1. Introduction
Concrete durability at high elevation has attracted the attentions of increasing researchers [
1,
2,
3,
4]. At high elevations, climate features including low atmospheric pressure, low relative humidity, low average temperature, and large range of diurnal temperature cycles are not friendly for concrete structures. Low average temperature coupling with a large range of diurnal temperature cycles delivers more annual freeze-thaw cycles over there [
4,
5]. The most effective way for concrete to improve frost resistance is increasing its air content by air-entraining agent (AEA). However, in engineering practice, some researchers and engineers found that the air content of fresh concrete in the Qinghai-Tibet Plateau, which is known as the roof of the world, was lower than that in low elevation regions. In their opinion, this phenomenon is owing to the atmospheric pressure difference [
6,
7]. To prove this opinion, many works have been done, which mainly include the following: (i) shaking/agitation of AEA solutions in different air pressures [
5,
8,
9]; (ii) concrete air content test at different elevations, corresponding to different atmospheric pressures [
3,
4,
5,
9,
10,
11]; and (iii) air content test of concrete mixed in different simulated air pressures [
12,
13,
14,
15,
16]. The majority of these works [
9,
10,
12,
13,
14,
15,
16] drew a common conclusion, stating that low atmospheric pressure is the main cause of low air content of concrete in the Qinghai-Tibet Plateau. However, there were huge differences between what different people claimed about how much the air content of concrete would be affected. Moreover, many of these works were not designed considerately and some of them even violated usual physics laws.
To study the effect of atmospheric pressure on the foaming properties of AEA, some researchers manually shook a glass tube that contained certain AEA solution at different elevations [
5,
9]. They found that the foam columns were lower and declining faster in Qinghai compared with those in Beijing or Xi’an. This means that both foaming capacity and foam stability of AEAs at higher elevation are weaker. However, this conclusion became debatable recently. Li [
8] conducted a similar test, in which he applied a mechanical agitation at a much higher speed. He found that neither the foam generation nor foam development in AEA solution was affected by the air pressure. Moreover, it is easy to understand that bubbles of solution cannot perfectly match bubbles in concrete:
- –
The nature of a bubble in the glass tube is a layer of spherical liquid film, which has two curved surfaces (outer liquid-inner gas surface and outer gas-inner liquid surface), while the air bubble in fresh concrete has only one curved surface (outer liquid-inner gas surface).
- –
In the glass tube, there is a gas-liquid two-phase system, while in fresh concrete there is a gas-liquid-solid (bubble-paste-aggregate) three-phase system.
- –
Bubbles observed and analyzed in the glass tube are at the top of liquid phase, however, in fresh concrete, bubbles that emerge to the surface would vanish quickly.
Therefore, the best way to study AEA properties in low atmospheric pressure is still concrete test, however, variables irrelevant to the atmospheric pressure should be strictly controlled.
Mix proportion, cementitious materials, and chemical admixtures were controlled in nearly all on-site studies [
3,
4,
5,
9,
10,
11], but only a few studies [
3,
4] controlled the aggregate sources. Moreover, few publications mentioned that the temperature of raw materials, instead of the temperature of the laboratory only, was kept the same when tests were conducted at different elevations. These two variables might induce significant errors to the research results. Firstly, a high clay content, which would negatively affect air entrainment, is a typical concern of aggregate quality in the Qinghai-Tibet Plateau [
5]. Secondly, raw materials stored at high elevation are much colder than those stored at low elevation. In some projects in the Qinghai-Tibet Plateau, the mixing water was drawn from local rivers [
5], which are the snowmelt from mountains. The highest temperature of the river water in the hottest summer is only 8–10 °C [
5]. The cold temperature of raw materials is an important factor that reduces the air content of concrete.
For the considerations of altitude stress risks, logistical and experimental conditions, irrelevant variables control, as well as travelling time and expense, some researchers studied the effect of atmospheric pressure on air content of air-entrained concrete by simulating different air pressures in a laboratory instead of conducting tests at different elevations [
12,
13,
14,
15,
16]. Li and Fu [
12,
13,
14] firstly applied this methodology in their works, which strongly influenced later researchers [
15,
16], whose methodologies were the same as that applied by Li and Fu [
12,
13,
14]. By the methodology, concrete is mixed in a sealed box with simulated air pressures, however, the simulated air pressures have to recover to local atmospheric pressure when mixing finishes, as, only if the concrete is taken out from the sealed box, can it experience subsequent tests. According to the Boyle-Mariotte’s law (P
1V
1 = P
2V
2), air bubbles entrained in low air pressure during mixing would be inevitably compressed when the ambient pressure rises to standard atmospheric pressure after mixing finishes. Unfortunately, these researchers [
12,
13,
14,
15,
16] ignored the compression by citing a paper of Ley [
17], which stated that the shell damage of air bubbles with a diameter less than 200 μm was difficult to observe in 500× magnification. The description from Ley [
17] was fine, but obviously, it was misinterpreted by theses researchers as that small air bubbles are not compressible. The ignorance of the compressibility of air bubbles inside concrete indicates a negation of the Boyle-Mariotte’s law, which is exactly the operational principle of air content measurement by the pressure method [
18,
19]. If the air bubbles inside concrete are incompressible, then the needle on the dial of pressure gauge would not move, which means the air content would never be detected. Paradoxically, these researchers still applied the pressure gauge method to test the air content of air-entrained concrete [
12,
13,
14,
15,
16]. A critical discussion about the outcomes of these studies is provided in
Appendix A.
This paper delivers a comprehensive study on air entrainment efficiency and stability in cement-based materials in both low and standard atmospheric pressures. Given the technical barrier that concrete cannot be tested in a sealed environment, the study was conducted on-site at different elevations instead of basing on simulated pressures. For tests on cement paste and mortar, specimens were prepared in Shigatse (3860 m, 64 kPa) and Harbin (150 m, 101 kPa), respectively. In these tests, the density of fresh cement paste, the air content of mortar at both fresh and hardened states, the air bubble stability in fresh mortar, and the air void characteristics of hardened mortar were investigated. However, concrete specimens for air content tests were prepared only in Lhasa (3646 m, 66 kPa) instead of at different elevations, because coarse aggregates were difficult to keep the same in two places separated by great distance. The study found that cement paste density, mortar air content, and air bubble stability in mortar were not significantly affected by atmospheric pressure change. Moreover, the air void characteristics of hardened mortar with SJ-2 or 303R type AEA were similar at either atmospheric pressures. Concrete with either SJ-2 or 303R type AEA, prepared in low atmospheric pressure, presented a satisfactory air content. Additionally, it is found that the low temperature of raw materials stored at high elevation should be an important factor causing poor air entrainment. These conclusions dispel excessive worries about the potentially negative effect of low atmospheric pressure on air entrainment in cement-based materials and remind one that people pay more attention to other factors that may adversely affect air entrainment at high elevation.