Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

A review and future application of Rankine Cycle to passenger vehicles for MARK
waste heat recovery
⁎
Feng Zhoua, , Shailesh N. Joshia, Raphael Rhote-Vaneyb, Ercan M. Dedea
a
Toyota Research Institute of North America, 1555 Woodridge Ave., Ann Arbor, MI 48105, USA
b
Siemens Product Lifecycle Management Software Inc., 38695 Seven Mile Road, Livonia, MI 48152, USA

A R T I C L E I N F O A BS T RAC T

Keywords: Rankine Cycle (RC) is a thermodynamic cycle that converts thermal energy into mechanical work, which is
Rankine Cycle (RC) commonly found in thermal power generation plants. Recently, there have been many studies focusing on
Waste Heat Recovery (WHR) applying Rankine Cycle to recover low-grade waste heat. On-road vehicles, which convert around one third of
Passenger vehicle the fuel energy into useful mechanical energy for propulsion, are moving energy conversion systems that
Two-phase cooling
generate considerable waste heat. It is found from prior research that the Rankine Cycle has great potential in
Exhaust gas
automobile waste heat harvesting applications. However, in contrast with other low-grade waste heat
Engine coolant
applications, vehicles have limited space for the RC system integration, and the waste heat is relatively
unsteady. In this work, the efforts in the past few decades to apply RC to on-road vehicles, specifically passenger
cars, are reviewed. Characteristics of the waste heat sources found in vehicles and the constraints put on the
automotive RC application are identified. Rankine Cycle architectures, system components, and working fluids
suitable to different applications are summarized, which provides a guideline for future RC system design in
automobiles. Lastly, a new concept and case study into the future application of Rankine Cycle to vehicle waste
heat recovery (WHR) is provided.

1. Introduction energy recovery.

In 2012, vehicle fuel-efficiency standards were announced by the
Obama administration that require all US cars and light trucks to reach
54.5 miles per gallon on average by model year 2025 [1]. As a result,
automobile manufacturers are working to improve fuel efficiency in
order to meet current and future fuel economy requirements plus
emissions regulations. Vehicle fuel efficiency is logically related to the
efficiency of the internal combustion engine, where approximately 60–
70% of fuel energy is wasted in the form of heat.
It has been estimated that the thermal efficiency of a modern
internal combustion (IC) engine is limited to 20–40% while 33% of the
fuel energy from a typical medium-size passenger vehicle is carried
away by exhaust gases and 33% is carried away by engine cooling water
in urban traffic conditions [2]. Depending on the engine type and
operating conditions, the IC exhaust gas temperature usually varies
from 500 to 900 ℃ and the engine cooling water temperature is around
100 ℃. It is reported that, for a typical light duty 4-cylinder spark
ignition engine, the waste energy carried by the exhaust gas ranges
from 4.6 to 120 kW and cooling water heat ranges from 9 to 48 kW [3],
which makes the exhaust gas and engine cooling heat very attractive for
Several technologies have been investigated for waste heat recovery
(WHR) including thermoelectric generators (TEG), turbochargers, six-
stroke cycle internal combustion engines, the Rankine Cycle (RC), etc.
The RC system has been identified as a promising solution to harvest-
ing part of this waste energy from vehicles as regenerated mechanical
or electrical power [4–7]. A RC system includes four main components:
1) an evaporator, 2) an expander, 3) a condenser, and 4) a pump; refer
to the flow loop of a typical RC system in Fig. 1. The pump drives the
working fluid to circulate through the loop, and the evaporator utilizes
a waste heat source to vaporize the working fluid. The fluid vapor
expands in the expander and converts thermal energy into mechanical
power output. Then, the expanded vapor flows through a condenser to
turn back into liquid phase, thus completing the cycle.
The first study examining the application of RC to an automotive
application was reported in the 1970 s [8], and since then extensive
work has been done to investigate the feasibility of applying RC to
automobiles (with a primary focus on heavy-duty diesel engine trucks
[9–33]). Generally, a 10–15% fuel efficiency improvement is reported
for diesel engine truck applications. The application of RC to passenger
vehicles is more challenging due to space limitations, which necessi-

⁎
Corresponding author.
E-mail address: feng.zhou@toyota.com (F. Zhou).

http://dx.doi.org/10.1016/j.rser.2016.11.080
Received 6 January 2016; Received in revised form 1 August 2016; Accepted 4 November 2016
Available online 14 November 2016
1364-0321/ © 2016 Elsevier Ltd. All rights reserved.
F. Zhou et al. Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

Fig. 1. Schematic of a general Rankine Cycle (RC) system layout including four main
components: evaporator, expander, condenser and pump.

Fig. 3. Structure 2 of the RC system for application to passenger vehicles, in which a

tates further investigation.
recuperator is added before the evaporator using the steam from the expander to preheat
In this article, existing architectures of RC systems, prior research the working fluid.
that has been done on RC passenger vehicle applications, expander
selection, suitable working fluids, and RC system vehicle integration The most common and simple Rankine Cycle system structure
strategies are summarized. Thus, this review provides a comprehensive (structure 1) is shown in Fig. 2, which utilizes the exhaust gas as the
guideline for future RC system design for passenger vehicles. only heat source to evaporate the working fluid. The heat from engine
Additionally, as a point of differentiation from prior studies, a unique coolant is dissipated to the environment through the radiator and is not
concept and case study on the future application of RC to passenger recovered by the RC system. The second structure design is shown in
vehicle waste heat recovery logically follows the state-of-the-art Fig. 3, in which another heat exchanger (i.e. recuperator) is added
summary. before the evaporator using the steam from the expander to preheat the
working fluid. Similar to structure 1, the engine coolant waste heat is
not utilized as well. A third structure is shown in Fig. 4; different from
2. RC architectures for passenger vehicle application
structure 2, waste heat from the engine coolant is used to preheat the
working fluid. The regenerative preheating of structure 2 requires a
For automotive applications, the RC system is usually installed
very complex liquid-gas heat exchanger with high surface area, while
downstream of the catalyst to avoid negative influence on emission
the preheater in structure 3 only requires a simple liquid-liquid heat
control by extending the time for the catalyst to reach light-off
exchanger. For some working fluids, e.g. organic fluids, structure 2
temperature at cold start and to utilize the extra energy produced by
(with a recuperator) is needed to cool down the vapor exiting from the
pollutant conversion within the catalyst [34]. Lecompte et al. [35]
expander, which is still superheated, to reduce the cooling load of the
conducted a thorough review of organic RC architectures for general
condenser.
WHR. However, heat sources on vehicles are different from other
There have been contradicting conclusions about the effect of
typical industrial heat sources, which are stable and have fewer space
preheating using engine coolant, e.g. structure 3, on the RC system
limitations for system implementation. To be technically feasible for
efficiency. Based on Vaja and Gambarotta's work [36], the RC system
vehicle RC application, the system should be kept as simple as possible
with the engine coolant preheating allows a net increase in power
and utilize components that already exist to reduce cost, weight, and
output, compared to structure 1, of 10–35%, depending on which
complexity. Another consideration requiring special treatment is
working fluid is chosen. Alberto Boretti [37,38] also showed a 8.2% fuel
related to the different temperature ranges of the two main heat
economy improvement using engine coolant to preheat the RC cycle,
sources found in vehicles. In this section, commonly used structures
compared to a 6.4% improvement when only exhaust gas is used to boil
and associated flow loops of RC systems for passenger vehicle applica-
tions are summarized.

Fig. 2. Structure 1 of the RC system for application to passenger vehicles, which utilizes Fig. 4. Structure 3 of the RC system for application to passenger vehicles, in which waste
the exhaust gas as the only heat source to evaporate the working fluid. heat from the engine coolant is used to preheat the working fluid.

1009
F. Zhou et al.
the working fluid. Arias et al. [39] also compared the combined exhaust
and engine coolant heat recovery system, structure 3, with the exhaust
only structure, structure 1. However, it was found that the additional
power recovered from the engine coolant system was 20 W out of a
total 2.14 kW, which is a ~1% improvement. It was argued that the
reason why it is not advantageous to build a combined engine coolant
and exhaust gas waste heat recovery system is due to the 100 ℃
constraint imposed on the engine coolant temperature which is not a
good match for preheating the working fluid. Duparchy et al. [40] also
analyzed the waste heat recovery potential for engine cooling water and
exhaust gas from both energy and exergy points of view. It was found
that the proportion of thermal energy carried by exhaust gas is
comparable with that dissipated by engine cooling water, which means
these two sources are roughly equivalent from the energy point of view.
However, exhaust gas shows a much greater recovery potential in terms
of exergy due to the greater temperature differential relative to
ambient. The main reason why opposite conclusions were drawn from
different studies is that different working fluids were selected for the
RC system. Vaja and Gambarotta [36] and Alberto Boretti [37,38]
adopted organic fluids, e.g. benzene, R11, R134a and 245fa, while
Arias et al. [39] and Duparchy et al. [40] selected water as the working
fluid. These organic fluids have moderate evaporating temperature and
low evaporation latent heat, compared to a higher boiling point and
higher latent heat of water. Therefore, for organic fluids, the preheating
energy ranges from 30% to 50% of the total obtained energy, while it is
only about 10% for water [3].
To utilize the engine waste heat, two more structure options could
be considered. The first option (structure 4) is shown in Fig. 5, which
was proposed by Arias et al. [39]. Instead of using engine coolant to
preheat the working fluid, the engine block, which has a higher
temperature, is used to partially evaporate the working fluid. For this
structure, complete evaporation of the working fluid inside the engine
block should be prevented. A design parameter, e.g. quality, was
recommended to be selected and given a value as an intermediate
state in the boiling process, so that the two-phase mixture would flow
out of the engine block and into the evaporator to complete the boiling
and superheating process via the exhaust gas [39].
The second option (structure 5) is to build separate system loops
and use different working fluids to match the specific characteristics of
the engine coolant and exhaust gas; see Fig. 6. The first generation
waste heat recovery system developed by BMW [34] adopted a similar
dual circuit system, in which water was used as the medium in the
high-temperature (HT) loop and ethanol as working fluid in the low-
temperature (LT) loop. The dual loop system can utilize the full
potential of the waste energy with the disadvantages of large system
Fig. 5. Structure 4 of the RC system for application to passenger vehicles, in which the
engine block at a higher temperature than the coolant is used to partially evaporate the
working fluid.
1010
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
Fig. 6. Structure 5 of the RC system for application to passenger vehicles, which utilizes
separate system loops and different working fluids to match the specific characteristics of
the engine coolant and exhaust gas.
volume, greater system complexity, and increased cost. Therefore, this
RC architecture is usually being applied to big vehicles, such as long-
haul trucks with diesel engine [33,41,42].
Another RC system which is integrated with engine cooling was
developed by Toyota [43], see Fig. 7. This system minimizes the
number of required components by adopting an evaporative engine
cooling structure and a refrigerant as the working fluid, so that the
engine functions as an evaporator and the existing radiator works as a
condenser. The only extra component added to the system is an
expander for power conversion. However, this structure requires a
modification of the existing engine cooling system, while the current
engine cooling technology is quite mature and has used 50/50, water/
ethylene glycol, as the coolant for decades.
There are some other RC architectures available which are not
documented here since they are generally variants or combinations of
the six different architectures. Kim et al. [44] proposed a single-loop
ORC structure for gasoline vehicle engine waste heat recovery from
both the exhaust gas and engine coolant. The proposed system is
basically a combination of Structure 2 and 3 with a low temperature
(LT) and a high temperature (HT) recuperator before and after the
engine coolant preheater. Therefore, the RC working fluid is preheated
by the LT recuperator, vaporized by the engine coolant, and further
heated by the HT recuperator and finally by the exhaust gas. Compared
to other single loop architectures, this system fully utilizes the waste
heat from both exhaust gas and engine coolant, and minimizes the
temperature differential between the heat source and working fluid,
thus providing higher thermal efficiency by sacrificing more space for
extra heat exchangers. Shi et al. [45] added an Exhaust Gas Mixture
Recirculation (EGMR) loop to the Structure 1 RC system, so that the
exhaust gas flowing out of the evaporator at lower temperature is
Fig. 7. Structure 6 of the RC system for application to passenger vehicles, which adopts
an evaporative engine cooling structure and a refrigerant as the working fluid to
minimize the number of required add-on RC components.
F. Zhou et al. Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

Table 1 WRC
ηm = .
Comparisons between different RC structures. Pe (2)
Pros Cons ηth Herein, all of the reported fuel economy improvement values are
Structure 1 Simple structure. Only recover exhaust gas 1–5%
converted to either thermal efficiency or mechanical efficiency for
waste heat. consistency and ease of comparison. Assuming approximately one third
Structure 2 Reutilize waste heat of An extra liquid-gas heat 1.5–5.5% of the total fuel energy is transferred to mechanical power output,
vapor from expander; exchanger. which is a typical value for modern IC engines, then
Reduce the cooling
load of the condenser. ηth ≈ ηm /3. (3)
Structure 3 Recover part of waste An extra liquid-liquid heat 1.5–5.5%
heat from engine exchanger; The first RC application to a small passenger vehicle was reported
coolant. Single working fluid to
in 1993 by Oomori and Ogino [43] from Toyota Motor Corporation.
match two heat sources at
different temperatures. The adopted system structure is structure 6, as shown in Fig. 7. Based
Structure 4 Recover waste heat from Modification of engine 2–5.5% on the study, approximately 3% of the engine output power was
engine more efficiently. block. recovered from the engine waste heat at an ambient temperature of
Structure 5 Maximize the energy Complex system. 3–6%
25 ℃.
recovered from both
engine coolant and Chammas and Clodic [3] conducted a rough estimation of fuel
exhaust gas. efficiency improvement for RC application to hybrid vehicles (HV). In
Structure 6 Simplest structure; Significant redesign of the 1–3% their work, analysis was conducted from the energy balance perspec-
engine and engine cooling tive. Different working fluids, e.g. water, R-245ca and isopentane, and
system;
different technical options, e.g. high pressure steam Rankine bottoming
Same coolant for engine No waste heat recovery
cooling and RC system. from exhaust gas. cycle, low pressure steam Rankine Cycle, organic Rankine Cycle, etc.,
are compared. It was estimated that using the available heat exchanger
and expander technologies, steam RC can offer up to 4% thermal
partially recirculated and mixed with the real-time high-temperature efficiency improvement while organic RC can provide up to 5% thermal
exhaust gas from the engine. In this way, both the flow rate and exhaust efficiency improvement since the heat released by the engine cooling
gas temperature at the evaporator inlet can be actively controlled circuit can be better recovered by organic fluids.
according to the driving profile to provide the optimal operating Following [3], Arias, Shedd and Jester [39] conducted theoretical
conditions for the ORC cycle. analysis of different structures of the RC system for waste heat recovery
Table 1 summarizes the pros and cons of the six different RC from a HV. Three different RC structures (i.e. structures 1, 3 and 4)
structures and the typical ranges of RC thermal efficiency for passenger were compared. It was found that structure 4 was the most efficient RC
vehicle applications; see Eq. (1) for the definition of RC thermal system and approximately 5.5% of the total inlet energy and 7.5% of
efficiency. It should be noted that the RC system thermal efficiency the total waste heat can be recovered when the engine block is used as
not only depends on the RC structure, but also the vehicle type, RC the preheater and the exhaust gas is used to super-heat the working
component type, etc. Therefore, the values reported by different groups fluid.
varied significantly. Besides, these values are based on work that has In 2007, Honda R & D Co., Ltd [46,47] reported an application of a
been reported publically to date. As component and integration RC system to a hybrid vehicle. To realize a compact design for vehicle
technology is being continuously developed, the RC thermal efficiency installation and high efficiency, densely layered multi-plate fins were
has great potential to be higher than the upper limit of the ranges adopted in the boiler and the unit was integrated with the engine
presented in the current paper. cylinder head to minimize the heat loss and system size. According to
For the structure design of the RC passenger vehicle application, the vehicle testing and numerical modelling, the thermal efficiency was
maximizing the recovered energy is not the only objective. System increased from 28.9–32.7% at 100 km/h constant cruising speed.
complexity, component volume and weight, and the related extra cost In 2008, BMW [34,48] unveiled their first generation “turbostea-
added to the vehicle are also big concerns. Accordingly, a comprehen- mer” to use the potential waste heat from the IC engine. A dual-circuit
sive review paper has logical benefits to automotive engineers/ system was developed to utilize the waste heat from both the engine
researchers and may facilitate greater adoption of associated waste coolant and the exhaust gas, with ethanol and water as the working
heat recovery technologies. fluids in the two circuits, respectively. An additional 10–15% power
output, which is approximately 5.7% thermal efficiency improvement,
was achieved. However, due to added weight, high system complexity,
3. Implementation of RC to passenger vehicle and space restrictions, the first generation turbosteamer was deemed
generally not feasible for mobile applications [49].
In this section, some typical applications of RC system to passenger The second generation turbosteamer [49], which simplified the
vehicles reported in open literature to date are reviewed and summar- prior system by only utilizing the waste heat from the exhaust gas to
ized in a rough chronological order. Two improvements in efficiency enable efficient vehicle integration, was reported in 2012. This second
were commonly used in the literature to show the impact of the RC generation system can reduce the mechanical engine load by approxi-
system to the vehicle efficiency, one of which is Thermal Efficiency, mately 6%, which is around 2% thermal efficiency improvement. BMW
ηth, defined by claimed that a higher fuel efficiency improvement can be reached with
intelligent heat management optimized for waste heat recovery. Some
WRC modelling work from BMW [50,51] followed, which underlined the
ηth = ,
ṁ fuel⋅LHV (1) importance of including vehicle integration effects when the fuel
efficiency improvement is estimated and foresaw an even higher fuel
in which WRC is net power output of the RC system, ṁ fuel is the fuel saving potential in hybrid-electric vehicles.
consumption rate and LHV is fuel low heating value. The other He et al. [52,53] conducted a steady-state experiment on a dual
definition is Mechanical efficiency, ηm, which is the ratio of the net loop waste heat recovery system for an IC engine, which is similar to
power output of the RC system, WRC, and the effective power produced structure 5. The difference is that a Kalina Cycle (KC) is used to harvest
by the engine, Pe, waste heat from the engine cooling loop and RC is only adopted to

1011
F. Zhou et al. Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

recover energy from exhaust gas. Different working fluids for the RC
loop were compared and a cycle thermal efficiency of 18–26% was
achieved, depending on RC fluid selection and evaporation tempera-
ture, which indicated a thermal efficiency improvement of 12–17.3% if
two-thirds of the total fuel energy is assumed to be wasted by the
engine coolant and exhaust gas. However, the proposed combined cycle
is more feasible for naval engine applications than for passenger
vehicles considering the system complexity and space requirement
[52].
Alberto Boretti [37,38] studied the potential application of RC to a
1.8L naturally aspirated HV. The waste heat recovery from exhaust gas
Fig. 8. BMW's specifically designed two-stage impulse turbine integrated with generator
and engine coolant increase the fuel efficiency by up to 6.4% and 2.8%,
for their 2nd generation turbosteamer [49].
respectively, and up to 8.2% combined with corresponding averaged
values to be 3.4%, 1.7% and 5.1% for steady operation. For transient
expander. In their work [54], a commercial Green Turbine™ [56] that
operation, the integrated system showed a 4.2% better fuel economy
works only with water as the working fluid was selected for one of their
over the full New European Driving Cycle (NEDC) and 6.4% better fuel
study cases. The main characteristics of the turbine include a maximum
economy at 120 km/h cruising speed after complete engine warm-up.
inlet stream pressure of 5.2 bar, maximum inlet stream temperature of
Wang et al. [41] conducted a performance analysis of a dual loop
200 ℃, a power output of 2.5 kW, a length of 25 cm and a diameter of
RC system for a gasoline engine, which is similar to structure 5. The
19 cm, and a total weight of 7 kg. The reduced mass and overall
only difference is that an extra heat exchanger was added to the system
dimensions made the turbine one of the most appropriate candidates
to use the vapor from the high temperature expander to preheat the
for a RC vehicle waste heat recovery application.
working fluid of the low temperature loop before it is heated by the
Due to the current lack of commercial micro-turbine expanders,
engine coolant. The combined cycle system augmented the net power
most researchers used an in-house specifically designed turbine to test
output by 14–16% in the peak effective thermal efficiency region and
their RC system [49,57]. BMW's second generation Turbosteamer [49]
this increment percentage went up to 30–50% in the small load region.
adopted a specifically designed two-stage impulse turbine integrated
The thermal efficiency improvement is within 3–6% range throughout
with an electric generator for their RC system; see Fig. 8. The turbine
the engine's operating region.
works in the pressure range of 6–10 bar and the rotary speed is about
Heidrich and Krisch [2] assessed various RC implementations for
50,000 to 100,000 rpm. The turbine-generator unit has an overall
waste heat recovery and found that the waste heat recovery benefit is
length of 150 mm and a diameter of 100 mm, with a total weight of
low (~0.2 kW) for passenger cars used for city driving since these cars
approximately 2 kg. The electrical power output of the unit is about
mostly operate at idle speed and low load conditions. On the other
2 kW. Compared to a scaled-down reaction turbine which is generally
hand, passenger vehicles that are used mostly on countryside high-
used in power plants, the impulse turbine design significantly reduces
ways demonstrate a significant waste heat recovery benefit.
the flow leakage. The lubrication free nature of the design is another
Domingues et al. [54] evaluated the vehicle exhaust gas waste heat
advantage when compared to volumetric expansion machines.
recovery potential using RC system. The simplest RC structure
In general, the advantages of a turbine expander include compact
(structure 1) was adopted in their study since they considered
structure, light weight and high efficiency. However, the design and
compactness an essential criterion for RC vehicle applications. It was
manufacturing of the turbine is very difficult, which leads to a relatively
found that, using the selected shell and tube heat exchanger at an
higher cost with the added risk that the turbine efficiency decreases
evaporating pressure of 2 MPa, the RC system increased the mechan-
significantly under off-design conditions. Furthermore, a turbine
ical efficiency by 2.64–6.96%. It was also predicted that improved
expander puts more constraints on the working fluid selection since a
expander designs allowing for higher evaporating pressures will further
turbine cannot handle two-phase conditions. Therefore, if the heat
improve the energy recovery potential of the RC system. Appropriately,
addition is variable, the wetness in the expansion process may become
the topic of expander selection is thus presented in the next section.
uncontrollable and liquid droplets may damage the turbine blades.
Also, a speed reduction gearbox might be required due to speed
4. Expander selection
mismatch if the turbine outputs the mechanical energy directly to the
crank shaft. It is suggested that turbine expanders might be preferred
The expander is one of the most critical components in an RC
in combination with electrical generators for energy conversion and
system and an inappropriately selected expander might decrease the
storage, thus making it a practical solution for HVs [5].
system efficiency significantly. Many factors need to be considered
during the expander selection process for an RC system targeting
passenger vehicles; such factors include the inlet and outlet working 4.2. Scroll expander
conditions, pressure ratio, power output, working fluid, weight, vo-
lume, cost, reliability, isentropic efficiency, lubrication requirement, A scroll expander is a displacement type of expansion machine that
complexity, and rotational speed, to name a few. In general, expanders is characterized by a lower flow rate, higher pressure ratio and much
may be placed in one of two categories: 1) velocity type, e.g. an axial lower rotational speed than a turbo-machine [3,58]. Recently, the scroll
turbine expander, and 2) volume type, e.g. a scroll expander, screw expander has become a candidate for micro- and small-scale RC
expander, or reciprocal piston expander [55]. These various types of applications due to its compact structure, reliability, fewer moving
energy conversion machines are discussed in this section. parts, and lower level of noise and vibration [58]. When compared with
turbine expanders, these displacement machines can tolerate two-
4.1. Turbine expander phase conditions that may be present at the end of the coolant
expansion process under dynamic vehicle operating conditions.
A turbine expander is rotary machine that converts the kinetic A schematic that shows the operational cycle of a scroll expander is
energy carried by a fluid stream into mechanical energy when it passes shown in Fig. 9 [59]. The machine is composed of two swirl-shaped
through a set of blades. So far, few micro-scale turbine expanders ( < scrolls, the mobile scroll and the fixed scroll. The rotational direction of
10 kWe) are commercially available. Domingues et al. [54] reported RC the mobile scroll is opposite to the swirl direction, therefore the space
application studies for vehicles that used a commercial micro-turbine between the two scrolls increases as the mobile scroll rotates. Then, the

1012
F. Zhou et al.
Fig. 9. Schematic showing the operational cycle of a scroll expander [59] that is composed of two swirl-shaped scrolls, the mobile scroll and the fixed scroll.
high pressure vapor flows into the machine and expands to a lower
pressure. The machine becomes a compressor if the rotational direction
changes. Currently, the majority of scroll expanders for RC vehicle
applications are modified from scroll compressors. There are two types
of scroll expanders: compliant and constrained. The former uses a
centrifugal effect to keep the mobile scroll in continuous contact with
the fixed scroll, and thus requires lubrication, while the latter is
constrained either radially or axially, or both, and can operate without
lubrication.
For an RC system with small power output, leakage of the working
fluid is the dominant cause of low efficiency for a scroll expander. The
first RC application to passenger vehicle study from Toyota [43]
adopted a scroll expander which had an inlet volume of 40 cc and an
expansion ratio of 2. The size of the scroll expander is 120 mm by
190 mm. The maximum expander efficiency reached in the study is
approximately 50%. It was found that the expander efficiency is
reduced due to deterioration of sealing performance and fluid leakage
at rotational speed below 1000 rpm.
Yamada [60] experimentally demonstrated a micro organic RC
generator for low-grade waste heat recovery. The overall size of the full
system is 385 mm×320 mm×275 mmand the net electrical power
output is around 5 W. The expander they adopted is an originally
developed scroll-type expander with a built-in expansion ratio of 1.99;
see Fig. 10. The scroll expander was selected because the efficiency of
displacement type expanders is relatively high for small power outputs
with minimal components compared to a turbine expander.
Regardless, the maximum achieved isentropic efficiency of the expan-
der was 4.55% due to the low speed of the expander and thus the high
working fluid leakage. The authors in [60] claimed that by increasing
the rotational speed of the expander the fluid leakage may be reduced
thus increasing expander efficiency.
4.3. Screw expander
The screw expander has been widely adopted as an expansion
Fig. 10. Photograph and schematic of scroll expander adopted to build a micro organic RC generator for low-grade waste heat recovery [60].
1013
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
machine in RC plants. This type of expander is composed of a pair of
meshing helical rotors that are packed within a housing with approxi-
mately 50 μm clearance; see Fig. 11 [55]. The fluid is admitted into the
expander from one end and expelled from the other. When the rotors
rotate, the volume trapped between the rotors and the housing
changes, which makes the fluid volume either increase or decrease,
depending on the rotational direction.
The screw expander has been extensively studied by Smith et al. at
City University London for decades, covering expander modelling,
performance prediction and experimental testing [61–66]. However,
implementing a micro-scale screw expander ( < 10 kWe) for a vehicle
application is challenging in the current market and few reports from
the open literature were found for screw expander applications in RC
systems with power output lower than 10 kWe [55]. The study reported
by Wang et al. [41] is one of the few exceptions where a screw expander
is adopted as the machine in the RC system for vehicle waste heat
recovery, and the net power output of the expander for the low
temperature loop (i.e. engine coolant) was 9.57 kW versus 26.39 kW
for the high temperature exhaust gas loop.
In general, a screw expander can tolerate two-phase flow and is
highly efficient in off-design conditions. However, like other positive
displacement devices, the seal is critical in the prevention of internal
leakage [67]. Lubrication is also required to avoid direct contact, while
at the same time, sealing between the lobes of the two rotors; this
feature makes the device relatively more expensive to fabricate than
scroll expanders.
4.4. Reciprocating piston expander
The piston expander is one of the oldest displacement machines.
There is currently a large regrowth of interest in piston expanders for
small-scale RC systems, e.g. waste heat recovery from internal combus-
tion engines [68]. A reciprocating piston expander has three main
processes: intake, expansion, and exhaust. The intake valve and
expansion valve open and close alternatively to allow the high pressure
F. Zhou et al. Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

Fig. 11. Screw expander composed of a pair of meshing helical rotors that are packed within a housing [55]. (a) View from rear and top; (b) View from front and bottom.

vapor to fill and expand in the chamber, driving the piston downward conducted more detailed comparisons of different expanders that can
for mechanical energy output. The expanded low pressure vapor is also be referred to.
expelled through the exhaust valve to finish the cycle. Reciprocating
machines have a good isentropic efficiency (about 70%) and a good
power output/size ratio. 5. Working fluid selection
Endo et al. from Honda R & D [46] chose a volumetric axial piston
as the expanding machine for their RC system due to its low operating When designing an RC system, special attention should also be paid
speed and comparatively flat efficiency characteristics compared to a to the working fluid selection according to the heat source temperature,
turbine expander. Their expander has a geometric expansion ratio of which has a significant effect on the system thermal and exergetic
14.7 and a maximum speed of 3000 rpm. A motor-generator was efficiency. The choice of the working fluid used in the RC depends on a
mounted coaxially in the expander unit to convert the mechanical number of factors including thermodynamic efficiency, environmental,
energy into electricity. A thermal efficiency of 8–13% was reached in safety, and process-related economic issues.
the study. Bao and Zhao [67], Chen et al. [69], Hung et al. [70], Lakew and
In general, piston expanders show some advantages over other Bolland [71], Lai et al. [72], Qiu [73] and Hærvig et al. [74] conducted
expansion machines, such as larger built-in volume ratio, higher thorough reviews on the working fluid selection for RC, in general.
achievable operating pressures and temperatures, and the ability to When implementation of a RC system in a passenger vehicle is
tolerate two-phase flow plus low rotational speeds. However, they also considered, stricter criteria should be adopted for the fluid selection
have some disadvantages, such as their heavy weight, valves and torque to maximize safety and minimize environmental concerns in case a
impulse, and many moving parts. fluid leakage occurs; as an example, a low flammability level is a major
Besides the four types of expanders discussed herein, there exist concern for passenger vehicle applications. Therefore, alcohols and
other types of expanders that have been considered in RC applications hydrocarbons, in spite of their good thermodynamic efficiencies, are
for vehicles. These additional types of expanders include Vane-cell arguably not ideal candidates. Instead, refrigerants, which are used in
expanders [34] and Wankel engines [52,53]. The advantages and automotive air conditioning systems, are usually better options.
disadvantages of four main types of expanders are summarized in Generally, according to the slope of the saturation curve, the working
Table 2 for comparison purpose. Qiu et al. [55], Bao and Zhao [67] fluid may be categorized into three different types: 1) wet fluid, 2) dry
fluid, and 3) isentropic fluid; see Fig. 12.
Most inorganic fluids are wet fluids, e.g. water, ammonia, etc.,
Table 2
Advantages and disadvantages of different types of expanders for RC application to which means the expansion ends in the two-phase region. Water, which
passenger vehicles. is the most environmentally friendly, safest, and exhibits the highest
evaporation enthalpy, has historically been the working fluid for many
Expander Advantages Disadvantages
large-scale power cycles. Water delivers higher work output and higher
Turbine Compact structure, light weight, Difficult to design and thermal efficiency than other working fluids, due to which Arias et al.
expander high efficiency, mature fabricate, high cost, low
manufacturability. off-design efficiency,
cannot bear two-phase.
Scroll expander Compact structure, high Leakage of the working
reliability, fewer moving parts, fluid leads to low
lower rotate speed and lower efficiency, low capacity,
level of noise and vibration, can require lubrication.
handle two-phase flow.
Screw expander Tolerate two-phase flow, low Mostly adopted in RC
rotate speed and efficient in off- plants, require seal to
design conditions. prevent internal leakage,
lubrication is required.
Reciprocating Larger built-in volume ratio, Heavy weight, have valves
piston higher achievable operating and torque impulse, many
expander pressures and temperatures, moving parts.
tolerate two-phase flow, low
rotational speeds.
Fig. 12. Three types of working fluid: dry fluid, wet fluid, and isentropic fluid.

1014
F. Zhou et al.
[39], auto manufacturers Honda [46] and BMW [34,49] selected water
as the working fluid for their RC system. It was found by Arias et al.
[39] that when water is selected as the working fluid, the waste heat
from the engine coolant is not effectively recovered by the system due
to the mismatch between the low temperature of the engine coolant
and the high boiling temperature of water. For BMW's first generation
RC system [34], water is used to harvest energy from exhaust gas only,
while ethanol is used in a separate loop for engine coolant waste heat
recovery. In the case of Honda's RC system [46] and BMW's second
generation turbosteamer [49], water is only heated by the exhaust gas.
Therefore, it is safe to conclude that water is a preferable working fluid
for high exhaust gas temperatures ranging from 500 to 800 ℃.
In terms of the disadvantages of water, there is the requirement for
superheating to avoid turbine blade erosion if such a machine is
selected for the expander. Also, the high degree of superheating makes
water less practical for automotive applications due to the variation of
exhaust temperature at different load conditions. Also, water's high
freezing point (0 ℃) cannot meet the standard automotive working
temperature range (−40~85 ℃). Therefore, practical solutions to solve
these issues need to be developed if water is chosen as the working fluid
for RC vehicle applications.
Most organic fluids are either dry fluids, e.g. R113, R245fa, R245ca,
etc., or isentropic fluids, e.g. R11, R134a, R123, etc. Associated CFCs/
HCFCs have been rejected due to environmental concerns and phased
out according to international protocols. Dry/isentropic refrigerants
are widely used in small-scale RC applications because of their good
heat transfer properties, excellent thermal stability and low viscosity.
They are generally non-flammable, which is a big advantage for
automotive application and compatible with most materials. Under
low temperature ambient conditions (0~−40 ℃) they also do not freeze,
which is a major benefit relative to water.
Chammas and Clodic [3] compared different organic fluids with
water for RC application to HVs and argued that using water for an RC
system to recover automotive waste heat could lead to a complex
system requiring large sized equipment and high investment cost,
which makes the study of organic working fluids necessary. He et al.
[52] and Zhang et al. [53] adopted R11 and R113 as the working fluid
for the RC loop of their combined thermodynamic cycle for IC engine
waste heat recovery. Boretti [37,38] selected R245fa as the working
fluid for their RC system in a hybrid passenger car with a naturally
aspirated gasoline engine. Wang et al. [41] chose R245fa and R134a for
the high-temperature and low-temperature RC loops, respectively,
because of their good safety and environmental properties.
Domingues et al. [54] compared R123 and R245fa with water as a
working fluid for vehicle RC waste heat recovery potential from exhaust
gas. The study revealed the advantage of using water as an RC working
fluid to recover waste heat from exhaust gas of vehicles equipped with a
spark-ignition engine. However, it was also found that the heat
exchanger effectiveness for R123 and R245fa is higher than that for
water, and consequently when the exhaust temperature is relatively
low, organic fluids can be considered appropriate for vehicle RC
application. Wang et al. [75] studied the RC system with nine different
pure organic working fluids for engine waste heat recovery. The results
indicated that R11, R141b, R113 and R123 presented slightly higher
thermodynamic performances, while R245fa and R245ca are the most
envrionment-friendly working fluids for engine waste heat recovery
applications.
Dry/isentropic organic fluids, on the other hand, have a few
shortcomings. First, the intrisic property of dry/isentropic fluids reduce
the area of net work in the temperature-entropy (T-S) diagram, which
means less power output compared to wet fluids, e.g. water. Second, to
reduce the cooling load of the condenser, a recuperator (liquid-gas heat
exchagner) is usually necessary to cool the superheated vapor to
saturated state, which increases the system complexity and cost.
Most organic fluids also have relatively low thermal instability tem-
peratures compared to water, which means that at high temperature
1015
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
and pressure they can undergo chemical decomposition and deteriora-
tion [5]. In addition, the current generation of refrigerants, e.g. HFCs
(hydroflurocarbons), has a high global warming potential, which means
that their use could be limited or banned in the near future.
In general, the working fluid selection for RC passenger vehicle
application should consider the following criteria: 1. Safety of passen-
gers is the first priority. Fluids with high flammability or high toxicity
should be avoided; 2. Environmental impact, including Ozone
Depletion Potential (ODP) and Global Warming Potential (GWP)
should be minimized. The EPA regulatory requirements for automotive
refrigerants must be followed; 3. Fluids should be compatible with
automotive application. The freezing point of the RC working fluid
must comply with typical automotive application requirements, e.g. the
operation of the RC system should guarantee engine idle and startup
temperatures as low as −40 ℃ [76]. Besides, if the RC system is
integrated with some electronic component, such as the application
proposed in Section 7, then the characteristics of the working fluid
should be compatible with electronics, e.g. the fluid most likely needs
be dielectric; 4. The fluid should be selected in a way that makes the
system thermo-economically optimal. From a thermodynamic perfor-
mance point of view, the fluid can be chosen such that it maximizes the
energy recovery [77] based on the characteristics of the waste heat
sources and properties of the fluid, e.g. critical temperature, critical
ideal gas molar heat capacity, etc. [78]. From the economic point of
view, the system size and cost also needs to be considered [79,80].
Following the general criteria summarized above, an appropriate
working fluid can be selected for the specific vehicle application of
interest.
6. System integration and challenges
To utilize the waste heat from vehicles via an RC system, a best
compromise between the system complexity and fuel savings potential
needs to be established. Regarding vehicle integration, it is widely
acknowledged that the approach taken in minimizing the system
complexity, weight, cost, and negative effects on existing components
is more important than maximizing the RC power output. Currently, as
an add-on system to an existing vehicle, RC system integration might
lead to the following important interactions and consequences.
First, an increased vehicle weight cannot be avoided due to extra
system components. This leads to more tractive effort required to
accelerate the vehicle and more rolling resistance from the tires. It is
roughly estimated that for every 100 pounds added to the vehicle, the
fuel economy is decreased by 1–2%. For small passenger vehicles, the
added-on weight has more impact to the fuel economy than it does on
long-haul diesel trucks.
Second, a heat exchanger is necessary to transfer the heat from the
exhaust gas to the working fluid of the RC system. The added pressure
drop from the heat exchanger causes the engine back pressure to
increase thereby affecting the engine performance, in terms of larger
horsepower loss. Therefore, the overall added exhaust back pressure
should be limited to avoid significant engine performance deteriora-
tion. Since the evaporator back pressure depends on the desired heat
exchanger capacity, Chammas and Clodic [3] recommended to define a
limit for the exhaust heat recovery, which is the maximum heat that
can be recovered before the RC system begins to lose its fuel economy
advantages. On the other hand, the evaporator lowers the exhaust gas
temperature, which decreases the exhaust gas velocity and thus a
smaller pressure loss results through the piping and components
following the evaporator [3]. Both effects should be considered when
the engine back pressure increment is calculated. Endo et al. [46]
integrated the evaporator of the RC system with the engine cylinder
head to reduce heat loss and size, which potentially could reduce the
engine back pressure increase. Boretti [37] claimed that turbocharged
engines may better work with an increased back pressure than
naturally aspirated engines. Refer to Hatami [81] for a detailed review
F. Zhou et al.
of different heat exchanger designs for exhaust gas waste heat recovery.
Third, that heat comes from the RC condenser needs to be
dissipated to the environment, which increases the cooling demand
of the vehicles. There are basically two ways to condense the RC
working fluid: 1) a separate air-cooled condenser and 2) integrating
into the existing engine cooling loop. Honda [46] adopted a separate
heat exchanger next to the radiator to condense the water vapor, which
requires extra space for installation and increased air resistance. BMW
[49,51] integrated the RC condenser with the engine cooling loop using
a liquid-to-liquid heat exchanger, so that no extra condenser is needed,
thus saving space in the engine compartment. The downside of this
integration strategy is that the engine coolant temperature at the
condenser inlet may be too high to completely condense the RC
working fluid. Also, the available cooling capacity of the radiator may
not be sufficient, which restricts the condenser heat rejection and thus
RC power output. Additionally, heat transferred to the engine cooling
loop also requires coolant circulation at a higher speed and higher
cooling air velocity; this leads to increased coolant pumping power and
increased air flow resistance.
Finally, the fourth interaction between the RC system and the
vehicle is how to utilize the RC expander power output. Generally, there
are two ways to use this recovered energy: 1) outputting the mechanical
energy directly to the crank shaft or 2) combining the RC system with
electrical generators to convert the mechanical energy into electricity.
In the former method, a speed reduction gearbox might be required if
the expander has a speed mismatch with the engine. For long-haul
diesel trucks, there is more room in the engine compartment and the
exhaust gas condition is more stable making these vehicles good
candidates for the first RC power utilization method [9–12,15–
18,20]. For small passenger vehicles, the driving profile is not as stable
leading to a fluctuating exhaust gas temperature. In this case, the
second energy recovery method seems to be a more reasonable option
[3,39,46,49].
Considering the RC interactions with an existing vehicle as an add-
on system, not all of the RC architectures summarized in Section 2 are
suitable for passenger vehicle application. Structure 5, which has two
separate RC loops, might be a good option for long-haul trucks, but
obviously is not a good choice for passenger vehicles considering its
high system complexity and volume. Structures 4 and 6 requires
modification to the existing power train components, e.g. engine,
which makes them the last to be considered as an add-on system.
Structures 1, 2 and 3 are the three to be first considered, and depending
on the available space for system integration, different structures may
be selected.
However, if the RC system is designed together with the vehicle
power train, all six and perhaps more structures might be possible.
Then, the design methodology would be to optimize the whole system
efficiency, volume, weight, stability, etc. in an integrated fashion,
applying a new concept of a combined-cycle power train, which is
similar to the combined-cycle power stations idea [76].
7. Future application of RC
As mentioned in the Introduction, US government along with other
governments worldwide are proposing legislative standards for fuel
economy and carbon dioxide (CO2) emissions. Conventional internal
combustion engine (ICE) vehicles are reaching their technical limits
and will not be able to meet the fuel economy requirement without
additional measures. Pure electric vehicles are possibly a more
sustainable option with low-to-zero CO2 emission, however their
currently short driving range and long battery charging time limit the
present acceptance of the pure EV architecture. On the other hand, a
hybrid electric vehicle, which recovers some braking energy to increase
the overall fuel efficiency, is a logical mid-term compromise involving a
traditional ICE vehicle that spurs the development of pure EV power
train elements. However, the HV is still limited by the thermal
1016
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
efficiency of ICE, which is in general below 40% for modern car
engines [82]. Nonetheless, by integrating a RC system into a HV to
recover a portion of the waste energy, a hybrid vehicle will break this
“40% limit” to further boost fuel economy. For RC application, a HV
also provides an ideal platform for transforming thermal energy to
electricity, storing the electricity in batteries, and feeding this electri-
city back to the vehicle. Another advantage of applying an RC system to
a HV is the relatively constant load conditions for the engine in a HV.
In this section, a concept and case study on future application of RC
to HVs (regardless of the engine and motor type) for greater compact-
ness and higher power output is provided. The general idea is to
integrate the RC system to the cooling loop of power electronics (PE) of
a HV and use the PE as an extra heat source. For future HVs, the
increasing demand of higher performance combined with smaller PE
module size results in a continuously growing heat dissipation require-
ment, which makes two-phase (2φ) cooling [83,84] attractive due to
enhanced thermal performance. A refrigerant, such as R245fa [85],
HFE7100 [86], or HFC 134a [87] is typically used as the working fluid
in a 2φ power electronics cooling loop. Different fluids may be selected
depending on the type of power device utilized and its maximum
operational temperature. In the current study, R245fa was selected as
the working fluid for demonstration purpose. A 2φ cooling loop has
three main components: 1) a two-phase cold plate, 2) a radiator/
condenser, and 3) a pump, refer to Fig. 13, which provides an excellent
platform for RC integration. If a RC system can be integrated with a 2φ
cooling loop, more waste heat may be utilized to generate power, and
under-hood space may be saved since only an extra boiler and
expander are required.
Fig. 14 shows the concept of combining the RC system with the PE
2φ cooling loop. In the concept, the pump, cold plate and HV radiator
are three main components for the 2φ cooling loop. To realize waste
heat recovery, only an evaporator and an expander need to be added to
the system. A recuperator is added in the current system, which is not
necessary depending on the available space and capacity of the HV
radiator. Also, the waste heat from the engine coolant is not utilized in
this system, but a liquid-to-liquid heat exchanger can be added to
recover part of the coolant heat, considering the low boiling point of the
refrigerant.
A 0-D numerical model of the RC-2φ system is developed on the
LMS ImagineLab Amesim platform [88,89] to analyze the impact of the
power electronics cold plate to the whole energy recovery system. The
sketch of the model is shown in Fig. 15. The model has six sub-models:
1) pump, 2) power electronics, 3) evaporator, 4) expander, 5) HV
radiator, and 6) recuperator. In the current work, one of the input
parameters of the current model is the mass flow rate of the working
Fig. 13. Structure of 2φ power electronics cooling loop that includes three main
components: a PE cold plate, a radiator and a pump [83,84].
F. Zhou et al. Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

Table 3
Assumptions and working fluid properties of the proposed RC-2φ integrated system.

Refrigerant R245fa Radiator coolant 50/50 water

ethylene glycol
R245fa flow rate RC system: Coolant flow rate 10 L/min
0.05, 0.06, 0.07, Coolant inlet temp 40 ℃
0.08 kg/s Pump efficiency 0.6
RC−2φ system: Evaporator 0.85
0.08, 0.09, 0.10, efficiency
0.11 kg/s
Exhaust 590 ℃ Expander 0.6
temperature efficiency
Exhaust flow rate 0.035 kg/s Radiator efficiency 0.9
Exhaust specific 0.9937 kJ/(kg K) Recuperator 0.6
heat efficiency
Power electronics 6 kW

purpose. In the current investigation, a specific driving cycle is not

Fig. 14. Integrated RC-2φ cooling and waste heat recovery system for hybrid vehicle
application which has dual functions of PE cooling and WHR from exhaust gas.
fluid, which is controlled by the pump speed. The total power in the
form of waste heat from the power electronics is vehicle dependent and
assumed to be a constant value, 6 kW, in this study. The efficiency
values are assumed to be constant and the expansion process is
considered to be adiabatic. Both the pump efficiency and expander
efficiency are specified to be 60%. More detailed sub-models could be
developed for these components if needed. However, the accuracy of
these assumptions appears to be sufficient for the proof-of-concept
considered. Therefore, the exhaust gas temperature and mass flow rate
are assumed to be constant as 590 ℃ and 0.035 kg/s, respectively. The
average specific heat of the exhaust gas is assumed to be 0.9937 kJ/
(kg K). 50/50 water/ethylene glycol is used as the coolant in the
condenser, which flows at 10 L/min at an inlet temperature of 40 ℃.
The system working fluid and working conditions are also summarized
in Table 3.
The overall system coefficient of performance (COP) is evaluated as
COP = Wexpander / Wpump, (4)
in which Wexpander is the work output of the expander and Wpump is

Fig. 15. Sketch of the zero-dimensional Amesim simulation model for the newly proposed RC-2φ dual-function system used to analyze the impact of the power electronics waste heat to
the whole energy recovery system.

1017
F. Zhou et al.
Fig. 16. Expander power output, COP and evaporator inlet vapor quality as a function of
expander pressure ratio for traditional RC system without PE cooling (with recuperator
as shown in Fig. 3 for Structure 2).
the work consumed by the pump system.
Fig. 16 shows the expander power output, COP and boiler inlet
vapor quality as a function of expander pressure ratio for the RC system
with recuperator (Structure 2 in Fig. 3) at different R245fa mass flow
rates, 0.05, 0.06, 0.07 and 0.08 kg/s. Observe that as the pressure ratio
increases the expander can generate more power, however the system
COP increases first and then decreases gradually. As for the R245fa
vapor quality at the boiler inlet, it decreases as the pressure ratio
increases. High expander pressure ratio leads to higher system
1018
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
Fig. 17. Expander power output, COP and cold plate outlet vapor quality as a function of
expander pressure ratio for the proposed RC-2φ integrated system with recuperator as
illustrated in Fig. 14.
pressure within the intercooler and boiler which can suppress boiling.
For different R245fa mass flow rates, the expander power output varies
between 1.2 kW and 1.5 kW and decreases as the mass flow rate
increases. The COP also decreases as the mass flow rate increases due
to higher power consumption by the pump. The vapor quality decreases
as the refrigerant mass flow rate increases, which is quite obvious.
The results for the proposed RC-2φ integrated system with
recuperator are shown in Fig. 17. For the current system with two-
phase power electronics cooling, the vapor quality at the cold plate
F. Zhou et al. Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021

outlet (i.e. boiler inlet) is an important parameter that requires
dedicated control. If the flow rate is excessively low, all of the
refrigerant flowing through the cold plate vaporizes, which might
result in dryout of the heat exchange surface within the cold plate.
The flow rate also cannot be excessively high otherwise the flow
through the cold plate becomes single-phase, which lowers cooling
performance. The power electronics is an additional heat source in the
cycle, so higher refrigerant flow rates are necessary to avoid dryout.
Therefore, for the current system structure, the studied flow rates
include 0.08, 0.09, 0.10 and 0.11 kg/s. The variation of the expander
power output, COP and cold plate outlet vapor quality follow the same
trends shown in Fig. 16. However, at the same flow rate (e.g. 0.08 kg/
s), the power electronics boosts the expander power output ~58% to
1.9 kW from 1.2 kW for the system without power electronics. The COP
also increases ~20% from 20 to 24 since additional heat is recovered by
the system. The impact of heat input from the power electronics to the
waste heat recovery system is well demonstrated.
Based on the above analysis, it seems that the application of a RC
system to a HV represents a promising solution to further improve fuel
efficiency. Furthermore, a HV with a turbocharged engine may better
work with an increased back pressure than naturally aspirated engines
[90]. On top of that, the motor and differential of a HV can potentially
be removed and replaced with in-wheel motors to save space for the RC
system [91]. Therefore, it can be foreseen that a turbocharged hybrid
vehicle with motors integrated in wheels and with a RC system co-
designed with a power electronics cooling system will significantly
increase the MPG of ICE driven vehicles and reduce tailpipe emissions.
However, breakthrough of this technology integration requires strong
collaboration between academia and industry, and interdisciplinary
technical know-how is a must.
8. Conclusions
In the current work, investigations from the literature conducted on
waste heat recovery from passenger vehicles using Rankine Cycle over
the past two decades are revisited and summarized, including the
system structure, research and developments to date, expander tech-
nologies, working fluid selection, system integration, etc. The key
studies that were reviewed herein are tabulated in Table 4. A few
major conclusions are drawn as follows.
The implementation of the RC for vehicle waste heat recovery was
originally considered in relation to heavy-duty long-haul diesel trucks.
Recently, a greater number of research groups and vehicle manufac-
turers (e.g. Toyota, Honda, BMW, etc.) are investigating RC applica-
tions to passenger vehicles and an increased application potential has
been shown.
The engine coolant and exhaust gas are the two main heat sources
in a vehicle, which carry a similar amount of thermal energy. However,
exhaust gas shows a much greater recovery potential in terms of exergy
because of a much higher temperature range. For this reason, recover
heat from the two sources using the same loop and same working fluid
might not be appropriate due to the potential mismatch between fluid
properties and heat sources temperature ranges. The waste heat from
the engine could be harvested either by pumping the working fluid
through the engine block or using a separate loop and different working
fluids.
The expander is a critical component in the RC system. The four
main types of expanders have their own advantages and disadvantages
for passenger vehicle applications. The turbine expander has a compact
structure, is light weight, and has high efficiency. However, design and
manufacturing is very difficult, thus leading to a higher cost. The
turbine expander also has lower efficiency under off-design conditions
and cannot tolerate two-phase conditions. It is more suitable to
combine a turbine expander with electrical generators for energy
conversion. The scroll expander is characterized by lower flow rates,
higher pressure ratios and much lower rotational speeds than turbo-
machines. It has a compact structure, fewer moving parts, lower levels
of noise and vibration, and the most cost efficient design. A screw
expander is highly efficient in off-design conditions. However, lubrica-
tion is required to avoid direct contact while achieving a seal between
the lobes of the two rotors; this feature makes the screw expander
relatively more expensive to fabricate than scroll expanders. Piston
expanders have larger built-in volume ratio, high achievable operating
pressures and temperatures, and low rotational speeds. All of the three
displacement types of expanders (scroll, screw, and piston types) have
the ability to tolerate two-phase flow.
Working fluid selection is a very important in the design of a RC
system for passenger vehicles. Water is the preferable working fluid for

Table 4
Summary of the work about the application of Rankine Cycle to passenger vehicles reported to public.

Ref. Year Institute Vehicle/ Heat Source Architect Expander Working fluid RC ηth RC ηm Research
engine method

[43] 1993 Toyota Motor Passenger car Coolant Structure 6 Scroll Expander HCFC123 ⁄ 3% Experiment
Corp.
[3] 2005 Ecole des Mines HEV Exhaust (high Structure 1 Turbine Water 2–4% ⁄ Numerical
de Paris pressure)
Exhaust (low Structure 1 Turbine Water 1.3–3% ⁄ Numerical
pressure)
Exhaust (ORC) Structure 1 Turbine R-245ca 4–5% ⁄ Numerical
[39] 2006 U. of Wisconsin– Prius 2004 Exhaust+cngine Structure 4 Turbine Water 5.5% ⁄ Numerical
Madison block
[46,47] 2007 Honda R & D HEV Exhaust Structure 1 Axial piston Water 3.8% ⁄ Experiment
[34] 2008 BMW-1st gen 3 Series Exhaust+coolant Structure 5 Vane-cell HT: Water; 5.7% ⁄ Experiment
expander LT: Ethanol
[40] 2009 IFP Engine Tech. HEV-Prius Exhaust+coolant Structure 1 Piston Expander Water ⁄ 9% Numerical
Dept. and 3
[52,53] 2011 Xi’an Jiaotong Toyota 8A-FE Exhaust, Structure 5 Refitted Wankel RC: R11, R113; 12–17.3% ⁄ Experiment
Univ. lubricant, coolant engine KC: Ammonia
[49–51] 2012 BMW-2nd gen 5 Series Exhaust Structure 1 Impulse Turbine Water ⁄ 6% Experiment
[37,38] 2012 Missouri Uni. Sci. 1.8L Engine Exhaust Structure 2 Unknown 245fa 3.4% ⁄ Numerical
& Tech Coolant Structure 2 1.7% ⁄ Numerical
Exhaust+coolant Structure 2 5.1% ⁄ Numerical
[41] 2012 Beijing Univ. BL18T Engine Exhaust+coolant Structure 5 Single Screw HT: R245fa; 3–6% ⁄ Numerical
Tech. LT: R134a
[54] 2013 Tech. Univ. Lisbon 2.8L VR6 Exhaust Structure 1 Green Turbine™ Water, R123, ⁄ 2.64– Numerical
R245fa 6.96% Experiment

1019
F. Zhou et al.
high exhaust gas temperature due to its high evaporation enthalpy and
thus higher power output and thermal efficiency. However, the down-
side of using water is the requirement for superheating and its high
freezing point. Organic fluid is more suitable to a low temperature heat
source, e.g. engine coolant, and a high temperature but unstable heat
source, e.g. exhaust gas of a passenger vehicle. Organic fluids usually do
not freeze under typical low temperature ambient conditions. However,
they exhibit a lower power output than water, require an extra
recuperator to reduce the cooling load of the condenser, have relatively
low thermal instability temperatures, and some of them have high
global warming potential. The selection of working fluid should be
considered together with the expander for the specific vehicle RC
application.
System integration is one of the biggest hurdles in applying the RC
system to passenger vehicles for waste heat recovery. The RC system
has many effects on the vehicle, such as increased weight, increased
engine back pressure, increased cooling demand of the vehicle, and
utilization of the RC power output. These effects should be considered
in determining the ultimate feasibility of the system. The application of
a RC system to a turbocharged or regular HV represents a promising
solution.
A novel aspect of the RC-2φ integrated system concept presented
herein is that it not only keeps the HV power electronics cool at high
efficiency, but also recovers more heat from the PE module for higher
total system power output. It is found that the RC-2φ integrated system
boosted the expander power output ~58% from 1.2 to 1.9 kW under the
same working conditions. The COP is also increased by 20% since an
extra heat source is added to the system without consuming additional
pumping power.
The present investigation and case study demonstrate that inte-
grating a RC system with a HV platform has great potential to further
boost the hybrid vehicle fuel economy to meet future emission
requirements. Co-designing the RC system and HV power train in
combination with other technologies, such as two-phase power electro-
nics cooling, turbocharging, and in-wheel motor, may drive the fuel
efficiency of the ICE-based vehicle to a higher level.
References
[1] The White House. 〈https://www.whitehouse.gov/the-press-office/2012/08/28/
obama-administration-finalizes-historic-545-mpg-fuel-efficiency-standard〉; 2012.
[2] Heidrich P, Krisch T. Assessment of waste heat recovery options in passenger car
applications by various rankine cycles. Heat Transf Eng 2015;36(14–15):1321–31.
[3] Chammas RE, Clodic D. Combined cycle for hybrid vehicles, in SAE-2005-01-1171;
2005.
[4] Saidur R, et al. Technologies to recover exhaust heat from internal combustion
engines. Renew Sustain Energy Rev 2012;16(8):5649–59.
[5] Wang T, et al. A review of researches on thermal exhaust heat recovery with
Rankine cycle. Renew Sustain Energy Rev 2011;15(6):2862–71.
[6] Tchanche BF, et al. Low-grade heat conversion into power using organic Rankine
cycles – a review of various applications. Renew Sustain Energy Rev
2011;15(8):3963–79.
[7] Sprouse Iii C, Depcik C. Review of organic Rankine cycles for internal combustion
engine exhaust waste heat recovery. Appl Therm Eng 2013;51(1–2):711–22.
[8] Morgan D , et al. Laboratory test results low emission rankine-cycle engine with
organic-based working fluid and reciprocating expander for automobiles. In:
Proceedings of the 8th Intersociety energy conversion engineering conference.
Philadelphia, PA, USA: AIAA; 1973.
[9] Patel PS, Doyle EF. Compounding the truck diesel engine with an organic Rankine
Cycle system, in SAE-760343; 1976.
[10] Leising CJ. et al. Waste heat recovery in truck engines, in SAE Paper, SAE-780686;
1979.
[11] Doyle E, DiNanno L, Kramer S. Installation of a diesel-organic rankine compound
engine in a class 8 truck for a single-vehicle test, in SAE-790646; 1979.
[12] DiBella FA, DiNanno LucoR, Kopiow MD. Laboratory and on-highway testing of
diesel organic rankine compound long-haul vehicle engine, in SAE-830122; 1983.
[13] Danov SN, Gupta AK. Modeling the performance characteristics of diesel engine
based combined-cycle power plants – Part I: mathematical model. J Eng Gas
Turbines Power 2004;126(1):28–34.
[14] Danov SN, Gupta AK. Modeling the performance characteristics of diesel engine
based combined-cycle power plants – Part II: results and applications. J Eng Gas
Turbines Power 2004;126(1):35–9.
[15] Teng H, Regner G, Cowland C. Achieving high engine efficiency for heavy-duty
diesel engines by waste heat recovery using supercritical organic-fluid rankine
1020
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
cycle, SAE-2006-01-3522, in SAE; 2006.
[16] Teng H, Regner G, Cowland C. Waste heat recovery of heavy-duty diesel engines by
organic rankine cycle Part I: hybrid energy system of diesel and rankine engines, in
SAE-2007-01-0537; 2007.
[17] Teng H, Regner G, Cowland C. Waste heat recovery of heavy-duty diesel engines by
organic rankine cycle Part II: working fluids for WHR-ORC, in SAE-2007-01-0543;
2007.
[18] Teng H, Regner G. Improving fuel economy for HD diesel engines with WHR
Rankine Cycle driven by EGR cooler heat rejection, in SAE-2009-01-2913; 2009.
[19] Teng H , et al. A Rankine Cycle system for recovering waste heat From HD diesel
engines - WHR system development, in SAE-2011-01-0311; 2011.
[20] Park T , et al. A Rankine Cycle system for recovering waste heat from HD diesel
engines - experimental results, in SAE-2011-01-1337; 2011.
[21] Dolz V, et al. HD Diesel engine equipped with a bottoming Rankine cycle as a waste
heat recovery system. Part 1: study and analysis of the waste heat energy. Appl
Therm Eng 2012;36:269–78.
[22] Serrano JR, et al. HD Diesel engine equipped with a bottoming Rankine cycle as a
waste heat recovery system. Part 2: evaluation of alternative solutions. Appl Therm
Eng 2012;36:279–87.
[23] Hountalas DT, et al. Improvement of bottoming cycle efficiency and heat rejection
for HD truck applications by utilization of EGR and CAC heat. Energy Convers
Manag 2012;53(1):19–32.
[24] Macián V, et al. Methodology to design a bottoming Rankine cycle, as a waste
energy recovering system in vehicles. Study in a HDD engine. Appl Energy
2013;104:758–71.
[25] Shu G-Q , et al. Simulations of a bottoming Organic Rankine Cycle (ORC) driven by
waste heat in a Diesel Engine (DE) in SAE-2013-01-0851; 2013.
[26] Yu G, et al. Simulation and thermodynamic analysis of a bottoming Organic
Rankine Cycle (ORC) of diesel engine (DE). Energy 2013;51:281–90.
[27] Furukawa T. et al. A study of the Rankine Cycle generating system for heavy duty
HV trucks, in SAE-2014-01-0678; 2014.
[28] Wang T, et al. Comparisons of system benefits and thermo-economics for exhaust
energy recovery applied on a heavy-duty diesel engine and a light-duty vehicle
gasoline engine. Energy Convers Manag 2014;84:97–107.
[29] Yang F, et al. Performance analysis of waste heat recovery with a dual loop organic
Rankine cycle (ORC) system for diesel engine under various operating conditions.
Energy Convers Manag 2014;80:243–55.
[30] Yue C, Han D, Pu W. Analysis of the integrated characteristics of the CPS
(combined power system) of a bottoming organic Rankine cycle and a diesel engine.
Energy 2014;72:739–51.
[31] Zhang Y-Q, et al. Development and experimental study on organic Rankine cycle
system with single-screw expander for waste heat recovery from exhaust of diesel
engine. Energy 2014;77:499–508.
[32] Kulkarni K, Sood A. Performance analysis of Organic Rankine Cycle (ORC) for
recovering waste heat from a heavy duty diesel engine, in SAE-2015-26-0037;
2015.
[33] Song J, Gu C-w. Parametric analysis of a dual loop organic Rankine cycle (ORC)
system for engine waste heat recovery. Energy Convers Manag
2015;105:995–1005.
[34] Freymann R, Strobl W, Obieglo A. The turbosteamer: a system introducing the
principle of cogeneration in automotive applications. MTZ Worldw
2008;69(5):20–7.
[35] Lecompte S, et al. Review of organic Rankine cycle (ORC) architectures for waste
heat recovery. Renew Sustain Energy Rev 2015;47:448–61.
[36] Vaja I, Gambarotta A. Internal Combustion Engine (ICE) bottoming with Organic
Rankine Cycles (ORCs). Energy 2010;35(2):1084–93.
[37] Boretti A. Recovery of exhaust and coolant heat with R245fa organic Rankine cycles
in a hybrid passenger car with a naturally aspirated gasoline engine. Appl Therm
Eng 2012;36:73–7.
[38] Boretti AA. Transient operation of internal combustion engines with Rankine waste
heat recovery systems. Appl Therm Eng 2012;48:18–23.
[39] Arias DA, Shedd TA, Jester RK. Theoretical analysis of waste heat recovery from an
internal combustion engine in a hybrid vehicle, in SAE-2006-01-1605; 2006.
[40] Duparchy A. et al. Heat recovery for next generation of hybrid vehicles: simulation
and design of a Rankine Cycle system. in International Battery. In: Proceedings of
hybrid and fuel cell electric vehicle symposium; 2009.
[41] Wang EH, et al. Performance analysis of a novel system combining a dual loop
organic Rankine cycle (ORC) with a gasoline engine. Energy 2012;43(1):385–95.
[42] Zhang HG, Wang EH, Fan BY. A performance analysis of a novel system of a dual
loop bottoming organic Rankine cycle (ORC) with a light-duty diesel engine. Appl
Energy 2013;102:1504–13.
[43] Oomori H, Ogino S. Waste heat recovery of passenger car using a combination of
rankine bottoming cycle and evaporative engine cooling system, in SAE-930880;
1993.
[44] Kim YM, et al. Single-loop organic Rankine cycles for engine waste heat recovery
using both low- and high-temperature heat sources. Energy 2016;96:482–94.
[45] Shi R, et al. System design and control for waste heat recovery of automotive
engines based on Organic Rankine Cycle. Energy 2016;102:276–86.
[46] Endo T. et al. Study on maximizing exergy in automotive engines, in SAE-2007-01-
0257; 2007.
[47] Kadota M, Yamamoto K. Advanced transient simulation on hybrid vehicle using
Rankine Cycle system, in SAE-2008-01-0310; 2008.
[48] Ringler J. et al. Rankine Cycle for waste heat recovery of IC Engines, in SAE-2009-
01-0174; 2009.
[49] Freymann R, et al. The second generation turbosteamer. MTZ Worldw
2012;73(2):18–23.
F. Zhou et al.
[50] Horst TA, et al. Dynamic heat exchanger model for performance prediction and
control system design of automotive waste heat recovery systems. Appl Energy
2013;105:293–303.
[51] Horst TA, et al. Prediction of dynamic Rankine Cycle waste heat recovery
performance and fuel saving potential in passenger car applications considering
interactions with vehicles' energy management. Energy Convers Manag
2014;78:438–51.
[52] He M, et al. A combined thermodynamic cycle used for waste heat recovery of
internal combustion engine. Energy 2011;36(12):6821–9.
[53] Zhang X , et al. Exhaust recovery of vehicle gasoline engine based on organic
Rankine Cycle, in SAE-2011-01-1339; 2011.
[54] Domingues A, Santos H, Costa M. Analysis of vehicle exhaust waste heat recovery
potential using a Rankine cycle. Energy 2013;49:71–85.
[55] Qiu G, Liu H, Riffat S. Expanders for micro-CHP systems with organic Rankine
cycle. Appl Therm Eng 2011;31(16):3301–7.
[56] TurbineTM G. 〈http://www.greenturbine.eu/〉; 2015.
[57] Kang SH. Design and experimental study of ORC (organic Rankine cycle) and radial
turbine using R245fa working fluid. Energy 2012;41(1):514–24.
[58] Song P, et al. A review of scroll expanders for organic Rankine cycle systems. Appl
Therm Eng 2015;75:54–64.
[59] Quoilin S, et al. Techno-economic survey of Organic Rankine Cycle (ORC) systems.
Renew Sustain Energy Rev 2013;22:168–86.
[60] Yamada N, Tominaga Y, Yoshida T. Demonstration of 10-Wp micro organic
Rankine cycle generator for low-grade heat recovery. Energy 2014;78:806–13.
[61] Kovacevic A, et al. Developments in modelling positive displacement screw
machines. In: Dudas L, editor. Engineering the future. Sciyo; 2010. http://
dx.doi.org/10.5772/10371 http://www.intechopen.com/books/engineering-the-
future/developments-in-modelling-positive-displacement-screw-machines.
[62] Smith IK SN, Kovacevic A. Screw Expanders Increase Output and decrease the cost
of geothermal binary power plant systems. In: Proceedings of geothermal resources
council annual meeting, September 25–28. Reno, Nevada-USA; 2005.
[63] Smith IK. , et al. Steam as the working fluid for power recovery from exhaust gases
by means of screw expanders. Proc. Inst. Mech. Eng. Part E: J. Process Mech. Eng.
225(2); 2011. p. 117–25.
[64] Read M, Smith IK, Stosic N. Optimisation of screw expanders for power recovery
from low grade heat sources. In ASME 2013 international mechanical engineering
congress and exposition. San Diego, California, USA: ASME; 2013.
[65] Leibowitz H, Smith IK, Stosic N. Cost effective small scale orc systems for power
recovery from low grade heat sources. In: 2006 ASME international mechanical
engineering congress and exposition. Chicago, Illinois, USA: ASME; 2006.
[66] Smith I, Stosic N, Kovacevic A. Power recovery from low grade heat by means of
screw expanders, 1st ed.. Sawston, Cambridge, UK: Chandos Publishing; 2014.
[67] Bao J, Zhao L. A review of working fluid and expander selections for organic
Rankine cycle. Renew Sustain Energy Rev 2013;24:325–42.
[68] Oudkerk JF. , et al. Experimental performance of a piston expander in a small-
scale organic Rankine cycle. In: Proceedings of IOP conference series: materials
science and engineering. 90(1); 2015. p. 012066.
[69] Chen H, Goswami DY, Stefanakos EK. A review of thermodynamic cycles and
working fluids for the conversion of low-grade heat. Renew Sustain Energy Rev
2010;14(9):3059–67.
[70] Hung TC, et al. A study of organic working fluids on system efficiency of an ORC
1021
Renewable and Sustainable Energy Reviews 75 (2017) 1008–1021
using low-grade energy sources. Energy 2010;35(3):1403–11.
[71] Lakew AA, Bolland O. Working fluids for low-temperature heat source. Appl Therm
Eng 2010;30(10):1262–8.
[72] Lai NA, Wendland M, Fischer J. Working fluids for high-temperature organic
Rankine cycles. Energy 2011;36(1):199–211.
[73] Qiu G. Selection of working fluids for micro-CHP systems with ORC. Renew Energy
2012;48:565–70.
[74] Hærvig J, Sørensen K, Condra TJ. Guidelines for optimal selection of working fluid
for an organic Rankine cycle in relation to waste heat recovery. Energy
2016;96:592–602.
[75] Wang EH, et al. Study of working fluid selection of organic Rankine cycle (ORC) for
engine waste heat recovery. Energy 2011;36(5):3406–18.
[76] Colonna P, et al. Organic rankine cycle power systems: from the concept to current
technology, applications, and an outlook to the future. J Eng Gas Turbines Power
2015;137(10), [100801].
[77] Yu H, Feng X, Wang Y. Working fluid selection for Organic Rankine Cycle (ORC)
considering the characteristics of waste heat sources. Ind Eng Chem Res
2016;55(5):1309–21.
[78] Brown JS, Brignoli R, Quine T. Parametric investigation of working fluids for
organic Rankine cycle applications. Appl Therm Eng 2015;90:64–74.
[79] Quoilin S, et al. Thermo-economic optimization of waste heat recovery Organic
Rankine Cycles. Appl Therm Eng 2011;31(14–15):2885–93.
[80] Darvish K, et al. Selection of optimum working fluid for Organic Rankine Cycles by
exergy and exergy-economic analyses. Sustainability 2015;7(11):15362.
[81] Hatami M, Ganji DD, Gorji-Bandpy M. A review of different heat exchangers
designs for increasing the diesel exhaust waste heat recovery. Renew Sustain
Energy Rev 2014;37:168–81.
[82] Kraljevic I , et al. Rankine cycle for hybrid vehicles: the potential for CO2 emissions
reduction. In: Proceedings of the ASME internal combustion engine division fall
technical conference. Vol 2, New York: Amer Soc Mechanical Engineers; 2016.
[83] Joshi SN, Dede EM. Effect of sub-cooling on performance of a multi-jet two phase
cooler with multi-scale porous surfaces. Int J Therm Sci 2015;87:110–20.
[84] Feng Zhou, Joshi SN, Dede EM. Visualization of bubble behavior for jet impinge-
ment cooling with phase change, Thermal and Thermomechanical Phenomena in
Electronic Systems (ITherm), 2014 IEEE Intersociety Conference on, Orlando, FL,
2014, pp. 897–903. http://dx.doi.org/10.1109/ITHERM.2014.6892376.
[85] Honeywell. Honeywell Enovate® 245fa: Morristown, NJ; 2012.
[86] 3M, 3M™ Novec™ 7100 Engineered Fluid. 2009: St. Paul, MN. p. 1–4.
[87] DuPont. DuPont™ Suva® refrigerants; 2004.
[88] Imagine.Lab, L. AMESim – advanced modeling environment for performing
simulations of engineering systems. 〈http://www.plm.automation.siemens.com/
en_us/products/lms/imagine-lab/amesim/〉; 2014.
[89] Galindo J, et al. Modeling and experimental validation of a volumetric expander
suitable for waste heat recovery from an automotive internal combustion engine
using an organic rankine cycle with ethanol. Energies 2016;9(4):279.
[90] Di Battista D, Mauriello M, Cipollone R. Waste heat recovery of an ORC-based
power unit in a turbocharged diesel engine propelling a light duty vehicle. Appl
Energy 2015;152:109–20.
[91] Hall J, et al. Through-the-road parallel hybrid with in-wheel motors, SAE technical
paper 2016-01-1160. in SAE World Congress. Detroit: SAE; 2016.