Eh

Pa

b

a

ARRA

1

t(ttntaechcaueta

dbpce

I

0h

Nuclear Engineering and Design 248 (2012) 108– 116

Contents lists available at SciVerse ScienceDirect

Nuclear Engineering and Design

jo u r n al hom epage : www.elsev ier .com/ locate /nucengdes

valuation methodology for advance heat exchanger concepts using analyticalierarchy process

iyush Sabharwall a,∗, Eung Soo Kimb, Mike Pattersona

Idaho National Laboratory, 2525 Fremont Avenue, Idaho Falls, ID 83415, USASeoul National University, Seoul, Republic of Korea

r t i c l e i n f o

rticle history:eceived 18 October 2011eceived in revised form 21 March 2012

a b s t r a c t

This study describes how the major alternatives and criteria being developed for the heat exchangersfor next generation nuclear reactors are evaluated using the analytical hierarchy process (AHP). This

ccepted 23 March 2012evaluation was conducted as an aid in developing and selecting heat exchangers for integrating powerproduction and process heat applications with next generation nuclear reactors. The basic setup forselecting the most appropriate heat exchanger option was established with evaluation goals, alternatives,and criteria. The two potential candidates explored in this study were shell-and-tube (helical coiled) andprinted circuit heat exchangers. Based on study results, the shell-and-tube (helical coiled) heat exchangeris recommended for a demonstration reactor in the near term, mainly because of its reliability.

. Introduction

Heat exchangers are used to transfer thermal energy from a hot-er medium to colder medium within a system. According to Sekulic1990), “heat exchangers are devices that provide for a change ofhe mutual thermal (energy) levels between two or more fluids inhermal contact without external heat and work interactions.” Theext generation nuclear reactors (NGNR), such as high tempera-ure gas-cooled reactors and advanced high temperature reactors,re intended to increase energy efficiency in the production oflectricity and/or provide high temperature heat for industrial pro-esses. In the case of advance high temperature reactor (AHTR) theeat is transferred from the reactor core by the primary liquid-saltoolant to an intermediate heat-transfer loop through intermedi-te heat exchangers (IHXs). The intermediate heat-transfer loopses an intermediate liquid-salt coolant through a secondary heatxchanger (SHX) to move the heat to a power conversion sys-em (Rankine cycle) or any potential industrial heat application,s shown in Fig. 1.

The efficient transfer of energy for industrial applicationsepends on the ability to incorporate effective heat exchangersetween the nuclear heat transport system and the industrial

rocess heat transport system. However, the need for efficiency,ompactness, and safety challenge the boundaries of existing heatxchanger technology, giving rise to this study. Previous studies

∗ Corresponding author at: Idaho National Laboratory, P.O. Box 1625, Idaho Falls,D 83415-3730, USA. Tel.: +1 208 526 6494; fax: +1 208 526 9683.

E-mail address: piyush.sabharwall@inl.gov (P. Sabharwall).

029-5493/$ – see front matter. Published by Elsevier B.V.ttp://dx.doi.org/10.1016/j.nucengdes.2012.03.030

Published by Elsevier B.V.

have been performed to update the intermediate heat exchangerlocated downstream from the primary heat exchanger, mostlybecause its performance is strongly tied to the ability to employmore efficient conversion cycles, such as the Brayton cycle usingsupercritical CO2 (Harvego, 2006). Fig. 2 presents a framework forheat exchanger selection and design, depending on the conditionand environment of the application.

Major alternatives and criteria are being developed for the sec-ondary heat exchangers (SHXs). This study describes how thesecriteria and the alternatives are evaluated using the analytical hier-archy process (AHP), which is a structured technique developed formulticriteria decision making, currently among the most widelyused multicriteria decision analysis techniques. The AHP helpsdecision makers find the best option that suits the goal and under-standing of a problem. It was first developed at Wharton Schoolof Business by Thomas L. Saaty in the 1970s based on mathemat-ics and psychology, and has been extensively studied and refinedsince then. It provides a comprehensive and rational frameworkfor structuring a decision problem, representing and quantifying itselements, relating those elements to overall goals, and evaluatingalternative solutions.

AHP enables decision-makers to derive ratio-scale prioritiesor weights as opposed to arbitrarily assigning values. AHP notonly supports decision-makers by enabling them to structurecomplexity and exercise judgment, it allows them to incorpo-rate both objective and subjective considerations into the decision

process.

The AHP method obtains weights and scores by structuring com-plexity as a hierarchy and deriving ratio-scale measures throughpairwise relative comparisons. The pairwise comparison process

P. Sabharwall et al. / Nuclear Engineering and Design 248 (2012) 108– 116 109

r pow

ctejmTfri

mtWtt

a1

Fig. 1. Thermal energy transfer in AHTR fo

an be performed using words, numbers, or graphical bars, andypically incorporates redundancy, thereby reducing measurementrror as well as producing a measure of consistency in comparisonudgments. This method is based on the fact that humans are much

ore capable of making relative rather than absolute judgments.he use of redundancy permits accurate priorities to be derivedrom verbal judgments even though the words are not very accu-ate. The weights or priorities are therefore not arbitrarily assignedn the AHP method.

The AHP involves the mathematical synthesis of numerous judg-ents about the decision problem at hand. It is not uncommon for

hese judgments to number in the dozens or even the hundreds.hile the math is pretty straight forward, it is far more common

o use one of several computerized methods for entering and syn-hesizing the judgments.

The procedure for using the AHP is summarized below (Bhushannd Kanwal, 2004; Forman and Gass, 2001; Hallowell, 2005; Saaty,996).

Fig. 2. Heat exchanger selection fram

er production or process heat applications.

1. Identify and describe the problem.2. Set up the final goal and objective.3. Select alternatives.4. Identify and list criteria.5. Model the problem as a hierarchy containing the decision goal,

the alternatives for reaching it, and the criteria for evaluating thealternatives.

6. Establish priorities among the elements of the hierarchy by mak-ing a series of judgments based on pairwise comparisons of theelements. For example, when comparing potential real-estatepurchases, the investors might say they prefer location over priceand price over timing.

7. Synthesize these judgments to yield a set of overall prioritiesfor the hierarchy. This would combine the investors’ judgmentsabout location, price, and timing for properties A, B, C, and D into

overall priorities for each property.

8. Check the consistency of the judgments.9. Come to a final decision based on the results of this process.

ework (Sabharwall et al., 2011).

110 P. Sabharwall et al. / Nuclear Engineerin

Table 1SHX design requirements and basic conditions for the AHTR.

Parameter Requirements

Reference system configuration AHTR + supercritical steam rankingPCS:

Heat exchanger type - Helical coiled- Printed circuit heat exchanger

Heat exchanger duty (MW) 3400/1700/1130Primary coolant KF-ZrF4

Secondary coolant Water/steamPrimary temperature (Tin/Tout) 679/587 ◦C (supercritical Rankine

cycle)679/586.1 ◦C (subcritical Rankinecycle)

Secondary temperature (Tin/Tout) 251/593 ◦C (supercritical Rankinecycle)241.7/550 ◦C (supercritical Rankinecycle)

Primary pressure (MPa) 0.103Secondary pressure (MPa) 25 (supercritical Rankine cycle)

17.3 (subcritical Rankine cycle)

2

sowhattd

3s

fia

•

•

•

•

•

•

•

•

hat(

i

Tube material Alloy NShell material Alloy N

. Basic requirements for the AHTR heat exchanger design

The operating environment for designing a heat exchangerhould be defined as specifically as possible. The heat exchangerf an AHTR is subjected to a unique set of conditions that bringith them several design challenges not encountered in standardeat exchangers. The somewhat corrosive molten salts, especiallyt temperatures in excess of 700 ◦C, require specialized materialshroughout the system to avoid corrosion, and adverse high-emperature effects such as creep. Table 1 summarizes the basicesign conditions and requirements for the AHTR SHX:

. Goal, alternatives, and criteria for heat exchangerelection

Table 2 summarizes general operating conditions and principaleatures of various heat exchanger types (Shah and Sekulic, 2003)n the current industry. The operating conditions covered in Table 2re defined as follows:

Compactness indicates surface area/heat exchanger core volume;when compactness is high, the heat exchanger can be smaller.System type indicates what fluid phases are generally used for acertain heat exchanger type in the industry.Material indicates what material has been used and experiencedby the current industry for a certain type of heat exchanger.Temperature range indicates the applicable temperature rangesof each type of heat exchanger.Maximum pressure indicates the applicable pressure ranges of acertain type of heat exchanger.Cleaning method indicates if the heat exchanger can be cleanedphysically or chemically.Multistream capability indicates if several independent flow loopscan be connected in a single heat exchanger.Multipass capability indicates if the heat exchanger can split flowinto several paths.

The goal of this study is to evaluate and select the best secondaryeat exchanger (SHX). The two heat exchanger types identifieds possible SHX options that meet the base requirements of high

emperature and pressure for advance reactors are shell-and-tubehelical coiled) and printed circuit heat exchangers (PCHE).

Shell-and-tube is the most typical type of heat exchanger. Its generally built of circular tubes and has considerable design

g and Design 248 (2012) 108– 116

flexibility. It can be designed for high pressures relative to theenvironment and high-pressure differences between the fluids.

The PCHE is a plate type heat exchanger. The plates are chemi-cally etched and then diffusion-welded. Fluid inlet/outlet headersare then welded on.

Table 3 summarizes the criteria for heat exchanger evaluationand selection. Details are explained in the subsections followingthe table.

3.1. Thermal performance

Effective heat transfer being the main purpose of a heatexchanger, thermal performance is the primary criterion for mostof the heat exchanger selection process. Thermal performance con-sists of the following subcriteria:

• High heat transfer performance: heat exchanger types have dif-ferent channel geometries and configurations that greatly affectheat transfer performance. Generally, a smaller channel size pro-vides a better heat transfer coefficient for the same flow-rates,but causes higher frictional losses.

• Effectiveness: when heat is transferred, the system requires mini-mizing loss of useful thermal energy (exergy), i.e. effectiveness isa ratio of the energy actually transferred to the maximum theo-retically possible. Therefore, the heat exchanger’s effectiveness isgenerally considered a very important design parameter. Highereffectiveness means the heat exchanger design is closer to theideal. Typical value for the effectiveness are 0.7–0.9 for conven-tional shell-and-tube design, and 0.9–0.98 for the compact heatexchanger design.

• Fouling: fouling can significantly degrade the thermal perfor-mance of a heat exchanger, especially for liquid coolants. Tocompensate for fouling, an oversized heat exchanger design isrequired and cleaning strategies are needed for specific applica-tions.

3.2. Structural performance

Structural performance is one of the most important criteria ofa heat exchanger because it must be operated safely at high tem-peratures and large pressure differences (if required) for long plantlifetimes. Heat exchanger integrity can be categorized primarily byhow it operates under steady-state and transient conditions. Heatexchangers are mainly exposed to two different stresses: mechan-ical and thermal. Vibration can also degrade the integrity of a heatexchanger, so a heat exchanger structure that is less prone to vibra-tional instabilities is preferred. Since the heat exchanger is operatedfor long periods of time at high temperatures, creep and fatigue arealso important issues that should be considered.

Validating the effects of system pressure and temperature onthe integrity of the joints (diffusion bonding, brazing, and weld-ing) in a prototypical environment under steady-state and transientconditions is necessary and would enable a better selection ofthe heat exchanger. Bonds are subjected to static and dynamicloading while the heat exchanger is operating; the one with bet-ter performance will be ranked higher. Brazed constructed heatexchangers are appropriate at lower temperatures, but there arepotential mechanical integrity problems at higher temperatureswith temperature cycling (Tatara, 1997). The heat exchanger withproven joining techniques will be preferred when compared withunproven techniques.

Materials being considered for the SHX – Alloys 617, 230, 242,800H, and Alloy N – all spontaneously form chromium rich oxidescales that will present problems in making diffusion bonds. Thediffusion bonding process needs to be further developed and

P. Sabharwall et al. / Nuclear Engineering and Design 248 (2012) 108– 116 111

Table 2Principal features of several types of heat exchangers (Shah and Sekulic, 2003).

Heat exchanger type Compactness(m2/m3)

System types Material Temperaturerange (C)a

Maximumpressure (bar)b

Cleaningmethod

Multistreamcapabilityc

Multipasscapabilityd

Shell and tube ∼100 Liquid/liquid,gas/liquid,2phase

s/s, Ti, Incoloy,Hastelloy,graphite,polymer

∼+900 ∼300 Mechanical,chemical

No Yes

Plate-and-frame(gaskets)

∼200 Liquid/liquid,gas/liquid,2phase

s/s, Ti, Incoloy,Hastelloy,graphite,polymer

−35 to +200 25 Mechanical Yes Yes

Partiallywelded plate

∼200 Liquid/liquid,gas/liquid,2phase

s/s, Ti, Incoloy,Hastelloy

−35 to +200 25 Mechanical,chemical

No Yes

Fully weldedplate (AlfaRex)

∼200 Liquid/liquid,gas/liquid,2phase

s/s, Ti, Ni alloys −50 to +350 40 Chemical No Yes

Brazed plate ∼200 Liquid/liquid,2phase

s/s −195 to +220 30 Chemical No No

Bavex plate 200–300 Gas/gas,liquid/liquid,2phase

s/s, Ni, Cu, Ti,special steels

−200 to +900 60 Mechanical,chemical

Yes Yes

Platular plate 200 Gas/gas,liquid/liquid,2phase

s/s, Hastelloy,Ni alloys

∼700 40 Mechanical Yes Yes

Compablocplate

∼300 Liquid/liquid s/s, Ti, Incoloy ∼300 32 Mechanical Not usually Yes

Packinox plate ∼300 Gas/gas,liquid/liquid,2phase

s/s, Ti,Hastelloy,Inconel

−200 to +700 300 Mechanical Yes Yes

Spiral ∼200 Liquid/liquid,2phase

s/s, Ti, Incoloy,Hastelloy

∼400 25 Mechanical No No

Brazed plate fin 800–1500 Gas/gas,liquid/liquid,2phase

Al, s/s, Ni alloy ∼650 90 Chemical Yes Yes

Diffusionbonded platefin

700–800 Gas/gas,liquid/liquid,2phase

Ti, s/s ∼500 >200 Chemical Yes Yes

Printed circuit 200–5000 Gas/gas,liquid/liquid,2phase

Ti, s/s −200 to +900 >400 Chemical Yes Yes

Polymer (e.g.channelplate)

450 Gas/liquid PVDF, PP ∼150 6 Water wash No No

Plate and shell – Liquid/liquid s/s, Ti ∼350 70 Mechanical,chemical

Yes Yes

s/s, stainless steel.a Heat exchanger operational temperature ranges.

r.

blnanjtO

•

•

•

ob

b Heat exchanger maximum applicable pressure.c Capability to connect several independent flow loops in a single heat exchanged Capability to split flow into several paths in the heat exchanger.

onding process parameters and controls identified. Technicaliterature addresses microstructure and mechanical properties, butot the parameters used to perform the bonding. Techniques suchs mating surface pickling, nickel plating, or a nickel foil interlayereed to be investigated. Mechanical testing of diffusion bonded

oints is needed to identify promising joining parameters such asemperature, applied pressure, and hold time for optimization.ther specific concerns that need to be addressed are:

Microstructural stability during the high temperature exposureassociated with diffusion bonding.The limited size of the bonding equipment used for productionof large components.Design and demonstration of adequately representative speci-mens because the components will require the joining of thinwalls that will need to meet complex stress conditions.

Creep–fatigue is mainly influenced by peak stresses becausef the stress concentration effects of coolant channels. The com-ination of creep–fatigue and high temperature environments

complicates creep–fatigue characterization. Creep data at thesehigh temperatures will be needed for the selected material in orderto develop a creep–fatigue interaction diagram.

Mechanical performance, as considered in this study, consists ofthe following subcriteria:

• Mechanical stress: the SHX could be exposed to large pressuredifferentials at high temperature between the primary and thesecondary sides, depending on the integrated systems. The abilityto handle mechanical stress is therefore an important consider-ation in selecting a heat exchanger. Mechanical stress in a heatexchanger is significantly affected by channel/tube configuration,size, and geometry. The bonding method is also very importantto the mechanical integrity of a heat exchanger, which should beable to withstand the mechanical stresses in the given environ-

ment.

• Thermal stress: thermal stress should be considered because largetemperature gradient exists in the heat exchanger. Thermal stressis affected by heat exchanger configurations and geometries.

112 P. Sabharwall et al. / Nuclear Engineerin

Table 3Criteria for heat exchanger selection.

Criteria Subcriteria

Thermal performance High heat transferperformance (heattransfer/pumping power)High effectivenessFouling

Structural performance (evaluated by AmericanSociety of Mechanical Engineers Boiler andPressure Vessel [ASME BPV] code design rules)

Mechanical stressThermal stressVibration

Material performance Geometry (heat exchangerwall thickness)Corrosion allowance indesign (uniform corrosion)Localizedcorrosion/environmentalcrackingFluid compatibility

Technology readiness MaterialFabrication methodASME BPV code statusIndustrial experience

System integration SizeAdaptability

Tritium permeation MaterialGeometry (total heattransfer area + wallthickness)

Inspection Ease of inspection(geometry) and field access

Maintenance CleaningWasteRepairing

Initial cost MaterialFabricationInstallation

Operability ReliabilityOperating and

•

3

buhrofwaphcw

••••

operational lifetime is therefore preferred. Nondestructive evalua-tion (eddy current testing, ultrasonic testing, radiography, pressure

maintenance

Therefore, the thermal stress should be minimized in the heatexchanger design.Vibration: vibration is one of the major failure mechanisms ofthe heat exchanger tube or channel. It is highly affected byheat exchanger geometry and configuration. The heat exchangershould therefore be designed to minimize vibration.

.3. Material performance

Corrosion is an important consideration in selecting the SHXecause of the potential corrosive effects of the secondary coolantsed, such as fluoride salts. A factor to consider in the lifetime of aeat exchanger will be its resistance to corrosion in the given envi-onment. Corrosion resistance will be a function of the materialsf construction, as well as the thickness of various sections. Designeatures of the heat exchanger and the material of constructionill also impact corrosion. Corrosion is resisted by using special

lloys in construction. If the selected material cannot effectivelyrevent corrosion, a better or a preferred option will be for theeat exchanger to have thicker walls (more forgiving) and moreorrosion allowance incorporated into its design when comparedith other heat exchangers. The subcriteria are:

Geometry.Corrosion allowance in design (uniform corrosion).

Localized corrosion/environmental cracking.Fluid compatibility.

g and Design 248 (2012) 108– 116

3.4. Technology readiness

Technology readiness is an important consideration for heatexchanger selection. A technology may look promising for a givenindustrial application but still need to be demonstrated underactual operating conditions. The subcriteria selected for technologyreadiness are:

• Material: the material used in the manufacture of the heatexchanger is an essential consideration in accurately estimatingthe overall lifetime of the heat exchanger.

• Fabrication method: some heat exchanger types, such as PCHEs,use unique fabrication methods such as photochemical etch-ing and diffusion bonding. Readiness for the fabrication methodshould be evaluated for certain candidate materials.

• American Society of Mechanical Engineers Boiler and Pressure Ves-sel (ASME BPV) code status: the material used in manufacturingthe heat exchanger should be supported by the ASME B&PVcode, even though it may exhibit excellent thermomechanicaland chemical performances. The readiness of the ASME BPV codeshould be considered in the heat exchanger evaluation process.

• Industrial experience: experience with a proven technology isalways recommended for nuclear systems applications becauseof low uncertainty.

3.4.1. System integrationThe main role of the SHX in the NGNR is to integrate the nuclear

reactor with the power conversion system and process applicationplant. Integrating the heat exchanger with both systems shouldtherefore be considered. The subcriteria are:

• Size: the size of the heat exchanger is important for integrating itwith both systems. A smaller size is generally preferred.

• Adaptability: the heat exchanger selected must be adaptable tothe system selected for integration with the heat exchanger.

3.4.2. Tritium permeationTritium is mainly generated in the advance reactor core by

ternary fission and various neutron reactions. A major health con-cern is tritium permeation through the high temperature metallicsurfaces because it is so small. Since tritium is radioactive iso-tope, it can eventually radioactively contaminate the industrialsystem and products. Less heat transfer surface area and thickerheat exchangers are therefore preferred in the design to mitigatetritium permeation. The subcriteria are:

• Material: tritium permeation of the heat exchanger is significantlyaffected by the type of tube or channel material used. It is alsohighly affected by existing oxide-layers produced by chemicalreactions between metals and steam/water or oxygen.

• Geometry: total tritium permeation through the heat transfer sur-face is affected by heat exchanger geometrical parameters such asthe heat transfer area and wall thickness. Tritium permeation isproportional to the heat transfer area and inversely proportionalto the wall thickness.

3.4.3. InspectionInspection should be easily accomplished to determine the state

of the equipment. Compact heat exchangers will be difficult toinspect. A heat exchanger design that allows joints to be examinedand cracks and crack growth to be more easily identified over its

leak testing, etc.) may be used to evaluate the structural integrity ofheat exchanger components. More insight is not possible because

P. Sabharwall et al. / Nuclear Engineering and Design 248 (2012) 108– 116 113

y for h

hi

3

ieeentw

•

•

•

3

c

Fig. 3. Decision hierarch

istorical inspection and operation data for these heat exchangerss nonexistent.

.4.4. MaintenanceEase of maintenance, such as cleaning, repair, and serviceability,

s important for a successful heat exchanger. All heat exchang-rs should be chemically cleanable, which is more effective andfficient than dismantling and physically cleaning them. The heatxchanger should also have the provision for replacing any compo-ents subject to corrosion, unless it is more economical to replacehe whole unit (Shah and Sekulic, 2003). Thus, a heat exchangerith these capabilities is preferred. The subcriteria are:

Cleaning: the heat exchanger should be cleaned regularly todecrease the probability of fouling inside a channel or tube. Heatexchangers can generally be cleaned by physical means, chemicalmeans, or both, depending on the type.Waste: when cleaning and repairing a heat exchanger, somewaste is discharged; this waste should also be considered in theevaluation process.Repairing: the heat exchanger should be easily repairable, shoulda tube rupture or other problems occur. Reparability is highlydependent on the type.

.4.5. Initial costIn one respect, the life-cycle cost for a given heat exchanger

an serve as a single criterion for comparison. Such aspects as

eat exchanger selection.

development, design, fabrication, installation, and maintenancecosts can be included in the life-cycle cost. However, at this point inthe project, the heat exchanger concepts are not sufficiently devel-oped to provide accurate cost information on which to base thesecomparisons. Therefore, the cost comparisons will be qualitativelyaddressed. The cost could be characterized as:

• Fabrication cost (costs associated with fabrication): complicateddesigns might cost more to fabricate than simpler designs. Inorder to increase the surface area density of the heat exchanger,the fluid channel diameter (or effective diameter) is reduced,which generally increases the net fabrication cost. The fabrica-tion method has to be acceptable and meet all the requirementsimposed by ASME. The heat exchanger designs for the NGNPare still in the development phase, so fabrication cost valuesare rough estimates. The equivalent (same thermal duty) heatexchanger will be compared based on the net fabrication cost;the one with the highest value for a specific thermal load in kW(quantitatively measured as Heat Load [Q] in kW per dollar [$]spent) per money spent will be ranked higher than the others.

• Materials cost (costs associated with construction materials): morematerial volume will be required for heavier heat exchanger, thus

increasing the cost.

• Installation cost: cost associated with installing the heatexchanger could potentially be higher for conventional designswhen compared with compact designs.

114 P. Sabharwall et al. / Nuclear Engineering and Design 248 (2012) 108– 116

Table 4Fundamental scale for pairwise comparison in AHP (Forman and Gass, 2001).

Intensity ofimportance

Definition Explanation

1 Equal importance Two elements contributeequally to the objective

3 Moderate importance Experience and judgmentslightly favor one elementover another

5 Strong importance Experience and judgmentstrongly favor one elementover another

7 Very strong importance One element is favoredvery strongly over another;its dominance isdemonstrated in practice

9 Extreme importance The evidence favoring oneelement over another is ofthe highest possible orderof affirmation

Intensities of 2, 4, 6, and 8 can be used to express intermediate values. Intensities1

3

maaicteP

3

w

3

anpnppsa

3

tbmeswiios

Table 5Subcriteria weight.

No. Criteria Scores

CR1 Thermal performance 9.44CR2 Structural performance

(evaluated by ASME BPVdesign rules)

22.62

CR3 Material performance 19.72CR4 Technology readiness 12CR5 System integration 6.08CR6 Tritium permeation 2.62CR7 Inspection 4.52CR8 Maintenance 5.9CR9 Initial cost 4.08CR10 Operability 13.03

Table 6Alternatives ranking.

Criteria A1 helical coiled A2 PCHE)

CR1 Thermal performance 2.79 6.65CR2 Structural performance

(evaluated by ASME BPVdesign rules)

7.99 14.62

CR3 Material performance 13.89 5.83CR4 Technology readiness 7.62 4.38CR5 System integration 3.34 2.73CR6 Tritium permeation 1.31 1.31CR7 Inspection 3.62 0.9CR8 Maintenance 4.26 1.64CR9 Initial cost 1.66 2.42

walls than PCHEs, making them preferable in corrosion environ-ments. PCHEs are generally constructed of a single material, while

.1, 1.2, 1.3, etc. can be used for elements that are very close in importance.

.4.6. OperabilityThe subcriteria for operability are reliability and operating and

aintenance. Operability would depend on operating conditionsnd environment, for example due to expansion of the outer wallnd inner tube in a helical coil heat exchanger could result in signif-cant fluid bypass, reducing the overall performance for the helicaloil heat exchanger when compared with PCHE. For the most ofhe conditions (steady state or transient) helical coil heat exchang-rs are more reliable and are always easy to maintain compared toCHE.

.4.7. Modeling of the heat exchanger selectionThe AHP for the heat exchanger selection is shown in Fig. 3,

hich is developed based on the criteria in Table 3 and alternatives.

.5. Pairwise comparisons

Once the AHP hierarchy has been constructed, the participantsnalyze it through a series of pairwise comparisons that deriveumerical scales of measurement for the nodes. The criteria areairwise compared against the goal for importance. The alter-atives are pairwise compared against each of the criteria forreference. The comparisons are processed mathematically, andriorities are derived for each node. Table 4 shows the fundamentalcale for pairwise comparison recommended in the AHP (Formannd Gass, 2001).

.6. AHP software (for evaluation of model): MakeItRational

This study uses MakeItRational software to evaluate and selecthe AHTR heat exchanger. MakeItRational is a well-known, Web-ased, multicriteria, decision-making software based on the AHPethod that uses pairwise comparisons to weight and rate pref-

rences. This software provides group decision evaluations andensitivity analysis results. Fig. 4 shows a screenshot of this soft-are. Currently, all goals, alternatives, criteria, and subcriteria are

mplemented into the software with the hierarchy structure shown

n Fig. 3. This setup will be finally used for evaluation and selectionf the heat exchanger in the next stage with adequate feasibilitytudies.

CR10 Operability 9.77 3.26Total 56.25 43.75

3.7. Results and discussion for secondary heat exchangerevaluation

Based on scores and weights obtained from pair wise com-parisons, final priorities of the alternatives were evaluated withrespect to the goal. Table 5 and Fig. 5 show the subcriteria weight,this weight were obtained by panel discussion on each weightand criteria and further translated into a matrix with the help ofMakeItRational AHP methodology. Structural performance (CR2),material performance (CR3), operability (CR10), and technologyreadiness (CR4) were weighted as the important criteria for theSHX. Tritium permeation (CR6), initial cost (CR9), inspection (CR7),and maintenance (CR8) were weighted relatively low. Table 6 andFig. 6 show the evaluated ranking of the two alternatives: Helicalcoiled heat exchanger (A1) and PCHE (A2). The helical coiled heatexchanger shows higher overall rating than the PCHE. The rate ofthe PCHE is 43.75 and the rate of the helical coiled heat exchanger is56.25. These ratings were estimated using the scores obtained fromthe pair-wise comparisons. The detailed evaluation methodologycan be referred to the book by Saaty (1996). Fig. 6 compares thetwo alternatives for various criteria, giving the following results:

• Thermal performance: PCHEs have better heat transfer perfor-mance and are more effective because they are more compactthan the helical coiled heat exchangers.

• Structural performance: PCHEs handle mechanical stress betterthan helical coiled heat exchangers because they are diffusionbonded and have relatively larger thickness-to-diameter ratios.They are also prone to less vibration than the helical coiled heatexchangers, which have lots of flow tubes/channels.

• Material performance: helical coiled heat exchangers have thicker

helical coiled heat exchangers consist of several materials fortheir different components.

P. Sabharwall et al. / Nuclear Engineering and Design 248 (2012) 108– 116 115

Fig. 4. Screenshot of MakeItRational.

Fig. 5. Subcriteria weight.

Fig. 6. Alternatives ranking.

1 eerin

•

•

•

•

•

•

mnop

4

aayita

16 P. Sabharwall et al. / Nuclear Engin

Technology readiness: helical coiled heat exchangers have a muchhigher technical readiness level than PCHEs.Tritium permeation: PCHEs have a smaller heat transfer area thanshell-and-tube heat exchangers, which is preferable for reducingtritium permeation, but they have thinner walls, which is notpreferred. Thus, the PCHE and helical coiled heat exchanger havea similar tritium permeation rating.Inspection: helical coiled heat exchangers are much easier toinspect than PCHEs, mainly because they are larger.Maintenance: helical coiled heat exchangers are much easier tomaintain (clean and repair) than PCHEs. They can be cleanedeither physically and chemically, while PCHEs can only be chem-ically cleaned.Initial costs: helical coiled heat exchangers cost less to fabricate,operate, and maintain. However, they cost more to manufactureand install because they are larger. In this study, cost was includedin the evaluations. However, in the complex decisions, the AHPusually sets the cost aside until the benefits of the alternativesare evaluated.Operability: helical coiled heat exchangers are more reliable andeasier to operate and maintain than PCHEs.

Overall, the helical coiled heat exchanger is preferred for itsaterial performance (corrosion resistance), technology readi-

ess, system integration ability, inspectability, maintainability, andperability. The PCHE is preferred for its thermal and structuralerformance.

. Conclusions

This study evaluated two heat exchanger options for next gener-tion nuclear reactors using the analytical hierarchy process (AHP),

structured technique developed for multicriteria decision anal-

sis (MCDA). AHP allows for the application of data, experience,nsight, and intuition in a logical way. AHP enables decision-makerso derive ratio-scale priorities or weights as opposed to arbitraryssigning values. In the AHP method, weights and scores are done

g and Design 248 (2012) 108– 116

by structuring complexity as a hierarchy and by deriving ratio-scalemeasures through pairwise relative comparisons. The pairwisecomparison process can be performed using words, numbers, orgraphical bars, and typically incorporates redundancy, resulting ina reduction of measurement error as well as producing a measureof consistency of the comparison judgments.

There are many variables to be evaluated in the selection ofa particular type of heat exchanger for a given application. Thisstudy presented the major characteristics that help in evaluationof a heat exchanger, such as: (1) thermal hydraulic performance,(2) structural performance, (3) material performance, (4) technol-ogy readiness, (5) system integration, (6) tritium permeation, (7)inspection, (8) maintenance, (9) initial cost, and (10) operability.Based on scores and weights obtained from pair wise comparisons,the helical coiled heat exchanger showed higher overall rating thanthe PCHE. The helical coiled heat exchanger was preferred for itsmaterial performance (corrosion), technology readiness, systemintegration, inspection, maintenance, and operability, while thePCHE was preferred for its thermal and structural performance.

References

Bhushan, N., Kanwal, R., 2004. Strategic Decision Making: Applying the AnalyticHierarchy Process. Springer-Verlag, London, ISBN 1-8523375-6-7.

Forman, E.H., Gass, S.I., 2001. The analytical hierarchy process—an exposition. Oper-ations Research 49 (4), 469–487.

Hallowell, D.L., 2005. Analytical Hierarchy Process (AHP) – Getting Oriented.Retrieved from: SixSigma.com (21.08.2007).

Harvego, E.A., 2006. Evaluation of Next Generation Nuclear Power Plant (NGNP)Intermediate Heat Exchanger (IHX) Operating Conditions. Idaho National Labo-ratory, Idaho Falls, Idaho.

Saaty, T.L., 1996. The Analytic Hierarchy Process. McGraw Hill, New York, NY (1980reprinted by RWS Publication, Pittsburgh).

Sabharwall, P., Kim, E.S., McKellar, M., Anderson, N., Patterson, M.W., 2011. ProcessHeat Exchanger Options for the Advanced High Temperature Reactor, INL-11-21584, June 2011.

Sekulic, D.P., 1990. A reconsideration of the definition of a heat exchanger. Int. J.

Heat Mass Transfer 33 (12), 2748–2750.

Shah, R.K., Sekulic, D.P., 2003. Fundamentals of Heat Exchanger Design. John Wiley& Sons, Hoboken, NJ.

Tatara, R., 1997. Heat Exchangers: An Overview of Maintenance and Operations,Duke Engineering and Services. EPRI.