This paper studies two different approaches for evanescent wave optical sensing: an horizontal one and a vertical one. In horizontal waveguides, the evanescent wave is distributed on the upper cladding. While in a vertical configuration, the evanescent wave is distributed on the left and right sides of the waveguide. In an horizontal configuration the evanescent wave can be also on both sides of the waveguide in order to increase the optical energy for sensing if the substrate under the waveguide is locally removed. However, in this configuration to achieve sensitive devices, the layers have to be freestanding and thin [1] limiting practical implementations of such approaches. Furthermore, very few materials can be defined as tall and thin in the case of a vertical configuration, as the deposition techniques often used (PECVD/LPCVD) are meant for films in the couple of micron range. In the following we will investigate the properties of the materials used but also the fabrication feasibility for both configurations.
Microlens Arrays in silicon are suitable for an important wavelength range within the IR spectrum, since silicon features relatively high refractive index and is transparent at the aimed wavelengths, leading to microlenses with a focal length short enough to allow compact systems, offering an alternative for applications where miniaturization and reduction of alignment and packaging costs are necessary. The microlenses are meant to sample and focus an IR beam on a focal plane array, which might be an image sensor, or a dedicated IR sensor, as for instance lab-on-chip, or a selective gas detection system. Nowadays refractive microlenses are manufactured using sophisticated techniques with relatively high costs and complexity of well controlled steps, like thermal reflow, and grayscale lithography. We hereby propose an alternative solution for microfabrication of silicon microlens arrays, with a single-mask step using KOH anisotropic etching of Si. The proposed technique solves many current demands, like achieving high reproducibility, fill-factor close to 100%, and higher precision of focal axes alignment. We have made optical profilometric measurements to estimate the shape, roughness and the focal distance. We have also observed the focal points imaging in the IR spectrum, proving that the silicon microlenses actually yield the results expected.
This work reports the surface functionalisation of evanescent waveguide sensors to immobilise E. coli. In biosensors, the
surface functionalisation is an important treatment to ensure that the sensor properly detects the cells of interest. In this
paper, we study the thin film surface functionalisation of a TiO2 evanescent waveguide sensor and their effect on light
transmission for the early detection of E. coli in post colon surgery. TiO2 deposited using atomic layer deposition (ALD)
is used as waveguide material. Four layers are used in the functionalisation : the self-assembled monolayer (SAM), the
protein, 1-ethyl-3-(3-dimethylaminopropyl) (EDC) and the antibodies. Aminopropyltriethoxysilane (APTES) is used as
SAM and reacts with -OH group (hydroxyl). The -OH group must be provided on substrate. In order to have the proper
-OH group we deposited 10 nm SiO2 on the waveguides using PECVD and then treated the samples in oxygen plasma
chamber for 2 minutes to create the groups. Afterward APTES is immediately applied on the surface after every layers of
the functionalisation process. The second layer (Protein A) of the functionalisation is then put on APTES as interlayer.
EDC is used as crosslink agent between APTES and antibodies. The light of Superluminescent light emitting diodes
(SLEDs) (λ = 1.3 μm, 400 mA) is channelled using an optical fibre into the functionalised waveguides. The transmitted
light is measured with a photodiode. The sensitivity of the sensor was evaluated using several different drain fluid
concentrations in medium.
Optical Coherence Tomography (OCT) is a promising medical imaging technique. It has found applications in many fields of medicine and has a large potential for the optical biopsy of tumours. One of the technological challenges impairing faster adoption of OCT is the relative complexity of the optical instrumentation required, which translates into expensive and bulky setups. In this paper we report an implementation of Time Domain OCT (TD-OCT) based on a silicon photonic platform. The devices are fabricated using Silicon-On-Insulator (SOI) wafers, on which rib waveguides are defined. While most of the components needed are well-known in this technology, a fast delay line with sufficient scanning range is a specific requirement of TD-OCT. In the system reported, this was obtained making use of the thermo-optical effect of silicon. By modulating the thermal resistance of the waveguide to the substrate, it is possible to establish a trade-off between maximum working frequency and power dissipation. Within this trade-off, the systems obtained can be operated in the kHz range, and they achieve temperature shifts corresponding to scanning ranges of over 2mm. Though the current implementation still requires external sources and detectors to be coupled to the Planar Lightwave Circuit (PLC), future work will include three-dimensional integration of these components onto the substrate. With the potential to include the read-out and driving electronics on the same die, the reported approach can yield extremely compact and low-cost TD-OCT systems, enabling a wealth of new applications, including gastrointestinal pills with optical biopsy capabilities.
Light delivery and optical monitoring during photodynamic therapy (PDT) is often limited by the need for a physical
optical link between the light source and detection devices and the treatment volume. This can be critical when sources
need to be implanted within the body for extended periods. We report on the latest developments for a telemetric PDT
delivery and monitoring device that can dynamically vary the local illumination parameters based on the in-situ fluence
rate within the PDT target volume. Local light delivery and collection is achieved using solid-state optodes, microfabricated
on a silicon substrate. Photodiodes have been produced using a standard bipolar process. Chip-form LEDs are
then assembled into micro-machined pits adjacent to the light fluence rate detectors. The devices (1.2×1.2mm2) are
bonded to a flexible PCB together with the remaining electronics. Power coupling and communications are achieved by
means of an inductive link while light delivery and fluence rate monitoring are digitally managed using a
microcontroller. These devices are being tested in optical phantoms and in pre-clinical models. Our results show that it is
possible to manufacture solid-state optodes of suitable dimensions and that it is feasible to telemetrically deliver and
control the local fluence rate using them. It can also be concluded from our work that while the optode is sufficiently
small to be useful as a light delivery and monitoring device, digital control, read-out electronics and power coupling can
benefit from further optimization and miniaturization.
Light transport in trabecular bone is not well understood despite its clinical interest. Recent experimental studies on optical bone biopsy are lacking models that relate their measurements to the underlying morphology and thus to tissue condition. Laser surgery can also benefit from a better understanding of energy distribution in cancellous bone. A Monte Carlo (MC) simulation environment, able to efficiently compute complex geometries and account for refraction and reflection on tissue boundaries has been developed to provide the missing insight. The geometry description is based on a 3D triangle mesh organised in a bounding-volume hierarchy. This efficient structure allows a fast photon-surface intersection test, ensuring a sufficient number of photon paths and thus a good signal-to-noise ratio. The simulation program has been validated against well-known problems of refractive optics and turbid media. This new tool has been applied to a set of numerical phantoms indicating that morphology may have a fundamental impact on long-range light transport. The simulation environment has also been used on high-resolution models of trabecular bone, based on micro-CT scans. Calculation of time resolved signals in transmission and reflectance geometries has been demonstrated, paving the way to numerical evaluation of new minimally invasive diagnostic techniques, and offering a link to evaluation of Optical Coherence Tomography (OCT) in complex heterogeneous geometries. Preliminary experimental results in support of the mentioned effects are presented.
Planar silicon carbide (SiC) waveguides are proposed for fabrication on a silicon substrate with a oxide isolation layer. Using post deposition annealing it is possible to achieve low Polarization-Dependent Loss (PDL) within optical SiC waveguides fabricated using a low temperature deposition technique. Those waveguides have been successfully used in power splitters and cantilivers. These first devices open the way for photonic sensing in harsh environment using SiC.
This paper presents an on-going work to develop micromachined silicon-based strain sensor inspired from the campaniform sensillum of insects. We present simple optical setup for the characterisation of a membrane-in-recess structure as an early stage in mimicking the natural sensor. The microstructure is a 500 nm-thick SiO2/SiN circular membrane, burried 13 μm from the surface of a 3x3 mm, 525 μm thick Si-chip. The chip was attached to a 45x10x0.525 mm Si beam. The simple optical characterisation setup is based on imaging the reflected laser beam from the biomimetic structure. Since an optical cavity between the membrane and the Si beams beneath was formed, ideal flat-parallel Fabry-Perot interferometer equation was applied to interpret the results semi-quantitatively. We obtained 2-D interference fringe pattern having 3 orders of maxima from the middle to the edge of the circular apperture, as a result of an initial internal membrane stress. The pattern changed non-linearly as we applied flexural strain from behind the beam up to 50 μm, most probably caused by nonlinear deflection of the membrane (i.e. the membrane did not deflect similarly as the beam beneath it). This phenomena might explain one of the strain-amplifying properties of this biomimetic strain sensing microstructure.
Standard heterodyne interferometer can be used as phase modulation subsystem in a novel interferometer designed to measure the figure of projecting mirrors with 0.1 nm accuracy. This article discusses possible operational principles of the sensor and presents experimental results for fast sampling type sensor prototype.
In this paper, free standing beams and trench structures are successfully fabricated in both n and p-type silicon using macro-porous silicon. Electrochemical etching in hydrofluoric acid has been demonstrated to make 3-D structures in n-type silicon, in a single etch step. A set-up for n-type silicon involves a light source to generate electronic holes in silicon substrate, and the width of the pores can be controlled by the light intensity. However, the use of light makes the set-up more complicated. Using HF, p-type silicon was found only to form random micro-porous structures, and thus not the trenches achieved in n-type silicon. Recently, in order to form a macro pore in p-type silicon, a new etchant has been proposed in which the macro pores. This technique has been used to obtain macro-porous structures in p-type material. This technique has been extended to make free standing structures, as done with n-type material. This work has shown that free standing structures can be fabricated using a single etch step in both n and p-type material. P-type has the advantage that no light source is required and also thermal oxide can be used as a masking layer. However, experiments performed to date have shown that the n-type material process is easier to control and is more flexible, in terms of the structures which can be fabricated. Work is continuing to further improve these processes and this paper will examine the processes for n- and p-type material giving the relative merits.
This paper investigates the effect of gas flow in holes on the squeeze film damping of perforated structures. An infinite perforated plate with circular holes is analyzed with an analytical model. The results show that there is a minimum damping ratio for a certain size ratio. The corresponding hole size can be defined as a critical size. When the hole size is smaller than the critical size, the hole effect dominates. The damping ratio increases drastically with hole size decreasing when the size ratio keeps constant. Some finite two-dimensional structures are analyzed with an equivalent circuit model. Similar results are obtained. The finite two-dimensional structures and some quasi three-dimensional structures are also simulated with ANSYS/FLOTRAN. The results are presented.
When developing new MEMS devices there are several decisions which have to be made. The first of these is which technology to use. Today there are a range of silicon based micromachining technologies available. In addition to backside bulk micromachining and surface micromachining, new processes available include epi-micromachining and deep plasma etching. These are all silicon based technologies. New processes using alternative materials have also been developed and these include processes such as LIGA and SU-8. The decision of which process to use is often based on the equipment available and the type of structure to be fabricated. The second question is often more difficult, and that is whether or not to integrate. The integration of sensor with electronics can present a number of advantages, in particular when dealing with small signals or noisy environments. However, there are many applications where environmental factor ensure that the electronics will not function correctly. It is also important to ensure that the processing used for the MEMS devices does not adversely affect the electronics. MEMS device scan be fabricated as pre- or post-processing modules, or integrated within standard processing. In each of these optics on the compatibility issues such as non-contamination of the processing line and thermal budget have to be considered. Furthermore the potential drop in yield of the total devices has to be weighed up against the advantages. This paper will examine these decision and how the different aspects of technology and applications effect the choices concerning whether or not to integrate.
The small size and possible low cost of micromachined sensors make them attractive for some medical applications. Minimally invasive therapy aims to reduce the damage done to healthy tissue by reaching the affected area through existing pathways through the body. However, information is scarce as direct view or touch is lacking. Small sensors are needed on catheters inside the blood vessels to gather the data such as blood pressure and flow. To this end a combined pressure and flow sensor is fabricated in an epi-poly process that uses a 4 micrometer thick polysilicon membrane grown during epitaxial growth, to form the diaphragm of the pressure sensor and the thermal insulation of the thermal flow sensor. Using RIE etching of holes through the membrane, sacrificial etching and closing of the etch holes by oxide depositions, a closed reference chamber is formed for an absolute pressure sensor. The process is compatible with standard bipolar electronics to enable integration of signal conditioning, multiplexing, etc. Measurements of the two sensors show that fabrication of flow and pressure sensors using epi-micromachining is possible and that the sensors have the required measurement range, but drift necessitates calibration before use.
This paper presents the development of accelerometers using the epitaxial layer as the mechanical structure. In this work epi-poly was chosen for the fabrication of the accelerometer structures. Epi-poly is a polycrystalline material deposited in an epitaxial reactor. This means that the mechanical structures can be deposited in the same step as the epitaxial layer used for the electronics. An extension to the epi-poly process has been used where after sacrificial etching to remove the oxide, anodic etching in HF is used to increase the airgap under the mechanical structures. This has the advantages of reducing vertical sticking and reducing parasitic capacitances. The paper describes the basic epi-poly process and the extension to a double sacrificial etching. Accelerometers have bene fabricated using both techniques and measurements have been made for both static and dynamic accelerations.
Electrochemical etching of silicon is commonly used for sensor fabrication processes. In fluoride-containing solutions porous silicon can be formed while in alkaline solutions silicon can be passivated by passing an anodic current through the silicon. For batch fabrication it is more convenient to have a contactless etch method. Galvanic etching can be used for this purpose. In this paper some general aspects of galvanic etching are considered. Examples of galvanically etched structures are presented.
In the fabrication of fully integrated sensors, which combine sensor and electronics on a single chip, it is essential to ensure that any additional process steps introduced for the sensor do not adversely affect the electronics. Where possible it is desirable to use existing processes or layers to fabricate the sensor. This is not always possible and therefore additional processing steps must be added to that of the electronic circuitry. In the fabrication of MEMS the additional processing is usually one of the forms of micromachining. Many processing steps are not compatible with the electronics and therefore they have to be adjusted which may lay considerable constraints on both materials and processing steps. In this paper the range of options to designer will be discussed.
The accuracy of piezoresistive based silicon sensors is partially determined by the uncertainty in the piezoresistive coefficients. In the production of sensors it is therefore desirable to measure the coefficients for the used technological process to avoid costly calibration of the sensors after production. A practical measurement method consists of bending a three-element rosette. The resulting resistance changes are used to calculate the independent piezoresistive coefficients (pi) 11, (pi) 12, and (pi) 44. A recognized problem of this method are the large uncertainties in the smallest calculated coefficients. These uncertainties arise partially from errors in the resistance measurements. This paper shows that the calculation of the coefficients is very sensitive to the measurement errors. This sensitivity is especially pronounced if there are large differences in magnitude between (pi) 11, (pi) 12, and (pi) 44, such as in p-type silicon. It appears, however, that the error sensitivity can be greatly reduced by optimizing the rosette orientation in the (001) crystal plane. For a p-type silicon rosette the optimum orientation is k*(pi) /2 (k equals 0, 1, 2,...) with respect to the [100] axis. Orientation along these angles may result in a reduction of the uncertainties up to a factor 48.
High stress levels can cause problems for both electronics and micromechanics. It is therefore important to characterize the stress and strain of the different mechanical layers used for processing. In order to predict the properties of a multilayered structure it is not sufficient to know the mechanical properties of each independent layer. The deposited layers may have a considerable effect on the mechanical properties of the underlying layer. This paper discusses the importance of the effect of each layer on underlying layers and on the total structure, using poly-silicon and silicon- nitride as the mechanical layers. The measurements show that at an anneal of 850 degrees Celsius the influence of nitride on the underlying poly is considerable. At an anneal of 1000 degrees Celsius this effect disappears. We also examined the stress in poly-nitride-poly layers in order to find out the influence of annealing in between the deposition of the layers. Measurements show that the temperature of the anneal is the dominant factor.
This paper presents a new technique of micromachining using macro porous silicon. Macro porous silicon is made by electrochemical etching in hydrofluoric acid. The etch rate and the morphology of the etched surface as a function of etch parameters, (current density, applied voltage and HF concentration) are investigated. Optimization of these parameters makes it possible to fabricate a micromechanical structures such as 45 micrometer deep, 3 micrometer wide and 8 micrometer pitch trenches. Furthermore the diameter of the pore is easy to control by adjusting the current density. During the pore formation an increase in the current density leads to an increase in the pore diameter. This does not effect the diameter of existing pores. This connection of the pores under the structure can be achieved. In this way, various kinds of single crystal silicon micromachined structures can be fabricated.
KEYWORDS: Chemical vapor deposition, Low pressure chemical vapor deposition, Surface roughness, Temperature metrology, Plasma enhanced chemical vapor deposition, Oxides, Silicon
The influences of surface characteristics, including adsorptive states led by different chemical treatments and surface roughness, on direct bonding between dissimilar CVD materials were investigated. The bonding procedures were carried out at temperature lower than 400 degrees Celsius. In this temperature range, LPCVD poly-silicon, PECVD oxide, and LPCVD silicon-nitride showed highly process dependent bonding behaviors, i.e., bondable or not bondable to another material under certain experimental conditions. Based on these facts, a selective bonding conception for Si-based CVD material is proposed and applied to fabricate new fluid structures and devices.
This paper presents the development of a pressure sensor based on epi-poly processing technology. The use of epi-poly creates an SOI structure and thus facilitates the use of an oxide etch-stop for the bulk micromachining. This avoids the requirement for the complicated electrochemical techniques. This simplification of the process greatly enhances the batch fabrication capabilities. Furthermore, the epi-poly deposition can be performed at the same time as the standard epitaxy used for the electronics. By defining the oxide layer only where a membrane is required, all other areas of the chip can be used for the fabrication of the read-out electronics. Etching from the back of the wafer, to define the membrane, is performed after completion of the bipolar processing. This etching is performed in KOH or TMAH which has a high etch selectivity over the oxide layer. This paper presents the fabrication and initial measurement results of the epi-poly pressure sensor.
This paper presents the development of a new technique to prevent occurrence of compressive stress in epipoly. The use of an epitaxial reactor to grow polysilicon enables the growth of monocrystalline silicon (for bipolar electronics) and polysilicon on top of oxide (for MEMS) in a single deposition step. However, after bipolar processing the early structures show compressive strain in the epipoly layer, which required careful MEMS design. We have found the cause of this compressive strain to be the oxidation steps in the bipolar process. The occurrence of this strain can be explained by the presence of oxygen in the epipoly. An alternative processing technique, where the epipoly is doped using implantation and shielded from oxygen by a nitride layer during further bipolar processing, yields epipoly layers without compressive strain. The full thermal budget of the bipolar process is used to diffuse and activate the implanted epipoly dopant.
In surface micromachining applications thick polysilicon layers are highly desirable. The low deposition rate of LPCVD poly-silicon severely limits the final thickness of the microstructures. This can be overcome by using an epitaxial reactor to grow polysilicon, in which very high deposition rates can be obtained. However, when a complete electronic process is used to fabricate smart sensors, the epipoly will be under compressive strain. The cause of this unwanted strain has been identified and the problem is solved by capping the epipoly with a nitride layer during further processing. In this way thick low-stress epipoly layers grown at the same time as the monocrystalline silicon epilayer required for the electronics, have been obtained. Several test structures, such as thermally actuated indicators, lateral accelerometers and lateral comb drives have been fabricated using this optimized process.
Micromachining technologies, thin film deposition and bonding technology, along with continued advances in silicon semiconductor circuitry, have resulted in miniature sensing devices whose performance is equal to and sometimes better than their full-sized counterpart. However, despite the obvious benefits and needs for microsystems solutions, the production and marketing thereof have proved to be more difficult than anticipated. The advantages related to the use of conventional IC processes to fabricate the various components of a microsystems have been often acknowledged. The problems resulting from the constrains of IC compatibility requirements have been extensively discussed. New upcoming silicon micromachining technologies and some interesting developments in the existing ones, seem to help meet these requirements. In this paper, some very promising techniques, such as epi-micromachining, laser micromachining, deep reactive ion etching of silicon and low temperature Si to Si bonding are reviewed. Examples of integrated micromachined sensors and devices which benefit from these new developments will be presented to underline the promise and expectation of these technologies. Further, developments made in CAD systems for microelectromechanical systems and the availability of MEMS technology in a multi- user environment are mentioned as they can have a positive impact on the integration of micromachined sensors.
The basic aim of this work is to obtain optical detectors with a spectral response programmable by design using the combined response of polysilicon and monocrystalline silicon photodiodes. Such an approach is needed in order to obtain a color sensor with improved flexibility and control of its characteristic parameters. In order to achieve this aim, the potential of different multilayer thin film light detectors has been evaluated. The results show that polysilicon diodes can be realized and used as light detectors and a simple test structure has been fabricated in order to demonstrate the possiblity of implementing thin film color detecting structures.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.