Experimental Studies to Evaluate the Use of Metal Foams in Highly Compact Air-Cooling Heat Exchangers

Purdue University

Purdue e-Pubs

International Refrigeration and Air Conditioning

Conference

School of Mechanical Engineering

2010

Experimental Studies to Evaluate the Use of Metal

Foams in Highly Compact Air-Cooling Heat

Exchangers

Kashif Nawaz

University of Illinois at Urbana Champaign

Jassie Bock

University of Illinois at Urbana Champaign

Zhengshu Dai

Institute of Reigeration and Cryogenics

Anthony M. Jacobi

University of Illinois at Urbana Champaign

Follow this and additional works at: h>p://docs.lib.purdue.edu/iracc

=is document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact epubs@purdue.edu for

additional information.

Complete proceedings may be acquired in print and on CD-ROM directly from the Ray W. Herrick Laboratories at h>ps://engineering.purdue.edu/

Herrick/Events/orderlit.html

Nawaz, Kashif; Bock, Jassie; Dai, Zhengshu; and Jacobi, Anthony M., "Experimental Studies to Evaluate the Use of Metal Foams in

Highly Compact Air-Cooling Heat Exchangers" (2010). International Reigeration and Air Conditioning Conference. Paper 1150.

h>p://docs.lib.purdue.edu/iracc/1150

2502, Page 1

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

Experimental Studies to Evaluate the Use of Metal Foams in Highly Compact

Air-Cooling Heat Exchangers

K. Nawaz1, J. Bock1, Z. Dai2, and A. Jacobi1

1Department of Mechanical Science and Engineering, University of Illinois, Urbana,

Illinois, U.S.A.

2Institute of Refrigeration and Cryogenics, Zhejiang University, Hangzhou, Zhejiang,

China

Abstract:

open-cell aluminum foam is considered as a highly compact replacement for

conventional fins in brazed aluminum heat exchangers. The experimental data needed

to for evaluation are obtained through wind-tunnel experiments. Using a closed-loop

wind tunnel, heat transfer and pressure drop measurements are undertaken, and the

results are characterized as Colburn j-factor and friction factor for geometric

configurations of foam and flat tubes that mimic current multi-louver, microchannel

geometries. The data obtained in the wind-tunnel are used to evaluate the potential for

using metal foam (of varying porosity) as a replacement to conventional fins for a

range of configurations and operating conditions typical for air-cooling applications.

Finally, we comment on some of the challenges that must be met for the adoption of

this technology in air-cooling systems.

Introduction:

Metal foams are porous media with low density and novel thermal, mechanical,

electrical, and acoustic properties [1]. They can be categorized as open-cell or closed-

cell foams, but only open-cell metal foams appear to have promise for constructing

heat exchangers. Open-cell metal foams have high specific surface area, relatively

high thermal conductivity, and a tortuous flow path to promote mixing. Metal foam

have been studies by a number of researchers for thermal applications; some were

focused on metal-foam heat exchangers (and heat sinks), and many others

investigated the basic thermal transport properties of metal foams. The basic

properties of the metal foams include the effective thermal conductivity, permeability,

and inertial coefficient. Calmidi and Mahajan [2] investigated the effective thermal

conductivity of high-porosity fibrous metal foams experimentally. An empirical

correlation was developed and a theoretical model was derived. The model

predictions agreed closely with the experimental data and were used for the evaluation

of metal foams as possible candidates for heat sinks in electronics cooling

applications. Boomsma and Poulikakos [3] see also [4] developed a one-dimensional

heat conduction model for use with open-cell metal foams, based on idealized three-

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

dimensional cell geometry of the foam. Their model showed that the fluid-phase

conductivity has a relatively small effect on the effective thermal conductivity, and

the overall effective thermal conductivity of the metal foam is controlled by the solid-

phase conductivity to a large extent. Bhattacharya et al. [5] conducted research on the

determination of the effective thermal conductivity, permeability, and inertial

coefficient of highly porous metal foams. A theoretical model was formulated and the

analysis showed that the effective thermal conductivity depends strongly on the

porosity and the ratio of the cross-sections of the fiber and the intersection, but no

systematic dependence on pore density was found. Fluid flow experiments were

conducted and the results showed that permeability increases with pore diameter and

porosity of the medium, and the inertial coefficient depends only on porosity. They

proposed a theoretical model for predicting inertial coefficient and a modified

permeability model; the models were shown to agree with experimental results.

Tadrist et al. [6] discussed the characteristics of randomly stacked fibers and metallic

foams and analyzed the transport properties for both materials.

Convection in porous media has been widely investigated, but most studies

focused on packed beds and granular materials with low porosities in the range 0.3-

0.6. The porosity of open-cell metal foams is much higher (Ȝ>0.90), and only during

the past decade has convection in high-porosity metal foams started to receive

attention. Calmidi and Mahajan [7] investigated forced convection in high-porosity

metal foams experimentally and numerically. Experimental results showed that the

transport enhancing effect of thermal dispersion is extremely low with foam-air

combinations, but for foam-water combinations it can be very high. In the numerical

study, a thermal non-equilibrium model was used and a Nusselt number correlation

was determined. Zhao et al. [8] studied natural convection and its effect on overall

heat transfer in highly porous open-cell FeCrAlY foams experimentally and

numerically. Experimental results showed that natural convection is significant in

metal foams due to the high porosity and inter-connected open cells. Numerical

calculations showed that the so-called non-equilibrium effect (the metal and fluid

being at different temperatures) cannot be neglected and hence a two-equation energy

model should be used instead of one-equation model for convection in metal foams.

Hetsroni et al. [9] studied natural convection heat transfer in metal foam strips with

internal heat generation by experiments. Infrared images on both the surface and the

inner region of the metal foam were analyzed, and the non-equilibrium temperature

distribution was estimated. The result indicated that the non-equilibrium effect is

significant.

Some studies have focused on metal-foam convective heat transfer devices.

Boomsma et al. [10] studied an open-cell aluminum foam heat sink for electronics

cooling applications. They found that compressed aluminum foams performed well,

offering a significant improvement in the efficiency over several commercially

available heat exchangers. They also found the metal foam can decrease the thermal

resistance to nearly half that of currently used heat exchangers in the same

application. Zhao et al. [11] and Lu et al. [12] analyzed forced convection heat

transfer performance in high-porosity, open-cell, metal-foam-filled heat exchanger

2502, Page 2

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

tubes and metal-foam-filled pipes using the Brinkman-extended Darcy momentum

model and the two-equation heat transfer model for porous media. The results showed

that, compared to conventional, finned-tube heat exchangers, the heat exchangers with

metal-foam-filled tubes have better heat transfer performance, and the metal-foam-

filled pipes have much better thermal performance than a plain tube, but at the

expense of higher pressure drop. Mahjoob and Vafai [13] have discussed the effects

of micro-structural metal foam properties on heat exchanger performance, and they

categorized and investigated the extant correlations for flow and thermal transport in

metal-foam heat exchangers. Tube and channel metal-foam heat exchangers were

used to evaluate thermal-hydraulic performance, and the results showed a

considerable improvement in performance by inserting the metal foam. Ejlali et al.

[14] numerically investigated the fluid flow and heat transfer of an air-cooled metal-

foam heat sink under a high speed laminar jet confined by two parallel walls at

Reynolds numbers from 600 to 1000. They compared the performance of the metal-

foam heat sink to that of conventional finned design and found that the heat removal

rate can be greatly improved without additional cost. Dai et al. [15] presented the

comparison of metal-foam heat exchangers to louver-fin heat exchangers based on

theȜ-N

method; the results showed that with the same thermal-hydraulic

performance, the metal-foam heat exchanger can be lighter and smaller, but much

more expensive.

As noted above, there are numerous studies of material properties and

transport phenomena, and fewer studies of metal-foam heat sinks. However, thermal-

hydraulic data which can be used to evaluate the use of metal foams in air-cooling

heat exchangers are currently scanty in the literature. Previous evaluations of this

technology have been limited by the dearth of experimental support for the

evaluation. In this study, a closed-loop wind tunnel is used to get the needed thermal-

hydraulic data for a metal-foam heat exchanger under dry conditions. The

measurements are compared to the results calculated by some existing correlations

from the open literature. Finally, some comments on some of the challenges that must

be met for the adoption of this technology in air-cooling systems are made.

Experimental Setup:

A closed-loop wind tunnel is used to assess the thermal-hydraulic performance of a

heat exchanger having metal foam as the fin material. As shown in Figure 1, air

downstream of test section passes through a set of electric strip heaters, past a steam

injection pipe, through an axial blower and another set of strip heaters, through a flow

nozzle, a mixing chamber, a flow conditioning section, a flow contraction, and the test

section, completing the loop. A variac controller is used to maintain the desired

upstream air temperature and dew point at steady state. Steam is generated by an

electric humidifier. The air temperature is measured using thermopile grids,

constructed using T-type thermocouples. Chilled-mirror hygrometers are used to

2502, Page 3

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

measure the upstream and downstream dew points. The cross-sectional flow area in

the test section is rectangular—30 cm wide and 20 cm high. An axial blower provides

an air flow with face velocities at the test section from 0.5 to 5 m/s. An ASME flow

nozzle, with a micro manometer, is used to measure air mass flow rate. Another micro

manometer is used to measure air-side pressure drop across the test section. A single-

phase liquid, an aqueous solution of ethylene glycol (DOWTHERM 4000), is used as

the tube-side heat transfer fluid. A chiller system with a commercial heat pump, two

large coolant reservoirs, a PID-controlled electric heater, and a gear pump supplies the

flow. The chiller system provides a coolant flow with a steady inlet temperature

(within 0.1°C) at a capacity up to 20 kW. Coolant inlet and outlet temperatures are

measured using thermocouples imbedded through supply tubes. Coolant flow mixing

devices are installed immediately upstream of the thermocouples to provide a well

mixed flow and a uniform coolant temperature. A Coriolis-effect flow meter located

in the downstream coolant pipe is used to measure mass flow rate. The significant

experimental uncertainties involved in the dry and wet wind-tunnel experiments are

listed in Table 1.

Table 1: Uncertainty in measurement for various parameters

Parameter

Uncertainty

Air temperature

r 0.1

Coolant temperature

r 0.1

Nozzle discharge coefficient

r 2%

Core pressure drop

r 0.17Pa

Nozzle pressure

0.17Pa

Coolant mass flow rate

0.1%

Dew point Temperature

0.1%

Fig. 1: Closed loop wind

tunnel

2502, Page 4

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

Before beginning wind-tunnel experiments, the heat exchanger specimens are

insulated using foam insulation tape. As the specimen has face dimensions different

from those of the test section, it is necessary to install within the tunnel an additional

flow contraction upstream and a diffuser downstream of the test specimen. The

specimen is mounted in the test section, the coolant hoses connected, and the gaps

between the specimen and the test section sealed with adhesive tape. The entire wind

tunnel, the test specimen, steam pipes, and coolant pipes are insulated to isolate the

system as much as possible from the environment. The thermal conductivity of the

insulation material is 0.03 W/m-K.

Sample Specifications:

The sample used for the experiments is a flat aluminum tube cross-flow heat

exchanger, aluminum foam is sandwiched between tubes as fins. Each tube ends in a

plastic manifold which distributes and collects coolant from flat tubes.

Fig. 2: Metal foam heat exchanger test

specimen

The sample consists of ten metal foam layers (length 200 mm, width 15mm and depth 15

mm) sandwiched between flat aluminum tubes (length 304.8 mm, width 25.4 mm height

3.2 mm and wall thickness 0.5 mm). The face area of the heat exchanger is

200 u 200m

.Transparent manifolds connect these flat tubes. All the connections are sealed checked to

avoid leakage problems. Important characteristic of the metal foam are summarized in the

following.

Table 2 Characteristics of the metal foam sample

orosity

)

(mm)

)

(×10

)

effective

/m k)

0.9272

0.25

3.13

0.097

1.2

4.10

2502, Page 5

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

Calculated(Calmidi)

Measured

heat transer coefficient (W/m^2-K)

In order to join the metal foam to the aluminum base plates Arctic Silver 5 High-

Density Polysynthetic Silver Thermal Compound was used, having a thermal

conductivity of approximately 5 W/m-K.

Results and discussion:

Figure (3) and (4) show the results obtained during the experimentation. The heat

transfer coefficient based on the total surface area of the metal-foam fin and the base

plate is plotted against the Reynold’s number based on the ligament diameter.

(1)







ℎ









ℎ







(2)

 

=  × ∆



(3)

 





(4)

∆



ℎ ,

−



, 

−(

ℎ

,

−

 ,

)

ln ⁡

ℎ , 

−



, 

(

ℎ

,

−



, 

)

(5)

F is the correction factor for a single pass, cross flow heat exchangers[19].Fin

efficiency is calculated based on the adiabatic tip condition.

Fig. 3: Heat transfer coefficient (measured and calculated Calmidi and Mahajan, 2000)

400

350

300

250

200

150

10 20 30 40 50 60 70 80 90

Reynold's No

2502, Page 6

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

(measured)

(Calculated)

Pressure drop per unit length(Pa/m)

To compare the results, equation (2) provided by Calmidi and Mahajan[7] is used.

The calculation performed using the relationship

0.52

0.5

0.37

(6)

The experimental data are compared to the model of Calmidi and Mahajan, 2000 [7]

in figure (3). The results show surprisingly good agreement and buttress both the data

and the model. The relationship(6) showed that the values differ less than 5% at low

Reynolds numbers(0.6 m/s) but the difference approaches 10 % at the highest

velocity (4.3 m/s)

Shown in the figure (4) is the pressure drop The pressure drop for metal foam is

assumed to be higher as compared to some other compact heat exchangers and similar

behavior was achieved during the experiments. The fanning friction factor for metal

foam with a face velocity of 3m/s is 0.15 which is 50% more than for multi- louver fin

heat exchanger under the same flow conditions. The results are compared with

equation (3) provided byBhattacharya et al.[5]





(7)

dx K

Fig. 4: Pressure drop per unit length (measured and calculated Bhattacharya et al. [5])

7000

6000

5000

4000

3000

2000

1000

10 20 30 40 50 60 70 80 90

Reynold's No

The relationship is widely used to estimate the pressure drop through porous media,

but the experimental results show that the difference varies from 76% (213 Pa/m

calculated 49 Pa/m measured) at low face velocity to 17 % (5629 Pa/m calculated

6739 Pa/m measured) at high face velocity. The uncertainty in the measuring device

for pressure difference is

2502, Page 7

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

Energy

Balance

Uncertainty Analysis:

The uncertainties in various instruments were reported in Table 1. While conducding

the wind tunnel experiments the first critical step is to achieve the energy balance on

the air side and on the coolant side. Such a balance supports the veracity of the data..

Even with significant care, these problems are unavoidable and resulting a slight heat

transfer mismatch on air and coolant sides. The problems become rather severe at

high face velocities. If the energy balance is defined by the following relationship,

then its variation with Reynolds No based on the ligament diameter is given as

follows (equation (8)).

b a la n ce





a ir co o la n t

a vg

(8)

Fig 5: Energy balance against Reynold’s no.

10 20 30 40 50 60 70 80 90

Reynold

The trend in the energy balance show that as the air mass flow rate is increased, the

problems like air leakage become more significant as compared to they were at low

air mass flow rates. The uncertainty for the heat transfer coefficient varies from 5.7

% at lowest face velocity (0.6 m/s) to 13.7 % at the highest face velocity (4.3 m/s).

2502, Page 8

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

Future work and challenges:

Current results are based on the experiments performed under dry conditions for just

one specimen, which is actually a drop-in-replacement configuration for some current

heat exchangers (louver fins are replaced with metal foam). In order to fully exploit

the performance-enhancing characteristics of metal foams, further experimental work

is required. For that purpose new samples with different configurations and having

different foam properties need to be tested. Bonding the metal-foam fins to the base

surface is perhaps the most challenging task while manufacturing and also the most

costly process if one wants the ideal thermal joint (soldering or brazing). Thermal

epoxies can also work well if they have good thermal conductivity, but the

performance of the brazed metal foam should be tested along with the specimen

employing thermal epoxy as adhesive. Furthermore the performance of the materials

should be tested under wet conditions, where there is condensation on the surface, to

see the water drainage behavior during operational conditions.

Conclusion:

Metal foams are novel materials which can perform well if they are used in compact

heat exchangers. For the two parameters, heat transfer performance and pressure drop,

one has to come up with some ideal conditions in terms of geometry and flow

conditions to get the best performance. There are correlations available to calculate

the heat transfer coefficient and pressure drop. The relationship for the heat transfer

coefficient by Calmidi and Mahajan [2] is accurate enough as the experimental values

are close enough. For the pressure drop performance the relationship by Bhattacharya

[5] work well at high velocities.

Nomenclature:

ff inertial coefficient





fin efficiency

h convective heat transfer coefficient, W/m

prendtl No

K permeability, m

k thermal conductivity of the fin, W/m.K

velocity, m/s

ȣ dynamic viscosity, N/ms

f fiber diameter,m

dp pore diameter,m

Ȩ



density, kg/m

2502, Page 9

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

Subscript:

a air side

c coolant side

m mean value

h,i

hot inlet

h,o

hot outlet

c,i cold inlet

c,o

cold outlet

REFERENCES

[1] Lu, T., Ultralight porous metals: from fundamentals to applications, Acta Mechanica

Sinica, Chinese J. Mech, vol.18, pp.457-479, 2002.

[2] Calmidi, V.V, and Mahajan, R.L., The effective thermal conductivity of high porosity

fibrous metal foams, J. Heat Transfer, vol. 121, pp. 466-471, 1999.

[3] Boomsma, K., and Poulikakos, D., On the effective thermal conductivity of a three-

dimensionally structured fluid-saturated metal foam, Int. J. Heat Mass Transfer, vol.44,

pp.827-836, 2001.

[4] Dai, Z., Nawaz, K., Park, Y., Bock, J., and Jacobi, A.M., Correcting and extending

the Boomsma-Poulikakos effective thermal conductivity model for three-dimensional,

fluid-saturated metal foams, Int. Comm. Heat Mass Trans., in press.

[5] Bhattacharya, A., Calmidi, V.V., and Mahajan, R.L., Thermophysical properties of

high porosity metal foam, Int. J. Heat Mass Transfer, vol. 45, pp.1017-1031,2002.

[6] Tadrist, L., Miscevic, M., Rahli, O., and Topin, F., About the use of fibrous materials

in compact heat exchangers, Experimental Thermal and Fluid Science,vol.28, pp. 193-

199, 2004.

[7] Calmidi, V.V., and Mahajan, R.L., Forced convection in high porosity metal foams, J.

Heat Transfer, vol.122, pp.557–565, 2000.

[8] Zhao, C.Y., Lu, T.J., and Hodson, H.P., Natural convection in metal foams with open

cells, Int. J. Heat Mass Transfer, vol.48, pp.2452-2463, 2005.

[9] Hetsroni, G., Gurevich, M, and Rozenblit, R., Natural convection in metal foam strips

with internal heat generation, Experimental Thermal and Fluid Science, vol. 32, pp.

1740-1747, 2008.

2502, Page 10

International Refrigeration and Air Conditioning Conference at Purdue, July 12-15, 2010

[10] Boomsma, K., Poulikakos, D., and Zwick, F., Metal foams as compact high

performance heat exchangers, Mechanics of Materials, vol.35, pp.1161-1176, 2003.

[11] Zhao, C.Y., Lu, W, and Tassou, S.A., Thermal analysis on metal-foam filled heat

exchangers. Part

ϩ: Tube heat exchangers, Int. J. Heat Mass Transfer, vol.49, pp.2762-

2770, 2006

[12] Lu, W., Zhao, C.Y., and Tassou, S.A., Thermal analysis on metal-foam filled heat

exchangers. Part

Ϩ: Metal-foam filled pipes, Int. J. Heat Mass Transfer, vol.49, pp.2751-

2761, 2006.

[13] Mahjoob, S., and Vafai, K., A synthesis of fluid and thermal transport models for

metal foam heat exchangers, Int. J. Heat Mass Transfer, vol.51, pp.3701-3711, 2008.

[14] Ejlali, A., Ejlali, A., Hooman, K., and Gurgenci, H., Application of high porosity

metal foams as air-cooled heat exchangers to high heat load removal systems, Int.

Commun Heat Transf, vol. 36, pp. 674-679, 2009.

[15] Dai, Z., Nawaz, K., Park, Y., Chen, Q.,and Jacobi, A.M., A comparison of metal-

foam heat exchangers to compact multi-louver designs for air-side heat transfer

applications, Proceedings of the Seventh International Conference on Enhanced,

Compact and Ultra-Compact Heat Exchangers: From Microscale Phenomena to

Industrial Applications, San Jose, Costa Rica, 2009, pp. 49–57.

[16] Park, Y., and Jacobi, A. M., The air-side thermal-hydraulic performance of flat-tube

heat exchangers with louvered, wavy, and plain fins under dry and wet conditions, J.

Heat Transfer, vol. 131, 061801, 2009.

[17] Chang, Y.J. and Wang, C.C., A generalized heat transfer correlation for louver fin

geometry, Int. J. Heat and Mass Transfer, vol. 40, pp.533-544, 1997.

[18] Kim, S.Y., Paek, J.W., and Kang, B.H., Flow and heat transfer correlations for

porous fin in a plate-fin heat exchanger, J. Heat Transfer, vol. 122, pp.572-578, 2000.

[19] Incropera, F.P., and DeWitt, D. P., Fundamentals of heat and mass transfer, 4thed.,

John Wiley & Sons, Inc, 1996

[20] Park, Y., Jacobi, A. M., Air-side heat transfer and friction correlations for flat-tube

louver-fin heat exchangers, J. Heat Transfer, vol. 131, 021801, 2009.

2502, Page 11