The Use of Phosphoric Acid to Stiffen Hot Mix Asphalt Binders

CHAPTER 5. LIME AND LIMESTONE AGGREGATES

LIME

When lime (calcium hydroxide) is used as an antistrip additive in asphalt binders, it is common to use 1 percent of lime based on the weight of aggregate. Concerns have been expressed that any phosphoric acid in the binder would react with the lime so that it would no longer be effective as an antistrip additive.

The chemistry of this reaction is given in the following equation:

Ca(OH)₂ + H₃PO₄ = CaHPO₄.2H2O

Stoichimetrically, 74 lb of lime would react with 98 lb of phosphoric acid yielding 172 lb of calcium phosphate. If the mix contains 1 percent of lime based on the aggregate and 1 percent of phosphoric acid based on the binder content, then there is a large excess of lime, about 25 times.

To determine whether or not the phosphoric acid in the binder would react with lime (calcium hydroxide), six different phosphoric acid Modified PG 64 binders were used; two from Citgo®, two from HollyFrontier®, one from Lion Oil, and one from BP Whiting. The binders were modified with 1-percent phosphoric acid and then mixed with 20-percent lime by weight of the asphalt at 165 °C using a propeller stirrer to blend the lime into the hot asphalt. To account for any possible reactions of lime with the binders themselves, samples of the unmodified binders were also treated with 20-percent lime.

The lime was removed by solvent extraction, the lime was filtered out, and the binder recovered by evaporation of the solvent. All the binders, both modified and unmodified, were put through the same extraction and recovery process.

The results given in figure 37 show the following:

• The PG of phosphoric acid-modified binders (black bars) was different for each binder and considerably higher than that of the unmodified binders.
• The PG of the recovered unmodified binders (purple bars) was slightly lower than the original.
• The PG of the recovered binders treated with lime alone (green bars) was lower than the original, particularly so with the binders from HollyFrontier®.
• The PG of the phosphoric acid treated binders after lime treatment and recovery to remove the lime (red bars) was lower than the phosphoric acid-modified binders and was close to the PG of the original binders.

Figure 37. Chart. Performance grades of binders after treatment with lime.

This study clearly shows that lime reacts with the phosphoric acid in a phosphoric acid-modified binder. To what extend this would occur in a mix is unknown.The stiffening effect of the acid is clearly lost after treatment with lime if the lime is removed during the binder recovery process. It is a complex issue because as figure 38 shows, the addition of an alkali like lime sometimes causes an increase in binder stiffness (PG). One might argue that if the mix is already treated with lime, then the addition of phosphoric acid to the binder might be superfluous.

Figure 38. Chart. Effect of lime content on asphalt stiffness.

As part of another research effort, the flow numbers of gyratory specimens containing lime, phosphoric acid, and SBS polymer were measured using AASHTO test method TP-79 "Standard Method of Test for Determining the Dynamic Modulus and Flow Number for Asphalt Mixtures using the Asphalt Mixture Performance Tester (AMPT)." The aggregate in these specimens was diabase, and the asphalts were grade PG 64 from Lion Oil (binder B) and HollyFrontier® (binder C). The results are given in figure 39 and figure 40.

Figure 39. Chart. Flow number for the mixture modified with binder from Lion Oil.

Figure 40. Chart. Flow number for the mixture modified with binder from HollyFrontier® asphalt.

Lime addition to the HollyFrontier® asphalt samples showed no increase in flow number for the control, and the Lion Oil showed a modest increase. The addition of 2-percent phosphoric acid (a very high level of modification not generally used in practice) to the lime-treated samples showed a slight increase for the HollyFrontier® asphalt sample and a slight decrease for the Lion Oil binder. Modification with 2-percent SBS polymer showed a significant increase in flow number as would be expected. The addition of lime to the SBS modified samples appears to have a synergisitc effect, with a substantial increase in flow number especially in the case of the HollyFrontier® asphalt binder. This synergy appeared to be almost totally absent when the binder was modified with 1.5-percent SBS and 0.5-percent phosphoric acid.

LIMESTONE

This experiment was designed to determine whether phosphoric acid in the binder would react with basic aggregates, e.g., limestone (primarily calcium carbonate). Several HMA samples were made with different limestone aggregates. The asphalt was recovered, tested for the high temperature PG, and the phosphorus content determined. The asphalt type and limestone aggregate sources were varied to see the effect.

Three asphalts from different sources were used: Citgo® (B6362), BP Whiting (B6364), and Lion Oil (B6367). Three aggregates were used: Maryland limestone (designated as MD) and two limestone samples from New York State Department of Transportation (NYSDOT) (designated NY3 and NY4).

The initial high temperature performance grade was 64 °C for all three asphalts. The asphalts were modified with 1-percent phosphoric acid by weight of the binder and DSR measurements of the original and RTFOT samples were made. The resulting high temperature PG of the phosphoric acid modified asphalts was 70 °C.

Loose mix was made with both the control and the phosphoric acid-modified asphalt for each aggregate. This resulted in 18 loose mix samples. All of the loose mix samples were short-term oven aged at 135 °C for 4 h. The job mix formula for NY3 and NY4 aggregate was provided by NYSDOT and is summarized in table 7.

Table 7 . Limestone mix designs.

The loose mix was extracted using AASHTO: T164 "Quantitative Extraction of Asphalt Binder from Hot-Mix Asphalt (HMA)" and recovered using AASHTO: T170 "Recovery of Asphalt from Solution by Abson Method." The extraction solvent was trichloroethylene (TCE). Every attempt was made to minimize the amount of aging that took place during the procedures by following the times and temperatures specified in the methods. The recovered binders were analyzed by Fourier transform infrared spectroscopy to ensure that no residual TCE solvent was present. DSR measurements were made, and the high temperature and continuous high temperature PG (using the RTFOT criteria of 2.2 kPa) was calculated. All of the recovered binder samples were analyzed for the presence of phosphoric acid with XRF as described earlier in this report.

The results summarized in table 8, show that all the recovered phosphoric acid-modified binders contained phosphorus.

Table 8 . Phosphorus in recovered asphalt binders.

While there was a slight increase in the continuous grade temperature in some of the modified binders recovered from the limestone mixes, this was less than 6 °C, indicating no change in the PG. The binder grade did not decrease in any of the samples in contrast to the case with lime.

Figure 41 shows a plot of the data grouped by aggregate type; the DSR results are in table 9.

Figure 41. Chart. PGs of phosphoric acid-modified binders recovered from limestone mixes.

Table 9 . High temperature PGs of phosphoric acid-modified binders recovered from limestone mixes.

CONCLUSIONS

• Lime (calcium hydroxide) in HMA can react with the phosphoric acid in phosphoric acid-modified binders.
• Limestone aggregates in HMA do not readily react with the phosphoric acid in phosphoric acid-modified binders.