Engineering Structures 156 (2018) 503–521

Contents lists available at ScienceDirect

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

Performance-based interpretation of in-plane cyclic tests on RC frames with T

strong masonry infills
⁎
Paolo Morandia, , Sanja Hakb, Guido Magenesa
a
University of Pavia and EUCENTRE, Via Ferrata 1, 27100 Pavia, Italy
b
University of Zagreb, Croatia and University of Pavia, Italy

A R T I C L E I N F O A B S T R A C T

Keywords: In line with the current European building practice, clay masonry infills are commonly adopted for the con-
Masonry infills struction of enclosures and partitions in RC frame structures. In order to improve further the understanding of
In-plane cyclic tests the seismic response of masonry infills in newly designed RC structures, within the scope of a systematic nu-
Experimental response merical and experimental research program, the in-plane experimental response of external masonry infills
In-plane damage
constructed with tongue and groove clay units has been studied. Particular attention has been devoted to the
Performance levels
infill performance, the related damage distribution, the lateral stiffness, strength and dissipation capacity and a
possible definition of performance limit states for future developments in the current design approach. This
paper reports the framework and discusses the results of in-plane static cyclic tests on full-scale, single-storey,
single-bay RC frame specimens with strong masonry infills with and without opening designed following
European seismic design provisions, carried out at the laboratory of the Department of Civil Engineering and
Architecture of the University of Pavia. Based on the observations during the tests, drift levels at the different
limit states have been evaluated, with the attainment of large values for the solid specimen (0.30%, 0.50% and
1.75% at the operation, damage limitation and ultimate limit state, respectively), but lower performance for the
panel with opening. The experimentally obtained lateral stiffness and strength have been compared with the
results of some of the most common analytical models based on single diagonal equivalent strut, underlining the
formulations that better fit the test outcomes.

1. Introduction
The construction of RC frame structures with masonry infills is part
of the traditional building practice in many European countries but also
worldwide. As interior partitions, exterior enclosures or façade veneers,
unreinforced masonry infills are often found to be the preferred solution
to appropriately fulfil architectural needs; durability, efficient sound
and thermal insulation of masonry, in addition to simple and cost-ef-
fective construction, are often underlined as advantages of this con-
struction technique. Masonry panels are commonly constructed in
complete contact with the surrounding RC frame, without the provi-
sions of any gaps or connections around the boundaries. Since the infills
are normally placed only after the surrounding RC frames have har-
dened, they are assumed as non-load bearing and in European design
practice commonly treated as non-structural elements. However, even
though the elements are deemed non-structural, damage to masonry
infills induced by earthquake ground motions can represent a threat to
human lives and lead to extensive economic losses.
The seismic vulnerability of inherently rather brittle masonry infills
is generally recognized and has been observed many times by field
experiences reported after damaging earthquakes, once more in recent
years, as in L’Aquila 2009 (see e.g., Ricci et al. [1]), in Emilia 2012 (see
e.g., Manzini and Morandi [2]) and in the very recent events in Central
Italy 2016 (see e.g., Morandi [3], Fig. 1). Nevertheless, although the
significance of specific measures for the damage control of masonry
infills has been widely acknowledged, code procedures for the design of
new buildings currently provide only few recommendations for non-
structural elements, still being in many aspects insufficient, incomplete
or not clearly defined.
For these reasons, in the past a significant number of numerical and
experimental investigations related to the general seismic performance
of masonry infilled frames has been carried out, however often focusing
on the analysis and assessment of RC frames, more than the effective
damage control of infills in newly designed structures. Moreover, the
experimental response of infills has been more extensively studied only
for selected masonry typologies, like clay bricks and concrete blocks
(i.e., Angel and Abrams [4], Meharabi and Shing [5], Mosalam et al.
[6]) and very limited experimental tests have been performed on real

⁎
Corresponding author.
E-mail address: paolo.morandi@unipv.it (P. Morandi).

https://doi.org/10.1016/j.engstruct.2017.11.058
Received 7 March 2017; Received in revised form 21 November 2017; Accepted 22 November 2017
0141-0296/ © 2017 Elsevier Ltd. All rights reserved.
P. Morandi et al.
Fig. 1. Examples of earthquake damage on external masonry infills (Amatrice, Italy – 2016).
scale infilled RC specimens (e.g., Calvi and Bolognini [7], Pujol et al.
[8], da Porto et al. [9]), as documented in Sassun et al. [10], where a
collection of the outcomes of several in-plane tests on infills performed
around the world has been summarized. Among these, only the research
reported in da Porto et al. [9] has been devoted to the study of masonry
infills with strong perforated clay units characterized by medium/large
thicknesses (i.e., larger than 25–30 cm) which, however, nowadays in
Europe are being adopted with increasing frequency, even in earth-
quake prone regions, in order to exploit their excellent thermal and
acoustic performances necessary for the fulfilment of the recent Eur-
opean regulations on the building energy efficiency and environmental
sustainability. On the other hand, specific attention needs to be devoted
on possible detrimental effects due to the thrust of the infill on the
frame, which can be particularly prominent for thick/strong masonry
typologies with the risk of generating local brittle failure on the RC
elements.
Moreover, the appropriate analysis and verification of masonry in-
fills plays a substantial role for the safe and efficient seismic design of
masonry infilled RC structures and modern codified approaches are
based on safety checks in terms of inter-storey displacement on bare
frame/infilled structures [11]. In order to assess and verify the expected
structural drift demands, which are associated to the level of damage
due to in-plane actions, during the design of new buildings, different
levels of drift, representing relevant performance limit state conditions,
need to be known. Therefore, the further evaluation of drift capacities
for commonly adopted masonry infill typologies is regarded as a subject
of considerable importance. However, not only values of drift limits are
not uniquely recognized, but even a commonly accepted definition of
possible damage limit states specifically related to performance re-
quirements for masonry infills has not yet been established. Further-
more, related specifications included in European and International
seismic standards and codes are rather inaccurate and often only re-
ferred to structural components; for instance, in EC8-Part 1 [12] in-
Fig. 2. Fully infilled (TA1, TA2 and TA3), partially infilled (TA4) and vertical infill stripe (TA5).
504
Engineering Structures 156 (2018) 503–521
plane drift verifications are only enforced for damage limitation re-
quirements in terms of inter-storey drift of the structure, to be limited at
0.50% or 0.75% respectively for “brittle” and “ductile” non-structural
elements attached to the structure.
Consequently, motivated by the necessity to improve further the
seismic design approach currently adopted for new RC structures, the
in-plane seismic response of rigid strong masonry infills characterized
by vertically perforated 35 cm thick clay units has been studied, based
on a systematic experimental campaign conducted on full-scale speci-
mens. The framework of the experimental work is presented and some
of the principal research objectives related to the in-plane response are
addressed, in particular referring to the examination of infill damage
patterns, failure mechanisms, energy dissipation and deformation ca-
pacity, along with the definition of performance levels suitable for in-
fills; a correlation of the test results with analytical expressions for infill
strength and stiffness is also provided. Infill-frame interaction effects
induced by the local thrust of the infill on the RC columns are also
briefly discussed.
2. Experimental campaign
2.1. Description of the test specimens
Within the scope of the experimental campaign, a series of cyclic
static in-plane and out-of-plane tests has been carried out at the la-
boratory of the Department of Civil Engineering and Architecture of the
University of Pavia on bare and fully or partially infilled full-scale
single-storey, single-bay RC frames (see Fig. 2), designed according to
European (and Italian) code provisions.
After a detailed characterization of all material components (i.e.,
concrete, reinforcing steel, mortar, masonry units and masonry), the
experimentation has been accomplished on six frame specimens, as
summarized in Table 1. Specifically, related to the cyclic in-plane tests,
P. Morandi et al. Engineering Structures 156 (2018) 503–521

Table 1 2.2. Tests of characterization on material components and masonry

Summary of performed experiments and achieved maximum drift or displacement levels. specimens
Infilled frame
A complete characterization of the relevant properties for all ma-
No. In-plane Out-of-plane Configuration terials utilized for the construction of the specimens has been carried
out, as reported in more detail in Morandi et al. [13]. Principally, the
TNT 3.50% – Bare frame
evaluation of clay unit, mortar and masonry properties for the selected
TA1 1.50% 75 mm Fully infilled
TA2 2.50% 75 mm Fully infilled infill typology was of primary importance. Additionally, the material
TA3 1.00% 75 mm Fully infilled characterization included compression tests on 6 concrete cubes (of
TA4 1.00% 75 mm Partially infilled 150 × 150 × 150 mm dimensions) for columns and 6 for beams, and
TA5 – 75 mm Infill stripe tension tests on reinforcement rebars sampled during the construction
phase; the related results are reported in Table 2.
Tests for the evaluation of the compressive strength of the clay units
one RC frame was tested without infill (TNT) up to maximum drift of
under vertical (fb, on 30 units) and lateral compression (fb′, on 10 units)
3.50% in order to reach the ultimate conditions of the specimen, while
according to EN 772-1 [18], as well as for the determination of flexural
three fully infilled specimens (TA1, TA2 and TA3) were tested at three
(ffl) and compressive strength (fm) of mortar on 6 hardened prisms (of
increasing maximum levels of drift, equal to 1.00, 1.50 and 2.50%. In
40 × 40 × 160 mm dimensions) according to EN 1015-11 [19] were
addition, a partially infilled frame configuration with a 1.44 m wide
carried out. Vertical and lateral compression tests on six masonry
and full-height opening in the middle of the span (TA4) was tested,
wallets were executed in accordance to EN 1052-1 [20] for the eva-
reaching a maximum in-plane drift of 1.00%. The in-plane infill per-
luation of the compression strength (fv, flat) and the elastic modulus (Ev,
formance at increasing levels of drift was aimed to approximately re-
Elat) of the masonry, the latter computed at a stress value of one third of
present different limit state conditions. The out-of-plane experiments
the maximum strength. Finally, tests for the determination of the initial
have been carried out on the specimens previously damaged in-plane,
shear strength (fv0) were performed according to EN 1052-3 [21],
in order to evaluate the related out-of-plane resistance reduction.
where three samples made of three units (triplets) were tested in the
Moreover, a 1.38 m wide stripe of the infill (TA5) was tested in the out-
direction parallel to the bed-joints, at three levels of horizontal com-
of-plane direction, with the aim to evaluate the out-of-plane strength
pression (fp), for a total of 9 triplets. The results of the mechanical
under vertical single-bending conditions.
characterization on units, mortar and masonry are summarized in
The dimensions of the RC frame specimen have been chosen with
Table 2.
the aim to realistically represent the part of a full-scale RC structure; a
clear span and height of 4.22 m and 2.95 m, respectively have been
2.3. In-plane test setup, instrumentation and testing protocol
adopted. Beam and column lengths have been extended beyond the
beam-column panel zone to provide sufficient space for the appropriate
For the in-plane cyclic static tests, a horizontal force has been ap-
anchorage of reinforcement rebars and to facilitate the introduction of
plied on the beam of the RC frame by means of a servo-controlled hy-
forces during the test. For the foundation of the frame an inverted T-
draulic actuator with an internal load cell, transferring the reaction to
shaped cross section has been adopted. The design of the RC frame
the existing reaction wall in the laboratory. To transmit the horizontal
specimen has been carried out considering the single-storey single-bay
force to the frame during reverse loading cycles, a system of two steel
frame being at the ground storey of a simple 4-storey bare frame
plates and four prestressed rebars was applied, resulting in a pre-
structural configuration, regular in plan and elevation; the design has
compression of about 800 kN in the beam. A concentrated vertical load
been developed according to the European code provisions (EC8-Part 1
was applied on the RC columns by means of two independent hydraulic
[12], EC1-Part 1-1 [14] and EC2-Part 1-1 [15]), supplemented with the
jacks, each placed between the column and a rigid transversal steel
Italian national code for what concerns the design spectra (NTC08
beam tied down to the foundation with two steel bars, resulting in a
[16]), applying a ductility class high (DCH), a behaviour factor of
self-equilibrated vertical load introduction system. The complete layout
q = 5.85 and a design PGA at the ultimate limit state (ULS) of 0.38 g on
of the in-plane experimental setup is illustrated in Fig. 5(a). Possible
soil class B. Material properties corresponding to concrete class C28/35
out-of-plane displacements of the specimen during the in-plane cyclic
and C35/45 have been adopted for the frame and the foundation, re-
loading were prevented using an additional steel reaction frame that
spectively, and reinforcing steel class B450C has been assumed for all
has been designed for the subsequent out-of-plane tests.
RC elements. The reinforcement details of the RC frame are given in
In order to measure the displacements and deformations of the
Fig. 3.
specimen during the test, displacement transducers (linear potenti-
The selected infill typology represents a commonly adopted, tradi-
ometers) have been adopted. In total, 22 potentiometers have been used
tional strong single-leaf unreinforced masonry infill of 35 cm thickness,
for the bare frame (TNT), 24 for the fully infilled frames (TA1, TA2 and
consisting of vertically hollowed lightweight tongue and groove clay
TA3), and 32 for the partially infilled frame (TA4). Moreover, during
block units, having nominal dimensions of 235 × 350 × 235 mm, a
the test on the bare frame (TNT) and on two infilled frame configura-
nominal volumetric percentage of holes of 50% and a minimum
tions (TA1, TA2) the deformations of the reinforcement rebars in po-
thickness of webs and shells equal to 6.8 mm and 4.8 mm, respectively,
tential plastic hinge regions of the RC structural elements have been
see Fig. 4(a). The application of a general-purpose mortar type “M5”
monitored my means of strain gauges attached during the construction
(compressive strength of 5 MPa) was considered a suitable choice with
of the specimens (see Fig. 5(b)). Additionally, an optical acquisition
respect to common construction practice. The specific weight of this
system was installed to measure the in-plane displacements of optical
masonry typology is 8.80 kN/m3. The infills have been constructed after
markers uniformly distributed (ca. 40–50 cm spaced) on the RC frame
full hardening of the RC frame, adopting traditional bed joints, having a
and on the masonry infill. Details on the installed instrumentation is
thickness of about 1.0 cm and dry head joints. The mortar bed joints
illustrated in Morandi et al. [13,17].
were applied in two stripes in the longitudinal direction of the wall with
For each in-plane test, the vertical load has firstly been imposed on
an intermediate cavity of about 2 cm, see Fig. 4(b). Full contact be-
both columns simultaneously by the hydraulic jacks at a velocity of
tween the infill and the surrounding RC members was achieved filling
about 2.0 kN/s. The achieved vertical load of 400 kN per column has
the remaining vertical gaps on the two sides of the infill and the hor-
been kept constant during the entire test. Finally, reverse cycles of
izontal gap at the top of the infill with mortar, see Fig. 4(c).
horizontal in-plane loading (first pull, then push) have been imposed on
the frame. Initially, two different levels of force-controlled loading were

505
P. Morandi et al.
accomplished, with force levels defined as fraction of the estimated
maximum resistance to get sufficient points describing the ascending
branch of the force-displacement curve, while subsequently displace-
ment-controlled loading cycles at increasing levels of in-plane drift
were imposed. For each level of loading (in both force- and displace-
ment controlled), three complete reverse loading cycles have been
carried out. The loading histories of the tests are reported in
Fig. 5(c) to (g), where “F” identifies force-controlled, whereas “D”
displacement-controlled loading cycles. Further information on the
loading protocol is reported in Morandi et al. [13,17].
3. Experimental test results
3.1. In-plane force-displacement response
Results of the cyclic in-plane tests on infilled and bare frame
Fig. 4. (a) Masonry clay unit; (b) Mortar bed joint; (c) Mortar joints adjacent to RC members.
506
Engineering Structures 156 (2018) 503–521
Fig. 3. Reinforcement details of the RC frame.
specimens have been evaluated representing the applied force in
function of the top displacement measured at half-height of the top
beam that corresponds to the centreline of the horizontal actuator.
Specifically, force-displacement curves and the corresponding envel-
opes per cycle obtained for the bare frame specimen (TNT), for the fully
infilled specimens (TA1, TA2, TA3) and for the infill with opening
(TA4) are shown in Fig. 6(a), (b) to (d) and (e), respectively; in Fig. 6(f)
the maximum envelopes (envelopes of the first cycles) for all in-plane
tests are presented. The data of the force-displacement hysteresis are
freely shared in Morandi et al. [13].
The results for the three fully infilled frames point towards con-
sistent values of initial stiffness and a comparable development of
nonlinear response due to the onset of degradation. Strength and
stiffness degradation occurred during loading cycles repeated at the
same displacement, in particular during the second cycle of the se-
quence (force reduction of about 4–6% between the first and the second
P. Morandi et al.
Table 2
Mechanical properties of the concrete, reinforcing bars, clay units, mortar and masonry
used in the specimens.
Property Symbol Mean
Compr. strength on cubes of concrete C28/35 Rc,col 34.0 MPa
(columns)
Compr. strength on cubes of concrete C28/35 Rc,beam 34.3 MPa
(beams)
Yield. tensile strength of steel rebars B450C (col. fy 521 MPa
& beams)
Ult. tensile strength of steel rebars B450C (col. & fu 606 MPa
beams)
Vertical normalized compr. strength of units fb,norm 9.81 MPa
Lateral normalized compr. strength of units f'b,norm 3.15 MPa
Flexural strength of mortar ffl 2.15 MPa
Compression strength of mortar fm 7.68 MPa
Vertical compr. strength of masonry fvert 4.64 MPa
Vertical elastic modulus of masonry Evert 5299 MPa
Lateral compr. strength of masonry flat 1.08 MPa
Lateral elastic modulus of masonry Elat 494 MPa
Initial shear strength of bed-joints fv0 0.359 MPa
cycle envelopes). In the case of partial infill, an asymmetric response is
observed due to different strut activation in the two panels, as high-
lighted in the description of the damage in Section 3.2. A numerical
simulation of the in-plane experimental response of the bare and in-
filled frames has been conducted by Oliaee et al. [22].
3.2. Sequence of the damage pattern
The damage propagation observed during the tests is discussed in
more detail for the bare frame (TNT), for the fully infilled frame sub-
jected to the highest level of drift (TA2) and for the infilled frame with
opening (TA4).
Following two force-controlled loading levels (40.0 kN and
80.0 kN), specimen TNT has been loaded at a series of target dis-
placements, reaching a maximum drift of 3.50%. Up to a drift of 0.40%,
no damage could be observed on the RC elements. Initial cracks became
visible at 0.50% drift at the connection between the top of the columns
and the beam-column joint; at 0.60% drift, first flexural cracks formed
at the column ends and continued to propagate progressively during the
following cycles. Cracks on the RC beam-ends were clearly visible at a
drift of 1.75%. At a drift of 2.50%, the concrete cover started to spall off
at the bottom of the columns and at the beam ends. The damage at the
final drift of 3.50%, see Fig. 7(a), points to the formation of a weak
beam, strong column response mechanism, as confirmed also by the
obtained values of strain in the reinforcement rebars, measured at the
ends of the RC elements using the installed strain gauges, which did not
attain yielding at the column top sections.
After the two initial force-controlled loading levels (100.0 kN and
200.0 kN), specimen TA2 was loaded at several levels of increasing
drift, reaching a maximum of 2.50%. Up to 0.10% drift, only light
vertical cracks occurred on the surface of the plaster covering the joint
between the masonry infill and the adjacent RC elements. First pieces of
a thin layer of about 10 cm wide plaster placed at the RC frame/panel
interface detached at 0.30% drift, see Fig. 7(b). Attaining a drift of
0.50%, see Fig. 7(c), pieces of outer shells of damaged blocks in the
upper left corner started to detach. Increasing the target drift to 0.80%,
first diagonal cracks were observed in the upper parts of the columns,
which continued to propagate at a drift of 1.00%, see Fig. 7(d). At
1.50% and in particular at 1.75% drift, see Fig. 7(e), the majority of
masonry blocks in the top course and several blocks in the lower central
part were strongly damaged. The width of existing column shear cracks
remained stable and no implications of a column shear failure were
observed. Reaching 2.00% drift, the infill was considerably damaged,
while after the last cycles, attaining 2.50% drift, see Fig. 7(f), the infill
Engineering Structures 156 (2018) 503–521
was substantially destroyed. For the other two fully infilled frame
specimens (TA1 and TA3), up to the maximum achieved drift (1.50%
and 1.00%, respectively), similar crack patterns as in the case of spe-
c.o.v. cimen TA2 were obtained.
For the partially infilled specimen (TA4), following the two levels of
3.6% force-controlled loading (≈60.0 kN and ≈120.0 kN), ten different
target displacements up to a maximum drift of 1.00% were applied. At
3.1%
the first cycles, light vertical cracks were observed on the plaster be-
5.2% tween the columns and the infill. At 0.20% of drift, see (a), the cracks
on the interface between the infill and the RC column extended further
3.2% and a crack formed in the bottom left corner above the foundation. At
0.40% drift, see Fig. 8(b), during the first cycle of loading in the ne-
9.2% gative direction an evident diagonal crack crossing the blocks in the left
10.8%
panel formed. The diagonal crack closed again during unloading, but at
19.4%
20.6% the following cycles, notable strength degradation was observed. At
14.1% 0.50% drift, see Fig. 8(c), the pronounced diagonal crack extended
8.6% noticeably. Nevertheless, almost equal values of force could be
14.5% achieved for both loading directions, although the force-displacement
32.8%
26.0%
curves reveals an unsymmetrical behaviour. On the bottom left side of
the right panel, diagonal cracking occurred, but no full crack along the
entire diagonal of the panel formed. Subsequently, at the 0.60% drift
loading cycles, a stepwise crack pattern developed starting from the
bottom left corner and reaching up to mid-height of the right panel. At
the last level of imposed drift (1.00%), see Fig. 8(d), the diagonal crack
on the substantially damaged left panel widened additionally, while no
complete diagonal strut formed in the right panel.
3.3. Deformed shapes of bare and infilled RC frames
Processing the results of the optical acquisition system, the de-
formed shapes of the bare frame (TNT), the fully infilled frame (TA2)
and the partially infilled frame (TA4) have been compared for two
different applied horizontal drift levels, i.e., at 1.00% and at 2.00%
drift, as illustrated in Fig. 9 (the infills of specimens TA2 and TA4 are
not shown).
A substantial difference in the deformed shapes of the RC elements
of the bare and the infilled specimens can be observed, showing that,
due to the contribution of the infill, the in-ward beam-column joint (left
joint in Fig. 9) rotates clockwise for specimen TNT and anticlockwise
for TA2, while the beams of the infilled specimens show a tendency to
deflect upwards, which is more pronounced for the full infill than for
the infill with opening. Such behaviour indicates a different distribution
of internal forces in the RC members of the bare frame, usually adopted
for seismic design of RC infilled structures, and of the corresponding
infilled configurations. However, it is important to underline that these
deformed shapes may differ from the ones of multi-storey buildings due
to different boundary conditions, in particular in relation with the
loading of the beam.
4. Evaluation of the main seismic structural parameters
4.1. Interpretation of the results from in-plane cyclic tests
The main structural parameters, significant for the seismic response,
such as the secant stiffness (k), the maximum force (F), the drift at
different force levels (δ) and the energy dissipation capacity (ξeq) have
been evaluated for all tested specimens.
A possible solution to evaluate the experimental stiffness and
strength values of the infill itself consists in the assessment of the
average infill contribution as the difference between the experimental
envelope of the infilled frame and the corresponding bare frame con-
figuration at corresponding displacements, as reported in detail in Hak
et al. [23] and also inferred in other past studies (i.e., Cavaleri et al.
[24], Sigmund and Penava [25], Tasnini and Mohebkhah [26]). Such
simplified approach allows a consistent comparison of the experimen-
tally obtained infill properties in terms of strength, stiffness and
507
P. Morandi et al.
Fig. 5. (a) Layout of the in-plane setup for cyclic static tests and loading history; (b) strain gauges; (c) TNT, (d) TA1, (e) TA2, (f) TA3 and (g) TA4.
deformation capacity to similar results for other masonry typologies, as
well as the experimentally based definition of parameters for simple
infill models founded on the widely adopted single diagonal strut
modelling approach. Moreover, even though in reality the effective
stiffness and distribution of internal forces due to lateral action in RC
members of an infilled frame differ from those in the corresponding
bare frame, at equal levels of imposed displacement, if a single strut
508
Engineering Structures 156 (2018) 503–521
model is considered, the RC elements of an infilled frame and those of
the corresponding bare frame provide the same contribution to the
stiffness and the same distribution of shear and bending moment.
Therefore, the simplified experimental evaluation of the infill con-
tribution can be, in principle, considered applicable for the given ob-
jectives.
The curves with maximum envelopes of specimens TNT, TA2 and
P. Morandi et al. Engineering Structures 156 (2018) 503–521

(a) (b)

(c) (d)

(e) (f) DriŌ [%]

-3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
600
500 TNT
TA1
400
TA2
300
TA3
200 TA4
Force [kN]

100
0
-100
-200
-300
-400
-500
-600
-109 -94 -78 -63 -47 -31 -16 0 16 31 47 62 78 94 109
Displacement [mm]

Fig. 6. Force-displacement curve and maximum envelope: (a) TNT; (b) TA1; (c) TA2; (d) TA3; (e) TA4; (f) comparison of envelopes.

TA4 and of the corresponding infill contributions (TA2-TNT and TA4-
TNT) are illustrated in Fig. 10. In addition, values of the secant stiffness
k corresponding to different levels of force (25%, 50%, 75% and 100%
of the maximum force Fmax), the maximum force (Fmax) and the corre-
sponding drift (δFmax), the maximum imposed drift (δmax) and the cor-
responding force (Fδmax) and the ratios Fmax/Fδmax and δmax/δFmax are
reported in Table 3 for each test, where the signs + and − refer to
positive (“pull”) and negative (“push”) loading direction, respectively.
The secant stiffness k is calculated as the slope of the line joining a point
on the envelope curve and the origin. The obtained stiffness degrada-
tion, represented by values of secant stiffness in function of the corre-
sponding drift, is illustrated in Fig. 11. It is worth underlining that
specimen TA2 is considered representative of the seismic performance
of the fully infilled frames and only the values for this specimen are
reported, since it has been pushed to larger values of in-plane drifts.
As expected, the initial lateral stiffness values (i.e., the secant
stiffness values in the ascending branch of the force-displacement en-
velopes) and the stiffness degradation vs. the imposed drift are much
larger for the infilled specimens than for the bare frame (TNT), being
the fully infilled specimen more rigid than the infill with opening; for
example, at approximately 0.10% in-plane drift, the stiffness of spe-
cimen TNT is about 7.5 times smaller than for the fully infilled spe-
cimen TA2 and about 3.5 times smaller than for the partially infilled
specimen TA4. Evidently, lower values of initial secant stiffness are
found for the infill contributions. The peak force of specimen TA2 is
about 80% larger than the one of the bare frame, whereas specimens
TA4 and TNT reach approximatively the same maximum force.
However, the peak force is being attained at different drift levels, at
about 3.00%, 1.50% and 1.00% for specimens TNT, TA2 and TA4, re-
spectively. If only the infill contribution is considered, the drift at peak
force decreases, in average between positive and negative branch, to
about 0.90% and 0.40% for specimens TA2 and TA4, therefore pro-
viding larger values of the ratio δmax/δFmax. It is worth underlining that
the strength degradation is stronger when the infill contribution only is
taken into account and it is equal to about 40% for specimen TA2 and
about 30% (in average) for specimen TA4 even if, for the latter, a
further strength reserve seems possible after 1.00% drift. Finally, a
comparison of the average positive and negative response of the strong
infill with opening (specimen TA4), characterized by a strongly asym-
metric behaviour, with the response of the solid infill (specimen TA2),
reveals a substantial infill strength reduction (to about 45%) due to the
presence of the opening.
For what concerns the evaluation of the dissipated hysteretic energy
of the tested infilled frames, possible simplified criteria consist in the
determination of the total dissipated energy Ehys (sum of the areas en-
closed in the hysteretic loop for each cycle Wd), the total energy

509
P. Morandi et al. Engineering Structures 156 (2018) 503–521

Fig. 7. Crack patterns at different levels of drift: (a) TNT at 3.50%; (b) TA2 at 0.30%; (c) TA2 at 0.50%; (d) TA2 at 1.00%; (e) TA2 at 1.75% and (f) TA2 at 2.50%.

Fig. 8. Crack patterns at different levels of drift: (a) TA4 at 0.20%; (b) TA4 at 0.40%; (c) TA4 at 0.50%; (d) TA4 at 1.00%.

demand Einp (sum of the amount of elastic energy at peak displacement at peak displacement We, following the expression given by Eq. (1),
stored in the same loop for each cycle We) and the equivalent viscous where signs + and − indicate the positive and the negative elastic
damping ξeq; given a single load–displacement cycle, ξeq can be ex- branch, respectively.
pressed as a function of the dissipated energy Wd and the elastic energy

510
P. Morandi et al. Engineering Structures 156 (2018) 503–521

Fig. 9. Deformed shape from optical acquisition (amplified): (a) TNT (in black), TA2 (in red) and TA4 (in blue), at 1.00% drift; (b) TNT (in black), TA2 (in red), at 2.00% drift. (For
interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

driŌ [%] Fig. 10. Maximum envelopes for specimens TNT,

TA2, TA4 and corresponding infill contributions
-3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
TA2-TNT, TA4-TNT.
600
TNT
500
TA2
400
TA2-TNT
300
TA4
200 TA4-TNT
100
Force [kN]

0
-100
-200
-300
-400
-500
-600
-109 -94 -78 -63 -47 -31 -16 0 16 31 47 62 78 94 109
Displacement [mm]

Table 3
Main seismic parameters evaluated from maximum envelopes of the in-plane tests.

k0.25Fmax [kN/ k0.5Fmax [kN/ k0.75Fmax [kN/ kFmax [kN/ Kδmax [kN/ Fmax [kN] δFmax [%] Fδmax [kN] δmax [%] Fδmax/Fmax δmax/
mm] mm] mm] mm] mm] δFmax

TNT 9.8 6.9 5.4 3.2 2.7 298 3.00 292 3.50 0.98 1.17
9.8 7.0 5.5 3.3 2.8 −304 −3.00 −301 −3.50 0.99 1.17
TA2 262 90 33 11.9 6.5 547 1.50 497 2.50 0.91 1.67
293 103 33 11.9 6.6 −555 −1.50 −512 −2.50 0.92 1.67
TA4 111 33 22 9.4 9.4 291 1.00 291 1.00 1.00 1.00
68 48 32 9.6 9.6 −297 −1.00 −297 −1.00 1.00 1.00
TA2-TNT 238 228 55 10.9 2.7 335 1.00 209 2.50 0.62 2.50
280 249 69 14.2 2.8 −347 −0.80 −216 −2.50 0.62 3.13
TA4-TNT 93 67 22 11.2 3.4 138 0.40 104 1.00 0.75 2.50
69 61 28 15.0 3.5 −177 −0.40 −107 −1.00 0.61 2.50

Wd demand are not considered realistic, given that the lateral response of
ξeq =
2π (|We+| + |We−|) (1) the frame is not characterized by any significant hysteretic dissipation
with the risk of providing ratios between the dissipated energy of the
Accordingly, the evolution of the hysteretic/input energy and of the
cycle over the elastic energy unrealistically high.
equivalent viscous damping ratios has been determined for the tested
In the case of the RC bare frame, the equivalent viscous damping
specimens and plotted against the imposed drifts, based on an estima-
increases substantially from about 3% up to 10% starting from drifts
tion from the experimentally obtained load-displacement loops con-
larger than 1.50% up to a drift of 3.50%. For the fully infilled specimen
sidering the first, the second and the third cycle at each target dis-
(TA2) the damping, after a first sharp decrease, lies in the range be-
placement; the values of the hysteretic/input energy and of equivalent
tween 6 and 7%, only increasing after 2.0% drift up to about 8%. The
viscous damping for specimens TNT, TA1 and TA2 corresponding to the
damping of the partially infilled specimen (TA4) is lower than for
first cycles are reported, respectively, in Fig. 12(a) and (b), in function
specimen TA2 and stays between about 4.0 and 5.5% without following
of drift. Although the evaluation of the damping has been computed for
a specific trend. An estimation of net energy dissipation capacity of the
all the applied cycles, the plots of Fig. 12 begin from a drift of 0.3%,
two infills was also obtained, computing the area enclosed by the
since the values of damping at very low level of in-plane displacement

511
P. Morandi et al.
hysteretic cycles of the infilled frames after subtracting the area of the
bare frame hysteretic curves at corresponding values of imposed drift.
The results show a substantial increase of equivalent damping coeffi-
cients, ranging around 8% for both the infills between 0.4 and 1.0%
drift and reaching a maximum value of about 11% for TA2. Notably, in
the case of the infill with opening (TA4) the damping sharply increases
between a drift of 0.30% and 0.40%, attesting the occurrence of sig-
nificant damage at that level of deformation demand, i.e., the formation
of the diagonal crack in one of the two panels of the specimen.
512
Engineering Structures 156 (2018) 503–521
Fig. 11. Secant stiffness vs. imposed drift for specimens TNT, TA2,
TA4 and corresponding infill contributions TA2-TNT, TA4-TNT at 1st
cycles of increasing levels of drift.
4.2. Experimental vs. analytical evaluation of stiffness and strength of
infilled frames
Numerical modelling strategies for the representation of infilled
frames are diversified and range from very detailed micro-modelling to
rather simple macro-modelling techniques. Omitting the issues re-
garding possible choices of most appropriate modelling approaches,
which are out of the scope of this paper, primary attention is given to
the simplified macro-modelling strategy assuming that infills are re-
presented with one single equivalent diagonal strut along each
Fig. 12. Cumulative hysteretic/input energy (a) and equivalent vis-
cous damping (b) vs. imposed drift for specimens TNT, TA1, TA4, and
corresponding infill contributions TA2-TNT, TA4-TNT at 1st cycles of
increasing levels of drift larger than 0.30%.
P. Morandi et al. Engineering Structures 156 (2018) 503–521

Fig. 13. (a) Equivalent diagonal single-strut model; (b) case for opening.

Fig. 14. Example of equivalent diagonal single-strut infill model in terms of stress-strain envelopes: (a) Klingner and Bertero [28]; (b) Crisafulli [29]; (c) Rodrigues et al. [30]; (d) Fardis
[31].

diagonal, pin-ended at the RC member centreline intersections (see mechanical properties of the materials coming from the tests of char-
Fig. 13). Since the mid-sixties of the last century, several different single acterization. For a consistent comparison with the experimental results,
strut models, usually expressed for the need of numerical analyses in the horizontal component of the strut force and stiffness have been
terms of a stress-strain (σ-ε) relationship of the diagonal strut section, assessed.
were proposed as outcome of the interpretation of in-plane test results For the evaluation of the stiffness kw in the horizontal direction, the
(see some examples in Fig. 14); the stress-strain relationship of the di- expression given by Eq. (2) has been adopted:
agonal strut can be derived from experimental envelopes of horizontal
Ewθ tw bw
force vs. displacement (or drift) response curves applying simple geo- kw = cos2 θ
dw (2)
metrical considerations, as reported in Hak et al. [27].
Independently of the diagonal strut model that is being im- where tw is the thickness of the panel, bw is the equivalent strut width,
plemented, the main parameters that characterize the monotonic be- dw is the diagonal length of the infill and Ewθ is the elastic modulus of
haviour of the infill are the stiffness, the maximum strength and the masonry assigned to the inclined direction (θ = tan−1(hw/Lw)) that has
displacement/drift capacity associated to selected characteristic points, been evaluated based on a model for orthotropic elastic materials
which may represent different limit states and correlate to the extent of subjected to bi-axial tensile stresses, as defined by Eq. (3), where ν is
masonry damage. Poisson’s ratio, and Evert, Elat and G the vertical, the lateral and the shear
Given that, a comparison between the experimentally obtained infill modulus of elasticity of the masonry, respectively.
contribution and the outcome of most common analytical expressions
−1
for infill stiffness and strength, including the influence of the opening cos4 θ sin4 θ 1 ν ⎞⎤
Ewθ = ⎡ + + cos2 θ sin2 θ ⎛ −2
⎜ ⎟
has been accomplished. The values of stiffness and strength are ob- ⎢ Elat Evert ⎝G Elat ⎠ ⎥ (3)
⎣ ⎦
tained applying existing analytical expressions for the diagonal strut
using the nominal dimensions of the frame specimens and the Commonly, the equivalent strut width bw is expressed in terms of a
bw/dw ratio, where dw is the diagonal length of the infill. Several models

Table 4
Evaluation of the diagonal infill strut width according to different models.

Holmes [32] bw
= 0.33
dw
Stafford Smith [33] 0.10 <
bw
< 0.25
dw
Mainstone [34] bw
= 0.175(λh)−0.4 Ewθ tw sin2θ
dw λ= 4
Adopted in FEMA 306 [35] 4Ec Ic hw
Liauw and Kwan [37] bw
=
0.95sin(2θ) (Stafford Smith, [36])
dw 2 λh Ec: elastic modulus of concrete
Bertoldi et al. [38] bw
=
K1
+ K2 λh < 3.14: K1 = 1.300; K2 = −0.178 Ic: moment of inertia adjacent columns
dw λh
3.14 < λh < 7.85: K1 = 0.707; K2 = −0.010
λh > 7.85: K1 = 0.470; K2 = 0.040
Paulay and Priestley [39] bw
= 0.25
dw
Cavaleri et al. [24] bw
=
kc Ewθ tw h h2 1 Ac L Coefficients:
λ∗ = ⎛ + ⎞
dw z (λ∗) β Ec Ac ⎝ L2 4 Ab h ⎠ c, β: accounting for Poisson’s ratio
Ec: elastic modulus of concrete k: accounting for the vertical load
Ac: column gross area z: geometrical parameter
Ab: beam gross area

513
P. Morandi et al. Engineering Structures 156 (2018) 503–521

Table 5
Evaluation of the infill strength according to different models.

Paulay and Priestley [39] Vw = min(Vs ;Vc ) fv 0 2

Vc = ztw flat
Vs = t L
π −1 1 − μ (h / L) w w 3
z= λ flat: compression strength in horizontal direction of masonry
2 fv0: initial shear strength of bed-joints
Ewθ tw sin2θ μ: friction coefficient (Compression failure of diagonal strut)
λ= 4
4Ec Ic hw

(authors propose μ = 0.3)

(Sliding shear failure of the infill)

Bertoldi et al. [38] ′ tw b w cosθ

Vw = fm 1.16fvert tanθ 1.12fvert sinθ cosθ
σw,cc = σw,ccorn =
K1 + K2 λh K1 (λh)−0.12 + K2 (λh)0.88
′ = min(σw,cc;σw,ccorn;σw,ss;σw,sd )
fm
fvert: vertical compr. strength of masonry fvert: vertical compr. strength of masonry
bw = ( K1
λh )
+ K2 dw (Compression at the center of the infill) (Compression at the corners of the infill)
Ewθ tw sin2θ (1.2sinθ + 0.45cosθ) fv 0 + 0.3σv 0.6ft + 0.3σv
λ= 4 σw,ss = σw,sd = bw
4Ec Ic hw bw
dw dw
fv0: initial shear strength of bed-joints ft: shear strength under diagonal compression
σv: vertical stress σv: vertical stress
(Sliding shear failure of the infill) (Diagonal cracking of the infill)

FEMA 306 [35] Vmf ⩽ Vw = min(Vs ;Vc;Vcr ) ⩽ Vmi Vs = (fv0 + μσv ) tw L w Vc = b w tw flat cosθ
b w = 0.175(λh)−0.4dw fv0: initial shear strength of bed-joints flat: compression strength in horizontal direction of masonry
Ewθ tw sin2θ μ: friction coefficient (μ = 0.4) (Compression failure)
λ= 4
σv: vertical stress
4Ec Ic hw
(Sliding shear failure of the infill)
2 2 tw Lw σcr Vmi = 2 0.0069fvert tw L w
Vcr =
⎛ Lw + hw ⎞
⎜ ⎟
fvert: compression strength in vertical direction (in MPa)
⎝ hw Lw ⎠
σcr: cracking strength of masonry Vmf = 0.3Vmi
(Diagonal cracking failure)

EC8/EC6 [12,40] Vw = fv tw L w fv = fv0 + 0.4σv

fv0: initial shear strength of bed-joints
σv: vertical stress

Table 6
Reduction coefficients of stiffness and strength due to the presence of the opening.

αa = 100(lphp)/(Lwhw); αl = 100lp/Lw; αh = 100hp/hw; αd = Lwhw − (dwsin(2θ) − d0sin(θ + ϕ))2/(2sin(2θ))

Sachansky [42] rp = 1−(0.004αl + 0.006αh)
Dawe and Seah [43] rp = 1−1.5αl/100 ; αl < 66%
Imai and Miyamoto [44] rp = min(1−0.01αl ; 1−0.1αa0.5)
Durrani and Luo [45] rp = 1−(αd/ L w hw )2
Al-Chaar [46] rp = 1−1.6(αa/100) + 0.6(αa/100)2 ; αa < 60%
Asteris [47] rp = 1−2(αa/100)0.54 + (αa/100)1.14
Mondal and Jain [48] rp = 1−2.6(αa/100) ; αa < 38.5%
Tasnimi and Mohebkhan [26] rp = 1−2.238(αa/100) + 1.49(αa/100)2 ; αa < 40%
Decanini et al. [41] rp = 0.55e (−0.035αa) + 0.44e (−0.025αl)

have been proposed in the past, many of which based on the di-
mensionless parameter λ, function of the relative stiffness between the
infill and the structure. The models studied within the scope of this
work are reported in Table 4.
Also for what concerns the lateral strength of the infill numerous
expressions have been proposed, some based on governing failure
modes (as for Paulay and Priestley [39], Bertoldi et al. [38], FEMA 306
[35]), other simply defined by a single expression (as for EC8-Part 1
[12]), as indicated in Table 5, referring to the horizontal component of
the strength (Vw) assigned to the diagonal strut. The expression pro-
posed in EC8-Part 1 [12], in relation to the assessment of local effects
due to the presence of masonry infills, recommends using the horizontal
component of the strut force of the infill, assumed to be equal to the
horizontal shear strength of the panel, as estimated on the basis of the
shear strength of bed-joints. For tongue and groove units with unfilled
head-joints, as in the case of the tested masonry, EC6 [40] requires
reducing the initial shear strength (cohesion) fv0 to one-half; hence, the
influence of this assumption has been assessed in the evaluation of the
analytical results.
The presence of an opening can be taken into account by reducing
the gross strength and stiffness of the infill through a unique reduction
coefficient rp, as reported in Decanini et al. [41]. Within the scope of
this study, expressions proposed by different authors for the case of
unreinforced opening, i.e. without reinforcing elements/lintels around
the opening, have been implemented (see Table 6 and Fig. 13(b)).
A comparison of the experimental results in terms of infill con-
tribution on the tested solid panel (specimen TA2) to the analytical
outcomes in terms of stiffness and strength of the diagonal strut has
been carried out and the influence of the opening on the infill stiffness
and strength coming from the experimental results for specimen TA4
has been assessed against the reduction factors proposed in previous
studies, as presented in Table 7 and in Fig. 15.
The analytical expressions lead to ratios of bw/dw in the range be-
tween 0.10 and 0.33 and the corresponding values of stiffness

514
 Table 7
Comparison between experimental and analytical stiffness and strength.

L [mm] h [mm] tw [mm] d [mm] Lw [mm] hw [mm] dw [mm] θ [rad] Evert [MPa] Elat [MPa] G [MPa] ν [–] Ewθ [MPa]
P. Morandi et al.

4570 3300 350 5637 4220 2950 5149 0.610 5299 494 2120 0.25 1228

σv [MPa] fvert [MPa] flat [MPa] fv0 [MPa] ft [MPa] σcr [MPa] Ec [MPa] Ic = Ib [m4] lp [mm] hp [mm] αaα = αc [%] αh [%] αd [m2]

0.013 4.64 1.08 0.359 – – 30072 0.00125 1440 2950 34.1 100 11.1

STIFFNESS [KN/mm]

EXP. (from TA2) ANALYTICAL (in brackets the values of bw/dw)

k0.25Fmax k0.5Fmax k0.75Fmax kFmax kFθmax Holmes [32] Stafford Smith [33] Mainstone [34] Liauw & Kwan [37] Paulay & Priestley [39] Bertoldi et al. [38] Cavaleri et al. [24]

238 228 55 11 3 95 (0.330) 29 (0.100) 32 (0.110) 72 (0.249) 72 (0.250) 66 (0.248) 78 (0.269)

280 249 69 14 3 72 (0.250)

STRENGTH [KN]

EXP. (from TA2) ANALYTICAL (in brackets the values with fv = fv0/2 according to EC6)

Fmax Paulay & Priestley [39] Bertoldi et al. [38] FEMA 306 [35] EC8/EC6 [12,40]

335 405 (3 3 8) 560 (2 8 3) 175 273

−347

515
REDUCTION COEFFICIEN T, rp[–]

EXP. (from TA4/TA2) ANALYTICAL

k0.75Fmax,TA4/k0.75Fmax,TA2 Fmax,TA4/Fmax,TA2 Sachansky [38] Dawe & Seah [43] Imai & Miyamoto [44] Durrani & Luo [45] Al-Chaar [46] Asteris et al. [47] Mondal & Jain [48] Tasnimi & Mohebkhan [26] Decanini et al. [41]

0.39 0.41 0.26 0.49 0.42 0.21 0.52 0.17 0.11 0.41 0.35
0.41 0.51
Engineering Structures 156 (2018) 503–521
P. Morandi et al.
approximately describe the experimental secant stiffness between
K0.75Fmax and KFmax (being the lower limit by Stafford Smith [33] and
the expression by Mainstone [34] the least stiff cases), as explicitly
specified in some studies (e.g., Bertoldi et al. [38]), rather than an in-
itial elastic stiffness. Moreover, these results have been derived com-
puting the elastic modulus of masonry for the inclined direction (Ewθ) as
defined in Eq. (3), even though for some of the adopted models it is not
clearly defined which elastic modulus of masonry needs to be used (Ewθ,
Evert or Elat). Hence, in some cases an erroneous estimation can imply
large differences in the obtained stiffness, above all for some specific
types of masonry with importantly different mechanical properties in
the directions perpendicular and parallel to the bed-joints, as for the
tested masonry with tongue and groove units. If for example, specifi-
cally in this case, the vertical modulus of elasticity of the masonry
(Evert = 5299 MPa) was employed instead of that for the inclined di-
rection (Ewθ = 1228 MPa), the values of analytically obtained stiffness
would differ substantially and would fit more closely the initial ex-
perimental stiffness of specimen TA2 than the secant stiffness at peak
force.
Regarding the lateral strength, the expressions proposed in Paulay
and Priestley [39] provide a very good agreement with the experi-
mental results, above all when one half of fv0 is assumed, as proposed in
EC6 [40]; the expression according to EC8/EC6 [12,40] leads to a fairly
good safe-sided result (about 15% below the experimental strength),
whereas the strength according to Bertoldi et al. [38] is in good
agreement with the test results only if for shear sliding, that governs the
failure mechanism, the cohesion fv0 is halved. FEMA 306 [35] leads to
over-underestimate the resistance since the governing expression is the
one related to the lateral compression strength of the masonry (flat),
which is limited due to the presence of units with dry head-joints.
Finally, the experimental reduction coefficient rp applied for the
reduction of stiffness and strength of specimen TA4 due to the presence
of the opening, although rather different for the two loading directions,
results to be in relatively good agreement with the expressions pro-
posed in Dawe and Seah [43], Imai and Miyamot [44], Tasnimi and
Mohebkah [26] and Decanini et al. [41], being the first slightly less and
the last slightly more conservative in comparison with the test results.
5. Performance levels for infills according to in-plane tests
In order to follow a systematic approach, performance levels in a
numerical analysis for a single masonry infill can be introduced as a
function of the inter-storey drift of the frame and/or of the
516
Engineering Structures 156 (2018) 503–521
Fig. 15. Comparison between the experimental
infill contribution envelope of specimen TA2 and
different analytical expressions for the evaluation
of stiffness and strength.
corresponding attained strain in the equivalent strut, evaluated through
interpretation of attained levels of damage. However, a commonly ac-
cepted definition of possible damage limit states specifically related to
performance requirements for masonry infills has not yet been estab-
lished within the scientific community and related specifications in-
cluded in international seismic standards and codes are rather in-
accurate.
Only few experimental investigations in the past have provided
clear definitions of performance limit states specifically applied to
masonry infills like, for example, Mehrabi and Shing [5], Calvi and
Bolognini [7] and, more recently, Hak et al. [27]. The definition of
different performance levels referring to the state of damage in struc-
tural and non-structural components (including infills) is commonly
also included in modern international seismic codes and standards,
both for the design of new buildings and for the assessment of existing
buildings. Nevertheless, only general definitions of the compliance
criteria are included in the Italian Norms for Construction [16], in the
European (EC8-Part 1 [12] and Part 3 [49]) and in the New Zealand
seismic codes (NZS1170.5:2004 [50]), whereas explicit damage defi-
nitions related to masonry infills are not available. Few additional in-
formation on the infill performance is instead included in the U.S.
seismic standards (i.e., FEMA E-74 [51], FEMA 273 [52] and FEMA 306
[35]).
5.1. Proposal for the definition of performance levels according to in-plane
tests
Based on the past research experience and the general requirements
provided by international standards, a definition of specific perfor-
mance levels suitable for in-plane response of non-structural masonry
infills is proposed within the scope of this study. The limit state defi-
nition relies on the performance of masonry infill evaluated from the
results of the in-plane cyclic tests on RC infilled frames. Hence, three
limit states, specifically referring to masonry infills are here introduced,
based on the increasing extent of infill damage due to in-plane actions:
• “Operational Limit State” (OLS). The infill is considered slightly
damaged. The occurrence of this level of damage, usually sub-
sequent to the detachment of the masonry panel from the RC frame
at the intrados of the top beam and along the (upper) height of the
columns, can be characterized by a very light and superficial
cracking in the masonry panel, mainly concentrated in the bed- and
in the head-joints, or by cracks in the plaster. A “cosmetic” damage
P. Morandi et al. Engineering Structures 156 (2018) 503–521

Table 8
Comparison of performance levels for single strong and slender masonry infills.

Limit State Operational (OLS) Damage Limitation (DLS) Life Safety (ULS)
Drift į = 0.30% į = 0.50% į = 1.75%
STRONG INFILL

OPENING (TA2)
WITHOUT

Damage

Drift į = 0.20% į = 0.35% į = 1.00%

STRONG INFILL
WITH OPENING
(TA4)

Damage

without the need for repair, like the occurrence of very light cracks
in the plaster also due to the detachment of the panel from the
frame, does not belong to this damage state but could eventually be
considered as a further minor damage state level (“almost zero da-
mage”).
• “Damage Limitation Limit State” (DLS). The infill is damaged, but
can be effectively and economically repaired. Damage of the infill
panel, through the formation of bi-diagonal cracking, involving both
the joints and the units or diagonal step-wise cracking affecting
mainly the bed and the head-joints, is expected. Sliding in the bed-
joints may also occur. Very limited crushing and spalling of a few
units, for instance at the upper corners and/or at the top edge of the
infill, can be presumed.
• “Life Safety or Significant Damage Limit State (Ultimate Limit
State)” (ULS). The infill is severely damaged and reparability is
economically questionable, however, lives are not threatened.
Detachment of large plaster area, if present, significant sliding in the
mortar joints and substantial development of cracks in the units is
expected. In addition, crushing and spalling of brick units are
widespread throughout the infill. The masonry is not repairable at
reasonable costs and it is more convenient to demolish and re-
construct the entire infill. However, the position and the weight of
masonry portions falling down is so limited as to exclude the risk for
the loss of human lives.
In the definitions above, specific reference to damage on the RC
members has intentionally been neglected in order to identify the
performance of the masonry infills independently by the response of the
structure.
Moreover, although the in-plane response is primarily considered,
the performance criteria should evidently also include a possible ad-
ditional limit state considering infill damage/collapse due to out-of-
plane actions that in principle could occur before the activation of one
of the in-plane limit states such that it may imply the ultimate condi-
tions of the panel.

driŌ [%] Fig. 16. Performance points on the infill con-

tribution curve for specimens TA2 and TA4.
-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
2
400
OLLS DLS ULS
TA2-TNT
300
TA4-TNT
200 DLS ULS
100
Force [[kN]]

OLS
0
OLS
-100

-200 ULS DLS

-300
ULS
DLS OLS
-400
-78 -63 -47 -31 -16 0 16 31 47 62 78
Displacement [m
mm]

517
P. Morandi et al.
Table 9
Comparison of damage patterns for single strong and slender masonry infills.
Limit State Operational (OLS)
Drift į = 0.30%
STRONG INFILL
OPENING (TA2)
WITHOUT
Damage
Drift į ~ 0.20%
SLENDER INFILL
Damage
5.2. Performance based interpretation for the tested infills
With reference to the definition provided in Section 5.1, the per-
formance levels for a single masonry infill can be identified in terms of
inter-storey drift (δ) associated with a specific degree of damage de-
rived from the available results of in-plane cyclic tests on infilled
frames, computing the drift as the ratio between the horizontal dis-
placement and the height at the beam centreline of the specimen.
Therefore, for each experimental test, a value of drift corresponding to
the attainment of a specific damage state can be identified based on the
extent of infill damage occurring during the test, implying that the
damage pattern for each imposed displacement needs to be recorded.
Such interpretation of in-plane cyclic tests, accounting primarily for
the correlation of damage levels and values of drift, could sometimes
result in biased and subjective conclusions, in particular, given the aim
of providing a qualitative definition of the damage states. A possible
solution to this issue could be the association of damage levels with
specific points on the experimental force-displacement curve. Based on
the experimental results of in-plane cyclic tests, such a simplified
Engineering Structures 156 (2018) 503–521
Damage Limitation (DLS) Life Safety (ULS)
į = 0.50% į = 1.75%
į ~ 0.30% į ~ 1.00%
Fig. 17. Comparison of performance points on the
infill contribution curve for strong infill (spe-
cimen TA2) and slender infill, both without
opening.
procedure consists in the evaluation of the response of the masonry
infill estimated as the difference between the experimental envelope of
the infilled frame and the experimental envelope of the corresponding
bare frame configuration. On the determined response curve, char-
acterizing the infill contribution to the lateral response of the infilled
frame, points representing the attainment of some of the defined da-
mage states can be identified; for example, damage limitation limit
state (DLS) may be as assigned to the peak of the envelope or to the
beginning of a possible strength plateau, as commonly assumed for
load-bearing masonry walls when first large cracking affecting units
and joints occurs and a damage limitation state is attained (see e.g.,
Morandi et al. [53]); moreover, life safety limit state (ULS) may cor-
respond to a certain post-peak strength degradation ratio. Such ap-
proach to interpret the infill contribution response curve, although
conventional and significantly dependent on the considered masonry
infill typology, can be used as an integration, and not as an alternative,
of the visual interpretation of damage. In any case, it allows a consistent
comparison of the experimentally obtained infill properties to similar
results for other masonry typologies, not only in terms of displacement
518
P. Morandi et al.
capacity but also in terms of strength and stiffness.
Hence, for the solid infill (specimen TA2) and for the infill with
opening (specimen TA4), values of drift (δ) corresponding to the at-
tainment of the specific damage states, described in Section 5.1, have
been selected based on the extent of infill damage observed during the
tests, as summarized in Table 8. In addition, the force-displacement
contribution of the two infills (specimen TA2 and TA4) with the in-
dications of the drift levels corresponding to the different limit states
are illustrated in Fig. 16.
In both cases, the operational limit state has been identified after
the detachment of the masonry panels from the RC frame, when the first
very light diagonal step-wise cracks have occurred in the bed- and head-
joints, the damage limitation limit state when diagonal cracking in the
joints has started to enlarge and limited cracking in the units has shown
involving also spalling of the blocks in the upper corners for the solid
infill and, finally, the life safety limit state when substantial develop-
ment of cracks and crushing in the clay units has started to be wide-
spread throughout the infill, however, without threatening life safety.
In the case of the partial infill, the damage and the creation of full
diagonal cracking has occurred only in one of the two panels.
Specifically, in the case of the fully infilled specimen (TA2), a drift
limit equal to 0.50% has been assigned to the attainment of damage
limitation limit state conditions, whereas a drift equal to 0.30% has
been proposed for the operational limit state, i.e. 2/3 of the drift value
obtained for damage limitation is assumed, which corresponds also to
the Italian seismic design provisions NTC08 [16]. Furthermore, always
supported by the observations during the test, a drift of 1.75% may be
reached before exceeding the ultimate limit state conditions, which
corresponds to a strength degradation of about 20% after the peak.
Similarly, for the strong infill with opening (TA4), drift values of 0.20%,
0.35% and 1.00% have been assigned to operational, damage limitation
and ultimate limit state conditions, respectively, showing a significant
reduction (of about 1.5 times) with respect to the corresponding de-
formation capacity of the solid infill due to the presence of the opening,
which implies a particularly brittle behaviour for this case of un-
reinforced opening (i.e., without reinforcing elements/lintels at all clear
height).
Applying the proposed interpretation to other in-plane tests on
simple infilled RC frames of the past, it is possible to compare the
seismic performance of different types of infills. For example, from an
experimental study reported in da Porto et al. [9] on a strong infill
typology made up by 30 cm thick vertically perforated clay units in-
plane cyclically tested on real scale specimens, similar values of drift at
the different limit states can be derived on solid infills, resulting to be
about 0.50% at damage limit state and 1.20% at ultimate limit state, the
latter being however a conservative choice, since no drift levels beyond
1.20% have been imposed during that experimental campaign; the drift
level at DLS was characterized by diagonal cracks, stepped through
joints, developed from the upper corners towards the centre of the
Table 10
Performance levels for a single masonry infill.
519
Engineering Structures 156 (2018) 503–521
walls. Comparing the obtained drift limits for the strong type of infill to
corresponding values obtained analogously from the tests described in
Calvi and Bolognini [7] on a slender weak infill (11.5 cm thick un-
reinforced horizontally hollowed clay masonry infill with 2.0 cm of
plaster), much smaller values of deformation capacity can be achieved,
being equal to 0.30% at DLS and 1.00% at ULS, albeit the ratio of drift
between damage limitation and ultimate limit state conditions is ap-
proximately preserved for the two typologies (3.0–3.5). A comparison
of the damage patterns and the performance points on the force-dis-
placement contribution curves for strong and slender infill typologies is
respectively shown in Table 9 and in Fig. 17, where it is also possible to
notice that, at ULS, the strength degradation after the peak is much
larger for the weak infill (about 85%) than for the strong one (about
20%).
5.3. Implications of infill performance interpretation for numerical analyses
In order to adopt a relatively simple single-strut model, reflecting
realistically the behaviour of strong masonry infills and being suitable
to study infilled RC structural configurations through extensive non-
linear static and dynamic analyses, focused primarily on the evaluation
of global inter-storey drift demands and the corresponding infill da-
mage distribution, a thorough model calibration should be performed
based on the obtained experimental test results. A masonry infill strut
model, defined by a suitable force-displacement (or, equivalently, axial
stress-strain) hysteretic rule in the case of cyclic loading and the cor-
responding envelope in the case of monotonic loading should be as-
sumed. The parameter describing the deformation capacity of the infill,
being of particular importance for the assessment of infill damage, must
be identified according to the interpretation of the experimentally ob-
tained response. Furthermore, particular attention needs to be devoted
to the definition of parameters describing suitably the hysteretic re-
sponse of strong infills, as well as to the evaluation of appropriate
strength and stiffness properties assigned to the diagonal strut.
It is important to underline that the proper use of these experi-
mental findings implies the awareness that the tests were carried out on
single span- single storey frames with vertical loads applied at the top of
the columns. The effects of the continuity of the beams and columns in
adjacent spans and in upper stories and the presence of vertical loads
directly applied on the beams (here, likely, little influencing the results
for the relatively high stiffness of the beams) that characterize real
cases, can be properly consider in the numerical modelling and analyses
of infilled buildings, starting from a suitable calibration on simple in-
filled frames, as done for example by Oliaee et al. [54] on this masonry
typology.
Therefore, independently of the infill model of the diagonal strut
that is being implemented (some models may be more suitable for
certain types of infills than others), for the characterization of the
monotonic behaviour of the infill the values of stiffness Kw, maximum
P. Morandi et al.
strength Vw and displacement/drift capacity associated to different
limit states needs to be identified according to the experimental in-
plane test results, as discussed in Section 5.2. In particular, the per-
formance levels in a numerical analysis for a single masonry panel must
be introduced as a function of the inter-storey drift δ of the frame and/
or of the corresponding attained strain ε in the equivalent strut, as
summarized in Table 10, and defined adopting the drift or strain limits
at OLS, DLS, and ULS evaluated through the interpretation of the da-
mage level.
During the numerical analyses of infilled structures the drift or
strain demand in each infill should be checked against the deformation
capacity at the different limit states in order to identify the performance
of the infill in terms of achieved damage.
6. Conclusions
The present paper is focused on the framework of an experimental
study carried out with the aim to improve the understanding of the in-
plane cyclic response of rigid strong masonry infills, commonly adopted
for building enclosure systems in seismic regions, considering the re-
lated damage distribution, the lateral stiffness and strength as well as
the dissipation capacity, in order to interpret the in-plane tests ac-
cording to a definition of limit states developed for the application in
current design/assessment approaches.
After a complete geometrical and mechanical characterization of
the materials, five in-plane cyclic tests on single bay-single storey RC
frames have been carried out: one on a bare RC frame (TNT), three on
fully infilled frames (TA1, TA2 and TA3) at different imposed drifts, and
one on an infill with a central unreinforced opening (TA4).
The response of the fully infilled frame TA2, taken as reference for
the solid infill specimens, was found to be characterized by a con-
siderable deformation capacity and a gradual attainment of a relatively
high peak resistance, accompanied by slow and rather low degradation
of strength, resulting in a substantial residual resistance, even at higher
levels of drift. Due to the relatively high resistance of the strong infill,
the creation of diagonal shear cracks in the RC columns, caused by the
infill strut action, was observed, even though no shear failure occurred.
In any case, possible shear failure mechanisms should be prevented
through adequate design measures for such types of strong masonry
(see e.g. Hak et al. [55]). On the other side, the infilled specimen with
opening (TA4) provided a strongly asymmetric behaviour due to the
occurrence of a corner-to-corner diagonal crack only on one of the two
panels, and revealed a substantially lower infill strength (about 45%) in
comparison with the solid infill. The values of equivalent viscous
damping contributed by the infill attained values of about 8% and 11%
for the specimens with and without opening.
Furthermore, particular attention has been devoted to the analysis
of the infill performance, resulting in the identification of characteristic
failure modes of the considered strong infill and the definition of per-
formance levels in function of attained drift values. Based on observa-
tions during the test and on a simplified representation of the infill
contribution, the attainment of operation, damage limitation and ulti-
mate limit state conditions has been found at drifts equal to 0.30%,
0.50% and 1.75%, respectively, for the fully infilled specimen and equal
to 0.20%, 0.35% and 1.00%, respectively, for the specimen with
opening. Clearly, the drift capacity is greatly reduced for the case with
opening, which was designed as unreinforced, although it is known (see
e.g., Decanini et al. [41] and Verlato et al. [56]) that the presence of
reinforcing elements around the openings (such as lintels bands or steel
reinforcement) can affect significantly the seismic behaviour of infills,
enhancing the strength and stiffness, reducing the crack pattern in the
infill and, therefore, increasing the deformation capacity. Even though
relatively large in-plane drift limits have been obtained for the con-
sidered case of strong infill, in particular for the solid specimen, the out-
of-plane infill stability has to be carefully assessed, accounting for
possible strength reduction due to in-plane damage as, for instance,
520
Engineering Structures 156 (2018) 503–521
discussed in Morandi et al. [57].
Common analytical models based on single diagonal equivalent
struts have been compared with the in-plane test results for the eva-
luation of the lateral stiffness and the strength. The models estimate
more closely the experimentally obtained secant stiffness between
K0.75Fmax and KFmax, rather than the initial elastic stiffness (as shown in
Fig. 15 and Table 7), and depend strongly on the choice of the elasticity
modulus of masonry (Ewθ, Evert or Elat); in particular, for masonry
typologies with different mechanical properties in the direction per-
pendicular and parallel to the bed-joints, as in this specific case, the use
of the elastic modulus of masonry for the diagonal direction Ewθ appears
to be the most reasonable choice. Regarding the lateral strength, the
expressions given in EC8/EC6 [12], [40], Bertoldi et al. [38] and Paulay
and Priestley [39] provide fairly good, safe-sided results, only if the
cohesion fv0 is halved as proposed in the EC6 for masonry units with dry
head-joints.
Applying the results of the performance-based interpretation of the
in-plane tests for the numerical analyses of infilled structures, the drift
or strain demand in each infill of the building should be checked against
the deformation capacity obtained at the different limit states, in order
to identify the performance of the infill in terms of damage. Although in
the seismic European norms the safety checks for the design/ assess-
ment of the in-plane performance of masonry infills are usually defined
only verifying the inter-storey drift demand against the drift limits at
the serviceability limit state (DLS and OLS for important structures), the
verification of the deformations at ultimate limit state is considered
essential, in order to control and limit severe damage of the infills and
prevent the risk for loss of human lives. Moreover, the drift limit at DLS
of 0.50% proposed by EC8 [12] and NTC08 [16] for masonry infills
attached to the structure appears to be conservative for these strong
infills without opening, assuming that the displacement demand is
evaluated on bare frames, as commonly done in the design practice.
More effective criteria for the estimation of drift demands in infilled
structures are to include the stiffening contribution of the panels di-
rectly in the structural models or to reduce the displacement demands
obtained on bare configurations, considering simple parameters ac-
counting for the structural properties and the presence of infills, as for
example proposed by Hak et al. [11].
Finally, although the attention of this study is mainly concentrated
on the in-plane response of infills in newly designed RC frames, it is
essential to take in duly consideration possible additional limit states
specific for the RC elements (i.e., ultimate deformation capacity of the
plastic hinges, local failures in the columns/beams and in the joints,
etc.) and/or possible consequences due to out-of-plane actions on the
masonry panels which could be activated before the attainment of the
proposed performance points defined specifically for the in-plane re-
sponse of masonry infills.
Acknowledgements
This work, conducted at the EUCENTRE Foundation and at
Department of Civil Engineering and Architecture of the University of
Pavia in Italy, was funded by ANDIL (Italian Association of Clay Brick
and Tile Producers) and by the Executive Projects DPC-RELUIS 2010-
2013 and DPC-RELUIS 2013-2016. The financial support received is
gratefully acknowledged. The authors would also like to thank Eng.
Carlo Filippo Manzini for his precious contribution in the post-proces-
sing of the experimental data.
References
[1] Ricci P, Manfredi V, De Luca F, Verderame GM. 6th April 2009 L’Aquila earthquake,
Italy: reinforced concrete building performance. Bull Earthq Eng 2011;9:285–305.
[2] Manzini CF, Morandi P. Rapporto preliminare sulle prestazioni ed i danneggiamenti
agli edifici in muratura portante moderni a seguito degli eventi sismici emiliani del
2012. Post-seismic survey, Eucentre, Pavia, Italy; 2012 [in Italian]. < http://www.
eqclearinghouse.org/2012-05-20-italy/ > .
P. Morandi et al.
[3] Morandi P. Pictures from the Central Italy earthquake; 2016.
[4] Angel R, Abrams D. Behavior of reinforced concrete frames with masonry infills
Structural Research Series No. 589 Urbana-Champaign (USA): Department of Civil
Engineering, University of Illinois; 2005.
[5] Meharbi AB, Shing PB. Seismic analysis of masonry-infilled reinforced concrete
frames. TMS J 2003.
[6] Mosalam KM, White RN, Gergely P. Static response of infilled frames using quasi-
static experimentation. ASCE J Struct Eng 1997;123:1462–4169.
[7] Calvi GM, Bolognini D. Seismic response of RC frames infilled with weakly re-
inforced masonry panels. J Earthq Eng 2001;5(02):153–85.
[8] Pujol S, Benavent-Climent A, Rodriguez ME, Smith-Pardo P. Masonry infill walls: an
effective alternative for seismic strengthening of low-rise reinforced concrete
building structures. Proceedings of the 14th world conference on earthquake en-
gineering, 12–17 October 2008, Beijing, China. 2008.
[9] Da Porto F, Guidi G, Dalla Benetta M, Verlato N. Combined in-plane/out-of-plane
experimental behaviour of reinforced and strengthened infill masonry walls.
Proceedings of the 12th Canadian masonry symposium, 2–5 June 2013, Vancouver,
British Columbia, Canada. 2013.
[10] Sassun K, Sullivan TJ, Morandi P, Cardone D. Characterising the in-plane seismic
performance of infill masonry. Bull New Zeal Soc Earthq Eng 2016;49(1):100–17.
[11] Hak S, Morandi P, Magenes G. Prediction of inter-storey drifts for regular RC
structures with masonry infills based on bare frame modelling. Bull Earthq Eng
2017. http://dx.doi.org/10.1007/s10518-017-0210-y.
[12] CEN Eurocode 8 – design of structures for earthquake resistance, Part 1: General
rules, seismic actions and rules for buildings. EN 1998-1. Brussels, Belgium: ECS;
2004.
[13] Morandi P, Hak S, Magenes G. Mechanical characterization and force-displacement
hysteretic curves from in-plane cyclic tests on strong masonry infills. Data in Brief,
submitted for publication.
[14] CEN Eurocode 1 – actions on structures, Part 1–1: General actions – densities, self-
weight, imposed loads for buildings. EN 1991-1-1. Brussels, Belgium: ECS; 2002.
[15] CEN Eurocode 2 – design of concrete structures, Part 1–1: General rules and rules
for buildings. EN 1992-1-1. Brussels, Belgium: ECS; 2004.
[16] NTC08 – norme tecniche per le costruzioni, D.M. 14 Gennaio 2008, Ministero delle
Infrastrutture, S.O. No. 30 alla G.U. del 4.2.2008, No. 29, Rome, Italy; 2008 [in
Italian].
[17] Morandi P, Hak S, Magenes G. Experimental and numerical seismic performance of
strong clay masonry infills. Research report EUCENTRE 2017/02. Pavia:
EUCENTRE FOUNDATION; 2017. ISBN: 978-88-85701-03-8. < http://www.
eucentre.it/report-di-ricerca/ > .
[18] EN 772-1. Methods of test for masonry units – determination of compressive
strength; 2011.
[19] EN 1015-11. Methods of test for mortar for masonry – determination of flexural and
compressive strength of hardened mortar; 2007.
[20] EN 1052-1. Methods of test for masonry – determination of compressive strength;
2001.
[21] EN 1052-3. Methods of test for masonry – determination of initial shear strength;
2007.
[22] Oliaee M, Morandi P, Magenes G. Macro-model calibration of a strong clay masonry
infill to in-plane cyclic tests. Proc. of the 5th international conference on compu-
tational methods in structural dynamics and earthquake engineering, 25–27 May
2015, Crete Island, Greece. 2015.
[23] Hak S, Morandi P, Magenes G. Evaluation of infill strut properties based on in-plane
cyclic tests. Građevinar 2013;65(6):509–22.
[24] Cavaleri L, Fossetti M, Papia M. Infilled frames: developments in the evaluation of
the cyclic behaviour under lateral loads. Struct Eng Mech 2005;21:469–94.
[25] Sigmund V, Penava D. Influence of openings, with and without confinement, on
cyclic response of infilled R-C frames – an experimental study. J Earthq Eng
2014;18(1):113–46.
[26] Tasnimi AA, Mohebkhah A. Investigation on the behavior of brick-infilled steel
frames with openings, experimental and analytical approaches. Eng Struct
2011;33(3):968–80.
[27] Hak S, Morandi P, Magenes G, Sullivan T. Damage control for clay masonry infills in
the design of RC frame structures. J Earthq Eng 2012;16(S1):1–35.
[28] Klingner RE, Bertero VV. Infilled frames in earthquake resistant construction Report
No. EERC 76-32 University of California; 1976.
[29] Crisafulli FJ. Seismic behaviour of reinforced concrete structures with masonry
infills [PhD. Dissertation]. University of Canterbury; 1997.
[30] Rodrigues H, Varum H, Costa A. Simplified macro-model for infill masonry panels. J
Earthq Eng 2010;14(3):390–416.
521
Engineering Structures 156 (2018) 503–521
[31] Fardis MN. Seismic design issues for masonry-infilled RC frames. Proceedings of the
1st European conference on earthquake engineering and seismology, Geneva. 2006.
[32] Holmes M. Steel frames with brickwork and concrete filling. Proc Inst Civ Eng
1961;19:473–8.
[33] Stafford Smith B. Behaviour of square infilled frames. ASCE J Struct Div
1966;92(1):381–403.
[34] Mainstone RJ. Supplementary note on the stiffnesses and strengths of infilled
frames. Building Research Establishment, Building Research Station; 1974.
[35] FEMA 306. Evaluation of earthquake damaged concrete and masonry wall buildings
– basic procedures manual. Washington D.C: Federal Emergency Management
Agency; 1998.
[36] Stafford Smith B. Methods for predicting the lateral stiffness and strength of mul-
tistorey infilled frames. Build Sci 1967;2:247–57.
[37] Liauw TC, Kwan K. Nonlinear behaviour of non-integral infilled frames. Comput
Struct 1984;18:551–60.
[38] Bertoldi SH, Decanini LD, Gavarini C. Telai tamponati soggetti ad azioni sismiche,
un modello semplificato: confronto sperimentale e numerico. Atti del VI Convegno
ANIDIS 1993;6:815–24. [in Italian].
[39] Paulay T, Priestley MJ. Seismic design of reinforced concrete and masonry build-
ings. New York: John Wiley & Sons; 1992.
[40] CEN Eurocode 6 – design of masonry structures, Part 1–1: Common rules for re-
inforced and unreinforced masonry structures. EN 1996-1-1. Brussels, Belgium:
European Committee for Standardisation; 2004.
[41] Decanini LD, Liberatore L, Mollaioli F. Strength and stiffness reduction factors for
infilled frames with openings. Earthq Eng Eng Vib 2014;13(3):437–54.
[42] Sachanski S. Analysis of earthquake resistance of frame buildings taking into con-
sideration the carrying capacity of the filling masonry. Proceedings of the 2nd
World conference on earthquake engineering, Japan, vol. 3. 1960. p. 2127–41.
[43] Dawe JL, Seah CK. Lateral load resistance of masonry panels in flexible steel frames.
Proceedings of the 8th international brick and block masonry conference, vol. 2.
London: Elsevier Applied Science; 1988. p. 606–16.
[44] Imai H, Miyamoto M. Seismic behavior of reinforced masonry walls with small
opening. Proceedings of the 5 Jornadas Chilenas de Sismología e Ingeniería
Antisísmica, vol. 2. 1989. p. 965–73.
[45] Durrani AJ, Luo YH. Seismic retrofit of flat-slab buildings with masonry infills
Technical report National Center for Earthquake Engineering Research; 1994.
p. 1–8.
[46] Al-Chaar G. Evaluating strength and stiffness of unreinforced masonry infill struc-
tures. No. erdc/cerl-tr-02-1. Engineer research and development center champaign
il construction engineering research lab; 2002.
[47] Asteris PG. Lateral stiffness of brick masonry infilled plane frames. ASCE J Struct
Eng 2003;129(8):1071–9.
[48] Mondal G, Sudhir KJ. Lateral stiffness of masonry infilled reinforced concrete (RC)
frames with central opening. Earthq Spectra 2008;24(3):701–23.
[49] CEN Eurocode 8 – design of structures for earthquake resistance, Part 3: assessment
and retrofitting of buildings. EN 1998-3. Brussels, Belgium: ECS; 2005.
[50] NZS1170.5. Structural design actions – Part 5 – earthquake actions. Standards New
Zealand (Code and Commentary); 2004.
[51] FEMA E74. Reducing the risks of non-structural earthquake damages. Federal
Emergency Management Agency, Practical Guide; 2012.
[52] FEMA 273. NEHRP guidelines for the seismic rehabilitation of buildings.
Washington DC: Federal Emergency Management Agency, U.S. Department of
Homeland Security; 1997.
[53] Morandi P, Albanesi L, Magenes G. In-plane experimental response of masonry
walls with thin shell and web clay units. Proceedings of the Vienna congress on
recent advances in earthquake engineering and structural dynamics, Vienna,
Austria. 2013.
[54] Oliaee M, Morandi P, Magenes G. In-plane seismic performance of a strong masonry
infill. Proceedings. of the 5th recent advances in mechanics and materials in design,
26–30 July 2015, Ponta Delgada/Azores, Portugal. 2015.
[55] Hak S, Morandi P, Magenes G. Local effects in the seismic design of RC frame
structures with masonry infills. Proceedings of the 4th ECCOMAS conference on
computational methods in structural dynamics and earthquake engineering, Kos
Island, Greece. 2013.
[56] Verlato N, Guidi G, da Porto F. Experimental testing and numerical modelling of
infill masonry walls subjected to in-plane damage. Proceedings of the 9th interna-
tional masonry conference, 7–9 July 2014, Guimarães, Portugal. 2014.
[57] Morandi P, Hak S, Magenes G. Simplified out-of-plane resistance verification for
slender clay masonry infills in RC frames. Atti del XV Convegno ANIDIS –
L’ingegneria Sismica in Italia; 30 Giugno – 4 Luglio 2013, Padova, Italy. 2013.