.

The 17th World Conference on Earthquake Engineering

2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020
Paper N° XXXX (Abstract ID)
Registration Code: S-XXXXXXXX

EXPERIMENTAL AND NUMERICAL ANALYSIS OF BEHAVIOUR IN

COMPRESSION AND SHEAR OF HANDMADE CLAY BRICK MASONRY

L. Lavado(1), J. Gallardo(2) C. Honma(3)

(1) Assistant Professor, National University of Engineering, [email protected]
(2) Associate Professor, National University of Engineering, [email protected]
(3) Researcher, Japan-Peru Center for Earthquake Engineering Research and Disaster Mitigation, [email protected]

Abstract
Masonry is a composite material consisting of an assemblage of brick and mortar each one with its own mechanical
properties. It is one of the most predominant structural material in Peru, as well as in many parts of the world because it
is a low-cost material, is a heat resistant and sound insulation, it has good mechanical properties and the material is
easily available.
The Japan–Peru Center for Earthquake Engineering Research and Disaster Mitigation (CISMID) of the National
University of Engineering from Peru was carried out several surveys about the condition of structures in various
districts of Lima city in order to build a database of structural systems and structural materials used in housing
buildings.
The results of this study provided information about the situation of the housing in recent years on the periphery of the
city by remarking that these buildings are mostly informal. Around 70% of the housing buildings in these areas of the
city are considered informal.
Regarding the brick used in informal masonry housing buildings, it has been observed that the use of handmade and
tubular bricks is predominant despite not complying with all the regulations given in the Peruvian technical standards.
This study develops two aspects in relation to the study of masonry buildings: the study of materials used in the
outskirts of the city of Lima and the numerical analysis of the behavior of diagonal compression and axial compression
that occur in masonry structures.
Compressive stress occurs in confined masonry structures due to dead and living forces, while flexion and shear effects
happen owing to seismic loadings. It was observed in wall subjected to cyclic lateral loading tests carried out at
CISMID that the modes of failures can be mainly generated by shear forces. Based on the previous information, it is
known that diagonal cracking and slip of mortar joints are the dominant failure mechanisms of confined masonry walls.
In the present study, the results of an experimental and numerical study of the mechanical behavior of non-reinforced
masonry are shown. It was carried out tests of axial compression, diagonal compression and shear direct tests on
specimens made by handmade bricks used in the city of Lima.
Numerical modelling of masonry structures requires the properties of each constituent of the masonry, in this case, the
constitutive relations of bricks and mortar. The parameters obtained from the above-mentioned experimental study are
used for numerical modelling by means of a finite element program.
Keywords: masonry, clay brick, compressive strength, prism test, numerical analysis

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

1. Introduction
Masonry units or bricks are used as a structural element for construction from about 11,000 years ago, bricks
are the oldest manufactured construction materials. Its simplicity, strength and durability caused its extensive
use, and gave it a dominant place in history along with the stone. Pre-ceramic Neolithic farmers were the
first to use them in the Levant to 9500 BC, who built homes with circular shape semi-excavated footing with
stone and wall made of adobe.
In Lima City, several archaeological sites have been built in adobe, such as “Huallamarca”, which was
probably a religious temple of cultural traditional Pinazo (100 BC to 200 AD) and “Huaca Pucllana”
belonging to the Lima culture (200-700 AD).
Currently, masonry buildings are built in many parts of the world. It is a low cost material, has good acoustic
and thermal insulation, easily available and labor is easy to obtain at each location.
In this paper, testing in handmade clay bricks and masonry specimens are shown in order to determine basic
mechanical properties and define the behavior of this kind of masonry used in the city of Lima. The
properties of masonry depends on the properties of bricks, type of mortar, physical properties of the sand
used for the mortar, state of bricks, manpower among other.
Analysis and design of buildings require the mechanic properties of masonry, for example, the modulus of
elasticity, stress-strain curves of masonry are required for detailed non-linear analysis of masonry structures.
Mathematical models of structures using masonry walls require material properties and constitutive relations
of the masonry and the elements that comprise it, i.e. bricks and mortar.

2. Background
In Peru, several studies have been developing to determine the characteristics of masonry at universities and
research centers. Since 1988, CISMID have performed full-scale trials tests in masonry specimens. In 1995
studies in handmade and industrial clay bricks specimens were conducted.
Since 2010, several districts in Lima have been assessed in seismic risk. These studies were conducted by
CISMID requested for some national government entities. For these studies, between 25% and 30% of total
blocks of each district were taken as a representative sample. Table 1 shows the percentage of houses built in
masonry and type of bricks used in walls and Fig. 1.

Table 1. Percentage of houses built in masonry

Blocks Hand made Bricks
District MasonryHouses (%)
Sample (%)
La Molina 438 95% 9%
Chorrillos 554 90% 26%
Villa El Salvador 957 89% 45%
Comas 825 91% 60%
Puente Piedra 732 80% 47%
San Juan de Lurigancho 1271 95% 44%
Cercado de Lima 403 75% 40%
Ventanilla 1080 50% 13%
Breña 126 68% 60%
Carabayllo 751 92% 56%
El Agustino 297 93% 25%
Independencia 415 93% 69%
Lurín 350 86% 35%

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

Fig. 1 – Masonry dwelling in Lima

3. Experimental program
In the present experimental study, handmade bricks from two factories were used, in order to differentiate
their origin bricks and specimens were named as ART1 and ART2. Several experimental tests on masonry
components have been carried out, starting by choose mix of mortar. Testing of mechanical properties of
handmade clay bricks as we can see in Fig. 2, mortar and masonry is performed according to E070 Peruvian
Standard. The masonry prisms and wallettes were built by experienced bricklayers, who selected only
uncracked bricks and controlled the 10mm thick of the bed and head joints by using wood templates and
plumbs. Twenty prisms for testing axial compression resistance and twenty wallettes for diagonal tension
tests were conducted.
Properties of bricks
According to standard the masonry units will not have foreign matter on their surfaces or inside, such as
pebbles, small shells or nodules of calcareous nature. The unit will be thoroughly cooked clay, will have a
uniform color and not submit vitrification. When it is hit with a hammer or a similar object will produce a
pinging sound. Masonry unit shall have no cracks, fractures, fissures or similar defects that degrade its
durability or resistance. About handmade masonry units, it has noticed that do not meet some of the
indications mentioned in the standard. For this reason, the units were selected, eliminating those who had
deterioration thus to avoid distortion of results by this parameter.
Physical properties bricks
Suction
The suction is related to the adhesion of the union of bricks and mortar, as with excessive suction is not
achieved a proper junction, because the brick absorbs fast so water from the mortar, hardening and
deforming this, obtaining unions of low resistance.
The standard specifies that according to the weather conditions where the work it is located, the units should
be watered for half an hour, between 10 and 15 hours before seat them. In this study, the results show that
bricks have high values of suction, so in work is necessary to soak the bricks before to bed them and prevent
dehydration of mortar.
Absorption
The absorption of masonry units is directly related to its resistance. To weathering, while the unit is more
porous will be more absorbent; therefore, be more vulnerable to moisture from weather. The units tested
meet the absorption requirements indicated in the standard, according to which the absorption of clay and
silico-calcareous units not be greater than 22%.
The test results for physical properties of units are shown in Table 2.

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

Table 2 - Test of physical properties of brick.

SUCTION ABSORPTION
ID
(g/200cm2) (%)
ART1 80.70 15.48
ART2 99.39 15.27

Fig. 2 - Handmade bricks

3.1. Mechanicals properties of bricks

3.1.1 Compressive strength in masonry units
Monotonic compression tests were carried out in ten units for each type of brick in order to determine the
compression strength (f'b). For calculating the resistance, the gross area of brick has been considered. Values
of 6.40 MPa for ART1 and 11.58 MPa for ART2 are obtained. The test results for compressive strength are
shown in Table 3.

Table 3 - Compression test results in

bricks.

ID f'b (Mpa)
ART1 6.40
ART2 11.58

Fig. 3 - Compression test in bricks.

3.2. Properties of mortar

Three types of mortar were adopted in this study. The grain size of the fine aggregate is not within the sieve
zone indicated by standard. However, no corrections are made in order to represent real conditions of
construction. Three different grades of mortar were used in the study. Since compressive strength of mortar
depends upon the water-cement ratio, mortar with cement-sand volumetric ratio 1:3, 1:4 and 1:5 were tested.
Mortar cubes of 50 mm size (obtained for each type of mortar during the construction of the specimens) were
cured in water for 28 days, and then they were tested to obtain their compressive strength. The test results for
mortar are shown in Table 4.

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

Table 4- Compression test result of mortar.

Modulus of
Type of Compressive
elasticity
mortar Strenght ( MPa)
(MPa)
Mortar 1:3 22.80 18603.7
Mortar 1:4 14.50 15320.5
Mortar 1:5 9.30 13736.0

Fig. 4 - Compression test in mortar

3.3. Axial Compression test of prism

Axial compression test result of prism can be used to determine the compressive capacity of a full-scale
masonry wall during lateral in plane cyclic.
The samples to determinate the compressive strength of masonry are prisms made with 5 bricks as we can
see Fig. 6. For each of the 2 types of bricks were prepared prisms using mortar with ratio cement: sand (1:3)
cement: sand (1:4) and prisms using mortar with ratio cement: sand (1:5).
Horizontal mortar joints had a nominal thickness of 1cm. The slenderness of the specimens varies between 4
and 4.7, so we had to make the appropriate correction.
Before testing, capping with mortar, made by cement and gypsum, is applied at the bottom and top of each
prism.
The tests started at an age of 28 days and compression testing was done following NTP 399.605. The test
equipment is composed by a compression machine of 300 ton and a data acquisition system as we can see
Fig. 7. For the measurement of deformations LVDT sensors was used.

Table 5 - Summary of compressive strength of

prism, f’m

Average Compressive Strength

Identification
f'm (MPa)
Mix 1:4 XV 6.89
Mix 1:5 XV 5.3
Mix 1:5 XIII 7.1
Mix 1:3 XIII 9.74

Fig. 5 - Average Stress strain curves

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

For data processing, inconsistent values were discarded previously. Compressive strength value of masonry,
f'm, for each set of prisms was calculated by average values minus the standard deviation, according to E070
Standard. The average compressive strength for each prism was calculated as the peak load divided by the
gross area of the prism. The results of the masonry prisms strength, are summarized in Table 5 and average
curves for prism with each type of mortar mix are shown in Fig. 5
Based on regression analysis of data obtained from an experimental program, relationship between elasticity
modulus (E) and compressive strength (f’m) of the prisms was developed as you can see in Fig. 8. and show
this relationship.
For handmade bricks prism, the relationship between modulus of elasticity and compressive strength is as
follow:
E=365.97f’m

Fig. 6 - Construction of prism. Fig. 7 - Compressive strength test.

Fig. 8 - Variation of modulus of elasticity of masonry with corresponding compressive strengths for
handmade bricks prisms.

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

(a) (b)
Fig. 9 - Type of failures in prisms. (a) Vertical cracking (b) vertical cracking and crushing.

3.4. Diagonal Tension Test

This test was developed to measure diagonal tension more precisely. The specimen is loaded in compression
along one of its diagonals causing a diagonal tension failure when the specimen splitting apart parallel to
load direction. Wallettes consist of six rows and each row consists of 2.5 bricks

(a) (b)
Fig. 10 - Test setup for diagonal tension test: (a) Front view (b) Back view

Fig. 11 - Diagonal tension test Fig. 12 -Average Stress strain curves

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

It was fixed one LVDT (gauge length of 200 mm) on each side of the specimen along the diagonal in order
to measure the shortening and elongation of the samples prepared. The experiment set-up of the system is
shown in Fig. 10 and Fig. 11.
Table 6 - Summary of Shear Strength, V’m
Shear Strength
Identification
f'm (MPa)
Mix 1:4 XV 0.78
Mix 1:5 XV 0.48
Mix 1:5 XIII 0.64
Mix 1:3 XIII 0.80
For processing data, inconsistent values were previously discarded. Shear stress V'm for each set was
calculated by average values minus the standard deviation. The average shear stress for each specimen is
calculated as the peak load divided by the gross area of the specimen. The results of the masonry shear
strength are summarized in Table 6 and Fig. 12.

4. Numerical analysis
Masonry is heterogeneous and anisotropic material because of its constituents as mortar and brick. There are
different possibilities to solve the problem of modelling masonry. These alternatives depend on how detailed
is the modelling and the purpose of the analysis. Usually, the alternatives are classified as: detailed micro
model, simplified micro model and macro model.
The first alternative to describe a masonry is the micro model (Fig. 13b). This type of modelling considers
the bricks and mortar as continuum elements with defined failure criteria. The interface between bricks and
mortar is modeled by special elements that represent the discontinuities. Because of the level of detail of this
model, it is possible to reach more accurate result.
Simplified micro model is the second alternative, the (Fig. 13c), the bricks are kept as in the “detailed micro
model”, but the mortar joints and interface elements are re-defined as individual elements to represent a
contact area. This means that the general geometry is maintained, but the individual elements that represent
joints and interface are not able to describe the Poisson’s effect of mortar over bricks. Because of this last
example, some types of failure mechanisms cannot be reproduced in this type of model.

(a) Masonry (b) Micromodel

(c) Simplified micromodel (d) Macromodel

Fig. 13 – Types of modelling masonry

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

The last alternative is the macro model (Fig. 13d). In this case, the masonry panel is considered as a
homogeneous element. Because of its characteristics, this type of model should be able to reproduce the
general structural behaviour of a masonry panel but it is not able to reproduce all the types of failure
mechanisms.
The damage model used in the present project is Concrete Damaged Plasticity (CDP), which is available in
ABAQUS library. CDP model has been developed to predict the behavior of concrete and other quasi-brittle
materials such as rock and mortar under cyclic loading. Cracks in tension or crushing in compression are the
main failure modes of this model. CDP model assumes that the uniaxial compressive and tensile response of
concrete is characterized by damaged plasticity (see Fig. 14).
In this study, handmade clay brick component model, meshed with element type of 3D stress linear, hex
shaped with reduced integration elements (C3D8R). For interaction contact between masonry components, a
friction contact element was used to be the general contact between bricks. Compression stress strain curve
of brick to be used in the CDP model is the same as obtained from the experimental study for brick type
ART1 (see Fig. 15), some parameter were obtained from the references. The explicit dynamic procedure has
been chosen.

(a) (b)

(a) Response of concrete to uniaxial loading in compression Fig. 15 – Experimental stress

(b) Response of concrete to uniaxial loading in tension strain curve for brick
Fig. 14 - Response of concrete to uniaxial loading

4.2. Numerical Analysis of Prism

For prism, the finite element simulation was under displacement control where axial displacement was
applied on the top of the prism. All the degrees of freedom at the bottom of the prism were constrained. S22
represent compression uniaxial stress along axial axe of the prism, S11 represent stress in the perpendicular
axe to the axial one. Fig. 16 shows state of stress generated due to application of load. Regarding to
compression strength, the value obtained from numerical analysis is 5.56 MPa and the value obtained from
test is 5.72 MPa. Regarding to strain at maximum compression strength, the value obtained from numerical
analysis is 0.0022 and the value obtained from test is 0.0045

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

State of stress S22 results Displacement U22 direction results

Fig. 16 – Result of numerical analysis of prism compression test

4.3. Numerical Analysis of wallettes

For wallettes, the finite element simulation was under displacement control where the displacement was
applied on a steel head placed on top. Fig. 17 shows the in-plane state of stress contour generated due to
application of load.
Regarding to shear strength, the value obtained from numerical analysis is 0.31 MPa and the value obtained
from test is 0.36 MPa.

State of stress S22 results Displacement U22 direction results

Fig. 17 - Result of numerical analysis of wallets diagonal tension

5. Conclusions
The handmade brick presents a high level of suction, which should be considered during the seating of units
in the constructions site.
For prisms, the relationship between modulus of elasticity and compressive strength obtained experimentally
is follows as:
E=365.97f’m
It can be observed that the computed values for compression strength of prism obtained with numerical
analysis are in good agreement with the corresponding experimental observations. Regarding to strain at
maximum compression strength, the value obtained from numerical analysis and the value obtained from test
are not very close. The difference in the computed and experimental values may be attributed to values

10

© The 17th World Conference on Earthquake Engineering - 2i-0141 -

. The 17th World Conference on Earthquake Engineering
2i-0141 17th World Conference on Earthquake Engineering, 17WCEE
Sendai, Japan - September 13th to 18th 2020

assumed for some parameter of contact element. The values of shear strength of wallet obtained with
numerical analysis are in good agreement with the corresponding experimental observations.

6. References
[1] T. Izquierdo, “Correlación para determinar las propiedades físicas y mecánicas de unidades o especímenes de
albañilería”, 1995.
[2] C. Zavala, M. Estrada, Z. Aguilar, F. Lázares, “Estudio De Microzonificación Sísmica y Vulnerabilidad en la
Ciudad De Lima”, Ministry of Housing, Construction and Sanitation, 2010.
[3] C. Zavala, M. Estrada, Z. Aguilar, F. Lázares, “Estudio de Microzonificación Sísmica, Mapas de Peligros Múltiples
y Análisis de Riesgo de los Distritos del Cercado De Lima, Ventanilla y de las Ciudades de Chincha y Contumazá”,
Ministry of Housing, Construction and Sanitation, 2012.
[4] C. Zavala, M. Estrada, Z. Aguilar, F. Lázares, “Evaluación del Riesgo Sismico del Distrito de Breña”,
CENEPRED, 2012.
[5] C. Zavala, M. Estrada, Z. Aguilar, F. Lázares, “Estudios de Microzonificación Geotécnica Sísmica y Evaluación
del Riesgo en Zonas Ubicadas en los Distritos de Carabayllo y El Agustino (Provincia Y Departamento De Lima);
Distrito del Cusco (Provincia y Departamento del Cusco); y Distrito de Alto Selva Alegre (Provincia y
Departamento de Arequipa)”, Ministry of Housing, Construction and Sanitation, 2013.
[6] C. Zavala, M. Estrada, Z. Aguilar, F. Lázares, “Reducción de Vulnerabilidad y Atención de Emergencia por
Desastres”, Ministry of Economy and Finance, 2013.
[7] Bolhassani M, Hamid A, Lau A, Moon F (2015): Simplified micro modeling of partially grouted masonry
assemblages. Construction and Building Materials 83(2015)159-173.
[8] Sümer Y, Aktas M (2015): Defining parameters for concrete damage plasticity model. Challenge Journal of
Structural Mechanics 1(3) (2015) 149-155.
[9] Mohamad A, Chen Z (2016): Experimental and numerical analysis of the compressive and shear behavior for a new
type of self-insulating concrete masonry system. Applied Sciences. MDPI.
[10] Lavado L, Taira J, Gallardo J (2014): Current state of masonry properties material on emerging zones in Lima city.
Journal of Disaster Research Vol. 0 N° 0, 2014.
[11] Lourenço P, Pina-Henriques J (2005): Validation of analytical and continuum numerical methods for estimating the
compressive strength of masonry. Elsevier.
[12] Vindhyashree, Rahamath A, Prema W, Prathap M (2015): Numerical simulation of masonry prism test using
ANSYS and ABAQUS. International Journal of Engineering Research & Technology (IJERT), ISSN: 2278-0181
[13] Santo C, Alvarenga R, Ribeiro J, Castro L, Silva R, Santos A, Nalon G (2017): Numerical and experimental
evaluation of masonry prisms by finite element method. Revista Ibracon de Estructuras e Materiales, Vol 10,
Number 2, 477-508 ISSN 1983-4195.
[14] Xiao Y, Chen Z, Zhou J, Leng Y, Xia R (2017): Concrete plastic-damage factor for finite element analysis:
concept, simulation, and experiment. Advances in mechanical engineering, vol. 9(9) 1-10.

11

© The 17th World Conference on Earthquake Engineering - 2i-0141 -