SOME ASPECTS OF THE MECHANICS OF MASONRY

E.Vintzileou
National Technical University of Athens
MECHANICS OF MASONRY
Within a structure, masonry (and masonry elements) is subjected to
a combination of
Compression
Tension
Shear (in- and out-of-plane)
Bending (in- and out-of-plane)
[The special case of slender elements subjected to vertical loads and
to out-of-plane bending moments: lateral instability-buckling]

In structures subjected to seismic actions, the axial loads are

changing in value (eventually, also in sign), whereas cyclic shear and
bending moments are developed.

We will first deal with basic aspects of the mechanical properties of

masonry and, then, with the bearing capacity of masonry elements.
Masonry is a heterogeneous anisotropic brittle material.

Made of masonry units The mechanical properties

and mortar. In many of masonry are different in
cases, other materials different directions Its deformation at failure
(iron, wood, …) are is small. A very steep
present falling branch follows

Masonry is sufficiently resistant in

compression. Its tensile strength is very
low.
Just as an indication…The compressive strength of a
historic masonry may vary between 1,0 and 2,0 MPa (in
most cases). Although this value may seem to be very
low, it is sufficient for the vertical loads to be safely
transferred from where they act to the foundation soil.

Calculate the compressive stress at the level of

foundation for a two-storey masonry building,
having bearing walls at distances of 5,0x5,0m.
MASONRY IN COMPRESSION
How do we expect masonry to fail in compression?

Take into account that (normally) the compressive

strength of stones or bricks is several times higher
than that of the mortar!

SURPRISINGLY ENOUGH, VERTICAL CRACKS CROSS

MASONRY UNITS!!!
WHY? σz Remember that the E-modulus of mortar is
significantly lower than the E-modulus of blocks and
that its Poisson’s ratio is significantly higher.
tb
tm
MASONRY UNIT

d
b

z σz MORTAR
y

x The mortar in the joint is subjected to (favourable) triaxial

compression, whereas the masonry unit is subjected to
(extremely unfavourable) heterosemous triaxial stresses.
 Thus, the vertical (compressive) stress that a masonry
unit can resist is significantly reduced.
Remember that in RC structures, when we calculate the
bearing capacity of the inclined struts (in regions close
to supports), we take the compressive strength of
concrete equal to ~0,25fc
fbc
Masonry does not
fail because of the
σz failure of the weak
material but due to
the failure of the
strong material!!!

0 σbx f Thus, the compressive strength of masonry is usually smaller than the
bt compressive strength of masonry units and larger than the compressive
strength of mortar.
Do experimental results confirm this interpretation of the failure
mode of masonry in compression?
Tests were carried out on specimens made of two bricks,
with the space between them filled with various materials.
Obviously, artificial cases that allow to check the validity of
the assumption made.

Some results: Smooth interface (without mortar): f/fb=0,89 (*)

Joint filled with mortar, f/fb=0,40, Joint filled with a soft paste,
f/fb=0,17
(*) If the two bricks were perfectly smooth, the f/fb ratio would
be equal to…?
If the material of the joint were steel, on the basis of the
assumption made before, the f/fb ratio would be equal to…?
Do experimental results confirm this interpretation of the failure
mode of masonry in compression?
Tests were carried out on specimens made of two bricks,
with the space between them filled with various materials.
Obviously, artificial cases that allow to check the validity of
the assumption made.

Some results: Smooth interface (without mortar): f/fb=0,89 (*)

Joint filled with mortar, f/fb=0,40, Joint filled with a soft paste,
f/fb=0,17
(*) If the two bricks were perfectly smooth, the f/fb ratio would
be equal to…?
If the material of the joint were steel, on the basis of the
assumption made before, the f/fb ratio would be equal to…?

For a steel joint, tests have given a f/fb ratio= 1,40!

It seems, therefore, that the compressive strength of masonry does
not depend much on the compressive strength of the mortar (fm).

Nevertheless, the mortar plays a primordial role on the behaviour of

masonry in compression, as it generates the (horizontal) tensile
stresses on blocks.

This effect is expressed also by means of the thickness of mortar

joints.

In order to better understand the effect of the mortar on the

compressive strength of masonry, as well as the effect of other
parameters, we will develop a simple model…
σz=fc and λ=f
λ=fbt/fbc

fc 1
=
f bc α (ν m − βν b )
+1
λ (αβ − αβν b + 1−
1 −ν m )
The negative effect of the thickness of
mortar joints is illustrated.
Test results confirm this effect.
However, one should bear in mind that
(a) This oversimplified model may be applicable only to modern
“single-leaf” masonry. Yet,
(b) To apply this relationship, one has to know mechanical
properties (such as moduli of elasticity and Poisson’s ratios) that
normally are not available to the Designer.
Thus, even in case of modern masonries, empirical formulae are
used instead of formulae based on physical models.
For example, in Eurocode 6, the main formula adopted for the
calculation of the compressive strength of masonry is the
following:

f k = kf b
0 , 70
f 0 , 30
m
k is a parameter depending on the construction type of masonry
(~0,40-0,60)
N.B.: This formula is valid only for mortar joints with a thickness NOT
EXCEEDING 15mm.
How to assess the compressive strength of a historic masonry?
Only empirical formulae are available, e.g.
2    Vm 
f wc = f bc + k1 f mc − k 2  : 1 + 3,50 − 0,30 
3    Vw 
Vm , Vw : the volume of the mortar and the volume of
masonry (their ratio is at least equal to 0,30)
fbc : compressive strength of blocks (not exceeding 100MPa)
fmc : compressive strength of mortar
k1 =0,60 for rubble stone masonry
=0,20 for brick or regular block stone masonry
k2 =0,0 MPa for brick or regular block stone masonry
=0,5 MPa for semi-regular block stone masonry
=2,5 MPa for rubble masonry

Source: Tassios, Chronopoulos: “Aseismic dimensioning of interventions on low-strength

masonry buildings”, Middle East and Mediterranean Regional Conference on low-strength
masonry in seismic areas, Middle East Univ., Ankara, 1986
Up to now, we talk about masonries that are “solid” within their
thickness. How often is this the case?
The most usual construction
type of masonry in historic
structures.
Two exterior leaves and filling
material (made of small pieces
of stones and mortar) placed
without compaction.
 Mechanics of three-leaf masonry:
Although the load may be
N/Aw σe σe uniformly distributed over the
σi σe/Ee σi/Ei
entire thickness of masonry, the
exterior (strong leaves) take up
most of the load, whereas the
fe,Ee,νe Fe (weak) filling material develops
fi,Ei,νi Fi low stresses. Nevertheless, the
filling material deforms more than
the exterior leaves (both vertically
te ti te te' ti' and horizontally) and they exert
The compressive strength of te'=te(1+νeσe/Ee) horizontal stresses to the exterior
the filling material may be as ti'=ti(1+νiσi/Ei) leaves (outwards!). Thus, the
low as 0,15MPa !!! (By an exterior leaves are subject to
order of magnitude smaller significant vertical load and to
than the compressive Again, failure is simultaneous horizontal stresses.
due to the
strength of the exterior failure of the In case bond between exterior
leaves) strong leaves and filling material is
“partner”
broken, the exterior leaves fail due
to out-of-plane bending.
Bond between exterior leaves and filling material is, by construction,
rather weak.
It may be broken due to ageing (decay of materials) and/or due to
mechanical actions.
Once again, we consult the results of experimental campaigns, with
the aim to check the validity of this working hypothesis.

FACE 1 SIDE 1 FACE 2 SIDE 2

Wallettes constructed for the needs of investigating the perimeter masonry of the
Katholikon of Daphni Monastery
 2.4 2.4
2.2 Wallette 3
Wallette 3 2.2
2 2
compressive stress, σ (MPa)
1.8 Wallette 1
1.8 Wallette 2

compressive stress, σ (MPa)

Wallette 1
1.6 1.6
1.4 1.4
Wallette 2
1.2 1.2
1 1
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0 0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 1 2 3 4 5 6 7 8 9
horizontal deformation at mid-height (mm) transverse deformations (mm)

~1,0mm vs. ~5,0mm

Vintzileou and Miltiadou “Mechanical properties of three-leaf stone masonry

grouted with ternary or hydraulic lime based grouts”, Engineering Structures, Volume
30, Issue 8, August 2008, Pages 2265-2276
Valluzzi (2004)
Prerequisite! We
Compressive strength of external leaves: Flat jacks know how
masonry is
Compressive strength of filling material: Core taking constructed
within its
and testing thickness

Compressive strength of masonry:

 Ve   Vi  Volume of filling material

f wc,0 =  θef c,e +  θif c,i
 Vw   Vw 
Volume of external leaves

Total volume of masonry Compressive strength of external leaves/infill

θe<1.0 and θi>1.0 empirical coefficients-interaction between external

leaves and infill. Value???

Again, due to the lack of an accurate physical model, we use empirical

formulae!

Valluzzi M.-R. 2004. Consolidamento di murature in pietra. Iniezioni di calce idraulica naturale.
Collana Scientifica REFICERE, Gruppo Editoriale Faenza Editrice S.p.a., 128pp.
 f wc = (2λeδf wc ,e + λi f wc ,i ) : (1 + 2δ )

Compressive strength of exterior leaves, of filling

material
Ratio between the
−4 / 3
λe ~ 1 − 0,06ζ t h e e w (te , hw in mm) Correction thickness of the exterior
leaf and that of the filling
ζ e = Ewe / f wc ,e factor~1,00
3 material
2.5

γRd~1.50
2
(fwc,0)exp(N/mm2)

Observe the
1.5 significant
1
scatter of data!

0.5

0
0 0.5 1 1.5 2 2.5 3
(fwc,0)pred(N/mm2)
Tassios,T.P. 2004. Rehabilitation of three-leaf masonry. In Evoluzione nella sperimentazione per le costruzioni, Seminario
Internazionale, 26 Sept- 3Oct., Centro Internationale di Aggiornamento Sperimentale – Scientifico (CIAS)
 Modulus of elasticity, E
(at stress level=?)
2.4
2.2 Vintzileou et al., 1995
Toumbakari et al., 2002
Wallette 3 Valluzzi, 2000
2 Vintzileou et al., 2006
2.8
1.8
compressive sttress, σ (MPa)

Wallette 1 2.6
1.6
2.4
1.4
Wallette 2

fwc,0(N/mm2)
1.2 2.2

1 2

0.8 1.8

0.6 1.6

0.4 1.4
Ew,0
0.2 1.2
0 250 500 750 1000 1250 1500 1750 2000 2250
0 Ew,0/fwc,0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
horizontal deformation at mid-height (mm) SCATTER!!!!!!!
Vintzileou E.: “Three-leaf masonry in compression before and after grouting: A review of literature”, to be published,
International Journal of Architectural Heritage
It seems that E-values are very scattered
also in case of modern masonry. Therefore,
in case deformations are of significance,
parameter analyses should be performed
(!!!)

Drysdale
Strain at failure
Eurocode 6 (for modern masonry) adopts a stress-strain diagram
similar to that used for concrete.
Vintzileou et al., 1995
Toumbakari, 2002
Valluzzi, 2000
Vintzileou et al., 2006
2.8
2.6 Historic masonries can be significantly
2.4 “softer”. This depends both on constituent
2.2 materials and on the construction type of
2 masonry.
1.8 In general, one should say that stronger
fwc,0(N/mm2)

1.6
masonries fail at smaller strain values than
1.4
weak masonries.
1.2
1
0.8
0.6
0.4
0.2
0
0 0.002 0.004 0.006 0.008
εwu,0
BEHAVIOUR OF MASONRY IN
SHEAR
In-plane shear is one of the two critical action-effects that affect
masonry structures subjected to earthquakes.

B B’ C C’ The length of the diagonal (AC’) is greater than

the initial length of the diagonal (AC).
On the contrary, the length of the diagonal (B’D)
is smaller than that of the diagonal (BD).
Therefore, there is tension along the diagonal
(AC) and compression along the diagonal (AC).
A D

Remember!

The oblique strut is simultaneously subjected to

compression and to transverse tension. Its
compressive strength is, therefore, significantly
reduced.
However, normally, failure is due to the opening of
cracks (perpendicular to the tensile stresses
The morphology of the shear cracks
(a) Are they diagonal or bi-diagonal? It depends on the
characteristics of the motion, as well as on the modified
characteristics of the structure.

The first diagonal crack has to close; the maximum displacement to be imposed in the
second loading direction has to be large enough for the first crack to close and for the
second to open.
Furthermore, if during the first large cycle the structure has other damages too (that
modify its dynamic characteristics), perhaps, during the displacement reversal the
second family of cracks does not appear.
(b) Do the cracks pass through mortar joints or through masonry
units as well?
It depends on the relative strength of the two constituent
materials.
Does the opening of diagonal or bi-diagonal cracks mean that
masonry has failed?

fv = c+μσ

σ Cohesion (shear
resistance under zero
normal stress)~0,10-
τ=μ σ 0,20MPa
12
G1
G2
10 G4

µmax and µres

6

Tests on interfaces between a 2

substrate (stone) and a grout

0
(mortar)-Ch.-E.Adami, Doctor Thesis 0 0.1 0.2 0.3 0.4 0.5 0.6
σ [MPa]
 Conservative μ-value~0,40-0,50

The friction coefficient, μ, may

be considered as almost
constant within a range of
cmb normal stress values.
What happens to masonry subjected to cyclic shear?

Disintegration of masonry

Significant force-response degradation

due to cycling
Narrow hysteresis loops
Very limited ductility
MASONRY IN TENSION
The tensile strength of masonry is very small.
We do not rely on the tensile strength of masonry (it is neglected)
Not only because it is small, but mainly because the appearance of
cracks in masonry (when its tensile strength is exceeded) does not
mean in itself that masonry has failed!!!
This is one reason for which (in contrast to what happens quite
often) the verification of masonry elements should be done in terms
of action-effects and not in terms of principal stresses!

…Although principal tensile stresses are very “instructive”, as they allow us to locate
regions of cracks…
ft

ft
ft =ζfmt
 Required!!!
Required !!!
σ0
tb

ft//
tm
ft//

lb σ0

f t 2(tb + t m ) = lb (cmb + µσ 0 ) + 2tbζf mt

//
 tb

ft//
tm
ft//

lb

f t 2(tb + t m ) = tb f bt + tbζf mt
//
 θ

ft

ftθ
θ ft//