Unsupported browser

For a better experience, please update your browser to its latest version.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Technical paper: Development of thermal steel sheet piles

sheet steel pile

Alex-Q Chen, Duncan Nicholson, Anton Pillai, Arup, and David Brown, Dawson Construction Plant

Arup Tech Paper logo (NB)

1.0 Introduction

The use of thermal bored piles is an established and proven foundation element that can be constructed with closed ground source heat pump (GSHP) systems to provide renewable heating and cooling energy to new buildings [1][2][3][4][5]. The thermal-hydro-mechanical (THM) interaction for thermal retaining walls has also been studied for Crossrail stations [6]. However, the use of steel sheet piles in the UK to similarly provide renewable heating and cooling energy is yet to be established.

To advance the adaptation of this technology to thermal steel sheet piles walls the Steel Piling Group funded a comparison study and installation trial to demonstrate feasibility. In this study, the application of steel sheet piles for their geothermal potential are investigated by numerical analysis. Comparison with equivalent concrete secant piles is also made. 

2.0 Steel sheet piles

Low-carbon steel has a thermal conductivity between 40 and 50Wm-1K-1; higher than concrete, which is typically in a range of 1 to 2Wm-1K-1 [9]. As a consequence of this higher conductivity, the soil mass can potentially be more efficiently activated via a steel sheet pile wall. To make use of the high thermal conductivity of steel all that is required is to establish a simple loop installation system.

In a concrete secant pile, the thermal U-loop heat exchanger system is fixed to the pile’s reinforcement cage and encased within the cast concrete. Refrigerant fluid (water with additives) is circulated through the loops from the heat pump down the piles where heat is exchanged with the surrounding soil [1].

Two steel sheet piling heat exchanger configurations have been considered: U-loop and concentric (coaxial) pipe, as shown in figures 1 and 2. For the sheet pile U-loop heat exchanger system, the pipework is held to the side of the sheet pile plate. For the sheet pile concentric pipe heat exchanger system, the pipe is welded to the side of the sheet pile plate to improve the heat conduction. Fluid is circulated through the inner concentric pipe and is returned to the surface through the annulus between the inner and outer pipes; the flow is regulated by a well head on top of the pipe.

figure 1

figure 2

In this study, the feasibility of installing thermal steel sheet piles and heat output rate are both investigated. The purpose is to find out whether steel sheet piles can provide a viable alternative for ground energy to traditional concrete secant piles.

3.0 Finite element modelling

Numerical modelling was carried out using finite element program LS-Dyna [8] to simulate thermal behaviour of steel sheet-piles and concrete secant piles. To simulate this thermal transfer behaviour of the piles and ground, a THM constitutive model was used to assess the cooling of the piles and the surrounding ground.

The two types of thermal pipe systems considered, concentric pipes and U-loops, were modelled for a hypothetical basement. The outputs were compared with traditional thermal concrete secant piles which have two U-loops per pile.

The hypothetical basement is assumed to be within the London Clay and overlain by made ground. The basement dimensions are 250m long, 250m wide and 8m deep, with a perimeter concrete secant pile wall or a permanent sheet pile wall. Both wall types were modelled as thermally activated below the formation level.

The sheet piles are installed to the same depth as the secant piles and fitted with ground heat exchangers.

The initial ground temperature was modelled as 12˚C and the circulating fluid had a temperature of 4˚C to simulate heating of a building. Harvesting heat from the ground becomes limited by the ground having to be kept above freezing. This avoids the risk of ice lenses developing which could affect the sheet pile’s load bearing performance.

The following three configurations were modelled and the model setup for each of the three configurations is shown in figure 3:

  • Secant thermal pile (U-loop) – comprising hard-firm secant piles of 750mm diameter at 600mm centres. Every secondary reinforced pile has two 25mm OD polyethylene U-loops per pile;
  • Thermal sheet pile (concentric) – consists of one concentric circular pipe per pair of AZ28-700 S355 steel sheets at 1400mm centres. The concentric pipe used in the model contained a 25mm OD polyethylene pipe encased within an outer 50mm OD steel pipe which is welded to the steel sheet pile; and
  • Thermal sheet-pile (U-loop) – consists of a single U-loop adjacent to the AZ28-700 S355 sheet pile at 700mm centres. The U-loop consists of a 25mm OD polyethylene pipe down the full length and held against the side of the steel sheet pile.

figure 3

 4.0 Results of finite element modelling

The temperature distribution for the concrete secant pile and the steel sheet piles after an arbitrary 100 hours of operation is shown in figure 4. At this 100 hour stage, prior to reaching quasi-steady state conditions, the thermal contours show the cooling patterns for the low volume high conductivity sheet piles and the high volume low conductivity secant piles configurations.

Figure 5 presents a graph of the heat output rate in watts per square metre of wall over 2,000 hours. As shown in the plot, both the steel sheet-pile configurations generate a similar heat output rate after 400 hours to that of the concrete secant piles. At 10 hours of operation the secant pile output rate is higher (88W/m2) than the sheet pile U-loops (79W/m2) and the concentric pipe (52W/m2). The output rates are shown to converge over time as more soil is thermally mobilised for sheet-piles. This is due to the faster heat conduction of the steel plate.

figure 5

After 100 hours the heat output rate for the three configurations was similar, with the secant pile and sheet pile – U-loops generating approximately 30W/m2 and the concentric pipe 22W/m2. After 1,000 hours the heat output rates in all configurations were almost the same, at approximately 10W/m2.

figure 6

Figure 6 presents a graph of cumulative heat output watt hour per square metre of wall over a period of 2,000 hours. Initial heat generation in all configurations are similar. After 100 hours the secant pile generating 5kWh of heat and the jetted U-loop 4kWh of heat and the concentric pipe 3kWh of heat. Over the next 1,900 hours the cumulative heat output difference between the configurations increases slightly with 26.5kWh for the secant pile, 24.5kWh for the jetted U-loops and 22.5kWh for the concentric pipe. This demonstrates that all models generate a comparative amount of heat, with the secant pile performing only slightly better than the sheet pile options due to the secant wall’s larger volume and greater total length of heat exchanger. However, the efficiency of the sheet pile systems could be improved with additional loops or narrower sheets giving closer spacing.

5.0 Field installation trial

An installation trial of a steel sheet-pile with a U-loop attached was carried out by Dawson Construction Plant in its Milton Keynes plant yard during 2016. The steel sheet was 7m long with a rectangular protective steel shoe of 90mm by 50mm by 550mm long attached at the toe for the plastic U-loop 32mm OD PE100-RC geothermal probes. An angled foot plate at the base of the protective shoe was built to facilitate ground penetration is shown in figure 7.

figure 7

Protective shoe with U-bend ready to be inserted

The U-loop runs the full length of the pile with the U-bend fixed in the shoe, with five clips used to fix the pipe in place. This attachment approach has been successfully adopted for mounting U-Loops onto the outside of precast concrete piles.

The ground conditions for the trial were stiff chalky boulder clay to 6m depth overlying stiff Oxford Clay.

To address the potential risk of damage occurring to the polyethylene U-loops during installation a retrievable steel shroud of 100mm by 100mm by 7000mm fitted over the U-loop was used in the trial. The steel shroud is shown in figure 8 and protected the loops from damage and scratching during installation.

figure 8

Sheet pile setup with shroud in place ready for installation

The installation method of using a retrievable steel shroud means that thermally enhanced grout could be injected into any void left on extraction of the shroud.

Thermally enhanced grout refers to grout with a higher thermal conductivity (1.3W/m-K) compared with soil. This would enhance the thermal contact potentially improving the output of the system. This enhancement was not modelled as part of this study.

The sheet pile was setup in a vibratory clamp at the pile head and successfully driven to the targeted depth. Vibratory installation method was selected to simulate a harsh installation method. The steel shroud was subsequently extracted.

6.0 Cost comparison

The estimated costs associated with installing a traditional thermal concrete secant pile and thermal steel sheet piles are presented in table 1. The cost of the secant and sheet piles is not included. Given the different spacing of each configuration the costs have also been broken down into per metre rate.

The analysis indicated that the sheet pile U-loop configuration is a viable alternative to thermal concrete secant piles, with a cost of approximately £125 per installation (approximately £180/m run in plan). This is comparable to the traditional secant pile of around £150/m run in plan. At £536 per pile (approximately £380/m run in plan) the sheet pile concentric pipe option was more than twice the cost of the other configurations and therefore deemed economically unfeasible due to the high cost of the well headset and steel protective pipe.

Table 1 - Estimated cost of the modelled thermal pile configurations
 Cost elementsSecant pile with double U-loop at 1.2m intervalsSheet-pile with concentric pipe at 1.4m intervals    Sheet-pile with jetted U-loop at 0.7m intervals      Sheet-pile with jetted U-loop in the trial
U-loops   £150  - £75   £75
Attachment to cage   £30  -


Concentric pipe including welded 50mm diameter outer pipe   £436  -  -
Headwork  £100  -  -
Jet hole and pull in piles   -  £50  -
Estimated cost per pile/installation   £180  £536  £125  £80*
Estimated cost per metre run along the piled wall  £150  £380  £180  -

* See breakdown below (data provided by DCP):


  • 12m long 100 x 100mm angle (shroud) c/w top plate @£225.00 each*, recoverable and reusable
  • Pipe shoe 90X50 rectangular box section c/w angled plate @£10.00 each*, disposable fixing
  • Retaining clips five to six per pile, assume £1.80*, disposable fixing
  • Fixing bolt to secure U-bend to protective shoe £0.25*, disposable fixing
  • Consumables per pile £20


Labour (welder):

  • Prepare materials and fabricate angled shroud. Approx. one to two hours
  • Prepare protective shoe (angled plate) Approx. 30mins
  • Labour: welder and piling hand
  • Fitting of pipe shoe, plastic piping, angle shroud to each sheet pile - 30mins
  • Total time offline per pile two to three hours for trial. Expect an hour in real production £60.00/pile

7.0 Thermal response test

The next stage in the development is to undertake in situ thermal response tests on the thermal loops to validate the U-loop integrity and the numerical model results and optimise the configuration.

The response test should install at least three interlocking sheet piles with the U-loop attached to the central sheet pile. This is to incorporate the heat conduction across the interlocks. The length of installed sheet pile will also need to be sufficient to demonstrate the heat conduction behaviour by steel and by soil mass.

The thermal grout could be used around the U-loop to ensure good thermal contact between the U-loop and the soil. The thermal response test for the sheet-pile, the loop circulation fluid is heated by an electrical heater or cooled by a refrigerator, the fluid is circulated at a rate to ensure turbulence flow. Data collection during the testing should include the inlet and outlet temperatures and the circulation rate.

8.0 Conclusions

This analysis found that thermal sheet piles provide a viable alternative to traditional thermal concrete secant piles with U-tubes. While the selected modelling geometries indicated that secant piles have a slightly higher heat output rate and cumulative heat output, the variation between all three configurations was relatively small.

With regards to installation cost, the concentric pipe method generated the least amount of heat and was more than twice the cost to install at approximately £536 per pile (£380/m) and thermally less efficient. The materials and installation cost analysis shows that the sheet pile U-loop option at around £180/m is comparable to secant piling method at approximately £150/m.

This study concludes that steel sheet piles can deliver similar performance to traditional secant piles, with the U-loops costing only slightly more than the secant pile option.

A thermal response test is now required to validate the numerical model results and optimise the configuration. U-loops attached to the sheet piles provides adequate ground source heat energy to supplement or fulfil a development’s heating and cooling requirements.

Water has a higher thermal conductivity than air. Therefore, in a water side environment like a port, canal or river front, the performance of thermal sheet piles would be better because the face of the sheet pile wall being submerged below flowing water rather than exposed to air as in the case of a basement.


The authors would like to acknowledge the funding provided by the Steel Piling Group to undertake this study as well and the assistance and review provided by the vice-chairman of the Steel Piling Group, Chris Barker, in the preparation of this paper.


[1] Ground Source Heat Pump Association (2012), Thermal Pile Design, Installation & Materials Standards, Issue 1.0

[2] Environment Agency (2009), Evidence: Ground Source Heating and Cooling Pumps – State of Play and Future Trends

[3] Loveridge, F and Powrie, W (2013), Pile Heat Exchangers: Thermal Behaviour and Interactions,

Proceedings of the ICE - Geotechnical Engineering, 166 (2), 178-196.

[4] Brandl H (2006), Energy foundations and other thermoactive ground structures. Geotechnique, 56(5): 81-122

[5] Nicholson, D, Chen, Q, Pillai, A, Chendorain, M (2013), Developments in Thermal Pile and Thermal Tunnel Linings For City Scale GSHP Systems, Proceeding of 38th Workshop on Geothermal Reservoir Engineering, Stanford University, Stanford, California, February 11-13, 2013

[6] Soga, K, Qi, H, Rui, Y, and Nicholson D, (2014). Some considerations for designing GSHP coupled geotechnical structures based on a case study. 7th International Congress on Environmental Geotechnics. Melbourne, Australia

[7] Omer, A M (2008). Ground-source Heat Pumps Systems and Applications, Renewable Sustainable Energy Reviews, issue 12, pp344-371

[8] Livermore Software Technology Corporation (2007), LS-Dyna Keyword User’s Manual

[9] Engineering Tool Box (2007)


Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.