top of the substrate concrete is carbonated to a depth 0.5 mm. The cement paste at the bottom of the core is typically not carbonated, although small areas are carbonated to a maximum depth of 0.2 mm.

In Core #3, the cement paste at the top of the substrate concrete is not carbonated. The cement paste at the bottom of the core is typically carbonated to depths of 0.5 mm or less, although small areas are carbonated to a maximum depth of 3.0 mm.

SUMMARY AND CONCLUSIONS

Three concrete cores from precast/prestressed hollow core planks the Ferry Terminal pier located at

Fort Sumter in Charleston, South Carolina, were tested and examined petrographically. Chloride content testing was performed on twenty concrete powders from the planks and on samples removed from the bottom of one of the cores. The testing and examination were prompted by the cracking and spalling distress exhibited by roughly 75 percent of the the planks.

General Quality of the Precast Concrete Planks

The fine and coarse aggregates are good quality and generally have a uniform distribution throughout the concrete. The cement paste is judged to be good quality having estimated water-cement ratios in the range of 0.40 to 0.47. The precast concrete is non-air-entrained with estimated total air void contents in the range of 1 to 3 percent.

Primary Cause of the Observed Distress

The cracking distress observed at the bottom of the planks appears to be related primarily to corrosion of the steel in the concrete. Provided site photographs show severe cracking and associated steel corrosion product (see Figure 3). Reportedly the lower level of prestressing strands is present at a depth of roughly 3 in. above the bottom surface of the planks. The chloride ion testing performed on powdered concrete samples show very high chloride ion contents in the lower portion of the planks and chloride contents well in excess for the initiation of steel corrosion at the depth of 3 to 4 in. above the bottom of the planks.

Each of the cores was received with horizontal fracture planes (see Figure 7). The upper fracture planes in Core #2 and Core #3 show very little associated cracking distress and these fractures are likely fresh fractures related to the coring.

The lower fracture planes and associated cracking distress in Core #1 and Core #3 and the cracking distress in the lower portion of Core #2 are likely related to the corrosion of the reinforcing steel in the lower portion of the planks, although none of the cracks contain steel corrosion products.

Core #1 was the only core that contained rebar in the lower portion of the core. The rebar is located

3.2 in. from the bottom of the plank. The steel exhibits only a small amount of corrosion and only very minor related cracking distress. Testing was performed to determine the level of chlorides within the lower portion of the core. The testing showed significantly lower chloride contents than most of the powdered concrete samples. However, the chloride contents at the depths of 3 in. and 4 in. are currently above the corrosion threshold and therefore continued corrosion of the steel is expected.

Delayed Ettringite Formation

The concrete represented by Core #2 exhibits features consistent with Delayed Ettringite Formation

(DEF). The primary feature is gaps around the aggregate particles that are completely filled with secondary ettringite. The ettringite filled microcracks throughout the concrete are also related to the delayed ettringite formation. The gaps around the aggregate particles are related to expansion of the cement paste caused by the DEF.

Ettringite is a normal component of concrete. When ettringite forms in the plastic state early in the hydration, it is termed primary ettringite. After hydration is complete, primary ettringite often dissolves in pore water and is recrystallized in air voids and cracks as secondary ettringite. These two forms of ettringite formation are innocuous in concrete.

Delayed ettringite formation can be a cause of deleterious cracking distress in concrete precast elements that are heat cured to temperatures in excess of about 70oC (~160oF). The elevated curing temperatures temporarily suppress the formation of ettringite. Over time, if the in-service concrete is exposed to moisture, delayed ettringite can form within the hardened cement paste, causing expansion of the paste and cracking distress. In delayed ettringite formation, an increase in the cement paste sulfate content is not expected since the ettringite is being produced from the original sulfates that were present within the cement.

It is unclear at this time if the DEF is causing any of the observed cracking distress at the exposed exterior surfaces of the planks. The plank represented by Core #2 should be examined in comparison with the planks represented by Core #1 and Core #3 to determine if any significant differences are observed. It is possible that other planks within the pier also exhibit DEF. Concrete affected by DEF is expected to exhibit a reduction of the compressive strength.

The expansion related to DEF is finite, with expansion ending once all of the ettringite is produced.

One laboratory study showed that the expansion of laboratory prepared mortar bars was complete in

1 to 2 years.1 The article goes on to state that expansion rate in field concrete would be expected to be much slower.

Trisha Piercey, Petrographer Nick Scaglione, President & Petrographer

Concrete Research & Testing, LLC Concrete Research & Testing, LLC

All concrete samples will be discarded 1 month after the report issue date unless otherwise requested.

References

1. Famy, C., Tayler, F.W. “Ettringite in Hydration of Portland Cement Concrete and its Occurrence in

Mature Concretes”, ACI Materials Journal, July-August 2020, pp. 350-355.

Page A1

Figure 1. Diagram provided by Childs Engineering showing typical cross section of the Ferry Terminal pier located at Fort Sumter in

Charleston, South Carolina.

Page A2

Figure 2. Site photographs provided by Childs Engineering showing the Ferry Terminal pier located at Fort Sumter in Charleston, South Carolina.

Page A3

Figure 3. Site photographs provided by Childs Engineering showing the end surface of the distressed, precast-prestressed hollow core planks. The lower photo is a close-up view of the red rectangle in the middle photo.

Page A4

Figure 4. As-received photographs of Core #1 which was removed from the Ferry Terminal pier located at Fort Sumter in South Carolina. The middle photos show the top and bottom surfaces of the core, and the lower photos show the fractured surfaces in the bottom of the core.

Page A5

Figure 5. As-received photographs of Core #2 which was removed from the Ferry Terminal pier surfaces of the core, and the lower photos show the fractured surfaces in the middle of the core.

Page A6

Figure 6. As-received photographs of Core #3 which was removed from the Ferry Terminal pier surfaces of the core, and the lower photos show the fractured surfaces in the middle and bottom of the core.

Page A7

Figure 7. Photographs of the lapped surfaces of the sections prepared from the cores. The fracture planes that were present in the received cores are indicated with arrows, and the red circles highlight the location of rebar in Core #1.

Core #3 Core #2 Core #1

Page A8

Figure 8. Photomicrographs of the top of the substrate concrete in Core #1; blue arrows mark the interface of the substrate and overlay concretes. The bond failure occurred at the interface of the materials or within weak material at the bottom of the overlay. The light-colored cement paste at the top of the substrate exhibits an elevated w/c relative to the underlying paste. The red arrow shows a bleed void associated with the elevated w/c paste.

Epoxy applied by CRT

Epoxy applied by CRT

Page A9

Figure 9. Photomicrographs of the top of the substrate concrete in Core #2; blue arrows mark the interface of the substrate and overlay concretes. Small amounts of the cement paste at the top of the substrate concrete is light-colored due to an elevated w/c relative to the underlying paste. The red arrow shows a bleed void associated with the elevated w/c paste.

Page A10

Figure 10. Photomicrographs of the top of the substrate concrete in Core #3; blue arrows mark the interface of the substrate and overlay concretes. The light-colored cement paste at the top of the substrate exhibits an elevated w/c relative to the underlying paste. Red arrows show bleed voids associated with the elevated w/c paste.

Page A11

Figure 11. Photomicrographs of the rebar in the bottom portion of Core #1. The lower photo is a higher magnification view of the boxed area in the upper photo. The red arrows highlight tight cracks extending from the rebar.

Page A12

Figure 12. Photomicrographs of ettringite present within gaps at the perimeter of coarse aggregate particles in Core #2. The lower photo is a higher magnification view of the boxed area in the upper photo. A tight, crack extends from the top of the aggregate (arrow).

Page A13

Figure 13. Photomicrographs of ettringite present within gaps at the perimeter of coarse aggregate particles in Core #2. The red arrows show tight, cracks filled with ettringite. The yellow arrow shows an ettringite filled air void.

Page A14

Figure 14. Photomicrographs of ettringite present within gaps at the perimeter of fine aggregate particles in Core #2.

Page A15

Figure 15. Photomicrographs of randomly-oriented cracks filled with ettringite in Core #2.

Page A16

Figure 16. Photomicrographs of randomly-oriented cracks filled with ettringite in Core #2. Ettringite can also be seen filling gaps around coarse aggregate particles.

Page A17

Figure 17. Photomicrographs of the secondary deposits present on the bottom surfaces of the cores in cross section. The deposits are judged to be calcite.

Core #1

Core #2

Core #3

Core 2

Core 1Core 3

Core 1 was taken from a plank with sound concrete with no visual deterioration.

Core 2 was taken from a plank with minor visual deterioration.

Core 3 was taken from a plank with moderate deterioration including cracking and delamination. The sample passed through a delaminated area.

Ryan Cavanaugh

Text Box

SAMPLE 2

SAMPLE 4

SAMPLE 3

SAMPLE 1

Sample 1 was taken from a plank with sound concrete with no visual deterioration.

Sample 2 was taken from a plank with moderate deterioration including cracking and delamination. The sample passed through a delaminated area.

Sample 3 was taken from a plank with moderate deterioration including cracking and delamination. The sample passed through a delaminated area.

Sample 4 was taken from a plank with sound concrete with no visual deterioration.

Sample 5 was taken from a plank with minor visual deterioration

SAMPLE 5

APPENDIX D

Structural Analysis

File No. 2983-22.01

Sheet 1 of 6

Client Date 26-May-22

Project Computed by LPL

Subject Checked by

PROBLEM STATEMENT & OBJECTIVE

EXECUTIVE SUMMARY

REFERENCES

1. ACI 222R - Protection of Metals in Concrete Against Corrosion

2. ACI 318 - Building code Requirements for Structural Concrete

3. PCI Design Handbook, 6th Edition

4. Fort Sumter (FOSU) Waterfront Dock Rehabilitation Supplementary Structural Assessment, March 2017, Childs Eng.

5. CRT Report No. P-1752, May 17, 2022

6. Dissertation: Residual flexural strength of corroded AASHTO Type II pretensioned concrete girder-deck system, May 2019, Bo Yan, Syracuse Univ.

7. Residual Flexural Capacity of Corroded Prestressed Reinforced Concrete Beams, Kioumarsi, 2021

8. Fundamental Study of Corrosion of Prestressing Strands in Concrete, Rumbayan, Jurnal Teknika, 2014

9. Predicting Concrete Deterioration due to Chloride Corrosion of Reinforcing Bars, Sasaki, 2004

10. Prestress and Bearing capacity loss of pre-tensioned concrete beams due to corrosion cracks, Dai, 2019

11. A global model for corrosion-induced cracking in prestressed concrete structures, Dai, 2015

12. Original Drawing Set for Fort Sumter Dock Replacement, 1990, ACOE

AVAILABLE INFORMATION

1. Per Ref. 12, all available plank geometry, and strand layout/data

2. Per Ref. 4 and 5, all history in regard to plank conditions, including core samples and crack locations and sizes

CONSIDERATIONS

Like mild reinforcing bars, the corrosion process for high strength prestressing strands is an electrochemical process.

Chloride ions penetrating the concrete cover dissipate the naturally occuring passivation layer surrounding the strand and in concert with exposure conditions, such as alternate drying and wetting, can rapidly initiate corrosion. It is well known that certain levels of chloride ions present in concrete surrounding the strand will initiate the corrosion process.

Ref. 8

Allowable live load vs. corrosion level graphs are presented to assist in evaluating the remaining service life of the existing precast prestressed deck plank system on the T section of the pier. Corrosion levels can be estimated based on observed surface cracks on the underside of the deck and at exposed plank ends, as well as cracking observed in the three concrete core samples.

The base line moment capacity of the deck planks is separated into two categories based on the condition of the bond between the concrete topping layer. Where evidence suggests that the bond has been compromised, a reduced plank moment capacity based on non-composite action is used to determine the base line moment capacity in lieu of the full composite section moment capacity. Fully composite planks have a moment capacity fMn = 216.3 ft-kips per plank (Ref. 4), while for non composite planks fMn = 155 ft-kips per plank. Core testing indicates that the concrete shear strength is not a governing factor in the overall strength of the planks.

CALCULATIONS

Fort Sumter

Fort Sumter Field Investigation Follow-up

Structural Evaluation of Existing Pre-stressed Deck Planks

Existing pre-stressed concrete deck planks exhibit various degrees of cracking distress at the bottom of the planks that appear to be related primarily to corrossion of the primary steel strands in the concrete. The objective of this calculation is to estimate the remaining allowable live load that the various planks can safely resist, based on concrete core samples and visual examination of the top, bottom, and in some cases exposed ends of the affected planks.

File No. 2983-22.01

Sheet 2 of 6

Client Date 26-May-22

Project Computed by LPL

Subject Checked by

From References 4 and 5, concrete sampling indicates chloride ion content far above these threshold levels and therefore verify that concrete corrosion of the strands has been ongoing for an extending period of time. Corrosion in strands is usually manifested in pitting on the exposed surface of the individual wires that make up the strand

(in this 6 outer strands wrapped around 1 inner strand).

Compared with non-prestressed steel, prestressed steel strands are more susceptible to corrosion for the following reasons:

1. Prestressed steel is continuously highly stressed due to the prestress (270 ksi compared to 60 ksi for rebar), and elevated stress in steel produces an increase the corrosion rate.

2. Prestress strands have a larger surface area as compared to plain bars, and include voids between the wires that comprise the strand which provide further room for corrosion to develop.

3. Prestressed strands twist, which results in all exterior surfaces of the strand to be on the side closest to the concrete cover, providing more opportunity relative to plain bars for surface pitting to occur. Instead of one single wire in the strand being on the side of the concrete cover, all the wires get a chance to be closest to the available corrosion producing elements.

Corrosion in prestressed strands affects performance in multiple ways:

1. The modulus of elasticity, the tensile strength and the failure strain will all decrease as the corrosion level increases.

2. As corrosion increases, deterioration of the bond strength between the strand and the surrounding concrete occurs, which leads to concrete crtacking, and hence accelerated corrosion. Bond deterioration and concrete cracking results in strand slip, and a loss of prestress.

3. In the strand anchorage zones (at the ends of the planks), corrosion will aggravate the already occurring high concrete anchorage hoop tension stresses (the hoyer effect) and cause large cracks to form at the ends of the planks.

It can't be emphasized enough that the loss of bond and hence prestress in the strands of the plank will potentially cause the ultimate strength failure mode to change from a desirable ductile type failure to a brittle non-ductile type failure mode.

Under these conditions the planks could possibly give no warning of imminent catastophic failure.

Historically prestressed concrete structures have been considered exceptionally resilient because the prestress keeps the concrete from cracking under service loads by maintaining a constant compression throughout the concrete section.

The prestress effectively keeps any occuring cracks tightly closed helping to prevent corrosion producing agents from getting to the prestressing strands. The downside to using prestressed sections has been found to be that if corrosion does manifest itself, the consequences can be more severe.

CALCULATIONS

CONCLUSION

Graphs have been generated to assist in estimating the available allowable live load capacity of existing prestressed concrete planks that are experiencing various levels of corrosion and cracking. The plank strength capacities are separated into two groups; one that considers composite action with the existing concrete topping, and the other that does not.