Fluidized bed technology

CONTENTS: CHAPTER 1: INTRODUCTION

1.1. Fluidized Bed technology1.2. Background1.3. Pressurized Circulating Fluidized Bed 1.4. Motivation

CHAPTER 2: LITERATURE REVIEW2.1. Hydrodynamics and Heat Transfer in a CFB 2.2. Mechanistic Model for prediction of heat transfer in a PCFB2.3. Table form of literature review2.4. Summary of the literature review

CHAPTER 3: OBJECTIVES OF THE WORK AND TIME ACTIVITY SCHEDULE

CHAPTER 4: METHODOLOGY

CHAPTER 5: WORK DONE DO FAR5.1. Preliminary experiment on the PCFB setup and observations5.2. Design of the distributor plate.5.3. Results and discussions.

CHAPTER 6: FUTURE WORK PLAN AND EXPECTED OUTCOME

REFERENCES

1

CHAPTER 1

Introduction:1.1. Fluidized Bed Technology:

There are various attractions to generate power from biomass derived fuels such as coal, biomass etc.

through fluidized bed technology [18]. Fluidized bed has emerged as an environmentally acceptable

technology for burning wide range of solid fuels to generate steam and electricity power because of

its unique in situ capture of SO2 and NOx. It is a compact, cheap and efficient method of using low

grade coals which are either difficult to be used or not possible to use in conventional processes [2].

Fluidization is a phenomenon where a granular material, such as sand, transforms from a solid-like

state into a fluid-like state [2]. Fluidization occurs when a fluid, liquid or gas, is passed up through

the granular material. Depending on the flow rate, the properties of the particles and the type of the

fluid several different states of fluidization are possible. If the flow rate is low, the fluid merely flows

through the empty spaces between the particles and the bed remains stationary. This situation is

called fixed bed. If the flow speed is increased, the drag forces between the fluid and the particles

become larger and the bed begins to expand in volume. Eventually a limit is reached where the drag

force is in balance with the gravitational force and the particles are suspended within the flow. At

this point the bed is considered to be at minimum fluidization state and the particles start to exhibit

fluid-like behaviour [2]. After the minimum fluidization point, in case of gas-solids systems, the

excess gas usually starts to form channels and bubbles within the bed. The bubbles are formed near

the gas entry points at the bottom and they rise up through the bed finally bursting up when they

reach the bed surface. On the way up the bubbles change in size and shape as they collide with other

bubbles or break up into smaller bubbles. If the bed is relatively tall and narrow, the bubbles can fill

up the entire width of the bed and the flow becomes slugged. If one continues to increase the flow

speed beyond what is needed for bubbling flow, the nature of the particle bed changes again. The bed

height increases and its surface becomes less clearly defined. The bubble shapes become more and

more distorted and quite suddenly instead of bubbling flow, one can observe very complex,

turbulent-like motion of particle strands and clusters. This state is known as turbulent fluidization

and in this regime the mixing of particles is very vigorous [2].

With even higher flow velocities than what is required for turbulent fluidization a substantial amount

of the particles start to leave the bed as they become entrained in the up-going flow. Stable operation

at this point requires either constant supply of new particles or a recirculation path for the escaped

particles. At this point the behaviour of the bed can be controlled by adjusting the feed rate of

2

particles. If the feed rate is kept relatively high, so that the particles are not completely blown away,

the bed is called a fast fluidized bed. A fast fluidized bed typically consists of a relatively dense,

turbulent bottom bed which gets more dilute higher up. The dilute region contains particle clusters

which are rapidly breaking apart and reforming [2].

With even higher flow velocities than what is required for turbulent fluidization a substantial amount

of the particles start to leave the bed as they become entrained in the up-going flow. Stable operation

at this point requires either constant supply of new particles or a recirculation path for the escaped

particles. At this point the behaviour of the bed can be controlled by adjusting the feed rate of

particles. If the feed rate is kept relatively high, so that the particles are not completely blown away,

the bed is called a fast fluidized bed. A fast fluidized bed typically consists of a relatively dense,

turbulent bottom bed which gets more dilute higher up. The dilute region contains particle clusters

which are rapidly breaking apart and reforming. The excellent mixing and good contact between the

fluid and particle phases makes fast fluidization interesting for many industrial applications [2].

A fast fluidized bed with a return channel is called a circulating fluidized bed (CFB). The particles

are fluidized by fast, up blowing air in the riser section. Some amount of the particles reach the top

of the riser, where they are separated from the up flowing gas in a cyclone. From the cyclone the

particles are dropped to the return channel. At the bottom of the channel there is a loop seal which

prevents the fluidization air entering the return channel. The particles are circulated in this fashion

until the desired reactions have been achieved. Many industrial processes also use fluidization in the

bubbling regime and these are called bubbling fluidized beds (BFB) [2].

3

Fig 1: Different regimes of fluidization [2].

1.2. B

ackground:Fritz Winkler from Germany was the first to obtain a patent for a fluidized bed hot gas generators in

1921 [3].Fluidized bed processes came into wide use in the petroleum industry in the 1940’s [1].

These processes are also extensively used in the chemical and metallurgical fields. The first CFB

boiler, designed exclusively for the supply of steam and heat, was built in the Vereingte Aluminum

Werke at Luenen, Germany in 1982 [18]. This plant generated 84 MW total (9 MW electricity, 31

MW process steam, 44 MW molten salt melt) by burning low-grade coal washery residues in the

presence of limestone [3]. Today there are a number of CFB plants operating all over the world and

size of the plant installations are increasing. The world’s largest power plant (Lagisza) of capacity

460MWe began commercial operation in 2009, which marked the beginning of a new era in the

evolution of this technology. This is also the world’s first once through unit supercritical CFB boiler

[2]. Along with the commercial CFB units, various laboratory and pilot CFB units are developed to

investigate and optimize the operating parameters for higher efficiencies at lower emission level.

Continued research and development in the field of fluidization has led to the development of

pressurized bubbling fluidized bed (PCFB). The first pilot PCFB plant of capacity 10MW was

developed by Ahlstrom at Karhula, Finland in 1989 [18]. Due to multiple advantages of CFB

technology such as multi-fuel flexibility, compactness, uniform temperature throughout etc. CFB

gasification has also been developed side by side with CFB technology. The first commercial

atmospheric gasifier was installed in 1983 in Jacobstad, Finland [18]. In 1991, Foster Wheeler

supplied a PCFB gasification system to a 17 MW biomass-fired Integrated Gasification Combined

Cycle (IGCC) demonstration project a Varnamo, Sweden and it is the first biomass fired IGCC plant

in the world [18]. The potential of fluidized bed technology appears to be almost unlimited. As the

technology refinement opens up new vistas, PCFB power generation holds a greater promise to

generate power in a most cost effective and environment friendly way.

4

Finally, if flow rate is high enough and if the

particle concentration is kept very low, i.e. the

volume fraction less than 1%, the situation is called

pneumatic transport. Under pneumatic transport

there is practically no interaction between the

particles. Pneumatic transport can be used for

instance to transport solids from one container to

another [2].

Fig 2: Schematic diagram of circulating fluidized bed [2]

1.3. Pressurized Circulating Fluidized Bed:Pressurized circulating fluidized bed (PCFB) combustor technology is a recent extension of the

fluidized bed technology family [1]. This type of circulating fluidized bed combustor operates at

elevated pressures, but still maintains all the advantages of the circulating fluidized bed system [18].

This technology is relatively new and is only at demonstration stage. An application of PCFB

combustor technology to combined cycle power generation has opened up an attractive alternative

due to its (1) high overall efficiency, (2) low emission and improved combustion, (3) good heat

transfer characteristics, (4) compact furnace size and (5) fuel flexibility. However, in order to fully

comprehend the advantages of pressurized circulating fluidized bed combustion, it is important to

understand the effect of operating pressure on different design parameters [18]. The bed-to-wall heat

transfer coefficient is one of the important parameters whose knowledge is required during both the

design and operating phase of a boiler [3]. There are two types of pressurized fluidized beds:

bubbling and circulating. While pressurized bubbling bed designs have been developed and are in

commercial operations, the circulating fluidized bed (CFB) designs are still in the pilot stage.

The PCFB unit, in general consists of a riser, a cyclone separator, a return leg called downcomer, a

loop seal and a feeding system. The performance of a PCFB unit is influenced by a number of

factors, including superficial velocity, solid circulation rate, solid inventory, and particle size

distribution [4]. Change of any of these parameters changes the bed hydrodynamics such as bed

voidage, and suspension density, and this causes a change in the heat transfer along the bed height.

Many researchers have reviewed the bed hydrodynamics and heat transfer at atmospheric conditions

[18].

1.4. Motivation:As mentioned earlier PCFB units have many advantages and the understanding of bed

hydrodynamics and heat transfer characteristics is very important for the proper design of PCFB

units. Very limited research has reported the effect of various parameters in a PCFB. In this view, the

present work concentrates in studying and investigating the effect of various operating parameters on

the hydrodynamic and heat transfer characteristics in a PCFB riser (laboratory unit).

5

CHAPTER 2

Literature Review:A critical review of the available literature on the bed hydrodynamics heat transfer characteristics in

pressurized fluidized bed units which are relevant to the scope of the present study are presented in

this chapter. The summary of the literature review is also presented at the end of the chapter.

2.1. Hydrodynamics and Heat Transfer in a CFB:For specific design of a fluidized bed the bed hydrodynamic studies are very essential as these would

provide information on basic flow patterns, mixing, particle attrition behavior, and heat and mass

transfer characteristics. The hydrodynamic condition influences the parameters like auxiliary power

consumption, heat absorption, temperature distribution, combustion condition, bed inventory and

erosion. The mechanism of heat transfer in a fluidized bed is very complicated and is quite different

from the usual system as it involves a large of number of variables and heavily depends on bed

hydrodynamics. Glicksman (1988) stated that the important hydrodynamic factors in a CFB are the

fraction of the wall covered by particles and gas and average contact time of particles at the wall.

The work of Weinstien et al. (1983) has confirmed the existence of core-annulus structure of the

dense phase of the fast fluidized bed. In simple terms, there exists a relatively dilute upflow core in

which solid particles are entrained upward by a high velocity gas stream and much denser annulus

layer near the column wall in which solid particles congregate and fall as dense structures similar to

waves of strands or streamers. As the heat transfer surfaces are commonly located at the column wall

for most CFB applications, the influence of down flowing wall layer on heat transfer is significant.

The velocity of descent of strands in the wall layer, the duration of their stay at the wall and the time

fraction of wall coverage are all important hydrodynamic parameters that affect the heat transfer

between the gas solids suspension and the wall. Glicksman (1988) considered voidage as the volume

fraction of the bed occupied by bubbles. The bed voidage ( ) at any cross-section has been estimated

from the measured pressure drop ( ) from a differential water filled U-tube manometer connected

across it.

=Voidage

1- =Fractional volume of the bed occupied by solid particles

For a fluidized bed, the pressure drop across the bed equals the weight of the bed, the fluid drag is written as

6

From this equation

Since, ρs >ρg , we may write,

Again,

Where : density of water = 1000 Kg/m3

: Difference of height in manometer fluid in cm of water.

Therefore from these equations we have,

Or,

Or, or

The axial variation of voidage is typically a flattened ‘S’ profile. At the bottom of the riser solid

fraction varies from 0.2 to 0.4, whereby it decreases and remains almost constant throughout the

height of the riser.

The amount of solids density at a particular section of the bed may be defined as suspension

7

Fig 3: Axial voidage profile [2]

density. Mathematically, it can be expressed as,

ρ sus=(1−ε ) ρs+ ε ρg

It is reported that, the particle size and the solid inventory have a substantial influence on the

suspension density profile in a CFB furnace (Basu, 2006). It is claimed that with the decrease in

particle size, the suspension density increases. One can influence the suspension density by

changing the bed inventory and this was suggested by Yue et al. (2005).The suspension density

along the height of a CFB boiler varies exponentially as it does in the freeboard region of a

bubbling fluidized bed (Liand Kwauk, 1980 and Kunii and Levenspiel, 1991).

Minimum fluidization velocity is the basic information required for the design and development of

a fluidized bed. Many researchers have studied how an increased pressure effects the minimum

fluidization velocity. They have experimentally found that the effect of pressure on the minimum

fluidization velocity depends on the particle size. The results of their experiments show a clear

decrease in the minimum fluidization velocity with increasing pressure for particles larger than

100µm. To estimate the effect of pressure on minimum fluidization velocity, Ergun (1952) has

developed an expression. The pressure drop per unit height of a packed bed of uniformly sized

particles, is correlated as

Type equation here .

For a fluidized bed, the pressure drop across the bed equals the weight of the bed, the fluid drag is

written as

The minimum fluid velocity at which the bed just becomes fluidized (Umf) may be obtained by

solving these two eq. simultaneously to obtain

The values of the empirical constants C1 and C2 as taken from experiments are 27.2 and 0.0408,

respectively (Grace, 1982).

8

Where Archimedes number,

Fig 4: Schematic of the mechanism of heat transfer

In a CFB, fine solid particles (Geldart Group A and B) agglomerate and form clusters or strands in

a continuum of generally up-flowing gas containing sparsely dispersed solids. The continuum is

called the dispersed phase, while the agglomerates are called the cluster phase. The majority of the

bed particles move upwards through the core of the bed, but flow downwards along the wall in the

form of clusters of particles or strands.

The heat transfer to the furnace wall occurs through

conduction from particle clusters, convection from

the dispersed phase, and radiation from both phases.

Solid clusters sliding down the wall experience

unsteady-state heat conduction to the wall. The

clusters cool down while losing heat to the wall

through conduction and radiation.

If f is the average fraction of the wall area covered

by clusters, the time-averaged overall heat transfer coefficient, h may be written as the sum of

convective, hconv and radiative, hr heat transfer coefficients (Subbarao and Basu, 1986; Dutta and

Basu, 2004):

h=hconv +hr = f(hc+hcr)+(1 - f)(hd+hdr)

where, hc and hd are the convective heat-transfer coefficients due to the cluster and dispersed

phase, respectively. The heat-transfer coefficients hcr and hdr are the radiative contributions of the

cluster and dispersed phase, respectively.

2.2. Mechanistic Model for prediction of heat transfer in a PCFB:

The bed-to-wall heat transfer coefficient is one of the important parameters whose knowledge is

required during both the design and operating phase of a boiler. A PCFB operates at several

atmospheric pressures. Higher gas densities under higher system pressures may have a major

bearing on the heat transfer coefficients. For the sake of scale up and parametric study, Basu et al.

(1996) developed a mechanistic model by extending an earlier model developed for atmospheric

pressure. Then on the basis of experiments conducted, an empirical relation is also developed for

prediction of heat transfer coefficients, within the present range of experiments. The riser of a

CFB operating under a fast fluidized regime comprises two phases: clusters and dispersed phase.

At any instant the wall of the bed is covered partially by clusters and partially by dispersed phase.

Thus the bed-to-wall convective heat transfer coefficient, hconv would have two components: one

9

due to the clusters hc and other due to the dispersed phase, hd. The fraction of the wall covered by

the clusters δc may be estimated as

δ c=K ¿¿

Where K = 0.5, ε wis the voidage near the wall, which can be obtained by ε w=ε (r i=R )=ε3.811,

ε c is the voidage within the clusters. The average solid concentration in the clusters at the wall

(1- ε c) of the bed was derived by Lints from the capacitance probe data of Wu et al. (1989) and

Longer et al.(1990) [1] as

1−εc=1.23(1−ε )0.54

Where ε is the cross-section average voidage of the solids. The volume fraction of the solid in the

dispersed phase Y may be taken as 0.001%.

Thus the bed-to-wall convective heat transfer coefficient hconv can be expressed as

hconv=δc hc+(1−δ c)hd

The convective heat transfer coefficient due to particles and clusters may be written as

hc=1

dp

10 K gf+( t c π

4 K c C c ρc)

0.5

where d p is the particle diameter, K gf is the thermal conductivity of the gas film evaluated at mean

gas film temperature, t c is the average residual time of clusters on the wall.

K c ,C c∧ρ c are the thermal conductivity, specific heat and density of cluster respectively.

C c=[ (1−εc ) Cp+εc Cg ] and ρc=[ (1−εc) ρp+εc ρg ] and the thermal conductivity of the cluster is

taken from the following equation

K c=K g( K p

K g )[0.28−0.757 log10 ε c−0.057 log10( K p

K g)]+0.1 ρg C g d pUmf

The gas thermal conductivity K gmay be assumed independent of pressure because it changes by

less than 2% whereas the pressure changes from 100 kPa 1000 kPa at the same bed temperature.

The minimum fluidization velocity Umf may be predicted by the Ergun equation. The maximum

fall velocity of clusters Umf is taken as 1.26 ms-2 in the model calculation. The convective heat

transfer due to the upflowing dispersed phase as given by Basu for atmospheric pressure could not

10

reflect the effect of the pressure very well. Xavier and Davidson (1985) and Martin (1985) found

that the gas convective heat transfer coefficient varies approximately as the square root of the gas

density. So the equation developed by Wen and Miller in 1961 used by Basu was modified as

hd=Kg

d p

C p

Cg( ρdis

ρp)

0.3

( U t

g d p)

0.21

pr ( ρg

ρgo)

0.2

Where the density of the dispersed phase is ρgY+ρg(1−Y ) and Ut is the terminal velocity of the

particle.

2.3. Table form of literature review:TITLE AUTHORS WORK CARRIED OUT OBSERVATIONS

REPORTEDHeat Transfer in a Pressurized Circulating Fluidized Bed Boiler

Basu and Cheng(1995)

Studied the heat transfer in Pressurized Circulating Fluidized bed. In their study they have modified the cluster renewal model for atmospheric circulating fluidized bed to account for the effect of pressure on heat transfer. Bed-to-wall heat transfer coefficients at different operating pressures were measured. The effects of system pressure, bed suspension density, particle size and superficial gas velocity were investigated in the tests. In addition to the modified model, a semi-empirical equation based on the test data is proposed for the prediction of heat transfer coefficients

(1) The bed-to-wall convective heat transfer coefficient increases with increasing system pressures and bed suspension density, but not with particle size for a short heat transfer surface. The effect of superficial gas velocity on the heat transfer coefficient is negligible.(2) The effect of system pressure on the heat transfer coefficient can be explained by its effect on the gas density and cluster thermal conductivity.(3) The modified cluster renewal model predicts the heat transfer coefficient with a reasonable accuracy.(4) An empirical relation is proposed to predict the heat transfer coefficient at room temperature.

Estimation of the Effect of the system pressure and CO2

concentration on

Reddy and Basu (2002)

Reports the effect of CO2

and system pressure on radiation heat transfer in a pressurized circulating fluidized combustor.

(1) They have investigated that for a given CO2

concentration in a pressurized circulating fluidized bed combustor, radiation heat

11

radiation heat transfer in a PCFB Combustor.

transfer increases slightly with system pressure due to increased gas partial pressure and gas emissivity.(2) The results demonstrate that for the given CO2

concentration, the effect of system pressure on radiation heat transfer is minimal in a pressurized circulating fluidized bed (PCFB) combustor.(3) The effect of variation in CO2 concentration levels on bed to wall radiation heat transfer during combustion in a pressurized circulating fluidized combustor is not significant for a narrow combustor.

Bed to wall heat transfer behavior in a PCFB

Gupta and Nag (1998)

An experimental investigation has been made to study the effect of pressure and other relevant operating parameters on bed hydrodynamics and bed-to-wall heat transfer in a pressurized circulating fluidized bed (PCFB) riser column of 37.5 mm internal diameter and 1940 mm height. A specially designed heat transfer probe is used to measure the bed-to-wall heat transfer co-efficient.

(1) The axial bed voidage along the height of the bed is observed to be less in the bottom zone and is high in the top zone.(2) It is also observed that the bed voidage increases in the bottom zone and decreases in the top zone with increase in operating pressure.(3) The heat transfer co-efficient is found to be increasing with the increase in operating pressure as well as increase in gas superficial velocity. It also increases monotonically with increasing bed temperature. The data show trends similar to those in published literature.

Hydrodynamics, erosion and heat transfer in a

Wiman and Almstedt

Measurements of hydrodynamics, local tube erosion and local

(1).Results shows that the hydrodynamic behavior is similar for two different

12

pressurized fluidized bed: influence of pressure, fluidization velocity, particle size and tube bank geometry

(1997) instantaneous bed-to tube heat transfer were carried out in a cold pressurized fluidized bed, with two horizontal tube banks having different tube packings. The influence of pressure, fluidization velocity,particle size and tube bank geometry was studied. Two size distributions of silica sand were used, one with a mean particle diameter of dp = 0.7 mm and one with dp = 0.45 mm. The bed has a cross-section of 0.2 m x 0.3 m, and was operated at pressures between 0.t and 1.6 MPa and at excess gas velocities of 0.2 and 0.6 m/s.

particle sizes if plotted against the excess gas velocity, but differs significantly between the two tube banks.(2).The smaller particles generally give rise to less erosion than the larger particles, as an effect of their momentum being lower at a given particle velocity.(3). The small particles also give a higher heat transfer than the large particles, as a result of a higher particle convection.(4). The denser tube bank causes an earlier transition to a turbulent bed behavior with increasing pressure or fluidization velocity. The dense tube bank gives rise to considerably less erosion but also gives a somewhat lower heat transfer than the more sparse tube bank, at corresponding operating conditions.(5). For the sparse tube bank investigated, at high pressures, the erosion decreases with increasing pressure. The bed-to-tube heat transfer coefficient generally increases with increasing pressure. Thus, it should be favorable to operate a bed at high pressure levels.

A model for heat transfer in a pressurized circulating fluidized bed furnace.

Reddy and Basu (2000)

Developed a mechanistic model based on cluster renewal approach to predict bed to wall heat transfer in pressurized circulating fluidized bed (PCFB). The

(1). Heat transfer coefficient increases with system pressure due to increased particle and dispersed phase convection heat transfer coefficient, as a result of

13

model takes into account the effect of pressure on cluster density, cluster thermal conductivity and particle convection heat transfer coefficient. The effect of bed pressure and temperature on bed to wall heat transfer coefficient is investigated. The model predictions are validated against the experimental data obtained from a PCFB riser of 52.4 mm in diameter and 2020 mm height. The experimental results are reported for pressures up to 600 KPa and bed temperature up to 6500C.

effect of pressure on cluster thermal conductivity, capacity and residence time.(2). Increase in suspension density results in higher cluster solid fraction and particle concentration near the wall, which results in higher heat transfer coefficient.

Effect of pressure and carbon dioxide concentration on heat transfer at high temperature in a PCFB combustor.

Winaya and Basu (2000)

Experiments are performed in an electrically heated 52.5 mm and 2020 mm high pressurized circulating fluidized bed to investigate the bed to wall heat transfer in it. The bed-to-wall heat transfer is determined from the radial temperature distribution in the wall. Carbon di-oxide content of the fluidizing gas is changed by either adding varying amounts of carbon dioxide to the air or burning coal and coke with and without addition of adding limestone.

Heat transfer coefficient increase with both system pressure and bed temperature due to increased contribution of gas convection and radiation. Heat transfer coefficient also increase with carbon dioxide concentration due to increased non-luminous radiation from carbon dioxide.

Effect of pressure on loop seal operation for a pressurized circulatingfluidized bed

Cheng and Basu (1999)

Studied the solids flow through a loop seal in a pressurized circulating fluidized bed (CFB) in an experimental setup in which the effects of system pressure, aeration rates in the two chambers of the loop

The results show that the solids flow rate increases with aeration rate as well as the system pressure. However solids are not recycled until the aeration rate reaches a threshold limit. The solids flow rate also increases with

14

seal, and particle size could be investigated.

decreasing particle size. Analysis of pressure drops through the standpipe, loop seal opening, recycle chamber and recycle pipe verifies that the recycle flow rate of the loop seal can be determined by an analysis of the pressure balance around the CFB loop.

Parametric study of hydrodynamics and heat transfer along the riser of a pressurized circulating fluidized bed unit.

Kalita et al.(2012)

In the present investigation, the effects of particle size, solid inventory and operating pressure on bed hydrodynamics and heat transfer in a pressurized circulating fluidized bed has been studied. Three different particle sizes of mean diameter of 278, 307 and 469 µm are considered. Experiments have been conducted at five different solid inventories such as 400, 500, 600, 800, and 1000 g, and at three different operating pressures such as 1, 3 and 5 bar. All the above studies have been made at three different superficial velocities of 5, 6 and 7 m/s. The axial variation of heat transfer coefficient with various percentage blending of sawdust in sand has also been studied.

Results show that, with the increase in operating pressure and solid inventory, the bed voidage decreases. The solid circulation rate is found to increase with an increase in superficial velocity as well as with an increase in solid inventory. The heat transfer coefficient decreases with an increase in particle size, while it increases with operating pressure from the bottom to the top of heat transfer probe. The heat transfer coefficient is also found to increase with the increase in % blending of sawdust in sand and the operating pressures.

Some studies on wall-to-bed heat transfer in a pressurized circulatingFluidized bed unit

Kalita et al. (2013)

In the present work, a pressurized circulating fluidized bed (PCFB) unit of 54 mm inner diameter and riser height of 2000 mm has been fabricated to investigate the effect of pressure on suspension density and heat transfer.

Results show that, the axial heat transfer coefficient increases from the bottom to the top of heat transfer probe with the increase in operating pressure. The radial variation of heat transfer coefficient decreases from the wall to the core of the heat transfer

15

The effects of blending of biomass in sand, and superficial velocity on bed hydrodynamics and heat transfer has also been studied. Experiments have been conducted at four different percentage blending of biomass such as 2.5 %, 7.5 %, 15 % and 20 % in sand with two different weight composition ratios and at a superficial velocity of 5 m/s. Operating pressure is varied from 1 to 5 bar in a step of 2 bar.

probe. The heat transfer coefficient is also found to be higher in between the 7.5 to 15 % biomass blending in sand.

Simulation and experimental studies on fluidization properties in a pressurized jetting fluidized bed

Cao.et al.(2008) The influence of pressure on the bubble size and average bed voidage has been investigated experimentally and computationally in a circular three-dimensional cold-flow model of pressurized jetting fluidized bed of 0.2 m i.d and 0.6 m in height with a central jet and a conical distributor, which roughly stands for the ash-agglomerating fluidized bed coal gasifier. The pressurized average bed voidage and bubble size in the jetting fluidized bed were investigated by using electrical capacitance tomography (ECT) technique. The time-averaged cross-sectional solids concentration distribution in the fluidized bed was recorded. The influence of pressure on the size of bubble and the average bed voidage in a

(1) Both experimental and theoretical results clearly indicate that there is, at the lower pressure, a small initial increase in bubble size decided by voidage and then a decrease with a further increase in pressure.(2) The average bed voidage increases gradually with the pressure at the same gas velocity.(3) There is a disagreement in the average bed voidage between the simulation results and experimental results as the pressure and gas velocity increase.(4) The change of average bed voidage is connected with the drag between gas and solid. The increase in drag may be because of the increase in the density of the gas, which increases with pressure.

16

pressurized fluidized bed was studied.

Effects of distributor design on solids distribution in a CFB riser.

Jing Xu et al.(2010)

The influence of gas distributor design on the gas-solids flow structure was investigated in a rectangular CFB riser. The gas distributor was altered five different ways. Five types of gas inlet arrangements were employed: (1) fully opened; (2) left opened; (3) right opened; (4) sides opened; (5) center opened. The solids distribution, in terms of solids concentration and particle velocity, was measured by optical fiber probes and investigated experimentally and numerically. The fully opened distributor was considered as the original condition, the other four types of special designed distributors were compared with the unaltered one.

The solids concentration and particle velocity along the lateral direction were quite uniform for fully opened distributor, while the left side opened or the right side opened gas inlet structure caused the profiles of solids concentration to be half-dilute half-dense, and the particle velocity to be half-high half-low; the center opened distributors leads to symmetric center-dilute wall-dense distribution of solids concentration and center-fast wall-slow pattern of particle velocity, whereas, reversed profiles of concentration and velocity were produced by the sides opened distributor. The results also show that the profile of the gas is the main reason dominating the solids distribution in the CFB riser, accounting for the profiles of solids concentration and particle velocity. The distribution of the gas dictates the flow structure in the riser column and might have more significant effects over the “inherent” wall effect.

The influence of distributor design on fluidized bed dryer hydrodynamics

Wormsbecker et al. (2007)

The influence of perforated plate, punched plate and Dutch weave mesh distributor designs on fluidized bed dryer hydrodynamics was studied for a range of bed loadings and superficial gas velocities.

The present study has found that the punched plate distributor design provides shorter drying times than the Dutch weave and perforated plate designs when wet bed loadings of 2.5 kg and above are dried at superficial gas velocities typical of the

17

pharmaceutical industry (1.0 to 1.5 m/s). This study was carried out on a laboratory scale fluidized bed dryer. It is possible that in larger diameter fluidized beds, such as clinical and production scale dryers, the positive influence of the punched plate may be more pronounced as bed loadings, and therefore bed depths, become greater.

Mixing patterns and residence time determination in a bubbling fluidized bedSystem.

Ghaly and Macdonald (2012)

Investigated the effects of sand particle size, distributor plate shape and angle, bed height and fluidizing velocity on particle mixing and residence time in the fluidized bed reactor.

(1) Greater values of the residence time were obtained with course sand whereas lower values were obtained with fine sand.(2) An increase in the angle of convex or a decrease in the angle of a concave of the distributor plate resulted in an increase in the residence time.(3) To improve the mixing properties of the binary mixture, which has great tendency for segregation due to density differences, an angled distributor plate (concave or convex) should be used.

Distributor effects near the bottom region of a turbulent fluidized beds.

Celia sobrino, Naoko ellis and Mercedes de vega (2008)

The distributor plate effects on the hydrodynamic characteristics of turbulent fluidized beds are investigated by obtaining measurements of pressure and radial voidage profiles in a column diameter of 0.29 m with Group A particles using bubble bubble-cap or perforated plate distributors. Distributor pressure drop measurements between the two distributors are

The perforated plate presented a lower Uc velocity and a higher decrease of the dense bed height with increasing the superficial gas velocity. This indicates that the rate of solids transferred from the dense bed to the free board is higher for the perforated plate, and begins at a lower superficial gas velocity. From the pressure measurement results, it is concluded that the solids

18

compared with the theoretical estimations while the influence of the mass inventory is studied. Comparison is established for the transition velocity from bubbling to turbulent regime, Uc, deduced from the pressure fluctuations in the bed using gauge pressure measurements. The effect of the distributor on the flow structure near the bottom region of the bed is studied using differential and gauge pressure transducers located at different axial positions along the bed. The radial voidage profile in the bed is also measured using optical fiber probes, which provide local measurements of the voidage at different heights above the distributor.

density near the bottom region is higher for the bubble-cap distributors; while a more homogeneous radial structure in terms of voidage is found. The results are in accordance with time mean average voidage obtained with optical probes; whereby, the voidage measured with these probes was also found to be smaller for all radial positions and the profiles obtained were flatter when using the bubble-cap distributor. A dilute core and a denser annulus structure was observed in the bottom region of the bed for the two distributors studied. Radial voidage profiles in the dense bed were found to be fitted by a quadratic profile, and presented similarity for different fluidizing velocities.

the influence of the distributor plate on thebottom zone of a fluidized bed approachingthe transition from bubbling toturbulent fluidization

j. m. paiva

c. pinho

r. figueiredo2004

The dynamics of the bottom zone of a narrow fluidized bed, from bubbling to turbulent regimes, was studied in a cold model of 0.1 m i.d. and 1.3 m high. Tested distributor types were perforated perspex plates, with six different perforation grids, metallic mesh and porous ceramic, with pressures drops ranging from 0.05 to 350 kPa, corresponding to superficial air velocities from 0.1 to 2.3 m sGroup B silica ballotini, within the range 0.355–

The results show that the flowchanges with variation of the operating conditions. Themeasurements at several different equal heights showchanges in the hydrodynamic behaviour of the gas–solidsuspension. For the tested group B particles and static bed height, there is no linear decrease, or even a sustaineduniform trend towards a decrease of the solids concentration with height in that region of the bed. Furthermore, using amodel for the calculation of the fraction of bubbles, the

19

0.425 mm, were used as bed material. The experimental data consisted of pressure drop and absolute pressure fluctuating signals, together with visual observations.

results include the detection of a fluctuation in the values of the porosity of the emulsion, with height and with the fluidization velocity.

Effect of Distributors onFluidised BedsHeat Transfer from Immersed Surfaces in Gas

d. sathiyamoorthy, ch. sridhar rao and m. raja rao(1987)

Heat transfer coefficients from a smooth tube of internal diameter 26 mm immersed horizontally in gas fluidised solids in theoverall size range 70 - 161 urn have been mea-sured as a function of gas velocity using multi- orifice distributors of various free areas (0.0888% - 0.52%).

The optimum gas velocity Uopt corresponding to the maximum heat transfercoefficient h,,, and the gas velocity U, at which all orifices in the distributor just become operational are found to be the same.For gas velocities U just below Uopt the wall heat transfer coefficient has been found to be influenced by the free area of the distributor, while the effective bed porosity remains close to the incipient state of fluidisation with a slight increase near optimum velocity. Correlations to predict Reopt and the corresponding Nu,,, have been proposed. An attempt has been made to explore the effects of distributor on heat transfer, but without a rigorous supporting theoretical analysis.

The Design of Distributors for G .as-Fluidized •-Beds

Geldart 1985 The performance of the gas distributor often determines the success or fai!ure of a fluidized bed and although much more is known now than 20 years ago, there are still many pitfalls for the designer. Particle and gas

The current state of the art .is reviewed in the light of recent research and industrial experience.

20

properties play a key, olein successful design together with the criticalpressure drop ratio, and hole size, geometry and spacing; these strongly influence jet penetration, dead zones, particle sifting,attrition and mixing.

Effect of Distributors on Gas-Solid Fluidization

s. c. saxena,a.chatterjee and r. c. patel1979

The efficient operation of a fluidized bed is very much dependent upon distributor performance, which in turn depends on its design parameters. The work reported here dealswith the characteristics of such distributors as are commonly employed in laboratories, pilot plant and large scale operations. Specifically, a porous plate distributor, two bubble cap distributors of different geometries and fourJohnson screen distributors of different percent open area have been investigated in a 30.5 cm by 30.5 cm square fluidized bed as a function of air fluidizing velocity and bed height. The pressure drop data for all the distributors have been correlated by a single equation with two unknown constants.

The distributor pressure drop was found to increase with fluidizing velocity, to decrease with percentage open area of the distributorplate, and to be independent of the bed weight or height for a given distributor design.

The pressure ratio, APd/APb, was found to increase rapidly with increase in fluidization velocity. The value of this pressure ratio at minimum fluidization was found to be dependent on the bed height, and its degree depends sensitively on distributor design,

The bed expansion ratio was found to increase with the increase in excess fluidizing velocity and decrease -with bed height. The quantitative magnitudes of these variations are dependent on the distributor design. A distributor with greater pressure drop across it gives rise to smaller bubbles for the same ex-cess fluidizing velocity than

21

a distributor with smaller pressure drop [15]. This is substantiated by the present measurements of bed ex- pansion ratio for the bubble cap, porous plate and Johnson screen distributors

Effect of Distributor Design on Heat Transfer from an Immersed Horizontal Tube in a Fluidized Bed

Grewal et al. 1979

Experiments have been carried out to examine the effect of distributor design on the total heat transfer coefficient between a 12.7 mm diameter electrically heated copper tube and a square fluidized bed (0.305 m X 0.305 m) of glass beads (a,, = 265, 357 and427 pm). The distributors are two perforated plates of different open area with a cloth sandwiched between them. The effect of distributor design on the total heat transfer coefficient is explained on the basis of its effect on the initial bubble size and frequency

2.4. Summary of the literature review:

Various literatures are available for investigation of heat transfer and hydrodynamic of an

atmospheric fluidized bed. However very limited literature is available for investigation of heat

transfer and hydrodynamics of pressurized circulating fluidized bed. In most of the cases, the

mechanistic models developed for atmospheric fluidized bed was used for prediction of heat transfer

22

Table 1: Literature Review

in pressurized fluidized bed. Some of the important observations related to heat transfer and

hydrodynamics of pressurized fluidized bed is presented below:

(1) With increase in operating pressure the bed voidage is increases in the bottom zone and decreases

in the top zone.

(2) The heat transfer coefficient is found to be increasing with the increase in operating pressure as

well as increase in gas superficial velocity.

(3) The minimum fluidization velocity, the bubbling velocity and the terminal velocity were

decreasing with increase of operating pressure

(4) The modified cluster renewal model predicts the heat transfer coefficient with a reasonable

accuracy. Increase in suspension density results in higher cluster solid fraction and particle

concentration near the wall, which results in higher heat transfer coefficient.

(5) The bed-to-wall convective heat transfer coefficient increases with increasing system pressures

and bed suspension density

(6) For a given CO2 concentration in a pressurized circulating fluidized bed combustor, radiation

heat transfer increases slightly with system pressure due to increased gas partial pressure and gas

emissivity.

(7) The distribution of the gas dictates the flow structure in the riser column.

(8) An increase in the angle of convex or a decrease in the angle of a concave of the distributor plate

resulted in an increase in the residence time.

CHAPTER 3

Objectives of the work and Time activity Schedule:

3.1. Objectives:In the prospect of the above discussion, the aim of the present work is to study the effects of various

operating parameters on the hydrodynamics and heat transfer characteristics in a PCFB laboratory

scale unit. The following tasks are taken as objectives for this study:

23

1. To investigate the variation of bed hydrodynamics along the riser and heat transfer

characteristics along the upper splash region of the riser under different heat flux and

operating pressure. 2. To fabricate the designed distributor plate.

3. To investigate the variation of the above mentioned parameters by using the designed gas

distributor plate and comparing the results with the straight hole distributor plate.

4. To perform a CFD simulation of the above experiment in a commercial CFD software Ansys

Fluent.

3.2. Time Activity Schedule:

WORK TIME Aug -

Sept.

Oct-

Nov.

Dec. Dec-

Jan.

January

– March.

April-

June.Literature review.

Learning of the CATIA and ANSYS

technical software.

Design of the distributor plate

Start of progress report and presentation.

Fabrication of the distributor plate.

Start of experimentation and data analysis.

CFD simulation of the experiment conducted and comparison

Thesis writing and final Submission

CHAPTER 4

Methodology:The following methodology is chosen to fulfill the objectives:-

1. Experiments will be conducted on a PCFB laboratory scale setup of 54 mm internal riser

diameter and 2000 mm height which is installed at the Energy Efficiency Laboratory, Center

for Energy, IITG.

24

Table 2: Time activity schedule

2. The experiments will be conducted by using two types of air distribution grid: (a) straight

hole distributor plate and (b) newly designed distributor plate.

3. The experiments will be performed at different superficial velocity, different heat fluxes and

different operating pressures.

4. The bed inventory will be ranging from 1.0 kg to 2.0 kg.

5. Sand particles of average size 300µm will be used as bed material.

6. The bed hydrodynamics and heat transfer characteristics will be investigated by using both

types of distributor plates at the bottom of the riser and finally the obtained results will be

compared.

7. The working formulas that will be used are:

a) Bed voidage ,

b) Suspension Density,ρ sus=(1−ε ) ρs+ ε ρg

c) Superficial Velocity, U ¿=ma

ρg X Ab

d) Solid circulation rate, Gs=ρ s X La X Ad X (1−εmf )

Ab t

e) Wall to bed heat transfer coefficient,h= V∗IAhtp∗(T bs−T bi)

Where,

ma = mass flow rate of the air through the orifice in kg/s, ε = bed voidage.

L= difference between two consecutive pressure taps across which pressure drops.

ρs= density of sand in kg/m3 , 𝛥hL= difference of height in manometric fluid measured in

cm of water column, ρg= density of air in kg/m3, Ab= cross-sectional area of the bed in m2.

La= solid accumulation height in m, Ad= cross-sectional area of the downcomer in m2.

ε mf= voidage at minimum fluidization, t= the time to accumulate particular height after

closing ball valve in s, V, I, Ahtp, Tbs and Tbi are the voltage supply, current supply, heat

transfer probe surface area, bulk surface temperatures and bed temperatures, respectively.

25

8. CFD simulation of the riser will be performed in a commercial CFD simulation software

named Ansys FLUENT and the results will be compared with the experimental results. All

CFD codes contain three main elements. a) Pre-processing b) Solver and c) Post-processing.

a). In Pre-processing, it consists of input of a flow problem by means of an operator friend

interface and subsequent transformation of this input into a suitable form which can be used

by the solver. This step is performed by software tools such a GAMBIT, TGRID and DM.

The pre-processing stage involves the following steps (Bakker, 2002).

Defining the geometry of the region for computational domain.

Generating the grids for subdivision of the domain into a number of smaller, non-

overlapping subdomains.

Specifying the appropriate boundary and continuum conditions at each cells, which

coincide with or touch the boundary.

b). The CFD solver does the flow calculations and produces the desired results. FLUENT

uses the finite-volume method to solve the governing equations for a fluid. It provides the

capability to use different physical models such as incompressible or compressible, inviscid

or viscous, lminar or turbulent,etc. Governing equations are non-linear and coupled, several

iterations of the loop are performed by solver before a converged solution is obtained

(Bakker, 2002). The main functions of the solver are:

Approximation of unknown flow variables by means of simple functions.

Discretization by substitution of the approximation into governing flow equations and

subsequent mathematical manipulations.

Solving the algebraic equations.

c) Post-processing is the final step in CFD analysis, and it involves the results and

interpretation of the predicted flow data.

CHAPTER 5

Work Done So Far:5.1. Preliminary experiment on the PCFB setup and observations:A trial experiment has been performed on the PCFB setup which is installed in the Energy Efficiency

Laboratory, Center for Energy, IITG. This experiment was performed to investigate the bed

26

hydrodynamics with 400gm of sand particles of average diameter 300µm as bed inventory. The

figure below shows the schematic diagram of the experimental setup.

The riser is made of stainless steel of ID 54 mm, height of 2000 mm and thickness of 3 mm. The

PCFB unit contains a mild steel cyclone separator of barrel diameter 80 mm and height 160 mm.

Entrained solids are recovered in a cyclone separator and are then sent to the bottom of the riser

column through a transparent return leg of ID 24.5 m. Air is supplied to the CFB unit through the

bottom of the riser by a high pressure centrifugal blower and a compressor.

The following observations were recorded during the experiment:

27

Pressure tap no.

InventoryPressureRotameter

lit/secDist. From the

Distributor plate

Diff. in height of the manometric

fluidLm Voidage

Suspension Density

1 400gm 1.5 7.2 0 0 0 0 02 400gm 1.5 7.2 0.12 0.57 0.12 0.979347826 48.577282613 400gm 1.5 7.2 0.1925 0.53 0.0725 0.968215892 74.168485764 400gm 1.5 7.2 0.37 0.7 0.1775 0.982853644 40.517758735 400gm 1.5 7.2 0.495 0.7 0.125 0.975652174 57.073217396 400gm 1.5 7.2 0.97 0.8 0.475 0.992677346 17.934050347 400gm 1.5 7.2 1.57 2 0.6 0.985507246 34.4173913

Fig 5: Schematic diagram of experimental setup [13]

Table 3: Observations of the preliminary experiment

The variation of bed voidage and suspension density were observed as follows:

0 0.2 0.4 0.6 0.8 1 1.20

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8Voidage Profile

Voidage

Dist

. fro

m th

e di

strib

utor

pla

te

0 10 20 30 40 50 60 70 800

0.20.40.60.8

11.21.41.61.8

Suspension Density Profile

Suspension Density

Dist

. fro

m th

e di

strib

utor

pla

te

5.2. Design of the

distributor plate:

The distributor was designed based on design procedure by Kunni and Levenspiel (1991), and Basu

(2006). The figure below shows the different views of the designed distributor grid.

28

Fig 6: Variation of voidage along the height of the riser

Fig 7: Variation of suspension density along the height of the riser

Specifications of the design distributor grid:-

Material that will be used:- Mild steel.

Diameter of the orifice: - 2mm.

Orifice discharge co-efficient: - 0.923.

No. of holes in the distributor plate: - 193.

Thickness of the distributor plate: - 5mm.

5.3. Results and discussions:

1) The experiment on the PCFB setup was performed at an operating pressure of 1.5 bar and

superficial velocity of 6 m/s. The bed inventory was 400gms and the average particle

diameter was around 300µm. The variation of bed voidage and suspension density along the

29

Fig 8: Different views of the designed distributor plate

height of the riser is shown in Fig 6. & Fig 7. It is being observed that bed voidage is found to

be in between 0.8 to 0.9. The bed voidage increases steeply during the first 200mm of the

riser height and then it remains almost constant for the rest of the height of the riser. This

may be due to the less amount of bed inventory. The suspension density is also found to be

fluctuating in nature as it increases rapidly and then decreases and again increases and then

keeps on decreasing along the height of the riser.

2) The distributor plate was designed according to the design procedure by Kunni and

Levenspiel (1991), and Basu (2006). The details of the design procedure is more elaborately

explained reference #18. CATIA V5 software was used in designing and post processing the

diagrams showing different views.

CHAPTER 6

Conclusions:Future work plan and expected outcome:

30

(1) The next set of experiments will be conducted in the next phase of the project. We will be

increasing the bed inventory and also the operating pressure which would certainly help us to

study the bed hydrodynamics more deeply.

(2) The heat flux to the heat transfer probe will also be varied accordingly.

(3) The newly designed distributor plate will be fabricated and it will be installed in the bottom

of the riser.

(4) The experiments will be conducted by using two types of distributor plates and the results

will be compared.

(5) CFD simulation of the above experiment will be done by using a commercial CFD software

known as FLUENT and the results will be compared with the experimental results.

REFERENCES:1) http://energy.gov/fe/fluidized-bed-technology-overview

31

2) Kunii, D. and Levenspiel, O. Fluidization Engineering. Butterworth-Heinemann, 2nd edition,

1991. ISBN 0-409-90233-0.

3) Basu. P, Combustion and Gasification in Fluidized Beds, Taylor & Francis Group, CRC

Press, New York, 2006.

4) Gupta. A.V.S.S.K.S, Nag. P.K, Bed-to-wall heat transfer behavior in a pressurized circulating

fluidized bed, Int. J. Heat Mass Transfer 45 (2002) 3429–3436.

5) Basu. P, Cheng. L, Heat transfer in a pressurized circulating fluidized bed, Int. J. Heat Mass

Transfer 39 (13) (1996) 2711–2722.

6) Basu .P, Nag. P.K, Heat transfer to walls of a circulating fluidized-bed furnace, Chem. Eng.

Sci. 51 (1) (1996) 1–26.

7) Winaya Nyoman S., and Basu Prabir, “Effect of pressure and carbon dioxide concentration

on heat transfer at high temperature in a pressurized circulating fluidized bed (PCFB)

combustor”, International Journal of Heat and Mass Transfer, 44, pp.2965-2971, 2001.

8) Reddy B.V., Basu and P., “A model for heat transfer in a pressurized circulating fluidized

furnace”, International Journal of Heat and Mass, Transfer, 44, pp.2877-2887, 2001.

9) Reddy B.V., and Basu P., “Estimation of the Effect of System Pressure and CO2

Concentration on Radiation Heat Transfer in a Pressurized Circulating Fluidized Bed

Combustor”, Institution of Chemical Engineers Trans IChemE, Vol 80, Part A, March 2002.

10) Cheng Leming and Basu Prabir, “Effect of pressure on loop seal operation for a pressurized

circulating fluidized bed”, Powder Technology, 103, pp.203–211, 1999.

11) Wiman J. and Almstedt A. E., “Hydrodynamics, erosion and heat transfer in a pressurized

fluidized bed: influence of pressure, fluidization velocity, particle size and tube bank

geometry”, Chemical Engineering Science, Vol. 52, No. 16, pp. 2677-2695, 1997.

12) Cao Jiantao, Cheng Zhonghu, Fang Yitian, Jing Huimin, Huang Jiejie, and Wang Yang,

“Simulation and experimental studies on fluidization properties in a pressurized jetting

fluidized bed”, Powder Technology, 183, 127–132, 2008.

13) Kalita P., Saha U. K., and Mahanta P, Parametric study on the hydrodynamics and heat

transfer along the riser of a pressurized circulating fluidized bed unit, Experimental Thermal

and Fluid Science, 2012. doi: http://dx.doi.org/10.1016/j.expthermflusci.2012.09.001.

14) Kalita P., Saha U. K., and Mahanta P., Some studies on wall-to-bed heat transfer in a

pressurized circulating fluidized bed unit, 2013. 5th BSME International Conference on

Thermal Engineering.

32

15) Zhu et al.,.The effects of distributor design on the solids distribution in a CFB riser, The 13th

International Conference on Fluidization - New Paradigm in Fluidization Engineering, Art.

53 [2010].

16) Wormsbecker et al.: Influence of Distributor Design on Dryer Hydrodynamics, the 12th

International Conference on Fluidization - New Horizons in Fluidization Engineering, Art.

100 [2007], http://dc.engconfintl.org/fluidization_xii/100.

17) Ghaly and Macdonald, Mixing patterns and residence time determination in a bubbling

fluidized bed System. American Journal of Engineering and Applied Sciences, 2012, 5 (2),

170-183 ISSN: 1941-7020 ©2012 Science Publication doi:10.3844/ajeassp.2012.170.183

18) Kalita P., Investigation of hydrodynamics and heat transfer characteristics with biomass

blends in a PCFB, PhD. Thesis, IIT Guwahati, 2013.

33