In most design codes, infill walls are considered as non-structural elements and thus are typically neglected in the design process. The observations made after major earthquakes (Duzce 1999, L’Aquila 2009, Christchurch 2011) have shown that even though infill walls are considered to be non-structural elements, they interact with the structural system during seismic actions. In the case of heavy infill walls (i.e. clay brick infill walls), the whole behaviour of the structure may be affected by this interaction (i.e. local or global structural failures such as soft storey mechanism). In the case of light infill walls (i.e. non-structural drywalls), this may cause significant economical losses. To consider the interaction of the structural system with the ‘non-structural ’infill walls at design stage may not be a practical approach due to the complexity of the infill wall behaviour. Therefore, the purpose of the reported research is to develop innovative technological solutions and design recommendations for low damage non-structural wall systems for seismic actions by making use of alternative approaches. Light (steel/timber framed drywalls) and heavy (unreinforced clay brick) non-structural infill wall systems were studied by following an experimental/numerical research programme. Quasi-static reverse cyclic tests were carried out by utilizing a specially designed full scale reinforced concrete frame, which can be used as a re-usable bare frame. In this frame, two RC beams and two RC columns were connected by two un-bonded post tensioning bars, emulating a jointed ductile frame system (PRESSS technology). Due to the rocking behaviour at the beam-column joint interfaces, this frame was typically a low damage structural solution, with the post-tensioning guaranteeing a linear elastic behaviour. Therefore, this frame could be repeatedly used in all of the tests carried out by changing only the infill walls within this frame. Due to the linear elastic behaviour of this structural bare frame, it was possible to extract the exact behaviour of the infill walls from the global results. In other words, the only parameter that affected the global results was given by the infill walls. For the test specimens, the existing practice of construction (as built) for both light and heavy non-structural walls was implemented. In the light of the observations taken during these tests, modified low damage construction practices were proposed and tested. In total, seven tests were carried out: 1) Bare frame , in order to confirm its linear elastic behaviour. 2) As built steel framed drywall specimen FIF1-STFD (Light) 3) As built timber framed drywall specimen FIF2-TBFD (Light) 4) As built unreinforced clay brick infill wall specimen FIF3-UCBI (Heavy) 5) Low damage steel framed drywall specimen MIF1-STFD (Light) 6) Low damage timber framed drywall specimen MIF2-TBFD (Light) 7) Low damage unreinforced clay brick infill wall specimen MIF5-UCBI (Heavy) The tests of the as built practices showed that both drywalls and unreinforced clay brick infill walls have a low serviceability inter-storey drift limit (0.2-0.3%). Based on the observations, simple modifications and details were proposed for the low damage specimens. The details proved to be working effectively in lowering the damage and increasing the serviceability drift limits. For drywalls, the proposed low damage solutions do not introduce additional cost, material or labour and they are easily applicable in real buildings. For unreinforced clay brick infill walls, a light steel sub-frame system was suggested that divides the infill panel zone into smaller individual panels, which requires additional labour and some cost. However, both systems can be engineered for seismic actions and their behaviour can be controlled by implementing the proposed details. The performance of the developed details were also confirmed by the numerical case study analyses carried out using Ruaumoko 2D on a reinforced concrete building model designed according to the NZ codes/standards. The results have confirmed that the implementation of the proposed low damage solutions is expected to significantly reduce the non-structural infill wall damage throughout a building.

This dissertation describes the analytical research as part of a comprehensive research program to develop a new floor anchorage system for seismic resistant design, termed the Inertial Force-limiting Floor Anchorage System (IFAS). The IFAS intends to reduce damage in seismic resistant building structures by limiting the inertial force that develops in the building during earthquakes. The development of the IFAS is being conducted through a large research project involving both experimental and analytical research. This dissertation work focuses on analytical component of this research, which involves stand-alone computational simulation as well as analytical simulation in support of the experimental research (structural and shake table testing). The analytical research covered in this dissertation includes four major parts: (1) Examination of the fundamental dynamic behavior of structures possessing the IFAS (termed herein IFAS structures) by evaluation of simple two-degree of freedom systems (2DOF). The 2DOF system is based on a prototype structure, and simplified to represent only its fundamental mode response. Equations of motions are derived for the 2DOF system and used to find the optimum design space of the 2DOF system. The optimum design space is validated by transient analysis using earthquakes. (2) Evaluation of the effectiveness of IFAS designs for different design parameters through earthquake simulations of two-dimensional (2D) nonlinear numerical models of an evaluation structure. The models are based on a IFAS prototype developed by a fellow researcher on the project at Lehigh University. (3) Development and calibration of three-dimensional nonlinear numerical models of the shake table test specimen used in the experimental research. This model was used for predicting and designing the shake table testing program. (4) Analytical parameter studies of the calibrated shake table test model. These studies include: relating the shake table test performance to the previous evaluation structure analytical response, performing extended parametric analyses, and investigating and explaining certain unexpected shake table test responses. This dissertation describes the concept and scope of the analytical research, the analytical results, the conclusions, and suggests future work. The conclusions include analytical results that verify the IFAS effectiveness, show the potential of the IFAS in reducing building seismic demands, and provide an optimum design space of the IFAS.

Dry etching is an important technique for pattern transferin fabrication of most opto-electronic devices, since it canprovide good control of both structure size and shape even on asub-micron scale. Unfortunately, this process step may causedamage to the material which is detrimental to deviceperformance. It is therefore an objective of this thesis todevelop and investigate low damage etching processes for InPbased devices. An ion beam system in combination with hydrocarbon (CH4) based chemistries is used for etching. At variousion energies and gas flows the etching is performed in twomodes, reactive ion beam etching (RIBE) and chemical assistedion beam etching (CAIBE). How these conditions affect both etchcharacteristics (e.g. etch rates and profiles, surfacemorphology and polymer formation) and etch induced damage (onoptical and electrical properties) is evaluated and discussed.Attention is also paid to the effects of typical post etchingtreatments such as annealing on the optical and electricalproperties. An important finding is the correlation betweenas-etched surface morphology and recovery/degradation inphotoluminescence upon annealing in PH3. Since this type of atmosphere is typical forcrystal regrowth (an important process step in III/Vprocessing) a positive result is imperative. A low ion energy N2/CH4/H2CAIBE process is developed which not onlysatisfies this criteria but also exhibits good etchcharacteristics. This process is used successfully in thefabrication of laser gratings. In addition to this, the abilityof the ion beam system to modify the surface morphology in acontrollable manner is exploited. By exposing such modifiedsurfaces to AsH3/PH3, a new way to vary size and density of InAs(P)islands formed on the InP surfaces by the As/P exchangereaction is presented. This thesis also proposes a new etch chemistry, namelytrimethylamine ((CH3)3N or TMA), which is a more efficient methyl sourcecompared to CH4because of the low energy required to break the H3C-N bond. Since methyl radicals are needed for theetching it is presumably a better etching chemistry. A similarinvestigation as for the CH4chemistry is performed, and it is found that bothin terms of etch characteristics and etch induced damage thisnew chemistry is superior. Extremely smooth morphologies, lowetch induced damage and an almost complete recovery uponannealing can be obtained with this process. Significantly,this is also so at relatively high ion energies which allowshigher etch rates. <b>Keywords:</b>InP, dry etching, ion beam etching, RIBE,CAIBE, hydrocarbon chemistry, trimethylamine, As/P exchangereaction, morpholoy, low damage, AFM, SCM, annealing

<p>Dry etching is an important technique for pattern transferin fabrication of most opto-electronic devices, since it canprovide good control of both structure size and shape even on asub-micron scale. Unfortunately, this process step may causedamage to the material which is detrimental to deviceperformance. It is therefore an objective of this thesis todevelop and investigate low damage etching processes for InPbased devices.</p><p>An ion beam system in combination with hydrocarbon (CH₄) based chemistries is used for etching. At variousion energies and gas flows the etching is performed in twomodes, reactive ion beam etching (RIBE) and chemical assistedion beam etching (CAIBE). How these conditions affect both etchcharacteristics (e.g. etch rates and profiles, surfacemorphology and polymer formation) and etch induced damage (onoptical and electrical properties) is evaluated and discussed.Attention is also paid to the effects of typical post etchingtreatments such as annealing on the optical and electricalproperties. An important finding is the correlation betweenas-etched surface morphology and recovery/degradation inphotoluminescence upon annealing in PH₃. Since this type of atmosphere is typical forcrystal regrowth (an important process step in III/Vprocessing) a positive result is imperative. A low ion energy N₂/CH₄/H₂CAIBE process is developed which not onlysatisfies this criteria but also exhibits good etchcharacteristics. This process is used successfully in thefabrication of laser gratings. In addition to this, the abilityof the ion beam system to modify the surface morphology in acontrollable manner is exploited. By exposing such modifiedsurfaces to AsH₃/PH₃, a new way to vary size and density of InAs(P)islands formed on the InP surfaces by the As/P exchangereaction is presented.</p><p>This thesis also proposes a new etch chemistry, namelytrimethylamine ((CH₃)₃N or TMA), which is a more efficient methyl sourcecompared to CH₄because of the low energy required to break the H₃C-N bond. Since methyl radicals are needed for theetching it is presumably a better etching chemistry. A similarinvestigation as for the CH₄chemistry is performed, and it is found that bothin terms of etch characteristics and etch induced damage thisnew chemistry is superior. Extremely smooth morphologies, lowetch induced damage and an almost complete recovery uponannealing can be obtained with this process. Significantly,this is also so at relatively high ion energies which allowshigher etch rates.</p><p><b>Keywords:</b>InP, dry etching, ion beam etching, RIBE,CAIBE, hydrocarbon chemistry, trimethylamine, As/P exchangereaction, morpholoy, low damage, AFM, SCM, annealing</p>

The Insurance Bureau of Canada released a report in 2013 that evaluated the seismic risk of two major metropolitan areas of Canada, with projected losses of $75bn in British Columbia along the Cascadia subduction zone, and $63bn in the east through the Ottawa-Montreal-Quebec corridor. Such reports should prompt researchers and designers alike to rethink the way that seismic design is approached in Canada to develop resilient and sustainable cities for the future. To mitigate the economic losses associated with earthquake damage to buildings in seismically active areas, controlled rocking steel braced frames have been developed as a seismically resilient low-damage lateral-force resisting system. Controlled rocking steel braced frames (CRSBFs) mitigate structural damage during earthquakes through a controlled rocking mechanism, where energy dissipation can be provided at the base of the frame, and pre-stressed tendons pull the frame back to its centred position after rocking. The result is a building for which the residual drifts of the system after an earthquake are essentially zero, and the energy dissipation does not result from structural damage. Design methods for the base rocking joint and the capacity-protected frame members in CRSBFs have been proposed and validated both numerically and experimentally. However, the is no consensus on how to approach the design of the frame members, questions remain regarding how best to design CRSBFs to prevent building collapse, and no experimental work has been done regarding how to connect the CRSBF to the rest of the structure to accommodate the rocking motion. Because the force limiting mechanism of a CRSBF is rocking only at the base of the frame, the frame member forces are greatly influenced by the higher-mode response, resulting in more complex methods to design the frame members. This thesis begins by outlining two new design procedures for the frame members in controlled rocking steel braced frames that target both simplicity and accuracy. The first is a dynamic procedure that requires a truncated response spectrum analysis on a model of the frame with modified boundary conditions to consider the rocking behaviour. The second is an equivalent static procedure that does not require any modifications to the elastic frame model, instead using theory-based lateral force distributions to consider the higher modes of the rocking structure. Neither method requires empirical calibration to estimate the forces at the target intensity. The base rocking joint design is generally in good agreement between the various research programs pioneering the development of the CRSBFs. However, the numerous parameters available to select during the design of the base rocking joint give designers an exceptional amount of control over the performance of the system, and little research is available on how best to select these parameters to target or minimise the probability of collapse for the building. This thesis presents a detailed numerical model to capture collapse of buildings with CRSBFs as their primary lateral force resisting system and uses this model to generate collapse fragility curves for different base rocking joint design parameters. The parameters include the response modification factor, the hysteretic energy dissipation ratio, and the post-tensioning prestress ratio. This work demonstrates that CRSBFs are resilient against collapse, as designing the base rocking joint with response modification factors as large as 30, designing the post-tensioning to prevent yielding at moderate seismic hazard levels, and using zero energy dissipation could lead to designs with acceptable margins of safety against collapse. While the design procedures are shown to be accurate for estimating the frame member force demand for the targeted intensity level, there is still a high level of uncertainty around what intensity of earthquake a building will experience during its lifespan, and there is no consensus on what intensity should be targeted for design. To address this, the ability of the capacity design procedures to provide a sufficiently low probability of collapse due to excessive frame member buckling and yielding is evaluated and compared to the probability that the building will collapse due to excessive rocking of the frame. The results of the research presented here suggest that the probability of collapse due to either frame member failure or excessive rocking should be evaluated separately, and that targeting the intensity with a 10% probability of exceedance in 50 years is sufficient for the design of the frame members. Finally, critical to the implementation of CRSBFs in practice is how they may be connected to the rest of the structure to accommodate the uplifting of the CRSBF while rocking under large lateral forces. An experimental program was undertaken to test three proposed connection details to accommodate the relative uplifts and forces. The connections that accommodate the uplifts through sliding performed better than that which accommodated the uplifts though material yielding, but the best way to transfer the forces and accommodate the uplifting without influencing the overall behaviour of the system is to position the connection such that it does not need to undergo large uplifts and carry lateral force simultaneously. A detailed numerical model of the experimental setup is presented and is shown to simulate the important response quantities for each of the tested connections. Using the results of this work, designers worldwide will be confident to design CRSBFs for structures from the base rocking joint to the selection of floor-to-frame connections for a complete system design while ensuring a safe and resilient building structure for public use and well-being. / Thesis / Doctor of Philosophy (PhD) / Traditional approaches to seismic design of buildings have generally been successful at preventing collapse and protecting the lives of the occupants. However, the buildings are often left severely damaged, often beyond repair. To address these concerns, controlled rocking steel braced frames have been proposed as part of a new construction technique to mitigate or prevent damage to steel buildings during earthquakes, but several aspects of the design and overall safety have yet to be explored or demonstrated. This thesis proposes and validates new tools to design controlled rocking steel braced frames and provides recommendations on how best to design them to achieve a safe probability against collapse. Details are proposed and presented for components to connect the controlled rocking steel braced frames into the rest of the structure. The findings of this thesis will aid practitioners looking to deliver resilient and sustainable structural designs for buildings in our cities of the future.