Click for next page ( 30


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 29
29 and to reduce the ingress of water, PDPs, and other poten- Knowing the mechanism(s) of damage will provide neces- tially deleterious contaminants into concrete. sary guidance for preventing or mitigating such distress and for developing the next generation of PDPs and airfield con- ASR has been conventionally controlled by limiting alkali crete pavement. Advances in technologies related to PDPs content in cement and selecting aggregates of good quality. and concrete pavement will make both understanding their Mortar bars prepared with nonreactive aggregates did not interaction and developing an appropriate compatibility test exhibit ASR distress when exposed to the standard NaOH or protocol a continued effort. deicer solutions, even when high alkali cement (0.82% Na2Oeq) was used in the mix (Rangaraju and Olek 2007). Based on these The IPRF closed a request for proposals in October 2006 to results, it appears that new concrete pavements should be pre- address these knowledge gaps. IPRF Project 01-G-002-05-7, pared with nonreactive aggregates, if feasible. Nonreactive Performance of Concrete in the Presence of Airfield Pavement aggregates can likely be identified by a modified concrete prism Deicers and Identification of Induced Distress Mechanisms, test (ASTM C1293, with deicer soaking solution) in which continues the investigation by Dr. Rangaraju that has previ- expansion after one year is limited to 0.04% (Rangaraju and ously implicated acetate/formate-based deicers in the ASR Olek 2007). increase in airfield pavements. Another IPRF project that recently underwent a request for proposals is Project 01-G-002- Furthermore, efforts have been made to mitigate ASR 06-5, Role of Dirty Aggregates in the Performance of Concrete by adding various supplementary cementitious materials or Exposed to Airfield Pavement Deicer, which will examine the chemical admixtures. Research sponsored by the FHWA used possibility of increased alkali content on dirty aggregates and lithium compounds to successfully reduce ASR induced its role in concrete durability in the presence of PDPs. by deicers. However, for existing concrete pavement, it is unlikely that the solution will penetrate significantly, and the IMPACT OF PAVEMENT DEICING PRODUCTS most benefit may only be seen on the surface with reduced ON ASPHALT PAVEMENT debris generation (Folliard et al. 2003). The potential for mit- igation of new concrete with lithium nitrate is more promis- This section will synthesize the information on the impact of ing, although additional research is needed to determine the PDPs on asphalt pavement. First, the potential role of PDPs appropriate dosage (Rangaraju 2007). The toxicity and envi- in the deterioration of asphalt pavement is described. The ronmental effects of lithium nitrate have not been evaluated current understanding of the mechanisms of damage is then (Materials Safety Data Sheet 2006). The effectiveness of min- discussed, followed by the associated standards and test eral admixtures was evaluated in reducing the ASR potential protocols, methods of prevention or mitigation, and finally in the presence of KAc (Rangaraju and Desai 2006). The knowledge gaps. effectiveness of fly ash in mitigating ASR in the presence of KAc was found dependent on the lime content (Rangaraju and In addition to the effects of PDPs on PCC pavement, their effects on asphalt pavement are also of increasing concern. Desai 2006). Fly ash with lower lime content was more effec- Canada's contribution to this research subject began in the tive in reducing the expansions, and greater amounts of fly ash 1990s with a laboratory study comparing cores of asphalt (up to 35% by weight) were needed to replace cement when mix immersed in distilled water and a 2.5% urea solution. more reactive aggregate was used. Ground granulated blast fur- After one freezethaw cycle, there was significantly more nace slag with a replacement level of 50% was needed to miti- loss of indirect tensile strength in the sample immersed in gate the expansions; 40% replacement was found ineffective urea compared with the one in distilled water. Field experi- (Rangaraju 2007). ence did not coincide with these findings, but later studies continued to include urea among the other deicers evaluated Knowledge Gaps (Hassan et al. 2000, 2002b). There is a need for research data from controlled field investi- A laboratory study found that the use of PDPs (NaCl, KAc, gations regarding the effects of alkali-metal-salt-based PDPs on NaF, as well as urea) was damaging to both aggregates and concrete pavement to help differentiate the contribution of such asphalt mixes (Hassan et al. 2002a). The PDPs were tested at a PDPs to concrete deterioration from other possible factors in concentration of 2% of full saturation, previously determined the field environment. One challenge is that the durability of to be a critical concentration capable of the greatest damage. PCC pavement is often significantly affected by the mix design Limestone and quartzite aggregate samples subject to freeze (water-to-cement ratio, type and amount of aggregates, air thaw testing showed more serious damage in all deicers than in content, etc.), construction, curing and maintenance practices, distilled water, as measured by accumulated weight loss. exposure to various climatic conditions (e.g., wetdry and Aggregates immersed in urea exhibited the most weight loss for freezethaw cycles), as well exposure to traffic loading. both types, whereas the least damaging deicer for limestone was NaCl and for quartzite was KAc. Asphalt pavement Furthermore, there is a need to unravel the specific mech- samples taken from the Ottawa MacdonaldCartier Inter- anism by which alkali metal salts cause or promote ASR. national Airport were subjected to freezethaw cycles in closed

OCR for page 29
30 containers. After 25 and 50 cycles, indirect tensile strength, problems emerged as degradation and disintegration of asphalt elastic modulus, and penetration tests were performed. The pavement, softening of asphalt binders, and stripping of asphalt indirect tensile strengths of the samples exposed to deicers were mixes occurring together with loose aggregates on the run- mostly higher than those exposed to distilled water. The lowest ways (Nilsson 2003; Pan et al. 2006; Seminar on the Effects of average elastic modulus was associated with the samples in De-Icing Chemicals on Asphalt 2006). Such problems were urea and visual inspection indicated significant damage by not identified before the airports changed from urea to KAc- urea. Based on weight measurements and density calculations, and KF-based deicers (Pan et al. 2006). In 2001, serious the asphalt mix sample immersed in NaF experienced the most asphalt durability problems were identified at airports in Nordic disintegration after 25 cycles, whereas urea (followed by KAc) countries that used acetate/formate-based PDPs (Pan et al. was the most detrimental deicer after 50 cycles. Exposure to 2006). Heavy binder bleeding and serious stripping problems freezethaw cycles and deicers was found to affect the viscos- were observed occurring together with loss of asphalt stabil- ity of the recovered asphalt binder and the gradation of recov- ity. Soft, sticky, and staining binder came to the surface, often ered aggregates. The freezethaw cycles seemed to result in leaving strong stains on electrical devices and on the airplanes. soft asphalt binder, whereas the deicers caused asphalt harden- The binder of the asphalt base layer was "washed" off, and the ing. However, the authors noted that these findings were incon- aggregates experienced severe loss of strength. In the labora- clusive owing to the difficulties involved in testing and the tory, the tests indicated chemical changes in the binder after inaccuracies in measuring the viscosity of the recovered exposure to the deicer, in the form of emulsification, distilla- asphalt. Overall, this laboratory investigation found urea to be tion, and an increased amount of polycyclic aromatic hydro- the most detrimental deicer, whereas the other deicers "induced carbons (PAHs). A field investigation was conducted there- relatively small damage, comparable to that caused by distilled after confirming the deleterious effects of acetate-based deicer water" (Hassan et al. 2000). However, it was noted that chem- on asphalt pavement. The bitumen and the mastic squeezed to ical reactions would occur slowly at the temperatures involved the surface of the core, and the concentration of the deicer had in this study and that damage in the field could occur as a result a clear influence on its solubility. Some bitumen was dissolved of reactions between PDP residues and asphalt during hot sum- into the pore liquid, and pure stone particles were found inside mer temperatures (Hassan et al. 2000). the core. The limestone filler was found fully dissolved by the PDP liquid and the rest of the mastic became brittle and grey- A follow-up study was conducted at higher temperatures colored. A large increase in the porosity of asphalt was also on asphalt pavement samples taken from the Dorval Interna- noticed. tional Airport (Montreal, Canada) to clarify the role played by the PDPs (NaCl, KAc, NaF, as well as urea) in asphalt To address the concerns over acetate-based deicers affect- deterioration, and to determine whether the damage was ing asphalt pavement, a joint research program--the JP attributable to the physical freezethaw action. Only 15 freeze Finnish De-icing Project--was established to conduct exten- thaw cycles were performed before subjecting some samples sive laboratory and field investigations on this subject. The to 40 wetdry cycles at 40C. This research confirmed the goal of JP was to provide answers to three fundamental previous finding that softening occurs during freezethaw concerns: how the damages are generated, how to determine and exposure to deicers causes hardening. After the freeze the compatibility between asphalt and de-icing materials, and thaw and wetdry cycles, the samples in NaAc showed the how to prevent damages by mix design (Pan et al. 2006). The lowest strength, followed by those in NaF. Interestingly, all research showed that formate/acetate-based deicers signifi- samples showed increased strength after the warm wetdry cantly damaged asphalt pavements. The damaging mecha- cycles and all except NaF and NaAc showed increased elas- nism seemed to be a combination of chemical reactions, ticity after the warm wetdry cycles. However, the dry sam- emulsification, and distillation, as well as generation of addi- ples not exposed to freezethaw or wetdry cycles had the tional stress inside the asphalt mix. Asphalt binders soaked greatest elasticity and nearly highest strength. Overall, the in the deicer solution were found to have lower softening Canadian studies did not indicate significant damaging points and tended to dissolve at temperatures as low as 20C. effects of KAc and NaF on asphalt pavement (Farha et al. Asphalt mixes soaked in the deicer solution were found to 2002; Hassan et al. 2002a). It should be cautioned, however, have lower surface tensile strength and lower adhesion (Nils- that these results were based on laboratory experiments on son 2003; Seminar on the Effects of De-Icing Chemicals on only two samples of asphalt pavement and the mix design for Asphalt 2006). It seemed to be clear that deicer (formate or each pavement was undeterminable from the reports. Addi- acetate), water or moisture, and heat were necessary for the tionally, the deicer concentration was low (2%), which may damage to occur. In the field, such damages mainly occurred not be conducive to simulate years of field exposure by during the repaving process or on hot summer days with means of accelerated laboratory testing. residual deicers from the winter season, as dynamic loading and unloading reduced the time it took for damages to occur. Concurrent to the use of acetate/formate-based deicers in the 1990s, asphalt pavement in Europe saw an increase in Recent laboratory studies by a research group at Montana pavement durability problems. At some Nordic airports, these State University were able to reproduce acetate-induced emul-

OCR for page 29
31 sification of asphalt similar to the field observations at Nordic water. Stripping occurred for both crushed gravel and lime- airports (Pan et al. 2008). Aqueous solution tests of asphalt stone aggregate particles included in the asphalt mix exposed binder in water and four NaAc solutions of different concen- to NaAc, suggesting that aggregate properties play at most a trations (5%40%) showed a bilinear trend of weight loss secondary role in asphalt emulsification (Pan et al. 2008). As increasing with the NaAc concentration. Both visual inspec- indicated in Figure 16a, significant amounts of aggregates tion and optical microscopy (as shown in Figure 15) indicated were stripped after exposure to the NaAc solutions and the that a significant amount of asphalt emulsification occurred in aggregate stripping followed a bilinear trend with weight loss NaAc, but not in water or aqueous solutions of NaCl or NaOH increasing with the NaAc concentration. with a pH of 9 (equivalent to the measured pH of 40% NaAc solution). For the two tested asphalt binders, PG 58-22 exhib- Phase I of Airfield Asphalt Pavement Technology Pro- ited slightly higher emulsification than PG 67-22. In the gram Project 05-03: Effect of Deicing Chemicals on HMA calcium magnesium acetate aqueous solution, asphalt emulsi- Airfield Pavements includes a literature review, interviews fication occurred similarly to that in NaAc. These results con- with 36 airports that use deicers and have asphalt pavement, firmed that asphalt emulsification should be attributed to the as well as laboratory testing. Seven airports indicated that acetate anion, CH3COOH and excluded the possibility that pavement deterioration had occurred, but the cause was high alkalinity was responsible for the asphalt emulsifica- unknown except in one case that was most likely attributable tion in NaAc. Asphalt emulsification also occurred in a NaF to the type and source of asphalt binder and aggregate. Pre- aqueous solution and more detailed laboratory testing is liminary laboratory testing was conducted of asphalt pave- being conducted. ment samples composed of either a chert gravel or diabase with two binders (PG 64-22 and PG 58-28) exposed to KAc The effects of NaAc on asphalt mixes were examined by and NaF. The presence of PAHs was inconclusive after conducting a modified ASTM D 3625-96 Boiling Water vacuum-induced saturated samples were stored for 4 days at Test, which was originally designed to test the susceptibility 60C. However, significant generation of carboxylate salts of asphalt mixes to moisture damage, by accelerating the had developed after the asphalt mixes were exposed to the effect of water on bituminous-coated aggregate with boiling deicers, although this may not be related directly to deicer- induced damage. Indirect tensile strength tests showed PG 64-22 to be "somewhat more resistant" (Advanced Asphalt Technologies 2007) and that chert gravel had significantly less strength when exposed to deicers compared with water. A long-term durability test developed by Advanced Asphalt Technologies also showed chert to be very susceptible to moisture damage, particularly when exposed to KAc or NaF. Soundness tests of both types of aggregate in magnesium sul- fate, KAc, and NaF were acceptable and also showed that direct attack on the aggregate by the deicers was not occur- ring (Advanced Asphalt Technologies 2007). (a) (b) 40 Percent Stripped Aggregate (%) 30 20 10 (c) (d) 0 FIGURE 15 Digital photos (left) and optical microscopic images 0 10 20 30 40 (right) showing the suspension solution of asphalt subsequent to Sodium Acetate Concentration (wt. %) the 60C aqueous solution test. (a) and (b) Twenty-four hour reservation of 60C aqueous solution test with the 20% acetate FIGURE 16 Percent stripped aggregates for an concentration; (c) and (d) Twenty-four hour reservation of the asphalt mix exposed to different NaAc solutions after 60C aqueous solution test with the 40% acetate solution the Modified Boiling Water Test [adapted from Pan [adapted from Pan et al. (2007)]. et al. (2007)].

OCR for page 29
32 Nature of the Effect of Modern Pavement Deicing includes more significant laboratory testing and field investi- Products on Asphalt Pavement Deterioration gations (Advanced Asphalt Technologies 2007). However, the research by Pan et al. (2008) shows: (1) asphalt emulsifi- Canadian research demonstrating the damaging effects of cation occurring to asphalt mixes with both reactive and non- urea on two types of asphalt pavement did not propose any reactive aggregates, and (2) asphalt emulsification not occur- mechanism or scientific explanation for the observations ring in NaOH solutions of the same pH values as the NaAc (Hassan et al. 2000, 2002b). solution. Thus, the research indicates that asphalt emulsifica- tion may be a more critical mechanism of asphalt mix deteri- The JPĭFinnish De-icing Project studied the ingredi- oration than ASR unless very reactive aggregates are used in ent materials in asphalt pavement individually and their roles the asphalt mix. played in the damaging mechanism were ranked accordingly (Alatypp 2005). The detailed test results of each ingredient Pan et al. (2008) proposed a detailed and specific mecha- material are included as follows, as summarized by Pan et al. nism of acetate-induced asphalt emulsification based on con- (2006). tact between acetate anions (CH3COOH) and asphalt, which can be greatly increased at high summer and/or repaving tem- Effects of Formate/Acetate-Based Deicers on Aggregates: The main reason for pavement damages was not due to poor peratures owing to the tendency of asphalt to swell. For NaAc, quality of aggregates. Mineral aggregate might be a reason aqueous solution tests of asphalt binder were performed at sev- secondary to asphalt binders in pavement damage. The decom- eral concentrations and temperatures and the resulting sus- position level of acidic aggregates was higher than for caus- pended substance was examined using the Fourier Transform tic aggregates, but was still acceptable. However attention Infrared Spectroscopy. No significant amounts of new chemi- should be paid to the weathering resistance of aggregates cals were identified, and intermolecular binding between the used in airfields to extend the life span of asphalt pavements. Physical Effects of Formate/Acetate-Based Deicers on acetate anion CH3COOH and the alkane component of Bitumen/Asphalt: (1) High density of deicer solution such as asphalt was inferred (Pan et al. 2008). Van der Waals forces 1.34 kg/dm3 for the 50 wt.% solution enabled the deicer solu- anchor the lipophilic organic chain (CH3) of the acetate anion tion to penetrate into bitumen by gravity. (2) Very low sur- to the molecular chain of asphalt (CH3CH2). At the same face tension between deicer chemicals and asphalt facilitated time, the hydrophilic polar end of the acetate anion (COO) stripping and emulsification of asphalt mixes. (3) Formate/ forms hydrogen bonds with water molecules and pulls on the acetate-based deicers had pH values usually between 9 and 11; and the higher the pH, the more aggressive the deicer would asphalt, overcoming the intermolecular forces within the be. (4) Formate/acetate-based deicers were very hygroscopic, asphalt. Asphalt emulsion is maintained by Brownian motion which kept the road surface constantly wet and retained water and repulsive forces on the floccules. The emulsification of inside asphalt to overfill the air voids. asphalt reduces the asphalt-aggregate bond and can lead to Chemical Effects of Formate/Acetate-Based Deicers on adhesion failure in the pavement. There is also a potential that Bitumen/Asphalt: When exposed to deicers, composition changes of bitumen/asphalt occurred in the hydrocarbon clas- the aggregate preferentially bonds with the acetate anion, sification C10-C40. When exposed to deicers, large organic which has a higher polarity than the asphalt molecules. molecules such as the PAHs grew in bitumen. Deicers in asphalt were found in both the liquid and gas phases. PAHs in the asphalt samples could migrate and become dissolved in Standards and Test Protocols the deicer. Failure Process of Asphalt Pavements: Deicers migrate into There are two existing Swedish test methods related to asphalt the asphalt after application onto pavements and saturate and deicers: the LFV Method 1-98, Bituminous Binders, Stor- asphalt mixes during the winter. The deicer solution intrudes age in De-icing Fluid, and the LFV Method 2-98, Effect of into asphalt due to gravity and for other unknown reasons, espe- cially when asphalt temperature rises significantly (a result of a De-icing Fluid on the Surface Tensile Strength of Asphalt hot asphalt layer laid or summer weather). Due to the low sur- Concrete for Airfields--Adhesion Test. In 2006, results of face tension between deicers and bitumen, the deicers are round-robin testing of LFV Method 2-98 by seven laborato- absorbed in the bitumen that in turn starts to emulsify. It is pos- ries were presented; however, the information was only avail- sible that the chemical composition of the bitumen changes able in the form of Microsoft PowerPoint Slides and thus dif- during emulsification. Due to emulsification the bitumen comes loose and the aggregate particles get cleaned, followed by ficult to follow. The confusion lies in the measurement units bleeding and stripping. reported. The LFV 2-98 method indicates the surface strength at failure should be reported to the nearest 0.1 N/mm2, but the Ongoing research by Advanced Asphalt Technologies standard deviation of the round-robin test for repeatability currently suggests that the damaging mechanism is mainly a was reported to be 130 N and 220 N for reproducibility (Nils- disruption of the asphalt-aggregate bond as a result of ASR. son 2006). One Norwegian airport reported the use of LFV Expansive pressures typical of ASR-damaged concrete are Method 2-98, according to the ACRP survey results. The not perceived to be the problem, but rather the bond disrup- Aqueous Solution Test as developed by Pan et al. (2008) tion and increased susceptibility to moisture damage. Well- showed high efficiency in examining the emulsifiability of drained pavements may provide some protection because asphalt, and can be potentially established as a standard accel- deicers are not applied during warm weather. Advanced erating test. The other test--the Modified Boiling Water Asphalt Technologies is currently working on Phase II that Test--also proposed by Pan et al. (2008) can be used as a rou-