Encyclopedia of Earthquake Engineering

Living Edition
| Editors: Michael Beer, Ioannis A. Kougioumtzoglou, Edoardo Patelli, Ivan Siu-Kui Au

Seismic Strengthening Strategies for Heritage Structures

  • Dina D’AyalaEmail author
  • Sara Paganoni
Living reference work entry
DOI: https://doi.org/10.1007/978-3-642-36197-5_199-1


The global seismic behavior of historic masonry buildings is highly influenced by the integrity of the connections among vertical and horizontal structural elements, to ensure the so-called box behavior. This, providing the transfer of inertial and dynamic actions from elements working in flexure out-of-plane to elements working in in-plane shear, leads to a global response best suited to the strength capacity of the constitutive materials and hence enhanced performance and lower damage level. Notwithstanding the importance of connections’ integrity, analytical checks of existing connecting elements, or design of new elements to strengthen existing connections, are generally based on qualitative rules or simplified overall checks, rather than rigorous analytical approach. The “Guidelines for Earthquake Resistant Non-Engineered Construction” were first published by the International Association for Earthquake Engineering (IAEE) in 1986 specifically with the objective of improving the seismic safety of non-engineered housing constructions.

A wide range of techniques and products for the seismic strengthening of heritage buildings are reported in the scientific literature and used in current practice to ensure the enhancement of existing connections. Heritage buildings require far more attention, especially when dealing with issues such as the compatibility between the chemical and mechanical properties of the strengthening system and the parent material. Many strengthening techniques, after an initial success and a strong commercial promotion, underperformed and showed unexpected drawbacks when put to the test of real seismic loading outside the controlled conditions of the laboratory environment. On the other hand, some existing strengthening systems can provide highly flexible applications and meet the expected requirements in terms of performance; some of these systems in fact draw on traditional reinforcement techniques, with the addition of innovative materials and a deeper insight in the laws governing the dynamics of structures.

The choice of the most suitable strengthening system for a given building, however, is determined not only by the set of constraints of the specific project but, very importantly, by the framework used for its assessment. The EC8 part 3 or ASCE41-06, dealing with assessment, repair, and strengthening, introduce the concept of knowledge level as the determining factor for the choice of alternative assessment procedures involving diverse levels of resources. The diagram in Fig. 1 shows that there is a strict correlation not just between the representation of the seismic input and the type of analysis that can be carried out with it but also and most importantly between the input and the way risk is quantified and measured and between such measures and the principle on which possible strengthening interventions operate. Hence the level of knowledge and availability of data to carry out the assessment becomes critical in determining and fully designing the most appropriate strengthening solution.
Fig. 1

Correlation between knowledge level, analytical tools, risk representation, and interventions. credits: this publication D’Ayala Paganoni 2014

Moreover, although European and national codes (e.g., Italian Ministry of Cultural Heritage and Activities 2006) suggest the use of various systems for the strengthening of connections, for example, ring beams, no detailed reference is made to specific procedures for the dimensioning and checks of such elements. The only indication in this sense can be found in Section 6.1 Retrofit Design Procedure for existing building of Eurocode 8 (EN 1998-3:2005), which states that the design process should cover:
  1. 1.

    Selection of techniques and/or materials, as well as of type and layout of intervention.

  2. 2.

    Preliminary sizing of additional structural parts.

  3. 3.

    Preliminary calculation of stiffness of strengthened elements.

  4. 4.

    Analysis of strengthened structure by linear or nonlinear analysis. The typology of analysis is chosen depending on the level of knowledge regarding the geometry detailing and materials of the structure.

  5. 5.

    Safety verifications for existing, modified, and new structural elements carried out by checking that the demand at three different limit states – damage limitation, significant damage, and near collapse – is lower than the structural capacity.


In the safety verifications, mean values of mechanical properties of existing materials derived from in situ tests and other sources of information, like available documentation or relevant sources, shall be used, taking into account the confidence factors (CFs) specified in 3.5 of Eurocode 8 (EN 1998-3:2005). Conversely, for new materials, nominal properties shall be used without modification by confidence factor. The code also states that in case the structural system, comprising both existing and new structural elements, can be made to fulfill the requirements of EN1998-1-2004, the verifications may be carried out in accordance with the provisions therein.

This last sentence indicates that for systems such as reinforced concrete (RC) ring beams or corner confinement, reference can be made to the specifications for RC members in the relevant sections of EC8 and other Eurocodes. However, this leaves open the problem of quantifying the interaction between original and new structural elements, and the assessment of the global seismic performance of the strengthened structure will still be affected by a large number of uncertainties.

The last 15 years have seen a steep increase in the use of new technologies for strengthening heritage buildings. Such fast development, as well as the high level of expertise and financial resources required for their application, often results in lack of standardization. Innovative technologies have not been extensively applied and validated in real-life situations yet, and the retrofit of a complex, precious building by means of unconventional systems is a difficult task that goes beyond the standard conservation practice. In fact, looking at the current scientific literature, it is clear that many projects of restoration and upgrade of monumental buildings are carried out by research organizations within the framework of specific projects or by large enterprises that specialize in the production and design of strengthening devices.

On the other hand, the lack of appropriate standards and procedures can be blamed for the incorrect application of some innovative strengthening techniques, notwithstanding recognized effectiveness and other benefits. One exception to this situation is the case of fiber-reinforced polymers for which guidelines and standards have been produced for selected countries (see, for instance, CNR-DT 200/04, CNR-DT 200R/13).

In the following an overview of methods to structurally strengthen the masonry elements of historic buildings is given, pointing out the advantages and pitfalls of each system and the fundamental physical parameters and design process required. Available design procedures are reviewed and referenced, when available, to the writers’ knowledge. The systems reviewed are applicable mainly to stone or brick masonry structures with timber floors and roof or with vaulted structures. The concepts underlying these methods are also usually applied to the strengthening of earth structures, although some of the details of the specific applications might differ due to the specific characteristics of the parent material.

The effective strengthening of a heritage building also relies on a correct redistribution of the inertia forces among vertical elements. For this to occur, the floor structures need to be acting as a whole and ensure diaphragm action. While there is a wealth of research and methods to assess and strengthen historic and traditional floor structures to ensure they develop such action, these are not currently included in this chapter.

Base isolation is also increasingly becoming a proposed option for the retrofitting of historic buildings; however, actual implementations are still very rare and very expensive. This technique is also not treated in this current edition.

In this current edition, only methods suggested in the Eurocodes or methods related to recent technological developments are included. Many traditional “vernacular” methods of strengthening historic building exist in earthquake-prone countries and some have been studied in detail. These are not addressed here either.

Strengthening systems are reviewed depending on the type of enhancement achieved, i.e., whether there is an increase in strength, a control of deformation and displacement, or a dissipation of energy.

Connections Between Vertical and Horizontal Macroelements

The Eurocode 8 states that to improve connection between intersecting walls, use should be made of cross-bonded bricks or stones. The connection can be made more effective in different ways (EN 1998-3:2005):
  1. I.

    Through construction of a reinforced concrete belt

  2. II.

    By addition of steel plates or meshes in the bed joints

  3. III.

    Through insertion of inclined steel bars in holes drilled in the masonry and grouting thereafter

  4. IV.

    Through post-tensioning


The addition of steel ties, along or transversely to the walls, external or within holes drilled in the walls, is an efficient means of connecting walls and improving the overall behavior of masonry buildings (EN 1998-3:2005).

In the following we review:
  • Stiffening of the wall system by corner and T-junctions confinement of wall panels achieved by inserting vertical concrete columns or steel meshes, fiber-reinforced horizontal strips, and polypropylene meshes

  • Stiffening of the wall and floor system and connection to the horizontal structures by ring beams

  • Connection of the walls and floor system by anchorage systems

  • Improvement of the performance of the wall system by including energy absorbers and dissipating devices to connect the walls

The design of effective connections between various structural elements of a masonry structures is a critical step for the achievement of a good structural response in case of seismic events. Current codes provide for carefully designed and detailed connections; for instance, the Eurocode states that: “Floor systems and the roof should be provided with in-plane stiffness and resistance and with effective connection to the vertical structural systems (EN 1998-1:2004).”

This rule is also applicable to interventions on existing structures as the Eurocode indeed states that: “The connection between the floors and walls shall be provided by steel ties or reinforced concrete ring beams (EN 1998-1:2004)” and “Specifically for masonry structures: non-ductile lintels should be replaced, inadequate connections between floor and walls should be improved, out-of-plane horizontal thrusts against walls should be eliminated (EN 1998-3:2005).”

Moreover, in relation to the connections between walls and floors, EN 1998-3:2005 states that: if existing tie-beams … are damaged, they should be repaired or rebuilt. If there are no tie-beams in the original building structure, such beams should be added.

Improvement of connections between vertical and horizontal structures can be achieved through a variety of methods, which are described in the following.

Confinement at Walls Junctions: Strength Enhancement

Column Ties

By confining plain masonry walls with vertical elements placed at all corners and wall intersections, as well as along the vertical frame of large openings, the seismic performance of a masonry building is improved as a result of the enhanced integrity of the structural system (Tomaževič,1999). This effectively changes an unreinforced masonry structure in a confined masonry structure. The disruption is significant and can affect very substantially the heritage value of the building. Moreover, such a technique is suitable only when horizontal RC tie beams in the bearing walls at floor level (see this IS section “Confinement at Wall to Floor Junctions: Ring Beams”) and stiff, monolithic floor diaphragms are in place; otherwise, the effect of the vertical confinement is scarce, or null, as shown by analysis conducted by Karantoni and Fardis (1992). Creation of corner confinement by column ties is not recommended for stonework masonry (Tomaževič 1999).

For the construction of the confinement elements, all the bricks in the intervention area are removed; the concrete of the existing tie beams is also removed to allow for the connection between existing and new reinforcement. Vertical rebars and stirrups are placed and concrete cast (Fig. 2).

Tie columns should be as thick as the wall where they are located, although in many cases of repair of existing buildings, they are of reduced dimensions due to on-site constraints. Furthermore, they may sometimes be replaced by sets of reinforcing bars placed in holes drilled in the masonry and connected to the surrounding substratum by stirrups.

The confinement prevents disintegration and improves ductility and energy dissipation of URM buildings but has limited effect on the ultimate load resistance (Chuxian et al. 1997). However, the real confinement effect mainly depends on the relative stiffness between the masonry wall and the surrounding resulting concrete frame. Before cracking, the confinement effect can in general be neglected (Chuxian et al. 1997; Karantoni and Fardis 1992). For very squat URM walls (geometrical aspect ratio of 0.33 and double fixed boundary conditions), the confinement increased the cracking load by a factor of 1.27 and the ultimate lateral capacity by a factor of 1.2 (Chuxian et al. 1997). For walls with higher aspect ratio, the confinement increased the ultimate capacity by a factor of 1.5. In addition, the confinement improved the lateral deformations and energy dissipation by more than 50 %.

The dimensioning procedure for column ties can be derived from the prescriptions specific to confined masonry structures. The process can be summarized as follows:
  • Calculation of the appropriate combination of static and seismic loads acting on the structural elements (EN 1991-1-1:2002 and EN 1998-1:2004)

  • Dimensioning and verification of the column ties for both static and seismic loading following the prescriptions provided for confined masonry structures (EN 1996-1-1:2005 and EN 1998-1:2004)

Special care should be taken in realizing the connection between the column ties and the horizontal structures in order to ensure a monolithic behavior of the resulting RC frame.

A much cheaper solution to confining masonry walls is achieved with Welded Wire Mesh (WWM). This consists in deploying a steel mesh on the two sides of a wall and around the corners, connecting it with bolts and then concrete over, by either shotcreting or other forms of plastering. The anchorage of the mesh to the foundation and horizontal structures is fundamental to achieve proper action transfer during shaking. Although less destructive than the construction of column ties, the system seriously affects the breathability of the walls and may cause deterioration due to moisture entrapment. The thickness of this reinforced plaster may vary between 25 and 100 mm depending on the regularity of the walls underneath. It cannot be applied in the presence of historic plasters or other valuable finishes. Its effectiveness highly depends on the relative stiffness between the original wall and the added layers.

Confinement by Fiber-Reinforced Plastic Strips

The application of fiber-reinforced polymer (FRP) material to the strengthening of masonry structures has to an extent mimicked application to RC structures. As a result, there are numerous examples in the literature relating to the confinement of masonry columns (e.g., Di Ludovico et al. 2010; Corradi et al. 2007). In the case of junctions between walls, horizontal strips of FRPs bonded at various levels along a masonry panel and anchored to the side walls can be used to restore corner connections, thus preventing overturning of façade walls. Optimal application is achieved when the whole perimeter of the structure is confined. Application of such techniques entails the removal of existing plaster and other superficial finish as a strong bond of the FRP strips to the masonry is necessary to ensure effectiveness of the system. The strips also need to be protected from UV radiations to prevent material deterioration. Their implementation might cause substantial loss of heritage value, and it is not recommended when valuable finishes are in place.

For the confinement to be effective, strips should be laid both horizontally and vertically. Tests by Hamoush et al. (2001) on walls strengthened with vertical and horizontal strips showed the effectiveness of the FRP systems as out-of-plane flexural strengthening elements, while Tumialan et al. (2001) have investigated the potential of near surface-mounted GFRP rods embedded into epoxy-based paste in the bed joints as shear enhancement. Cyclic testing of walls strengthened with vertical and horizontal GFRP or CFRP strips was performed by Marcari et al. (2003) and by Krevaikas and Triantafillou (2006). Those tests have highlighted a general decrease in strength and ductility of the confined members with the increase of the aspect ratio of the confined cross section. This is due to a reduction in ultimate strain at failure when increasing the aspect ratio. Glass fibers are more effective than carbon fibers, owing to lower stiffness. However, its effectiveness in connections between orthogonal walls has not been tested to the authors’ knowledge. Issues of debonding should be thoroughly investigated on a case by case basis as failure mode and extent will be highly affected by the integrity and mechanical characteristics of the parent material external strata in relation to the composite characteristics.

FRP confinement is dimensioned by verifying:
  1. 1.

    The tensile capacity of the composite material

  2. 2.

    The capacity of the anchorage area

In detail this is done by (CNR-DT 200/04):
  1. 1.
    Checking that:
    $$ {F}_d\le 2{F}_{Rd}=2\left({A}_f\cdot {f}_{fd}\right) $$


    F d is the design action applied to the FRP stripes by the front wall undergoing overturning caused by seismic equivalent action Q d ; this is calculated by:
    $$ {F}_d=\frac{1}{2{h}^{*}}\left({Q}_dh-{N}_dt-{P}_dt\right) $$

    Dimensions and weight involved in the overturning mechanism are shown in Fig. 3:

    F Rd : design tensile capacity of the FRP strengthening

    f fd : design ultimate strength of the FRP strengthening

    h*: distance between level where FRPs are bonded and bottom hinge

    A f : FRP area

  2. 2.
    Cheking the rip-off of FRP stripes from side walls:
    $$ {F}_d<2{F}_{pd}=2\left({A}_f\cdot {f}_{pd}\right) $$


    F d is calculated as above.

    F pd is maximum anchorage capacity of FRP to one of the two side walls.

    f pd is design debonding strength of FRPs.

    A f is FRP area.

Fig. 3

Static scheme for the calculation of the tensile capacity of FRP strengthening to prevent overturning of front wall (CNR-DT 200/04)

The second check is generally more demanding than the first, but it is not needed when full confinement of the structure is achieved by wrapping FRP stripes around the whole perimeter of the building and a suitable anchorage length is ensured (CNR-DT 200/04). Anchorage length should be at least 300 mm; otherwise, mechanical fixings can be used; however, the guideline does not specify suitable types and the procedure for sizing and positioning the fixings.

Additional calculations of the strengthened structure are necessary to ensure that the strengthened structure will be able to withstand other actions, such as the increased in-plane shear loading, now transferred from the walls orthogonal to the walls parallel to the action. This should be done according to the specific indications of codes dealing with masonry structures in general.

Of course in order to prevent damage in the historic building, the capacity of the strengthening system needs to be less than the shearing capacity of any masonry course in the façade wall. Given the absence of ductility in the fiber composite, this system also presents the fundamental drawback of brittle failure, modest ductility, and relying on the damaging process of the masonry for energy dissipation. An improved version in this respect is provided by SRP strips, strips reinforced with steel, which show a greater ductility (see Fig. 4).
Fig. 4

Corner confinement of masonry walls by use of SRP strips anchored with SRP connectors (Photo Paolo Casadei 2008, product FIDSTEEL)

Confinement by Polypropylene (PP) Mesh

PP meshing uses common packaging straps (PP-bands) to form a mesh, which is then used to encase masonry walls, preventing both collapse and the fall of debris during earthquakes. PP-bands are commonly used for packaging and are therefore cheap and readily available while the retrofitting technique is simple and suitable for local builders. PP meshing was developed by Meguro Lab, Tokyo University, and has had application in Kashmir and Nepal.

The retrofitting installation procedure is as follows (Mayorca and Meguro 2004):
  1. 1.

    The PP-bands are arranged in a mesh and connected at their crossing points (Fig. 5a).

  2. 2.

    Two steel rods are placed at the edges of the mesh; these bars are used to anchor the mesh at the foundation and at the top edge of the wall (Fig. 5b).

  3. 3.

    The walls are cleaned and, if possible, the paint is removed. Any loose brick is removed and replaced.

  4. 4.

    Six millimeter diameter holes are drilled through the wall at approximately 250–300 mm distance. The holes are cleaned with water spray or air.

  5. 5.

    The meshes are installed on both sides of the wall and wrapped around the corners and wall edges. An overlapping length of approximately 300 mm is needed.

  6. 6.

    Wire is passed through the holes and used to connect the meshes on both wall sides. In order to prevent the wires from cutting the PP-band mesh, a plastic element is placed between the band and the wire. Connectors are placed in proximity of the wall intersections and of the wall edges.

  7. 7.

    The top and bottom edges of the mesh are glued to the foundation and top of the wall by epoxy resin. The epoxy is used to connect the bars and the wall and it is not directly applied to the mesh. The bands, which are rolled around the bars, transfer their load through friction.

  8. 8.

    The overlapping parts of mesh are glued together so as to ensure the continuity of the strengthening (Fig. 5c).

  9. 9.

    A layer of mortar is laid on top of the mesh to protect it from UV radiation and rain and also to provide further bond.

Fig. 5

Procedure for installation of PP meshing: (a) PP-band mesh, (b) detail of top/bottom connection, and (c) retrofitted wall before application of final layer of mortar (Mayorca and Meguro 2004)

Shear tests on walls strengthened by PP mesh (Mayorca and Meguro 2004) showed that although the strengthening does not increase the peak strength, it contributes to improving the structural performance after crack occurrence. The strengthened walls exhibit larger post-peak strength, while the mesh helps to spread the diagonal cracks over a wide region. The presence of connectors and the mortar layer, ensuring bond to the masonry, are critical to the performance of the wall.

Confinement at Wall-to-Floor Junctions: Ring Beams

A ring beam is a structural element built on top of the masonry structure to the purpose of:
  • Creating a continuous connection between the roof structure and walls to better distribute the vertical loads

  • Improving the connection between orthogonal walls and the three-dimensional behavior of structure

Reinforced concrete (RC) ring beams have been widely used for the retrofit of masonry structures from the aftermath of the Friuli, Italy 1976 earthquake onwards. The construction of RC ring beams involves:
  • Consolidating the top part of the existing masonry by injections

  • Drilling vertical holes and inserting steel rods for the vertical connection between the masonry and new element

  • Building the formwork and placing the reinforcement of the beam

  • Casting the concrete

Such system improves the seismic capacity of structures by:
  • Distributing the vertical loads

  • Transferring the horizontal loads from the floors to the bearing walls

  • Connecting the bearing walls so as to create a boxlike behavior and prevent out-of-plane failure

However, during a seismic event, the high stiffness of RC beams may induce out-of-plane bending of the portions of wall between the restrained floors (Borri et al. 2009) with consequent bulging of walls and activation of out-of-plane damage mechanisms. Furthermore, such technique is often time-consuming, not cost-effective, and adds mass to the structure, which increases the earthquake-induced inertia forces and consequently requires strengthening at the basis of the walls.

All these collateral effects represent major drawbacks for heritage buildings due to the large deformability and scarce cohesion of historic masonry as opposed to the high stiffness and weight of reinforced concrete. Moreover, the construction of a ring beam has a high impact from the aesthetic point of view and may require the removal of large portions of the original material of the walls and of part of the structure of the roof.

Indeed, several surveys carried out in European historic centers after earthquakes (e.g., Spence and D’Ayala 1999; D’Ayala and Paganoni 2011) highlighted how RC beams with poor connections to the underlying material negatively affected the buildings rather than improving its structural response. Damage generally consisted of a shear failure at the interface between roof ring beam and wall, with more or less extensive damage to the masonry and eventual collapse of walls.

Nevertheless, the use of RC ring beam is allowed for the connection between walls and roof structure (Italian Ministry of Cultural Heritage and Activities 2006), providing that the dimensions of the ring beam itself do not cause an excessive increase of mass. However, conversely to steel ring beam, the use of RC beams for the strengthening of the connection between walls and intermediate floor structures is advised against (Ministero deri Beni Architettonici e Culturali, Italia, 2006), as it is far too invasive and deeply affects the performance of the structural system.

The 2007 edition of the California Historical Building Code (2007), instead, generally recommends the use of tie beams but does not specify the use of reinforced concrete.

In the scientific literature, references to procedures for the sizing of reinforced concrete ring beams can be hardly found, and it is highly likely that in common practice a large majority of additional concrete elements are simply dimensioned by the rule of thumb, using standard geometry and amount of reinforcement of ring beams in RC frames.

An improvement on RC ring beams are reinforced masonry ring beams, combining good quality brickwork (generally made of solid bricks) to steel or composite reinforcement bonded together by a binder-like grout (Ministero deri Beni Architettonici e Culturali, Italia, 2007).

In the case of steel reinforcement, the brickwork of the ring beam is laid leaving an internal cavity where the reinforcement bars and stirrups are positioned before casting the mortar-crete.

The creation of a reinforced masonry ring beam aims to:
  • Provide a continuous connection between the roof structure and walls to better distribute the vertical loads.

  • Improve the connection between orthogonal walls so as to enhance the three-dimensional behavior of structure.

The advantage of the technique in respect to RC ring beams is that stiffness and mass are closer to the original masonry. A RM ring beam has also good vertical deformability, and therefore it spreads the vertical load to the masonry beneath. Furthermore, it has a lower aesthetic impact, as the additional element can match the appearance of the original masonry.

RM beams are realized by:
  • Building two wallets of brickwork or stonework on top of the existing wall, leaving a cavity between the two

  • Placing reinforcement bars within the cavity

  • Casting the concrete within the cavity, as to create a beam

Steel plate and transversal bricks connect the outer wallets.

In some cases, consolidation of the masonry under the ring beam by injections is needed to ensure a sufficient shear capacity and similar stiffness.

Like in the case of RC ring beams, examples of the sizing of RM ring beams can be hardly found in the scientific literature.

A systematic dimensioning procedure for RM ring beams could be derived from the prescriptions specific to RM structures. The process can be detailed as follows:
  • Calculation of the appropriate combination of static and seismic loads acting on the structural elements (EN 1991-1-1:2002 and EN 1998-1:2004).

  • Dimensioning and verification of the beam for both static and seismic loading following the prescriptions provided for reinforced masonry structures (EN 1996-1-1:2005 and EN 1998-1:2004). This is done because a RM ring beam is in fact a bearing element and should therefore be calculated for vertical loading as beams are. Furthermore, the RM ring beams should be dimensioned for tensile load as the confining action of the beam consists indeed in resisting the thrust generated by walls that separate under horizontal actions; this is done by ignoring the cross-sectional area of concrete and masonry and sizing the steel reinforcement so that it can bear the whole in-plane action.

  • Dimensioning and checks of metallic connectors for both static and seismic loading (EN 1993-1:2005).

  • Dimensioning and checks of connection of the metallic fasteners to the concrete element of the beam for both static and seismic loading (EN 1992-1-1:2004 – Section 8 Detailing of Reinforcement; DD CEN/TS 1992-4-1:2009).

  • Dimensioning and checks of connection of the metallic fasteners to the masonry of walls for both static and seismic loading.

  • Dimensioning and checks of connection of the metallic fasteners fixed to the timber elements of the roof for both static and seismic loading (EN 1995-1-1:2004 – Section 8 Connections with Metal Fasteners).

Borri et al. (2009) propose a system combining FRPs and steel-reinforced grout (SRG), the latter made of high strength steel wires forming cords (Fig. 6) embedded in a cementitious grout. The ring-beam system, called LATLAM, is obtained by overlapping several layers of bricks and laminates embedded within a polymeric matrix or a cementitious grout (Fig. 7). By superposition of blocks and composite sheets, it is possible to achieve a structural element of any size and length, similarly to glue-laminated timber beams. The assembly acts as a single structural element and combines the compressive strength of masonry units with the good tensile properties of composite materials, avoiding the problem of the increase of mass typical of RC ring beams. A simplified model for the evaluation of the ultimate strength of a LATLAM section under bending stresses is available in Borri et al. (2007). The collapse is assumed to occur by crushing of masonry.
Fig. 6

Example of steel cord used in SRG (Borri et al. 2007)

Fig. 7

Assembly of LATLAM ring beam (Borri et al. 2007)

Strengthening by Way of Ties: Displacement Control

The insertion of ties to connect walls to walls, walls to floor, or walls to vaults is a traditional system, still in large use nowadays.

Post earthquake observation, experimental evidence and computational results, all show that crossties installed at the intersection of perpendicular sets of walls are able to prevent the overturning of whole façades without interfering with the original structural layout (Tomaževič 1999) (D’Ayala and Yeomans 2004; D’Ayala and Paganoni 2011).

For the ties to be effective, proper anchorage within or against the masonry is necessary. This might be achieved, either by using end plates or by using grouted ends. The tie is usually a passive element which becomes active when cracks open between orthogonal walls or timber beams tend to slide off their seat.

The dimensioning procedure of metallic ties connecting timber joist to walls can be derived from the prescriptions specific to steel structures. The process can be detailed as follows:
  1. (a)

    Calculation of the appropriate combination of static and seismic loads acting on the structural elements (EN 1991-1-1:2002 and EN 1998-1:2004).

  2. (b)

    Dimensioning and checks of the ties following the prescriptions provided for steel structures (EN 1993-1-1:2005).

  3. (c)

    Dimensioning and checks of connection between metallic elements and timber elements of the horizontal structure (EN 1995-1:2004).

  4. (d)

    Dimensioning and checks of connection of the metallic elements to masonry. Depending on the type of connection – by binder or by mechanical locking – checks should be performed in a different way as discussed in the next two subsections.


In the case of ties connecting masonry to masonry point c is not relevant. However, it might be important to proceed to a local consolidation of the masonry by use of mortar grout before implementing the tie. The major pitfall of crossties is the possibility of pullout damage at the head of the anchorage due to the different deformabilities of anchor and masonry. The use of high-ductility systems recommended by codes would overcome this issue, thus achieving the objective of protecting culturally valuable finishes and preserving life and safety.

Ties with End Plates

The dimensioning process of metallic ties with end plates (Fig. 8) consists in:
Fig. 8

Examples of metallic ties with end plates (Giuffrè 1993)

  • Sizing the cross section of the metallic rod to resist the axial load deriving from the pulling action of the portion of wall constrained by the tie in case of out-of-plane action. The minimum diameter of the rod is calculated by (Tomaževič 1999):
    $$ {D}_{\min }=\sqrt{\frac{H_{u,\mathrm{seg}}}{n}\frac{4}{\pi}\frac{1}{f_y}} $$

    H u,seg: ultimate seismic resistance of a critical part of a building,

    n: number steel ties,

    f y : yield stress of steel.

    The horizontal load that activates the overturning mechanism in a portion of wall, Hu,seg, needs to be equilibrated by the action of crossties and can be calculated by limit analysis. The position of hinges and the collapse load factor, for the portion of wall involved in the out-of-plane mechanism, are calculated so as to satisfy rotational equilibrium and the distribution of stress assumed in the masonry section at collapse as shown in Fig. 9. In deciding the static scheme for the calculation of the mechanism, the type of connections should be considered as they influence the constraints of the ideal beam that represents the wall: for instance, a wall with no positive connections to the floor structures can be modeled as a cantilever, while the positive effect of well-connected horizontal structures should be accounted for by using a simply supported beam scheme.

  • Sizing the end plate to prevent punching failure. This is done by verifying that the tensile strength of the parent material is sufficient to withstand the 45° component of the axial load acting on the tie. The rupture surface to consider for the calculation of the acting stress is a conical/pyramid surface with sides inclined at 45° in respect to the plain of the wall (Fig. 10).

Fig. 9

Static scheme of multistory wall panel for determination of hinge position and collapse load multiplier (D’Ayala, Speranza, 2003)

Fig. 10

Rupture surface for the calculation of tensile strength acting on parent material and sizing of the end plate in case of: (a) rectangular key or (b) circular end plate (http://postterremoto.altervista.org/Dati/i/i.html, last accessed 7th March 2011)

Ties Without End Plates

Anchors without end plates (Fig. 11) consist of a metallic profile that is embedded in the masonry and can transmit loads to the substratum by mechanical locking, friction, bond, or a combination of these three (Eligehausen et al. 2006), rather than use of an element such as a plate, a key, or a peg. Mechanical locking can be obtained, for instance, by undercut, namely, shaping the end part of the hole and introducing an anchor with an end shape so as to result larger than the rest of the shaft. Friction systems, instead, consist of torsion or displacement-expansion anchors, which are designed to introduce a radial pressure between the anchor and the hole. Finally bond systems rely on the use of a binding agent, which is either injected or released through a capsule system (Fig. 12).
Fig. 11

Grouted anchor injection scheme (Gigla, Wenzel 2000)

Fig. 12

Post-installed anchors: (a) systems for the transmission of load to substratum. (b) Typologies of binding materials for bond anchors (Eligehausen et al. 2006)

A fairly common commercial product consists of metallic profiles shaped as a coil; such profiles can be dry screwed in the masonry, thus mainly relying on a mechanical and frictional mechanism, or injected with resins/grout, in which case they work through the bond established between masonry and binder. Another popular system, increasingly used in conservation, is formed by steel sections provided with a fabric sleeve; profiles are installed in holes and then grout/resin is injected in the sleeve, so that the sleeve molds itself to the spaces and voids in the wall and creates a system combining mechanical locking and bond. A further advantage of this latter system is the control over the diffusion of the binder within the parent material, while the grout can expand to the cavity around the ties, particularly useful for rubble double leaf masonry.

A variety of materials, ranging from mineral binders to polymers, can be used as binder. However, materials based on polymers, such as epoxy or polyester resins, which are often used for concrete structures, should undergo careful assessment before use in historical masonry, due to possible issues regarding the mechanical and physical compatibility. Much more common for the repair of historical buildings are mineral binder systems based on cement or hydraulic lime with the addition of admixtures and fillers or aggregate. To inject bore holes, usually pure water/binder systems are used with typical w/b values of 0.8–1.0. However, the w/b ratio has to be adjusted according to the volume to be injected and to the moisture content of the substrate.

Post-installed anchors are a very common method for the strengthening of historic masonry since they allow for a wide range of applications, from restoring the through-thickness cohesion of multilayered masonry to the increase of tensile and shear capacity of panels; furthermore, they can be used at the joints between structural elements either in the form of short connectors or as longitudinal elements. These latter have the same function as metallic crossties, but, rather than relying on an end plate for the connection to the wall, pullout loads are transmitted by a shear mechanism – for friction and bond anchors – or through mechanical locking within the masonry itself.

The diameter of the anchors is chosen as a function of the expected pullout force applied, the tensile strength of the masonry, and the bond strength of the grout to masonry interface. Bonded anchors can be passive or active, i.e., prestressed or posttensioned. Passive anchors are used for up to 4 m in length; prestressed anchors can be used up to 35 m in length (Gigla and Wenzel 2000).

The installation of anchors is carried out by:
  • Drilling holes in the parent material; due to the weakness and preciousness of the parent material, dry/wet diamond rotary drilling rather than percussive drilling is recommended.

  • Removing all cores from the bore hole and checking the depth. Removing dust and debris.

  • Placing anchors and injecting the binder.

The anchors utilized are usually stainless steel, threaded rods, or special prestressing bars. The diameter of the hole drilled to accommodate the anchor is a function of the strength differential between parent material and grout and this and the steel itself. The bond of the anchor inside the masonry unit depends on the type and properties of the grouting material and on the type of masonry block units.

The failure mechanisms of injection anchors in masonry are discussed in Gigla (2004, 2010), based on an extensive test program of over 500 pullout tests. The studies consider five different failure modes including failure mechanisms related to the (a) poor strength of the injected grout, (b) exceeded tensile strength of surrounding parent material, (c) bond failure between the outer surface of the grout and the parent material, (d) a combination of b and c, and (e) failure of the steel or grout in tension (under-designed).

Recommendations for the design of injection anchors are given in Gigla (2004). The design value for the bond strength is computed as follows:
$$ {X}_{A,d}=\frac{\Phi_J}{\gamma_m}\left(\frac{f_{G,c}^2}{500}+{X}_{B,W}\right) $$
X A,d : Design value of bond strength, as a function of the compressive strength of grout and independent of bond length.

Required minimum bond length: Lb = 150 mm inside monolithic stone, Lb = 190 mm in bed or head joints of brick, and Lb = 430 mm in bed and head joints of blockwork.

f G,c : Compressive strength of grout. Minimum value: f G,c = 16.6 N/mm2. Maximum value covered: f G,c = 38.7 N/mm2. A minimum bending tensile strength of f G,ct = f G,c /8 = 2.0 N/mm2 is required.

f G,ct : Flexural strength of grout obtained from standard.

Φ J : Reduction factor for bond in bed or head joints, Φ J = 0.5.

X B,W : Term indicating the increase of bond strength inside water-absorptive stone material. X B,W = 0.15 N/mm2.

γ M : Partial factor for property, recommended: γ M = 1.35

The term X B,W considers the water absorption of the masonry substrate. Pullout studies revealed that materials with higher water absorption capacity yielded higher pullout strength than materials with low or no water absorption capacity.

Furthermore, the anchor capacity can be designed by first computing Ra,d which refers to the bond capacity between grout and rod. The equation considers the ratio of bed and head joints of the borehole surface across bond length and limits RA,d to bond inside full stone sections (AB/AG,d = 1 completely in stone; AB/AG,d < 1 in joint):
$$ {\mathrm{R}}_{\mathrm{A},\mathrm{d}} = {\mathrm{X}}_{\mathrm{A},\mathrm{d}}\cdot \frac{{\mathrm{A}}_{\mathrm{B}}}{{\mathrm{A}}_{\mathrm{G},\mathrm{d}}}\cdot {\mathrm{A}}_{\mathrm{A},\mathrm{d}} $$
with RA,d, design capacity of the injected anchor; AB, portion of the cylindrical surface of injected mortar grout made of stones or units; AG,d, total cylindrical surface of injected mortar grout; and AA,d, interface of tensile element and injected mortar plug (cylindrical surface of steel bar), calculated with nominal bar diameter and bond length. The design strength of the anchor shall be smaller than the tensile capacity of the surrounding masonry so as to cause failure by yielding of the metallic element and prevent brittle failure of the masonry. This threshold can be computed as
$$ \mathrm{F}\le \frac{1,9\cdot {\mathrm{f}}_{\mathrm{B},\mathrm{t}}\cdot {\mathrm{L}}_{\mathrm{b}}\cdot \uppi \cdot {\mathrm{d}}_{\mathrm{B}}\cdot {{\Big(\mathrm{h}}_{\mathrm{S}}}^2-{{\mathrm{d}}_{\mathrm{B}}}^2\Big)}{\upgamma_{\mathrm{M}}\cdot \tan \left(\upvarphi \right)\cdot {{\Big(\mathrm{d}}_{\mathrm{B}}}^2+{{\mathrm{h}}_{\mathrm{S}}}^2\Big)} $$

F: anchor tensile force

fB,t: tensile strength of surrounding stone

dB: borehole diameter

hS: minimal distance to edge of surrounding masonry

Lb: bond length

tan(φ): tangent of angle of friction for force transmission between anchor’s grout and borehole, about 50° in water-absorptive material and about 60° in non-water-absorptive material

γM: partial factor of safety for stone tensile strength; recommendation, 1,5

While these values are appropriate for an axial pullout test, they do not apply to failures of anchors loaded in shear or bending as a result of relative movements between the structural elements that they reconnect. Furthermore, as the family of dowel anchors also includes short fixings such as pins and nail-like fixings, failures connected to loads transmitted from other elements and in proximity to edges or openings should also be analyzed, these being highly relevant to various typologies of strengthening, such as confinement or end plates, where the use of fixings is required but rarely regulated.

Whereas the dimensioning procedure of bonded anchors in concrete has been extensively studied and commented (Eligehausen et al. 2006) and European guidelines do exist (EOTA TR029 2010; EOTA TR045 2013; DD CEN/TS 1992-4-1:2009), anchors in masonry lack specifically dedicated codes or recommendations. Modes of failure and hence the procedure for dimensioning could be partly derived from DD CEN/TS 1992-4-1:2009. Possible modes of failure due to axial load may be bond between metal and binder or between binder and parent material; cone pullout of parent material, due either to the failure of bond in the mortar joints or to tensile failure of masonry units; and failure of steel, although this is highly unlikely unless the metallic section is severely under-designed. Possible modes of failure due to loading transversal to the anchor axis may be crushing failure of either the binder or the parent material, edge failure of masonry when the load is applied in proximity of openings and corners, pryout failure of parent material, or failure of steel for combined bending and shear action. Each of these failures is considered by the guidelines and the elements of the anchorage assembly dimensioned accordingly.

Other factors considered in the guidelines are the performance of anchors when undergoing fatigue loading and the dimensioning criteria for seismic loading. For anchors embedded in concrete, ductile failure is achieved by controlling the dimensioning of steel so as to avoid the fragile failure of the substratum, and this concept should be extended to masonry structures. However, prescriptions in EOTA TR029, EOTA TR045, and DD CEN/TS 1992-4-1:2009 rely on a series of parameters, such as bond strength or minimum distance to edge, that should be either provided by the producer of the anchors or derived experimentally. In case of anchors for masonry, due to the lack of standardization as well as the variability of parent materials and masonry fabrics, not every producer is able to provide all the required information and extensive tests covering the full range of parameters needed are missing from the scientific literature.

Energy Dissipation Systems

Systems like crossties have been and are still commonly applied in rehabilitation practice throughout Europe for providing connection at the joints of perpendicular sets of masonry walls where out-of-plane damage is most likely to occur. Nonetheless pullout damage at the head of the anchorage and increased in-plane diagonal cracking may affect valuable finishes and precious frescoes.

The current codes encourage the use of these conventional stiffness-based systems (EN 1998 Eurocode 8; Italian Ministry of Cultural Heritage and Activities 2006) because innovative techniques drawing on performance-based principles (Priestley 2000), despite their effectiveness in new structures, rarely meet the requirements of reversibility and low impact required for historic structures.

As widely experienced in many applications carried out in recent years, an effective alternative to conventional strengthening techniques is the use of passive energy control and dissipation techniques.

Energy dissipation systems indeed comply with the concept expressed in EC 8 (EN 1998-3:2005) that:

The selection of the type, technique, extent and urgency of the intervention shall be based on the structural information collected during the assessment of the building. The following aspects should be taken into account:


d) Increase in the local ductility supply should be pursued where required;

e) The increase in strength after the intervention should not reduce the available global ductility.

In the field of cultural heritage, such techniques must aim to enhance the seismic performance of structures as well as limit the aesthetic impact on the building and protect elements that, despite not being relevant to the structural behavior, have cultural and historic value.

Energy control systems can either provide the structure with additional dissipation capacity and/or reduce the amount of ground input energy transferred to the structure. However, base isolation systems require heavy interventions on the bearing structure (base cut, new foundation structure, etc.) and are therefore rarely suitable for the retrofit of heritage buildings.

Energy dissipation devices can instead be placed in a number of key locations within the structure where relative displacements between members are expected to occur; thus, from the point of view of intrusiveness, they do not differ from many other standard techniques. A suitable position for the installation of dissipative devices is at the wall-to-floor interface of masonry buildings, where relative motion can occur without impairing the global structural integrity.

Energy Absorbers

Benedetti (2004) developed a series of energy-absorbing devices for existing masonry buildings drawing on experimental results that had showed that the more energy is absorbed through damage by the noncritical elements of the structure, the later global failure occurs. The main requirements set for the devices were:
  • Activation of the device for very low level of damage and cracking

  • Sensitivity to both in-plane and out-of-plane movements

  • Limitation of forces transmitted to the parent material at the fixings of the dissipative devices, so as to avoid localized damage and detachment of the devices

The RAG energy absorber (Fig. 13a) consists of four arms hinged at their end so as to form a square element. Hinges are made of lead cylinders press-fitted into the arms. When two (or four) of the corners of the device, which are connected to the masonry wall, displace relatively to each other as a consequence of damage in the parent material, the hinges deform plastically in torsion (in-plane action of the device) or bending (out-of-plane action of the device), thanks to the low yield strength of lead. Conversely, the arms are made of a stiffer material so as to remain in the elastic range. Besides a full characterization of the device, RAG energy absorbers were also tested inserted in full-scale masonry specimens on shaking table in an out-of-plane configuration (Fig. 13b), where the devices were used in series with traditional metallic crossties to connect two parallel walls. The holes that can be observed in the arms of the RAG device aim at reducing the weight of the element and avoiding a delay in the activation of the yielding mechanism. A pretensioning element was positioned in series with the anchor to further ensure the lack of any initial deformation in the crosstie.
Fig. 13

RAG energy absorber: (a) prototype (Benedetti 2004); (b) setup in series with metallic crosstie installed on a masonry specimen (Benedetti 2004)

Shock Transmission Units

STUs, or viscous dampers, consist of a hollow cylinder filled with fluid, this typically being silicone based. As the damper piston rod and piston head receive an impact, the fluid is forced to flow through orifices either around or through the piston head. The resulting differential in pressure across the piston head can produce very large forces that resist the relative motion of the damper, while the input energy is dissipated in form of heat due to friction between the piston head and fluid particles flowing at high velocities. Conversely, since this type of devices is velocity dependent, slow movements such as thermal expansions are allowed.

The viscous devices are characterized by a nominal strength: for high load rates and for input forces below their capacity, they show high stiffness; above it they feature a perfect plastic behavior, with dissipation of hysteretic energy in case of cyclic loads. The main drawback of this type of devices is the risk of leakage of the fluid, which involves the need for regular inspections and further costs involved with the maintenance of the dissipative system. Mandara and Mazzolani (2001) investigated two different methods for the dimensioning of viscous devices. In the Plastic Threshold Approach (PTA), devices are conceived and sized to limit the magnitude of force transmitted across connected members to a maximum value, determined according to the design resistance of structural elements involved. Beyond this threshold, the hysteretic energy dissipation takes place, while below the threshold the behavior of structural members is virtually rigid, which ensures the maximum degree of redundancy to the structure under serviceability load conditions. According to the Optimal Viscous Approach (OVA), the interaction between connected members is controlled by the viscous properties of the devices, which are dimensioned so as to minimize the magnitude of the force acting on them independently from its value. Contrary to PTA, the connection between elements is never fully rigid, so that energy can be dissipated under moderate intensity earthquakes too.

Shape Memory Alloys

Shape Memory Alloys (SMAs) are metallic materials that show special thermomechanical properties due to a reversible transformation between two crystalline configurations, austenite and martensite, without degradation of the crystal structure.

For some SMAs, the phase transformation is temperature-correlated, while for others, such as nitinol (NiTi SMA), the phase change can be stress induced at room temperature.

The stress–strain curve, measured during a monotonic tension test on a NiTi SMA wire, shows:
  1. (a)

    An almost linear portion corresponding to the elastic deformation of the material in its austenitic phase.

  2. (b)

    A loading plateau due to the transformation in martensite.

  3. (c)

    A new elastic phase that initiates after the complete transition to the martensitic phase (whose upper boundary is called maximum superelastic strain).

  4. (d)

    A final plateau, related to the true yielding of the alloy, which ends with failure.

  5. (e)

    Stress removal causes a reverse phase transformation, where the material goes back to the austenitic phase, stable for lower load’s values.

  6. (f)

    Strains are almost completely recovered. This property is known as superelasticity.


These characteristics make SMAs particularly suitable for use in seismic dampers, especially considering that some alloys, such as nitinol, have very good corrosion-resistance characteristics.

Shape Memory Alloy Devices (SMADs), which consist of metallic ties and groups of wires of SMAs, were developed within the framework of the ISTECH project (Indirli et al. 2001) and patented by Fip Industriale, Padova, Italy. Their application has featured in high-profile projects such as the repair and strengthening of the Upper Basilica of St. Francis in Assisi, San Feliciano cathedral in Foligno, and San Serafino Church in Montegranaro, Italy. In the Upper Basilica of St. Francis, 47 SMADs were used to connect the roof to the tympanum of the transept; the SMADs were connected on one end to a threaded bar attached to an anchorage plate in the façade and on the other end to a plate bolted to a counter plate embedded in the RC rib that was cast to stiff the existing concrete roof (dating back to the 1950s). SMADs were three-plateau self-balanced devices of three different sizes, with design capacity ranging from 17 to 52 kN and maximum allowable displacement between ±8 and ±25 mm.

Dissipative Anchor Devices

Given the difficulty of ensuring ductile failures of anchors embedded in masonry, D’Ayala and Paganoni (2014) in collaboration with CINTEC International developed two prototypes of dissipative anchor devices to address the problem of out-of-plane mechanisms.

The devices are conceived to be inserted at the joint between perpendicular walls, as part of longitudinal steel anchors grouted within the thickness of the walls. Devices are designed to work in tension, similarly to traditional crossties, which experience axial loading when one wall panel tilts outwards as a consequence of ground acceleration. This type of installation ensures a low impact on the aesthetics of the building as it does not affect the finishes.

One device relies on the plastic behavior of steel and consists of a metallic element designed to achieve lower capacity than the standard tie to which it is connected (Fig. 14a). The second device (Fig. 14b) is made of a set of small plates; bolts are used to apply a perpendicular pressure creating friction and adjusting the level of slip load beyond which relative sliding occurs. Specially designed stops ensure that the sliding motion remains in the desired range.
Fig. 14

Dissipative anchor devices: (a) hysteretic and (b) friction based (D’Ayala, Paganoni 2014)

While the metallic profiles improve the boxlike behavior of the building, contributing to an increase in stiffness that improves the structural response to low excitations, the devices allow small relative displacements between orthogonal sets of walls; additionally, for higher horizontal loads, they dissipate part of the energy input into the structure so that problems of localized damage can be avoided. Therefore, the design focuses on control of displacements and reduction of accelerations and stress concentration.

The two dissipative devices have been validated by cyclic pseudo-static and dynamic tests on the isolated devices and by pullout tests on the devices anchored in series to a conventional post-installed anchor embedded in low-shear capacity masonry (D’Ayala and Paganoni 2014).

Experimental results confirmed that both the yielding and frictional elements are able to provide the anchorage with ductility and prevent damage to occur in either the grouted sleeve or the masonry, while traditional anchors, which were tested for comparison purposes, display a high stiffness and fail at the interface between grout and parent material or in the masonry units. In the case of friction devices, it was observed that, due to tolerances of the pieces composing the assembly, a locking phenomenon occurred for the higher levels of perpendicular pressures; consequently, load-deflection curves feature additional stiffness in respect to the flat branch typical of a friction mechanism. Such behavior will need to be corrected so that the device can provide homogeneous performance over the required range of perpendicular forces.

The dimensioning procedure is based on the idea that a strengthening system can be divided into subcomponents to which one type of failure controlled by a single parameter can be associated. These parameters can be used in the formulae prescribed by design codes to calculate the capacities of components and hence determine the hierarchy of failure. The parameter values can be obtained via a number of sources: producers’ specifications, recommended or limit values provided by codes, and, if any, laboratory and/or on-site tests.

Table 1 exemplifies the procedure for the calculation of the tensile capacity of the dissipative anchoring devices in series with a grouted metallic anchor. Maximum capacity is reached if one of the components fails or when the dissipative device is activated. Further checks can also be performed in terms of displacement and ductility, as discussed in the following.
Table 1

Design of dissipative anchors: parameters that identify the tensile capacity, value range achieved during experimental validation, and range prescribed/recommended by design codes (From D’Ayala and Paganoni 2014)

The demand in terms of tensile capacity of metallic anchors is calculated as:
$$ {\mathrm{F}}_{\mathrm{D}}=M\cdot {\mathrm{a}}_{\mathrm{i}} $$

M: mass of structure that bears on the ith anchor of the strengthening system. M depends on the geometry and construction arrangement of the building, including horizontal structures, and on the layout of the set of anchors to be designed.

ai: horizontal acceleration experienced by the mass M. An estimate of the natural period of the system can be used to determine the correct design spectral ordinate and the distribution of amplification over the height of the structure. An accurate estimate of the acceleration at the height of the anchor can be obtained either by using the first natural modal shape or the procedure proposed by Miranda and Taghavi (2005).

The reference acceleration is calculated as function of the three limit states defined in EN 1998-3:2004, so that the design demand is:
  • FDNC: near collapse, calculated for a seismic action with a probability of exceedance of 2 % in 50 years

  • FDSD: significant damage, calculated for a seismic action with probability of exceedance of 10 % in 50 years

  • FDDL: damage limitation, calculated for a seismic action with probability of exceedance of 20 % in 50 years

All the subcomponents of the strength-only portion of the anchor assembly are brittle or, in the case of the grouted steel elements, are not supposed to experience large deformations; therefore, they are dimensioned in terms of strength for near collapse limit state, according to EN 1998-3:2004. The minimum capacity in the assembly must be:
$$ \mathrm{M}\mathrm{i}\mathrm{n}\;\left({\mathrm{F}}_{\mathrm{steel}},\ {\mathrm{F}}_{\mathrm{a}/\mathrm{b}\ \mathrm{bond}},\ {\mathrm{F}}_{\mathrm{b}/\mathrm{p}\ \mathrm{bond}},\ {\mathrm{F}}_{\mathrm{masonry}}\right)>{\mathrm{F}}_{\mathrm{DU}} $$
Capacities are calculated using information in the relevant row in Table 1. It is important to notice that while certain values in Table 1 are generally applicable, others are not: the standardization of steel production ensures highly repeatable performance, whereas bond strength, for instance, largely varies depending on the substratum. Therefore, it is advisable that tests be performed each time; if this cannot be done, lower limit values provided by the code and reported in Table 1 can be used.

Table 1 does not include the connections’ anchor/dissipative devices and the stops of the friction device. However, these components should also be dimensioned to resist FDNC as also required by EN 15129:2009.

The dissipative elements of the devices, either hysteretic or frictional, are designed to be activated at the threshold of damage limitation, when cracks start opening and allow for the dissipative elements to become active. When the hysteretic devices enter the plastic field, or the friction devices start sliding, the connection between wall panels is still maintained, but the pullout of the head of the anchorage is prevented and drift controlled.

Hence, from the point of view of force design, depending on which dissipative device is used, the following requirements apply:
$$ \begin{array}{l}{\mathrm{F}}_{\mathrm{yield}}={\mathrm{F}}_{\mathrm{DD}}\;\mathrm{or}\\ {}{\mathrm{F}}_{//}={\mathrm{F}}_{\mathrm{DD}}\end{array} $$
where Fyield is the yielding capacity of the hysteretic device and F// is the slip load of the frictional device.

The dissipative devices should also comply with requirements for interstorey drift of buildings undergoing seismic action. The chosen value of maximum allowable drift, dr = 0.003, for damage limitation is taken from OPCM (2005) which provides drift limits for damage limitation in walls of masonry buildings. This limit is also in line with the expected drift stated in FEMA 356 (BSSC 2000) for unreinforced masonry buildings at the limit state of immediate occupancy.

For historic buildings, the allowable drift is increased by using the reduction factor ν = 0.4, which is taken from EN 1998:2004 and accounts for the fact that devices are designed to be used in heritage structures, which fall in the importance category III of Eurocode 8 (EN 1998:2004).

It is therefore
$$ {\mathrm{d}}_{\mathrm{r}}={\Delta}_{\mathrm{e}}<0.003\cdot \mathrm{h}/v $$

Δe: the elongation of the device before yielding, in case of the hysteretic device, or before activation of the friction mechanism, in case of the frictional device,

h: interstorey height, or vertical distance of installation of anchors. A standard distance of 3 m has been assumed in the calculations, but anchors might need to be spaced more closely along the height of the wall to prevent substratum failures.

The first threshold of the hysteretic device, identified at 0.5 % elongation of the dissipative element, coincides with 46 % of the hysteretic device load capacity. The second threshold, at 5 % elongation and 72 % maximum load, should instead be used to verify the capacity of the dissipative device for the limit state of significant damage, so that dissipation of energy is ensured during low-to-medium seismic excitations. Beyond this limit, the device has a further margin of safety given by buckling, meaning that the device can reach the limit of near collapse with damage, but not complete failure, and it could still be substituted, as it has been proved by experimental testing and in response to the requirements of EN 1998:2004.

In the case of a frictional device, the drift limit is ensured by default because: before activation of the friction mechanism, deformations are negligible and, beyond activation of sliding, the device displacement is limited by the assembly stops. The device can therefore perform for all limit states, as long as the connections and stops in the assembly are designed to resist up to the state of near collapse.


In the last decades, a number of technical solutions for the improvement of structural connections have been developed in response to the increasing demand for strengthening systems specifically designed for the protection of heritage assets from earthquake-induced damage.

Notwithstanding recommendations from seismic standards and conservation codes of practice, and the availability of alternative technical solutions on the market, much emphasis when choosing strengthening solutions of connections is still placed on their capacity to enhance the overall strength of the system, rather than improve its ductility or ability to dissipate energy. When connections between vertical elements, and between vertical and horizontal elements, as well as vaults and domes have sufficient capacity, vertical and horizontal loads are better distributed, out-of-plane mechanisms of damage are prevented, and, hence, failure of floors and roofs can be avoided, substantially reducing the probability of collapse and fatalities in an earthquake.

Historically, builders mastered the art of careful detailing of joints from a process of trial and error; today’s engineers possess the necessary insight into physical and mechanical laws governing the dynamics of structures to control the process of structural upgrading at the design level. Thus, when connections, in existing buildings, lack adequate capacity, further elements can be added to the original structure to achieve the desired performance.

The review of modern techniques to enhance connections presented in the previous section highlights aspects specific to each applications as well as advantages and pitfalls typical of each system. A large majority of the discussed methods are applicable both at the local and global scale for repair and upgrade of single structural elements or of whole structures.

In respect to the past, strengthening systems for structural connections rely on a more accurate design, modern structural systems, and innovative and more durable materials, for example, stainless steel or titanium, which substituted iron in crossties, with considerable advantages in terms of issues related to material deterioration (e.g., expansion due to corrosion, failures due to reduced resisting section).

Some of the techniques described have undergone revision after observation of their performance in the aftermath of major earthquakes proved that they are unsuited to certain applications. For instance, RC ring beams need to be especially well connected to bearing walls; otherwise, they tend to create shear failure in walls and collapse; furthermore, it has been shown that they are not a viable solution in case of multileaf masonry not well bonded.

Recent strengthening solutions are evolving to include issues of environmental compatibility while remaining economically feasible, especially for vulnerable communities in developing countries.

The most important shift for future implementation is the increased awareness and technological research for economic solutions for ductility and energy dissipation-based devices, which can allow the improvement of performance of entire building stock in historic centers of earthquake-prone countries.

The performance of each system can be verified by determining the parameters and carrying out the checks summarized in Table 2. The parameters listed refer, when available, to dimensioning procedures prescribed by codes and guidelines and otherwise from the available technical literature included in this review. Some of the systems in the table are not codified; therefore, the dimensioning procedure is not standardized or sometimes not reported for intellectual property reasons; in these cases parameters are suggested by the authors on the basis of their expertise and of the available information on the behavior of the various devices.
Table 2

Performance parameters typical of each strengthening technique

Typology of strengthening

Performance parameters

Ring beams

Flexural, tensile, and shear capacity of beam (kNm/kN)

Capacity of fixings between ring beam and parent material (crushing or pryout of parent material, shearing off of fixings, shearing of parent material, etc.) (kN)

Horizontal sliding shear capacity of the masonry underneath the level of the fixings of the ring beam into the parent material (−)

Wall plates

Flexural, tensile, and shear capacity of wall plates (kNm/kN)

Capacity of fixings between wall plates and parent material (crushing or pryout of parent material, shearing off of fixings, shearing of parent material, etc.) (kN)

Horizontal sliding shear capacity of the masonry underneath the level of the fixings of the ring beam into the parent material (−)

Steel connectors for floor/roof structures

Tensile capacity (kN) as minimum depending on

Anchoring with end plates

 Combined tensile and bending strength of connector/tie (N/mm2)

Anchoring without end plates and with grouting

 Tensile capacity of fixings in timber elements (kN)


 Bond strength (for bonded systems) connector/binder (N/mm2)

 Bond strength (for bonded systems) binder/parent material (N/mm2)

 Tensile capacity of fixings (frictional or mechanical fixings) in parent material (kN)

 Tensile strength of parent material for cone pullout or punching shear (N/mm2)

 Reduction of tensile capacity due to distance to edge (%)

 Reduction of tensile capacity due to distance between anchors, nails, and connectors (%)

 Compressive strength of parent material under action of end plate (N/mm2)

 Flexural capacity of end plate (kN)

Shear capacity (kN) as minimum depending on

 Shear strength of anchor (N/mm2)

 Tensile strength of anchor for combined axial and bending load (N/mm2)

 Bearing capacity of parent material (N/mm2)

 Bearing capacity (for bonded system) of binder (N/mm2)

 Shear capacity of fixings in timber elements (kN)

 Reduction of shear capacity due to distance to edge (%)

 Reduction of shear capacity due to distance between anchors (%)

Corners confinement

Column ties

Flexural, tensile, and shear capacity of columns (kNm/kN)

Capacity of fixings between ring beam and parent material (crushing or pryout of parent material, shearing off of fixings, shearing of parent material, etc.) (kN)


Tensile capacity of FRPs (kN)

Bond strength between FRP/parent material (N/mm2) and the tensile and compressive capacity of the parent material (N/mm2)

Polymer mesh, PP mesh, steel mesh

Tensile capacity (kN) as minimum depending on

 Tensile strength of mesh (N/mm2)

 Tensile strength of fixings (N/mm2)

 Bond strength fixing/binder (N/mm2)

 Bond strength binder/parent material (N/mm2)

 Shear strength of parent material for cone pullout (N/mm2)

Shear capacity (kN) as minimum depending on

 Shear strength of parent material at the interface strengthened/unstrengthened material

 Shear strength of fixings (N/mm2)

 Bearing capacity of binder (N/mm2)

 Bearing capacity of parent material (N/mm2)

Energy dissipation systems

Yielding anchor devices

Yielding load of device (kN)

Allowable displacement after yielding (mm)

Dissipated energy (kJ)

Change in material properties during cyclic loading (e.g., hardening of metal, progressive crushing of parent material)

Buckling of device

Ultimate tensile capacity (kN) as minimum depending on

 Tensile strength of device (N/mm2)

 Other checks same as traditional anchors

Shear capacity (kN) as minimum depending on

 Shear strength of device (N/mm2)

Other checks same as traditional anchors

Frictional anchor devices

Load for activation of frictional mechanism (kN)

Ultimate tensile load of device (kN)

Ultimate allowable displacement for sliding (mm)

Dissipated energy (kJ)

Change in material properties during cyclic loading (e.g., hardening of metal, deterioration of frictional surfaces, effect of thermal expansion, presence of debris on frictional surfaces, progressive crushing of parent material)

Ultimate tensile capacity (kN) as minimum depending on

 Tensile capacity of device (kN)

 Other checks as traditional anchors

Shear capacity (kN) as minimum depending on

Checks as per traditional anchors

Energy dissipation system

RAG energy absorbers

Axial yielding load that leads to yielding of lead pins (kN)

Allowable displacement after yielding (mm)

Dissipated energy (kJ)

Change in material properties during cyclic loading (e.g., hardening of metal, progressive crushing of parent material)

Ultimate tensile capacity (kN) as minimum depending on

 Tensile strength of anchor (N/mm2)

 Tensile strength of RAG device

 Tensile strength of parent material to punching failure (N/mm2)

 Compressive strength of parent material under action of end plate (N/mm2)

Flexural capacity of end plate (kN)

RETE energy absorbers

Yielding load of the lead wire (kN)

Ultimate tensile load of device (kN)

Allowable displacement after yielding (mm)

Dissipated energy (kJ)

 Change in material properties during cyclic loading (e.g., hardening of metal, progressive crushing of parent material)

Shock Transmission Units (STUs) and viscous dampers

1st threshold – thermal deformations

 Allowable displacement (mm)

 Transmitted load for low-rate loading (kN)

2nd threshold – low seismic excitation, connections expected to perform as pinned connections

 Allowable displacements (mm)

 Capacity (kN)

 Capacity of connectors between STUs and parent material at 1st threshold (kN)

3rd threshold – higher seismic excitation, connections expected to have hysteretic response

 Allowable displacement (mm)

 Transmitted load (kN)

 Dissipated energy for hysteretic response (kJ)

Shape Memory Alloy Devices (SMADs)

1st threshold – thermal deformations

 Allowable displacement (mm)

 Transmitted load for low-rate loading (kN)

2nd threshold – low seismic excitation, connections expected to perform as pinned connections

 Allowable displacements (mm)

 Capacity (kN)

 Capacity of connectors between STUs and parent material (e.g., pullout or tensile failure of fixings) (kN)

3rd threshold – higher seismic excitation, connections expected to have hysteretic response

 Allowable displacement (mm)

 Transmitted load (kN)

 Dissipated energy for hysteretic response (kJ)

ith threshold: same parameters as per 2nd and 3rd threshold repeated for the higher plateaus of Shape Memory Alloys

The major advantage of Table 2 is that parameters have been homogenized, using a common nomenclature, to facilitate the comparison of various techniques.



  1. Benedetti D (2004) Increasing available ductility in masonry buildings via energy absorbers. Shaking table tests. Eur Earthq Eng 18(3):3–16MathSciNetGoogle Scholar
  2. Borri A, Castori G, Grazini A, Giannantoni A (2007) Performance of masonry elements strengthened with steel reinforced grout. In: Fiber-reinforced polymer reinforcement for concrete structures – FRPRCS-8, Patras, 16–18 JulyGoogle Scholar
  3. Borri A, Castori G, Grazini A (2009) Retrofitting of masonry building with reinforced masonry ring-beam. Constr Build Mater 23(5):1892–1901CrossRefGoogle Scholar
  4. Building Seismic Safety Council (BSSC) (2000) Prestandard and commentary for the seismic rehabilitation of buildings, FEMA, 356. Federal Emergency Management Agency, Washington, DCGoogle Scholar
  5. California Historical Building Code (2007) California code of regulations. Title 24, part 8. California Building Standards Commission, International Code Council, Washington, DCGoogle Scholar
  6. Casadei P, Agneloni E (2008) Advance composites applications on historical structures in Italy: Case studies and future developments 6th International Conference on Structural Analysis of Historical ConstructionGoogle Scholar
  7. Chuxian S, Guiqiu L, Wenchao W (1997) The design of brick masonry structure with concrete column. In: XI brick and block masonry conference, ShanghaiGoogle Scholar
  8. CNR-DT 200 R1/2013. Guide for the Design and Construction of Externally Bonded FRP Systems for Strengthening Existing Structures, ROMA – CNR October 10th 2013 – release of May 15th 2014Google Scholar
  9. Corradi M, Grazini A, Borri A (2007) Confinement of brick masonry columns with CFRP materials. Compos Sci Technol 67:1772–1783CrossRefGoogle Scholar
  10. D’Ayala D, Paganoni S (2011) Assessment and analysis of damage in L’Aquila historic city centre after 6th April 2009. Bull Earthq Eng 9(1):81–104CrossRefGoogle Scholar
  11. D’Ayala D, Paganoni S (2014) Protocol for testing and design of dissipative devices for out-of-plane damage Struct Build 167(1):26–40Google Scholar
  12. D’Ayala D, Speranza E (2003) Definition of Collapse Mechanisms and Seismic Vulnerability of Historic Masonry Buildings. Earthquake Spectra 19(3):479–509. doi:10.1193/1.1599896Google Scholar
  13. D’Ayala D, Yeomans D (2004) Assessing the seismic vulnerability of late Ottoman buildings in Istanbul. In: IV international seminar on structural analysis of historical constructions – SAHC04, PaduaGoogle Scholar
  14. DD CEN/TS 1992-4-1:2009. Design of fastenings for use in concreteGoogle Scholar
  15. Di Ludovico M, D’Ambra C, Prota A, Manfredi G (2010) FRP confinement of tuff and clay brick columns, experimental study and assessment of analytical models. J Compos Constr 14(5):583–596CrossRefGoogle Scholar
  16. Eligehausen R, Mall R, Silva JF (2006) Anchorage in concrete construction. Ernst Sohn, BerlinGoogle Scholar
  17. EN 1991-1:2002. Eurocode 1- Actions on structuresGoogle Scholar
  18. EN 1992-1:2004. Eurocode 2- Design of concrete structuresGoogle Scholar
  19. EN 1993-1:2005. Eurocode 3- Design of steel structuresGoogle Scholar
  20. EN 1995-1:2004 + A1:2008. Eurocode 5- Design of timber structuresGoogle Scholar
  21. EN 1996-12005. Eurocode 6- Design of masonry structuresGoogle Scholar
  22. EN 1998-1:2004. Eurocode 8- Design of structure for earthquake resistance. Part 1: General rules, seismic actions and rules for buildingsGoogle Scholar
  23. EN 1998-3:2005. Eurocode 8- Design of structure for earthquake resistance. Part 3: Assessment and retrofitting of buildingsGoogle Scholar
  24. EN 15129:2009: Anti-seismic devicesGoogle Scholar
  25. EOTA, European Organisation for Technical Assessments (2010) TR 029, Design of bonded anchors, Sept 2010. Downloaded at www.eota.eu. Accessed 8 Jan 2014
  26. EOTA, European Organisation for Technical Assessments (2013) TR 045, Design of metal anchors for use in concrete under seismic actions. Downloaded at www.eota.eu. Accessed 8 Jan 2014
  27. Gigla B (2004) Bond strength of injection anchors as supplementary reinforcement inside historic masonry. In: XIII international brick and block masonry conference, Amsterdam, pp 1–10Google Scholar
  28. Gigla B (2010). Comparison of failure of injection anchors as supplementary reinforcement inside masonry and concrete. In: VIII international masonry conference, Dresden, pp 2019–2028Google Scholar
  29. Gigla B, Wenzel F (2000) Design recommendations for injection anchors as supplementary reinforcement of historic masonry. In: XII international brick and block masonry conference, Madrid, pp 691–706Google Scholar
  30. Giuffrè A (1993) Sicurezza e conservazione dei centri storici: il caso di Ortigia. Laterza, Roma-BariGoogle Scholar
  31. Hamoush SA, McGinley MW, Mlakar P, Scott D, Murray K (2001) Out-of-plane strengthening of masonry walls with reinforced composites. J Compos Constr 5(3):139–145CrossRefGoogle Scholar
  32. Indirli M, Castellano MG, Clemente P, Martelli A (2001) Demo-application of shape memory alloy devices, the rehabilitation of S. Giorgio Church Bell-Tower. In: VI international symposium on smart structures and materials – SPIE 2001, Newport BeachGoogle Scholar
  33. Ministero deri Beni Architettonici e Culturali, Italia, Guidelines for evaluation and mitigation of seismic risk to cultural heritage : (2007) Cangemi Editore S.p.A.Google Scholar
  34. Karantoni F, Fardis M (1992) Effectiveness of seismic strengthening techniques for masonry buildings. ASCE 118(7):1884–1902CrossRefGoogle Scholar
  35. Krevaikas TD, Triantafillou T (2006) Masonry confinement with fiber-reinforced polymers. ASCE J Compos Constr 9(2):128–135CrossRefGoogle Scholar
  36. Mandara A, Mazzolani FM (2001) Energy dissipation devices in seismic up-grading of monumental buildings. In: III international seminar on structural analysis of historical constructions – SAHC01, GuimarãesGoogle Scholar
  37. Marcari G, Manfredi G, Pecce M (2003) Experimental behaviour of masonry panels strengthened with FRP sheets. In: Tan KH (ed) 6th international conference on fibre-reinforced plastics for reinforced concrete structures, World Scientific, Singapore, pp 1209–1218Google Scholar
  38. Mayorca P, Meguro K (2004) Proposal of an efficient technique for retrofitting unreinforced masonry dwellings. In: XIII world conference on earthquake engineering, VancouverGoogle Scholar
  39. Miranda E, Taghavi S (2005) Approximate floor acceleration demands in multistory buildings. I: formulation. J Struct Eng 131:203–211CrossRefGoogle Scholar
  40. OPCM (2005) No. 3431, May 3, 2005. Official Bulletin no. 107, May 10, 2005 (in Italian). Gazzetta Ufficiale – Serie Generale n. 107Google Scholar
  41. Priestley MJN (2000) Performance based seismic design. In: XII world conference of earthquake engineering, AucklandGoogle Scholar
  42. Spence R, D’Ayala D (1999) Damage assessment and analysis of the 1997 Umbria-Marche earthquakes. Struct Eng Int 9(3):229–233CrossRefGoogle Scholar
  43. Tomazevic M (1999) Earthquake-resistant design of masonry buildings. Imperial College Press, LondonGoogle Scholar
  44. Tumialan G, Huang P-C, Nanni A, Silva P (2001) Strengthening of masonry walls by FRP structural repointing. In: Burgoyne CJ (ed) 5th international conference on fibre-reinforced plastics for reinforced concrete structures, Thomas Telford, Cambridge, pp 1033–1042Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  1. 1.Department of Civil, Environmental and Geomatic EngineeringUniversity College LondonLondonUK
  2. 2.Ziegert|Roswag|Seiler Architekten IngenieureBerlinGermany