lable at ScienceDirect

Journal of Cleaner Production 145 (2017) 180e187

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier .com/locate/ jc lepro

High performance overhead power lines with carbon nanostructuresfor transmission and distribution of electricity from renewable sources

S. Kumar a, *, G. Pal a, Tushar Shah b

a Institute Center for Energy (iEnergy), Department of Mechanical and Materials Engineering, Masdar Institute of Science and Technology, PO Box 54224,Abu Dhabi, United Arab Emiratesb Applied Nanostructured Solutions, LLC 2323 Eastern Blvd., MP 50, Baltimore, MD, USA

a r t i c l e i n f o

Article history:Received 19 July 2015Received in revised form2 January 2017Accepted 10 January 2017Available online 13 January 2017

Keywords:Energy efficiencyPower transmission linesCarbon nanostructure-epoxy compositesSmart gridsMultiphysics finite element analysis

* Corresponding author.E-mail addresses: s.kumar@eng.oxon.org, skumaar

http://dx.doi.org/10.1016/j.jclepro.2017.01.0530959-6526/© 2017 Elsevier Ltd. All rights reserved.

a b s t r a c t

Integration of information, communication and materials technologies into the electricity Smart Grids isthe key to sustainable clean energy future. This study focuses on materials innovation and design ofpower transmission lines. Conventional bi-material power transmission lines consist of aluminumconductors with steel reinforcement. Increase in temperature of transmission lines due to Joule heatinglimits their current carrying capacity and the efficiency of power transmission. New design of trans-mission wires proposed in this study consists of carbon nanostructure (CNS)-epoxy composites in amultilayered architecture to enable multifunctional capabilities. Excellent thermal transport propertiesof CNS-epoxy composites are utilized to optimally dissipate heat from the outer surface of transmissionwire in order to maximize its performance. A coupled electrical-thermal finite element (FE) analysis ofAluminum Conductor Steel Reinforced (ACSR) wires is performed and the results are benchmarked withthose obtained from relevant IEEE standards. The validated model is then extended to new transmissionline composed of aluminum conductor - composite core with CNS-epoxy composite coating (ACCC-CNS).Steady-state coupled FE analyses of ACSR and ACCC-CNS wires indicate that the proposed design ofACCC-CNS wire enables transmission of larger currents than an equivalent ACSR wire for the sameamount of conductor material due to reduced operating temperature. The proposed design of powerlines would enable development of Smart Grids for more efficient utilization of electricity fromrenewable sources.

© 2017 Elsevier Ltd. All rights reserved.

1. Introduction

1.1. Smart grids and electric transmission

Increasing urbanization, developing digital economy and busi-ness environment requiring electronic transmission of data are themajor factors contributing to the growth in electricity demand.Electrical power research institute (EPRI) report on electricitytechnology roadmap shows that electrification has promoted adrop in energy usage intensity while promoting the economic andsocietal development (Fig. 1). According to International AtomicEnergy Agency (IAEA) report, by 2030 electricity would constitute22% of the total energy consumedworldwide as compared to 18% in2000 (International Atomic Energy Agency (2004)). Since most of

@mit.edu (S. Kumar).

the high growth economies of 21st century are located in devel-oping part of the world, it is projected that by 2030 these devel-oping countries will utilize 43% of global electricity productioncompared to 27% in 2000 (International Atomic Energy Agency(2004)). This high demand for electric power is met by numerouspower generation plants, supported by strong transmission sys-tems, supplying safe, reliable and economic electricity to its com-mercial, residential and industrial clients.

Three major technologies namely generation, transmission, anddistribution have been growing under the umbrella of Smart Grids(Cardenas et al. (2014); Nejadfard-Jahromi et al. (2015); Zahurulet al. (2016); Santo et al. (2015); Selvam et al. (2016)). Efficient andreliable transmission of electricity is intricately connected to livingstandards. Integration of information and communication tech-nology into the electricity smart grids is the key for meeting thefuture electricity demands and efficiency requirements (Bouhafset al. (2012); Gungor et al. (2011); O'Driscoll and O'Donnell(2013); Battaglini et al. (2009); Electrical Power Research Institute

Fig. 1. Energy intensity and economic development as a function of electrification(Electrical Power Research Institute (2003)).

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187 181

(2003)). The ever increasing electric power demand and stringentenvironmental regulations necessitate the integration of renewableenergy power plants with new electric transmission lines andreplacement of aging transmission lines to improve the grid sta-bility and reliability (Edison Electric Institute (2013); World Bank(2011); StatPlan Energy Limited (2012)). Lund et al. (2015) hasrecently emphasized the importance of Smart energy systemdesign for large clean power schemes in urban areas and predictedthat increasing the renewable electricity share beyond a limitwithout a smart design adds only limited benefit. Governments andprivate entities all over the world are addressing the issue byproviding strategic funding and stimulus to the industrial enter-prises interested in developing efficient electricity transmissionand distribution (T&D) systems. Responding to this trend, the po-wer transmission industry is searching for high performance al-ternatives to conventional transmission lines in order to meet theprojected demand for the bulk transmission of electricity.

1.2. Materials technology

From the technology development point of view the EuropeanUnion in 2010 established an energy technology policy for Europeto support the 2020 energy and climate change targets. One of itskey initiatives was to address materials technology that will beneeded in order to achieve the 2020 targets (EuropeanCommission (2010); Giordano et al. (2013)). A recent review onmaterials used for generators in wind turbines by Lacal-Ar�antegui(2015) indicates that the improvement in materials specificationsis challenging but achievable in most areas. However, the costreduction has been identified as a crucial aspect of renewableenergy production. Overhead electricity transmission wires are theprimary carrier of bulk electricity from generation facilities to thedistribution sub-stations (over 90% of total market). Most of theexisting transmission lines use aluminum conductor - steel rein-forced (ACSR) wires developed in 1920s. The increased demand forelectricity constrains the existing transmission and distribution(T&D) systems to operate at higher efficiency and reliability levels.However, with the current state of electricity transmission tech-nology, approximately 9% of the total electricity generated (1.8trillion kilowatt-hours out of 20.3 trillion kilowatt-hours) is lost inthe form of line losses during transmission of electricity globally(World Bank (2014)). Therefore, the electric utility companies aresearching for high performance alternatives to construct new

transmission lines and replace existing transmission lines. Thiswarrants improved design of transmission wires with innovativematerials capable of offering higher transmission efficiency. Thecurrent carrying capacity and efficiency of transmission lines areprimarily limited by their elevated operating temperature owingto Joule heating (Kova�c et al. (2006); Banerjee (2014); Chen et al.(2012)). The amount of energy lost due to Joule heating dependsupon the resistivity of the conductor, conductor temperature andthe amount of current. Steady-state temperature of the trans-mission lines is resultant of energy balance between heat gener-ation due to Joule heating and heat loss from the wire to thesurrounding (IEEE (2013)) via radiation and convection. Higherconductor temperature promotes thermal deterioration of wiresand increases the wire sag. These two have often been cited as thereasons for physical failure of transmission lines causing brown-outs and blackouts (Harvey (1972); Burks et al. (2010); Bhuiyan(2011)). The life-cycle analysis of 11 kV electrical overhead linesand underground cables concluded that the system with lowestconductor resistance should be selected to minimize the deterio-rating life-cycle impacts (Jones and McManus (2010)). Neverthe-less, the choice of conductor material is limited by the materialstechnology.

Several electrical equipment manufacturers attempted todevelop high capacity solutions to reduce power transmissionbottlenecks. For instance, 3M offered high ampacity AluminumConductor e Aluminum Composite Reinforced (ACCR) trans-mission wires with aluminum-zirconium (Al-Zr) alloy as primaryconductor (3M Publication (2003)). However, these transmissionlines cost 3e6 times more than the conventional ACSR trans-mission lines (Iowa State University (2011)). In 2005, CTC Globalintroduced a high temperature - low sag Aluminum Conductor eComposite Core (ACCC) transmission wire with Al-1350-O alloy asthe conductor in place of conventional Al-1350-H19 alloy (CTCGlobal Corporation (2011)). Both of these wires received limitedenthusiasm from highly conservative power transmission in-dustry because introduction of new conductor material for thedesign of transmission lines would require capacity re-evaluationof all the accessories of transmission lines. These issues arealleviated in the new design of transmission lines proposed inthis study.

This study examines the new transmission line architecture as ahigh performance alternative to conventional ACSR transmissionwires. Superior electrical, thermal and mechanical properties ofCNS materials (Pal and Kumar (2016a,b); Arif et al. (2016);Kundalwal and Kumar (2016)) are utilized to design such a highperformance wire. The new transmission line (ACCC-CNS) pro-posed in this study consists of conventional Al-1350-H19 alloy asconductor and glass fiber reinforced epoxy composite core to sus-tain thermo-mechanical load with CNS-epoxy composite outerlayer. The inner-CNS layer protects the composite core from strayradio frequency (RF) generated by the electromagnetic pulse (EMP)emanating from high electric current carrying aluminumconductor. The outer CNS-epoxy layer with increased thermalconductivity and high emissivity is expected to facilitate optimaldissipation of the heat generated in the conductor. Incorporation ofouter CNS composite layer in the transmission wires may alsoenable in-built deicing capability and protection from lightningstrike and foreign object damage (Al-Saleh and Sundararaj (2009);Chu et al. (2014)). In this study, the performance of the new ACCC-CNS and ACSR transmission lines is investigated in terms of line lossdue to Joule heating using both transient and steady-state coupledthermal-electrical finite element (FE) analyses. For both wires(ACCC-CNS and ACSR), conductor surface temperature is consid-ered as the measure of energy loss due to Joule heating.

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187182

2. Energy efficient power lines and their design

2.1. CNS-fabrication

Carbon nanostructure (CNS) is a new form of nanomaterial andit contains clusters of highly aligned multiwall CNTs. CNS used inthis study was produced by one of the authors at ANS, USA. CNSproductionmethod is briefly described as follows. The CNSmaterialis grown on glass fiber substrate through plasma-enhancedchemical vapor deposition (CVD) process (Shah et al. (2012);Malet and Shah (2014)). The glass substrate used for CNS growthcan be of various shapes and formations such as spoolable glassfiber tows, glass tapes, woven glass fabric and glass fiber mats. Inthe first step, glass fibers (or woven fabric) are treated with aplasma etching process in order to facilitate surface bonding of thecatalyst and the CNS to substrate glass fibers. The etched glass fibersare then immersed into the aqueous solution of hydrogen peroxide,ferrous acetate and cobalt acetate. The deposited ferrous and cobaltsalts work as the precursor catalyst for CNT growth. Following thedeposition, the glass fibers are heated to 300e450 oC to convertcatalyst precursors into intermediate catalyst [g-Fe3 O4 (maghe-mite), a-Fe3 O4 (hematite), cobalt ferrite (CoFe3 O4 ) and cobaltousoxide (CoO)]. The heated fibers are further heated up to 550e800 oCin the presence of acetylene or ethylene gas. At 700 �C, pyrolysis ofacetylene (or ethylene) takes place and hydrogen gas and atomiccarbon are released (Malet and Shah (2014)). At this temperature,hydrogen gas reduces the intermediate catalysts in active transi-tional metal carbide nanoparticles and the presence of atomiccarbon with metal catalytic particles result in the formation andgrowth of carbon nanotubes forest on the substrate glass fibers. Inthe last step, these bundles of carbon nanotubes are sheared offfrom glass fibers in the form of CNS material. Fig. 2 shows high-resolution scanning electron micrograph (SEM) of CNS materials.High degree of CNT alignment offers improved physical andtransport properties to CNS material as compared to those ofrandomly cross-linked CNT clusters. This presents a unique op-portunity to develop high performance ACCC-CNS power

Fig. 2. Scanning electron microscope (SEM) image of carbon nanostructure (CNS)material. The SEM image in inset shows well aligned multiwall carbon nanotubeswhich impart improved transport properties to CNS material as compared to regularCNT clusters.

transmission lines with improved heat dissipation capabilities.

2.2. CNS-characterization

CNS material was characterized to obtain both thermal andelectrical properties. The electrical conductivity of CNS compositelayer is measured using van der Pauw method and Lake ShoreCryogenics's integrated Hall effect measurement system (HMS).The van der Pauwmethod is a technique to determine the electricalresistivity of heterogeneous semiconducting materials bymeasuring direct current (DC) resistance of the specimen betweenthe HMS probe needle contact points on the specimen. In thismethod, 1 cm � 1 cm square specimen is cut from the CNS-Epoxycomposite sheet. In order to facilitate low resistance electricalconnections, four small dots are made with conductive silver painton the four corners of the square specimen. The four tungstenneedle probes of standard samplemounting card of HMS are placedon the four silver dots. The card is then placed in the sample holderand mounted on the HMS to measure the electrical resistivity. Themeasured electrical conductivity of CNS-Epoxy composite is850e1000 S/m. The thermal conductivity of CNS-epoxy compositeis measured using laser flash method and is in the range of 15e20W/(m.K). Uniaxial tensile test was performed to determine theYoung's modulus of CNS/Epoxy composite. The average Young'smodulus of CNS/Epoxy is 3.5 GPa. The coefficient of thermalexpansion of the CNS lies in the range of 20�10�06/K - 30�10�06/K.The CNS content in CNS-epoxy coating is approximately 3e4% byweight. The electrical and thermal conductivities of aluminum andsteel are extracted from the experimental results reported in theliterature.

3. Multiphysics Finite Element modeling

Energy loss due to Joule heating of transmission line is a coupledthermal-electrical problem. Consider a transmission line carryingcurrent, I, at line voltage, V, and electrical field intensity, E. Thecurrent density, j, and the heat generated due to Joule heating,W, inthe transmission line with resistance, R, can be given, respectively,by

j ¼ I=A (1)

W ¼ j:VV ¼ j:E (2)

since,

j ¼ s:E (3)

The expression for the electrical energy lost in the form of heatcan be rewritten as

W ¼ E:s:E ¼ s���VV���2 (4)

where

s ¼ L=ðRAÞ (5)

is the electrical conductivity of the conductor material, A is totalcross sectional area of the conductors in the wire and L is the lengthof wire. The electrical conductivity, and hence the resistivity, of theconductor is a temperature dependent material property. As thetemperature of the line increases, the conductivity of the trans-mission lines decreases. Therefore, when higher current is passedthrough transmission wire, the wire's steady-state temperatureincreases not only due to higher conductor current but also due to

Table 1Transport properties for materials used in FE modeling (Davis (1993); TheAluminum Association (1989); Edison Electric Power Institute (2007);Spitalsky et al. (2010); Hendren and Ghorashi (1994)).

Material Temp, K Electricalconductivity, S/m

Thermal conductivity,W/(m-K)

Aluminum 300 3.54E+07 237400 2.51E+07 240500 1.94E+07 236600 1.58E+07 231

Steel 300 3.42E+06 47.2400 2.56E+06 47.8500 2.05E+06 48.1600 1.61E+06 47.2

CNS-epoxycomposite

Ambienttemp

1.0E+03 15

Glass-epoxycomposite

Ambienttemp

1.10E-13 0.1

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187 183

drop in electrical conductivity of the conductor at elevated tem-perature. In a coupled thermal-electrical problem, heat conductionin the conductor is governed by the following energy balancerelation

VðkTVTÞ þ s���VV���2 ¼ rCpðvT=vtÞ (6)

where, r is density, kT is thermal conductivity and Cp is the heatcapacity of the conductor material. In the above heat conductionequation, s

���VV���2 works as source term and represents the electricalpart in a coupled thermal-electrical problem. When electric cur-rents flows, conductor starts heating up. Within a brief period oftime, the entire cross section of transmission line attains a steady-state temperature in equilibrium with its surrounding. For a givenoverhead transmission line, the value of steady-state temperatureis a function of conductor material, electric current and ambientconditions such as temperature, wind speed, etc. In general, theoverhead conductors are exposed to surrounding environment (i.e.air) and lose a fraction of heat generated to the surroundingthrough radiation and convection. The radiative heat loss flux qrad(W/m2) from the outer surface of the transmission line is given by

qrad ¼ nε�T40 � T4cond

�(7)

where, ε is the emissivity of the material at the outer surface, n isStefan-Boltzmann constant (5.6704 � 10�8 W=m2K4), T0 and Tcondare ambient and transmission line outer surface temperatures in K,respectively. The convective heat loss flux qconv (W/m2) from theouter surface of the transmission line is given by

qconv ¼ hcðT0 � TcondÞ (8)

where, hc is the heat transfer coefficient of convection and dependson the type of surrounding medium, shape of heat sink andwhether the heat dissipation is through free convection or forcedconvection. The value of film coefficient for natural convection ingases varies between 5 and 10. The value of film coefficient fornatural convection in air is taken equal to 7 and is applied asinteraction condition on outer wire surface for both ACCC-CNS andACSR wires. In the present study, the emissivity of the CNS/Epoxycomposite at the outer surface of ACCC-CNS wire is taken equal to0.8 and that of aluminum (oxidized) at outer surface of ACSR wire istaken equal to 0.1. Heat gain from the sun is determined using theprocedure outlined in the IEEE Standard 738: 2006 and is applied asheat flux on the outer surface of the line. Its value varies with ge-ography from 850 to 1350 W

m2, but a value of 1000 Wm2 was used in this

study.

3.1. Coupled thermal-electrical FE analysis of transmission wires

The FE model for transmission wire considers temperaturedependent material properties for aluminum conductor and steelcore to appropriately account for drop in conductor's electricalconductivity as the conductor attains higher temperature due toincreased current load (see Table 1). Three dimensional transientand steady-state coupled thermal-electrical multiphysics finiteelement (FE) studies with ‘DC3D8E’ elements are performed usingAbaqus FEAVersion 6.12. Fig. 3(a) and (c) show typical cross sectionof ACSR and ACCC-CNS wires, respectively. The ACSR wire consistsof two layers; namely, steel reinforcing strands at the center tosustain thermo-mechanical loads, including self-weight, andcurrent-carrying aluminum conductor strands at the circumferenceof the wire. Proposed design of ACCC-CNS wire has a multi-layered(four layers) architecture. The core of ACCC-CNS wires consists of

unidirectional glass fiber reinforced epoxy composite (glass fibervolume fraction z 60% by weight). The application of unidirec-tional glass fiber-epoxy composite offers a light weight alternativeto conventional steel strands core so as to sustain the thermo-mechanical load. The second layer, i.e. the inner CNS-epoxy layerprotects the composite core from stray radio frequency (RF) ab-sorption as high current carrying aluminum will generate electro-magnetic pulses (EMP). The third layer in ACCC-CNS wire consistsof conducting aluminum strands. The conductor material for bothcases, ACSR and ACCC-CNS, is aluminum Al-1350-H19 alloy. Thefourth layer, i.e. outer CNS-epoxy layer absorbs heat generated inthe conductor and dissipates it to the surrounding through radia-tion and convection.

Fig. 3b and (d) show the FEmodels of ACSR and ACCC-CNSwires,respectively. In these FE models, both core and the conductorportions of the wire are modeled as homogenized sections withperfect interfacial contact between them. The area of each section(aluminum and steel) is kept equal to their respective areas inconventional ACSR wire. FE analyses are conducted for differentelectric current loads applied to the conductor (aluminum) of thewire in order to study the thermal response of wires. The modelaccounts for heat dissipation to the surrounding through radiationand natural convection interactions (Eqs. (7) and (8)).

4. Results and discussion

4.1. Validation of ACSR FE model

IEEE (2013) is used for validating FE models developed to studythe Joule heating in overhead ACSR transmission lines. According tothe IEEE standard, the steady-state heat balance for unit length oftransmission line is given by

qr þ qc ¼ qs þ I2R (9)

I ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiqr þ qc � qs

R

r(10)

where, I is line electric current, R is line resistance per unit lengthwhich is a function of temperature, qr is radiated heat loss, qc isconvected heat loss rate through natural or forced convection andqs is heat gain from sun. The three heat fluxes (qr, qc and qs) aremeasured per unit length of the transmission line. Natural con-vection removes the heat generated during the power transmissionat a slower rate than the forced convection. As a result, heat

Fig. 3. Design configuration of (a) typical ACSR and (c) ACCC-CNS architecture and Finite element (FE) model of (b) ACSR and (d) ACCC-CNS transmission wires.

Fig. 4. Verification of FE results for transmission line's outer surface temperature usingIEEE standard 738-2006.

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187184

dissipation due to natural convection offers a conservative estimateof conductor steady-state temperature. Therefore, natural convec-tion is considered in this study tomodel convective heat dissipationfrom transmission wires and is given by

qc ¼ 0:0205rairD0:75ðTc � TaÞ1:25 (11)

where, rair is density of air (kg/m3), Tc and Ta are conductor surfaceand ambient air temperatures (�C) and D is the wire diameter(mm), respectively. Similarly, the radiative heat loss rate, qr, fromtransmission lines outer surface is given by

qr ¼ 0:0178Dε

"�Tc þ 273

100

�4

��Ta þ 273

100

�4#

(12)

Based on the procedure outlined in IEEE 738e2006 standard, itis straight forward to calculate the amount of electric currentrequired to generate a given conductor surface temperature usingthe above expressions (Eqs. (9)e(12)). However, in order to validatethe FE model, one needs to solve the inverse problem of calculatingthe conductor temperature for a given electric current. Therefore,this inverse problem is solved following an iterative procedureinvolving Eqs. (9)e(12) using Matlab. The conductor temperatureobtained using IEEE standard is taken as reference in order tovalidate ACSR FE results. In the FE model, the steady-state tem-perature at the outer surface of ACSR wires is computed for threerepresentative ACSR transmission lines (Parakeet, Drake andBittern) for a range of conductor currents and is compared withthose obtained using IEEE standard. The range of current load foreach wire is representative of their standard current carrying ca-pacities. Fig. 4 shows the comparison for all three cases where solidlines represent the conductor surface temperature obtained usingIEEE 738e2006 standard and dotted lines represent the same ob-tained from FE simulations. In all three cases, the ACSR wire surfacetemperature obtained from the two methods are in excellentagreement over the entire range of electric currents. The agreement

between surface temperature values obtained from the two ap-proaches validates the FE approach. Similar FE models withnecessary modifications are used to study steady-state response ofACCC-CNS transmission lines.

4.2. Transient analysis

Fig. 5(a) - 5(f) show the temperature distribution (K) across thecross-section in Parakeet ACSR and the equivalent ACCC-CNS wiresfor a given electric current (400 A) at different time intervals in atransient analysis performed using Eq. (6). In this analysis, thethickness of outer and inner coating was 1 mm. From these figures,it can be seen that the addition of CNS layer in ACCC-CNS wires

Fig. 5. Comparison of temperature distribution (in Kelvin) across the cross section ofParakeet ACSR and the equivalent ACCC-CNS wires for a given electric current (400 A)at different time intervals.

Table 2Steady-state conductor temperature [�C] for Parakeet ACSR and Parakeet ACCC-CNSwires as a function of conductor current [Ampere].

Conductor Current[ampere]

Parakeet ACSR,[�C]

Parakeet ACCC-CNS, [�C]

% Drop intemperature

400 48 38 21500 60 46 23600 75 57 24700 94 69 27800 118 84 29

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187 185

increases the time required to attain thermal steady-state. Such adelay in attaining steady-state temperature could be helpful inimproving wire's emergency ratings.

4.3. Steady-state analysis

In order to perform steady-state analysis of coupled thermal-electrical problem, the term contributing to internal energy of thesystem (term with Cp as coefficient) is omitted from the governingequation (Eq. (6)). Fig. 5(g) and (h) show the steady-state temper-ature distribution in representative Parakeet ACSR and equivalentACCC-CNS wires for the same current (400 A), respectively. Asbefore, the thickness of outer and inner coating was assumed to be1 mm. These results show that the steady-state temperature forACSR cable (48�C) is 10�C higher than that of its equivalent ACCC-CNS wire (38�C). This implies that for a given conductor current,faster heat dissipation rate due to the presence of CNS layer inACCC-CNS wires keeps the operating temperature of the wire wellbelow that of an equivalent conventional ACSR wire. Table 2 com-pares the steady-state conductor temperature [�C] for Parakeet

ACSR and Parakeet ACCC-CNS wires for different electric currentloads and reports the percentage of reduction in conductor tem-perature for Parakeet ACCC-CNS. From Table 2, it can be seen thatalthough the wire current increases linearly, the wire steady-statetemperature increases nonlinearly. This incremental deviationfrom linearity for wire's steady-state temperature is the result ofdrop in conductor's electrical conductivity with increase in wiretemperature. The comparative reduction in ACCC-CNS's steady-state temperature also indicates that the wires with new archi-tecture can support more current than their ACSR counterpartbefore reaching the upper limit of operating temperature of ACSRwires. This indicates that an ACCC-CNS wire would require lessamount of conductor (aluminum) in order to attain the operatingtemperature range typical of equivalent ACSR wires for a givencurrent. Fig. 6a and b shows the comparison of maximum operatingtemperature attained in Parakeet ACSR and Drake ACSR trans-mission wires and their equivalent ACCC-CNS counterparts fordifferent volume fractions (vf ) of conductor (aluminum) based onsteady-state FE analyses, respectively. It can be seen from Fig. 6aand b that the for same amount of current through the conductor,the conductor (aluminum) volume in ACCC-CNS can be reduced byapproximately 20e30% to maintain the operating temperaturetypical of comparable ACSR wires. In other words, for a givenamount of aluminum, ACCC-CNS transmission line has higher cur-rent carrying capacity than its ACSR counterpart and thereforeACCC-CNS wires can be utilized as a drop-in alternative whereexisting ACSR transmission line is incapable of meeting theincreased electricity demands.

The reduction in steady-state operating temperature of ACCC-CNS wires is due to the superior thermal conductivity of outerCNS layer and the higher emissivity of CNS-epoxy coating. Theemissivity of oxidized aluminum is in the range of 0.1e0.3 whereasthe same for cured epoxy varies between 0.8 and 0.9. The highthermal conductivity of outer CNS-epoxy coating facilitates thetransfer of heat generated in the conductor to the outer surface ofwire efficiently. Due to higher emissivity coefficient, CNS-epoxycoating radiates the heat generated to the surrounding at ahigher rate than the aluminum outer surface of conventional ACSRwires. This results in reduced operating temperature of ACCC-CNSwires. Fig. 7 shows the effect of change in coating thickness onthe performance of Parakeet ACCC-CNS wires compared withequivalent ACSR wires. It can be seen that in steady-state, thechange in operating temperature is insignificant with increase inthickness of the outer coating. Therefore, other factors such asmechanical and dielectrical strength of the coatings should also beconsidered in determining the optimal thickness of epoxy-CNScoatings.

5. Conclusion

CNS material was fabricated and its thermal and electricalcharacteristics were experimentally obtained. Subsequently,coupled thermal-electrical FE analyses of the proposed

Fig. 6. Comparison of operating temperature of conventional ACSR transmission wireswith that of equivalent ACCC-CNS wires of different volume fractions (vf) of aluminumconductor as a function of conductor current.

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187186

multilayered ACCC-CNS transmission wires were conducted. FEanalysis indicates that the ACCC-CNS wires would requireapproximately 20e30% less conductor material than those of

Fig. 7. Effect of outer coating thickness on outer surface temperature due to Jouleheating in ACCC-CNS conductors.

equivalent conventional ACSR wires for the same current rating inthe limit of ACSR wires' operating temperature. Furthermore,multilayered design of transmission lines with CNS-compositesreduces the lines temperature by 10�C for the parameters usedherein. The wire surface temperature is considered as a measure ofenergy loss. Reduction in conductor's temperature also preventsthermal deterioration of lines and hence minimizes the line sag.Fig. 6a in the manuscript shows the performance of the ACCC-CNSand ACSR Parakeet wires. For a current load of 650 amps, theoperating temperature for ACCC-CNS and ACSR wires are 60�C and85�C, respectively, leading to a 30% reduction in energy loss. Asdiscussed in the manuscript, the ACCC-CNS wire would requireabout 20e30 vol% less conductor material to transmit the samecurrent, this directly translates into reduced cost. However, thecoating requires small quantity of CNS material but continuouslydecreasing cost of CNS justifies the economic viability of proposedsolution for overhead power transmission lines. The existingtransmission infrastructure is on an average 40 þ years old withmany transmission lines expected to reach the end-of-useful life inthe next few years. Building new transmission line is often veryresource consuming and takes exceedingly long time for federaland state governments approval as a result of growing public op-position to building new infrastructure near their homes or work-places. Therefore, many electric utilities are constrained to replacethe existing transmission infrastructure with higher capacity lines,so new transmission towers and civil construction can be avoided.The high ampacity design solution provided in this study would bea way forward to achieve sustainable and resilient electrical powergrids by replacing existing power lines. Use of CNS material inconjunction with multi-layered architecture can lead to light-weight, stronger, durable and damage tolerant design of powertransmission wires. CNS coating would also provide EMI shielding,and de-icing capabilities. Thus, the ACCC-CNS wires offer aneffective solution to reduce the line losses, number of new right-of-ways and power generation requirements (and, therefore, theassociated energy cost and emissions) while helping the energytransmission and distribution industries in reducing the carbonfootprint by increasing the capacity of existing right-of-way. Thisstudy further suggests that the material design and multilayeredarchitecture of the tranmission wires is one of the most importantaspects for developing Smartgrids aimed at efficient utilization ofelectricity from renewable sources. Detailed computational mate-rial design of CNS/polymer composites requires a dual step FEanalysis in which a coupled electrical-thermal analysis is needed toobtain steady-state temperature of the wires for a given currentrating along with subsequent thermo-mechanical analysis to esti-mate the line sag. Geometric parameters of the wire could befurther optimized by allowing the stresses and operating temper-ature to attain their maximum allowable limits so as to reduce thewire weight, material consumption and line sag. Back and forthinformation flow between these two steps in an iterative mannerwill lead to a truly optimal design.

Acknowledgment

SK and GP would like to acknowledge the financial support fromLockheed Martin Corporation for this work under Award No:EX2014-000027. The authors would like to thank Ms. MaribethMalloy, Mr. Sunil Pancholi, and Mr. Scott Stallard of LockheedMartin Corporation, USA, for their constructive suggestions andhelpful hints.

References

3M Publication, 2003. Aluminum Conductor Composite Reinforced (ACCR)

S. Kumar et al. / Journal of Cleaner Production 145 (2017) 180e187 187

Technical Notebook (795 kcmil Family): Conductor and Accessory Testing.Al-Saleh, M.H., Sundararaj, U., 2009. Electromagnetic interference shielding

mechanisms of cnt/polymer composites. Carbon 47 (7), 1738e1746.Arif, M.F., Kumar, S., Shah, T., 2016. Tunable morphology and its influence on

electrical, thermal and mechanical properties of carbon nanostructure-bucky-paper. Mater. Des. 101, 236e244.

Banerjee, K., 2014. Making the Case for High Temperature Low Sag (Htls) OverheadTransmission Line Conductors. Ph.D. thesis. Arizona State University.

Battaglini, A., Lilliestam, J., Haas, A., Patt, A., 2009. Development of supersmart gridsfor a more efficient utilisation of electricity from renewable sources. J. Clean.Prod. 17 (10), 911e918.

Bhuiyan, M.M.I., 2011. Md. Mafijul Islam Bhuiyan. Ph.D. thesis. University of Alberta.Bouhafs, F., Mackay, M., Merabti, M., Jan 2012. Links to the future: communication

requirements and challenges in the smart grid. IEEE Power Energy Mag. 10 (1),24e32.

Burks, B., Armentrout, D., Kumosa, M., 2010. Failure prediction analysis of an acccconductor subjected to thermal and mechanical stresses. Dielectrics and Elec-trical Insulation. IEEE Trans. 17 (2), 588e596.

Cardenas, J.A., Gemoets, L., Rosas, J.H.A., Sarfi, R., 2014. A literature survey on smartgrid distribution: an analytical approach. J. Clean. Prod. 65, 202e216.

Chen, L., Zheng, Z., Liu, S., Guo, L., Dun, C., 2012. Thermal grading around overheadtransmission line under various environments and its influence to load ca-pacity. In: Electrical Insulation (ISEI), Conference Record of the 2012 IEEE In-ternational Symposium on. IEEE, pp. 545e548.

Chu, H., Zhang, Z., Liu, Y., Leng, J., 2014. Self-heating fiber reinforced polymercomposite using meso/macropore carbon nanotube paper and its application indeicing. Carbon 66, 154e163.

CTC Global Corporation, 2011. Engineering Transmission Lines with High CapacityLow Sag ACCC Conductors.

Davis, E.J.R., 1993. ASM Specialty Handbook: Aluminum and Aluminum Alloys. ASMInternational.

Edison Electric Institute, 2013. Transmission Projects: at a Glance.Edison Electric Power Institute, 2007. Carbon Steel Handbook. Edison Electric Power

Institute.Electrical Power Research Institute, 2003. Electrical Technology Roadmap - Meeting

the Critical Challenges of 21st Century: 2003 Summary and Synthesis.European Commission, 2010. Energy 2020: a Strategy for Competitive, Sustainable

and Secure Energy.Giordano, V., Meletiou, A., Covrig, F.C., Mengolini, A., Ardelean, M., Fulli, G., 2013.

Smart Grid Projects in Europe: Lessons Learned and Current Developments.Gungor, V.C., Sahin, D., Kocak, T., Ergut, S., Buccella, C., Cecati, C., Hancke, G.P., Nov

2011. Smart grid technologies: communication technologies and standards.IEEE Trans. Indus. Inf. 7 (4), 529e539.

Harvey, J., 1972. Effect of elevated temperature operation on the strength ofaluminum conductors. IEEE Trans. Power Apparatus Syst. 5 (PAS-91),1769e1772.

Hendren, G., Ghorashi, H., May 18 1994. High Strength Papers from Floc and Fibers.EP Patent 0,446,870. URL https://www.google.com.ar/patents/EP0446870B1?cl¼en.

IEEE, 2013. Ieee Standard for Calculating the Current-temperature Relationship ofBare Overhead Conductors, pp. 1e72.

International Atomic Energy Agency, 2004. Power to the People: the World Outlookfor Electricity Investment. Iaea bulletin.

Iowa State University, 2011. Transmission Line Loading: Sag Calculation and High-temperature Conductor Technologies.

Jones, C.I., McManus, M.C., 2010. Life-cycle assessment of 11kv electrical overheadlines and underground cables. J. Clean. Prod. 18 (14), 1464e1477.

Kova�c, N., Saraj�cev, I., Poljak, D., 2006. Nonlinear-coupled electric-thermal modelingof underground cable systems. Power Delivery. IEEE Trans. 21 (1), 4e14.

Kundalwal, S., Kumar, S., 2016. Multiscale modeling of stress transfer in continuousmicroscale fiber reinforced composites with nano-engineered interphase.Mech. Mater. 102, 117e131.

Lacal-Ar�antegui, R., 2015. Materials use in electricity generators in wind turbi-nesestate-of-the-art and future specifications. J. Clean. Prod. 87, 275e283.

Lund, P.D., Mikkola, J., Ypy€a, J., 2015. Smart energy system design for large cleanpower schemes in urban areas. J. Clean. Prod 103, 437e445.

Malet, B., Shah, T., Jul. 22 2014. Glass substrates having carbon nanotubes grownthereon and methods for production thereof. US Patent 8,784,937. URL https://www.google.com.ar/patents/US8784937.

Nejadfard-Jahromi, S., Rashidinejad, M., Abdollahi, A., 2015. Multistage distributionnetwork expansion planning under smart grids environment. Int. J. Electr. Po-wer Energy Syst. 71, 222e230. URL. http://www.sciencedirect.com/science/article/pii/S0142061515001040.

O'Driscoll, E., O'Donnell, G.E., 2013. Industrial power and energy meteringea state-of-the-art review. J. Clean. Prod. 41, 53e64.

Pal, G., Kumar, S., 2016a. Modeling of carbon nanotubes and carbon nano-tubeepolymer composites. Prog. Aerosp. Sci. 80, 33e58.

Pal, G., Kumar, S., 2016b. Multiscale modeling of effective electrical conductivity ofshort carbon fiber-carbon nanotube-polymer matrix hybrid composites. Mater.Des. 89, 129e136.

Santo, K.G.D., Kanashiro, E., Santo, S.G.D., Saidel, M.A., 2015. A review on smart gridsand experiences in Brazil. Renew. Sustain. Energy Rev. 52, 1072e1082. URL.http://www.sciencedirect.com/science/article/pii/S1364032115008291.

Selvam, M.M., Gnanadass, R., Padhy, N., 2016. Initiatives and technical challenges insmart distribution grid. Renew. Sustain. Energy Rev. 58, 911e917. URL. http://www.sciencedirect.com/science/article/pii/S1364032115016408.

Shah, T., Gardner, S., Alberding, M., Apr. 17 2012. Cnt-infused fiber and methodtherefor. US Patent 8,158,217. URL https://www.google.com.ar/patents/US8158217.

Spitalsky, Z., Tasis, D., Papagelis, K., Galiotis, C., 2010. Carbon nanotubeepolymercomposites: chemistry, processing, mechanical and electrical properties. Prog.Polym. Sci. 35 (3), 357e401.

StatPlan Energy Limited, 2012. The Global Market for Wire and Cable 2012-2050:Insulated Cable and Bare Conductors.

The Aluminum Association, 1989. Aluminum Electrical Conductor Handbook. TheAluminum Association.

World Bank, 2011. Transmission Expansion for Renewable Energy Scale-up:Emerging Lessons and Recommendations.

World Bank, November 2014. Electric Power Transmission and Distribution Losses(Kwh) and Electric Power Consumption (Kwh). Accessed on 17-11-2014 @ 15:11hours. URL. http://search.worldbank.org/data?qterm¼transmission.

Zahurul, S., Mariun, N., Grozescu, I., Tsuyoshi, H., Mitani, Y., Othman, M., Hizam, H.,Abidin, I., 2016. Future strategic plan analysis for integrating distributedrenewable generation to smart grid through wireless sensor network: Malaysiaprospect. Renew. Sustain. Energy Rev. 53, 978e992. URL. http://www.sciencedirect.com/science/article/pii/S1364032115009909.