Reducing cementitious paste volume of slipformed pavement concrete by blending aggregates

A high cementitious paste volume (CPV) can increase the early cracking tendency of the concrete and reduce the durability of concrete pavement. This study investigated the effects of minimized CPV in slipformed pavement concrete (SPC) with blended aggregates (BA). Based on the laboratory results, the performance of pavement concrete with different CPV was evaluated. The CPV of standard SPC can be reduced to 25.0% without affecting its properties as evaluated by compressive strength, drying shrinkage and surface resistivity tests However, the CPV of SPC with optimized aggregate gradation (OAG) using BA technique can be further reduced to 22.5% with satisfactory properties. The SPC mixes with OAG was noted to have better potential performance as a pavement concrete. SPC concrete using Portland limestone cement can give similar properties as those of the conventional concrete using ordinary Portland cement.


Introduction
The concrete industry in Florida is facing two challenges, namely the rising cost of cement and the shortage of fly ash. The ascending price of cement has increased the initial cost of concrete. According to the forecast from American Road and Transportation Builders Association, the demand for fly ash will increase by at least 53% for the next two decades [1]. Reducing the cementitious paste volume (CPV) is a way to mitigate these problems.
Previous research indicated that most of Florida Department of Transportation concrete mixtures have an excess of cementitious paste. The high cementitious content will increase shrinkage and permeability, and thus decrease the durability of concrete [2,3]. The main reason for this is because of the mistaken assumption that the strength of the concrete can always be made stronger by increasing the cementitious content. However, Abram's concrete mix principles stated that the strength of the concrete increases as the water to cementitious materials (w/cm) ratio decreases so long as the concrete mix is workable [4]. At a fixed w/cm ratio, the strength of concrete increases with CPV up to a certain amount, after which the strength does not increase but higher absorption and chloride penetration may result [3,5]. The role of CPV in concrete is to fill the voids between the aggregates, provide workability, and bind the aggregates together with cementitious hydration products. The amount of cement used in concrete affects the shrinkage and permeability of the hardened concrete, and not the strength. The concrete with higher shrinkage and permeability usually has early cracking issues which reduce its service life. Based on the American Association of State Highway and Transportation Officials (AASHTO) Designation: PP84-17 Developing for Performance Engineered Concrete Pavement Mixtures, the suitable paste volume should be lower than 25% for pavement concrete to mitigate cracking [6].
Rigid Pavements could be an economical and sustainable type of pavement for developing countries under tropical climate [7]. Slipformed pavement concrete (SPC) has been widely used in Florida, instead of traditional pavement concrete. SPC is a lowslump concrete with high productivity and smooth riding surface in pavement concrete. However, the SPC must be heavily vibrated because the workability is low. The heavy vibration of the SPC would run the risk of segregation and cause low durability of the concrete due to the air void system being compromised [8]. Well graded mixtures not only keep the concrete from segregation but also need less vibration efforts to consolidate and finish the slab [8,9]. Therefore, the blended aggregate (BA) system of the concrete is needed. The BA technique alone cannot affect the properties of hardened concrete. There are serval reasons that make BA important for producing quality concrete. First, BA technique could be used to design concrete containing a high volume of aggregate and a low volume of cementitious paste.
Previous studies concluded that BA technique could improve the workability of fresh concrete [8][9][10][11]. Moreover, better workability can reduce the w/cm ratio of the concrete mix to achieve the target slump. The strength of the concrete increased when w/cm ratio of concrete decreased. BA technique can not only reduce the CPV but also decrease the w/cm ratio of the concrete. On the other hand, Crouch et al. found it necessary to use BA technique in order to meet the goal of the high-performance concrete mix. The highperformance concrete mixes were able to lower the w/cm by 8.3% with no detrimental effects on plastic properties [12]. The BA technique could further reduce the CPV of the concrete with similar properties of fresh concrete and no loss in the properties of hardened concrete.
This research applied the portland limestone cement in the SPC concrete. In U.S., the American Society for Testing and Materials (ASTM) ASTM C150 allows ordinary portland cement (OPC) to contain up to 5% limestone powder [13]. According to ASTM C595, blended hydraulic cement can contain up to 15% limestone powder [14]. In 2017, Florida Department of Transportation (FDOT) allowed portland limestone cement (PLC) containing up to 15% limestone powder (by weight) to be used in approved concrete mixture designs. The use of up to 15% limestone content, by weight, in PLC typically does not affect the properties of concrete substantially because the cement manufacturers tailor the properties of PLC to match those of OPC. Since limestone powder is a filler material, its function is not like that of an supplementary cementitious materials, which is to improve hardened properties by forming additional C-S-H by the aqueous pozzolanic reaction of the silica in the pozzolan with the calcium released from cement hydration. Hardened properties of PLC may be very similar to those of OPC in typical concrete mixes, but their use in mix designs incorporating optimized aggregate gradations and reduced paste contents needs investigation. Moreover, the PLC concrete could reduce the carbon dioxide emission level and reduce the cost of concrete. Because of the lack of studies for evaluating the effect of CPV on the fresh and hardened properties of PLC concrete, it is important to assess the optimal CPV for PLC concrete to produce SPC concrete with sufficient properties. To fill this knowledge gap, a research program was designed for PLC concrete incorporating BA technique to find out the optimal CPVs for SPC concrete. In this program, the concrete mixes were made using PLC, and the CPV was varied from 22.5% to 27.5 % and the w/cm ratio was 0.44.

Cementitious materials
PLC was used in all the concrete mixes in this research. The physical and chemical properties for the cement used are shown in Tables 1 and 2. Tables 1 and 2 show that the PLC also passed the requirements of ASTM C595/C595M [14]. The PLC contained 15% limestone powder blended with 85% portland cement. Table 3 shows the physical and chemical properties of the Class F fly ash used [15].

Aggregates
The coarse aggregates used was a #57 limestone and the intermediate aggregates is a #89 limestone, based on the AASHTO M43 specifications [16]. The fine aggregate used was a silica sand  (with a fineness modulus of 2.45), which is typically used in Florida. The size 57 stone, size 89 stone, and silica sand were the coarse, intermediate, and fine aggregates used, respectively. The absorption percent of #57 stones was 2.8% and its bulk specific gravity was 2.45. The absorption percent of #89 stones was 4.7% and its bulk specific gravity was 2.43. The absorption percent of the sand was 0.4%, its bulk specific gravity was 2.64, and its fineness modulus was 2.45. The admixtures used included an airentraining admixture [17] and two water-reducing admixtures (Type D and Type F) [18]. Dosage rates selected were based on the original mix design per cementitious content and mixing condition.

Blending aggregates methods
The Individual Percent Retained Chart (IPRC) is a plot of the individual percent of the total aggregate content retained on each of the different sieves. Designed aggregate distributions can be classified by the content ranges bracketing the percentages retained for each sieve. The 8-18 distribution, also referred to as the Haystack distribution, is shown on the IPRC method. The intent is to keep the individual retained percentages between 8 and 18 percent for sieves No.30 through the sieve one size below the nominal maximum aggregate size (NMAS), and to keep all sieve sizes below 18 percent retained [19]. The IPRC plot should not have a significant valley between the 3/8 in. and the lowest specific sieve size. The well-graded aggregate from an 8-18 distribution can reduce the total surface area of the aggregate, so it can reduce the water demand of the concrete. Moreover, ACI 302.1 R-96 recommends 8 to 18 percent retained on each sieve for a 1½-in. NMAS gradation, but 8 to 22 percent for ¾-and 1-in. NMAS [20]. Since the development of the 8-18 chart in 1974, some research has shown that it may not always produce a mix with adequate workability. Ley et al. reported that the 8-18 method was insufficient to ensure adequate workability for slip-form-paver mixtures [21]. The Box Test was developed in 2012 to ensure that slip-form mixes had sufficient workability but would be stiff enough to hold straight-formed edges [22]. Taylor studied the aggregate combinations used in over 400 concrete mix designs and developed specifications that are summarized in what is now referred to as the Tarantula Curve ( Fig. 1) [9]. This aggregate distribution varies from the 8-18 distribution in that for most fractions, the upper and lower bounds are broadened, except for those on the #8 and #16 sieves, which are reduced.
The Tarantula Curve (TC) was developed using historical concrete pavement mix designs from the Minnesota Department of Transportation. Contractors refined the mix designs as the corresponding concrete performance was improved through trial and error. The fit to the Tarantula Curve was found to improve in relation to the refinement of the mix designs; increases in performance were mirrored by better fits to the Tarantula Curve. Similar results have been reported for mixture designs in Iowa and North Dakota. Research in Texas also verified that concrete with aggregate gradation optimized using the TC showed excellent response to vibration for concrete with low cementitious content [9]. Most research confirmed that the TC is a reliable tool for BA, but most of the concrete mixtures used in this research were pavement mixtures. Therefore, this research would use the TC to blend the aggregate gradation of the SPC.

Mix design
In this research, there were a total of six concrete mix designs with different CPVs to be evaluated, and the mixes contained the standard concrete (SC) mix and blended aggregates (BA) mix. The purpose of the experimental design was to find out the optimized CPV of SPC concrete and to explore the benefits of BA in SPC concrete. The BA methodology used in this research is Tarantula Curve (TC). Federal Highway Administration (FHWA) recommended Tarantula Curve for SPC concrete mix [8]. Therefore, this research used the TC to optimize aggregate gradation. Fig. 2 showed all the SC and BA mixes plotted in the TC. Table 4 showed the details of the mix designs for each mix. The identification code used for the samples consisted of the group designation, followed by the different class concrete and CPV. For example, the designation SC_275 was used to indicate a mix using a CPV of 27.5% of concrete. The designation BA_225 was used to indicate a mix using a CPV of 22.5% of concrete and BA technique.

. Tests of fresh concrete
The tests on fresh concrete include slump (ASTM C143/C143M), air content (ASTM C231/C231M), density (ASTM C138/C138M), temperature (ASTM C1064/C1064M) and box test [9,[24][25][26]. Slump tests were performed immediately after the concrete was produced to verify the workability of the mixtures. If the workability was not achieved, then some water-reducing admixture was added to make the concrete more workable. As the target slump was achieved, the remaining tests on fresh concrete were performed in accordance with the ASTM standards and AASHTO methods. For the pavement concrete mixtures, the box test was conducted after the slump test. For slip-form paving process, the paving concrete must ensure consolidation while maintaining a vertical edge without slumping. The box test was a useful and consistent tool in evaluating the response of a pavement concrete mixture to vibration. It is important to note that the mixtures investigated are for low-slump slip-formed pavement concrete. Fig. 3 shows the different possible ratings from the box test [9]. The results of the box test indicate that all the concrete mixtures can be used for pavement concrete.

Tests of hardened concrete
The tests of hardened concrete include compressive strength (ASTM C39/C39M), surface resistivity (AASHTO T95) and drying shrinkage (ASTM C157/C157M) [27][28][29]. Concrete cylindrical specimens for compressive strength and surface resistivity tests were 100 × 200 mm (4 × 8 inches). Three specimens for each test at different ages were used. Three specimens for each strength test and each mix, and at different ages were used. Concrete prism specimens used for the shrinkage test were 76 × 76 × 275 mm (3 × 3 × 11.25 inches) The compressive strength of concrete test was performed at 7, 14 and 28 days. The surface resistivity of concrete was performed at 28 days. The test for shrinkage of concrete was done at 7, 28, 90, and 180 days.

Properties of fresh concrete
The fresh concrete properties of the concrete mixes, namely, slump, air content, temperature, density and box test are shown in Fig. 3. Concrete appearance for various ratings in the box test. Table 5. Based on the requirement of FDOT specification for normal concrete, the slump of the concrete should between 25 to 75 mm [23]. Superplasticizer (Type F) was applied in this research to improve the slump. The superplasticizer was used in all the concrete mixes to obtain the desired workability of the concrete. Based on the results of Tables 4 and 5, it can be observed that the dosage decreased, and the slump increased for the concrete using the BA technique. It means that BA could improve the workability of concrete. On the other hand, the slump increased when the concrete contains higher CPV. The workability of the PLC concrete was affected by the CPV, and aggregate gradation.
In this research, the concrete was designed for hot weather conditions. The requirement of the air content is below 6% for SPC. The air content of most concrete mix was similar to each other. The air content decreased when the concrete incorporated the BA technique. The better packing of aggregates would reduce the air content of the fresh concrete. All the density results were in the range of 2240 to 2400 kg/m 3 , which was within the range for normal-weight concrete. It can be concluded that the density results of PLC concrete are close to those of the OPC concrete. The temperatures of the fresh concrete were between 20.5 and 24.5℃. The results of the box test indicate that most of the concrete mixtures can be used for pavement concrete, except the SC with 22.5% CPV.

Compressive strength
The compressive strength results for the SPC concrete mix are taken as the average values from three replicate specimens. The compressive strength results of the concrete mixes at 7, 14, and 28 days are shown in Fig. 4. It can be seen that for all the BA mixes, the compressive strengths are about the same at CPV of 27.5, 25.0 and 22.5%. For the SC mixes, the compressive strength drops substantially when the CPV drops to 22.5%. According to the compressive strength results, the minimum CPV of the SC is 25.0% and the minimum CPV of mixes incorporating BA is 22.5%. When BA technique is used to optimize the aggregate gradation, the CPV can be further reduced to 22.5% without loss in compressive strength. According to the 28-day strength requirement of FDOT, the required strength of pavement concrete is 21 MPa (3000 psi) and the overdesigned compressive strength of it is 29 MPa (4200 psi) [23]. Therefore, the strength development of the PLC concrete is similar to the OPC concrete. The PLC can be used in pavement concrete. For the SC mix and BA mixes, the compressive strengths were about the same when the concrete has sufficient CPV. BA technique could not improve the compressive strength of the concrete.

Surface resistivity
The surface resistivity of the SC and BA mixes are shown in Fig.  5. The surface resistivity of concrete increased when the paste volume decreased. The concrete with BA technique cannot significantly improve the surface resistivity. For the SC mixes, the surface resistivity drops substantially when the CPV drops to 22.5%. It could be concluded that the permeability of the concrete would be affected when the CPV of the concrete is insufficient. Thus, the minimum CPV of the SC mixes is 25.0% However, the BA mix with 22.5% CPV provides the highest surface resistivity. When BA technique is used to optimize the aggregate gradation, the CPV can be further reduced to 22.5% without loss in surface resistivity.

Drying shrinkage
The average drying shrinkage of SPC concrete, measured up to 180 days, are shown in Fig. 6. The specimens were stored in the curing room for 28 days before they were left to dry and drying   shrinkage was measured. For all the mixes, drying shrinkage decreases with reduced paste volume. Concrete with a higher paste volume will have an increased chance for shrinkage cracking. The BA concrete mixes showed significantly lower drying shrinkage when the mixes had reduced paste volume. Better packing density and higher aggregate volumes can mitigate the drying shrinkage of concrete. Thus, the use of the BA technique could be an efficient way to mitigate the shrinkage cracking of pavement concrete.

Conclusions
In this study, SPC concrete mixes with various CPV were tested to assess the effects of CPV on properties of PLC concrete with or without the incorporation of BA technique. The fresh concrete properties which were evaluated included slump, air content, density, temperature, and finishability (box test). The hardened concrete properties evaluated included compressive strength, surface resistivity, and drying shrinkage. According to the results, the following conclusions were made.
1. Concrete using portland limestone cement containing 15% limestone powder can provide similar properties as the concrete using ordinary portland cement and meet the current requirement. 3. Increased CPV and adjusted aggregates gradation of concrete could improve the workability, but the air content would be reduced because of the high packing density. 4. Average surface resistivity results showed that the surface resistivity of the concrete was lower when the concrete had a higher paste volume. It indicates that the concrete with higher paste volume could have higher permeability, which reduces the durability of the concrete. However, the surface resistivity would also be reduced when the concrete had insufficient CPV. 5. Average drying shrinkage results showed that the drying shrinkage of the concrete was lower when the concrete had a lower paste volume. The concrete with lower paste volume could reduce the shrinkage, which keeps the concrete from early cracking. 6. The use of BA technique could not significantly improve the strength and surface resistivity of the concrete at an early age. However, when BA technique was used in the concrete, the drying shrinkage could be reduced.

Acknowledgments
A compilation of this nature could not have been completed without the help and support of others. FDOT is gratefully acknowledged for providing the financial support for this study. FDOT State Materials Office provided the additional testing equipment, materials, and personnel needed for this investigation.

Disclaimer
Any findings, opinions, and conclusions or recommendations expressed in this report are those of the author(s) and do not necessarily reflect the views of the Florida Department of Transportation.
This article is licensed under a Creative Commons Atribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwisein a credit line to the material. If the material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to optain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/