Transition Plan for AASHTO
May 2000

THE PRESENT

Successful Accomplishments

The ASR Lead State Team takes credit for accomplishing several key elements of their objectives. These include conducting a national survey on ASR; developing and maintaining an internet web site; updating the SHRP-C-315 publication; drafting a guide specification on ASR-Resistant Concrete for AASHTO (these last two items were posted on the ASR web site); and incorporating many of the state-of-the-art practices of ASR mitigation into their own states' specifications and policies. Each of these items is discussed below.

ASR Survey
State highway agencies were surveyed to assess the extent of ASR in the nation. The full text and results of the survey are included in Appendix D. A separate Microsoft Excel file (Appendix D ASR-Survey.xls) is included that shows the actual survey results, with embedded comments. Conclusions drawn from the responses showed an increased awareness of ASR both in detecting it and confirming its presence. The original SHRP studies found ASR generally evident in 35 states. This survey found that of the 38 responding agencies (including 3 Canadian provinces) 8 reported widespread ASR, 13 thought their ASR was localized, 11 felt it was not applicable, 2 did not know, and 4 did not respond to the question.

Web Site
The ASR Lead State Team provided information to AASHTO to include on the Internet web site of /leadstates/asr and kept it updated and current. The information available to the public included an updated version of the SHRP-C-315 publication, the AASHTO Guide Specification on ASR-Resistant Concrete, a publications list of ASR related documents, a questions and answer forum, the ASR Lead State contacts, ASR related terminology, a bibliography of ASR-related research; online training materials; databases on aggregates; a list of resources, equipment, and sources of funding for ASR-related projects; and a bulletin board for obtaining technical assistance from Lead States team members.

Updated SHRP-C-315
Published under SHRP, the Handbook for the Identification of ASR in Highway Structures was perhaps the most popular ASR product with simple text and plenty of photographs. Without modifying any of the original text, the ASR Lead State team elaborated on the detection technique. This updated text (but not photographs) of the publication is included in Appendix E. A separate Adobe Acrobat file (Appendix E ASR-C315.pdf) is included that shows the actual web pages with modified text and photographs.

ASR Guide Specification
The ASR Lead State team produced an ASR guide specification for AASHTO to incorporate the current state-of-the-art practices in producing ASR-resistant concrete. The Team included information from the Mid-Atlantic Regional Technical Committee, SHRP recommendations, practices from the Canadian Standards Agency, the Portland Cement Association, and individual states' experiences. This ASR guide specification for Portland Cement Concrete Resistant to Excessive Expansion Caused by Alkali-Silica Reaction is included in Appendix F. It was designed to help highway agencies and others draft their own specifications for mitigating ASR, and was submitted to the AASHTO Subcommittee on Construction for balloting and approval.

Technical Assistance and Research
Team members have provided technical assistance to highway agencies in California, Delaware, Maryland, Minnesota, Nevada, New Jersey, and New England and electric companies in New Mexico, Utah, and Wyoming. The team also participated in ASR-related research projects, including:

• Tests on in-service pavements in Pennsylvania, New Mexico and South Dakota of concrete mixes and additives that prevent ASR-related damage.
• Tests in South Dakota, New Mexico, North Carolina, and Maryland of ways to slow or stop ASR-related damage in existing pavements.
• Development of a new, environmentally safe test for ASR.
• A study on preventing the reoccurrence of ASR when using recycled aggregate from ASR-affected concrete.
• Laboratory tests of improved methods of determining the effectiveness of ASR mitigation techniques.

State Highway Agency Specifications and Policies
All of the state agencies represented on the ASR Lead State Team modified their specifications to include parts of their own recommendations. Descriptions of these are listed below for each of the five states of New Mexico, North Carolina, Pennsylvania, South Dakota, and Virginia.

New Mexico Department of Transportation
New Mexico implemented performance based ASR mitigation requirements in January 1999, over a year ago. Since that time, all of the suppliers in the state have been required to determine if their aggregates are reactive, and if so, how they will mitigate the potential reactions.

North Carolina Department of Transportation
The following is a brief summary of NCDOT implementation for prevention of ASR for new construction, as included in the specifications as the Standard Special Provision effective June 1977.

There are six aggregates in the state that are identified as highly or moderately reactive.
1) For these six aggregates, use cement alkali content equal to or less than 0.4%.
2) For other aggregates use cement alkali content, equal to or less than 0.6%.
3) Alternatively, use class F fly ash or other pozzolans (e.g., silica fume, and GGBFS) of specified amounts.

Pennsylvania Department of Transportation
Policy Concerning Use of Portland Cement Concrete Aggregates With the Potential to Cause Alkali-Silica Reactivity

The Pennsylvania Department of Transportation has had a specification in effect since August of 1992 which addresses the use in Portland cement concrete of aggregates which have the potential to cause alkali-silica reactivity (ASR). This specification has been revised twice since it was first written. At first, the specification concerning the use of aggregates with ASR potential was published as a special provision for all contract work, which required the use or Portland cement concrete. With the printing of Form 408/2000, "Specifications", the ASR special provision was incorporated into Section 704, "Cement Concrete", as Section 704.1(h), Mix Designs Using Potentially Reactive Aggregate.

The Pennsylvania Department of Transportation, Materials Testing Division has tested all Department-approved Portland cement concrete aggregates for the potential to cause ASR. The results of this testing are published in Bulletin #14, "Aggregate Producers". If the mortar bar expansion resulting from testing an aggregate by AASHTO TP 14 (now AASHTO T 303) is greater than 0.10% linear expansion, then any concrete mix utilizing this aggregate must be mitigated for ASR. The methods accepted for mitigation are as follows:

1. Use of a Portland cement meeting the optional requirement in AASHTO M 85 for a maximum alkali content of 0.60%.
2. Use of a blended hydraulic cement, Type IS or IP, ASTM C 595, from a manufacturer listed in Bulletin 15, "Approved Construction Materials".
3. Substitution of Class F fly ash, meeting the optional chemical requirement in AASHTO M 295 for a maximum alkali content of 1.5%, in an amount constituting between 15% and 25% of the total cementitious material by mass. If the AASHTO TP 14 mortar bar expansion of both coarse and fine aggregates proposed for use in a fly ash mix are greater than 0.10%, but less than 0.41%, then the minimum amount of fly ash may be used. If the AASHTO TP 14 mortar bar expansions or either or both aggregates proposed for use in the mix are 0.41% or greater, then the fly ash must constitute a minimum of 20% of the total cementitious material by mass.
4. Substitution of ground granulated blast furnace slag in an amount between 25% and 50% of the total cementitious mass. When the TP 14 mortar bar expansions of both aggregates proposed for use in the mix are greater than 0.10%, and less than 0.41%, the minimum substitution level may be used. If the TP 14 mortar bar expansion of either aggregate proposed for use in the mix is 0.41% or greater, then the ground granulated blast furnace slag must constitute a minimum of 40%, by mass, of the total cementitious material.
5. Substitution of between 5% and 10%, by mass, of the total cementitious material by silica fume.

South Dakota Department of Transportation
ASR Specifications
South Dakota DOT has specified the use of Type II LA (Low Alkali) cement for concrete since 1983 as a means of mitigating ASR. We also currently specify the use of Modified Class F fly ash on all paving projects excluding isolated small volume pours at a 1.25 replacement rate for a 15% reduction in cement content. This equates to a fly ash content of 18.75%. The reason for this blanket specification is the significant incidence of reactive fine aggregate throughout the state and the fact that one of our three quarried coarse aggregate (the only three used) is also ASR-reactive. A Modified Class F fly ash meets all the general AASHTO M295 requirements except as modified by the following:

Loss on ignition.............................................2.0% Maximum
Moisture content............................................2.0% Maximum
Sum of Oxides (SiO2 + Al2O3 + Fe2O3..........66.0% Minimum
SiO2 content.................................................40.0% Minimum

ASTM C1260 is used to screen fine aggregate sources and monitor possible changes in reactivity over time for a given source. If the 14 day expansion value from the test is > 0.250%, the cement used on the project employing this source is altered from a Type II LA to a Type V LA cement to further insure a substantial reduction in reactivity. The State is also in the process of reviewing the incorporation of lithium nitrate into new concrete containing reactive aggregate in conjunction with Modified Class F fly ash as additional insurance against potential reactivity. A series of test sections containing various lithium admixtures with fly ash were constructed in 1996 and are being evaluated.

Virginia Department of Transportation
ASR Implementation
In the early 1980's, prior to the start of SHRP, the Virginia Department of Transportation (VDOT) began to notice durability problems involving extensive cracking of some Portland cement concrete pavements and structures. Alkali-silica reaction (ASR) was identified as a primary cause of distress in these cases and they were included in the SHRP ASR project.

In 1990 the Virginia Transportation Research Council (VTRC) began a project determine the extent of the ASR problem confronting VDOT and assess strategies to avoid this durability problem. During the course of this research project, it became clear that a number of aggregate sources were potentially reactive and that the most suitable screening test for aggregates (now standardized as AASHTO T 303 and ASTM C 1260) was not sufficiently effective in discriminating between non-reactive and deleteriously reactive aggregates.

Consequently, to avoid future durability problems, VDOT changed its specification for concrete in December 1991. This change required the use of 20% (by mass of cementitious material) Class F fly ash (low-lime content), 30-50% ground slag, or 7% silica fume when the alkali content of the Portland cement to be used exceeded 0.40% Na2O equivalent (Lane, 1993a). This change was also supported by the knowledge that these mineral admixtures would produce concretes with low permeability that were thus more resistant to the other major concrete durability problem confronted by most DOTs, chloride-induced corrosion of reinforcing steel. In the Virginia market and many others, the use of Class F fly ash or ground slag offer economical alternatives to producing concrete with Portland cement as the sole cementitious material.

In 1992, VIRC initiated a study to (1) evaluate the effectiveness of particular mineral admixtures in combination with Portland cements of varying alkali content and (2) to establish the minimum amount of mineral admixture necessary to control a given cement alkali level. The findings of this study were published (Lane & Ozyildirim, 1995 and 1999a). Based on these findings, the VDOT concrete specification was revised in 1997 to require that the cementitious materials used in concrete mixtures limit the expansion of ASTM C 441, mortar (with Pyrex glass aggregate) to 0.1% at an age of 56 days. Compliance with the specification could also be met by using materials that conformed to requirements listed in Table 1.

Recognizing that these specification requirements were based on the testing of mortars made with Pyrex glass, VTRC initiated a new project to evaluate the performance of these cementitious materials in concretes produced with construction aggregates. This project encompassed a broader approach by evaluating important concrete properties such as strength and permeability as well as resistance to freezing and thawing, deicer scaling and ASR. The ASR testing was performed using a modified ASTM C 1293 test with reactive Virginia aggregates. The findings were published (Lane & Ozyildirim, 1999b) and clearly indicate that lower amounts of Class F fly ash (15-20%) or slag (35% min.) than currently specified should control the reactivity of Virginia aggregates with cements having alkali contents of 1.0% Na20eq or less. This suggested that an expansion limit of 0.15% at 56 days might be more appropriate for evaluating cementitious materials in C 441 intended for use with aggregates of similar reactive potential to those found in Virginia. The findings of this study support the continued use of mineral admixtures to produce concretes resistant to ASR and chloride-induced corrosion thus providing the durable concrete needed for an economical transportation system. Various aspects of VTRC's research efforts in ASR can also be found in Lane & Ozyildirim, 1999c; Lane, 1999a; Lane, 1999b; Lane, 1996; and Lane, 1993b.

Table 1. Material combinations accepted as conforming to the 1997 VDOT concrete specification without testing.

Material	Maximum Portland cement alkali content, % Na20eq
Portland cement only	0.45
Cement with min. 15% Class F fly ash	0.60
Cement with min. 20% Class F fly ash	0.68
Cement with min. 25% Class F fly ash	0.75
Cement with min. 30% Class F fly ash	0.83
Cement with min. 25% ground slag	0.60
Cement with min. 35% ground slag	0.90
Cement with min. 50% ground slag	1.00
Cement with min. 3% silica fume	0.60
Cement with min. 7% silica fume	0.90
Cement with min. 10% silica fume	1.00

In 1997, VTRC initiated a laboratory investigation to evaluate the effectiveness of lithium compounds in minimizing ASR-related expansion. This study should be completed in the near future. Preliminary findings indicate that lithium-bearing admixtures appear more effective with rapidly reactive materials than they do with the more slowly reactive aggregates found in Virginia presumably because a certain portion of the lithium is consumed during the hydration of the cementitious material. This suggests that the determination of lithium admixture dosage should be evaluated using longer-term tests following C 1293 with the construction aggregates, rather than in rapid tests such as C 441.

VDOT began permitting the use of fly ash or ground slag as cementitious materials in concrete in the mid-1980. Consequently there are now structures and pavements in place with up to 15 years service. VTRC has recently initiated a project to evaluate the performance of concretes containing fly ash or ground slag relative to the performance of Portland cement concretes. This study will provide the information necessary to assess the durability of these materials under field conditions and guidance for further specification development.

Management of pavements and structures affected by ASR has not differed significantly from the approach used with other deterioration mechanisms VDOT has experienced the largest maintenance problems with pavements, likely because of the repeated, active loading they receive. The section of I-64 at Charlottesville serves as the best example. This section of continuously reinforced concrete pavement (CRCP) was constructed in the early 1970's. By the late '70's extensive map cracking with a predominant longitudinal trend was noticed and the cause was determined to be ASR, primarily involving the coarse aggregate. This pavement also suffered from CRCP punch-outs at closely spaced (0.3-1.0 m) transverse cracks. Spalling began at the transverse cracks, and where punch-outs were repaired with full-depth patches, the rate of spalling increased on either side of the patch. By 1987, the worst stretch had become such a continual maintenance problem, that a decision was made to remove and reconstruct the section. The new CRCP was produced using the same aggregate sources, the same Portland cement source (having an alkali content of ~0.60% Na2Oeq), but including ground slag as 20% by mass of the cementitious material. Today, approaching 13 years in service, this section of pavement is performing quite well.

As maintenance problems mounted on the remaining sections of the 1970's CRCP at Charlottesville, a decision was reached in the early 1990's to quickly proceed with full--depth repairs of punch-outs and associated delaminated sections, installation of edge drains, followed by an asphalt overlay. This approach also appears to be working well, presumably because the overlay isolates the cracked concrete from the constant traffic- loading that would otherwise lead to its breakup. From this experience it would appear that when spalling at edges and joints or transverse cracks becomes common, it signals the beginning of a rapid decline in pavement condition requiring increasingly frequent repairs. Consequently, the onset of spalling is the point at which rehabilitation such as overlays or removal and replacement should be considered to avoid an excessive, burdensome, and eventually futile series of repairs.

Structures in Virginia have presented fewer problems, presumably because most elements are not dynamically loaded by traffic. Decks, which are dynamically loaded, usually are not cracked as extensively as the associated concrete because they lack a continual source of moisture. Routine repair and maintenance of joints, as well as providing for good drainage away from concrete elements should be emphasized to extend service-life of ASR-affected structures.

VDOT has not conducted any field trials using the topical application of lithium solutions to reduce the expansion of ASR-affected concrete. At this point in time this approach seems problematic. Much research is needed to ascertain not only if, and under what conditions, such treatment might be beneficial; but also to develop the methodologies needed to accurately assess the concrete condition to make such an assessment. Published findings of field trials conducted in other states will be of interest.

References for the Virginia Department of Transportation: