Experimental and numerical study of basalt fiber ...

Engineering Structures 152 (2017) 901–919

Contents lists available at ScienceDirect

Engineering Structures journal homepage: www.elsevier.com/locate/engstruct

Experimental and numerical study of basalt fiber reinforced polymer strip strengthened autoclaved aerated concrete masonry walls under vented gas explosions Zhan Li a,b, Li Chen a,⇑, Qin Fang a, Wensu Chen b,⇑, Hong Hao b, Yadong Zhang a a b

State Key Laboratory of Disaster Prevention & Mitigation of Explosion & Impact, PLA Army Engineering University, Nanjing, Jiangsu 210007, China Centre for Infrastructural Monitoring and Protection, School of Civil and Mechanical Engineering, Curtin University, Kent Street, Bentley, WA 6102, Australia

a r t i c l e

i n f o

Article history: Received 19 June 2017 Revised 26 September 2017 Accepted 26 September 2017

Keywords: Autoclaved aerated concrete (AAC) Basalt fiber reinforced polymer (BFRP) Two-way masonry wall Vented gas explosion Field test Numerical simulation

a b s t r a c t Ten full-scale field tests were conducted to study the performance of basalt fiber reinforced polymer (BFRP) strip strengthened autoclaved aerated concrete (AAC) masonry walls subjected to vented gas explosions. Three walls i.e. unstrengthened, rear-face strengthened and front-face strengthened wall specimens were prepared for blast tests. The testing data including overpressure time histories of vented gas explosions, displacement time histories, damage characteristics and fragment distribution of each wall specimen were recorded and analyzed. It was found that the rear-face strengthened wall specimen showed the best blast-resistant performances. Three wall specimens under vented gas explosions experienced damage modes of typical two-way flexural deformation along with shattering of AAC blocks at the latter stage of gas explosions. A detailed micro model for masonry walls was developed in LSDYNA, incorporating material parameters obtained from material tests. The accuracy of numerical model in predicting the responses of masonry walls was validated with the testing data. Parametric studies were also conducted to explore the influences of BFRP strip layouts, strip thickness and fiber types on the performances of masonry walls. It is found that the BFRP strip layout, strip thickness and fiber types affect the resistance capacity of masonry walls significantly. Ó 2017 Published by Elsevier Ltd.

1. Introduction Blast loads can be generated from various explosion sources (e.g. high explosive charges, vapor gas explosions, bursting pressure vessels, etc.) in industrial, civilian and military fields, and they impose great hazards to existing structures [1,2]. Masonry wall has been widely used in the constructions of masonry buildings as load-bearing components and in reinforced concrete (RC) frame buildings as non-load-bearing components. The unreinforced masonry wall is vulnerable to blast loads due to its low resistance capacity. The failure of masonry walls may claim lives and enormous loss of properties. Finding a cost-effective method to strengthen masonry walls is a necessity and draws attentions in recent decades [3–5]. Strengthening techniques by using different strengthening materials have been developed for masonry walls against blast loading. The masonry wall subjected to out-of-plane loads (static and high explosive loads) were studied by conducting field blasting ⇑ Corresponding authors. E-mail addresses: [email protected] (L. Chen), [email protected] (W. Chen). https://doi.org/10.1016/j.engstruct.2017.09.055 0141-0296/Ó 2017 Published by Elsevier Ltd.

tests, numerical simulations and theoretical analysis, as summarized in Table 1. Composite materials including fiber reinforced polymers (e.g. CFRP and GFRP) and spray-on polyurea, and steel wire mesh, have been used to strengthen concrete masonry unit (CMU), clay brick and AAC masonry walls. The composite structures made of composite material and masonry wall can make use the full advantage of their own characteristics. When the applied loading is not devastating leading to wall collapse, the existence of the composites enhances the equivalent section area of structures, increases the stiffness significantly and in turn reduces the deflection of structures. By means of absorbing strain energy, the failure of composites can mitigate the damage of main structures effectively. When intensive loading is applied, the composites can decrease the number of fragments and can also be a catching system to effectively mitigate the threats from flying debris. As an ‘ultra-lightweight’ concrete material, autoclaved aerated concrete (AAC) has been used as an alternative to conventional normal-weight or lightweight concrete products [22,23]. The micro cellular structures of AAC bring excellent thermal and sound insulation properties, but also result in strength reduction and heterogeneity of the material. Currently, the performance of AAC

902

Z. Li et al. / Engineering Structures 152 (2017) 901–919

Table 1 Summary of previous studies on the masonry wall strengthening. Researchers

Year

Oswald and Wesevich [6] Crawford et al. [7,8] Muszynski and Purcell [9] Myers et al. [5] Baylot et al. [4] Davidson et al. [10] Urgessa et al. [11,12] Hrynyk and Myers [13,14]

2001 2002, 2003 2003, 2005 2005 2008, 2007,

Tan and Patoary [15] Irshidat et al. [16,17] Chen et al. [3,18] Elsanadedy et al. [19] S.H. Alsayed et al. [20] Wang et al. [21]

2009 2010, 2011 2013, 2014 2016 2016 2016

2003 2004

2009 2008

Structures

Strengthening material

Experimental

CMU walls CMU walls AEC walls CMU walls CMU walls CMU walls CMU walls Clay brick CMU walls Clay brick walls CMU walls Clay brick CMU walls CMU walls Clay brick, AAC

Aramid CFRP CFRP GFRP GFRP Polymer spray GFRPs, CFRP GFRP

Shock tube

CFRP and GFRP Nanoparticle polymer CFRP, steel wire mesh, steel plate GFRP GFRP Polymer spray

Explosive Explosive (blast simulator) Explosive Static Explosive Explosive

masonry walls strengthened with composite materials under blast loads is mainly studied by using experimental method. Muszynski and Purcell [9] conducted field tests on the air-entrained concrete (AEC) masonry walls strengthened with CFRP subjected to high explosive detonations. However, the test data were not reported in detail. Their study only concluded CFRP laminate strengthened AEC masonry wall exhibited less residual displacement than the bare walls. Wang et al. [21] performed field tests to obtain the dynamic response of spray-on polyurea reinforced AAC block masonry walls under TNT explosions. The results showed that the existence of ployurea layer improved the blast resistance of walls significantly, prevented the wall collapse and reduced the number of high-velocity fragments effectively. Yankelevsky and Avnon [24] carried out tests to study the localized responses of AAC under contact explosion and the effectiveness of surface treatment by gluing textile fabric on the ACC block performances. The test results showed that the surface treatment enhanced the tensile strength and ductility of AAC blocks, and contributed to higher resistance to tensile wave spalling. The above-mentioned studies focused on the performances of AAC blocks and strengthened AAC masonry walls under blast loads generated by high explosives. However, the study of AAC masonry walls subjected to gas explosions is very limited in the open literature. In the previous study [25], the responses of unreinforced AAC masonry walls under gas explosion loads were investigated by conducting experimental, numerical and simplified analytical studies. The behavior and failure mode of unreinforced AAC masonry walls under gas explosion loads were investigated and discussed in detail, which are not necessarily the same as the walls under blast loads from high explosives owing to the unique characteristics of gas explosion loads (such as lower amplitude, longer rise time, longer duration and possibly multiple peak pressures). No study of the responses of strengthened ACC masonry walls subjected to gas explosions has been reported in the open literature yet. As compared with CFRP and GFRP, BFRP, which is made from basalt rocks through melting process, is a relatively new type of FRP composite. BFRP is usually inferior to CFRP in terms of the ultimate strength and Young’s modulus and to GFRP in terms of the ultimate strain. BFRP also has good fire resistance and is costeffective therefore a potential material for strengthening structures [26,27]. Limited studies were carried out to research the material properties of BFRP products under dynamic loads especially blast loads [28,29]. The static and dynamic tensile properties of BFRP strips were studied by Chen et al. [28] at the strain rate from 4.68 e5 s1 to 259.0 s1. The dynamic enhancements on the strength, elastic modulus and failure strain of BFRP strips at different strain rates were well captured and the empirical formulae were proposed. Basturk et al. [29] investigated dynamic

Numerical

Theoretical

DYNA3D Explosive Explosive Explosive Explosive Explosive Static

DYNA3D YES

AUTODYN LS-DYNA

YES YES YES

AUTODYN

behaviors of laminated basalt composite plates under blast loads based on large deflection theory of thin plate. The theoretical result showed that the basalt fiber composite plate might serve as an alternative to composite structures. However, no study on the performance of BFRP strengthened masonry walls and other RC structures (beam, column and slab) under blast loads can be found in the literature. Therefore, it is necessary to investigate the performance of BFRP strengthened structures under blast loads. In this study, a series of full-scale field tests on BFRP strip strengthened AAC masonry walls subjected to vented gas explosions were conducted. A detailed micro model was also developed to predict dynamic responses of strengthened AAC masonry walls by using LS-DYNA. The parameters of the material models for AAC blocks, mortar and FRP strips were obtained from material tests. The predictions from the numerical simulations were compared with the testing data to validate the numerical model. Parametric studies were carried out to evaluate the effects of strip layouts, strip thickness and fiber types on the response of strengthened masonry walls subjected to vented gas explosions. 2. Experimental program A 3 m 2 m 2 m reinforced concrete (RC) chamber with two openings was built for the field gas explosion tests. The small opening of 0.8 m 0.8 m was installed with vent covers and acted as a vent window. The larger opening with dimensions of 3 m 2 m was used to place wall specimens for testing. 2.1. Preparation of wall specimens Three wall specimens, including one unstrengthened specimen, one rear-face strengthened specimen and one front-face strengthened specimen were prepared according to the standard [30]. The wall specimens with dimensions of 3 m 2 m 120 mm were lumped with AAC blocks with dimensions of 590 mm 120 mm 240 mm and 5 mm-thick mortar in ‘‘Running” pattern, as shown in Fig. 1. The wall specimens were pre-built inside the RC frames for the convenience of installation. The RC frame made of C30 concrete has the internal dimensions of 3 m-high and 2 m-wide and crosssection area was 30 cm 30 cm. To fasten the RC frame onto the RC chamber, a total of 24 holes with the diameter of 70 mm were predesigned and uniformly distributed along the edges of RC frame as shown in Fig. 1(b). Correspondingly, 24 steel bars with the diameter of 50 mm were pre-embedded into the RC chamber. The RC frames were fastened on the RC chamber by using 24 nuts. The right and left boundaries of wall specimens were bolted into the RC frames by using steel bars at an equal space of 25 cm. The

903


RC frame

2m

120 mm 3m

240 mm 590 mm

Predesigned holes (dia:70 mm)

(a) AAC blocks

(b) Wall specimens

Fig. 1. Dimensions of AAC blocks and wall specimens.

length of anchoring steel bars was 20 cm and the diameter was 8 mm. The uniaxial tensile strength of steel bars was 235 MPa. The holes with the depth of 10 cm were created along the right and left sides of RC frames for the steel bars. Then the steel bars were inserted into the holes and the glues were used to ensure the bond of steel bars and RC frames. Unidirectional BFRP strip (Tongzheng, Shanghai) with a unit weight of 300 g/m2 was used to strengthen wall specimens, as shown in Fig. 2. As provided by the supplier, the tensile strength of dry fiber was 2100 MPa and the nominal thickness of the fiber was 0.12 mm. The corresponding density of basalt fiber was 2.5 g/cm3. The design of strengthened wall specimens is presented in Fig. 3. The width of BFRP strips was 50 mm. Seven strips were distributed at a distance of 250 mm along the vertical direction of wall specimens and eleven strips were also applied along the transverse direction of wall specimens at a distance of 250 mm, as shown in Fig. 3(a). The epoxy resin (Tongzheng, Shanghai) consisting of two components (i.e. main agent and hardener) with a ratio of 2:1 was used as bonding agent for FRP application. The epoxy resin had a tensile strength of 42 MPa, tensile modulus of 2.7 GPa and rupture tensile strain of 1.6%. The ends of BFRP strips were fixed into the RC frame by using angle irons. The dimensions of angle irons used in tests were 50 mm 50 mm 5 mm and the design strength was 235 MPa. Expansion screws were used to attach angle irons and RC frames and specific structural glue (Tongzheng, Shanghai) was also used to attach FRP strips, angle irons and RC frames. The BFRP strip distribution for the two strengthened specimens were the same. The rear-face strengthening and the front-face strengthening were prepared by switching the strengthened side facing inside or outside chamber during installation as shown in Fig. 3(b).

2.2. Experimental setup The methane with the concentration of 99.9% (Nanjing Chenhong, China) was chosen as the flammable gas in the tests and piped into the RC chamber from the cylinder gas tanks. An infrared gas analyzer (QGS-08C, Nanjing Xinfen, China) equipped with a thermostat system was employed to monitor the methane concentration. The measuring range and sensitivity of the gas analyzer were 0–15 vol% and 0.1 vol%, respectively. An explosion-proof fan (CBF-300, Zhejiang Dafeng Blowers, China) was installed in the chamber to ensure the homogeneity of gas mixtures. The methane/air mixture was ignited by the igniting pill with the energy of 100 mJ, which was triggered by the high voltage power source. At the same time of ignition, the high voltage power source also triggered the data acquisition system and the high speed camera. Fig. 4 shows the test setup for the rear-face strengthened case. All specimens were tested under the vented gas explosion ignited at the center of the RC chamber. 2.3. Instrumentation The testing data including gas explosion loads, displacements and failure modes of wall specimens were captured by using pressure sensors, displacement transducers and high-speed camera, respectively. Four piezoresistive pressure sensors with a measuring range of 20 to 150 kPa (CYG 1409, Kunshan Shuangqiao, China) were used to record the overpressure. Three pressure sensors were mounted on the front wall at the height of 0.2 m, 0.8 m and 2.5 m along the midline of the wall, respectively. One pressure sensor was mounted 0.9 m off the center line at the height of 1.5 m as shown in Fig. 4(b). It should be noted that the pressure sensors

Fig. 2. Unidirectional BFRP strip.

904


BFRP strips: 50 mm in width

Angle irons 11 @250 mm

RC chamber Wall specimen

Ignition point Front face

Rear face

7 @250 mm

(a) Layout of BFRP strips on wall specimens

(b) Schematic diagram of specimen position

Fig. 3. Design of strengthened wall specimens.

Displacement sensor Pressure sensors

Water tank RC chamber

RC chamber

Gas analyser probe

Wall specimen

Vent window

Fan

(a) Front view

(b) Back view

Fig. 4. Test setup (rear-face strengthened cases).

were installed to record the pressure inside the chamber. A water circulation system was used for the four pressure sensors to prevent the possible damage and data distortion by the relativelylong-time high temperature (about 1 s and 2000 °C) during gas explosions. The water tank and pump, key parts of water circulation system, were located on the top the RC chamber as shown in Fig. 4(a). Five displacement transducers with a measuring range of 0–300 mm (WYJL, Xian Xinmin, China) were fixed along the midline of wall specimens at an equal space of 75 cm to acquire the displacement time histories. A high speed camera (FASTCAM SA-Z, Photron Co., Ltd., Japan) was placed behind the wall about 15 meters away from one side of the wall to record the failure process of wall specimens. The signals from the pressure sensors and displacement transducers were logged by the data acquisition instrument (DongHua 5927, Donghua Testing Technology Co., Ltd., China) at a sampling rate of 30 kHz. 2.4. Testing scheme A total of ten tests were conducted to investigate the performance of AAC walls strengthened by BFRP strips under gas explo-

sion loads, as listed in Table 2. Three wall specimens, i.e. unstrengthened wall as control specimen, rear-face strengthened specimen and front-face strengthened specimen were tested. Different gas concentrations and vent covers were used to achieve the desired gas explosion loads. 2.5. Test results and discussions 2.5.1. Vented gas explosion loads In the previous studies [25,31–33], the characteristics of gas explosion loads have been analyzed and discussed. In this study, only pressure-time histories recorded by one pressure sensor are presented, and the pressure time profiles are filtered to remove high-frequency pressure oscillations. It should be noted that pressure oscillation of high frequency is one of the characteristics of gas explosion loads and the oscillation is filtered for the convenience sake in the discussion as such high-frequency oscillations have only minimum effect on structural responses. The overpressure-time histories acquired from the ten tests are shown in Fig. 5 and summarized in Table 3. In Test 7 and Test 10, the vent cover installed on the vent window survived the gas

905

Z. Li et al. / Engineering Structures 152 (2017) 901–919 Table 2 Testing scheme. Test no.

Wall no.

Basalt strips fiber

Gas concentration (%)

Vent cover

1 2 3

W1

Unstrengthened

12.5 12.5 9.5

6 lm plastic film 4 mm glass panel 1.5 lm plastic film

4 5 6 7

W2

Strengthened at rear-face

12.5 12.5 12.5 9.5

6 lm plastic film 4 mm glass panel 5 mm glass panel 12 mm glass panel

8 9 10

W3

Strengthened at front-face

12.5 12.5 7.5

6 lm plastic film 4 mm glass panel 12 mm glass panel

explosions and wall specimens acted as the vent covers instead. It is found that vented gas explosion loads are sensitive to the test conditions. By comparing the gas explosion loads of Test 1/2/4/5/6/8/9 with 12.5% gas concentration, it is found that the profiles of pressure at the rising stage from all these tests agree with each other well, which indicates that the gas concentration control is reliable and accurate. It can be noted that Tests 2/5/9 with the same gas concentration of 12.5% and the vent cover of 4 mm glass panel generated gas explosion loadings with different peak values. They are 13.25 kPa, 9.27 kPa and 7.57 kPa, respectively for Tests 2/5/9. This might be caused by the slight difference of boundary conditions during the installation of vent covers, which results in different initial stress of glass panels and in turn the variation of gas explosion loads, and/or the strength variations of the glass panel that lead to different venting pressures. These results demonstrate the great sensitivity of the gas explosion loads on venting conditions. The effect of venting conditions on the gas explosion loads can be examined by comparing the data of Test 3 and Test 7 with the same gas concentration of 9.5% but different venting conditions. For Test 3, the vent window with vent area of 0.64 m2 was covered with 1.5 lm plastic film. Whereas the wall specimen W2 acted as the vent cover with a vent area of 6 m2 in Test 7, which resulted in a higher vent pressure in the test. The higher vent pressure of Test 7 contributed to the pressure increase and flame expansion inside the chamber to a large extent. The larger vent area facilitated the combustion products to flow out of the chamber. Therefore, only one peak pressure (DP1 ) appeared in the pressure time history of Test 7. However, the smaller vent area and low vent pressure of vent cover in Test 3 generated a smaller first peak and enhanced turbulence combustion, which led to a much higher peak (DP4 ) at the latter stage of gas explosion. Therefore, the vent area and vent pressure should be carefully considered in the venting design to ensure the safety of structures and equipment. This, however, is not a topic of discussion in the present study. The present study concentrates on investigating the performance of BFRP strip strengthened ACC masonry wall subjected to different gas explosion loads. 2.5.2. Displacement of wall specimens The mid-span displacement-time histories of wall specimens are also shown in Fig. 5, and the peak and residual displacements are given in Table 3. It should be noted that the displacements were not measured in the last destructive test of each wall specimen (W1, W2 and W3) to avoid destroying displacement transducers, i.e., the transducers were removed before destructive tests. As shown, the curves of displacement histories have similar trend with the corresponding overpressure-time history profiles. It indicates a quasi-static characteristic of the vented gas explosion loads. Because the rising stage of overpressure is much longer than that of the typical impulsive load. What is more, the responses of wall specimens were elastic and no residual displacement of wall specimen was detected in the first test of each wall specimen (Test 1,

Test 4 and Test 8) as listed in Table 3. In Test 5, the second test of W2, the residual displacement was only 0.23 mm, which is ignorable compared with the thickness of wall specimens. Therefore, neglecting the accumulative damage of wall specimens after Test 1/4/5/8 should be acceptable. The effect of strengthening methods on the elastic responses (i.e. low response level) of wall specimens is studied by comparing the test data of Test 1, Test 4 and Test 8. Because it was not possible to accurately control such tests, the recorded peak pressures in the three tests are different. Therefore, the ratio of P/D (i.e. the ratio of peak pressure over maximum displacement) is used as a parameter to compare the strengthening performance. They are 1.31, 2.87 and 2.72 for the unreinforced wall W1, the rear-face strengthened wall specimen W2 and the front-face strengthened wall specimen W3, respectively. The elastic response of walls is determined by structural stiffness which is associated with boundary conditions, crosssection geometry and material properties. The stiffness changes of W2 and W3 are owing to the changed boundary conditions by adding angle irons and the changed equivalent section area by applying BFRP strips. Both adding angle iron and BFRP strips affect the strengthening effectiveness. The P/D ratios for the strengthened wall W2 and W3 with the same boundary condition, section geometry and material properties are close to each other. However, W2 with the BFRP strips pasting on the backside of the wall gives slightly higher P/D ratio against the explosion. It is attributed to the large tensile stiffness of BFRP strips pasting on the back face. It increases flexural stiffness of the wall-BFRP composite structure. BFRP strips pasting on the front face of the wall does not increase the flexural stiffness because the BFRP strip has little compressive strength. The effect of strengthening methods on the inelastic responses (i.e. high response level) of wall specimens is studied by comparing the testing data of Test 2, Test 5 and Test 9. Under the same testing condition, the maximum displacement of W2 in Test 5 is 4.29 mm, which accounts for 13.2% of the maximum displacement of W1 (i.e. 32.62 mm) and 75.8% of that of W3 (i.e. 5.66 mm). In terms of the ratio of P/D, W2 has the maximum ratio of 2.17, followed by W3 (1.33) and W1 (0.41). Therefore, it can be concluded that the rear-face strengthened wall specimen W2 has the better resistance capacity than the front-face strengthened wall specimen W3, and unstrengthened wall specimen. These results are expected, but confirmed in the tests with quantitative data that will be used for further analyses. It is noted that little debonding of BFRP strips pasting on the back face was observed in the wall W2 in Test 6, in comparing with field tests under high explosive loads [3]. It also strongly demonstrates the long rising characteristic of the overpressure induced by vented gas explosions and the reflection of stress wave in the wall is not strong enough to induce local debonding of BFRP strips backside. The wall specimens discussed in this study were strengthened by changing boundary conditions through adding angle irons and increasing cross section through applying BFRP strips. Both factors

906


(a) Pressure-time histories for W1

(b) Displacement-time histories for W1

(c) Pressure-time histories for W2

(d) Displacement-time histories for W2

(e) Pressure-time histories for W3

(f) Displacement-time histories for W3

Fig. 5. Testing results for the wall W1, W2 and W3.

affect the strengthening effectiveness and their individual contribution can be distinguished by taking the ratio of P/D as an indicator. W1 without strengthening is the control specimen. W2 is the specimen strengthened with angle irons and BFRP strips at the rear. W3 is the specimen strengthened with angle irons and BFRP strips at the front, where the BFRP strips have negligible strengthening efficacy. It is found that the strengthening efficacy is different when the strengthened specimens experience elastic or

inelastic response. For instance, the wall specimens (W1, W2 and W3) experienced elastic response when they were subjected to Test 1, Test 4 and Test 8, respectively. The P/D ratio of W1, W2 and W3 is 1.31, 2.87 and 2.72, respectively. The improvement contributed from angle irons is 107% (comparing Test 1 and Test 8) and the improvement contributed from angle irons and BFRP strips is 119% (comparing Test 1 and Test 4). Therefore, the improvement contributed from BFRP strips is 12% when the wall specimens

907

Z. Li et al. / Engineering Structures 152 (2017) 901–919 Table 3 Vented gas explosion loads and mid-span displacements of wall specimens.

DP

Test

Wall

Residual displacement (mm)

Peak (P: kPa)

Duration (ms)

Impulse (I: Pa s)

Maximum displacement (D: mm)

P/D (kPa/mm)

1 2 3

W1

2.91 13.25 85.88

357 783 80

525 3353 3106

2.22 32.62 –

1.31 0.41 –

0.0 3.8 Collapse

4 5 6 7

W2

2.96 9.27 21.26 43.17

347 814 914 202

446 2183 5373 3353

1.03 4.29 20.26 –

2.87 2.16 1.04 –

0.0 0.21 1.23 Collapse

8 9 10

W3

3.11 7.57 32.3

684 1017 274

641 2079 3418

1.14 5.66 –

2.72 1.33 –

0.0 0.49 Collapse

Note: Only the dominate peak pressure are presented.

experience elastic response. The wall specimens (W1, W2 and W3) experienced inelastic response when they were subjected to Test 2, Test 5 and Test 9. The P/D ratio of W1, W2 and W3 is 0.41, 2.16 and 1.33. The improvement contributed from angle irons only is 224% (comparing Test 2 and Test 9) and the improvement contributed from angle irons and BFRP strips is 427% (comparing Test 2 and Test 5), which indicates the improvement contributed from BFRP strips only is 203%. It can be concluded that the existence of angle irons improves the strengthening efficacy for both elastic and inelastic response ranges. The BFRP strips make minimal contribution to the strengthening efficacy when the wall specimen experiences elastic response while the BFRP strips make significant

(a) Wall specimen W1 in Test 2

contribution to the strengthening efficacy when the wall specimen experiences inelastic response.

2.5.3. Damage modes of wall specimens The crack/damage patterns of wall specimens (W1 W3) after Test 2, Test 6 and Test 9 are shown in Fig. 6. As shown in Fig. 6(a), the unstrengthened wall specimen W1 experienced an ‘‘X” pattern through crack in Test 2. With respect to the rear-face strengthened specimen W2, it can be found that the extension of crack was constrained by the BFRP strips and there was no through crack as shown in Fig. 6(b). The front-face strengthened wall specimen

(b) Wall specimen W2 in Test 6

(c) Wall specimen W3 in Test 9

Fig. 6. Damage modes of wall specimens.

Fig. 7. Failure process of unstrengthened wall specimen W1 in Test 3.

908


W3 in Test 9 experienced the minimal visible damage with the residual displacement of 0.49 mm. The failure processes of wall specimens in the destructive tests (i.e. wall collapse) are shown in Figs. 7–9. All three wall specimens experienced the ‘‘X” pattern cracks which divided the wall into four major sections at the initial stage of structural response. Three wall specimens under vented gas explosion showed typical twoway flexural response mode. Due to the high peak overpressure and the low strength of AAC material, the AAC blocks failed and three wall specimens turned into small pieces at the latter stage of the destructive gas explosions. It should be noted that the accumulative damage brings about the degradation of structural stiff-

ness and has limited effect on the failure mode of wall specimens due to the maintained load-deflection envelope [34]. No visual damage of BFRP strips can be found in Test 4/5/6/8/9. The failure of BFRP strips were found after the collapse of rear-face strengthened wall specimen W2 in Test 7 as shown in Fig. 10. As shown in Fig. 10(a), the rupture of BFRP strips occurred at the boundary of wall specimens with the combination of tensile and shear stress and the debonding failure can be observed along the boundary. In the mid-span area with maximum tensile stress, the rupture can be also observed as highlighted in Fig. 10(b). As to the front-face strengthened wall specimen W3, the BFRP strips experienced typical debonding failure along the edges, and BFRP

Fig. 8. Failure process of rear-face strengthened wall specimen W2 in Test 7.

Fig. 9. Failure process of front-face strengthened wall specimen W3 in Test 10.

(a) FRP strip rupture near the boundary

(b) Rupture of BFRP strips

Fig. 10. Damage of BFRP strips of wall specimen W2 in Test 7.


(a) Fragment distribution of W1

909

(b) Maximum fragment size

Fig. 11. Fragments of W1 after Test 3.

(a) Wall specimen W2 after Test 7

(b) Wall specimen W3 after Test 9

Fig. 12. Distribution of fragments after the failure of wall specimens.

strips were pulled out directly. Therefore, different strengthening methods cause different failure modes of BFRP strips and eventually affect the resistance capacity of structures. 2.5.4. Fragment distributions The fragment distributions of wall specimens (W1 W3) after gas explosions are shown in Fig. 11 and Fig. 12. As shown in Fig. 11(a), W1 totally collapsed after Test 3 and the maximum flying distance of fragment was about 40 m. The maximum size of fragment was about 30 cm, as shown in Fig. 11(b). The fragment distributions of W2 and W3 after Test 7 and Test 10 are shown in Fig. 12. It was observed that the maximum size of fragment

was about 60 cm for the strengthened wall specimens. The maximum flying distances of fragments in W2 and W3 were 33 m and 26 m, respectively, which were much less than 40 m of the unstrengthened specimen W1. Therefore, it can be concluded that the two strengthening techniques are able to decrease the number of fragments and retain the flying and scattered fragments. However, the strengthening effectiveness of BFRP strips on the AAC masonry walls under vented gas explosions is not as significant as that on the traditional masonry walls (clay brick and UMC block) under high explosive loads [3,9,10]. There might be several reasons: (1) AAC blocks with low strength are prone to be shattered into small pieces; (2) The loading conditions are not comparable

910


(a) AAC Block

(b) Mortar

Fig. 13. Material tests of AAC block and mortar specimens.

Table 4 Material parameters for AAC block and mortar [37–40]

AAC block Mortar

Density (kg/m3)

Young’s Modulus (MPa)

Poisson’s ratio

Tensile strength (MPa)

Shear strength (MPa)

Compressive yield strength (MPa)

Fracture toughness (N/m)

Shear retention factor

625 2100

530 2211

0.2 0.25

0.7 0.75

1.0 1.1

3 7.5

80 140

0.03 0.03

25 mm

250 mm

56 mm

56 mm

138 mm Fig. 14. BFRP specimen for quasi-static tests.

(a) Shimadzu AGS-300kNX

(b) Epsilon Extensometer

Fig. 15. Setup for quasi-static tensile tests.

between gas explosion and high explosive loads. Gas explosion with higher impulse in this study might result in higher velocity of flying fragment. In addition, it is interesting to observe that the rear-face strengthened specimen W2 had a higher blast resistance capacity in elastic and inelastic response range but the ejected fragments flew a longer distance than the front-face

strengthened specimen W3. It might be caused by the coupling effect between structural response and gas explosion loads. In Test 7 and Test 10, both wall specimens acted as vent covers during gas explosions and the gas explosion loads with similar rising rates are obtained. The strengthened wall specimen W2 with higher blast resistance capacity still resulted in a higher peak pressure of gas

911


(a) Full wall model

(b) Mortar model

(c) BFRP strip model

Fig. 16. Geometric models of strengthened masonry walls.

explosion loads and in turn augments the flying distance of fragments. Therefore, the strengthening of masonry walls under vented gas explosions should be designed with caution for better protective performance. The excessive strengthening might augment the peak pressure of gas explosion loads if no other venting is available. It also should be noted that the accumulative damage may reduce the energy absorbed by structures deformation, mitigate the gas explosion loads due to the coupling of structural response and gas explosion loads and in turn decrease the fragment flying distance. Though the wall specimen W2 experienced more damage than W3 in term of residual displacement, the fragment flying distance of W2 was still longer than that of W3 due to the excessive strengthening of BFRP strips. 3. Numerical simulation 3.1. Material model and its parameters 3.1.1. Masonry material A series of material tests, as shown in Fig. 13, were conducted to determine the parameters of material models for the AAC block

Fig. 18. Comparison of mid-span displacement between numerical results and test data (Test 6).

RC frame

RC frame

Masonry wall

Masonry wall

Basalt fiber strip

Basalt fiber strip

Steel angle

Contact surface

(a) Simplified boundary condition

Contact surface

(b) Detailed boundary condition

Fig. 17. Boundary conditions of two-way wall models.

912


and mortar according to the testing standards i.e. GB/T 119711997 [35] and JGJ/T70-2009 [36]. The dimensions of AAC block specimen and mortar cube specimen were 100 mm and 70.7 mm, respectively. MTS servo-hydraulic testing machine with accurate loading control was used in the material tests. The mean uniaxial compressive strengths of AAC block and mortar were measured as 3.07 MPa and 7.53 MPa, respectively. The anisotropic material model ⁄Mat_96 (MAT BRITTLE DAMAGE) in LS-DYNA was employed to model the AAC blocks and mortar. The compressive yield strength was taken from the material tests. The density and Young’s modulus of blocks and mortar were also acquired from the material tests, as summarized in Table 4. The Poisson’s ratio, tensile strength, shear strength, fracture toughness and shear retention factor were obtained from the Refs. [37–40]. The parameters used in the numerical simulations are listed in Table 4. The erosion criteria were used to simulate the failure of materials. The maximum principal strains at the fail-

ure of block and mortar were defined as 0.04 and 0.02, respectively, which were calibrated with testing data of Test 2 and Test 6 by trial and error approach. 3.1.2. Basalt FRP strip Quasi-static tensile test was carried out to investigate the mechanical properties of unidirectional BFRP strips. The quasistatic specimens were prepared according to the guidelines ASTM 3039-14 [41] and ASTM 7565-10 [42]. As shown in Fig. 14, the specimen was 250 mm long and 25 mm wide. The gauge length was 138 mm long and the aluminum tabs with 56 mm long were glued onto the specimen to avoid possible damage when gripping the specimen. The Shimadzu AGS-X 300 kN Series universal testing machine was used in the tensile tests and the extensometer was installed to measure the deformation of the specimen as shown in Fig. 15. The loading rate of 0.5 mm/min (i.e. 8.3 e6 m/s) was applied for the quasi-static test. Based on the testing data, the

(a) Testing observations

t=200 ms

t=260 ms

t=277 ms

t=300 ms

(b) Displacement contour of wall specimen

t=200 ms

t=260 ms

t=277 ms

(c) Displacement contour of BFRP strips Fig. 19. Failure modes of wall specimen W2 (Test 7).

t=300 ms

913


tensile strength, failure strain and elastic modulus of BFRP strips were calculated as 1642.2 MPa, 0.021 and 77.9 GPa, respectively. The isotropic material model ⁄Mat_3 (MAT PLASTIC KINEMATIC) was used to model the unidirectional BFRP strips. Without defining the kinematic hardening plasticity, BFRP strips were simplified as an isotropic and elastic-brittle material, which is a costeffective method to simulate the BFRP strips. It should be noted that the strain rate effect of BFRP strip is negligible when it is subjected to gas explosions, in which rising stage of overpressure is long. The quasi-static tensile strength, failure strain and elastic modulus of BFRP strips were determined according to the material testing data. 3.2. Modelling strategy and geometric model 3.2.1. Detailed micro model For the detailed micro model, the AAC blocks and mortar are modelled separately and the nodes of the mortar and AAC blocks along their interfaces are merged. The wall models with 120 mm thick, 3.0 m height and 2.0 m width are shown in Fig. 16. The size of AAC blocks is 590 240 120 mm and the thickness of mortar joints is 5 mm. Mesh convergence test was carried out and the 20 mm mesh size for the AAC blocks and BFRP strips and one layer for mortar joint were adopted for the numerical model in the sub-

(a) Layout A

sequent analyses. The default settings in LS-DYNA (i.e. hourglass coefficient of 0.1) were used for hourglass control. Gravity was not included in the numerical model. Solid 164 (with element formulation of constant stress solid element) was used for solid elements of AAC and mortar. Shell 163 (with element formulation of Belytschko-Tsay) was used for angle irons and BFRP strips. The finite element model had a total of 120,078 solid elements and 7920 shell elements for the wall and BFRP strips, respectively. 3.2.2. Contact surface The bonding between BFRP strips and AAC masonry walls was simulated as a tie-break contact in the numerical model. In LSDYNA, the tie-break contact is governed by the stress-based failure criterion as follows [39],

2 2 jrn j jss j þ P1 fn fs

ð1Þ

where rn and ss represent the normal stress and shear stress on the contact surface respectively; f n and f s represent the normal failure stress and shear failure stress, respectively. Actually, the bonding and bond-slip behavior between BFRP strips and AAC masonry walls is complex and determined by lots of factors, such as the strength of materials, quality of specimen and even the strain rate during gas explosions and so on. This method is just a simplification

(b) Layout B

(c) Layout C

Fig. 20. Description of different strip layouts.

(a) 35 kPa

(b) 40kPa

Fig. 21. Displacement of walls with different strengthening layouts.

914


of the reality. In this study, the static and dynamic coefficients of friction were 0.7. The normal failure stress f n and shear failure stress f s were 42 MPa and 2 MPa, respectively. The above parameters were acquired from trial calculations and the test report of resin provided by the supplier and validated by the testing data of Test 6. 3.2.3. Simplification of boundary conditions Properly modelling boundary condition is critical to achieve reasonable numerical results. Two kinds of boundary conditions, i.e. simplified and detailed boundary conditions, were considered

in the simulations. In the simplified boundary condition, as shown in Fig. 17(a), the RC frame was included in the model and the interface between the RC frame and the masonry wall was simulated as a tie-break contact. Since no damage of RC frames was observed in the field tests, the RC frames were simplified as elastic solid elements and no reinforcement was used to improve the computational efficiency. The boundary conditions were assumed as fixed in three directions. This method for the unstrengthened wall has been verified by the testing data of Test 2 in the previous study [25]. The static and dynamic coefficients of friction for the top and bottom edges were defined as 0.4 and the normal failure stress

(a) Layout A (40kPa)

(b) Layout B (40kPa)

(c) Layout C (40kPa) Fig. 22. Displacement contour of masonry wall strengthened with different BFRP strip layouts under 40 kPa gas explosion load.

915


f n and shear failure stress f s were set as zero. As to the left and right boundaries, the static and dynamic coefficients of friction were also set as 0.4. The normal failure stress f n and the shear failure stress f s were defined as 0.5 MPa. For the strengthened walls, the predicted peak displacement is about two times higher than that from testing by using the simplified boundary conditions, which means that the simplified method underestimates the effectiveness of boundary condition of the strengthened wall specimen significantly. Therefore, a more realistic boundary condition was used to simulate the strengthened wall specimens. As shown in Fig. 17(b), the angle irons were modelled along the boundary. The tie-break contact was defined between the angle iron and the masonry walls and between BFRP and angle irons. The bonding between angle irons and concrete was also simulated as tie-break contact. The arrangement of angle irons was the same as experimental setup. The parameters were acquired from trial calculations and validated by the testing data of Test 6. 3.3. Model validation Using the actual recorded pressure time history as input, the responses of the above wall model were calculated and Test 6 and Test 7 were used to calibrate the numerical model. Fig. 18 compares the numerical prediction and test data of the specimen W2 in Test 6. The predicted period of vibration and the peak value of mid-span displacement agree well with the testing data. In addi-

tion, the predicted damage mode of W2 was also compared with the testing results of Test 7 as shown in Fig. 19. It is found that the damage characteristics of wall specimen at the early stage and the yield line could be well simulated. However, this model cannot well predict severe shattering of AAC blocks at the latter stage of the gas explosion loading and debris flying. Therefore, it can be concluded that the numerical model developed in this study can give reasonably good prediction of structural response of strengthened AAC masonry walls against vented gas explosions, but it cannot well predict the subsequent debris range, velocity and distribution. 3.4. Parametric studies Parametric studies were carried out to investigate the effects of strip layouts, strip thickness and fiber types on the performance of strengthened AAC masonry walls under vented gas explosion. The masonry wall used for parametric studies was 3.0 m in height and 2.0 m in width. The dimension of AAC blocks was 590 mm 240 mm 120 mm and the thickness of the mortar was 5 mm. A single-peak overpressure wave with a triangular shape was adopted for the parametric studies. The rising rate and decreasing rate of the pressure wave were set as 400 kPa/s and 1000 kPa/s based on the testing data. The structural responses including maximum displacement and failure mode are compared herein.

(a) 35 kPa

(b) 40 kPa

(c) 45 kPa Fig. 23. Comparisons of displacement of walls strengthened with different strip thicknesses.

916


Table 5 Material properties of BFRP, GFRP and CFRP [43]

Failure stress (MPa) Young’s Modulus (GPa) Failure strain (%)

Basalt fiber

Glass fiber (E)

Carbon fiber

1642 77.9 2.1

3407 72.5 4.7

4100 212 1.74

3.4.1. Effect of different strip layouts The effect of strip layouts on the response of strengthened AAC masonry walls is compared and discussed in this section. Three strip layouts for strengthening AAC masonry walls, i.e. Layout A, Layout B and Layout C are presented in Fig. 20. Two layers (i.e. 0.24 mm) of BFRP strips are used for Layout C to keep the total amount of BFRP strips applied to the wall the same for the three layouts. The displacement time histories at the center of masonry walls strengthened with different strip layouts under two peak pressures of 35 kPa and 40 kPa are shown in Fig. 21. It is found that the specimen with Layout C experiences the largest displacement, followed by Layout A and Layout B. When the peak pressure is 35 kPa, the maximum displacements of masonry walls with Layout A, Layout B and Layout C are 54.3 mm, 46.7 mm and 106.3 mm, respectively. Due to the characteristic of gas explosion loads (i.e. uniform distribution on the surface of structures), midspan area of two-way masonry walls experiences the maximum

moment in the whole structure. In Layout B, the BFRP strips are distributed more evenly around the mid-span area, which changes the equivalent section area and in turn improves the stiffness of the whole structures. When the peak pressure is 40 kPa, similar observations on the displacement histories of the walls with three strengthening measures can be drawn. As shown in Fig. 22(b), the concentrated distribution of BFRP strips in Layout B results in the minor damage of masonry walls and no BFRP strip rupture occurs. While in Layout A and Layout C, the rupture of strips occurs in the high stress regions (i.e. yield line). Distinct damage of masonry walls with layout A and C can be found in Fig. 22(a) and (c). It can be concluded that proper arrangements of strengthening layouts can more effectively enhance the resistance capacity of masonry walls. Two-way AAC masonry walls should be strengthened in two directions and the concentrated layout can improve blast resistance capacity of walls more effectively since the primary response mode of masonry wall under gas explosion is flexural bending. However, it should be pointed out that on the contrary, the distributed strengthening layout is usually more beneficial to prevent local damage of walls under high explosive loads. 3.4.2. Effect of strip thickness The effect of strip thickness on the response of masonry walls with Layout A is examined under three gas explosion loads with peak overpressure of 35 kPa, 40 kPa and 45 kPa, respectively. The masonry walls strengthened with three strip thicknesses i.e.

(a) 35 kPa

(b) 40 kPa

(c) 45 kPa Fig. 24. Comparison of displacement of walls strengthened with different fibers.


0.12 mm, 0.24 mm and 0.36 mm are considered. The displacement histories at the center of masonry walls are shown in Fig. 23. It is obvious and expected that increasing the BFRP strip thickness reduces the wall responses. When the peak pressure is 45 kPa, the masonry wall strengthened with 0.12 mm-thick strips fails while the other two walls survive the gas explosion loads. The strengthening strips contribute to the equivalent section area and the stiffness of the strengthened walls. With the increase of strip

917

thickness, the blast resistance capacity of masonry walls is enhanced. 3.4.3. Effect of different fibers Three types of fiber strips (i.e. BFRP, CFRP and GFRP) are used to strengthen masonry walls with Layout A in this section. The material properties of three fiber strips are summarized in Table 5. The displacement time histories of masonry walls strengthened with

(a) CFRP

(b) GFRP

(c) BFRP Fig. 25. Displacement contours of walls strengthened with different fibers under 45 kPa gas explosion load.

918


three materials with thickness of 0.12 mm under peak pressure of 35 kPa, 40 kPa and 45 kPa are shown in Fig. 24. As shown the peak displacement of the wall strengthened with CFRP is the smallest, while that of the BFRP and GFRP are comparable when the peak explosion load is 35 and 40 kPa. This is because CFRP strip used has the largest modulus and strength, while BFRP and GFRP have comparable modulus, but GFRP has higher strength, as given in Table 5. As shown, the peak displacements of masonry walls strengthened with BFRP, CFRP and GFRP under the peak pressure of 35 kPa is 54.3 mm, 42.4 mm and 55.4 mm, respectively. When the peak pressure increases to 40 kPa, the peak displacement of masonry wall with CFRP strips is 58.5 mm, with BFRP strips is 90.60 mm, and with GFRP strips is 88.58 mm. The masonry wall with BFRP strips fails when the peak pressure is 45 kPa, but the masonry wall with CFRP and GFRP survive the load because of the high strength of the two FRPs. Fig. 25 shows the damage modes of masonry wall and strips at the time of 0.146 s. As shown in Fig. 25(b), although the GFRP does not rupture due to its good ductility, the damage of masonry wall is more severe than the masonry wall strengthened by CFRP strips. Therefore, appropriate strengthening materials should be carefully selected depending on the strengthening performance requirement of structures. 4. Conclusions In this study, the performance of BFRP strip strengthened AAC masonry walls under vented gas explosions were investigated through full-scale field tests and numerical simulations. The conclusions are summarized as follows: 1) The unstrengthened and strengthened AAC masonry wall experienced typical two-way flexural failure along with fragmentation of AAC blocks due to the high peak overpressure and the low strength of AAC material. The failure mode of BFRP strips for rear-face and front-face strengthened specimens were tensile rupture on the edges and mid-span area and debonding failure along the boundaries, respectively. The wall specimen strengthened at rear-face had the best resistance capacity and reduces the response of wall specimens by 87.8% in terms of maximum displacement against vented gas explosions because little debonding of BFRP strips improved the combination between strips and AAC masonry and in turn enhanced bending-resistant stiffness and capacity of walls. 2) The existence of BFRP strips can decrease the number of fragments and retain the flying and scattered fragments. The rear-face strengthened wall with higher blast resistant capacity resulted in higher peak pressure acting on the wall and hence led to longer flying distance of fragments. Therefore, the strengthening of masonry walls and other structure types against vented gas explosions should be designed with caution if there are no other vents to relief the gas pressure. 3) The numerical model of rear-face strengthened walls was developed and validated. Parametric studies show that increasing strip thickness and choosing an appropriate layout can significantly reduce the wall responses. The CFRP strips show better strengthening effectiveness than BFRP and GFRP strips because of its higher modulus and strength.

Acknowledgement The authors acknowledge the financial support from the National Key Research and Development Program of China (Nos. 2016YFC0305200, 2015CB058000) and the Australian Research

Council (Grant No. LP150100259). The first author would also like to acknowledge the financial support from the National Natural Science Foundation of China (No. 51622812) and the China Scholarship Council, as well as good advices from Dr. Xiaoqin Li. References [1] Hao H, Hao Y, Li J, Chen W. Review of the current practices in blast-resistant analysis and design of concrete structures. Adv Struct Eng 2016;19:1193–223. [2] Lees F. Lees’ Loss prevention in the process industries: Hazard identification, assessment and control. Oxford, UK: Elsevier Butterworth-Heinemann; 2012. [3] Chen L, Fang Q, Fan J, Zhang Y, Hao H, Liu J. Responses of masonry infill walls retrofitted with CFRP, steel wire mesh and laminated bars to blast loadings. Adv Struct Eng 2014;17:817–36. [4] Baylot JT, Bullock B, Slawson TR, Woodson SC. Blast response of lightly attached concrete masonry unit walls. J Struct Eng 2005;131:1186–93. [5] Myers JJ, Belarbi A, El-Domiaty KA. Blast resistance of FRP retrofitted unreinforced masonry (URM) walls with and without arching action. Masonry Soc J 2004;22:9–26. [6] Oswald C, Wesevich J. Structural retrofits to increase the blast capacity of roofs and masonry walls. In: Proceedings of the SAME national symposium on comprehensive force protection. Charleston, USA. [7] Crawford JE. Modeling blast-resistant protection systems composed of polymers and fabric. Model Performance Eng Struct Mater II 2002;I:61–75. [8] Crawford J, Morrill K. Retrofit techniques using polymers and FRPs for preventing injurious wall debris. In: Fibre-reinforced polymer reinforcement for concrete structures: (In 2 Volumes). World Scientific; 2003. p. 1199–08. [9] Muszynski LC, Purcell MR. Use of composite reinforcement to strengthen concrete and air-entrained concrete masonry walls against air blast. J Compos Constr 2003;7:98–108. [10] Davidson JS, Porter JR, Dinan RJ, Hammons MI, Connell JD. Explosive testing of polymer retrofit masonry walls. J Performance Constr Facilities 2004;18:100–6. [11] Urgessa GS, Maji AK. Dynamic response of retrofitted masonry walls for blast loading. J Eng Mech 2009;136:858–64. [12] Maji AK, Brown JP, Urgessa GS. Full-scale testing and analysis for blastresistant design. J Aerospace Eng 2008;21:217–25. [13] Hrynyk T, Galati N, Myers JJ. An analytical approach to predict the out-of-plane behavior of FRP retrofitted masonry infill walls with arching action. In: Proc., 3rd Int. Conf. Advanced Composites in Construction; 2007. p. 320–27. [14] Hrynyk TD, Myers JJ. Out-of-plane behavior of URM arching walls with modern blast retrofits: Experimental results and analytical model. J Struct Eng 2008;134:1589–97. [15] Tan KH, Patoary M. Blast resistance of FRP-strengthened masonry walls. I: Approximate analysis and field explosion tests. J Compos Constr 2009;13:422–30. [16] Irshidat M, Al-Ostaz A, Cheng A, Mullen C. Blast vulnerability evaluation of concrete masonry unit infill walls retrofitted with nano particle reinforced polyurea: modelling and parametric evaluation. In: Ames D, Droessler TL, Hoit M, editors. Proc Struc Cong, Las Vegas, NV; 2011. p. 2126–41. [17] Irshidat M, Al-Ostaz A, Cheng A-D, Mullen C. Nanoparticle reinforced polymer for blast protection of unreinforced masonry wall: laboratory blast load simulation and design models. J Struct Eng 2010;137:1193–204. [18] Chen L, Fang Q, Jiang C, Fan J, Hao H. Response and damage level of confined masonry walls to blast. Disaster Adv 2013;6:380–94. [19] Elsanadedy H, Al-Salloum Y, Al-Zaheri Z, Alsayed S, Abbas H. Behavior and design aspects of FRP-strengthened URM walls under out-of-plane loading. J Compos Constr 2016;04016048. [20] Alsayed SH, Elsanadedy H, Al-Zaheri ZM, Al-Salloum YA, Abbas H. Blast response of GFRP-strengthened infill masonry walls. Constr Build Mater 2016;115:438–51. [21] Wang J, Ren H, Wu X, Cai C. Blast response of polymer-retrofitted masonry unit walls. Compos B Eng 2016. [22] Wittmann FH, Balkema A. Advances in autoclaved aerated concrete, Citeseer; 1992. [23] Narayanan N, Ramamurthy K. Structure and properties of aerated concrete: a review. Cement Concr Compos 2000;22:321–9. [24] Yankelevsky DZ, Avnon I. Autoclaved aerated concrete behavior under explosive action. Constr Build Mater 1998;12:359–64. [25] Li Z, Chen L, Fang Q, Hao H, Zhang Y, Chen W, et al. Study of autoclaved aerated concrete masonry walls under vented gas explosions. Eng Struct 2017;141:444–60. [26] Sim J, Park C, Moon DY. Characteristics of basalt fiber as a strengthening material for concrete structures. Compos B Eng 2005;36:504–12. [27] Wu Z, Wang X, Wu G. Advancement of structural safety and sustainability with basalt fiber reinforced polymers, CICE2012. Rome 2012;13:15–29. [28] Chen W, Hao H, Jong M, Cui J, Shi Y, Chen L, et al. Quasi-static and dynamic tensile properties of basalt fibre reinforced polymer. Compos B Eng 2017. [29] Basßtürk S, Uyanık H, Kazancı Z. An analytical model for predicting the deflection of laminated basalt composite plates under dynamic loads. Compos Struct 2014;116:273–85. [30] Ministry of Construction of the PR China. Code for design of masonry structures (GB50003-2011). Beijing, China: China Architecture and Building Press; 2011.

Z. Li et al. / Engineering Structures 152 (2017) 901–919 [31] Bao Q, Fang Q, Zhang Y, Chen L, Yang S, Li Z. Effects of gas concentration and venting pressure on overpressure transients during vented explosion of methane–air mixtures. Fuel 2016;175:40–8. [32] Bao Q, Fang Q, Yang S, Zhang Y, Xiang H, Chen L, et al. Experimental investigation on the deflagration load under unconfined methane-air explosions. Fuel 2016;185:565–76. [33] Li Z, Chen L, Fang Q, Hao H, Zhang Y, Xiang H, et al. Experimental and Numerical Study of Unreinforced Clay Brick Masonry Walls Subjected to Vented Gas Explosions. Int J Impact Eng 2017;104:107–26. [34] Albert ML, Elwi AE, Cheng JR. Strengthening of unreinforced masonry walls using FRPs. J Compos Constr 2001;5:76–84. [35] General Administration of Quality Supervision Inspection and Quarantine of PR China, Test methods for mechanical property of aerated concrete (GB/T 11971–1997), Standards Press of China, Beijing, China; 1998. [36] Ministry of Construction of the PR China. Standard for test method of performance on building mortar (JGJ/T70-2009). Beijing, China: China Architecture and Building Press; 2009.

919

[37] Argudo JF, Fern J, Argudo O. Evaluation and synthesis of experimental data for autoclaved aerated. Concrete 2003. [38] Govindjee S, Kay GJ, Simo JC. Anisotropic modelling and numerical simulation of brittle damage in concrete. Int J Numer Meth Eng 1995;38:3611–33. [39] Halquist J. LS-DYNA keyword user’s manual version 971. Livermore, CA: Livermore Software Technology Corporation; 2007. [40] Ferretti D, Michelini E, Rosati G. Mechanical characterization of autoclaved aerated concrete masonry subjected to in-plane loading: Experimental investigation and FE modeling. Constr Build Mater 2015;98:353–65. [41] ASTM D 3039-14. Standard test method for tensile properties of polymer matrix composite materials; 2014. [42] ASTM D 7565-10. Standard test method for determining tensile properties of fiber reinforced polymer matrix composites used for strengthening of civil structures; 2010. [43] Singha K. A short review on basalt fiber. Int J Textile Sci 2012;1:19–28.