A Review of Phase Change Materials for Vehicle Component Thermal Buffering ⇑ Nicholas R

Applied Energy 113 (2014) 1525–1561

Contents lists available at ScienceDirect

Applied Energy

journal homepage: www.elsevier.com/locate/apenergy

Review A review of phase change materials for vehicle component thermal buffering ⇑ Nicholas R. Jankowski a,b, , F. Patrick McCluskey b a Sensors and Electron Devices Directorate, U.S. Army Research Laboratory, 2800 Powder Mill Road, Adelphi, MD 20783, USA b Department of Mechanical Engineering, University of Maryland, 3135 Glenn L. Martin Hall, College Park, MD 20742, USA highlights

A review of latent heat thermal energy storage for vehicle thermal load leveling. Examined vehicle applications with transient thermal proﬁles from 0 to 800 °C. >700 materials from over a dozen material classes examined for the applications. Recommendations made for future application of high power density materials. article info abstract

Article history: The use of latent heat thermal energy storage for thermally buffering vehicle systems is reviewed. Vehicle Received 2 August 2012 systems with transient thermal profiles are classified according to operating temperatures in the range of Received in revised form 24 July 2013 0–800 °C. Thermal conditions of those applications are examined relative to their impact on thermal buf- Accepted 9 August 2013 fer requirements, and prior phase change thermal enhancement studies for these applications are dis- Available online 4 October 2013 cussed. In addition a comprehensive overview of phase change materials covering the relevant operating range is given, including selection criteria and a detailed list of over 700 candidate materials Keywords: from a number of material classes. Promising material candidates are identified for each vehicle system Review based on system temperature, specific and volumetric latent heat, and thermal conductivity. Based on the Phase change material Thermal management results of previous thermal load leveling efforts, there is the potential for making significant improve- Vehicle systems ments in both emissions reduction and overall energy efficiency by further exploration of PCM thermal Thermal buffering buffering on vehicles. Recommendations are made for further material characterization, with focus on Energy efficiency the need for improved data for metallic and solid-state phase change materials for high energy density applications. Published by Elsevier Ltd.

Contents

1. Introduction ...... 1526 2. Vehicle thermal buffer applications ...... 1527 2.1. Low temperature vehicle applications, T < 100 °C...... 1528 2.1.1. Energy storage for cold start improvement ...... 1528 2.1.2. Cabin climate system thermal buffering ...... 1528 2.1.3. Absorption air conditioning loop thermal buffering...... 1529 2.1.4. Cabin payload systems, electronics thermal protection ...... 1529 2.1.5. Vehicle battery thermal buffering ...... 1530 2.2. Medium temperature vehicle applications, 100 °C 200 °C ...... 1532

⇑ Corresponding author. Address: U.S. Army Research Laboratory, ATTN: RDRL- SED-E, 2800 Powder Mill Road, Adelphi, MD 20783-1197, USA. Tel.: +1 301 394 2337; fax: +1 301 394 1801. E-mail address: [email protected] (N.R. Jankowski).

0306-2619/$ - see front matter Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.apenergy.2013.08.026 1526 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Nomenclature TEG thermoelectric generator Acronyms/abbreviations U.S. United States USABC U.S. Advanced Battery Consortium AC air conditioning CAS Chemical Abstracts Service Symbols C2 command and control c specific heat at constant pressure (kJ/kg K) CO carbon monoxide p k thermal conductivity (W/mK) DOD U.S. Department of Defense th H latent heat (kJ/kg) DOE U.S. Department of Energy n formula number ECU engine control unit T temperature (°C) EHR exhaust heat recovery wt% weight percentage EV electric vehicle HC hydrocarbon Greek symbols HEV hybrid electric vehicle 3 HTF heat transfer fluid q density (kg/m ) kph kilometers per hour (km/h) LPG liquefied petroleum gas Subscripts PCM phase change material M melting PHEV plug-in hybrid electric vehicle f fusion RoHS Restriction of Hazardous Substances l liquid SWaP size, weight and power s solid TES thermal energy storage t transition TCU temperature control unit v volumetric TE thermoelectric

2.3.1. High temperature exhaust energy storage for cold start improvement ...... 1532 2.3.2. Exhaust energy buffering for waste energy electrical conversion ...... 1532 2.4. Summary of vehicle appropriate thermal buffer applications ...... 1533 3. Phase change material overview ...... 1534 3.1. PCM reviews and material coverage ...... 1534 3.1.1. Earlier studies and PCM selection criteria...... 1534 3.1.2. Recent PCM coverage and temperature range expansion ...... 1535 3.1.3. Comments on reported commercial phase change materials ...... 1536 3.2. Phase change materials by category ...... 1537 3.2.1. Organic materials ...... 1537 3.2.2. Inorganic materials...... 1538 3.2.3. Solid–solid phase transition materials ...... 1540 3.3. Summary of available PCMs ...... 1541 4. PCM comparison for vehicle applications ...... 1541 4.1. Low temperature materials, T < 100 °C ...... 1542 4.2. Medium temperature materials, 100 °C 200 °C...... 1544 4.4. Comments on PCM thermal buffer cost implications ...... 1545 5. Recommendations and conclusions ...... 1546 Disclaimer ...... 1557 Appendix A. Full material tables ...... 1557 References ...... 1557

1. Introduction improving the management of vehicle heat is critical to achieving higher platform efficiency [2,3]. Depending on operating condi- Fuel economy has long been a dominant design goal for com- tions, typical vehicles reject approximately 65–75% of the fuel’s en- mercial vehicles, but recently issued U.S. Department of Defense ergy as waste heat through the exhaust or radiator, and in current (DOD) policy has set increased energy efficiency and fuel economy combat vehicles about 10–15% of the useful energy is devoted to as immediate priorities for military vehicles as well, putting running the cooling system [4,5]. emphasis on the strategic and operational impact of the military’s A number of investigations have been directed at improving overall energy usage [1]. System level analyses by both the DOD overall vehicle thermal efficiency, but these efforts are complicated and the U.S. Department of Energy (DOE) have recognized that by the transient nature of the vehicle’s thermal load. As shown in N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1527

Fig. 1, over any given drive cycle a vehicle can have peak velocity Past studies have attempted to cover PCMs and their use in the and power demands much higher than the average values, with aforementioned applications, but the overall vehicle system has sudden load changes occurring multiple times over the cycle been only tangentially addressed, prompting this report which at- [6,7]. Most vehicle thermal management systems are designed to tempts to provide a consolidated review of vehicle component handle expected worst case conditions as steady-state require- thermal buffering. Section 2 of this paper classifies vehicle systems ments. However, using the drive cycle in Fig. 1b as an example, with transient thermal profiles according to likely phase change running at a constant 100% cooling capacity for such a transient temperature range, examines the thermal conditions of those loading condition would represent more than a 2Â system overde- applications relative to their impact on thermal buffer require- sign relative to the average requirement, with resultant size, ments, and reviews prior PCM thermal enhancement studies. In weight, and power (SWaP) penalties to the vehicle. Thus, there ex- Section 3, we provide a PCM overview including selection criteria ists a need to find better ways to accommodate the unsteady ther- and a detailed list of candidate materials from several material mal load to improve thermal efficiency and overall fuel economy in classes spanning the full range of temperatures appropriate to modern vehicles. vehicle systems. Finally, some of the most promising materials Phase change thermal energy storage (TES) has received much for each vehicle application are identified in Section 4 and recom- attention for non-vehicular applications to load-level transient mendations are made for future material and application research thermal behavior, including facility climate control [8–10], power directions. generation and cogeneration enhancement [11,12], electrical power grid demand side reduction [13], and electronic component thermal protection [14–18]. Solid-to-liquid latent heat absorption 2. Vehicle thermal buffer applications and release has a benefit over sensible heat absorption due to the ideally isothermal phase front that acts as temporary, nearly Civilian and military vehicle components are subjected to a infinite thermal capacitance, reducing overall temperature rise wide range of thermal conditions that either are imposed by the with minimal material volume [19–21]. Properly engineered, external operating environment or are the byproduct of the waste phase change materials (PCMs) can buffer thermal transients and heat from high power components. These components must be allow the system to be designed for the average, rather than the able to operate from sub-freezing temperatures to peak tempera- peak, thermal load. Put in other terms, thermal buffering can allow tures that could reach hundreds of degrees Celsius as seen in cooling systems to be designed based on total energy, rather than Fig. 2 for a modern commercial vehicle [22]. As mentioned previ- peak power, requirements. This could reduce steady-state thermal ously, these temperatures likely do not represent constant condi- overdesign and improve overall vehicle SWaP. tions for the vehicle. Certain component temperatures can be One of the main difficulties in designing a PCM-based thermal expected to increase during surges and drop during idling condi- system, however, is matching an appropriate material to the ex- tions. Engine compartment temperatures may be highly dependent pected thermal condition. Both civilian and military vehicles pres- on vehicle speed if primary cooling is from intake air. In addition, ent a varied set of challenging thermal environments for electrical no matter what the driving profile is for the vehicle, it will likely and mechanical components [22]. Because a material’s melting undergo an ambient-to-operating temperature swing at least once temperature is generally fixed, every application with a different if not several times every day. Many components are designed for desired operating temperature will require a unique PCM, and efficient operation at the nominally steady operating temperature, the component design for one application will not necessarily but the vehicle warm-up can be a highly inefficient operating translate directly to another. Thus, two things must be identified period. for successful implementation of a vehicle-based PCM system: Whatever the cause of the transient vehicle temperatures, (1) what are the thermal conditions for the various vehicle compo- selecting a PCM to assist in the thermal management of any partic- nents or systems and which ones could benefit from thermal buf- ular system requires identifying expected operating temperature fering, (2) what PCMs are available for use in those TES systems, ranges, component temperature limits, and any configuration con- and is sufficient thermophysical data available on which to base straints. Although detailed PCM selection criteria will be discussed design decisions. in Section 3, the choice would be made primarily by identifying a melting temperature range for thermal buffering, whether for ther-

140

120

100

80 speed [mph] speed

60 ,

Speed [kph] Speed [kph] 40

0 power [kW]

time [sec] (a) (b)

Fig. 1. (a) Average and maximum vehicle speeds under different standard driving cycles, adapted from [6], (b) speed and modeled engine power for the highway portion of the US06 drive proﬁle, from [7]. 1528 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

phate dibasic dodecahydrate (TM 36 °C) was used to keep a lique- ﬁed petroleum gas (LPG) engine warm between usage cycles and reduce the need to use a more polluting gasoline combustion stage for cold starts. The PCM kept the engine above LPG evaporation temperature (30 °C) for an additional 10 h relative to the standard system reducing cold start CO and HC emissions by 17% and 29%, respectively [26]. Addressing the issue of diesel fuel component solidiﬁcation at sub-zero temperatures as described in [27],an

unspecified PCM blend (TM 70 °C) was used with a small thermoelectric (TE) heater to initiate reflow in a clogged diesel fuel filter [28]. In this case the PCM was not explicitly designed to prevent fuel solidification, but it provided a long duration thermal reservoir for the TE device to pump heat into the solidified fuel. Finally, in a military heavy vehicle application, the oil pan and filter on a U.S. Army M925 5-ton truck was surrounded by a low temperature

PCM (hexadecane, TM 18 °C) keeping the oil warm for over 12 h after shutdown. This allowed faster cold starts, improved engine lubrication, and about 2–6 times less cold-start cranking energy from the battery [29]. Fig. 2. Peak temperatures likely to be seen by electronics and sensors in various parts of the vehicle [22]. 2.1.2. Cabin climate system thermal buffering Beyond just engine performance, during cold start a significant mal protection or load leveling, and then selecting a material with- amount of energy is used to either raise or lower a vehicle’s cabin in that range that maximizes energy storage. The remainder of this temperature to an occupant acceptable condition. Climate control section examines vehicle systems or components spanning low needs vary greatly with specific vehicle, application, and climate, (<100 °C), mid (100–200 °C), and high (>200 °C) operating temper- but they are becoming more of a concern for the increasingly pop- ature ranges. Thermal conditions for each component are identi- ular electric and plug-in hybrid electric vehicles (EVs and PHEVs) fied along with examples from prior studies and any particular which must draw on limited battery power for cabin conditioning. constraints the application imposes on the PCM. A study on HEV operating strategies identified the effects of ambient temperature as the most complex factor affecting successful prediction of power demand, severely impacting overall vehicle 2.1. Low temperature vehicle applications, T < 100 °C efficiency [30]. Here we divide the application set based on whether the climate control need is for passenger comfort or The low temperature applications are primarily concerned with sub-ambient refrigeration. adapting the vehicle to the environment, including buffering of vehicle components for hot or cold ambient temperatures and sta- 2.1.2.1. Passenger cooling or heating. Commercial vehicles target ca- bilization of in-cabin thermal conditions. bin temperatures of around 25 °C regardless of external conditions, while U.S. military systems are specified by MIL-STD-1472F to pro- 2.1.1. Energy storage for cold start improvement vide cold air to the cabin at a maximum of 29.5 °C [31]. While these The higher oil viscosity and decreased combustion efficiency are steady-state requirements, a review of passenger thermal com- present during vehicle cold start makes it a prime target for ther- fort in vehicles concluded that a dominant complicating factor in mal improvement. A number of non-PCM studies have demon- providing adequate cooling is the transient nature of the passenger strated the thermal benefit of using exhaust heat recovery (EHR) compartment thermal/fluid conditions [32]. In addition, even with to more quickly bring the engine up to efficient operating temper- conventional vehicles the demand for cool air from the cabin air ature. One study by Hyundai Motor Corporation showed that a conditioning (AC) system can be greatest when the vehicle is least 2.5% fuel economy improvement could be achieved by preheating able to provide it, such as when driving at low speeds or idling [33]. both the coolant and gear box oil in a commercial vehicle, with Addressing these transient conditions through direct thermal buf- higher savings expected in heavy diesel and hybrid electric vehicle fering of a vehicle’s cabin cooling system has received little atten- (HEV) systems [23]. Another study on cold temperature HEV oper- tion in the literature, where focus has mainly centered on ation showed a 9% fuel economy boost due to EHR [24]. In both residential and commercial climate control. A general AC system cases, however, the cold-start duration is not eliminated, only de- improvement review by Chua described PCM integration as a novel creased while awaiting exhaust temperature rise. improvement to cooling, but any efficiency improvement was Attempting to eliminate the cold-start penalty altogether, sev- highly dependent on matching both thermal conductivity and eral groups have used PCMs to maintain component warmth long phase change temperature to expected load profile [34]. In addi- after engine shut down. Such a PCM would need a melting temper- tion, Al-Abidi reviewed a number of approaches and configurations ature just high enough to sufficiently warm the component, but for implementing such a system, with most focus given to off-peak low enough to ensure PCM melting even in cooler operating condi- cold storage to achieve ‘‘free cooling’’ during the peak load portion tions. In addition, a lower PCM temperature reduces heat loss to of the day [35]. While actively storing energy would be unlikely in the environment, potentially increasing the time it can maintain a powered-off vehicle, thermal buffering a vehicle’s active cooling component temperature. In a pair of studies by Gumus and Ugurlu, system could either provide sufficient cold air during idling periods a lower temperature PCM was used to retain heat and improve cold or reduce the cooling load during start-up after short vehicle-off start emissions and efficiency. First, sodium sulfate decahydrate periods. PCM choice would necessarily differ from facility applica-

(TM 32 °C) was used on a gasoline engine coolant loop and kept tions, where ice-storage has received most attention, as too-cold engine temperature about 17 °C above ambient after a 12 h wait, direct air impingement could impact passenger comfort. and CO and HC emissions were decreased during startup by about It is also worth mentioning that a signiﬁcant amount of facility 64% and 15%, respectively [25]. In the second study, sodium phos- PCM research over the past few decades has examined direct PCM N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1529 integration with structural material to augment insulating proper- tainer volume) to keep a non-refrigerated container below critical ties with additional thermal capacity. This has mainly focused on temperatures to simulate a passive storage condition or active wallboard materials, although other construction material demon- refrigeration power failure, showing durations almost twice as strations have shown promising results [36,37]. Material sets long as without a PCM [44]. Taking PCM refrigeration further, Liu would vary signiﬁcantly for vehicle applications, but a similar de- et al. [45], and Sharma and Buddhi [46] both experimentally eval- sign approach could prove worthwhile for a vehicle cabin, and the uated the ability of a PCM to maintain a refrigerated van below reader is referred to a comprehensive review of the subject by critical temperatures in the absence of an on-board refrigerator, Tyagi and Buddhi [38]. suggesting use of the warehouse pre-charging approach. Liu devel-

There have been a few successful demonstrations of passenger oped a proprietary low-temperature PCM (TM À26.7 °C), and climate control thermal buffering. In one case, an airport passenger showed they could keep frozen goods below 18 °C for 10 h, using shuttle cooling system was modified with a PCM loaded booster a same weight of PCM as a conventional on board refrigerator with tank [39]. PCM characteristics were not specified, but as a surge half of the energy cost due to off-peak charging. Sharma’s work buffer on the radiator water-glycol loop it would have required a investigated a variety of parameters, such as total PCM mass, temperature between the nominal loop temperature and cabin PCM placement, and air circulation, and the PCM (TM À15.2 °C) temperature (indicated as 18–22 °C in different tests). Previously, was shown to keep the refrigerated space below 5 °C for 2–7 h the shuttle had suffered from poor cabin cooling due to frequent depending on configuration. Finally, an alternative approach was and prolonged idling periods resulting in long temperature pull reported in 2010 by Ahmed et al. [47], where rather than coupling down times. During peak cooling system demand, the thermal to the refrigeration system, the group investigated using a paraffin- storage unit was able to provide the equivalent of about half of based PCM (TM 7 °C, just above the 4 °C compartment tempera- the required cooling capacity with faster cabin temperature pull- ture) to enhance the chilled compartment wall insulation. Com- down. This reduced AC energy consumption by 56% relative to pared to a non-PCM enhanced control trailer, the presence of the available standard AC systems while providing increased passen- PCM reduced the daily peak heat transfer rate by an average of ger comfort. More recently, Delphi Automotive demonstrated a 29%, and total daily heat transfer by 16%, giving credence to the no- ‘‘thermal storage evaporator’’ for a hybrid vehicle that acts as a tion of improving thermal performance through structural material PCM buffered air-side heat exchanger for the vehicle cabin [40]. integration.

Using a small PCM-volume with TM 7 °C, the AC could be turned off and vent temperature maintained below the 15 °C comfort limit 2.1.3. Absorption air conditioning loop thermal buffering for 34–42 s under high thermal loading conditions, and extended As the AC system can draw from 12% to 17% of available engine PCM ‘charge time’ could increase that duration by as much 4Â. power in small commercial vehicles, with most of that energy The company claimed that under certain drive cycles, the technol- being used to drive the compressor [33], several studies have ex- ogy could reduce HVAC fuel usage by 50% while maintaining pas- plored eliminating the electrical cooling load by driving the AC sys- senger comfort [41]. tem from reclaimed exhaust heat, primarily through absorption In addition to the cabin cooling requirement, cabin heating based thermal cycles [48]. A detailed analysis and design of vehicle could draw 4–6 kW from the battery during the start-up period. exhaust driven absorption air conditioning can be found in reports This significantly decreases useful battery range and lifetime and by Lambert and Jones [33,49], and the system concept is shown in has a major impact on EV design, as described in a report on the Fig. 3. Typical refrigeration loop cold and hot side temperatures at impact of using a PCM with an electric passenger car in cold cli- steady state can be about 5 °C and 60 °C, respectively. They showed mates [42]. The study concluded that battery energy draw over that the required air conditioning can be provided via recovered the drive cycle could be reduced by up to 21% by loading the cabin exhaust heat even at steady idle, completely eliminating the AC with approximately 5 L of paraffin (Tm =56°C). It should be noted load from the diesel engine. In that design, as well as in [50], en- that pre-charging of the PCM during vehicle battery charging was ergy was recovered from the exhaust by an oil heat exchange fluid necessary to guarantee a preconditioned cabin. with parallel PCM thermal storage in a shell and tube heat exchanger. PCM temperature in both designs was chosen to be just below 2.1.2.2. Sub-ambient refrigeration. A unique climate control applica- the designed loop hot side temperature (60 °C) to keep the oil tion in modern commercial society is the high use of refrigerated loop warm enough to decrease start-up time, but not necessarily shipping of perishable goods. It has been proposed in several stud- enough to drive the cycle by itself in the absence of sufficient ex- ies to augment or remove on-board diesel-driven vapor compres- haust heat. However, sufficient energy was stored in the PCM to re- sion refrigeration systems with an on-board sub-ambient TES duce the initial cooling delay and load-level the cooling system system. As many cases of warehouse-to-warehouse shipment have during variable driving conditions. Additionally, high performance fairly predictable transport times, a PCM-based refrigeration sys- absorption systems in non-vehicular applications typically require tem could permit at-warehouse pre-charging, possibly removing a higher temperature condensing side, perhaps from 85 to 95 °C. the need for the vehicle to carry the weight of a vapor refrigeration Future development of PCMs for this application could include compression system. Warehouse pre-charging could be done with materials spanning this higher range, perhaps enabling higher effi- a higher efficiency refrigeration system, and using off-peak electri- ciency systems on vehicles, or even higher to permit the PCM to di- cal power could further reduce operating costs. rectly drive the adsorption cycle during low temperature exhaust There have been a number of analytical and experimental stud- conditions. ies into PCM refrigeration systems. In 2003, Simard and Lacroix performed an analytical analysis of PCM refrigeration, including 2.1.4. Cabin payload systems, electronics thermal protection the effects of frost build-up on the heat exchanger [43]. Even in While generally less extensive in commercial vehicles, many non-ideal conditions, they showed that using a very low tempera- military vehicles are configured with numerous high power elec- ture PCM (TM À43 °C) a refrigerated truck could provide the tronic components within the crew cabin. A recent power study equivalent of 3500 W of cooling for 8 h using only 3% of the refrig- of a M2A3 Bradley vehicle detailed several electrical load scenarios erated compartment volume for thermal storage. In 2012, Oro et al. with total electrical power ranging from 4 to 10 kW [51]. These and Liu et al. investigated the use of PCMs as alternatives to active loads included command and control (C2) computers and displays, refrigeration in transport vehicles. Oro investigated the ability of a radio systems, navigation systems and others, where the total small amount of PCM (TM À21 to À18 °C, again about 3% of con- equipment electric load can far exceed the approximately 100– 1530 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Fig. 3. (a) Vehicle absorption air conditioning system with a PCM thermal buffer driven by waste heat recovery from the exhaust stream and (b) notional implementation layout, from [33].

150 W metabolic thermal load per crew member [52]. Most of vibration and shock imposed by the military environment, but a these systems are not integrated into the vehicle cabin, and instead similar set of battery materials and are being explored [60,61]. vent heat to the cabin air which must then be rejected through an While Li-Ion technology is meeting the USABC near term goals, overburdened, high thermal resistance AC system. In addition, optimal battery performance requires operation over a much nar- many electronic systems being loaded into military vehicles are rower temperature range. As reported in an overview of the Chev- derived from commercial components with lower maximum tem- erlet Volt Li-Ion battery system, the available output power rolls perature limits than militarized hardware, further increasing the off very quickly outside of a 20–40 °C temperature band, falling cooling problem. to 60% of rated power below 0 °C and above 50 °C [62]. Addition- A recent study modeled two improvements to such a situation ally, several battery pack thermal studies have shown that both on a Cougar mine protected armored vehicle Mine Protected Ar- absolute temperature and temperature uniformity during use can mored Vehicle loaded with C2 equipment [53], an example of signiﬁcantly affect battery pack operation and lifetime [63,64]. which is shown in Fig. 4 [54]. First, a cooling rail architecture Non-uniform heating can occur during rapid discharge creating was implemented, whereby each electronic system was provided highly variable pack temperature distributions, and under certain a heat exchanger to a separate liquid cooling loop to avoid passing conditions localized temperature excursions can lead to thermal the heat through the cabin air. Second, a PCM was integrated with runaway in lithium-based battery packs [65,66]. An in-depth re- the heat sink to provide local thermal buffering. Models showed view of battery thermal management requirements and ap- that, with the improved system, components never reached critical proaches covering these and other concerns was compiled by Rao temperature (85 °C), and the use of the PCM extended operating and Wang [67], and a series of necessary battery thermal control time by one hour in the case of coolant failure. In this system, systems have been examined by Pesaran [68]. Most recent com- the modeled PCM was acetamide embedded within aluminum foam to enhance thermal heat spreading. The acetamide had a melting temperature just below the assumed electronics critical temperature.

2.1.5. Vehicle battery thermal buffering The move toward more complex vehicle electrical architectures including electric and hybrid electric vehicles has driven the development of lighter batteries with higher power and energy density. These batteries make up one of the larger cost elements in newer commercial vehicles, and their performance and lifetime are strongly inﬂuenced by operating temperature [55,56]. A summary of common vehicle battery technologies, their typical operating temperature ranges and other relevant parameters is given in Table 1. The U.S. Advanced Battery Consortium (USABC) has set stan- dards for commercial electric vehicle batteries to be able to operate over the range of À30 °Cto52°C, with a future technology goal of À40 °Cto85°C [58,59]. For military applications, additional imple- Fig. 4. Cougar Mine Protected Armored Vehicle similar to that studied in [53], mentation difﬁculty comes from the wider range of temperature, image from [54]. N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1531

Table 1 using a configuration similar to that shown in Fig. 5 [77,78]. These Common vehicle battery chemistries and performance details [55,57]. studies demonstrated the ability to downsize the cooling system Battery type Op. temp Typical energy storage Energy density capacity by about 1/3 while staying below maximum engine tem- limits (°C) efficiency (%) (W h/kg) perature limits during power surges. Moderate melting tempera- Lead-acid À5 to 40 72–78 25 ture materials were investigated, mainly focusing on the Nickel À40 to 50 72–78 45–80 polyalcohol erythritol (TM 118 °C), although in [78] it was men- cadmium tioned that a slightly lower melting temperature PCM would pro- Nickel metal 10 to 40 90 80 À vide more margin for thermal gradients to develop around the hydride Lithium ion À30 to 50 100 90–190 cooling system. The melting temperatures of an alternate PCM would have to be above the steady coolant temperature (80–100 °C for the primary engine loop, lower temperatures for other loops), but below the loop boiling limit with enough margin for any expected thermal gradients between the loop and critical mercial HEVs, including the Honda Insight and Toyota Prius, have components. made use of conditioned cabin air [69], while the recently released Chevrolet Volt PHEV has a lower temperature antifreeze loop ded- icated to the battery [70]. 2.2.2. Vehicle power electronics thermal buffering Several investigators have explored using a PCM to augment the Increasingly electrified vehicles require more power conversion battery cooling system. The presence of a PCM, typically paraffin electronics to move energy to the various vehicle systems, includ- waxes embedded in a metal or carbon foam, was shown through ing battery packs, traction motors, and accessory payloads. Core both modeling and experiment to decrease peak temperature rise elements of these conversion electronic systems are semiconductor and increase thermal uniformity under high-rate discharge condi- (typically silicon) power transistor and diode array modules that tions [71,72]. In addition, it was suggested that the thermal capac- are used as the switching elements, an example of which is shown ity of the PCM could both reduce the need for cooling system in Fig. 6 [79]. While the power conversion efficiency of these com- overdesign and enable higher battery power output at elevated ponents can be high (>95%), their high power density still creates a ambient temperatures [73]. The paraffin waxes chosen by the challenging thermal management problem. For high power elec- researchers had melting temperatures ranging from 42 °Cto tronics, heat is removed through several packaging layers to a liquid 56 °C, which falls just below the 50–60 °C upper temperature limit cooled baseplate connected to the vehicle’s cooling loop. for NiMH and Li-Ion battery technology. Because significant but As mentioned in Section 2.2.1, future vehicle designs are target- predictable battery heating occurs during the recharging cycle of ing a single, high temperature (>100 °C) cooling loop for the engine PHEVs, especially for rapid-charging systems [74], incorporating and electronics, which does not provide much margin for silicon PCM based thermal protection in a vehicle battery pack could play devices with typical operating temperature limits of only 125– an important role enabling customer-demanded high-rate charg- 150 °C [22]. As a result, current commercial HEVs have had to ing while maintaining battery lifetime and reliability. use separate liquid loops for electronics with temperatures around 60–70 °C [80]. While semiconductor devices made from wide- 2.2. Medium temperature vehicle applications, 100 °C < T < 200 °C bandgap materials, such as silicon carbide and gallium nitride, are being developed with operating temperature limits as high as The medium temperature applications are generally related to 400 °C, other package material limits keep operation of those thermal protection of the underhood drive components, including devices under 200 °C [81]. In addition, while showing significant the engine cooling loop and high-power electronics. progress, those semiconductor devices remain cost-prohibitive for commercial applications, and even military circuit demonstra- 2.2.1. Engine coolant loop thermal buffering tions are still in the early prototype phase [79,82]. Most vehicle engines use an antifreeze coolant loop consisting Additionally, while both DOE and the U.S. Army have been mak- of an ethylene or propylene glycol and water mix that is pumped ing numerous efforts to reduce overall package thermal resistance to a radiator for heat rejection to maintain temperature between [80,83–86], previous studies by the authors [87] and others [88,89] 80 and 100 °C. In some configurations the loop will also cool other have shown that the associated material reduction also tends to re- engine components (e.g., the transmission or gear-box), while pos- duce package thermal capacity which can actually result in higher sibly providing some cabin heating [75]. This loop is generally temperatures under transient loading. Incorporating a PCM into pressurized to maintain the coolant in the liquid phase with the the electronics package to restore thermal capacity has been antifreeze boiling temperature being just above 120 °C. More complicated vehicle systems, especially EVs and HEVs, may have one or more additional liquid loops at lower temperatures for cooling electrical components. Because higher cooling loop temperatures decrease the total radiator size for a given heat load, some future civilian and military vehicle designs are targeting a single 100– 105 °C vehicle coolant loop for all electrical and mechanical components [3,76]. Higher loop temperatures would further decrease radiator size, but such high temperatures can cause performance issues in the engine and reliability issues with elements such as the pump, seals, and hoses [77]. Typically, cooling loop capacity is sized to account for surge and transient loads. Because those peak loads are present for only a fraction of total operating time, design oversizing is reflected in the system pumps, radiators, heat sinks, and other associated hardware. Several studies have looked at shifting some or all of Fig. 5. Engine cooling system with a heat accumulator incorporating a phase the transient load to a PCM thermal buffer or heat accumulator, change material to thermally buffer the coolant temperature, from [77]. 1532 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Fig. 6. (a) Dual switch module, and (b) internal view of module showing transistor and diode arrays bonded down to substrate and baseplate, from [79]. shown to significantly slow this temperature rise [14,15,90].In over 300 °C for 24 h, reducing certain emissions by 84–96%. Be- addition, directly buffering the electronic device creates a shorter cause the U.S. Environmental Protection Agency has estimated that thermal path to the PCM than buffering the cooling loop, poten- there is less than 24 h of off-time between 98% of all automobile tially allowing higher rate transients to be mitigated. trips, a thermally buffered catalytic converter could drastically im- As before with coolant loop thermal buffering, the lower limit prove overall vehicle emissions [96]. Additionally, future emissions for acceptable PCM melting temperature is set by the temperature control trends are leading toward lower temperature catalysts of the cooling medium, and the upper limit by the limiting temper- (potentially below 200 °C) to reduce pollution from large diesel ature of the electronic device. Additional margin must be allowed vehicles with lower exhaust temperatures [97], which could open on both the lower and upper temperature limits to account for the application to a wider range PCM materials. the thermal gradient between the PCM and coolant at steady state and between the PCM and device during the transient. Indepen- 2.3.2. Exhaust energy buffering for waste energy electrical conversion dent DOE and U.S. Army studies have identified erythritol as a A large fraction of fuel energy goes directly out the vehicle ex- candidate PCM for a power electronics thermal buffer heat sink haust due to combustion process inefficiency, and there has been a due to both its melting temperature just below the aforementioned lot of interest in reclaiming this energy to improve overall vehicle silicon thermal limit (TM = 118 °C) and its relatively high latent efficiency. Gasoline engine exhaust peak temperatures can vary heat [7,91]. The DOE study also identified the PCM thermal buffer from 500 to 900 °C, while heavy duty vehicles (typically trucks heat sink concept as a key enabler for using silicon electronics on a with diesel engines) might only have exhaust temperature peaks 105 °C cooling loop, while permitting 30–50% reduction in elec- from 500 to 650 °C [98]. As shown in Fig. 7, the temperature also tronics heat sink size and weight relative to a non-PCM baseline. varies significantly along the exhaust path. A number of investigators have explored converting that waste heat into electricity, 2.3. High temperature vehicle applications, T > 200 °C which could augment or even replace the vehicle alternator to help drive the increasing number of vehicle electrical loads. The high temperature applications all relate to thermally cou- Little attention has been given to using a PCM thermal buffer to pling to the hot engine exhaust for vehicle efficiency improvement. augment electrical conversion of recovered vehicle exhaust energy. The range of available temperatures, shown for a passenger vehicle However, a recent DoE modeling effort examined waste heat in Fig. 7, provides a number of options for PCM selection and inte- recovery on four vehicle classes, from light cars to heavy trucks, gration with the exhaust system. and found that the most limiting condition was insufficient available waste heat during cold starts and intermittent acceleration 2.3.1. High temperature exhaust energy storage for cold start (city versus highway driving) [99]. These are the same conditions improvement that made thermal buffering an important part of the absorption While the cold start improvement efforts described in Sec- cooling analysis mentioned in Section 2.1.3. In fact, Fu et al. re- tion 2.1.1 mainly focus on keeping components above ambient cently performed a detailed power and temperature mapping of conditions in cold environments, the catalytic converter is de- vehicle exhaust as a function of engine speed and load [100], com- signed to run at much higher temperatures to burn off polluting menting on the energy available for WHR. These maps show that compounds from the exhaust stream. Conventional catalysts have significant variation from the peak design point will occur in real light-off temperatures of about 350 °C, below which conversion vehicle operation, which can significantly reduce any recovered efficiency quickly degrades, but they often operate at higher tem- power and increase the payback period of a WHR system. There peratures for peak performance [92]. Even in warm environments, the catalytic converter will cool quickly resulting in an undesirable cold-start condition on the next use. Studies estimate that 60–80% of certain pollutants emitted during a standard drive cycle occur in the first two minutes while the catalytic converter comes up to temperature [93]. Several advanced emissions controls techniques have been proposed, some of which have included use of a high- temperature PCM to maintain converter temperature over an extended off-time. Reports from Burch et al., described the use of a high conductivity aluminum/silicon PCM material (TM 580 °C) and a lower temperature proprietary PCM (TM 350 °C) jacketing the catalytic material on a conventional gasoline engine [94,95]. They found that the lower temperature PCM was more likely to melt completely and had lower losses to ambient than the higher temperature system. It was able to keep converter temperature Fig. 7. Range of available exhaust temperatures in a passenger vehicle, from [4]. N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1533

Fig. 8. (a) p-Type and (b) n-type thermoelectric materials and temperature dependent thermoelectric performance, from [106].

have been case studies on hot-side thermal buffering for thermo- 2.3.2.2. Mechanical–electrical conversion. Researchers have also electric [101] and mechanical conversion technologies [102] for proposed using mechanical thermodynamic cycles instead of (or other high temperature applications (primarily concentrated solar in addition to) TEGs for waste heat recovery, showing potential to- power generation). They examined using a PCM thermal buffer to tal fuel economy improvements near 20% [107,108]. These systems maintain a stable hotside temperature, finding that sufficient en- are primarily Rankine based cycles, which are a mature technology ergy storage allowed for extended operation even after solar input that may better tolerate a wider operating temperature span than had ceased. As it is likely that there may be significant similarities TEGs, but they are typically used in steady conditions and will be between TES material sets and heat exchange requirements for limited in temperature and performance by the chosen working vehicle WHR and high temperature, large scale power generation fluid. BMW recently modeled and tested a steam-based Rankine

(and in particular solar thermal generation), the reader is referred WHR system (Tmax = 350 °C), focusing on the system complexities to a detailed review of the subject by Medrano et al., in [103]. Both created by the highly transient exhaust heat exchanger conditions TE and mechanical generation WHR techniques on vehicles are dis- and the potential for resulting system inefficiency or even critical cussed in the following sections. failure [109]. They focused on the need for a complex control system to adjust for high rate transients, but it is possible that a PCM could buffering the input could mitigate some of those concerns. In a recent historical review of Rankine cycles for WHR, Sprouse and 2.3.2.1. Thermoelectric conversion. One of the most often proposed Depcik [110] indicated that organic fluids are generally chosen vehicle recovery methods involves coupling a thermoelectric gen- over water/steam for cycles with hot side temperatures below erator (TEG) to a portion of the exhaust pipe at the appropriate 370 °C, and there are a significant number of available fluids below temperature for the material. A thorough analysis of thermoelec- 200 °C [108]. Thus, despite the fact that storing energy at higher tric exhaust heat recovery including the impact of materials and temperature is a major factor in achieving net efficiency improve- thermal interface conditions can be found in [98], and an overview ment [111], the expected variation in exhaust temperature and of recent design techniques can be found in [4,104]. A recent cri- operating conditions could mandate selecting a lower working tique of TEG technology identified vehicle waste heat recovery as temperature working fluid and open the application up to a wide one of the few power generation applications where thermoelec- range of potential PCMs. tric usage makes sense relative to more mature conversion technologies, primarily because efficiency of mechanical alternatives is significantly decreased at that power level [105]. In addition, 2.4. Summary of vehicle appropriate thermal buffer applications TEGs are attractive because they are solid-state and lightweight, potentially providing higher reliability and lower mechanical com- The vehicle applications described in this section cover a wide plexity than other conversion options. temperature range and are summarized in Table 2. The desired A number of TE materials are under development, and their PCM melting temperature for any of those applications will depend thermoelectric performance is extremely temperature dependant upon whether it is more for thermal protection or thermal control. as shown in Fig. 8 [106]. The expected variation in both exhaust For example, for thermal protection of current battery systems, a heat and temperature means that both the TEG hotside tempera- buffer temperature of just under 52 °C might be appropriate. How- ture and the magnitude of any thermal gradients will see wide ever, for optimal performance it might be more desirable to main- variation over a realistic drive profile. The TEG would likely be tain a lower battery temperature, which may lie from 20 to 40 °C. off-optimum and operate under inefficient conditions for a Also, it is important to remember for any particular design that a majority of its operating life. A well designed thermal buffer thermal gradient will always exist between the PCM and the item could maintain the material near the optimal performance tem- being buffered, and PCM choice must allow some margin between perature. For the materials shown, this could fall almost any- the target melting temperatures. where within the exhaust temperature span, but would likely Beyond appropriate melting temperature selection, all of the center around the 400–500 °C temperature range. In addition, be- applications share additional thermo-mechanical challenges to cause of the thermal gradient expected down the length of the successful implementation. Packaging the thermal storage element exhaust path, a multi-PCM solution might provide for a spatially will require accommodation for a number of potentially conflicting tailored system with higher efficiency. factors. The component will have to provide adequate heat ex- 1534 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table 2 Summary of vehicle system thermal buffer requirements.

Application Component temperature Likely PCM Comments temperature range Range Limit

Cabin/payload <0 °C (froz.) À50 to 5 °C Tt either just below nominal air temperature if primary cooler, or just above if expected to act refrigeration <5 °C (refr.) as an upper limit buffer during thermal excursions Cold start buffering, <10 °C 10–35 °C Lower PCM temperature reduces loss to ambient and insulation requirements engine ﬂuids Cabin air conditioning 25 °C (com) 5–25 °C Buffering AC cabin fan side of loop, target for PCM is slightly above nominal ﬂuid being loop buffered: either evaporator refrigerant temperature or vent discharge temperature to ensure charging during normal operation 29.5 °C (mil) Cabin heating Similar 40–70 °C Should melt just below heat exchange temperature, whether charging from electrical cabin heater, engine block, or exhaust gasses Vehicle battery pack, 20–40 °C À30 to 52 °C 30–40 °C PCM designed to add thermal inertia either within optimum band to enhance performance, or current (buffer) just below limiting temperature for thermal protection (USABC) 50–55 °C (protect.) Vehicle battery pack, est 20–40 °C À40 to 85 °C 30–40 °C Assumes newer batteries have the same optimal operating point, but higher limiting future (buffer) temperature (USABC) 65–80 °C (protect) Absorption AC loop from 60–100 °C 55–65 °C Current designs based on conventional AC temps. High performance numbers estimated for exhaust (current) improved absorption loops 85–95 °C (high perf.) Cabin electronics 65–85 °C 60–80 °C Commercial electronics derate after 85 °C, militarized electronics impose similar limits Cooling loop

Lower temperature 60–80 °C <125 °C 65–85 °C Tt must be greater than nominal loop temperature, but below coolant limit, pressure relief setpoint, or component limit. Higher temperature loop similar provides less margin for component temperature rise Higher temperature 100–120 °C 110–120 °C Silicon power 125–150 °C 110–120 °C Most silicon electronics will operate above 125 °C, but are ‘derated’ or run at lower power electronics High temp power 175–200 °C 150–175 °C Wide-bandgap electronics can run well at much higher temperatures, but other packaging electronics restrictions limit temperatures Engine exhaust heat 200 °C (org) 180–200 °C Available exhaust temperature varies from the engine to the tailpipe. PCM temperature based recovery, mechanical on working ﬂuid temperature limit

350 °C(H2O) 300–350 °C Engine exhaust heat 200–800 °C 400–500 °C PCM temperature based PbTe based materials. Higher temperature TE would increase thermal recovery, potential thermoelectric Cold start buffering, 320–560 °C >300 °C Temperature should be above the light-off temperature for the catalyst. Future low-temp

catalytic converter catalysts could allow much lower TM

change between the heat source and PCM, including sufficient heat atures below 100 °C. This section attempts to provide a consolidated spreading to overcome conductive resistance in charging/discharg- review of phase change materials covering the full range of vehicle ing the typically low thermal conductivity materials. At the same temperatures of interest. Following a summary of previous PCM re- time, it will have to maintain proper PCM form and containment views, the various PCM types will be described along with a sum- in both solid and liquid state, despite potentially significant PCM mary of materials of interest and available thermal properties. Full volume change over numerous repeated cycles. Finally, whatever tables of potential PCMs will be listed in Appendix A. This data will solution is selected to satisfy these requirements, it must meet then be used to identify PCM candidates warranting further investi- the size and weight restrictions that can be tolerated on what are gation for the vehicle applications described in Section 2 based on typically tightly constrained vehicle systems. While solutions to the material selection factors below. these individual problems may be available, finding an effective comprehensive solution for the vehicle system poses a significant engineering challenge. 3.1. PCM reviews and material coverage

3.1.1. Earlier studies and PCM selection criteria 3. Phase change material overview There have been a number of broad phase change material and application reviews performed over the last fifty years. Earlier It is apparent from the review of vehicle systems that the thermal PCM reviews, including those by Bentilla et al. [14], Hale [112], environment varies greatly across components. Unfortunately, this Lorsch [19], and Abhat [113], provide comprehensive overviews means that no general solution for PCM-based thermal buffering of the design of latent heat TES systems, including material exists, as tailored material selection is required for each specific selection, container or heat exchanger considerations, and sample implementation. Past research has identified a wide variety of PCMs applications. Common to those reviews is the identification of across the temperature ranges of interest for vehicle components, criteria for the selection of a particular PCM. In order of importance, with most attention given to materials with phase change temper- the material must: N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1535

Table 3 Pure parafﬁns from n = 10–100.

3 3 PCM name n CAS # TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) q (kg/m ) kth (W/mK) cp (kJ/kg K) Ref. Decane 10 124-18-5 À29.65 202 147 726 (l) – – [127] Undecane 11 1120-21-4 À25.6 177 130 737 (l) – – [127] Dodecane 12 112-40-3 À9.6 216 161 745 (l) – – [127] Tridecane 13 629-50-5 À5.4 196 148 753 (l) – 2.21 (l) [127] Tetradecane 14 629-59-4 5.5–5.8 227–228 187–188 825 (s) 0.15 (s) 2.07 (s) [112,127] 771 (l) Pentadecane 15 629-62-9 10 205–210 157–161 (l) 765–768 (l) – – [14,112,127] Hexadecane 16 544-76-3 16.7–18.1 236–237 200 835 (s) 0.15 (s) 2.11 (s) [112,127] 770–776 (l) Heptadecane 17 629-78-7 21.7–22.0 213–216 166 775–778 – – [14,112,127] Octadecane 18 593-45-3 27.5–28 243–244 198–211 814–865 (s) 0.15–0.36 (s) 1.9–2.14 (s) [112,120,122,127,131] 774–780 (l) 0.15 (l) 2.3–2.66 (l) Nonadecane 19 629-92-5 32 222 174 782 – – [112,127] Eicosane 20 112-95-8 36.7 247 192 785 (s) 0.15 (s) 2.21 (s) [112,127] 778 (l) 2.01 (l) Heneicosane 21 629-94-7 40–41 200–216 152–170 758–788 – – [14,112,127] Docosane 22 629-97-0 44 246–252 188–199 763–791 – – [14,112,127] Tricosane 23 638-67-5 47.5 232–234 177–186 764–793 – – [112,127] Tetracosane 24 646-31-1 50.6 255 203 796 – – [127] Pentacosane 25 629-99-2 53.5–54 238–240 190–192 798 – – [14,127] Hexacosane 26 630-01-3 56–56.3 250–258 193–206 770–800 – – [14,112,127] Heptacosane 27 593-49-7 58.8 235 181–189 773–802 – 1.92 (s) [112,127] 2.44 (l) Octacosane 28 630-02-4 61–61.6 252–254 196–204 779–803 – – [14,112,127] Nonacosane 29 630-03-5 63.2 239 192 805 – – [112,127] Triacontane 30 638-68-6 65.4 252 203 806 – – [127] Hentriacontane 31 630-04-6 67.9 242 196 808 – – [127] Dotriacontane 32 544-85-4 69.43 266 215 809 – – [127] Tritriacontane 33 630-05-7 71.4 256 207 810 – – [127] Tetratriacontane 34 14167-59-0 73.1–75.9 268–269 217–218 811 – – [121,127] Pentatriacontane 35 630-07-9 74.7 257 209 812 – – [127] Hexatriacontane 36 630-06-8 76.2 269 219 814 – – [127] Heptatriacontane 37 7194-84-5 77.7 259 211 815 – – [127] Octatriacontane 38 7194-85-6 79 271 221 815 – – [127] Nonatriacontane 39 7194-86-7 80.3 271 221 816 – – [127] Tetracontane 40 4181-95-7 81.5 272 222 817 – – [127] Dotetracontane 42 7098-20-6 84.17 273 223 817 – – [127] Tritetracontane 43 7098-21-7 85.5 273 224 819 – – [127] Tettratetracontane 44 7098-21-7 86.4 274 225 820 – – [127] Hextetracontane 46 7098-24-0 88.3 276 227 822 – – [127] Octatetracontane 48 7098-22-8 90.3 276 227 823 – – [127] Pentacontane 50 6596-40-3 92 276 228 825 – – [127] Hexatcontane 60 7667-80-3 99 279 232 831 – – [127] Heptacontane 70 7719-93-9 105.5 281 235 836 – – [127] Hectane 100 6703-98-6 115.25 285 241 846 – – [127]

(a) have a melting temperature within the desired operating materials, including fundamental phase change processes temperature range; and specific heat exchange configurations can be found in (b) have a high latent heat of fusion to minimize the required [114,115]. amount of material; Depending on the specific application, other desirable material (c) melt congruently, or as a single constituent material, so that properties may include non-corrosiveness, low toxicity, and low the solid and liquid phases have the same composition; cost. While some investigators place increased importance on ther- (d) change phase stably and repeatably, with minimal superco- mal conductivity as a limiting factor to heat absorption and rejec- oling or degradation. tion, most PCM implementations have shown that the low thermal conductivity problem can at least partially be mitigated by engi- In addition to those requirements, proper design of a PCM sys- neering some form of heat spreader into the material or container. tem requires sufficient knowledge of material properties. At a This does come as a cost to the system in the form of increased size minimum, a properly formulated transient PCM thermal model and weight for heat spreaders and other components. A compre- or simulation would require melting/transition temperature hensive review of structures used to enhance heat conduction

(TM,t), latent heat of fusion (Hf) or transition (Ht), speciﬁc heat through the PCM can be found in [116]. In addition, it has been (cp), density (q), and thermal conductivity (kth). The latter three shown that placing the PCM in contact with the heat source but properties should also be known for both solid and liquid states, outside the primary thermal path can allow a properly designed and liquid state viscosity would enable modeling of convective ef- system to take advantage of the high latent heat without severely fects in the melt for large PCM volumes. Finally, the ideal material degrading overall heat removal [117,118]. property set would also include temperature dependence of the properties, as they can often change signiﬁcantly over large tem- 3.1.2. Recent PCM coverage and temperature range expansion perature ranges and around the material’s melting point. Two re- Although the aforementioned review studies covered a large cent reviews of modeling and simulation of phase change number of phase change materials, the International Energy 1536 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table 4 Parafﬁn waxes and blends.

3 3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref. Tetradecane + octadecane À4.02 to 2.1 227.52 – – – – [132] 91.67% Tetradecane + 8.33% hexadecane 1.70 156.20 – – – – [125] Tetradecane + docosane 1.5–5.6 234.33 – – – – [125] Tetradecane + heneicosane 3.54–5.56 200.28 – – – – [125] Paraffin blend (n = 14–16) 5–6 152 119 783 – – [20] Pentadecane + heneicosane 6.23–7.21 128.25 – – – – [125] Paraffin blend (n = 15–16) 8 147–153 115–119 751.6–809.5 – – [20] Pentadecane + docosane 7.6–8.99 214.83 – – – – [125] Pentadecane + octadecane 8.5–9.0 271.93 – – – – [125] Paraffin blend (n = 16–18) 20–22 152 – – – – [120] Octadecane + heneicosane 25.8–26 173.93 – – – – [125] Octadecane + docosane 25.5–27 203.80 – – – – [125] ‘‘Paraffin wax’’ 32–32.1 251 208 830 0.514 (s) 3.26 (s) [122] 0.224 (l) 1.92 (l) ‘‘Medicinal paraffin’’ 40–44 146 121 830 0.5 (s) 2.2 (s) [122] 2.1 (l) 2.3 (l) ‘‘Commercial paraffin wax’’ 52.1 243.5 197.1 809.5 (s) 0.15 (s) 2.89 (s) [122] 771 (l) ‘‘Paraffin wax’’ 54.4 146 128 880 – – [112] Beeswax 61.8 177 168 950 – – [112] ‘‘Paraffin wax’’ 64 173.6 159 916 (s) 0.339–0.346 (s) – [120] 790 (l) 0.167 (l)

Table 5 Fatty acid PCM candidates.

3 3 PCM name Formula CAS # TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) q (kg/m ) kth,l (W/mK) cp (kJ/kg K) Ref.

Caprylic acid (octanoic acid) CH3(CH2)6ÁCOOH 124-07-2 16–16.7 148.18–149 145.4–153.9 981–1033 (s) 0.145–0.149 2.11 (s) [133] 862–901 (l) 1.95 (l)

Capric acid (decanoic acid) CH3(CH2)8ÁCOOH 334-48-5 31.5–32 152.7–162.83 153.3–163.5 1004 (s) 0.149–0.153 2.10 (s) [133] 853–886 (l) 2.09 (l)

Lauric acid (dodecanoic acid) CH3(CH2)10ÁCOOH 143-07-7 41–44.2 177.4–211.6 178.6–213.1 1007 (s) 0.139–0.192 1.76–2.14 (s) [113,120,133] 848–870 (l) 2.15–2.27 (l)

Elaidic acid C8H17C9H16ÁCOOH 112-79-8 47 218 185.5 (l) 851 (l) – – [112]

Myristic acid (tetradecanoic acid) CH3(CH2)12ÁCOOH 544-63-8 49–58 186.6–204.5 184.7–202.5 990 (s) – 1.59–2.8 (s) [112,113,133] 844–861 (l) 2.16–2.7 (l)

Pentadecanoic acid CH3(CH2)13ÁCOOH 1002-84-2 52.5 178 – – – – [121]

Palmitic acid (hexadecanoic acid) CH3(CH2)14ÁCOOH 57-10-3 55–64 163–211.8 161.2–209.5 989 (s) 0.103–0.172 2.06–2.2 (s) [21,112,120,133] 845–850 (l) 1.7–2.48 (l)

Stearic acid (octadecanoic acid) CH3(CH2)16ÁCOOH 57-11-4 55–71 186.5–210 175.5–202.7 941–965 (s) 0.097–0.172 2.07–2.83 (s) [21,113,120,133] 839–848 (l) 1.9–2.38 (l)

Agency’s Energy Conservation through Energy Storage Implementing 3.1.3. Comments on reported commercial phase change materials Agreement working group reported in 2004 that the lack of a sin- Numerous commercial PCM products are reported along with gle source for information was hindering scientific development standard materials in the aforementioned material and application [119]. As such, Zalba et al., compiled one of the more comprehen- reviews. While such reporting provides an overview of the range of sive material datasets up to that time from data available in the available materials, the limited traceability and verifiability of scientific literature that included 173 PCMs and referenced over many commercial products severely limits the value of their inclu- 230 other reports, however, only 28 of the materials melted at sion in the general PCM literature. This lack of traceability can be temperatures above 100 °C [120]. Since that time, subsequent re- due to materials no longer being available in their reported form, views by Sharma [121], Agyenim [122] and Halliott et al. [12] ex- references to material sources being no longer valid, or companies panded the set covered by Zalba, mainly between 100 °C and themselves having changed sufficiently to degrade the value of 200 °C. More recently, three high temperature PCM studies were previously reported material data. performed separately by Kenisarin [123], Nomura et al. [124], Three examples encountered by the authors when compiling and Gil et al. [115], which combined cover more than 300 mate- this review highlight these problems. First, the paraffin blend P- rials from 100 °C to over 1000 °C. On the other end of the temper- 116 (SunTech P116 or P-116 Wax, originally produced by the Sun ature spectrum, lower temperature applications such as Oil Company, later Sunoco, Inc.) has been heavily cited over the refrigeration and building environmental control has prompted past 30 years [113,127–130], yet it has not been commercially pro- reviews of lower temperature PCMs. These include a survey of duced since 2006, and Sunoco has since sold off its specialty wax materials for residential and commercial buildings by Cabeza product line to another company. Second, no longer available is et al., bringing together over 300 materials from À30 to 120 °C the TH-series of PCMs from TEAP Energy (Perth, Australia) reported with a significant coverage of commercial materials [125], and in many reviews but only by citing the corporate website, which an overview of cold thermal storage materials for applications be- was bought by a completely unrelated PCM company sometime low 20 °C by Oró et al. [126]. around 2009 (PCM Energy P, Ltd., Mumbai, India). It appears TEAP N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1537

Table 6 Fatty acid derivatives, alcohols and esters.

3 3 PCM name CAS # Formula TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) q (kg/m ) kth,s (W/mK) cp (kJ/kg K) Ref.

1-Decanol 112-30-1 C10H21OH 6 206 171 830.1 (s) – – [19]

Propyl palmitate 2239-78-3 C19H38O2 10 186 – – – – [133]

Isopropyl palmitate 142-91-6 C19H38O2 11 95–100 – – – – [120]

Ethyl myristate 124-06-1 CH3(CH2)12COOC2H5 11 184 – – – – [112]

Isopropyl stearate 112-10-7 C21H42O2 14–18 140–142 – – – – [120]

Butyl stearate 123-95-5 C22H44O2 18–23 123–200 – – 0.21 – [125]

1-Dodecanol 112-53-8 C12H26O 26 200 – – – – [125]

Ethyl palmitate 628-97-7 C18H36O2 23 122 – – – – [125]

Vinyl stearate 111-63-7 C20H38O2 27–29 122 – – – – [120]

Methyl palmitate 112-39-0 C17H34O2 29 205 – – – – [112]

Methyl stearate 112-61-8 C19H38O2 29 169 – – – – [125]

Cetyl caprate 29710-34-7 C26H52O2 29.38 182.5–190.2 – – – 2.68 (s) [134] 2.15 (l)

Trimyristin 555-45-3 (C13H27COO)3C3H3 33 204 – – – – [14]

1-Tetradecanol 112-72-1 C14H30O 38 205 – – 0.358 – [120]

Cetyl laurate 20834-06-4 C28H56O2 38.24 192.2–198.9 – – – 1.65 (s) [134] 2.17 (l)

Stearyl laurate 3234-84-2 C30H60O2 42.21 201.03–201.53 1.97 (s) [135] 2.31 (l)

Methyl 12-hydroxystearate 141-23-1 C19H38O3 42–43 120–126 – – – – [120]

Methyl eicosanoate 1120-28-1 C21H42O2 45 230 – – – – [112]

Stearyl myristate 3234-81-9 C32H64O2 48.86 203.39–203.53 2.07 (s) [135] 2.33 (l)

Cetyl alcohol 36653-82-4 CH3(CH2)15OH 49.3 141 – – – – [121]

Cetyl myristate 2599-01-1 C30H60O2 49.44 222.0–228.4 – – – 1.97 (s) [134] 2.44 (l)

Cetyl palmitate 540-10-3 C32H64O2 51.21 214.6–220.3 – – – 2.51 (s) [134] 2.93 (l)

Methyl behenate 929-77-1 C24H46O2 52 234 – – – – [112]

Ethyl tetracosanoate 24634-95-5 C26H52O2 54 218 – – – – [112]

Methyl oxalate 553-90-2 C4H6O4 54.3 178 – – – – [121]

Cetyl stearate 1190-63-2 C34H68O2 54.63 212.1–216.3 – – – 1.99 (s) [134] 2.6 (l)

Tristearin 68334-00-9 (C17H35COO)3C3H5 56 190.8 164.4 (l) 862 (l) – – [112]

Stearyl palmitate 2598-99-4 C34H68O2 57.34 219.74–219.88 1.55 (s) [135] 1.89 (l)

Stearyl stearate 2778-96-3 C36H72O2 59.22 214.75–214.93 1.86 (s) [135] 2.15 (l)

Cetyl arachidate 22413-05-4 C36H72O2 59.32 224.2–228.7 – – – 1.98 (s) [134] 2.31 (l)

Ethyl cerotate 29030-81-7 C28H56O2 60 226 – – – – [112]

Stearyl arachidate 22432-79-7 C38H76O2 64.96 226.12–226.23 1.93 (s) [135] 2.35 (l)

Dimethyl fumarate 624-49-7 (CHCO2CH3)2 102 242 253 (l) 1045.2 (l) – – [112]

Energy re-incorporated as Phase Change Products Pty, Ltd., and has have qualities that could make them uniquely suited for some some similar products, but all previous TH-series references are applications. In this section we will briefly describe each material misleading dead ends. Finally, even though Environmental Process class, including any class-specific material studies and specific Systems, Ltd. (EPS, Ltd.) is still operating, it’s oft-cited E-series PCM qualities for phase change applications, followed by a list of mate- line is no longer advertised. All of its PCMs appear to be sourced rials and thermal properties for that class. It should be noted that from another company (PCM Products, Ltd.) which does provide because of the varied and sometimes inconsistent naming of or- an extensive material set, but even if the E-series is just rebranded, ganic materials in the literature, attempts were made to list each the material properties cannot be simply traced to those reported pure organic material with its associated CAS registry number1 to for EPS in previous reviews. Material data from all three examples reduce ambiguity. have propagated through material reviews as recently as 2012. Because material data in archival reports should maintain trace- 3.2.1. Organic materials ability to the best extent possible, this review does not include 3.2.1.1. Paraffins and waxes. Paraffins and waxes have been the commercial PCM products, with the exception of some metallic most studied PCMs due to their availability over a wide range of materials where full compositional information has been provided. temperatures for commercial applications, moderate latent heat, Interested readers are referred to [125,132] for reviews including chemical compatibility, low toxicity, and relatively low cost. Pure relevant commercial materials. paraffins (alkanes) are defined by a chemical composition of the

form CnH2n+2 (also referred to as n-alkanes). Both the melting tem- 3.2. Phase change materials by category perature and latent heat content of the materials generally increase with increasing molecular weight. A parafﬁn wax is a Most PCMs are classiﬁed by chemical type as set forth by Abhat blend of n-alkanes (typically >75 wt%) and other hydrocarbon mol- [113], primarily dividing them into organic and inorganic materials ecules (typically <25 wt%). This blending allows a wax to be tai- and their major subtypes. In this report, we have also attempted to lored to a particular melting range while broadening the phase give additional focus to both metallic and solid–solid PCMs, as they transition width. A comprehensive study of pure alkanes as phase 1538 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

change materials, covering a range from C1 to C100 with a melting heat values, several of the sugar alcohols, primarily xylitol, eryth- range of about À183 °C to 115 °C, can be found in [127]. Table 3 ritol, and mannitol, have much higher values than other materials shows thermophysical data for pure alkanes ranging from À30 °C in the same range. Some of the first proposed uses of polyalcohols to 115 °C, while Table 4 lists non-commercial paraffin wax blends. as PCMs are described in a patent by Hormansdorfer on the use of Latent heat values for most pure alkanes lie between 200 and pure or blended polyols for tailored TES [136], but more detailed 285 kJ/kg, while most commercial paraffin blends show signifi- phase transition behavior was reported separately by Talja and cantly reduced latent heats relative to the alkane series. Most par- Roos [137], and Kaizawa et al. [138]. They noted that some of affins do expand by about 10–15% on melting, which can the polyalcohols have volumetric latent heat values as much as complicate container design, and their low specific gravity (typi- twice that of other organic materials, as shown in Table 7, while cally 0.8–0.9) results in higher PCM volume for a given heat stor- still possessing the non-toxic and non-corrosive properties of age. Additionally, most have thermal conductivities under paraffins. 0.25 W/mK which slows thermal transport within the package Among sugar alcohols, erythritol has received the most atten- [113,120]. tion as a TES material due to its well positioned melting temperature and high latent heat [139,140]. However, many sugar alcohols, 3.2.1.2. Fatty acids. Fatty acids are some of the primary components including erythritol, have been shown to exhibit a much larger de- of naturally occurring vegetable and animal oils, and are character- gree of supercooling than lower temperature organics, whereby resolidification occurs at temperatures much lower than the melt- ized by the general formula CH3(CH2)(nÀ2)COOH. The natural rather than synthetic source of these materials reduces concerns about ing temperature. This can cause efficiency and repeatability prob- long term material availability relative to petroleum based paraf- lems in TES systems, and may render a material unusable as a fins. A specific review of fatty acids as sustainable PCMs was per- PCM. The supercooling of erythritol and its mitigation has been ad- formed by Rozanna et al. [133], addressing pure material dressed in a number of reports, and chemical, mechanical, and properties in low melting temperature applications as well as the electrical mitigation methods have been suggested with some suc- wider applicability of tailored binary mixtures of fatty acids. A cess [141–143]. In addition to supercooling, these materials also number of these materials from Rozanna and others are shown undergo the 10–15% volume expansion on melting typical of or- in Table 5. ganic materials, and decomposition at excessive temperatures Fatty acids have well behaved phase change characteristics has been observed [138,144]. with demonstrated long term stability. They share many of the characteristics of organic paraffins, including low thermal conduc- 3.2.1.4. Other organic PCMs. While the majority of the investiga- tivity and significant expansion on melting, but they typically have tions of organic PCMs have focused on the previously described lower latent heats than paraffins in similar melting ranges while categories, numerous other materials covered in the literature having higher costs. In addition, while generally nontoxic some have shown the potential for utility as PCMs. In particular, a num- fatty acids can be mildly corrosive complicating container design ber of carboxylic acid based materials are listed in Table 8 with [21,121]. As with paraffins, blends of fatty acids provide shifted high volumetric latent heat values at temperatures up to 160 °C. melting temperatures, but generally with the cost of significant la- Another series of materials shown in Table 9 are congruently melt- tent heat reduction. A list of examined fatty acid blends can be ing clathrate hydrates, which are ice-like structures stabilized at found in Appendix A. Derivative products of fatty acids, including temperatures above the melting point of water by additional alcohols and esters with other compounds, have been investigated ‘guest’ molecules. Although not much information on TES with as well, some of which are shown in Table 6. Some have higher la- these materials is available outside of the review by Lorsch et al., tent heats than their base acids, including recently reported fatty they present the possibility of use as high energy PCMs in low tem- acid esters of cetyl and stearyl alcohol [134,135], although they perature applications [19]. Finally, a variety of additional organic are still generally lower than for comparable paraffins. materials have been investigated for use as a PCM, including poly- mers, ketones, phenols, amines and others, and those with 3.2.1.3. Sugars and sugar alcohols. Higher temperature organic Hf > 200 kJ/kg are listed in Table 10. Additional organic materials materials have received less attention in the TES literature, but are listed in Appendix A. the majority of sugars and sugar alcohols, also called polyalcohols or polyols, have melting temperatures between 90 and 200 °C 3.2.2. Inorganic materials putting them outside the range of the majority of paraffins and 3.2.2.1. Salt hydrates. Hydrated salts are solutions of salts and fatty acids. While sugars themselves only have moderate latent water that form a crystalline solid with chemical formula of the

Table 7 Sugar and sugar alcohol PCMs.

3 3 PCM name Formula CAS # TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref.

Glycerol C3H8O3 56-81-5 17.9 198.7 250 (l) 1260 (l) – – [112]

Xylitol C5H12O5 87-99-0 92.7–94.5 232–263.3 353–400 1520 – – [137–139]

Sorbitol C6H14O6 50-70-4 95–97.7 110–185 165–278 1500 – – [137–139]

Erythritol C4H10O4 149-32-6 117–118 315–344 466–509 1480 (s) 0.733 (s) 1.383 (s) [137–139] 1300 (l) 0.326 (l) 2.765 (l)

Glucose C6H12O6 50-99-7 141 174 269 1544 – – [112]

Fructose-D C6H12O6 57-48-7 144–145 145 – – – – [12]

Isomalt C12H24O11 64519-82-0 145 170 – – – – [12]

Maltitol C12H24O11 585-88-6 145–152 173 – – – – [12]

Lactitol C12H24O11 585-86-4 146–152 135–149 – – – – [12]

Xylose-D C5H10O5 58-86-6 147–151 216–280 330–428 1530 – – [12]

Xylose-L C5H10O5 609-06-3 147–151 213 326 1530 – – [12]

d-Mannitol C6H14O6 69-65-8 165–168 294–341 438–518 1489–1520 – – [112,138,139]

Galactitol C6H14O6 608-66-2 188–189 351.8 517 1470 – – [139] N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1539

Table 8 Carboxylic acid based PCMs.

3 3 PCM name CAS # Formula TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref. Formic acid 64-18-6 HCOOH 7.8 247 303 1226.7 – – [112]

Acetic acid 64-19-7 CH3COOH 16.7 187 196 (l) 1050 (l) 0.18 (l) 2.04 (s) [112] 1.96 (l)

d-Lactic acid 10326-41-7 CH3CHOHCOOH 26 184 230 1249 – – [112]

beta-Chloroacetic acid 79-11-8 C2H3ClO2 56 147 – – – – [121]

Chloroacetic acid 79-11-8 C2H3ClO2 56 130 205 1580 – – [112]

Heptadecanoic acid 506-12-7 C17H34O2 60.6 189 – – – – [112]

alpha-Chloroacetic acid 79-11-8 C2H3ClO2 61.2 130 – – – – [121]

Glycolic acid 79-14-1 HOCH2COOH 63 109 – – – – [112]

Acrylic acid 79-10-7 CH2@CHCO2H 68 115 – – – – [121]

Phenylacetic acid 103-82-2 C8H8O2 76.7 102 – – – – [121]

Glutaric acid 110-94-1 (CH2)3(COOH)2 97.5 156 223 1429 – – [112]

Benzoic acid 65-85-0 C6H5COOH 121.7 143 181 1266 – – [112]

Sebacic acid 111-20-6 (HOOC)(CH2)8(COOH) 130–134 228 290 1270 – – [12] Maleic acid 110-16-7 HOOC–CH@CH–COOH 131–140 235 374 1590 – – [12]

Malonic acid 141-82-2 HOOC–(CH2)–COOH 132–136 – – 1620 – – [12]

trans-Cinnamic acid 140-10-3 C9H8O2 133 153 191 1250 – – [12]

Chrolobenzoic acid 118-91-2 C7H5ClO2 140 164 253 1540 – – [12]

Suberic acid 505-48-6 (CH2)6(COOH)2 141–144 245 250 1020 – – [12]

Adipic acid 124-04-9 (CH2)4(COOH)2 151–155 260 354 1360 – – [12]

Salicylic acid 69-72-7 HOC6H4COOH 159 199 287 1443 – – [112]

Table 9 latent heat of the storage element. Finally, Lane reported that long Clathrate hydrates from [19]. term use in several residential solar applications was terminated

PCM name CAS # Formula TM Hf (kJ/ due to the corrosive PCM gradually destroying the storage con- (°C) kg) tainer [8]. While additional studies have shown that common materials like stainless steel are generally resistant to corrosion Tetrahydrofuran clathrate 18879-05-5 C4H8OÁ17.2H2O 4.4 255 hydrate from salts undergoing phase change [149], such heavy materials Trimethylamine semi 15875-97-5 (CH3)3NÁ10.25H2O 5.9 239 may not be ideal for heat transfer in mobile applications. Recent clathrate hydrate studies on salt corrosion resistant ceramics for high temperature Sulfur dioxide clathrate –SO2Á6.0H2O 7 247 hydrate phase change applications support the idea that appropriate mate-

Tetrabutylammonium –Bu4NCHO2Á32H2O 12.5 184 rials can be identified for salt hydrate compatibility [150], while formate 32-hydrate low temperature studies have also found acceptable polymer Tetrabutylammonium –Bu4NCH3CO2Á32H2O 15.1 209 materials options [151]. acetate 32-hydrate 3.2.2.2. Fused/molten salt PCMs. Thus far, all of the PCMs described have phase transition temperatures below 200 °C. One of the op- form MÁnH2O, where M is the salt molecule and n represents the tions for higher temperature applications are PCMs composed of number of water molecules in solid solution with the salt. The fused salts (or molten salts, depending on their predominant state). phase change process of the hydrated salt is a dehydration process Although several salts or blends were mentioned in other reviews, where the water molecules come out of the solid solution. Salt hy- they are covered in depth by Kenisarin in [123] including approx- drates are one of the more commonly studied classes of phase imately 200 salts and eutectic blends spanning melting tempera- change materials due to the fact that they generally have higher la- tures from 105 °C to 1680 °C. A selection of those and others are tent heats, densities, and thermal conductivities than paraffins and shown in Tables 12 and 13, with a full material list available in fatty acids and are also relatively inexpensive [121]. A number of Appendix A. While the data available for most of the fused salts salt hydrates with high latent heat values are listed in Table 11. is limited, heat of fusion tends to increase along with melting tem- A full listing of salt hydrates and salt hydrate blends are available perature, and so working with high temperature salts can give ac- in Appendix A. cess to latent heat values 2–3Â that found in other material Earlier uses of salt hydrates for solar and climate control appli- categories. In addition, density and thermal conductivity are gener- cations were covered extensively by Lane [8], with significant de- ally much higher than most organic materials. However, as with tail given to difficulties encountered with those systems. In the salt hydrates container design for molten salt PCMs is compli- particular, being multi-constituent materials many salt hydrates cated by the high application temperatures, the corrosive nature of do not melt congruently. The dehydration process may leave either molten salts, and measured melting expansions for different salts a separated salt and water solution, or a similar salt hydrate with a that vary between 1% and 30% [123]. lower order of n and excess water. Settling of the different salt compounds can occur, and phase segregation within the mixture 3.2.2.3. Metals as PCMs. Metallic PCMs present a significantly dif- can develop over repeated phase change cycles appearing exter- ferent engineering challenge for TES design. Their high densities nally to be a reduced system latent heat [146]. In addition to the and low specific latent heat (kJ/kg) make them unsuitable for incongruent melting, many salt hydrates have been shown to exhi- weight sensitive applications. Because of this, metals and alloys bit significant supercooling [147]. Both of these problems can be have only received cursory attention in the lower temperature mitigated to some degree by adding extra material to the PCM mix- PCM review literature, primarily by Hale [112], with Kenisarin ture, including excess water to reduce the phase segregation prob- providing slightly more focus for high temperatures [123].In lem [113], nucleating agents to reduce supercooling [113], and addition, a study by Gasanaliev provides some data on lower thickening agents to increase long term phase change stability temperature Gallium-based metals [154]. Only recently did a [148]. Both of these solutions, however, generally reduce the net review by Ge et al., address in detail low melting temperature 1540 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table 10

Other organic PCM materials with Hf > 200 kJ/kg.

3 3 PCM name CAS # Formula TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref.

Diethylene glycol 111-46-6 (HOCH2CH2)2O À10 (À7) 247 296 (l) 1200 (l) – – [125]

Triethylene glycol 112-27-6 C6H14O4 -7 247 296 (l) 1200 (l) – – [125]

Tetrahydrofuran (THF) 109-99-9 (CH2)4O 5 280 272 970 – – [126]

Camphenilone 13211-15-9 C9H14O 39 205 – – – – [112]

1-Bromodocosane 6938-66-5 C22H45Br 40 201 – – – – [112]

Caprylone 818-23-5 (CH3(CH2)6)2CO 40 259 – – – – [112]

1-Cyclohexyloctadecane 4445-06-1 C24H48 41 218 – – – – [112]

8-Heptadecanone 14476-38-1 C17H34O 42 201 – – – – [112] Cyanamide 420-04-2 HNCNH 44 209 226 1080 – – [112]

2-Heptadecanone 2922-51-2 C17H34O 48 218 – – – – [112]

3-Heptadecanone 84534-29-2 C17H34O 48 218 – – – – [112]

Camphene 79-92-5 C10H16 50 238 201 (l) 842 (l) – – [112]

9-Heptadecanone 540-08-9 C17H34O 51 213 – – – – [112]

Hypophosphoric acid 7803-60-3 H4P2O6 55 213 – – – – [112]

Acetamide 60-35-5 CH3CONH2 81 241 280 1159 (s) – – [112] 998.6 (l)

Ethyl lithium 811-49-4 LiC2H5 95 389 – – – – [112] High density polyethylene 9002-88-4 – 100–150 200–233 – – – – [12,120]

Catechol 120-80-9 C6H4(OH)2 104.3 207 283 1370 – – [112]

Acetanilide 103-84-4 C8H9NO 115–119 152–222 184–269 1210 – – [112,121]

Succinic anhydride 108-30-5 (CH2CO)2O 119 204 225 1104 – – [112]

Tris(hydroxymethyl)-aminomethane (TAM, Trometanol) 77-86-1 H2NC(CH2OH)3 132 285 385 1350 – – [12]

Urea 57-13-6 CO(NH2)2 133–135 170–258 228–346 1340 – – [12]

Hydroquinone 123-31-9 C6H4(OH)2 172.4 258 351 1358 – – [112]

metallic PCMs as a promising alternative to traditional organic heat, thermal conductivity, density, and specific heat), and most and inorganic PCMs, although the study only examines a small only specify melting temperature. Table 14 details the RoHS non- number of potential materials [155]. Despite this limited cover- compliant and Table 15 a set of select RoHS compliant metals that age, there have been a number of individual studies using low at least have latent heat and melting temperature data available. A temperature solders as PCMs for electronics thermal protection full set of RoHS compliant metals can be found in Appendix A. Fur- [17,18,90,118,156,157]. The metal’s high thermal conductivity ther investigation of metallic PCMs will require improving the and lower specific heat allowed the metallic thermal buffer to re- available material data for RoHS compliant metals in both solid spond quickly to fast transients, even down into sub-millisecond and liquid states, especially in the 30–138 °C gap where only RoHS timescales in [90]. These studies benefited from metallic PCMs’ non-compliant materials are known to have significant volumetric high volumetric latent heat due to their high density, and weight latent heats. was not a primary consideration. The weight penalty of using metallic PCMs, though, can be partially offset by the reduced total 3.2.3. Solid–solid phase transition materials PCM volume for a given energy store, and the high thermal con- A number of materials undergo a solid-state phase transforma- ductivity reduces the need for complicated, and potentially heavy tion below the melting temperature with a significant heat of trans- heat spreading structures. formation. Because these solid–solid PCMs, also referred to as Another consideration for lower melting temperature metals is ‘Plastic Crystals’ or ‘Dry PCMs’ by some researchers, do not need that most contain lead or cadmium making them non-compliant to melt to absorb heat, they avoid some of the liquid-state system with Restriction of Hazardous Substances (RoHS) directives and complexities including containment and expansion on melting other similar international environmental legislation [158]. While [162]. Multi-constituent solid-state PCMs are also far less suscepti- some equipment using PCMs may be exempt from these require- ble to phase segregation and component settling than solutions or ments, especially military applications, many manufacturers are mixtures that go into a liquid phase. While a number of these mate- developing RoHS-compliant hardware even if not legally required rials have latent heats lower than equivalent solid–liquid PCMs, the to do so. Reasons for this include eliminating separate production reduced packaging complexity could make up for some of the lost lines, maintaining an environmentally aware public image, or stay- system energy density. In addition, the solid–liquid latent heat of ing ahead of a continually shrinking list of exempted products many of the materials is still significant, and could be designed as [159]. In any case, PCMs with banned substances will be unlikely an additional thermal buffer stage at higher temperatures. to successfully transition into a product for automotive applications. 3.2.3.1. Polyalcohol solid–solid PCMs. Early reviews covering solid– While some material data can be obtained from the aforemen- solid PCMs include Hale et al. [112] which covered ten organic tioned studies, the number of metallic materials with sufficient compounds with phase transitions from 68 to 184 °C, and a more thermal data for PCM design is low in proportion to the immense comprehensive study by Benson et al. for the U.S. Department of number of potential metals and alloys available. Some lower tem- Energy covering six high-melting temperature polyalcohols and perature material data can be found in studies of liquid metals for their binary solutions [163]. A summary of organic solid–solid electronics cooling applications, such as [160]. As a more extensive PCMs is shown in Table 16, and the properties of binary solid solu- example, the Indium Corporation has a published dataset of over tions of pentaerythritol, pentaglycerine, and neopentyl-glycol are 250 metals and metal alloys with melting temperatures from summarized in Fig. 9. The blends showed high repeatability of 8 °C to just over 1000 °C, with half of those below 200 °C making phase change, but latent heat was reduced below the expected them prime candidates for a number of TES applications [161]. weight based average for some compositions. Melting temperature Unfortunately, as it is a solder database only 11 listed alloys in- was shown to be smoothly adjustable over from 24 °C to 187 °C, clude the data needed for comprehensive thermal modeling (latent with latent heats up to 270 kJ/kg. In addition, the study showed N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1541

Table 11 3 Salt/water and salt hydrate PCM candidates with Hf > 200 kJ/kg or Hf,v > 300 MJ/m .

3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/ q (kg/m ) kth (W/mK) cp (kJ/ Ref. m3) kg K)

NaCl (22.4 wt%) + H2O À21.2 222 246.0 1108 (s) – – [125] 1165 (l)

KCl (19.5 wt%) + H2O À10.7 283 312.7 1105 (s) – – [125] 1126 (l)

Lithium chlorate trihydrate (LiClO3Á3H2O) 8.1 253 435 1720 (s) – – [120]

Potassium ﬂuoride tetrahydrate (KFÁ4H2O) 18.5 231 336 1455 (s) – 1.84 (s) [113] 1447 (l) 2.39 (l)

Calcium chloride hexahydrate (CaCl2Á6H2O) 27.45–30 161.15–192 272–346 1682.4–1802 1.088 (s) 1.4 (s) [19,21,120,122] (s) 1496–1620 (l) 0.53–0.56 (l) 2.2 (l)

Lithium nitrate trihydrate (LiNO3Á3H2O) 29.9– 287–296 452–460 1550–1575 (s) 0.74–0.8 (s) 1.8 (s) [112,145] 30.2 1372–1430 (l) 0.56–0.59 (l) 2.8 (l) Sodium sulfate decahydrate, ‘‘Glauber’s Salt’’ 31–32.4 251.1–254 372 1485 (s) 0.544 (s) 1.93 (s) [112,113,120]

(Na2SO4Á10H2O) 1458–1460 (l)

Sodium carbonate decahydrate (Na2CO3Á10H2O) 32–36 246.5–251 355–362 1440–1442 (s) – – [21,112,120]

Calcium bromide hexahydrate (CaBr2Á6H2O) 34 115.5–138 253–303 2194 (s) – – [120,121] 1956 (l)

Sodium phosphate dibasic dodecahydrate (Na2HPO4Á12H2O) 35–40 265–281 422–427 1507.1–1522 0.514 (s) 1.69 (s) [19,112,113,120,121] (s) 1442 (l) 0.476 (l) 1.94 (l)

Iron chloride hexahydrate (FeCl3.6H2O) 37 226 – – – – [112]

Sodium thiosulfate pentahydrate (Na2S2O3.5H2O) 45–51.3 200–217.2 344–376 1720–1730 (s) – 1.46 (s) [20,21,112,113] 1660–1690 (l) 2.39 (l)

Magnesium sulfate heptahydrate (MgSO4Á7H2O) 48.4 202 – – – – [112]

Sodium acetate trihydrate (Na(CH3COO)Á3H2O) 58–58.4 226–264 – – – – [120]

Lithium acetate dihydrate (Li(CH3COO)Á2H2O) 58–70 150–377 – – – – [112,121]

Sodium hydroxide monohydrate (NaOHÁH2O) 64.3 227.6–272 385–468 1690–1720 (s) – – [112,120]

Barium hydroxide octahydrate (Ba(OH)2Á8H2O) 78 265.7–301 657 2070–2180 (s) 1.255 (s) 1.17 (s) [112,120] 1937 (l) 0.653–0.678 (l) Ammonium aluminum sulfate hexahydrate 95 269 – – – – [120]

((NH4)Al(SO4)Á6H2O) that pentaglycerine supercooling of up to 43 °C could be reduced the past focus on domestic and lower temperature industrial/com- by more than 50% by using a very small amount of nucleating mercial applications. Over half of the studied materials have tran- agent (<0.1 wt%). Again, the lack of a liquid state reduces the like- sition temperatures below 100°, and only salts and metallics have a lihood of the nucleating agent settling over time, increasing the significant number of materials above 200 °C. It is also worth not- effectiveness of such as solution. A number of other studies into ing that of the 704 materials indicated in the figure, only 89 in- binary and tertiary organic material blends have generally showed cluded the complete set of thermal properties described in that transition temperature is adjustable between, and even lower Section 3 (TM, Hf, q, kth, and cp), while over 400 of the others only than, that of the pure constituents, but there is usually a corre- include TM and Hf. As a result, while the figures suggest a broad sponding reduction in latent heat [164–166]. range of available materials, individual efforts to make use of any particular material will likely also require an investment in addi- 3.2.3.2. Other solid–solid transition materials. Other materials have tional material characterization. also received attention for solid–solid TES applications. A series of organometallic compounds called layered perovskites, represented 4. PCM comparison for vehicle applications as (n-CnH2n+1NH3)2MX4 where M is a metal atom, X is a halogen, and n varies from 10 to 16, have been characterized in several re- With the large number of materials and blends identified in this ports for X = Cl, shown in Table 17 [167–169]. Binary mixtures of report, down-selecting a PCM for any particular application can be a number of these materials show behavior similar to the afore- a difficult task. In this section, we examine the phase change mate- mentioned polyalcohol blends, with the ability to adjust transition rial candidates according to low, medium and high vehicle temper- temperature for a C10–C16Co blend from 77 to 100 °C [170]. Finally, ature range and identify the more promising materials for each of Steinert et al., measured solid–solid transition characteristics of a the individual vehicle applications summarized in Table 2. For the number of dialkyl ammonium salts of the form C2nH4nXNH4, with lower temperature applications, the selection process is compli- X being a salt anion, which exhibit solid–solid phase transitions cated by the fact that selection can be made from every material from À2 to 190 °C and transition enthalpies up to 186 kJ/kg, shown category surveyed, as shown by the temperature spans in Fig. 11, in Table 18 [171]. Many of the materials in these sets exhibit multi- while the medium and high temperatures show a significant ple, discrete solid–solid transitions over the ranges shown, with the drop-off in the number of permissible categories. It should be listed latent heat being the sum over those transitions. noted that throughout this section, materials with a range of material parameters (primarily transition temperature and latent heat) 3.3. Summary of available PCMs are listed and graphed according to their mid-range value. Following transition temperature, the primary selection criteria The literature surveyed in this section of the report covers over for any application will be the latent heat of transition. However, 700 individual candidate materials. As stated previously, and whether the specific or volumetric latent heat value is most impor- shown in Fig. 10, the temperature distribution of the PCMs reflects tant will differ based on application. For example, weight may not 1542 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table 12 3 Fused salt PCM candidates up to 600 °C with Hf > 300 kJ/kg or Hf,v > 450 MJ/m .

3 3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref.

Aluminum chloride (AlCl3) 192–192.4 272–280 664–683 2440 – – [21,112]

Lithium nitrate (LiNO3) 250–254 360–373 857–888 2380 – – [21,123,124,152]

Sodium nitrate (NaNO3) 306–310 172–199 388–450 2257–2261 0.5 (s) 1.1 (s) [120,122–124,153] 1.82 (l)

Potassium nitrate (KNO3) 330–336 88–266 186–561 2109–2110 0.5 (s) 0.94 (s) [120,153] 0.5 (l) 1.22 (l)

Sodium peroxide (Na2O2) 360 314 – – – – [21] Lithium hydroxide (LiOH) 462 875 1269 1450 – – [124,152]

Potassium perchlorate (KClO4) 527 1253 3158 2520 – – [21,152]

Table 13 Selected fused salt blends spanning 200–600 °C.

3 3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp,s (kJ/kg K) cp,l (kJ/kg K) Ref.

87%LiNO3 + 13%NaCl 208 369 – – – – – [123]

68.1%KCl + 31.9%ZnCl2 235 198 491 2480 0.8 (s) 2.25 [120,122]

97.4%LiNO3 + 2.6%Ba(NO3)2 253 368 – – – – – [123] 36%LiCl + 63%LiOH 262 485 752 1550 1.1 (s) 2.4 [123]

7.8%NaCl + 6.4%Na2CO3 + 85.8%NaOH 282 316 673 2130 – 2.51 [123]

(46.8–47.0)%KCl + (3.2–3.4)%LiCO3 340–343 375–380 – – – – – [123] + (47.4–47.7)%LiCl + (2.1–2.4)%LiF 41.3%KCl + 58.7%LiCl 352.7 251.5 473 1880 – 0.95 1.33 [92]

20.4%KCl + 60%MgCl2 + 19.6%NaCl 380 400 720 1800 – 0.96 – [153]

14%KCl + 63%MgCl2 + 22.3%NaCl 385 461 1037 2250 0.95 (l) 0.96 [123]

35%K2CO3 + 32%Li2CO3 + 33%Na2CO3 397 276 635 2300 2.02 (l) 1.67 1.63 [123]

(40.0–40.4)%KCO4 + (8.6–8.7)%KCl 422–426 407–412 – – – – – [123] + (33.2–33.8)%KF + (17.6–17.7)%LiF 20%LiF + 80%LiOH 426 869 1390 1600 – 0.88 1 [123]

61%KCl + 39%MgCl2 435 351 741 2110 0.81 (l) 0.8 0.96 [123] (2.8–3.0)%KCl + (41.0–43.0) %KF 440–448 682–692 – – – – – [123] + (42.5–45.5)%LiF + (10.7–11.5)%NaF 67%KF + 33%LiF 442–493 458–618 1159–1564 2530 3.98 (l) 1.34 1.63 [123] 59%KF + 29%LiF + 12%NaF 454 590 1493 2530 4.5 (l) 1.34 1.55 [123]

27%CaCl2 + 25%KCl + 48%MgCl2 487 342 865 2530 0.88 (l) 0.8 0.92 [123]

53%K2CO3 + 47%Li2CO3 488–491 321–342 706–752 2200 1.99 (l) 1.03 1.34 [123]

44%Li2CO3 + 56%Na2CO3 496 370 858 2320 2.09 (l) 1.8 2.09 [123]

67%CaCl2 + 33%NaCl 500 281 607 2160 1.02 (l) 0.84 1 [123]

65%K2CO3 + 35%Li2CO3 505 344 777 2260 1.89 (l) 1.34 1.76 [123]

21%KF + 62%K2CO3 + 17%NaF 520 274 652 2380 1.5 (l) 1.17 1.38 [123]

40%KCl + 23%KF + 37%K2CO3 528 283 645 2280 1.19 (l) 1 1.26 [123]

37%MgCl2 + 63%SrCl2 535 239 664 2780 1.05 (l) 0.67 0.8 [123]

3.3%BaF2 + 8.1%BaMoO4 + 18.5%CaF2 536 653 – – – – – [123] + 36.1%LiF + 34%NaF

20%K2CO3 + 20%Li2CO3 + 60%Na2CO3 550 283 674 2380 1.83 (l) 1.59 1.88 [123]

95.2%NaCl + 48%NiCl2 573 558 – – – – – [123]

(27.0–27.25)%CaF2 + (25.67–25.76)%LiF 593–595 510–515 – – – – – [123]

+ (10.63–10.67)%MgF2 + (36.45–36.57)%NaF be a significant concern relative to volume for a fixed installation few of the clathrate hydrates and other organics standing out just or ground vehicle TES system, whereas weight is a critical factor above 0 °C. These high specific latent heat values are largely in air and space applications. Also, the total system weight and vol- responsible for the large amount of attention they have received ume, and not just those of the PCM, need to be considered, as PCMS in past studies. However, with respect to volumetric latent heat, from different categories may drastically change system interface shown in Fig. 13b, performance of all paraffins drops significantly and containment requirements. As such, while the majority of past due to their low density. The combination of low conductivity studies have simply focused on specific latent heat, both values and low volumetric latent heat increases both the size of the en- should be considered in the selection process. These values for ergy storage container, and the difficulty in spreading heat to the the surveyed materials are shown in Fig. 12a showing the aggre- larger volume, and it may explain some of the less than impressive gate specific latent heat data for transition temperatures up to system performance of some past TES systems using these materi- 1000 °C, and Fig. 12b showing volumetric latent heat for materials als. The salt hydrates also have an acceptably high energy density, with sufficient information available. however the gallium-based metallic PCMs between 10–30 °C have In the following sections, materials will be identified for the dif- the highest volumetric latent heat below 75 °C. These materials ferent applications first by transition temperature, then by the spe- have been largely unexamined likely because of their low specific cific or volumetric latent heat, and, in some cases, also by thermal latent heat, but being RoHS compliant and having a relatively high conductivity. thermal conductivity could make them notable PCMs in this temperature range. 4.1. Low temperature materials, T < 100 °C Table 19 lists noteworthy PCMs and their material properties of interest in this temperature range with potential vehicle appli- Shown in Fig. 13a, apart from water the highest latent heat cations indicated as per Table 2. Again, gallium and a number of materials in this range are the paraffins and salt hydrates, with a gallium-based alloys have some of the highest volumetric latent N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1543

Table 14 RoHS non-compliant metallic PCM candidates.

3 3 PCM formula (Name) TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth,s (W/mK) cp,s (kJ/kg K) Ref. Mercury À38.87 11.4 154.4 (l) 13,546 8.34 0.139 [160,155] 74Ga/22Sn/4Cd 20.2 75.2 120.9 5983 – – [154] 93Ga/5Zn/2Cd 24.6 85.03 148.1 6020 – – [154] 44.7Bi 22.6Pb 19.1In 8.3Sn 5.3Cd (Indalloy 117) 47 36.8 337 9160 15 0.163 (s) [161] 0.197 (l) Bi/Pb/Sn/In (no composition given) 57 29.5 267 9060 (s) 33.2 (s) 0.323 (s) [156] 8220 (l) 10.6 (l) 0.721 (l) 47.5Bi 25.4Pb 12.6Sn 9.5Cd 5In (Indalloy 140) 57–65 36 341 9470 15 0.159 (s) [161] 0.188 (l) 49Bi 21In 18Pb 12Sn (Indalloy 136, Cerrolow Eutectic) 58 28.9 260 9010 10 0.167 (s) [112,161] 0.201 (l) 33Bi/16Cd/51In 61 25 201–251 8040 – 10,040 – – [112] 50Bi 26.7Pb 13.3Sn 10Cd (Indalloy 158, Wood’s metal) 70 32.6–45.8 287–439 9400–9580 18–19 0.146–0.167 (s) [112,154,161] 0.167–0.184 (l) 52Bi/26Pb/22In 70 29 234–293 8069 – 10,103 – – [112] 42.5Bi 37.7Pb 11.3Sn 8.5Cd (Indalloy 160–190) 71–88 34.3 336.483 9810 – 0.146 (s) [161] 52Bi 30Pb 18Sn (Indalloy 39) 96 34.7 333 9600 13 0.151 (s) [161] 0.167 (l) 55.5Bi 44.5Pb (Indalloy 255) 124–125 20.9 218 10,440 4 0.126 (s) [112,161] 0.155 (l) 85Pb 10Sb 5Sn (Indalloy 233) 245–255 0.9 9 10,360 – 0.15 (s) [161] heat values in the lower temperature range, possibly making 4.2. Medium temperature materials, 100 °C < T < 200 °C them suitable for the cold start, air conditioning, and battery buffering applications. There are also a number of lithium-based There are far fewer materials in the medium temperature range, compounds with high specific and volumetric latent heats, with only the highest temperature paraffins, fatty acids, and salt including the organometallic Ethyl Lithium with the highest spe- hydrates extending into it. But other organics, including sugars, su- cific latent heat in the temperature range, although packaging gar alcohols, and carboxylic acids do span almost all the way to will need to account for chemical reactivity. Other metallic mate- 200 °C, and blends of fused salts and metallics cover the full range rials in this range have decent latent heat and thermal conductiv- as well. Specific and volumetric latent heat of the materials in this ity values, but they are all RoHS non-compliant, decreasing range are shown in Fig. 14a and b, respectively. The sugar alcohols commercialization potential. This makes the salt hydrates and have the highest specific latent heat and, given their high density, possibly the lower melting temperature sugar alcohol, Xylitol, also the highest specific latent heat for most of this temperature the most likely candidates for applications like electronics and span, with the exception of the lowest temperature fused salt, alu- cooling loop buffering. minum chloride. With respect to vehicle applications, it is worth

Table 15 select RoHS compliant metallic PCM candidates up to 600 °C.

3 3 PCM formula (Name) TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth,s (W/mK) cp (kJ/kg K) Ref. 67Ga–20.5In–12.5Zn 10.7 67.2 415 6170 – – [154] 78.55Ga–21.45In 15.7 69.7 432 6197 – – [154] 82Ga–12Sn–6Zn 18.8 86.5 516 5961 – – [154] 86.5Ga–13.5Sn 20.55 81.9 482 5885 – – [154] 96.5Ga–3.5Zn 25 88.5 526 5946 – – [154] Gallium 29.8–30 80.12–80.3 473–474 5903–5907 (s) 33.7 (s) 0.34 (s) [112,160,155,161] 6093 (l) 29.4 (l) 0.37–0.397 (l) 66.3In 33.7Bi (Indalloy 162) 72 25 200–202.5 7990–8100 – – [112,161] 60Sn 40Bi (Indalloy 281–338) 138–170 44.4 361 8120 30 0.18 (s) [161] 0.213 (l) 58Bi 42Sn (Indalloy 281) 138.3 44.8 384 8560 19 0.167 (s) [161] 0.201 (l) Indium 156.7–156.8 28.47–28.59 208 7310 (s) 86 (s) 0.243 (s) [160,155,161] 7030 (l) 36.4 (l) 0.27 (l) 91Sn 9Zn (Indalloy 201) 199 71.2 518 7270 61 0.239 (s) [161] 0.272 (l) Tin 232 60.5 440.4 7280 (s) 73 0.222–0.257 [160,155,161] 6940 (l) Bismuth 271–271.4 53.3 521.8–522.3 9790–9800 8.1 0.122 [155,161] 46.3Mg 53.7Zn 340 185 851 4600 – – [123,154] 96Zn–4Al 381 138 915 6630 – – [123,154] 59Al–35Mg–6Zn 443 310 738 2380 – 1.63 (s) [123] 1.46 (l) 60Mg–25Cu–15Zn 452 254 711 2800 – – [123] 64.6Al–5.2Si–28Cu–2.2Mg 507 374 1646 4400 – – [123,154] 66.92Al–33.08Cu 548 372 1339 3600 – – [123,154] 87.76Al–12.24Si 557 498 1265 2540 – – [123,154] 46.3Al–4.6Si–49.1Cu 571 406 2257 5560 – – [123,154] 88%Al–12%Si 576 560 1512 2700 160 1.038 (s) [123] 1.741 (l) 1544 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table 16 Single constituent organic solid–solid transition PCMs.a

3 PCM name CAS # Tt (°C) Ht (kJ/kg) TM (°C) Hf (kJ/kg) Htotal (kJ/kg) qs (kg/m ) cp,s (kJ/kg K) Ref. Neopentane 463-82-1 À133 35.9 À16.54 45.2 81.1 – – [163] Neopentyl alcohol (NPA) 75-84-3 À31 50.6–53.3 51–55 45.9–46.1 96.5–99.4 – 2.79 [163,164] Neopentyl glycol (NPG) 126-30-7 40–48 110.4–131 125–126 44.2–45.3 – 1060 – [112,163,164] Diaminopentaerythritol 36043-16-0 68 184 – – – – – [112] Ammediol (AMPD) 115-69-5 78–80 223.9–264 110–112 28–31.7 251.9–295.7 – 1.79 [112,163,164] Nitroisobutylglycol (NMPD) 77-49-6 79–80 190–201 149–153 28–32 218–233 – – [112,163] Tris(hydroxymethyl) nitromethane (Tris Nitro) 126-11-4 80–82 149 – – – – – [163] Pentaglycerine (PG) 77-85-0 81–89 139–193 197–198 44.6–46 – 1160–1220 1.71 [112,163,164] Monoaminopentaerythritol 36043-15-9 86 192 – – – – – [112] Tris(hydroxymethyl) acetic acid 2831-90-5 124 205 – – – – – [112] Tris(hydroxymethyl)-aminomethane (TAM) 77-86-1 131–135 269.9–285.3 166–172 25–27.6 294.9–312.9 – 1.80 [112,163,164,166] Form-stable HDPE 9002-88-4 133 188 – – – 960 – [163] Dimethylolpropionic acid (DMPA) 4767-03-7 152–155 287–289 194–197 26.8–27 313.8–316 – – [112,163] Pentaerythritol (PE) 115-77-5 182–188 269–303.3 258–260 36.8–37.2 325.8–340.5 1390 – [112,163,164,166]

a Range in values represents spread in reported transition data from referenced sources.

listed as per Table 2. Despite the application gap from 120 to 150 °C, materials in that range were still highlighted to cover potential expansion of application ranges. Again, the highest volumetric latent heat materials have values exceeding 500 MJ/m3, but there are large gaps in this range (below 117 °C and from 118 to 165 °C) that could benefit from the identification of additional, high energy density materials. Apart from loop buffering and electronics buffering with erythritol, polyalcohols also cover the the higher temperature electronics and exhaust buffering applications. In addition, four separate materials types are identified for the 180–200 °C mechanical EHR application, each with different strengths for thermal energy storage. For example, while having lower latent heat, the solid–solid PCM pentaerythritol would not require liquid containment, and could better tolerate Fig. 9. Summary of the solid–solid transition temperatures (solid lines) and latent additives to improve lifetime cycling performance. The metallic heat values (dashed lines) of binary blends of PE, PG, and NPG from [163],asa PCM is RoHS compliant, and the high thermal conductivity would function of the molar fraction of the first constituent component. reduce the need for heat spreading structures. Finally, the sugar alcohols and fused salts represent a trade-offs between latent heat and container chemical compatibility. Optimizing a system for noting that while there has been much interest in PCMs in the these different material classes could produce very different re- range of 100–120 °C for cooling loops and electronics protection, sults, and should most definitely not be based just on a single there are few high energy density materials in that range. Erythri- material property. tol is the exception to this, which explains why it has been the subject of many TES studies in recent years. It was suggested in some 4.3. High temperature materials, T > 200 °C loop cooling studies that using a slightly lower melting temperature PCM would be desirable, however doing so would require Above 200 °C, PCMs are almost entirely comprised of fused salts the use of a material with significantly lower energy density. Again, and metals, as shown in Fig. 15a and b. Again, the lack of data for a with many materials in this temperature range having unexplored large number of both RoHS compliant and non-compliant metallic material properties, including a number of metallics, further work PCMs in this range make it difficult to conclusively state that the identifying other high volumetric latent heat materials could be salts dominate to the extent it appears in the figures. In general, worthwhile. volumetric latent heats get much higher in this temperature range Noteworthy materials in the medium temperature range are due to the high density of both material groups, with metallics listed in Table 20 along with their potential vehicle applications becoming dominant above 500 °C, except for potassium perchlo-

Table 17 layered perovskite solid–solid PCMs – bis-alkylammonium tetrachlorometallates (II) – (n-CnH2n+1NH3)2MCl4.

a 3 a 3 PCM name CAS # Tt (°C) Ht (kJ/kg) qs (kg/m ) Ref. PCM name CAS # Tt (°C) Ht (kJ/kg) qs (kg/m ) Ref.

C10Mn 58675-50-6 32.8 70.34 – [167] C16Cu 63643-59-4 72.8–96.0 79.76 – [167]

C10Cu – 33.8–36.9 62.57 – [167] C16Mn 53290-99-6 73.1–91.0 104.48 – [167] b C12Cu 71163-11-6 52.5–63.8 70.15, 147.13 1111 [167,168] C10Co 56104-89-3 77.7, 82 74.28 – [167,170]

C12Mn 75899-75-1 54.1–56.4 80.80 – [167] C10Zn – 80.1–162.8 100.92 – [167]

C12Co 56104-91-7 60.7–88.0 92.89 – [167] C12Zn 57947-14-5 88.2–156.0 120.23 – [167]

C14Cu – 69.2–79.5 163.99 1186 [169] C16Co 56104-95-1 93.4–164.1 153.79 – [167,170]

C15Cu – 72.3–87.8 126.42 1245 [169] C16Zn 57947-17-8 99.1–160.5 137.52 – [167]

a Temperature range represents minimum and maximum temperatures over multiple solid transitions. Ht is total for all transitions. b Multiple latent heat values represent discrepancy in reported heat release data between sources. N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1545

Table 18

Dialkyl ammonium salt solid–solid PCMs – C2nH4nXNH4 – from [171].

Pcm name Formula Tt (°C) Ht (kJ/kg) Pcm name Formula Tt (°C) Ht (kJ/kg)

a Dioctylammonium iodide DC8I À2to6 58 Didodecylammonium iodide DC12I6580 a Dioctylammonium hydrogen sulfate DC8HSO4 16–190 30 Didodecylammonium chloride DC12Cl 65 123

Dioctylammonium chloride DC8Cl 21 132 Didodecylammonium nitrate DC12NO3 66 185 a Dioctylammonium perchlorate DC8ClO4 23 108 Dioctadecylammonium hydrogen sulfate DC18HSO4 70–82 176

Dioctylammonium bromide DC8Br 30 78 Didodecylammonium bromide DC12Br 73 113 a Dioctylammonium chlorate DC8ClO3 32 122 Didodecylammonium hydrogen sulfate DC12HSO4 74–100 66.5

Dioctylammonium nitrate DC8NO3 45 176 Didecylammonium dihydrogen phosphate DC10H2PO4 81 153 a Didecylammonium chloride DC10Cl 48 119 Dioctadecylammonium chlorate DC18ClO3 84–91 154 a Dioctylammonium dihydrogen phosphate DC8H2PO4 50–70 – Dioctadecylammonium perchlorate DC18ClO4 88.1 185

Didecylammonium chlorate DC10ClO3 55.5 154 Didodecylammonium dihydrogen phosphate DC12H2PO4 90 183

Didecylammonium bromide DC10Br 57 100 Dioctadecylammonium chloride DC18Cl 91.2 174

Didecylammonium nitrate DC10NO3 60 179 Dioctadecylammonium iodide DC18I 93 116

Didodecylammonium chlorate DC12ClO3 61.5 159 Dioctadecylammonium nitrate DC18NO3 93.5 186

Didodecylammonium perchlorate DC12ClO4 62 159 Dioctadecylammonium bromide DC18Br 98 135

a Range in temperature values represent minimum and maximum temperatures over multiple solid-state transitions. Hf is total for all transitions.

120

100

material count 40

PCM Phase Transition Temperature [ C] solid-solid PCMs salt hydrates and blends paraffins and blends fatty acids and derivatives sugars and sugar alcohols other organics fused salts and blends metallic PCMs

Fig. 10. Transition temperature distribution of 740 PCM candidates, sorted by material category.

rate, the salt with the highest speciﬁc and volumetric latent heat. It should also be noted that both fused salts and metallics can present signiﬁcant containment issues due to corrosiveness and/or reactivity at high temperatures. Table 21 lists noteworthy materials in this high temperature solid-solid organic PCMs range. While the energy density of materials in this range is gener-

salt hydrates ally higher than that for materials in lower temperature ranges, there is a notable lack of high energy density materials from

misc. organic PCMs 200 °C to 250 °C, which could have an impact on TES systems for organic Rankine based WHR systems, as well as future catalytic

paraffins converter buffering if useful catalysts with light-off temperatures below 300 °C are developed. For thermoelectric exhaust heat fatty acids & recovery applications, the PCM of choice will highly depend on derivatives the TEG hot side material chosen for the system. Although there sugars and are lithium based materials with high speciﬁc latent heat below sugar alcohols 500 °C, there are some very high volumetric latent heat PCMs

metallic PCMs above this temperature that could warrant developing a system to make use of those materials.

fused salts i 4.4. Comments on PCM thermal buffer cost implications

-200 -100 0 100 200 1500 1600 1700 It goes without saying that implementing a PCM thermal buffer Transition Temperature [ C] system will impact overall vehicle cost. How much it does so, how- Fig. 11. Temperature ranges of the surveyed material categories with low, medium ever, is nearly impossible to state in general terms. In most of the and high vehicle application temperature ranges indicated by the shaded regions. vehicle studies previously described, little attention was paid to 1546 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

more with speciﬁc application and integration method than with PCM type. Factors that will determine whether net cost impact will be signiﬁcantly positive or negative (along with some possible examples) include:

Is the PCM thermal buffer simply an add-on vehicle component, or is it replacing/augmenting another system? – Refrigeration (supplement or replace vapor compression system). – Battery thermal management (replace active with passive cooling). Does the thermal buffer permit downsizing, elimination, or cost reduction in another system? – Coolant loop (reduce capacity on radiator, pumps, etc.). – Electronics (extend use of traditional, lower cost silicon components). – Emissions (reduce need for alternative, more expensive emissions abatement). Does the PCM, or PCM enabled component impact energy usage? What is the net payback? – Engine cold start (reduced engine losses, heating/cooling loads). – Waste heat recovery (reduced climate control costs or increased electrical output). What is the PCM material cost, and does the material choice impact other components? – Metallics cost much more than waxes, but low wax latent heats increase cost per energy stored, and low conductivity can increase heat exchanger size and complexity.

Generally, using some early estimates from Abhat, low temperature paraffins and hydrated salts have the lowest material cost per mass, followed closely by fatty acids and other organics, then Fig. 12. Aggregate values of (a) specific latent heat and (b) volumetric latent heat by higher temperature salts and metals [113]. Of course, these for the 276 materials with that information available. costs will also vary greatly between laboratory grade materials at small quantity, and commercial grade materials at volume. Thus, it is outside the scope of this broad survey to identify the cost im- PCM system cost. This is in sharp contrast to many of the facility pact for any particular PCM application. Specific designs will need thermal management studies, where often the primary focus is to identify net impact of both application and material selection to net energy usage, savings, and system payback period (hence the determine economic feasibility. focus on designing for ‘free cooling’ conditions), where case studies could be more feasibly evaluated for cost. However, even in one of 5. Recommendations and conclusions these studies by Zalba et al., the PCM itself accounted for less than 20% of total system cost, with the balance going to heat exchange Developing future vehicles that are both more energy dense and components and other integration elements [172]. Thus, it is sim- more efficient will require replacing traditional techniques of over- ilarly expected that vehicle PCM thermal buffer costs will vary designing thermal systems to accommodate peak transient loads ]

400 3 700

350 600 300 500 250 400 200 300 150 100 200 50 100

Specific Latent Heat [kJ/kg] 0 0

-50 -25 0 25 50 75 100 Volumetric Latent Heat [MJ/m -50 -25 0 25 50 75 100 Transititon Temperature [ C] Transititon Temperature [ C] (a) (b)

solid -solid PCMs salt hydrates paraffins and blends fatty acids & derivatives other organics sugars & sugar alcohols fused salts & blends metallic PCMs

Fig. 13. (a) Speciﬁc and (b) volumetric latent heats of low transition temperature PCMs. For reference water is marked with a blue circle at 0 °C. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.) N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1547

Table 19

Notable PCMs with Tt from À30 to 100 °C, and potential vehicle applications.

3 a PCM name Type TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) kth,s (W/mK) Potential application

NaCl (22.4 wt%) + H2O Aqueous salt À21.2 222 246 – 1

KCl (19.5 wt%) + H2O Aqueous salt À10.7 283 313 – 1 Diethylene/triethylene glycol Organic À10 (À7) 247 296 – 1 Water Inorganic 0 333–335 305–307 2.3 – Tetrahydrofuran clathrate hydrate Clathrate hydrate 4.4 255 – – 1, 2, 3 Lithium chlorate trihydrate Salt hydrate 8.1 253 435 – 2, 3 82Ga–12Sn–6Zn Metallic 18.8 86.5 516 – 2, 3 96.5Ga–3.5Zn Metallic 25 88.5 526 – 2, 3 Gallium Metallic 29.8–30 80.1–80.3 473–474 33.7 2, 4 Lithium nitrate trihydrate Salt hydrate 29.9–30.2 296 460 0.8 2, 4 44.7Bi 22.6Pb 19.1In 8.3Sn 5.3Cd (Indalloy 117) b Metallic 47 36.8 337 15 4, 5 Lithium acetate dihydrate Salt hydrate 58–70 150–377 – – 4, 5, 6 Sodium hydroxide monohydrate Salt hydrate 64.3 227.6–272 385–468 – 4, 5, 6, 7, 8 50Bi 26.7Pb 13.3Sn 10Cd (Wood’s metal)b Metallic 70 32.6–45.8 287–439 18–19 4, 5, 6, 7, 8 Barium hydroxide octahydrate Salt hydrate 78 265.7–301 657 1.26 4, 5, 6, 7, 8 Xylitol Sugar alcohol 92.7–94.5 232–263.3 353–400 – 6 Ethyl lithium Organometallic 95 389 – – 6 52Bi 30Pb 18Sn (Indalloy 39) b Metallic 96 34.7 333 13 6

a Potential vehicle applications, as per Table 2: 1 – refrigeration, 2 – engine cold start buffer, 3 – cabin air conditioning output buffer, 4 – battery pack buffer/protection, 5 – cabin heating, 6 – exhaust driven absorbtion AC buffer, 7 – cabin/commercial electronics buffer, 8 – low temp. cooling loop buffer. b non-ROHS compliant metals due to Pb and/or Cd content. ]

400 3 700

350 600 300 500 250 400 200 300 150 200 100 50 100

Specific Latent Heat [kJ/kg] 0 0 100 120 140 160 180 200 Volumetric Latent Heat [MJ/m 100 120 140 160 180 200 Transititon Temperature [ C] Transititon Temperature [ C] (a) (b)

solid -solid PCMs salt hydrates paraffins and blends fatty acids & derivatives other organics sugars & sugar alcohols fused salts & blends metallic PCMs

Fig. 14. (a) Speciﬁc and (b) volumetric latent heats of medium transition temperature PCMs.

Table 20

Notable PCMs with Tt from 100 to 200 °C, and potential vehicle applications.

3 a PCM name Type TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) kth,s (W/mK) Potential application Heptacontane Parafﬁn 106 281 235 – 1, 2 Erythritol Sugar alcohol 117–118 315–344 466–509 0.733 1, 2 Tris(hydroxymethyl)-aminomethane Organic 132 285 385 – – Maleic acid Organic 131–140 235 374 – – 58Bi, 42Sn (Indalloy 281) Metallic 138 44.8 384 19 – Xylose-D Sugar 147–151 216–280 330–428 – 3 Adipic acid Organic 151–155 260 354 – 3 Dimethylolpropionic acid (solid–solid) Polyol 152–155 287–289 – – 3 d-Mannitol Sugar alcohol 165–168 294–341 438–518 – 3 Pentaerythritol (solid–solid) Polyol 182–188 269–303.3 374–391 – 4 Galactitol Sugar alcohol 188–189 351.8 517 – 4 Aluminum chloride Fused salt 192–192.4 272–280 664–683 – 4 91Sn, 9Zn (Indalloy 201) Metallic 199 71.2 518 61 4

a Potential vehicle applications, as per Table 2: 1 – high temp. cooling loop buffer, 2 – silicon power electronics buffer, 3 – high temp. power electronics buffer, 4 – lower temp. mechanical exhaust heat recovery buffer.

with more intelligent thermal management methods. The use of with respect to both cooling supply and power generation. Exam- thermal buffering can overcome the temporal mismatch between ination of past research into vehicle systems has identiﬁed a num- thermal supply and demand commonly found in vehicle systems, ber of applications with either a transient mismatch or a thermal 1548 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 ] 3 1200 3000

1000 2500

800 2000

600 1500

400 1000

200 500

Specific Latent Heat [kJ/kg] 0 0

200 300 400 500 600 Volumetric Latent Heat [MJ/m 200 300 400 500 600 Transititon Temperature [ C] Transititon Temperature [ C] (a) (b)

solid -solid PCMs salt hydrates paraffins and blends fatty acids & derivatives other organics sugars & sugar alcohols fused salts & blends metallic PCMs

Fig. 15. (a) Speciﬁc and (b) volumetric latent heats of high transition temperature PCMs.

Table 21

Notable PCMs with TM from 200 to 600 °C, and potential vehicle applications.

3 a PCM Name Type TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) kth,s (W/mK) Potential application

87%LiNO3 + 13%NaCl Salt blend 208 369 – – 1

68.1%KCl + 31.9%ZnCl2 Salt blend 235 198 491 0.8 1 Lithium nitrate Salt 250–254 360–373 857–888 – 1 36%LiCl + 63%LiOH Salt blend 262 485 752 1.1 1,2 46.3 Mg 53.7Zn Metallic 340 185 851 – 1,2 (46.8–47.0)%KCl + (3.2–3.4)%LiCO3 + (47.4–47.7)%LiCl + (2.1–2.4)%LiF Salt blend 340–343 375–380 – – 1,2 14%KCl + 63%MgCl2 + 22.3%NaCl Salt blend 385 461 1037 0.95 (l) 2 20%LiF + 80%LiOH Salt blend 426 869 1390 – 2,3 59%KF + 29%LiF + 12%NaF Salt blend 454 590 1493 4.5 (l) 2,3 Lithium hydroxide Salt 462 875 1269 – 2,3 64.6Al–5.2Si–28Cu–2.2Mg Metallic 507 374 1646 – 3 Potassium perchlorate Salt 527 1253 3158 – 3 46.3Al–4.6Si–49.1Cu Metallic 571 406 2257 – 3 88%Al, 12%Si Metallic 576 560 1512 160 3

a Potential vehicle applications, as per Table 2: , 1 – mechanical exhaust heat recovery, 2 – cold start buffering, catalytic converter, 3 – thermoelectric exhaust heat recovery.

Table A.1 List of fatty acid blends.

Composition TM (°C) Hf (kJ/kg) Ref. Composition TM (°C) Hf (kJ/kg) Ref. 65% Capric acid + 35% Lauric acid 13–18 116.76–148 [120,133] 75.5% Lauric acid + 24.5% Stearic acid 37.3 171 [133] 73% Capric acid + 27% Lauric acid 18.2 120 [173] 91% Lauric Acid + Acetamide 39.4 183 [20] 61.5% Capric acid + 38.5% Lauric acid 19.1 132 [133] 51% Myristic acid + 49% Palmitic acid 39.8 174 [174] 45% Capric acid + 55% Lauric acid 21 143 [120] 58% Mystiric acid + 42% Palmitic acid 42.6 169.7 [132] 73.5% Capric acid + 26.5% Myristic acid 21.4 152 [133] 65.7% Myristic acid + 34.3% Palmitic acid 44 181 [174] 75.2% Capric acid + 24.8% Palmitic acid 22.1 153 [133] 89% Myristic Acid + Acetamide 48.7 199 [20] 34% Myristic acid + 66% Capric Acid 24 147.7 [120] 64.9%Palmitic acid + 35.1% Stearic acid 50.4 179 [133] 86.6% Capric acid + 13.4% Stearic acid 26.8 160 [133] 72.5% Palmitic acid + 27.5% Stearic acid 51.1 159 [173] 62.6% Lauric acid + 37.4% Myristic acid 32.6 156 [174] 64.2% Palmitic acid + 35.8% Stearic acid 52.3 181.7 [132] 80% Lauric acid + 20% Palmitic acid 32.7 147 [173] 89% Palmitic acid + Acetamide 57.2 172 [20] 64% Lauric acid + 36% Palmitic acid 32.8 165 [133] 81% Palmitic acid + Acetamide 59.1 177 [20] 77% Lauric acid + 23% Palmitic acid 33 150.6 [132] 83% Stearic acid + 11% Palmitic acid + other 60–66 206 [125] 85% Lauric acid + 15% Stearic acid 34 152 [173] 95% Stearic acid + 5% Palmitic acid 65–68 209 [125] 66% Lauric acid + 34% Myristic acid 34.2 166.8 [125] 83% Stearic acid + Acetamide 65.4 213 [20] 69% Lauric acid + 31% Palmitic acid 35.2 166.3 [122] ––––

protection requirement that could beneﬁt from phase change TES. applications according to likely performance based on material These applications cover the entire range of temperatures present properties. on the vehicle, from near freezing up to 800 °C. A consequence of As is true for most all material investigations, candidate selec- using a PCM thermal buffer is that a unique material must be used tion is complicated by the fact that the data is dispersed over a for each system that has a different temperature requirement. Sur- large number of publications with disparate nomenclature, mate- veying over 700 materials from over a dozen material classes, we rial focus, and level of detail. Signiﬁcant variation between sources attempted to match the most promising PCMs to the vehicle in measured values of melting temperature, latent heat, and other N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1549

Table A.2 Full list of clathrate hydrates from [19].

PCM name CAS # Formula TM (°C) Hf (kJ/kg)

Tetrabutylammonium benzoate 32-hydrate – Bu4NC6H5CO2Á32H2O 3.5 –

Tetrahydrofuran clathrate hydrate 18879-05-5 C4H8OÁ17.2H2O 4.4 255

Tetrabutylammonium nitrate 32-hydrate – Bu4NNO3Á32H2O 5.8 –

Trimethylamine semi clathrate hydrate 15875-97-5 (CH3)3NÁ10.25H2O 5.9 239

Sulfur dioxide clathrate hydrate – SO2Á6.0H2O 7 247

Ethylene oxide clathrate hydrate – C2H4OÁ6.9H2O 11.1 –

Sulfur dioxide clathrate hydrate – SO2Á6.1H2O 12.1 –

Tetrabutylammonium bromide 32-hydrate – Bu4NBrÁ32H2O 12.5 –

Tetrabutylammonium formate 32-hydrate – Bu4NCHO2Á32H2O 12.5 184

Tetraisoamylammonium formate 40-hydrate – i-Am4NCHO2Á40H2O 15–20 –

Tetrabutylammonium acetate 32-hydrate – Bu4NCH3CO2Á32H2O 15.1 209

Tetrabutylammonium chloride 32-hydrate 37451-68-6 Bu4NClÁ32H2O 15.7 –

Di-tetrabutylammonium oxalate 64-hydrate – (Bu4N)2C2O4Á64H2O 16.8 –

Di-tetrabutylammonium hydrogen phosphate 64-hydrate – (Bu4N)2HPO4Á64H2O 17.2 –

Tetrabutylammonium bicarbonate 32-hydrate – Bu4NHCO3Á32H2O 17.8 –

Tetrabutylammonium ﬂuoride 32-hydrate 22206-57-1 Bu4NFÁ32H2O 24.9 –

Tetraisoamylammonium chloride 38-hydrate – i-Am4NClÁ38H2O 29.8 –

Tetrabutylammonium hydroxide 32-hydrate 147741-30-8 Bu4NOHÁ32H2O 30.2 –

Tetraisoamylammonium hydroxide 40-hydrate – i-Am4NOHÁ40H2O31–

Tetraisoamylammonium ﬂuoride 40-hydrate – i-Am4NFÁ40H2O 31.2 –

properties can be due to differing measurement techniques or (4) The majority of potential metal PCM candidates were not accuracy, or commonly transcription errors and propagation of er- developed for thermal applications, and lack sufficient prop- rors through the literature, all of which reduces confidence and in- erty characterization for even the most basic comparison. creases the effort required to develop PCM-based systems. This Further material analysis is warranted to develop a more review has attempted to consolidate PCM information, but addi- comprehensive metallic PCM dataset for future use, espe- tional material measurements will be needed for any rigorous cially over the 0–200 °C temperature range. component development. In fact, it is the recommendation of the (5) Solid–solid PCMs have also received only limited attention, authors that the PCM research community put effort into a ‘‘PCM and those that have been studied generally have low-to- materials handbook’’ of sorts, as a central resource of material info moderate specific and volumetric latent heat. The lack of a that can be updated and corrected through a managed editorial liquid phase, however, gives them a number of advantages and editioning process. Apart from these technicalities, several over traditional PCMs, such as elimination of void space for observations can be made regarding the state of material data volume change and freedom to use additives or multi-con- within the PCM literature: stituent blends without concern for liquid phase segregation. While there has been more study into heterogeneous (1) Focusing on the primary PCM function of energy storage, ‘form stable’ blends of PCMs with non-melting materials, many reviews limit material data to melting temperature further research into homogenous solid-state PCMs could and specific latent heat, which is insufficient for modeling produce materials with useful system level performance. or providing a proper material comparison. Both volumetric (6) By its nature, the primary technical challenge to PCM-based and specific latent heat values are needed for a minimum thermal buffering of any application involves accommodat- material comparison, and thermal conductivity and specific ing the significant transient thermal and mechanical heat are additionally required for transient thermal changes that are designed to occur within the component. modeling. Even assuming full understanding of the external transient (2) Because of the focus on specific latent heat, until recently condition, all of the following must still be provided for or metals have received very limited attention in the PCM liter- the system will fail: solid and liquid containment, accommo- ature. This is despite their generally high volumetric latent dation of large volume changes, maintenance of adequate heat which, in combination with their high thermal conduc- heat spreading and thermal contact, and continued presence tivity, could result in much higher system energy density of sufficient thermal capacity. Ensuring these criteria is a with proper design. Specifically at low temperatures, metals challenging enough engineering task with sufficient material appear almost inconsequential when specific latent heat is knowledge, but will be nearly impossible with the current the only metric, but become dominant materials in the range state of partial material information available to the PCM when energy density is considered. design engineer. (3) Similarly, paraffin waxes have received the dominant research focus in the field, despite the fact that they have Successful implementation of any thermal buffering strategy low volumetric energy density and very low thermal con- will require better understanding of the application’s transient ductivities. Both of these factors would significantly increase conditions. Designing PCM-based systems requires accurate the size and weight of the container and heat spreading knowledge not only of average and peak temperatures, but also structures needed to make efficient use of the wax, reducing transient duration and frequency. Desired total energy storage the net (packaged) system energy density. Primarily for and absorption rate will dictate the size of any PCM energy these reasons, the candidate PCMs listed in Section 4 store as well as required heat exchanger performance. While includes few paraffins, or other low temperature organics. several of the application studies examined here used standard While those materials may be convenient for testing and commercial vehicle drive cycles for determining some or all of prototyping, they generally produce mediocre results due these parameters, most were limited to isolated case studies to a lack of any outstanding material properties. and proof experiments. In addition, most made use of available 1550 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table A.3 Full list of other organic PCM materials.

3 PCM name CAS # Formula TM (°C) Hf (kJ/kg) Hf,v (MJ/ q (kg/m ) kth (W/mK) cp (kJ/ Ref. m3) kg K)

Ethylene glycol 107-21-1 HOCH2CH2OH À13 146 162 (l) 1113 (l) – – [112]

Diethylene glycol 111-46-6 (HOCH2CH2)2O À10 247 296 (l) 1200 (l) – – [125] (À7)

Triethylene glycol 112-27-6 C6H14O4 À7 247 296 (l) 1200 (l) – – [125]

Tetrahydrofuran (THF) 109-99-9 (CH2)4O 5 280 272 970 (s) – – [126]

Polyglycol 400 25322-68- H(OCH2CH2)nOH 8 99.6 122 1228 (s) 0.185–0.187 – [120] 3 (l) 1125 (l)

Tetrabutylammonium bromide 1643-19-2 C16H36BrN 10–12 193–199 – – – – [125]

Dimethyl sulfoxide 67-68-5 (CH3)2SO 16.5 85.7 86 1009 (s) – – [120]

Polyglycol 600 25322-68- H(OCH2CH2)nOH 20–25 127.2– 157–180 1232 (s) 0.16–0.189 (l) 2.26 (s) [112,120] 3 146 1100–1126 (l)

Lithium chloride ethanolate – LiClÁ4C2H6O 21 186 – – – – [14]

Dimethyl sebacate 110-40-7 C14H26O4 21 120–135 – – – – [120]

Octadecyl 3-mercaptopropionate 31778-15- C21H42O2S 21 143 – – – – [125] 1

Octadecyl thioglycolate 10220-46- C20H40O2S2690– – – – [125] 9

13-Methyl pentacosane 22331-48- C26H54 29 197 – – – – [112] 2

Polyglycol 900 25322-68- H(OCH2CH2)nOH 34 150.5 181 1200 (s) 0.188 (s) 2.26 (s) [122] 3 1100 (l) 0.188 (l) 2.26 (l)

2-Dimethyl-n-docosane – C24H50 35 198 – – – – [14]

Camphenilone 13211-15- C9H14O 39 205 – – – – [112] 9

1-Bromodocosane 6938-66-5 C22H45Br 40 201 – – – – [112]

Caprylone 818-23-5 (CH3(CH2)6)2CO 40 259 – – – – [112]

1-Cyclohexyloctadecane 4445-06-1 C24H48 41 218 – – – – [112]

4-Heptadecanone 53685-77- C17H34O 41 197 – – – – [112] 1

7-Heptadecanone 6064-42-2 C17H34O 41 198 – – – – [14]

Phenol 108-95-2 C6H5OH 41 120 – – – – [121]

8-Heptadecanone 14476-38- C17H34O 42 201 – – – – [112] 1

p-Toluidine 106-49-0 C7H9N 43.3 167 – – – – [121] Cyanamide 420-04-2 HNCNH 44 209 226 1080 (s) – – [112]

2-Heptadecanone 2922-51-2 C17H34O 48 218 – – – – [112]

3-Heptadecanone 84534-29- C17H34O 48 218 – – – – [112] 2

Hydrocinnamic acid (3-Phenylpropionic acid) 501-52-0 C9H10O2 48 118 – – – – [121]

O-Nitroaniline 88-74-4 C6H4(NH2)(NO2)50 93 – – – – [121]

Camphene 79-92-5 C10H16 50 238 201 (l) 842 (l) – – [112]

9-Heptadecanone 540-08-9 C17H34O 51 213 – – – – [112]

Thymol 89-83-8 C10H14O 51.5 115 – – – – [121]

Diphenylamine 122-39-4 (C6H5)2NH 52.9 107 – – – – [121]

p-Dichlorobenzene 106-46-7 C6H4Cl2 53.1 121 – – – – [121]

o-Xylene dichloride 612-12-4 C8H8Cl2 55 121 – – – – [121]

Hypophosphoric acid 7803-60-3 H4P2O6 55 213 – – – – [112]

Nitro naphthalene 86-57-7 C10H7NO2 56.7 103 – – – – [121]

p-Bromophenol 106-41-2 C6H5BrO 63.5 86 – – – – [121]

Polyglycol 6000 25322-68- H(OCH2CH2)nOH 66 190 230 1212 (s) – – [120] 3 1085 (l)

Azobenzene 103-33-3 C12H10N2 67.1 121 – – – – [121]

p-Chloroaniline 106-47-8 ClC6H4NH2 69 156 189 1213 (s) – – [112]

2,4-Dinitrotoluene 121-14-2 C6H3(CH3)(NO2)2 70 111 – – – – [121]

Biphenyl 92-52-4 (C6H5)2 71 119.2 139 1166 (s) – – [120] 991 (l)

Thiosinamine 109-57-9 C4H8N2S 77 140 – – – – [121]

Bromocamphor 76-29-9 C10H15OBr 77 174 252 (l) 1449 (l) – – [112]

Benzylamine 100-46-9 C6H5CH2NH2 78 174 – – – – [121]

Propionamide 79-05-0 C3H7NO 79 168.2 – – – – [120]

Durene (1,2,4,5-tetramethylbenzene) 95-93-2 C10H14 79.3 156 131 838 (s) – – [112]

Napthalene 91-20-3 C10H8 80 147.7 169 1145 (s) 0.310–0.341 – [120] (s) 976 (l) 0.132 (l)

Acetamide 60-35-5 CH3CONH2 81 241 280 1159 (s) – – [112] 999 (l)

Methyl 4-bromobenzoate 619-42-1 BrC6H4CO2CH3 81 126 – – – – [112]

Diethyl tartrate 87-91-2 (COOCH3)2CHOH 87 147 191 1300 (s) – – [112] N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1551

Table A.3 (continued)

3 PCM name CAS # Formula TM (°C) Hf (kJ/kg) Hf,v (MJ/ q (kg/m ) kth (W/mK) cp (kJ/ Ref. m3) kg K)

Ethyl lithium 811-49-4 LiC2H5 95 389 – – – – [112]

1-Naphthol 90-15-3 C10H7OH 96 163 178 1095 (s) – – [112]

p-Xylene dichloride 623-25-6 C8H8Cl2 100 138.7 – – – – [112] High density polyethylene 9002-88-4 – 100– 200–233 – – – – [12,120] 150

Catechol 120-80-9 C6H4(OH)2 104.3 207 283 1370 (s) – – [112]

Quinone 106-51-4 C6H4O2 115 171 226 1318 (s) – – [112]

Acetanilide 103-84-4 C8H9NO 115– 152–222 184–269 1210 (s) – – [112,121] 119

Mandelic acid 90-64-2 C6H5CH(OH)CO2H 118– 161 209 1300 (s) – – [12] 121

Succinic anhydride 108-30-5 (CH2CO)2O 119 204 225 1104 (s) – – [112]

Picric acid 88-89-1 C6H3N3O7 121– 75 – – – – [12] 122

E-Stilbene 103-30-0 (C6H5CH)2 124 167 – 970–1164 (s) – – [12,112]

Benzamide 55-21-0 C6H5CONH2 127.2 169.4 227 1341 (s) – – [112]

Phthalic anhydride 85-44-9 C8H4O3 131 159 243 1530 (s) – – [12]

Tris(hydroxymethyl)-aminomethane (TAM, 77-86-1 H2NC(CH2OH)3 132 285 385 1350 (s) – – [12] Trometanol)

Urea 57-13-6 CO(NH2)2 133– 170–258 228–346 1340 (s) – – [12] 135

Phenacetin 62-44-2 C10H13NO2 134– 137 – – – – [12,112] 137

Dimethyl terephthalate 120-61-6 C6H4(CO2CH3)2 142 170 219 1290 (s) – – [12]

Trans-1,4-polybutadiene (TPB) 25038-44- C4H6 145 144 – – – – [120] 2

p-Acetotoluidide 103-89-9 C9H11NO 146– 180 – – – – [12,112] 151

Anthranilic acid (2-Aminobenzoic acid) 118-92-3 C6H4(NH2)COOH 147 148 209 1410 (s) – – [12]

Benzaldehyde phenylhydrazone 588-64-7 C6H5CH2N2HC6H5 155 134.8 – – – – [112]

Benzanilide 93-98-1 C6H5CONHC6H5 161 162 – – – – [112]

Hydroquinone 123-31-9 C6H4(OH)2 172.4 258 351 1358 (s) – – [112]

p-Aminobenzoic acid 150-13-0 H2NC6H4COOH 187 153 – 1113 (l) – – [112]

Table A.4 List of organic PCM blends.

3 3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) q (kg/m ) kth (W/mK) cp (kJ/kg K) Ref. H2O + polyacrylamid 0 292 305.7 (l) 1047 (l) 0.486 (l) – [120] Oleﬁn blend (n = 15/16) 3–4 36 29 806 (s) – – [20] 90% Capric–lauric acid + 10% Pentadecane 13.3 142.2 – – – – [120] 38.5% Trimethylolethane + 31.5% Water + 30% Urea 14.4 160 187 1170 (s) 0.66 (s) 4.22 (s) [123] 1140 (l) 0.37 (l) 3.09 (l) 48% Butyl palmitate + 48% Butyl stearate + 3% Other 17 140 – – – – [125] 65–90% Methyl palmitate + 35–10% Methyl stearate 22–25.5 120 – – – – [125]

34%C14H28O2 + 66%C10H20O 24 147.7 – – – – [121]

50%CH3CONH2 + 50%NH2CONH2 27 163 – – – – [121]

50%CH3CONH2 + 50%NH2CONH2 27 163 – – – – [125] 62.5% Trimethylolethane + 37% Water 29.8 218 244 1120 (s) 0.65 (s) 2.75 (s) [123] 1090 (l) 0.21 (l) 3.58 (l)

40%CH3COONaÁ3H2O + 60%NH2CONH2 30 200.5 – – – – [121] Acetamide + 91% Stearic acid 39.4 183 – – – – [20]

50%Na(CH3COO)Á3H2O + 50%HCONH2 40.5 255 – – – – [120]

53%NH2CONH2 + 47%NH4NO3 46 95 – – – – [121] Acetamide + 89% Myristic acid 48.7 199 – – – – [20] Acetamide + 89% Palmitic acid 57.2 172 – – – – [20] Acetamide + 81% Palmitic acid 59.1 177 – – – – [20]

50%CH3CONH2 + 50%C17H35COOH 65 218 – – – – [121] Acetamide + 83% Stearic acid 65.4 213 – – – – [20] 67.1% Naphthalene + 32.9% Benzoic acid 67 123.4 – – 0.257–0.282 (s) – [120] 0.130–0.136 (l)

66.6%NH2CONH2 + 33.4%NH4Br 76 151 – – – – [121]

or convenient PCMs, typically paraffins for low temperatures, require more comprehensive vehicle and component level mod- rather than materials that might provide the highest perfor- eling, all of which would provide insight into potential benefits mance for the application. Understanding the large number of and trade-offs involved in incorporating a specific PCM-based internal and external application and material parameters will system. 1552 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table A.5 Full list of aqueous and hydrated salt PCM candidates.

3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/ q (kg/m ) kth (W/mK) cp (kJ/kg K) Ref. m3)

Al(NO3)3 (30.5 wt%) + H2O À30.6 131 163.9 1251 (s) – – [125] 1283 (l)

NaCl (22.4 wt%) + H2O À21.2 222 246.0 1108 (s) – – [125] 1165 (l)

KCl (19.5 wt%) + H2O À10.7 283 312.7 1105 (s) – – [125] 1126 (l)

31%Na2SO4 + 13%NaCl + 16%KCl + 40%H2O 4 234 – – – – [126]

Lithium chlorate trihydrate (LiClO3Á3H2O) 8.1 253 435.2 1720 (s) – – [120]

Dipotassium phosphate hexahydrate (K2HPO4Á6H2O) 13–14 109 – – – – [120,121]

Potassium ﬂuoride tetrahydrate (KFÁ4H2O) 18.5 231 336 1455 (s) – 1.84 (s) [113] 1447 (l) 2.39 (l)

Iron bromide hexahydrate (FeBr3Á6H2O) 21–27 105 – – – – [121]

Manganese nitrate hexahydrate (Mn(NO3)2Á6H2O) 25.5– 125.9–148 226–266 1795 (s) – – [120,121] 25.8 1728–1738 (l)

Calcium chloride hexahydrate (CaCl2Á6H2O) 27.45– 161.15– 272–346 1682.4–1802 1.088 (s) 1.4 (s) [20,21,120,122] 30 192 (s) 1496–1620 (l) 0.53–0.56 (l) 2.2 (l)

Lithium nitrate trihydrate (LiNO3Á3H2O) 29.6– 296 460 1550 (s) 0.8 (s) 1.8 (s) [112,131] 29.9 1430 (l) 0.56 (l) 2.8 (l) Sodium sulfate decahydrate, Glauber’s salt 31–32.4 251.1–254 372 1485 (s) 0.544 (s) 1.93 (s) [112,113,120]

(Na2SO4Á10H2O) 1458–1460 (l)

Sodium carbonate decahydrate (Na2CO3Á10H2O) 32–36 246.5–251 355–362 1440–1442 ––[112,21,120] (s)

Potassium iron sulfate dodecahydrate (KFe(SO4)2Á12H2O) 33 173 – – – – [121]

Lithium bromide dihydrate (LiBr2Á2H2O) 34 124 – – – – [121]

Calcium bromide hexahydrate (CaBr2Á6H2O) 34 115.5–138 253–303 2194 (s) – – [120,121] 1956 (l) Disodium hydrogen phosphate dodecahydrate 35–40 265–281 422–427 1507.1–1522 0.514 (s) 1.69 (s) [112,20,113,120,121]

(Na2HPO4Á12H2O) (s) 1442 (l) 0.476 (l) 1.94 (l)

Zinc nitrate hexahydrate (Zn(NO3)2Á6H2O) 36–36.4 134–147 269 1937–2065 0.464–0.469 1.34 (s) [20,113,120] (s) (l) 1828 (l) 2.26 (l)

Iron chloride hexahydrate (FeCl3Á6H2O) 37 226 – – – – [112]

Manganese nitrate tetrahydrate (Mn(NO3)2Á4H2O) 37.1 115 – – – – [121]

Cobalt sulfate heptahydrate (CoSO4Á7H2O) 40.7 170 – – – – [112]

Potassium ﬂuoride dihydrate (KFÁ2H2O) 41.4–42 162 – – – – [120,121]

Magnesium iodide octahydrate (MgI2Á8H2O) 42 133 – – – – [121]

Calcium iodide hexahydrate (CaI2Á6H2O) 42 162 – – – – [121]

Calcium nitrate tetrahydrate (Ca(NO3)2Á4H2O) 42.7–47 142–153 259 – – – [20,120,121]

Zinc nitrate tetrahydrate (Zn(NO3)2Á4H2O) 45 110 – – – – [121]

Dipotassium phosphate heptahydrate (K2HPO4Á7H2O) 45 145 – – – – [121]

Sodium thiosulfate pentahydrate (Na2S2O3Á5H2O) 45–51.3 200–217.2 344–376 1720–1730 375.756 (s) 1.46 (s) [112,113,21,20] (s) 1660–1690 (l) 2.39 (l)

Magnesium nitrate tetrahydrate (Mg(NO3)2Á4H2O) 47 142 – – – – [121]

Iron nitrate nonahydrate (Fe(NO3)3Á9H2O) 47–47.2 155 261 1684 (s) – – [112,121]

Dipotassium phosphate trihydrate (K2HPO4Á3H2O) 48 99 – – – – [121]

Sodium silicate tetrahydrate (Na2SiO3Á4H2O) 48 168 – – – – [121]

Magnesium sulfate heptahydrate (MgSO4Á7H2O) 48.4 202 – – – – [112]

Calcium nitrate trihydrate (Ca(NO3)2Á3H2O) 51 104 – – – – [121]

Zinc nitrate dihydrate (Zn(NO3)2Á2H2O) 55 68 – – – – [121]

Iron chloride dihydrate (FeCl3Á2H2O) 56 90 – – – – [121]

Nickel nitrate hexahydrate (Ni(NO3)2Á6H2O) 57 169 – – – – [121]

Manganese chloride tetrahydrate (MnCl2Á4H2O) 58 151 – – – – [121]

Magnesium chloride tetrahydrate (MgCl2Á4H2O) 58 178 – – – – [112]

Sodium acetate trihydrate (Na(CH3COO)Á3H2O) 58–58.4 226–264 – – – – [120]

Lithium acetate dihydrate (Li(CH3COO)Á2H2O) 58–70 150–377 – – – – [112,121]

Iron nitrate hexahydrate (Fe(NO3)2Á6H2O) 60.5 126 – – – – [121] Sodium aluminum sulfate decahydrate 61 181 – – – – [121]

(NaAl(SO4)2Á10H2O)

Sodium hydroxide monohydrate (NaOHÁH2O) 64.3 227.6–272 385–468 1690–1720 ––[112,120] (s)

Sodium phosphate dodecahydrate (Na3PO4Á12H2O) 65–69 190 – – – – [121,125]

Aluminum nitrate nonahydrate (Al(NO3)2Á9H2O) 70 155 241 1555 (s) – – [20]

Sodium pyrophosphate decahydrate (Na2P2O7Á10H2O) 70 184 – – – – [120]

Barium hydroxide octahydrate (Ba(OH)2Á8H2O) 78 265.7–301 657 2070–2180 1.255 (s) 1.17 (s) [112,120] (s) 1937 (l) 0.653–0.678 (l) N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1553

Table A.5 (continued)

3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/ q (kg/m ) kth (W/mK) cp (kJ/kg K) Ref. m3)

Magnesium nitrate hexahydrate (Mg(NO3)2Á6H2O) 89–95 149.5– 245–267 1636–1640 0.611–0.669 0.9 (s) [20,120,122] 162.8 (s) (s) 1550 (l) 0.490–0.502 (l) Aluminum potassium sulfate dodecahydrate 91 184 – – – – [112]

(AlK(SO4)2Á12H2O) Ammonium aluminum sulfate hexahydrate 95 269 – – – – [120]

((NH4)Al(SO4)Á6H2O)

Magnesium chloride hexahydrate (MgCl2Á6H2O) 115– 165–172 265–269 1560–1570 .694–.704 (s) 1.72–2.25 [112,20,113,120,122] 117 (s) (s) 1442–1450 (l) 0.57–0.598 2.61–2.82 (l) (l)

Table A.6 Salt hydrate PCM blends.

3 3 PCM mixture TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) Ref.

6.16%NH4Cl + 6.66%NaCl + 32.5%Na2SO4 + 41.4%H2O 13 146 – – – [21]

50%CaBr2Á6H2O + 50%CaCl2Á6H2O 14 140 – – – [163]

55%CaBr2Á6H2O + 45%CaCl2Á6H2O 14.7 140 – – – [121]

45–52%LiNO3Á3H2O + 48–55%Zn(NO3)2Á6H2O 17.2 220 – – – [125]

55–65% LiNO3Á3H2O + 35–45% Ni(NO3)2Á6H2O 24.2 230 – – – [125]

45%Ca(NO3)2Á4H2O + 55%Zn(NO3)2Á6H2O 25 130 251 1931 (s) – [20]

66.6%CaCl2Á6H2O + 33.3%MgCl2Á6H2O 25 127 202 1590 (s) – [120]

50%CaCl2Á6H2O + 50%MgCl2Á6H2O2595–––[121]

48%CaCl2Á6H2O + 4.3%NaCl + 0.4%KCl + 47.3%H2O 26.8–27 188 – 1640 (s) – [125] 1530 (l)

67%Ca(NO3)2Á4H2O + 33%Mg(NO3)2Á6H2O 30 136 228 1676 (s) – [20]

18%Mg(NO3)2Á6H2O + 82%Zn(NO3)2Á6H2O 32 130 249 1915 (s) – [20]

28%Al(NO3)2Á9H2O + 72%Ca(NO3)2Á4H2O 35 139 240 1727 (s) – [20]

61.5%Mg(NO3)2Á6H2O + 38.5%NH4NO3 52 125.5 213 1696 (s) 0.552 (s) [120] 1515 (l) 0.494–0.515 (l)

58.7%Mg(NO3)2Á6H2O + 41.3%MgCl2Á6H2O 58–59 132–132.2 215 1630 (s) 0.678 (s) [120,121] 1550 (l) 0.510–0.565 (l)

50%Mg(NO3)2Á6H2O + 50%MgCl2Á6H2O 58–59.1 132.2–144 215–235 1630 (s) – [121,163]

80%Mg(NO3)2Á6H2O + 20%MgCl2Á6H2O 60 150 – – – [125]

53%Mg(NO3)2Á6H2O + 47%Al(NO3)2Á9H2O 61 148 249 1682 (s) – [20]

59%Mg(NO3)2Á6H2O + 41%MgBr2Á6H2O 66 168 – – – [121]

86%Mg(NO3)2Á6H2O + 14%LiNO3Á3H2O 72 180 – 1610 (s) – [120] 1590 (l)

Table A.7 Fused salt PCM candidates up to 1000 °C.

3 3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref.

Aluminum chloride (AlCl3) 192–192.4 272–280 664–683 2440 – – [21]

Lithium nitrate (LiNO3) 250–254 360–373 857–888 2380 – – [21,123,124]

Sodium nitrite (NaNO2) 270–282 180–216 – – – – [124] Zinc chloride (ZnCl2) 280 75 218 2907 0.5 (s) 0.74 (l) [122]

Sodium nitrate (NaNO3) 306–310 172–199 388–450 2257–2261 0.5 (s) 1.1 (s) [120,122,124,123,153] 1.82 (l)

Rubidium nitrate (RbNO3) 312 31 – – – – [123] Sodium hydroxide (NaOH) 318 159–165 334–347 2100 0.92 (l) 2.08 (l) [122,124]

Potassium nitrate (KNO3) 330–336 88–266 186–561 2109–2110 0.5 (s) 0.935 (s) [120,153] 0.5 (l) 1.22 (l) Potassium hydroxide (KOH) 360–380 134–150 273.4–307 2040–2044 0.5 (s) 1.34 (s) [120,123,122,124,153] 0.5 (l) 1.47 (l)

Sodium peroxide (Na2O2) 360 314 – – – – [21]

Cesium nitrate (CsNO3) 409 71 – – – – [123] Lithium hydroxide (LiOH) 462 875 1269 1450 – – [124,152]

Lithium chromate (Li2CrO4) 485 168 – – – – [123]

Potassium perchlorate (KClO4) 527 1253 3158 2520 – – [21,152]

Strontium iodide (SrI2) 538 57 – – – – [123] Lithium bromide (LiBr) 550 203 – – – – [123] Rubidium iodide (RbI) 556 104 – – – – [123]

Calcium nitrate (Ca(NO3)2) 560 145 – – – – [123]

(continued on next page) 1554 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table A.7 (continued)

3 3 PCM name TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) qs (kg/m ) kth (W/mK) cp (kJ/kg K) Ref.

Barium nitrate (Ba(NO3)2) 594 209 – – – – [123] Lithium chloride (LiCl) 610 416 – – – – [123] Cesium iodide (CsI) 632 96 – – – – [123]

Magnesium iodide (MgI2) 633 93 – – – – [123] Cesium bromide (CsBr) 638 111 – – – – [123] Cesium chloride (CsCl) 645 121 – – – – [123]

Strontium nitrate (Sr(NO3)2) 645 231 – – – – [123]

Strontium bromide (SrBr2) 657 41 – – – – [123] Sodium iodide (NaI) 661 158 – – – – [123] Potassium iodide (KI) 681 145 – – – – [123]

Sodium molybdate (Na2MoO4) 688 109 – – – – [123] Rubidium bromide (RbBr) 692 141 – – – – [123] Lithium hydride (LiH) 688–699 2678–3260 2800 859 – – [21,175]

Sodium tungstate (Na2WO4) 696 107 – – – – [123]

Cesium ﬂuoride (CaF2) 703 143 – – – – [123]

Lithium molybdate (Li2MoO4) 703 281 – – – – [123]

Barium iodide (BaI2) 711 68 – – – – [123]

Magnesium bromide (MgBr2) 711 214 – – – – [123]

Magnesium chloride (MgCl2) 714 452–454 211–212 2140 – – [120,123,124] Rubidium chloride (RbCl) 723 197 – – – – [123]

Lithium carbonate (Li2CO3) 732 509 – – – – [123] Potassium bromide (KBr) 734 215 – – – – [123]

Lithium tungstate (Li2WO4) 740 157 – – – – [123]

Calcium bromide (CaBr2) 742 145 – – – – [123] Sodium bromide (NaBr) 742 255 – – – – [123] Potassium chloride (KCl) 770–771 353–355 – – – – [123,124]

Calcium chloride (CaCl2) 772 253 – – – – [123]

Calcium iodide (CaI2) 783 142 – – – – [123]

Sodium chromate (Na2CrO4) 794 146 – – – – [123] Rubidium ﬂuoride (RbF) 795 248 – – – – [123] Sodium chloride (NaCl) 800–802 466.7–492 1008–1063 2160 5 (s) – [120,123,124] Lithium ﬂuoride (LiF) 848–868 932–1041 – – – – [21,124]

Sodium carbonate (Na2CO3) 854–858 165–275.7 2163–2173 2533 2 (s) – [120,123]

Barium bromide (BaBr2) 857 108 – – – – [123] Potassium ﬂuoride (KF) 857–858 452–507 1071–1202 2370 – – [120,123,124]

Lithium sulfate (Li2SO4) 858 84 – – – – [123]

Strontium chloride (SrCl2) 875 103 – – – – [123]

Sodium sulfate (Na2SO4) 884 165 – – – – [123]

Potassium carbonate (K2CO3) 897–900 200–235.8 458–540 2290 2 (s) – [120,123,124]

Potassium tungstate (K2WO4) 923 86 – – – – [123]

Potassium molybdate (K2MoO4) 926 163 – – – – [123]

Cesium molybdate (Cs2MoO4) 935 75 – – – – [123]

Rubidium tungstate (Rb2WO4) 952 78 – – – – [123]

Cesium tungstate (Cs2WO4) 953 63 – – – – [123]

Rubidium molybdate (Rb2MoO4) 955 140 – – – – [123]

Barium chloride (BaCl2) 961 76 – – – – [123]

Potassium chromate (K2CrO4) 973 41 – – – – [123]

Cesium chromate (Cs2CrO4) 975 94 – – – – [123]

Magnesium carbonate (MgCO3) 990 698 – – – – [123] Sodium ﬂuoride (NaF) 993–996 750–801 – – – – [21,123,124]

Table A.8 Full list of fused salt blends up to 1000 °C.

PCM name TM Hf Hf,v qs kth cp Ref. (°C) (kJ/kg) (MJ/m3) (kg/m3) (W/mK) (kJ/kg K)

25%LiNO3 + 65%NH4NO3 + 10%NaNO3 80.5 113 – – – – [121]

14.9%KNO3 + 26.4%LiNO3 + 58.7%NH4NO3 81.5 116 – – – – [121]

27%LiNO3 + 5%NH4Cl + 68%NH4NO3 81.6 108 – – – – [121]

49.4%KNO3 + 29%LiNO3 + 17%NaNO3 + 4.6%Sr(NO3)2 105 110 – – – – [123]

67%KNO3 + 33%LiNO3 133 170 – – – – [124]

68.3%KNO3 + 31.7%LiNO3 135 136 – – – – [123]

53%KNO3 + 40%NaNO2 + 7%NaNO3 142 80 – – – – [124]

40.1%KCl + 55.4%LiNO3 + 4.5%NaNO3 160 266 – – – – [123]

41.9%KCl + 58.1%LiNO3 166 272 – – – – [123] 50%KOH + 50%NaOH 169–171 202–213 – – – – [123,124]

1.4%LiCl + 47.9%LiNO3 + 50.7%NaNO3 180 265 – – – – [123]

57%LiNO3 + 43%NaNO3 193 248 – – – – [123]

49%LiNO3 + 51%NaNO3 194 265 – – – – [124]

45%LiNO3 + 47%NaNO3 + 8%Sr(NO3)2 200 199 – – – – [123]

87%LiNO3 + 13%NaCl 208 369 – – – – [123] 30%LiOH + 70%NaOH 210–216 278–329 – – – – [123,124]

50%NaNO3 + 50%KNO3 220–222 100–100.7 192–193.3 1920 0.56 (s) 2.25 (s), 1.35 (l) [122,124]

46%KNO3 + 54%NaNO3 222 117 – – – – [123] N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1555

Table A.8 (continued)

PCM name TM Hf Hf,v qs kth cp Ref. (°C) (kJ/kg) (MJ/m3) (kg/m3) (W/mK) (kJ/kg K) 80%NaNO2 + 20%NaOH 230–232 206–252 – – – – [123,124]

68.1%KCl + 31.9%ZnCl2 235 198 491 2480 0.8 (s) 2.25 (s) [120,122]

27%NaNO2 + 73%NaOH 237–238 249–295 – – – – [123,124]

3.6%NaCl + 18.3%NaNO3 + 78.1%NaOH 242 242 – – – – [123]

72%NaNO3 + 28%NaOH 246–247 182–257 – – – – [123,124]

70%NaNO3 + 30%NaOH 247 158 – – – – [123]

4.2%NaCl + 40.2%NaNO3 + 55.6%NaOH 247 213 – – – – [123]

97.4%LiNO3 + 2.6%Ba(NO3)2 253 368 – – – – [123] (28.5–28.9)%LiCl + (43.5–44.5)%CsCl + (13.7–14.1)%KCl 253–259 375–380 – – – – [123] + (13.3–13.5)%RbCl

93.6%LiNO3 + 6.4%NaCl 255 354 – – – – [123]

18.5%NaNO3 + 81.5%NaOH 256–258 251–292 – – – – [123,124] 36%LiCl + 63%LiOH 262 485 752 1550 1.1 (s) 2.4 (s) [123]

NaNO2ÁNaOH 265 313 – – – – [124]

41%NaNO3 + 59%NaOH 266 221–278 – – – – [123,124]

40.85%Ca2 + 59.15%LiCl 270 167 – – – – [123]

NaNO3Á2NaOH 270 295 – – – – [123]

NaNO3ÁNaOH 271 265 – – – – [123] 34.5%LiCl + 65.5%LiOH 274 339 – – – – [123]

5.3%NaCl + 6.4%Na2CO3 + 88.3%NaOH 282 279 – – – – [123] 1.5KCl + 36.5%LiCl + 62%LiOH 282 300 – – – – [123]

7.8%NaCl + 6.4%Na2CO3 + 85.8%NaOH 282 316 673 2130 – 2.51 (s) [123]

8.4%NaCl + 86.3%NaNO3 + 5.3%Na2SO4 287 177 – – – – [123]

8%NaCl + 5%NaF + 87%NaNO3 288 224 – – – – [123]

6.1%NaCl + 6.6%Na2CO3 + 87.3%NaOH 291 283 – – – – [123]

16.2%NaCl + 6.6%Na2CO3 + 77.2%NaOH 318 290 – – – – [123]

6.4%BaCl2 + 39.4%KCl + 54.2%LiCl 320 170 – – – – [123]

95.5%KNO3 + 4.5%KCl 320 74 155.4 2100 0.5 (s) 1.21 (s) [153]

41.5%LiCl + 7%LiF + 16.4%LiVO3 + 35.1%Li2CrO4 340 177 – – – – [123]

(46.8–47.0)%KCl + (3.2–3.4)%LiCO3 + (47.4–47.7)%LiCl 340–343 375–380 – – – – [123] + (2.1–2.4)%LiF 42%KCl + 58%LiCl 348 170 – – – – [123]

28.7%KCl + 45%MnCl2 + 26.3%NaCl 350 215 – – – – [123] 41.3%KCl + 58.7%LiCl 352.7 251.5 473 1880 – 0.95 (s), 1.33 (l) [92]

(23.4–24.2)%LiCl + (24.8–25.3)%LiVO3 + (27.1–27.6)%Li2MoO4 + 360–363 278–284 – – – – [123]

(17.3–17.8)%Li2SO4 + (6.1–6.2)%LiF

42%LiCl + 17.4%LiF + 11.6%Li2MO4 + 17.4%LiVO3 + 11.6%Li2SO4 363 277–291 – – – – [123]

20.4%KCl + 60%MgCl2 + 19.6%NaCl 380 400 720 1800 – 0.96 (s) [153]

21.6%KCl + 45.4%MgCl2 + 33%NaCl 385 234 – – – – [124]

14%KCl + 63%MgCl2 + 22.3%NaCl 385 461 1037 2250 0.95 (l) 0.96 (s) [123]

(18.5–22.5)%KCl + (57.0–53.0)%MgCl2 + (22.5–26.5)%NaCl 385–393 405–410 – – – – [123]

45.5%KCl + 34.5%MnCl2 + 20%NaCl 390 230 – – – – [123]

22%KCl + 51%MgCl2 + 27%NaCl 396 290 – – – – [123]

20%KCl + 50%MgCl2 + 30%NaCl 396 291 – – – – [123]

25%K2NO3 + 43.5%Li2CO3 + 31.5%Na2CO3 397 274 – – – – [124]

35%K2CO3 + 32%Li2CO3 + 33%Na2CO3 397 276 635 2300 2.02 (l) 1.67 (s), 1.63 (l) [123]

37.7%KCl + 37.3%MnCl2 + 25%NaCl 400 235 – – – – [123]

51.5%LiCl + 16.2%LiF + 16.2%Li2MoO4 + 16.2%Li2SO4 402 291 – – – – [123]

(40.0–40.4)%KCO4 + (8.6–8.7)%KCl + (33.2–33.8)%KF 422–426 407–412 – – – – [123] + (17.6–17.7)%LiF 20%LiF + 80%LiOH 426 869 1390 1600 – 0.88 (s), 1. (l) [123] 50%LiF + 50%LiOH 427 512 – – – – [21]

25%LiF + 16.5%Li2MoO4 + 14.8%Li2SO4 + 43.8%LiVO3 428 260 – – – – [123]

44%MgCl2 + 56%NaCl 430 320 – – – – [123] 80%LiF + 20%LiOH 430 528 – – – – [123]

55%MgBr2 + 45%NaBr 431 212 740 3490 0.9 (l) 0.5 (s), 0.59 (l) [123]

54%KCl + 46%ZnCl2 432 218 525 2410 0.83 (l) 0.67 (s), 0.88 (l) [123]

2KClÁMgCl2 435 184 – – – – [124]

61%KCl + 39%MgCl2 435 351 741 2110 0.81 (l) 0.8 (s), 0.96 (l) [123] 38.5%MgCl + 61.5%NaCl 435 328 708 2160 – – [120]

1.8%BaF2 + 41.2%KF + 45.7%LiF + 11.3%NaF 438 332 – – – – [123] (2.8–3.0)%KCl + (41.0–43.0) %KF + (42.5–45.5)%LiF + (10.7– 440–448 682–692 – – – – [123] 11.5)%NaF 67%KF + 33%LiF 442–493 458–618 1159–1564 2530 3.98 (l) 1.34 (s), 1.63 (l) [123]

58.5%LiCl + 17.9%Li2MoO4 + 23.6%Li2SO4 445 327 – – – – [123]

36%KCl + 64%MgCl2 448–470 236–388 517–850 2190 0.83 (l) 0.84 (s), 0.96 (l) [123]

49%LiCl + 12.75%Li2SO4 + 38.25%LiVO3 449 450 – – – – [123]

55.1%KF + 27.1%LiF + 5.9%MgF2 + 11.9%NaF 449 699 – – – – [123]

50%MgCl2 + 50%NaCl 450 429 961 2240 0.96 (l) 0.93 (s) [123]

52%MgCl2 + 48%NaCl 450 430 959 2230 0.95 (l) 0.92 (s), 1. (l) [123]

39.9%MgCl2 + 60.1%NaCl 450 293–328 – – – – [123,124] 42%KF + 46.5%LiF + 11.5%NaF 454 400 – – – – [124] 59%KF + 29%LiF + 12%NaF 454 590 1493 2530 4.5 (l) 1.34 (s), 1.55 (l) [123]

47.6%CaCl2 + 8.1%KCl + 41.3%NaCl + 2.9%NaF 460 231 – – – – [123]

(continued on next page) 1556 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Table A.8 (continued)

PCM name TM Hf Hf,v qs kth cp Ref. (°C) (kJ/kg) (MJ/m3) (kg/m3) (W/mK) (kJ/kg K)

41.6%CaCl2 + 2.2%KCl + 8.8%MgCl2 + 47.4%NaCl 460 245 – – – – [123]

50%CaCl2 + 7.25%KCl + 42.75%NaCl 465 245 – – – – [123]

42%KCl + 58%MgCl2 470 392 – – – – [124]

8.7%BaCl2 + 52.3%KCl + 18.2%MgCl2 + 20.7%NaCl 475 248 – – – – [123]

13.1%BaCl2 + 16.9%CaCl2 + 47.3%KCl + 22.7%NaCl 478 208 – – – – [123]

9.3%BaCl2 + 22.2%CaCl2 + 42.7%KCl + 25.8%NaCl 479 217 – – – – [123]

69.5%LiCl + 26.5%LiF + 4%MgF2 484 157 – – – – [123] 26.4%LiCl + 73.6%LiF 485 403 – – – – [123]

27%CaCl2 + 25%KCl + 48%MgCl2 487 342 865 2530 0.88 (l) 0.8 (s), 0.92 (l) [123] 50%KF + 50%LiCl 487 344 – – – – [123]

38%K2NO3 + 62%Li2CO3 488 370 – – – – [124]

53%K2CO3 + 47%Li2CO3 488–491 321–342 706–752 2200 1.99 (l) 1.03 (s), 1.34 (l) [123]

50%CaCl2 + 1.5%CaF2 + 48.5%NaF 490 264 – – – – [123]

(52.3–55)%CaCl2 + (45–47.2)%NaCl 490–500 233–239 – – – – [123]

18%LiF + 29.0%Li2MoO4 + 53.0%LiVO3 493 297 – – – – [123]

44%Li2CO3 + 56%Na2CO3 496 370 858 2320 2.09 (l) 1.8 (s), 2.09 (l) [123]

53.3%Li2CO3 + 46.7%Na2CO3 496 372 – – – – [124]

72%K2CO3 + 28%Li2CO3 498 263 589 2240 1.85 (l) 1.46 (s), 1.8 (l) [123]

71.5%K2CO3 + 28.5%Li2CO3 498 316 – – – – [123]

52.8%CaCl2 + 47.2%NaCl 500 239 – – – – [123]

67%CaCl2 + 33%NaCl 500 281 607 2160 1.02 (l) 0.84 (s), 1. (l) [123]

13%KCl + 19NaCl + 68%SrCl2 504 223 613 2750 1.05 (l) 0.67 (s), 0.84 (l) [123]

66%CaCl2 + 5%KCl + 29%NaCl 504 279 600 2150 1 (l) 1.17 (s), 1. (l) [123]

65%K2CO3 + 35%Li2CO3 505 344 777 2260 1.89 (l) 1.34 (s), 1.76 (l) [123]

43%NaBr + 2%NaF + 55%Na2MoO4 506 241 – – – – [123] 20.1%NaF + 79.9%ZrF4a 510 255 – – – – [123]

21%KF + 62%K2CO3 + 17%NaF 520 274 652 2380 1.5 (l) 1.17 (s), 1.38 (l) [123]

40%NaBr + 5%NaCl + 55%Na2MoO4 524 215 – – – – [123]

40%KCl + 23%KF + 37%K2CO3 528 283 645 2280 1.19 (l) 1. (s), 1.26 (l) [123]

37%MgCl2 + 63%SrCl2 535 239 664 2780 1.05 (l) 0.67 (s), 0.8 (l) [123]

3.3%BaF2 + 8.1%BaMoO4 + 18.5%CaF2 + 36.1%LiF + 34%NaF 536 653 – – – – [123]

3.5%CaMoO4 + 59.8%Li2SO4 + 36.7%Li2MoO4 538 406 – – – – [123]

53%BaCl2 + 28%KCl + 19%NaCl 542 221 667 3020 0.86 (l) 0.63 (s), 0.8 (l) [123]

20%K2CO3 + 20%Li2CO3 + 60%Na2CO3 550 283 674 2380 1.83 (l) 1.59 (s), 1.88 (l) [123]

62%K2CO3 + 22%Li2CO3 + 16%Na2CO3 550–580 288 674 2340 1.95 (l) 1.8 (s), 2.09 (l) [123]

47%BaCl2 + 29%CaCl2 + 24%KCl 551 219 642 2930 0.95 (l) 0.67 (s), 0.84 (l) [123]

94.5%LiCl + 5.5%MgF2 573 131 – – – – [123]

95.2%NaCl + 48%NiCl2 573 558 – – – – [123] 60%KBr + 40%KF 576 315 – – – – [123]

(27.0–27.25)%CaF2 + (25.67–25.76)%LiF + (10.63– 593–595 510–515 – – – – [123]

10.67)%MgF2 + (36.45–36.57)%NaF 45%KCl + 55%KF 605 407 – – – – [123]

23%NaBr + 38.5%NaCl + 38.5%Na2MoO4 612 168 – – – – [123]

26.5%CaF2 + 35.2%LiF + 38.3%NaF 615 636 – – – – [123]

13%CaF2 + 52%LiF + 35%NaF 615 640 – – – – [123]

65%KBr + 35%K2MoO4 625 90.5 – – – – [123]

46%LiF + 10%MgF2 + 44%NaF2 632 858 1922 2240 1.2 (s) 1.4 (s) [123] 73%NaBr + 27%NaF 642 360 – – – – [123]

33.4%LiF + 17.1%MgF2 + 49.9%NaF2 650 860 2425 2820 1.15 (s) 1.42 (s) [123]

(24.5–25.0)%CaF2 + (34.51–34.79)%LiF + (37.25– 651–657 460–470 – – – – [123]

37.6)%MgF2 + (3.21–3.31)%NaF 60%LiF + 40%NaF 652 816 – – – – [123]

38.5%CaCl + 4%CaMoO4 + 11%CaSO4 673 224 – – – – [123] 66.5%NaCl + 33.5%NaF 675 572 – – – – [123]

6.56%CaMoO4 + 11.44%CaSO4 + 82%Li2SO4 680 207 – – – – [123]

62%LiF + 19%MgF2 + 19%NaF 693 690 – – – – [123]

51%K2CO3 + 49%Na2CO3 710 163 391 2400 1.73 (l) 1.67 (s), 1.56 (l) [123]

47.8%K2CO3 + 52.2%Na2CO3 710 176 – – – – [123]

70%LiF + 30%MgF2 728 520 – – – – [123]

23%CaF2 + 12%MgF2 + 65%NaF 745 574 907 1580 – 1.17 (s) [123]

67%LiF + 33%MgF2 746 947 2491 2630 – 1.42 (s) [123]

13%KF + 74%LiF + 13%MgF2 749 860 – – – – [123] 20%CeF3 + 80%LiF 756 500 – – – – [123]

21%CaF2 + 79%LiF 765 757 – – – – [124]

19.5%CaF2 + 80.5%LiF 767–769 816–820 1950–1960 2390 3.8 (s), 1.7 (l) 1.77 (l) [122,123]

15%CaF2 + 85%KF 780 440 – – – – [123]

85%KF + 15%MgF2 790 520 – – – – [123]

16%KF + 20%MgF2 + 64%NaF 804 650 – – – – [123]

15%KF + 22.5%MgF2 + 62.5%NaF 809 543 – – – – [123]

32%CaF2 + 68%NaF 810 600 – – – – [123]

33%MgF2 + 67%NaF 832 616 1318 2140 4.65 (l) 1.42 (s), 1.38 (l) [123]

25%MgF2 + 75%NaF 832 650 1742 2680 4.66 (l) 1.42 (s) [123]

49%CaF2 + 9.6%CaMoO4 + 41.4%CaSO4 943 237 – – – – [123]

64%MgF2 + 36%NaF 1000 794 – – – – [124] N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1557

Table A.9 RoHS compliant metallic PCMs up to 1000 °C.

3 3 PCM formula (Name) TM (°C) Hf (kJ/kg) Hf,v (MJ/m ) (s) qs (kg/m ) kth,s (W/mK) cp,s (kJ/kg K) Ref. 67Ga–20.5In–12.5Zn 10.7 67.2 415 6170 – – [154] 78.55Ga–21.45In 15.7 69.7 432 6197 – – [154] 82Ga–12Sn–6Zn 18.8 86.5 516 5961 – – [154] 86.5Ga–13.5Sn 20.55 81.9 482 5885 – – [154] 96.5Ga–3.5Zn 25 88.5 526 5946 – – [154] Cesium 28.65 16.4 29.5 (l) 1796 (l) 17.4 (l) 0.236 (l) [160,155] Gallium 29.8–30 80.3 469 5903 (s) 33.7 (s) 0.34 (s) [112,160,155,161] 6093 (l) 24 (l) 0.397 (l) Rubidium 38.85 25.74 37.8378 1470 (l) 29.3 (l) 0.363 (l) [160,155] Potassium 63.2 59.59 39.56776 664 (l) 54 (l) 0.78 (l) [160,155] 66.3In 33.7Bi (Indalloy 162) 72 25 200–203 7990–8100 – – [112,161] Sodium 97.83 113.23 104.95289 926.9 (l) 86.9 (l) 1.38 (l) [160,155] 60Sn 40Bi (Indalloy 281–338) 138–170 44.4 361 8120 30 0.18 (s) [161] 0.213 (l) 58Bi 42Sn (Indalloy 281) 138.3 44.8 384 8560 19 0.167 (s) [161] 0.201 (l) Indium 156.7 28.47 208 7310 86 0.243 [161] 91Sn 9Zn (Indalloy 201) 199 71.2 518 7270 61 0.239 (s) [161] 0.272 (l) 46.3Mg 53.7Zn 340 185 851 4600 – – [123,154] 52Zn 48Mg 340 180 – – – – [123] 96Zn–4Al 381 138 915 6630 – – [123,154] 55Mg–28Cu–17Zn 400 146 330 2260 – – [123] 59Al–35Mg–6Zn 443 310 738 2380 – 1.63 (s) [123] 1.46 (l) 60Mg–25Cu–15Zn 452 254 711 2800 – – [123] 52Mg–25Cu–23Ca 453 184 368 2000 – – [123] 34.65Mg–65.35Al 497 285 614 2155 – – [123,154] 60.8Al–33.2Cu–6.0Mg 506 365 1113 3050 – – [123,154] 64.6Al–5.2Si–28Cu–2.2Mg 507 374 1646 4400 – – [123,154] 54Al–22Cu–18Mg–6Zn 520 305 958 3140 – 1.51 (s) [123] 1.13 (l) 68.5Al–5.0Si–26.5Cu 525 364 1069 2938 – – [123,154] 64.3–34.0Cu–1.7Sb 545 331 1324 4000 – – [123,154] 66.92Al–33.08Cu 548 372 1339 3600 – – [123,154] 83.14Al–11.7Si–5.16Mg 555 485 1213 2500 – – [123,154] 87.76Al–12.24Si 557 498 1265 2540 – – [123,154] 46.3Al–4.6Si–49.1Cu 571 406 2257 5560 – – [123,154] 65Al–30Cu–5Si 571 422 1152 2730 – 1.3 (s) [123] 1.2 (l) 86.4Al–9.4Si–4.2Sb 575 471 1272 2700 – – [123,154] 88%Al–12%Si 576 560 1512 2700 160 1.038 (s) [123] 1.741 (l) 80%Al–20%Si 585 460 – – – – [124]

Zn2Mg 588 230 – – – – [123] 49Zn–45Cu–6Mg 703 176 1526 8670 – 0.42 [123] 91Cu–9P 715 134 750 5600 – – [123] 69Cu–17Zn–14P 720 368 2576 7000 – – [123] 74Cu–19Zn–7Si 765 125 896 7170 – – [123] 56Cu–27Si–17Mg 770 420 1743 4150 – 0.75 [123] 84Mg–16Ca 790 272 375 1380 – – [123] 47Mg–38Si–15Zn 800 314 – – – – [123] 80Cu–20Si 803 197 1300 6600 – 0.5 [123] 83Cu–10P–7Si 840 92 633 6880 – – [123]

Mg2Cu 841 243 – – – – [123] 49Si–30Mg–21Ca 865 305 686 2250 – – [123] 56Si–44Mg 946 757 1438 1900 – 0.79 [123] Gold 961 104.6 1098 10,500 – – [124,161]

Disclaimer operational energy strategy; 2011 May. DTIC accession no.: ADA544100. . [2] Kelly TK, Peters JE, Landree E, Moore LR, Steeb R, Martin A. (RAND National The views expressed are those of the authors and do not reflect Defense Research Institute). The U.S. combat and tactical wheeled vehicle the official policy or position of the Department of Defense or the fleets: issues and suggestions for congress. Technical report. Santa Monica U.S. Government. (CA): Office of the Secretary of Defense (US). LCCN: 2011921602. . [3] United States, Department of Energy, Office of Energy Efficiency and Appendix A. Full material tables Renewable Energy. FreedomCAR and fuels partnership. Electrical and electronics technical team roadmap; 2010 December [cited 10.01.12]. Technical report. . [4] Chau KT, Chan CC. Emerging energy-efficient technologies for hybrid electric References vehicles. Proc IEEE 2007;95(4):821–35. http://dx.doi.org/10.1109/ JPROC.2006.89011410.1109/JPROC.2006.890114. [5] Khalil G, Barshaw E, Danielson E, Chait M. Power supply and integration in [1] United States, Department of Defense, Office of Assistant Secretary of Defense future combat vehicles. In: Proceedings of the NATO RTO AVT Symposium on for Operational Energy, Plans, and Programs. Energy for the warfighter: 1558 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

Functional and Mechanical Integration of Weapons and Land and Air vehicles. on Thermoelectrics. Pamplona, Spain; 2002 October. p. 1–4. . 7–9. p. 39-1–12. RTO-MP-AVT-108. DTIC accession no.: ADA468635. . Institute). Diesel engine cold start improvement using thermal management [6] Kwon J, Kim J, Fallas E, Pagerit S, Rousseau A. Impact of drive cycles on PHEV techniques. Final report. Research Triangle Park, NC: Army Research Office component requirements. In: Proc 2008 SAE World Congress. Detroit, MI; (US), Army Research Laboratory; 2000 October. DTIC accession no.: 2008 April 14–17. http://dx.doi.org/10.4271/2008-01-1337. ADA384223. Contract no.: DAA55-9801-0035. . In: FY2007 annual progress report for the advanced power electronics and [30] Shams-Zahraei M, Kouzani AZ, Kutter S, Bäker B. Integrated thermal and electric machinery program. United States Department of Energy, Office of energy management of plug-in hybrid electric vehicles. J Power Sources Energy Efficiency and Renewable Energy; 2007. p. 16-23. . [31] Department of Defense Design Criteria Standard – Human Engineering MIL- [8] Lane GA. Solar heat storage: latent heat materials. Background and scientific STD-1472F. . principles, vol. I. Boca Raton (FL): CRC Press; 1983. [32] Walgama C, Fackrell S, Karimi M, Fartaj A, Rankin GW. Passenger thermal [9] Khudhair AM, Farid MM. A review on energy conservation in building comfort in vehicles – a review. Proc Inst Mech Eng Part D: J Automob Eng applications with thermal storage by latent heat using phase change 2006;220(5):543–62. http://dx.doi.org/10.1243/09544070D00705. materials. Energy Convers Manage 2004;45(2):263–75. http://dx.doi.org/ [33] Lambert MA, Jones BJ. Automotive adsorption air conditioner powered by 10.1016/S0196-8904(03)00131-6. exhaust heat. Part 1: conceptual and embodiment design. Proc Inst Mech Eng [10] Zhou D, Zhao CY, Tian Y. Review on thermal energy storage with phase Part D: J Automob Eng 2006;220(7):959–72. change materials (PCMs) in building applications. Appl Energy [34] Chua KJ, Chou SK, Yang WM, Yan J. Achieving better energy-efficient air 2012;92:593–605. http://dx.doi.org/10.1016/j.apenergy.2011.08.025. conditioning – a review of technologies and strategies. Appl Energy [11] Kaizawa A, Kamano H, Kawai A, Jozuka T, Senda T, Maruoka N, et al. Technical 2013;104:87–104. http://dx.doi.org/10.1016/j.apenergy.2012.10.037. feasibility study of waste heat transportation system using phase change [35] Al-Abidi AA, Mat SB, Sopian K, Sulaiman M, Lim C, Th A. Review of thermal material from industry to city. ISIJ Int 2008;48(4):540–8. http://dx.doi.org/ energy storage for air conditioning systems. Renew Sustain Energy Rev 10.2355/isijinternational.48.540. 2012;16(8):5802–19. http://dx.doi.org/10.1016/j.rser.2012.05.030. [12] Haillot D, Bauer T, Kroner U, Tamme R. Thermal analysis of phase change [36] Cabeza LF, Castellón C, Nogués M, Medrano M, Leppers R, Zubillaga O. Use of materials in the temperature range 120–150 °C. Thermochim Acta microencapsulated PCM in concrete walls for energy savings. Energy Build 2011;513(1–2):49–59. http://dx.doi.org/10.1016/j.tca.2010.11.011. 2007;39(2):113–9. http://dx.doi.org/10.1016/j.enbuild.2006.03.030. [13] Arteconi A, Hewitt N, Polonara F. State of the art of thermal storage for [37] Castell A, Martorell I, Medrano M, Pérez G, Cabeza LF. Experimental study of demand-side management. Appl Energy 2012;93:371–89. http://dx.doi.org/ using PCM in brick constructive solutions for passive cooling. Energy Build 10.1016/j.apenergy.2011.12.045. 2010;42(4):534–40. http://dx.doi.org/10.1016/j.enbuild.2009.10.022. [14] Bentilla EW, Sterrett KF, Karre LE (Northrop Space Laboratories, Northrop [38] Tyagi VV, Buddhi D. PCM thermal storage in buildings: a state of art. Renew Corporation, Hawthorne, CA). Research and development study on thermal Sustain Energy Rev 2007;11(6):1146–66. http://dx.doi.org/10.1016/ control by use of fusible materials. Final report. Huntsville (AL): George C. j.rser.2005.10.002. Marshall Space Flight Center, National Aeronautics and Space Administration; [39] Bilodeau S. Thermal storage to optimize passenger shuttle cooling systems 1966 April. Report no.: N66-26691. Contract no.: NAS 8-11163. . Congress. Rosemont, IL; 2005 November 1–3. http://dx.doi.org/10.4271/ [15] Lu TJ. Thermal management of high power electronics with phase change 2005-01-3638. cooling. Int J Heat Mass Transfer 2000;43(13):2245–56. http://dx.doi.org/ [40] Craig T, O’Brien J, Polisoto D, Wolfe N. Integrated air conditioning evaporator 10.1016/S0017-9310(99)00318-X. with phase change material for thermal storage (presentation). In: Proc SAE [16] Vesligaj MJ, Amon CH. Transient thermal management of temperature Automotive Refrigerant and System Efficiency Symposium. Scottsdale, AZ; fluctuations during time varying workloads on portable electronics. IEEE 2010 July 13–15. . Trans Compon Pack Technol 1999;22(4):541–50. http://dx.doi.org/10.1109/ [41] Shahidinejad S, Bibeau E, Filizadeh S. Design and simulation of a thermal 6144.814970. management systems for plug-in electric vehicles in cold climates. In: Proc [17] Krishnan S, Garimella SV, Kang SS. A novel hybrid heat sink using phase 2012 SAE World Congress and Exposition. Detroit, MI; 2012 April 24–26. change materials for transient thermal management of electronics. IEEE http://dx.doi.org/10.4271/2012-01-0118. Trans Compon Pack Technol 2005;28(2):281–9. http://dx.doi.org/10.1109/ [42] Barnitt, RA, Brooker, AD, Ramroth L, Rugh J, Smith KA. Analysis of off-board TCAPT.2005.848534. powered thermal preconditioning in electric drive vehicles. In: Proc 25th [18] Yoo D, Joshi YK. Energy efficient thermal management of electronic World Battery, Hybrid and Fuel Cell Electric Vehicle Symposium and components using solid–liquid phase change materials. IEEE Trans Device Exposition. Shenzhen, China; 2010 November 5–9. . [19] Dincer I, Rosen M. Thermal energy storage: systems and applications. 2nd [43] Simard A, Lacroix M. Study of the thermal behavior of a latent heat cold ed. West Sussex (United Kingdom): John Wiley & Sons; 2010. storage unit operating under frosting conditions. Energy Convers Manage [20] Lorsch HG, Kauffman KW, Denton JC. Thermal energy storage for solar 2003;44(10):1605–24. http://dx.doi.org/10.1016/S0196-8904(02)00183-8. heating and off-peak air conditioning. Energy Convers 1975;15(1–2):1–8. [44] Oro E, Miro L, Farid MM, Cabeza LF. Thermal analysis of a low temperature http://dx.doi.org/10.1016/0013-7480(75)90002-9. storage unit using phase change materials without refrigeration system. Int J [21] Hasnain SM. Review on sustainable thermal energy storage technologies. Part Refrig 2012;35(6):1709–14. http://dx.doi.org/10.1016/j.ijrefrig.2012.05.004. I: heat storage materials and techniques. Energy Convers Manage [45] Liu M, Saman W, Bruno F. Development of a novel refrigeration system for 1998;39(11):1127–38. http://dx.doi.org/10.1016/S0196-8904(98)00025-9. refrigerated trucks incorporating phase change material. Appl Energy [22] Johnson RW, Evans JL, Jacobsen P, Thompson JR, Christopher M. The changing 2012;92:336–42. http://dx.doi.org/10.1016/j.apenergy.2011.10.015. automotive environment: high-temperature electronics. IEEE Trans Electron [46] Sharma MK, Buddhi D. Performance studies of refrigerated van having PCM Pack Manuf 2004 Jul;27(3):164–76. http://dx.doi.org/10.1109/ for generating off site refrigeration effect. J Pure Appl Sci Technol 2013; 3(1): TEPM.2004.843109. 1–12. . recovery system for a hybrid electric vehicle. In: Proc 2011 SAE World [47] Ahmed M, Meade O, Medina MA. Reducing heat transfer across the insulated Congress & Exhibition. Detroit, MI; 2011 April 12–14. http://dx.doi.org/ walls of refrigerated truck trailers by the application of phase change 10.4271/2011-01-1171. materials. Energy Convers Manage 2010;51(3):383–92. http://dx.doi.org/ [24] Chiew L, Clegg MW, Willats RH, Delplanque G, Barrieu E. Waste heat energy 10.1016/j.enconman.2009.09.003. harvesting for improving vehicle efficiency. In: Proc 2011 SAE World [48] Manzela AA, Hanriot SM, Cabezas-Gomez L, Sodre JR. Using engine exhaust Congress & Exhibition. Detroit, MI; 2011 April 12–14. http://dx.doi.org/ gas as energy source for an absorption refrigeration system. Appl Energy 10.4271/2011-01-1167. 2010;87(4):1141–8. http://dx.doi.org/10.1016/j.apenergy.2009.07.018. [25] Gumus M. Reducing cold-start emission from internal combustion engines by [49] Lambert MA, Jones BJ. Automotive adsorption air conditioner powered by means of thermal energy storage system. Appl Therm Eng exhaust heat. Part 2: detailed design and analysis. Proc Inst Mech Eng Part D: 2009;29(4):652–60. http://dx.doi.org/10.1016/ J Automob Eng 2006;220(7):973–89. http://dx.doi.org/10.1243/ j.applthermaleng.2008.03.044. 09544070JAUTO222. [26] Gumus M, Ugurlu A. Application of phase change materials to pre-heating of [50] Pandiyarajan V, Pandian MC, Malan E, Velraj R, Seeniraj RV. Experimental evaporator and pressure regulator of a gaseous sequential injection system. investigation on heat recovery from diesel engine exhaust using finned shell Appl Energy 2011;88(12):4803–10. http://dx.doi.org/10.1016/ and tube heat exchanger and thermal storage system. Appl Energy j.apenergy.2011.06.053. 2011;88(1):77–87. http://dx.doi.org/10.1016/j.apenergy.2010.07.023. [27] Claudy P, Létoffé JM, Neff B, Damin B. Diesel fuels: determination of onset [51] Conover J, Husted H, MacBain J, McKee H. Logistics and capability crystallization temperature, pour point and filter plugging point by implications of a Bradley fighting vehicle with a fuel cell auxiliary power differential scanning calorimetry. Correlation with standard test unit. In: Proc 2004 SAE World Congress. Detroit, MI; 2004 March 8–11. http:// methods. Fuel 1986;65(6):861–4. http://dx.doi.org/10.1016/0016- dx.doi.org/10.4271/2004-01-1586. 2361(86)90082-7. [52] Pahinkar DG, Kumar V. Analytical model for human thermal comfort in [28] Vázquez J, Palacios R, Sanz-Bobi MA. Thermoelectric device to allow diesel passenger vehicles. SAE Int J Mater Manuf 2011;4(1):204–15. http:// engine start-up at cold weather conditions. In: Proc 7th European Workshop dx.doi.org/10.4271/2011-01-0130. N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1559

[53] Finn J, Ewing DJ, Ma L, Wagner J. Thermal protection of vehicle payloads using Lubricants Meeting. Rio de Janeiro, Brazil; 2010 May 5–7. http://dx.doi.org/ phase change materials and liquid cooling. In: Proc 2010 American Control 10.4271/2010-01-1509. Conference. Baltimore, MD; 2010 June 30–July 2. p. 1204–10. [76] Hitchcock J. Ground vehicle power and mobility overview (Presentation). [54] Coveny D. United States Navy. Navy news service – eye on the fleet [Internet]; Warren (MI): U.S. Army, TACOM Research Development and Engineering 2007 November 30 [cited 30.07.12]. VIRIN #: 071130-N-4198C-001. . docs/citations/ADA470330>. [55] Meissner E, Richter G. The challenge to the automotive battery industry: the [77] Vetrovec J. Engine cooling system with a heat load averaging capability. In: battery has to become an increasingly integrated component within the Proc Convergence 2008. Detroit, MI; 2008 October 20–22. http://dx.doi.org/ vehicle electric power system. J Power Sources 2005;144(2):438–60. http:// 10.4271/2008-01-1168. dx.doi.org/10.1016/j.jpowsour.2004.10.031. [78] Kim K, Choi K, Kim Y, Lee K, Lee K. Feasibility study on a novel cooling [56] Divya KC, Østergaard J. Battery energy storage technology for power systems technique using a phase change material in an automotive engine. Energy – an overview. Electr Power Syst Res 2009;79(4):511–20. http://dx.doi.org/ 2010;35(1):478–84. http://dx.doi.org/10.1016/j.energy.2009.10.015. 10.1016/j.epsr.2008.09.017. [79] Urciuoli D, Green R, Lelis A, Ibitayo D. Performance of a dual, 1200 V, 400 A, [57] Alamgir M, Sastry AM. Efficient batteries for transportation applications. In: silicon-carbide power MOSFET module. In: Proc IEEE Energy Conversion Proc Convergence 2008. Detroit, MI; 2008 October 20–22. http://dx.doi.org/ Congress and Exposition. Atlanta, GA; 2010 September 12–16. p. 3303–10. 10.4271/2008-21-0017. http://dx.doi.org/10.1109/ECCE.2010.5618324. [58] Pesaran AA, Markel T, Tataria HS, Howell D. Battery requirements for plug-in [80] Kelly K, Abraham T, Bennion K, Bharathan D, Narumanchi S, O’Keefe M. hybrid electric vehicles - analysis and rationale. In: Proc 23rd International Assessment of thermal control technologies for cooling electric vehicle power Electric Vehicle Symposium. Anaheim, CA; 2007 December 2–5. . CA; 2007 December 2–5. . (presentation). In: Proc CALSTART webinar; 2011 April 21. Report no.: NREL/ [81] McCluskey FP, Grzybowski R, Podlesak T. High temperature electronics. The PR-5400-51474. . [82] Stevanovic LD, Matocha KS, Losee PA, Glaser JS, Nasadoski JJ, Arthur SD. [60] Gao Y, Ehsani M. Investigation of battery technologies for the army’s hybrid Recent advances in silicon carbide MOSFET power devices. In: Proc Applied vehicle application. In: Proc 56th IEEE Vehicular Technology Conference, vol. Power Electronics Conference and Exposition. Palm Springs, CA; 2010 3. Vancouver, Canada; 2002 September 24–28. p. 1505–9. http://dx.doi.org/ February 21–25. p. 401–7. http://dx.doi.org/10.1109/APEC.2010.5433640. 10.1109/VETECF.2002.1040467. [83] Boteler L, Urciuoli D, Ovrebo G, Ibitayo D, Green R. Thermal performance of a [61] Heuvers JK. Energy storage commonality military vs. commercial trucks. In: dual 1.2 kV, 400 a silicon-carbide MOSFET power module. In: Proc Proc Hybrid Truck Users Forum National Conference. Atlanta, GA; 2009 Semiconductor Thermal Measurement and Management Symposium. Santa October 27–29. . Clara, CA; 2010 February 21–25. p. 170–5. http://dx.doi.org/10.1109/ [62] Matthe R, Turner L, Mettlach H. VOLTEC battery system for electric vehicle STHERM.2010.5444297. with extended range. SAE Int J Eng 2011;4(1):1944–62. http://dx.doi.org/ [84] Jankowski NR, Everhart L, Geil BR, Tipton CW, Chaney J, Heil T, et al. 10.4271/2011-01-1373. Stereolithographically fabricated aluminum nitride microchannel substrates [63] Pesaran AA, A V, Burch SD. Thermal performance of EV and HEV battery for integrated power electronics cooling. In: Proc 11th Intersociety Thermal modules and packs. In: Proc 14th International Electric Vehicle Symposium. and Thermomechanical Phenomena in Electronic Systems Conference. Orlando, FL; 1997 December 15–17. . ITHERM.2008.4544269. [64] Lukic SM, Cao J, Bansal RC, Rodriguez F, Emadi A. Energy storage systems for [85] Sharar DJ, Jankowski NR, Morgan B. Thermal performance of a direct-bond- automotive applications. IEEE Trans Ind Electron 2008;55(6):2258–67. http:// copper aluminum nitride manifold-microchannel cooler. In: Proc dx.doi.org/10.1109/TIE.2008.918390. Semiconductor Thermal Measurement and Management Symposium. Santa [65] Kim GH, Pesaran A. Analysis of heat dissipation in Li-ion cells & modules for Clara, CA; 2010 February 21–25. p. 68–73. http://dx.doi.org/10.1016/ modeling of thermal runaway (presentation). In: Proc 3rd International j.rser.2007.10.005. Symposium on Large Lithium Ion Battery Technology and Application. Long [86] Boteler L, Jankowski N, Geil B, McCluskey P. A micromachined manifold Beach, CA; 2007 May 15–18. Report no.: NREL/PR-540-41531. . and Exposition. Lake Buena Vista, FL; 2009 November 13–19. p. 61–8. http:// [66] Lisbona D, Snee T. A review of hazards associated with primary lithium and dx.doi.org/10.1115/IMECE2009-11780. lithium-ion batteries. Process Saf Environ Prot 2011;89(6):434–42. http:// [87] Jankowski NR, McCluskey FP. Modeling transient thermal response of pulsed dx.doi.org/10.1016/j.psep.2011.06.022. power electronic packages. In: Proc IEEE Pulsed Power Conference. [67] Rao Z, Wang S. A review of power battery thermal energy management. Washington, DC; 2009 June 28–July 2. p. 820–5. http://dx.doi.org/10.1109/ Renew Sustain Energy Rev 2011;15(9):4554–71. http://dx.doi.org/10.1016/ PPC.2009.5386368. j.rser.2011.07.096. [88] Meysenc L, Jylhakallio M, Barbosa P. Power electronics cooling effectiveness [68] Pesaran AA. Battery thermal management in EVs and HEVs: issues and versus thermal inertia. IEEE Trans Power Electr 2005;20(3):687–93. http:// solutions. In: Proc Advanced Automotive Battery Conference. Las Vegas, NV; dx.doi.org/10.1109/TPEL.2005.846548. 2001 February 5–8. . Hybrids Manuf Technol 1980;3(4):588–94. http://dx.doi.org/10.1109/ [69] Kelly KJ, Mihalic M, Zolot M. Battery usage and thermal performance of the TCHMT.1980.1135667. Toyota Prius and Honda Insight during chassis dynamometer testing. In: Proc [90] Evans AG, He MY, Hutchinson JW, Shaw M. Temperature distribution in 7th IEEE Battery Conference on Applications and Advances. Long Beach, CA; advanced power electronics systems and the effect of phase change materials 2002 January 15–18. p. 247–52. http://dx.doi.org/10.1109/ on temperature suppression during power pulses. J Electron Pack BCAA.2002.986408. 2001;123(3):211–7. http://dx.doi.org/10.1115/1.137037. [70] General Motors Corporation. Chevrolet Volt’s revolutionary Voltec electric [91] Jankowski NR, Morgan BC, McCluksey FP. Design and fabrication of a drive system delivers efficiency with performance (press release); 2010 substrate integrated phase change thermal buffer heat sink. In: Proc October 10 [cited 10.01.12]. . IMECE2009-12713. [71] Sabbah R, Kizilel R, Selman JR, Al-Hallaj S. Active (air-cooled) vs. passive [92] Korin E, Reshef R, Tshernichovesky D, Sher E. Improving cold-start (phase change material) thermal management of high power lithium-ion functioning of catalytic converters by using phase-change materials. In: packs: limitation of temperature rise and uniformity of temperature Proc SAE International Congress and Exposition. Detroit, MI; 1998 February distribution. J Power Sources 2008;182(2):630–8. http://dx.doi.org/10.1016/ 23–26. http://dx.doi.org/10.4271/980671. j.jpowsour.2008.03.082. [93] Laing PM. Development of an alternator-powered electrically-heated catalyst [72] Lukic SM, Khateeb SA, Al-Hallaj S, Selman JR, Emadi A. On the suitability of a system. In: Proc SAE International Congress and Exposition. Detroit, MI; 1994 new high-power lithium ion battery for hybrid electric vehicle applications. February 28–March 3. http://dx.doi.org/10.4271/941042. In: Proceedings of the Future Transportation Technology Conference & [94] Burch SD, Potter TF, Keyser MA, Brady MJ, Michaels KF. Reducing cold-start Exhibition. Costa Mesa, CA; 2003 June 23–25. http://dx.doi.org/10.4271/ emissions by catalytic converter thermal management. In: Proc SAE 2003-01-2289/2003-01-2289. International Congress and Exposition. Detroit, MI; 1995 February 27– [73] Kim GH, Gonder J, Lustbader J, Pesaran A. Thermal management of batteries March 2. http://dx.doi.org/10.4271/950409. in advanced vehicles using phase-change materials. World Electr Vehicle J [95] Burch SD, Keyser MA, Colucci CP, Potter TF, Benson DK, Biel JP. Applications 2008; 2(2): 46–59. . International Spring Fuels & Lubricants Meeting. Dearborn, MI; 1996 May [74] Gross O, Clark S. Optimizing electric vehicle battery life through battery 6–8. http://dx.doi.org/10.2172/266686. thermal management. SAE Int J Eng 2011;4(1):1928–43. http://dx.doi.org/ [96] United States, Environmental Protection Agency. Federal test procedure 10.4271/2011-01-1370. review project: preliminary technical report; 1993 May. Report no.: EPA [75] Lin W, Sunden B. Vehicle cooling systems for reducing fuel consumption and 420-R-93-007. . 1560 N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561

[97] Johnson TV. Diesel emission control in review. In: Proc SAE World Congress [120] Zalba B, Marin JM, Cabeza LF, Mehling H. Review on thermal energy storage and Exhibition. Detroit, MI; 2006 April 3–6. http://dx.doi.org/10.4271/2006- with phase change: materials, heat transfer analysis and applications. Appl 01-0030. Therm Eng 2003;23(3):251–83. http://dx.doi.org/10.1016/S1359- [98] Hendricks TJ, Lustbader JA. Advanced thermoelectric power system 4311(02)00192-8. investigations for light-duty and heavy duty applications. I. In: Proc [121] Sharma A, Tyagi VV, Chen CR, Buddhi D. Review on thermal energy storage International Conference on Thermoelectrics. Long Beach, CA; 2002 August with phase change materials and applications. Renew Sustain Energy Rev 25–29. p. 381–6. http://dx.doi.org/10.1109/ICT.2002.1190343. 2009;13(2):318–45. http://dx.doi.org/10.1016/j.rser.2007.10.005. [99] Smith K, Thornton M. Feasibility of thermoelectrics for waste heat recovery in [122] Agyenim F, Hewitt N, Eames P, Smyth M. A review of materials, heat transfer conventional vehicles. Technical report. Golden (CO): United States and phase change problem formulation for latent heat thermal energy Department of Energy, National Renewable Energy Laboratory; 2009 April. storage systems (LHTESS). Renew Sustain Energy Rev 2010;14(2):615–28. Report no.: NREL/TP-540-44247. Contract no.: AC36-99-GO10337. http:// http://dx.doi.org/10.1016/j.rser.2009.10.015. dx.doi.org/10.2172/951806. [123] Kenisarin MM. High-temperature phase change materials for thermal energy [100] Fu J, Liu J, Feng R, Yang Y, Wang L, Wang Y. Energy and exergy analysis on storage. Renew Sustain Energy Rev 2010;14(3):955–70. http://dx.doi.org/ gasoline engine based on mapping characteristics experiment. Appl Energy 10.1016/j.rser.2009.11.011. 2013;102:622–30. http://dx.doi.org/10.1016/j.apenergy.2012.08.013. [124] Nomura T, Okinaka N, Akiyama T. Technology of latent heat storage for high [101] Dashevsky Z, Kaftori D, Rabinovich D. High efficiency thermoelectric unit temperature application: a review. ISIJ Int 2010;50(9):1229–39. http:// within an autonomous solar energy converter. In: Proc IEEE International dx.doi.org/10.2355/isijinternational.50.1229. Conference on Thermoelectrics. Nagoya, Japan; 1998 May 24–28. p. 531–4. [125] Cabeza LF, Castell A, Barreneche C, de Gracia A, Fernández AI. Materials used http://dx.doi.org/10.1109/ICT.1998.740434. as PCM in thermal energy storage in buildings: a review. Renew Sustain [102] Saitoh TS, Hoshi A. Proposed solar Rankine cycle system with phase change Energy Rev 2011;15(3):1675–95. http://dx.doi.org/10.1016/j.rser.2010.11.01. steam accumulator and CPC solar collector. In: Proc 37th Intersociety Energy [126] Oro E, de Gracia A, Castell A, Farid MM, Cabeza LF. Review on phase change Conversion Engineering Conference. Washington, DC; 2002 July 29–31. p. materials (PCMs) for cold thermal energy storage applications. Appl Energy 725–730. http://dx.doi.org/10.1109/ICT.1998.740434. 2012;99:513–33. http://dx.doi.org/10.1016/j.apenergy.2012.03.058. [103] Medrano M, Gil A, Martorell I, Potau X, Cabeza LF. State of the art on high- [127] Himran S, Suwono A, Mansoori GA. Characterization of alkanes and paraffin temperature thermal energy storage for power generation. Part 2 – case waxes for application as phase change energy storage medium. Energy studies. Renew Sustain Energy Rev 2010;14(1):56–72. http://dx.doi.org/ Source 1994;16(1):117–28. http://dx.doi.org/10.1080/00908319408909065. 10.1016/j.rser.2009.07.03. [128] Wang XQ, Yap C, Mujumdar AS. A parametric study of phase change material [104] Crane DT, LaGrandeur JW. Progress report on BSST-led US Department of (PCM)-based heat sinks. Int J Therm Sci 2008;47(8):1055–68. http:// Energy automotive waste heat recovery program. J Electron Mater dx.doi.org/10.1081/j.ijthermalsci.2007.07.016. 2010;39(9):2142–8. http://dx.doi.org/10.1007/s11664-009-0991-0. [129] Alawadhi EM, Alqallaf HJ. Building roof with conical holes containing PCM to [105] Vining CB. An inconvenient truth about thermoelectrics. Nat Mater reduce the cooling load: numerical study. Energy Convers Manage 2009;8(2):83–5. http://dx.doi.org/10.1038/nmat2361. 2011;52(8–9):2958–64. http://dx.doi.org/10.1016/j.enconman.2011.04.004. [106] LaLonde AD, Pei Y, Wang H, Snyder GJ. Lead telluride alloy thermoelectric. [130] Stupar A, Drofenik U, Kolar J. Optimization of phase change material heat Mater Today 2011;14(11):526–32. http://dx.doi.org/10.1016/S1369- sinks for low duty cycle high peak load power supplies. IEEE Trans Compon 7021(11)70278-4. Pack Manuf Technol 2012;2(1):102–15. http://dx.doi.org/10.1109/ [107] Weerasinghe WMSR, Stobart RK, Hounsham SM. Thermal efficiency TCPMT.2011.2168957. improvement in high output diesel engines a comparison of a Rankine [131] Shamberger PJ, Forero DE. Towards high energy density, high conductivity cycle with turbo-compounding. Appl Therm Eng 2010;30(14–15):2253–6. thermal energy storage composites. In: Proc ASME Micro/Nanoscale Heat & http://dx.doi.org/10.1016/j.applthermaleng.2010.04.028. Mass Transfer International Conference. Atlanta, GA; 2012 March 3–6. [108] Miller EW, Hendricks TJ, Peterson RB. Modeling energy recovery using [132] Kenisarin M, Mahkamov K. Solar energy storage using phase change thermoelectric conversion integrated with an organic Rankine bottoming materials. Renew Sustain Energy Rev 2007;11(9):1913–65. http:// cycle. J Electron Mater 2009;38:1206–13. http://dx.doi.org/10.1007/s11664- dx.doi.org/10.1016/j.rser.2006.05.005. 009-0743-1. [133] Rozanna D, Chuah TG, Salmiah A, Choong TSY, Sa’ari M. Fatty acids as phase [109] Horst TA, Rottengruber HS, Seifert M, Ringler J. Dynamic heat exchanger change materials (PCMs) for thermal energy storage: a review. Int J Green model for performance prediction and control system design of automotive Energy 2005;1(4):495–513. http://dx.doi.org/10.1081/GE-200038722. waste heat recovery systems. Appl Energy 2013;105:293–303. http:// [134] Aydin AA, Aydin A. High-chain fatty acid esters of 1-hexadecanol for low dx.doi.org/10.1016/j.apenergy.2012.12.060. temperature thermal energy storage with phase change materials. Sol Energy [110] Sprouse III C, Depcik C. Review of organic Rankine cycles for internal Mater Sol Cell 2012;96:93–100. http://dx.doi.org/10.1016/ combustion engine exhaust waste heat recovery. Appl Therm Eng j.solmat.2011.09.013. 2013;30(1–2):711–22. http://dx.doi.org/10.1016/ [135] Aydin AA. High-chain fatty acid esters of 1-octadecanol as novel organic j.applthermaleng.2012.10.017. phase change materials and mathematical correlations for estimating the [111] Gopal NK, Subbarao R, Pandiyarajan V, Velraj R. Thermodynamic analysis of a thermal properties of higher fatty acid esters’ homologous series. Sol Energy diesel engine integrated with a PCM based energy storage system. Int J Mater Sol Cell 2013;113:44–51. http://dx.doi.org/10.1016/ Thermodynam 2010; 13(1): 15–21. . [136] Hormansdorfer G, inventor. Latent heat storage material and use thereof. [112] Hale DV, Hoover MJ, O’Neill MJ. Phase change materials handbook. Technical United States patent 4,795,580; 1989 January 3. report. Huntsville (AL): National Aeronautics and Space Administration; 1971 [137] Talja RA, Roos YH. Phase and state transition effects on dielectric, mechanical, September. Report no.: NASA-CR-61363. Contract no.: NAS8-25183. http:// and thermal properties of polyols. Thermochim Acta 2001;380(2):109–21. dx.doi.org/2060/19720012306. http://dx.doi.org/10.1016/S0040-6031(01)00664-5. [113] Abhat A. Low temperature latent heat thermal energy storage: heat storage [138] Kaizawa A, Maruoka N, Kawai A, Kamano H, Jozuka T, Senda T, et al. materials. Sol Energy 1983;30(4):313–32. http://dx.doi.org/10.1016/0038- Thermophysical and heat transfer properties of phase change material 092X(83)90186-X. candidate for waste heat transportation system. Heat Mass Transfer [114] Dutil D, Rousse DR, Ben Salah N, Lassue S, Zalewski L. A review on phase- 2008;44(7):763–9. http://dx.doi.org/10.1007/s00231-007-0311-2. change materials: mathematical modeling and simulations. Renew Sustain [139] Kakiuchi H, Yamazaki M, Yabe M, Chihara S, Terunuma Y, Sakata Y, et al. A Energy Rev 2011;15(1):112–30. http://dx.doi.org/10.1016/j.rser.2010.06.01. study of erythritol as phase change material. In: Proc 2nd workshop on PCMs [115] Gil A, Medrano M, Martorell I, Lázaro A, Dolado P, Zalba B, et al. State of the and chemical reactions for thermal energy storage. International Energy art on high temperature thermal energy storage for power generation. Part 1 Agency ECES Annex 10. Sofia, Bulgaria; 1998 April 11–13. – concepts, materials and modellization. Renew Sustain Energy Rev [140] Jesus AJL, Nunes SCC, Silva MR, Beja AM, Redinha JS. Erythritol: crystal 2010;14(1):31–55. http://dx.doi.org/10.1016/j.rser.2009.07.03. growth from the melt. Int J Pharm 2010;388(1–2):129–35. http://dx.doi.org/ [116] Fan L, Khodadadi JM. Thermal conductivity enhancement of phase change 10.1016/j.ijpharm.2009.12.043. materials for thermal energy storage: a review. Renew Sustain Energy Rev [141] Kakiuchi H, Chihara S, Yamazaki M, Isaki T, inventors. Mitsubishi Chemical 2011;15(1):24–46. http://dx.doi.org/10.1016/j.rser.2010.08.007. Corporation, assignee. Heat storage material composition. United States [117] Krishnan S, Garimella SV. Analysis of a phase change energy storage system patent 5,785,885; 1998 July 28. for pulsed power dissipation. IEEE Trans Compon Pack Technol [142] Ona EP, Zhang X, Kyaw K, Watanabe F, Matsuda H, Kakiuchi H, et al. 2004;27(1):191–9. http://dx.doi.org/10.1109/TCAPT.2004.825758. Relaxation of supercooling of erythritol for latent heat storage. J Chem Eng [118] Pal D, Joshi YK. Application of phase change materials to thermal control of Jpn 2001;34(3):376–82. http://dx.doi.org/10.1252/jcej.34.376. electronic modules: a computational study. J Electron Pack [143] Jankowski NR, McCluksey FP. Electrical supercooling mitigation in erythritol. 1997;119(1):40–50. http://dx.doi.org/10.1115/1.2792199. In: Proc 14th International Heat Transfer Conference, vol. 7. Washington, DC; [119] Hauer A, Mehling H, Schossig P, Yamaha M, Cabeza L, Martin V, et al. 2010 August 7–13. p. 409–16. http://dx.doi.org/10.1115/IHTC14-22306. Advanced thermal energy storage through phase change materials and [144] Kohno H, Maruyama Si, Yasaki H, Uraji T, inventors. Nikken Chemicals chemical reactions – feasibility studies and demonstration projects. Final Company, Ltd., assignee. Heat storage composition. European Union patent report. International Energy Agency, Implementing Agreement on Energy application EP0722997; 1996 January 10. Conservation through Energy Storage, Annex 17; 2004. . dx.doi.org/10.1021/je3000469. N.R. Jankowski, F.P. McCluskey / Applied Energy 113 (2014) 1525–1561 1561

[146] Kubota M, Ona EP, Watanabe F, Matsuda H, Hidaka H, Kakiuchi H. Studies on [162] Zheng N, Wirtz RA. A hybrid thermal energy storage device, Part 1: design

phase change characteristics of binary mixtures of erythritol and MgClÁ6H2O. methodology. J Electron Pack 2004;126(1):1–7. http://dx.doi.org/10.1115/ J Chem Eng Jpn 2007;40(1):80–4. http://dx.doi.org/10.1252/jcej.40.80. 1.1646419. [147] Ryu HW, Woo SW, Shin BC, Kim SD. Prevention of supercooling and [163] Benson DK, Webb JD, Burrows RW, McFadden JDO, Christensen C. Materials stabilization of inorganic salt hydrates as latent heat storage materials. Sol research for passive solar systems: solid-state phase-change materials. Energy Mater Sol Cell 1992;27(2):161–72. http://dx.doi.org/10.1016/0927- Technical Report. Golden (CO): United States Department of Energy, Solar 0248(92)90117-8. Energy Research Institute; 1985 March. Report no.: SERI/TR-255-1828. [148] Lane GA. Phase change materials for energy storage nucleation to prevent Contract no.: AC02-83CH10093. http://dx.doi.org/10.2172/5923397. supercooling. Sol Energy Mater Sol Cell 1992;27(2):135–60. http:// [164] Chandra D, Chien WM, Gandikotta V, Lindle DW. Heat capacities of ‘‘plastic dx.doi.org/10.1016/0927-0248(92)90116-7. crystal’’ solid state thermal energy storage materials. Z Phys Chem [149] Porisini FC. Salt hydrates used for latent heat storage: corrosion of metals and 2002;216(12):1433–44. http://dx.doi.org/10.1524/zpch.2002.216.12.1433. reliability of thermal performance. Sol Energy 1988;41(2):193–7. http:// [165] Chandra D, Chellappa R, Chien WM. Thermodynamic assessment of binary dx.doi.org/10.1016/0038-092X(88)90136-3. solid-state thermal storage materials. J Phys Chem Solids 2005;66(2– [150] Guillot S, Faik A, Rakhmatullin A, Lambert J, Veron E, Echegut P, et al. 4):235–40. http://dx.doi.org/10.1016/j.jpcs.2004.08.047. Corrosion effects between molten salts and thermal storage material for [166] Yan Q, Liang C. The thermal storage performance of monobasic, binary and concentrated solar power plants. Appl Energy 2012;94:174–81. http:// triatomic polyalcohols systems. Sol Energy 2008;82(7):656–62. http:// dx.doi.org/10.1016/j.apenergy.2011.12.057. dx.doi.org/10.1016/j.solener.2007.12.008. [151] Oró E, Miró L, Barreneche C, Martorell I, Farid MM, Cabeza LF. Corrosion of [167] Li W, Zhang D, Zhang T, Wang T, Ruan D, Xing D, et al. Study of solid–solid

metal and polymer containers for use in PCM cold storage. Appl Energy phase change of (n-CnH2n+1NH3)2MCl4 for thermal energy storage. 2013;109:449–53. http://dx.doi.org/10.1016/j.apenergy.2012.10.049.. Thermochim Acta 1999;326(1–2):183–6. http://dx.doi.org/10.1016/S0040- [152] Haynes WM, editor. CRC handbook of chemistry and physics. Boca Raton, 6031(98)00497-3. FL: CRC; 2011. [168] He DH, Di YY, Yao Y, Liu YP, Dan WY. Crystal structure, low-temperature [153] Michels H, Pitz-Paal R. Cascaded latent heat storage for parabolic trough solar heat capacities, and thermodynamic properties of bis(dodecylammonium)

power plants. Sol Energy 2007;81(6):829–37. http://dx.doi.org/10.1016/ tetrachlorocuprate (C12H28N)2CuCl4(s). J Chem Eng Data j.solener.2006.09.008. 2010;55(12):5739–44. http://dx.doi.org/10.1021/je100699g. [154] Gasanaliev AM, Gamataeva BY. Heat-accumulating properties of melts. Russ [169] He DH, Di YY, Tan ZC, Yi FF, Dan WY, Liu YP. Crystal structures and

Chem Rev 2000;69(2):179–86. http://dx.doi.org/10.1070/ thermochemistry on phase change materials (n-CnH2n+1NH3)2CuCl4(s) (n=14 RC2000v069n02ABEH000490. and 15). Sol Energy Mater Sol Cell 2011;95(10):2897–906. http://dx.doi.org/ [155] Ge H, Li H, Mei S, Liu J. Low melting point liquid metal as a new class of phase 10.1016/j.solmat.2011.06.014. change material: an emerging frontier in energy area. Renew Sustain Energy [170] Ruan D, Li W, He L, Hu Q. Phase diagrams of binary systems of Rev 2013;21:331–46. http://dx.doi.org/10.1016/j.rser.2013.01.008. alkylammonium tetrachlorometallates(II). J Therm Anal Calorim [156] Fukuoka Y, Ishizuka M. Thermal analysis of a new high density package 1995;45:235–42. http://dx.doi.org/10.1007/BF02548685. cooling technology using low melting point alloys. Jpn J Appl Phys [171] Steinert S, Voigt W, Glausch R, Neuschutz M. Thermal characteristics of solid- 1991;30(Pt 1):1313–9. http://dx.doi.org/10.1143/JJAP.30.1313. solid phase transitions in long-chain dialkyl ammonium salts. Thermochim [157] Gurrum SP, Joshi YK, Kim J. Thermal management of high temperature pulsed Acta 2005;435(1):28–33. http://dx.doi.org/10.1016/0165-1633(89)90054-3. electronics using metallic phase change materials. Numer Heat Transfer Part [172] Zalba B, Marin JM, Cabeza LF, Mehling H. Free-cooling of buildings with phase A 2002;42(8):777–90. http://dx.doi.org/10.1080/10407780290059800. change materials. Int J Refrig 2004;27(8):839–49. http://dx.doi.org/10.1016/ [158] Directive 2011/65/EU of the European Parliament and of the Council of 8 June j.ijrefrig.2004.03.015. 2011 on the restriction of the use of certain hazardous substances in [173] Feldman D, Shapiro MM, Banu D, Fuks CJ. Fatty acids and their mixtures as electrical and electronic equipment. Ofﬁcial J Eur Union 2011; L174:54: 88– phase-change materials for thermal energy storage. Sol Energy Mater 110. . [174] Peippo K, Kauranen P, Lund PD. A multicomponent PCM wall optimized for [159] Eveloy V, Ganesan S, Fukuda Y, Wu J, Pecht MG. WEEE, RoHS, and what you passive solar heating. Energy Build 1991;17(4):259–70. http://dx.doi.org/ must do to get ready for lead-free electronics. In: Proc 6th International 10.1016/0378-7788(91)90009-R. Conference on Electronic Packaging Technology. Shenzhen, China; 2005 [175] Peippo K. Faasimuutosvaraston termodynaamiset perusmekanismit ja August 30–September 2. p. 27–44. http://dx.doi.org/10.1109/ energiasovelluksia (The basic thermodynamic phenomenon in phase ICEPT.2005.1564718. change material storage and energy applications). Technical report. [160] Ma K, Liu J. Liquid metal cooling in thermal management of computer chips. Helsinki, Finland: Helsinki University of Technology, Department of Front Energy Power Eng Chin 2007;1(4):384–402. http://dx.doi.org/10.1007/ Physics; 1989. Report no.: TKK-F-B125 (in Finnish). s11708-007-0057-3. [161] Indalloy Specialty Alloys [Internet]. Clinton (NY): Indium Corporation. c2010 [cited 28.01.12]. .