Microelectromechanical Systems Or Mems

Print   

02 Nov 2017

Disclaimer:
This essay has been written and submitted by students and is not an example of our work. Please click this link to view samples of our professional work witten by our professional essay writers. Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of EssayCompany.

CHAPTER 2

MicroElectroMechanical Systems, or MEMS, represent an extraordinary technology that promises to transform whole industries and drive the next technological revolution . MEMS is an acronym that originated in the United States, and also known as Micromachines which often used in Japan and more broadly as Microsystem Technology (MST) in Europe . In most general form, MEMS can be defined as miniaturized mechanical and electro-mechanical elements (devices and structures) that are made using the techniques of microfabrication. MEMS are made up of components between 1 to 100 micrometres in size (0.001 to 0.1mm) while MEMS devices usually range in size from 20µm to a millimetre . These devices or systems have the ability to sense, control, and actuate on the micro-scale, and generate effects on the macro scale . The first MEMS products were developed in the 1960’s, when accurate hydraulic pressure sensors were needed for aircraft . In the early 90’s, MEMS accelerometers for car airbags were developed as a less expensive, more reliable, and more accurate replacement for a conventional crash sensor . Today’s commercial MEMS – also known as Micro System Technologies (MST), Micro Machines (MM), or M3 (MST, MEMS & MM) have more manufacturable, reliable and accurate . Other commercialisation of MEMS devices are listed in Table 2.1:

Table 2.1

Commercialisation of selected MEMS devices

Product

Discovery

Evolution

Cost Reduction/ Application Expansion

Full Commercialisation

Pressure sensors

1954-1960

1960-1975

1975-1990

1990-present

Accelerometers

1974-1985

1985-1990

1990-1998

1998

Gas Sensors

1986-1994

1994-1998

1998-2005

2005

Valves

1980-1988

1988-1996

1996-2002

2002

Nozzles

1972-1984

1984-1990

1990-1998

1998

Photonics/displays

1980-1986

1986-1998

1998-2004

2004

Bio/Chemical sensors

1980-1994

1994-1999

1999-2004

2004

RF switches

1994-1998

1998-2001

2001-2005

2005

Rate(rotation) sensors

1982-1990

1990-1996

1996-2002

2002

Micro relays

1977-1982

1993-1998

1998-2006

2006

The major categories of MEMS devices are microsensors and microactuators. Microsensors and microactuators are appropriately categorized as "transducers", which are defined as devices that convert energy from one form to another . A sensor is a device that measures information from a surrounding environment and provides an electrical output signal in response to the parameter it measured . Sensors have been categorized in terms of the type of energy domains but MEMS devices generally overlap several domains and some do not even belong in any category . These energy domains include: 1) mechanical (force, pressure, velocity, acceleration, position), thermal (temperature, entropy, heat, heat flow), radiant (electromagnetic wave intensity, phase, wavelength, polarization reflectance, refractive index, transmittance), magnetic (field intensity, flux density, magnetic moment, permeability), chemical (concentration, composition, reaction rate) and electrical (voltage, current, charge, resistance, capacitance, polarization). An actuator is a device that converts an electrical signal into an action . It can create a force to manipulate itself, other mechanical devices, or the surrounding environment to perform some useful function.

Many applications of MEMS have requirements on material basis, geometry, aspect ratio, dimensions, shape, accuracy of microstructure and number of parts that cannot fulfilled easily by mainstream silicon-based micromachining technologies . Manufacturing a successful MEMS device need a well understanding of basic physics and operating principles including scaling laws at both a macro and micro level. Moreover, the increased surface area (S) to volume (V) ratios at microscales has both considerable advantages and disadvantages in MEMS devices. For example, microsensors such as micro-gyroscopes tend to have no moving parts that operate under the conditions of sliding friction . On the other hand, microactuators such as micromirrors, microengines and gears represent an entirely different class of devices that requires the characterization of their electrical and mechanical properties . Some of these micro level issues include:

Friction is greater than inertia. Capillary, electrostatic and atomic forces as well as stiction at a micro-level can be significant.

Heat dissipation is greater than heat storage and consequently thermal transport properties could be a problem or, conversely, a great benefit.

Fluidic or mass transport properties are extremely important. Tiny flow spaces are prone to blockages but can conversely regulate fluid movement.

Material properties (Young’s modulus, Poisson’s ratio, grain structure) and mechanical theory (residual stress, wear and fatigue etc.) may be size dependent.

Integration with on-chip circuitry is complex and device/domain specific. Lab-on-a-chip systems components may not scale down comparably.

Miniature device packaging and testing is not straightforward. Certain MEMS sensors require environmental access as well as protection from other external influences. Testing is not rapid and is expensive in comparison with conventional IC devices.

Cost – for the success of a MEMS device, it needs to leverage its IC batch fabrication resources and be mass-produced. Hence mass-market drivers must be found to generate the high volume production.

Thus, new and improved techniques of fabrication are desired based on the specific applications and designs. High aspect ratio microstructures and the demand of nanopowder as one of new nanotechnology in microfilling are reviews in this chapter. Also, several methods of fabricating MEMS are discussed with their advantages and limitations.

2.2 HIGH ASPECT RATIO MICRSTRUCTURES (HARMS)

High aspect ratio microstructure (HARMS) is today of very importance in microelectronics and MEMS technology. They are essential when deep structures or thick layer are required in addition maintaining the microscopic lateral dimensions of the structures. HARMS are typically hundreds of micrometres in height, with width ranging from a few micrometres to ten micrometres, and they can be manufactured from a variety of materials such as metals, polymers and ceramics . The increasing demand for products with higher aspect ratio and larger spatial resolution are created to meet consumers’ needs in making micro parts, such as the dies for making an IC lead frame, the miniature thin film used in the semiconductor industry and medical devices and microelectronic medical implants in the biotech industry . Different implementations of electrostatic transduction mechanisms, including in-plane, out of plane and torsional oscillations with enough force requires high aspect ratio microstructures . HARMS are often desirable in MEMS technology as they offer a number of benefits including:

Increase mechanical robustness , reduce out-of plane motion of flexures or sensitivity, and increase surface area for functionalisation or capacitive actuation and detection

Low driving voltage and increase structural rigidity, torque, actuation force and angular displacement in actuator systems

Higher sensitivity in sensor applications for virtue of large mass

Larger magnetic forces for magnetic MEMS due to the larger volume

Applications for deep microstructures exist in many sectors of R&D activity and as industrial products, worldwide. The presence of thick, deep and highly precise microstructures with high aspect ratios is a requirement for a number of micromechanical, optical or packaging applications, as well as in other areas. Large surface areas (tens of square centimeters to square meters) covered with HARM’s have potential applications in a variety of fields, including heat transfer, fluid mechanics, composite materials, bearings, and catalytic systems . Large capacitance and compliance in wafer plane can be attained with rigidity in normal direction is another reason why high aspect ratio microstructures are employed in many MEMS devices.

In microfluidic and biomedical applications, the micro/nano pillar array has been used ranging from microfluidic chips for separation of long DNA molecules and adhesion substrates for cell growth , to the investigation of hydrophilicity and hydrophobicity on nanostructured surfaces . ) found that high aspect ratio microstructures can be used to either encourage or discourage cell attachment, depending on the structure; to restrict or direct cell morphology, though not necessarily in an intuitive manner; and to generate three dimensional cell cultures systems without the need for mixing or other bioreactor enhancements.

Another example is HARMS with vertical sidewalls can increase the output force of a micro-actuator, and hence be used as a mold for replica multichannel polymer chips . The high aspect ratios possible with the process are helpful in increasing the force output of the comb actuator (or the sensitivity of a comb variable capacitor) by increasing the thickness of the comb array while allowing small lateral features for high overall capacitance. Devices for high frequencies based on diffraction (e.g. Fresnel zone plates and diffraction gratings) and on refraction (e.g. wave plates and retarders) require both nanoscale dimensions and high aspect ratios for optimal operation . Inductors with taller via structures (higher aspect ratio) had better radio frequency (RF) characteristics than those of lower via structures, which are highly desirable in many communication applications .

2.2.1 NEED OF HIGH ASPECT RATIO MICRO HOLES

It is well known that the micro structural characteristics of micro holes in electronic devices determine its mechanical performances . By increasing the aspect ratio of micro holes, the capability of the micro devices to function efficiently will be increased. Micro holes are widely used in many fields, for example, as gas detection in the environment monitoring. Micro hole and strip plate (MHSP) detector used as a micro pattern detector for the detection of thermal and epithermal neutrons . Whilst micro holes array structures ensures that gases could further diffuse into the other columns quickly in gas sensor which increase the sensor’s sensitivity with quick responses . For chemical analysis in biochemistry, thin silicon membranes with micro-via holes are used as filters in miniature chemical reactors . The technology of filters of this type utilizes net anisotropic etching, connected with the technique of etch stop diffusion or wet double-side etching. In micro-capillary array injector, the micro holes are used to attract and isolate cells towards the active parts by means of aspiration and to inject gene into cells . This can improve the efficiency of DNA insertion into cells in gene therapy.

The micro-holes have become the basic structure in the micro-fluidic devices . Micro-holes are able to transport liquid without any additional power input . The self-pumping capability of micro-holes is due to the pressure and capillary force. In common, micro-holes with smaller effective diameter of holes usually have larger capillary . However, the friction between the structure of holes and liquid will increase rapidly as the effective diameter of holes shrinks . The smaller characteristic size of micro-holes will not always have the better ability in transporting liquid. Besides, the gravitational field also affects the behaviour of micro-holes. The well-designed micro-holes devices must overcome the effect of gravitation and can be operated under any inclination angles from horizontal.

For micro-electromechanical systems (MEMS) devices, if two nearby planar structures have relative motion in their normal direction, the thin layer of air between them will produce squeeze-film effects, which can alter the dynamic response of MEMS devices by adding stiffness and damping to the system. To reduce squeeze-film effects, micro-holes are widely used in MEMS devices . Actually, micro-holes have several functions as mentioned in : (1) to reduce squeeze-film effects; (2) to control the dynamic response of MEMS devices; and (3) to enhance the etching rate of sacrificial layers in the micro-fabrication process. In drug delivery system, the drug release rate can be controlled at a stable level in the whole release process with proper micro holes in bonding membrane . has found that nozzle holes with high aspect ratio are suitable for higher resolution inkjet print head. Micro holes with high aspect ratio are also used to create coolant flow paths through the substrate in a cold plate of electronic devices (heat transfer engineering) .

Micro holes/ micro vias technology has also becoming more important within the packaging and the semiconductor devices since it allows a high density interconnection (HDI) technology, an excellent electrical performance and can configure multilayer structures and assemble components on a microdevice . Therefore, more functions can be integrated in a compact volume due to shorter wire length that allows the use of higher frequencies , less power consumption and smaller device.

2.3 METAL NANOPOWDER

2.3.1 POWDER METALLURGY

Powder metallurgy process using nanopowders is expected to be applicable to fabricate tiny and complex structures. Powder metallurgy provides metal components with good tolerances in a wide variety of metals and alloys cost effectively. The net or near-net shape production of parts with tailored properties like high strength, corrosion resistance, wear resistance and also possible at high temperatures . The transfer of this technique to MEMS technology has broadened the range of materials and improves the properties for components and tooling.

2.3.2 METAL NANOPOWDER

So far MEMS based on polymeric and silicon materials have successfully entered the market. However, metal has more practical industrial applications in many aspects which they can provide better material properties than those of polymers and silicon. It is believed that metal-based microsystems enjoy some inherent advantages over silicon-based micro devices, especially in the area of harsh environment-compatible devices. Examples include the devices subjected to high temperatures (micro heat exchangers) , placed in contact with corrosive chemicals (micro chemical reactors) , and under high mechanical stresses . It is believed that one important factor to commercial realization of metal-based microdevices is efficient and economical fabrication.

2.4 MICROFABRICATION

Most of the metal-based active or passive microdevices envisioned require the construction of microscale structures with larger ratios of height to lateral dimensions as compared to what is typical for Si-based MEMS . Despite the considerable practical implications if metal-based microdevices can be manufactured economically, current techniques for making metallic, high-aspect-ratio, microscale structures (HARMS) are much less well developed as compared to the mature, Si-based, integrated-circuit, fabrication technologies.

Microfabrication is an enabling technique in the development of technologies that allow the fabrication of micro and nano scale components with high precision and reproducibility. Moreover, microfabrication is a technology development process to manufacture small three dimensional structures. There are some differences between microfabrication and IC fabrication such as :

Aspect ratio in microfabrication is generally much greater than IC fabrication.

The device sizes in microfabrication are often much larger than in IC processing.

The structure produces in microfabrication often include cantilevers and bridges, whilst other shapes requiring gaps between layers.

Many efforts from academies institutions and industrial companies have been put in the exploration and study of the microfabrication technologies. Hence, quite a few technologies have been developed and put into practice successfully such as photolithography, Excimer laser micromachining, silicon bulk and surface micromachining, LIGA technique and hot embossing . Therefore, increasing interest in microfabrication have been driving the variety of applications of MEMS. In the following, the major MEMS microfabrication methods will be discussed.

2.4.1 LIGA

LIGA comes from German names (Lithographie, Galvanoformung, Abformung) which means lithography, electroforming and molding methods which are used to produce microscale structures . LIGA represents an important approach toward producing metal-based HARMS. In X-ray LIGA, an X-ray sensitive polymer photoresist typically poly methyl methacrylate (PMMA) is bonded to an electrically conductive substrate. PMMA is exposed to parallel beams of high-energy X-rays from synchrotron radiation source through a mask which covered with a strong X-ray absorbing material. Then, followed by chemical dissolution of the exposed portion of photoresist and filled by the electrodeposition of metal. After recess filling, the remaining resist is chemically stripped away to produce a metal mold insert which can be used to produce parts in polymers or ceramics through injection molding . Fig 2.1 shows the flow steps of LIGA process:

Fig 2.1: The flow steps of LIGA process

LIGA process is well-known technique to create HARMS with aspect ratios up to 100:1 with a few millimetres in height . However, LIGA process is limited to materials that can be electroplate such as gold, copper, nickel and a few nickel alloy . LIGA requires a synchrotron facility and X-ray radiation source to fabricate HARMS. Thus, the operational cost highly increased.

2.4.2 LASER MICROMACHINING

Laser micromachining is the material removal process accomplished through laser and target material interactions . Laser machining processes transport photon energy into the target material in the form of thermal energy or photochemical energy; and they remove material by melting and blow away, or by direct vaporization/ablation . Lasers are effective material processing tools offering distinct advantages, including choice of wavelength and pulse width to match the target material properties, and one step direct and locally confined structural modification . Pulsed lasers with wavelengths in different regions of the electromagnetic spectrum have been used for material removal in a non-contact fashion, including excimer lasers operating in the ultraviolet (UV) radiation and femtosecond pulsed Nd:YAG laser operating in the near infrared (IR) . Molecular dynamics (MD) simulations have been used to study the non-thermal and thermal mechanisms of material removal by interaction with a fast laser pulse .

Recent studies include the use of nanosecond laser for nanoscale ablation of thin metal films and nanoscale patterning of Au nanoparticle films . For microscale machining of metals, the material removal rates are observed to be on the order of 1μm per pulse, with laser beam diameter on the order of 10μm . Serial scanning of the laser beam is therefore necessary for forming extended microscale features in silicon and metals with depths on the order of a few hundred microns. At typical pulse repetition rates of 1 kHz or less, creation of microscale structures covering practical dimensions will be time consuming. Laser beam induced material removal creates elevated surface roughness and high aspect ratio hole drilling on as-machined metal structures . Additional surface treatments may therefore by required depending on the nature of the intended application. The existence of an upper limit in the maximum thickness of metal that can be laser machined is also highly depending on the incident laser intensity and average laser power .

2.4.3 ELECTRODEPOSITION

Electrodeposition or electrochemical deposition of metals and alloys onto metallic substrates plays an important role in many modern technologies. In MEMS, electrodeposition can provide an ideal approach to fill topologically complex structures because electrodeposition fills the structure from the bottom up. For example, Cadmium selenide (CdSe) is electrodeposited through a filter membranes and filling the pores completely from the bottom to the top in high aspect ratio microstructures . Today, electrodeposition is a very important feature from the viewpoint of environmental protection and material conservation due to the process benefits including :

Low process temperature and equipment cost

A negligible waste of material

Control capability of composition and morphology

Ability to deposited films on a complex surface

Instead of those benefits, there are a few limitations occurring such as uniform current distribution and ionic solution are needed in this technique. Moreover, a conductive surface of silicon as a seed layer is required to avoid poor quality of plating . Electrodepositions are restricted in their choice of electrolytes which allow microscale structures with high aspect ratios to be deposited. At present, this limits the choice of materials to Au, Cu, Ni, and a few Ni alloys . In using such methods for production, one has also to contend with the slow speed of electrodeposition, sometimes led to void formation and the electroplated layers tend to separate from microstructure during the process .

2.4.4 WET CHEMICAL ETCHING

Wet-chemical etching is a technique that utilizes liquid chemicals to remove material. In this way a thin/thick film can be patterned or completely removed. The technique involves immersion of a substrate in a pure or mixture of chemicals for a given amount of time. The time required is dependent on the composition and thickness of the layer to be etched, as well as the etchant and temperature to be used . Typically the strong anisotropic etching solutions are potassium hydroxide (KOH:H2O), ethylene diamine and pyrocatecol (EDP) and tetramethyl ammonium hydroxide (TMAH) whilst isotropy in HF:HNO3:H2O solution .

Anisotropic wet chemical etching remains the most widely used processing technique in silicon technology. Used in combination with a multitude of other processes, it has a wide range of applications in MEMS. Its wide presence is not only due to its ease of use and low cost, but also due to the fact that it provides fairly smooth surfaces with no physical damage to the bulk . However, several disadvantages discovered, for example the slow rate of etching and dependent to the temperature around 100°C in the process .

During recent years, much effort has been dedicated to the characterization and understanding of the surface morphology during anisotropic etching, both experimentally and theoretically. Specific studies of the most frequent surface in homogeneities, such as pyramidal hillocks (fig. 2.2(a)) and shallow round pits (fig. 2.2(b)) on Si(100), nosed zigzag structures (fig. 2.2(c)) on (vicinal) Si(110) and triangular pits (fig. 2.2(d)) on Si(111), have provided important insights into the problem .

Fig. 2.2: Most frequent (simulated) morphological features during anisotropic etching (a) Pyramidal hillocks on (100). (b) Round shallow pits on (100). (c) Nosed zigzags on vicinal (110). (d) Triangular pits on (111). (e) Polygonal steps on terraced vicinal (111). (f) Straight steps on terraced vicinal (111) .

2.4.4.1 TMAH ETCHING

In response to the requirement for damage less and highly directional etching, a conventional orientation-dependent anisotropic etchant of tetra methyl ammonium hydroxide, TMAH, is utilized to realize high aspect ratio microholes structure for MEMS. TMAH has strong alkalinity with molecular formula of (CH3)4N+OH-. It has been widely used as a photoresist developer in LSI industry resulting in stable supply with guaranteed quality and reasonable price . There is no notable problem concerning hazardous impurities. As the etch rate of TMAH for (111) Si plane relative to those of other planes is less than 1/100 in appropriate etching conditions, V-shaped grooves and straight grooves are formed on (100) silicon surface and (110), revealing etched surfaces of (111).

TMAH is commonly used in the microelectronic industry for about 20 years. TMAH based solutions provide smooth surfaces with concentrations exceeding 22-25wt% of TMAH in water solution . TMAH is relatively high anisotropy and an interesting high silicon/ silicon dioxide etching selectivity compared to KOH. Silicon dioxide (SiO2) and silicon nitride (Si3N4), which are used as etch masks, are etched at a significantly lower rate than Si. TMAH was studied decades ago and only at high temperature (>60C) is known to ensure a higher etching rate, thus limits the etching time . TMAH solutions are non toxic and can be handled easily. Since no alkali ions are present, TMAH presents no danger to electrical circuits . Therefore, TMAH is increasingly recognized as a viable alternative for MEMS process applications.

2.4.5 MICROMILLING

Micromilling is a direct operation to manufacture net-shaped small parts offering alternative to other micromanufacturing processes. It is a flexible method of fabricating three-dimensional (3-D) features including micro molds/dies and fully functional metal devices specifically with recently developed miniature machine tools . Increasing popularity of micro-milling has sparked the interest of researchers to study the micro-milling processes to improve the quality, reliability and productivity. Micromilling tools are available commercially in diameters larger than 50μm. Custom-built tools as small as 20μm have been fabricated by the focused ion beam (FIB) machining process . Controlled machining over small feature sizes is made possible through the use of computer numerically controlled (CNC) ultraprecision machines. CNC ultraprecision machines have been used to create microscale features while maintaining submicron tolerance over a span of millimeters. The bottom of machined trenches has a typical root-mean-square (rms) roughness of ~100nm .

Several issues exist regarding micromilling as a mass fabrication technique. One issue is tool wear, which has a strong effect on the quality of the machined parts . Progressive wearing of the cutting edge influences the geometry and the surface roughness of the machined parts. High temperature and high pressure during cutting promote the formation of cutting edge wear or build-up, and cause eventual tool breakage. Burr occurrence during the micromilling is another issue, and is difficult to avoid . It may be minimized by select smaller milling tools, at the expense of increased machining time and tooling costs. The sharpness of interior corners in machined parts is limited by the diameter of the milling tool, and is therefore limited to tens of microns . For these reasons, micromilling is more suitable for machining metal parts with small and medium lot sizes, with minimum dimensions about tens of microns.

2.5 FOCUSED ION BEAM (FIB) MICROMILLING

Focused ion beam (FIB) technique uses a focused beam of ions to scan the surface of a specimen, analogous to the way electrons are used in a scanning electron microscope (SEM). The use of a gallium ion is a well developed technology in the semiconductor industry for doing modifications on integrated circuits . The FIB system is able to cross-section devices and provides a high resolution image of the internal structure of a semiconductor device (Kirk et al., 1988). Micromilling in silicon is especially useful for micro-electromechanical systems (MEMS), micro-optomechanics systems, microsensors and microactuators. Material modifications and patterning are to be controlled at the address technical level, thus the rapid migration of the critical dimensions necessitates a more detailed knowledge of the fundamental mechanisms involved in milling process to enable full use of the opportunities offered by the focused ion beam technique .

There has been considerable interest in FIB technologies (e.g. FIB direct milling, FIB projection, FIB-assisted deposition and others) to produce complex shaped silicon structures for NEMS/MEMS application. Among FIB technologies, the FIB direct milling process, also known as a FIB direct writing, has drawn much interest because of the high flexibility in the working shapes, the dimensions ( a scale ranging from a few tens of nanometers to hundreds of micrometers), and the material selectivity . Because of the very short wavelength and very large density, the FIB has the ability for direct milling of structures that have features sizes at or below 1µm. As a result, the FIB has recently become a popular candidate in making high-quality microdevices of high precision microstructures .

2.5.1 ADVANTAGES AND LIMITATIONS OF FIB MICROMILLING

Several advantages of FIB milling has been observed by recent findings, including:

Has high flexibility in the working shapes, the dimensions and the material selectivity .

The pattern desired can be milled directly on metal, silicon, glass, carbon substrate without any pattern exchange and electroplating. However, FIB milling is more suitable for hard brittle materials than metals in terms of the achieved surface quality .

Rapid dry etching as compared to the electron beam, as it can remove and deposit the target materials in nano-scale using focused gallium ion beam instead of electron beam

The high resolution image obtained by detecting the secondary ions that from the sample are forced out .

Can reduce various hassles and defects caused by the masks and resists in pattern transfer and directly write a very narrow line because ions are much heavier and the lateral scattering of FIB is relatively low, resulting in striking only the intended regions .

8. FIB can work at room temperature and has no influence upon the performance of substrate material

9. The density of patterns can be controlled and produce a hole geometry of high aspect ratio up to 15

A few limitations also occurred during FIB milling process, such as mass production is too slow because of the low milling rate. Thus, milling rate of FIB milling has to be improved to increase the throughput and the ability to be used in production . Imaging process in FIB requires that the ion beam continuously remove material from sample, which can damage the crystal structure of a device especially in nanometer scale . A continuous electrical path is required from the area being limited to the measuring instrument as the end point is detected by monitoring the change of the current signal during ion milling .

During analysis, noise from the side walls of the milled structure prevent from measuring depth accurately using atomic force microscopy (AFM) and profilometer. Both AFM and profilometer cannot access the bottom of high aspect ratio structure . Other is the formation of the outer amorphous layer induced by knock-on damage in the case of a crystalline specimen . The size of the structures that can be obtained is also limited by the available processing time. Dimensions up to some tens of micrometers are easily feasible, but above 100µm, the typical processing time become unacceptably high .

2.5.2 FIB MICROMILLING USING FIB-SEM SYSTEM

Formation of the outer amorphous layer induced by knock-on damage in the case of a crystalline specimen is one of drawbacks occurs during FIB milling . The imaging process requires ion beam with continuously remove material from sample can cause damage crystal structure . Today, a well designed FIB-SEM system is able to integrate the ion beam’s milling and deposition capabilities with the electron beams high resolution, non-destructive imaging, without compromise to either column’s performance . The FIB-SEM system is a combination of a focused ion beam, an electron beam and secondary ion and/or secondary electron detectors on the same platform . Gallium (Ga+) ions at low beam currents is used for imaging and high beam currents are used for site specific in situ sputtering or milling. By combining these functions a significant amount of time is saved because there is no need to transfer the sample from one system to another.

Furthermore, FIB-SEM system allows precise monitoring of FIB operation through the SEM by using the slice-and-view technique . The signals from the sputtered secondary ions or secondary electrons are collected to form an image . The two beams are co-focused at the coincidence point, typically with a 5mm working distance (WD), which is the optimized position for the majority of operations taking place within the machine. The stage can be controlled to tilt, allowing changes in the sample beam orientation. For best performance, the ion beam is tilted 45-54° from the electron beam, allowing SEM imaging and FIB sample modification without having to move the sample . This purpose will help the operator to decide when to end the milling process.

This project used a Zeiss Auriga FIB-SEM crossbeam system. Fig.2.3 shows a photograph of the FIB-SEM crossbeam workstation, consists of the Scanning Electron Microscopy (SEM) column, the Focused Ion Beam (FIB) column with ion source of liquid gallium (Ga+), detectors, Energy Dispersion X-Ray Spectrometer (EDS) facilities and Gas Injection System (GIS) functions. The adjustment of focus, stigmators, contrast and/or brightness can be made via the operation panel. The sample stage is driven by a 5 axis motorized and computer controlled stage (e.g., X, Y, Z, T (tilt), and R (rotate)).

Fig.2.3: FIB-SEM System (Auriga Zeiss FIB-SEM Crossbeam Workstation)

The main features of the equipment are described below:

1. SEM Column and FIB Column

The scanning electron microscopy (SEM) features Schottky field emission Gemini electron column operating between 100V and 30kV, capable of resolutions of 1.0nm at 15kV and 1.9nm at 1kV. The SEM column is coupled with an Orsay Physics "Cobra" Ga+ ion focused ion beam (FIB). The FIB column operates between 1kV and 30kV with a range of ion beam currents between 1pA and 20nA. It also has the capability of imaging to 2.5nm resolution. 

2. GEMINI FE-SEM Column

The GEMINI FE-SEM column has special in-lens Energy-selected Backscatter (EsB) detector generates images with superior material contrast.

3. Vacuum chamber

This system has a newly designed vacuum chamber, which includes a total of 15 ports for full analytical flexibility. An unrivalled charge compensation system enables the local application of an inert gas flush. In this way, charge build-up on non-conductive samples is neutralized and detection of secondary electrons(SE) as well as backscattered electrons(BSE) become feasible.

4. Gas Injection System (GIS)

Gas injection system(GIS) is fitted, which enables the deposition of carbon, platinum and silicon oxide (SiO2). Additionally, GIS also capable for selective etching of dielectrics with XeF2

5. Omniprobe sample manipulator

This is an insert-able probe that is intended for sample lift-out (such as for TEM sample preparation) but also can be used for in-situ circuit testing and sample manipulation.

6. Leica Cryogenic Stage

A Leica Cryogenic stage that will allow the operation to FIB mill a specimen at a cryogenic temperature. This is particularly useful for the milling of biological specimens, gels, or low Tg polymers.

7. Energy Dispensive X-ray Spectrometer(EDS)

EDS that can be used to analyze the atomic composition of a sample.

8. Electron Backscatter Detector (EBSD)

EBSD that can be used to characterize the crystalline elements of a material as well measure their grain size and orientation.

9. Patterning software

This system also provides independent patterning software, which allows the user to mill complex shapes.

2.6 PARAMETERS OF FIB MILLING

The FIB milling technology primarily aims to obtain accurate milled pattern shapes by removing a required amount of material from a scanned area . The main operation parameters of the FIB should be control with the proper beam size, shape, current and acceleration voltage as well as secondary parameters of dwell time, pattern size, scan mode, and pixel spacing . The combined effects of physical sputtering, material re-deposition and pattern geometry are the goal of the milling process by control of parameters . According to various parameters, the efforts to attain the precision high aspect ratio micro holes milling will be reviewed in these following subsections; 1. Sputter yield, 2.Acceleration voltage, 3. Beam current, 4. Dwell time, 5. Scan mode, 6.Pixel spacing, 7.Pattern size, and 8. Milling mode

2.6.1 SPUTTER YIELD

The physical sputtering is the major mechanism for material removal, and its efficiency is normally represented by the sputter yield, defined as the number of atoms ejected from the target surface per incident ion at given ion beam parameters (atoms/ion) . Sputter yield is a statistical property of the material and a measure of the efficiency of material removal . The yield is normally in the range of 1-50 atoms per ion and is a function of many variables, including masses of ions and target atoms, ion energy, direction of incidence to the surface of the target, target temperature and ion flux . Initially, the sputter yield increases as the ion energy increases, but the yield starts to decrease as the energy is increased past the level where the ions can penetrate deep into the substrate .

Typical sputter yield varies for various materials, incident angle and energy . However, these number cannot be used directly to calculate the etch rate, because, depending on the scanning style, re-deposition occurs, which drastically reduces the effective etch rate. The sputtering yield is roughly increases with 1/cos(), with the angle between the surface normal and the ion beam direction . The changes in the sputtering yield can be accounted for by changing the dwell time (the time the ion beam remains fixed on one pixel position) at the pixels during the milling . The sputtering yields of various materials for 30keV Ga+ ions are shown in Table 2.2.

Table 2.2: The sputtering yield of various materials for 30keV Ga+ ions

Material

Range (nm)

Sputter Yield (atom/ion)

Silicon

27

2.6

Aluminium

24

4.4

Copper

10

11.0

Silver

11

14.0

.

The sputter yield cannot be directly used to determine the material removal rate in milling because there are some factors that cause deviations from the ideal case. These include the shape of the ion beam and the re-deposition of sputtered material . A high sputtering yield leads to the increase of the milling depth and it produces more material removed from the root of the sidewall that was accumulated along the ion sputtering direction . When the depth of the hole exceeds the crater width (the aspect ratio is more than 1), re-deposition occurs and the milling depth becomes less than proportional to the dose, resulting in lower sputter yield. However, the aspect ratio can be improved further by selecting the substrate material with higher sputter yield .

2.6.2 ACCELERATION VOLTAGE

Acceleration voltage (ion energy) is the source of Gallium (Ga+) ion held at a positive potential relative to ground . When the ions are extracted from the source, it is the force exerted by this difference in electrical potential that causes them to accelerate through the column to reach ground potential. The higher the acceleration voltage, the faster the ions are travelling as they exit the column and the greater the energy they impart to the specimen . Increasing the acceleration voltage appears advantageous because it does not influence the probe current. With ion energy of more than 30keV, the probe size decrease to 1nm but the milling time increases for a feature of fixed area .

Acceleration voltage of 30keV has been used for most experiments in order to obtain the highest possible lateral resolution together with the high sputtering yield . At this stage of interaction, implantation or doping can take place in which the ions become trapped in the substrate as their energy is expanded. As a result, the proper ion energy for sputtering is between 10 and 100keV for most of the ion species used for milling . found that FIB milling with Ga+ ion at an energy of 30keV produced amorphisation damage of ~28nm thick along a silicon which have low sputtering yield, and up to 20wt% Ga+ ions may be present within the damage region. However, the use of reduced acceleration voltage is shown to reduce the damage from higher energy ions .

2.6.3 BEAM CURRENT

Beam current is another factor for the aspect ratio. Different beam current determines the different beam spot size. Normally, it is too difficult for the operator to know the real spot size in practice. Therefore, usually the beam current is used as an important process parameter to represent the influence of the spot size . Beam current defines the spot size of the beam which in turn defines resolution and feature size . Table 2.3 shows the different beam currents and corresponding milling spot sizes of FIB-SEM Auriga Crossbeam.

Table 2.3: Different beam currents and corresponding milling spot sizes (at 30keV)

Beam current (pA)

Milling spot size (nm)

1

3

2

7

5

9

10

13

20

13

50

22

120

25

240

40

600

59

1000

200

4000

300

16000

900

20000

1100

In general, the higher the primary beam current, the faster material is sputtered from the surface. Therefore, if only high magnification imaging is desired, a low current beam must be used. For milling applications, high current beam operation is used to sputter or remove material from the surface, whilst a lower beam current is used for fine polishing. The sputtering rate can be easily and accurately controlled by altering the beam current or using smaller spot sizes . A lower beam current can also produce channels with a finer quality, but the number of repetitive passes required to mill the channels is much higher, which can significantly reduce the productivity .

2.6.4 DWELL TIME

Dwell time is a period of time the beam stays at a particular position, and also defines as how long the ion beam stops at each pixel (td) . Dwell time needs to be optimised for a particular material and a particular application. The change of the dwell time generally influences the broadening effect and sputtering depth in silicon milling . During line by line scanning, the ion beam pauses on a particular point (dwell point or pixel point) for a certain time (dwell time) and then moves to the next pixel point . With fast scanning, the dwell time at any pixel point is short, during which time the ion beam converts any gas on the surface to a deposited product. According to FIB manual (1996), long dwell time is suitable for small structures such as vias. It may be possible to achieve a higher enhancement with dwell time of 10µm or even higher . The ion intensity outside the core region of the FIB is sufficiently high to produce a sizeable amount of sputtering at longer dwell times . also conclude that the aspect ratio can be improved further by increasing the dwell time.

However, has found that, although the entire feature sizes increase with the dwell time varying from 5 to 50ms, the increase rates of the sizes are gradually reduced as the dwell time increases. This may indicate that a large amount of sputtered materials is deposited into not only the ridge (outside the channel) but also inside the channel since the mouth width and the milling depth should increase proportionally with the dwell time without re-deposition. It is also often reported that using slow single scans (long dwell time) leads to the formation of structures with inclined bottoms . On the other hand, there was significant influence between parameters with low dwell time, large digitally controlled scan size, and above 1:1 aspect ratio for the FIB milling process . studied the correlation between the absorbed current and the secondary electron as a function of the FIB dwell time and found that the detection accuracy can be significantly improved by reducing the dwell time

2.6.5 SCAN MODE

There are two types of scanning modes, raster and serpentine scans, which are normally used to drive the FIB movement . They are performed in a sequential manner, one scan line after another. In raster scanning, beam returns to the initial point of the next scanning line after finishing any line (retracing), or in simple words, the scan moves in the same direction. Since the raster scan is the most commonly used scan procedure in computer graphics, each series of the horizontal pixels is also called a raster line or scan line . In raster scan, sputtered materials are continuously re-deposited into the region milled earlier and the re-deposited materials accumulate more on the regions milled earlier in the pass, resulting in the inclined bottom of milled surface. However, the raster scan is an effective process to fabricate V-shaped channels or cavities with inclined bottom surface .

In serpentine scanning mode, the beam moves in opposite directions from one row to the next and eliminates the long retrace between the last pixel of one row and the first pixel of the next . The re-deposition is proportionally reduced in each pass and a portion of the re-deposition from the earlier passes is removed by the subsequent milling. Thus, increasing the number of passes can reduce the effects of re-deposition. A sharp defined region with a uniform depth and smooth surface can be milled. The serpentine scan procedure is indispensable for making cavities with vertical sidewalls and flat channel bottom or making any high aspect ratio or curved structure . For best performance, a raster or serpentine scan is used in control of the first beam pass to trace all the regions to be milled. Then, a repetitive pass is applied following the pass milled previously.

2.6.6 PIXEL SPACING

Pixel spacing is a distance between the centres of two adjacent pixels (Ps) . In order to mill a smooth profile with a constant rate of material removal or milling rate, the ion intensity rate or ion flux with respect to the scanning direction has to be uniform or unwavering . To achieve this, the pixel spacing must be small enough to allow a proper overlap between adjacent pixels so that a smooth profile can be milled . In addition, to mill a smooth surface between scan lines, the pixel spacing between adjacent scan lines must be also small enough to allow a proper overlap between adjacent lines .

If the ion distribution of a FIB is approximately by a Gaussian distribution, the scanning ion flux becomes steady and unwavering when the normalized pixel spacing (Ps/α) is smaller than 1.5, where α is the standard deviation of the Gaussian distribution. In order to have a uniform scanning ion flux in channel milling, the normalized pixel spacing should be equal to or smaller than 1.5 (or Ps/dr = 0.637). However, a full drilling process for making an array of holes would require both the normalized pixel spacing to be larger than 8. As a result, the drilling process can be considered as a special case of channel milling . The roughness on the milled surface is also greatly influenced by pixel spacing. The reason is that a sufficient number of spots by digitally controlled pixel can lead to a superior milling surface .

2.6.7 PATTERN SIZE

In many results, one of the greatest influencing parameter on sputtering depth is the pattern size. The reason is the dwell time and pixel spacing are dependent on the pattern size. found that the milled hole/box was not uniform in width and showed rough damages in the silicon samples with decreasing the pattern size and the size of the neighbouring beam points. Thus, each device required adjustments in the pattern size before fabrication.

2.6.8 MILLING MODE

There are two types of milling modes, bitmap function and pixel space mode . For bitmap function, a designed pattern with bitmap format of *.XBM is prepared first. Then, this pattern can be called in a defined area. After start milling, the FIB can direct write the pattern by deflection of the beam in terms of bitmap file. Micro-holes were directly fabricated using this method. For pixel space mode, the pixel spaces in both horizontal and vertical directions are set to be large enough that centre distance is more than two times of the hole’s diameter. Both bitmap function and pixel space mode are limited by the vibration of the stage occurred more or less during the process . However, it is difficult to fabricate the micro-hole array with size below 100nm by bitmap function in the area as large as 100x100µm2 and even more due to the resolution limitation of beam deflection in the large area. It only can be realized by use of the pixel space mode with large pixel in both horizontal and vertical scanning directions.

2.7 EFFECTS OF FIB MILLING ON MICRO-HOLES

The understanding of FIB damages is important to ensure that the region being milled is indeed representative of the holes pattern, and is not due to a specimen preparation artefact. The interaction between the incident ions (Ga+ ions) and the target material during FIB milling may lead to surface damage and consequently limit the ability to achieve high quality of milled holes.

2.7.1 RE-DEPOSITION

For FIB milling the aspect ratio is limited by the re-deposition of sputtered material on the surfaces of the structures and it is difficult to estimate. When energy transferred from the Ga+ ions to the target substrate, this may lead to a collision cascade involving substrate atoms at or near the surface where sputtering occurs . As the sputtered atoms are not in their thermodynamic equilibrium state, they tend to condensed back into the solid phase after their collision with any nearby surface. Sputtered atoms are re-deposited at the walls of patterns and they distort the pattern profiles. This distortion depends on the beam scanning procedure. The amount of re-deposited material is dependent on the geometry of the structure being milled and the process of re-deposition is also nonlinear with time.

In normal ion milling, it is found that re-deposition on the sidewall is controllable by adjusting the FIB scan speed (dwell time) and direction . Sputtered atoms are re-deposited more on the sidewall facing the slope having the gradient (Ѳ) on the sidewall in the back of the slope. By varying the scan speed and the gradient (Ѳ), hence sputter re-deposition can be controlled. The sidewall too causes the variation of ion sputtering that leads to re-deposition at the bottom and to groove forming .

At an aspect ratio >1, more than half of the sputtered atoms from the bottom surface will redeposit. Therefore, the actual milling depth is less than the designed value owing to the accumulated material at the bottom . The higher the aspect ratio, the more re-deposited material is accumulated on the convex surface, because the nearer the ion beam is to the edge, the higher is the sputtering yield from the sidewalls which leads to increase the milling depth. Furthermore, the re-deposited material will adhere to the sidewall for milling a 3D structure with an aspect ratio >1, and land on the bottom causing a variation of milling depth for a structure with a low aspect ratio (normally <1). According to Tseng AA (2004), the earlier the region is milled, the more re-deposited material is accumulated in that same region. As a rule of thumb, the region where a greater milling depth or more material removal is desired should be milled last in the scanning sequence. The propensity for re-deposition also increases when the FIB parameters are used that contribute to factors that increase the sputtering rate (eg. using a higher beam current) .

High aspect ratio cuts cannot be produced unless the material removed is volatile and does not deposit on the walls of the cut, which often limits the obtainable aspect ratio of a micromachined feature to 10 . Because of the effects of re-deposition of the sputtered materials, the holes grow deeper but also narrower. It can be characterized by the Orloff model . The flux density of material sputtered from the bottom of the hole and deposited to the sidewall of the hole, F(h), can be represented by

where F0 is the total number of atoms sputtered per unit length, d is the hole diameter, h is the depth of the processed hole, r=(x2+h2)1/2.

ideal

actual

Fig. 2.3: Aspect ratio vs normalized re-deposition rate, F/Fo

The FIB reached aspect ratio of the micro-hole is decreased exponentially with normalized re-deposition, F (h) / F0 , as shown in Fig. 2.3. At an aspect ratio <1 more than half of the sputtered atoms from the bottom surface redeposit on the sidewall from the bottom surface. For a material with higher sputter yield, the ratio of F (h) / F0 will be degraded. Therefore, the aspect ratio can be improved further by selecting the substrate material with high sputter yield .

In addition, this redeposit material can result in short circuit after IC edit . Since the intent of circuit editing is usually to edit or alter the electrical conducting metallization of the IC by cutting and/or rerouting conductors, any re-deposition of micro-machined material which may result in unwanted conduction paths or interfere with the electrical connection of FIB deposited conductors to circuit metallization cannot be tolerated.

2.7.2 AMORPHISATION

Amorphisation of a FIB milled crystalline surface may occur due to sufficient atom displacement within the collision cascade resulting in the less of long range order when the density of point defects reaches a critical value. If the dose level of milling ions is not enough, amorphisation may occurs in the bombarded area of a crystalline substrate and induce the substrate to swell . For a crystallized Si substrate bombarded by Ga+ ions, the dose level to cause amorphisation is of the order of 1015 ions cm-2, while the effective milling dose should be at least two orders of magnitude higher than the amorphisation dose as indicated by .

Since the most of the FIB roughly resembles a Gaussian ion distribution, the intensity at the fringe (tail) of the beam is much smaller than that at the core (centre region) and it is not strong enough to sputter materials but is sufficient to cause amorphisation that induces substrate swelling. The FIB always has a tail region that possesses the right dose to cause maximum swelling during channel milling. has found that FIB milling with Ga+ ions at an energy of 30keV will produce amorphization damage along a Si sidewall that is ~28nm thick and up to 20wt% Ga+ may be present within the damage region. To prevent top surface damage, one may deposit a layer of platinum of other metal using an FIB instrument prior to the start of any milling by using low energy ions . Fig.2.4 shows the re-deposition and amorphization on FIB milled holes.

Amorphization

Re-deposition

Fig. 2.4: Effect of FIB milling, re-deposition and amorphization on FIB milled holes.

2.8 MICRO HOLES/VIAS FILLING

Micro holes or micro vias filling is a standard process used in IC substrate production where generally no through via are present, and also used in other high density interconnection (HDI) applications in particular for hand held devices. High aspect ratio micro holes/vias has becoming more important within the packaging of semiconductor devices especially MEMS since it allows a compact connection technology with an excellent electrical performances and a possibility of 3D integration packaging. However, filiing high aspect ratio micro holes/vias is a challenge for conventional filling technology, which restricts the improvement of the aspect ratio and density of micro holes/vias.

2.8.1 FILL MATERIAL

There are many fill materials that can be used to fill high aspect ratio micro holes/vias. The need of fill material is depend from the large range of devices used in a wide variety of conditions and applications. The conducting material to be used for holes/vias filling must be corrosion resistant, electrically conducting, easily adaptable to the existing manufacturing process such as screen printable, low shrinkage after processing, high strength, good adhesion and yet, low cost.

Copper and gold are the most materials used, as their ability to be seeded and plated, and for their high current capabilities. Copper is the obvious material for the holes filling in high density interconnect (HDI) technology, which gives improved reliability. Copper filling gives a more efficient thermal and electrical conductive path in a substrate that allows potentially small holes/vias with thermal characteristics that release space on a substrate for active circuit pathways. However, nowadays copper does not have a suitable coefficient of thermal expansion match with silicon. Thus, other metals need to be considered for this purpose. Moreover, the effects such as cost, resistivity and compatibility (referred to the volumetric coefficient of thermal expansion, β) must be the main considerations. A fill material with the best matches to the volumetric coefficient of expansion of silicon may produce a higher yield for long term, extreme temperature and high current density devices. A variety of possible filling materials are listed and compared in Table

Table : Cost, resistivity and volumetric coefficient for holes/vias filling.

Material

Cost (S/lb)

Resistivity (Ω.cm) at 20°C

Volumetric coefficient of thermal expansion (3x linear coefficient),β at 20°C (β/°C)

Silver

561.16

1.59x10-8

5.4x10-5

Copper

3.99

1.68x10-8

5.1x10-5

Gold

21,746.66

2.44x10-8

4.2x10-5

Tungsten

15.00

5.6x10-8

1.35x10-5

Nickel

3.33

6.844x10-8

4.74x10-5

Silicon

1.29

6.4x102

0.9x10-5

2.8.2 METAL NANOPOWDER FILLING

The use of nanopowder filling to produce metallic/ceramic microstructures has great potential as demand for this structures increased, due to nature of net-shaping in which machining is not required . Moreover, nanopowder has a higher catalytic activity and better sinterability compared with coarse grained bulk materials . Nanopowder used in micro-Powder Injection Molding (µPIM) was found to differ from conventional PIM of coarse powders which provide good dimensional control and flowability. The widely used metals in the creation of MEMS are gold, copper, titanium, silver, aluminium, and nickel. These metals can be deposited by the processes like electroplating, sputtering, micro injection molding and hot embossing. A good example is 316 stainless microgear, microneedles for selective nerve stimulation and micromolds made by hardmetals for plastic micro injection molding and hot embossing .

found that by inserting silver nanoparticles-filled photonic crystal hole structure array, the accumulated enhancement in the photoluminescence (PL) intensity from a layer of quantum dots (QD), 84%, was achieved due to hybrid effect of silver nanoarray-induced localized surface plasmon resonance (SPR) and outcoupling of wave-guided light in two-dimensional nanopattern array. High aspect ratio microstructures of electrodeposited copper-γ-Al2O3 nanocomposites were successfully performed into deep recesses with uniform concentration of alumina nanoparticles along the length of the microposts . A gas bearing greenpart with tiny and complex structures has successfully fabricated by filling the 17-4 PH stainless steel nanopowders mixed with polymer binder into the PDMS soft mold .

There are several challenges in applying metal nanopowder to high aspect ratio microfilling. The particle size of the metal nanopowder is a very important parameter which limits the producible HARMS . Furthermore, agglomeration becomes the critical issue during fabrication process. Highly agglomerating nanopowders leads to gas trapped in HARMS and cavitations may occur .

2.8.3 NICKEL NANOPOWDER

Nickel is commonly used in MEMS industry. It has good mechanical properties, such as hardness, yield strength and Young’s modulus and good resistance against corrosion. Nickel and Ni based alloys have been found to have good mechanical properties that can be exploited to realize movable structures in MEMS devices. Moreover the magnetic properties of Ni have been widely used in magnetic . Usually nickel is filled through holes by electroplating to make electric contacts. A fabrication technology of high density electrical feed-through in Pyrex glass wafers is successfully fabricated by filling the through holes with electroplated nickel . Nickel and nickel alloys were used as the tip and structure materials fabricated by electroplating for a new cantilever type of MEMS probe card . mentioned the effect of annealing on the morphologies and conductivities of sub-micrometer sized nickel particles used for electrically conductive adhesive.

Another findings are a single layer of Ni nanoparticles with average diameter 5nm and estimated density 1012/cm2 has successfully embedded into MOS capacitors fabricated by using EBV and RTA for non-volatile memory applications . Because Ni has a larger work function than that of semiconductor, the MOS capacitor with Ni nanoparticles is expected to have a deeper potential than that of semiconductor nanoparticles. An investigation for characterization of pure nickel nanopowders prepared by anodic arc discharging plasma method with homemade experimental apparatus has been carried out by . The samples distributed uniformly in spherical chain shapes, the crystal structure of the samples is FCC structure as same as the bulk materials and the average particle size is about 47nm with the size distribution ranging from 20 to 70nm. The nickel nanopowders obtained with uniform size, monodisperse with a narrow size distribution and have low impurity contamination. The specific surface area is 14.23m2/g, pore volumes are 0.09cm3/g and average pore diameter is 23nm. Although nickel has a slightly high resistivity than copper, the cost is affordable with higher volumetric coefficient of thermal expansion, plus the large capabilities of nickel in nano-size, which promise an alternative material of holes/vias filling.

2.9 METHODS OF HOLES/VIAS FILLING

2.9.1 ELECTROPLATING

Plating method is generally uses for holes/vias filling in high density or through holes interconnection (HDI) technology. Due to the high deposition rates, relatively low cost and easiness in satisfying high aspect ratio features, plating adoption rates have been increased in the fields of semiconductors, especially in MEMS. Plating can be divided into electroplating, which requires a seed layer, and electroless plating which does not. Electroplating is more suitable than electroless plating since the method is simpler and provides good adhesion, in order to get an adhesive strength of plating with the holes wall and prevent the plating from being separated from the wall.

Electroplating of copper is preferred as filling material because of high electrical conductivity of copper and the high electrodeposition rate. Moreover, electroplating is amenable to process at close to room temperature, as well as a wide available tool vendor support and process maturity. However, it has some problems to be solved due to its complexity in terms of process controllability, reliability and throughput. In particular, high aspect ratios of holes/vias filling with void free metal cores are difficult to implement. The filling up of holes/vias with high aspect ratio up to 15 is stil



rev

Our Service Portfolio

jb

Want To Place An Order Quickly?

Then shoot us a message on Whatsapp, WeChat or Gmail. We are available 24/7 to assist you.

whatsapp

Do not panic, you are at the right place

jb

Visit Our essay writting help page to get all the details and guidence on availing our assiatance service.

Get 20% Discount, Now
£19 £14/ Per Page
14 days delivery time

Our writting assistance service is undoubtedly one of the most affordable writting assistance services and we have highly qualified professionls to help you with your work. So what are you waiting for, click below to order now.

Get An Instant Quote

ORDER TODAY!

Our experts are ready to assist you, call us to get a free quote or order now to get succeed in your academics writing.

Get a Free Quote Order Now