otechnique 56, No. 2, 81122 Brandl, H. (2006).

Ge

Energy foundations and other thermo-active ground structures

H. BRANDL* Energy foundations and other thermo-active ground structures, energy wells, and pavement heating represent an innovative technology that contributes to environmental protection and provides substantial long-term cost savings and minimised maintenance. The paper focuses on earth-contact concrete elements that are already required for structural reasons, but which simultaneously work as heat exchangers. Absorber pipes lled with a heat carrier uid are installed within conventional structural elements (piles, barrettes, diaphragm walls, basement slabs or walls, tunnel linings), forming the primary circuit of a geothermal energy system. The natural ground temperature is used as a heat source in winter and for cooling in summer. Hence no additional elements have to be installed below surface. The primary circuit is then connected via a heat pump to a secondary circuit within the building. Free cooling may even run without a heat pump. The paper describes heat transfer in the ground, and between absorber uid and concrete/soil. Temperature-induced changes of soil properties or of foundation behaviour are also discussed, and recommendations for design and operation are given. Pilot research projects and case histories bridge the gap between theory and practice, and special applications reveal the wide eld of geothermal geotechnics.
KEYWORDS: basements; case history; design; diaphragm and in situ walls; environmental engineering; footings/foundations; piles; soil/structure interaction; temperature effects; time dependence

nerge tiques et autres structures de sol Les fondations e nergie et le chauffage des thermoactives, les puits de sentent une technologie innovante qui controttoirs, repre ` la protection de lenvironnement et permet des tribue a conomies de cou ` long terme tout en ts substantielles a e duisant la maintenance. Cet expose e tudie les e le ments re ton au contact du sol qui sont ne cessaires pour des de be me temps raisons structurales mais qui servent en me changeurs de chaleur. Des tuyaux absorbeurs remplis de s dans des dun liquide conducteur de chaleur sont installe le ments structuraux conventionnels (piles, barrettes, rie deaux souterrains, dalles ou murs de sous-sol, doublures ` me de nerde tunnels) formant le circuit primaire du syste othermique. La tempe rature naturelle du sol est gie ge e comme source de chaleur en hiver et source de utilise te . Il nest donc pas ne cessaire dinrefroidissement en e le ments sous la surface. Le circuit pristaller dautres e par linterme diaire dune pompe maire est alors connecte ` chaleur a ` un second circuit a ` linte rieur du ba timent. Ce a me fonctionner sans refroidissement gratuit peut me ` chaleur. Cet expose de crit le transfert de chaleur pompe a ton. dans le sol et entre le liquide absorbeur et le sol/be galement les changements dus a ` la temNous analysons e rature des proprie te s du sol ou des comportements des pe ` fondations et nous faisons des recommandations quant a ` lexploitation. Des projets de recherche la conception et a pilotes et des histoires de cas comblent le vide entre la orie et la pratique et des applications spe ciales re ve ` lent the tendue du domaine de la ge otechnique ge othermique. le

INTRODUCTION Subsurface geothermal resources represent a great potential of directly usable energy, especially in connection with (deep) foundations and heat pumps. Environmental and economical aspects were the incentive for the invention of the heat pump by the Austrian mining engineer Peter Ritter von Rittinger in 1855. Already two years later he could prove an annual saving of 293 000 m3 rewood if the heat pumps were installed in all Austrian saltworks. Geothermal energy can also be obtained by means of at collectors, trench collectors, or borehole heat exchangers. These systems have been widely used for many years in Austria (Fig. 1). Since the beginning of the 1980s, geothermal energy has also been increasingly obtained from foundation elements in Austria and Switzerland: at rst from base slabs, then from piles (1984) and diaphragm walls (1996). This innovation makes use of the high thermal storage capacity of concrete. Moreover, these concrete members are already required for structural reasons, and need not be installed as additional
Manuscript received 26 April 2005; revised manuscript accepted 27 October 2005. Discussion on this paper closes on 1 August 2006, for further details see p. ii. * Institute for Soil Mechanics and Geotechnical Engineering, Vienna University of Technology, Austria.

Earth collector

Energy piles

Fig. 1. Scheme for heating/cooling a small one-family house with energy foundations. Also indicated are additional thermoactive ground-source systems

elements like conventional thermal energy utilisation systems. Fig. 2 shows the increasing use of energy piles in Austria since the year 1984. With combined geothermal cooling/heating systems heat energy is fed into and withdrawn from the ground via energy foundations or other thermo-active ground structures. This innovative method is signicantly more cost-effective 81

82
6000

BRANDL
26 000
5511 22843

24 000 22 000 20 000

5000 Energy piles per year Energy piles (cumulative)

12956 11449 9085 9961 1842 1488 1507 970 876 2534

Energy piles per year

4000

17332 14798

18 000 16 000 14 000 12 000 10 000 8000 6000

3000

2000
5511 6486 975 1003 961 4741 770 3780 571 2777 2206

8268 7298

1000
6 17 45 88

664

701 685 1521

812

817

4000 2000

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Fig. 2. Number of energy piles installed in Austria

than conventional systems, and it is environmentally friendly because it uses clean, renewable energy. In Austria nearly 300 buildings are tted with energy piles or energy diaphragm walls. The number of foundation slabs, retaining walls, basement walls, garage walls etc. used for geothermal heating or cooling is of similar magnitude. Furthermore, this technology and tting techniques have been exported to several countries, even to tropical regions. Conventional ground heat exchangers consist of one or more U-shaped loops of plastic tubes (absorber pipes) inserted into a borehole. The heat transfer from the surrounding soil/rock to the heat carrier uid (or vice versa) takes place via the absorber pipes and the groundwater or material that lls the borehole. The heat transfer of such a closed system is consequently not as efcient as for an open system that consists of a single plastic tube, through which the uid is transported from/to the bottom of the borehole. Energy foundations represent a closed system too, but owing to the good thermal properties of concrete the heat transfer is signicantly higher than in boreholes. Furthermore, thin pipes are used, installed as loops, whereby the bending radius of the absorber pipes in the bottom zone is larger than in the case of conventional heat extraction/ storage boreholes. Thus the ow resistance of the heat carrier uid is smaller, and consequently the operation costs are lower. Energy foundations contain closed coils of plastic piping through which a heat carrier uid is pumped that exchanges energy from a building with the ground. The same principle is used for other thermo-active elements such as energy walls (retaining walls, basement walls), energy wells and energy tunnels. The essential difference from conventional earth-collector systems or ground heat exchanger boreholes is that the earth-contact concrete elements that serve as heat exchangers are already required for structural reasons and need not be constructed separately. Furthermore, concrete has a higher thermal conductivity than soil. Energy wells are wells temporarily used for groundwater lowering and/or groundwater recharging but simultaneously adapted for energy extraction/storage purposes. The energy systems therefore have a double function, and they work most efciently if the thermo-active elements are in contact with mobile groundwater in the case of heating or cooling only. However, for seasonal operation (i.e. heating in winter and cooling in summer) rather steady groundwater conditions with a low hydraulic gradient are favourable for

seasonal energy extraction and feeding (recharging, storage). Nevertheless, a sufcient seasonal performance factor of the system is available even without closed groundwater and, moreover, the use of renewable clean energy favours environmental protection. Thermo-active foundations, tunnels etc. use earth-contact structural elements with closed circuits. In contrast, open systems use water from an aquifer, which is pressed through a heat exchanger or heat pump. These are simpler but hardly used in Austria because of operational problems such as clogging or bio-fouling in the wells and heat exchangers. Clogging may occur by precipitation of dissolved minerals caused by temperature changes and precipitation of ironmanganese hydroxides. It increases with temperature variations in the aquifer and with air entering the wells or pipework. The latter can be avoided by operating the system with a slight overpressure. Furthermore, such wells need submersible pumps that can be lifted for maintenance. PRINCIPLES OF GEOTHERMAL UTILISATION OF FOUNDATIONS (ENERGY FOUNDATIONS) Energy foundations may comprise base slabs, piles, barrettes, slurry trench systems (single elements or continuous diaphragm walls) and concrete or grouted stone columns (energy columns). Combinations with near-surface earth collectors or retaining structures are also possible. Energy foundations can be used for heating and/or cooling buildings of all sizes, as well as for road pavements, bridge decks etc. Concrete has a good thermal conductivity and thermal storage capacity, which makes it an ideal medium as an energy absorber (heat exchanger). To use these properties for energy foundations, high-density polyethylene plastic pipes of 20 or 25 mm diameter, with 2.0 or 2.3 mm wall thickness respectively, have to be installed within the concrete. They are placed to form several individual closed coils or loops, which circulate a heat carrier uid (heat transfer medium) of either water, water with antifreeze (mainly glycol), or a saline solution. The plastic piping can be xed to the reinforcement cages of the energy foundation in a plant or on the site (Figs 3 to 5). The latter is more common, whereby the piping is delivered to site on reels, and a special working area is needed. At the start (uid inow) and return end (outow) of the pipework in each pile, diaphragm wall panel or other individual foundation element a locking valve and a man-

2004

820

Energy piles (cumulative)

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

83

Fig. 5. Absorber pipes attached to the reinforcement cage of a diaphragm wall (see Fig. 53) Fig. 3. Absorber pipes tted to the reinforcement cage of a large-diameter bored energy pile. The connecting ends are protected by a tube at the pile head

Fig. 6. Installation of absorber pipes (lled with heat-carrier uid) on the subconcrete of a piled raft foundation

Fig. 4. Reinforcement with attached absorber pipes inserted within casing of an auger pile; ready for casting concrete

ometer are xed. This allows the pipe circuit to be pressurised to (normally) 8 bar for integrity check and to resist the head of the wet concrete without collapsing. This pressure is typically maintained until after the foundation concrete is a few days old, and is again applied before the entire primary circuit is denitely closed. Figure 6 gives a partial view of the laying system of absorber pipes on the subconcrete of a piled raft foundation.

Connected to this primary circuit is the secondary circuit within the building where the thermal energy is distributed. Until recently prefabricated driven piles of reinforced concrete with integrated heat exchanger (absorber pipes) represented the majority of energy piles. But the percentage of (large-diameter) bored piles has been steadily increasing since the year 2000. Pile excavation may be supported by casing or uid. Continuous ight auger piling and plunging of the heat exchanger with the reinforcement cage into wet pile concrete may affect the nal integrity of the plastic absorber pipes. The risk of absorber pipe damage can be lowered by a stiff reinforcement cagethat is, by welding the helical reinforcement to the vertical reinforcement bars. Merely connecting with wires would allow excessive deformations of the reinforcement cage during lifting and inserting into the pile concrete. Consequently, the rotary bored technique or excavation by grab should be preferred for energy piles, and even in this case a relatively stiff reinforcement cage is recommended. An increasingly used alternative to driven piles of prefabricated reinforced concrete is ductile cast iron piles with integrated heat exchangers. Contrary to grey cast iron with lamellar graphite, ductile cast iron contains spherical graphite. It therefore exhibits higher stresses at failure and a ductile behaviour. The piles consist of 5 m long standard elements that can easily be assembled to longer sections during the driving procedure. The tubes are lled under pressure with concrete. Shaft grouting to increase friction is also possible. Since 1985 more than 1 million metres of

84

BRANDL

ductile cast iron piles have been installed in Austria; at present about 130 000 m are driven every year, with an increasing proportion of energy piles. The heat exchangers are inserted into the fresh concrete, and have to be secured against uplift until the concrete has sufciently hardened. The standard diameter of such driven piles is d 170 mm, but this can be increased signicantly by shaft grouting. Nevertheless, the geothermal effectiveness of such thin energy piles is smaller than that of driven precast concrete piles or large-diameter bored piles, despite the high thermal conductivity of cast iron. The small diameter enables the installation of only one pipe loop and no coiled piping (similar to Fig. 6). Moreover, the contact area with the ground is relatively small. In soft soils, buckling of the piles also has to be considered. Other alternatives for energy piles are (a) driven steel tube piles lled with concrete and inserted heat exchangers (absorber pipes) (Fig. 7) (b) vibrated concrete columns installed after the vibrootation technique and then tted with absorber pipes. Partially grouted vibro stone columns can also be used as energy columns but have a lower geothermal efciency. The primary circuit contains closed pipework in earthcontact concrete elements (piles, barrettes, diaphragm walls, columns, base slabs, etc.) through which a heat carrier uid is pumped that exchanges energy from the building with the ground. The heat carrier uid is a heat transfer medium of either water, water with antifreeze (glycol) or a saline solution. Glycolwater mixtures have proved most suitable, containing also additives to prevent corrosion in the header block, of valves, of the heat pump, etc. Once cast, the pipings within the underground-contact concrete elements are individually joined to a header and manifold block. They are joined by connecting pipes which, in the case of energy foundations, are normally laid within the blinding beneath the base slab. The secondary circuit is a closed uid-based building heating or cooling network (secondary pipework) embedded in the oors and walls of the structure or in bridge decks, road structures, platforms etc. (Fig. 8). Commonly, primary and secondary circuits are connected via a heat pump that increases the temperature level, typically from 10158C to a level between 258C and 358C (Fig. 9). All that is required for this process is a low application of electrical energy for raising the originally non-usable heat
Fig. 8. Absorber pipes entering the thermo-active concrete slab in the rst oor of a building

resources to a higher, usable temperature. The principle of a heat pump is similar to that of a reverse refrigerator (Fig. 10). In the case of the heat pump, however, both the heat absorption in the evaporator and the heat emission in the condenser occur at a higher temperature, whereby the heating and not the cooling effect is utilised. The coefcient of performance, COP, of a heat pump is a device parameter and is dened by COP energy output after heat pump kW energy input for operation kW (1)

A value of COP 4 means that from one portion of electrical energy and three portions of environmental energy (from the ground) four portions of usable energy are derived (Fig. 9). The efciency of a heat pump is strongly inuenced by the difference between extracted and actually used temperature. A high user temperature (inow temperature to the heating system of the secondary circuit) and a low extraction temperature (due to too low a return-ow temperature) in the heat exchanger (primary circuit) reduce its efciency. For economic reasons a value of COP > 4 should be achieved. Therefore the usable temperature in the secondary circuit should not exceed 35458C, and the extraction temperature in the absorber pipes should not fall below 058C. Consequently, this technology tends to be limited to lowtemperature heating (and cooling). Commonly, electric heat pumps are used, less often heat pumps with internal combustion and occasionally absorption heat pumps. For environmental reasons only refrigerants without ozone reduction potential and with a minimum greenhouse potential are allowed for heat pumps. Therefore halogenated uorinated hydrocarbons should not be used. The seasonal performance factor (SPF) of a thermo-active system with a heat pump is the ratio of the usable energy output of the system to the energy input required to obtain it. Therefore SPF includes not only the heat pump but also the other energy-consuming elements (e.g. circulation pumps). At present, values of SPF 3.84.3 are achieved with standard electric heat pumps; special devices with direct vaporisation increase SPF by 1015%. SPF usable energy output of the energy system kWh energy input of the energy system kWh (2)

Fig. 7. Installation of absorber pipes in a driven steel tube pile lled with concrete

In Austria about 200 000 heat pumps are running at present, and their number increases by more than 5000 per year. Their main purpose was at rst provision of warm water, but

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

External power (electricity) 1 4 4 4 (from the ground) Primary circuit 3 Energy flux 4

85

Secondary circuit

Connecting lines

Manifolds

P Header block: collector (for heating) distributor (for cooling) Heat pump

Absorber pipes (5 ground heat exchangers) incorporated in energy foundations (e.g. energy piles)

Fig. 9. Scheme of a geothermal energy plant with energy piles and an energy ux for COP 4 of the heat pump. COP coefcient of performance dening the heat pump efciency
Primary circuit Heat source
0C Refrigerant medium 23C/365 bar

Electrical power Compressor

Secondary circuit
135C

Refrigerant medium 162C/132 bar

Heating circuit water

Heat from compression Evaporator

25%

Heat from heat source 75% Expansion valve

24C Refrigerant medium 28C/365 bar

100%

Condenser

Refrigerant medium 133C/132 bar

130C

Fig. 10. Scheme of a compression heat pump with temperatures and pressures for the refrigerant medium R290 (as example). Heat exchange occurs from primary circuit to refrigerant medium in the evaporator and from refrigerant medium to secondary circuit in the condenser

since the year 2000 heating (and cooling) of buildings has dominated. It is estimated that these heat pumps save more than 250 000 t of fuel oil per year. Experience has shown that these geothermal cooling/ heating systems from energy foundations and other thermoactive ground structures may save up to two-thirds of conventional heating costs. Moreover, they represent an effective contribution to environmental protection by providing clean and self-renewable energy. If only heating or only cooling is performed, highpermeability ground and groundwater with a high hydraulic gradient are of advantage. However, the most economical and environmentally friendly is a seasonal operation with an energy balance throughout the year, hence heating in winter (i.e. heat extraction from the ground) and cooling in summer (i.e. heat sinking/recharging into the ground). In this case low-permeability ground and groundwater with only low

hydraulic gradients are favourable. Dry soil makes deeper piles and a larger area of the heat exchanger necessary. Depending on soil properties and the installation depth of the absorbers, 1 kW heating needs roughly between 20 m2 (saturated soil) and 50 m2 (dry sand) of the surface of concrete structures in contact with soil or groundwater. There is no limitation to the depth of piles or diaphragm walls as far as the installation of energy absorber systems is concerned. The energy potential increases with depth: hence deeper foundations are advantageous. The economically minimum length of piles, barrettes or diaphragm wall panels is about 6 m. The production of electric current from energy foundations and other thermo-active ground structures is theoretically possible but not effective. This is similar to biomasses as base materials: they exhibit a high efciency for heating (85%) but an extremely low efciency for producing electric current (25%). This would be an inexpedient application of environmental technologies, whereas combined heat and power plants reduce CO2 production signicantly. The combustion of wood is CO2 -neutral and replaces fossil resources. In Vienna the largest biomass-operated power station in the world is operating. Waste incineration for heating purposes also saves fossil fuels, especially in connection with district heating. Furthermore, combinations of geothermal and solar heating and cooling have proved suitable to contribute effectively to environmental protection. The economically optimal solution depends on numerous local conditions (e.g. twin circuits and/or photovoltaically operated heat pumps).

HEAT TRANSFER IN THE GROUND General Soil is a multiple phase system with a complex heat transfer mechanism involving (a) conduction (b) radiation (c) convection

86

BRANDL If the thermal conductivity and the temperature gradient are constant over the area and in its normal direction, respectively, equation (7) can be modied for an energy pile with radius R and length l to _ 2 R l d T [W] Q dr (10)

(d) vaporisation and condensation processes (e) ion exchange ( f ) freezingthawing processes. Heat transfer in unfrozen soil occurs mainly by conduction and secondly by convection. Heat convection is also possible if there is a phase change of water (latent heat during vaporisation and condensation). In soils, radiation usually contributes only negligibly to heat transfer; its effect in sand is less than 1% of the overall heat transfer (Rees et al., 2000). Freezing and thawing, however, can also transfer signicant heat, but such processes should be avoided for thermo-active ground structures. Heat convection occurs between thermo-dynamic systems that move relative to each other (i.e. by means of circulation ows). In the soil the solid phase is static: hence only convection with water and convection with gas has to be distinguished. Heat transfer by uid convection can be described by _l,conv cw rw vw T T 9 q (3) where cw is the specic heat capacity of the soil water, rw is the density of the soil water, vw is the vector of water velocity ( vw ki), and T 9 is a reference temperature. A similar equation is obtained for vapour (pore gas) convection _v,conv cv rv vv T T 9 q (4) where cv is the specic heat capacity of soil vapour. The latent heat transfer that occurs as a result of a phase change of water (vaporisation) depends mainly on the quantity of vapour transfer occurring in the soil pores. It increases with decreasing water content, and can be expressed by _lat L0 rw vv q (5) where L0 is the latent vaporisation heat at temperature T9. _tot , may be Thus the total heat transfer in the soil, q dened as (Rees et al., 2000) _cond q _l,conv q _v,conv q _lat _tot q q (6) _l,conv is the heat _cond is the conductive heat ux, q where q _v,conv is the heat ux ux generated by liquid convection, q _lat is the heat ux due generated by vapour convection, and q to latent heat. Heat conduction is a process whereby energy is passed from one region of a medium to another by molecular transfer. According to Fouriers law, the heat ux for a heat volume Q through an arbitrary area during time t, that is, the heat ux density q _cond , can be written _cond q _ Q Q @T At A @n (7)

This is indicated in Fig. 11 for an energy pile utilised for cooling ( heat transfer into the ground) or heating ( heat extraction from the ground), describing a steady state where the temperature does not change within a period of time t. A temperature change is caused by an alternation of the heat ux density within this period, thus leading to a change of the internal energy r c _ @q _ @q _ @T @q @t @x @ y @z (11)

Differentiating equation (7) with respect to spatial coordinates and combining it with equation (11) yields ! @T @2 T @2 T @2 T a a div grad T a T @t @ x2 @ y2 @ z2 (12) with the thermal diffusivity a (m2 /s) given by a rc (13)

where (W/m K) is the thermal conductivity, c (J/kg K) is the specic heat capacity (thermal capacity), and r (kg/m3 ) is the density of the solid medium. If an inner heat source (internal heat generation) exists in a considered soil volume, the basic heat conduction equation becomes _i @T Q a T @t rc (14)

Equation (12) in rectangular coordinates can be transformed into cylindrical coordinates with radius r, azimuth and axis z ! @T @2 T 1 @ T 1 @2 T @2 T a (15) @t @ r 2 r @ r r 2 @ j2 @ z 2
T Cooling: temperature increase 5 thermal conductivity dT 5 temperature dr gradient

where is the thermal conductivity and @ T/@ n is the temperature gradient in the actual ow direction n @T @T @T @T ex ey e z grad T @ n @x @y @z (8)
Heating: temperature decrease

If the grain sizes of the soil particles and the pore sizes are negligibly small in relation to the considered soil volume, the complex heat transfer process may be reduced to only conduction, which dominates in the case of thermo-active foundations, retaining walls, and tunnels. Equation (7) can be written in rectangular coordinates as   @T @T @T _ q ex ey e z grad T (9) @x @y @z

Thermal flow . dT Q 5 2 rl dr

Fig. 11. Temperature curves in the soil around an energy pile utilised for heating or cooling. Schematic; constant heat ux assumed

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES In spherical coordinates with radius r, angle of declination and angle of ascension , equation (12) becomes ! @T @2 T 2 @ T 1 @2 T cos @ T 1 @2 T a @t @ r 2 r @ r r 2 @ 2 r 2 sin @ r 2 sin @ j2 (16) The full mathematical description of ground heat transfer as a physical problem requires the assumption of initial and boundary conditions. The initial state considers the temperature distribution at time t 0. Regarding the boundary conditions, three different assumptions can be made (Adam & Markiewicz, 2002; Honger, 2002) (a) Dirichlets boundary condition. Surface temperature TS is constant or depends only on time t. _ at (b) Neumanns boundary condition. Heat ux density q the surface is constant or depends only on time t. (c) Cauchys or mixed Neumanns boundary condition. On the surface of the considered body a heat exchange occurs with the liquid or gaseous surroundings of the body exhibiting a temperature TU. According to Newtons law of cooling the heat ux is proportional to the temperature difference between the surroundings and the surface temperature, Tinterface. The factor of proportionality is dened as the heat transfer coefcient (W/m2 K)   @T Tinterface TU (17) @ n interface Because of mathematical difculties, analytical solutions of these equations are possible only for simple cases. Onedimensional problems can easily be solved, since equation (12) depends only on one coordinate then. Two and threedimensional problems can be solved in some cases by combining one-dimensional solutions. Examples are two or three-dimensional edges/corners of bodies or other bodies gained by Booles mathematical operations of basic forms.

87

(a) Case 1: Semi-innite body. Simulates plane interfaces between soil and atmosphere or foundations (slabs; basement walls, retaining walls, pile walls, diaphragm walls; horizontal earth collectors and trench collectors). (b) Case 2: Innite body with cylindrical gap. Simulates long vertical ground heat exchangers, and hence heat extraction/storage boreholes, energy piles and energy wells; also, the lower part of pile or diaphragm walls embedded on either side. (c) Case 3: Innite body with spherical gap. Simulates spherical thermo-active ground openings (underground structures), and approximates the toe zone of heat extraction/storage boreholes and energy piles. Analytical solutions are available only for case 1 (semiinnite body) considering simple boundary conditions. More complicated boundary conditions require semi-analytical solutions that reduce the differential equation to a semianalytically solvable function, the so-called Gaussian error function (Adam & Markiewicz, 2002; Honger, 2002). The complement of the Gaussian error function is dened as follows to be solved numerically 2 2 (18) erfc 1 p e d 0 where is the integration limit and is the integration variable. This error function is also the basis for the solution of case 3 (innite body with spherical gap). Independently of boundary conditions, case 2 (innite body with cylindrical gap) can only be found numerically using a differential equation solver.

Analytical and numerical calculations Absorbers for geothermal heat extraction or feeding/ storage exhibit different forms depending on the shape of the structures used as thermo-active elements. Energy piles or vertical bored heat exchangers can be modelled by cylinders, whereas complicated foundations or energy tunnels require sophisticated geometric considerations. In the following, three basic cases are discussed, which can be used to simulate most of the commonly designed absorber elements (Table 1).

Semi-innite body (case 1) The problem was solved for different boundary conditions and heat supply or loss at the surface (Adam & Markiewicz, 2002; Honger, 2002). First, a harmonic temperature oscillation on the surface was assumed (Cauchys or mixed Neumanns boundary condition) with heat transfer between soil and air. Thus the daily or seasonal temperature uctuations between ground and atmosphere can be treated, whereby a time lag has to be considered. This depends on the considered depth in the half space, and may even cause an anti-cyclic behaviour in the seasonal histogram of the mean daily air temperature ( Tm,out ). The surface of a semi-innite body is assumed to exchange heat with the air, which performs a sinusoidal temperature oscillation F(t): see Fig. 28 for an example.

Table 1. Basic cases for heat conduction in soil Case 1: Semi-innite body Sketch Differential equation @2 1@ x, t x, t a @t @ x2

2: Innite body with cylindrical gap

@2 1 @ 1@ r, t r, t r, t r @r a@t @ r2

3: Innite body with spherical gap

@2 2 @ 1@ r , t r , t r, t r @r a @t @ r2

88

BRANDL Pyearly p 365 19:1 Pdaily (25)

Furthermore, assuming the transient period of the initial vibrations to be over, a steady state therefore exists. Under these conditions the soil temperature uctuates according to the mean yearly air temperature Tm,out if radiation effects and geothermal temperature gradient are neglected. However, the amplitudes decrease depth with owing to the thermal inertia of the soil. A function that fulls the differential equation (12) with the corresponding boundary conditions is ! z (19) T z, t Tm,out T out e z= d cos t d where r s 2a aP d 1 p 1 2 k 2k2 k arctan 1 k r k d aP (20) (21) (22) (23)

Equation (25) reveals that the yearly temperature wave penetrates 19 times deeper into the ground than the daily wave. Furthermore, the heat ux in the ground reaches its maximum earlier than the temperature. The second solution of the basic case 1 is achieved by assuming a sudden temperature fall on the surface of the semi-innite halfspace (Dirichlets boundary condition). This allows the investigation of the heat transfer between a plane thermo-active element and the ground in contact. Assuming a temperature fall from T0 to TS at time t 0, the solution of equation (12) is obtained as follows   z (26) T z, t TS erfc p 2 at T z, t is the temperature function with T0 as reference (initial) temperature T z, t T ( z, t) T0 Hence T z, 0 0 (28) (27)

Tm,out is the mean yearly air temperature (8C), Tm,out is the mean daily air temperature (8C), Tout is the temperature amplitude (8C), 2= P (1/s), P is the period duration of temperature oscillation (s), d is the damping depth, and is the heat transfer coefcient from soil to air (W/(m2 K)). On the surface (z 0) the solution reduces to T 0, t Tm,out Tout cos t (24)

Comparative calculations of the basic cases The basic cases 1 to 3 were simulated by numerical models and the results compared with the (semi-) analytical solutions. As an example for timetemperature curves of different heat exchanger shapes the following assumptions are made. (a) The soil, initially at constant temperature T0, is suddenly heated up to TS T0 + 258C on its surface (i.e. the interface between soil and heat exchanger). (b) The soil parameters are as given in Table 2. (c) The radius of the cylindrical underground space (case 2) and spherical underground space (case 3), R 30 cm. (d) The thermal conditions and properties are as given in Fig. 11 and Table 2. (e) The time period considered is half a year, that is, 182.5 days. Figure 13 shows the results plotted for two-week intervals. In order to keep the surface temperature constant, a certain heat supply must be provided. This is given by   @ TS _0, t T z, t p (29) q  @z at
z0, t>0

It is evident that the amplitude of the surface temperature decreases by a factor , 1 in relation to the air temperature and, moreover, undergoes a time lag of . Fig. 12 illustrates the timedepth curves for a full period at different times t. The damping depth d depends on the number of periods and temperature conductivity. In soils with high a values (moist sand) the temperature wave penetrates relatively deeply. Comparing the response of the soil body to yearly and daily periods yields the following ratio
Soil parameters: Thermal conductivity s 5 25 W/m K Density rs = 2700 kg/m3 Spec. heat capacity cs 5 800 Ws/kgK Oscillation parameters: Tm,air 5 10C Tair 5 15K s 5 30 W/m2 K P 5 365 3 24 3 3600 5 31536000 s 5 2/P 5 199238 3 1027 s21 t 5 2P/8 t 5 3P/8 t 5 7P/8

25 20
Temperature: C

t50 t 5 P/8

_ is integrated over the considered time period of 182.5 If q

Table 2. Example parameters for calculating the basic cases of Table 1 Property Thermal conductivity, : W/m K Density, r: kg/m3 Specic heat capacity, c: J/kg K Temperature conductivity, a: m2 /s Radius of cylindrical and spherical gap: m Sudden temperature rise, S : K Observed period, t Value 2.5 2700 800 1.1574 3 106 0.3 25 Half a year 182.5 days

15 10 5 t 5 5P/8 25

t 5 6P/8

50

25 t 5 P/2

75 Depth, z: m

100

125

150

Fig. 12. Temperature distribution within soil at different times caused by a harmonic oscillation of outdoor temperature

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

25 20 Temperature: C 15 10 5 t 5 14 days 2 4 6 8 Depth: m (a) 25 20 Temperature: C 15 10 5 t 5 1825 days t 5 14 days 2 4 Radius: m (c) 6 8 10 10 12 14 t 5 1825 days 25 20 Temperature: C 15 10 5 t 5 14 days 2 4 6 8 Radius: m (b) 10 12 14 Numerical solution

89

(x,t) 5 Serfc

x 2!at

t 5 1825 days

(r,t) 5 S

R r2R erfc r 2!at

for R < r < `

Fig. 13. Comparison of the basic cases of heat conduction at equal boundary conditions and thermal and material properties. Temperatures in soil due to a temperature change from T 0 to 258C on the surface for half a year: (a) case 1, semi-innite body; (b) case 2, innite body with cylindrical gap; (c) case 3, innite body with spherical gap

days, the total heat supply to be provided is obtained. This is 72.37 kWh/m2 for case 1. The heat supply needed for case 2 results from equation (15), whereby differentiation with respect to r on the surface and integration over time were performed numerically. This yielded 287.04 kWh/m2 . The solution for the spherical underground space with radius R 30 cm (case 3) can be found from equation (16) as follows T r, t T S R rR erfc p for R < r < 1 r 2 at (30)

Table 3. Results of comparative calculations of basic cases 1 to 3 (to Tables 1 and 2) Basic case 1 2 3 Specic heat exchange: kWh/m2 72.367 287.040 984.807

most effective, whereas plane elements such as base slabs or earth collectors are less productive. THERMAL SOIL PROPERTIES In most regions of Europe the seasonal ground temperatures remain relatively constant below a depth of 1015 m. Values between 108C and 158C predominate to a depth of about 50 m (Fig. 14). Such temperatures permit economical heating and cooling by thermo-active ground structures, and they represent ideal conditions for heat pumps. Substantial temperature uctuations during the year would reduce the efciency of both heat pumps and absorber systems. In the tropics the constant ground temperature at a depth of more than 1015 m below the surface varies between 208C and 258C (locally even 288C), which still allows cooling of buildings. The thermal conductivity , heat capacity c and density r of the soil are temperature-dependent parameters, and are coupled in the basic equation of heat conduction (equation (13)) with the thermal diffusivity a. This parameter (cm2 /h) describes the depth and velocity of penetration of a temperature wave into the ground. If the ground temperature is

The heat ux density is given by     @ TS TS p _0, t T r, t : q  @r R at r R, t > 0

(31)

If R ! 1 the solution for the semi-innite body according to equation (29) is obtained (case 1). Independently of the time a basic heat ux TS /R must occur to keep the temperature of the spherical underground space constant (at t ! 1). Integrating the heat ux over half a year yields the required heat supply of 984.81 kWh/m2 . Table 3 summarises the results of the comparative investigations. A comparison of these three basic cases illustrates that there are signicant differences of the heat penetration velocity caused by temperature changes at the surface. Higherdimensional heat conduction leads to a more efcient heat transport. This means that spherical underground spaces can extract more heat energy per unit area. Spherically shaped absorber elements or heat-exchanging structures are therefore

90
Winter Summer

BRANDL specic heat capacity of ice is only ci 1884 J/kg K compared with cw 4186 J/kg K for water. The volumetric capacity CV is derived from the specic heat capacity and the bulk density of the soil, and represents the weighted arithmetic average of the particular soil components. Hence CV
T 5 const. 1015 (Europe) 2025 (tropics)

1 Frost

0C

0C

z 5 10215 m

ri ci xi

(34)

or
Heat transfer: Conduction Convection Radiation

CV rs

w cs cw 100

 (35)

Fig. 14. Heat transport and geothermal situation for deep foundations (schematic). Ground temperature at depth z > 1015 m tends to be constant (Europe, T 10158C; tropics, T 20258C, locally 288C)

measured at at least two depths z1 and z2 , the thermal diffusivity can be calculated either from the amplitude decrease or from the time lag. The proportionality factor that relates the rate at which heat is transferred by conduction to the temperature gradient is known as the thermal conductivity (SI units W/m K). is strongly inuenced by the water content and density of the soil, and hence also by the mineralogical components and by the chemical properties of the pore water. Freezing increases the thermal conductivity signicantly, because water 0.57 W/m K changes to ice 2.18 W/m K. Consequently, the thermal conductivity of the soil can only be expressed approximately (Rees et al., 2000). The heat capacity c (J/kg K) denes the amount of energy stored in a material per unit mass per unit change in temperature. It is required when non-steady solutions are to be determined. The heat capacity does not depend on the microstructure. In most cases, therefore, it is satisfactory to calculate the heat capacity of soils from the specic heat capacities of the different constituents according to their volume ratios c cs xs cw xw ca xa (32) where x is the specic volume: hence xs 1 n for the solid phase, xw nS for the pore water, and xa n(1 S) for the pore air, where S is the degree of saturation and n is the soil porosity. The heat capacity of a soil with more than three constituents (e.g. contaminated ground) can be calculated by simply adding more terms into equation (32) (Rees et al., 2000). The heat capacities of solid and liquid constituents can be considered constant, and xa is negligible. Thus equation (32) reduces approximately to (W s=kg K) (33) c 2:0 xs,min 2:5 xs,org 4:2 xw where xs,min refers to the mineral components and xs,org to the organic components of the solid fraction. As the mineral and organic solid components have somewhat similar thermal capacities, only the water content remains as a relevant variable, at least in the short term. In the long term consolidation or shrinkage processes of soils (under external loads or self-weight, or due to excessive heat extraction) may play a role because the volume ratios change. The overall thermal capacity increases with the water content and decreases in the case of freezing. The

where cs is the specic heat capacity of mineral components (most minerals have cs 1000 W s/kg K at a temperature of 108C); cw is the specic heat capacity of water (cw 4186 W s/kg K); and w is the water content related to the dry mass in percent. The most important thermal soil parameter is the thermal conductivity . For the preliminary design of complex energy foundations, or for the detailed design of standard geothermal systems, it can be taken with sufcient accuracy from diagrams considering water content, saturation density and texture of the soil (Fig. 15). For large complex projects should be determined from laboratory or/and eld tests, whereby a soil body has to be exposed to a temperature gradient. However, this may cause a signicant moisture transfer in unsaturated soils, which should be taken into account when interpreting the measured results. In the eld, the thermal response test has proved suitable, as it can be performed directly with vertical ground heat exchangersthat is, with heat extraction/storage boreholes, energy foundations or energy wells. In the laboratory both the steady state method and the transient method are used, but the latter is preferred, especially the hot wire method. In this method an electrical wire is implanted in the experimental soil sample. A steady current is supplied to the electric wire, and the temperature rise and fall of the heating wire is measured by a thermocouple and recorded over a short heating and cooling interval. Comparative tests have shown that the thermal conductivity during heating is commonly slightly higher than that during cooling (Abu-Hamdeh et al., 2001). The specic heat capacity can be determined in the laboratory by mixing water and soil of different temperatures. If the total thermal energy of both components remains constant, and the specic heat capacity of one component is known (e.g. cw of water), then the specic heat capacity of the soil (cs ) can be achieved. Commonly, test temperatures of 08C for soil and 208C for water are used, and the steady-state temperature of the mixture is measured. Moist soil requires a correction of the cs value according to its water content. Considering the temperature dependence of the specic heat capacity c(T) the energy balance of the watersoil mix can be written as (Leu et al., 1999) cs Ts Ts ms cw Tw Tw mw cs Tmin ms cw Tmin mw Tmin (36) where Ts is the temperature of the cooled soil (8C); Tw is the temperature of the water before mixing (8C); Tmin is the minimum temperature of the mixture (8C); ms , mw are the mass of the soil and the water (g); and cs , cw are the specic heat capacities of the soil and the water (J/g K).

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

25 24 23 22 21 rd: t/m3 20 19 18 17 16 15 14
0 6

91

100% saturation
4 8 5 4 5 4 2
1

25 24 23 22 21 rd: t/m3 20 19 18 17
3 3
27 24 21

100% saturation
9 3 4 2 3

: W/mK

: W/mK

3 6 3
0 3
2 7

39

16
3 0

21 18

24

36
2 1 5 1 8

15 14 40 45 13 0
9 0

15

9 0

13

10

15 20 25 30 Water content: % (a)

35

10

15 20 25 30 Water content: % (b)

35

40

45

20 19 18 17 rd: t/m3 16 15 14

100% saturation : W/mK

1 2

20 19 18 17 rd: t/m3 16 15 14 13 12 11 10
03

100% saturation 21
15

18

: W/mK

13
0

21

09

0 9

12 11 10 0

6 0 3

06

10

15 20 25 30 Water content: % (c)

35

40

45

10

15 20 25 30 Water content: % (d)

35

40

45

Fig. 15. Thermal conductivity against dry density and water content for frozen and unfrozen soils (Jessberger & Jagow-Klaff, 1996): (a) coarse-grained soil, frozen; (b) coarse-grained soil, unfrozen; (c) ne-grained soil, frozen; (d) ne-grained soil, unfrozen

TEMPERATURE-INDUCED CHANGES OF SOIL PROPERTIES The range of temperature changes in the ground caused by geothermal energy utilisation is rather small. Nevertheless, they should be checked, and environmental assessments are frequently required by national laws. Moreover, the thermal properties of soils vary considerably in a freezing soil, where the transfer coefcients across the freezing front change greatly. Consequently, excessive heat extraction from energy foundations causing soil freezing should be avoided. Stressstrain properties Unlike the linear stressstrain relationship of ideal elastic materials, the thermal expansion coefcient of soil cannot be determined on the basis of Hookes law, but only by means of tests whereby the water content has to be kept constant to avoid shrinkage effects. Pure minerals do not undergo signicant strains until very high temperatures (> 3008C). Consequently, the thermal expansion coefcient of very moist ground corresponds to that of the soil water. Thermal processes in the ground induce water migration towards the colder regions. In ne-grained sensitive soils this may gradually cause shrinkage in the warm zone and expansion in the cold one. A thermal expansion of pore water

increases the pore water pressure, and consequently decreases the effective stress of the soil. Furthermore, an increasing temperature reduces the internal viscosity, and hence the shear resistance. Oedometric tests on silt and sands have shown larger settlements at higher temperatures; however, with increasing load the modulus went up more (Ennigkeit, 2002). These tests also revealed an accelerated consolidation at higher temperatures caused by lower viscosity of the pore water and thus a higher hydraulic conductivity. Organic constituents increase the temperature sensitivity of soils, especially clays. Nevertheless, experience has shown that properly operated energy systems do not affect the loadbearing capacity of thermo-active piles embedded in such strata. Commonly, the interactions are rather negligible in practice, but they need to be considered if the above-ground part of a building is extremely sensitive to differential settlements. Hydraulic properties Lowering the groundwater temperature causes an increase in viscosity, and hence a decrease of hydraulic conductivity. This leads to lower ow velocities and to smaller ow gradients of the groundwater. However, comprehensive

92

BRANDL
uo u(r,x) r R u(r)

parametric studies have shown that in practice these effects are negligible in relation to geothermal energy systems. Physico-chemical properties Too intensive cooling of the groundwater (due to excessive energy extraction for heating buildings) increases the pH value and reduces calcium solubility, which favours the clogging of pores. On the other hand, the solubility of gaseous substances such as CO2 increases, and hence also the hardness of the groundwater. Hitherto investigations have shown that temperature changes smaller than T 58C have a negligible effect within a temperature range of 0208C. Biological properties Temperature is one of the most important environmental factors for the microorganisms in groundwater. Many of them can exist only within a very limited temperature range. In particular, the activity of bacteria-consuming microorganisms drops signicantly at temperatures below 108C. Therefore, for instance, the mortality rate of pathogens is reduced by about half if the temperature drops from 78C to 28C. HEAT TRANSFER BETWEEN ABSORBER FLUID AND CONCRETE/SOIL General Assuming that the walls of the absorber pipes of a ground heat exchanger have the same temperature as the surrounding concrete or soil respectively reduces the complex thermal problem (Fig. 16) to the heat transfer from pipe wall to absorber uid (heat carrier uid). This is essentially inuenced by the ow behaviour of the uid, that is, laminar or turbulent. Pipe ow is described by two zones (Fig. 17): the transient inow zone, where ow velocity and temperature prole change with pipe length; and the steady-state condition with a constant hydrodynamic and thermal prole. The heat transfer does not then change (at constant thermal conductivity). In absorber pipes in thermo-active foundations, retaining walls, tunnels, pipe wells and roads the steady-state phase dominates. Furthermore, this state is reached after only a short distance. Hence the following theoretical considerations are limited to the steady-state ow and heat transfer problem (Adam & Markiewicz, 2002).

x
Hydrodyn. entry length Lhyd To Hydrodyn. stabilised

Tw T(r,x)

Tw T(r)

r R

x
Thermal entry length Lth Thermal stabilised

Fig. 17. Flow velocity and temperature distribution in absorber pipes lled with heat carrier uid

Laminar ow in a pipe is based on ow paths with different velocities u and interface friction , which is proportional to the velocity gradient du/dx perpendicular to the ow direction. The coefcient of proportionality is the viscosity , which increases with temperature. Fig. 18 shows an example of a typical absorber uid for energy foundations (an antifreezewater mixture). For this purpose Newtons friction law can be applied du dx (37)

1000 800 600 400 200 100 80 60 Kinematic viscosity: mm2/s 40 20 10 8 6 4 2 100% (v/v) Antifrogen L

Energy pile

Soil

80 57 47 38 25 16 0 5 water 5 Anti-freezing limit 240 220 60 120 140 60 80 100 Temperature: C 5 Boiling point 140 180

Heat transfer soil f (T,T,t,r,,c,e,GW) Heat transfer soil concrete f (T,T,) Heat conduction concrete f (,s,T,t) Heat transfer concretepipe (Nu,,s,T) Heat conduction pipe f (,s,T, t) Heat transfer pipe absorber fluid f (Nu,,s,T) Heat transport absorber fluid f (R,L,,,c,r,,u)

1 08 06 04 02

Fig. 16. Heat transport from soil to heat carrier uid within the absorber pipe of an energy pile (GW, ground water)

Fig. 18. Kinematic viscosity against temperature for different mixtures of water and Antifrogen L

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES 93 The mean velocity of laminar ow is umean 0.5umax , and Heat transfer by forced convection for turbulent ow umean (0.800.85)umax . The transition The absorber pipes of a heat exchanger are part of a closed circuit (primary circuit in Fig. 9) where the ow is from laminar to turbulent ow condition is described by the created by a pump. Therefore this is called forced convecReynolds number, Re tion. Commonly, the calculation is based on steady-state ud with (38) Re conditions, whereby the ow velocities should not be too r low. The ow velocity u(r) within a circuit is different at each point of the cross-section. Consequently, the period t for where u is the mean velocity (m/s), d is the pipe diameter which individual uid particles remain within a certain (m), is the kinematic viscosity (m2 /s), is the dynamic absorber pipe section differs. According to equation (12) viscosity (kg/m s), and r is the density (kg/m3 ). Below the one-dimensional conditions can be described by critical Reynolds number Re 2300 laminar ow occurs; above Re . 104 full turbulence exists. Between these bound@T @2 T ary values transient conditions occur. Turbulence increases 2 (43) rc @t @r the diffusive transfer of energy, impulse and mass. This effect increases with ow velocity. where r is the density, c is the specic heat capacity and is the thermal conductivity of the owing medium. The different time periods of the uid absorber staying in particular sections are t x/u(r) with the radial distance r Heat transfer by convection according to Fig. 19. This leads to Heat transfer through contact is based on Fouriers law of   molecular heat transport, which can be written for one@T 1 @ @T r r c u r (44) dimensional problems as @x r @r @r   @T _ (39) q Furthermore, another dimensionless coefcient is used for @x parametric studies, the Prandtl number Pr, which has a material-dependent value and is dened as _ is the heat ux density (W/m2 ), is the thermal where q conductivity of the owing medium (W/(m K)), T is the r c (45) Pr temperature, and x is the local coordinate. a Heat transfer between masses not moving relative to each other occurs by conduction. Heat transfer by convection is For Pr ! 0 the velocity prole along a ow path x, where based on differential movements. The latter occur between the heat is transferred, is equivalent to the prole of a piston pipe wall and absorber uid, whereby molecular heat transfer ow. In the case of Pr ! 1 the velocity prole corresponds takes place at the interface to the HagenPoiseuille ow (Fig. 19). Common absorber   uids exhibit values of about Pr 7 for clean water close @T to the freezing point, and Pr 70 for a viscous uid, such _W (40) q @ r wall as waterglycol mixture as anti-freeze medium Commonly, the heat transfer from concrete or soil to the _W is the heat ux density at the pipe wall (W/m2 ). where q absorber uid occurs at a widely constant pipe wall temperaThe heat transfer between the pipe wall and the uid can ture (Tw ) along the entire pipe length, if laminar ow be described by the heat transfer coefcient conditions prevail. For a mean uid temperature Tm the molecular heat transfer can be then be described by equation _W q @ T =@ rwall (41) (47) and Fig. 20 (left) Twall Tfluid T @T TW T d Tm (46) @ x TW Tm d x and by the Nusselt number Nu, which is dened as follows     TW T dTm 1 @ @T d @ T =@ rwall (47) r c u r r Nu (42) TW Tm dx r @r @r T = D Under turbulent conditions the heat transfer velocity depends not only on the self-velocity of the energy carrier (heat carrier uid) but also on that of the turbulent uctuations, which is connected to the average ow velocity of the absorber uid. Consequently, the heat transfer depends also on the ow velocity, and the heat transfer coefcient is a function of material properties, geometric dimensions, length of heat transfer occurrence, and ow velocity of absorber uid. Calculation of the temperature gradient (@ T/@ r)wall at the pipe wall is possible only as long as equation (39) is valid at each point of the absorber uid. But this applies only to laminar ow without friction, or to motionless media. For turbulence, equation (39) is valid only for the pipe wall (according to equation (40)) but not for the interior of the owing medium. To date there is no exact theory for this thermal problem; it can be solved only by equations based on experimental data (VDI, 1997). _w (Tw Tm ) In the case of constant heat ux density q the heat transfer coefcient is constant  ! qW @ TW T (48) TW Tm R @ z= R TW Tm wall In this case the temperature difference Tw Tm is also constant, leading to
um u(r ) r x

Fig. 19. Flow velocity distribution in pipes (HagenPoiseuilles parabola)

94
T . qw Tw 5 const

BRANDL
T . qw Tw const Tm Tm . qw . qw5 const

Tw 5 const (a)

. qw 5 const (b)

Fig. 20. Heat conditions at the wall of an absorber pipe for: (a) constant temperature; (b) constant heat ux density

@ T d TW d Tm @x dx dx

(49)

_ r , and the oscillating balls have a mass ux density of m shear stress 9 along l is _ r uc u9 9 m (52)

This nally gives a molecular heat transfer according to equation (50) and Fig. 20 (right):   dTm 1 @ @T r r c u r (50) r @r @r dx Turbulent ow conditions in the absorber pipe include a laminar zone close to the pipe wall where the local ow velocity is nally zero (Fig. 21). Equation (37) is then valid only along the pipe wall but not within the core of the ux. The shear stress W along the wall is u9 W r u 2 l 8 c (51)

_ transferred by these In a similar way the heat ux density q turbulence balls can be expressed by _9 m _ r c T 9 Tc q (53)

The temperature prole in a cross-section is similar to the _w ow velocity prole (Fig. 21). Thus the heat ux density q at the pipe wall becomes _ W q _W T9 q Tc or l TW Tc (54) the the of for

where uc is the ow velocity in the core of the uid, u9 is the ow velocity along the laminar (viscous) edge zone, is the coefcient of ow pressure loss within the pipe system, and l is the thickness of the laminar (viscous) edge zone. Energy and impulse transfer within this core is achieved by so-called turbulence balls continually entering and leaving the laminar edge zone, and thus undergoing an average velocity change from ucore to u9 and vice versa. These
Turbulent flow in the core

where Tc is the temperature in the core of the uid, T 9 is temperature at the laminar (viscous) edge zone, Tw is temperature on the wall, and l is the thickness the laminar (viscous) edge zone Equations (51)(54) lead to Prandtls basic equation the relationship between heat transfer and ow resistance Nu 1 Re Pr 8 1 Pr 1 u9= uc

(55)

p The velocity ratio u9/uc is then substituted by 12.7 =8, which represents a suitable approach. Further sophistications are based on experimental data, leading nally to a formula that also considers the length of the pipe system (Oertel, 2001) Num,T with s 2 3 d f 1 L (57) =8 Re 1000 Pr f pp 1 12:7 =8 3 Pr 2 1 (56)

Laminar zone

um u Tm T

. mr s R r

Fig. 21. Turbulent ow conditions with a laminar ow area along the wall of an absorber pipe

for 0.5 , Pr , 104 and 2300 , Re ,106 , and for 0 , D/L , 1. Figure 22 considers laminar and turbulent ow conditions depending on the dimensionless Reynolds, Nusselt and Prandtl numbers. When determining these parameters a possible temperature dependence of the material properties has to be taken into account, whereby in practice only the dynamic viscosity is inuenced by temperature changes in a relevant way. Thus the Nusselt number becomes

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

100000 10000
D/L 5 00001 D/L 5 0001

95

Turbulent flow

Cpile

q
Cpipe

Nu 5 3 D/

1000 100 10 1

D/L 5 001 D/L 5 01

Re 5 2300 Laminar flow

10

100 1000 Re 3 Pr 3 D/L

10000

100000

. qw

Fig. 22. Heat transfer from absorber pipe wall to heat carrier uid for different conditions. Prandtl number Pr 75

Nu Num

m w

0:14

(58)

n absorber pipes

where m is the dynamic viscosity at a caloric mean temperature Tm, and w is the dynamic viscosity at a wall temperature of TW. Summarising remarks The material properties should be related to a mean temperature of Tm (Tinflow + Treturn flow )/2 considering the inow and return ow of the absorber uid into/from the primary circuit. The heat transfer coefcient depends on the pipe diameter d, the pipe length L, the ow velocity u, the viscosity , the density r, and the specic heat capacity c or thermal conductivity respectively. For laminar ow it can be determined theoretically; turbulence, however, requires experimental data. The heat transfer coefcient of turbulent ow is always higher than that of laminar ow: laminar < f (u) whereas turbulent f (u 3=4 ), if equal boundary conditions are assumed. The ow velocity is a criterion not only for the contact period but also for the intensity of a turbulent mixing. The Nusselt number Nu is a valuable criterion to describe the heat transfer intensity from the absorber uid to a particular section of the absorber pipe, but it does not describe the overall heat extraction (or storage) Q of the entire absorber system. For that variable the time period for an absorber uid circulates within the heat exchanger has also to be considered. Some guidelines for geothermal energy utilisation recommend the creation of turbulent ow in the absorber pipes. However, this should not be generalised. In the case of longer heat extraction (or storage) the critical point is not the heat transfer but the quantity of heat energy economically extracted from or stored in the surrounding soil. Highperformance pumps, required to create turbulent conditions, would therefore reduce the seasonal performance factor (SPF) of the overall geothermal system. Figure 23 illustrates the performance balance of soil and absorber (heat exchanger) per metre run of an energy pile _ of the under steady-state conditions. The heat ux density q _w , whereby the soil is compared with that of the absorber, q geometric conditions (circumference of pile Cpile , circumference of absorber pipe Cpipe ) and the number of absorber pipes have to be considered _w Cpipe n Tw Tm Cpipe _ Cpile $ n q q (59) where n is the number of absorber pipes lled with heat carrier uid.

Fig. 23. Heat volume balance for an energy pile: C, circumfer_, heat ux density ence of pile or pipe; q

Equation (59) does not include the heat transfer through the concrete cover and the pipe wall. This can only be simulated numerically. Therefore numerous parametric studies were conducted comprising the parameters discussed below.

Numerical simulations In order to investigate the inuence of individual parameters and their interaction, comprehensive comparative studies were conducted (Markiewicz, 2004). They referred to the following parameters: concentration of absorber uid (waterglycol), Conc.; cooperation temperature, Tm ; pump performance, P; efciency of pump, p ; inner pipe diameter, d; roughness of inner pipe wall, ; length of pipe for a particular pump performance, LP ; specic heat ux resistances, Ri ; pipe length for heat transfer, L; and temperature of pipe wall, TW. Figure 24 illustrates the scheme for a one-family house as an example. The ground heat exchanger is idealised as one

Lp(m) P (W); P [-]

d (m) (mm) absorber fluid concentration (%) Tm (C) Ts (C)

L (m)

Absorber units (e.g. energy piles)

Fig. 24. Scheme of an energy system for a house; symbols for parametric studies

96

BRANDL Furthermore, it increases with operating temperature, whereas under turbulent conditions the ow velocity tends to be independent of the temperature. Under laminar conditions the pipe diameter does not inuence the residence period of the heat carrier uid within the ground heat exchanger (absorber pipes), whereas under turbulence this duration increases with pipe diameter. In small-diameter pipes and at low operating temperatures laminar ow occurs in practically all cases. If small-diameter pipes are used, Nu can hardly be increased by increasing the pump performance. Hence installing pumps with higher capacity is then of no use. However, for large-diameter pipes turbulent ow conditions can be achieved rapidly by increasing the pump performance. This also increases the heat transfer from absorber pipe wall to absorber uid. With decreasing pipe diameter the total ow resistance that has to be overcome by the pump increases, as pipe wall friction related to the diameter increases. For diameters of d 24 cm the pressure remains nearly constant, but for smaller diameters it increases overproportionally. Furthermore, the pressure increases with reduced operating temperature because the viscosity of the uid increases.

absorber structure. For thermal-ow-relevant investigations it is sufcient to consider only the primary circuit of the geothermal energy system. The secondary circuit is optimised by the buildings HVAC expert (for heating, ventilation, air-conditioning) considering the interaction with the primary circuit. The calculations are based on a clear limit between laminar and turbulent ow state for Re 2300. Hence a step function is assumed theoretically, whereas a transient zone occurs in practice. Prior investigations revealed that there is no qualitative difference between Nu and the heat transfer coefcient per metre run of the absorber pipe. Thus Nu is the most suitable for comparing heat transfer processes in parametric studies. The results of these numerical simulations can be summarised as follows. Heat transfer from the absorber wall to the absorber uid increases with Nu. It depends widely on the ow conditions: zones of laminar or turbulent ow have do be considered in a different way. In principle, heat transfer under turbulent conditions is signicantly higher than in laminar zones. Hence the parameters of an earth-source geothermal system ought to be chosen such that turbulence occurs in the absorber pipes. However, economical aspects have to be considered too: increasing the pump performance to achieve turbulence causes higher costs for operation. Consequently, an economical optimisation requires an overall view, balancing ground-extracted and equipment-supplied energy. Heat transfer under laminar ow conditions depends on how long the uid stays in the absorber pipes, on the pipe diameter, and on the density, thermal conductivity and specic heat capacity of the absorber uid. Thus heat transfer between absorber pipe wall and uid increases with (a) (b) (c) (d) (e) (f) decreasing operating temperature decreasing concentration of the waterglycol mixture decreasing pipe length increasing pipe diameter increasing pump performance and ow velocity increasing temperature of the pipe wall.

(d )

(e) (f)

(g )

Heat transfer under turbulent ow conditions depends on Re and Pr, and on the diameter and length of the absorber pipes, but Re number has the greatest inuence. Heat transfer between absorber pipe wall and carrier uid increases or decreases qualitatively similarly to laminar conditions with one exception: decreasing the operating temperatures of the pipe uid reduces heat transfer. When laminar and turbulent ow conditions are compared directly, the following conclusions can be drawn. (a) Heat transfer under turbulent conditions depends signicantly more on the input parameters. Under laminar conditions Nu tends rather constant for all parameters. (b) Increasing the concentration of the waterglycol mixture increases the laminar zone, and hence reduces the turbulent zone and therefore heat transfer. If pure water is used as absorber uid, turbulence occurs in nearly all cases, and Nu number is about 15 times higher than for a waterglycol mixture. Nevertheless, in most cases an anti-freeze is unavoidable, because temperatures below 08C are possible during operation. It should be considered thatwhen using a heat pumpthe uid temperature at the vaporiser is still about 28C lower than the inow temperature. This involves the danger of freezing/icing in the heat pump. (c) Under laminar conditions the ow velocity is independent of the pipe diameter. It depends only on pump performance, pump efciency and pipe length, and on the ow parameters (kinematic viscosity and density).

DESIGN ASPECTS OF THERMO-ACTIVE SYSTEMS AND INSTALLATION SCHEMES Early ecological energy planning for building can in many cases prevent costly refurbishment and renovation in the future. High-quality energy design involves not only heating and cooling (rooms, water) but also lighting, and it requires a multi-objective optimisation. An optimised energetic-thermal design should also consider the seasonal heat loss from (un-)insulated slab-ongrade oors or basement walls. Far more energy and costs are expended in running an inefciently laid out building than in constructing an efcient one. A proper design should consider the efciency of the overall building process, including the sustainability of all elements. Proper design of energy foundations for a complex thermoactive system requires detailed ground investigation andat least for large projectsa numerical simulation of the entire system, including the secondary energy system within the building. Accordingly, the following parameters should be considered. (a) Hydrogeological ground properties, especially (i) depth and seasonal uctuation of the groundwater table (ii) ow direction and velocity of groundwater. (b) Geotechnical soil properties, especially (i) soil layering (ii) water content (iii) density and void ratio (iv) permeability (saturated and unsaturated hydraulic conductivity) (v) swellingshrinking behaviour (if intensive heat extraction is required) (vi) freezingthawing behaviour (if intensive heat extraction is required) (vii) shear parameters and stressstrain behaviour (for foundation design). (c) Geothermal soil properties, especially (i) thermal conductivity and heat capacity at specic temperature levels (ii) in situ ground temperature (iii) thermal gradient.

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES (d) Mineralogical composition and geochemical soil properties (only in exceptional cases). (e) Structural details of the building (i) type and size of energy foundations (length, width, thickness, diameter) (ii) depth of foundation below original ground surface and basement oor (iii) position, arrangement and spacing of foundation elements (iv) pile pattern (v) method of installation of foundation, construction sequence (with regard to absorber pipework installation) (vi) details of reinforcement, concrete properties of energy foundations. ( f ) Installation details of geothermal heating/cooling system (i) available space for connecting lines (ii) position of header block (distributor/collector) (iii) position of heat pump and technical service centre. (g) Building physics (i) insulation thickness of roof, walls oors (ii) size and quality of windows (iii) location and design of staircase (closed/open) (iv) design of heat bridges (v) temperature conditions in primary and secondary energy circuits. (h) Climate conditions, energy concept, and optional concept (i) monthly heating/cooling demand and peak demands within building (ii) temperature conditions in primary and secondary energy circuits (iii) type of heating/cooling system in building (iv) type/mixture and velocity of circulating uid (heat carrier uid) within energy system (v) heating/cooling intervals, operation plan. Most parameters are widely interacting. For instance, the installation of pipe coils within energy piles can mean that the most cost-efcient piling technique for the particular ground conditions may not be possible, and advice must be sought from an experienced piling contractor at the beginning of the design process. In practice it is frequently sufcient to supplement the standard site investigation with empirically established geothermal values as shown in Fig. 15. However, detailed geothermal investigation is recommended for large, complex projects. For general feasibility studies and pre-design of energy foundations the following assumptions can be made regarding the energy volume that can be extracted from thermoactive energy foundations (a) pile foundations with piles D 0.30.5 m: 4060 W per metre run (b) pile foundations with piles D > 0.6 m: 35 W per m2 earth-contact area (c) diaphragm walls, pile walls (fully embedding the soil): 30 W per m2 earth-contact area (d) base slabs: 1030 W per m2 earth-contact area. The heat that can be extracted from or fed into/stored in the ground depends on the maximum possible heat ux density in the absorber pipe system. There, the heat transport occurs by forced convection of the uid (usually an antifreezewater mixture). In order to optimise the absorber pipe system the following parameters have to be considered (a) diameter and length of pipes

97

(b) properties of pipe wall (roughness) (c) heat conductivity, specic heat capacity, density and viscosity of uid circulating in absorber pipes (d) ow velocity and ow conditions (laminar-turbulent) within absorber pipes. Figure 25 gives a schematic overview of the heat transport within a thermo-active system consisting of energy piles. It _ prim transported by heat carrier illustrates that the heat ux Q uid in the primary circuit is given by the specic heat _ prim and the temperature capacity cprim , the mass ow m difference Tprim: Complex ground properties and pile groups require numerical modelling of the geothermal heating/cooling system. Fig. 26 shows for example the temperature distribution within a section of a piled raft foundation during the winter and summer maxima. Figure 27 shows the daily mean temperatures in Vienna for the year 2001. Such data are needed to design a heatingcooling system whereby it is assumed that heating typically starts at external temperatures lower than 128C. This provides the heating period for the unsteady numerical models. Fig. 28 illustrates that the seasonal course of the air temperature can be simulated by a sinusoidal curve according to the following equation ! 2 (60) TGS Tm,out Tout cos t t P where TGS is the ground surface temperature, t is time, Tm,out is the average yearly temperature, Tout is the temperature amplitude, P is the duration period, and t is the phase displacement. In the end, the monthly heating and cooling demands have to be compared with the available output, as indicated in Fig. 29 (and Fig. 60). Moreover, the seasonal course of the absorber uid temperature (heat carrier uid temperature) should be predicted. Usually, a numerical simulation of the geothermal system is recommended for buildings with a heating and cooling demand of more than 50 kW. This rough value decreases to about 20 kW for buildings where rooms have to be cooled throughout the year. Geometric simplication may lead to signicant errors in heat calculation. Therefore three-dimensional analyses should be conducted. The simulation should comprise the expected inow and outow temperatures at the energy foundations and the temperature distribution in the ground. Numerical models and computer programs should be reliably calibratedthat is, on the basis of longterm measurements and experience from other sites, and on physical plausibilityotherwise wrong results may be gained, even from well-known suppliers. Experience has shown that the results are very sensitive to even small changes in the nite element mesh. Consequently, the importance of numerical simulations lies rather in parametric studies (to investigate the inuence of specic parameters) than in gaining exact quantitative results. Calculation of the temperature distribution in the ground due to energy foundations is increasingly being demanded by local authorities for environmental risk assessment. This refers mainly to possible inuences on adjacent ground properties and on the groundwater by the long-term operation of thermo-active deep foundations. Proper geothermal energy utilisation requires an interdisciplinary design, especially in the case of houses. The geotechnical engineer, architect, building equipment (sanitation) designer and installer, heating engineer and specialised plumber should cooperate as early as possible to create the most economical energy system. However, the tender for

98

BRANDL

Soil Heat flux to energy pile Ts 5 e.g. 1 10C . Qs

Primary circuit Heat flux to absorber pipes Heat transport to heat pump

Heat pump Heat exchange within heat pump

Secondary circuit Heat transport to heating installation Heat flux for room heating Troom5 e.g. 120C

. Qroom

T Prim in 5

e.g. 4C

T sec in 5

e.g. 135C

. Qprim . qW Tprim
T Prim out 5 e.g. 0C

. Qsec Tsec

T sec out 5 e.g. 130C

l pile l pile . . . Qs 5 #qsCpiled l pile # npipesqwCpipedLpipe 5 l pile . . 5 # npipes(Tw 2 Tprim)CpipedLpipe 5 Qprim 5cprimmprimTprim

PHP 5 PHP(HP ,HP)

. . Qprim 1 PHP 5 Qsec

. . . Qsec 5 csecmsecTsec Qroom

Analytical calculations FEM calculations

Heat pump technology Carnot 5 HP 5

sec T in sec prim T in 2 T out

Heating technology

. Qsec PHP

. Qprim 1 PHP PHP

where

. Q 5 heat flux (W) . q 5 heat flux density (W/m) C 5 circumference (m) l pile 5 pile length (m) Lpipe 5 length absorber pipe (m)
5 heat transfer coefficient (W/(m2K)) npipes 5 quantity of absorber pipes (units)

c 5 specific heat capacity (Ws/(kgK)) . m 5 mass flow (kg/s) T 5 temperature (C) PHP 5 electrical power heat pump (W) HP 5 efficiency factor of heat pump
HP 5 coefficient of performance for heat pump

Carnot 5 coefficient of performance of Carnot process

Fig. 25. Scheme of heat ux balance for heating in an energy pile plant

construction should clearly specify the individual performances on the site. It has proved suitable to entrust the geothermally experienced plumber with all details of the primary and secondary circuits, beginning with the mounting of the absorber pipe systems in the foundation elements. Figure 30 shows an installation scheme for heating with energy foundations (piles or diaphragm wall elements) and a heat pump comprising the primary energy circuit (in the ground) and the secondary energy circuit (in the building). Reversible heat pumps can achieve heating and cooling. Fig. 31 illustrates an installation scheme for free cooling. In this case the necessary energy input is limited to the electricity required to operate a circulation pump, whereby about 1 kWh of electricity is needed to obtain up to 50 kWh of cooling energy. The same uid, which cools as it passes through the absorber loops in the foundation, is pumped through the cooling system of the building. Coolness can be transferred into the building and distributed there by a ceiling, oor, or wall cooling system. PILOT RESEARCH PROJECT A: HEATING AND COOLING OF A REHABILITATION CENTRE Project Figure 32 shows the ground plan of a large rehabilitation centre, with areas A to J, and Fig. 33 depicts area F, where most of the energy piles were installed. The building has a

volume of 90 000 m3 , a usable area of 21 500 m2 , plus a swimming hall, a tness centre, etc. It comprises seven oors, two of them beneath the ground surface, and it had to be constructed on a slide-prone, unstable slope. Extensive ground investigations during the design stage revealed that the operating costs of this centre could be minimised by extracting/sinking geothermal energy for heating/cooling and by using groundwater from the subsoil for non-drinking purposes. The foundation work was performed in the years 1994 1995. The bored piles were excavated with a casing, and their structural integrity was checked by dynamic testing (TNO). In all 175 bored piles (D 1.2 m) were installed, having three functions: (a) foundation of the statically critical area F (b) retaining structures for the slide-prone slope (c) retaining walls for the 14 m deep excavation pit. Owing to the sloped surface of the building area and the deep-seated collector galleries, the piles had to be installed at rather different levels (Fig. 34). The pile depth varied according to static requirements and local soil characteristics: 911 m for the foundation piles and 918 m (mean value 14 m) for the retaining piles. Most of the retaining structures had to be tied back with prestressed anchors (Fig. 34).

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

99

26 24 22 20 Temperature: C 18 16 14 12 10 8 0 566 113 170 226 283 339 396 452 429 275 352 509 566 622 679 429 120 198 Width: m 2426 2224 2022 1820 1618 1416 1214 1012 810

Length: m

(a)

14 12 10 Temperature: C 8 6 4 2 0 0 566 113 170 226 283 339 396 452 429 429 120 198 275 Width: m 352 509 566 622 679 1214 1012 810 68 46 24 02

Length: m

(b)

Fig. 26. Detail of temperature distribution around energy piles: results of numerical calculations for a pile group used for (a) cooling in summer and (b) heating in winter

Of the piles, 143 are tted with heat exchangers and hence act as energy piles, as indicated in Fig. 33. The piles of the retaining structures/walls also form a part of the heating/cooling system, as shown in Fig. 34. The relevant technical data are as follows (a) (b) (c) (d) (e) (f) (g) (h) 2050 m run of energy piles HDPE absorber pipes: do 25 mm, di 20 mm 260290 m absorber pipes per pile absorber uid: 70% water, 30% antifreeze L required absorber uid passage: 77.4 m3 /h condensation capacity of heat pump: 270 kW minimum absorber uid inow temperature: +18C minimum absorber uid return ow temperature: +48C.

Some 40 000 m of polyethylene piping were tted in the reinforcement cages of the piles. Throughout the construction period the absorber units were pressurised at 8 bar so that any defects could be detected immediately. The pipes leading from the piles to the distributors were placed beneath the raft of the building and behind the pile walls. Ground properties The subsoil consists of weathered talus (mainly loam) down to 25 m below original ground level. It varies from silty sand to clayey silt, and is underlain by tertiary sediments without a clear interface. The piles are fully embedded in these sediments, which are weathered (oxidised)

100
30 Mean daily outdoor temperature: C 25 20

BRANDL (g) residual angle r > 108 (h) compressive strength qu 20500 kN/m2 . The density and strength-deformation properties of the soil improve with depth. The permeability coefcient k of undisturbed silty soil samples varied between 108 and 1010 m/s. The overall hydraulic conductivity was much higher owing to joints and sandy interlayers, generally about k 5 3 107 m/s and locally k 104 m/s. The mineral composition of the nes (siltclay) varied as follows: mica group, 2030%; montmorillonite, 1030%; quartz, 1020%; calcite, 1020%; chlorite, 510%; kaolinite, 510%. Groundwater was found 45 m below the original surface, though not as a continuous aquifer but locally in sandy interlayers and joints. Accordingly, predominantly local water ingress occurred during piling and excavation of the 14 m deep pit. The maximum was about 20 l/s during the construction stage. This groundwater situation required a drainage blanket beneath the (piled) raft foundation. Longterm monitoring disclosed a steady state of about q 7 l/s as a mean value over the year. This is that portion of groundwater from the slope that is collected in the drainage systems along the retaining structures and beneath the foundation raft. It is used to a high percentage for the rehabilitation centres non-drinking water purposes. Measurements The (long-term) stability of the retaining structures has been monitored since the initial construction period by inclinometers in the piles and by pressure cells on the anchor heads. The pipes of the absorber system have been monitored with pressure gauges. One load-bearing pile of the piled raft foundation was tted with the following measuring devices (a) pressure cells on pile toe and head (b) ssuremeters at three levels (c) thermo-elements at ve levels. The aim of the measuring pile (Fig. 33) was to investigate the effects of temperature changes within an energy pile on its bearing capacity, and especially on its shaft resistance during hydration and subsequent energy extraction. Moreover, the inuence of natural temperature uctuation in the ground should be monitored.
35 Heating 30 231 24 246 207 178 5000 127 153 Heat carrier fluid (return flow) 186 25 20 15 10 Temperature of heat carrier fluid: C

01 May

01 Sep

5 0

25 210 215

Heating period

Heating period

Fig. 27. Mean daily outdoor temperatures in Vienna, 2001. If temperature drops below 128C, room heating is usually necessary

Mean daily outdoor temperature: C

30 25 20 15 5 0 25

Vienna 2001 (measured) Sinusoidal curve

Start of simulation Amplitude:11 Amplitude:11

Annual mean 10 temperature:11

210 1 May 1 Sep 1 Nov 1 Aug 1 Mar 1 Feb 1 Apr 1 Jan 1 Jun 1 Oct 1 Jul 215

01 Dec

01 Aug

01 Nov

01 Mar

01 Feb

01 Jan

01 Jun

01 Apr

01 Oct

01Jul

15 12 10

Time 1 Dec
229 10 000

Fig. 28. Mean daily outdoor temperatures in Vienna, 2001, with idealised sinusoidal curve for numerical calculations

and locally severely jointed to a depth of 1015 m. They consist predominantly of clayey to sandy silt, with a natural water content near the plasticity limit. The main soil characteristics are (rounded) (a) (b) (c) (d) (f) natural water content wn 2540% plasticity index Ip (0) 1035% density rd 1.31.7 g/cm3 moduli (oedometric) E 535 MN/m2 friction angle 25358
20 000 Heating/cooling demand Heating/cooling output 15 000 Heating/cooling: kWh/month

Au g

Se p

Fe b

Ja n

Ju n

Ju l

M ay

25000 Cooling 210 000

N ov

D ec

M ar

Ap

O ct

67

71

8 5 0 25

Fig. 29. Example of energy demand and output for heating and cooling (annual distribution) of a building founded on energy piles. Temperature of heat carrier uid is also shown

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

Floor 1 Floor 2 Floor 3 Floor 4

101

Cooling machine Expansion Buffer tank Pressure vessel

Fluid-based cooling of building (concrete core cooling)

Blower Cool air

Manifolds

Expansion

Air-based cooling (air-conditioning) of building

Ground

Groundwater (Temp. 1025C) Concrete absorber piles, diaphragm walls, barrettes

Fig. 30. Heating of buildings with energy foundations and heat pump: installation scheme with primary and secondary circuit. With reverse heat pumps this energy system can also be used for cooling

The head of the measuring pile is situated in an easily accessible place, which facilitates long-term monitoring. The pile length is 9 m and the calculated total load Q 2000 kN. In fact, up to now the maximum theoretical life load has not yet been imposed, so that the measured head loads are clearly smaller. Moreover, part of the load is transferred directly from the piled raft into the subsoil. The rst testing phase was conducted during the winter of 1996/1997 and provided initial data that could be used to optimise the absorber system. The rehabilitation centre was opened in late 1997, and the energy piles have been in full operation for the oor heating since then. Figure 35 shows the load changes on the head and toe of the measuring pile against time. The rough structure of the building was completed in the autumn of 1996, providing about 90% of the dead load. The initial results illustrate that residual stresses were imposed on the pile before any loading, caused by heat development in the fresh pile concrete due to hydration: thermal contraction after peak temperature caused a temporary reduction of the base pressure in the pile toe. To a certain extent this continued for a short period and, simultaneously, shaft friction was increasingly mobilised. Over a long-term period the point load in the base of the pile, Qb , remained constant, independently of the increase in

the total load and temperature variations within and around the energy pile. This proved to be the rst evidence that a proper operating of energy piles has no relevant inuence on the shaft resistance, as could then be conrmed on several other sites and by numerical modelling. In Fig. 36 the temperature uctuations during the rst years are plotted against pile depths. The rst measurement was performed during the hydration phase of the pile concrete and showed a temperature of up to 608C. The second measurement (27 September 1995) provided typical summer results and the third one (12 February 1996) winter results, whereby the measuring pile was already loaded by four oors. At that time, only the concrete structure was under construction, and no temperature insulation had been installed. This explains the value of 08C on the pile head. Measurement No. 4 was taken during a test phase of operating the energy piles, after the structure was nished at that time. The measurement on 9 February 1998 was taken in the rst winter period during excessive heat extraction for testing purposes. The minimum pile temperature was close to the allowable limit of +28C. Since autumn 1997 the energy piles have been under full operation. Long-term monitoring has revealed typical seasonal temperature uctuations with relatively large amplitudes

102

BRANDL
Floor 1 Floor 2 Floor 3 Floor 4

Heat exchanger

Buffer tank

Pressure vessel

Fluid-based cooling of building (concrete core cooling)

Manifolds

Expansion

Ground

Groundwater (Temp. 1015C) Concrete absorber piles, diaphragm walls, barrettes

Fig. 31. Free cooling of buildings with energy foundations and circulation pump: installation scheme with primary and secondary circuit

40 15

30

40 Energy piles

G
34

D C

A J

21

120 m

Fig. 32. Ground plan of the rehabilitation centre at Bad Schallerbach, Austria, with foundation areas A to J and part of the retaining walls, tted with energy piles. Area F rests on a piled raft foundation; the other areas are founded only on rafts

of maximum and minimum pile temperatures in summer and winter, as intended for this pilot project. Consequently, the pile temperatures in the subsequent winter were somewhat higher than in the quasi-steady state before. Moreover, Fig. 37 shows that daily outdoor mean values are needed for a reliable interpretation. Single outdoor temperature measurements parallel to pile temperature measurements are not sufcient (indicated in the diagram from April 2004 to December 2005). In Fig. 38 some temperature curves are selected as examples of heating and cooling periods in relevant years. Immediately after commencing, continuous operation of the geothermal system started (with excessive heating); the year 2003 had a very high cooling demand, and there was a quasi-steady state in the year 20042005. Figure 39 shows the total strains, comprising load-induced and temperature-induced parts. The load inuence dominates in the upper zone of the pile and diminishes toward the pile toe. Consequently, the shaft resistance in the lower part of the pile is quite small. The skin friction distribution along

81

34

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

Creeping slope A Section 22 Shotcrete arches Inclinometer

103

Pile no. 116 Section 4 4 152

Bored piles (D 5120 m)

Pile no. 29 l1 5 69 m

20

Pile no. 69

Section 11 (West)

Section 33

342 Energy piles (D 5 120 m)

l2 5 69 m Section 11 (East)

Measuring pile

60 A

Fig. 33. Detail of Fig. 32 with energy piles (black) and longitudinal sections of the energy retaining walls. Also shown is the location of the measuring pile and the piles with inclinometers (and temperature sensors) for monitoring. Piles without absorbers (white) for reference

184

Capping beam Continuous beam 139 Multistorey building

Silt (clayey, sandy)

270

30 280 Pile depth variable

2110

Prestressed anchors Tallow. 5 750 kN lA 5 2738 m

2130

Piles for retaining structures D 5 12 m

Foundation piles D 5 12 m

Fig. 34. Partial view of anchored energy pile wall on uphill side of building, being part of the heating/cooling system. Corresponds to cross-section AA in Fig. 33

the pile shaft was back-calculated from ssuremeter measurement data by approximately assuming Hookes law of elasticity. Between the retaining structure and building an accessible gap has remained. It avoids horizontal forces from the creeping slope onto the building and enables long-term monitoring of the tied-back retaining wall (including thermal effects). Thus the inuence of excessive heat extraction

from the ground could be clearly observed along sections 11 of the piled retaining structure (Fig. 33). Operational temperatures between 28C and 38C (temporarily even 58C) caused the formation of ice lenses in the ground and thus a frost heave, H, of the surface behind the piles. The maximum of H 15 cm was observed near the collection shaft, and it decreased with distance from the heat extraction sources, similar to Fig. 11. After stopping

104
1200 1000 800 Load, Q: kN 600 400 200 0 29.04.1995 Test run of energy system Full operation of energy system Pile weight Qhead

BRANDL
0 21 22 23 24 25 26 27 28 29 0 Intensive heating
09.02.1998 25.08.2003 01.02.2004 14.06.2004 09.08.2004 24.08.2004 18.10.2004 07.02.2005

Temperature: C 10

15

20

Depth: m

Qbase 29.04.1996 Time 29.04.1997 29.04.1998

Long period of intensive cooling

Fig. 35. Loads on head and toe of measuring energy pile against time at different construction stages. Permanent, full operation of energy pile system since autumn 1997; previously, only temporary test runs of energy system
Temperature: C 10

Fig. 38. Temperature within energy pile against pile depth; data from long-term operation, illustrating seasonal uctuation

201 15
27.09.1995 12.02.1996 29.04.1997 09.02.1998 29.01.1999

Strain, : 0 21 22 23 24 25 26 27 28 29
Load Temperature Load 1 temperature

101 1

0 21 22 23 24 25 26 27 28 29

20

Depth: m

Fig. 36. Temperature within energy pile against pile depth at different construction stages (see Fig. 35) and during initial operation period

Fig. 39. Strains in measuring pile between 12 February 1996 and 9 February 1998. Total values and load- and temperatureinduced values

this improper test run of the energy system, the temperatures increased again, also favoured by warmer weather. Figs 40 and 41 show the temperature proles along the depth of two piles and in the subsoil behind the pile wall. A comparison of the temperature proles along the pile embedment illustrates the signicant inuence of the groundwater ow, which is much stronger around pile No. 29 than No. 69. Also clearly visible are the inuence of air temperature above the embedment of the piles and the
30 25 20
Temperature T : C

temperature decrease in the soil behind the retaining structure. The deep temperatures were caused partly by the low air temperature in the winter and partly by the low operation temperature in the energy piles. Hydration of fresh pile concrete Measurements within the research programme for deep energy foundations have also shown that the hydration of

15 10 5 0 25

00 25 Pile temperature 45 at depth [m] 65 90 Outdoor (daily average) Outdoor (singular)

Fig. 37. Outdoor temperature near rehabilitation centre and temperature within energy pile (measuring pile in Fig. 33). Strong heatwave in summer 2003; more normal temperature distribution in 2004; cold summer and warm autumn in 2005

Feb.03 Mar.03 Apr.03 May.03 Jun.03 Jul.03 Aug.03 Sep.03 Oct.03 Nov.03 Dec.03 Jan.04 Feb.04 Mar.04 Apr.04 May.04 Jun.04 Jul.04 Aug.04 Sep.04 Oct.04 Nov.04 Dec.04 Jan.05 Feb.05 Mar.05 Apr.05 May.05 Jun.05 Jul.05 Aug.05 Sep.05 Oct.05 Nov.05 Dec.05

210

Depth z: m

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES pile or diaphragm wall concrete causes a heat development, which reaches its maximum within 1524 h after casting the concrete. In large-diameter cast in situ piles, temperatures up to 708C develop. The subsequent decrease of concrete temperature can last from several days up to one month, until the ground temperature is reached. Temperature increase and decrease depend on the thermal and hydrological properties of the surrounding ground, and on the diameter of the pile. Furthermore, they are strongly inuenced by the pile concrete with regard to its composition, sort and grinding neness of cement, and additives (e.g. plasticiser, retarders). Hydration of the cast in situ deep foundation may cause a temporary decrease of the base pressure due to thermal contraction of the young concrete. Increasing external loads then lead to an increase of the base pressure once more. Shrinkage due to the hydration of in situ cast concrete piles may cause thermal strain-induced cracking. Therefore tension piles should have sufcient reinforcement. Figure 42 illustrates the alteration of the shaft resistance in the initial phase of a large-diameter bored pile. When casting the concrete, hydrostatic stress conditions exist along
Temperature, T: C 20 15 10 5

105

25

0 Multistorey building 2 4 Depth z: m 6 8 10

Local frost heave

the pile depth, and the shaft resistance is practically negligible (Fig. 42(a)). The following hydration of the young concrete causes shrinkage, and hence a contraction of the pile. In homogeneous ground this leads to a distribution of skin friction according to Fig. 42(b), if the pile weight is neglected. Superposing the effects of shrinkage and pile weight leads to a reduction of negative skin friction along the lower part of the pile and to an increase in positive skin friction along the upper part (Fig. 42(c)). Fig. 42(c) gradually develops into Fig. 42(d) when external loads are transferred into the pile. In Fig. 42 the vertical displacement of the pile head and the movement of the pile base are also sketched, on an enlarged scale. When comparing Figs 42(a), (b) and (c), those measuring results become clear, which show a temporary decrease of the base pressure shortly after casting the pile, and then a gradual increase due to external loads (e.g. Fig. 35). The timebase pressure curve of a cast in situ concrete pile that exhibits strong thermally induced contraction is shown schematically in Fig. 43. This scheme depends not only on physical and thermal concrete and ground properties and on external loads, but also on the slenderness of the piles.

Prestressed anchors

27.2.1998 3.3.1998 16.3.1998 1.4.1998 14.5.1998

12 14 16 Strong groundwater flow q 5 6 l/s

Fig. 40. Cross-section along pile to No. 29 (see Fig. 33) with depthtemperature diagram (in pile centre). Also indicated is the amount of groundwater ow in wall section 11, west

25

Temperature, T: C 20 15 10 5

0 Depth z: m Multistorey building 2 4 6 8 10

T: C
15 10 5

Local frost heave

PILOT RESEARCH PROJECT B: FIRST THERMOACTIVE TRAFFIC TUNNEL (ENERGY TUNNEL) Overview To upgrade the railway line between Vienna and western Europe for four-track operation, a tunnel through the northern Vienna woods, a hilly landscape consisting of ysch and molasse formations, is being built. The core section is the 12.8 km long Lainzer Tunnel, which partly also serves as large energy-absorbing tube. A tunnel activates a signicantly larger quantity of usable geothermal heat than deep foundations. The energy can be used for heating and/or cooling of railway stations, administration and residential buildings, and for keeping platforms, bridges, passages etc. free from ice in winter. Consequently, shallow tunnels (especially in cut and cover) make a wider application possible than deep-seated tunnels, because the heat transfer between source and user is easier. However, deep-seated road or railway tunnels may be used simultaneously for heat transport if hot groundwater occurs. But this represents another technology, completely different from thermo-active tunnel linings based on energy exchange by means of earth-contact structures. The Lainzer Tunnel has been constructed in several sections and by different methods
Q qs: kN/m2 qs: kN/m2 1 W W 1 qs: kN/m2 1

x: m
10 10

27.2.1998 3.3.1998 16.3.1998 1.4.1998 14.5.1998

12 14 16

Poor groundwater flow q 5 06 l/s

2 (a) (b)

2 (c) (d)

Fig. 41. Cross-section along pile to No. 69 (see Fig. 33) with depthtemperature diagram (in pile centre). Also indicated are temperature uctuations behind pile wall, and groundwater ow in wall section 11, east. Prestressed anchors not drawn

Fig. 42. Shaft resistance during initial phases of a cast in situ large-diameter bored pile (schematic; for homogeneous ground): (a) initial state, concrete just cast; (b) hydration-induced contraction (self-weight neglected); (c) hydration-induced contraction and self-weight; (d) as (c), plus initial loading Q

106

BRANDL cages of piles, diaphragm walls or base slabs could not be used. Therefore a completely new technology had to be developed.
Q50 Q.0 l 5 pile length c 5 unit weight of fresh concrete

Base pressure, qb: kN/m2

Hydration c.l

Selfweight W

Construction period W 1 dead load Qd Time, t

W 1 total load Qt

Fig. 43. Timebase pressure curve of a cast in situ concrete pile exhibiting relevant initial contraction due to hydration

(a) cut and cover, consisting of large-diameter bored piles, reinforced concrete base slab and roof (b) New Austrian Tunnelling Method (NATM), with a primary support of reinforced shotcrete, rockbolts and anchors, and a secondary lining of reinforced concrete. To optimise the energy design of the tunnel, and for research purposes, the following geothermal projects were carried out. (a) Energy plant: LT24, HadersdorfWeidlingau. The section LT24 was selected as a test plant, to investigate both the technical and the economic aspects. In this section the tunnel was constructed using the cut and cover method, allowing the application of already proven absorber techniques combined with structural engineering. (b) Energy well: Hetzendorferstrae. Another test plant was constructed in order to investigate the performance of energy wells that were simultaneously used for lowering the groundwater level. (c) Energy membrane: LT22, Bierha uslberg. The section LT22 is located in an NATM section, where the proven system of attaching absorber pipes to the reinforcement

Cut and cover tunnel In section LT24 the primary side wall lining of the tunnel consists of bored piles, whereby each third pile is used as an energy pile (Figs 44 and 45). Thus the energy plant LT24: HadersdorfWeidlingau comprises 59 bored piles with a diameter of 1.2 m and an average pile length of about 17.1 m. The intermittent pile wall exhibits jet-grouting columns between the piles. Pile excavation (by grab) was supported by casings using rotating equipment. The energy piles are equipped with absorber pipes connected to collector/distributors, which are located at a central point of the tunnel. The pipes leading from the piles to the collector/ distributors are placed alongside the cover of the tunnel. The connecting pipes lead into a collector/distributor room that is easily accessible on top of the cut and cover tunnel. Fig. 46 shows the header block with the collector/distributor for all collecting pipes. The manometers allow a detailed watertightness check of all absorber pipes. A manifold with a diameter of 150 mm connects the collector/distributors with heat pumps in an adjacent school in order to heat the building. Table 4 gives the relevant technical data. Preliminary calculations yielded an extractable thermal power of about 150 kW in the long term. In one heating period an energy amount of 214 MWh can be gained. Furthermore, the benets of this new energy concept are that it is both environmentally friendly and economical: the reduction of natural gas of 34 000 m3 per year leads to a decrease of annual CO2 emissions of 30 t. Furthermore, annual savings in operation costs of A10 000 will be achieved, compared with the old natural gas heating system of the school building. The plant was constructed as a demonstration project in the context of a major research initiative by the Austrian government. Because of this scientic background, the plant is intensively instrumented with measurement devices. Six energy piles are tted with 18 temperature gauges at different levels; additionally, one pile is tted with combined straintemperature gauges at ve levels for measuring strains and temperature. The aim of this measuring system is

Fig. 44. Schematic cross-section of energy plant LT24, HadersdorfWeidlingau. One side wall of cut-and-cover tunnel used as energy wall

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

Soil 150 Jet grouting d~80 cm S-07-20
D1-A T1

107

Table 4. Technical data of the energy plant LT24: Hadersdorf Weidlingau

150

S-07-21

S-08-01

Annual heating output: MWh Energy piles Mean length: m Heating capacity: kW Required antifreeze passage: m3 /h Absorber pipes Absorber circuits Absorber pipes in piles, total length: m Connecting lines, total length: m

D1-I

214 59 piles 17.1 150 51.6 HDPE, do 25 mm, di 20 mm 80 units 9709 13 754

Absorber pipes

Inner lining d~45cm Tunnel

Dowels

Shotcrete

Fig. 45. Detail of Fig. 44: longitudinal section through energy tunnel wall. The primary lining consists of bored piles with jetgrouted columns in between and is connected to the watertight secondary lining by dowels. Every third pile is equipped with absorber pipes that are situated behind reinforcement bars and thus protected from damage through dowel installation for the inner lining (reinforced concrete panels). Also shown is the location of measurement instruments in pile S-07-20

perature surrounding the energy plant is also registered (at different distances). Temperature differences between energy piles (thermo-active and bearing function) and standard piles (bearing function only) have been checked by heat picture photographs. The differences can be registered even after the placement of the secondary lining, that is, the reinforced concrete cover (Fig. 47). The hydration heat caused by concreting the secondary tunnel lining could be clearly observed in the energy piles. Furthermore, the base slab of the tunnel had a signicant temperature separation function (Fig. 48). Fig. 48 also shows thateven without energy operationtemperature differences exist between the pile head and toe, and between the inner and outer side of the pile skin. The operation of the energy plant started in February 2004, and during the rst testing phase initial data were obtained that could be used to optimise the absorber system. About 40 MWh of heating energy could be extracted from the energy piles during the rst six weeks of operation. Since autumn 2004 the energy system has run permanently for a school near the tunnel. The external air temperature (outdoor temperature) is used as criterion for regulating the energy system. Down to 58C the school building can be fully heated with ground source energy. At lower temperatures the existing gas boiler furnace is added. Figure 49 shows for example the specic strain in measurement pile S-7-20 at different times plotted for the inner and outer side and the central axis. The zero reading was before soil excavation in front of the pile wall. Therefore the curves include both mechanical and thermal effects, as can be clearly seen from the seasonal differences. The tempera-

*>100C 100 90 80 70 60 50

Fig. 46. Header block to Fig. 44 where the connecting lines and manifolds are collected (see also Fig. 9)

40 30

to investigate the effects of temperature changes within an energy pile on its bearing capacity and the temperature uctuation in the energy piles during operation. Moreover, heat carrier uid passage, total extracted heat, and temperatures in the manifold are monitored. The groundwater tem-

20 *<20C

Energy piles

Fig. 47. Photo taken with heat camera, showing (equally spaced) colder areas of inner tunnel lining caused by energy piles

108
Continuous operation High energy demand 22.01.04: Start of data logging 25 05.02.04: Start of energy operation 19.02.04: Continuous operation

BRANDL
Continuous operation Low energy demand Continuous operation High energy demand

t 5 100 m; inner 11.10.04: Manifolds thermal insulated Temperature above base slab (inner side of pile) t 5 100 m; outer t 5 500 m; inner t 5 500 m; outer t 5 900 m; inner t 5 900 m; outer t 5 1400 m; inner t 5 1400 m; outer t 51800 m; inner t 5 1800 m; outer

20

Temperature: C

15

12.02.04: Concreting of inner lining

10

Power failure

0 22.01.04 05.02.04 19.02.04

Nearly same temperatures within pile because of energy operation 04.03.04 18.03.04 01.04.04 15.04.04 29.04.04 13.05.04

Temperature above base slab (outer side of pile)

Temperature below base slab

time 10.06.04 24.06.04 08.07.04 22.07.04 05.08.04 19.08.04 02.09.04 16.09.04 30.09.04 14.10.04 28.10.04 11.11.04 25.11.04 09.12.04 800

Fig. 48. Temperature distribution at inner and outer side of an energy pile at different depths (see Fig. 44)

27.05.04 2200

2800

2600

2400

Relative strain, 200 400 600 0

0 1 2 3 4

Cover slab

Depth below pile head: m

5 6 7 8 9 10 11 12 13 14 15 16 17 18

Base slab

06.03.03-inner 06.03.03-outer 06.03.03-centre 13.08.03-inner 13.08.03-outer 13.08.03-centre 22.01.04-inner 22.01.04-outer 22.01.04-centre 24.06.04-inner 24.06.04-outer 24.06.04-centre 20.08.04-inner 20.08.04-outer 20.08.04-centre

Fig. 49. Relative strain along inner and outer (earth) side and along centre line of an energy pile

ture-induced deformations are signicantly smaller than those caused by earth pressure, and the natural uctuation of the tunnel temperature has a greater inuence than the temperature changes due to energy extraction/storage in the energy piles. Energy operation even creates a more uniform temperature in the piles, that is, smaller temperature differences between pile head/toe and inner/outer side. This thermal balancing-out reduces the temperature-induced moments in the energy piles. Of course, the energy operation causes a stronger cooling or heating of the piles, but this occurs uniformly, and hence causes a volumetric deformation

without constraints, and therefore no additional load on the structure. NATM section The utilisation of tunnels excavated according to the New Austrian Tunnelling Method or sprayed concrete method for heat extraction/storage requires special absorber elements. The construction sequences make the installation of continuous absorber pipes in the longitudinal direction rather complicated.

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES In principle, three structural elements can be used for thermal energy extraction/storage (i.e. heating/cooling; Fig. 50) (a) energy anchors or nails that thermally activate the surrounding soil or rock (b) energy geosynthetics that make use of the ground around the tunnel circumferencemainly non-woven geotextiles and geocomposites (Fig. 51), but also some geomembranes (c) thermo-active secondary lining (inner reinforced concrete).

109

PILOT RESEARCH PROJECT C: COOLING AND HEATING OF METRO STATIONS Until now the rooms of stations on the Vienna Metro have usually been actively air-conditioned (monitoring rooms, transformer rooms, switchrooms, storage rooms, shops etc.). District heating or electric current serve for heating, and cooling is achieved by electric refrigeration systems. Gas is excluded for safety reasons. The extension of the Metro line U2 offered the chance to supply four stations with geothermal energy from earth-contact structures for both heating and cooling purposes. These projects represent the rst fullscale application of the thermo-active technology in metro engineering worldwide. Hence comprehensive measurement and monitoring systems were installed. They serve for optimisation and quality assurance (similar to the observational method), for maintenance control and for further research. The design was based on previous experience with geothermal energy systems. The stations were constructed by the cut and cover method. Consequently, absorber pipes are situated in diaphragm walls, in bottom slabs and between the primary and secondary lining of the station tunnels. Feasibility studies, including numerical analyses of tempera-

Fig. 51. Energy geotextile (geocomposite) installed in an energy uselberg) tunnel (testing plant LT22 Bierha

ture ow in the ground, revealed that geothermal energy extraction/storage would have only a very limited inuence on the soil close to the absorber elements (i.e. within a few metres). Unfavourable thermal effects within the surrounding soil and groundwater will not occur. Therefore the project passed all legal proceedings regarding environmental impact, compatibility and risk assessment without problems, and it was greatly supported by public opinion. The rst station was U2/2-Taborstrae, where signicant waste heat is created, causing very high room temperatures all the year round. Therefore a geothermal cooling system involving 1865 m2 of energy diaphragm wall and 1640 m2 of energy base slab was designed. If the surplus energy cannot be used for heating, it is transferred into the soil via the absorber system, thus avoiding noisy or unsightly outdoor cooling towers. Thermal simulations revealed that the heat carrier uid will have temperatures between 108C and 288C. A total uid volume (waterglycol mixture) of about 10 m3 circulates in the absorber system, providing a maximum cooling capacity of 81 kW. In the design the following energy demand of the Metro station was assessed (a) (b) (c) (d) maximal heating 95 kW maximal cooling 67 kW average yearly heating energy 175 MWh average yearly cooling energy 437 MWh.

Fig. 50. Scheme of energy tunnel excavated with the New Austrian Tunnelling Method

Figure 52 shows the cooling capacitytime histogram and the seasonal uid temperature in the absorber pipes. It shows that sufcient energy (heat) can be transferred into the ground to achieve yearly cooling of the entire Metro station. Figure 53 gives a section of the Metro station, and also indicates the ground prole and the instrumentation in a measuring panel of the diaphragm wall. The right wall serves for heating, and hence for energy extraction from the ground. The standard measuring programme for all Metro stations with geothermal heating/cooling systems comprises temperature sensors installed in the diaphragm walls, in the tunnel tubes, and 5.0 m below the bottom slabs. These devices serve not only for quality control and safety assessment but also for regulating, controlling and optimising the operation of the geothermal system. In addition to this standard equipment, several measuring devices were installed to investigate the thermal inuence on

110
50 45 40 Cooling capacity Cooling demand
329 43 438

BRANDL
40
41 45 451 45 454 396 38 39.2 39

Cooling: MWh/month

35 30 25 20 15 10 5 0 Jan. Feb. Mar.

17175 102 20 199 141 28 291 167

32

231

267

27.5

280

284 253 225 23 234

30 25 20
149 13 139

209

Temperature heat carrier fluid

15 10 5

Apr.

May

Jun.

Jul.

Aug.

Sep.

Oct.

Nov.

Dec.

Fig. 52. Energy demand/capacitytime histogram and seasonal temperature of heat carrier uid at metro station U2/2-Taborstrae

Ground profile: Manmade fills Second part of reinforcement 1445 Sandy silt Quaternary sediments GL Sandy gravel Clayey sand, silt overlap

Diaphragm wall with measurement instruments W2-4 W1-2

D1L, D1E D2L, D2E

Temperature of heat carrier fluid: C

424

35

GW
D3L, D3E

D4L, D4E

W8-10 W6-8 W3-6

D5L, D5E

First part of reinforcement

Track 1 direction Karlsplatz

D6L, D6E D7L, D7E D8L, D8E

Track 2 direction Stadion

Tertiary sediments

Silty sand bottom slab

D1L strain gauge measurement depth 1, inner side T12E temperature gauge measurement depth 12, earth side W1-2

D9L, D9E

D10L, D10E

W11-12 W10-11

Tein, Taus

Connection point
D11L, D10E

extensometer-segment from measurement depth 1 to 2

T12L, T12E

D 5 80cm

Energy extraction by energy diaphragm wall

Fig. 53. Cross-section of U2/2 metro station Taborstrae with ground prole and thermo-active diaphragm walls (energy diaphragm walls). Monitoring equipment installed in one panel of diaphragm wall

the loaddeformation behaviour of the underground structure, and especially of diaphragm wall panels. The measuring panel in Fig. 53, for instance, was equipped with (a) 21 strain gauges to register tension and compression strains and temperature along the front and rear zone of the diaphragm wall on the supporting points and in the middle between them (b) chain extensometers consisting of seven segments to measure the longitudinal deformations between the individual support points and below the bottom slab, down to the toe of the diaphragm wall (c) temperature sensors to measure the temperature on either side of the diaphragm wall (d) temperature sensors to monitor inow and return-ow temperatures of both absorber circuits. Already it can be concluded that geothermal absorber

systems for cooling and heating Metro stations and lines represent an innovation that is not only environmentally friendly but also highly economical. FURTHER CASE HISTORIES Cooling and heating of a paper-processing plant This case history refers to the piled raft foundation of a paper-processing plant where 400 employees are printing daily 50 000 m2 of paper and other materials. Such plants require special temperature and humidity conditions within their production halls. The overall structural volume of this plant is about 130 000 m2 . The subsoil consists of weak silt, underlain by sandy gravel. The deep foundation comprised groups of two to six energy piles in a different pattern according to structural requirements; pile spacing within the groups is typically 1.4 m. The external walls are founded on single piles at

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES 111 variable spacing of 1.44.0 m. In all 570 driven reinforced before the energy system was installed and after its operaconcrete piles (0.4 m 3 0.4 m) were installed, thus thermally tion was started. It shows clearly that the increase due to activating a heat storage volume of about 100 000 m3 in the cooling exhibits its maximum in the centre of the building ground. The plant was opened in autumn 1995, and the but decreases with depth. energy system has been running without problems since that Figure 55 shows the uctuation of used and generated time. energy during the rst heating period in winter 1995/1996. Cooling and heating are based on multiple energy systems The diagram clearly shows that the portion of groundcomprising not only the energy piles but also cooling with extracted energy using the absorber piles prevails. In all, near-surface earth collectors andto cover peaksan ice 95 000 absorber pipes (d 20 mm) provide 520 kW of coolstorage refrigeration machine. For heating the building, at ing and heating. They have been positioned in the top 14 m rst the waste heat from compressors and from the printing of the 24 m long piles. machine cooling system is used, then the energy piles and Cooling of rooms, structural elements and machines in the heat pump are activated, and, nally, only peaks are this industrial plant also requires a relatively high amount of covered by fossil fuel (oil) heating. energy. In comparison to Fig. 55, Fig. 56 shows the cooling Figure 54 shows the temperature uctuation in the subsoil energy generated within the same period. The portion of
Winter 94/95 Summer 95 Winter 95/96 Summer 96 Depth
50 m 100 m 145 m

Jul. 95 Aug. 95

Sep. 95 Oct. 95

Dec. 94 Jan. 95

Nov. 95 Dec. 95

Aug. 96 Sep. 96

Feb. 95

Nov. 94

Feb. 96

Jun. 95

Jan. 96

Jun. 96 Jul. 96

(a) 180 170 160 150 140 130 120 110 100 90 80

Nov. 96 Depth
50 m 100 m 145 m

Mar. 95 Apr. 95 May 95

Mar. 96 Apr. 96

May 96

Temperature: C

Jul. 95 Aug. 95

Aug. 96 Sep. 96

Sep. 95 Oct. 95

Feb. 95

Feb. 96

Jun. 96 Jul. 96

Oct. 96 Oct. 96

180 170 160 150 140 130 120 110 100 90 80

Temperature: C

Start of operation

Nov. 95 Dec. 95

Nov. 94

Dec. 94 Jan. 95

(b) 150 140 130 120 110 100 90 Jul. 95 Aug. 95 Aug. 96 Sep. 96 Sep. 95 Oct. 95 Feb. 95 Feb. 96 Jun. 96 Jul. 96 Nov. 95 Dec. 95 Nov. 94 Dec. 94 Jan. 95 Nov. 96 Jun. 95 Jan. 96 Mar. 95 Apr. 95 May 95 Mar. 96 Apr. 96 May 96 Oct. 96 80

Nov. 96 Depth
50 m 100 m 145 m

Jan. 96

Jun. 95

Mar. 95 Apr. 95 May 95

Mar. 96 Apr. 96

Temperature: C

(c) 150 Temperature: C 140 130 120 110 100 90 Jul. 95 Aug. 95 Aug. 96 Sep. 96 Sep. 95 Oct. 95 Feb. 95 Feb. 96 Jun. 96 Jul. 96 Nov. 95 Dec. 95 Nov. 94 Dec. 94 Jan. 95 Nov. 96 Jun. 95 Jan. 96 Mar. 95 Apr. 95 May 95 Mar. 96 Apr. 96 May 96 Oct. 96 80 Depth
50 m 100 m 145 m

(d)

Fig. 54. Soil temperature against time at three levels (5.0, 10.0, 14.5 m) below energy pile heads (piled raft foundation); location of temperature sensors within ground area of the building and among the piles as parameter: (a) in pile groups, 1 m from centre piles; (b) between pile groups, 3.5 m from centre piles; (c) 0.5 m next to pile at buildings edge; (d) between two piles, 1.5 m to buildings edge

May 96

112

BRANDL hygrothermal behaviour of the building necessary. The allowable variations in daily and long-term temperature and humidity are very small, < 28C and 3% respectively. The thermal energy for conditioning is extracted from the concrete absorber system integrated in the diaphragm walls, which serve as retaining structure for the excavation pit and as perimeter foundation of the building (Fig. 57). In total 24 000 m of DN 25/2.3 mm PE pipes were installed within 4 500 m3 of diaphragm wall concrete, forming 249 units, each approximately 100 m in length, which are connected to ow and return circuits placed around the building next to the diaphragm wall. The absorber units were mounted in loops within the reinforcing cages of the diaphragm wall panels and were then lowered into the slurry trenches before the concrete was cast. Throughout the construction phase the absorber units were pressurised at 8 bar and monitored, so that any pipe failure could be detected immediately. A total water volume of 26 m3 circulates in the absorber system and provides a maximum cooling capacity of 120 kW, owing to the heat exchange between the absorber units in the diaphragm walls and the ground (i.e. primary circuit). The secondary circuit within the building comprises some 30 000 m of 16/2 mm piping positioned in the concrete oors and walls. Moreover, the solar energy entering through the windows into the concrete of the structure is transferred into the diaphragm walls and stored there (and in the surrounding soil). The benets of this energy concept for the Bregenz Arts Centre are both environmental and economical. The saving in investment costs was A1.32 million, and the annual savings in energy and operation costs are A22 700 compared with a conventional air-conditioning system. Furthermore, excessive energy can be temporarily stored in the ground for short periods.

heat pumps/absorber piles predominates, followed by cooling with near-surface earth collectors if the ground temperature is not too low. Hence the energy needed from refrigeration machines could be reduced to a negligible minimum. The geothermal energy system has been running now for 10 years without any problem. The pile temperature uctuates with a steady envelope corresponding to the maximum and minimum values already registered during the rst two years of operation. Hence no relevant change of the groundwater properties has occurred.

Heating and cooling of an arts centre The second case history refers to an example with energy piles and diaphragm walls, both installed for the foundation of the Arts Centre in Bregenz, Austria. The building has a structural volume of 28 000 m3 and an area of 33 500 m2 . Its construction required an excavation depth of up to 11 m. The foundation elements comprised a perimeter diaphragm wall of 0.5 m, 0.9 m, and 1.2 m thickness around the excavation pit, and diaphragm wall panels (barrettes) and piles of 1.2 m diameter. The wall depth reached up to 28 m, and the pile depth varied between 17 m and 25 m. The subsoil consisted of 3.5 m of near-surface gravel covering loose sands and weak clays to a depth of about 21 m below ground. These young sediments were underlain by a moraine and nally by rock. Hence the foundations were designed as end-bearing elements. The groundwater level was about 1 m below surface. The building consists of a cast concrete structure, incorporating piping loops for heating in the winter and cooling in the summer to provide a comfortable room climate. Moreover, the sensitive art collection makes an optimal

Compressed air Energy heating: MWh 60 50 40 30 20 10 0 W40 W44 W48 1995

Energy piles 1 heat pump 1996

Oil

W52 W4 Time: weeks

W8

W12

Thermo-active concrete slabs

Fig. 55. Piled raft foundation with multiple energy system for heating and cooling of a paper processing plant: generated ( required) energy for heating (including hot water)
Earth collectors Refrigeration machine 40 Energy cooling: MWh 30 20 10 0 W40 Piles Piles Energy diaphragm walls 1995 Heat pump Free cooling 1996 Secondary thermal circuit

with energy piles

Primary thermal circuit

W44

W48

W52 Time: weeks

W4

W8

W12

Fig. 56. Similar to Fig. 55, but energy generated for cooling

Fig. 57. Cross-section through an arts centre with geothermal cooling and heating (energy diaphragm walls); piled raft foundation with perimeter wall

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES Heating and cooling of a multipurpose hall and of a spa hotel A multipurpose hall with a capacity of 8 000 persons was designed for exhibitions and fairs and as a sports hall, especially as an ice rink. The latter required intensive cooling and temporary heating. The complex energy management could be solved with energy piles, because piles were already needed for a deep foundation of the structure resting on weak clays. The deep foundation comprises 320 cast in situ concrete piles (bored piles, d 0.5 m) 18 m long. The piles contain in total about 65 km absorber pipes (HDPE; d 25 mm). This cooling/heating system provides an annual saving of 85 000 m3 of natural gas, which is equivalent to an environmental relief of 73 t of CO2. In the same region a 43 m high spa hotel with geothermal heating and cooling was built. The core of the spa centre comprises four oors with 6500 m2 of spa and tness zones, and a 2000 m2 bath and sauna area. The energy foundation consists of 357 auger piles, 30 m long, and includes 69 000 m of plastic pipes. Groundwater temperature is constant at 128C. Primary and secondary energy circuits are connected by a 400 kW heat pump. In winter 1.6 GWh are extracted from the ground, corresponding to the energy demand of about 160 modern one-family houses. About the same heat volume is then sunk back into the ground when cooling the building in summer. Heating and cooling of Keble College in Oxford The rst energy pile project in the United Kingdom started in 2001, using Austrian geothermal piling technology. It is a new building for Keble College at Oxford University, a six-storey structure including a basement up to 7 m below existing ground level, providing a new lecture theatre, teaching rooms and study bedrooms. The soil prole can be roughly summarised as follows: made ground (3 m); rm alluvial clay (1 m); medium dense Thames river deposits (3.5 m); very stiff to hard Oxford clay (below). There is a perched groundwater table in the upper part of the river deposits. According to Figs 58 and 59 foundation piles (d 0.75 m and 0.45 m) and secant piles (d 0.6 m) of the retaining wall for the excavation pit were designed as energy piles tted with absorber pipes. Because of an alternative bid, the retaining wall was executed as a hard/soft secant pile wall, eventually comprising three pile types (Suckling & Smith, 2002): 115 primary piles (D 0.60 m, l 9 m); 53 secondary hard piles (D 0.75 m, l 15 m) in areas of higher ground retention; and 55 secondary hard piles (D 0.60 m, l 12 m) in areas of lower ground retention. Moreover, 61 bearing piles (D 0.45 m, l 12 m) were installed within the basement to take the structural loads. Thus the original pile design was somewhat modied by the contractor, but only with regard to the required structural or geotechnical applied loads. The amount of pipework required within the piles was designed to t within the structural requirements. Hence no pile diameter or length was increased to suit the geothermal requirements above that required for the applied loads. Installation details are given by Suckling & Smith (2002). Piping comprises 6 500 m, whereby 5 200 m of absorber pipes were installed in the piles. The heating load of the building is 85 kW and the cooling load 65 kW. Thus the monthly costs for heating and cooling can be fully covered by this geothermal system (Fig. 60). Heating and cooling of ofce centres in Vienna Table 5 gives an overview of some ofce centres in Vienna that use energy foundations for heating and cooling.
2990 m Bored piles 075 m (at higher level) Reinforced piles Not reinforced piles

113

D C 1785

D C

B 90 Energy piles A Bored piles 06 m (at lower level)

Fig. 58. Energy piles for foundation and pit retaining wall for new building of Keble College, Oxford (rst design)

1400 E 6115 D C

6450 B

4870 A

Fig. 59. Cross-section of Fig. 58

For comparison, relevant data of the Lainzer Railway Tunnel (pilot research project B) and of the Vienna Metro (pilot research project C) are added. All thermo-active energy systems have been running without problems since their inception.

SPECIAL APPLICATIONS Overview Subsurface geothermal resources can be widely used for heating and/or cooling all sorts of building and trafc area by utilising structural elements for energy extraction or storage (a) piles, barrettes and diaphragm walls as foundation elements of bridges (b) shallow foundations (c) retaining walls (d) embankments (e) tunnel linings (especially near the portals). Special applications are (a) (b) (c) (d) (e) heating/cooling of multipurpose buildings heating/cooling of bridge decks heating/cooling of road pavements and parking places heating of airport runways energy tunnels for heating/cooling of buildings near tunnel portals ( f ) energy wells for heating/cooling of buildings near

114
20000
16235

BRANDL
15000 10000 Energy: kWh/month 5000 0 25000 210000 215000 220000 225000 Heating/cooling loads Heating/cooling capacity Fluid temperature out 6 4 3
220902 221700 13642 15045 12026 10187 8262 5694 4550

22 Heating 17
220555 12 26260

23 21
6294 13600 13550 11114 8585

25

Heat carrier fluid 17

1961

20 Temperature, T: C

15 11 10 6 12 10

2 3 Time: months

4 8

22 747 23720

10

210566 211238 214814

215505

Cooling

Fig. 60. Energytime histogram (demand and output) and seasonal heat carrier uid temperature in absorber pipes (return ow)

Table 5. Examples of some ofce centres in Vienna using energy foundations for heating and cooling. Tunnel and metro projects with geothermal energy systems for comparison Building Year of Energy foundation construction 1995 2001/2002 2002 2002 2002 2003/2004 6000 m2 diaphragm walls 242 piles and 6000 m2 of foundation raft 7800 m2 diaphragm walls and foundation raft 12 400 m2 diaphragm walls and 300 piles 60 piles (cut and cover) 1425 m2 diaphragm walls and 14 piles 2650 m2 diaphragm walls, 820 m2 of foundation raft and 6 piles 7125 m2 diaphragm walls and 3744 m2 of foundation raft 2348 m2 of foundation raft Absorber pipes: m 30 000 78 000 90 000 90 000 18 000 13 000 27 000 59 000 12 000 Thermal power: kW Annual output: MWh

Generali Ofce Tower Strabag Ofce Centre Uniqa Tower Columbus Centre Lainzer Tunnel Four sections of Vienna Metro U2 U2/1 U2/2 U2/3 U2/4

Cooling: Heating: Cooling: Heating: Cooling: Heating: Cooling: Heating: Heating: Cooling: Heating: Cooling: Heating: Cooling: Heating: Cooling: Heating:

400 600 2026 1680 240 420 1428 370 150 22 85 67 95 105 208 37 62

Cooling: Heating: Cooling: Heating: Cooling: Heating: Cooling: Heating: Heating: Cooling: Heating: Cooling: Heating: Cooling: Heating: Cooling: Heating:

220 630 1450 840 819 647 1678 660 214 119 171 437 175 641 411 270 110

groundwater extraction wells (e.g. for temporary or permanent groundwater lowering) The Austrian railway authorities plan to use geothermal energy from thermo-active foundations, retaining walls and tunnels for heating (and cooling) railway stations, platforms, pedestrian ramps and staircases along the new high-speed railway lines. Heating of turf in sport stadiums is another possibility for the use of geothermal energy. This is increasingly being performed in countries with long, cold winters, but is rather expensive because of its low efciency.

(c) reducing the temperature-induced rutting of asphalt pavements caused by dense, heavy trafc (d) reducing temperature constraints in bridge decks, which increases the service life of the superstructure and pavement (e) reducing maintenance costs ( f ) reducing environmental impacts.

Heating and cooling of road pavements Geothermal technology in road engineering refers mainly to the heating of pavements during the winter months, and comprises the following goals (a) road surface free from ice, and hence increased trafc safety (b) reduced winter road clearance (c) increased environmental protection, because salt or grit for icy roads is not necessary (d) increased lifetime of the road pavement/surface (e) increased trafc comfort (no tting of snow chains) ( f ) minimisation of freezethaw damage to the road structure, especially in the case of frost-susceptible sub-bases

Heating and cooling of bridge decks In countries with cold winters and hot summers the heating and cooling of bridge decks (Fig. 61) provides numerous environmental, technical, and economical advantages (a) keeping the pavement free from ice and snow, thus signicantly reducing trafc hazards for road users (b) substituting a clean, renewable energy for gritting and the use of de-icing salt

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

Cross-section Pipes (HDPE or copper) Sealing
6 3
5

115

Asphalt wearing course Asphalt base cource Protective layer isolation Prestressed reinforced concrete girder

ing deeply into the surrounding ground can activate a relatively large mass for geothermal utilisation. Energy tunnels may be excavated as closed systems, for example by the NATM (Figs 50 and 51), or by the cut and cover method (Fig. 44). Near the portals of transportation tunnels with geothermal equipment the following groups may take the available energy (a) the owner or operator of the tunnel (b) private users (large residential blocks in particular, but also one-family houses) (c) commercial and industrial users (d) public users (municipal and federal). An example from a railway tunnel in Vienna underlines these advantages: about 1200 private ats could be supplied with geothermal energy, but also large public buildings. Energy tunnels are an exciting challenge to geotechnical engineering, whereby the optimisation of energy extraction or feed/storage, transfer and distribution requires multidisciplinary cooperation. Ground investigation and geotechnical design should incorporate geothermal aspects at an early stage. The main advantages of this innovative technology are as follows. (a) Commonly, tunnels are situated at a depth where the seasonal ground temperature is widely constant. (b) Tunnels exhibit large interfaces between structure and ground, and thus favouring the extraction and/or feed (and hence storage) of geothermal energy. (c) Very deep-seated mountain tunnels can make use of great geothermal gradients. (d) In long tunnels, signicant inner heat is available, mainly thanks to the waste heat of transportation. In metro tunnels, for instance, temperatures of more than +208C are possible even during the winter months. (e) Utilising clean and self-renewable energy from tunnels is environmentally friendly and economical. Therefore energy tunnels have high public acceptance and political support, which makes the approval procedures easier. Until now only cut and cover tunnels and open face tunnels (excavated using the NATM) have been equipped with thermo-active geosynthetics or anchors/nails, but bored tunnels with segmental lining can also be used as energy tunnels, as they exhibit earth-contact structural elements. Furthermore, optimised energy tunnels may use not only the ground temperature from their large underground contact area but also inner heat sources from trafc, lighting etc. Energy wells There are three different groups of geothermal well that serve for environmentally friendly heating/cooling at low cost (a) exploiting hot water from the ground by boreholes reaching to a depth up to $2000 m (b) conventional ground heat exchanger boreholes drilled up to about 300 m deep (c) wells that are required for temporary groundwater lowering, but which can serve simultaneously as heat extraction/storage systems. Drilling deep geothermal boreholes for (a) and (b) requires specialised equipment, considerable skill and experience. Such systems serve only for heat exchanging. Energy wells of group (c), however, represent a two-purpose system, and hence a technology involving geotechnical and geothermal engineering. Many construction sites require wells for

Pipes leading to distributor

Fig. 61. Heating/cooling of a bridge deck: watertight passage of absorber pipes through superstructure

(g) cost savings for the road authorities/owners. The use of snow chains increases the costs of winter road clearance and road maintenance, for the following reasons (a) mechanical abrasion of the road surface (b) no self-regeneration of bituminous pavements in tunnels and galleries (c) dust development due to abrasion (d) rapid pollution of tunnel and gallery walls due to road surface abrasion. It is estimated that savings of about 50% of maintenance costs would be possible along numerous road sections in the Alpine regions if no snow chains are required. This could be achieved by geothermal heating of the pavement. Furthermore, winter-long safe access to the skiing centres would be highly appreciated by tourists. In order to keep a road surface free from ice, its temperature should be higher than +28C. The critical range of air temperature lies between 08 and 108C. Lower temperatures allow the heating system to be operated intermittently or even turned off, because commonly there is no snow fall then. At present a long-term research project is running in Austria in order to determine the optimal position of absorber pipes from the thermal, energetic and structural points of view (Fig. 62). In part, these aspects exhibit contrary optima, and thus certain compromises are required. The scheme of thermo-active road pavements can be used also for airelds and basically even for turf heating of sports stadiums (e.g. soccer matches during winter). Energy tunnels Until recently geothermal heating from tunnels was used only in connection with hot water, mostly without heat pumps. But the heat potential along a tunnel can also be utilised by using the tunnel support and lining as energy absorbers. These may be anchors, rock/soil nails, geosynthetics or secondary concrete lining. Anchors or nails reach-

20

Heating loops (spacing 5 33 cm; in 10 cm sand)

033 Return flow

033 Inflow

Fig. 62. Heating/cooling of road pavements with geothermal energy: example of heating loops embedded in sand

116

BRANDL an optimal heat exchange between absorber and surrounding soil the spacing is typically lled with a cementbentonite suspension for common energy wells. However, in this case only sand and gravel were used, because the wells should also be used againat least temporarilyfor groundwater extraction. Moreover, the inuence of different contact media should be investigated. In well GB 2/97 (Fig. 63) sand was lled between tube and lter; in well GB 4/97 the heat transfer into the absorber pipes should occur mainly by contact with the groundwater. Thermal energy was taken from the heat source GB 2/97 (discharge well) and transported to a heat pump, where the energy was raised to a higher temperature level, and then transferred back to the ground by the heat sink GB 4/97 (recharge well) (Fig. 63). Figure 64 shows the temperature uctuation in the absorber system after the start of operation on 28 March 2001. In the heat source (GB 2/97) the temperature difference of the heat carrier uid between absorber inow and outow was about T 1.58C, whereby the temperature level was very low and uctuated at about T 58C. This was caused by a high-performance circulating pump, which had too much power in relation to the relatively short absorber pipes in the well. But the over-capacity of the pump had been chosen deliberately to investigate the effects of an over-design. RECOMMENDATIONS FOR PRACTICE General High-permeability soil and groundwater ow are especially suitable if only heating or cooling is performed. However, energy balance is the ideal form of seasonal heating/ cooling in winter/summer. In this case low-permeability soil and a low hydraulic gradient of the groundwater are favourable. Moreover, the smaller the temperature difference between ground source energy and used energy, the higher is the seasonal performance factor (SPF), and hence the efciency of the thermo-active system. Consequently, detailed soil investigation is essential for optimising an absorber system for thermal energy extraction/storage. Usually, temperature uctuations caused by energy foundations have no relevant effect on the surrounding soil, assuming its temperature remains higher than +28 C. However, cooling below 08C, as a result of improper operation,

groundwater lowering. Sometimes discharge wells are coupled with recharge wells in order to minimise ground settlement. These temporary measures can also be used for heating and cooling adjacent buildings. This may be performed temporarily during the construction period but also permanently after ceasing groundwater lowering. Experience has revealed that the public acceptance of metros, railways and roads by neighbouring people increases if they are provided with cheap, renewable energy from such energy wells or other geothermal systems. Figure 63 shows the scheme of a large-scale test with energy wells for heating and cooling. Simultaneously, these wells were used for long-term groundwater lowering along a new railway line under construction. Hence the tests could run from 2001 to 2003. The following investigations were carried out for research purposes and to optimise an adjacent energy tunnel in cut and cover (a) in situ determination of thermal soil parameters (thermal conductivity, specic heat capacity) (b) maximum amount of extractable heat and energy inux, and storage capacity (c) long-term behaviour and temperature conditions (d) inuence of groundwater ow. The 45 m deep test wells were installed in 50 m deep boreholes of 600 mm diameter. The subsoil consisted of manmade lls (down to 4.7 m) underlain by heterogeneous tertiary sediments: silty clay, sand, sandy silt, gravel and wide-grained silty-sandy gravel, locally with sandstone and boulders. Below 23.0 m stiff cohesive layers with interlayers of sand and clay dominate. The hydraulic conductivity was extremely scattered: locally from k 103 to 1010 m/s with an overall value of about k 106 m/s. Owing to the layered ground prole the horizontal permeability exceeded the vertical one: hence kh (1050)kv . Seepage water occurred between 9.5 and 13.4 m depth, the closed groundwater table lay at 2023 m depth (seasonal amplitude), and at 36 m depth artesian groundwater was found. Both boreholes were tted with U-pipe heat exchangers consisting of HDPE pipes of 25 mm outer diameter and located within the lter tube of the wells. In order to achieve
Expansion vessel P
T2 V1 P1 W1 T1

Expansion vessel T0 Outdoor temperature

T4 V2 W2 T3

Return

Return
P2

T1 Inflow to heat extraction well T2 Return from heat extraction well T3 Inflow to heat sink well T4 Return from heat sink well V1 Flow counter heat extraction well

Inflow
T7

Heat pump

Inflow
T10

V2 Flow counter heat sink well P1 Circulating pump heat extraction well P2 Circulating pump heat sink well T5 T7 Soil/GW-temperature (heat extraction well)
Heat sink well (GB 4/97)

Heat extraction well (GB 2/97)

T6

T9

T8 T10 Soil/GW-temperature (heat sink well) P Power input of heat pump W1 Heat meter for heat extraction well W2 Heat meter for heat sink well

Datalogger

T5

T0

Computer

Modem (GSM)

T8

Fig. 63. Scheme of testing plant energy well with GB 2/97 as heat extraction well and GB 4/97 as heat sink well

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

45 40 35 30 Temperature: C 25 20 15 10 5 0 25 210 28.03.01 Absorber entry (to well GB 2/97) Absorber return (from well GB 2/97) Outdoor temperature Absorber return (from well GB 4/97) Absorber entry (to well GB 4/97)

117

04.04.01

11.04.01

18.04.01

25.04.01

02.05.01

09.05.01

Fig. 64. Operating temperatures in absorber system of energy well

should be avoided, as it may cause freezethaw impacts in the soil, and hence affect the load-bearing behaviour of the piles, barrettes or diaphragm walls. Moreover, thermal conductivity increases and thermal storage capacity decreases with freezing, especially in soil with high water content. Freezing is avoided if the absorber system is operated only with water without antifreeze additives, but experience has shown that this reduces the efciency of the energy system signicantly. Therefore it has proved suitable to use a waterglycol mixture and to limit the freezing temperature in the core of piles or diaphragm walls to about 18C. Commonly, a temperature difference of about T 28C between inow and return-ow temperature of the absorber uid is sufcient for economical operation of the energy system. Operational uctuation of the groundwater temperature should be kept as low as possible (T < 58C). Lowering the groundwater temperature causes an increase in viscosity, and hence a decrease of the hydraulic conductivity. For T < 58C this inuence is practically negligible. Too intensive cooling of the groundwater increases the pH value, reduces calcium solubility, and raises the solubility of gaseous substances such as CO2. Too intensive heating results in a relatively large reduction in oxygen solubility, which may make the groundwater unt for drinking. Furthermore, temperature is one of the most important environmental factors for the microorganisms in water. Many of them can exist only within a very limited temperature range. In particular, the activity of bacteria-consuming microorganisms drops signicantly at temperatures below 108C. If there is sufcient heat supply from the ground, intermittent operation of the heating/cooling system is possible. This means, for instance, one to two days of operation and turn-off, alternately. In the case of a piled raft foundation, the raft should be properly isolated in order to minimise heat loss in winter and cooling reduction in summer. Geothermal utilisation of concrete retaining walls is also possible: gravity walls, cantilever walls, pile walls, diaphragm walls, etc. In this case, a proper design of the absorber system has to take into consideration the natural temperature uctuations along the free-standing front face of energy walls, which differ widely from those in the fully embedded zone beneath surface. Pile walls may consist of secant piles, contiguous or intermittent piles. The spacing of the piles has a strong inuence on the efciency of the energy system. A generally valid rule about which pattern provides the highest efciency cannot be given, because this depends on several interacting factors

(a) the surface area and volume of the concrete elements (b) the heat conductivity and thermal storage capacity of the piles and ground (c) the hydrogeological conditions.

Installation of absorber pipes in energy foundations or walls The installation of reinforcement cages tted with absorber pipes into bored piles, barrettes or diaphragm walls requires the following measures. (a) Protection from mechanical damage, especially in the case of cutting by machine and non-deburred reinforcement bars. (b) Protection from thermal damaging (during reinforcement welding). (c) Exact positioning of the reinforcement cages (orientation of the connecting box). (d) The construction of stiff reinforcement cages for deep foundations (e.g. welding of helical reinforcement to vertical rebars of deep piles, barrettes etc.). (e) Lifting long reinforcement cages at both ends to prevent damage to the pressurised absorber pipe loops. ( f ) The use of full tremie pipes to place concrete in pile bores; also for dry rotary-bored piles, where commonly self-compacting concrete is placed via a short tremie pipe from the ground surface. (g) Upon completion of the pipework xing on the reinforcement cage, a visual check on the nal location of the pipes is imperative to ensure that the oor of the wet pile concrete through the reinforcement cage would not be impaired. The pipe ends near the bottom of the cage should be placed at different levels to help this. (h) Very long reinforcement cages tted with absorber pipes have to be installed in sections, which should be coupled by screwing, not welding. The pipes are extended/coupled by electrically welded sleeves. Welding of the reinforcement sections is unavoidable only if a lightning protection element is attached. In such cases the absorber pipes have to be protected during welding by welding mats. (i) Careful insertion and withdrawal of the tremie pipes. ( j) Protection from torsion and heave of the reinforcement cage during concreting and steel pipe withdrawal. (k) Sufcient distance of the absorber pipes from the reinforcement on the head and toe of the piles, barrettes, or diaphragm wall panels. (l) Special precautions have to be taken for energy piles or

Heat source

Heat sink

118

BRANDL diaphragm walls used for cut and cover tunnels or retaining walls if they are covered by a secondary lining. In order to avoid damage of the absorber pipes when installing the connecting nails or dowels, the pipes should be protected by twin bars. buildings are superior to passive houses as they allow more natural air circulation in the rooms and avoid too intensive insulation.

For energy rafts (base slabs) the following construction sequence has proved suitable. (a) First lean concrete layer (blinding) on ground. (b) Place a wire mesh to facilitate xing of the absorber pipes. (c) Install the absorber pipes, which must be clearly and durably marked. (d) Second lean concrete layer to protect the absorber pipes. (e) Place the reinforcement of the raft and casting concrete. The absorber pipes in the foundation elements should be kept under pressure during all construction stages. This enables prompt localisation of possible defects and repair in time. The collectors and distributors of the absorber pipes should be tted with optical owmeters for long-term monitoring of the uid circulation in the geothermal system. Interaction of primary and secondary energy systems The temperature felt by persons in a room consists of the air temperature and the radiation temperature (i.e. the temperature of walls and oors), whereby the ratio between air and surface temperature is essential. Compensating too low a wall or oor temperature by higher air temperature is felt as uncomfortable (e.g. in temporarily uninhabited cold houses that have to be warmed up). Surface temperatures of 20258C are optimal, corresponding to an air temperature between 168C and 218C. Low-temperature heating systems, such as wall and oor heating with a large surface radiation, full this prerequisite, whereby the feed temperature of wall heating should not exceed 408C and the oor temperature a maximum of 288C. This heating system can be coupled in an ideal way with energy foundations, retaining walls and other thermo-active ground structures. Thermo-active foundations, retaining walls, etc. should be combined with thermo-active building systems (tabs). The temperature difference between rooms and structure should not exceed 238C in order to create a comfortable room climate. This is effectively achieved by optimal glazing, basic ventilation and minimal thermal resistance (e.g. no thick oor covers, double oor system). Low and locally different inow temperatures are essential: consequently, corner areas should have their own feed circuits. The energy demand during day/night and seasonally does not always coincide with the available renewable energy. Therefore energy must be stored temporarily, either in the ground or within the structure of the building. The installation of plastic pipes in the concrete elements is easy and economical during the construction of new buildings. However, thermo-active installations in existing buildings are quite complicated. Innovative technologies are thin oor elements of light weight with a high heat storage capacity using phase-change materials. In many cases the restoration of old buildings or their heightening by one or more oors requires an underpinning of the existing foundation. This can be used to install energy foundations and couple them with the thermo-active heating (and cooling) system of the improved building. The optimal low-energy building is an energy-active house using the entire structural envelope and foundation and even garden walls for heat absorption and emission, and hence as a low-temperature thermal storage system. Such

Proper operation of geothermal energy system Temperature changes of the soil directly bordering on thermo-active foundations have negligible inuence on bearing-deformation behaviour if the energy system is properly operated. Granular soil is affected less than clay or silt; furthermore, the temperature sensibility of soil increases with its organic contents. Proper heat extraction for heating a building reduces the ground temperature locally from 10158C to about 5108C, whereas excessive heat extraction with freezing-stable absorber uids may create sub-zero temperatures. This causes the formation of ice lenses in the soil whereby clayey to sandy silt is especially affected, depending on its mineralogical composition. Soil freezing hardly has any inuence on deep energy foundations, but may cause severe heave of base slabs, or lateral deformations (and cracks) in retaining walls. During the subsequent thawing period the ground loses strength, and settlements of the building are possible. Experience and long-term monitoring have shown that the temperature along the skin of energy piles or diaphragm walls should not fall below +28C. Intensive heat input into the ground for cooling a building is less critical for the foundation but possibly so for the quality of groundwater and microorganisms. Temperature increase causes a slight decrease of effective stress, and hence of shear strength; furthermore, shrinkage may also lead to settlements. Numerous papers on general temperature effects between 28C and 658C on soils show rather divergent results (e.g. Ennigkeit, 2002). This can be explained mainly by the different behaviour of clay minerals, the inuence of water content/saturation, density etc. Even the speed of warming up has an inuence, but less than the freezing speed in the case of excessive heat extraction. However, these investigations refer only to general laboratory tests, and cannot be applied directly to thermo-active ground structure operation. Here the temperature differences are rather small, and have less physical-mechanical inuence than heat extraction. Seasonal operation of the energy systemthat is, heat extraction and heat input (recharging) by alternate heating and coolingcauses a superposition of the above-mentioned effects, and hence cumulative settlements occur until a steady state is reached. Cumulative heave would occur only in the case of freezing of frost-susceptible soil below base slabs. When energy piles, barrettes or diaphragm walls are exposed to geothermal temperature changes, their head and toe react in a different way: Heating of the buildingthat is, cooling of the deep foundationcauses its shrinkage, and hence head settlement and toe heave. Cooling of the building leads to reverse movements. Numerical simulations of such temperature-induced deformations have yielded temporary settlements of up to 10% of the static settlements for low- and high-rise buildings. Ennigkeit (2002) predicted somewhat higher values. However, in situ monitoring has clearly shown smaller values, for both seasonally maximal and cumulatively residual values; moreover, differential settlements have been negligible from a statical point of view. This refers also to temperatureinduced differential settlements among central piles, edge piles and corner piles of pile groups, and to temperatureinduced changes of the skin friction of thermo-active deep foundations.

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES Promotion of geothermal energy utilisation Early ecological energy planning for buildings can in many cases prevent costly refurbishment and renovation in the future. High-quality energy design involves not only heating and cooling (rooms, water) but also lighting. Building biology (including building ecology) is gaining increasing importance in the ght against global warming, depletion of the ozone layer and the exploitation of material resources. Building biology has become (or should become) a multidisciplinary science combing architecture, civil and geotechnical engineering, physics and chemistry, installation engineering, medicine and related sciences. It considers not only interactions of buildings and human health, but also energy concepts, the life cycle of building materials, sustainability etc. Geothermal geotechnics offers a promising alternative to conventional heating/cooling systems, providing solutions to the challenges of todays energy policies. The targets for renewable energy and for energy buildings can in general be reached only by political measures. (a) High taxes on fossil fuels are the most important prerequisite for energy saving and promotion of renewable energy sources. (b) In order to promote the installation of thermo-active systems or other heating/cooling systems based on renewable energy, the economic incentives for private investors, house owners, companies, and also for public administrators to invest in renewable energy systems should be improved in many countries. Strong support by European Union policy is necessary. (c) Legislation. (d) Public grants. Since January 2004 each person who wants to build a family house in Austria has received nancial support by the local government only if they present an energy performance certicate with a low energy number. This number describes the energy consumption (provided by heating energy minus heating losses) and is expressed in kWh/m2 and year. Promotion by public funds is granted only if this energy number is smaller than 50 kWh/m2 for each oor. At values less than 40, 30, 25, 20 and 15 kWh/m2 the grant increases step by step. However, if a building is heated/cooled by means of clean, renewable energy, for example by geothermal systems, the allowable limit value for energy consumption may be increased. The target of multidisciplinary innovation should approach heat-and-light systems combined with groundsourced or solar residual heating/cooling. Thermo-active structures (including energy foundations) are therefore very helpful in reaching this low energy number. Their installation is widely supported by politicians and media. Consequently, about 500 buildings with energy foundations or retaining/basement walls already exist in Austria. This philosophy is fully supported by Directive 2002/91/ EC of the European Parliament and of the Council on the energy performance of buildings, which will come into force in the European Community on 4 January 2006 at the latest. Thus an energy performance certicate has to be presented if a building with more than 500 m2 is sold or rented. CONCLUSIONS World supplies of fossil fuels are rapidly being depleted. Consequently, multidisciplinary efforts are needed to develop innovative building practices using renewable energy including new energy storage technologies. Near-surface geothermal and deep geothermal energy, solar photovoltaic and solar thermal energy, and wind energy are promising

119

alternatives (like conventional hydropower energy). Optimum economical efciency and environmental protection are gained in most cases by an energy mix from different sources. Local climate and ground properties, technological level, the specic use of a building, seasonal uctuations, environmental conditions and actual energy prices are the main inuencing parameters. Integral planning and balancing of buildings means considering the technical, economic, aesthetic and ecological aspects. An integral design, therefore, is always a sustainable design too, and it requires multi-objective optimisation. Balancing refers to materials, energy, emissions, waste water, waste/rubbish and its disposal or recycling, costs (investment, maintenance, demolition), and life cycle. Both thermo-active ground structures (energy foundations, retaining walls, tunnels, etc.) and energy wells are innovations that promise sustainable and clean energy consumption. A signicant advantage of such systems is that they are installed within elements that are already needed for statical/ structural or geotechnical reasons. Hence no additional structural or hydraulic measures are required. Foundations, walls (below and above ground) and tunnel linings can be used directly for the installation of absorber pipes for heat exchange. Moreover, concrete has a higher thermal conductivity than soil. Wells for groundwater lowering may be simultaneously used for heat extraction/storage, thus becoming energy wells. This innovation is a key improvement over the conventional geothermal methods such as (deep) borehole heat exchangers or near-surface earth collector systems. At a certain depth, ground temperature remains widely constant throughout the year (e.g. 10158C below 1015 m in most European regions), and a heat exchanger allows it to be used as a heat source in winter and for cooling in summer. Energy systems based on earth-contact structural elements have a double function, and they work most efciently if the thermo-active elements are in contact with groundwater. Nevertheless, a sufcient seasonal performance factor of the system is achievable even without groundwater, especially for seasonal operationthat is, heating in winter and cooling in summer. Energy balance is the ideal form of heating and cooling. Moreover, the smaller the temperature difference between ground source energy and used energy, the higher is the seasonal performance factor, and hence the efciency of the thermo-active system. Usually, a temperature difference of only T 28C between absorber uid inow and return ow from the primary circuit is sufcient for economical operation of the energy system. Consequently, such geothermal systems represent low-temperature systems. Experience has shown that the electricity required for operating the entire system commonly varies between 20% and 30% of the total energy output. If no heat pump is necessary (e.g. for free cooling) this value drops to 13% for merely operating a circulation pump. Costbenet analyses depending on the climatic conditions in the region concerned have shown that the investment payback period for such heating/cooling systems usually varies between 2 and 10 years, depending on the particular ground properties, foundations systems, building characteristics and energy prices. In special cases even the initial cost outlay may be lower than for conventional heating/cooling systems. Nearly 20 years of experience with energy foundations has shown that heating or cooling costs can be reduced by up to two thirds over the design life of a building. Balanced seasonal operation of heating and cooling may even save up to 75% in electricity compared with conventional air-conditioning systems. The shaft resistance and base pressure of energy piles, barrettes or diaphragm walls, and the bearing capacity of the

120

BRANDL (p) The cost of geothermal energy is not prone to unpredictable price uctuations. (q) The consequent reduction of energy imports means a reduced dependence on external economical or political situations. (r) Geothermal energy has a positive public image, and in several regions is supported by government grants.

soil around their toe, are not affected by the heat absorption process in a statically relevant magnitude. Hence temperature-induced settlement or heave of buildings is negligible in relation to displacements caused by static loads (this has been shown by detailed site measurements since 1994), but exposure of the surface of deep foundations to temperatures below freezing should be avoided. Fine-grained soils with a high content of active clay minerals (e.g. montmorillonite) are especially critical in this context. The temperature uctuations around an energy foundation decrease signicantly with distance from the heat exchanger (roughly an exponential decay). Thermal properties and efciency of energy foundations can be inuenced not only by their design (always coupled with statical/structural and geotechnical requirements) but also by the concrete characteristics. Proper geothermal energy utilisation requires an interdisciplinary design, especially in the case of houses. The geotechnical engineer, structural engineer, architect, building services designer and installer, heating engineer and specialised plumber should cooperate as early as possible to create the most economical energy system. In the rst phase of operation precise adjustment is recommended to optimise the performance of the engineering system. Furthermore, some operation rules have to be considered (as mentioned in this paper). The benets of energy foundations and other thermoactive earth-contact structures may be summarised as follows. (a) They are environmentally friendly (non-polluting, sustainable energy). (b) They offer a reduction of fossil energy demand, and hence of CO2 emissions. (c) They promote compliance with international environment obligations, such as the Kyoto and Toronto targets. (d) They are economical, at least in the long term. (e) Although thermo-active earth-contact structures commonly require investment costs that are similar to or slightly higher than those of conventional systems, they have lower running costs and hence lower life cycle costs. ( f ) They are low maintenance, long lifetime systems. (g) Geothermal energy systems run fully automated. (h) Thanks to the low temperature and pressure in the heat carrier circuits, geothermal heating/cooling systems can be operated without risk. (i) The closed primary heat carrier circuit embedded in concrete prevents damage of pipework or groundwater pollution. ( j) They offer increased personal comfort in buildings (indoor rooms). The temperature personally felt there that is, the ambience experiencedconsists of air temperature and radiation temperature, which are inuenced by wall and oor temperatures. Comfort is enhanced by low-temperature heating of walls and oors exhibiting a large heat-radiating surface. (k) Optimal hygrothermal behaviour of buildings is possible (e.g. for museums and arts centres). (l) There is no storage of fossil fuel, no stove or chimney, and no visible radiators are needed. (m) Geothermal cooling may replace conventional airconditioning, which is often felt to be loud and unhygienic. (n) Geothermal energy may be easily combined with other energy systems. (o) Unlike hydroelectricity, geothermal energy is not vulnerable to droughts.

ACKNOWLEDGEMENTS The author would like to thank his assistants, Associate Professor Dr D. Adam and Dr R. Markiewicz at the Institute for Soil Mechanics and Geotechnical Engineering, Vienna University of Technology, for their valuable contributions to this paper. In the framework of Universities meet research interested companies the long-established cooperation with the engineering company Naegelebau/enercret is also acknowledged. Moreover, the Vienna Authorities (Wiener Linien), the Austrian Federal Ministry for Transport, Innovation and Technology, and the Austrian Railway Authorities (HLAG) have strongly supported the research project Thermo-active ground structures. Finally, the author is grateful to the iC consultants, Austria, for discussions on metro subjects of this paper.

NOTATION
A a COP C CV c cv cw D d ex , e y , ez i k L l m _r m Nu n P P Pr Q _ Q _ q q _cond _l,conv q _lat q _tot q qu _v,conv q R, r Re S SPF T T9 Tm Tm,out Tm,out TS TU T0 Tout t u area thermal diffusivity coefcient of performance circumference volumetric heat capacity specic heat capacity specic heat capacity of soil vapour specic heat capacity of soil water pile diameter; thickness of diaphragm wall damping depth; pipe diameter unit vectors hydraulic gradient hydraulic conductivity; permeability coefcient pipe length length mass mass ux density Nusselt number ow direction; soil porosity; quantity of pipes pump performance period duration of temperature oscillation Prandtl number heat volume; total load heat ux heat ux density conductive heat ux density heat ux density generated by liquid convection heat ux density due to latent heat total heat transfer in soil compressive strength heat ux generated by vapour convection radius Reynolds number degree of saturation seasonal performance factor temperature reference temperature caloric mean temperature mean daily air temperature mean yearly air temperature surface temperature surrounding temperaure reference (initial) temperature temperature amplitude time ow velocity of absorber uid

ENERGY FOUNDATIONS AND OTHER THERMO-ACTIVE GROUND STRUCTURES

vv vw w x z l t r rv rw vector of vapour velocity vector of water velocity water content specic volume axis heat transfer coefcient roughness of inner pipe wall thickness of laminar edge zone phase displacement integration limit dynamic viscosity; efciency thermal conductivity kinematic viscosity coefcient of ow pressure loss within pipe system density density of soil vapour density of soil water shear stress azimuth; angle of declination; friction angle angle of ascension integration variable

121

(2000). Ground heat transfer effects on the thermal performance of earth-contact structures. Renewable and Sustainable Energy Rev. 4, No. 3, 213265. Suckling, T. P. & Smith, P. E. H. (2002). Environmentally friendly geothermal piles at Keble College, Oxford, UK. Proc. 9th Int. Conf. Exhib. on Piling and Deep Foundation, Nice, 445452. Verein Deutscher Ingenieure (1997). VDI-Wa rmeatlas. Berlin/ Heidelberg: Springer-Verlag.

FURTHER READING
Adam, D., Honger, J. & Ostermann, N. (2001). Utilization of geothermal energy from railway tunnels. Proc. 15th Int. Conf. Soil Mech. Geotech. Engng, Istanbul 3, 20292034. Andersland, O. B. & Anderson, D. M. (1978). Geotechnical engineering for cold regions. New York: McGraw-Hill. Brandl, H. (1980). The inuence of mineral composition on frost susceptibility of soils. Proc. 2nd Int. Symp. on Ground Freezing, Trondheim, 815823. Brandl, H. (1998a). Energy piles for heating and cooling of buildings. Proc. 7th Int. Conf. Exhib. Piling and Deep Foundations, Vienna, 3.4.13.4.6. Brandl, H. (1998b). Energy piles and diaphragm walls for heat transfer from and into ground. Proc. 3rd Int. Geotech. Seminar, Deep Foundations and Auger Piles III, Ghent, 3760. Brandl, H. & Markiewicz, R. (2001). Geothermische Nutzung von sterr. Ing. ArBauwerksfundierungen (Energiefundierungen). O chit. Z 146, No. 56, 216222. Brandl, H., Adam, D. & Kopf, F. (1999). Geothermische Energienutzung mittels Pfa hlen, Schlitzwa nden und Stu tzbauwerken. Pfahl-Symposium 1999, Braunschweig, 329356. Brandl, H., Adam, D. & Markiewicz, R. (2004). Nutzung von Verkehrstunneln als Absorberbauwerke fu r die Gewinnung geothermischer Energie, Research Report. Vienna: Federal Ministry for Transport, Innovation and Technology. Brauner, M. (2002). Obera chennahe Nutzung der Geothermischen Energie: Analyse ihrer Umweltauswirkungen und der rechtlichen Rahmenbedingen im internationalen Vergleich. Masters thesis, Institute for Soil Mechanics and Geotechnical Engineering, Technical University of Vienna. Enercretna gele (2004). Thermoaktive Fundamente. Na gele-Reports, Ro this, Austria. Ennigkeit, A. & Katzenbach, R. (2001). The double use of piles as foundation and heat exchanging elements. Proc. 15th Int. Conf. Soil Mech. Geotech. Engng, Istanbul 2, 893896. Farouki, O. T. (1986). Thermal properties of soils, Series on Rock and Soil Mechanics, Vol. 11. Clausthal-Zellerfeld: Trans Tech Publications. Ferrari, R. (1998). The exploitation of energy extracted from ground-contact concrete members. Na gele Reports, Ro this, Austria. Gehlin, S. (1998). Thermal response test: In situ measurements of thermal properties in hard rock. Division of Water Resources Engineering, Department of Environmental Engineering, Lulea University of Technology, Sweden. Hellstro m, G. (1991). Ground heat storage: thermal analyses of duct storage systems I. Theory. Doctoral dissertation, Lund Institute of Technology, Department of Mathematical Physics, Lund University, Lund, Sweden. Hellstro m, G. (1994). Fluid-to-ground thermal resistance in duct ground heat storage. Proc. Calorstock 94, Espoo, pp. 373380. Holdsworth, B. (2003). Cool thinking. European Foundations, Spring, 1011. von der Hude, N. & Kapp, Chr. (1998). Einsatz von Energiepfa hlen am Beispiel des Main Tower in Frankfurt am Main. Reports of the Institute for Geotechnics, Technical University of Darmstadt, Vol. 39, pp. 1528. von der Hude, N. & Vo lkner, R. (2003). Erdwa rmenutzung und Ausfu hrung auf der U-Bahnbaustelle U2/1, Schottenring in sterreichische Vereinigung fu Wien. O r Beton- und Bautechnik 54, 1511. Hueckel, T. & Pellegrini, R. (1991). Thermo-plastic modelling of undrained failure of saturated clay due to heating. Soils Found. 31, No. 1, 116. Kaltschmitt, M. (ed.) (1999). Energie aus Erdwa rme. Stuttgart: Deutscher Verlag fu r Grundstofndustrie.

Subscripts
a c GS GW h i i m min o p prim s s,min s,org sec v W w air; pore air core (of uid) ground surface ground water horizontal ice; water inner mean minimum outer pump primary circuit soil; solid mineral component of solid fraction organic component of solid fraction secondary circuit vertical wall water; pore water

REFERENCES
Abu-Hamdeh, N., Khadair, A. & Reeder, R. (2001). A comparison of two methods used to evaluate thermal conductivity for some soils. Int. J. Heat Mass Transfer 44, No. 14, 10731078. Adam, D. & Markiewicz, R. (2002). Nutzung der geothermischen sterr. Ing. Archit. Z. Energie mittels erdberu hrter Bauwerke. O 147, No. 4, 5 and 6. Ennigkeit, A. (2002). Engerieanlage mit Saisonalem Thermospeicher, Mitteilungen des Institutes und der Versuchsanstalt fu r Geotechnik der Technischen Universita t Darmstadt, No. 60. Honger, J. (2002). Nutzung geothermischer Energie und Umweltwa rme auf niedrigen Temperaturniveau mittels erdberu hrter Bauteile. Masters thesis, Institute for Soil Mechanics and Geotechnical Engineering, Technical University of Vienna. Jessberger, H. L. & Jagow-Klaff, R. (1996). Frost im Baugrund. Grundbau-Taschenbuch. Teil 1, 5th edn. Berlin: Ernst & Sohn Verlag. Leu, W., Keller, B., Matter, A., Scha rli, U. & Rybach, L. (1999). Geothermische Eigenschaften Schweizer Molassebecken. Bern: Schlussbericht Bundesamt fu r Energie. Markiewicz, R. (2004). Numerische und experimentelle Untersuchungen zur Nutzung von geothermischer Energie mittels erdberu hrter Bauteile und Neuentwicklungen fu r den Tunnelbau. Doctoral thesis, Institute for Soil Mechanics and Geotechnical Engineering, Technical University of Vienna. Oertel, H. Jr (Ed.) (2001). Prandtl: Fu hrer durch die Stro mungslehre. Braunschweig/Wiesbaden: Friedr. Vieweg & Sohn Verlagsgesellschaft mbH. Rees, S. W., Adjali, M. H., Zhou, Z., Davies, M. & Thomas, H. R.

122

BRANDL Much of Professor Brandls work has been to do with the construction and stabilisation of transport infrastructure in alpine country. Many of the projects are extraordinarily daring and audacious, in conditions where the ground properties vary enormously and precise calculations are not helpful. This work has bred in Professor Brandl a deep reliance on understanding the basic mechanisms of behaviour, coupled with an insistence on observation and measurement. His work is also characterised by a willingness to use innovative solutions for dealing with very challenging problems. These alpine projects have also given rise to a highly developed sense of the need for compatibility of deformations between ground and structurea sense of when rigidity is benecial and when exibility and robustness are what is requiredparticularly in ground-retaining structures. In this evenings lecture we have seen displayed all these characteristics of Professor Brandls work. There are three topics that are exercising society as a whole right now, and will do so for the foreseeable future. They are: conservation and enhancement of the environment; risk assessment; and sustainable development. All three of these involve many disciplines. Professor Brandl has touched on all three of these topics, and has demonstrated that geotechnical engineering has major roles to play in all of them. The idea of energy foundations is exciting, and the energy savings look attractive and viable. I believe that the concepts and case histories outlined by Professor Brandl will stimulate study and discussion and hopefully further successful applications worldwide. With his background in alpine construction it is natural that aesthetics should be a key issue, and Professor Brandl has shown us that it is possible to make high retaining structures using novel and varied techniques that are very attractive. He has certainly offered us a challenge in this respect. Professor Brandl touched on landll engineering, which is of course a signicant aspect of environmental geotechnics but by no means the dominant part of it. He has given us much detailed valuable advice and a number of alternative strategies. What I found particularly interesting in this part of his lecture is his perception of risk and in particular the importance of involving laypeople, those directly affected, in dialogue. In many senses, the assessments of technical and commercial risk are the easy bit. It is dealing with society that is the difcult bit. The communication of risk in terms that are readily understood, together with informed presentation of the options available, is something that the engineering profession is just beginning to learn. I believe that we should not be reluctant to enter into dialogue with the lay public (the stakeholders) because it is as much about understanding and making opportunities as it is about preventing conict and mitigating risk. The key is to do it at the right time and to know when to call on the artist or psychologist! I know that you will all agree with me that we have listened to a challenging and stimulating 41st Rankine Lecture. The published version will, sadly, be more condensed, but I believe that it will be a seminal paper. It is with the utmost pleasure that I propose this vote of thanks to Professor Brandl.

Katzenbach, R., Arslan, U. & Ru ckert, A. (1998). Das Prinzip des saisonalen Thermospeichers, Report No. 39 of the Institute for Geotechnics. Technical University of Darmstadt. Kuntiwattanakul, P., Towhata, I., Ohishi, K. & Seko, I. (1995). Temperature effects on untrained shear characteristics of clay. Soils Found. 35, No. 1, 116. Laloui, L., Moreni, M., Fromentin, A., Pahud, D. & Vulliet, L. (1999). In-situ thermo-mechanical load test on a heat exchanger pile. Proc. 4th Int. Conf. on Deep Foundation Practice, Singapore, 273279. Lippuner Energie- u. Metallbautechnik (1996). Erfolgskontrolle Energiepfahlanlage Pago, Zwischenbericht, Grabs. Unpublished report. Lund, P. (1985). Effect of ground water ow on the performance of long-term pipe heat storage in the ground, Report TKK-F-A586 (1985). Helsinki University of Technology Finland. Markiewicz, R. & Adam, D. (2003). Utilisation of geothermal energy using earthcoupled structures: theoretical and experimental investigations, case histories. Proc. 13th Eur. Conf. Soil Mech. Geotech. Engng, Prague 2, 310. Merker, G. P. & Eiglmeier C. (1999). Fluid- und Wa rmetransport Wa rmeu bertragung. Leipzig: B. G. Teubner Stutgart. Mu llner Energieberatung & Haustechnik (1990). Anlagenbeschreibungs Ph. Ma gele ser-Maselli Strickwarenfabrik, Dornbirn. Na Reports, Ro this, Austria. zisik, M. N. (1980). Heat conduction. New York: John Wiley & O Sons. Preg, R. (1997). Energy savings with concrete: Heating and cooling with energy from subsoil. Na gele-Reports, Ro this, Austria. Sanger, F. J. (1979). Degree-days and heat conduction in soils. Proc. 1st Int. Conf. Permafrost, Purdue Univ., Indianapolis, 253262. Sauty, J. P. (1980). An analysis of hydrodispersive transfer in aquifers. Water Resour. Res. 18, No. 2, Purdue Univ., Indianapolis, 253265. Stai, F. (2001). Jahrbuch Erneuerbare Energien. Radebeul: Bieberstein-Fachbuchverlag. Suckling, T. & Cannon, R. (2004). Energy piles for Pallant House, Chichester, UK. Ground Engng 37, No. 7, 2729. Unterberger, W. & Honger, J. (2003). Erdwa rmenutzung sterreiWirtschaftliche Aspekte und Zukunftsperspektiven. O chische Vereinigung fu r Beton- und Bautechnik, Wien 54.

VOTE OF THANKS Professor J. B. BURLAND, Imperial College of Science, Technology and Medicine, London Over the last four years I have had the great pleasure of listening to the following four major lectures presented by Professor Brandl: the Special Lecture to the Hamburg Conference; the Young Geotechnical Engineers Conference in Santorini; the First International Conference on Geotechnical Education in Sinaia, Romania; and GeoEng 2000 in Melbourne, Australia. Tonights lecture is the fth that I have attended. I know that Professor Brandl has given a number of other major lectures as well, and I am full of admiration for the way he has been able to nd the time to do this on top of a most demanding professional and teaching workload. Each of these lectures has been on very different topics, ranging from the practical through to the ethical and even philosophical, all of them thought-provoking and all of them challenging. I am sure you will agree that this evenings Rankine Lecture has also been thought-provoking and challenging, and has covered a remarkable range of topics.