Alan Auld Member Name
Senior Consultant
Two of the greatest challenges for deep excavation and tunnelling are ground stability and control of groundwater. The risks associated with these challenges extend beyond immediate excavation stability issues, as there can be significant cost and programme implications for the project if these risks are not appropriately managed.
There are numerous ways to manage stability and groundwater inflows for deep excavation or tunnelling projects, but one that is sometimes overlooked hails straight from nature’s winter wonders. However, thanks to the development of artificial freezing techniques, the excavation doesn’t need to be located close to the polar regions or at high altitudes to reap the benefits of frozen ground.
For more than 150 years, engineers have applied the benefits of freezing to mining and tunnelling, using artificial ground freezing (AGF) techniques to stabilise the ground and limit groundwater inflow.
How does ground freezing work?
Anyone who has used a home freezer, let alone dug into frozen ground, will have observed one of the simplest and most obvious natural truths: frozen material is hard, stable and impermeable.
Freezing causes any water within the ground to turn to ice. In saturated low-strength soils (particularly sands, but also silts and clays), this increases the strength and hence the stability of the ground. As the frozen ground is now solid, it resists the ground and groundwater from flowing into the excavation.
Two systems for artificially freezing the ground are available, and the choice depends on the project’s circumstances. Both require a series of cased freeze holes to be drilled around the perimeter of the excavation. In the first system, known as calcium chloride freezing, brine is chilled in a refrigeration plant and pumped in a closed-circuit through pipes in the freeze holes at a temperature of between minus 250C and minus 350C. The brine-filled pipes cause the ground around them to freeze, gradually creating a frozen ‘wall’. The second system, a faster yet often more expensive alternative, is liquid nitrogen (LNG) freezing, in which liquid nitrogen is pumped into the holes at minus 1900C for a more rapid freeze, then exhausted to the atmosphere.
Regardless of the type of coolant chosen, the whole process is temporary: once construction is complete, the system is turned off, the ground thaws and the groundwater system returns to its original condition.
Starting out with a site investigation
It’s one thing to install and operate the ground-freezing equipment, but it’s quite another thing to understand the characteristics and behaviour of frozen ground and to design appropriate structures for the scheme. Specialist capability is required here – and the first step is procuring a thorough site investigation.
The site investigation needs to gather a range of data relating to the ground conditions from borehole samples, in-situ tests and laboratory testing of core samples. The specialist is trying to build a strong understanding of the geological setting (including the stratigraphic sequence and the spatial inter-relationships, thickness, strength and stability of the units) and the hydrogeological setting (number and nature of aquifers, hydraulic properties of individual geological units, direction of groundwater flow and background water quality).
Designing the freeze
Managing Underground Construction Risks Through Site Investigation
The landscape below our feet is unseen and uncertain – so any construction project involving major subsurface excavation will unearth a range of complexities and risks. A comprehensive site investigation allows us to proactively manage risk by providing insight into the likely conditions to be encountered underground. This is the key to a solid start and a stable project with a lower likelihood of unwanted surprises.
The ground-freezing contractor will determine the appropriate coolant type (e.g. brine or LNG), the capacity of the refrigeration plant, the temperature and flow rates for the coolant and the configuration of the freeze tubes. They will also design a system to monitor ground temperature.
Specialist input is needed to address the thermal design, which predicts the freeze rates through the ground. Thermal design confirms the minimum time before the excavation can take place, which will depend on forming a fully frozen wall with sufficient strength to resist ground and water pressures. The predicted growth rates of the frozen ground are compared with the actual rates observed by in-situ monitoring.
To conduct the thermal design, information is required on the temperature of the coolant, the initial ground temperature, the radius of the freeze tubes, and the thermal conductivities and heat capacities of the unfrozen and frozen ground.
The growth of the freeze wall can then be calculated, and an estimate made of the power required and number of refrigeration units needed.
Once the thermal design is completed, excavation design will require specialists who can predict the behaviour of frozen ground and design structures for frozen ground. The design of the frozen ground structure will include the determination of the unfrozen and frozen ground properties, drawing on the information gathered during the site investigation. Strength calculations and predictions of the expected deformation during excavation can then be made based on the design loads.
A further specialist element in the design is predicting ground heave and thaw settlement. Predictions can be made using specialist finite element software based on soil type and soil frost susceptibility.
Ground freezing in action
Many deep mine shafts have been sunk in the UK, particularly for the coal industry, using artificial ground freezing to stabilise the ground and prevent groundwater ingress during excavation. For example, during the sinking of British Coal’s Selby Mine shafts, 10 concrete-lined shafts, with internal diameters generally between seven and eight metres and depths of between 383 and 1043 metres, were sunk using ground freezing to be able to penetrate the water-bearing Bunter Sandstone and the unstable Basal Sands. Freeze depths ranged from 273 to 305 metres.
Many of the German coal mine shafts, with similar internal diameters and depths of more than 1000 metres, required the use of ground freezing to facilitate the penetration of unstable, water-bearing, strata comprising sands, silts and clays. Freeze depths of up to 600 metres were employed on these projects.
Similarly, in Canada during sinking of the shafts for the potash mines in Saskatchewan artificial ground freezing was required to deal with the unstable, water-bearing Blairmore strata conditions at depth. Freeze depths adopted were over 600 metres and, more recently, 720 metres.
As well as involvement in projects using artificial ground freezing for deep mine shafts, artificial ground freezing has been used for civil engineering works such as access and ventilation shafts, cross passage construction, TBM recovery and box-jacked tunnels. In many such situations, artificial ground freezing has been adopted because more conventional methods of temporarily controlling the ground and groundwater have been deemed not feasible.
Harnessing the power of nature has proved to be very reliable and effective in a range of applications and soil types. However, to get the best out of artificial ground freezing, upfront investment in specialist site investigation and design will help avoid the pitfalls associated with the method when it is not properly designed or implemented.
This article is part of a series of insights to help you manage ground risk when planning, procuring, designing and constructing underground infrastructure. We’ll explore other detailed aspects in the series, giving you a broader picture of a best-practice approach to managing ground risk. Follow us on LinkedIn or subscribe to our emails to be notified about upcoming articles in this series and other valuable content.
