B08_Attachment_14_GERO_289951_Concrete_Exam_Report.pdf

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

Concrete Examination Report

Lincoln Memorial Bridge Vincennes, IN

Prepared for Jablonski Building Conservation, Inc.

Client ID

JABL001

Report No.

SL1813-01

Report Date 07/20/23

Jablonski Building Conservation, Inc. Report #: SL1813-01 Lincoln Memorial Bridge Page 2 of 29

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

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

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

Cover Image Photograph of the Lincoln Memorial Bridge ca. 1936. Image courtesy of Mr. Bob Gotthelf of DHM Design.

Respectfully submitted, John J. Walsh Heather Hartshorn President/ Senior Petrographer Chemist/ Staff Scientist Highbridge Materials Consulting, Inc. Highbridge Materials Consulting, Inc.

Lincoln Memorial Bridge Page 3 of 29

1. Executive Summary This report presents the results of petrographic examination, chloride analysis, and compressive strength testing on a suite of concrete cores recovered from the approach at the Lincoln Memorial Bridge in Vincennes, IN.

The petrography was performed on Cores C-1, C-2, and C-9. These three samples contain similar materials mixed at approximately the same proportions. The samples consist of conventional normal weight portland cement concrete with no supplementary cementitious materials and no air-entrainment. Water to cementitious materials ratios (w/cm) are estimated to have been moderately low resulting in a dense cementitious hydrate. The coarse aggregate is a natural gravel consisting of various sedimentary rock grains. Aggregate gradations are not consistent across the sample set.

The fine aggregate is a well-graded natural sand. No significant workmanship deficiencies are identified in the samples examined petrographically.

Water-soluble chloride contents were measured for Cores C-1, C-2, C-3, and C-9. Chloride values are low and uniform throughout each core indicating that the original materials were not contaminated and that no secondary ingress has occurred during the service life. Carbonation depths are also minimal. With the exception of significant cracking in Core C-9, there are no concrete material conditions observed that would leave any embedded reinforcement especially vulnerable to corrosion.

Core C-9 exhibits a moderately advanced alkali-silica reaction due to the presence of chert in the coarse aggregate.

This concrete may be considered to have exceeded its useful service life. Though the other two cores represent concrete that has remained sound, these contain the same aggregate and may be considered susceptible to the reaction. The high quality of the cement paste has likely allowed the concrete to resist moisture ingress. This may explain the lack of reaction in Cores C-1 and C-2.

Compressive strength testing was performed on Cores C-4, C-5, and C-8. The strengths vary widely from 4240 to 11440 psi.

Lincoln Memorial Bridge Page 4 of 29

2. Introduction On April 26, 2023, Highbridge received eight (8) concrete core samples from Cody Hege of ABC Cutting Contractors on behalf of Ms. Helen Thomas Haney of Jablonski Building Conservation and Mr. Nathaniel Hicks of Silman. The samples are reported to have been taken from the Lincoln Memorial Bridge, a ca. 1832 reinforced concrete bridge in Vincennes, IN.

In Field Report 3, dated April 20, 2023, Mr. Hicks requested a test distribution for the samples. The author had requested a redistribution to better suit the size and condition of the samples. Mr. Hicks agreed to the amendment. The request includes petrographic examination, water-soluble chloride determination, and compressive strength testing. The petrographic study includes an identification of constituents, assessment of overall condition, and an investigation of the potential causes of any observed distress. No specific emphasis has been requested for the petrography. The sample identifications, locations identified in the field report, and final test distributions are indicated in the following table.

Table 2.2: Summary of Core Samples and Test Distribution

Core ID Location Testing performed C-1 #1 Petrographic examination, water-soluble chloride determination at four depths C-2 #2 Petrographic examination, water-soluble chloride determination at four depths C-3 #4 Water-soluble chloride determination at three depths C-4 #1 Compressive strength testing C-5 #2 Compressive strength testing C-7 #2 No testing C-8 #4 Compressive strength testing C-9 #3 Petrographic examination, water-soluble chloride determination as a full-depth average

3. Methods of Examination The petrographic examination was conducted in accordance with the standard practices contained in ASTM C856/C856M-20. Data collection is performed or supervised by a degreed geologist who by nature of their education is qualified to operate the analytical equipment employed. Analysis and interpretation are performed or directed by a supervising petrographer who satisfies the qualifications as specified in Section 4 of ASTM C856.

Water-soluble chloride determinations were performed in accordance with ASTM C1218/C1218M-20.

Compressive strength testing was performed in accordance with ASTM C39/C39M-21 as modified by ASTM C42/C42M-20.

Capping methods follow those of ASTM C617/C617M-15. The laboratory does not have the accreditation necessary to perform compression testing for the certification or acceptance of new construction.

The following personnel contributed to the examination:

Technicians: A. Ledwitch M. Pattie Laboratory Assistant: M. Cortez Chemist: H. Hartshorn Petrographer: J. Walsh

Lincoln Memorial Bridge Page 5 of 29

4. Laboratory Findings and Discussion

4.1 - Materials The three core samples contain similar materials and appear to represent the same design despite variations in aggregate gradation. The proportioning of the concrete constituents is within the range expected for conventional normal weight portland cement concrete. Mix water contents were appropriately low to ensure a dense matrix that is not excessively water-permeable. At the same time, the water contents were sufficient to ensure workable mixtures. The cement appears sound and there is no evidence for excessive free lime or other unstable components. The sand quality is suitable for the application.

While rounded gravels are generally adequate for conventional concrete mixtures, the aggregate is considered deficient due to the presence of potentially reactive components.

The material in all three cores is identified as a normal weight portland cement concrete prepared with moderately low water to cementitious materials ratio (w/cm). Though the ratios cannot be quantified petrographically, all are estimated to have been in the mid to high 0.4s. The portland cement has microtextural characteristics consistent with products manufactured during the first few decades of the twentieth century. No supplementary cementitious materials are identified in any sample.

The concrete does not exhibit an air structure that would indicate intentional air-entrainment. Of course, the ca. 1932 construction date just predates the introduction of air-entraining admixtures. No evidence for any other admixtures are detected petrographically.

Aggregate contents are relatively high in all samples and there is a good packing arrangement of particles. The coarse aggregate is a clean, rounded, natural gravel consisting primarily of sedimentary rock grains. These include limestone, dolostone, chert, and sandstone. The rock types are hard and inelastic. The aggregate is generally non-porous aside from some minor secondary porosity in the limestone and chert. Though the aggregate is physically suitable for stone to be used in portland cement concrete, the relatively abundant chert has a high potential for alkali-silica reactivity and its presence in the concrete is considered a significant deficiency.

The aggregate gradations are not very consistent throughout the sample set. In fact, there does not appear to have been an effort to grade the stone at all. It is possible that the natural deposit was washed, and the extreme sizes screened out.

Otherwise, the particle size distributions were likely accepted as quarried. Table 4.1a summarizes some of the key features of each gradation profile. Note that nominal top sizes are estimated to vary between 1" and 2".

Table 4.1a: Summary of Aggregate Gradations The information presented in this table is estimated through qualitative observations made petrographically (Tables 7.1 through 7.3).

Sample ID Approximate gradation Nominal top size Remarks C-1 No. 6 1" Majority retained on 3/8" C-2 No modern category 2" with most if not all passing Possibly bimodal with a minima near 3/8" C-9 No. 57 or No. 6 At least 1.5" Most coarser than 3/8"

The fine aggregate is a well-graded natural sand derived from the same source as the coarse aggregate. The only compositional difference is the presence of monocrystalline quartz grains. There are no clay linings, friable materials, or obvious organic impurities. The sand contents are moderate and appropriate for conventional portland cement concrete.

Grain shapes are suitable for sand to be used in cementitious mixtures. The gradations are estimated to comply with moderate standards for concrete sand with a majority of grains sized between the No. 16 and No. 50 sieves.

Lincoln Memorial Bridge Page 6 of 29

4.2 - Original Placement and Hydration Based on the examined samples, the concrete constituents were well-mixed, properly deposited, and adequately cured. The mixtures were fully consolidated and there is no segregation of ingredients. Fine water-voids are identified in Core C-1 along what is assumed to be the undersides of many coarse aggregate particles. These voids represent places where bleed water became entrapped as it migrated upward through the fresh mixture. All else being equal, such voids reduce the paste-aggregate bond and may cause the concrete to be somewhat weaker relative to its design strength. That said, this type of water-void is very common in larger concrete placements. Assuming the concrete meets the structural requirements of the application, these are not considered a significant defect.

4.3 - Features Related to Potential for Steel Corrosion An assessment of the corrosion potential of steel reinforcement is outside the scope of a petrographic examination. Still, some features identified petrographically and chemically have a bearing on corrosion potential.

Water-soluble chloride contents were measured for all three cores and results are presented in Section 5. Chloride values are low throughout the full depth of each core and there is no apparent risk for corrosion due to salt attack. The low values indicate that no chloride-based accelerating admixtures were used in the concrete mixes. In fact, the chloride contents are essentially identical for Cores C-1, C-2, and C-3. This is more likely to reflect the minor amount of chloride normally present in portland cement. It is noted that Core C-9 contains about twice as much chloride though still well below a level that would be concerning. Admittedly, it was necessary to take an average value across this core rather than testing at more precise intervals. It may simply be that there is a significant amount of contamination just at the surface. Alternatively, the pervasive cracks in this concrete may have allowed chloride salts to permeate readily into the element where this was not possible in the sound concrete.

Similarly, there is no risk for pH-related depassivation of any properly embedded steel based on the carbonation depths identified in the examined core samples. In most cases, carbonation of the cement paste extends no more than 2 millimeters from the formed surface. The interior side of Core C-1 is the one exception with a carbonation depth that extends 25 millimeters into the concrete. However, this profile is gradational such that the lowest pH is only present within the outermost few millimeters.

Finally, all concrete is estimated to have been mixed with a sufficiently low mix water content. This has resulted in a dense, well-formed cementitious matrix that is more likely to resist the ingress of moisture, oxygen, salts, or other chemicals. Only where there is cracking is there a pathway for significant water intrusion. Based on the several cores examined, this is likely to be the case wherever the concrete is suffering from alkali-silica reaction.

With all this said, a more thorough evaluation of corrosion potential must be deferred to a corrosion specialist. The statements made here are tentative and offered without any firsthand knowledge of the jobsite conditions or electrical properties of the concrete and any embedments.

4.4 - Existing Condition and Durability The concrete examined for this report is a typical example of conventional normal weight concrete as it would have been designed in the 1930s when the bridge was built. Where not affected by secondary reactions, the concrete remains sound as represented by Cores C-1 and C-2. These two samples are free of any significant cracking or chemical distress with the exception of very minor microcracking within the interiors of chert aggregate grains.

Lincoln Memorial Bridge Page 7 of 29

As expected for a pre-1940 structure, there is no air-entrainment in the mixtures that might mitigate the susceptibility to freeze-thaw distress. To some extent, the resistance is aided by suitably low water to cement ratios that result in lower permeability. However, some resistance to freeze-thaw distress may also be attributed to the quality of the cement. Cements in the early twentieth century were more coarsely ground than those manufactured today, and also tended to contain less of the more rapidly hydrating alite phase. These earlier cements gained strength more slowly than modern cements. This in turn reduced the tendency for microscopic shrinkage cracking. All else being equal, concrete made with early twentieth century portland cement is likely to be more durable than the same mixture made with modern portland cement. With all this said, the concrete examined for this report is understood to have been taken horizontally meaning that its exposure class relative to modern ACI categories would not be higher than an F1 (American Concrete Institute, 12). The author is not aware if there is concrete more exposed to saturation that has experienced some freeze-thaw distress.

The one feature that raises a concern for the long-term durability of the represented concrete is the occurrence of an alkali-silica reaction (ASR) in Core C-9, and quite probably in Core C-6 that was only observed in photographs. ASR is a reaction in which unstable varieties of silica in the aggregate react with cement paste alkalis to form a hygroscopic gel product. This gel can expand greatly upon the absorption of water and this expansion can often be sufficient to exceed the tensile strength of the concrete. In this case, chert within the coarse aggregate is the reactive phase.

Core C-9 includes only the outermost 4" of the concrete element. Nevertheless, a moderately advanced ASR reaction is identified in the core. Cracks are subparallel to the face of the concrete and have spacings between 0.25" and 1". Reaction rims and cracking are common in most chert grains identified petrographically. However, the reaction gels that extend into the adjacent cement paste are not that advanced. Nevertheless, it may be fair to state that the material has exceeded its useful service life based on the pervasiveness of the reaction.

As stated above, the other two cores examined for this report exhibit only a trace reaction though they both contain chert in the coarse aggregate. Assuming an extended service life is intended for the bridge, it is prudent to consider all concrete susceptible to eventual ASR even if no significant reaction has yet occurred. There are several parameters that must be considered when evaluating the causes for differences in the degree of reaction:

• The chert in Core C-9 could be present at a pessimum proportion meaning that it is at an optimum content to generate maximum ASR expansion. This would imply that the chert content in Cores C-1 and C-2 are either too low or too high to have produced a reaction. Admittedly, it is not practicable to quantify the chert contents in each concrete sample. Further, it is impossible to determine the pessimum proportion for this particular aggregate once it is mixed into the concrete. Nevertheless, the amount of chert present should be enough in all cores to have resulted in some type of reaction if all other parameters were identical.

• ASR requires the cement paste to be sufficiently alkaline to allow decomposition of the reactive stone. The bulk of the concrete is uncarbonated in all cores. Consequently, the concrete has a sufficiently high alkalinity to promote dissolution of the unstable silica.

• Reactions can only proceed if there is soluble sodium and potassium in the hydrated cement paste. Since chloride contents are low in all cores, it might be reasonable to assume that there has been little ingress of environmental alkali. However, the chloride contents do not necessarily relate to initial or residual alkali contents. Further testing would be necessary to evaluate this particular parameter (Fournier, 28).

• There must be sufficient pore water available to facilitate the various stages of ASR. It is not unusual for concrete that has highly reactive aggregate and more than sufficient alkali loading to exhibit no reaction at all if the concrete has remained dry in service.

Lincoln Memorial Bridge Page 8 of 29

It is unlikely that any concrete having these same constituents is immune to ASR. Saturation frequency is likely to have been the critical parameter in determining whether any area of concrete has begun to react. Of course, this is something that would be better verified in the field. It assumed that concrete experiencing any appreciable reaction would be exhibiting map-cracking or other pattern-cracking depending on the element's geometry. With all this said, the susceptibility to ASR does not indicate that concrete that is currently sound cannot remain serviceable indefinitely.

4.5 - Miscellaneous Other secondary service effects are described in the petrographic observations tables in Section 7. These include some very minor reaction rims around dolomitic aggregate and associated "blooms" of calcium carbonate in the adjacent cement paste.

These are dedolomitization reactions that are found in all three examined cores. Alkali-carbonate reactions (ACR) that have been called into question as a root cause of concrete distress often exhibit such dedolomitization reactions. However, the reactions are considered completely innocuous in these samples. The aggregates do not have compositions thought to be susceptible to ACR.

In Core C-9, there is significantly more secondary ettringite in the pervasive cracks than there is ASR gel. However, the passive deposition of ettringite often accompanies ASR. There is no evidence to suggest that any type of sulfate attack has occurred in the concrete represented by this sample.

References ACI Committee 201. Guide to Durable Concrete: (ACI 201.2R-16). Farmington Hills, MI: American Concrete Institute, 2022.

Fournier, B., Bérubé, M., Folliard, K.J., and Thomas, M., “Report on the Diagnosis, Prognosis, and Mitigation of Alkali-Silica Reaction (ASR) in Transportation Structures,” U.S. Department of Transportation, Federal Highway Administration, Report No. FHWA-HIF-09-004, 2010.

Lincoln Memorial Bridge Page 9 of 29

5. Chloride Content Determination For samples C-1, C-2, and C-3, water-soluble chloride analysis was performed at three to four depths. Each core was cut down the middle longitudinally, and half of the core reserved for chloride testing. From this core half, saw cuts were made to extract half-round wafers of approximately 0.5" thickness spaced evenly throughout the depth of the concrete. In the table below, the 0" depth represents the outer 0.5" of the concrete. The remaining reported depths represent the approximate center of each test wafer. For Core C-3, the deepest interval is somewhat irregular due to the core break surface.

Because Core C-9 was the only one from Location #3, and the only one for which a significant chemical distress was visible, the laboratory considered it important to determine chloride contents for this core. However, there was little material available. Instead of subsampling multiple depth intervals, a single thin wedge was sawn down the full length of the core.

Chloride content was averaged from this section.

Water-soluble chloride determinations were performed in accordance with ASTM C1218/C1218M-20. The air-dry density is assumed to be 3960 lbs./yd.3 for normal weight concrete. This represents an average determined from the compressive strength testing presented in the following section. The density value is used in calculating the chloride content per cubic yard of concrete. To calculate chloride by cement mass, a cement content of 14% by weight was assumed.

Table 5.1 - Chloride Content Results

Sample ID Depth Chloride content Chloride by concrete mass Chloride by cement mass (in.) (ppm) (lbs./yd.3) (%)

C-1

0.0 32.6 0.129 0.023

4.0 28.5 0.113 0.020

8.0 27.0 0.107 0.019

11.75 30.3 0.120 0.022

C-2

0.0 28.5 0.113 0.020

5.0 30.9 0.122 0.022

9.0 27.6 0.109 0.020

14.5 27.2 0.108 0.019

C-3

0.0 33.0 0.131 0.024

3.0 34.5 0.136 0.025

6.0 30.2 0.120 0.022

C-9 0 to 4.5 66.4 0.263 0.047

Notes:

1. For context, ACI 222R-19 recommends a water-soluble chloride limit by cement mass of 0.15% for newly placed reinforced concrete in dry conditions.

Lincoln Memorial Bridge Page 10 of 29

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0 2 4 6 8 10 12 14 16

Ch lo rid e co nt en t b y ce m en tit io us m as s (

Depth interval (in.)

Core C-1

Core C-2

Core C-3

Core C-9

Chart 5.1: Chloride contents are plotted as a percentage of the total weight of estimated cementitious materials as a function of depth in each core. Note that an average chloride content was determined for the full thickness of Core C-9. That value is plotted as a constant on the graph though it is possible that chloride varies with depth in this sample. The red line represents the 0.15% limit that ACI 222R-19 recommends for newly placed reinforced concrete in dry conditions. This value is offered for comparative purposes only.

Lincoln Memorial Bridge Page 11 of 29

6. Compressive Strength Testing All cores used for compressive strength testing were received in two contiguous pieces. The following represent interruptions to the testable length of each core sample:

• Core C4 has a core break at 7.5" depth. There is a No. 3 reinforcing rod with 1.75" cover from the inner surface.

There is also some minor honeycombing toward the front surface and this is included in the fabricated test piece.

• Core C5 has a core break at 9" depth.

• Core C8 has a core break at approximately 1" depth.

Table 6.1 - Compressive Strength Results

Sample ID C4 C5 C8 Length as drilled (in.) 11.75 13.75 6.75 Length cut (in.) 6.50 7.30 4.70 Length capped (in.) 6.75 7.55 4.92 Diameter (in.) 3.67 3.67 3.67 L/D ratio 1.84 2.06 1.34 Weight (lbs.) 5.9 6.5 4.2 Density (lbs./ft.3) 1 148 145 146 Area (in.2) 10.57 10.59 10.57 Load (lbf) 71650 44920 128220 Fracture type 2 4 4 2 Compressive strength (psi) 6777 4243 12131 Correction factor 3 n/a n/a 0.94 Corrected compressive strength (psi) 6780 4240 11440

1. The density is computed by dividing the as-prepared sample weight by the sample volume. The latter is determined simply from the measured length and diameter. The determined values may differ somewhat from more accurately measured densities using immersion methods.

2. Core C4 exhibited a fracture type that was close to a Type 4 except that the fracture extended into the cap on one side of the sample.

3. A correction factor is necessary when the L/D ratio is less than or equal to 1.75.

Lincoln Memorial Bridge Page 12 of 29

7. Petrographic Data

Table 7.1: Petrographic Data (Core C-1)

Sample ID C-1

As-received description Dimensions/details The sample consists of a 5.75" diameter core with a 12.0" length. The core was received as one intact piece.

The core is understood to have been taken horizontally though the orientations were not identified by the coring contractor. The outer direction was assumed by the laboratory based on the location of the coring burr. The vertical plane was assumed by the orientation of board form impressions as indicated by the contractor and Silman by email. The upward direction remains unknown.

Outer surface The outer face is a compact and roughly planar formed surface with board form impressions. No bugholes or other defects of forming or form removal are identified.

Inner surface The inner face is a mostly compact and roughly planar formed surface with board form impressions. There is some shallow clumping and rock-pocketing. However, no honeycombing is observed in cross section.

Core circumference Smooth and untextured.

Embedded items Three levels of embedded reinforcement are included in the sample. All are well-encased with no corrosion product evident. The innermost bar has approximately 1" of cover from the inner surface. The reinforcing includes:

• A No. 3 bar located 4.0" on center from the outer face

• A No. 3 bar located 8.0" on center from the outer face

• A No. 5 bar located 10.7" on center from the outer face

Visible cracks No cracks are visible in hand sample.

Other details Some minor white mineral scale is observed along the outer face of the concrete

Coarse aggregate Type Natural gravel Estimated volume Approximately 35-40% by hardened concrete volume Lithology The stone consists mostly of sedimentary rock types including micritic limestone, dolostone, chert, quartz arenite

(± carbonate), various wackes (mostly quartz wackes). Granite is a minor constituent.

Nominal top size 1" Grading There is a broad distribution of particles from the 1" to the No. 8 sieve. A majority of the stone is estimated to be retained on the 3/8" sieve. The particle size distribution would probably comply with a No. 6 gradation profile.

Roundness Rounded Aspect Mostly equidimensional Coatings and rims Darker reaction rims are observed in lighter-colored chert grains. A minor proportion of dolostone grains exhibit dedolomitization rims.

Cracks None significant

Fine aggregate Type Natural sand Estimated volume Approximately 35% by hardened mortar fraction Lithology The sand consists of quartz along with rock grains that have essentially the same composition as the coarse aggregate.

Nominal top size The top size is difficult to discern due to the similarity between coarse and fine aggregates. Given the narrowness of the fine aggregate gradation, the top size is probably no coarser than the No. 8 sieve.

Grading The sand is somewhat narrowly graded with a majority of grains present between the No. 16 and No. 50 sieves.

Roundness Mostly rounded Aspect Mostly equidimensional Clay coatings or friable materials None observed Organic impurities None observed

Lincoln Memorial Bridge Page 13 of 29

Table 7.1 (cont'd.): Petrographic Data (Core C-1)

Cement paste Hand sample qualities Intentional fractures are mostly sharp with a little crumbliness. The fractures mostly deflect around aggregate grains. The freshly exposed cement paste is medium gray with a moderately dull luster. The paste is moderately hard and somewhat rapidly water-absorptive.

Paste microstructure The cured paste has a microscopically mottled texture defined by patches of denser paste surrounded by a more permeable matrix. However, the binder has a moderate capillary porosity on average.

Calcium hydroxide Primary calcium hydroxide is mostly depleted throughout the paste.

Evidence for admixtures 1 No admixtures are obvious based on indirect petrographic evidence.

Residual cement 2 The cement hydration is extreme and very little hydraulic calcium silicate remains unhydrated. Nevertheless, there is a moderately high concentration of residual cement particles in the form of so-called "Williamson grains". These are pores in the cement paste filled with cementitious hydrate that retain the form of the now-hydrated portland cement particles. The impressions indicate medium-sized agglomerates containing both alite and belite with an interstitial ferrite matrix. Many of the agglomerates would have been retained on a No. 200 sieve.

SCMs 3 None identified Other inclusions None identified Variations The paste hydration characteristics are mostly uniform throughout the examined cross section.

W/cm estimate 4 The water to cement ratio is estimated to have been moderately low. The ratio was likely somewhere in the mid to high 0.4s.

Air-void system Entrained air None identified Entrapped air Entrapped air-voids are estimated at approximately 2-3%. These are generally rounded and sometimes deformed voids up to several millimeters in size. These are evenly distributed throughout the full depth of the core. There are no honeycombs or other coarse void structures.

Paste-aggregate interfaces Description Microscopic water-voids are relatively common consistently along one side of coarse aggregate grains. The sectioning of these gravity-related microstructures is probably not optimized given the uncertainty of the core orientations. No significant microporous zones are observed in the paste adjacent to aggregate particles.

Original placement features Mixing of dry ingredients There are no cement lumps, sand streaks, or aggregate concentrations. All constituents are estimated to have been well mixed.

Mix water The mix water was well incorporated into the fresh mixture and there is no evidence for inappropriate retempering.

Consolidation/ compaction The concrete mixture was adequately consolidated.

Segregations None detected Preferential orientations None significant Bleed water features The water-voids present along one side of many aggregate grains represents bleed water that became entrapped underneath aggregate surfaces as it was migrating upwards through the concrete.

Cold joints/ other contacts None detected Laitance None detected Forming The outer formed surface is compact and well-formed. The inner formed surface has some shallow clumping and rock-pocketing but no significant defect.

Finish Not applicable Curing The portland cement is thoroughly cured and the cementitious paste is uniformly developed.

Lincoln Memorial Bridge Page 14 of 29

Table 7.1 (cont'd.): Petrographic Data (Core C-1)

Secondary service effects Surface erosion/weathering None significant Carbonation (outer) Carbonation of the cement paste extends no more than 1 to 2 millimeters from the outer surface.

Carbonation (inner) A gradational carbonation reaction extends up to approximately 25 millimeters from the interior surface.

Cracking None identified Chemical alteration in paste None identified Mineral deposits Traces of ettringite are observed in air-voids. Note that other soluble salts are typically distributed throughout the cement paste capillaries and may not be detectable petrographically.

Aggregate reactivity Aggregate reactions are minor. Reaction rims are fairly common in chert particles. No cracking or reaction gel is observed in association. Some dedolomitization is noted in the carbonate aggregate. No cracking is associated with this reaction either.

Other materials Details None observed

1. Organic admixtures or soluble inorganic additives cannot be detected petrographically, though their effects on concrete microstructure can sometimes be discerned.

2. The presence of unhydrated cement grains does not necessarily indicate poor hydration of the binder. A greater abundance of unhydrated cement is typical with reduced w/cm.

3. SCMs = supplementary cementitious materials.

4. Water to cementitious materials ratio (w/cm) cannot be quantified petrographically. Estimates are made qualitatively based on many characteristics of the hardened cement paste including behavior of the paste when fractured, capillary porosity, concentration of residual unhydrated cement, and microtextural features of the primary calcium hydroxide formed during the cement hydration. The estimates are complicated by the presence of SCMs, organic additives, or through chemical changes that occur with age.

Lincoln Memorial Bridge Page 15 of 29

Table 7.2: Petrographic Data (Core C-2)

Sample ID C-2

Dimensions/details The sample consists of a 5.75" diameter core with a 15.0" length. The core was received as one intact piece.

The core is understood to have been taken horizontally though the orientations were not identified by the coring contractor. The inner direction was assumed by the laboratory based on the presence of a membrane on this side. The vertical plane was assumed by the orientation of board form impressions as indicated by the contractor and Silman by email. The upward direction remains unknown.

Outer surface The outer face is a compact and roughly planar formed surface with board form impressions. No bugholes or other defects of forming or form removal are identified. There is a dark bituminous? coating below a layer of carbonate scale.

Inner surface The inner face has an adhered membrane over a cementitious coating. The original surface appears relatively planar but not easily viewed.

Core circumference Smooth and untextured.

Embedded items One No. 3 reinforcing bar is located approximately 10.5" on center from the outer surface. The bar is well-encased and exhibits no corrosion product.

Visible cracks No cracks are visible in hand sample.

Other details Some minor white mineral scale is observed along the outer face of the concrete.

Coarse aggregate Type Natural gravel Estimated volume Approximately 35-40% by hardened concrete volume Lithology The stone consists mostly of sedimentary rock types including micritic limestone, dolostone, chert, quartz arenite

(± carbonate), various wackes (mostly quartz wackes). Granite is a minor to lesser constituent. Traces include diabase, basalt, and cordierite gneiss.

Nominal top size 2" with most if not all passing Grading There is a broad distribution of particles from the 1" to the No. 8 sieve. The aggregate appears to be somewhat bimodal with a trough near the 3/8" sieve. The particle size distribution does not appear to comply with modern aggregate gradation categories.

Roundness Rounded Aspect Mostly equidimensional Coatings and rims Reaction rims are evident on chert grains. A minor proportion of dolostone grains exhibit dedolomitization rims.

Cracks Minor internal microcracking is detected in some chert and rarely in dolostone.

Fine aggregate Type Natural sand Estimated volume Approximately 35% by hardened mortar fraction Lithology The sand consists of quartz along with rock grains that have essentially the same composition as the coarse aggregate.

Nominal top size The top size is difficult to discern due to the similarity between coarse and fine aggregates. Given the narrowness of the fine aggregate gradation, the top size is probably no coarser than the No. 8 sieve.

Grading The sand is somewhat narrowly graded with a majority of grains present between the No. 16 and No. 50 sieves.

Roundness Mostly rounded Aspect Mostly equidimensional

Lincoln Memorial Bridge Page 16 of 29

Table 7.2 (cont'd.): Petrographic Data (Core C-2)

Hand sample qualities Intentional fractures are mostly sharp with a little crumbliness. The fractures mostly deflect around aggregate grains. The freshly exposed cement paste is medium gray with a moderately dull luster. The paste is hard and somewhat slowly water-absorptive.

Paste microstructure The cured paste has a microscopically mottled texture defined by patches of denser paste surrounded by a more permeable matrix. However, the binder has a moderate capillary porosity on average.

Calcium hydroxide Primary calcium hydroxide is mostly depleted throughout the paste.

Evidence for admixtures 1 No admixtures are obvious based on indirect petrographic evidence.

Residual cement 2 The cement hydration is extreme and very little hydraulic calcium silicate remains unhydrated. Nevertheless, there is a moderately high concentration of residual cement particles in the form of both "Hadley grains" and "Williamson grains". The former are empty pores in the cement paste that retain the form of the now-hydrated portland cement particles. The Williamson grains are similar but are filled with cementitious hydrate. The impressions indicate medium-sized agglomerates containing both alite and belite with an interstitial ferrite matrix. Many of the agglomerates would have been retained on a No. 200 sieve.

SCMs 3 None identified Other inclusions None identified Variations The paste hydration characteristics are mostly uniform throughout the examined cross section.

W/cm estimate 4 The water to cement ratio is estimated to have been moderately low. The ratio was likely somewhere in the mid to high 0.4s.

Air-void system Entrained air None identified Entrapped air Entrapped air-voids are estimated at approximately 2-3%. These are generally rounded and sometimes deformed voids up to several millimeters in size. These are evenly distributed throughout the full depth of the core. There are no honeycombs or other coarse void structures.

Description No significant water-voids or microporous zones are observed in the paste adjacent to aggregate particles.

Original placement features Mixing of dry ingredients There are no cement lumps, sand streaks, or aggregate concentrations. All constituents are estimated to have been well mixed.

Mix water The mix water was well incorporated into the fresh mixture and there is no evidence for inappropriate retempering.

Consolidation/ compaction The concrete mixture was adequately consolidated.

Segregations None detected Preferential orientations None significant Bleed water features None significant Cold joints/ other contacts None detected Laitance None detected Forming The outer formed surface is compact and well-formed. The inner formed surface appears sound but it is not readily viewed due to the presence of a cementitious coating.

Lincoln Memorial Bridge Page 17 of 29

Table 7.2 (cont'd.): Petrographic Data (Core C-2)

Surface erosion/weathering None significant Carbonation (outer) Carbonation of the cement paste extends no more than 1 to 2 millimeters from the outer surface.

Carbonation (inner) None significant Cracking Cracking is limited to minor microcracks within chert particles.

Chemical alteration in paste None identified Mineral deposits Traces of ettringite are observed in air-voids. Note that other soluble salts are typically distributed throughout the cement paste capillaries and may not be detectable petrographically.

Aggregate reactivity Aggregate reactions are minor. Reaction rims are fairly common in chert particles. Microcracking is observed within these same particles. However, the cracks do not extend into the adjacent cement paste. Reaction gels are observed in exceedingly trace abundance as fine deposits along aggregate surfaces. Some dedolomitization is noted in the carbonate aggregate. No cracking is associated with this reaction.

Details A 5 millimeter thick cementitious coating is present along the inner surface. This material was not observed petrographically.

1. Organic admixtures or soluble inorganic additives cannot be detected petrographically, though their effects on concrete microstructure can sometimes be discerned.

2. The presence of unhydrated cement grains does not necessarily indicate poor hydration of the binder. A greater abundance of unhydrated cement is typical with reduced w/cm.

3. SCMs = supplementary cementitious materials.

4. Water to cementitious materials ratio (w/cm) cannot be quantified petrographically. Estimates are made qualitatively based on many characteristics of the hardened cement paste including behavior of the paste when fractured, capillary porosity, concentration of residual unhydrated cement, and microtextural features of the primary calcium hydroxide formed during the cement hydration. The estimates are complicated by the presence of SCMs, organic additives, or through

Lincoln Memorial Bridge Page 18 of 29

Table 7.3: Petrographic Data (Core C-9)

Sample ID C-9

Dimensions/details The sample consists of a 3.75" diameter core broken along an oblique fracture surface. The short length is 2.8" and the long length is 4.5". The core is understood to have been taken horizontally though the orientations were not identified by the coring contractor. The outer face is assumed to be the unfractured surface. There is no evidence to help identify the vertical direction.

Outer surface The outer face is roughly planar though mostly hidden by a scabbly mortar residue over a dark veneer.

Inner surface The inner face is an oblique crack transecting approximately 12 coarse aggregate grains most of which display reaction rims. The mortar within the crack is coated with white mineralizations. There is also some darker brown "bleeding" around reacted aggregate grains.

Core circumference Smooth and untextured.

Embedded items No reinforcement is included with the sample.

Visible cracks Surface parallel cracks are observed in many places around the core circumference. There is at least one continuous crack through the concrete at approximately 2.25" depth. The core break is also interpreted to represent a pre-existing crack.

Coarse aggregate Type Natural gravel Estimated volume Approximately 35-40% by hardened concrete volume Lithology The stone consists mostly of sedimentary rock types including micritic limestone, dolostone, chert (and chalcedony), quartz arenite (± carbonate), various wackes (mostly quartz wackes). Granite is a minor constituent. Traces include diabase, basalt, and associated greenstone.

Nominal top size Only a small amount of material was available for study. The nominal top size is at least 1.5".

Grading Most of the visible aggregate is estimated to be coarser than a 3/8" sieve. Based on the small area visible, the particle size distribution would likely comply with either a No. 57 or a No. 6 stone.

Roundness Rounded Aspect Mostly equidimensional Coatings and rims Reaction rims are evident on chert grains. A minor proportion of dolostone grains exhibit dedolomitization rims.

Cracks Cracking related to alkali-silica reaction is observed in some aggregate grains.

Fine aggregate Type Natural sand Estimated volume Approximately 35% by hardened mortar fraction Lithology The sand consists of quartz along with rock grains that have essentially the same composition as the coarse aggregate.

Nominal top size The top size is difficult to discern due to the similarity between coarse and fine aggregates. Given the narrowness of the fine aggregate gradation, the top size is probably no coarser than the No. 8 sieve.

Grading The sand is somewhat narrowly graded with a majority of grains present between the No. 16 and No. 50 sieves.

Roundness Mostly rounded Aspect Mostly equidimensional

Lincoln Memorial Bridge Page 19 of 29

Table 7.3 (cont'd.): Petrographic Data (Core C-9)

Hand sample qualities Intentional fractures are crumbly and fail across pre-existing cracks related to alkali-silica reaction. The fractures mostly transect aggregates. The freshly exposed cement paste is medium gray with a moderately subvitreous translucent luster. The paste is hard and somewhat slowly water-absorptive.

Paste microstructure The cured paste has a microscopically mottled texture defined by patches of denser paste surrounded by a more permeable matrix. However, the binder has a moderate capillary porosity on average.

Calcium hydroxide Primary calcium hydroxide is mostly depleted throughout the paste.

Evidence for admixtures 1 No admixtures are obvious based on indirect petrographic evidence.

Residual cement 2 The cement hydration is extreme and very little hydraulic calcium silicate remains unhydrated. Nevertheless, there is a moderately high concentration of residual cement particles in the form of both "Hadley grains" and "Williamson grains". The former are empty pores in the cement paste that retain the form of the now-hydrated portland cement particles. The Williamson grains are similar but are filled with cementitious hydrate. The impressions indicate medium-sized agglomerates containing both alite and belite with an interstitial ferrite matrix. Many of the agglomerates would have been retained on a No. 200 sieve.

SCMs 3 None identified Other inclusions None identified Variations The paste hydration characteristics are mostly uniform throughout the examined cross section.

W/cm estimate 4 The water to cement ratio is estimated to have been moderately low. The ratio was likely somewhere in the mid to high 0.4s.

Air-void system Entrained air None identified Entrapped air Only a few inches of cross section are available for evaluation. In this area, entrapped air-voids are estimated at less than 2% by volume. Most are less than three millimeters in diameter. There are no honeycombs or other coarse void structures.

Description No significant water-voids or microporous zones are observed in the paste adjacent to aggregate particles.

Original placement features Mixing of dry ingredients There are no cement lumps, sand streaks, or aggregate concentrations. All constituents are estimated to have been well mixed.

Mix water The mix water was well incorporated into the fresh mixture and there is no evidence for inappropriate retempering.

Consolidation/ compaction The concrete mixture was adequately consolidated.

Segregations None detected Preferential orientations None significant Bleed water features None significant Cold joints/ other contacts None detected Laitance None detected Forming The outer formed surface is compact and well-formed.

Lincoln Memorial Bridge Page 20 of 29

Table 7.3 (cont'd.): Petrographic Data (Core C-9)

Surface erosion/weathering None significant Carbonation (upper) Carbonation of the cement paste extends no more than 1 to 2 millimeters from the outer surface.

Carbonation (lower) The interior of the concrete element was not included with the sample.

Cracking See "Aggregate reactivity" below Chemical alteration in paste None identified Mineral deposits See "Aggregate reactivity" below Aggregate reactivity • The chert has undergone a moderately advanced alkali-silica reaction. Most grains have reaction rims and contain internal cracks. Some have gel plugs that extend into the adjacent cement paste though the plugs are relatively thin at this stage. Macroscopically, expansive cracks are observed that are subvertical in orientation with spacings from about 0.25" to 1". Secondary ettringite deposits are common where the cracks travel through the mortar fraction. Ettringite is observed within cracks in the chert as well in some places. Opening widths of the cracks are typically no more than 0.1 millimeters.

• Some dedolomitization is noted in the carbonate aggregate. No cracking is associated with this reaction.

Details None identified

1. Organic admixtures or soluble inorganic additives cannot be detected petrographically, though their effects on concrete microstructure can sometimes be discerned.

2. The presence of unhydrated cement grains does not necessarily indicate poor hydration of the binder. A greater abundance of unhydrated cement is typical with reduced w/cm.

3. SCMs = supplementary cementitious materials.

4. Water to cementitious materials ratio (w/cm) cannot be quantified petrographically. Estimates are made qualitatively based on many characteristics of the hardened cement paste including behavior of the paste when fractured, capillary porosity, concentration of residual unhydrated cement, and microtextural features of the primary calcium hydroxide formed during the cement hydration. The estimates are complicated by the presence of SCMs, organic additives, or through

Lincoln Memorial Bridge Page 21 of 29

Appendix I: Photographs and Photomicrographs

Microscopic examination is performed on an Olympus BX-51 polarized/reflected light microscope and an Olympus SZ40 stereoscopic microscope. The polarized light microscope is fitted with a Tucsen MIchrome 5 Pro 5MP digital camera. The stereoscopic microscope is used for simple magnification. Sample types examined under this microscope include fractured surfaces, fine constituents extracted through chemical or physical means, or honed or polished cross sections. The polarized light microscope (PLM) magnifies but also employs principles of optical crystallography. The most common sample preparation for the PLM is the petrographic thin section.