Condition monitoring and operational decision making in semiconductor manufacturing

Today, the majority of semiconductor fabrication plants (fabs) conduct equipment preventive maintenance based on statistically-derived time- or wafer-count-based intervals. While these practices have had relative success in managing equipment availability and product yield, the cost, both in time and materials, remains high. Condition-based maintenance has been successfully adopted in several industries, where costs associated with equipment downtime range from potential loss of life to unacceptable affects to companies’ bottom lines. In this paper, we present a method for the monitoring of complex systems in the presence of multiple operating regimes. In addition, the new representation of degradation processes will be used to define an optimization procedure that facilitates concurrent maintenance and operational decision-making in a manufacturing system. This decision-making procedure metaheuristically maximizes a customizable cost function that reflects the benefits of production uptime, and the losses incurred due to deficient quality and downtime. The new degradation monitoring method is illustrated through the monitoring of a deposition tool operating over a prolonged period of time in a major fab, while the operational decision-making is demonstrated using simulated operation of a generic cluster tool.

Degradation modeling and monitoring of machines using operation-specific hidden Markov models

In this paper, a novel data-driven approach to monitoring of systems operating under variable operating conditions is described. The method is based on characterizing the degradation process via a set of operation-specific hidden Markov models (HMMs), whose hidden states represent the unobservable degradation states of the monitored system while its observable symbols represent the sensor readings. Using the HMM framework, modeling, identification and monitoring methods are detailed that allow one to identify a HMM of degradation for each operation from mixed-operation data and perform operation-specific monitoring of the system. Using a large data set provided by a major manufacturer, the new methods are applied to a semiconductor manufacturing process running multiple operations in a production environment.

Perylene-based profluorescent nitroxides for the rapid monitoring of polyester degradation upon weathering: An assessment

A profluorescent nitroxide possessing an isoindoline nitroxide moiety linked to a perylene fluorophore was developed to monitor radical mediated degradation of melamine-formaldehyde crosslinked polyester coil coatings in an industry standard accelerated weathering tester. Trapping of polyester-derived radicals (most likely C-radicals) that are generated during polymer degradation leads to fluorescent closed-shell alkoxy amines, which was used to obtain time-dependent degradation profiles to assess the relative stability of different polyesters towards weathering. The nitroxide probe couples excellent thermal stability and satisfactory photostability with high sensitivity and enables detection of free radical damage in polyesters under conditions that mimic exposure to the environment on a time scale of hours rather than months or years required by other testing methods. There are indications that the profluorescent nitroxide undergoes partial photo-degradation in the absence of polymer-derived radicals. Unexpectedly, it was also found that UV-induced fragmentation of the NO–C bond in closed-shell alkoxy amines leads to regeneration of the profluorescent nitroxide and the respective C-radical. The maximum fluorescence intensity that could be achieved with a given probe concentration is therefore not only determined by the amount of polyester radicals formed during accelerated weathering, but also by the light-driven side reactions of the profluorescent nitroxide and the corresponding alkoxy amine radical trapping products. Studies to determine the optimum probe concentration in the polymer matrix revealed that aggregation and re-absorption effects lowered the fluorescence intensity at higher concentrations of the profluorescent nitroxide, but too low probe concentrations, where these effects would be avoided, were not sufficient to trap the amount of polyester radicals formed upon weathering. The optimized experimental conditions were used to assess the impact of temperature and UV irradiance on polymer degradation during accelerated weathering.

The synthesis and application of novel profluorescent nitroxides as probes for polymer degradation

This PhD project has expanded the knowledge in the area of profluorescent nitroxides with regard to the synthesis and characterisations of novel profluorescent nitroxide probes as well as physical characterisation of the probe molecules in various polymer/physical environments. The synthesis of the first example of an azaphenalene-based fused aromatic nitroxide TMAO, [1,1,3,3-tetramethyl-2,3-dihydro-2-azaphenalen-2-yloxyl, was described. This novel nitroxide possesses some of the structural rigidity of the isoindoline class of nitroxides, as well as some properties akin to TEMPO nitroxides. Additionally, the integral aromatic ring imparts fluorescence that is switched on by radical scavenging reactions of the nitroxide, which makes it a sensitive probe for polymer degradation. In addition to the parent TMAO, 5 other azaphenalene derivatives were successfully synthesised. This new class of nitroxide was expected to have interesting redox properties when the structure was investigated by high-level ab initio molecular orbitals theory. This was expected to have implications with biological relevance as the calculated redox potentials for the azaphenalene ring class would make them potent antioxidant compounds. The redox potentials of 25 cyclic nitroxides from four different structural classes (pyrroline, piperidine, isoindoline and azaphenalene) were determined by cyclic voltammetry in acetonitrile. It was shown that potentials related to the one electron processes of the nitroxide were influenced by the type of ring system, ring substituents or groups surrounding the moiety. Favourable comparisons were found between theoretical and experimental potentials for pyrroline, piperidine and isoindoline ring classes. Substitution of these ring classes, were correctly calculated to have a small yet predictable effect on the potentials. The redox potentials of the azaphenalene ring class were underestimated by the calculations in all cases by at least a factor of two. This is believed to be due to another process influencing the redox potentials of the azaphenalene ring class which is not taken into account by the theoretical model. It was also possible to demonstrate the use of both azaphenalene and isoindoline nitroxides as additives for monitoring radical mediated damage that occurs in polypropylene as well as in more commercially relevant polyester resins. Polymer sample doped with nitroxide were exposed to both thermo-and photo-oxidative conditions with all nitroxides showing a protective effect. It was found that isoindoline nitroxides were able to indicate radical formation in polypropylene aged at elevated temperatures via fluorescence build-up. The azaphenalene nitroxide TMAO showed no such build-up of fluorescence. This was believed to be due to the more labile bond between the nitroxide and macromolecule and the protection may occur through a classical Denisov cycle, as is expected for commercially available HAS units. Finally, A new profluorescent dinitroxide, BTMIOA (9,10-bis(1,1,3,3- tetramethylisoindolin-2-yloxyl-5-yl)anthracene), was synthesised and shown to be a powerful probe for detecting changes during the initial stages of thermo-oxidative degradation of polypropylene. This probe, which contains a 9,10-diphenylanthracene core linked to two nitroxides, possesses strongly suppressed fluorescence due to quenching by the two nitroxide groups. This molecule also showed the greatest protective effect on thermo-oxidativly aged polypropylene. Most importantly, BTMIOA was found to be a valuable tool for imaging and mapping free-radical generation in polypropylene using fluorescence microscopy.

Sustainable subdivisions – ventilation monitoring

Australia’s current pattern of residential development is typified by relatively low-density subdivision of land and highlights the necessity for development to be more sustainable to avoid unnecessary demand on natural resources and to prevent environmental degradation and to safeguard the environment for future generations. What role can climatically appropriate sub-division design play in decreasing the use of energy required to cool premises by maximising access to natural ventilation? How can this design be achieved?

Intelligent prognostics of machinery health utilising suspended condition monitoring data

The ability to forecast machinery failure is vital to reducing maintenance costs, operation downtime and safety hazards. Recent advances in condition monitoring technologies have given rise to a number of prognostic models for forecasting machinery health based on condition data. Although these models have aided the advancement of the discipline, they have made only a limited contribution to developing an effective machinery health prognostic system. The literature review indicates that there is not yet a prognostic model that directly models and fully utilises suspended condition histories (which are very common in practice since organisations rarely allow their assets to run to failure); that effectively integrates population characteristics into prognostics for longer-range prediction in a probabilistic sense; which deduces the non-linear relationship between measured condition data and actual asset health; and which involves minimal assumptions and requirements. This work presents a novel approach to addressing the above-mentioned challenges. The proposed model consists of a feed-forward neural network, the training targets of which are asset survival probabilities estimated using a variation of the Kaplan-Meier estimator and a degradation-based failure probability density estimator. The adapted Kaplan-Meier estimator is able to model the actual survival status of individual failed units and estimate the survival probability of individual suspended units. The degradation-based failure probability density estimator, on the other hand, extracts population characteristics and computes conditional reliability from available condition histories instead of from reliability data. The estimated survival probability and the relevant condition histories are respectively presented as “training target” and “training input” to the neural network. The trained network is capable of estimating the future survival curve of a unit when a series of condition indices are inputted. Although the concept proposed may be applied to the prognosis of various machine components, rolling element bearings were chosen as the research object because rolling element bearing failure is one of the foremost causes of machinery breakdowns. Computer simulated and industry case study data were used to compare the prognostic performance of the proposed model and four control models, namely: two feed-forward neural networks with the same training function and structure as the proposed model, but neglected suspended histories; a time series prediction recurrent neural network; and a traditional Weibull distribution model. The results support the assertion that the proposed model performs better than the other four models and that it produces adaptive prediction outputs with useful representation of survival probabilities. This work presents a compelling concept for non-parametric data-driven prognosis, and for utilising available asset condition information more fully and accurately. It demonstrates that machinery health can indeed be forecasted. The proposed prognostic technique, together with ongoing advances in sensors and data-fusion techniques, and increasingly comprehensive databases of asset condition data, holds the promise for increased asset availability, maintenance cost effectiveness, operational safety and – ultimately – organisation competitiveness.

Utilising Reliability and Condition Monitoring Data for Asset Health Prognosis

The ability to forecast machinery health is vital to reducing maintenance costs, operation downtime and safety hazards. Recent advances in condition monitoring technologies have given rise to a number of prognostic models which attempt to forecast machinery health based on condition data such as vibration measurements. This paper demonstrates how the population characteristics and condition monitoring data (both complete and suspended) of historical items can be integrated for training an intelligent agent to predict asset health multiple steps ahead. The model consists of a feed-forward neural network whose training targets are asset survival probabilities estimated using a variation of the Kaplan–Meier estimator and a degradation-based failure probability density function estimator. The trained network is capable of estimating the future survival probabilities when a series of asset condition readings are inputted. The output survival probabilities collectively form an estimated survival curve. Pump data from a pulp and paper mill were used for model validation and comparison. The results indicate that the proposed model can predict more accurately as well as further ahead than similar models which neglect population characteristics and suspended data. This work presents a compelling concept for longer-range fault prognosis utilising available information more fully and accurately.

Characterising in situ activation and degradation of hindered amine light stabilisers using liquid extraction surface analysis-mass spectrometry

Changes in the molecular structure of polymer antioxidants such as hindered amine light stabilisers (HALS) is central to their efficacy in retarding polymer degradation and therefore requires careful monitoring during their in-service lifetime. The HALS, bis-(1-octyloxy-2,2,6,6-tetramethyl-4-piperidinyl) sebacate (TIN123) and bis-(1,2,2,6,6-pentamethyl-4-piperidinyl) sebacate (TIN292), were formulated in different polymer systems and then exposed to various curing and ageing treatments to simulate in-service use. Samples of these coatings were then analysed directly using liquid extraction surface analysis (LESA) coupled with a triple quadrupole mass spectrometer. Analysis of TIN123 formulated in a cross-linked polyester revealed that the polymer matrix protected TIN123 from undergoing extensive thermal degradation that would normally occur at 292 degrees C, specifically, changes at the 1- and 4-positions of the piperidine groups. The effect of thermal versus photo-oxidative degradation was also compared for TIN292 formulated in polyacrylate films by monitoring the in situ conversion of N-CH3 substituted piperidines to N-H. The analysis confirmed that UV light was required for the conversion of N-CH3 moieties to N-H - a major pathway in the antioxidant protection of polymers - whereas this conversion was not observed with thermal degradation. The use of tandem mass spectrometric techniques, including precursor-ion scanning, is shown to be highly sensitive and specific for detecting molecular-level changes in HALS compounds and, when coupled with LESA, able to monitor these changes in situ with speed and reproducibility. (C) 2013 Elsevier B. V. All rights reserved.

Integration of science and monitoring of river ecosystem health to guide investments in catchment protection and rehabilitation

1. Stream ecosystem health monitoring and reporting need to be developed in the context of an adaptive process that is clearly linked to identified values and objectives, is informed by rigorous science, guides management actions and is responsive to changing perceptions and values of stakeholders. To be effective, monitoring programmes also need to be underpinned by an understanding of the probable causal factors that influence the condition or health of important environmental assets and values. This is often difficult in stream and river ecosystems where multiple stressors, acting at different spatial and temporal scales, interact to affect water quality, biodiversity and ecosystem processes. 2. In this article, we describe the development of a freshwater monitoring programme in South East Queensland, Australia, and how this has been used to report on ecosystem health at a regional scale and to guide investments in catchment protection and rehabilitation. We also discuss some of the emerging science needs to identify the appropriate scale and spatial arrangement of rehabilitation to maximise river ecosystem health outcomes and, at the same time, derive other benefits downstream. 3. An objective process was used to identify potential indicators of stream ecosystem health and then test these across a known catchment land-use disturbance gradient. From the 75 indicators initially tested, 22 from five indicator groups (water quality, ecosystem metabolism, nutrient cycling, invertebrates and fish) responded strongly to the disturbance gradient, and 16 were subsequently recommended for inclusion in the monitoring programme. The freshwater monitoring programme was implemented in 2002, funded by local and State government authorities, and currently involves the assessment of over 120 sites, twice per year. This information, together with data from a similar programme on the region's estuarine and coastal marine waters, forms the basis of an annual report card that is presented in a public ceremony to local politicians and the broader community. 4. Several key lessons from the SEQ Healthy Waterways Programme are likely to be transferable to other regional programmes aimed at improving aquatic ecosystem health, including the importance of a shared common vision, the involvement of committed individuals, a cooperative approach, the need for defensible science and effective communication. 5. Thematic implications: this study highlights the use of conceptual models and objective testing of potential indicators against a known disturbance gradient to develop a freshwater ecosystem health monitoring programme that can diagnose the probable causes of degradation from multiple stressors and identify the appropriate spatial scale for rehabilitation or protection. This approach can lead to more targeted management investments in catchment protection and rehabilitation, greater public confidence that limited funds are being well spent and better outcomes for stream and river ecosystem health.