Hydroxide catalysis and indium bonding research for the design of ground-based gravitational wave detectors

Q Science (General);QB Astronomy;QC Physics

Phelps, Margot Hensler ; Rowan, Sheila

University:University of Glasgow

Department:School of Physics and Astronomy

关键词: Physics and astronomy, gravitational waves, material science.;   

Others  :  http://theses.gla.ac.uk/30604/7/2018PhelpsPhD.pdf

来源: University of Glasgow

【 摘 要 】

In 2015, a gravitational wave (GW) signal from a binary black hole merger passed through the arms of the US-based Advanced LIGO (aLIGO) interferometers, resulting in the first direct detection of gravitational waves. This long-awaited observation made worldwide news one hundred years after Einstein first predicted the existence of GWs in 1916. Since the first detection, four more binary black hole inspiral events have been detected, as well as the ground-breaking GW observation of a binary neutron star inspiral. To detect these signals, ground-based GW detectors like aLIGO and the French-Italian detector, Advanced Virgo, need to be sensitive to changes in separation of close to 10^-19m between freely suspended test masses spaced up to 4km apart. This has always been a challenge to achieve, thus 50 years of technological developments were needed to make these first detections possible.Following the first observations of coalescing black holes and neutron stars, it is essential to pursue technological advancements that improve the sensitivities of ground-based detectors. Doing so will increase the signal-to-noise ratio of future detectors, which will allow for the better extraction of astrophysical source parameters. Observing more types of astrophysical sources, and at greater distances from the Earth will further the field of GW astronomy. One such area of advancement is to pair the operation of detectors at cryogenic temperatures with improvements in mirror and suspension design, with the aim of improving sensitivities by lessening the effects of thermal noise. Fused silica, currently used for the mirror substrates and suspension fibre elements in all detectors that operate at room temperature, cannot be used in detectors that operate at cryogenic temperatures due to its unfavourable thermo-mechanical properties. Thus a change of mirror substrate and suspension material is necessary for the construction of cryogenic detectors. There are two promising candidates for cryogenic mirrors and suspension elements, sapphire and silicon. Currently one cryogenic detector, the Japan-based KAGRA observatory, is under construction using sapphire as a material for its mirrors and some suspension elements. Other future detectors currently in the design phase, such as the Einstein Telescope (ET) in Europe and Voyager, in the USA may use silicon or sapphire material in their mirror suspensions.In all ground-based detectors the test masses are supported in multi-stage pendulum suspensions, where the last stages are quasi-monolithic. In the quasi-monolithic stage, the test masses are suspended from penultimate masses via fibres, welded to an interface piece, or "ear". Currently these ears are connected to the test masses using a method called hydroxide catalysis bonding, which creates a strong, low noise joint. This bonding technique has been used successfully in room temperature detectors for 17 years. This thesis details research into hydroxide catalysis bonding, with a focus on its use to create cryogenic crystalline suspensions for future ground-based detectors. The use of indium as an alternative bonding technology for joints in low temperature crystalline suspensions is also investigated. The aim of this study is to research possible ways to implement indium bonding into suspension design along with hydroxide catalysis bonds to create a more versatile and easily repairable system. This work was completed with the aim of investigating novel ways of implementing bond techniques into GW detectors, and studying their material properties. The breaking stress and stability of different bond technologies were investigated, as well as their thermal noise levels and impact on overall detector sensitivity. The majority of substrate materials used in this thesis were sapphire and silicon, as these are the two materials of choice for use in future cryogenic detectors. Measurements of the Young's modulus of hydroxide catalysis bonds between fused silica were also completed and used to model the thermal noise contribution of bonds in a prototype test mass for the possible room temperature upgrade to aLIGO, A+.In Chapter 1 an overview of the field of gravitational wave research is given. An explanation of GW sources and a history of the different types of ground-based GW detectors are summarised here, with a focus on Michelson-type interferometric detectors, used to make the first direct GW detections. The noise sources that affect the sensitivity of interferometric detectors are also reviewed.In Chapter 2 there is a summary of several different bonding techniques that could be considered for making joints between the test masses and suspension elements of GW detectors. The mechanisms of bond formation as well as the advantages and disadvantages to each approach are covered, especially in the context of the requirements for use in a GW detector. Finally hydroxide catalysis and indium bonding are introduced as possible techniques to join the suspension and mirror elements in GW detectors.In Chapter 3 the breaking stresses of hydroxide catalysis bonds between c-plane sapphire substrates as a function of time is studied. The aim of this experiment is twofold. The breaking stress of bonds that have been allowed to cure for shorter lengths of time is investigated to gain insight into the chemical processes of the bonds as they develop. Additionally, it is crucial to know the breaking stress over longer periods of curing time to be assured that they will not fail in the long term. In fact, this study found that hydroxide catalysis bonded sapphire shows an initial drop in breaking stress, which then levelled off at 15-16MPa. These results agree with similar trends found in shorter curing time tests on sapphire and fused silica completed in the past.In Chapter 4 the effect of crystal orientation on the tensile strength of hydroxide catalysis bonded sapphire is investigated. Specifically, the breaking stress of bonds between a-a and m-m planes of sapphire jointed with hydroxide catalysis bonds is studied, using samples of the same geometry and jointed using the same bonding procedures as those presented in Chapter 3. These samples were allowed to cure at room temperature for 4 weeks, then the samples were strength tested. The breaking stresses were recorded and compared with the breaking stress results of c-c plane sapphire, also cured for 4 weeks at room temperature, reported in the previous chapter.In Chapter 5 a non-destructive technique of measuring the Young's modulus of hydroxide catalysis bonds between silica and between sapphire is developed. This approach uses acoustic pulses from an ultrasonic transducer transmitted through the bonded samples, and the portion of the acoustic wave that is reflected back from the embedded bond layer is recorded and studied. The bond Young's modulus was extracted from the data by analysis of the amplitudes of the acoustic pulses reflected from the bonds. A Young's modulus value of 15.3+/-5.2GPa for \hcbed sapphire and 21.5+/-6.6GPa for bonded fused silica was found with this approach. A Bayesian analysis model of the reflected acoustic signal and the underlying noise background was developed to analyse the low SNR signals of bonds between fused silica. A value of 18.5+/-2GPa, with a 90% confidence range was found with this approach, agreeing well with the results from the pulse amplitude analysis.In Chapter 6 the new Young's modulus value found in Chapter 5 is used to assess the mechanical loss and thermal noise budgets of hydroxide catalysis bonds in different mirror suspension geometries. Two room temperature test masses were modelled; a bonded aLIGO mass and a bonded prototype test mass, of a design suitable for use in A+. Three different cryogenic masses were also modelled; first a sapphire KAGRA mass, followed by a prototype sapphire ET mass, and a prototype silicon ET mass. The thermal noise budgets of the bonds in all of these cases were found to be below the anticipated technical noise requirement for bonds, which is based on each detector's current design sensitivity curves. This indicates that hydroxide catalysis bonds are suitable for use in current detectors and for the design of future ones.In Chapter 7 different approaches to creating indium bonding procedures for use in cryogenic ground-based detectors are studied. Hybrid suspension designs that utilize both indium and hydroxide catalysis bonding are being considered in cryogenic detector designs such as KAGRA or ET. It is proposed that the \hydroxide catalysis bonds would be used to fix the test masses to the suspension elements. This takes advantage of their high breaking stress under shear and peeling, as has been successfully demonstrated in the past for room temperature detectors such as Virgo, aLIGO, or the Germany-based detector GEO600. Indium's low tensile strength means it cannot be used as a joint under tensile or shear load. However it is being considered for use in compressive joints, such as between the fibres and ears or between the fibres and blade springs. This would be done for contingency reasons, since indium can be de-bonded and re-bonded relatively easily, whereas hydroxide catalysis bonds cannot. In the event of a fibre break or a test mass upgrade, the whole bonded test mass assembly could be removed by de-bonding the indium bond interface. It could then be replaced by re-bonding it, making it a good option for future cryogenic mirror suspensions.Two indium bonding approaches are investigated, diffusion bonding and induction bonding. In both cases the substrates used were polished silicon, and the indium layers between them were made with different combinations of thin thermally deposited films and foils. The tensile strength and a post-break visual inspection of the indium bonds were used as a standard by which to judge bond quality and repeatability.