WO2023087054A1

WO2023087054A1 - Cooling and fracturing a rock mass

Info

Publication number: WO2023087054A1
Application number: PCT/AU2022/051367
Authority: WO
Inventors: Gabriel Jose AMORER
Original assignee: Newcrest Mining Limited
Priority date: 2021-11-16
Filing date: 2022-11-16
Publication date: 2023-05-25

Abstract

A method of fracturing and cooling a rock mass includes injecting a cryogenic fluid into a hole in a floor of a pit in a mine, with the cryogenic fluid fracturing a rock mass surrounding the hole and cooling the surrounding rock mass via heat exchange with the rock mass.

Description

COOLING AND FRACTURING A ROCK MASS

TECHNICAL FIELD

The invention relates to the mining industry.

The invention relates to cooling a rock mass.

The invention relates to fracturing a rock mass.

The invention relates particularly although by no means exclusively to injecting a cryogenic fluid, such as a cryogenic liquid, such as liquid nitrogen, into a rock mass and cooling and fracturing the rock mass.

BACKGROUND ART

The invention is concerned with a problem of increasing pit temperatures faced by the Newcrest Lihir mine in PNG.

The problem is relevant to the current mine as mining progresses to greater depths in the mine.

The problem is also relevant to possible mine expansions.

The problem is also relevant to other mines in “Ring of Fire” countries that have hot pits.

The invention is also relevant to other mines where the pits are not hot as is the case with mines in “Ring of Fire” countries but nitrogen and other gases are a viable rockbreaking option that are an opportunity to reduce the reliance on explosives.

The current mining process at the Newcrest Lihir mine includes (a) drilling holes in the pit floor, (b) filling the holes with emulsion explosives, (c) positioning detonators in the holes, (d) detonating the explosives, and (e) excavating blasted ore.

The Newcrest Lihir mine is located in a volcano crater and the pit floor temperatures are currently quite high and will continue to increase as mining continues in the current mine.

The hot pit limits the options for explosives and detonators.

Currently, available explosives and detonators can be used at temperatures up to 130- 150°C.

The above description is not an admission of the common general knowledge in Australia or elsewhere. SUMMARY OF THE DISCLOSURE

In broad terms, the invention is a method of fracturing and cooling a rock mass that includes injecting a cryogenic fluid into a hole in a floor of a pit in a mine, with the cryogenic fluid fracturing a rock mass surrounding the hole and cooling the surrounding rock mass via heat exchange with the rock mass.

The pit may be a hot pit.

The term “hot pit” is understood herein to mean any pit where pit temperature limits the options for the selection of explosives for drilling and blasting operations in the pit.

The cryogenic fluid may be a liquid when it is injected into the hole and undergo a phase change and form a gas in the hole.

The cryogenic fluid may be a gas when it is injected into the hole.

Further to the above, when the cryogenic liquid includes a cryogenic liquid and the pit is a hot pit, the method may include injecting the cryogenic liquid into the hole (for example, a hole drilled to a depth of 10-14 m or any other suitable depth) in the floor of the hot pit and forming a cooled gas in the hole that fractures the rock mass surrounding the hole, with the gas cooling the surrounding rock mass via heat exchange with the rock mass.

The term “cryogenic liquid” is understood herein to mean any liquid that is capable of cooling a section of a rock mass.

The cryogenic liquid may be liquid nitrogen.

Other cryogenic liquids may include liquid argon, liquid carbon dioxide, and liquid helium.

The method provides an opportunity to cool a surrounding rock mass sufficiently so that a wider range of available explosives can be packed into the hole and detonated with detonators in the hole than would otherwise be the case.

The method also provides an opportunity to fracture a surrounding rock mass to an extent that it is possible to reduce the amounts of explosives that are required to blast the rock mass for subsequent excavation of broken rock mass.

By way of example, when nitrogen is the cryogenic fluid, nitrogen expands approximately 700 times when changing phase from a liquid to a gas and this level of expansion can cause substantial fracturing of the surrounding rock mass when the expansion occurs in a hole.

In some embodiments, it may not be necessary to use any explosives, and the rock mass may be fractured to an extent that direct removal of the rock mass is possible without the use of explosives. The method may include continuously or periodically discharging gas from the hole to remove heat from the hole and the surrounding rock mass.

The method may include continuously or periodically discharging gas from the hole at least while injecting the cryogenic fluid into the hole.

The method may include:

(a) drilling a hole in the rock mass,

(b) positioning a casing in at least an upper section of the drilled hole to line the hole,

(c) positioning a well-head in the opening of the hole and thereby closing the opening, the well-head having an injection opening for the cryogenic liquid and optionally a pressure relief valve, and

(d) injecting the cryogenic fluid, such as a cryogenic liquid, into the hole.

The casing may be formed from fibreglass or any other suitable material that can retain mechanical properties, such as toughness, at the temperatures in the drilled hole after injection of the cryogenic liquid.

The method may include selecting the cryogenic fluid and the amount of the cryogenic fluid for injection into the hole having regard to factors, such as (but not limited to) the depth of the hole and the pit geology and the amount of the rock mass to be cooled to a selected temperature for a selected time.

The method may include continuously or periodically discharging gas from the hole to remove heat from the hole and the surrounding rock mass optionally at least while injecting the cryogenic fluid into the hole.

By way of example, the well-head may include an outlet opening for gas to flow from the hole.

The method may include injecting the cryogenic fluid into a plurality of spaced-apart holes in a section of a mine pit to be mined and injecting the cryogenic fluid into the holes and fracturing a rock mass surrounding each hole and cooling the surrounding rock mass via heat exchange with the rock mass.

The spacing of the holes may be any suitable spacing.

The method may include selecting the spacing of the holes to be the spacing required for a subsequent explosive blasting program so that the same holes can be used for injecting the cryogenic fluid and for explosives.

Typically, the spacing is at least 6 m between adjacent holes.

The plurality of holes may be in any suitable array. The invention is also a method of drilling and blasting a section of a mine pit that includes:

(a) injecting a cryogenic fluid, such as a cryogenic liquid, as described herein, into a rock mass in a section of a mine pit to be mined and lowering the temperature of the rock mass; and

(b) blasting the rock mass.

The method may include drilling a plurality of spaced-apart holes in the section of the mine, injecting the cryogenic fluid into the holes and fracturing a rock mass surrounding each hole and cooling the surrounding rock mass via heat exchange with the rock mass, positioning explosives in the same or newly-drilled holes, and blasting the rock mass.

The section of the mine to be drilled may be any suitable size depending on the mine and the mine plan for the mine.

The invention is also a method of mining that includes:

(a) injecting a cryogenic fluid, such as a cryogenic liquid, as described herein, into a rock mass in a section of a mine pit to be mined and lowering the temperature of the rock mass,

(b) blasting the rock mass,

(c) excavating the blasted rock mass, and

(d) transferring the excavated rock mass to downstream processing operations.

The method may include drilling a plurality of holes in the section of the mine, injecting the cryogenic fluid into the holes and fracturing a rock mass surrounding each hole and cooling the surrounding rock mass via heat exchange with the rock mass, positioning explosives in the same or newly-drilled holes, and blasting the rock mass.

The method may include lowering the temperature of the rock mass at least 10°C.

The method may include lowering the temperature of the rock mass at least 15 °C. The method may include lowering the temperature of the rock mass between 20°C and 50°C, typically between 20°C and 30°C.

The method may include lowering the temperature of the rock mass to a suitable temperature for drilling and blasting the rock mass for a period of time of up to 8 hours, typically up to 10 hours, and more typically up to 12 hours, before the temperature rises above the suitable blasting temperature.

The term “suitable temperature for safe drilling and blasting the rock mass” is understood herein to mean a rock mass temperature that is within an accepted temperature range for commercially available explosives to be used to blast the rock mass.

The downstream processing operations may be any suitable processing operations for recovering a valuable metal from the excavated rock mass. The valuable metal may be any one of gold, copper, molybdenum and other metals.

The invention is also a mine that includes a system for selectively fracturing and cooling a rock mass in a section of the mine to be mined, the system including:

(a) a plurality of drilled holes with openings and well-heads that close the openings, each well-head having an injection opening for a cryogenic fluid, such as the cryogenic liquid, and optionally a pressure relief valve; and

(b) an apparatus for injecting the cryogenic fluid, such as the cryogenic liquid, into the holes via the well-heads.

Each well-head may include an outlet opening for gas to flow from the hole at least while injecting the cryogenic fluid into the hole.

The apparatus (b) may include a source of the cryogenic fluid, such as the cryogenic liquid, and pipe works, such as a network of pipes, and pumps for transferring the cryogenic fluid to the well-heads.

The pipeworks may include a network of pipes in fluid communication with the wellheads for supplying the cryogenic fluid, typically under pressure, to the holes via the wellheads.

In a situation in which the cryogenic fluid is liquid nitrogen, the source of the cryogenic fluid may include (i) an oxygen plant for producing oxygen and nitrogen in gas forms from air and (ii) a unit for liquefying nitrogen gas.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention is described further below by way of example only with reference to the accompanying drawings, of which: Figure 1 is a simplified perspective view of an embodiment of a system for cooling and fracturing a rock mass in a section of a mine in accordance with the invention;

Figure 2 is an enlarged perspective view of a section of Figure 1 ;

Figure 3 is a schematic diagram and horizontal and vertical cross-sections of the diagram illustrating an embodiment of a closed circulating system for cryogenic fluid injection into a drilled hole in a rock mass and the fracture and cooling mechanism in the section as a consequence of the use of the system of the type shown in Figures 1 and 2;

Figure 4 is a schematic diagram of a tri-axial loading system used in laboratory and modeling studies carried out by the Colorado School of Mines;

Figure 5 is a series of diagrams of a concrete sample with embedded thermocouples used in the laboratory studies;

Figure 6 is a series of images of a wellhead device used in the laboratory studies;

Figure 7 is a schematic of liquid nitrogen (LN2) injection in the laboratory studies;

Figure 8 is graph of the temperature profile of the 2^nd LN2 circulation treatment on sample Cl in the laboratory studies;

Figure 9 is a graph of the results of pressure decay tests before and after the 2^nd LN2 circulation treatment on sample Cl;

Figure 10 is a graph of water loss versus time of sample C2;

Figure 11 is graph of the temperature profile of the LN2 circulation treatment on sample C3;

Figure 12 is a graph of pressure decay tests before and after the LN2 circulation treatment on sample C3;

Figure 13 is graph of the temperature and pressure profiles of the first LN2 high- pressure injection on sample C4;

Figure 14 is a graph of the results of pressure decay curves before and after the two LN2 treatments on sample C4;

Figure 15 is graph of the temperature profile of the LN2 circulation on sample C5;

Figure 16 is a graph of pressure decay curves before and after the LN2 circulation treatments on sample C5;

Figure 17 is a graph of the temperature profile of the LN2 circulation on sample C6:

Figure 18 is graph of an early stage temperature profile of the LN2 circulation on sample C6;

Figure 19 is a graph of pressure decay curves before and after the LN2 circulation treatments on sample C6; Figure 20 is a graph of pressure and stress profiles during fracturing attempts and pressure decay curves before and after each fracturing treatments on sample C7;

Figure 21 is a graph of pressure decay curves of sample C8 before and after the fracturing treatment;

Figure 22 is a graph of temperature and pressure profiles of the LN2 circulation on sample C8;

Figure 23 is a graph of temperature and pressure profiles of the LN2 circulation on sample C9;

Figure 24 is a graph of temperature and pressure profiles of the LN2 injection on sample CIO;

Figure 25 is a graph of temperature profiles of the 1st LN2 treatment on sample Cl 1;

Figure 26 is a graph of temperature profiles of the 2nd LN2 treatment on sample Cl 1;

Figure 27 is a graph of temperature profiles of the 3rd LN2 treatment on sample Cl 1;

Figure 28 is a graph of pressure decay curves of sample Cl before and after the 2^nd LN2 treatment;

Figure 29 is a graph of temperature profiles of the 1^st LN2 treatment on sample C12;

Figure 30 is a graph of temperature profile of the 2^nd LN2 treatment on sample C12;

Figure 31 is a graph of pressure decay curves of sample C12 before, after the 1st and 2^nd LN2 treatment;

Figure 32 is a graph of temperature profiles and LN2 consumption of the LN2 circulation treatment on C14;

Figure 33 is a graph of temperature profiles and LN2 consumption of the 1st LN2 injection treatment on sample C13;

Figure 34 is a graph of temperature profiles and LN2 consumption of the 2^nd LN2 injection treatment on sample C13;

Figure 35 is a graph of temperature profiles and LN2 consumption of the 3^rd LN2 injection treatment on sample C13;

Figure 36 is a graph of temperature and pressure profile of the LN2 circulation on sample C15;

Figure 37 is a graph of temperature and pressure profile of the LN2 circulation on sample Cl 6;

Figure 38 is a graph of temperature and pressure profile of the LN2 circulation on sample C17; Figure 39 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample Cl 8;

Figure 40 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C19;

Figure 41 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C20;

Figure 42 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C21;

Figure 43 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C22;

Figure 44 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C23;

Figure 45 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C23;

Figure 46 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C25;

Figure 47 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C26;

Figure 48 is a thermal image of sample C26 during LN2 injection after LN2 was escaping from the sample;

Figure 49 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C28s;

Figure 50 is a graph of pressure decay curves before and after the LN2 injection treatment on C28s;

Figure 51 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample C29s;

Figure 52 is a thermal image of sample C29s during LN2 injection;

Figure 53 is a graph of pressure decay curves at room temperature, before and after each LN2 treatment on sample Fl;

Figure 54 is a graph of temperature profile and LN2 consumption of the 1^st LN2 injection treatment on sample Fl;

Figure 55 is a graph of temperature profile and LN2 consumption of the 2^nd LN2 injection treatment on sample Fl; Figure 56 is thermal images of the surfaces 2, 3 and top of Fl before the end of injection;

Figure 57 is a graph of pressure decay curves at room temperature, before LN2 treatment on sample F3;

Figure 58 is images of the results a bubble test during the first pressure decay test at room temperature on sample F3;

Figure 59 is a graph of pressure decay curves at room temperature, before and after LN2 circulation on sample F2;

Figure 60 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample F2;

Figure 61 is a graph of pressure decay curves before and after LN2 circulation on sample F4;

Figure 62 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample F4;

Figure 63 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample F5;

Figure 64 is a graph of pressure decay curves before and after LN2 circulation on sample F6;

Figure 65 is a graph of temperature profile and LN2 consumption of the LN2 injection treatment on sample F6;

Figure 66 is a comparison of warm-up time from LN2 circulation treated rock samples;

Figure 67 is a graph of pressure decay curves of concrete samples with induced vertical fractures at 200 °C before any LN2 treatment;

Figure 68 is a Comparison of warm-up time and induced fracture morphology of the LN2 injection treated concrete samples with vertical fractures;

Figure 69 is a graph of the pressure decay curve of concrete samples with induced lateral fractures at 200 °C before any LN2 treatment;

Figure 70 is a comparison of warm-up time and induced fracture morphology of the LN2 injection treated concrete samples with lateral fractures; and

Figure 71 is a graph of pressure decay curves of field samples 200 °C before any LN2 treatment. DESCRIPTION OF EMBODIMENTS

As described above, the invention is a method of cooling and fracturing a rock mass that includes injecting a cryogenic fluid such as a cryogenic liquid into a hole in a floor of a hot pit in a mine (such as the Lihir mine of the applicant) and fracturing a rock mass surrounding the hole via heat exchange with the rock mass.

It is noted that the invention also extends to mines where the pits are not hot as is the case with mines in “Ring of Fire” countries but nitrogen and other gases, for example derived from a cryogenic fluid, such as a cryogenic liquid, are a viable rock-breaking option that are an opportunity to reduce the reliance on explosives.

Cryogenic cooling of a hot pit makes it possible to use a wider range of commercially available explosives in the mine than is the case without pit cooling.

As noted above, it is necessary to cool a surrounding rock mass of a hole in a hot pit because the heat in this mass will quickly increase the hole temperature and reduce the options for explosives for use in the hole.

An objective of the invention is to reduce the hole temperature and a surrounding rock mass sufficiently to allow the use of explosives and detonators not only immediately after drilling a hole but also for a period of at least 8 hours, typically for periods of 12-24 hours, that is needed typically to position explosives and detonators in a series of such holes and initiate the explosives in the holes.

The method also provides an opportunity to fracture a rock mass surrounding a hole to an extent that it is possible to reduce the amounts of explosives that are required to blast the surrounding rock mass for subsequent excavation of broken rock mass. There may be situations in which the extent of the rock fractures is such that it is not necessary to blast the surrounding rock mass at all.

The use of cryogenic liquids (such as liquid nitrogen) as a specific type of cryogenic fluid is for a quite specific reason. Liquid nitrogen (and other suitable cryogenic liquids) evaporates in a hole and expands quickly and generates pressure in the hole, with the resultant gas pressure fracturing (i.e. cryogenic fracturing) a surrounding rock mass to allow the gas to penetrate and cool the rock mass via heat exchange.

Figures 1-3 illustrate embodiments of a system for fracturing and cooling a rock mass in a section of a mine pit 1.

The section may be any suitable area in the pit 1. The selection of the location and the size of the area in the pit 1 will depend on a mine plan for a given mine. Figures 1-3 are described in the context of the Lihir mine.

As noted above, the Lihir mine has a hot pit, which limits the selection of explosives that can be used in the mine.

In the embodiment described in relation to Figures 1-3, the cryogenic fluid is liquid nitrogen, noting that the invention is not confined to this cryogenic fluid.

With reference to Figures 1-3, the system includes:

(a) a plurality of spaced-apart drilled holes 3 (for example, at depths of 10-14 m at spacings of 6-10 m, but could be any suitable depth and spacing) located in a section of a mine pit 1,

(b) well-heads 5 closing upper openings of the holes, with each well-head 5 having an injection opening for liquid nitrogen (not shown), a pressure relief (not shown) and a manually-operated shut-off valve 7 - all of which are standard components of well-heads used in the oil and gas industry; and

(c) an apparatus for injecting the liquid nitrogen into the holes 3 via the well-heads 5, with the apparatus including (i) a source of the liquid nitrogen 9 (see Figure 1), including a vehicle-mounted pump unit for pumping liquid nitrogen 9 to the wellheads 5 and (ii) a network of pipes 11 in fluid communication with the well-heads 5 for transferring liquid nitrogen from the tanks to the well-heads 5.

The drilled holes 3 may be in any suitable arrangement of holes.

The location of the drilled holes 3 may be selected to be the spacing required for a subsequent explosive blasting program so that the same holes can be used for injecting the cryogenic fluid and for explosives.

With reference to Figure 3, each well-head 5 is configured to inject liquid nitrogen into a hole 3 and to allow heated nitrogen gas to flow from the hole. The embodiment shown in Figure 3 is a different albeit similar embodiment to that shown in Figures 1 and 2. There are common elements.

The Figure 3 embodiment is a closed circulating system, with:

(a) liquid nitrogen being pumped downwardly through a central tube extending into the hole from the well-head 5 and evaporating in the hole and forming a gas, with substantial volume expansion,

(b) heated gas produced via heat exchange between (i) the surrounding rock mass and (ii) liquid nitrogen and evaporated liquid nitrogen flowing upwardly in the annulus around the central tube and from the well-head 5; and (c) the heated gas being passed through a heat exchanger that cools the heated nitrogen gas to a liquid and returns liquid nitrogen to the hole.

In the embodiment shown in Figures 1 and 2, the heated gas is allowed to vent to atmosphere from the hole via the well-heads 5.

With reference to Figure 1, the source of the liquid nitrogen 9 is in the form of a series of storage tanks 13 located in the pit 1 and the above-mentioned vehicle-mounted pump unit.

In other, although not the only other, embodiments, the liquid nitrogen is piped directly into the pit 1 from the storage facility external to the pit 1.

The Lihir mine includes autoclaves (not shown) to oxidise gold-bearing sulphides excavated from the pit and a plant (not shown) to produce oxygen from air for use in the autoclaves. The oxygen plant also produces significant amounts of nitrogen gas. In the embodiment described in relation to Figures 1-3, the system includes an oxygen plant and a unit (not shown) for liquefying nitrogen gas produced in the oxygen plant.

Liquid nitrogen has a boiling point of minus 195.8 °C. Therefore, it is necessary to store and transport liquid nitrogen below this temperature.

In use, in accordance with a mine plan, a section of the pit floor is selected to be the next section to be drill and blasted. Holes 3 are dilled into the pit floor in the section and the holes are lined and cased and well-heads 5 are positioned to close the upper open ends of the holes. A network of pipes 11 is positioned to supply liquid nitrogen to the holes. Liquid nitrogen tanks are positioned in the section and connected to the pipework via the vehiclemounted pump unit. The apparatus is then operated to deliver liquid nitrogen to the holes 3 and thereby cryogenically fracture the rock mass surrounding each hole. When cryogenic fracturing and cooling in the section of the pit is completed, the pipework 11, well-heads 5 and tanks are then moved to another section of the pit and the above-described method is repeated. The fractured holes 3 are then filled with suitable amounts of explosives, typically emulsion explosives, detonators are placed in the holes, and the explosives are detonated to blast the holes. Excavators then remove blasted rock mass from the pit, and the blasted rock mass is processed to recover gold and copper.

It is noted that in some embodiments, it may not be necessary to use any explosives, and the rock mass may be fractured to an extent that direct removal of the rock mass is possible without the use of explosives.

Laboratory and Modeling Studies The applicant engaged the Colorado School of Mines to conduct extensive laboratory and modeling studies of cryogenic fracturing and cooling of bore holes (in use, these will typically be drilled holes) via injection of liquid nitrogen. The following description of the studies is drawn from a report prepared by Colorado School of Mines for the applicant.

1.0 Introduction

The cooling process via liquid nitrogen (LN2) injection into a drilled hole in a geothermally active area is a complex process that involves factors including liquid flow, thermodynamics, heat transfer, and geomechanics (including fracturing of the formation rocks).

The Colorado School of Mines investigated cooling via LN2 injection and warm-up periods after LN2 injection into holes.

The Colorado School of Mines designed and conducted an extensive set of laboratory experiments to study the LN2 cooling effect in 20x20x20 cm cubic blocks of (a) concrete and (b) actual field rock samples under various temperature, pressure, and fracture conditions. The sample blocks are shown in Figure 5. The sample blocks have a central hole extending into the block from a top face 5. The other faces of the block are identified by the numerals 1, 2, 3, 4, and 6 in the Figure. Figure 5 also indicates the locations of thermocouples 1. 2, 3, 4, and 5 in the top face 5 of the sample block.

Table 1 is a summary of the complete experimental conditions and results for the samples, with the description after the Table providing more information.

Table 1

LN2

S Oamp ile Tests Cond Ji-.ti-on Fracture Ambient Heating ° „ Proced ,ure Tre . ating ° Max E „xp ,lod ,ed , i .ni .ected , N ..ote direction temp rerature device time p rressure amount

Intact / vertical / , Circulation/ , . __Z., _T ,

~ , degC . . . seconds psi Y/N kg fractured lateral ° injection ^r °

1 intact 19 N/A circulation 305 20 N Failed due to burst TC

Cl

2 intact 19 N/A circulation 900 19 N TC on casing

No 1N2 test. Heated in the oven

C3 1 intact 19 N/A circulation 1800 16 N TC on casing

0.95L 1N2, pressurize with

1 intact 20 N/A injection 1440 512 N ,, 1.22 g ^&N2, n .o p ,re-coold ,ow , n, no warm back recorded, TC on

C4 casing immediately after test 1 , pre-

2 intact 20 N/A injection 1184 425 N 1.22 cooldown to 1N2 temperature, TC on casing

C5 1 intact 20 N/A circulation 1730 15 N 4 embedded TC

C6 1 intact 20 N/A circulation 2400 17 N 4 embedded TC

2.0 Laboratory equipment

The equipment used for the hydraulic fracturing experiments included a tri- axial loading system, an injection pump, and data acquisition devices. Other methods to investigate and evaluate the induced fracture extent and conductivity were also adopted, such as dye solution injection, pressure decay tests, etc. In addition to the fracturing experiments, rock samples were also characterized by their thermal, hydrological, and mechanical properties via core measurement to provide enough physical properties input for later result analysis and simulations.

2.1 Tri-axial loading system

The tri-axial loading system was a pneumatic powered hydraulic press frame.

The tri-axial loading press at the Colorado School of Mines was employed to confine rock samples to desired polyaxial stress conditions. It consisted of three hydraulic pistons. Two of the pistons were in horizontal directions and perpendicular to each other. Both pistons were in a carbon steel containment ring with counter blocks to provide counter stress. Another one was in the vertical direction and fixed on the loading frame. The hydraulic pistons were powered by three manually controlled pneumatic pumps. The tri-axial loading system could provide up to 31.0 MPa stress in the horizontal direction and 41.4 MPa stress in the vertical direction on a 20x20x20 cm cubic block located in the center of the containment ring, as shown in the schematic drawings of the loading system in Figure 4.

2.2 Injection pump

The injection pump used for the hydraulic fracturing experiments was an ISCO 500 HPx, manufactured by Teledyne ISCO. The ISCO 500HPx is a large capacity high-pressure syringe pump. It can provide precise, predictable flow, and pressure control at flow rates from sub-microliter to 408 mL/min. The maximum output volume of one stroke was 507.38 ml before refilling. This pump could work within a pressure range of 69 KPa to 34.5 Mpa. This pump was capable of injecting various fluids, such as water, oil, and CO2, into boreholes of rock samples via stainless steel tubing.

2.3 Liquid nitrogen transfer vessel

Due to the ISCO pump’s limitation, it could handle the injection of cryogenic liquids like LN2. Thus, a liquid transfer vessel was assembled and installed to inject LN2 at high pressure up to 3.45 Mpa. The vessel was made of annealed stainless-steel tubing with 5.08 cm OD, 4.11 cm ID, and 70 cm length. The 5.08 cm OD tubing was reduced to 0.635 cm OD tubing using multiple tube fittings. The vessel tubing was rated for 6.9 Mpa, but 3.4 Mpa was the maximum recommended pressure at the vendor’s cryogenic temperature. The vessel was heavily insulated to minimize heat transfer and installed with multiple pressure relief valves (3.4 Mpa) for safety. The internal storage volume was 0.95 L.

2.4 Data acquisition and monitoring devices

The data acquisition and monitoring devices included the data acquisition device, a power supply, temperature, pressure sensors, and a pc. The sensors included temperature sensors and pressure transducers. Temperature sensors used in the experiments were Type T thermocouples (TC) made of copper and constantan and were suited for temperature measurements in the -200 to 350 °C range. The total number of temperature sensors was limited to nine by the available slots in the data acquisition device. Pressure transducers were used to monitor the pressure inside the drilled bore of a sample. The pressure transducers were rated up to 20.7 Mpa and were connected to the data acquisition system, which could provide real-time reading or monitoring at 2 Hz frequency while testing. 2.5 Heating oven

A heating oven was used to provide an experiment environment with elevated temperature. The heating oven was a model 51-550ER manufactured by Quincy Lab. It had large enough interior space to accommodate the rock samples along with flow lines. The temperature range could be as high as 287 °C with ±0.5 °C control stability, which satisfied the requirement for the test work.

An MTS loading frame manufactured by MTS Systems Corporation was employed to measure rock samples’ mechanical strength. The MTS loading frame could measure both load (force) and displacement simultaneously during tests. The test methodology for measuring tensile strength and uniaxial compressive strength in the laboratory followed the American Society of Testing and Materials (ASTM) standards.

3.0 Properties and preparation of samples

The rock samples were (a) concrete and (b) actual field rock samples.

Concrete samples were chosen due to their analogical properties with real rocks and convenience to make in the laboratory condition.

The use of concrete samples made it possible to improve experimental procedures and obtain more data from actual experiment runs without wasting field rock samples.

The actual field samples provided a more accurate representative situation for real- world applications.

Properties for both concrete and field rock samples were measured for characterization, analysis, and simulation of the LN2 cooling processes.

3.1 Concrete samples

Concrete samples were prepared by mixing Portland cement (commercial grade Portland Cement Quikrete Type I/II #1124) with coarse sand (Table 2) and water. The sand/cement and water/cement ratios were 2.5 and 0.55, as listed in Table 3. After mixing, the concrete slurry was first poured into a mold and left to cure overnight.

For samples embedded with thermocouples (TC), five TCs were embedded into the concrete slurry by jackets to the designated locations and depth (10 cm from the top) of the sample (see Figure 5). Two of the TCs were provided to measure the internal temperature variation along the y-axis, while another three were provided to measure the temperature variation in the diagonal direction. The concrete samples were cured in water. The underwater curing lasted one month to reach the maximum strength required for the concrete samples.

After water curing, the samples were removed from water and dried in ambient conditions before use. Table 1: Sieve analysis of sand used for concrete preparation

Sand Mesh Percentage

Size

4 100.0

8 99.4

16 86.8

30 58.4

50 27.6

100 6.9

200 2.4

Table 2: Sieve analysis of sand used for concrete preparation

Composition Amount Ratio to Cement

(kg)

Sand 4 2.5

Cement 1.6 1

Water 0.88 0.55 Mechanical properties, density, porosity, permeability, etc. of the concrete samples were measured before use.

Most of the mechanical properties were obtained from acoustic measurement, while others were measured independently with various procedures and equipment. The results are shown in Table 4 and Table 5. Table 3: Mechanical properties of concrete samples by acoustic measurement

Property Values

Compressional Velocity, 4215 m/s v_P

Shear Velocity, V_s 2455 m/s

Constraint Modulus, M 36.3 GPa

Shear Modulus, G 12.5 GPa

Bulk Modulus, K 20.1 GPa

Young’s Modulus, E 30.0 GPa

Poisson’s Ratio, u 0.243 Table 4: Other properties of concrete samples

Property Values

Density, p 2.041 g/cc

Porosity, (p 9.56%

Permeability, k 0.009 mD

Specific heat, c_p 891 J/(kg-K)

Tensile strength, c_L 2.878 MPa

Uniaxial Compressional 37.343 MPa

Strength, G_C

3.2 Field samples

The field samples were obtained from the Newcrest Lihir mine. Two types of field samples were obtained, i.e. smaller and larger field samples.

The as-received small field samples were cut into shapes suitable for measurement of the permeability, tensile strength, uniaxial compressional strength, and heat capacity of the samples. The tensile strength and uniaxial compressional strength were measured in accordance with ASTM standards. The heat capacities were measured by isolating circumferential boundaries and exposing one side evenly to a heat source while measuring the temperature change on the other side.

The averaged measurements for the field samples are listed in Table 6.

Table 5: Properties of field samples

Property White Dark mineral mineral

Density, p 2.898 g/cc 2.566 g/cc

Porosity, (p 2.39% 3.94%

Permeability, k 8.50E-4 mD 2.27E-3 mD

Specific heat, c_p 590 J/(kg-K) 911 J/(kg-K)

Thermal conductivity, k_t 0.778 1.060

W/(m-K) W/(m-K)

Thermal diffusivity, a 4.64E-7 m²/s 4.68E-7 m²/s

Tensile strength, st 5.240 MPa 7.196 MPa

Uniaxial Compressional 46.145 MPa 38.11 MPa

Strength, G_C The large field samples were big chucks of rock consisting mostly of dark mineral, which is the target formation for the actual field operations.

The large field samples were cut to be approximately the same size as the concrete samples. Due to the limitation of the saw dimension, the cutting quality is not as expected. We then relocated rest field samples to a commercial rock shop for cutting.

After the large field samples were cut, 2 mm inner diameter holes were drilled at the specific locations shown in Figure 2 and thermocouples were embedded to a depth of around 10 cm in the drilled holes in all of the samples, except for the first field sample (Fl).

3.3 Wellhead device installation

Before any actual tests, a borehole with 1.3 cm diameter and 12.7 cm depth was drilled in a top surface of each sample.

A wellhead device was assembled in the laboratory with tubing and fittings manufactured by Swagelok. The assembly comprised a 1.27 cm OD tubing as an outside casing and a 0.635 cm OD tubing as an inside tubing. This device was installed on each sample as shown in the images in Figure 6. In each case, the end of the tubing was around 10 cm deep, which was approximately the center of the cubic block, leaving an open-hole section of 5 cm.

The device resembled a wellbore setup in an oil well. The device was designed so that injectant could flow into the borehole through the tubing and flow out via the annulus between the tubing and casing to the atmosphere if the valve was open. Two outlets were connected to the annulus on the union cross. One of the outlets was connected to a pressure transducer, which measured the borehole pressure, while the other was connected to the atmosphere via a needle valve.

4.0 Experiment procedures

In order to characterize a LN2 treatment effect on rock samples, several types of tests were designed and conducted, including a pressure decay test, a fracturing treatment, a LN2 circulation test, and a LN2 injection test.

4.1 Pressure decay test

The objective of the pressure decay test was to evaluate the overall permeability of the rock samples.

It is similar to a reversed buildup test in oil field operations. In a pressure decay test, the borehole pressure is increased to a certain value and then allowed to decay naturally. The real-time pressure data is plotted against time. By performing these tests before and after different treatments, e.g. fracturing and LN2 treatment, the change of overall permeability can be seen from the plotted curves. For LN2 treatment, any permeability enhancement effect could be demonstrated. For fractured rock samples, the induced fracture conductivity could be estimated.

In addition, a so-called bubble test comprising spraying soapy water onto the sample surfaces during the pressure decay tests was also carried out. This test was designed to show any leaking points or traces of induced fractures on sample surfaces.

4.2 Fracturing treatment

The fracturing treatment was designed to induce fractures in rock samples.

The tri-axial loading frame was used to confine rock samples at planned stress conditions, which controlled the direction and morphology of induced fractures. For most of the cases, nitrogen gas was used as a fracturing fluid. Water was used to fracture the rest of the samples.

During a fracturing treatment, the fracturing liquid was injected into the borehole of a confined rock sample. The injection continued until a peak pressure was reached. Then, based on experience, more injection was allowed at a large flow rate to further enhance fractures.

The borehole temperature, pressure, and confining stress in all three directions were monitored in real-time at 2 Hz sampling rate during the fracturing treatments.

4.3 LN2 circulation

LN2 circulation is one type of LN2 treatment that was carried out on rock samples.

In each test, LN2 was forced into the drilled hole through the inner tubing of the wellhead device, contacted and cooled the open hole section of the borehole, and then exited through the annulus of the wellhead device into the atmosphere. The pressure of LN2 in circulation was low to avoid any potential damage to the rock, such as unexpected fracturing.

4.4 LN2 injection

LN2 injection is the other type of the LN2 treatment that was carried out on rock samples.

In LN2 injection tests, LN2 was injected into the boreholes through the inner tubing of the wellhead device, then entered the rock material either through connected pores or induced fractures, and cooled the rock mass, as shown in the schematic diagram in Figure 7. Depending on the overall permeability of the rock sample, the pressure should be elevated to drive LN2 to flow into the borehole at a rate sufficient to decrease the borehole temperature. Due to the heat loss in the flow line, it often takes much longer time to cool the borehole to the LN2 temperature. To increase the cooling process, a pre-cooling circulation may be conducted to fill the borehole with LN2 before the injection.

4.5 LN2 injection tests on concrete samples

Due to the analogical properties of natural rocks and easy accessibility in a laboratory environment, the concrete samples provided more data points for analysis and numerical simulation of the cryogenic cooling process around boreholes than actual rock samples.

The concrete samples were used to test the equipment setup used for LN2 circulation and LN2 injection at room temperature, 100 °C and 200 °C. To account for the existence of pre-existing fractures, multiple concrete samples were fractured with gas nitrogen (gN2) in vertical (perpendicular to y-axis) and lateral (perpendicular to x-axis) directions.

4.6 LN2 treatment at room temperature

The experiments conducted at room temperature were designed for primary tests of the testing procedures. Sample Cl to C4 are plain samples without any TCs embedded in the samples.

Sample Cl

Sample Cl was a plain concrete block without embedded TCs. It was treated twice with LN2 circulation. The first attempt failed due to a burst thermocouple. In the second attempt, two TCs were used. TCI was attached to the casing at an edge of the top surface, the TC2 was attached to the center of face 4 of the sample block (see Figure 5). The temperature profile is shown in Figure 8. The circulation lasted for about 900 s. The initial cooling process was relatively slow due to the low pressure of the LN2 tank. It took approximate 500 s to reach the minimum temperature in the drilled hole judging from TCl’s reading. TC2 on the center of face 4 started to show decreasing temperature after 800 s, which was possibly also affected by the room temperature drop. TCI showed fairly fast warm-up slope after the circulation stopped. As the borehole temperature rose, its rate of increase also decreased. The pressure decay tests before and after the 2nd LN2 circulation show a very small enhancement on the overall permeability of Cl (Figure 9). Sample C2

Sample C2 was heated in the oven at 100 °C to study the water loss rate with time to avoid potential problems during heated LN2 treatment. The weight of C2 was measured at different times and the water loss vs time is plotted in Figure 10. After heating for about one week, the weight of sample C2 basically reached a constant weight that did not change with further heating.

Sample C3

Sample C3 was treated with LN2 circulation. Compared to Cl, the temperature warmup process was more complete. The LN2 circulation followed the procedures conducted on Cl, with one additional TC attached to the top surface at the location 2.5 cm away from the casing in the diagonal direction (Figure 11). It also took about 500 s to reach the minimum temperature in the borehole. The circulation then continued for about 30 mins. TC2 showed noticeable temperature drop about 300 s after the borehole reached minimum temperature and stabilized around 5 °C after LN2 circulation stopped. TC3 at the center of face 4 of the sample block (see Figure 5) did not show much difference due to the insulation on outer surface this time. The pressure decay tests showed a fairly large enhancement from the LN2 circulation treatment as in Figure 12. The bubble test revealed a major leak through the epoxy and some minor leaks via the concrete around the drilled hole.

Sample C4

Two rounds of high-pressure LN2 injection treatments were conducted on sample C4. To avoid potential breakdown under high pressure, 0.69 MPa (100 psi) of confining stress was applied in each direction of C4. Two TCs were used, with one on the casing and the other on the center of face 4 of the sample block (see Figure 5).

For the first high-pressure LN2 injection, LN2 were first circulated through the drilled hole starting around 1200 s in Figure 13 until liquid nitrogen started coming out of the wellhead outlet. The borehole temperature quickly dropped to about -65 °C when the outlet was shut. In the meantime, the valve on the LN2 transfer vessel was opened to allow LN2 to flow into the drilled hole. The pressurization process was controlled to avoid a pressure wave. However, due to the low injection rate, the borehole temperature was still above LN2 temperature, and the borehole temperature started to gradually increase. The fluctuating pressure between 2200 s to 2600 s was caused by the pressure relief valve on the transfer vessel, which opened at pressure around 3.5 MPa. When the borehole pressure approached 0 °C, the injection stopped.

In the 2^nd high-pressure injection treatment, the transfer vessel and borehole were filled at the same time so that the borehole could be cooled down to the LN2 temperature and filled with LN2 when injection started. The injection started at the same time when the outlet was closed around 650 s in Figure 14. The borehole temperature again started rising. The pressure in the transfer vessel increased to 400 psi by high-pressure gN2. Further pressure increment was because of the vaporization of LN2. Due to the non-ideal cooling effect, the outlet on the wellhead was opened slightly to allow more flow of LN2, which caused the temperature drop from 1160 s to 1550 s.

These two rounds of LN2 injection showed that the LN2 at maximum 3.5 Mpa cannot generate enough flow rate or fractures for LN2 to flow into the concrete sample due to its low permeability. The permeability enhancement was also not ideal, as shown in Figure 14.

Sample C5

Sample C5 had 4 embedded TCs and 2 additional TCs (one on the casing and one on the center of face 4 of the sample block (see Figure 5)). The LN2 circulation started around 20 s, as shown in Figure 15. The borehole temperature reached LN2 temperature around 335s. The LN2 ran out at about 1700 s and the outlet was closed 100 s thereafter. During the LN2 circulation, the borehole temperature dropped quickly to the LN2 temperature with a large amount of LN2 flow through the borehole, and was maintained for about 1400 s. The readings from the TCs basically reflects their distance from the borehole. Once the outlet was closed, the borehole temperature quickly went up. The turning points for the other TCs were shortly after the time of closing outlet, because the inner portion of the sample were colder. The warm-up process slowed down as the temperature increased. The final temperature was slightly lower than initial room temperature due to large amount of LN2 evaporated into the atmosphere. The pressure decay tests showed opposite results in Figure 16 due to a post circulation re-epoxy of the near borehole region. But the permeability enhancement effect was minor.

Sample C6

C6 had 4 embedded TCs, 2 in the borehole (one attached to the casing and one suspended in the borehole), and 2 additional TCs (one on the casing and one on the center of face 4 of the block (see Figure 5)). The LN2 circulation was conducted at room temperature (20 °C). The circulation started at the 30 s in Figure 17 and Figure 18. The borehole temperature quickly dropped to the LN2 temperature at around 150 s. Before the borehole was cooled to around -190 °C, the TC at the bottom hole fluctuated several times to the LN2 temperature, which was likely cause by LN2 stream shooting from the tubing to the bottom hole and cooled that TC. When the borehole was cooled enough, LN2 accumulated in the borehole around 150 s. As the liquid level increased to the height of the free hanging TC, the reading from it dropped to LN2 temperature instantaneously. This phenomenon also applies to the reading from the TC attached to wellhead. The circulation stopped around 2500 s.

Pressure decay tests were conducted before and after LN2 treatment (Figure 19). The overall permeability was enhanced, although not much. The major leaking revealed by bubble test was an open pore in the upper left top face of the wellbore. There was also some minor leaking through epoxy in the wellbore.

Sample C7

C7 was the first sample fractured to create conductivity for LN2 to flow into the concrete samples. Multiple attempts of fracturing were conducted on C7 to achieve a fracture conductivity large enough for LN2 to flow into the borehole before vaporization. The first two used water while the last used gN2. The pressure and stress profiles, along with the pressure decays, are show in Figures 20 (a) to (d). The first fracturing with water were conducted at injection rate of 1 mL/min and reached the peak pressure of 1724 psi. The second were at 40 mL/min and reached 2236 psi. The last gN2 fracturing reached 1572 and 2428 psi with increasing gas flow rate.

There were 5 embedded TCs, 1 TC in the borehole, and 2 additional TCs (one on the casing and one on the center of face 4 of the sample block (see Figure 5)). When injection started, the injection pressure was solely by vaporization of LN2 in the transfer vessel. The flow rate was very low, which cause the borehole temperature started to go up. Then, the transfer vessel was pressured with gN2 to 500 psi, and then maintained around 490 psi. The pressurized injection was capable of stopping the increasing trend, but not able to increase the flow rate enough to continue cool the borehole to LN2 temperature due to the low fracture conductivity. At 2120 s, the sample C7 broke down into two pieces due to the pressure. TC2 burst during the breakdown, and other TCs all showed minor temperature drop, which may be caused by the Joule-Thomson effect (gas volume expansion). The fracture was slightly deviated from the preferred fracturing plane leaning more towards the top of the sample. Sample C8

C8 was fractured with gN2 under the confining stress of 1500-1000-2000 psi with a peak pressure of 1156 psi. Then the flow rate of the gN2 was increased to improve fracture width. The pressure decay tests before and after the fracturing showed a good fracture conductivity in Figure 21. A fracture trace was visible from the left side of the borehole on face 5 to face 3 of the sample block (see Figure 5), which was along x-axis.

C8 had 5 embedded TCs, 1 TC in borehole and 2 additional TCs (one on the casing and one on the center of face 4 of the sample block (see Figure 5)). The LN2 treatment started with a 200 s LN2 circulation starting from 50 s in Figure 22. At 260 s, the outlet on the wellhead was closed, which converted the circulation to injection. The borehole temperature immediately went up due to the low flow rate of LN2. To increase the flow rate for better treatment, the injection pressure was increased gradually. When the injection pressure reached slightly over 200 psi, the sample C8 broke down in to two pieces. The temperature reading from the borehole, the wellhead and the face 4 dropped after breakdown. TC 3, 4, and 5 burst during the breakdown. The fracture plane was significantly tilted away from the preferred fracture plane, which exhibits more characteristics of shear failures and may be caused by the low viscosity of gN2.

4.7 LN2 treatment at 100 °C

In order to replicate the high-temperature environment of the mining site, several concrete samples were tested under 100 °C to establish the testing procedures and partially provide some data points at relatively high temperature. Thermal pads were initially used for maintaining the samples at 100 °C, but later was found unstable. The heating oven then used with some modifications for steady testing temperature.

Sample C9

C9 was the first sample treated with LN2 at elevated temperature. It had 5 embedded TCs, 1 TC in borehole and 2 additional TCs (one on the casing and one on the center of face 4 of the sample block (see Figure 5)). C9 was preheated in the heating oven overnight at 100 °C and tested only insulated but not heated during LN2 treatment.

LN2 was circulated at low pressure (about 18 psi) through the borehole of C9 for around 20 mins, as shown in Figure 23. The borehole temperature was quickly cooled to the LN2 temperature due to the large flow rate during LN2 circulation. The embedded TC 2 and 3 also showed obvious temperature drop since they were the closet TCs to the borehole. After the circulation stopped at 1400 s, the borehole temperature quickly warmed back to the internal temperature of concrete sample C9, which was around 50 °C at 3000 s. The temperature of the concrete block dropped faster than expectation even with thermal insulation, which yielded that continuous heating is essential during LN2 treatment.

Sample CIO

CIO was gN2 fractured under a confining stress of 1000-2000-2000 psi, which resulted in a planar fracture along y-axis with slight tilt. The maximum pressure during fracturing reached 1323 psi. The post-fracturing pressure decay test could not build the borehole pressure higher than 20 psi, which indicated a very good fracture conductivity.

CIO had 5 embedded TCs, 1 TC in borehole and 2 additional TCs (one on the casing and one on the center of face 4 of the sample block (see Figure 5)). With the good fracture conductivity, the LN2 injection were conducted under the LN2 tank’s pressure, which was around 13 psi. The heating pads were used to maintain the elevated temperature during injection and warm-up period. The LN2 injection started at 215 s and stopped at 655 s. The borehole temperature quickly dropped to the LN2 temperature after the injection started due to the large flow rate through induced fracture (Figure 24). The embedded TCs all showed significant temperature drop during the injection. The temperature change followed the distance of these TCs away from the fracture plane. But for TC 5, the temperature increased to over 100 °C, which was due to the poor temperature monitoring of the thermal pad controllers. Although the thermal pads were maintaining the temperature of the concrete sample surface, but it was not stable and affected by the locations of the TCs used by the controllers for temperature monitoring.

Sample Cl l

Cl 1 was fractured by gN2 under the confining stress of 2000-1000-2000 psi.

A bubble test showed that the induced fracture was planar and along the x-axis. The induced fracture covered the majority of the sectional area. However, the post-fracturing pressure decay test showed relatively slow decay curve.

Cl l had 5 embedded TCs, 1 TC in borehole and 2 additional TCs (one on the casing and one on the center of face 4 of the sample block (see Figure 5)). For the purpose of establishing a standard testing procedure, Cl l had three different LN2 treatments at 100 °C with heating pads. Due to the low fracture conductivity, the 1^st LN2 treatment on Cl 1 was complicated. The treatment comprised 2 rounds of LN2 injection with pre-circulation to fill the borehole with LN2, as shown in Figure 25. The first pre-cooling circulation started at 470 s. The borehole temperature quickly dropped to the LN2. At 760 s, the wellhead outlet was closed, converting the treatment to injection. The injection pressure was the LN2 tank pressure, which was around 18 psi. The flow rate at this injection pressure was too small to keep the borehole temperature at -190 °C. The borehole temperature quickly increased and then gradually slowed down after switching to injection. The outlet was re-opened at 1200 s, switching back to circulation, due to the poor cooling performance of the injection at low pressure. The borehole temperature rapidly dropped until the borehole was filled with LN2. After 200 s of circulation, the outlet was closed again. The 2nd injection lasted for 1000 s, the borehole temperature increased as in the 1st injection but with slower rate. After the 2nd injection stopped at 2400 s, the borehole temperature jumped again. During this LN2 treatment, the thermal pads performed within approximately 5 °C of 105 °C.

The 2^nd LN2 treatment was pressurized LN2 injection using the transfer vessel without any pre-cooling circulation. There was a pre-cooling circulation started at 2500 s in Figure 26. After the borehole temperature was stabled at LN2 temperature, the treatment switched to injection. The injection pressure was from the evaporation of LN2 in the transfer vessel. The borehole temperature first jumped to -120 °C, then gradually decreased to LN2 temperature, and remained stable for 250 s. The maximum pressure during injection was 43 psi. After the LN2 in the transfer vessel was exhausted, the borehole temperature quickly increased to around 50 °C and then gradually approached to 100 °C. The readings from embedded TCs showed various temperature drop during injection. TC3 had the lowest reading of 49 °C since it is the closet TC to the induced fracture plane. The TC on face 4 of the sample block (see Figure 5) was attached to a hot spot on the heating pad on that face, thus its reading was much higher than the average temperature of that face. The heating pads in this treatment actually performed reasonably.

The 3^rd LN2 treatment was a pressurized injection from the transfer vessel. The injection pressure again relied on the evaporation of LN2 in the transfer vessel. There was no pre-cooling circulation. The injection started at 1000 s in Figure 27. Since Cl l had already gone through 2 LN2 treatment, the injectivity was greatly improved. The borehole temperature quickly dropped to the LN2 temperature. At 1486 s, Cl l broke into two pieces at maximum pressure of 64 psi. The remaining LN2 ran through the open fracture and evaporate in the following a few seconds and cooled the borehole to -195 °C. After breakdown, TC 4 gave a fluctuating reading of temperature. The warm-up period was short due to short LN2 cooling time.

The pressure decay curves before the 1st treatment and after the 2nd treatment are shown in Figure 28. After 2 rounds of LN2 treatment, there was a noticeable enhancement on the fracture conductivity of Cl 1.

Sample C 12

C12 was fractured by gN2 under the confining stress of 2000-1000-2000 psi.

A bubble test showed a fracture on face 3 of the sample block (see Figure 5) from the up left extended to the middle of the bottom edge. There was also a planar fracture with a slightly inclined upper portion along the y-axis. The fracture covers approximately 50% of the sectional area.

C12 had 5 embeded TCs, 1 TC in borehole and 2 additional TCs (one on the casing and one on the center of face 1 of the sample block (see Figure 5)). Two rounds of LN2 treatments were conducted on C12 at 100 °C with heating pads, including one LN2 injection with pre-cooling circulation and one direct injection.

In the 1^st LN2 treatment, the pre-cooling LN2 circulation started at 200 s in Figure 29. After liquid nitrogen started coming out of the wellhead outlet and the borehole temperature was stable at LN2 temperature, the circulation was stopped at 700 s with the assumption that the borehole was filled with LN2. The injection pressurized by LN2 evaporation in the transfer vessel started right after. The borehole temperature first jumped to -105 °C and then gradually declined to -175 °C as the fracture conductivity was enhanced by low temperature of LN2. The maximum pressure during injection was 75 psi. At 1575 s, the LN2 ran out thus the injection stopped. The readings from embedded TCs followed their distance from the fracture plane. The TC3 had the largest drop and reached as low as 50 °C.

The 2^nd LN2 treatment was via direct injection without any pre-cooling circulation. The injection was also pressurized by the LN2 evaporation in transfer vessel. The injection started at 530 s in Figure 30. The borehole temperature dropped slowly during the initial stage of injection. From the time of 530 s to 1050 s, most of the injected LN2 was cooling the pipeline and borehole. After the peak of 95 psi, the injection pressure started to decrease at 943 s. The borehole temperature decreased faster shortly after that, indicating large flow rate hence larger fracture conductivity. The borehole reached -175 °C around 1150 s and remained stable. At approximately the same time, a leaking sound was noticed from the sample. At around 1300 s, steam was observed coming out of the sample C12 surface, indicating that LN2 was escaping the concrete sample through fractures formed in liquid phase injection. The LN2 injection stopped at 1350 s.

Pressure decay tests were performed before, between, and after the two LN2 treatments at heated conditions, as shown in Figure 31. The enhancement effect is significant during each test.

Sample C14

C14 was the first concrete sample tested in the heating oven for a more constant temperature boundary. It had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). Only one LN2 circulation were conducted at 100 °C to provide a baseline case for simulation. In addition, the amount of LN2 consumption were also measured during circulation.

The LN2 circulation started at 50 s in Figure 32. The circulation lasted for 1800 s (30 mins) and stopped at 1850 s. The embedded TC 2 and 3 showed large temperature drop. Temperature reading from TC 2 was even lower that it from TC 3, despite its location is actually further away from wellbore. The total LN2 consumed in the LN2 circulation was 7.6 kg, with a relatively stable rate across the whole treatment.

Sample C 13

C13 was fractured with gN2 under the confining stress of 2000-1000-2000 psi. During the post-fracturing pressure decay test, a bubble test showed a fracture along the x- axis from the borehole on the top surface, with additional leaking points along the x-axis on face 1 of the sample block (see Figure 5).

C13 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). The TC2 was failed before any LN2 treatment. There were 3 rounds of LN2 injection treatment conducted on C13 due to the low injectivity. The temperature profile of the 1^st LN2 injection treatment is shown in Figure 33. The injection pressure was only around 20 psi from the LN2 tank. The treatment started with a pre-cooling circulation at 20 s and stopped at 500 s, when the borehole was filled with LN2 after circulation. During the injection period, which was from 500 s to 1500 s, there were very minimum flow of LN2. The temperature jumped right after injection started and then gradually went up. After the injection stopped, the temperature reading from TCs on borehole wall and free hanging separated due to faster warm-up of the rock materials. The total LN2 consumption for this treatment is 0.95 kg. The 2^nd LN2 treatment was an injection with pre-cooling circulation at elevated pressure around 70 psi. The circulation started at 10 s in Figure 34. After the borehole was filled with LN2, the injection started at 650 s by closing the outlet. The borehole temperature immediately went up, since the LN2 flow rate was too low. At around 300 s, the LN2 flow rate increased a little. But the bore hole temperature was still increasing gradually. The injection stopped at 650 s with the total LN2 consumption of 0.95 kg.

The 3^rd LN2 treatment was an injection with pre-cooling circulation at 100 psi. The circulation started at 10 s in Figure 35 and stopped at 200 s to switch to injection. The borehole temperature jumped at the beginning of injection and then quickly decreased to LN2 temperature. The increment in injection pressure enhanced the flow rate of LN2. But this cannot rule out the effect of longer time of heating and multiple round of LN2 treatment before this one. C13 broke down into two pieces at 440 s and damaged the TC attached to the borehole wall. The free hanging TC was blown out of borehole into the middle of the induced fracture, with direction contact with ambient environment. Hence, its readings were not stable and no longer representative for the borehole temperature. The injection lasted to 520 s with total LN2 consumption of 2.8 kg.

4.8 LN2 treatment at 200 °C

Samples C15 to C29s were tested with LN2 under 200 °C in the oven. Three of them (C15, C16, C17) were intact samples that only treated with LN2 circulation to establish a baseline with pure heat conduction. Six samples (C16, C18-22) were fractured for vertical fractures and injected with LN2 under pressure. Four samples (C23-26) were fractured for horizontal fractures and injected with LN2. In addition, three concrete samples were made with weak and high permeability middle layer (so-called “sandwich” samples) to investigate the heterogeneity effect on LN2 treatment.

8.8.1 Normal samples

Sample C15

C15 was the first concrete sample tested in the heating oven at 200 °C. It had 5 embedded TCs and two additional TCs in the borehole, including one attached to the borehole bottom and one suspended in the borehole. Only one round of LN2 circulation was conducted to provide a baseline case for the simulation. Thus, the sample remained intact and unfractured. During the circulation, the LN2 consumption (by weight reduction of the LN2 tank) was measured.

The LN2 circulation started at 50 s in Figure 36. Since TC2 failed before the circulation, the readings are not in the results. The circulation lasted for 3960 s (66 mins) and stopped at 4010 s. The borehole temperature dropped to LN2 temperature right after the start of the circulation (around 100 s). The embedded TCI, 3, and 4 showed large temperature drops. TC3 showed the most significant temperature drop among them, which is reasonable considering its smallest distance to the borehole. TC5 failed at 727 s and we removed its reading after this point as it is no longer representative of the local temperature. The total LN2 consumed during the LN2 circulation was 15.85 kg, with a relatively stable consuming rate across the whole treatment. After the circulation, we measured the warm-up time of the borehole temperature, as shown in Table 7.

Table 6: Warm-up time of borehole temperature in C15

Temperature Warm-up time

(°C) (min)

130 17.5

140 26.8

150 58.7

160 91.3

Sample C16

C16 was fractured with gN2 at confining stress of 2000-1000-2000 psi for a vertical fracture perpendicular to y-axis. A bubble test showed a fracture on face 1 of the sample block (see Figure 5) from the middle of the upper edge to the lower right. After dissection, it was found that the fracture pattern of the sample fits the description above. The fracture plane covered approximately 50% of the sectional area.

C16 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). The LN2 injection treatment with precirculation was conducted in the oven at 200 °C. The TC hanging in the borehole failed before LN2 injection. Initially, the LN2 injection treatment was performed at an injection pressure of 75 psi. The test failed because the flow rate of LN2 was too high and LN2 quickly leaked through the sample into the oven. The oven door popped open due to the evaporation of the escaped LN2. Therefore, another round of LN2 injection was conducted at a lower injection pressure of 20 psi after the sample C16 was re-equalized at 200 °C. Figure 37 shows the temperature profile during the treatment. The circulation prior to the LN2 injection started at 50 s and stopped at 120 s. There was a small increment in the borehole temperature when switching to LN2 injection. Then it quickly decreased back to the LN2 temperature. For the first 30 s of the injection (from 120 s to 150 s), the internal temperatures were relatively steady. After 150 s, the reading from TC3 and TC4 quickly dropped. TC4’s reading even dropped to the LN2 temperature at the end of the injection, indicating that the fracture might extend to its location. We stopped the LN2 injection at 430 s with a total LN2 consumption of 7.6 kg. The warm-up time of the borehole temperature is summarized in Table 8.

Table 7: Warm-up time of borehole temperature in C16 Temperature Warm-up time (°C) (min) 130 17.5

140 26.8

150 58.7

160 91.3

Sample C17

C17 was the other unfractured sample besides C15 used to provide a baseline case for the simulation.

It had 5 embedded TCs and two additional TCs in the borehole, including one attached to the borehole bottom and one suspended in the borehole. A LN2 circulation was conducted on this sample in the oven at 200 °C.

Figure 38 shows the temperature profile of C17 during the LN2 circulation. The circulation started at 50 s and the borehole temperature quickly dropped to LN2 temperatures. The readings of TC2 and TC3 showed significant temperature reductions at the corresponding locations. The temperature reduction at each location basically reflected its distance to the borehole. The LN2 circulation stopped at 1850 s, ending up with a total LN2 consumption of 7.65 kg. The warm-up time of the borehole temperature is summarized in Table 9.

Table 8: Warm-up time of borehole temperature in C17 Temperature Warm-up time

(°C) (min) 130 19.5 140 23.5

150 28.8

160 36.4

Sample C18

Concrete sample C18 was fractured with gN2 at confining stress of 2000-1000-2000 psi for a vertical fracture that perpendicular to y-axis. A bubble test showed that multiple fractures on the sample surfaces. On face 1 (front) of the sample block (see Figure 5), there was a vertical fracture along the x-axis locating one inch from the borehole to the right, and also a horizontal fracture from the left edge to the center. In addition, there were fractures noted on faces 2 and 4 of the sample block (see Figure 5). After dissection, it was found that the sample had a complex fracture pattern that was dominated by the vertical fracture along the x-axis.

C18 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). A LN2 injection treatment was conducted in the oven at 200 °C. Due to the low fracture conductivity, a pre-circulation was carried out before the injection to fill the borehole with LN2, as shown between 50 s to 140 s in Figure 39. At the start of the LN2 injection (140 s), the LN2 flow rate was low, which resulted in a temperature increment in the borehole. At around 350 s, the flow rate dramatically increased, and the injection pressure dropped from 80 psi to nearly 0 psi. This demonstrates that fractures were reopened under the low temperature. The borehole temperature thereby reduced back to LN2 temperature. All embedded TCs showed significant temperature reductions at the corresponding locations. The reading of TC5 even dropped to LN2 temperature. Considering the complicated fracture pattern, the fractures might extend to that location. The injection stopped around 1217 s with a total LN2 consumption of 11.05 kg. Such a high LN2 flow rate reveals that the fracture conductivity was improved significantly after the LN2 treatment, which explains the long warm-up time shown in Table 10.

Table 9: Warm-up time of borehole temperature in C18

Temperature Warm-up time (°C) (min) 130 98.3

140 113.0

150 129.9 160 149.8

Sample C19

C19 was fractured with gN2 at confining stress of 2000-1000-2000 psi for a vertical fracture that perpendicular to y-axis. A bubble test showed a fracture on the top surface along the x-axis, across the wellhead. After dissection, it was found that a tilted fracture from the top surface to the lower right corner. The post-fracturing pressure decay shows that the sample was of good fracture conductivity.

C19 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). A LN2 injection was conducted in an oven at 200 °C, with no pre-circulation considering the good fracture conductivity. However, the LN2 flow rate was initially very low, as reflected by the LN2 consumption in Figure 40 (LN2 injection started at 50 s). Until 2500 s, the borehole temperature finally reached LN2 temperature. Then the LN2 flow rate increased dramatically, and the reading of embedded TCs started to decrease. Their temperature readings basically followed their distance to the borehole, except for TC 2 who gave an unstable and fluctuating reading in a short period from 3250 to 3360 s. At 3400 s, the LN2 injection stopped with a total LN2 consumption of 8.35 kg. Note that the oven had an auto shutdown mechanism, which can automatically trigger a shutdown the oven when there is a dramatic temperature drop. During the LN2 treatment, the oven was shut down because a lot of LN2 got into the oven through fractures. We then turned on the oven again to allow the sample to warm-up. The warm-up time of the borehole temperature is summarized in Table 11.

Table 10: Warm-up time of borehole temperature in C19

Temperature Warm-up time

(°C) (min)

130 62.5

140 74.9

150 89.4

160 106.9

Sample C20

C20 was fractured with gN2 at confining stress of 2000-1000-2000 psi for a vertical fracture that perpendicular to y-axis. A bubble test showed a fracture on the top surface that deviated around 15 degrees from the x-axis, across the wellhead. The fracture did not reach either the top or the bottom edge of the top surface. After dissection, it was found that there was a planar fracture along the x-axis. The post-fracture pressure decay shows that the sample’s fracture conductivity is very good.

C20 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). The TC hanging in the borehole failed before any treatment. A LN2 injection was carried out on C20 in the oven at 200 °C, with no pre-circulation owing to the good fracture conductivity. The LN2 injection started at 50 s, as shown in Figure 41. At 380 s, the borehole temperature reduced to the LN2 temperature. Then LN2 injection rate dramatically increased, indicating the reopen of the induced fractures. The 5 embedded TCs all showed temperature reductions. The temperature in the oven also dropped significantly, which eventually led to the shutdown of the oven at 600 s. The LN2 injection stopped at 650 s with a total LN2 consumption of 8.95 kg. We then turned the oven on to heat the sample. The warm-up time of the borehole temperature is summarized in Table 12. The warm-up time is short because the LN2 quickly escaped from the sample through fractures and did not cool the inside of the sample efficiently.

Table 11: Warm-up time of borehole temperature in C20

Temperature Warm-up time

(°C) (min)

130 10.5

140 11.1

150 11.6

160 14.9

Sample C21

C21 was fractured with water instead of gN2, intending to have a more consistent fracture conductivity. The confining stress used during the water fracturing was 1500-1000- 2000 psi. Water was injected into the sample at a constant flow rate of 40 ml/min. The breakdown pressure was around 1220 psi. The post-fracturing pressure decay showed a slow pressure reduction. Also, there was no visible fracture on the sample surfaces, during the bubble test. These two facts indicated that C21 had an undesirable fracture conductivity.

C20 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). The one attached to the borehole bottom failed before any LN2 treatments. Three rounds of LN2 injections were performed on C21 at 200 °C. However, they all failed due to the poor fracture conductivity of the sample, even pre-circulations were introduced. Taking the 2nd round injection (Figure 42) as an example, the borehole temperature gradually recovered after switching circulation to LN2 injection at 140 s. At 1480 s, we decided to stop the injection since there was no sign of temperature drops. Only 0.5 kg of LN2 was consumed during the injection. Considering the unsatisfactory performance of water fracturing, the following samples were all fractured with gN2, as usual.

Sample C22

C22 was fractured with gN2 at confining stress of 2000-1000-2000 psi for a vertical fracture perpendicular to y-axis. The sample was broken down during the post-fracturing decay. It revealed a planar fracture induced along the x-axis.

C22 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). One round of LN2 injection was conducted on C22 in the oven at 200 °C. No pre-circulation was included because the fracture conductivity was near infinite after the fracturing treatment. The LN2 injection started at 50 s in Figure 43, and the borehole temperature dropped to LN2 temperature only in a few seconds. After that, all the embedded TCs, especially TC3, showed significant temperature reductions at their corresponding locations. The TC placed in the oven gave unstable temperature reading due to the fierce evaporation of the LN2. The Injection ended around 450 s with a total LN2 consumption of 7.75 kg. The oven was then turned on to heat the sample. The warm-up time of the borehole temperature is summarized in Table 13. As a large amount of LN2 evaporated into the oven, the oven took a much longer time to reach the set temperature. This eventually led to a long warm-up time.

Table 12: Warm-up time of borehole temperature in C22

Temperature Warm-up time

(°C) (min)

130 48.0

140 61.5

150 77.9

160 97.8

Sample C23 C23 was fractured with gN2 under confining stress of 2000-2000-1000 psi for a horizontal fracture. It was the first sample with a fracture induced horizontally. A bubble test showed fractures trace at the middle of faces 2 and 3 of the sample block (see Figure 5), deviating about 10 degrees from the horizontal direction. There was also a fracture on the top surface that deviated around 15° from the x-axis, across the wellhead. After dissection, it was found that there was a near-planar fracture across the sample horizontally. The postfracturing pressure decay conducted in the heating oven showed good fracture conductivity.

C23 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). One round of LN2 injection was conducted on C23 in the oven at 200 °C. The injection of LN2 started around 150 s in Figure 44. The injection continued to 1600 s, which lasted for about 24 min. Sample C23 broke down shortly after the LN2 entered the borehole with an injection pressure around 76 psi. The LN2 injection stopped at 1660 s with a total LN2 consumption of 5.85 kg. The maximum pressure during the injection was 83 psi. Several TCs showed incorrect temperature readings, e.g. TC 1, 2, 3, and the one attached to the borehole bottom (BH wall). This issue attributes to the thermal degradation of the TC connectors, with a thin layer of metal oxidation formed on the surface of the metal plate of these connectors. Around 1750 s we turned on the oven to heat the sample. The warm-up time for C23 is listed in Table 14.

Table 13: Warm-up time of borehole temperature in C23

Temperature Warm-up time

(°C) (min)

130 15.4

140 20.5

150 29.2

160 44.1

Sample C24

C24 was fractured with gN2 under confining stress of 2000-2000-1000 psi for a horizontal fracture. A bubble test showed a fracture on face 2 of the sample block (see Figure 5). It deviated about 10 degree from the horizontal direction, reaching the upper 1/3 of the right edge. After dissection, it was found that there was a tilted planar fracture that slanted through the sample horizontally. The post-fracturing pressure decay conducted in the heating oven showed good fracture conductivity. C24 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). The TC hanging in the borehole failed before any LN2 treatment. A LN2 injection treatment was performed on C24 in the oven at 200 °C. As shown in Figure 45, the injection started at 250 s with an injection pressure around 80 psi. At 760 s, the LN2 filled the borehole. Meanwhile, the injection pressure built to 148 psi, reopening the induced fracture. The embedded TCs showed temperature drops at their locations since then. The injection stopped at 900 s with a total LN2 consumption of 6 kg. Then we turned on the oven to heat the sample. The warm-up time (Table 15) was relatively short due to the short-lived LN2 treatment.

Table 14: Warm-up time of borehole temperature in C24

Temperature Warm-up time

(°C) (min)

130 7.4

140 8.5

150 11.2

160 18.3

Sample C25

Sample C25 was fractured with gN2 under confining stress of 2000-2000-1000 psi for a horizontal fracture. A bubble test showed a fracture on the top surface from the middle of the left edge to the upper edge. After dissection, it was fond that there was a tilted planar fracture that deviated about 20 degrees from the horizontal direction to the top surface. The post-fracturing pressure decay showed good fracture conductivity of the sample.

C25 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). One round of LN2 injection was conducted on C25 in the oven at 200 °C. As shown in Figure 46, the injection started at 50 s. The initial injection pressure was 75 psi. At 1837 s, the borehole temperature reached LN2 temperature. The LN2 flow rate dramatically increased since then, and the injection pressure dropped to 0.5 psi, showing that the fracture was reopened with almost infinite conductivity. Meanwhile, the sample was broken down, and TC 4, 5 as well as the one attached on the borehole wall burst. We thereby removed their readings after this point. The LN2 injection stopped at 1917 s with a total LN2 consumption of 7.35 kg. The warm-up times were summarized in Table 16. Table 15: Warm-up time of borehole temperature in C25

Temperature Warm-up time (°C) (min) 130 96.1

140 107.8

150 124.4

160 144.2

Sample C26

C26 was fractured with gN2 under confining stress of 2000-2000-1000 psi, which should result in a horizontal fracture. However, A bubble test showed one additional vertical fracture on face 1 of the sample block (see Figure 5), from the middle of the top edge to the center. After dissection, the fracture pattern was found to be a vertical fracture that covered the upper half of the sample and a horizontal fracture that covered the entire cross-section. The post-fracturing pressure decay reveals a very good fracture conductivity from the fracturing treatment.

C26 had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). The TC hanging in the borehole failed before any LN2 treatment. A LN2 injection was conducted on C26 in the oven at 200 °C. As shown in Figure 47, the injection started at 200 s. The initial injection pressure was 80 psi. In about 50 s, the sample was broken down. The borehole temperature reached LN2 temperature almost right after the breakdown. The embedded TCs showed significant temperature reductions at their locations. The LN2 injection stopped at 300 s with a total LN2 consumption of 3.95 kg. The warm-up of the borehole temperature was slow since most of the LN2 evaporated into the oven (Table 17). The oven took a long time to reach the set temperature (200 °C).

Table 16: Warm-up time of borehole temperature in C26 Temperature Warm-up time (°C) (min) 130 36.6

140 45.8

150 56.2

160 71.0 During the LN2 injection, we used a thermal camera to capture the temperature distribution of sample’s surfaces. The blue circle in Figure 48 demonstrates that the LN2 manly flowed through the induced horizontal fracture. The dark area above the fracture corresponds with the leak of the LN2 from the epoxy sealing. The brighter colors (red, orange, and yellow) indicate warmer temperatures while the purples and dark blue/black indicate cooler temperatures.

4.8.2 “Sandwich” samples

A “sandwich” sample contained three layers in order to account for effect of formation heterogeneity on LN2 treatment. During molding, the top and bottom layers were made with the original cement/sand ratio. The middle layer was made with half cement to reduce the strength and increase permeability. The three-layer structure resembles a sandwich, hence the name. The open hole section of the borehole is all in the middle low- strength high-permeability layer.

Sample C27s

Concrete sample C27s was fractured with gN2 under confining stress of 2000-1000- 2000 psi, which should result in a vertical planar fracture. During the post-fracturing pressure decay, the pressure was fairly stable showing a poor fracture conductivity. In addition, there was no visible fracture on the sample surfaces, but only a minor leakage from the epoxy sealing. As a result, we decided to discard this sample.

Sample C28s

C28s was another “sandwich” sample that had the same composition as C27s. It was fractured under confining stress of 2000-1000-2000 psi for a vertical fracture perpendicular to y-axis. A bubble test showed a fracture on the top surface along the x-axis. After dissection, it was found that there was a vertical planar fracture.

C28s had 5 embedded TCs and 2 additional TCs in the borehole (one attached to the borehole bottom and one suspended in the borehole). One round of LN2 injection was performed in the oven at 200 °C. The injection started at 200 s, as shown in Figure 49. The initial injection pressure was around 52 psi. Due to the low fracture conductivity, the initial LN2 flow rate was very low, and the borehole temperature drop was insignificant. At around 3740 s, the LN2 entered the sample through the fracture, as reflected by the increase in LN2 flow rate. Meanwhile, the local temperature of each embedded TC started to reduce. The TC attached to the borehole wall gave incorrect readings from 3750 to 4300 s. We thereby remove the corresponding data from the result. The injection ended at 4180 s with a total LN2 consumption of 4.6 kg. We then turned on the oven to heat the sample at 4380 s. The warm-up time of the borehole temperature is listed in Table 18.

Table 17: Warm-up time of borehole temperature in C28s

Temperature Warm-up time

(°C) (min)

130 22.5

140 27.9

150 35.8

160 52.7

Pressure decay tests were conducted at a heated condition (200 °C) before and after the LN2 treatment. The results in Figure 50 yielded intermediate initial fracture conductivity and high conductivity after the LN2 injection.

Sample C29s

Sample C29s was the last “sandwich” sample. It was fractured under confining stress of 2000-1000-2000 psi by gN2, which should result in a horizontal planar fracture. During a bubble test, there were no visible fractures noted on the sample surfaces, but only a minor leak through TCs (BH wall and Hanging). During the post-fracturing pressure decay, the pressure reduced at an acceptable rate. It indicated that there were fractures generated internally. After dissection, it was found that there was a tilted horizontal fracture along the x-axis. It deviated around 30 degrees from the horizontal direction to the upper part of face 1 of the sample block (see Figure 5).

C29s also had 5 embedded TCs and 2 additional TCs in the wellbore (one attached to the borehole bottom and one suspended in the borehole). TC5 failed before any LN2 treatment. The LN2 injection started at 400 s, as shown in Figure 51. In the beginning, the LN2 flow rate was low due to the poor fracture conductivity. We then built the injection pressure from 50 to 80 psi and restarted the injection at 1540 s. The LN2 fill the borehole and reopened the induced fractures at 2700 s. Then, the LN2 flowing rate dramatically increased, and the injection pressure dropped to 0.5 psi. All embedded TCs showed significant temperature reductions at their corresponding locations. The LN2 injection stopped at 2762 s with a total LN2 consumption of 5.35 kg. The sample was broken down at 2690 s. Then the TC on the borehole wall gave wrong readings and returned to normal until 2920 s. The oven was turned on again at 2805 s to heat the sample. The warm-up time of the borehole temperature was relatively long as exhibited in Table 19. After the dissection of the sample, a large amount of LN2 evaporated into the oven. It thus took a long time to reach the set temperature.

Table 18: Warm-up time of borehole temperature in C29s

Temperature Warm-up time

(°C) (min)

130 98.8

140 115.9

150 138.0

160 159.7

During the LN2 injection, a thermal camera was used to capture the temperature distribution on the sample’s surface. In Figure 52, there is a major fracture pathway for LN2 to flow on Face 2 of the sample - see of the sample block (see Figure 5). In addition to that, Face 3 of the sample block (see Figure 5) also had a major conduit leaking LN2, which was not captured due to the limitation of the angles.

5.0. LN2 injection tests on actual rock field samples

The field samples tested in this section were obtained from the actual mining site at Lihir mine. The original big chunks of rocks were cut in to 8-inch cubic shape for consistency with the concrete samples so that the experiment results can be comparable. Five TC holes with 2 mm diameter and 10 cm depth were drilled from the top surface at specific locations in on each of the field samples except the sample Fl. The results of LN2 treatment on these field samples provide a more realistic physical simulation of the cooling effect around wellbore at 200 °C. However, due to the limitation on dimensions, these results required an up-scaling before applied to any field application.

Among all field samples, Fl was treated with LN2 circulation (half-hour) and intermediate pressure injection; F3 was treated with high pressure injection; F2 was treated with LN2 circulation (one-hour); F4 and F5 were treated with high pressure LN2 injection; and F6 was treated with low pressure LN2 injection.

5.1 Field sample Fl The field sample Fl was cut at the Colorado School of Mines. Due to the limitation of the cutting equipment, the cutting quality was not ideal with surfaces not exactly perpendicular to each other and obvious cutting marks. By observing from outside of the sample (Figure 61), Fl is highly heterogeneous. It composes mostly the dark mineral with multiple intrusion of the white mineral. The boundary between both minerals seems to be very strong. There are also multiple natural fractures visible from the surfaces of Fl. There are two large fracture on faces 2 and 4 of the sample block (see Figure 5). On face 2, the fracture is diagonal from the top right corner to bottom left comer. On face 4, the major fracture is horizontal at about 1/4 position from the bottom with another two small branches in diagonal direction. There are also some more shorter fractures visible on face 1 and on the top surface of the sample block (see Figure 5).

Due to Fl being the first field sample, the depth of TC holes was shallower than the planned depth and varied due to the difficulty of drilling small and long holes in such hard rocks. The depths of these TC holes i.e. shown in Table 20. The actual depth of the embedded TCs can be assumed to be these values, which differ from the planned 10 cm at the designated locations.

Table 19: Depth of TC holes in field sample Fl

Location TCI TC2 TC3 TC4 TC5

Depth 1 cm 1 cm 1 cm 3.5 cm 0.7 cm

Before heating field sample Fl, a pressure decay test with pressure up to 50 psi was conducted on Fl. The purpose of low decay pressure was to avoid new induced fracturing. The borehole pressure decayed much faster than expected (Figure 53), which indicated that the fracture conductivity of natural fractures in Fl was fairly good. A bubble test was also conducted during this pressure decay test. The gas bubbles on surfaces of Fl show the conduit of the leaking gas from the wellbore. On the top surface, the bubbles showed another major fracture near the wellbore.

After the sample Fl was heated in the oven overnight at 200 °C, another pressure decay was conducted before any LN2 treatment to evaluate the general permeability of Fl at high temperature (Figure 53). The borehole pressure decayed slightly faster than room temperature decay, which may have been due to the sealing or particles inside natural fractures expanded at high temperature and thus propped the fracture width.

Fl was treated with two rounds of LN2. The first round of LN2 treatment is simply a LN2 circulation through borehole, which lasted for 30 mins. This process resembled the treating process on concrete sample C17. The circulation started around 220 s in Figure 64. The thermocouple attached to the borehole wall (TC 17) failed at the time of LN2 entered wellbore at 340s. With the large flow rate, the borehole temperature of Fl dropped to LN2 temperature within around 300 s after the circulation started and kept stable through the whole circulation process. The borehole pressure during the treatment peaked at 15 psi at the beginning of the circulation and was stable at 10 psi for majority of the time. The embedded TCs all showed different temperature drop, where TC3 gave the lowest reading around 120 °C. Considering the burial depths of these embedded TCs were much shallower than planned, the temperature at originally planned depth should be even lower than these readings. The circulation treatment stopped at 2020 s in Figure 54 with a total LN2 consumption of 8.1 kg. The warm-up times of the borehole temperature were listed in Table 21.

Table 20: Warm-up time of borehole temperature of Fl in first LN2 circulation treatment

Temperature Warm-up time

(°C) (min)

130 13.8

140 16.6

150 19.7

160 25.8

The second LN2 treatment on Fl was a pressurized injection with pre-circulation at high temperature. The LN2 pre-circulation started at 50 s and lasted for about 100 s, as shown in Figure 55. With the tank pressure of 40 psi, the borehole temperature quickly decreased to LN2 temperature at the end of the pre-circulation. Then, the outlet of the borehole was shut, which converted the treatment mode to injection. The borehole temperature quickly climbed to around 90 °C. In the meantime, the injection pressure increased gradually to planned 100 psi. After the borehole pressure reached 80 psi, the borehole temperature stopped increasing and began to decline. At approximately 3300 s, the borehole temperature reached LN2 temperature by injection. After that, the temperature readings from embedded TCs started to dip. Reading from TC 3 dropped to LN2 temperature at the later stage of the injection. The TC 2 and 4 gave very similar temperature curve and reached lowest temperature around 115 °C. The injection finished at 3758 s, with total LN2 consumption of 5.2 kg. The borehole were fully submerged with LN2 for around 5 min. Table 22 listed the warm-up time for the borehole temperature.

Table 21: Warm-up time of borehole temperature of Fl in 2^nd LN2 injection treatment

Temperature Warm-up time

(°C) (min)

130 20.6

140 28.2

150 43.2

160 66.4

Before the injection ended, several photographs were taken with a thermal camera, - see Figure 56. On the top surface, the area around wellbore showed very low temperature on the thermal image. Surface 2 also showed a very low temperature area at the upper middle section. These low temperature areas could be the major pathway for injected LN2. Due to the limitation of angles, we did not capture thermal images on the other surfaces.

For field sample Fl, pressure decay tests were conducted before and after each heating and LN2 treatment. All the pressure decay tests were conducted at elevated temperature (200 °C) after the sample’s internal temperature were fully equalized. The comparison of these pressure decay curves indicated that the slow and even heating process does not affect the conductivity of the natural fractures much. For the thermal shock processes, including the LN2 circulation and injection, could enhance the fracture conductivity significantly. This observation may indicate that the thermal shocks may have induced enough thermal stresses and created new fractures or extended the original fractures.

5.2 Field sample F3

After Fl, all the other field samples were cut at a commercial rock shop. The cutting quality improved noticeably when compared to Fl, with some corners chipped off due to the existence of natural fractures. Observing from outside, sample F3 was also very heterogeneous with white and dark minerals mingled with each other. Unlike Fl, there was no large chunk of white mineral intrusion. Close to the edge between faces 1 and 2 on the face 1 of the sample block (see Figure 5), there is a small piece of crystallized mineral embedded in the rock matrix, potentially to be gypsum crystals. A planar natural fracture was found on faces 1 and 2 with an angle of 30° from horizontal direction. The pressure decay test at room temperature before heating in the oven showed that F3 had fairly good overall permeability, as shown in Figure 57. The bubble test during this pressure decay test revealed three major liquid pathways on faces 2, 4 and 5 of the sample block (see Figure 5). Figure 58 shows the fractures on faces 2, 4 and 5. On the top face 5, the fracture was diagonal from the corner of faces 1 and 2 to the corner of faces 3 and 4. On face 2, a fracture in the middle section generated some gas bubble. This may indicate that only some part of this natural fracture was open at this condition. On face 4, there was a fracture tilted about 30° from the vertical direction in the center. After F3 was fully heated in the oven, the pressure decay test showed almost the same borehole pressure decline curve, which agreed with the finding of slow and even heating process did not affect general permeability of the field samples much.

Only one successful LN2 injection treatment was conducted on F3. The planned injection pressure was 160 psi to ensure enough flow rate of LN2. Unfortunately, the data file corrupted during the logging process, which caused all auto-logged data (including temperature and pressure readings) were lost. Therefore, there is no detailed data for temperature curves and further analysis, exception some descriptive notes. At the beginning of the injection, the borehole temperature decreased very slowly. But, the injectivity of LN2 increased exponentially after the borehole temperature dropped below 100 °C. Shortly after the borehole was filled with LN2, sample F3 exploded, which happened about 2.2 hours after injection started. The injection was stopped subentry with a total LN2 consumption of 2.05 kg-

5.3 Field sample F2

The field sample F2 was cut into shape with fairly good quality. But the corner of faces 2 and 3 of the sample block (see Figure 5), and bottom chipped off due to an existing fracture. F2 composed mostly the dark mineral. Despite the heterogeneous nature, there was only very small percentage of white mineral intrusion as planes in F2. Two visible natural fracture traces were visible when observing from outside on faces 1 and 4 of the sample block (see Figure 5). Judging from their locations, they could be connected with each other. The pressure decay test at room temperature showed that the overall permeability of F2 was not very ideal for injection type of LN2 treatment (Figure 59). The bubble test did not reveal any major leaking point on the sample’s surface, except some very minor gas bubble around wellbore. Thus, we decided to conduct a long-time circulation on this sample. After heated in the oven at 200 °C, the pressure decay test before test in Figure 612 showed a slightly decreased fracture conductivity in F2.

The long-time LN2 circulation through the wellbore was planned to last around 1 hour (64.5 mins) in order to match the experiment condition with concrete sample C15. The circulation started at 950 s in Figure 60. Due to the LN2 tank was at 70 psi, the borehole temperature declined very fast. Within 200 s, the borehole of F2 was filled with LN2. In the meantime, the temperature readings from embedded TCs started to decline. As the flow rate of LN2 was fairly large, the treating pressure quickly declined to less than 10 psi after about 10 mins of the start and remained at this level throughout the whole circulation. The circulation lasted 64.5 minutes and stopped at 4830 s. The total LN2 consumption was 15.4 kg. All embedded TCs showed significant temperature drop. TC 3 reached sub-zero degree since it was the closet TC from the wellbore. The temperature readings from other TCs generally followed their distances away from wellbore.

However, due to a technical error, the data logging was stopped at 8733 s. By that time, the borehole temperature only reached 120 °C. Therefore, the warm-up time for F2 was estimated by fitting the temperature curves at warm-up stage with adjusted time line of sample C17. The estimated warm-up time are shown in Table 23.

Table 22: Warm-up time of borehole temperature of F2 in LN2 circulation treatment

Temperature Warm-up time

(°C) (min)

130 *71

140 *86

150 *107

160 *135

*: The warm-up times are estimated and not accurate.

Thermal images were taken of sample F2 at the end of the circulation. The low temperature areas coincided with the fracture traces noted before treatment. This indicated that the internal natural fractures connected with not only each other but also the wellbore. Since both of these two traces did not shown any gas bubble during bubble test, it could be concluded that the circulation of LN2 opened up the closed natural fractures during experiment. The post-treatment pressure decay curve showed a significant enhancement on the overall permeability of sample F2. The enhancement should be attributed mostly to the thermal shock effect by circulating of the LN2 since the treating pressure during the experiment was low and should not induce any new fracture.

5.4 Field sample F4

The field sample F4 composed with approximately the same amount of the white and dark minerals. On the top, the areas of dark minerals were more like impregnated in the white minerals. The amount of white minerals gradually decreased to the bottom. On the bottom face, the composition became mostly the dark minerals. Overall, there was only one visible fracture trace on face 1 of the sample block (see Figure 5). The later bubble test revealed another lateral fracture on face 2. The pressure decay test at room temperature showed a fairly fast borehole pressure decline rate. The major leaking points were on faces 1 and 2 of the sample block (see Figure 5). After heated in the oven at 200 °C, the pressure decay test before the LN2 treatment showed good fracture conductivity (Figure 61).

Since the planned injection pressure for the high pressure LN2 injection was 100 psi, we used three steel hose clamps to laterally fasten the sample. In addition, two steel plates vertically held the sample. In this way, we could hold the sample as one piece even if the sample break into pieces during high pressure injection.

The high pressure LN2 injection started around the 60 s in Figure 62. With the aid of high injection pressure, the borehole temperature began to decline shortly after the injection started even without a pre-circulation of LN2. Although we attempted to maintain the injection pressure around 100 psi, the actual injection pressure was not stable due to the higher pressure in the LN2 tank. In the early stage (before 2700 s), the injection pressure was around 125 psi. But in the later stage (after 2700 s), the injection pressure increase to around 150 psi with occasionally spikes up to 280 psi. This phenomenon was caused by LN2 drops entered the flowline which was not cooled enough to maintain the LN2 at liquid state. The LN2 drops would quickly evaporate causing the pressure spikes. On temperature curves, it would cause fluctuated borehole temperature curves, which can be seen between 4700 and 5660 s in Figure 62. The embedded TCs all showed significant temperature drop during the treatment. TC 3 and 5 reached sub-zero degrees to the end of the injection. Since TC 5 located furthest from the wellbore, the fracture on face 1 of the sample block (see Figure 5) should be a main liquid flow path. The almost synchronized temperature changes in TC 3, 4, and 5 also agreed with this. The injection stopped at 5556 s, with a total LN2 consumption of 4.3 kg. The warm-up times of the borehole temperature were listed in Table 24. Table 23: Warm-up time of borehole temperature of F4 in the high-pressure LN2 injection treatment

Temperature Warm-up time

(°C) (min)

130 120.9

140 128.1

150 138.7

160 154.7

The post-treatment pressure decay test (Figure 61) also showed a significant enhancement on the overall permeability of F4. The borehole pressure only took about 20 s to atmosphere pressure from 50 psi. Thermal images taken at the end of injection revealed a circular low temperature area on face 1 (dark blue region) of the sample block (see Figure 5), which proved that the fracture on face 1 was open to liquid flow during the LN2 treatment.

5.5 Field sample F5

The field sample F5 was composed mostly of the dark minerals, with some intrusions of the white minerals. By observing from outside, it seems that the original rock matrix of dark minerals was naturally fractured. And then the fractures were cemented with the white mineral. There were several fracture traces visible on the surfaces of F5. But the bubble test at room temperature showed that there was only one fracture leaking gas from the wellbore, which was on the face 3 of F5. The leaking fracture was generally in vertical direction with leveled head and tail, forming an S -shape.

The pressure decay test at 200 °C showed very good fracture conductivity of F5. It only took less than 40 s for the borehole pressure to decay from 50 psi to atmosphere pressure. One round of high-pressure LN2 injection was conducted on F5. In order to keep the sample F5 from explosion into pieces, three steel hose clamps along with the two steel plates were employed to contain sample F5 in the oven.

The high-pressure LN2 injection started at 50 s in Figure 63. The embedded TC 3 failed before any LN2 treatment. With the planned high injection pressure around 150 psi and good fracture conductivity of F5, the LN2 injection rate was fairly high. The borehole temperature declined very quickly after the injection started. At 235 s (185 s after injection started), the borehole temperature reached the LN2 temperature and remained constant throughout the injection process. Due to large amount of LN2 released into the oven through highly conductive fractures, the ambient temperature in the oven also decreased very quickly. Since the major flow path was on face 3 of the sample block (see Figure 5), which was far away from any of the embedded TCs, all the remaining embedded TCs only showed mild temperature drop during the injection process. At 345 s, the injection stopped. The total time of borehole filled with LN2 was around 2.5 mins. The total LN2 consumption of this high- pressure LN2 injection was 5.25 kg. The warm-up times of the borehole temperature were listed in Table 25. Due to the majority of the injected LN2 quickly escaped from sample F5 through the highly conductive fracture and did not have sufficient time to contact with the rock matrix, the cooling effect of this LN2 injection treatment was not ideal. Therefore, the warm-up process was relatively short when compared with F4.

Table 24: Warm-up time of borehole temperature of F5 in the high-pressure LN2 injection treatment

Temperature Warm-up time

(°C) (min)

130 31.2

140 40.0

150 55.6

160 77.9

Even though the original overall permeability of F5 was fairly high, the pressure decay test post-injection showed even faster decline rate of borehole pressure. The total time for the decay process only took around 5 seconds, which demonstrated a significant enhancement on the fracture conductivity. The thermal images revealed that the fracture on face 3 was the major flow path for injected LN2.

5.6 Field sample F6

The field sample F6 also composed of mostly the dark minerals with some white minerals as intrusions into the rock matrix. It was very heterogeneous with a lateral plane filled with white mineral. In the meantime, there were many cemented-fracture-like white mineral scattering all over the sample. By observing from outside, there was one major lateral fracture plane in the middle of faces 3 and 4 of the sample block (see Figure 5).

The pressure decay test before LN2 treatment at 200 °C showed very good fracture conductivity (Figure 64). The borehole pressure decayed from 50 psi to atmosphere pressure within 6 s. The fast decay rate proved that the lateral fracture plane connected to the wellbore and remained open at the elevated temperature. Similar to F4 and F5, sample F6 was also fixed with steel hose clamps and plates to avoid sample breakdown during LN2 treatment.

For the LN2 treatment on F6, one round of low pressure LN2 injection was conducted with planned injection pressure of 70 psi. The injection started at 50 s in Figure 65. Since the fracture conductivity was very good, the injection rate of LN2 was high. The borehole temperature reached LN2 temperature around 200 s. However, the TC hanging in the wellbore failed when LN2 entered the wellbore of F6 and somehow recovered when the injection stopped. This may be caused by the poor soldering quality when making the TC. When temperature was low enough, the two threads of the metal wire in the TC detached due to shrink at low temperature. When temperature recovered, the threads expanded and recontacted with each other again. During the injection, TC 1, 3, and 4 showed significant temperature drop. Both TC 3 and 4 dropped to sub-zero. The minimum temperature reading from TC 1 was 8 °C. However, readings from TC 2 and 5 remained above 100 °C. The discrepancy from the embedded temperature showed an extended lateral fracture within F6. The injection stopped at 323 s with a total LN2 consumption of 3.75 kg. The total time of borehole filled with LN2 was only 1.9 mins. Similar to F5, the injected LN2 did not have enough time to effectively cool the sample down before escaping from fractures. The warmup times of the borehole temperature were listed in Table 26.

Table 25: Warm-up time of borehole temperature of F5 in the high-pressure LN2 injection treatment

Temperature Warm-up time

(°C) (min)

130 31.9

140 39.1

150 51.2

160 68.0

The post-injection pressure decay test showed even faster borehole pressure decline rate (Figure 64). The overall permeability of LN2 treated sample F6 was so high that it was very difficult to build the borehole pressure to target 50 psi. Thermal images taken at the end of injection revealed two major flow path. Despite the high conductivity, the lateral fracture was only partially open at some region (marked as dark blue). And due to the short injection time, the cooled region within sample F6 was also locally constrained before extending to larger volume.

6.0 Analysis

The experiment data was analysed to correlate the potential factors that affect the borehole temperature warm-up time under the laboratory environment. The majority of the analysis was conducted with a concrete sample due to more predictable and homogeneous properties and more complete experiment data set.

6.1 Circulation time

For the cases of LN2 circulation treatment, the process was much easier than with LN2 injection due to a simpler physical process involved.

During the LN2 circulation treatment, LN2 just flowed into the wellbore through tubing, contacted and cooled the inner wellbore surface, and then escaped from the annulus. The samples treated with LN2 circulation were typically fairly dense (intact without any fractures for concrete samples, lower injection rates than for field samples). Generally, there was very limited liquid flow into the rock matrices since the treatment pressures were typically low and the formation was dense (with no fractures). Thus, the liquid flow in porous media or fractures could be neglected. The whole process could be modeled with a simple heat transfer model. And the warm-up time for borehole temperatures could be positively related with treatment time.

Table 27 lists the warm-up times for samples C15, C17, Fl, and F2, which were all treated with LN2 circulation. Note that the warm-up time for sample F2 was estimated using data from concrete samples due to logging file corrupted during warm-up period.

Table 27: Warm-up time of four LN2 circulation treated rock samples

Warm-up time (mins)

Borehole temperature C17 Fl C15 F2

(degC) (30 (30 (66 (64 mins) mins) mins) mins)

130 19.5 13.8 17.5 71*

140 23.5 16.6 26.8 86*

150 28.8 19.7 58.7 107*

160 36.4 25.8 91.3 135* Figure 71 compares the borehole temperature warm-up time. By comparing the curves from two concrete samples, their borehole temperature reached 130 °C almost at the same time. This may have been caused by a stainless steel casing installed in the borehole. During the warm-up time, the casing might have been the main heat source inside wellbore instead of the rock matrix, which was cooled during the circulation treatment. Since the LN2 treatment time on C15 is more than twice that of C17, its internal cooled area was larger and colder than that of C17. The warm-up speed of C15 at later stage was much slower than that of C17.

On the other hand, when comparing the field sample Fl and concrete sample C17, their warm-up curves in Figure 66 had very similar curves. But the warm-up speed of Fl was generally faster. Due to very low circulation pressure was applied during the circulation treatment, the circulated LN2 was not likely to enter the existing fractures in Fl. Therefore, this observation agrees with the larger thermal diffusivity of field samples when compared with concrete samples.

However, when comparing the F2 with C15 for effect of long-time LN2 circulation, the warm-up speed of F2 was much slower. Despite the data for F2 being estimated due to the data corruption, it took 65 mins for its borehole temperature to reach 120 °C, which yielded a reasonable estimation. The longer warm-up period of F2 was mostly caused by the higher circulation pressure (around 68 psi at beginning and then decreased to 10 psi). The higher circulation pressure pushed LN2 into the existing natural fractures, thus resulting in a better cooling effect and longer warm-up period, regardless that most LN2 was released to the atmosphere through the annulus of the wellbore.

The comparison of the circulation results from field and concrete samples showed that the higher thermal diffusivity of the field sample resulted in easier cooling for area around wellbore, but also shorter warm-up time. The conductive natural fractures could tremendously enhance the cooling effect and slow down the warm-up process.

6.2 Concrete sample with induced vertical fracture

For concrete samples treated with LN2 under 200 °C, C16, C18-20, C22, and C28s were successful experiments with artificially induced vertical fractures. Due to the complicated nature of the rock samples (including synthetic concrete samples), the injection induced fracturing processes was not consistent even when the experimental conditions were the same. The induced vertical fractures in these samples were all in the direction perpendicular to the y-axis (min stress direction during fracturing treatment). But their conductivity, propagation direction, and morphology varied from each other. For example, Figure 67 shows a comparison of the pressure decay curves of these samples under 200 °C before any LN2 treatment. The induced fracture conductivity varied across a relative large range.

With the variations in the induced fracture, the LN2 treatments also resulted in different cooling effect and warm-up time. Table 28 and Table 29 sum the potential affecting experiment results of the LN2 injection treatments on concrete samples with vertical fractures along with their warm-up time after injection finished. As the correlation coefficient indicates, the trend of warm-up time was more correlated with the amount of LN2 for injection, with a correlation coefficient of 0.4958. The injection time and the time for borehole filled with LN2 on the other hand were not playing an important role for the warmup time.

Table 26: LN2 injection results of concrete samples with vertical fractures

Sample #

Corr. Ceof.

C16 C18 C19 C20 C22 C28s

LN2 amount (kg) 7.6 11.1 8.4 9.0 7.8 4.6 0.4958

Injection time (mins) 6.3 19.5 55.7 10.0 6.7 69.7 -0.06861

Time for borehole filled with

LN2 (mtns) 0.0754

Max injection pressure (psi) 74.0 91.0 84.0 118.0 0.0 52.0 -0.2109

However, even with LN2 amount, which is the most influential factor for warm-up time, the correlation was not high enough to find a correlation with the warm-up time. This is due to the limitation of the size of the experimental samples. Since only the 8-inch cubic blocks were used for these LN2 cooling experiment, the injected LN2 did not have enough time to fully contact the internal surfaces of the induced fracture before escaping to the ambient environment. After a borehole was filled with LN2, it was observed that liquid nitrogen was coming out of the fractures from the sample surfaces and flowing onto the floor of the oven. This resulted in wasting LN2.

This observation also explained the negative correlation coefficient (-0.2109) between maximum injection pressure and warm-up time. With higher pressure, the flow rate would be higher. Thus, the time for LN2 to flow through the internal surfaces of induced fracture would be less, resulting in less cooled volume inside the rock samples and less warm-up time. Table 27: Warm-up time of LN2 injection treated concrete samples with vertical fractures

Borehole Warm-up time (mins) temperature

(degC) _{C16 C18 C19 C20} C22 C28s

130 17.5 98.3 62.5 10.5 48.0 22.5

140 26.8 113.0 74.9 11.1 61.5 27.9

150 58.7 129.9 89.4 11.6 77.9 35.8

160 91.3 149.8 106.9 14.9 97.8 52.7

In addition to the mentioned affecting factors above, the fracture morphology was also very important for the warm-up time of the borehole temperature.

Figure 73 compares the warm-up time of these concrete samples along with their internal fracture morphology. The warm-up time increased from left to right in Figure 68. Thus, there was an observed trend of the fracture morphology, which was that the dyed internal fracture area generally increased from left to right. For example, the induced fracture in sample C20 was mostly located around the wellbore and gradually widened to the top surface. This force the injected LN2 to quickly escape without cooling a large volume. C28s had slightly larger internal fracture surface area which was partially generated during the injection (dyed with blue and not as prominent as red color in Figure 68).

However, the right-most sample C18 seemed not to have the largest dyed internal fracture surface area. There might be two reasons result in this smaller fracture area with better cooling-effect. The first one is the fracture conductivity was high enough to conduct sufficient LN2 flow rate while low enough to enable injected LN2 to sufficiently contact with the fracture surface and exchange heat. Its induced fracture had the conductivity in an optimal range to ensure both injectivity and heat exchange efficiency for LN2. The injected LN2 could flow into the fractures in a liquid state before evaporating in the line while having enough time to absorb heat when flowing through fractures. In addition, the LN2 consumption on C18 was the largest among all concrete samples mentioned in this section, which also contribute to its long warm-up time.

6.3 Concrete sample with induced lateral fracture

The concrete sample C23-C26 and C29s were successful experiments with artificially induced lateral fractures. Similar to the samples with vertical fractures, the induced fracture morphology and conductivity were not consistent. Though most samples had induced fractures in a horizontal direction or at an angle, C26 had an additional vertical fracture passing through the wellbore on its top half along x-axis. Figure 69 shows the comparison between the pressure decay curves of these samples at 200 °C before any LN2 treatment. Note that C26 broke down during the pressure decay test in the oven, hence its pressure decay curve is almost a vertical line.

The concrete samples with induced lateral fractures showed some interesting differences when compared with samples with vertical fractures. Table 30 and Table 31 shows the potential affecting factors and warm-up time respectively. Unlike the previous section, the four factors were more correlated with warm-up time (both positively or negatively). The LN2 amount became the least relevant parameter with a correlation coefficient of 0.2297. The injection time became the most important factor with a correlation coefficient of 0.7298. This was likely caused by the direction of the induced fracture.

As shown in Figure 70, almost all induced lateral fractures were symmetric about the wellbore and covered relatively large internal area across the lateral cross-sectional plane. This enable the injected LN2 to cool more internal volume of these concrete samples during the entire injection process, no matter whether it was in gas or liquid state when flowing through the induced fractures. The large, cooled volume around the wellbore retarded the heat conduction in the later warm-up stage. Hence, it made the injection time more relevant to the warm-up time. While in the vertical fracture cases, the cooled volume can only partially hinder the warm-up process.

Table 28: LN2 injection results of concrete samples with lateral fractures

Sample # Corr.

C23 C24 C25 C26 C29s Ceof.

LN2 amount (kg) 5.9 6.0 7.4 4.0 5.4 0.2297

Injection time (mins) 26.7 10.8 31.1 4.2 39.4 0.7298

Time for borehole filled with

LN2 (mins)

Max injection pressure (psi) 83.0 144.6 110.0 53.0 76.0 -0.3146

Table 29: Warm-up time of LN2 injection treated concrete samples with lateral fractures

Borehole Warm-up time

“ c>“^re C23 C24 C25 C26 C29_S

130 15.4 7.4 96.1 36.6 98.8

140 20.5 8.5 107.8 45.8 115.9

150 29.2 11.2 124.4 56.2 138.0

160 44.1 18.3 144.2 71.0 159.7 For the LN2 amount, due to the high injection pressure and fracture conductivity of C23 and C24, the LN2 did not sufficiently cool the fracture surface before escaping to the ambient. Thus, despite the LN2 consumptions are higher, the warm-up times were shorter due to the waste of the injected LN2. For C26, which also had high fracture conductivity, its internal fracture surfaces were much larger than C23 and C24 and the injection pressure was slightly lower. Its internal cooled rock volume would be much larger than C23 and C24. Therefore, its warm-up time would be longer.

The better cooling efficiency of the injected LN2 of C25 and C29s could be attributed to their low fracture conductivity (compared to other later fracture cases) and the direction of induced fracture (compared to vertical fracture cases). Their low conductive fractures make injected LN2 slowly flow through the induced fractures and better cool the rock volume around wellbore, which greatly reduces the waste of LN2 and extends the warm-up time of the borehole temperature.

6.4 Difference between concrete and actual rock field samples

Since the actual rock field samples were more complex geologically than convrete samples, with much higher heterogeneity and more discontinuity than concrete samples, it was difficult to compare the experimental results directly with both types of rock samples. However, some similarities and differences were found by comparing the experimental results with both types of rock with similar fracture direction, distribution, and conductivity.

This comparison provides some qualitative insight although may not be very accurate or quantitative.

The field sample F5 had a major natural fracture partially in the vertical direction, while F4 and F6 have natural fractures in horizontal direction. Table 32 and Table 33 conclude the affecting factors and the warm-up time from these field samples. The correlation coefficients are omitted due to it had less meaning when compare the samples with different fracture directions. Figure 71 shows the comparison of the pressure decay curves of these field samples at 200 °C before any LN2 treatment.

Table 30: LN2 injection results of field samples

Sample # _ F4 _ F5 _ F6 LN2 amount (kg) 4.3 5.3 3.8

Injection time (mins) 91.6 4.9 4.6 T

_T ime for borehole filled with , . 6.5 _c 2.5 _c 1.9

LN2 (mins)

Injection pressure (psi) 160.0 150.0 70.0

Table 31: Warm-up time of LN2 injection treated concrete samples with lateral fractures

Borehole Warm-up time temperature ... ...

(degC) ^{F4 F5 F6}

130 120.9 31.2 31.9

140 128.1 40.0 39.1

150 138.7 55.6 51.2

160 154.7 77.9 68.0

Based on the information shown, the field sample F6 had more similar testing parameters with concrete sample C26. They both have very good fracture conductivity (C26 is better than F6), similar injected amount of LN2 (3.8 kg for F6 and 4 kg for C26), and injection time (4.6 mins for F6 and 4.2 for C26). The internal fracture surface area should be larger than F6 considering the induced fracture in C26 propagated in two distinct directions, while F6 only have 2 small fracture outlets on its faces 3 and 4 of the sample block (see Figure 5).

In relation to the cooling effect of the LN2 injection treatment, the concrete sample C26 had advantages of larger fracture surface area, lower thermal diffusivity, and very slightly more injected LN2 (~5%). On the other hand, the field sample F6 had advantages of lower fracture conductivity and longer injection time. In addition, the steel plates holding F6 during injection treatment may also have slowed down the heat exchange on its top and bottom surfaces.

Considering these affecting factors, C26 should have better performance during the warm-up process mostly due to its large internal fracture surface area.

A comparison of the warm-up time for these two samples showed that the warm-up time for C26 was slightly longer (about 4.4% longer for temperature of 160 °C).

6.5 From laboratory scale to actual rock field scale

The laboratory experiments using 20x20x20 cm cubic block samples provided insights for LN2 cooling and subsequent warm-up of samples.

As described above, several key factors for the warm-up of LN2 treated rock samples identified in the laboratory experiments are as follows: Although higher thermal diffusivity is preferred for the cooling process, it also makes the warm-up process faster since heat can be conducted faster through high thermal diffusivity medium.

The existence of fractures has a significant impact on warm-up.

Fracture direction was an important factor affecting warm-up times of LN2 injection treated samples.

Lateral fractures have a greater impact than vertical fractures on warm-up times. Lateral fractures cool a rock volume around a borehole, which retards warm-up from almost all directions of the borehole, while the heat transfer impact of vertical fractures is only in a direction perpendicular to the fracture plane.

The mechanisms of fracture conductivity are complex.

Fracture conductivity should be large enough to ensure injection of LN2 before LN2 evaporates. On the other hand, fracture conductivity should not be too large, otherwise injected LN2 will quickly flow through fractures and escape from a sample without sufficiently cooling down the internal volume of the rock in the sample.

Generally speaking, the lower the fracture conductivity, the higher the utilization efficiency of LN2.

The internal fracture surface area is also important. Injected LN2 flowed through fractures and cooled the surface area down during LN2 injection treatment. With larger fracture areas, more internal rock volume was exposed for cooling, which lead to longer warm-up times. Larger fracture areas resulted in higher fracture conductivity. In the case of the same fracture conductivity, larger fracture areas produce better results for warm-up times.

The amount of injected LN2 is not important under laboratory conditions. This observation contradicts original expectations. However, due to the dimension limitation of the rock samples, most injected LN2 could absorb the heat inside the rock samples before exiting, resulting in some LN2 being wasted. The controlling factor on this process is fracture conductivity.

Injection time is more important than the amount of LN2 for lateral fractures because vaporized nitrogen cooled the near borehole area before LN2 entered fractures. Thus, under the condition of low utilization efficiency of LN2, injection time become a more important factor in samples with lateral fractures. When applying LN2 cooling to actual rock samples, there were differences between the actual rock sample results and the laboratory results with concrete samples which were partly due to complexity of geological properties.

The following are observations and recommendations to be taken into account when applying the experimental findings described above to actual rock field operations:

Judging from the actual rock field samples, the compositions of rocks in the field can be extremely heterogeneous. As lower thermal diffusivity is preferred for longer warm-up times, it is likely to be difficult to predict warm-up times accurately due to the heterogeneity. An overestimation of warm-up times might be important to ensure operation safety, despite upscaling of dimensions resulting in averaging varying rock properties.

Since natural fractures are very common in target formations, the directions of these fractures are a major factor affecting the cooling and warm-up processes. Lateral fractures are likely to have a greater impact than vertical fractures on slowing down the warm-up process. Considering thermal gradients, a lateral fracture near the bottom hole is an ideal case.

Fracture conductivity is likely to be much less important for warm-up times process. In the laboratory, fracture conductivity determined the amount of wasted LN2. However, in a larger scale operation, fractures are unlikely to reach a surface and cause LN2 leaks to the atmosphere. Almost all injected nitrogen is likely to stay in the formation within the time period of interest and cool the formation. The amount of injected LN2 is likely to be a controlling factor for warm-up time. Since heat conduction is a fairly slow process in large scale operations, the more LN2 introduced into the system will lead to a longer time for temperature to recover.

The injection time is not likely to be a major factor for warm-up time.

7.0 Conclusions

LN2 treatment is an opportunity for an efficient, environment friendly, and economical option for borehole cooling operations in a hot pit.

The above-described series of experiments can be summarised as follows:

The laboratory equipment was set up to accommodate LN2 circulation/inj ection treatment on 20x20x20 cm cubic rock blocks at high temperature (100 and 200 °C). The basic thermal, mechanical, and hydrological properties of both concrete and actual rock samples were measured.

A set of characterization methods was designed, proved, and conducted to quantify the LN2 treatment experiments. - In total, more than 30 rock samples were treated with LN2 to evaluate the warmup process after LN2 injection under different conditions.

Several key factors affecting the warm-up process were pin-pointed for the 20x20x20 cm rock samples treated with LN2 under laboratory condition based on the experiment results. - Potential differences between lab-scale experiment and field-scale operation were identified.

For the laboratory study, the most important factors affecting the warm-up process include thermal diffusivity, fracture direction, fracture conductivity, fracture surface area, and injection time (for later fracture cases). The laboratory study was limited to an extent by the small dimensions of the 8-inch blocks, which lead to a significant amount of LN2 being wasted during LN2 treatment.

For the field samples, the most important factors included heterogeneity, natural fracture direction, and total amount of injected LN2.

Claims

1. A method of fracturing and cooling a rock mass that includes injecting a cryogenic fluid into a hole in a floor of a pit in a mine, with the cryogenic fluid fracturing a rock mass surrounding the hole and cooling the surrounding rock mass via heat exchange with the rock mass.

2. The method defined in claim 1 wherein the cryogenic fluid includes a cryogenic liquid and the pit is a hot pit, and the method includes injecting the cryogenic liquid into the hole in the floor of the hot pit forming a cooled gas in the hole that fractures a rock mass surrounding the hole, with the gas cooling the surrounding rock mass via heat exchange with the rock mass.

3. The method defined in claim 1 or claim 2 includes continuously or periodically discharging gas from the hole to remove heat from the hole and the surrounding rock mass.

4. The method defined in any one of the preceding claims includes continuously or periodically discharging gas from the hole at least while injecting the cryogenic fluid into the hole.

5. The method defined in any one of the preceding claims wherein the cryogenic fluid is liquid nitrogen.

6. The method defined in any one of the preceding claims includes:

(a) drilling a hole in the rock mass,

(c) positioning a well-head in an upper opening of the hole and thereby closing the opening, the well-head having an injection opening for the cryogenic fluid and optionally a pressure relief valve, and

(d) injecting the cryogenic fluid into the hole.

63

7. The method defined in any one of the preceding claims includes selecting the cryogenic fluid to be a cryogenic liquid and selecting the amount of the cryogenic fluid for injection into the hole having regard to factors, such as (but not limited to) the depth of the hole and the geology and the amount of the rock mass to be cooled to a selected temperature for a selected time.

8. The method defined in any one of the preceding claims includes continuously or periodically discharging gas from the hole to remove heat from the hole and the surrounding rock mass optionally while injecting the cryogenic fluid into the hole.

9. The method defined in any one of claims 6 to 8 includes injecting the cryogenic fluid into a plurality of spaced-apart holes in a section of a mine pit to be mined and injecting the cryogenic fluid into the holes and fracturing a rock mass surrounding each hole and cooling the surrounding rock mass via heat exchange with the rock mass.

10. The method defined in claim 9 includes selecting the spacing of the holes to be the spacing required for a subsequent explosive blasting program so that the same holes can be used for injecting the cryogenic fluid and for explosives.

11. A method of drilling and blasting a section of a mine pit to be mined that includes:

(a) injecting a cryogenic fluid into a rock mass in a section of a mine pit and lowering the temperature of the rock mass; and

(b) blasting the rock mass.

12. The method defined in claim 11 includes drilling a plurality of spaced-apart holes in the section of the mine, injecting the cryogenic fluid into the holes and fracturing a rock mass surrounding each hole and cooling the surrounding rock mass via heat exchange with the rock mass, positioning explosives in the same or newly-drilled holes, and blasting the rock mass.

13. The method defined in claim 12 includes selecting the spacing of the holes to be the spacing required for a subsequent explosive blasting program so that the same holes can be used for injecting the cryogenic fluid and for explosives.

64

14. A method of mining that includes:

(a) injecting a cryogenic fluid into a rock mass in a section of a mine pit to be mined and lowering the temperature of the rock mass,

(b) blasting the rock mass,

(c) excavating the blasted rock mass, and

(d) transferring the excavated rock mass to downstream processing operations.

15. The method defined in claim 14 includes drilling a plurality of spaced-apart holes in the section of the mine, injecting the cryogenic fluid into the holes and fracturing a rock mass surrounding each hole and cooling the surrounding rock mass via heat exchange with the rock mass, positioning explosives in the same or newly-drilled holes, and blasting the rock mass.

16. The method defined in claim 15 includes selecting the spacing of the holes to be the spacing required for a subsequent explosive blasting program so that the same holes can be used for injecting the cryogenic fluid and for explosives.

17. The method defined in any one of the preceding claims includes lowering the temperature of the rock mass at least 10°C and typically by at least 15°C.

18. The method defined in any one of the preceding claims includes lowering the temperature of the rock mass between 20°C and 50°C, typically between 20°C and 30°C.

19. The method defined in any one of the preceding claims includes lowering the temperature of the rock mass to a suitable temperature for drilling and blasting the rock mass for up to 8 hours, typically up to 10 hours and more typically up to 12 hours.

20. A system for selectively cooling and fracturing a rock mass in a section of the mine to be mined, the system including:

65 (a) a plurality of drilled holes with openings and well-heads that close the openings, each well-head having an injection opening for a cryogenic liquid and optionally a pressure relief valve; and

(b) an apparatus for injecting the cryogenic fluid into the holes via the well-heads.

21. The system defined in claim 20, wherein the apparatus (b) includes a source of the cryogenic fluid and pipe works for transferring the cryogenic liquid to the well-heads.

66