Recent Advances in High-Flux, Two-Phase Thermal Management: Issam Mudawar
Recent Advances in High-Flux, Two-Phase Thermal Management: Issam Mudawar
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-1
Copyright VC 2013 by ASME
Downloaded from https://thermalscienceapplication.asmedigitalcollection.asme.org/data/journals/jtsebv/927027/tsea_5_2_021012.pdf by guest on 03 January 2020
Fig. 1 Examples of computer, medical, transportation, energy, aerospace, and defense
applications demanding high-heat-flux cooling schemes
Let us now consider a typical two-phase correlation for critical heat Notice here that, unlike the single-phase correlation in Eq. (1), the
flux (CHF) in tubes. Clearly numerous, such correlations are available CHF correlation (a) consists of a dimensionless group that is a
in the literature, but a typical correlation takes the form of [4] function of numerous other independent dimensionless groups,
and (b) each of the independent groups is valid over a finite range.
! The development of a CHF correlation is hindered by the high
q00m qf G2 L cp;f DTsub L G cost of conducting two-phase experiments compared to their
¼f ; ; ; ; pffiffiffiffiffiffiffi ; …
G hfg qg r qf hfg D qf g D single-phase counterparts, which can have a strong bearing on the
8 number of data points available from a given source. With such a
>
> P1;min < P1 < P1;max
>
> limitation, the coverage of individual independent parameters is
>
> P2;min < P2 < P2;max also very limited. This implies that a database that is consolidated
>
>
>
> from different experimental studies to construct a correlation of
< P3;min < P3
> < P3;max
the form of Eq. (2) will be restricted to relatively narrow ranges of
¼ f ðP1 ; P2 ; P3 ; P4 ; P5 ; :::Þ P4;min < P4 < P4;max
>
> many independent parameters. In other words, the correlation is
>
> P5;min < P5 < P5;max valid over a very small region of the multidimensional space
>
>
>
> encompassing all these independent parameters. This is the reason
>
> :
>
: why thermal system designers are often confronted with the
: necessity to utilize correlations to design cooling systems with
(2) parameter ranges outside those for which the correlation is
Fig. 2 Fundamental challenges to the understanding and prediction of fluid flow and heat
transfer in mini/microchannel heat sinks: (a) difficulty ascertaining two-phase flow and heat
transfer regimes, (b) two-phase flow constraints and instabilities, and (c) heat transfer limits
and anomalies
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-3
practical interest possess diameters in the range of 0.1–0.6 mm, likelihood of premature CHF occurrence. This phenomenon is
while diameters in minichannel heat sinks are typically in the range closely associated with two-phase flow instabilities and is encoun-
of 1.0–3.0 mm [2,10–12]. These heat sinks are therefore very com- tered mostly at low mass velocities, where a larger volume of
pact and lightweight, and by allowing the coolant to undergo phase vapor is produced inside the channels. Illustrated in Fig. 2(c), it is
change along the channels, they provide heat transfer coefficients triggered when the momentum of incoming liquid from the heat
far greater than those of their single-phase counterparts by capital- sink’s upstream plenum is momentarily too weak to overcome the
izing upon the coolant’s latent heat content rather than sensible pressure drop across the channel. This causes vapor from the
heat alone. This greatly reduces the coolant flow rate required to microchannels to flow backwards into the inlet plenum [16]. Two
dissipate the same amount of heat compared to a single-phase heat effective means to overcoming premature CHF are to (a) increase
sink, which also helps reduce coolant inventory for the entire cool- mass velocity, which helps prevent vapor backflow, and (b)
ing system. Two-phase heat sinks also provide better temperature increase inlet subcooling, which reduces vapor accumulation in
uniformity by maintaining surface temperatures close to the cool- the upstream plenum by condensation.
ant’s saturation temperature. However, two-phase microchannel The compressibility and flashing effects and likelihood of two-
heat sinks are not without shortcomings, and their implementation phase choking are all reasons behind the need for accurate predic-
is hindered by the relatively limited understanding of two-phase tive tools for pressure drop in two-phase mini/microchannel heat
phase region (for subcooled inlet conditions), saturated two-phase effectiveness in predicting pressure drop data for different cool-
region, downstream vapor region, and outlet pressure recovery. ants as shown in Fig. 3.1
Most studies on pressure drop in two-phase mini/microchannel Despite the overwhelming contribution of the two-phase region,
heat sinks are centered on the saturated two-phase region that a complete pressure drop model must be based on accurate models
contributes the largest percentage of total pressure drop. Three for all component of pressure drop, as well as be able to accu-
different approaches have been used for this purpose: (1) homoge- rately predict the axial extent of individual flow and heat transfer
neous equilibrium model (HEM), used in conjunction with an regimes. The extent of the upstream developing and fully devel-
appropriate two-phase friction factor or two-phase viscosity oped portions of the single-phase liquid region can be determined
model [10], (2) semiempirical models, based mostly on the from a relation such that of Shah and London [19],
Lockhart–Martinelli separated flow model (SFM) [17], and (3)
theoretical models [15]. Lockhart–Martinelli-type formulations
1
have been especially popular in recent years because of their References cited in Fig. 3 are [15,18].
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-5
Table 2 Relations for determination of heat transfer coefficient in mini/microchannel heat
sinks
!
GDh 2
L1/;u;dev ¼ 0:06 þ 0:07b 0:04b Dh (3)
lf
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-7
understanding of this confined flow configuration is of vital impor- this and many other published databases for two-phase jet
tance to the design of the cooling module. By far, the most difficult impingement, to the author’s best knowledge, no reliable correla-
task in designing a practical two-phase jet-impingement cooling tion has been recommended for the nucleate boiling regime.
system utilizing multiple circular or slot jets (Fig. 5(c)) is the com- Figure 6(b) shows CHF data from the Johns and Mudawar study
plexity of fluid interactions between jet impingement zones and successfully correlated by the relation
likelihood of instabilities in the flow of spent fluid.
!0:264 !2=3 0:611
Johns and Mudawar [34] examined the influence of jet q00m qf U 2 L dj qf dj
diameter, jet velocity, surface width, and subcooling on jet- ¼ 0:250
qg U hfg r qg L dj
impingement cooling of FC-72 for single circular nozzle configu-
rations. Figure 6(a) shows the influence of jet velocity on the !2=3
qf cp;f DTsub;in cp;f DTsub;in 1=3
boiling curve. Increasing jet velocity is shown increasing the heat 1 þ 0:028 1þ
transfer coefficient in the single-phase region. There is a noticea- qg hfg hfg
ble convergence of data in the nucleate boiling region regardless (6)
of flow velocity. But the velocity influence is very noticeable
again in the upper region of nucleate boiling, which is manifest by where U, L, and dj are mean jet velocity, width of impingement
Fig. 7 (a) Confined slot-jet impingement cooling modules using single jets (Mudawar and
Wadsworth [35]) and multiple jets (Meyer et al. [37]). (b) Correlation of CHF data for single FC-72
slot jets for different jet velocities, jet widths, nozzle-to-surface distances, and subcoolings
(Mudawar and Wadsworth [35]).
4 Spray Cooling
Designing a spray cooling system is by no means a straightfor-
ward endeavor. Cooling system designers are quickly confronted
with the task of configuring a system whose performance depends
on an usually large number of parameters. Aside from nozzle type
and coolant, cooling performance is influenced by droplet size,
droplet velocity, volumetric flux, and the spatial distributions of Fig. 8 (a) Microgroove and microstud surfaces used to enhance
these three parameters; in addition to spray angle, nozzle orienta- single confined slot-jet impingement cooling. (b) Variation of
tion angle, and orifice-to-surface distance [37]. Clearly the design CHF with jet velocity for smooth and enhanced surfaces (Wads-
of a spray cooling system requires a comprehensive, systematic worth and Mudawar [38]).
methodology that tackles the influences of all these parameters
with a high degree of accuracy, and which is applicable to cool-
ants with drastically different thermophysical properties.
Early research on spray cooling centered mostly on high- droplets having large surface-area-to-volume ratio, which acquire
temperature cooling in metal processing operations. Emphasis multiple trajectories and impact a broad surface area. Despite
shifted during the last three decades to the implementation of these benefits, implementation of spray cooling has been hindered
spray cooling in relatively low temperature applications, such as by relatively poor understanding of the influence of the spray
those depicted in Fig. 1. Unfortunately, the spray cooling litera- parameters on cooling performance, and the lack of long-term
ture for low temperature applications is quite sparse. Most pub- repeatability of performance because of both corrosion and ero-
lished studies point to volumetric flux Q00 as the dominant spray sion of the intricate flow passages of the spray nozzle [46].
parameter influencing cooling [38,39]. This parameter is defined Figure 9 provides a summary of fundamental mechanisms that
as the flow rate impacting an infinitesimal portion of the surface govern spray cooling. Effective droplet breakup is key to achiev-
divided by the area of the same portion. Mudawar and Valentine ing large surface-area-to-volume ratio in the form fine droplets.
[37] measured and correlated spray cooling data for all regimes of As shown in Fig. 9(a), the breakup commences with formation of
the boiling curve (single-phase cooling and nucleate, transition, unstable liquid sheets, which break up further first into ligaments
and film boiling) for water sprays. Estes and Mudawar [32] devel- and eventually fine droplets. A sufficiently large orifice-to-surface
oped an empirical CHF relationship for FC-72, FC-87, and water distance is therefore required to achieve fully developed liquid
sprays based on local volumetric flux, Q00 , and Sauter mean diam- breakup. Another fundamental issue is the spatial distributions of
eter (SMD) d32. Other noteworthy literature includes studies on key droplet parameters, such as Sauter mean droplet diameter d32,
nucleate boiling by Ghodbane and Holman [40], Holman and mean droplet velocity Um, and especially volumetric flux Q00
Kendall [41], and Rini et al. [42], and on CHF by Cho and Wu (Fig. 9(b)). The spatial variations of these parameters are compli-
[43], Chen et al. [44], and Lin and Ponnappan [45]. cated by interaction between adjacent sprays (Fig. 9(b)) which is
Spray cooling provides several important benefits, such as high often used to promote surface temperature uniformity. Cooling
heat transfer coefficients, high CHF, and superior surface temper- effectiveness is also a function of droplet impact with the surface
ature uniformity for a relatively large surface. Cooling effective- (Fig. 9(c)) which is further complicated by liquid buildup on
ness is realized by breaking the liquid into a dispersion of fine the surface.
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-9
Downloaded from https://thermalscienceapplication.asmedigitalcollection.asme.org/data/journals/jtsebv/927027/tsea_5_2_021012.pdf by guest on 03 January 2020
Fig. 9 Fundamental physics of spray cooling: (a) Droplet breakdown, distance
from orifice required to ensure fully developed breakup, and influence of nozzle
type, orifice size, and pressure drop on mass flow rate and mean droplet size and
velocity. (b) Volumetric flux distribution on impact surface and required overlap
between impact areas to ensure uniform surface temperature. (c) Impact dynamics
and heat transfer of individual droplets, liquid film buildup (dense versus light
sprays), nucleate boiling heat transfer, and critical heat flux (CHF).
Presented below is a summary of available predictive models As shown in Fig. 10, the spatial distribution of volumetric flux
and correlations for spray cooling, as well as recommendations for a spray impacting a flat surface located a distance H from the
for future research. The methodology presented here concerns orifice can be described with the aid of a point source model
pressure spray nozzles that are preferred for electronic cooling which assumes that the volumetric flux is uniform along any
applications. spherical surface centered at the orifice [48]. The uniform flux for
For full-cone spray nozzles, d32 is determined from the nozzle a spherical surface of radius H yields a volumetric flux on the
pressure drop DP and orifice diameter do based on the nozzle flow impact surface that decreases radially from the spray centerline.
Weber and Reynolds numbers according to the following relation This model yields the following distribution for local volumetric
by Estes and Mudawar [47]: flux Q00 along the surface as a function of H, radial distance r, and
h i0:259 the spray’s cone angle h:
d32 1=2
¼ 3:67 Wed0 Redo (8)
do
Q00 1 tan2 ðh=2Þ 1
where ¼
r 2 3=2 (11)
Q
00
2 1 cosðh=2Þ
qg 2DP=qf d0 1þ
Wedo ¼ (9) H
r
Fig. 11 Correlations for (a) nucleate boiling and (b) CHF for upward-facing, downward-fac-
ing, and inclined PF-5052 sprays, downward-facing FC-77 sprays, and downward-facing
water sprays (Visaria and Mudawar [52])
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-11
Downloaded from https://thermalscienceapplication.asmedigitalcollection.asme.org/data/journals/jtsebv/927027/tsea_5_2_021012.pdf by guest on 03 January 2020
Fig. 13 Hybrid cooling scheme combining jet-impingement with
mini/microchannel cooling using (a) circular jets and (b) slot jets
(Sung and Mudawar [53,54])
detailed discussion of how the function f10:30 =f2 in Eq. (13) is
determined for different orientations is provided by Visaria and
Mudawar [50]. The same reference also provides a technique for
assessing the influence of overlap. In general, increasing volumet-
ric flux for overlapping spray impact areas increases the nucleate
boiling heat transfer coefficient but not CHF.
Overall, the heat transfer models and correlations discussed
above provide a fairly complete methodology for determining the
spray’s key parameters of volumetric flux Q00 and Sauter
mean diameter d32, as well as the spatial distributions of these
parameters based on nozzle inlet temperature and pressure
Fig. 12 (a) Images of normal and inclined sprays. (b) Model of drop, orifice diameter, cone angle, orientation angle, and orifice-
inclined spray impacting a square surface. (c) Variation of spray to-surface distance. This information can be used to predict both
impact area with inclination angle (Visaria and Mudawar [52]). cooling performance in the nucleate boiling region and CHF. One
spray parameter that requires further study is the orifice-to-surface
distance required to ensure fully developed droplet breakup prior
!0:3 !0:35 to impact.
q00m qf qf Q002 d32
00 ¼ 2:3
qg hfg Q qg r
! 5 Hybrid High-Heat-Flux Cooling Schemes
qf cp;f DTsub f10:30
1 þ 0:0050 (13) As cooling demands challenge the capabilities of even the most
qg hfg f2 promising mini/microchannel, jet, and spray cooling schemes,
there is now a new focus on boosting thermal performance by
includes a multiplier f10:30=f2 that accounts for the influence of combining key attributes of these schemes in the form of “hybrid”
spray orientation on CHF; f10:30 =f2 ¼ 1 for normal sprays. cooling schemes. Figure 13 shows one type of hybrid cooling that
Several changes from the simple spray configuration depicted combines some of the advantages of both jet impingement and
in Fig. 10 have been attempted in the implementation of spray mini/microchannel cooling. Here the coolant is supplied into each
cooling. The two most relevant are the use of spray orientations of a number of parallel mini/microchannels either gradually, in
other than normal, and sprays with overlapping impact areas. the form of small circular jets, or as a single slot jet [51,52]. There
Inclined sprays have been proposed for electronics cooling to are many possible variations of this hybrid configuration. For
reduce the height of the cooling package by reducing the orifice- example, the jets supplying coolant to each channel could have
to-surface distance as depicted in Fig. 12(a). Figure 12(b) shows a equal diameters, or diameters that either increase or decrease from
schematic of an inclined spray whose orifice-to-surface distance is the centerline [53]. Supplying the coolant gradually with these
adjusted such that the major axis of the impact area just inscribes hybrid schemes greatly decreases temperature gradients along the
a square surface in order to maximize CHF. Figure 12(c) shows channels compared to conventional mini/microchannels, where
the variation of spray impact area with orientation angle a. A the coolant is supplied from one end of the channel to the other
end. The gradual introduction of coolant suppresses the growth of dictive tools for cooling system design. Key findings from the
void fraction along the channel, and therefore helps achieve study are as follows:
higher CHF. The hybrid configurations also control the flow of
(1) Mini/microchannel cooling has received considerable atten-
spent jet fluid in individual channels, compared to uncontrolled
tion in recent years, resulted in fairly comprehensive methods
flow in between impingement zones when using multiple conven-
to predicting pressure drop, heat transfer coefficient, and
tional jets. Using HFE 7100 as coolant, this hybrid scheme proved
CHF. Additional work is needed to predict conditions that
capable of achieving an unprecedented heat flux of 1127 W/cm2
trigger premature CHF and those that preclude its occurrence.
without incurring CHF [53].
(2) Despite the attention two-phase jet-impingement cooling
Figure 14 shows another example of hybrid cooling that com-
has received in recent years, no reliable correlations are
bines benefits of (a) microchannel flow, (b) jet impingement, (c)
available for the nucleate boiling region. There are suffi-
highly subcooled boiling, and (d) surface enhancement to greatly
cient tools to predict CHF for both confined circular jets
increase CHF [54]. This configuration consists of a series of large
and confined slot jets.
studs, each attached to a heat dissipating device, with the coolant
(3) A number of models and correlations are available that are
forced to first impinge along the front of the stud and afterwards
crucial for spray cooling system design. These include tools
flow in two parallel circumferential microchannels with the aid of
to predict Sauter mean diameter d32 and the spatial distribu-
a specially grooved cover plate. The stud itself combines three
tion of volumetric flux Q00 , based on nozzle pressure drop,
levels of surface enhancement: (a) centimeter-scale extended stud
orifice diameter, and cone angle. These parameters are then
that contributes an appreciable increase in surface area in addition
incorporated into available correlations to predict both nu-
to the aforementioned control of coolant flow, (b) millimeter-scale
cleate boiling and CHF. Predictive tools are also available
grooves that contribute a further increase in surface area in
for the effects of nozzle orientation and overlap between
addition to providing sharp corners that are highly favorable for
spray impact areas when using multiple sprays. Further
bubble nucleation, and (c) microsurface texture (achieved by solid
research is needed to investigate droplet breakup mecha-
particle blasting or vapor blasting) to provide an abundance of
nisms and determine the orifice-to-surface distance required
surface cavities for efficient bubble nucleation. This hybrid
to ensure fully developed breakup.
cooling technique is an example of how standard cooling schemes
(4) Overall, the design of cooling systems utilizing mini/micro-
can be combined advantageously to achieve superior cooling
channel, jet, or spray cooling schemes can benefit from
performance.
consolidation of world databases for different coolants and
broad ranges of operating conditions. Such efforts provide
a systematic basis for recommending any necessary future
6 Conclusions experiments and developing reliable “universal” correla-
This study explored the implementation of two-phase mini/ tions for performance parameters of interest.
microchannel, jet, and spray cooling schemes into practical cool- (5) The design of cooling systems can benefit from the devel-
ing packages, and the advantages and disadvantages of each opment of efficient software tools and robust algorithms
scheme. The key focus of the study is assessment of available pre- that are based on available models and correlations.
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-13
(6) Use of hybrid cooling schemes that combine the merits of c, c0 ¼ angles used in uniform point source spray model
mini/microchannel, jet, spray and other cooling schemes is h¼ spray cone angle
a very effective means to achieving cooling performances l¼ viscosity
unattainable with the individual cooling schemes. P¼ dimensionless group
q¼ density
Nomenclature r¼ surface tension
rc ¼ contraction parameter
A ¼ area measured along sprayed surface re ¼ expansion parameter
A0 ¼ area measured along spherical surface centered at orifice /¼ two-phase multiplier
of spray nozzle
Bo* ¼ modified boiling number
C ¼ empirical constant
Subscripts
cp ¼ specific heat at constant pressure A¼ acceleration
D ¼ tube diameter c¼ contraction
Dh ¼ hydraulic diameter ch ¼ channel
Journal of Thermal Science and Engineering Applications JUNE 2013, Vol. 5 / 021012-15