INTRODUCTION

1.1Introduction
Heat transfer in sizeaffected domains has become one of the most widely studied areas in
thermalscienceandengineeringinrecenttimes.Theeverincreasingquestforminiaturization,in
many critical technologies, has made this a topic of considerable interest over the past decade.
The possibility of including heat sinks as an integral part of individual components, and the
fabrication and operational challenges in ensuing effective thermal management have inspired
investigatorstofocustheirattentiononunderstandingheattransferatmini/microscales.Thishas
ledtoatremendousgrowthofexperimentalaswellascomputationalresearchandtheresultant
publications in the field of micro heat exchangers, microreactors, electronics thermal
management,highheatfluxcoolingdevices,labonchipsystems,MEMSandmicrofluidicsetc.
Figure1.1presentscomparisonofvariousheattransfertechniquesbasesonthemechanisms
and working fluids used in them (Lin et al. (2002)). From this figure it is clear that highest heat
transfer rates can be achieved with phasechange processes, e.g. via pool boiling or flow boiling
mechanisms. Phasechange heat transfer is a broad field that finds applications in almost all the
engineeringdisciplines.Boilingisoneofthemostimportantliquidvaporphasechangeprocess,as
it is generally associated with very high heat transfer rates. Heterogeneous boiling is a phase
1

change process in which vapor bubbles are formed either on a heated solid surface and/or in a
superheated liquid layer adjacent to the heated surface. The practical application of boiling is
probably one of the most ancient arts, dating back to the era of the caveman. Yet, very few
scientificstudieswerereporteduntilaround1935,whenNukiyama(1934)observedtheexistence
of several modes of boiling over an electrically heated wire. Boiling is a very complex process as
well,andhasbeeninvestigatedextensivelyoverthelasthalfofthetwentiethcentury,particularly
onmacroscalesystems,asapplicabletoconventionalpowerplantoperation.

Figure1.1Heattransfercoefficientsforvariouscoolingtechnologies(Linetal.(2002)).

Flowboilingoffluidsinnarrowchannelsisoneofthemostefficienttechniquesforremoving
highheatfluxes,oftheorderof100W/cm
2
orevenhigher.Workingfluidscanbepureormixtures
of pure liquids, depending on the application. For example, many electronic cooling systems
employ pure water, ammonia (for space applications), dielectric fluids, Flourochloro fluids,
organicmixtures,etc.Anotherexampleofmixtureboilingisseeninamicroscalefuelprocessorfor
distributedhydrogengeneration,whichemploysboilers/preheatersforcatalyticsteamreforming
ofethanol(Menetal.(2007);Haryantoetal.(2005)).Becauseofconsiderablecomplexityofflow
during flow boiling of pure fluids and their mixtures, this has been a topic of intense research in
therecentpast.Aneffectivedesignofmicroheatexchangersystemsdemandspreciseknowledge
and understanding of the characteristics of fluid flow and heat transfer mechanisms on the
relevantscales.Thisisoneoftheaimsofthepresentwork.
2

1.2Whymicroheatexchangers?
Asthechannelhydraulicdiameterbecomessmaller,theratioofheattransfersurfaceareato
the fluid flow volume increases in inverse proportion to the channel hydraulic diameter. For
example,incaseofacircularchannel,thisratioisgivenby,
t
= =
t
s
2
A DL 4
v D
D L
4
(1.1)
The heat transfer coefficient also increases as the channel hydraulic diameter becomes smaller.
For fully developed laminar singlephase flows under constant temperature or constant heat flux
boundary condition, the magnitude of Nusselt number is constant; thus the dependence of heat
transfercoefficientonthechannelhydraulicdiameterisgivenby,
=
k
h Nu
D
(1.2)
Theheattransferrateperunitflowvolumeisgivenby,
A A
= =
t
s
2
2
4k TNu hA T q
v D
D L
4
(1.3)
Foragivenfluidandgiventemperaturedifference,thevolumetricheattransferratethusdepends
inverselyonthesquareofthechannelhydraulicdiameter.
The pressure drop penalty increases significantly as the channel diameter becomes smaller.
So, generally the resulting flow rates are kept low. The low flow rate, combined with a small
hydraulicdiameter,leadstolowvaluesofresultingflowReynoldsnumber,typicallyinthelaminar
flow region; thus the assumption of constant Nu is quiet reasonable for micro heat exchangers
providedtheflowisfullydeveloped.

1.3Whytoworryaboutdimensions?
Some of the conventional transport theories applicable for bulk fluid need to be revisited for
validation. There are several factors responsible for departure from the conventional theories
whenappliedtomicroscalesystems.Theyareasfollows:
(a)Changeinthefundamentalprocess
Deviationfromthecontinuumassumption,typicallyformicroscalegasflows
Slipflows:changeintheboundaryconditionatthewall,ascomparedtomacrochannels
Increasedinfluenceofsomeadditionalforces,viz.,surfacetension,electrokineticforce,etc.
3

(b)Uncertainty regarding the applicability of empirical factors derived from experiments
conductedatlargerscales
Entranceandexitlosscoefficientsforfluidflowinpipes,etc.
(c)Experimentaluncertaintiesinmeasurementsatmicroscale
Geometricaldimensions/Effectofsurfaceroughness
Ambiguousoperatingparameters
Limitationsofinstrumentation;intrusivevsnonintrusivemeasurements
Dominanceofconjugateheattransfereffects
Illdefinedoruncontrolledboundaryconditions
Before proceeding any further, it is important to define what is meant by the terms mini,
meso and micro, etc., as opposed to normal sized macro channels? Unfortunately, an explicit
agreement has not been established in the literature; there are no clear demarcations of
mini/meso/microgeometricalregimes.Theclassificationsproposedintheliteraturearebasedon
sizeaswellassomerelevantnondimensionalnumbersofinterests.Atentativeschemebasedon
the smallest channel dimension, as proposed by Kandlikar et al. (2006), is presented in Fig. 1.2
where the values of the channel diameter D are in m. In the case of noncircular channels, it is
recommendedthattheminimumchanneldimension;forexample,theshortsideofarectangular
crosssectionshouldbeusedinsteadofdiameterD.

Figure1.2Channelclassificationbasedonthesmallestchanneldimension.

Adetailedreviewofvariousproposedchannelclassificationschemesbasedonsizeaswellas
basedondimensionlessnumbersisgivenbyChengandMewes(2006).Thenotabledimensionless
numbersusedinliteratureforthispurposesare(a)Laplaceconstant(b)Confinementnumber(c)
Etvsnumber:criterionforthedominanceofsurfacetension(d)Bondnumber:ameasureofthe
relative importance of the buoyancy force to surface tension force. Additionally Cheng and Wu
(2006) proposed the Bond number, Bo, as the criteria to distinguish between microchannels,
minichannels,andmacrochannels.TheBondnumberisdefinedas,
4

( ) | |
= =
|
o
\ .
2
2
l v h h
d
c
g D D
Bo
L
(1.4)
which is a measure of the relative importance of the buoyancy body forcetosurface tension
force. Cheng and Wu (2006) distinguished between microchannels, minichannels, and
macrochannelsasfollows:
(a) Microchannels:IfBo<0.05wherethegravityeffectcanbeneglected;
(b) Minichannels:If0.05 <Bo<3.0 where thesurfacetensioneffectbecomesdominantandthe
gravitationaleffectissmall;
(c) Macrochannels: If Bo > 3.0 where the surface tension is small in comparison with the
gravitationalforce.

Figure1.3Variationofthresholddiameteraspervariousdefinitions,for(a)ethanoland(b)water.
Thereisalargevariationinthedefinitionswhichspecifytheacceptablelimitswhereinmicroscale
effectsbecomepredominant.

5

Asnoted,thereisnounifiedandexplicitdefinitionofsmalldiameterchannelsuntilnow.The
threshold diameters based purely on a specified size remains fixed, whereas the threshold
diametersbasedondimensionlessnumbersaregreatlyaffectedbythephysicalpropertiesofthe
fluids. Figure 1.3 (a) and (b) shows the relevant threshold diameters, i.e. transition diameters for
considering a channel as micro, for ethanol and water, respectively, for the entire range of
operatingpressures.Itisclearthatthereareconsiderabledifferencesinthethresholddiameters,
asdefinedbyvariouscriteria.Thisexemplifiesthedifficultiesinclassificationofchannelsbasedon
the comparison of relevant forces. The issue becomes more important when working fluid
mixturesarethefocusofstudy.
Flowboilingheattransferisintimatelylinkedtothewalltransportinthepresenceofaliquid
vaporinterface.Asthechanneldiameterdecreases,theabsolutemagnitudesofalltheforceswill
godown.However,inmostcircumstancesofinterest,therelativemagnitudesoftheforcesareof
interest rather than the actual absolute values. A detailed scalingeffect of various applicable
forcesonflowboilingheattransferinmicrochannelsisdiscussedbyKandlikar(2010).

Figure 1.4 Scale effect of channel diameter on various forces during flow boiling of water and
ethanolatG=50kg/m
2
sandq=1MW/m
2
.

The relative magnitudes of five major forces at atmospheric pressure are plotted in Fig. 1.4,
forwaterandethanol,toillustratetheeffectofchangingscale,i.e.,flowdiameter.Inadditionto
thevalueofmassfluxGandq,theforcesarefluiddependentalso.Shiftingfromwatertoethanol
as the working fluid, the gravity and the surface tension force decreases comparatively where as
6

other three forces, i.e., inertia, shear and evaporative momentum, comparatively increases.
Secondly,ascanbeseen,therelativeincrease/decreaseisratherarbitrary.

1.4Applicationsofmini/microchannelsinheattransferdevices
Thefollowingisapartiallistoffieldswheremini/microchannelheattransferdevicesareused:
Compactheatexchangers/Evaporators
Automotiveandaeroindustry
Biomedicalapplications
Spacethermalmanagement
Heatremovalfromhighheatfluxdevicessuchas
Microprocessors
Laserdiodes/Powerelectronics
Electronicdevicesetc.
Pharmaceuticalandchemicalindustry
Like every system, heat transfer devices consisting of mini/microchannels have some
advantagesaswellasdisadvantages.Theadvantagesofusingmini/microchannelsare:
Highheatfluxhandlingcapability
Economicfromthepointofviewofspaceandmaterialcost
Compactdesigns
Thedisadvantagesofusingmini/microchannelsare
Needforverycleanworkingfluids
Complexityinmanifolddesign
Manufacturingchallenges
Pressuredroppenalty
Fouling

1.5Motivationforthepresentwork
Unless definitive measures are implemented soon enough, the world is heading towards a
severeandprolongedenergycrunchinthenottoodistantfuture.Inthehistoricalbackgroundof
the decarbonization pattern of primary fuel usage, hydrogen is emerging as a natural choice for
moresecureandcleanerenergycarrier(Barretoetal.(2003)).Hydrogencanbeproducedfroma
varietyofsourcesincludingfossilfuels(coal,naturalgas,LPG,gasoline,diesel,methane,propane,
wind,solar,methanol,ethanol,biodieseletc.)usingavarietyofmethods;abriefreviewofthese
7

methodsispresentedinMoharanaetal.(2011b).Todecreasethedependenceonfossilfuelsfully
or partially, current situation strongly demands that hydrogen be produced from a renewable
source.Inthiscontext,ethanol/bioethanolisapromisingbiomassderivedliquidfuelforhydrogen
production (Balat et al. (2008); Kim and Dale (2004)).Secondly, with ever increasing demand for
energy and changing lifestyle, there is also a need of decentralized portable power generation
deviceswhichseemstobeachievablethroughintegratedreactorsystems.

1.5.1Whytouseethanol?
All the traditional methods of hydrogen production have some inherent limitations (such as
dependenceonfossilfuel,lowrates/processefficiency,etc.)whichhindertheiruseforsustainable
energy generation, at least at this point of time. In this context, ethanolsteam reforming is a
viable process for generating energy from biomassbased fuel. Ethanol is a promising future bio
fuelbecauseofitsrelativelyhighhydrogencontent(onmolarbasis),nontoxicity,easeofstorage
and handling safety. Ethanol can be readily produced from renewable feed stocks such as
sugarcane, municipal solidwaste and agrowaste.Another advantage ofusing ethanol is that the
CO
2
produced during steam reforming reaction is equal to the CO
2
required for biomass growth
and thus provides a closed cycle for CO
2
consumption. Therefore, no net pollutants are released
intotheenvironment.Theonlyprimarydemeritincludeshighreformingtemperatures(~600
o
C).

1.5.2Whytousemicroreactor?
During the last two decades, technological developments in micromanufacturing systems
have gained considerable momentum GadelHak (2002). Recently, microdevices and systems
(popularly known as microreactors), having their characteristic dimensions typically of the order
of few micrometers, have attracted the interest of chemical process industry because of certain
inherentadvantages(KiwiMinskerandRenken(2005);Ehrfeldetal.(2000);Ehrfeldetal.(2007)).
Theobviousdiscerningcharacteristicsofmicroreactorsarecompactsize,lightweight,andlower
material and energy consumption. Secondly, smaller linear dimensions lead to increased specie
gradients (momentum flux, concentration, temperature), which are particularly important for
chemical reactor processing. This results in rapid heat and mass transport, and short diffusion
lengths. Faster system response gives better process control and high product yields. Besides
these, the attractive feature of microreactors is their high surface to volume ratio compared to
conventional chemical reactors. Because of smaller passage size, flow through microreactors
usually remains laminar. Thus, the heat/mass transfer coefficients become inverselyproportional
8

to the channel hydraulic diameter; of course there is increase in corresponding momentum flux
transfer i.e. pressure drop. Nevertheless, overall heat/mass transfer to pressure drop ratio is
increased. The higher heat and mass transfer characteristics of micro heat exchangers are an
added advantage for utilizing the full potential of the catalysts used, especially during
endothermic/exothermic reactions. Thus, local hotspot formations are avoided because of high
heatremovalcapacity.Secondly,becauseofhigherreactiontemperature,catalysiscanbeutilized
effectively requiring a smaller volume. This leads to low operating cost and higher system
efficiency. Finally, because of smaller reactants and products inventories, high level of safety is
achieved.Figure1.5showsexamplesofvarioustypesofmicroreactorswhichhavebeenreported
inrecenttimes.

Figure 1.5 Micro reactor and its components: (a) micro structured reactor (b) micro mixer (c)
microchannel(d)catalyticmicroreactor(source:http://www.immmainz.de).

9

1.5.3Distributedhydrogenproductionfromethanolinamicrofuelprocessor
The reaction pathways and thermodynamics of ethanol steam reforming (for producing
hydrogen) have been studied extensively recently. The possible reaction pathways of ethanol
steam reforming were summarized in Ni et al. (2007). Hydrogen production varies significantly
withdifferentreactionpathways.Inordertomaximizehydrogenproduction,itiscrucialtoensure
sufficientsupplyofsteamandtominimizeethanoldehydrationanddecomposition.So,thereisa
need of fuel processor for producing hydrogen by steam reforming of ethanol for fuel cell
applications.

Figure1.6Schematicofanintegratedreactorsystemfordistributedgenerationofhydrogenfrom
ethanol and water (Moharana et al. (2011b)). Four major building units i.e., (i) mixer, (ii)
preheater,(iii)reformer,and(iv)COcleanupstagesarerequiredtodevelopanintegratedreactor
system.

The production of H
2
from ethanol includes the reaction of ethanol with steam, followed by
stepstoreducetheCOcontentoftheproductstream.AsshowninFig.1.6,waterandethanolare
mixed together and then fed to a preheater where flow boiling of the mixture takes place by
10

application of external heat input. The gasphase mixture is further superheated to the desired
reaction temperature. The preheated vapor is fed to the reformer, where the mixture reacts to
formhydrogenaswellasothercompounds,suchasCO
2
,COandCH
4
.

Astheproducedhydrogen
isrichinCO(~10mol%),acleanupunitneedstobeattachedafterthereformer.Asshowninthe
figure, three possible routes are being investigated for reducing the CO content in the product
stream: (a) membrane separation; where the effluent is passed through a membrane after
reforming, which allows only H
2
molecules to pass through it. In this way 99.99% pure hydrogen
canbeproduced(OckwigandNenoff(2007)).(b)&(c)arechemicalroutesinwhichCOisreduced
toppmlevelsbyWGSfollowedbyeitherPrOXorselectivemethanationofCO.
The routes described above involve several thermal and chemical transport processes i.e.
endothermic/exothermiccatalyticreactions,flowboilingofliquidmixtures,singlephaseandtwo
phase heat exchange, etc. In contrast to conventional reactor and heat exchanger systems,
microreactors provide unique features not only in the integration of various thermochemical
processesdescribedabove,butalsobymeetingthedemandsofspaceoptimizationindistributed
systems. Heat resistant materials like Klingersil

can be used to avoid unwanted heat exchange

betweenthedeviceswhilemakingtheoverallsystemcompact.Inrecenttimes,therelativemerit
ofmicroreactorshasbeenstudiedbymanygroups(Holladayetal.(2004);Kolbetal.(2006);Kolb
etal.(2007)).
It is envisaged to develop an ethanolbased hydrogen production device of the order of one
kWusingsteamreformingofethanol(SRE)coupledwithfundamentalsofmicroscaleengineering.
The stoichiometry of the reaction for maximum hydrogen production by catalytic ethanol steam
reformingprocessis:
+ +
2 5 2 2 2
C H OH 3H O 6H 2CO : (1.5) A
o
298
H =174kJ/mol
Theproducedhydrogenismadetoreactwithoxygen(orair)toformwaterandthermalenergy:
+
2 2
H 0.5O H O
2
+241.8kJ/molofH
2
O (1.6)
Thus, stoichiometry requires a mixture of 2.4 ml/min of C
2
H
5
OH and 2.23 ml/min of H
2
O for
producing 1kW of thermal energy equivalent. In realtime practice, to avoid coke formation and
to enhance the selectivity towards reforming reaction, a steamrich mixture is desired (Ni et al.
2007a). Thus, assuming water to ethanol molar ratio of 1:6 (Stoichiometry = 1:3), the amount of
waterrequiredwouldbe4.5ml/min.ByassumingaW/F
A0
of1.00,(W=weightofthecatalystand
F
A0
=molarflowrateofthereactantA(i.e.ethanol),theamountofcatalystrequiredwouldbe2.5
g. With this data, if the cross section area of the microchannel is of the order of ~500 m
2
, the
11

length of reactor is of the order of ~50 mm, catalyst layer thickness of ~30m, number of
microchannels per reactor plate ~25, the number of reactor plates required works out be about
30.
Majorchallengesaretodevelop:
(i) Microheatexchangerforethanolwaterbinarymixturetomaintaintemperatureof873K.
(ii) Highlyactive,selectiveanddurablecatalysisforthereactionsinvolved.
Inthisbackground,DepartmentofScienceandTechnology,GovernmentofIndiahasassigned
aprojecttitledMicrodevicesforprocessapplicationstotheresearchgroupsatIITKanpur,India
for developing a microdevice for potable power generation, especially producing hydrogen from
ethanol through steam reforming. As discussed above, this involves development of an efficient
micro heat exchanger for singlephase and twophase heat transfer. The present work is
motivatedbythisassignment.TheresearchgroupatthedepartmentofChemicalEngineering,IIT
Kanpur is working on development of catalysis. In this work attention is focused only on the
development and thermohydrodynamic study of micro heat exchanger for mixture boiling of
ethanolwater.

1.5.3.1Preheaterdesignissues
With the advent of micro and meso scale thermofluidics and chemical systems, the
development of ultracompact process units such as heat exchangers, micro pumps and turbines
and micro thermal systems has become an important research activity area, especially in the
distributed power production parlance. The production for mixture feed for the ethanol steam
reformernecessitatesapplicationofamicroheatexchangerintheupstreamlocation,beforethe
catalytic reactions can commence. The catalytic steam reforming of ethanol for hydrogen
production is highly endothermic (H
673
= 208.4 kJ/mol), which accounts for the requirement of
high reforming temperatures usually around ~873 K (Casanovas et al. (2008)). A preheater,
consisting of a phasechange micro heat exchanger and a superheater, are therefore needed to
converttheliquidphasereactants(waterandethanol)tosuperheatedgasphase,beforeentering
the microreactor. In this background, the fundamental understanding of thermohydrodynamics
offlowboilingofpurefluidsandbinarymixturesisaprecursortodesignthecatalyticreformer.

1.6Objectiveofthepresentwork
Recent developments in singlephase and twophase flow in enhanced microchannels have
provided significantly higher performance with high heat flux dissipation and high heat transfer
12

coefficients. In order to improve the performance of flow boiling systems with microchannels to
matchandexceedthislevel,afundamentalunderstandingoftheunderlyingtransportprocessesis
needed. In this background a simultaneous experimental, numerical and theoretical investigation
is carried out to understand the thermohydrodynamics of mini/microchannels. The outcomes of
thisworkarepresentedasdetailedbelow.Thecontentsaredividedintodifferentchapters.

Chapter 2 details the thermohydrodynamics of singlephase in an array of minichannels

consisting of fifteen rectangular channels of hydraulic diameter 0.907 mm. In this work first an
experimentalinvestigationiscarriedouttounderstandthethermohydrodynamicsinthearrayof
minichannels. Simultaneously, a threedimensional numerical simulation is also carried out
depicting the exact experimental condition. Conservation equations have been solved in the
conjugate domain to explore conjugate heat transfer effects leading to deviation from true
constant heat flux conditions. It is observed that at higher value of the axial conduction number
(M), the experimental setup is prone to conjugate effects as a consequence of axial back
conductioninthesubstrate.Duetothis,localexperimentalNusseltnumbersarenotonlysmaller
thantheactualcounterpartspredictedbythemodel,butalsotheiraxialvariationisaffected.This
motivatedtoundertakefurtherdetailedstudy,asdescribedinchapter3.

Chapter 3 details a threedimensional numerical study of axial heat conduction in

microchannel substrate to understand and highlight the effects of axial wall conduction in a
conjugate heat transfer situation involving simultaneously developing laminar flow and heat
transfer.Inthefirstpart,asquaremicrochannelisconsideredcoveringawiderangeofparametric
variations, typically encountered in microfluids/microscale heat transfer domains. The results
showthattheconductivityratio(k
sf
)isthekeyfactorinaffectingtheextentofaxialconductionon
theheattransportcharacteristicsatthefluidsolidinterface.Itisfoundthat,allotherparameters
remaining the same, there exists an optimum value of k
sf
which maximizes the average Nusselt
number. Such a phenomenon is also observed in circular microtubes. Secondly, the effect of
channel aspect ratio on axial conduction in the solid substrate is studied and it is found that the
averageNusseltnumberisminimumcorrespondingtochannelaspectratiointherange1.82.0.

Chapter 4 Considering that it is very difficult to get standard cross sectional geometry of
microchannelforwhichmodelsareavailableforpredictingpressuredropandrelatedparameters.
Therefore it was felt that there is a need of developing a generalized formulation for estimating
13

14
pressure drop in fully developed laminar flow in arbitrarily shaped crosssection, which is
addressed in this chapter. This chapter details a twodimensional semianalytical investigation of
pressure drop in fully developed laminar flow in arbitrary crosssection channels (both singly as
well as doubly connected) using boundary collocation technique where collocation points are
considered on the arbitrarily shaped perimeter. The developed model is only a function of
geometrical parameters of the channel crosssection. Fluid velocity contours, shear stress
distributionalongthechannelwallsandthefrictionfactor(Poiseuillenumber)hasbeenestimated
in several complicated cross sections along with validation for some standard crosssections. An
experimental test case is also presented wherein, the application of present methodology in
accuratelypredictingthefullydevelopedvelocityprofile,andthereforethenetpressuredrop,ina
microchanneletchedonasiliconsubstrateisdiscussed.

Chapter5detailsanexperimentalinvestigationofpressuredropandheattransferofethanol
water binary mixture in a single microchannel. Ethanolwater mixture at different proportions
includingpureethanolandpurewaterareusedastheworkingfluid.Microchannelwasfabricated
on stainless steel substrate using laser machining processes. The experimental pressure drop in
microchannels was found to be within 10% of theoretical prediction correlations available in the
literature. The experimental study of ethanolwater mixture in microchannel indicated that the
boiling heat transfer process was significantly influenced by (a) the composition of the ethanol
water mixture (b) flow rate. For singlephase heat transfer, the heat flux was found to be
decreasing with increasing concentration of ethanol. When twophase flow starts, the heat flux
abruptly increased. The heat transfer coefficient in the singlephase fluid flow was found to be
almost independent of the excess wall temperature (wall super heat). Secondly, the singlephase
heat transfer coefficient was found to be decreasing with increasing concentration of ethanol;
maximum for pure water and minimum for pure ethanol, for a fixed mass flux. The axial
conduction effects in the combined singlephase liquid flow, twophase liquidvapor flow and
singlephasevaporflowalongthefullchannellengthareexplored.Thestudyrevealsthatthetwo
phase region will not be generally influenced by axial back conduction in the substrate as
comparedtothesinglephaseregions.

Chapter 6 details the summary and conclusion of the present work followed by possible
suggestionsforfuturework.