Powered By Blogger

Saturday, August 11, 2012


Design and Production Process of CMOS Fabrication


Submitted By
Lokanath Tripathy(115)
Satya Swarup(89)
Sumeet Singh(98)
Branch of Engineering: Electronics and Instrumentation Engineering


Introduction
The following discussion will concentrate on the well-established CMOS fabrication technology, which requires that both n-channel (nMOS) and p-channel (pMOS) transistors be built on the same chip substrate. To accommodate both nMOS and pMOS devices, special regions must be created in which the semiconductor type is opposite to the substrate type. These regions are called wells or tubs. A p-well is created in an n-type substrate or, alternatively, an n-well is created in a p-type substrate. In the simple n-well
CMOS fabrication technology presented here, the nMOS transistor is created in the p-type substrate, and the pMOS transistor is created in the n-well, which is built into the p-type substrate. In the twin-tub CMOS technology, additional tubs of the same type as the substrate can also be created for device optimization.
The simplified process sequence for the fabrication of CMOS integrated circuits on a p-type silicon substrate is shown in Fig. 1.1. The process starts with the creation of the n-well regions for pMOS transistors, by impurity implantation into the substrate. Then, a thick oxide is grown in the regions surrounding the nMOS and pMOS active regions. The thin gate oxide is subsequently grown on the surface through thermal oxidation. These steps are followed by the creation of n+ and p+ regions (source, drain, and channel-stop implants) and by final metallization (creation of metal interconnects).
Fig.1.1 Simplified process sequence for the fabrication of the n-well CMOS integrated circuit with a single polysilicon layer, showing only major fabrication steps.
Fabrication Process Flow: Basic Steps
Note that each processing step requires that certain areas are defined on chip by appropriate masks. Consequently, the integrated circuit may be viewed as a set of patterned layers of doped silicon, poly-silicon, metal, and insulating silicon dioxide. In general, a layer must be patterned before the next layer of material is applied on the chip. The process used to transfer a pattern to a layer on the chip is called lithography. Since each layer has its own distinct patterning requirements, the lithographic sequence must be repeated for every layer, using a different mask. To illustrate the fabrication steps involved in patterning silicon dioxide through optical lithography, let us first examine the process flow shown in Fig. 1.2. The sequence starts with the thermal oxidation of the silicon surface, by which an oxide layer of about 1 m thickness, for example, is created on the substrate (Fig. 1.2(b)). The entire oxide surface is then covered with a layer of photoresist, which is essentially a light-sensitive, acid-resistant organic polymer, initially insoluble in the developing solution (Fig.1.2(c)). If the photoresist material is exposed to ultraviolet (UV) light, the exposed areas become soluble so that they are no longer resistant to etching solvents. To selectively expose the photoresist, we have to cover some of the areas on the surface with a mask during exposure. Thus, when the structure with the mask on top is exposed to UV light, areas which are covered by the opaque features on the mask are shielded. In the areas where the UV light can pass through, on the other hand, the photoresist is exposed and becomes soluble (Fig. 1.2(d)).
Fig.1.2 Process steps required for patterning of silicon dioxide.
The type of photoresist which is initially insoluble and becomes soluble after exposure to UV light is called positive photoresist. The process sequence shown in Fig. 1.2 uses positive photoresist. There is another type of photoresist which is initially soluble and becomes insoluble (hardened) after exposure to UV light, called negative photoresist. If negative photoresist is used in the photolithography process, the areas which are not shielded from the UV light by the opaque mask features become insoluble, whereas the shielded areas can subsequently be etched away by a developing solution. Negative photoresists are more sensitive to light, but their photolithographic resolution is not as high as that of the positive photoresists. Therefore, negative photoresists are-used less commonly in the manufacturing of high-density integrated circuits. Following the UV exposure step, the unexposed portions of the photoresist can be removed by a solvent. Now, the silicon dioxide regions which are not covered by hardened photoresist can be etched away either by using a chemical solvent (HF acid) or by using a dry etch (plasma etch) process (Fig. 1.2(e)). Note that at the end of this step, we obtain an oxide window that reaches down to the silicon surface (Fig. 1.2(f)). The remaining photoresist can now be stripped from the silicon dioxide surface by using another solvent, leaving the patterned silicon dioxide feature on the surface as shown in Fig. 1.2(g).
The sequence of process steps illustrated in detail in Fig. 1.2 actually accomplishes a single pattern transfer onto the silicon dioxide surface, as shown in Fig. 1.3. The fabrication of semiconductor devices requires several such pattern transfers to be performed on silicon dioxide, polysilicon, and metal. The basic patterning process used in all fabrication steps, however, is quite similar to the one shown in Fig. 1.2. Also note that for accurate generation of high-density patterns required in sub-micron devices, electron beam (E-beam) lithography is used instead of optical lithography.  In the following, the main processing steps involved in the fabrication of an n-channel MOS transistor on a p-type silicon substrate will be examined.

Fig. 1.2 Process steps required for patterning of silicon dioxide (continued).

 
Fig. 1.3 The result of a single lithographic patterning sequence on silicon dioxide, without showing the intermediate steps. Compare the un-patterned structure (top) and the patterned structure (bottom) with Fig. 1.2(b) and Fig. 1.2(g), respectively.
Fabrication of the nMOS Transistor
The process starts with the oxidation of the silicon substrate (Fig. 1.4(a)), in which a relatively thick silicon dioxide layer, also called field oxide, is created on the surface (Fig. 1.4(b)). Then, the field oxide is selectively etched to expose the silicon surface on which the MOS transistor will be created (Fig. 1.4(c)). Following this step, the surface is covered with a thin, high-quality oxide layer, which will eventually form the gate oxide of the MOS transistor (of fig 1.4(d)). On the top of thin oxide layer, a layer of polysilicon (polycrystalline silicon) is deposited (Fig. 1.4(e)). Polysilicon is used both as gate electrode material for MOS transistors and also as an interconnect medium in silicon integrated circuits. Un-doped polysilicon has relatively high resistivity. The-resistivity of polysilicon can be reduced, however, by doping it with impurity atoms. After deposition the polysilicon layer is patterned and etched to form the interconnection and the MOS transistor gates (Fig. 1.4(f)). The thin gate oxide not covered by polysilicon is also etched away, which exposes the bare silicon surface on which the source and drain junctions are to be formed (Fig. 1.4(g)). The entire silicon surface is the n-doped with a high concentration of impurities, either through diffusion or ion implantation (in this case with donor atoms to produce n-type doping). Fig. 1.4(h) shows that the doping penetrates the exposed areas on the silicon surface, ultimately creating two n-type regions (source and drain junctions) in the p-type substrate. The impurity doping also penetrates the polysilicon on the surface, reducing its resistivity. 
Fig. 1.4 Process flow for the fabrication of an n-type MOSFET on p-type silicon.

Note that the polysilicon gate, which is patterned before doping, actually defines the precise location, of the channel region and, hence, the location of the source and the drain regions. Since this procedure allows very precise positioning of the two regions relative to the gate, it is also called the self-aligned process.
Fig. 1.4 Process flow for the fabrication of an n-type MOSFET on p-type silicon (continued).

Once the source and drain regions are completed, the entire surface is again covered with an insulating layer of silicon dioxide (Fig. 1.4(i)). The insulating oxide layer is then patterned in order to provide contact windows for the drain and source junctions (Fig. 1.4(j)). The surface is covered with evaporated aluminium which will form the interconnection (Fig. 1.4(k)). Finally, the metal layer is patterned and etched, completing the interconnection of the MOS transistors on the surface (Fig. 1.4(l)). Usually, a second (and third) layer of metallic interconnect can also be added on top of this structure by creating another insulating oxide layer, cutting contact (via) holes, depositing, and patterning the metal.
Device Isolation Techniques
The MOS transistors that comprise an integrated circuit must be electrically isolated from each other during fabrication. Isolation is required to prevent unwanted conduction paths between the devices, to avoid creation of inversion layers outside the channel regions of transistors, and to reduce leakage currents. To achieve a sufficient level of electrical isolation between neighbouring transistors on a chip surface, the devices are typically created in dedicated regions called active areas, where each active area is surrounded by a relatively thick oxide barrier called the field oxide. One possible technique to create isolated
active areas on silicon surface is first to grow a thick field oxide over the entire surface of the chip, and then to selectively etch the oxide in certain regions, to define the active areas. This fabrication technique, called etched field-oxide isolation, is already illustrated in Fig. 1.4(b) and Fig. 1.4(c). Here, the field oxide is selectively etched away to expose the silicon surface on which the MOS/h transistor will be created. Although the technique is relatively straightforward, it also has 29 some drawbacks. The most significant disadvantage is that the thickness of the field oxide leads to rather large oxide steps at the boundaries between active areas and isolation Fabrication (field) regions. When polysilicon and metal layers are
 
Fig. 1.4 Process flow for the fabrication of an n-type MOSFET on p-type silicon (continued).
deposited over such boundaries in of MOSFETs subsequent process steps, the sheer height difference at the boundary can cause cracking of deposited layers, leading to chip failure. To prevent this, most manufacturers prefer isolation techniques that partially recess the field oxide into the silicon surface, resulting in a more planar surface topology.
Local Oxidation of Silicon (LOCOS)
The local oxidation of silicon (LOCOS) technique is based on the principle of selectively growing the field oxide in certain regions, instead of selectively etching away the active areas after oxide growth. Selective oxide growth is achieved by shielding the active areas with silicon nitride (Si3N4) during oxidation, which effectively inhibits oxide growth. The basic steps of the LOCOS process are illustrated in Fig. 1.5. First, a thin pad oxide (also called stress-relief oxide) is grown on the silicon surface, followed by the deposition and patterning of a silicon nitride layer to mask (i.e., to define) the active areas (Fig. 1.5(a)). The thin pad oxide underneath the silicon nitride layer is used to protect the silicon surface from stress caused by nitride during the subsequent process steps. The exposed areas of the silicon surface, which will eventually form the isolation regions, are doped with a p-type impurity to create the channel-stop implants that surround the transistors (Fig. 1.5(b)). Next, a thick field oxide is grown in the areas not covered with silicon nitride, as shown in Fig. 1.5(c). Notice that the field oxide is partially recessed into the surface since the thermal oxidation process also consumes some of the silicon. Also, the field oxide forms a lateral extension under the nitride layer, called the bird's beak region. This lateral encroachment is mainly responsible for a reduction of the active area. The silicon nitride layer and the thin pad oxide layer are etched in the final step (Fig. 1.5(d)), resulting in active areas surrounded by the partially recessed field oxide. The LOCOS process is a popular technique used for achieving field oxide isolation with a more planar surface topology. Several additional measures have also been developed over the years to control the lateral bird's beak encroachment, since this encroachment ultimately limits device scaling and device density in VLSI circuits.




Fig. 1.5 Basic steps of the LOCOS process to create oxide isolation around active areas.


The CMOS n-Well Process
Having examined the basic process steps for pattern transfer through lithography and having gone through the fabrication procedure of a single n-type MOS transistor, we can now return to the generalized fabrication sequence of n-well CMOS integrated circuits, as shown in Fig. 1.1. In the following figures, some of the important process steps involved in the fabrication of a CMOS inverter will be shown by a top view of the lithographic masks and a cross-sectional view of the relevant areas. The n-well CMOS process starts with a moderately doped (with impurity concentration typically less than 1015 cm-3) p-type silicon substrate. Then, an initial oxide layer is grown on the entire surface. The first lithographic mask defines the n-well region. Donor atoms, usually phosphorus, are implanted through this window in the oxide.
Layout Design Rules
The physical mask layout of any circuit to be manufactured using a particular process must conform to a set of geometric constraints or rules, which are generally called layout design rules. These rules usually specify the minimum allowable line widths for physical objects on-chip such as metal and polysilicon interconnects or diffusion areas, minimum feature dimensions, and minimum allowable separations between two such features. If a metal line width is made too small, for example, it is possible for the line to break during the fabrication process or afterwards, resulting in an open circuit. If two lines are placed too close to each other in the layout, they may form an unwanted short circuit by merging during or after the fabrication process. The main objective of design rules is to achieve, for any circuit to be manufactured with a particular process, a high overall yield and reliability while using the smallest possible silicon area. Note that there is usually a trade-off between higher yield, which is obtained through conservative geometries, and better area efficiency, which is obtained through aggressive, high-density placement of various features on the chip. The layout design rules which are specified for a particular fabrication process normally represent a reasonable optimum point in terms of yield and density. It must be emphasized, however, that the design rules do not represent strict boundaries which separate "correct" designs from "incorrect" ones. A layout which violates some of the specified design rules may still result in an operational circuit with reasonable yield, whereas another layout observing all specified design rules may result in a circuit which is not functional and/or has very low yield. To summarize, we can say, in general, that observing the layout design rules significantly increases the probability of fabricating a successful product with high yield.
The design rules are usually described in two ways:
        i.            Micron rules, in which the layout constraints such as minimum feature sizes and minimum allowable feature separations  are stated  in  terms  of absolute dimensions in micrometres, or,
      ii.            Lambda rules, which specify the layout constraints in terms of a single parameter (λ) and thus allow linear, proportional scaling of all geometrical constraints.
Lambda-based layout design rules were originally devised to simplify the industry-standard micron-based design rules and to allow scaling capability for various processes. It must be emphasized, however, that most of the submicron CMOS process design rules do not lend themselves to straightforward linear scaling. The use of lambda-based design rules must therefore be handled with caution in submicron geometries. In the following, we present a sample set of the lambda-based layout design rules devised for the MOSIS (MOS Implementation System) CMOS process and illustrate the implications of these rules on a section of a simple layout which includes two transistors (Fig. 1.6). The complete set of MOSIS CMOS scalable design rules are also illustrated in colour in Plate 6 and Plate 7.

MOSIS Layout Design Rules (sample set)

Rule number                                      Description                                                     λ-Rule
Active area rules
R1                                Minimum active area width                                             3λ
R2                                Minimum active area spacing                                          3λ

Polysilicon rules
R3                                Minimum poly width                                                       2λ
R4                                Minimum poly spacing                                                    2λ

R5                                Minimum gate extension of poly over active                     2λ
R6                                Minimum poly-active edge spacing                                 1λ
(poly outside active area)
R7                                Minimum poly-active edge spacing                                 3λ
(poly inside active area)

Metal rules
R8                                Minimum metal width                                                    3λ
R9                                Minimum metal spacing                                                 3λ

Contact rules
R10                              Poly contact size                                                          2λ
R11                              Minimum poly contact spacing                                      2λ
R12                              Minimum poly contact to poly edge spacing                  1X
R13                              Minimum poly contact to metal edge spacing                 1λ
R14                              Minimum poly contact to active edge spacing                3λ
R15                              Active contact size                                                        2λ
R16                              Minimum active contact spacing                                    2λ
(on the same active region) 
R17                              Minimum active contact to active edge spacing             1λ
R18                              Minimum active contact to metal edge spacing              1λ
R19                              Minimum active contact to poly edge spacing                3λ
R20                              Minimum active contact spacing                                    6λ
(on different active regions)


Fig. 1.6 Illustration of some of the typical MOSIS layout design rules.





MOSFET Layout Design

            Fig. 1.6  Illustration of some of the typical MOSIS layout design rules (continued).


Summary on

CMOS Image Sensors for High Speed Applications


(Munir El-Desouki  , M. Jamal Deen  , Qiyin Fang  , Louis Liu  , Frances Tse   and David Armstrong )
Sensors 2009, 9(1), 430-444,Published: 13 January 2009

Introduction
  For an Image to be captured 2 basic steps are involved

  •  Image Acquisition
  • Image processing
Recent advances in deep submicron CMOS technologies and improved pixel designs have enabled CMOS-based imagers  to surpass charge-coupled devices (CCD) imaging technology for mainstream applications. The parallel outputs that CMOS imagers can offer, in addition to complete camera-on-a-chip solutions due to being fabricated in standard CMOS technologies, result in compelling advantages in speed and system throughput. Since there is a practical limit on the minimum pixel size (4~5  μm) due to limitations in the optics, CMOS technology scaling can allow for an increased number of transistors to be integrated into the pixel to improve both detection and signal processing. Such smart pixels truly show the potential of CMOS technology for imaging applications allowing CMOS imagers to achieve the image quality and global shuttering performance necessary to meet the demands of ultrahigh-speed applications.
In this paper, a review of CMOS-based high-speed imager design is presented and the various implementations that target ultrahigh-speed imaging are described. One of the main advantages of CMOS image sensors is that they are fabricated in standard CMOS technologies, which allows for full integration of the image sensor along with the processing and control circuits on the same chip and at a low cost. This camera-on-chip system leads to reduction in power consumption, cost and sensor size and allows for integration of new sensor functionalities. Since digital transistors take more advantage of CMOS scaling properties, digital pixel sensors (DPS) have become very attractive. Such smart pixels truly show the potential of CMOS technology for imaging applications allowing CMOS imagers to achieve the image quality and global shuttering performance necessary to meet the demands of ultrahigh-speed applications.

  1.      Array Level Technique


(Array-Level Techniques)



 Pixel-Level Techniques
A DPS(Digital pixel sensor)  integrates an ADC into each pixel resulting in a massively
Parallel readout and conversion that can allow very high speed operation, where digital data is read out of each pixel.
        (3 MOSFET transistor)               (Operational Amplifier)           (DigitalOutput)                                                                  
                          (Photo Diode)

 Using a DPS(right side diagram) will only require one ADC conversion cycle for all pixels in parallel, which results in a great increase in capturing rate. The high speed readout makes CMOS image sensors suitable for very high-resolution imagers (multi-megapixels) especially for video applications.

Analog Readout Architectures

In order to achieve the fastest FR possible for a certain high-speed experiment, a number of extremely fast consecutive images can be captured and stored in analog form. By doing so, the inter-frame delay caused by the ADC conversion time and array readout can be avoided. The FR (Frame rate)only depends on the speed of the devices and transistors used within the pixel, assuming a large enough illumination exists on the object being imaged. Depending on the type of experiment and the Speed of capture, there will be a minimum number of frames that is acceptable.
Ø  Ultrahigh-Speed CMOS Imager can capture 8 frames at an acquisition rate of 1.25 billion fps.
Ø  Contain 38 transistor(complex circuit, Not included here)
Ø  The basic idea is to utilize 8 analog memory units to temporarily hold 8 frames at a very high speed, avoiding the delay time in analog-to-digital conversion and readout.( Things are done parallel, hence delay time get reduced as compared to using 1 memory unit).

Conclusions
Existing  results with different CMOS imaging architectures have achieved thousands up to even millions of fps(frame per second).By combining a number of different methods, which  include parallel per-column or per-pixel ADCs, image compression, parallel output port readout, high  readout clock rates and simultaneous capture and processing, researchers have managed to push frame-rates to 10,000 fps.
Using an ultrahigh-speed imager design for a 1-D line-scan imager can increase the number of consecutive images that can be captured at rates of over a billion fps.
Research on high speed image capturing using Cmos and other technology is going on largely and the time to acquire and process the high speed image is getting decreased over the years



Reference: www.mdpi.com/journal/sensors.

Thursday, August 9, 2012

Summary of Research Paper on IE -Optimization of cutting Wood


Optimization of cutting in primary wood transformation industries 

Raïdandi Danwé, Isaac Bindzi, Lucien Meva’a
National Advanced School of Engineering, the University of Yaounde I (CAMEROON)
(JOURNAL OF INDUSTRIAL ENGG. AND MANAGEMENT,2012,VOL-5,115-132)

Introduction:
The Cameroon forest covers about 20 millions of hectares, representing 44% of the national territory. Cameroon possesses one of the most developed primary conversion wood industry. The loss of raw materials in wood cutting industries has reached high proportions (30 to 36% of volume yield).
 Statistical analyses have shown that there exists a considerable difference between the volume of logs bought by primary conversion wood factories in Cameroon and the volume of the cuts leaving the industries. This low efficiency results in the waste of raw material, which is burnt as biomass. In order to tackle this problem, the exploitation and the primary conversion of wood in  Cameroon was studied.  The various methods of cutting and the different products obtained were studied. We then proceeded with the formulation of the log cutting  optimization problem based on a real shape model of the logs. The solution to this problem then led to the design of a software package to be used as a cutting optimizer. The automation of the cutting operation leads to an accelerated work and an increase in the volume of the cuts produced daily.
In this Paper, the work they  studied the optimization of wood production in industries of primary conversion. Primary wood conversion comprises of  the activities of sawing and wood planning, unrolling and trenching, drying, and finally  wood impregnation, in which they  studied the various methods of cutting and the  different products obtained. We then proceeded with the formulation of the log cutting optimization  problem based on a real shape model of the log. Finally  design and the presentation of a software package called cutting optimizer  were done.
Geometrical Modeling and Cutting up
The achievement of a better profit in sawing industries depends mainly on the mathematical representation of the logs of wood during cutting up. Geometric modeling of logs serves as a preliminary stage in the automation of operations in a sawmill .The automation also requires real time simulations. Time required for the analysis of data related to logs should then be reduced. This additional constraint imposes a compromise between the accuracy of the model and the quantity of data to be processed.
It is quite relevant to mention that in modeling  the external shape of the log, swellings (which generally characterize the roughness of logs), should not be taken  in account in the perspective of optimizing  the volume output of logs being cut up. Only smooth logs are considered here.
Cutting
The mode of cutting  depends on the characteristics of the log and the equipment’s of the company. In order to optimize cutting decisions, an automated real-time (i.e. at production speed) evaluation of material dimensions and quality is required. Some requirement for the first cut is to lead to a maximum value yield of a log are a knowledge of the log  geometry (length, diameters, eccentricity, sweep...), the log quality, the  determination of the best orientation and position of the log at the first cut, and  the use of a cutting system that enables to obtain plane surfaces during log  cutting.

Discrete Geometric Model of the log


Products from the first stages of primary conversion of wood
·        Best Opening Face: The BOF uses the principle that the first cut is the key of the sawing strategy for a given sawing pattern, in order to maximize the yield of the saw log. It considers the volume-yield of the saw log, leaving out the effect of internal characteristics.
Production Optimization: The achievement of a better profit in sawing industries depends mainly on the mathematical model of logs during cutting. With respect to the small logs, loss of material can be observed due to the effects of the wood feature, the precision, the method used in cutting, and the thickness of the saw.
Cutting Process and Classification: For local sawing industries, we encounter three major classes. For the first class is made up of cuts presenting no defect (absence of nodes, dimensional conformity according to the functional condition of contract). The second class is made up of cuts having at most one node. For the third one, the number of nodes is greater than or equal to two and in addition, we can also find dimensional defects.
Formulation of Optimization Problem: The optimization model resorts to wide-spread techniques of linear programming or operational research. The above optimization problem can be assimilated to a knapsack problem with non-bounded variables which can be formulated thus: We have N types of object (product from log cutting) in an infinite number for each type.  An object of type k has a positive integer value Ck (its market price) and occupies a volume ak (for a cut cross-section Sk). Let uk be the number of type k objects in the bag. The problem is thus expressed as: Finding the subset of objects of maximum value, whose volume does not exceed the capacity b of the bag.
Resolution of Optimizing Problem: knapsack model is used in following situation
·         When loading ships or planes: all luggage have to be packed without  overloading
·          When cutting materials: to minimize scraps when cutting rods into iron bars.
Optimization software package for log cutting: It aims at determining the number of cuts of type k that we  can extract from a log of wood in a way so as to optimize the material output and maximize its  commercial value.
The Cutting Optimizer: The optimization program provides a cutting schedule. This program calculates the first position and the next positions depending on the memorized log dimensions and the thickness of the blade.
The  optimization software used by these sawmills present some deficiencies which are
·         The software does not take into consideration the ordering list of the  enterprise and the market prices in optimizing of the cutting process.
·         Cutting is done uniquely with one of the cutting methods, whatever the  operator’s order.
·         The software neither takes into account  the real shape of the log, nor does it take into consideration the defects on it. It deals only with the principal cross-section of the log. We run the risk of having much scrap. 
The command principle of the cutting optimizer:

The command principle of the optimizer
The length of each log varies  depending on the order. At the output of our system , we obtain the cuts.
The optimization system for the log carriage semi-modern software (piloting and optimization devices), a decision making block which receives data from the input (characteristic dimensions of each log), analyses them  together with the data from the command computer and then takes a decision on  the cutting method to be implemented. This decision is transmitted to the  automaton that commands the rotational and translational movements of the
carriage. The command computer receives at its input a feedback from the market (database on the prices of the cuts) and the commands received by the enterprise.
Conclusion: The modeling  of the real shape of the log enabled us to identify the input parameters to generate the log in real shape. The formulation of the optimization  problem assimilated to a knapsack problem was solved using the method of dynamic programming. That enabled to set up an algorithm for optimum cutting of the log, alongside a cutting optimizer. This work has led to a possible solution for increasing material output, productivity, and quality of cuts, regardless the type of wood.

Go through this link
To access the Journal