Air Void Spacing Factor Research Articles

Long-term performance of magnesium phosphate cement concrete prepared in the winter

Previous studies have established the strength development of magnesium ammonium phosphate cement (MAPC) concrete and magnesium potassium phosphate cement (MKPC) concrete at sub-zero temperatures. However, the long-term performance of these magnesium phosphate cement (MPC) concrete, prepared at low temperatures and exposed to complex environmental effects of cold regions, remains underexplored. This study monitored the mechanical properties and durability of MAPC concrete at different ages for 520 d, which fully demonstrated the feasibility of MAPC concrete for winter construction in cold regions. At the age of 520 d, MAPC concrete had a compressive strength of over 65 MPa and excellent resistance to chloride ion penetration with a measured charge of 129.1 C, and could endure more than 250 rapid freeze-thaw cycles. In contrast, MKPC concrete exhibited significant disadvantages in terms of freeze-thaw resistance during natural freeze-thaw cycles. No expansive substances or damage were found in MKPC concrete that was prepared at low temperatures and subsequently experienced temperature increases along with accelerated hydration. The relatively poor freeze-thaw resistance of MKPC concrete could be attributed to a combination of factors including larger capillary size and volume, reduced air content, greater air-void spacing factor, wider and weaker interface zone between cement mortar and coarse aggregate.

Study on the influence of fractal dimension and size effect of coarse aggregate on the frost resistance of hydraulic concrete

In cold regions, hydraulic concrete structures are exposed to harsh environmental conditions that greatly impact their durability and safety. The role of aggregates in influencing concrete performance is paramount. This study aims to investigate how the fractal distribution and size of aggregates affect the frost resistance of hydraulic concrete, with air void characteristics, mass loss rate, Elastic Modulus of dynamic, and strength serving as primary references. Two experiment series, fractal distribution influence and aggregate size influence of concrete frost resistance, were conducted to analyze the frost resistance of hydraulic concrete with different aggregates. Based on the fractal theory, four grading fractal distributions were utilized to fractalize aggregate gradation, aiming to investigate the influence of aggregate fractal dimension on the frost resistance of hydraulic concrete. Simultaneously, concrete specimens of fully graded aggregate and wet-screened aggregate were also constructed to simultaneous freeze-thaw cycle tests, the impact of aggregate size on concrete's frost resistance performance was also studied by analyzing the air void structure in the frost-resistant concrete specimens. Results indicate that the specimens exhibit varying frost resistance performance under different aggregate gradation fractals, ranking from strongest to weakest as follows: D=2.5, D=2.3, D=2.1, D=2.7, which is attributed to the differences of air void spacing factors. Increasing aggregate size leads to greater internal structural heterogeneity in concrete, resulting in an increase in defects and weak points. This causes microcracks to develop more rapidly in the transition zone of fully graded concrete during freeze-thaw cycles, leading to more severe damage. The findings of this study can serve as a fundamental scientific basis for the designing and constructing of hydraulic concrete structures in cold areas.

Synergistic effects of different ionic surfactants and temperature on rheological behavior and air-void structure of cement mortar

The influence of air entrainment on rheological behavior of fresh cementitious materials has attracted a lot of attention in the past years, but the results are still in controversy. In this research, cement mortars were prepared with equivalent air contents using different air-entraining admixtures (AEAs) to measure their fresh-state performance and hardened air-void system at variable temperature conditions, which aimed to clarify and unveil the coupled effects of different surfactants and temperature for the first time. The adsorption amount of AEA, zeta potential of cement, along with hydration kinetics of mixtures were evaluated to characterize the interaction of air bubbles with cement grains. It was found that for mixture without superplasticizer (SP), the impact of air entrainment on rheological parameters was dependent on the interaction intensity of AEA molecules with cement. For SP-plasticized mixture, air entrainment tended to cause a shear-thinning behavior with decreased plastic viscosity, while its influence on yield stress was in accordance with the situation without SP. The hardened samples with anionic surfactant contained more small voids and lacked big voids, whereas the samples with cationic surfactant showed opposite outcomes. With temperature decreasing, both the average chord length and the spacing factor of air voids increased owing to the reduced yield stress of mixture. The results will provide in-depth illustrations for understanding how air bubbles change the rheological behavior of mixture, and more importantly it will explain the seemingly inconsistent findings reported by previous literatures.

Preliminary exploration of a novel high-performance air entraining agent based on sodium dodecyl sulfate added self-made fluorocarbon surfactant

Sodium dodecyl sulfate (SDS), an air entraining agent (AEA), is used to improve the frost resistance of concrete, but the effectiveness of SDS is urgent to be improved because it is easily weakened by many factors. In this work, a novel high-performance AEA (HPAEA) was prepared by compounding sodium dodecyl sulfate (SDS) with the self-made fluorocarbon surfactant (S-FC). The performance of HPAEA in cement paste solution and in mortar were evaluated and the improvement mechanism was explored. The results of foamability and foam stability in cement paste solution, air content and air-void parameters in cement mortar showed that S-FC could greatly promote the foam properties of SDS and optimize the air-void structure in entrained mortar. When the concentration of S-FC reached 0.2 wt%, foamability, foam stability, fresh air content of HPAEA was 2.9 times, 16.5 times, and 1.6 times higher than that of SDS, respectively. The air-void spacing factor in hardened mortar was reduced by more than 50%. The improvement mechanism of the effectiveness of HPAEA was revealed by the lower surface tension and better comprehensive property of the bubble liquid film, because there was synergistic effect between SDS and S-FC. It was concluded that the novel HPAEA obtained by compounding SDS with S-FC had a potential to significantly improve the frost resistance of concrete.

Freeze-thaw resistance of concrete containing azodicarbonamide expansive agent

Azodicarbonamide expansive agent (ADC) and air-entraining agent (AEA) have similarities in introducing pores into concrete and can both be applied in constructions served in severe cold environments. This paper investigates the effect of ADC on the freeze-thaw resistance of concrete and compares the differences between ADC and AEA in the anti-freezing mechanism. The results show that ADC improves the freeze-thaw resistance of concrete, while its effect is weaker than that of AEA. The non-hydrophobic bubble shell formed by ADC cannot reduce the freezing temperature and delay freezing time as well as the shell consisted of hydrophobic calcium salt formed by AEA. A rosette-shaped pore structure and non-dense shell of air voids formed by ADC have both positive and negative effects on freeze-thaw resistance: they release hydraulic pressure to improve freezing resistance, while they reduce the suction force resulting in weakened freezing resistance. The air-void spacing factor instead of the air content at fresh stage is more suitable for assessing the freeze-thaw resistance of concrete with ADC.

Correlating field and laboratory investigations for preventing ASR in concrete – The LNEC cube study (Part I – Project plan and laboratory results)

As part of the Norwegian R&D project “ASR - Reliable concept for performance testing (KPN-ASR)” and of the RILEM TC 258-AAA activities, a study was launched in 2015 with the aim of correlating the results of field and laboratory investigations for selecting appropriate preventive measures against deleterious alkali-silica reaction (ASR) in concrete. Three coarse aggregates, i.e. New Mexico polymictic gravel, Ottersbo cataclasite, and a non-reactive limestone aggregate, were used to manufacture cubes (300×300×300 mm3) from air-entrained, control (high-alkali cement; normal alkali cement) and ASTM Class F fly ash (20 and 30 % replacement levels of the high-alkali cement) concrete mixtures. The cubes were then shipped to be placed on the study participants’ outdoor exposure sites, where they are being monitored for long-term expansion and cracking development (Canada; France; Germany; Iceland; Norway; Portugal; USA). Companion laboratory cubic specimens (for compressive strength and air-void spacing factor determination on hardened concrete) and test prisms were cast from the above mixtures. The prisms were subjected to RILEM AAR–3.1 and RILEM AAR-4.1 testing at LNEC's laboratory; their expansion was monitored over 205 and 28 weeks, respectively. Selected control and fly ash mixtures, without air-entrainment, were repeated at SINTEF following the 38 °C Norwegian concrete prism test (NCPT, prism size 100×100×450 mm3) and RILEM AAR-4.1. This paper, the first of the LNEC cube study, gives an overview of the project, lists the project's main objectives, provides a summary of the activities already developed in the project, and presents and discusses the results obtained thus far in the laboratory investigations. As expected, excessive expansions were obtained for the control high-alkali concretes, whereas addition of fly ash reduced the expansion to a low level. However, the expansion attained for the fly ash mixtures depended on the CPT procedure used.

Air void clustering in concrete and its effect on concrete strength

ABSTRACT Air voids in concrete tend to cluster around coarse aggregate particles and potentially weaken the concrete. This study explored key factors influencing air void clustering and evaluated the effect of air void clustering on concrete strength. The six considered variables were cement type (low alkali cement and limestone cement), fly ash (with the low and high loss on ignition), coarse aggregate type (limestone and gravel), admixture combination type (stable and unstable air-entraining agents and compatible and incompatible water reducers), mixing temperature (70 and 90°F), and workability (with and without retempering). A total of 64 mixtures were prepared and tested. The fresh concrete’s slump, unit weight and air content were studied along with its compressive strength after 7 and 28 days. Air-void spacing factor, clustering and the voids’ specific surface were evaluated. The results showed that retempering was the most significant factor influencing air void clustering. It was found that retempering can increase air content, create finer air voids and aggravate air void clustering. To reduce the risk of air void clustering, retempering and the mixing water temperature of 90°F should be avoided. When the clustering rating exceeded 0.6, some strength loss was observed.

Influencing factors of parameters of air-void structure in fresh cement mortars

There is a limitation to apply air content in fresh concrete as the only index to pre-control frost resistance of cast-in-place concrete. By using an air void analyzer (AVA), the research investigated the influence laws of factors including water-cement ratio, addition amount of fly ash, vibration time, and time-dependent behaviors on a series of characteristics of fresh cement mortars. These characteristics include content, spacing factor, specific surface area (SSA), and size of air voids. The results demonstrated that when the air content was in the range of 7≈8%, the water-cement ratio insignificantly influenced the air-void structure. As fly ash can adsorb the air void, so its addition substantially reduced the content of large air voids and make the size distribution of voids more uniform. Compared with poker vibration, high-frequency vibration exhibited better effects in more uniformly distributing the air voids in the mixture. As large air voids escaped and split, the air content decreased, which was in favor of reducing the air-void size and the air-void spacing factor; Air-void structure in the mixture underwent complex change over time. On the whole, the air-void structure showed the following change trend: small air voids combined to large ones, air voids escaped, the number of air voids and the air-void SSA declined, and the air-void spacing factor enlarged. Therefore, while preparing concrete in practical engineering, it is particularly necessary to consider the adsorption effect of admixture, the time-delay effect of air-entraining agents, and the time-dependent behaviors of air voids, so as to control air-void parameters of hardened concrete.

Field and laboratory validation of the sequential air method

This work compares the sequential air method (SAM) and the results from the hardened air-void analysis (ASTM C 457) for 488 different concrete mixtures from the lab and field. The results show that there is a wide variation of air contents that correlate with a Spacing Factor of 200 μm. These results show the inadequacy of using air content to decide the quality air void system of the concrete. In fact, 25% of the field data was shown to have a Spacing Factor higher than the recommended values. The results from the SAM, Spacing Factor, and volume of fine air voids (chords less than 300 μm) exhibited good agreement for both the laboratory and field data. Since the SAM can be used to test concrete before it sets and it can give important insight into the bubble size and spacing, this makes it a valuable tool to design and evaluate the air void system of fresh concrete and provide insight into the air void system in the hardened concrete.

Effects of Parameters of Air-Avid Structure on the Salt-Frost Durability of Hardened Concrete

Through laboratory testing, this research studied the connection between air-void structures of hardened concrete and fresh concrete and discussed the effects of the air-void structure on the salt-frost durability of the concrete. The results demonstrate that, in comparison with fresh concrete, the air-void spacing factor shows a close correlation with hardened concrete air-content and decreases in the form of a power function as the air-content increases. When the fresh concrete air-content is more than 6% and the hardened concrete air-void spacing factor is less than 0.18 mm, the influence of parameters of air-void structure on the salt-frost resistance of the concrete reduces. The air-void spacing factor more significantly affects the salt-frost resistance of the concrete compared with air content and the correlation reaches 0.93. Therefore, air-content and air-void spacing factor are recommended for dual control.

The Influence of Atmospheric Pressure on Air Content and Pore Structure of Air-entrained Concrete

To study the effect of atmospheric pressure on the properties of fresh and hardened air-entrained concrete, three kinds of air entraining agents were used for preparing air-entrained concrete in the plateaus (Lhasa, 61 kPa) and the plains (Beijing, 101 kPa). Air content, slump, compressive strength and pore structure of the three air-entrained concretes were tested in these two places. It is found that the air content of concrete under low atmospheric pressure (LAP) is 4%-36% lower than that of concrete under normal atmospheric pressure (NAP), which explaines the decrease of slump for air-entrained concrete under LAP. Pore number of hardened concrete under LAP is reduced by 48%-69%. While, the proportion of big pores (pore diameter >1 200 μm) and air void spacing factor are increased by 1.5%-7.3% and 51%-92%, respectively. The deterioration of pore structure results in a 3%-9% reduction in the compressive strength of concrete. From the results we have obtained, it can be concluded that the increase of critical nucleation energy of air bubbles and the decrease of volumetric compressibility coefficient of air in the concrete are responsible for the variation of air content and pore structure of concrete under LAP.

Probabilistic evaluation of concrete freeze-thaw design guidance

A novel limit-state function using Powers’ models is developed to assess current freeze-thaw exposure categories and design criteria for concrete placements established by American, Canadian, and European standards organizations. Based upon performance assessments by standardized accelerated testing, the current specifications are shown to provide sufficient levels of reliability pending an appropriate mean air-void spacing factor. Sensitivity assessments of the model demonstrate that the spacing factor, saturation state, permeability, and freezing rate significantly influence the response of the air-entrained concrete. The model is validated with a large dataset derived from standard freeze-thaw tests, and an equation is developed to probabilistically design concrete for freeze-thaw resistance.

Determination of air-void system and modified frost resistance number for freeze-thaw resistance evaluation of ternary blended concrete made of ordinary Portland cement/silica fume/class F fly ash

Use of fly ash (FA) in concrete subjected to freeze-thaw (F-T) cycles in a cold region has been limited because of its slow hydration and low scaling resistance. One of the solutions to increase the utilization of FA in this area may adopt ternary blended concrete to overcome the shortcoming of FA. In this paper, the F-T resistance of plain concrete, binary concrete, and ternary blended concrete mixtures made of ordinary portland cement (OPC), 5%-fixed silica fume (SF), and various Class F FA contents up to 45% as replacement of cement by weight were evaluated. Durability factor (DF), both frost resistance number (FRN) and modified FRN, and critical air void spacing factor (CAVSF) were determined to characterize the F-T resistance of all concrete mixtures. Except for non-air-entrained concrete containing OPC/5SF/15FA, air-entrained binary concretes containing OPC/0SF/35FA and OPC/0SF/45FA, and air-entrained ternary blended concrete containing OPC/5SF/45FA, all concretes had a good F-T resistance which exceeded DF of 60%. The critical air-void spacing factor for binary and ternary blended concretes for 60% DF was determined to be 198.2 μm and 305.2 μm, respectively. Moreover, ternary blended concretes resulted in higher FRN than plain and binary concrete mixtures. Finally, air voids under 300 μm size play a critical role in reducing the F-T damage of ternary concrete mixtures. Based on these results, it can be concluded that construction of the proper air-void system is required to ensure the F-T resistance of ternary blended concrete. Moreover, the modified FRN can be successfully used to evaluate the F-T resistance of ternary blended concrete.

Freezing and thawing resistance of cellular concrete containing binary and ternary cementitious mixtures

Cellular concrete (CC) is a foamed low-density and low-strength material made with cement and/or lime, silica-rich material (sand, slag, or fly ash), water, fine aggregate and a foaming agent. The CC containing millions of evenly distributed, uniformly sized macroscopic air-voids of approximately 0.1–1 mm in size is considered to have good freeze-thaw (F-T) resistance. In the present study, the CC of binary and ternary cementitious mixtures with varying proportions of portland cement, fly ash, and lime were explored in a comprehensive laboratory test program related to porosity, water absorption, dry density, compressive strength, and resistance to F-T including durability factor and loss of mass. For selected mixtures, air-void spacing factor and air-void distribution had been determined. Test results showed that compressive strength of CC was primarily as a function of the porosity and density regardless of type of cementitious material with respect to the combination of binary and ternary cementitious mixture. It was also found that higher porosity did not necessarily result in higher water absorption. CC was generally found to present good F-T resistance compared to non-aerated concrete although the CC with high porosity did not necessarily result in higher resistance of F-T. The addition of fly ash to mixture led to a decrease in the number of air voids smaller than 300 µm. It was also found the F-T resistance of CC was more affected by the size of the air-void. The number of air-voids smaller than 300 µm played a critical role on reducing the F-T damage in CC.

Evaluation of Modifications to the ASTM C672 Deicer Salt Scaling Test for Concrete Containing Slag Cement

Abstract The standard ASTM C672/C672M-12 deicer salt scaling resistance test has been found to be overly aggressive to concretes containing slag cement or fly ash. It was compared to the recently adopted CSA A23.2-22A test method, based on the Quebec BNQ test, as well as several modifications, including use of an accelerated curing regime developed by Virginia (VADOT). Sixteen concrete mixtures were studied using high-alkali cement, low-alkali cement, grade 100 slag and grade 120 slag with slag contents of 0, 20, 35, and 50 %. Vinsol resin air-entraining admixture (AEA) was compared to a synthetic AEA. Modifications to the test method used in this study resulted in improved deicer scaling performance of concretes containing slag and many of these modifications have been incorporated into the CSA A23.2-22A test method. While it was found that increasing the level of slag replacement resulted in increased scaling, it was found that for slag mixtures, the synthetic AEA provided an improved air void spacing factor, higher hardened air content, and improved scaling resistance compared to Vinsol resin AEA.

Determining optimum air-void spacing requirement for a given concrete mixture design using poromechanics

The frost resistance of concrete is a function of the concrete constituent properties, entrained air-void system parameters and environmental exposure history. However, only a single maximum value for the void spacing factor is specified for all types of concrete by code writing bodies for successful protection against freezing damage. The advent and utilisation of new materials over the recent years warrant reevaluation of the validity of this single pass/fail criteria established more than 50 years ago. Here, a poromechanical model, capable of incorporating concrete constituent properties, environmental exposure and air-void spacing factor, has been used to determine the role of various concrete constituents and air-void system on the damage propensity of concrete exposed to freezing temperatures. It is found that a maximum threshold of acceptance, for instance a 0.2 mm spacing factor, may not be adequate for all concrete mixture designs subject to various cooling conditions. The model also suggests that concrete with low-porosity, low-permeability mortar matrix, a characteristic property of mortar containing supplementary cementitious materials and/or low water to cement ratio, can perform satisfactorily under freezing temperatures even with a spacing factor greater than the recommended value. If utilised for design, this model will give more freedom to practitioners in ensuring concrete durability by controlling multiple factors including the concrete mixture components and proportions rather than just satisfying a single pass/fail criterion for void spacing factor for all concrete mixtures.

Experimental Study on Correlation Analysis of Air-void, Air-spacing factor and Long-term Durability for Roller-compacted Concrete pavement

PURPOSES : The use of roller-compacted concrete pavement (RCCP) is an environmentally friendly method of construction that utilizes the aggregate interlock effect by means of a hydration reaction and roller compacting, demonstrating a superb structural performance with a relatively small unit water content and unit cement content. However, even if an excellent structural performance was secured through a previous study, the verification research on the environmental load and long-term durability was conducted under unsatisfactory conditions. In order to secure longterm durability, the construction of an appropriate internal air-void structure is required. In this study, a method of improving the long-term durability of RCCP will be suggested by analyzing the internal air-void structure and relevant durability of roller-compacted concrete. METHODS : The method of improving the long-term durability involves measurements of the air content, air voids, and air-spacing factor in RCCP that experiences a change in terms of the kind of air-entraining agent and chemical admixture proportions. This test should be conducted on the basis of test criteria such as ASTM C 457, 672, and KS F 2456. RESULTS : Freezing, thawing, and scaling resistance tests of roller compacted concrete without a chemical admixture showed that it was weak. However, as a result of conducting air entraining (AE) with an AE agent, a large amount of air was distributed with a range of 2~3%, and an air void spacing factor ranging from 200 to (close to ) coming from PCA was secured. Accordingly, the freezing and thawing resistance was improved, with a relative dynamic elastic modulus of more than 80%, and the scaling resistance was improved under the appropriate AE agent content rate. CONCLUSIONS : The long-term durability of RCCP has a direct relationship with the air-void spacing factor, and it can be secured only by ensuring the air void spacing factor through air entraining with the inclusion of an AE agent.

Freeze–thaw resistance of sprayed concrete in tunnel linings

Freeze–thaw exposure of wet-mix sprayed concrete for permanent tunnel linings is characterized by thermal and moisture gradients between wet temperate rock mass/concrete and cold dry concrete/air surface. Hardened sprayed concrete with water/binder ratios in the range of 0.45–0.47 has been studied. The standard freeze/thaw tests like e.g. salt frost scaling according to CEN/TS 12390-9 and rapid freeze/thaw cycling according to ASTM C666 represent much more severe exposure conditions than the worst possible cases in tunnel linings. The sprayed concrete did not pass the salt frost scaling test with 3 % NaCl and rapid freeze/thaw cycling in water but showed no deleterious freezing dilation during sealed freezing following 4–6 months submersion in water. The porosity characteristics, air void content, spacing factor and protective pore factor also indicate a certain freeze–thaw resistance. A new functional performance freezing test method simulating the moisture and freezing gradient exposure conditions in a tunnel lining in a more realistic manner is presented. Reproducible cyclic freeze–thaw exposure during the different test series was measured. Frost damage was measured by resonant frequency on disc shaped specimens prepared from the core sample after gradient frost exposure. The new gradient frost method also indicates frost durability at full capillary saturation on concrete that deteriorated in salt frost scaling- and rapid freeze/thaw tests. The findings of this study suggest that the sprayed concrete in a sprayed concrete lining structure waterproofed with membrane based on ethyl-vinyl-acetate copolymer, is not prone to freeze–thaw damage under the tested moisture and thermal conditions.

정제유지류 및 팽창성 인공 기포 조합이 고로슬래그 다량치환 콘크리트의 내구성 및 미시적 특성에 미치는 영향

본 연구에서는 고로슬래그 다량치환 콘크리트의 탄산화 저항성 및 내동해성 향상을 위하여 팽창성 인공 기포인 Expancel과 ERCO를 조합 사용하고 그에 따른 일련의 실험을 진행하고자 한다. 연구결과에 따르면 ERCO 치환에 따라 비누화반응에 의한 모세관공극 충전효과로 탄산화 저항성이 향상되었으며, 이 때 문제점으로 지적된 내동해성 저하현상을 개선하기 위하여 Expancel을 사용하였는데, Expancel 치환율이 증가함에 따라 내동해성이 대폭 개선되는 것으로 나타났다. 이는 Expancel이 외부압력에 의해 선택적으로 체적축소현상과 기포간격계수의 감소에 기인한 결과로 판단된다. In this paper, a series of experiments was conducted to evaluate the resistibility of carbonation and freeze-thawing damage of the high-volume blast furnace slag concrete using expancel, the expandable microsphere, and ERCO, emulsified refine cooking oil. The concrete mixture of 0.45 water-to-binder ratio with 60% of blast furnace slag was evaluated for carbonation, freeze-thawing resistibility, SEM, and porosity. According to the previous research, replacing ERCO contributes on improving carbonation resistibility with capillary pore filling effect by soap foaming reaction of ERCO while significantly decreased freeze-thawing resistibility. To improve this decreased freeze-thawing resistibility, expancel was used, and thus freeze-thawing resistibility was improved as the replacement ratio of expancel was increased. It is considered that the selective volume shrunken effect of expancel due to the external pressure and decreased air void spacing factor due to expancel.

Effect of powdered activated carbon on the air void characteristics of concrete containing fly ash

Abstract One of the common methods to reduce mercury vapor emissions from coal-fired power plants involves the injection of powdered activated carbon (PAC) into the flue gases to adsorb the mercury vapor. If the PAC then settles along with the fly ash in the electrostatic precipitator, the ability to use the collected fly ash as a supplementary cementitious admixture in concrete must be established. This study examined the influence of PAC in fly ash on the air void networks and consequent properties of concrete produced with an air entraining admixture. Samples included those with PAC added to the fly ash at the laboratory as well as PAC-containing fly ash collected at a generating station. The air void content, specific surface area and spacing factor of hardened concrete were determined using an image analysis technique on epoxy-impregnated and ink-prepared samples. The PAC-containing fly ash had an insignificant effect on the air void content and spacing factor compared to samples with laboratory-added PAC. Image analysis using ink-prepared samples provided improved reliability of the air void network results compared to epoxy-prepared samples.