| Front page | | Contents | | Previous | | Next |

Hydraulisk frakturering udført ved vandret boreteknik - Design og anlæg

Summary and conclusions

The objective of this project has been to establish whether hydraulic fracturing will be a cost-effective remediation technique in low permeable glacial deposits. At the test site, fracturing has been applied to improve the hydraulic effect of horizontal drainpipes inserted into the moraine clay. Remedial pumping from the drainpipes was carried out by use of Dual Phase Extraction (DPE).

The report gives an account of the design of the remedial system, experiences during the physical establishment of the system, and the results of a series of short-term hydraulic tests of the performance of the installed drain pipes. The long term effectiveness and economy for the drain system and experience acquired during operation will be reported in a separate report after a total remediation period of approximately 3 years.

At the test site, a heavy contamination with chlorobenzene and anilines has been detected from the soil surface to a depth of approx. 5 m below ground level. The horizontal extent of the "hot-spot" has been estimated at approx. 500 m2.

The site geology comprises fill in the upper 1-2 meters with moraine clay beneath this to a depth of approx. 23 m, where the "Danien" chalk layer is found.

In the moraine clay layer, any chalky deposits have generally been leached from upper 0.5 m, while the redox boundary is observed as a colour shift from brown to grey approx. 2.3 m below the clay surface. The moraine clay is assumed to be naturally fractured to a depth of 3 - 5 m.

At a depth 16 - 18 m, there are lenses of moraine gravel in the moraine clay. This is a water-bearing layer, separated from the primary water reservoir (in the chalk aquifer) by a 5-7 m thick clay layer. The water table in the primary reservoir is artesian with a hydraulic pressure level 1.5-2 m below ground level.

The quaternary deposits of moraine clay constitute a hydraulically linked secondary reservoir with a downward vertical pressure gradient of approx. 0.03 (m/m). The hydraulic pressure level measured in the top of the moraine clay (2 m below ground level) thus shows a pressure level approx. 0.5-0.75 m higher than measured in the primary chalk reservoir. Slug tests carried out in short filters (l = 30 cm, diameter = 25 mm) show a hydraulic conductivity in the moraine clay of approx. 1.5-3.2e-07 m/s in the upper 2-3.5 m below ground level, decreasing to approx. 3.5e-08 m/s in a depth of 4.3 m.

On basis of a risk assessment, it was initially assessed that the most suitable remediation technique for this locality would be to hinder the naturally occurring downward transport of chlorinated solvents. It was assessed that this could be done by creating an upward gradient by remediation pumping from 3 -5 parallel and horizontally installed drain pipes in a depth of approx. 4.5 m below ground level in the hot-spot area.

During the detailed system design, it was decided to replace the 3-5 drainpipes originally planned with 2 horizontal drainpipes installed with hydraulic fracturing of the moraine clay. These 2 horizontal hydraulic fractured drain systems were expected to give the same hydraulic effect as the "standard" drains, and furthermore would be more economical during operation of the remediation system.

Based on experiences in other lands, it was expected that the fractures induced by hydraulic fracturing would spread to a distance of 4 - 8 meters from the drainpipes (corresponding to a total fracture zone width of 8 -16 meters).

Furthermore, a groundwater model (MODFLOW) was used to simulate the effect of hydraulic fracturing. Compared to a "standard" drain system, the simulation of the hydraulic fractured horizontal drains indicated that the inflow of water could be increased 3 - 5 times, and that a 2 - 3 times significantly larger range of effect on the water table could be achieved.

The design of the fracturing process and the description of the special tools required for the process have to a great extent been based on assistance rendered by an American sub-consultant. These aspects are described in detail in this report, and are of importance for future projects.

As part of the design, lab tests were carried out on a number of core samples of the intact moraine clay from the test site. The objective was to define the determining geological and geo-technical parameters for the moraine clay. On the basis of these tests, the direction of the induced fractures in the moraine clay was expected to be mainly horizontal.

Based on the experiences from the construction phase, the following conditions require special attention:

1. Handling of the fracturing slurry used for injection in the pre-cut fractures demands a great deal of experience and care by the contractor.
2. The work requires special equipment for pre-cutting of fractures (crack cutting tools), fracturing (fracturing machinery), etc., as well as a special pump that can cope with the high viscosity slurry.

These conditions are described in detail in the report.

Eight of the ten planned fractures were established along the 2 horizontal borings, and 35 m of filter screen were installed in each of the borings.

It was initially planned that a minimum of 120 l "propant" (quartz sand mixed in the slurry) per fracture would be injected to achieve the desired average fracture radius of 3 mm for a minimum of 4 m length of the fracture. In the actual test system, 4 of the 8 fractures were injected with 160-240 l propant, while the remaining 4 were injected with approx. 120 l.

The price for the installed horizontal drain pipe in unfractured and hydraulically fractured moraine clay amounted to approx. kr. 3,400 and 7,100 per linear metre, excl. VAT respectively. The hydraulically fractured drain pipe system is therefore at present at least twice as expensive to install as "standard" drainpipes.

It is assessed that in future projects, an even greater price difference can be expected, and it is suggested that a metre price of about 10-15,000 kr. excl. VAT per metre installed fractured drainpipe should be used for economic proposals, until more experience concerning the fracturing technique has been gained.

The actual spreading of the fractures has been assessed by visual inspection as well as mineralogical analysis of core samples taken at a distance of up to 4.5 m perpendicular to the drainpipes. Three fractured zones out of a total of eight were investigated by these techniques.

An essentially horizontal fracture zone spreading to a distance of at least 4 m from the drainpipe was localised for one zone. The other two fracture zones showed steeply inclined fractures rising 40-50? from the horizontal plane and with relative horizontal distances of at least 3 m from the drainpipes.

Furthermore, some fractures extended to the soil surface ("blow-up"), which might be due to formation of short-circuits via natural fractures and other inhomogenities in the moraine clay.

It can be concluded that it is possible to establish 1-20 mm sand-filled fractures with a range of at least 3-4 m on either side the drain (corresponding to a total fracture zone width of 6-8 metres). The average aperture is estimated at 2-4 mm, which indicates a surface area of 30-120 m2 has been established in the fracture zone. Furthermore the investigation has shown that it is extremely difficult to predict the orientation of the fractures.

To be able to assess the hydraulic effect obtained by hydraulic fracturing of the two drain systems, a reference drain was also established. The reference drain was placed at the same level and constructed of the same materials and in the same dimension as the fractured drain systems, but was placed at a distance of approx. 27m from the fractured systems. It is assessed that the geology surrounding the reference drain is representative for the conditions around the fractured drain systems.

A DPE-test over the course of 2-7 days has been carried out on both the two fractured drains and the reference drain. During the tests, the water yield from the drains and water table measurements in a number of observation wells – all installed with screens at different levels - have been recorded. The water yield from the fractured drain systems is approx. 40-80 litres/filter screen/day, while the water yield from the reference drain was approx. 90 litres/filter screen/day. Contrary to all expectations, no significant increase in the water yield from the fractured drain compared to the reference drain was observed during the short DPE-test. And whether this lack of increase in water yield also applies over a longer time scale can only be determined after the first year of monitoring.

The lowering of the groundwater level around the fractured drain system occurs rapidly in the filter screens with direct contact to the established fractures – a lowering of several meters is achieved after only few hours of draining. The actual propagation of fall in the water table along the fractures and into the moraine matrix only happens slowly, and it is assessed that the water level is only lowered by 1-10 cm at a distance of approx. 1 m perpendicular to the fractures after 2-3 days.

For the non-fractured drain, only a limited propagation of the fall in water table is observed in that the pressure is primarily transmitted through the matrix itself. Only a total fall of 1-10 cm at a distance of 1 – 2 m’s from the drain has been observed.

From the short DPE-tests, it can thus be concluded that the unfractured and fractured drain systems affect a zone on either side of the drains of approx. 1 m and 3- 4 m respectively (corresponding to a total zone width of 2 and 4 - 8 m, respectively). This is also in accordance with the model calculations and experiences in USA. However the model calculations had predicted increases of 3 - 5 times the water yield in the hydraulically fractured drain systems and this increase in yield was not observed in the practical tests. Furthermore, it was not possible to obtain steady state potentiometric measurements during the very short tests, and therefore documentation concerning the changes in water table and reversal of the hydraulic gradient can only be collected during the on-going monitoring during the first year of operation.