ASCE 2020 - Strengthening of In-Plane and Out-of-Plane Capacity of Thin Clay Masonry Infills Using Textile and Fiber-Reinforced Mortar

Strengthening of In-Plane and Out-of-Plane Capacity
of Thin Clay Masonry Infills Using Textile-

and Fiber-Reinforced Mortar
Massimiliano Minotto, S.M.ASCE1; Nicolò Verlato, Ph.D.2; Marco Donà, Ph.D., M.ASCE3;
and Francesca da Porto, Ph.D.4
Downloaded from ascelibrary.org by CORNELL UNIV LIBRARIES on 08/20/20. Copyright ASCE. For personal use only; all rights reserved.
Abstract: This paper presents an overview of the experimental results obtained by combined in-plane/out-of-plane (IP/OOP) tests carried
out on reinforced concrete (RC) frames infilled with thin clay masonry walls. After preliminary characterization tests on building materials,
combined IP/OOP tests on eight full-scale, one-bay, one-story infilled RC frames were carried out. Three external strengthening solutions
were investigated considering three types of lime-based plasters, reinforced by means of dispersed fibers and/or bidirectional basalt meshes,
applied on both sides of each wall. The experimental results of the tested frames are presented, discussed, and compared to evaluate the per-
formance and the effectiveness of each strengthening solution. DOI: 10.1061/(ASCE)CC.1943-5614.0001067. © 2020 American Society of
Civil Engineers.
Author keywords: Clay masonry infills; Textile-reinforced mortars (TRM); Fiber-reinforced mortars (FRM); Experimental tests;
Combined in-plane/out-of-plane behavior; Infilled RC frames.
Introduction damage of the infill, induced by the RC frame deformation, yields

to a reduction of the OOP strength, thus increasing the probability
The construction of reinforced concrete (RC) frames with thin clay of infill out-of-plane ejection (Donà et al. 2017, 2019). As a conse-
masonry infill walls represents a widespread building practice in quence, the seismic vulnerability of infill walls increases consider-
many European countries and worldwide (INSYSME 2014). These ing their combined IP/OOP response. The damage observed
types of infill walls, also used as internal partition walls and classified after recent seismic events is a demonstration of this problem,
as nonstructural elements, are often neglected in the current design e.g., L’Aquila 2009, Italy (Bazzurro et al. 2009; Ricci et al. 2011;
procedures. During earthquakes, the seismic action affects these non- Braga et al. 2011); Lorca 2011, Spain (Hermanns et al. 2014;
structural walls, causing relevant economic losses (Chiozzi and De Luca et al. 2014); Simav 2011, Turkey (Doğ angün et al. 2013);
Miranda 2017; De Martino et al. 2017; Di Ludovico et al. 2017a, Emilia 2012, Italy (Penna et al. 2014; Verderame et al. 2014);
b) and loss of building functionality (Dolce and Goretti 2015) in me- Kefalonia 2014, Greece (Manos et al. 2015); and Central Italy 2016
dium to severe earthquakes. Among the most relevant seismic events, (Masi et al. 2017, 2019; Fragomeli et al. 2017; De Risi et al. 2018).
it is possible to mention Loma Prieta (1989) and Northridge (1994) in From the second-half of the 20th century, several experimental
California where extensive damage to nonstructural elements, in campaigns (among the others, Thomas 1952; Polyakov 1960;
buildings with slight structural damage, accounted for a total amount Holmes 1961; Stafford-Smith 1962; Mainstone and Weeks 1970;
of about $30 million (Vicente et al. 2010). Mainstone 1971; Fiorato et al. 1970; Zarnic and Tomazevic
The out-of-plane (OOP) load can induce partial or total collapse 1988; Mehrabi et al. 1994, 1996; Crisafulli 1997; Anil and Altin
of the infill walls, which may occur also for low seismic intensities. 2007; Blackard et al. 2009; Kakaletsis et al. 2011; Bergami and
The brittle behavior of clay masonry infill walls leads to an uncon- Nuti 2015; Gazic and Sigmund 2016) focused on the IP behavior
trolled collapse that may occur without any warning, representing a of infill walls, investigating the formation of the strut mechanism,
serious hazard for human lives. Furthermore, the in-plane (IP) the infill–frame (IF) interaction, and the IF overall response.
Few experimental studies focused on the OOP response of un-
1 reinforced infill walls (Drysdale and Essawy 1988; Dawe and
Ph.D. Student, Dept. of Civil, Architectural and Environmental Engi-
neering, Univ. of Padova, Via Marzolo 9, 35131 Padova, Italy. Email:
Seah 1989; Dafnis et al. 2002; Tu et al. 2010), whereas a consistent
[email protected] number of works developed and experimentally investigated vari-
2
Research Fellow, Dept. of Civil, Architectural and Environmental ous strengthening solutions to improve the OOP performance of
Engineering, Univ. of Padova, Via Marzolo 9, 35131 Padova, Italy. masonry infill walls, to avoid brittle collapse. Tumialan et al.
Email: [email protected] (2003), Saatcioglu et al. (2005), El-Dakhakhni et al. (2006), Kalali
3
Post-Doc, Earthquake Engineering Research & Test Center, (2012) and Lunn and Rizkalla (2011) introduced the application of
Guangzhou Univ., Guang Yuan Zhong Rd. 248, 510405 Guangzhou, fiber-reinforced polymer (FRP) sheets eternally bonded with or-
China (corresponding author). Email: [email protected] ganic matrixes (or epoxy resins). The reduced thickness, high stiff-
4
Full Professor, Dept. of Geosciences, Univ. of Padova, Via Gradenigo
ness, and relative ease of application give to FRP a huge range of
6, 35131 Padova, Italy. Email: [email protected]
Note. This manuscript was submitted on December 24, 2019; approved
application; nevertheless, to improve the physical and mechanical
on May 27, 2020; published online on August 18, 2020. Discussion period compatibility with masonry, further retrofitting techniques, based
open until January 18, 2021; separate discussions must be submitted for in- on textile-reinforced mortar (TRM), were developed.
dividual papers. This paper is part of the Journal of Composites for Con- The most common reinforcing textiles are bidirectional compos-
struction, © ASCE, ISSN 1090-0268. ite meshes made of steel or natural/mineral fibers (e.g., flax, hemp,
© ASCE 04020059-1 J. Compos. Constr.
J. Compos. Constr., 2020, 24(6): 04020059

basalt), carbon, glass, and polyester, applied with inorganic ma- hysteresis loops, IP damage distribution, OOP force–displacement
trixes directly on the masonry surface. The effectiveness of curves and, finally, in terms of OOP strength reduction along with
TRMs applied on non-load-bearing masonry infill walls has the increase of IP interstory drift. Lastly, a discussion on possible de-
been experimentally confirmed by Papanicolaou et al. (2007), sign formulations for this type of strengthening solutions is given.
Kakaletsis et al. (2011), Vasconcelos et al. (2012), Martins
et al. (2015), Koutas and Bournas (2019), and De Risi et al.
(2019). An extended experimental campaign was carried out
by Valluzzi et al. (2014) to compare the FRP and TRM strength-
Description of the Strengthened Specimens
ening solutions and evaluate their effectiveness.
The seismic performance of three FRM/TRM strengthening solu-
Only a relatively recent series of studies evaluated the interac-
tions were experimentally investigated through combined IP/OOP
tion of the out-of-plane behavior of masonry infill walls with the
tests carried out at the Laboratory of the University of Padova.
concurring effects of in-plane damage. In particular, the earlier
The specimens were full-scale RC frame (one-bay and one-story)
works of Angel et al. (1994), Flanagan and Bennett (1999),

Calvi and Bolognini (2001), and Furtado et al. (2016) were the fully infilled with masonry panels. Each infill wall was made of
first to deal with this issue. They demonstrated how the OOP ca- clay bricks with horizontal holes (60% holes percentage) and dimen-
pacity degrades as the IP displacement demand increases, and in sions of 250 × 120 × 250 mm3 (length × width × height). According
some cases, they also tested some reinforcement/strengthening to EN 1996-1-1 (CEN 2005), clay units were of Group 4. The me-
techniques, such as horizontal joint reinforcement or external ap- chanical properties of the blocks were stated by the manufacturer ac-
plication of ferrocement jacketing. cording to EN 771-1 (CEN 2015); the average compressive
More recently, also Ricci et al. (2018) dealt with the combined strengths, parallel and orthogonal to the holes, were equal to 8.0
IP/OOP behavior of weak clay infill walls, studying the response of and 2.0 N/mm2, respectively. The masonry walls were built with
original infill walls in unreinforced conditions. Some other works both horizontal (12 mm thick) and vertical (8.8 mm thick) joints,
concentrated on the use of horizontal joint reinforcement or ferroce- fully filled with mortar of declared class M5 (EN 998-2, CEN
ment or other steel devices (Pereira et al. 2011; Silva et al. 2016), 2016b). Each infill wall had a thickness of 150 mm considering
whereas others also tested strengthened solutions based on the ex- that it was rendered on both façades with a 15 mm thick plaster
ternal application of fiber-reinforced or textile-reinforced mortars layer. A total number of eight specimens, strengthened with three re-
(da Porto et al. 2015; Ismail et al. 2018; Sagar et al. 2019). inforcing solutions identified as F, FB, and RBB, were experimen-
Lastly, a more recent series of works studied the combined IP/ tally investigated. Fig. 1 shows the details of each strengthening
OOP behavior of more robust, eventually internally reinforced with solution, and Fig. 2 shows some details of the construction phases.
steel bars, clay masonry infill walls (da Porto et al. 2013; Vintzileou • Strengthening type F (fiber plaster): characterized by a 15 mm
et al. 2016; Palieraki et al. 2018), and a consistent number of works layer of FRP containing alkali-resistant (AR) and corrosion-
has focused on the development of innovative solutions to increase resistant glass fibers with a high content (above 16%) of zirconium
the infill wall ductility and/or decrease the infill wall/RC frame inter- oxide. The plaster is of class CS III (EN 998-1, CEN 2016a), with
action (Mohammadi and Akrami 2010; Mohammadi et al. 2011; Baio a characteristic compressive strength at 28 days of 5.0 MPa, made
Dias et al. 2014; Preti et al. 2015; da Porto et al. 2016; Verlato et al. of natural hydraulic lime (NHL 3.5) with a quite fine grading
2016; Morandi et al. 2018; Marinković and Butenweg 2019). (0÷1.8 mm). The plaster was applied directly on the raw masonry
These researches concerned different types of masonry infill surface, without any reinforcing textile mesh.
walls to be used for new constructions. In this context, the EN • Strengthening type FB (fiber plaster and basalt mesh): char-
1998-1 (CEN 2004) simplistically suggests to pay particular atten- acterized by the same type of plaster used for strengthening
tion to infill walls characterized by a slenderness ratio greater than type F. The only difference is that a balanced biaxial mesh,
15, and to adopt measures to improve both IP and OOP behavior of made of alkali-resistant basalt fibers, was embedded into
the infill walls, such as light wire meshes on one face of the wall, the two external plaster layers, each 15 mm thick. The basalt
wall ties fixed to the columns, and concrete post and belts. The net, produced with square mesh size 22 × 22 mm 2, is charac-
use of light plaster meshes on two faces of the infill walls, intercon- terized by an equivalent thickness of 0.023 mm, a character-
nected through the masonry thickness and to the frame elements, or istic tensile strength greater than 1,600 MPa, a Young’s
the use of bed-joint reinforcement, is also suggested by the more modulus of 56,000 MPa, a minimum value of ultimate strain
recent Italian Standard (Circ. 2019 no. 7, MIT 2019), as a defensive of 2.5%, and a weight of 0.12 kg/m2; the failure load is there-
measure against OOP collapse. However, recent earthquakes dem- fore greater than 30 kN/m (information certified by the man-
onstrated the ineffectiveness of some of these measures (e.g., the ufacturer). Like the previous strengthening type F, the plaster
concrete post and belts, Vintzileou and Palieraki 2007), and in was applied on the raw masonry surface.
some other cases, these measures can be hardly applied for improv- • Strengthening type RBB (render, basalt mesh, and helicoidal
ing existing masonry infill walls. In addition, both codes do not ex- bars): composed of a first plaster layer (shortly named R) of
plicitly refer to the problems of IP/OOP interaction. class CS II and thickness 10 mm, with quite fine grading
Therefore, this research work focuses on the experimental char- (0÷1.4 mm) and made of pure NHL 3.5, directly applied on
acterization of three FRM/TRM strengthening solutions, designed the raw masonry surface to simulate the presence of a preexist-
for strengthening thin masonry infill walls in existing buildings. ing, low-quality plaster. Subsequently, a second plaster layer of
The reinforcement was externally applied on both sides of the ma- class CS IV and thickness 5 mm was applied on the first one
sonry panel. The experimental campaign consisted of tests for char- with interposition of the previously mentioned balanced biaxial
acterizing the building materials and combined IP/OOP tests on basalt fiber mesh. The second plaster type is a smoothing–
real-scale specimens with strengthened infill walls. Eight full-scale, levelling plaster type NHL 3.5 with the same grading of the
one-bay, one-story infilled RC frames were tested by means of initial one. In addition, the strengthening was fixed to the RC
quasi-static cyclic procedures to evaluate the damage progression frame top beam by means of 8 Ø8 mm helicoidal stainless
due to IP cyclic displacements and the effect on the OOP response. steel ties, which are made of steel type AISI316 and have a char-
The three solutions are compared in terms of envelope curves of the acteristic tensile failure load greater than 12.7 kN.
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 1. Strengthening solutions (F, FB, RBB).
Fig. 2. Application of RBB. (Images by Nicolò Verlato.)
The basalt mesh (used in the FB and RBB solutions) is supplied Table 1 summarizes the specimens and the type of test that was
in 1 m-wide rolls; it was embedded into the plaster layer with a ver- carried out. For each type of strengthening solution, one specimen
tical strip arrangement and an overlap width of 20 cm, without was directly (without previous in-plane damage) tested out-of-plane,
adopting any anchoring device to the RC frame besides the helicoi- by monotonic loading on eight points. Other three specimens were
dal bars in the RBB solution. These helicoidal bars were dry in- in-plane tested imposing incremental cyclic displacements at the
stalled, with an inclination of 45° and a penetration depth in the frame top beam, until 1.2% drift, and only after this IP damaging,
beam of about 5 cm. After inserting the bars into the beam, they they were tested out-of-plane until collapse. The same test procedure
are bent to be embedded into the second plaster layer. Compared was applied, only for strengthening type FB and RBB, also on spec-
with the use of traditional bars, helicoidal bars are more convenient, imens tested until 0.5% drift.
as they do not have to be grouted. The name of the specimens in Table 1 is given by the type of
These bars act only as further restraints, avoiding the rigid rota- strengthening solution (F, FB, and RBB) and by the maximum in-
tion of the infill panel and thus allowing a better activation of the plane drift θmax to which the sample was cyclically tested before the
resistant OOP mechanism (vertical arching and/or vertical bend- out-of-plane monotonic test.
ing). These bars were used in the RBB reinforcement only, which
being applied on an existing (poor) plaster layer, is the one that
needs the greater aid to improve the OOP capacity of the masonry Material Characterization
panel, as well as to reduce its IP damage. In the other solutions (F
and FB), which were applied on the masonry and column surface Flexural and compression tests on plaster specimens and on ma-
directly, it was decided to rely on the new plaster adherence sonry assemblages were performed to characterize the mechanical
only. In this case, the main aim was to check whether the reduced properties of the masonry substrate and of the strengthening mate-
time- and cost-consuming strengthening solutions F and FB could rials. Other common tests, such as compression tests on clay units,
still increase the OOP capacity of the infill walls, although not sig- were not carried out, as the same units had already been tested dur-
nificantly improving the fragile OOP collapse mode. ing a previous experimental campaign (Valluzzi et al. 2014).
J. Compos. Constr., 2020, 24(6): 04020059

Table 1. Experimental specimens
Specimen Preexisting plaster Strengthening Test type
F.00 — Fiber plaster of class CS III OOP
F.12 IP (θmax = 1.2%) + OOP
FB.00 — Fiber plaster of class CS III + basalt fiber mesh OOP
FB.05 IP (θmax = 0.5%) + OOP
FB.12 IP (θmax = 1.2%) + OOP
RBB.00 Class CS II Class CS IV + basalt fiber mesh + helicoidal stainless steel ties OOP
RBB.05 IP (θmax = 0.5%) + OOP
RBB.12 IP (θmax = 1.2%) + OOP
(a) (b) (c)
Fig. 3. (a) Flexural; (b) compressive; and (c) elastic modulus characterization tests on plaster specimens.
The preparation of the samples and the test procedures were in ac- Table 2. Mechanical properties of plasters (CoV in brackets)
cordance to specific standard requirements.
Plaster Flexural strength Compressive strength Elastic modulus
type [N/mm2] [N/mm2] [kN/mm2]
Tests on Plasters M 3.9 (23.3%) 12.8 (29.4%) 5.9 (19.7%)
F 2.6 (28.3%) 8.6 (30.4%) 7.4 (16.4%)
Plaster specimens were sampled during the construction of both the R 1.3 (16.1%) 3.6 (22.0%) 4.5 (10.6%)
masonry assemblages to be tested in compression and the infill walls
for the combined IP/OOP tests and were tested after 28 days curing,
according to EN 1015-1 (CEN 1998a) and EN 13412 (CEN 2006b). of 1.0 mm/min. As in the case of the subsequent flexural tests, one
With reference to the strengthening solutions described before, six type of specimen was built with the low-quality plaster type R and
specimens for each mortar type M (CS IV), F (CS III), and R (CS without any type of reinforcement, whereas the other two specimen
II) were tested. According to EN 1015-11/A1 (CEN 2006a), speci- types were made with strengthening types RBB and F.
mens of dimensions 160 × 40 × 40 mm3 were monotonically loaded Table 3 shows a summary of the results. As expected, the com-
with a uniform rate of 50 N/s to evaluate the flexural strength pressive strength and the elastic modulus of the specimens were not
[Fig. 3(a)]. After the failure of the sample, the two portions were significantly affected by the quality of the plaster and the presence
kept for the determination of the compressive strength [Fig. 3(b)] of the basalt mesh.
using the same testing machine of the previous flexural tests and ap- Two types of flexural tests, both having plane of failure parallel
plying a constant loading rate of 200 or 400 N/s until failure. No spe- to the bed joints, were conducted. The first type of test was carried
cific standards are available to conduct tests for the determination of out on specimens having the low-quality plaster that simulate a pre-
the mortars Young’s modulus; thus, the standard for hardened con- existing material, without any reinforcement. In this case, a bending
crete EN 12390-13 (CEN 2013) was considered. Elastic modulus test on a mortar joint was carried out, as larger specimens could have
tests were carried out on one of the three prismatic specimens of collapsed under the effect of the dead-load only. Conversely, in the
each mortar sample only, measuring on each face of the specimen case of specimens with a strengthening system applied, an ordinary
the vertical strains during several loading cycles [Fig. 3(c)]. Table 2 four-point bending test was carried out.
lists the average values of flexural strength, cubic compressive • Mortar joint bending tests: nine masonry assemblages with
strength, and elastic modulus of the tested plaster. plaster type R (CS II) and dimensions 250 × 515 × 150 mm3
[Fig. 5(a)] were tested applying the monotonic load directly
on the mortar joint. The specimens were placed on two steel
Tests on Masonry Assemblages
supports with a diameter of 40 mm, spaced by 415 mm, and a
To characterize the basic mechanical properties of the masonry third steel roller of the same geometry was located centrally
constituting the infill wall systems analyzed in this work, 12 ma- for loading the specimen [Fig. 5(b)] under displacement control
sonry specimens with a shape ratio b:h = 1:1 (dimensions 775 × (velocity rate of 0.5 mm/min). The typical failure mode oc-
780 mm2, Fig. 4) were tested in compression following the stan- curred along the mortar joint, as shown in Fig. 5(c). The average
dard EN 1052-1 (CEN 1998b). The load was applied orthogonally maximum nondimensional bending moment per meter length of
to the masonry unit holes, in displacement control, with a velocity specimens was 1.1 kN · mm/mm.
J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)
Fig. 4. Compression tests: (a) geometry with dimensions in mm; (b) test setup; and (c) failure mode.
Table 3. Compression test results (CoV in brackets)

Specimen Max. load [kN] Compressive strength [N/mm2] Elastic modulus [kN/mm2]
R 293 2.7 (12.2%) 4,198 (12.7%)
RBB 358 2.9 (14.0%) 4,777 (22.2%)
F 325 2.8 (6.1%) 4,752 (37.5%)
(a) (b) (c)
Fig. 5. Flexural tests on mortar joints: (a) geometry with dimensions in mm; (b) test setup; and (c) failure mode.
• Four-point bending tests: four specimens of dimensions 1,300 × by means of dispersed glass fibers (F) provides significantly different
390 mm2 [Fig. 6(a)] for each type of strengthening solution results. Indeed, despite the test results may have been partly influ-
(RBB, FB, F) were tested with plane of failure parallel to the bed enced by the different test set-ups used, the obtained results reflect
joint (EN 1052-2, CEN 1999). During the application of the mono- the difference in the basic mechanical properties of the two coating
tonic load, the deflection was measured by six displacement sensors materials (see Table 2) that constitute a significant portion of the entire
[Fig. 6(b)]. Fig. 7 shows test results. In specimen type F, the col- cross section, on both the tension and compression sides of the spec-
lapse was fragile and occurred immediately after reaching peak imens. Conversely, once a reinforcing mesh is embedded into the
strength. Conversely, RBB and FB showed a more controlled col- plaster, the results obtained on RBB and FB are practically the same.
lapse, governed by the failure of the basalt mesh fibers along the
main tensile crack [Fig. 6(c)], which developed close to one of
the load application lines. The average value of the nondimensional In-Plane/Out-of-Plane Tests Set-Up and Procedure
bending moment per meter length was 2.7 kN · mm/mm in the case
of specimens F and 3.1 kN · mm/mm in the case of RBB and FB. The combined IP-OOP tests were conducted on full-scale, one-bay,
Table 4 summarizes the results of the flexural tests. It is evident and one-story, RC frames (Fig. 8) fully infilled with the above de-
that the use of a low-quality plaster (R) and of a plaster reinforced scribed masonry type. The choice to adopt full-scale specimens was
J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)
Fig. 6. Four-point bending tests: (a) geometry with dimensions in mm; (b) test setup; and (c) failure mode.
(a) (b) (c)
Fig. 7. Four-point bending test results: (a) RBB; (b) FB; and (c) F specimens.
Table 4. Flexural test results (CoV in brackets) bottom beam. An axial load of 200 kN on each column was kept
constant during the IP and OOP tests. One servo-controlled hydrau-
Specimen Dim. less bending moment [kN] Flexural strength [N/mm2]
lic actuator is placed at the height of the top beam for applying hor-
R 1.1 0.3 (30.9%) izontal IP cyclic displacements. Lastly, the loading system for the
RBB 3.1 0.9 (14.4%) OOP test is made of commercial steel profiles connected to another
FB 3.1 0.8 (5.7%) hydraulic jack. The system allows applying eight punctual forces of
F 2.7 0.7 (22.0%)
equal intensity at every third of the width and at each fifth of the
height of the masonry infill wall.
To measure IP deformations, 11 potentiometric transducers and 1
made to avoid any scaling effect due to the relative stiffness ratio of magnetostrictive transducer were installed on the RC frame. In partic-
the RC frame and infill wall in the scaled specimens and the real ular, the magnetostrictive transducer was anchored to the top beam to
situation. The specimens were designed following the criteria de- measure the horizontal displacements applied to the specimens and to
scribed in da Porto et al. (2013), and their main characteristics retroactivate the horizontal actuator. In addition, to measure global
are shortly reported. The infill wall dimensions were 4.15 × OOP deflections, four potentiometric transducers and nine draw
2.65 mm2 (length × height) and the RC frame was designed as wire sensors were installed on the infill walls, and two potentiometric
part of the ground level frame of a regular three-story typical Italian transducers were installed on the top beam for control purposes.
residential building (class of use II) with columns spaced 4.5 m by Fig. 8 shows the position of the sensors for the measurement of IP/
4.5 m and 3.0 m high storys. The frame was designed in a high duc- OOP deformations. Load cells were also applied on each actuator
tility class (class “A” according to MIT 2008), considering a peak stem to measure forces and to retroactivate vertical and OOP loading
ground acceleration (PGA) of 0.25g. Concrete frame cross sections actuators. Fig. 9 shows some details of the IP/OOP test set-up.
were designed to respect the hierarchy of strength. The test procedure adopted was quasi-static. For IP tests, the
The experimental set-up consisted of two servo-controlled hy- displacement history shown in Fig. 10(a) was applied (with a max-
draulic actuators placed over the beam-column nodes for applying imum stroke speed of less than 0.5 mm/s), which consists of cyclic
vertical loads. The load is transferred to the columns through a self- displacements of increasing amplitude that are repeated three times
locking device, hinged to a reaction steel beam at the top of the ac- for each amplitude; the target displacements were defined based on
tuator and connected by ball-and-socket joints to the RC frame the following interstory drifts: ±0.1%; ±0.2%; ±0.3%; ±0.4%;
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 8. RC frame geometry with dimensions in mm.
(a) (b) (c)
Fig. 9. Experimental setup: (a) actuators for vertical loading; (b) actuator for IP horizontal displacements; and (c) OOP thrusting system.
(a) (b)
Fig. 10. IP test protocols adopted for (a) infilled frames; and (b) bare frames.
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 11. Crack pattern and view of specimens at end of IP tests.
±0.5%; ±0.6%; ±0.8%; ±1.0%; ±1.2%. At the end of the IP test, Fig. 12. The hysteresis loops were not completely symmetrical
the specimens were brought back to null horizontal displacement due to a nonsymmetrical damage on the reverse loading cycles.
and then the OOP test was performed by applying a monotonic in- However, for the strengthening type RBB (and FB), for which
cremental force until the collapse of the infill. At the end of the two specimens were tested at different maximum drift levels, it
OOP test, the infill walls were completely removed and the bare can be seen that the initial part of the two tests gives comparable
frames (BF) were tested in-plane [see Fig. 10(b)]; it was thus pos- results, demonstrating the reliability and repeatability of the tests.
sible to evaluate the contribution of the RC frame to the global IP Fig. 12 shows the force–displacement envelope curves associ-
response of the infilled specimens. ated with the three loading cycles. As can be seen, the execution
of three loading cycles per each displacement level allows charac-
terizing the initial response of the specimen after damage accumu-
Combined IP/OOP Test Results lation and allows checking how stable the specimen response is at
each target displacement. Indeed, damage accumulation decreases
rapidly as the number of cycles increases at the same drift level, be-
In-Plane Tests coming negligible after the third cycle. For this reason, some results
Fig. 11 shows the crack patterns and some detailed pictures of the (e.g., the load–drift envelope curves of the infill walls) are given on
observed damage at the end of the IP tests. The data acquired dur- the third cycle, in order to conservatively take into account the in-
ing the tests were used to obtain the hysteresis loops shown by elastic cyclical response of a structure during an earthquake.
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 12. Hysteresis loops and envelope curves of IP tests.
The choice of having closer target displacements in the drift range the IP behavior of the infill wall as the difference between the
from 0% to 0.6%, where the panel undergoes the various limit force–displacement envelope curve of the infilled frame and that
states and quickly changes its stiffness, allowed to obtain accurate of the bare frame [Fig. 13(b)]. Fig. 13 shows that the bare frame re-
force–displacement curves in the most significant test phases, while sponse is almost linear in the range of analyzed displacements,
maintaining reasonable testing times. which means that the frame is not significantly affected by the pre-
Fig. 13(a) shows the hysteresis loops and envelope curve ob- vious test carried out on the infilled frame. This is reasonable con-
tained from the test of a bare frame. This test was performed sidering that the frame was designed in high ductility class for a
after testing the associated infilled frame, and allowed to derive PGA of 0.25g (hence, with severe reinforcement details), and tested
J. Compos. Constr., 2020, 24(6): 04020059

(a) (b)
Fig. 13. (a) IP hysteresis loops of bare frame; and (b) infill wall contribution: subtraction of bare frame from infilled frame.
(a) (b)
Fig. 14. IP tests (all specimens): (a) load-drift envelope curves; and (b) stiffness degradation.
up to relatively low levels of drift, at which no significant damage of the infill, due to increasing IP displacements. Figs. 15 and 16
was visible. The obtained results can be thus considered reliable. show the comparison of the FRM/TRM solutions in terms of, re-
The IP performance of the investigated FRM/TRM solutions spectively, equivalent viscous damping ratio (ξ) and energy dissipa-
can therefore be compared in terms of load–drift envelope curves tion capacity (Ehyst/Einp) of the infill walls per loading cycle. The
and stiffness degradation (Fig. 14), equivalent viscous damping equivalent damping ratio is obtained with the simplified method
ratio and energy dissipation capacity (Figs. 15 and 16), and global proposed by Jacobsen (in 1930), i.e., as: ξ = Ehyst/(4πEsto), where
parameters at maximum and ultimate capacity (Table 5). Con- Ehyst = energy dissipated per-cycle and Esto = elastic energy stored
versely, since the adopted external reinforcement systems tend to at the peak displacement. In particular, Ehyst is calculated as the sum
hide the internal portion of the infill walls, and thus do not allow of the products between the imposed displacement increments and
appreciating the evolution of damage, the analysis of crack patterns the related measured loads (for the whole cycle). Instead, Einp is
was carried out at the end of the IP tests only (Fig. 11) and was not the input energy, i.e., the total energy provided to the specimen dur-
given as a function of the drift level. Further considerations and ob- ing the test, and it is obtained with the same procedure as Ehyst but is
servations were then made possible by comparing all specimens limited to positive products only. Table 5 summarizes the following
with a reference unreinforced infill wall (URM), which was tested results: maximum IP capacity Fmax, drift θFmax at peak strength, in-
in a previous experimental campaign (da Porto et al. 2015). The fill damping ratio ξ and energy dissipation capacity Ehyst/Einp at
URM panel was geometrically identical to this research samples maximum strength, ultimate force Fult evaluated as a 20% decrease
and it was realized using the same type of masonry units. The ex- of Fmax and related drift θFult, and the maximum stiffness Kmax.
ternal plaster layer was made of natural hydraulic lime without Observing the test results, the embedded fiber meshes seems to
strengthening mesh. It was cyclically IP tested up to 1.2% drift. provide a better distribution of IP damage. Indeed, specimen F.12,
Fig. 14(a) shows the force–displacement envelope curves of the the only one without reinforcing mesh, developed the first well-
infill walls, while Fig. 14(b) reports the secant stiffness degradation defined cracks along the infill–frame interface since 0.2% drift.
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 15. IP tests (specimens θ = 1.2%): equivalent viscous damping ratio (ξ) per loading cycle.
Fig. 16. IP tests (specimens θ = 1.2%): energy dissipation capacity (Ehyst/Einp) per loading cycle.
Table 5. In-plane test results

Maximum capacity Ultimate capacity
Kmax Average θFult/θFmax
Specimen Fmax [kN] θFmax [%] ξ [%] Ehyst/Einp [%] Fult [kN] θFult [%] [kN/mm] [—]
F.12 230 0.30 5.9 41.8 184 0.53 58.8 1.77
FB.05 273 0.30 10.6 56.0 218 0.42 69.4
1.62
FB.12 259 0.30 9.0 53.5 207 0.55 79.6
RBB.05 178 0.20 5.7 38.5 142 0.44 53.6
2.31
RBB.12 190 0.20 8.5 48.3 152 0.48 58.3
URM.12 181 0.20 6.0 32.7 144 0.30 49.7 1.52
Damage increased until attaining maximum drift θFmax = 1.2%, at specimens, reaching comparable peak strength at the same IP
which widespread damage with detachment of plaster portions drift level (i.e., 0.2%), although the post-peak behavior of RBB
and crushing of masonry blocks at the infill corner was observed was characterized by a more gradual strength decrease.
(see Fig. 11). The infill reached its maximum IP capacity 230 kN Indeed, in all specimens characterized by the presence of the ba-
at 0.3% drift, and its ratio of θFult/θFmax was 1.77 (Table 5). The salt mesh (RBB and FB strengthening solutions), IP damage devel-
reference unreinforced infill wall showed thin cracks and plaster oped in a completely different manner. Detachment was limited to
detachment concentrated in the upper beam–column joint area little plaster portions and the only visible cracks were at the col-
and along the infill–frame interface up to 0.5% drift. The peak umn–infill interface. The mesh could not prevent IP damage from
strength of 181 kN was achieved at 0.2% drift after which infill occurring, but it could effectively counteract masonry and plaster
strength decreased quite sharply. At 1.2% drift, the masonry corner portions from falling off the wall (see Fig. 11).
was disintegrated, and significative lateral portions collapsed Coupling basalt mesh with glass fiber plaster allowed reaching
out-of-plane (see Fig. 17). Compared with the F.12 specimen, it higher maximum strength. FB.05 and FB.12 reached their peak ca-
had more extended damage of the wall which was not prevented pacity of 273 and 259 kN at a drift of 0.3% and the average ratio of
by the external plaster with lower mechanical properties. In terms θFult/θFmax was 1.62. As expected, the lower values of Fmax were
of IP capacity, the URM infill wall behaved similarly to the RBB related to the application of the basalt mesh directly on an existing
J. Compos. Constr., 2020, 24(6): 04020059

(a) (b)
Fig. 17. Reference URM infill wall at (a) 0.3% drift; and (b) 1.2% drift.
Fig. 18. OOP force-displacement curves.
plaster of lower mechanical properties. Indeed, specimens RBB.05 Out-of-Plane Tests

and RBB.12 reached their peak strength (178 and 190 kN respec-
As previously mentioned, specimens F.00, FB.00, and RBB.00
tively) at a drift value (0.2%) lower than that found for the previous
were tested under OOP monotonic loading without previous appli-
solutions due to the thick low-quality plaster layer between the ma- cation of IP displacements, whereas FB.05 and RBB.05 were tested
sonry panel and the strengthening system. For RBB specimens, the OOP after attaining 0.5% IP drift and F.12, FB.12, and RBB.12
average ratio of θFult/θFmax was 2.31. after 1.2% drift. Fig. 18 shows the force–displacement curves of
The post-peak behavior is very important in evaluating the over- the OOP tests, whereas Figs. 19–21 show the main failure modes
all performance of the investigated solutions. In general, all of the eight specimens. Table 6 lists the main experimental results
strengthening solutions showed a better post-peak branch with rein terms of maximum OOP capacity Fmax, relevant displacement
spect to the URM infill wall, with higher ultimate strength (espe- δFmax, and observed collapse mode.
cially for FB and F solutions) and/or with a more gradual strength All undamaged specimens obtained similar values of maximum
degradation (as shown by the average values of θFult/θFmax). In OOP capacity. Thanks to the combination of fiber plaster with good
terms of stiffness degradation [Fig. 14(b)], all specimens showed mechanical properties and of basalt mesh, FB.00 had the stiffer
similar trends and the URM infill wall, due to its significative IP OOP behavior, reaching its maximum capacity (96.01 kN) with
damage, showed the lowest stiffness values. an OOP deflection of 5.35 mm. F.00 showed an initial stiffness
Lastly, the energy dissipation capacity of the strengthening so- equal to that of FB.00 (the two specimens have the same plaster
lutions with embedded mesh (RBB and FB) is greater than that of type) but, starting from 30 kN, cracking induced a significant
the F and URM cases (Figs. 15 and 16). In particular, this better loss of stiffness. The infill reached a capacity of 90.05 kN at
performance is more evident for the loading cycles following the 14.60 mm. Lastly, RBB.00 showed an intermediate behavior. In
first and in the drift range between 0% and 0.6%, which includes this TRM solution, probably due to the low-quality plaster, it
the ultimate limit state of these infill walls. Considering that the had an initial stiffness that was lower than those of FB.00 and
RBB and FB solutions are those associated with the lowest level F.00 but, thanks to the basalt mesh, the hardening branch devel-
of observed damage (see Fig. 11), these results further confirm oped until the peak strength of 101.31 kN, attained with an OOP
the overall effectiveness of these solutions. Such behavior can be deflection of 12.14 mm.
explained in relation to the improved distribution of IP damage al- The experimental results demonstrated that IP damage is re-
lowed by the strengthening mesh. sponsible for a reduction of OOP stiffness and strength. This
J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)
Fig. 19. OOP failure mode: (a) F.00; (b) FB.00; and (c) RBB.00.
(a) (b)
Fig. 20. OOP failure mode: (a) FB.05; and (b) RBB.05.
(a) (b) (c)
Fig. 21. OOP failure mode: (a) F.12; (b) FB.12; and (c) RBB.12.
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 22. OOP failure mode of URM infill wall (da Porto et al. 2015).
Table 6. Out-of-plane test results Table 6 also lists the values of OOP equivalent acceleration (aeq)
Out-of-plane capacity
reached by each infill type, calculated assuming the total mass of the
infill wall (variable between 1.6 and 1.9 t depending on FRM/TRM
Specimen Fmax [kN] δFmax [mm] aeq [g] Failure mode solution) as participating mass. The proposed values of aeq take into
F.00 90 14.6 5.7 Material ejection account the soil amplification factor S, the position of the infill along
FB.00 96 5.4 6.0 Controlled the building height H, and the interaction between infill wall and RC
RBB.00 101 12.1 5.4 Controlled frame in terms of Ta/T1 ratio, where Ta and T1 are the fundamental
FB.05 70 10.6 4.4 Controlled vibration period in the OOP direction of the infill and of the
RBB.05 68 21.5 3.6 Controlled frame, respectively. It must be taken into account that the dynamic
F.12 60 14.0 3.7 Material ejection effects associated to the real seismic action may reduce the previous
FB.12 81 13.0 5.1 Controlled
OOP strength values, obtained by means of a quasi-static test.
RBB.12 62 12.8 3.3 Controlled
Influence of In-Plane Damage on Out-of-Plane

means that IP damage induced lower values of maximum OOP Strength
strength with higher values of OOP deflections, as all tested spec-
imens generally confirmed. An anomalous behavior was observed Analyzing the OOP tests, the specimens tested until an IP drift level
for specimen FB.12. Indeed, this infill wall reached an OOP capac- of 0.5% (FB.05 and RBB.05) showed a significant OOP strength
ity of 80.54 kN, which was higher than that of FB.05 (i.e., degradation (33%) compared with the specimens tested only
69.51 kN). This was caused by a more accurate construction of under OOP loads (i.e., specimens F.00, FB.00, and RBB.00). It is
the infill wall, due to the fact that it was a replica of another one, worth noting that the specimens tested until a definitely higher IP
that failed even before carrying out the experimental test. drift level of 1.2% (except for the anomalous FB.12 specimen), ex-
Considering the observed collapse mechanisms (Figs. 19–21), perienced an only slightly higher OOP strength degradation (39%)
the FRM solution (without the embedded basalt mesh) showed a compared with the previous. It thus seems that the most relevant
more fragile collapse mode, characterized by the uncontrolled ejec- strength degradation occurs at low IP drift values (until 0.5%), con-
tion of masonry and plaster portions [see Figs. 19(a) and 21(a)]. In sistently with the fact that all masonry infill walls reached their IP
addition, the URM infill wall, after being tested IP up to 1.2% drift, maximum capacity for a drift level between 0.2% and 0.3%, with
showed a rather fragile collapse. As shown in Fig. 22, the masonry consequent damage propagation and strength degradation.
wall subdivided into three portions separated by the diagonal Fig. 23 shows the experimental results of the combined IP/OOP
cracks which had formed during the in-plane test. The effectiveness tests plotting the OOP strength reduction of all specimens versus the
of the embedded mesh was again demonstrated by the FB and RBB previously attained IP drift level. These values represent the ratio be-
TRM solutions, whose out-of-plane collapse occurred maintaining tween the OOP strength of each FRM/TRM solution and its corre-
the overall integrity of the masonry panel. The reinforcing mesh is sponding value obtained on the undamaged wall. In the same
highly effective, even when it is applied directly on the existing chart, the OOP strength degradation values of other experimental
low-quality plaster. This aspect is particularly relevant, especially tests available in literature are summarized (i.e., Angel et al. 1994;
in all those situations where a rapid and low impact rehabilitation Calvi and Bolognini 2001; Furtado et al. 2016; Ricci et al. 2018).
intervention is required. Furthermore, the additional use of helicoi- Only tests on unreinforced thin clay masonry infill walls are reported.
dal anchorages that connect the strengthening system to the upper The latter show a significative OOP strength degradation caused by
beam (RBB solution) represents a further defense against the OOP the previous IP damage and, starting from a drift level of 0.4%, they
overturning of the masonry panels. It is worth noting that, despite show, in general, a strength degradation about 73%, which is twice
the considerable contribution in the OOP behavior, the helicoidal than that of the strengthened panels tested in our research.
bars did not significantly influence the infill IP response, both in According to Verlato et al. (2014), an OOP strength reduction
terms of damage reduction and post-peak behavior control. factor βa,exp, expressing the reduction of masonry strength as a
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 23. Out-of-plane strength degradation due to in-plane damage.
function of the in-plane drift θIP, has been proposed:
1
βa, exp = [(a − 1)e−(b·θIP ) + 1]
2
(1)
a
where a = 1.65 and b = 2.80. The reduction factor can be proposed

also for URM infill walls assuming a = 5.90 and b = 3.70. This re-
sult can be useful for design approaches that require the OOP ver-
ification of the infill wall, which should be related to actual
properties of the masonry walls, accounting for the eventual prop-
erty degradation due to the occurrence of in-plane damage.
In this study, the effect of infill wall openings, such as windows
and doors, was not evaluated. The shape and size of these openings
clearly affects the overall IP behavior of infilled frames, reducing
strength, stiffness, and dissipative capacity, as shown for example
by Kakaletsis and Karayannis (2008, 2009) (testing 1/3-scale, one-
bay, one-story RC frames infilled with weak or strong URM panels)
and by Verlato et al. (2016) (testing full-scale, one-bay, one-story RC Fig. 24. Distribution of seismic forces.
frames infilled with strong and ductile URM panels, with deformable
joints). Instead, as regards the OOP behavior of the infill walls, it is horizontal seismic force Fa acting on a masonry infill wall is
reasonable to assume, in the hypothesis of a vertical arch or vertical given by
bending resistant mechanism, that this behavior is not significantly
affected by the presence of openings, i.e., that the OOP capacity Sa Wa
per unit of length of infill wall remains substantially unchanged. Fa = (2)
qa
However, the presence of openings induces greater IP damage on
the infill walls which, in turn, influences their OOP resistance, thus
where Fa = seismic force acting in the center of mass of the non-
requiring a different calibration of Eq. (1) (as shown by Verlato
structural element, resulting from the distributed forces propor-
et al. 2014 with IP/OOP tests of full-scale, one-bay, one-story RC
tional to mass, Wa = weight of the wall, and Sa = maximum
frames infilled with strong unreinforced and reinforced masonry pan-
OOP acceleration, which can be calculated, according to Circ.
els). In this framework, it is very likely that the proposed strengthen-
2019 no. 7 (MIT 2019), by means of response spectra or through
ing solutions increase in a fairly proportional way the OOP
simplified formulations of proven validity. The behavior factor qa
performance of the infill walls even when there are openings, as
can be assumed equal to two (veneer walls, internal partitions, in-
these solutions, besides increasing the OOP resistance, significantly
fill walls, etc.), according to Circ. 2019 no. 7 (MIT 2019). As
limit the capacity reduction due to IP damage.
shown in Fig. 24, the OOP seismic action can be represented
as a pressure pa due to the horizontal force Fa distributed along
the free height of the wall hw.
Non-Structural Elements Design Formulations For practical design purposes, it is important to propose and val-
idate adequate formulations for the calculation of the OOP capacity
The ultimate limit state (ULS) verification of nonstructural ele- of masonry infill walls; in our case, in particular, for the various
ments is satisfied when the OOP capacity is greater than the seismic strengthening solutions proposed by this research. It is noteworthy
demand corresponding to the considered design limit state. The that the current EN 1998-1 (CEN 2004) and Italian DM 2018 (MIT
J. Compos. Constr., 2020, 24(6): 04020059

Fig. 25. Out-of-plane capacity models (1 and 2).
2018) do not provide specific recommendations for the calculation To satisfy the ULS verification, the design lateral resistance qlat,d
of the infill out-of-plane capacity at all. In the case of unreinforced must be higher than the seismic lateral load. The lateral capacity for-
masonry infill walls, it is possible to adopt pure flexural models, as mulation provided by EN 1996-1-1 (CEN 2005) assumes a unitary
well as to evaluate the out-of-plane bending capacity, taking into factor, instead of 0.72 as in Eq. (4), entailing an overestimation of
account the infill wall self-weight at the section to be verified. Nev- the lateral capacity of about 50%. Considering the type of masonry
ertheless, one of the most consistent way to evaluate the OOP infill infills addressed in this research, it seems more reliable to apply the
wall capacity, relies on the model of wall arching between supports, formulation of Eq. (4), which is derived from pure mechanical con-
proposed in EN 1996-1-1 (CEN 2005). Starting from these consid- siderations, also according to Drysdale et al. (1999). In the follow-
erations, five models are described and compared in the following ing, the arching mechanism will be identified as Model 1.
subsections, which are: the arch model (Model 1), the
modified-arch models according to Hak et al. (2013) (Model 2)
and to FEMA-306 1999 (Model 3), the flexural model of the rein- Model 2: Arching Mechanism with Simplified
forced section (Model 4), and the flexural model with simplified Reinforcement Contribution
vertical load contribution (Model 5). The formulation provided by Model 1 does not consider the con-
tribution given by an external reinforcement to the lateral OOP
capacity. Hak et al. (2013) proposed a formulation which allow
Model 1: Out-of-Plane Arching Mechanism
to calculate the OOP capacity of strengthened thin masonry infill
The ULS verification of masonry infill panels can be carried out as- walls by adding the contribution of the external reinforcement
suming that a horizontal or vertical arch develops within the wall Mr,r to the resistant moment due to the arching mechanism Mr,a.
thickness. As demonstrated by Angel et al. (1994), it is possible to The approach is based on the simplified assumption that the neu-
rely on this mechanism when the masonry panel is built in adherence tral axis depth can be assumed equal to that used for the calcu-
to the structural elements. The resulting arching mechanism depends lation of the arching mechanism. The reinforcement resisting
on the slenderness ratio of the wall λ = hw/tw and on the masonry moment is defined in Eq. (5), where Ar and fyd are, respectively,
compressive strength fmd. The analytical model is based on a simpli- the cross-section area and the design yield strength of the verti-
fied configuration of a three-hinged arch with an arch support cal reinforcement:
footprint at the extremities and on the central hinge equal to 0.1
times the thickness of the wall tw, as shown in Fig. 25. Assuming Mr,r = 0.9 · tw · Ar · fyd (5)
that the deflection of the arch under the lateral loads may be ne-
glected, the masonry wall resistant moment Mr,a can be calculated as Expressing the resistant moment in terms of equivalent lateral
pressure per unit length of the wall Lw, the design lateral resistance
Mr,a = 0.9 · tw · fmd · 0.1 · tw = 0.09 · fmd · tw2 (3) can be calculated as
2
The lateral capacity qlat,d of the infill panel can be derived by as- tw tw
qlat,d = 0.72 · fmd + 7.2 Ar fyd (6)
suming the resistant moment equal to the acting moment, i.e., hw Lw h2w
Mr,a = Ms = qlat,d · h2w /8, which leads to
2 Model 3: Modified-Arch Model According to FEMA-306
tw
qlat,d = 0.72 · fmd (4)
hw According to FEMA-306 (FEMA 1999, chapter 8), the uniform
pressure causing the infill wall OOP failure can be estimated as
J. Compos. Constr., 2020, 24(6): 04020059

tw and Ar = the Young modulus and the equivalent area of the rein-
qlat,d = 2 · fmd ρR1 R2 (7) forcement, respectively:
hw
where ρ = slenderness parameter dependent on the hw/tw Rc (x) = αβfmd Lw x (8)

ratio and defined in Table 8-5 of the FEMA-306, and R1 and d−x
R2 = OOP strength reduction factors that take into account, Rt (x) = εmu Er Ar (9)
x
respectively, the existing IP damage and the confining frame
flexibility (for further details, please refer to FEMA 1999). In The position of the neutral axis is determined imposing the equi-
particular, for the purposes of comparison with the experimental librium of the section against vertical translations (Rc = Rt + NSd) as
case studies, the values adopted are: ρ = 0.027 (hw/tw = 17.7),
R1 = 1 and R2 = 1. x(NSd )

NSd − Er Ar εmu + (Er Ar εmu − NSd )2 + 4αβfmd Lw Er Ar εmu d
Model 4: Out-of-Plane Flexural Mechanism =

2αβfmd Lw
According to DM 2018 (MIT 2018) and CNR-DT 200 (CNR 2004), (10)
the OOP flexural capacity of a reinforced masonry section can be
calculated by assuming a compressed stress block with a height Therefore, assuming NSd = 0, the resisting moment is defined as
equal to αfmd and a depth of βx, where α = 1.00 (0.85 for design),
αβfmd Lw x εmu tw
β = 0.80 and x is the depth of the neutral axis (see Fig. 26). The sec-
Mr,c = (tw − βx) + (d − x)Er Ar d − (11)
tion failure may occur on the masonry side, exceeding the ultimate 2 x 2
deformation ɛmu, as well as on the reinforcement side, exceeding the
ultimate deformation ɛrd. which is correct when the condition ɛr ≤ ɛrd is respected.
In the case of compressive failure of masonry, the resultants of In the case of tensile failure of reinforcement, the resultant of com-
compression forces on the masonry (Rc) and tensile stresses on pression forces on the masonry (Rc) is the same as described in
the reinforcement (Rt) are given by Eqs. (8) and (9), where Er Eq. (8), whereas the resultant of tensile stresses on the reinforcement
(a) (b)
Fig. 26. Cross-section failure mode: (a) compressive failure of masonry; and (b) tensile failure of reinforcement.
J. Compos. Constr., 2020, 24(6): 04020059

(Rt) is given by
Rt (x) = εrd Er Ar (12)
The position of the neutral axis is determined imposing the equi-

librium of the section against vertical translations (Rc = Rt + NSd) as
NSd + Er Ar εrd
x(NSd ) = (13)
αβfmd Lw
Therefore, assuming NSd = 0, the resisting moment is defined by

αβfmd Lw x tw
Mr,t = (tw − βx) + εrd Er Ar d − (14)
2 2
which is correct when the condition 0 ≤ ɛm ≤ ɛmu is respected.
The flexural capacity can be taken as the lesser of the above cal-
culated values as
Mr = min {Mr,c ; Mr,t } (15)

Fig. 27. OOP lateral capacity: comparison between experimental and
Expressing the resistant moment in terms of equivalent lateral
analytical values.
pressure per unit length of the wall Lw, the design lateral resistance
can be calculated by
8Mr load lines. In the case of plaster type F, the area and the yield strength
qlat,d = (16) of the reinforcement are considered to be, respectively, the cross sec-
Lw h2w
tion and the tensile strength of the external plaster layer.
As expected, Model 1 underestimates the experimental lateral ca-
Model 5: Out-of-Plane Flexural Mechanism pacity (with 33–38% deviation) since it neglects the contribution of
with Simplified Vertical Load Contribution the reinforcement layer. On the other hand, Model 2, although imple-
menting a simplified evaluation of the reinforcement contribution to
The formulation provided by Model 4 does not consider the con- the masonry wall capacity, gives results that are very close to the ex-
tribution given by the relevant vertical load that, in the case of in- perimental ones (with 4–7% overestimation). Model 3, i.e., the
fill walls built in adherence to the structural elements, develops modified-arch model of FEMA-306 (FEMA 1999), provides the
within the masonry thickness as soon as the wall deflects out of closest results (with ±4% deviation); however, considering that it
its plane, due to the action of lateral loads and boundary condi- does not specifically take into account the contribution of the rein-
tions. Neglecting the vertical load, or taking into account a forcement layer (as for Model 1), it could be non-conservative
value of vertical load equal to the infill wall self-weight at the sec- when applied to typical weak masonry infill walls made of hollow
tion to be verified, may bring to excessive underestimation of this clay units, such as the URM infill walls of our research.
contribution that, in turn, means excessively conservative estima- Model 4, neglecting the contribution of the vertical load acting
tions of the OOP infill wall capacity. on the section, significantly underestimate the experimental lat-
To take into account the actual vertical load acting at the wall eral capacity (with 55–61% deviation). Conversely, Model 5,
mid-height section, a rigorous procedure would require an iterative
which takes this contribution into account, albeit in a simplified
calculation. However, a simplified procedure for design purpose
way, provides better results than Model 4, with an overestimation
can assume that the vertical load acting on the mid-height section
of the experimental results of about 8%–10%, i.e., slightly higher
is evaluated according to the arching mechanism (as given by
than that of Model 2 despite the more detailed evaluation of the
Model 1), adding the self-weight of the wall acting on such section.
neutral axis in Model 5.
The position of the neutral axis of Model 4 can be then evaluated on
the basis of this value of vertical load and, as a consequence, the
values of resisting moment and flexural capacity of the wall in pres-
ence of a vertical load can be calculated. Conceptually, Model 5 is Conclusions
not much different from Model 2, although the results of Model 5
are affected by the relative quantities of masonry and reinforcement Considering the results of IP tests, it can be concluded that:
which influence the position of the neutral axis. • All specimens, although strengthened with various solutions,
reached the peak strength at low drift levels (0.2%–0.3%).
These outcomes are compatible with other results found in the
Validation of the Proposed Formulations literature, for thin clay masonry infill walls. These results con-
The proposed formulations were implemented and validated on the firm that, particularly in the case of masonry infill walls made
experimental results of tested strengthened masonry infills (i.e., F, with weak fragile clay units, the inter-story drift values proposed
FB and RBB), in case of undamaged masonry. The results are sum- by EN1998 for the damage limit state (θ = 0.5%), should be re-
marized in Fig. 27. All calculations were carried out considering the garded as values to be found from analyses on bare frames, and
mean mechanical values of the materials, i.e., neglecting partial should be carefully taken into account.
safety factors. Furthermore, the calculated values of lateral resis- • In general, IP capacity and stiffness of thin and weak clay ma-
tance, consistently with the experimental conditions, were obtained sonry infill walls increase with the use of fiber-reinforced plas-
by evaluating the out-of-plane thrust applied along four horizontal ters of high mechanical properties. However, the use of a glass
J. Compos. Constr., 2020, 24(6): 04020059

fiber plaster alone (F ), cannot significantly reduce the damage only slightly overestimate the experimental results, with devia-
due to IP displacements. tions of up to 7% and 10%, respectively.
• Embedding a strengthening mesh on the external plaster layers • Finally, the arch-modified model proposed in FEMA-306 (FEMA
produces a confinement effect, which reduces IP damage of the 1999; Model 5) provides the lateral capacity values that best ap-
infill wall. The basalt mesh cannot prevent crushing of masonry proximate the experimental ones. However, it does not explic-
units and detachment of little plaster portions, nevertheless it has itly take into account the effect of external reinforcement,
an evident beneficial effect in preventing the expulsion of ma- therefore it seems that the models reported in the previous
sonry and plaster portions. The mesh also allows controlling point could be more suitable for the scope of evaluating thin
the initial post-peak branch and obtaining higher values of masonry infill walls strengthened with various types of exter-
equivalent viscous damping and energy dissipation capacity nally applied systems.
(despite damage reduction).
• The previous mesh effects are irrespective of the quality of the
plaster, as they also occur in specimens (RBB) with low-quality Data Availability Statement
plaster, simulating the effect of a preexisting material.
• When taking into account the behavior of URM infill walls, all of Some or all data, models, or code that support the findings of this
the previous considerations are confirmed. Indeed, at the end of study are available from the corresponding author upon reason-
the IP test, the URM specimen showed a significative damage able request.
distribution compared with the specimens of this research, espe-
cially along the infill–frame interface. In addition, compared
with the URM specimen, the strengthening solutions allowed Acknowledgments
an increase in strength (Fmax and therefore Fult) of about 30%
(F) or 50% (FB) and an increase in deformation capacity, This research activity was funded by Kerakoll S.p.A. This study
in terms of θFult, of about 50% (RBB) or 70% (F, FB); the initial has been also partly supported by a program funded by the Presi-
stiffness Kmax was also increased by about 15% (RBB, F ) or 50% dency of the Council of Ministers Department of Civil Protection
(FB). (2019–2021 DPC- ReLUIS). The experimental tests were carried
Considering the results of the OOP tests, it can be concluded that: out at the Laboratory of Structural Materials Testing of the Univer-
• The comparison among strengthened infills and the reference un- sity of Padova, Italy.
reinforced one showed a relevant increase of the OOP strength in (All photos: University of Padova and Kerakoll S.p.A., Italy.)
the first case. Indeed, all the strengthened specimens, after reach-
ing an IP drift of 1.2%, showed an OOP capacity at least 26%
higher than that of the reference unreinforced specimen. Notation
• Although the final values of OOP capacity, at each IP drift level,
do not differ significantly, it was possible to notice that the use of The following notation symbols are used in this paper:
an embedded mesh, in FB and RBB solutions, improved the OOP Ar = cross-section area of infill wall reinforcement;
response of the infill wall, preventing its possible brittle collapse. aeq = equivalent seismic acceleration;
Conversely, also taking into account the actual dynamic behavior, d = effective depth of reinforcement;
Er = Young modulus of reinforcement;
type F strengthening could yield to a more fragile collapse.
Fa = horizontal seismic force acting on a masonry infill;
• The use of anchorage bars embedded into the plaster layer and fixed Fmax = maximum resistance of infill wall;
to the upper beam (RBB specimens), although not significantly in-
Fult = ultimate resistance of infill wall, corresponding to
fluencing the OOP capacity of the infill, allowed better control of
resistance degradation of 20%;
the OOP failure mode after the peak strength. This addresses at a
fmd = design compressive strength of masonry;
combined solution (mesh + anchorage bars) as particularly promis-
fyd = design tensile strength of reinforcement;
ing for controlling the masonry infill wall OOP behavior. hw = height of infill wall;
• The experimental tests allowed evaluation of the reduction of Lw = length of infill wall;
OOP capacity due to IP damage. The strength degradation is Mr = flexural capacity of infill wall;
around 39% at 1.2% drift, and it seems to be already stabilized Mr,a = bending moment, calculated per unit length, due to
after an initial relevant strength degradation of 33% at 0.5% arching mechanism;
drift. The definitely higher strength degradation values (i.e., Mr,c = flexural capacity of the infill wall due to compressive
73% at 1.2% drift) of similar URM walls further demonstrate failure of masonry;
the effectiveness of the proposed strengthening solutions. Mr,r = bending moment, calculated per unit length, due to tensile
Lastly, some analytical models for the calculation of the OOP strength of reinforcement;
capacity of masonry infill walls were described and validated on Mr,t = flexural capacity of the infill wall due to tensile
the experimental results of IP undamaged panels: reinforcement failure;
• As expected, both the arching (Model 1) and the out-of-plane Ms = calculated acting moment per unit length of infill wall;
flexural (Model 4) mechanisms significantly underestimate the NSd = vertical load acting on the wall;
OOP infill wall capacity, with a reduction up to 61% compared pa = seismic pressure acting on infill wall surface;
with the experimental values. qa = behavior factor of infill wall;
• Other hybrid formulations which account for, in a simplified qlat,d = design lateral strength, expressed as a uniform load
way, the contribution of the external reinforcement (Model 2) distributed over the wall length;
and the contribution of the vertical load acting at wall mid- Rc = resultant of compression forces on masonry;
height (Model 5), provide more realistic values of the lateral ca- Rt = resultant of tensile stresses on reinforcement;
pacity and can be proposed as reliable formulations for design/ R1 = out-of-plane strength reduction factor taking into account
verification purposes. In particular, both Model 2 and Model 5 the in-plane damage of infill;
J. Compos. Constr., 2020, 24(6): 04020059

R2 = out-of-plane strength reduction factor taking into account CEN (European Committee for Standardization). 1998b. Methods of test
the confining frame flexibility; for masonry, Part 1: Determination of compressive strength. EN
S = soil amplification factor; 1052-1. Brussels, Belgium: CEN.
Sa = maximum out-of-plane acceleration acting on the infill CEN (European Committee for Standardization). 1999. Methods of test for
masonry, Part 2: Determination of flexural strength. EN 1052-2.
wall;
Brussels, Belgium: CEN.
Ta = fundamental vibration period of the infill wall in the
CEN (European Committee for Standardization). 2004. Design of struc-
out-of-plane direction; tures for earthquake resistance, Part 1: General rules, seismic ac-
T1 = fundamental vibration period of the building in the tions and rules for buildings. Eurocode 8, EN 1998-1. Brussels,
considered direction; Belgium: CEN.
tw = thickness of infill wall; CEN (European Committee for Standardization). 2005. Design of masonry
Wa = weight of infill wall; structures, Part 1-1: General rules for reinforced and unreinforced ma-
x = depth of neutral axis; sonry structures. Eurocode 6, EN 1996-1-1. Brussels, Belgium: CEN.
α = reduction coefficient to take into account long-term
CEN (European Committee for Standardization). 2006a. Methods of test

effects on the compressive strength; for mortar for masonry—Determination of flexural and compressive
β = coefficient to calculate the depth of the stress block; strength of hardened mortar. EN 1015-11/A1. Brussels, Belgium:
βa,exp = masonry strength reduction factor, function of the CEN.
CEN (European Committee for Standardization). 2006b. Products and
in-plane drift;
systems for the protection and repair of concrete structures—Test
δ = out-of-plane displacement; methods—Determination of modulus of elasticity in compression.
δFmax = out-of-plane displacement at the maximum out-of-plane EN 13412. Brussels, Belgium: CEN.
strength; CEN (European Committee for Standardization). 2013. Testing hardened
ɛmu = ultimate deformation of masonry; concrete, Part 13: Determination of secant modulus of elasticity in
ɛrd = ultimate deformation of reinforcement; compression. EN 12390-13. Brussels, Belgium: CEN.
θ = in-plane inter-story drift ratio; CEN (European Committee for Standardization). 2015. Specification for
θFmax = inter-story drift ratio at maximum resistance; masonry units, Part 1: Clay masonry units. EN 771-1. Brussels,
θFult = inter-story drift ratio at ultimate resistance; Belgium: CEN.
θIP = maximum level of in-plane drift ratio applied to the CEN (European Committee for Standardization). 2016a. Specification for
specimen; mortar for masonry, Part 1: Rendering and plastering mortar. EN
998-1. Brussels, Belgium: CEN.
λ = slenderness ratio of infill wall; and
CEN (European Committee for Standardization). 2016b. Specification for
ρ = slenderness parameter of infill wall. mortar for masonry, Part 2: Masonry mortar. EN 998-2. Brussels,
Belgium: CEN.
Chiozzi, A., and E. Miranda. 2017. “Fragility functions for masonry infill
References walls with in-plane loading.” Earthquake Eng. Struct. Dyn. 46 (15):
2831–2850. https://doi.org/10.1002/eqe.2934.
Angel, R. E., D. P. Abrams, D. Shapiro, J. Uzarski, and M. Webster. 1994. CNR (National Research Council). 2004. Guide for the design and con-
Behavior of reinforced concrete frames with masonry infills. struction of externally bonded FRP systems for strengthening existing
Technical Report. Champaign, IL: Univ. of Illinois at structures. [In Italian.] CNR-DT 200. Rome: CNR.
Urbana-Champaign. Crisafulli, F. J. 1997. “Seismic behaviour of reinforced concrete struc-
Anil, Ö, and S. Altin. 2007. “An experimental study on reinforced concrete tures with masonry infills.” Ph.D. thesis, Civil Engineering, Univ. of
partially infilled frames.” Eng. Struct. 29 (3): 449–460. https://doi.org Canterbury.
/10.1016/j.engstruct.2006.05.011. da Porto, F., G. Guidi, M. Dalla Benetta, and N. Verlato. 2013. “Combined
Baio Dias, A., F. da Porto, E. Fehling, P. B. Lourenço, P. Morandi, E. in-plane/out-of-plane experimental behaviour of reinforced and
Vintzileou, and A. Yakut. 2014. “Innovative systems for earthquake re- strengthened infill masonry walls.” In Proc., 12th Canadian
sistant masonry enclosures in RC buildings.” In Proc., 9th Int. Masonry Masonry Symp. Vancouver, BC: Canadian Masonry Symposium
Conf., 1–12. Minho, Portugal: University of Minho and ISISE and (CMS).
International Masonry Society (IMS). da Porto, F., G. Guidi, N. Verlato, and C. Modena. 2015. “Effectiveness of
Bazzurro, P., et al. 2009. Learning from earthquakes: The Mw 6.3 plasters and textile reinforced mortars for strengthening clay masonry
Abruzzo, Italy, Earthquake of April 6, 2009. EERI Special infill walls subjected to combined in-plane/out-of-plane actions””
Earthquake Report. Oakland, CA: Earthquake Engineering Research [Wirksamkeit von Putz und textilbewehrtem Mörtel bei der
Institute. Verstärkung von Ausfachungswänden aus Ziegel]. Mauerwerk 19 (5):
Bergami, A. V., and C. Nuti. 2015. “Experimental tests and global model- 334–354. https://doi.org/10.1002/dama.201500673.
ing of masonry infilled frames.” Earthquakes Struct. 9 (2): 281–303. da Porto, F., N. Verlato, G. Guidi, and C. Modena. 2016. “The INSYSME
https://doi.org/10.12989/eas.2015.9.2.281. project : Innovative construction systems for earthquake resistant ma-
Blackard, B., K. Willam, and M. Sivaselvan. 2009. Experimental observa- sonry infill walls.” In Vol. 6 of Proc., 16th Int. Brick and Block
tions of masonry infilled reinforced concrete frames with openings, Masonry—Trends, Innovations and Challenges Conf., edited by C.
199–221. ACI Special Publication 265. Farmington Hills, MI: Modena, F. da Porto, and M. R. Valluzzi, 1173–1178. London:
American Concrete Institute. Taylor & Francis Group.
Braga, F., V. Manfredi, A. Masi, A. Salvatori, and M. Vona. 2011. Dafnis, A., H. Kolsch, and H. G. Reimerdes. 2002. “Arching in masonry
“Performance of non-structural elements in RC buildings during the walls subjected to earthquake motions.” J. Struct. Eng. 128 (2): 153–
L’Aquila, 2009 earthquake.” Bull. Earthquake Eng. 9 (1): 307–324. 159. https://doi.org/10.1061/(ASCE)0733-9445(2002)128:2(153).
https://doi.org/10.1007/s10518-010-9205-7. Dawe, J. L., and C. K. Seah. 1989. “Behaviour of masonry infilled steel
Calvi, G. M., and D. Bolognini. 2001. “Seismic response of reinforced frames.” Can. J. Civ. Eng. 16 (6): 865–876. https://doi.org/10.1139
concrete frames infilled with weakly reinforced masonry panels.” /l89-129.
J. Earthquake Eng. 5 (2): 153–185. https://doi.org/10.1080 De Luca, F., G. M. Verderame, A. Pérez-garcía, and F. Gómez-martínez.
/13632460109350390. 2014. “The structural role played by masonry infills on RC building
CEN (European Committee for Standardization). 1998a. Methods of test for performances after the 2011 Lorca, Spain, earthquake.” Bull.
mortar for masonry, Part 1: determination of particle size distribution Earthquake Eng. 12 (5): 1999–2026. https://doi.org/10.1007/s10518
(by sieve analysis). EN 1015-1. Brussels, Belgium: CEN. -013-9500-1.
J. Compos. Constr., 2020, 24(6): 04020059

De Martino, G., M. Di Ludovico, A. Prota, C. Moroni, G. Manfredi, and M. Hak, S., P. Morandi, and G. Magenes. 2013. Damage control of masonry.
Dolce. 2017. “Estimation of repair costs for RC and masonry residential Infills in seismic design. Pavia, Italy: IUSS.
buildings based on damage data collected by post-earthquake visual in- Hermanns, L., A. Fraile, E. Alarcón, and R. Álvarez. 2014. “Performance
spection.” Bull. Earthquake Eng. 15 (4): 1681–1706. https://doi.org/10 of buildings with masonry infill walls during the 2011 Lorca earth-
.1007/s10518-016-0039-9. quake.” Bull. Earthquake Eng. 12 (5): 1977–1997. https://doi.org/10
De Risi, M. T., A. Furtado, H. Rodrigues, J. Melo, G. M. Verderame, A. .1007/s10518-013-9499-3.
Arêde, H. Varum, and G. Manfredi. 2019. “Experimental analysis of Holmes, M. 1961. “Steel frames with brickwork and concrete infilling.”
textile reinforced mortars strengthening strategies against the Proc. Inst. Civ. Eng. 19 (4): 473–478. https://doi.org/10.1680/iicep
out-of-plane collapse of masonry infill walls.” In Proc., 18th .1961.11305.
CONVEGNO ANIDIS, SG13–57. Pisa, Italy: Pisa University Press srl. INSYSME. 2014. Report on types of structural frames, related enclosure
De Risi, R., A. Sextos, P. Zimmaro, A. Simonelli, and J. P. Stewart. 2018. wall systems, and requirements for the construction systems.
“The 2016 central Italy earthquake sequence: Observations of incre- INSYSME Project D3.1. Padova, Italy: INSYSME.
mental building damage.” In Proc., 11th U.S. National Conf. on Ismail, N., T. El-Maaddawy, N. Khattak, K. Q. Walsh, and J. M. Ingham.
2018. “Out-of-plane behaviour of in-plane damaged masonry infills
Earthquake Engineering. Los Angeles, California: Earthquake

Engineering Research Institute (EERI). strengthened using fibre reinforced matrix.” In Proc., Int. Masonry
Di Ludovico, M., A. Prota, C. Moroni, G. Manfredi, and M. Dolce. 2017a. Society Conf., 2200–2216. Milan, Italy: International Masonry
“Reconstruction process of damaged residential buildings outside his- Society (IMS) and Technical University of Milan.
torical centres after the L’Aquila earthquake: Part I—“light damage” re- Kakaletsis, D. J., K. N. David, and C. G. Karayannis. 2011. “Effectiveness
construction.” Bull. Earthquake Eng. 15 (2): 667–692. https://doi.org of some conventional seismic retrofitting techniques for bare and in-
/10.1007/s10518-016-9877-8. filled R/C frames.” Struct. Eng. Mech. 39 (4): 499–520. https://doi
Di Ludovico, M., A. Prota, C. Moroni, G. Manfredi, and M. Dolce. 2017b. .org/10.12989/sem.2011.39.4.499.
“Reconstruction process of damaged residential buildings outside his- Kakaletsis, D. J., and C. G. Karayannis. 2008. “Influence of masonry
torical centres after the L’Aquila earthquake: Part II—“heavy damage” strength and openings on infilled R/C frames under cycling loading.”
reconstruction.” Bull. Earthquake Eng. 15 (2): 693–729. https://doi.org J. Earthquake Eng. 12 (2): 197–221. https://doi.org/10.1080
/10.1007/s10518-016-9979-3. /13632460701299138.
Doğ angün, A., A. Ural, H. Sezen, Y. Güney, and F. K. Fırat. 2013. “The Kakaletsis, D. J., and C. G. Karayannis. 2009. “Experimental investigation
2011 earthquake in Simav, Turkey and seismic damage to reinforced of infilled reinforced concrete frames with openings.” ACI Struct. J.
concrete buildings.” Buildings 3 (1): 173–190. https://doi.org/10.3390 106 (2): 132–141.
/buildings3010173. Kalali, A. 2012. “Cyclic behavior of perforated masonry walls strengthened
Dolce, M., and A. Goretti. 2015. “Building damage assessment after the with glass fiber reinforced polymers.” Scientia Iranica 19 (2): 151–165.
2009 Abruzzi earthquake.” Bull. Earthquake Eng. 13 (8): 2241–2264. https://doi.org/10.1016/j.scient.2012.02.011.
https://doi.org/10.1007/s10518-015-9723-4. Koutas, L. N., and D. A. Bournas. 2019. “Out-of-plane strengthening of
Donà, M., M. Minotto, E. Bernardi, E. Saler, N. Verlato, and F. da Porto. masonry-infilled RC frames with textile-reinforced mortar jackets.”
2019. “Macro-modelling of combined in-plane and out-of-plane seis- J. Compos. Constr. 23 (1): 04018079. https://doi.org/10.1061/(ASCE)
mic response of thin strengthened masonry infills.” In Proc., 7th Int. CC.1943-5614.0000911.
Conf. on Computational Methods in Structural Dynamics and Lunn, D. S., and S. H. Rizkalla. 2011. “Strengthening of infill masonry
Earthquake Engineering, 2449–2463. Athens, Greece: Institute of walls with FRP materials.” J. Compos. Constr. 15 (2): 206–214.
Structural Analysis and Antiseismic Research School of Civil https://doi.org/10.1061/(ASCE)CC.1943-5614.0000088.
Engineering, National Technical University of Athens (NTUA). Mainstone, R. J. 1971. “On The stiffness and strengths of infilled frames.”
Donà, M., M. Minotto, E. Saler, G. Tecchio, and F. da Porto. 2017. Proc. Inst. Civ. Eng. 49 (2): 57–90. https://doi.org/10.1680/iicep.1971.6267.
“Combined in-plane and out-of-plane seismic effects on masonry infills Mainstone, R. J., and G. A. Weeks. 1970. “The influence of a bounding frame
in RC frames.” Int. J. Earthquake Eng. 34: 157–173. on the racking stiffness and strengths of brick walls.” In Proc., 2nd Int.
Drysdale, R. G., and A. S. Essawy. 1988. “Out-of-Plane bending of con- Brick Masonry Conf., Building Research Establishment, 165–171.
crete block walls.” J. Struct. Eng. 114 (1): 121–133. https://doi.org/10 Stoke-on-Trent, UK: British Ceramic Research Association.
.1061/(ASCE)0733-9445(1988)114:1(121). Manos, G., D. Naxakis, and V. Soulis. 2015. “The dynamic and earthquake
Drysdale, R. G., A. A. Hamid, and L. R. Baker. 1999. Masonry structures: response of a two story old r/c building with masonry infills in
Behaviour and design. Boulder, CO: Prentice Hall. Lixouri-Kefalonia, Greece including soil-foundation deformability.”
El-Dakhakhni, W. W., A. A. Hamid, Z. H. R. Hakam, and M. Elgaaly. In Proc., 5th Int. Conf. on Computational Methods in Structural
2006. “Hazard mitigation and strengthening of unreinforced masonry Dynamics and Earthquake Engineering, 435–459. Athens, Greece:
walls using composites.” Compos. Struct. 73 (4): 458–477. https://doi Institute of Structural Analysis and Antiseismic Research School of
.org/10.1016/j.compstruct.2005.02.017. Civil Engineering, National Technical University of Athens (NTUA).
FEMA. 1999. “Evaluation of earthquake damaged concrete and masonry Marinković , M., and C. Butenweg. 2019. “Innovative decoupling system
wall buildings—Basic procedures manual.” Chap. 8 in Infilled frames, for the seismic protection of masonry infill walls in reinforced concrete
183–214. FEMA-306. Washington, DC: FEMA. frames.” Eng. Struct. 197: 109435. https://doi.org/10.1016/j.engstruct
Fiorato, A. E., M. A. Sozen, and W. L. Gamble. 1970. An investigation of .2019.109435.
the interaction of RC frames with masonry filler walls. Technical Martins, A., G. Vasconcelos, R. Fangueiro, and F. Cunha. 2015.
Report. Champaign, IL: Univ. of Illinois at Urbana-Champaign. “Experimental assessment of an innovative strengthening material for
Flanagan, R. D., and R. M. Bennett. 1999. “Bidirectional behavior of struc- brick masonry infills.” Composites, Part B 80: 328–342. https://doi
tural clay tile infilled frames.” J. Struct. Eng. 125 (3): 236–244. https:// .org/10.1016/j.compositesb.2015.06.012.
doi.org/10.1061/(ASCE)0733-9445(1999)125:3(236). Masi, A., L. Chiauzzi, G. Santarsiero, M. Liuzzi, and V. Tramutoli. 2017.
Fragomeli, A., et al. 2017. “Performance of masonry buildings in the seis- “Seismic damage recognition based on field survey and remote sens-
mic sequence of Central Italy 2016—Part 1 : Overview.” Progettazione ing: General remarks and examples from the 2016 Central Italy earth-
Sismica 8 (2): 49–77. quake.” Nat. Hazards 86 (S1): 193–195. https://doi.org/10.1007
Furtado, A., H. Rodrigues, A. Arêde, and H. Varum. 2016. /s11069-017-2776-8.
“Experimental evaluation of out-of-plane capacity of masonry Masi, A., L. Chiauzzi, G. Santarsiero, V. Manfredi, S. Biondi, E. Spacone,
infill walls.” Eng. Struct. 111: 48–63. https://doi.org/10.1016/j C. Del Gaudio, P. Ricci, G. Manfredi, and G. M. Verderame. 2019.
.engstruct.2015.12.013. “Seismic response of RC buildings during the Mw 6.0 August 24,
Gazic, G., and V. Sigmund. 2016. “Cyclic testing of single-span weak 2016 central Italy earthquake: The amatrice case study.” Bull.
frames with masonry infill.” J. Croatian Assoc. Civ. Eng. 68 (8): Earthquake Eng. 17 (10): 5631–5654. https://doi.org/10.1007/s10518
617–633. https://doi.org/10.14256/JCE.1614.2016. -017-0277-5.
J. Compos. Constr., 2020, 24(6): 04020059

Mehrabi, A. B., P. Benson Shing, M. P. Schuller, and J. L. Noland. Saatcioglu, M., F. Serrato, and S. Foo. 2005. “Seismic performance of
1996. “Experimental evaluation of masonry-infilled RC frames.” masonry infill walls retrofitted with CFRP sheets.” In Proc., 7th Int.
J. Struct. Eng. 122 (3): 228–237. https://doi.org/10.1061/(ASCE) Symp. on Fiber-Reinforced Polymer Reinforcement for Concrete
0733-9445(1996)122:3(228). Structures, 341–354. Farmington Hills, Michigan: American
Mehrabi, A. B., P. B. Shing, M. P. Schuller, and J. L. Noland. 1994. Concrete Institute.
Performance of masonry-infilled RIC frames under in-plane lateral Sagar, S. L., V. Singhal, and D. C. Rai. 2019. “In-plane and out-of-plane
loads. Technical Report. Boulder, CO: Univ. of Colorado at Boulder. behavior of masonry-infilled RC frames strengthened with
MIT (Ministry of Infrastructure and Transport). 2008. New technical stan- fabric-reinforced cementitious matrix.” J. Compos. Constr. 23 (1):
dards for constructions 2008. DM 2008/01/14, “S.O. No. 30 alla G.U. 04018073. https://doi.org/10.1061/(ASCE)CC.1943-5614.0000905.
del 4 Febbraio 2008, No. 29”. [In Italian.] Rome: MIT. Silva, L. M., G. Vasconcelos, P. B. B. Lourenço, and F. Akhoundi. 2016.
MIT (Ministry of Infrastructure and Transport). 2018. Update of the “Experimental evaluation of a constructive system for
“technical standards for constructions” 2018. DM 2018/01/17, earthquake-resisting masonry enclosure walls.” In Proc., 16th Int.
“S.O. No. 8 alla G.U. del 20 Febbraio 2018, No. 32”. [In Italian.] Brick and Block Masonry Conf., 1333–1340. Leiden, The
Rome: MIT. Netherlands: Taylor & Francis Group.
MIT (Ministry of Infrastructure and Transport). 2019. Instructions for the Stafford-Smith, B. 1962. “Lateral stiffness of infilled frames.” J. Struct.
application of the “update of the “technical standards for construc- Div. 88 (6): 183–226.
tions” 2018”. Circ. 2019/01/21 no. 7, “S.O. No. 5 alla G.U. del 11 Thomas, F. C. 1952. “The strength of brickwork.” J. Struct. Eng. 31 (2):
Febbraio 2019, No. 35”. [In Italian.] Rome: MIT. 35–46.
Mohammadi, M., and V. Akrami. 2010. “An engineered infilled frame: Tu, Y.-H., T. H. Chuang, P.-M. Liu, and Y. S. Yang. 2010. “Out-of-plane
Behavior and calibration.” J. Constr. Steel Res. 66 (6): 842–849. shaking table test on unreinforced masonry panels in RC frame.” Eng.
https://doi.org/10.1016/j.jcsr.2010.01.008. Struct. 32 (12): 3925–3935. https://doi.org/10.1016/j.engstruct.2010
Mohammadi, M., V. Akrami, and R. Mohammadi-Ghazi. 2011. “Methods .08.030.
to improve infilled frame ductility.” J. Struct. Eng. 137 (6): 646–653. Tumialan, J. G., N. Galati, and A. Nanni. 2003. “Fiber-reinforced polymer
https://doi.org/10.1061/(ASCE)ST.1943-541X.0000322. strengthening of unreinforced masonry walls subject to out-of-plane
Morandi, P., R. R. Milanesi, and G. Magenes. 2018. “Innovative solution loads.” ACI Struct. J. 100 (3): 321–329.
for seismic-resistant masonry infills with sliding joints: In-plane exper- Valluzzi, M. R., F. da Porto, E. Garbin, and M. Panizza. 2014.
imental performance.” Eng. Struct. 176: 719–733. https://doi.org/10 “Out-of-plane behaviour of infill masonry panels strengthened with
.1016/j.engstruct.2018.09.018. composite materials.” Mater. Struct. 47 (12): 2131–2145. https://doi
Palieraki, V., C. Zeris, E. Vintzileou, and C.-E. Adami. 2018. “In-plane .org/10.1617/s11527-014-0384-6.
and out-of plane response of currently constructed masonry infills.” Vasconcelos, G., S. Abreu, R. Fangueiro, and F. Cunha. 2012. “Retrofitting
Eng. Struct. 177: 103–116. https://doi.org/10.1016/j.engstruct.2018 masonry infill walls with textile reinforced mortar.” In Proc., 15th
.09.047. World Conf. on Earthquake Engineering. 30588–30597. Red Hook,
Papanicolaou, C. G., T. C. Triantafillou, M. Papathanasiou, and K. Karlos. NY: Curran Associates, Inc.
2007. “Textile reinforced mortar (TRM) versus FRP as strengthening Verderame, G. M., P. Ricci, F. De Luca, C. Del Gaudio, and M. T. De Risi.
material of URM walls: Out-of-plane cyclic loading.” Mater. Struct. 2014. “Damage scenarios for RC buildings during the 2012 Emilia
41 (1): 143–157. https://doi.org/10.1617/s11527-007-9226-0. (Italy) earthquake.” Soil Dyn. Earthquake Eng. 66: 385–400. https://
Penna, A., P. Morandi, M. Rota, C. F. Manzini, F. da Porto, and G. doi.org/10.1016/j.soildyn.2014.06.034.
Magenes. 2014. “Performance of masonry buildings during the emilia Verlato, N., G. Guidi, and F. da Porto. 2014. “Calcolo della resistenza
2012 earthquake.” Bull. Earthquake Eng. 12 (5): 2255–2273. https:// fuori piano di tamponature realizzate con il sistema Antiespulsione :
doi.org/10.1007/s10518-013-9496-6. Modellazione analitica e proposte progettuali.” Murature Oggi
Pereira, M. F. P., M. F. N. Pereira, J. E. D. Ferreira, and P. B. Lourenço. 116 (3): 8–16.
2011. “Behavior of masonry infill panels in RC frames subjected to Verlato, N., G. Guidi, F. da Porto, and C. Modena. 2016. “Innovative sys-
in plane and out of plane loads.” In Proc., 7th Int. Conf. AMCM. tems for masonry infill walls based on the use of deformable joints:
Krakow, Poland: Wydawnictwo Politechniki Krakowskiej im. Combined in-plane/out-of-plane tests.” In Proc., 16th Int. Brick and
Tadeusza Koś ciuszki. Block Masonry Conf., 1359–1366. Leiden, The Netherlands: Taylor
Polyakov, S. V. 1960. “On the interaction between masonry filler walls and & Francis Group.
enclosing frame when loaded in plane of the wall.” In Translations in Vicente, R., H. Rodriguez, A. Costa, H. Varum, and J. A. R. Mendes da
Earthquake Engineering, 36–42. Oakland, CA: Earthquake Silva. 2010. “Masonry enclosure walls: Lessons learnt from the recent
Engineering Research Institute. Abruzzo Earthquake.” In Proc., 14th European Conf. of Earthquake
Preti, M., L. Migliorati, and E. Giuriani. 2015. “Experimental testing of en- Engineering, 1094–1101. Red Hook, NY: Curran Associates, Inc.
gineered masonry infill walls for post-earthquake structural damage Vintzileou, E., C.-E. Adami, and V. Palieraki. 2016. “In-plane and
control.” Bull. Earthquake Eng. 13 (7): 2029–2049. https://doi.org/10 out-of-plane response of a masonry infill divided into smaller wallettes.”
.1007/s10518-014-9701-2. In Proc., 16th Int. Brick and Block Masonry Conf., 1367–1373. Leiden,
Ricci, P., M. Di Domenico, and G. M. Verderame. 2018. “Experimental as- The Netherlands: Taylor & Francis Group.
sessment of the in-plane/out-of-plane interaction in unreinforced ma- Vintzileou, E., and V. Palieraki. 2007. “Perimeter infill walls: The use
sonry infill walls.” Eng. Struct. 173: 960–978. https://doi.org/10.1016 of bed joint reinforcement or RC tie-beams.” Masonry Int. 20 (3):
/j.engstruct.2018.07.033. 117–128.
Ricci, P., F. De Luca, and G. M. Verderame. 2011. “6th April 2009 Zarnic, R., and M. Tomazevic. 1988. “An experimentally obtained method
L’Aquila earthquake, Italy: Reinforced concrete building performance.” for evaluation of the behavior of masonry infilled RC frames.” In Proc.,
Bull. Earthquake Eng. 9 (1): 285–305. https://doi.org/10.1007/s10518 9th World Conf. on Earthquake Engineering, 163–168. Tokyo: Japan
-010-9204-8. Association for Earthquake Disaster Prevention.
J. Compos. Constr., 2020, 24(6): 04020059

ASCE 2020 - Strengthening of In-Plane and Out-of-Plane Capacity of Thin Clay Masonry Infills Using Textile and Fiber-Reinforced Mortar

Uploaded by

Copyright:

Available Formats

ASCE 2020 - Strengthening of In-Plane and Out-of-Plane Capacity of Thin Clay Masonry Infills Using Textile and Fiber-Reinforced Mortar

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ASCE 2020 - Strengthening of In-Plane and Out-of-Plane Capacity of Thin Clay Masonry Infills Using Textile and Fiber-Reinforced Mortar

Uploaded by

Copyright:

Available Formats

Strengthening of In-Plane and Out-of-Plane Capacity

of Thin Clay Masonry Infills Using Textile-

Introduction damage of the inﬁll, induced by the RC frame deformation, yields

© ASCE 04020059-1 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

works of Angel et al. (1994), Flanagan and Bennett (1999),

© ASCE 04020059-2 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 1. Strengthening solutions (F, FB, RBB).

Fig. 2. Application of RBB. (Images by Nicolò Verlato.)

© ASCE 04020059-3 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)

© ASCE 04020059-4 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)

Table 3. Compression test results (CoV in brackets)

(a) (b) (c)

© ASCE 04020059-5 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)

(a) (b) (c)

© ASCE 04020059-6 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 8. RC frame geometry with dimensions in mm.

(a) (b) (c)

© ASCE 04020059-7 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 11. Crack pattern and view of specimens at end of IP tests.

© ASCE 04020059-8 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 12. Hysteresis loops and envelope curves of IP tests.

© ASCE 04020059-9 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

© ASCE 04020059-10 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Table 5. In-plane test results

© ASCE 04020059-11 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 18. OOP force-displacement curves.

plaster of lower mechanical properties. Indeed, specimens RBB.05 Out-of-Plane Tests

© ASCE 04020059-12 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

(a) (b) (c)

(a) (b) (c)

© ASCE 04020059-13 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Influence of In-Plane Damage on Out-of-Plane

© ASCE 04020059-14 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 23. Out-of-plane strength degradation due to in-plane damage.

function of the in-plane drift θIP, has been proposed:

where a = 1.65 and b = 2.80. The reduction factor can be proposed

© ASCE 04020059-15 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

Fig. 25. Out-of-plane capacity models (1 and 2).

© ASCE 04020059-16 J. Compos. Constr.

J. Compos. Constr., 2020, 24(6): 04020059

where ρ = slenderness parameter dependent on the hw/tw Rc (x) = αβfmd Lw x (8)