See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/323992102

Local effects on RC frames induced by AAC masonry infills through FEM

simulation of in-plane tests

Article  in  Bulletin of Earthquake Engineering · March 2018

DOI: 10.1007/s10518-018-0353-5

CITATIONS READS

4 213

3 authors:

Riccardo Milanesi Paolo Morandi

University of Pavia University of Pavia
17 PUBLICATIONS   42 CITATIONS    74 PUBLICATIONS   508 CITATIONS   

SEE PROFILE SEE PROFILE

Guido Magenes
University of Pavia
207 PUBLICATIONS   3,262 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Contribution in the preparation of the Italian Norms for Construction View project

Seismic behaviour of innovative masonry infills View project

All content following this page was uploaded by Riccardo Milanesi on 26 March 2018.

The user has requested enhancement of the downloaded file.

 Local effects on RC frames induced by AAC masonry infills through
FEM simulation of in-plane tests
Riccardo R. Milanesi*1, Paolo Morandi1,2 a, and Guido Magenes1,2 b
1
Department of Civil Engineering and Architecture, University of Pavia, Via Ferrata 3, 27100 Pavia, Italy
2
EUCENTRE, Via Ferrata 1, 27100 Pavia, Italy

Abstract. Unreinforced masonry infills are widely used in many parts of the world and it is com-
mon practice for seismic design to use simplified methods that usually do not take into account the
interaction between the infill and the structure. Starting from the 1950s, many researchers have in-
vestigated the lateral response of masonry infills focusing on several different topics. The scientific
interest on masonry infills is continuously raising due to the unsatisfactory seismic response of the
infilled frame structures observed during post-event inspections and to the difficulty to contrive a
widely scientifically and practical recognized solution. Although some modern codes consider the
presence of infills with some specifications to prevent damage in the masonry panels and global and
local effects on the structure, an effective evaluation of these detrimental effects has not been
achieved yet. Within this paper, a FEM simulation of in-plane pseudo-static cyclic tests on a RC
frame specimen infilled with unreinforced Autoclaved Aerated Concrete (AAC) masonry infill has
been performed in order to study accurately the influence and the interaction of the infill with the RC
structure. The experimental results performed by Calvi and Bolognini (1999), and Penna and Calvi
(2006) on one-bay one-storey full scale specimens are taken as reference. Non-linear static analyses
using a “meso-modelling” approach have been carried out. The masonry used in the model has been
calibrated according to tests of mechanical characterization and to in-plane cyclic tests on load-
bearing AAC masonry conducted by Costa et al. (2011). The analyses performed have allowed to
investigate the local effects on the frame and, in particular, the changes in the moment and shear de-
mands on the RC elements due to the presence of the AAC infill in comparison with the ones in the
bare structure, and to estimate the thrust and the contact length activated by the infill on the frame.

Keywords: local effects; AAC masonry; infilled structure; FEM simulation; infill-structure in-
teraction.


Corresponding author, Ph.D. Researcher,
E-mail: riccardo.milanesi@unipv.it
a
Ph.D. Senior Researcher, E-mail: paolo.morandi@unipv.it
b
Full Professor, E-mail: guido.magenes@unipv.it

Acknowledgements

The research, upon which this work is based, has been carried out at the University of
Pavia/EUCENTRE and at the Middle East Technical University, and it has been sponsored by the Erasmus
Mundus Programme and by the Executive Projects DPC-RELUIS 2013-2016 and DPC-RELUIS 2017. The
financial support for this study is gratefully acknowledged.
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

1 1 Introduction
2
3 Unreinforced masonry infills are nowadays widely present and used in European countries as non-
4 structural panels in RC frame structures. Despite the fact that field observations after major earthquakes
5 repeatedly point towards the high seismic vulnerability of masonry infills (see, i.e., Figure 1), not exclu-
6 sively in older buildings but also in more recently designed RC structures (Braga et al., 2011; Ricci et al.,
7 2011; Magenes et al., 2012; Manzini et al., 2012; Decanini et al., 2012; Celano et al., 2016), the current
8 code provisions provide few recommendations for these elements, being often insufficient, incomplete or
9 not clearly defined for the specific prevention of the damage and loss of human lives.
10 Therefore, the scientific interest on the seismic behaviour of the infills is extremely relevant, and many
11 studies have been accomplished during the last decades. Various distinct aspects have been investigated
12 through experimental (Mehrabi, 1994; Crisafulli, 1997; Calvi et al., 1999; Da Porto et al., 2012; Morandi et
13 al., 2017) and/or numerical (Asteris et al., 2011; Tarque et al., 2015; Lourenço, 1995; Stavridis, 2009)
14 studies.
15
(a) (b)

16 Figure 1: Masonry infill damage (a) after the Emilia earthquake in 2012 (Magenes et al., 2012); (b) after the Amatrice
17 earthquake in 2016 (Celano et al., 2016).
18
19 The poor seismic behaviour of the infills may not only be due to the inherent weakness of the traditional
20 unreinforced masonry but, in some cases, also be caused by deficiencies in design criteria. On the other
21 hand, the wide range of innovative technologies available in the market and possible improvements of the
22 seismic performance due to new investigation on this topic are leading to the development of new products
23 and innovative systems (i.e., Preti et al., 2012; Totoev et al., 2012, Morandi et al., 2016, Verlato et al.,
24 2016, Vintzileou et al., 2016).
25 The increased use of Autoclaved Aerated Concrete (AAC) masonry for both structural and non-
26 structural purposes thanks to its lightness, excellent thermal and acoustic insulation and the fire resistance
27 highlights the need of improving the knowledge about the lateral response of such infill material with fur-
28 ther experimental campaigns combined with adequate numerical studies.
29 In this paper, FEM simulation of in-plane pseudo-static cyclic tests on a RC frame specimen infilled
30 with unreinforced AAC masonry infill tested by Penna and Calvi (2006) has been performed, with the main
31 objective of investigating the influence and the interaction of the infill with the RC structure. Non-linear
32 static analyses using a “meso-modelling” approach have been carried out and the masonry constitutive law
33 has been calibrated according to tests of mechanical characterization and to in-plane cyclic tests on load-
34 bearing AAC masonry conducted by Costa et al. (2011). The analyses aim at an evaluation of the local ef-
35 fects on the frame regarding, in particular, the distribution of moment and shear demands on RC elements
36 due to the presence of the AAC infill compared to the bare structure, and the estimation of the additional
37 shear and the contact length activated by the infill on the frame.
38 The evaluation of the local effects of AAC masonry infills on RC columns represents a relevant and
39 quite novel aspect compared to past studies, more devoted on other materials, for instance clay masonry
40 (Hak et al., 2013b). Moreover, this study provides an original process for the estimation of the additional
41 shear demand on RC columns, characterized by a refined numerical simulation of a complete full-scale ex-
42 perimental campaign (from characterization to in-plane cyclic tests) that can be reproduced on other infill
43 typologies following the same framework.
44
2
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

45 2 Current seismic provisions for infilled structures according to European codes

46
47 According to Eurocode 8 (CEN, 2004c), the structures located in seismic prone regions have to be de-
48 signed and constructed such that the fundamental performance requirements associated with adequate de-
49 grees of reliability related to life safety and damage limitation are satisfied. Consequently, Ultimate Limit
50 States (ULS), associated with no-collapse or life safety requirements, have to be verified at a reference de-
51 sign seismic action for which a reference return period of TULS is defined in the National Annex of each
52 country. Eurocode 8 states that the structures shall be designed and constructed to withstand the ULS seis-
53 mic actions without local or global collapse, thus retaining the structural integrity and the residual load
54 bearing capacity after the seismic event. The verification of the Damage Limit State (DLS) is required for
55 the corresponding seismic action with a return period TDLS, having a larger probability of occurrence than
56 the design seismic action. Structures shall be designed and constructed to withstand the DLS seismic action
57 in terms of displacements, in order to limit the occurrence of damage and limitations of use, the costs of
58 which would be disproportionally high in comparison with the costs of the structure itself. In Figure 2 the
59 design procedure to be followed according to Eurocode 8 provisions for the seismic design of RC structures
60 is reported.
61

62
63 Figure 2: Current design procedure for masonry infilled RC structures according to Eurocode 8.
64
65 The damage limitation requirements at DLS, according to Eurocode 8, are considered satisfied if the in-
66 ter-storey drift dr,DLS,j/hj (dr,DLS,,j is the inter-storey displacement for storey j induced by the damage limita-
67 tion seismic action, and hj is the storey height) do not exceed the inter-storey drift limitation dDLS equal to
68 0.50% for buildings with brittle non-structural elements attached to the structure, equal to 0.75% with duc-
69 tile non-structural elements and equal to 1.00% in case of structures without non-structural elements or
70 with elements that do not interfere with the structure. No further specifications are provided referring to the
71 definition of brittle and ductile non-structural elements. These codified drift limits are deduced to adjust the
72 results obtained from common design approach where the infills are usually neglected in the global analy-
73 sis.
74 At the ULS, the safety verification has to be accomplished in terms of resistance to seismic action ef-
75 fects for both structural and non-structural elements. In particular, for non-structural elements that might, in
76 case of failure, cause risk for human life or affect the main structure of the building or services of critical
77 facilities, as the infills, the verification is usually performed checking that the out-of-plane resistance is
78 larger than the out-of-plane seismic force acting on the infill. Thus, the effects of the seismic action may be
79 determined applying a horizontal out-of-plane force Fa, acting at the centre of mass of the infill, and a
80 structural inter-storey drift limitation is usually considered to avoid widespread damages within the infill.
81 For the local effects, Eurocode 8 (CEN, 2004c) suggests a critical length equal to the total height for the
82 columns at the ground floor or in case the infill is in full contact with the column only on one side (Figure
83 3). Different criteria for a simplified evaluation of the local effects based on the estimation of the strut
84 forces and of the contact length included in some seismic codes and research studies are discussed in Sec-
85 tion 6.1.
86

3
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

(a) (b) (c)

87 Figure 3: Critical length of the columns: (a) in full contact with the infill on one side, and particular additional cases: (b) col-
88 umns at the ground floor, and (c) columns in partial contact with the infill (Hak et al., 2013a).
89
90 3 Description of the adopted Finite Element Model
91
92 Masonry infills can be modelled with different modelling approaches developed in the past. Crisafulli et
93 al. (2000) classified the models used for RC infilled structured into two groups: micro-model, also called
94 local models, where the structure is divided into numerous elements to take into account the local effects in
95 details; and the macro-models, also called simplified models, which are based on a physical understanding
96 of the behaviour of the infill panel. Furthermore, Lourenço (1995) proposed to classify the micro-model
97 depending on the level of refinement. The first possibility is to represent the masonry as a homogenous ma-
98 terial and, sometimes, these models are defined as “meso-models”. It is also possible to model the masonry
99 as composed by masonry unit continuous elements and interface elements as mortar joints. A third ap-
100 proach consists in modelling the masonry as two-phase materials; masonry units and mortar joints are mod-
101 elled as two different continuous elements and their mechanical behaviour is specified separately. For the
102 present study, a meso-model has been used with the aid of TNO Diana software (2010).
103 Finite element approaches for the modelling of a masonry infill in a reinforced concrete frame may usu-
104 ally refer to two well known crack models: discrete and smeared (Roots, 1988). The discrete crack criterion
105 appears to be able to reach more detailed results, however, it suffers from two major drawbacks: it necessi-
106 tates a continuous change in nodal connectivity, which unfits the nature of the finite element displacement
107 method and force the crack to follow a predefined path along the element edges, which may decrease the
108 accuracy of the approach. In the smeared crack criterion the materials are instead idealized to be a homoge-
109 neous continuum. The smeared crack criterion can be categorized into a fixed and rotating smeared crack
110 approach. In the present work, the total strain crack model, originally proposed by Vecchio and Collins
111 (1986), has been used. In both existing approaches (fixed and rotating), the compressive stress is evaluated
112 in the rotated coordinate system. However, in the fixed approach case, the behaviour in compression is
113 evaluated in the fixed coordinate system which defines the direction of cracks. Moreover, in the fixed crack
114 model it is necessary to model the degradation of the shear behaviour along the crack whereas, in the rotat-
115 ing crack model, the direction of the principal stress is assumed to coincide with the direction of the princi-
116 pal strains; in fact, no shear strain occurs perpendicular to the cracking and only two normal stress
117 components may be calculated using stress-strain models. Here, the concrete and the AAC masonry infill
118 have been modelled using the smeared total strain rotating crack model.
119

120
121 Figure 4: Reference coordinate system for elements (TNO Diana, 2010).
122
123 The RC frame and the masonry infill have been modelled with an eight-node quadrilateral iso-
124 parametric plane stress element (Figure 5a) based on quadratic interpolation and Gauss integration with 4
125 integration points per element. The frame-infill interface has been modelled using six-node interface ele-
126 ments (Figure 5b).
127

4
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

(a) (b)

128 Figure 5: (a) 8-node quadrilateral iso-parametric plane stress element; (b) Frame-infill 6-node interface element.
129
130 The same constitutive model has been used for the concrete of the frame and for the AAC masonry infill.
131 A parabolic behaviour under compression has been assumed for the concrete and the AAC masonry (Figure
132 6a) and the compressive strength, along with the compressive fracture energy, has been defined. The “crack
133 bandwidth” (h), which is contingent on the element size and the type of element used, has been computed
134 automatically by the finite element software in order to avoid the “size effect” and in this particular case
135 corresponds to the characteristic element length. The reduction of the compressive strength due to lateral
136 cracking has been considered according to the relationship proposed by Vecchio and Collins (1986) (Figure
137 6b). In addition, the increase of the compressive strength due to lateral confinement has been taken into
138 account, using the model proposed by Selby and Vecchio (1993).
139 The tensile behaviour has been idealized according to the model developed by Hordijk (1991) (Figure
140 6c). For the concrete, the tensile strength fctm has been computed according to Eurocode 2 (CEN, 2004a),
141 and the Mode-I fracture energy has been considered according to CEB-FIP Model Code 2010 (CEB-FIP,
142 2010), as reported in Equation 1:
GF I  0.70  73 f cm
0.18
 1.30  73 f cm
0.18
(1)
143 where fcm is the mean value of concrete cylinder compressive strength and the minimum value of the inter-
144 val of the Mode-I fracture energy has been assumed in the model. For the masonry, the fracture energy has
145 been estimated through a numerical calibration on diagonal compressive tests; as an alternative, an elastic-
146 brittle tensile constitutive model has also been used, where the material has a perfect elastic behaviour up to
147 the tensile strength (ft), after which the stress drops to zero.
148 For the reinforcing steel, the Von Mises yield criterion was adopted, with a bilinear stress-strain curve
149 in uniaxial tension. The reinforcement steel bars have been considered as “embedded reinforcement” which
150 add stiffness to the finite element model, as discussed by several authors (i.e Barzegar et al., 1994 and
151 Markou et al., 2012). In this modelling technique, reinforcing bars, which are explicitly modelled according
152 to their real location and not spread, are assumed not to occupy any space or mass in the finite element
153 model and the addition of embedded bars do not provide any weight to the mother element. The strains of
154 the reinforcement are computed from the displacement field of the mother elements assuming perfect bond
155 with the surrounding material.
156 The frame-infill interface elements employ a Coulomb Friction criterion (Figure 6d). The linear normal
157 and tangential stiffness of the interface elements were assumed to be equal. The cohesion c and the friction
158 coefficient tan were computed from initial shear tests according to EN 1052-3 (EN, 2007). This model
159 assumes that a gap forms if the tensile strength is exceeded.
160
(a) (b) (c) (d)

161 Figure 6: (a) Parabolic compressive behaviour which can be used in rotating crack model; (b) Vecchio and Collins relation-
162 ship for reduction factor due to lateral cracking; (c) Hordijk tensile constitutive law used in rotating crack model (red line);
163 brittle tensile constitutive law (green line, only for masonry); (d) Coulomb friction criterion with tensile strength limit.
164

5
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

165 4 Calibration of the AAC masonry according to tests of characterization and to in-
166 plane cyclic tests of load-bearing walls
167
168 In order to evaluate the mechanical characteristics of the AAC masonry, different types of test on small
169 specimens were performed in the study on the AAC infills by Penna and Calvi (2006). In this work, the
170 results of the vertical, horizontal and diagonal compressive strength tests have been taken into considera-
171 tion. In addition, in order to investigate the seismic response of load-bearing masonry piers, different in-
172 plane cyclic tests have been executed in the framework of the research conducted by Costa et al. (2011).
173 Some of the most relevant results of the monotonic push-over analyses obtained by the FEM modelling
174 calibrated on the results of the tests of characterization and of the in-plane cyclic tests on masonry piers are
175 hereafter presented.
176 The masonry studied was composed by 30 cm thick AAC unit and both horizontal and vertical thin
177 layer mortar (about 1 mm thick).
178
179 4.1 Calibration of the model of the AAC masonry on vertical, horizontal and diagonal compression tests
180
181 Figure 7 illustrates some phases of the vertical and the horizontal compression tests on masonry wallets
182 conducted according to EN 1052-1 (EN, 2001). The values of deformations were measured through linear
183 potentiometers. Three tests of cyclic vertical and three of cyclic horizontal compression on 0.94 m x 1.25 m
184 x 0.30 m (length l, width b, thickness tw) specimens have been conducted (see Figure 7), along with seven
185 monotonic tests of diagonal compression on 0.63 m x 0.75 m x 0.30 m specimens (Figure 9a). A mean
186 value of about 2.0 MPa was found for the compressive strength of the wallets and of 0.28 MPa for the shear
187 strength for diagonal cracking (=0.5∙Fmax /(tw ∙(b + l)/2), where Fmax is the maximum applied load).
188 The masonry has been modelled as an isotropic homogenous material. The elastic-brittle and the
189 Hordijk constitutive models for the tensile behaviour of the masonry have been used and compared in order
190 to select the more suitable tensile constitutive law.
191 For both vertical and lateral compression, the stress-strain relations of the three cyclic tests are reported
192 in Figure 8, together with the results of the FEM pushover analyses (Figure 8a for the vertical and Figure
193 8b for the horizontal compression). The results appear to be in good agreement, although the peak stress is
194 not perfectly attained in both the cases (vertical and horizontal compression). The two laws used for model-
195 ling the tensile behaviour of the material have provided similar results in the case of the vertical compres-
196 sion tests but a different deformation capacity for the case of the horizontal compression tests.
197
(a) (b)

198 Figure 7: Tests on AAC masonry carried out at the University of Pavia (Penna and Calvi, 2006): (a) vertical compression test;
199 (b) horizontal compression test.
200

6
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

Comparison between experimental and FEM analysis - Vertical Compression Comparison between experimental and FEM analysis - Horizontal Compression
(a) 2.5
(b) 2

1.8
2 1.6

1.4

Stress (MPa)
Stress (MPa)

1.5 1.2

1 0.8

0.6

0.5 0.4 FEM analysis - Brittle

FEM analysis - Brittle
FEM analysis - Hordijk 0.2 FEM analysis - Hordijk
Experimental Experimental
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 0 1 2 3 4 5 6
Strain (-) -3 Strain (-) x 10
-3
x 10

201 Figure 8: Comparison between experimental and FEM analysis: (a) vertical compression; (b) horizontal compression.
202
203 For the diagonal compression tests, the comparison between the experimental and the numerical results
204 with Hordijk and elastic-brittle tensile behaviour is shown in Figure 9b. The same parameters adopted for
205 the vertical/horizontal compression tests are used. The Hordijk tensile behaviour presents a better accuracy
206 in the descending branch, whereas the brittle model appears to predict slightly better the peak and its loca-
207 tion in the load-displacement coordinates. In both cases, the stiffness and the obtained ultimate displace-
208 ment have found to be accurate. The Hordijk tensile behaviour has been preferred to the elastic-brittle one
209 because of better results of the descending branch in the post-peak region. In Table 1, the obtained maxi-
210 mum loads are reported.
211
Max Load Horizontal displacement at peak
Experimental average 115.1 kN 0.281 mm
Elastic-brittle constitutive law in tension 116.4 kN 0.311 mm
Hordijk constitutive law in tension 118.0 kN 0.315 mm
212 Table 1: Comparison of the peak values between experimental results and the one obtained from the FEM analysis with the
213 two tensile constitutive models.
214 Comparison between experimental and FEM analysis - Diagonal Compression
150
(a) (b) FEM analysis - Brittle
FEM analysis - Hordijk
Experimental

100
Load (kN)

50

0
0 0.2 0.4 0.6 0.8 1
Displacement (mm)

215 Figure 9: (a) Diagonal compression test on AAC masonry carried out at the University of Pavia; (b) Comparison between
216 experimental and FEM analysis of the diagonal compression tests.
217
218 4.2 Calibration of AAC masonry according to in-plane cyclic experimental results on load-bearing piers
219
220 Two of the in-plane cyclic tests on load-bearing AAC walls conducted in the study performed by Costa
221 et al. (2011) have been considered. The two piers (1 and 2) are 2.75 m high, 1.5 m long and 0.30 m thick
222 with different vertical applied loads (200 kN and 300 kN, respectively) and same “cantilever” boundary
223 conditions.
224 For the numerical simulation of the tests, the masonry has been modelled using the tensile constitutive
225 law proposed by Hordijk and the parameters calibrated through the FEM analyses conducted on the vertical,
226 the horizontal and the diagonal compression tests as discussed in the previous sections.

7
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

227 The numerical analysis conducted on wall 1 (applied vertical load of 200 kN) has shown a “rocking-
228 type” failure with a horizontal crack at the bottom part of the masonry wall; such a mechanism is found to
229 be very similar to the one observed during the tests. The comparison between experimental and analysis
230 data highlights a similar overall F-D envelope response, as shown in Figure 10a, where also the stiffness
231 obtained from finite element analysis is very similar to the real one. The maximum force attained is ap-
232 proximately 10% higher than the experimental one: 54.0 kN in the FEM analysis and 48.8 kN in the test.
233 The results acquired from pushover analysis conducted on wall 2, with an applied vertical load of 300
234 kN, are also capable to predict with enough precision the stiffness, the peak lateral force and the overall
235 behaviour of the specimen, as illustrated in Figure 10b, where the F-D envelope from the analysis matches
236 rather accurately the envelope of the test.
237
Specimen 1 - Slender wall - 200 kN Specimen 2 - Slender wall - 300 kN
(a) 60 (b)
60

40
40
Lateral Force (kN)

Lateral Force (kN)

20 20

0 0

-20
-20

-40
-40
FEM Analysis FEM Analysis
Experiment -60 Experiment
-60
-0.6 -0.4 -0.2 0 0.2 0.4 0.6 -0.6 -0.4 -0.2 0 0.2 0.4 0.6
Drift (%) Drift (%)

238 Figure 10: Comparison between the experimental and the pushover analysis results. (a) Specimen 1; (b) Specimen 2.
239
240 5 Numerical simulation of the in-plane cyclic experimental responses of a RC bare
241 frame and a RC frame with AAC unreinforced masonry infill
242
243 The nonlinear static FEM analyses of a full scale one-storey one-bay RC frame without infill and of a
244 fully infilled RC frame with AAC unreinforced masonry have been performed, using the parameters of the
245 materials described in Sections 3 and 4.
246
247 5.1 Numerical simulation of the in-plane seismic behaviour of a RC bare frame
248
249 In the late '90s, experimental tests on different weak clay masonry infills in reinforced concrete frames
250 have been executed in an extensive research performed by Calvi and Bolognini (1999 and 2001). Full scale,
251 one-storey, one-bay frame specimens were designed according to modern seismic codes and subjected to
252 in-plane and out-of-plane tests. The experimental loading scenario consisted of a vertical constant load of
253 400 kN applied at the top of each column and a horizontal displacement imposed to the beam in order to
254 perform a cyclic pseudo-static in-plane test up to 3.6% drift for the bare frame. Figure 11 illustrates the di-
255 mensions and the reinforcement of the bare frame (Figure 11a), the point of application of the loads during
256 the tests and the Force-Displacement curve of the bare frame obtained from the in-plane test (Figure 11b).
257 The in-plane cyclic response of the bare frame has been simulated firstly in order to obtain a reliable fi-
258 nite element model of the RC frame, and, at a later stage, to study the influence of the AAC masonry infill
259 (see Section 6). Therefore, it is extremely important to correctly interpret the behaviour of the bare RC
260 frame to better understand the influence of the infill on the structure.
261 A phased analysis has been performed to replicate the experimental testing procedure; firstly the vertical
262 loads were applied, secondly an incremental imposed displacement was imposed (pushover analysis). Four
263 materials have been used according to the tests of characterization on the materials: the concrete of the
264 beam, the concrete of the column, the steel of the longitudinal reinforcement and the steel of the transversal
265 reinforcement. According to the reinforcement details, a clear cover of the concrete elements equal to 27
266 mm has been used. Only some of the experimental data on the mechanical characteristics of the concrete
267 were available and, therefore, some values have been assumed making reference to the Eurocode 2 (CEN,

8
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

268 2004a) and the CEB FIP 2010 Model Code (CEB/FIP, 2010). The properties of the concrete and of the re-
269 inforcement steel used in the FEM model are reported in Table 2.
270
(a)

(b)

271 Figure 11: (a) Dimensions and reinforcement of the RC frame; (b) Force-Displacement curve of the bare frame obtained from
272 in-plane cyclic pseudo-static test and application points of the loads during the test. (Calvi and Bolognini, 1999 and 2001)
273

9
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

Experimental FEM analysis Standard formulae

COLUMN CONCRETE
Ecm from Table
Young Modulus [MPa] --- 24303 0.8 ∙Ecm=0.8 ∙22000∙[(fcm)/10]0.3
3.1 EC2
Poisson ratio --- 0.2 0.2 3.1.3(4) EC2
Density [kg/m3] --- 2500 --- ---
Tension Strength (fctm) [MPa] --- 2.307 fctm=0.30 ∙fck(2/3) Table 3.1 EC2
Tension Fracture Energy 0.7 ∙73∙ fcm0.18 ≤ GFI ≤
--- 0.0939 CEB FIP 2010
(GFI) [Nmm/mm2] 1.3∙ 73 fcm0.18
Compression Strength (fcm) From tests of
29.32 29.32 ---
[MPa] characterization
BEAM CONCRETE
Ecm from Table
Young Modulus [MPa] --- 25532 0.8 ∙Ecm=0.8 ∙22000∙[(fcm)/10]0.3
3.1 EC2
Poisson ratio --- 0.2 0.2 3.1.3(4) EC2
Density [kg/m3] --- 2500 --- ---
Tension Strength (fctm) [MPa] --- 2.671 fctm=0.30∙ fck(2/3) Table 3.1 EC2
Tension Fracture Energy 0.7 ∙73∙ fcm0.18 ≤ GFI ≤
--- 0.0967 CEB FIP 2010
(GFI) [Nmm/mm2] 1.3∙ 73 fcm0.18
Compression Strength (fcm) From tests of
34.56 34.56 ---
[MPa] characterization
LONGITUDINAL REINFORCEMENT STEEL
Young Modulus [MPa] --- 200000 200000 3.2.7(4) EC2
Poisson ratio --- 0.3 0.3 3.2.5(1) EC3
From tests of
Yield Strength [MPa] 558 558 ---
characterization
From tests of
Ultimate Strength [MPa] 649 649 ---
characterization
Ultimate Strain --- 0.023 --- 3.3.6(7) EC2
TRANSVERSAL REINFORCEMENT STEEL
Young Modulus [MPa] --- 200000 200000 3.2.7(4) EC2
Poisson ratio --- 0.3 0.3 3.2.5(1) EC3
From tests of
Yield Strength [MPa] 557 557 ---
characterization
From tests of
Ultimate Strength [MPa] 630 630 ---
characterization
Ultimate Strain --- 0.023 --- 3.3.6(7) EC2
274 Table 2: Material properties values used for concrete and reinforcement steel.
275
276 Calvi and Bolognini (1999 and 2001) estimated the initial stiffness of the RC bare frame equal to 12087
277 kN/m. The maximum horizontal force reached in the test, according to the envelope curves, was 227.7 kN.
278 As reported in Figure 12a, the pushover analysis conducted on the FEM model has been able to accurately
279 predict the overall response of the bare frame, especially the peak strength (equal to 225.5 kN) and the
280 post-yielding behaviour. In order to match the results in terms of initial stiffness (in the FEM model equal
281 to 14316 kN/m), a value for the Young elastic modulus E equal to 80% of the average elastic modulus
282 (Ecm), evaluated according to the expressions in the EC2, was used; this approximation is however in line
283 with the common dispersion of the values of this parameter. In Figure 12b, the deformed shape at the last
284 drift obtained from numerical analysis (4.3% drift) has been reported.
285
10
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

Comparison between Force-Deformation curves

(a) 300
FEM Analysis
(b)
Experimental Cyclic
200

100
Force (kN)

-100

-200 [1]
-300
-5 -4 -3 -2 -1 0 1 2 3 4 5
Drift (%)

286 Figure 12: (a) Comparison Force-Displacement curves: pushover FEM analysis versus experimental cyclic results. (b) De-
287 formed shape at 4.3% drift.
288
289 5.2 Numerical simulation of the in-plane seismic behaviour of a AAC masonry infilled RC frame
290
291 In 2006, several tests on AAC masonry infills have been conducted by Penna and Calvi (2006), using
292 the same RC frames of the research by Calvi and Bolognini (1999 and 2001). Different AAC infills were
293 tested: unreinforced, reinforced with a RC mid-height beam, reinforced with steel bars in the bed joints and
294 with and without opening. Within this study, only the in-plane test on the unreinforced AAC infill has been
295 considered.
296 In Figure 13 the cracking pattern at the end of the test (Figure 13a) and the corresponding Force-
297 Displacement curve (Figure 13b) are reported; the specimen has been tested up to a drift of 1.2%. This test
298 demonstrated that the masonry infilled frame resulted in a significantly different lateral response when
299 compared to the bare frame. A much lower maximum drift was reached at ultimate conditions and the in-
300 filled RC frame presented stiffer behaviour and higher maximum force peak.
301
(a) (b)

302 Figure 13: Unreinforced AAC infill. (a) Crack pattern, (b) Force-Displacement curve (Penna and Calvi, 2006).
303
304 The same phased FEM analysis conducted for the model described in Section 5.1 has been carried out
305 herein. The material properties of the concrete and of the steel rebars assumed for the bare frame have been
306 used for this model. As shown in Figure 14a, an interface material was included between the masonry and
307 the RC frame and the infill was considered fixed at the base; although interface elements could be adopted
308 also at the bottom of the masonry, the assumption of fixed boundary condition is justified by the absence of
309 any visible crack in the performed in-plane test.
310 The properties of the AAC masonry have been obtained as described in Section 4, whereas the charac-
311 teristics of the interface mortar (cohesion, friction angle, tensile strength) have been calibrated according to
312 the results of the in-plane cyclic test on the infilled frame. However, the tensile strength of the interface
313 does not seem to significantly affect the lateral response of the AAC masonry infilled frame and it is, in any
314 case, very difficult to be evaluated, also due to uncertainties in the execution of the mortar joints adjacent to
315 the columns. The property values of the AAC masonry and interface mortar are reported in Table 3.

11
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

316 The maximum strength reached during the test, according to the average envelope, was 439 kN and the
317 horizontal load peak for the finite element analysis is 432 kN. Figure 14b shows that the results of the FEM
318 analysis are able to predict fairly well the overall behaviour of the structure.
319
Experimental FEM analysis
AAC MASONRY
Young Modulus [MPa] 1498 1498 From tests of characterization
Density [kg/m3] 484 484 From tests of characterization
Tension Strength [MPa] 0.279 0.279 From tests of characterization
Tension Fracture Energy [Nmm/mm2] --- 0.0567 Calibrated from diagonal tests
Compression Strength [MPa] 2.0 2.0 From tests of characterization
INTERFACE MORTAR
Cohesion [MPa] --- 1.15 Calibrated on cyclic tests on infills
Friction Angle --- 0.31 Calibrated on cyclic tests on infills
Tensile Strength [MPa] --- 0.10 Calibrated on cyclic tests on infills
320 Table 3: Material property values used for masonry and interface mortar.
321
Comparison between Force-Deformation curves
(a) (b) 500 FEM Analysis
400 Experimental Cyclic
300

200

100
Force (kN)

-100

-200

-300

-400

-500
-1.5 -1 -0.5 0 0.5 1 1.5
Drift (%)

322 Figure 14: (a) Sketch of the model adopted for the infilled RC frame; (b) Comparison Force-Deformation curves.
323
324 Figure 15a, b and c represent the crack pattern at 0.40%, 0.75% and 1.10% drift, respectively, and show the
325 creation of a diagonal strut. The cracking pattern obtained by the numerical simulation appears to be in line
326 to that of the cyclic tests, as reported in Figure 16a for the case of 0.40% drift. Figure 16b shows a com-
327 parison between the experimental and analytical crack patterns considering the experimental cracks due to
328 the load applied in the same direction of the pushover analysis. Moreover, Figure 16b indicates that the
329 results of the FEM analysis is able to capture the main diagonal cracks, although it cannot predict in a pre-
330 cise way some of the minor cracks. The damage pattern in the RC frame shown in Figure 15 and Figure 16,
331 also highlights the formation of inclined light cracks, visible after 0.30% drift, at the edges of the columns.
332 The progression of such shear cracking as a function of the in-plane drift was sensibly lower than the evo-
333 lution of the damage pattern in the masonry of the infill and no shear failure in the RC elements occurred.
334
(a) (b) (c)

335 Figure 15: Crack pattern at 0.40% drift (a); 0.75% drift (b); 1.10% drift (c).
336

12
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

(a) (b)

337 Figure 16: (a) Experimental crack pattern at 0.40% drift [3]; (b) Comparison between experimental and analytical crack pat-
338 terns at 0.40% drift.
339
340 Several studies present in literature have clearly demonstrated, both experimentally and numerically,
341 how the global behaviour of the structure can be modified due to the presence of infills. Indeed, the varia-
342 tion of the deformed shape between the bare and the infilled frames is significant and the results at different
343 drift (0.5% and 1.2%), reported in Figure 17, show that for increasing levels of drift the response of the
344 structure diverges from the bare frame configuration. Moreover, Figure 18 reports how the applied force is
345 subdivided into the base shear absorbed by the infill and the one taken by the RC columns.
346
(a) (b)

347 Figure 17: Comparison of the deformed shape of the RC bare frame (blue) and the infilled RC frame (red) at different drifts:
348 (a) drift 0.50%, (b) drift 1.20%.
349
(a) (b)

350 Figure 18: Base shear subdivision among RC columns and infill at different drifts: (a) drift 0.50%, (b) drift 1.20%.
351
352 6 Study on the local effects on the RC columns due to the presence of the infill
353
354 As reported by Hak et al. (2013a), the possible occurrence of local effects on structural elements of in-
355 filled RC frames with masonry infills is widely recognized and supported by field evidence from recent
356 earthquakes events. In particular, local damage and eventually brittle failure can commonly be caused on
357 columns, which are in partial contact with masonry infill, causing a reduction of the clear height of the col-
358 umn and, hence, inducing locally increased shear and displacement demands. However, detrimental effects
359 on columns of RC frames structures can be caused by the interaction with masonry infills also in the case of

13
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

360 full contact along the height of the column, in particular when masonry infill typologies of high strength
361 and stiffness properties are used in construction, and/or when the infill is located only on one side of the
362 column. Specifically, additional concentrated shear demands may be imposed on the column at its ends in
363 the region of contact with the masonry infill due to the activation of compressive diagonal strut forces, pos-
364 sibly causing local column damage or shear failure.
365 Criteria for the evaluation of infill strength and contact length are present in some modern codes and in
366 literature and are briefly described in the following sub-section. Afterwards, the results coming from the
367 FEM pushover analysis are discussed and compared with the outcomes of these simplified criteria, in order
368 to understand which is able to better fit the results of the numerical analyses, in terms of estimation of the
369 additional forces and contact length.
370
371 6.1 Criteria for the evaluation of infill strength and contact length
372
373 The influence due the presence of the infill and the local effects on RC members have been object of
374 various studies. Since the infill has been usually represented as single or multiple equivalent struts through
375 macro-modelling analysis, the need to define, through experimentally proved analytical criteria some key
376 values such as the maximum resistance, the dimensions (width and thickness) of the equivalent strut and
377 the stiffness, has been fulfilled starting from the earliest studies on the seismic behaviour of infilled struc-
378 tures. Due to the fact that is out of the scope of the present work to provide a comprehensive literature re-
379 view on the topic, which can be found in many works such as Asteris et al. (2011) and Tarque et al. (2015),
380 only few of the most relevant approaches to consider the local effects are discussed within this paper.
381 The contact length lc of the column, over which the equivalent diagonal strut force of the infill acts
382 causing a local pressure fs and introducing a concentration of forces (Figure 19), should be verified in shear
383 for the demand VC,Ed,lc evaluated, according to Eurocode 8, as the smaller of the following two shear forces
384 (Equation 2):
385  the shear demand VC,Ed,w equal to the horizontal component Vw,hor of the infill, that can be assumed
386 equal to the strength of the panel according to Eurocode 6 - Part 1 (CEN, 2004b) (Equation 3);
387  the shear demand VC,Ed,M determined following the capacity design principle (Equation 4), assuming
388 that the flexural column capacity develops at the ends of the contact length lc, as shown in Figure 19.
389
VC , Ed ,lc  min VC , Ed ,w ;VC , Ed , M  (2)
VC ,Ed ,w  Vw,hor  f v 0twlw (3)
2M C , Rd
VC , Ed , M   Rd (4)
lc
390 where fv0 is the initial shear strength of bed-joints, tw is the thickness of the panel, Lw is the length of the
391 panel, MC,Rd is the design resisting moment of the column, Rd is the over-strength coefficient and lc is the
392 contact length.
393

394 Figure 19: Local effects on an infill adjacent column.

395

14
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

396 In addition to the procedure of the Eurocodes, also other international codes present indications on how
397 to deal with the presence of the infills and the related local effects on RC members.
398 For example, in FEMA 306, that includes part of the studies conducted by Mainstone (1971 and 1974)
399 and Saneinejad and Hobbs (1995), the width of the equivalent strut bw is expressed in terms of ratio with
400 the diagonal length of the infill dw according to Equation 5 ( is the slenderness of the infill according to
401 Stafford Smith, 1967 and h is the centreline height of the frame, see Figure 19) and the maximum shear of
402 the infills (Vw,hor) is calculated as the minimum between the expressions for the sliding shear failure of the
403 infill, the compression failure and the diagonal cracking failure.
bw
 0.175(h)0.4 (5)
dw
404 Another widely used approach is the one proposed by Bertoldi et al. (1993), where bw/dw is expressed
405 according to Equation 6 and the infill strength is computed as the minimum between four different failure
406 modes (ultimate compression at the centre of the infill, ultimate compression at the corners of the infill,
407 sliding shear failure of the masonry and diagonal cracking of the infill).
bw K1
  K2 (6)
d w h
408 Finally, Paulay and Priestley (1992) have proposed a method for the evaluation of the equivalent strut
409 width according to Equation 7, and of the infill strength as the minimum between the sliding shear failure
410 of the infill and the compression failure of the diagonal strut.
bw
 0.25 (7)
dw
411 Further detail on these simplified criteria can be found in Morandi et al. (2018).
412 From an interpretation of the different approaches, the contact length on the column can be computed
413 by considering the vertical component of the equivalent strut width bw (lc=bw/cos) in the case proposed by
414 Mainstone and by EC8 (see Figure 20a), whereas, in the other considered approaches, the contact length is
415 estimated subdividing it between the column and the beam (lc=0.5bw/cos, Figure 20b). However, EC8
416 does not provide any explicit indication on the estimation of the strut width bw.
417
(a) (b)

418 Figure 20: Equivalent strut and contact length depending from the width of the equivalent strut for: (a) Mainstone (1971-1974)
419 and EC8, (b) Bertoldi et al. (1993) and Paulay and Priestley (1992) approaches.
420
421 6.2 Local effects on RC elements from pushover analysis on the AAC masonry infill
422
423 The bending moment (Figure 21 and Figure 23) and shear demands (Figure 22 and Figure 24) on the two
424 RC columns of the bare and the infilled frame have been evaluated from the pushover analysis as a function
425 of the imposed drifts. A complete comparison of the moment and shear demands of bare and infilled frame
426 at increasing drifts is reported in Table 4. Relevant differences in the intensity and distribution of the inter-
427 nal forces between the infilled and the bare frame are evident. The changes start from the beginning, at very
428 low levels of in-plane drift demand. For the infilled frame, the shear is found to be constant up to the point
429 where the local pressure of the infill begins to be activated and then it increases almost linearly up to the

15
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

430 column edges; the maximum shear on the RC columns is always much larger than the one of the bare frame.
431 The moment distribution in the columns of the infilled frame is obviously very different from the one of the
432 bare frame; the maximum values are included, with few exceptions, between the maximum positive and
433 negative moments of the bare structure.
434 Furthermore, the results reported in Table 4 and Figure 24 show a clear change of the shear demand de-
435 pending on the imposed displacement, since the local pressure produced by the concentration of forces at
436 the edges of the columns shows a gradual decrease of the maximum peak stresses for higher level of drifts,
437 maybe also related to previous damages within the masonry, despite an increase of both the contact length
438 and the additional shear.
439 The large increase in shear action applied to the RC frame columns with respect to the case without infills
440 are somehow in accordance with experimental and numerical findings of past studies conducted on differ-
441 ent infill materials and typologies (i.e., Buonopane et al., 1999, Asteris 2003, D'Ayala et al., 2009, Bolis et
442 al., 2017).
443 Moreover, although the actual shear cracking resistance of the RC columns has been exceeded after 0.30%
444 drift (see Figure 24), in agreement with the damage pattern discussed in Section 5.2, the ultimate shear re-
445 sistance of the section has never been attained, as observed in the experimental test, where no shear failure
446 has occurred.
447
(a) Bare Frame - Moment demand - Column 1 (b) Bare Frame - Moment demand - Column 2

0.1% drift
2.5 2.5
0.2% drift
0.3% drift
0.5% drift
2 2
0.75% drift
1% drift
Height (m)

Height (m)

1.2% drift
1.5 1.5

0.1% drift
1 0.2% drift 1
0.3% drift
0.5% drift
0.5 0.75% drift 0.5
1% drift
1.2% drift
0 0
-100 -50 0 50 -100 -50 0 50
Moment (kNm) Moment (kNm)

448 Figure 21: RC bare frame moment demand at different drifts (a) Column 1 (windward), (b) Column 2 (leeward).
449
(a) Infilled Frame - Moment demand - Column 1 (b) Infilled Frame - Moment demand - Column 2

0.1% drift
2.5 2.5
0.2% drift
0.3% drift
0.5% drift
2 2
0.75% drift
1% drift
Height (m)

Height (m)

1.2% drift
1.5 1.5

0.1% drift
1 0.2% drift 1
0.3% drift
0.5% drift
0.5 0.75% drift 0.5
1% drift
1.2% drift
0 0
-100 -50 0 50 -100 -50 0 50
Moment (kNm) Moment (kNm)

450 Figure 22: AAC infilled RC frame moment demand at different drifts (a) Column 1 (windward), (b) Column 2 (leeward).
451

16
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

(a) Bare Frame - Shear Demand - Column 1 (b) Bare Frame - Shear Demand - Column 2

0.1% drift
2.5 2.5
0.2% drift
0.3% drift
0.5% drift
2 2
0.75% drift
1% drift
Height (m)

Height (m)
1.2% drift
1.5 1.5

0.1% drift
1 0.2% drift 1
0.3% drift
0.5% drift
0.5 0.75% drift 0.5
1% drift
1.2% drift
0 0
0 100 200 300 400 0 100 200 300 400
Shear (kN) Shear (kN)

452 Figure 23: RC bare frame shear demand at different drifts (a) Column 1 (windward), (b) Column 2 (leeward).
453
(a) Infilled Frame - Shear Demand - Column 1 (b) Infilled Frame - Shear Demand - Column 2

0.1% drift
2.5 2.5
0.2% drift
0.3% drift
0.5% drift
2 2
0.75% drift
1% drift
Height (m)

Height (m)
1.2% drift
1.5 1.5

0.1% drift
1 0.2% drift 1
0.3% drift
0.5% drift
0.5 0.75% drift 0.5
1% drift
1.2% drift
0 0
0 100 200 300 400 0 100 200 300 400
Shear (kN) Shear (kN)

454 Figure 24: AAC infilled RC frame shear demand at different drifts (a) Column 1 (windward), (b) Column 2 (leeward).
455

17
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

Moment demand comparison Shear demand comparison

Moment demand comparison at 0.10% drift Shear demand comparison at 0.10% drift

Bare
2.5 2.5
Infilled

2 2
Drift 0.10%

Height (m)

Height (m)
1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-30 -20 -10 0 10 20 30 40 0 50 100 150 200
Moment (kNm) Shear (kN)
Moment demand comparison at 0.20% drift Shear demand comparison at 0.20% drift

Bare
2.5 2.5
Infilled

2 2
Drift 0.20%

Height (m)

Height (m)
1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-60 -40 -20 0 20 40 0 50 100 150 200 250 300 350
Moment (kNm) Shear (kN)
Moment demand comparison at 0.30% drift Shear demand comparison at 0.30% drift

Bare
2.5 2.5
Infilled

2 2
Drift 0.30%

Height (m)

Height (m)

1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-60 -40 -20 0 20 40 -50 0 50 100 150 200 250 300 350
Moment (kNm) Shear (kN)

456
457

18
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

Moment demand comparison Shear demand comparison

Moment demand comparison at 0.50% drift Shear demand comparison at 0.50% drift

Bare
2.5 2.5
Infilled

2 2
Drift 0.50%

Height (m)

Height (m)
1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-80 -60 -40 -20 0 20 40 60 -50 0 50 100 150 200 250 300 350
Moment (kNm) Shear (kN)
Moment demand comparison at 0.75% drift Shear demand comparison at 0.75% drift

Bare
2.5 2.5
Infilled

2 2
Drift 0.75%

Height (m)

Height (m)
1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-100 -80 -60 -40 -20 0 20 40 60 -100 0 100 200 300 400
Moment (kNm) Shear (kN)
Moment demand comparison at 1.00% drift Shear demand comparison at 1.00% drift

Bare
2.5 2.5
Infilled

2 2
Drift 1.00%

Height (m)

Height (m)

1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-150 -100 -50 0 50 100 -100 0 100 200 300 400
Moment (kNm) Shear (kN)
Moment demand comparison at 1.20% drift Shear demand comparison at 1.20% drift

Bare
2.5 2.5
Infilled

2 2
Drif 1.20%

Height (m)

Height (m)

1.5 1.5

1 1

0.5 0.5
Bare
Infilled
0 0
-150 -100 -50 0 50 100 -100 0 100 200 300 400
Moment (kNm) Shear (kN)

458 Table 4: Comparison of the moment and shear demand on RC column 1 (windward) of the RC bare frame and the AAC in-
459 filled RC frame at different drifts.
460
461
462 Figure 25 reports the local pressure distributions along the height of the column at different drifts for both
463 columns. The maximum compression stress of the masonry (2.0 MPa) has never been attained and, from a
464 drift of about 0.50%, the local pressure can be assimilated to an almost rectangular “stress block”.
19
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

465
(a) Infilled Frame - Local pressure - Column 1 (b) Infilled Frame - Local pressure - Column 2

2.5 0.1% drift

2.5
0.2% drift
0.3% drift
2 0.5% drift
2
0.75% drift
1% drift
Height (m)

Height (m)
1.2% drift
1.5 1.5

0.1% drift
1 0.2% drift 1
0.3% drift
0.5% drift
0.5 0.75% drift 0.5
1% drift
1.2% drift
0 0
0 0.5 1 1.5 2 0 0.5 1 1.5 2
Local pressure (MPa) Local pressure (MPa)

466 Figure 25: Local pressure on the RC columns due to the presence of the AAC infill at different drifts (a) Column 1 (wind-
467 ward), (b) Column 2 (leeward).
468
469 Figure 26b reports the additional force in the infilled RC frame due to the presence of the infills (Fadd,infilled)
470 for different imposed drifts, computed as the product of the integral of the pressure over the contact length,
471 by the thickness tw of the masonry (300 mm) or, equivalently, by the following equation:
(8)
472 where Vmax, infilled is the maximum shear at the edges of the columns (at the top for windward and at the base
473 for leeward columns) and VC,infilled is the constant part of the shear on the column, as shown in Figure 26a
474 which reports schematically the shear and moment distribution of a windward column in the case of infilled
475 and bare configurations.
476 The shear distribution along the height of the columns can be therefore evaluated through conditions of
477 equilibrium applying such additional force (or pressure) over the contact length, together with appropriate
478 values of top and bottom bending moments MC,infilled,top and MC,infilled,bottom, as for example reported in Preti
479 et al., 2017.
480
(a) (b)

481 Figure 26: (a) Shear and moment demands: comparison between bare and infilled frame (case of windward columns); (b)
482 Additional force in the infilled frame vs drift.
483
484 For the use of such results in the case of a simplified evaluation of the local effects, for example in de-
485 sign/assessment of RC frames, when the lateral contribution of infills is usually neglected, the additional
486 shear due to the thrust of the infills, Vadd,bare , to be added to the one of the bare frame, can be calculated as:
(9)
487 where Vc,bare is the shear demand of the bare frame. In Figure 27a the additional shear due to the presence
488 of the infills has been plotted as a function of drifts, where VC,bare is taken from the results of the pushover
489 analysis on the bare frame.
20
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

490 Finally, Figure 27b reports the evolution of the contact length (“normalized” over the RC column clear
491 height) as a function of the imposed drifts. The contact length has been computed as the distance where the
492 local pressure is applied on the columns. This length increases with drifts from about 30% to more than 50%
493 of the column clear height, the latter value attained at 1.2% in-plane drift.
494 The contact length and the maximum additional shear derived by the results of the FEM analyses have been
495 then compared according to the aforementioned simplified approaches (Mainstone 1971-1974/FEMA 306
496 1998, Paulay and Priestley, 1992, Bertoldi et al., 1993, Eurocodes 6/8). The comparisons are reported in
497 Figure 27a and b. The expected contact length was found to be, for all the considered approaches, approxi-
498 mately half than the maximum one provided by the analysis. The predicted maximum additional shear has
499 ranged in a wide interval, to be in some case safe-sided (i.e., Paulay and Priestley, 1992), and, in other
500 cases, un-conservative (for example for the Bertoldi et al. approach). The approach suggested within Euro-
501 codes seems to be the one that provides results more aligned with the ones of the numerical analysis. Table
502 5 reports the input data, derived from the experimental test, and Table 6 the obtained results according to
503 the discussed approaches.
504
(a) (b)

505 Figure 27: (a) additional shear to be added to the bare frame vs drift and comparison with analytical expressions in literature;
506 (b) Contact length ratio vs drift and comparison with analytical expressions in literature.
507
L h Lw hw tw dw  Ic h  v
4 3
[mm] [mm] [mm] [mm] [mm] [mm] [rad] [dm ] [-] [kN/m ] [MPa]

4500 2875 4200 2750 300 5020 0.580 6.75 3.59 4.84 6.66 · 10-3

fvert flat ft fv0  Evert Evat G  Ec Ew

[MPa] [MPa] [MPa] [MPa] [-] [MPa] [MPa] [MPa] [-] [MPa] [MPa]

2.00 2.00 0.28 0.24 0.57 1600 1600 640 0.25 24303 1600
508 Table 5: Input data of the tested RC frame infilled with AAC masonry.
509

21
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

Approach Strut width Contact length Infill resistance

Eurocodes --- --- VC,Ed,w = Vw,hor = 302.4 kN
FEMA 306 bw = 527 mm lc = 630 mm Vw,hor = 264.4 kN
(Mainstone; Saneinejad and Hobbs) bw / dw = 0.10 lc / hw = 0.23
Bertoldi et al. bw = 1038 mm lc = 620 mm Vw,hor = 214.2 kN
bw / dw = 0.21 lc / hw = 0.23
Paulay and Priestley bw = 1255 mm lc = 750 mm Vw,hor = 475.6 kN
bw / dw = 0.25 lc / hw = 0.27
510 Table 6: Results from the discussed approaches by Eurocodes, FEMA 306 (1998), Bertoldi et al. (1993) and Paulay and
511 Priestley (1992).
512
513 7 Conclusions
514
515 In the present work, a numerical simulation of the experimental in-plane test conducted at the Univer-
516 sity of Pavia by Penna and Calvi in 2006 on a RC frame infilled with AAC masonry has been performed
517 and the local frame-panel interaction effects derived from the analysis have been discussed.
518 Non-linear static analyses using a FEM “meso-modelling” approach have been carried out, where the ma-
519 sonry has been calibrated according to tests of mechanical characterization and to in-plane cyclic tests on load-
520 bearing AAC masonry of the same typology. A good fit between the numerical and the experimental re-
521 sponse on the bare and on the infilled frame has been observed, both in terms of F-D envelope and of
522 cracking pattern.
523 The analyses performed have also allowed to investigate the local effects on the frame and, in particular, the
524 changes in the moment and shear demands on the RC elements due to the presence of the AAC infill in compari-
525 son with the ones of the bare structure. The estimation of the additional shear and of the contact length activated
526 by the infill on the RC members has been also computed.
527 Significant differences in the distribution of the internal forces have been obtained between the bare and
528 the infilled frame, already from low levels of in-plane drift. The presence of the infill clearly produces an
529 additional shear demand along the contact length of the RC columns and originates a local pressure charac-
530 terized by an almost rectangular stress block shape starting from 0.50% drift, when a complete activation of
531 the diagonal strut occurs and the first evident damage in the masonry was observed. Moreover, these results
532 show a dramatic increase of the shear action experienced in the RC columns of the infilled configuration
533 compared to bare frames (even more than 4 times at 1.0-1.2% in-plane drift), as also remarked in past ex-
534 perimental and numerical experience on different infill types and materials, and a strong dependency on the
535 in-plane drift. The ratio between the additional shear due to the infill contribution as respect to the shear of
536 the bare frame decreases as a function of the drift, although larger additional shear has been found at larger
537 imposed displacements.
538 The contact length has been found to increase with the imposed drift, from about 30% to more than 50%
539 of the column clear height, being the latter value attained at the maximum in-plane drift imposed in the test
540 (1.2%).
541 Moreover, the additional shear to be added to the bare frame depending from the imposed drift has been
542 computed for each column and compared with some well-recognized approaches, which aim to evaluate the
543 infill resistance and the contact length to be used in macro-models and in seismic design/assessment proce-
544 dures. These criteria tend to underestimate the contact length of about 50%, whereas provide infill strength
545 values that lie in a wide range, being in some cases also un-conservative. For this study, the method sug-
546 gested by Eurocodes seems to provide an additional force value that reasonably matches the results ob-
547 tained from the numerical analysis.
548 In conclusion, it appears that further work is needed in order to better comprehend the local effects ap-
549 plied to the RC elements. A wider research including different aspect ratios and infill solutions might allow
550 to improve the understanding of the issues regarding the infill-structure interaction.
551

22
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

552 References
553
554 Asteris PG (2003), “Lateral Stiffness of Brick Masonry Infilled Plane Frames”, Journal of Structural
555 Engineering, 129(8), August 1, 2003.
556
557 Asteris PG, Kakaletsis DJ, Chrysostomou CZ, Smyrou E, (2011), “Failure modes of infilled frames”,
558 Electronic Journal of Structural Engineering, 11(1), 11-20.
559
560 Barzegar F, Maddipudi S, (1994), “Generating reinforcement in FE modeling of concrete structures”,
561 Journal of Structural Engineering, 120(5), 1656-1662.
562
563 Bertoldi SH, Decanini LD, Gavarini C, (1993), “Telai tamponati soggetti ad azione sismica, un modello
564 semplificato: confronto sperimentale e numerico” [in Italian], Atti del 6 convegno nazionale ANIDIS, 6,
565 815-824, Perugia, Italy.
566
567 Bolis V, Stavridis A, Preti M, (2017), “Numerical investigation of the in-plane performance of ma-
568 sonry-infilled RC frames with sliding subpanels”, Journal of Structural Engineering, 143(2): 04016168.
569
570 Braga F, Manfredi V, Masi A, Salvatori A, Vona M, (2011), “Performance of nonstructural elements in
571 RC buildings during the L'Aquila 2009 earthquake”, Bulletin of Earthquake Engineering, 9, 307-324.
572
573 Buonopane SG, White RN, (1999), “Pseudodynamic testing of masonry infilled reinforced concrete
574 frame”, Journal of Structural Engineering, 125(6), 578-589.
575
576 CEB/FIP (2010), “Model Code 2010”.
577
578 Celano F, Cimmino M, Coppola O, Magliulo G, Salzano P, (2016), “Report dei danni registrati a segui-
579 to del terremoto del Centro Italia del 24 Agosto 2016 (Release 1) ”, [in Italian], http://www.reluis.it .
580
581 CEN (2004a), “Eurocode 2 - Design of concrete structures, , Part 1-1: General rules and rules for build-
582 ing”, Brussels, Belgium, European Committee for Standardisation, Part 1.1: General rules and rules for
583 building.
584
585 CEN (2004b), “Eurocode 6 - Design of masonry structures, , Part 1-1: Common rules for reinforced
586 and unreinforced masonry structures” Brussels, Belgium, European Committee for Standardisation.
587
588 CEN (2004c), “Eurocode 8 - Design of structures for earthquake resistance, Part 1: General rules, seis-
589 mic actions and rules for building”, Brussels, Belgium, European Committee for Standardisation.
590
591 Calvi GM, Bolognini D, (1999), “Seismic response of reinforced concrete frames infilled with weakly
592 reinforced masonry panels”, Journal of Earthquake Engineering, 5, 153-185.
593
594 Calvi GM, Bolognini D, (2001), “Seismic response of R.C. frames infilled with weakly reinforced hol-
595 low masonry panels”, University of Pavia, Italy.
596
597 Cavaleri L, Di Trapani F, (2015), “Prediction of the additional shear action on frame members due to
598 infills”, Bulletin of Earthquake Engineering, 13, 1425-1454.
599
600 Costa AA, Penna A, Magenes G, (2011), “Seismic Performance of Autoclaved Aerated Concrete (AAC)
601 Masonry: From Experimental Testing of In-Plane Capacity of Walls to Building Response”, Journal of
602 Earthquake Engineering, 15, 1-31.
603
604 Crisafulli FJ, (1997), “Seismic behaviour of reinforced concrete structures with masonry infills”, PhD
605 Thesis, University of Canterbuty - Christchurch, New Zealand.
606

23
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

607 Crisafulli FJ, Carr AJ, Park R, (2000), “Analytical Modeling of Infilled Frame Structures - A General
608 Review”, Bulletin of the New Zealand Society for Earthquake Engineering, 33.
609
610 D'Ayala D, Worth J, Riddle O, (2009), “Realistic shear capacity assessment of infill frames: compari-
611 son of two numerical procedures”, Engineering Structures, 31, 1745-1761.
612
613 Da Porto F, Guidi G, Dalla Benetta M, Verlato N, (2012), “Sistemi costruttivi e risultati sperimentali
614 Reluis Research Report” [in Italian], Experimental Report, University of Padova, Italy.
615
616 Decanini LD, Liberatore D, Liberatore L, Sorrentino L, (2012), “Preliminary Report on the 2012, May
617 20, Emilia Earthquake”, v.1, http://eqclearinghouse.org/2012-05-20-italy-it/, 2012.
618
619 EN (2001), “UNI EN 1052-1 - Methods of test for masonry, Part 1: Determination of compressive
620 strength”, Brussels, Belgium, European Committee for Standardisation.
621
622 EN (2007), “UNI EN 1052-3 - Methods of test for masonry, Part 3: Determination of initial shear
623 strength”, Brussels, Belgium, European Committee for Standardisation.
624
625 FEMA 306, (1998), “Evaluation of Earthquake Damaged Concrete and Masonry Wall Building: Basic
626 Procedures Manual”, Federal Emergency Management Agency, Washington D.C., USA.
627
628 Hak S, Morandi P, Magenes G, (2013a), “Local effects in the seismic design of rc frame structures with
629 masonry infills”, 4th ECCOMAS Thematic Conference on Computational Methods in Structural Dynamics
630 and Earthquake Engineering.
631
632 Hak S, Morandi P, Magenes G, (2013b), “Evaluation of infill strut properties based on in-plane cyclic
633 tests”, Gravedinar, 65, 411-422.
634
635 Hordijk DA, (1991), “Local Approach to Fatigue of Concrete”, PhD Thesis, Delft, University of Tech-
636 nology, Netherland.
637
638 Lourenço PB, (1995), “An Orthotropic Continuum Model for the Analysis f Masonry Structures”, Delft,
639 University of Technology, Netherland, 1995.
640
641 Magenes G, Bracchi S, Graziotti F, Mandirola M, Manzini CF, Morandi P, Palmieri M, Penna A, Rosti
642 A, Rota M, Tondelli M, “Preliminary damage survey to masonry structures after the May 2012 Emilia
643 earthquakes”, v.1, http://eqclearinghouse.org/2012-05-20-italy-it, 2012.
644
645 Mainstone RJ, (1971), “On the stiffnesses and strengths of infilled frames”, Proceedings of the Institu-
646 tion of Civil Engineers, Suppl. (iv), 57-90.
647
648 Mainstone RJ, (1974), “Supplementary note on the stiffness and strengths of infilled frames”, Current
649 Paper CP 13/74, Garston, Watford, UK.
650
651 Manzini CF, Morandi P, (2012), “Rapporto preliminare sulle prestazioni ed i danneggiamenti agli edi-
652 fici in muratura portante moderni a seguito degli eventi sismici emilani del 2012” [in Italian], v.1, Eucentre,
653 http://eqclearinghouse.org/2012-05-20-italy/.
654
655 Markou G, Papadrakakis M, (2012), “An efficient generation method of embedded reinforcement in
656 hexahedral elements for reinforced concrete simulations”, Advances in Engineering Software, 45, 175-187.
657
658 Mehrabi AB, (1994), “Behavior of masonry-infilled reinforced concrete frames subjected to lateral
659 loading”, PhD Thesis, University of Colorado, USA.
660

24
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

661 Morandi P, Milanesi RR, Magenes G, (2016), “Innovative seismic solution for clay masonry infills
662 with sliding joints: principles and details”, Proceedings of the 16th IBMAC (International Brick and Block
663 Masonry Conference), 26-30 June 2016, Padova, Italy.
664
665 Morandi P, Hak S, Magenes G, (2018), “Performance-based interpretation of in-plane cyclic tests on
666 RC frames with strong masonry infills”, Engineering Structures, 156, 503-521. DOI:
667 10.1016/.engstruct.2017.11.058.
668
669 Paulay T, Priestley MJ, (1992), “Seismic design of reinforced concrete and masonry building”, John
670 Wiley & Sons, New York, NY, USA.
671
672 Penna A, Calvi GM, (2006), “Campagna sperimentale su telai in c.a. con tamponamenti in Gasbeton
673 (AAC) con diverse soluzioni di rinforzo” [in Italian], Experimental Report, University of Pavia, Italy.
674
675 Preti M, Bettini N, Plizzari G, (2012), “Infill Walls with Sliding Joints to Limit Infill-Frame Seismic
676 Interaction: Large-Scale Experimental Test”, Journal of Earthquake Engineering, 16(1), 125-141.
677
678 Preti M., Bolis V., Stavridis A. (2017), “Seismic infill–frame interaction of masonry walls partitioned
679 with horizontal sliding joints: analysis and simplified modelling”, Journal of Earthquake Engineering, DOI:
680 10.1080/13632469.2017.1387195.
681
682 Ricci P, Manfredi V, De Luca F, Verderame GM, (2011), “6th April 2009 L'Aquila earthquake, Italy:
683 reinforced concrete building performance”, Bulletin of Earthquake Engineering, 9, 285-305.
684
685 Roots JG, (1988), “Computational modeling of concrete fracture”, PhD Thesis, Delft, University of
686 Technology, Netherland.
687
688 Saneinejad A, Hobbs B, (1995), “Inelastic Design of Infilled Frames”, Journal of Structural Engineer-
689 ing, 121, 634-650.
690
691 Selby RG, Vecchio FJ, (1993), “Three-dimensional Constitutive Relations for Reinforced Concrete”,
692 Technical Report 93-02, University of Toronto, Toronto, Canada.
693
694 Stafford Smith B, (1967), “Methods for predicting the lateral stiffness and strength of multistorey in-
695 filled frame”, Building Science, 2, 247-257.
696
697 Stavridis A, (2009), “Analytical and Experimental Study of Seismic Performance of Reinforced Con-
698 crete Frames Infilled with Masonry Walls”, PhD Thesis, University of California, USA.
699
700 Tarque N, Candido L, Camata G, Spacone E, (2015), “Masonry infilled frame structures: state of the art
701 and review of numerical modeling”, Earthquakes and Structures, 8(1).
702
703 Totoev Y, Lin K, (2012), “Frictional energy dissipation and damping capacity of framed semi-
704 interlocking masonry infill panel”, Proceedings of the 15th IBMAC (International Brick and Block Ma-
705 sonry Conference), Florianopolis, Brazil.
706
707 TNO-Diana, (2010), “User's Manual - Version 9.4.3”, Analysis procedure, TNO-Diana, Delft, Nether-
708 land.
709
710 Vecchio FJ, Collins MP, (1986), “The Modified Compression-Field Theory for Reinforced Concrete
711 Elements Subjected to Shear”, ACI Structural Journal, 83, 219-231.
712
713 Verlato N, Guidi G, Da Porto F, Modena C, (2016), “Innovative systems for masonry infill walls based
714 on the use of deformable joints: Combined in-plane/out-of-plane tests”, Proceedings of 16th IBMAC (Inter-
715 national Brick and Block Masonry Conference), 26-30 June 2016, Padova, Italy.
716
25
 Local effects on RC frames induced by AAC masonry infills through FEM simulation of in-plane tests

717 Vintzileou E, Adami CE, Palieraki V, (2016), “In-plane and out-of-plane response of a masonry infilled
718 divided into smaller wallettes”, Proceedings of 16th IBMAC (International Brick and Block Masonry Con-
719 ference), 26-30 June 2016, Padova, Italy.

26

View publication stats