**1. Introduction**

LNG (Liquefied Natural Gas) has been regarded as the most realistic alternative to reduce global warming from petroleum energy because it emits very little sulfurous acid gas (SO2) recognized as a major cause of environmental problems. LNG demand is expected to be determined by the climate change response activities of each country around the world. Most of all, Asia is the world's largest importing region of LNG and accounts for two-thirds of global consumption [1–4].

Natural gas is cooled to about −163 ◦C to convert the gaseous gas to a liquid state for storage. Therefore, many kinds of research have been conducted to minimize heat loss and ensure safety against gas leakage [5–7].

LNG storage tanks can be largely divided into above-ground, in-ground and underground depending on the installation location. According to the type of inner tank, it can be roughly classified into a 9%-nickel steel tank and membrane tank. Based on the definition of NFPA 59 A (2001), BS EN 1473 (1996) and BS 7777 (1993), a 9%-Ni steel tank is classified into single, double and fully containment LNG storage tanks. The full containment LNG storage tank with relatively high safety is a double tank structure in which the inner tank and the outer tank can independently store LNG at cryogenic temperatures [8].

**Citation:** Lee, G.; Na, O. Assessment of Mechanical, Thermal and Durability Properties of High-Volume GGBS Blended Concrete Exposed to Cryogenic Conditions. *Materials* **2021**, *14*, 2129. https://doi.org/10.3390/ ma14092129

Academic Editor: Alessandro P. Fantilli

Received: 1 April 2021 Accepted: 20 April 2021 Published: 22 April 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

The inner tank stores cryogenic LNG under normal operating conditions and the outer tank is located between 1 m and 2 m from the inner tank and functions as a dike as well as has a function to support the outer roof. The tank has a prestressed concrete outer container with a flexible inner container and insulation supported by an outer tank wall. Prestressed concrete (PC) structures are suitable for storing LNG and their design incorporates special loads and a special performance at cryogenic temperatures [4,8–10].

Nevertheless, the inner and outer storage tanks of LNG involve several potential risks. Sudden failure of LNG storage tank is not acceptable, since the escaped liquid would vaporize, mix with air and form an explosive cloud. The explosions or fires resulting from such an event could lead to an unacceptable loss of life and damage to the plant and environment. To reduce the potential risks and safety issues of a concrete storage tank, design guidance is given in BS 7777, NFPA 95A and BS EN 1473 [10,11]. Moreover, Jeon et al. (2003) studied the liquid tightness design associated with the cryogenic temperature under the emergency condition of LNG leakage [6]. After then, Jean et al. (2004) focused on the major factors deciding the shape of the large LNG tank [12]. Hoyle (2013) in Chevron presented the modular design of the precast concrete outer wall, instead of the in-situ concrete wall for the full containment storage tank. It concluded that this modular concept could replace the 9%-nickel steel with concrete and reduced the material cost and construction time [13]. Based on the composite concrete cryogenic tank (C3T) of Chevron, Jeon et al. (2014) and Jo et al. (2015) also studied the precast concrete module with outer liners to shorten the construction period [14,15]. Even though the concrete module could make construction time to be saved, the connection between concrete panels and countermeasures of emergency leakage should be supported with sufficient researches.

Generally, impermeable insulations such as PUF (Poly Urethane Foam) have been located between the inner tank and outer concrete wall to prevent direct contact as demonstrated in Figure 1. Recently, to remove the impermeable insulation, cryogenic rebars as reinforcement at the inner surface of the concrete wall were partially used. Yoon (2012) suggested the application of cryogenic rebar to LNG storage outer tank [16]. Cryogenic steel rebar is used to prevent the brittle failure of concrete outer wall when the leakage of LNG occurs. Cryogenic steel rebar is a specially-designed concrete reinforcing steel for cryogenic applications and is suitable for use in storage tanks with temperatures down to −170 ◦C in accordance with EN 14620-3:2006. In fact, cryogenic rebar has been manufactured by Commercial Metals Company (CMC) in USA and ArcelorMittal in Luxembourg City.

**Figure 1.** Full containment LNG storage tank.

Concrete used as the outer shell can be directly exposed to the LNG leakage and the inner surface of outer concrete can be cooled lower than −165 ◦C of cryogenic temperature. The concrete used to contain the liquefied natural gas must withstand sub-arctic

temperatures as low as −165 ◦C and is called "cryogenic concrete". However, concrete behavior at cryogenic conditions has been not elucidated. Kogbara et al. (2013) reviewed the concrete properties under cryogenic temperatures such as permeability, coefficient of thermal expansion (CTE), tensile strength, bonding strength, compressive strength and so on [17]. Kogbara et al. (2014) investigated the damaged microstructure of concrete due to cryogenic temperature including the effect of aggregate type and introduced the design method of a damage-resistance cryogenic concrete. To demonstrate the damage effects before and after freezing, acoustic emission (AE) and X-ray computed tomography (XRCT) methods were employed. The results indicated that microcracking resistance of concrete after the cryogenic condition was very related to the type of coarse aggregate [18]. Dahmari et al. (2007) mentioned the basic cause of concrete failure under cyclic freezing resulted from the transition to ice from free water in the pores and lead to the reduction in strength and structural damage [19]. Especially, Kwak et al. (2008) studied to measure the change in mechanical properties of concrete exposed to a specific temperature range from −20 to −60 ◦C [7]. To reveal the fracture properties at temperatures ranging from 20 to −170 ◦C, Rocco et al. (2001) conducted three-point bending tests on notched beams and determined the fracture parameters with the cohesive crack model, in terms of tensile strength, fracture energy, softening curve (stress vs. crack opening), characteristic length and modulus of elasticity [20]. Recently, for developing outer concrete, Kim et al. (2018) investigated the flexural and cracking behavior of ultra-high-performance fiber-reinforced concrete (UH-PFRC) before and after exposure to cryogenic temperatures through four-point bending tests. The test results indicated that UHPFRC had higher resistance to microcrack formation and better flexural performance rather than normal concrete [21]. Moreover, Mazur et al. (2019) also carried out laboratory tests to reveal the negative effect under low temperature and suggest the improved ways of mix design with respect to decease in *w/c* ratio, type of cement and aggregate and use of aeration admixture [22].

As shown in Figure 1, concrete outer wall in LNG tank is generally designed as mass concrete with 1 m thick and more and most of LNG tanks are located in coastal area. Thus, this concrete outer wall can be easily exposed to chloride-rich environment. To enhance the concrete durability in corrosive environment, high-volume of GGBS (Ground Granulated Blast-furnace Slag) should be added. Based on ACI 233 and some references, basically, more than 50% of GGBS replacement in concrete mixture has an influence on the improvement of the durability and the reduction of the heat of hydration in mass concrete [23–27]. Rashad et al. (2017) and Rachel et al. (2019) investigated the mechanical and durability properties of the high-volume GGBS mixture with metakaolin and flyash. Even though the replacement of GGBS increased, test results of RCPT (Rapid Chloride Permeability Test), sorptivity and water permeability were lower than those of conventional concrete mixture [25,26]. Recently, Lee et al. (2020) evaluated the optimal CaO content range to secure the durability performance. As a result, the optimal CaO content was within range of about 55% and the replacement ratio of GGBS was about 50% [27]. Therefore, highvolume of GGBS replacement would be necessary to improve the durability performance of concrete outer tank installed on the coastal area.

Despite the progress of many types of research about the properties of cryogenic concrete, there are some limitations of the researches using the composition of high-volume GGBS binder in mix design. In addition, sufficient research results about the practical procedure of cryogenic tests have not been provided significantly with focus on the concrete properties. Therefore, the purpose of this study is to suggest the optimum mix design with a high volume of GGBS replacement and the procedure of the cryogenic test to consider mechanical and thermal properties, and durability performance based on the review of ACI 376 [28].

Above all, ACI 376 was reviewed to define the investigation items about mechanical and durability properties under cryogenic environment. Then, all raw materials used in mix design were tested to compare the test results with requirements in ASTM and BS codes. Particularly in this study, high-volume of GGBS and air entrainer admixture were used to the control of heat of hydration and durability for the increase of freeze-thawing resistance in accordance with emergency condition of LNG leakage. Two types of cryogenic conditions were employed, and various specimens were tested to measure the mechanical, thermal and durability properties. Finally, with mock-up specimens, productivity and semi-adiabatic tests were carried out.

#### **2. Experimental Plan**

#### *2.1. Materials and Mix Design*

All materials used in concrete followed the related standards in Table 1. Cement was combined with Ground Granulated Blast-furnace Slag (GGBS) in conformity to ASTM C 595 [29]. Table 2 shows the requirements for cement from ASTM C 150 and all test results satisfied with the requirements are provided by the supplier [30].

**Table 1.** Concrete Materials and Standards [30–35].



In order to achieve the strength and durability of the concrete, the replacement of GGBS as a mineral admixture is necessary. GGBS decreases a permeability of concrete and improves chemical resistance such as chlorides and sulfates. It also reduces the heat of hydration related to Delayed Ettringite Formation (DEF). The slag constituent shall not exceed 70% of the mass of total cementitious material in the concrete mix. Table 3 shows the requirements for GGBS from ASTM C 989 and all test results satisfy the requirements [31].


**Table 3.** Properties, requirements, and test results for GGBS according to ASTM C 989.

Unwashed original sand contains many fine particles less than the sieve number 200 (0.075 mm) which induce more water and admixture consumption and then weaken the strength and durability of concrete. Sand as fine aggregate was washed at the washing plant before supplied for the test. The grading and the requirement of fine aggregates are shown in Figure 2 and Table 4, respectively. Coarse aggregates were washed gravels or crushed stones in accordance with Table 5. Coarse aggregates were combined with two types of single size, 20 mm and 10 mm and supplied from local providers. In Table 5, all test results for coarse aggregate were compared with the requirements.

In order to improve workability, strength and setting time, a high range and retarding super-plasticizer as a chemical admixture was used, complying with ASTM C 494. Air entrainer admixture shall comply with ASTM C 260. Micro-air 100 as an air entrainer admixture was used for improving a freeze-thaw resistance under a cryogenic environment. Mixing water was used without oil, acid, alkaline and organic matters or deleterious substances, complying with ASTM C 94 as shown in Table 6.

**Figure 2.** Grading of fine aggregate.


**Table 4.** Properties, requirements and test results for fine aggregates [36–42].

**Table 5.** Properties, requirements and test results for coarse aggregates [36,43–48].


**Table 6.** Properties, requirements and test results for mixing water.


### *2.2. Mix Design for Cryogenic Concrete*

The specified compressive strength of cryogenic concrete was 40 MPa and target compressive strength was 50 MPa which was determined by adding 10 MPa to the specified compressive strength due to variations in materials, operations and testing. The maximum size of aggregate was 20 mm. The target slump and slump flow were 220 ± 25 mm and 620 ± 75 mm, respectively. The water/binder ratio was selected with 0.28, and total cementitious content was varied between 475 kg/m<sup>3</sup> to 495 kg/m3. Water content was started from 128 to 133 kg/m<sup>3</sup> for cryogenic concrete. For water reduction, two types of high-range water-reducing chemical admixtures were applied: Daracem 208 (GCP applied technologies, Cambridge, MA, USA) as naphthalene type and Baxel PC 650 (Baxel, Sharq, Kuwait) as polycarboxylate type. Air content of 4 ± 1.5% was also achieved using a proper air-entraining admixture. Mix proportions are listed as shown in Table 7. Concrete materials were mixed by following ASTM C 192.



Type I: Naphthalene type (Daracem 208), Type II: Polycarboxylate type (Baxel PC 650).

*2.3. Preparation of SPECIMENS and Test Methods*

For fresh concrete, air content and retention time of concrete slump were measured. For hardened concrete, mechanical properties such as compressive strength and elastic modulus were carried out. A total of 15 specimens were prepared in each mixture including reserved samples as shown in Table 8.


**Table 8.** Test method and concrete specimens.

For one-time cryogenic test (Test A method), a total of 38 specimens were prepared as referred to Table 9. Test A method consisted of five specified tests: compressive and tensile strength, elastic modulus, thermal expansion coefficient and thermal conductivity. Each test was conducted under four different temperature conditions. For freeze-thaw cyclic test (Test B method), a total of 12 specimens were cast as referred to Table 10. Test B method was conducted to investigate the compressive strength and elastic modulus after 50-times freeze-thaw cycles.
