Writing and Publications

Disposal of Dirty Liquids Using Slurry Fracture Injection

Dusseault , M . B ., Bilak , R . A ., and Rodwell , G . L ., SPE 37907, Proc. 1997 SPE/EPA Expl. and Prod. Env. Conf. , Dallas, TX, March 3-5, 1997, pp. 193-202.

This paper was prepared for presentation at the 1997 SPE/EPA Exploration and Production Evironmental Conference held in Dallas, Texas, 3-5 March 1997.

Abstract

An automated waste liquid injection system for dirty liquids was designed and implemented in an unconsolidated sandstone formation in southeastern Alberta. This development was based on previous waste sand injection in central Alberta. Waste liquids consisting of contaminated water, heavy oil, and fine-grained particulate matter are being diluted and injected under formation fracture pressures. Monitoring for environmental security and process control includes precision tiltmetres and comprehensive pressure analysis. Infectivity, transmissivity, reservoir pressure evolution, and fracture closure pressure calculations based on pressure fall-off and step-rate tests allow the formation state to be continuously re-evaluated. The reservoir is not being impaired by the continued injection of dirty liquids.

This article first describes Slurry Fracture Injection (SFI) history and principles, the site geology, and recompletion strategy. Then, data from the trial period and the summer operations are presented and analyzed. The use of SFI in this new application seems to be highly successful, and the nature of the process and controllable factors are discussed in the context of preserving target formation injectivity.

Introduction

Slurry Fracture Injection (SFI) for waste solids disposal was developed over the last decade in Western Canada for the large volumes of sand produced along with heavy oil in the production process known as "Cold Production", SFI is a variation of hydraulic fracturing with modifications to cope with requirements for episodic large-volume injection at moderate rates for periods of months to years in the same well. A major aspect of SFI implementation is continuous monitoring to demonstrate containment and environmental security; this has aided in approval by regulatory agencies in Alberta, Saskatchewan, and more recently in California. Monitoring is also used for process optimization, and allows the reservoir state to be carefully tracked during the process.

The first experimental SFI trial was executed in Saskatchewan by Mobil Canada at their Celtic Project in the years 1989-1990. Over a period of several years they injected approximately 10,000 m 3 of fine-grained produced sand as a diluted aqueous slurry into a 35 m thick unconsolidated quartzose sandstone at a depth of 690 m. The target formation was an oil-free sand with 1-4 Darcy permeability, and injection took place at about the third point from the base. This experiment showed that formation injectivitity to a sand-water slurry could be re-established for an episodic injection strategy, providing that the injection well was properly operated.

The SFI approach was accepted for produced sand by the Alberta Energy and Utilities Board during the period 1993- 95, and five sand disposal projects have taken place since then, with several more having recently been approved. The waste injected at these sites is a fine-grained sand contaminated with 1-4% (weight percent) of viscous, asphaltene-rich heavy oil.

On three projects, it was necessary to cope not only with sand and the waste water used for making the slurry, but also with two materials known locally as "gorp" and "slops".

On the Celtic Project, the gorp was a stable emulsion of variable composition generated during heavy oil (9-18 AP1) exploitation using the Cold Production technique. It is composed of approximately 40-85% H 2 O, a few percent of fine-grained silicate minerals (clays, SiO 2 , etc.), and an asphaltene-rich fraction of the heavy oil being produced. Chemical or thermal breaking proved difficult and costly, therefore disposal by SFI was considered. Inclusion of large quantities of gorp in the slurry and injection at typical SFI rates of 0.8-1.5 m 3 /min caused problems in the target formation, and in one case even generated small pressure responses in monitoring wells (from an abandoned heavy oil field) 90 m above the 690 m deep injection point.

For a complete copy of this article, click here to initiate download.